a AUS AE LOT) | ET LOT, EX TE = kalt DRE sa tre \ aus i asl 4 Ait ath ok ET De von RER 68 Sy ni Ju DELA he rs rs 4 . ‘ AR LA AD IR NETTER A a à i à M8 ns a al Er ne À a AS Pee Er DAUERTE TEE ad abr a 2 de Las tt UE ETS hy x i 0 à TENTONTA x 2 Bo Rae OAS nee Rm 2 1 Pt A À We Dante Nee EX oh Mr 4 aA 40 DR} CS US RTE 0 tL wath aa \* Hye Mae ay LO Mary CRUE RONA NL RDA al feat Te CANCER CANON EN EN ER ER I DOTE . 0 : Sak ee dS Da See Le INT 7: ere: EIERN ; man ë CURE WN tle We PTE Ay a da ORO Le TA MS at ab KA rs PLANE bey 4 KALT RETRO . Ran Ast IR die œ Bahl mn ua lan mihnne? 2 ü M: 2 “ia Hie and Ehre D “i À ayo tall he N I Von atthe Pt ot om a GOUT LETTRE , ni a a & à dd 6 a lt LAC PIED WIE vt J D \ io ’ AIS STV foe gi i A SE EST p | Re Meal RC ET thors nnd ‘ i AL ee et LA 6 N RUE DATE he hee ere Puree ae eat 4: ñ F : An Mp dr, Dean de Nee Le Ho ne ed BS ete Se he LI Ae ant. Pith ahs LE Ê MAMIE (RUES ET Te ea h Mn ne En DL EEE ae IR RS Er Im 3 DT hund > AN Te de Katie an. J Nie beh EA ADIL ML ihe PRT yn SIN PE LA TS SC teary ’ DIR An RAA) CES RAR edi (a 46 A fe HAE à ; 34 DVS Satie Dh 9 ¥ si De DEN Ne NN ben Hann dés de thy, aus r SPY AAR AP MAIR ENTER STA Ta ke abt hae Der ah BN Saal + Ier é SI Rit de Le MOP JL be IK Hi a Lis Sead m DES et BR TR bt ; à ‘ 9 ME an Na D RER à eh LI Free Ve DA dE HA A NS HA Made DEL R 3 er 5 ! EASTERN ni Da mb ie m Amine mann DE COMFORT E DEEE TES) PACD Bs de od Hen Dl DAL Tn PHORM mm EA am Me halvun à ann CETTE # à r . Pro Ne B® Final tell Rhian ln 2 24 ag a 3: 10 4 db. Eu tot Ji Sarg DRE MN en enable D ¥ DATE SE ty, N nen Ya NN te Fu Rn one HM In Bite LA PAN E x à ‘ e 2 IN Fendt ain Panel Ve Paume nn Le POINTER à D Ka < * u DORE Man Zi Nid Gat ofa tan bow a el} PS 10 6 54 aoe kam fay entr QUE TES ROLE CE DER CS DLL HOS NSE aM sey We Sing ENE Ak SBN A ie 5 Dar in aa MONIES Se 5 2 we Bag! ’ 3 PRES ce IES on E + # 4 He Ti te J N tn No a he ihe Le ea ee wel a Mal at a ee, itor ibang ae i BeMhagit Akon Ce Eee ee à heh Mn most : * i Pt betel td ku ren À 3 ft i ù Tail elton aay ~ MIN ORO yd 29 Pate Zug À É F Me te dite TELE A - « : PRAT ewe, D ha the 2 AN... $ J 5 oe I “ Ann Wi ten aad as a EE va 3 F : TEN BR a Bo tod oe Be ste Em Pa Mente Blan mee Tee yy hn te ch, ah S A LOUIS he Mis Sh ak Te FN Hane an Anh Ue Rate tye “ Just Aah Med 8e r nn De a no Hi, Tete Matin anale à à LL DA N De don I on adhe te in RE En a IL re fee Sanne at Tome ON en AS, 2 à 2 E La nn 2. À Dur ar ns + EN do an ANDY Win or et N SRE 2 Run eins le de Bo Te de da nein Da De HE on ee y : r = \ nt. men Aie Vu x ; x > . RE À Le Ve Al M et on, i i MLAS Sewn, Dm com nas onde À nn, > . men PUR Er) . N 4 : DEE Er Tee DAC Mee VERT SEEN RER En er ellen Don Tea do I Tann > poet : ue Pe Nee Vin a on >. 7 > ES md, . . Sree ran BEER LER CARE sbi a pa at hie FD WT WS : RP TRES eat me ont | aM a ARTEN N. Sorted mn Po a Ve SRT a nie Reta tie we ET ea ot Tann lan : cpap eran reir ete dre began in SS Sot i Or SS a Snir ee 2 Ge Wok Pei ms at 7 2 IE IE ae TxA IR eer Se CS irs THA ya RE ee ee Mr Pe a ithe Be rl i ame EA teen de ER BELLA ; ds RR 29 ae ET : EEE: Ex fas abat à = EN HEN ER ES : 34 AU my \ ¥. Pe ce RER a is D D Ve er SR 4 ... le t . = “ Se aaa Per a a en nn {AFTER se N SY = WSS om CMA = \E ] CAS f2 AX ea | al z ay z LIBRARIES SMITHSONIAN INSTITUTION NOILNLILSNI NVINOSHLINS S31uvVYa17 L | = = z - 2 ri | 2 mee) o = NE oO D | = ‘i Pe] = D S N = D 5 | . > hes : 2, E = = 3 En = = - = m Cons = (47) » one! [47] 4 NOILNLILSNI SAlYVUdIT LIBRARIES SMITHSONIAN Ni | 2 wm <= eee n z nO . = = = N = < = \ = | = | = = F a g à 8 à 2; = 2 = 2 = x de > "€ ee DURS LIBRARI ES SMITHSONIAN _ INSTITUTION, NOILNLILSNI NVINOSHLINS Sa Iyvy g iu i > > te 23 ul AT) a NW | x. Là wy es: cc = NN œ 4 / LP, =. < À } NN < = a Greg c œ = N c\% = 2% = = 70 6 ! — À. z ac ar i 2 2 ay \ NOILNLLLSNI_NVINO HLIWS Sa Iuvyg itl BRARI ES_ SMITHSONIAN_ INSTITUTION "NO En z Ir Z 5 rm z 2 zit 2 > > ae > | SANE E > = <> a = WOE > EGG: E -IBRARIES SMITHSONIAN INSTITUTION NOILNLILSNI NVINOSHLIWS S3IyvuagıT LIE . 2 > Z *« un = un ; = 4 = < = ; < = < N = 5 2 Le = 5 . ” en an Ç A on on D h : : 8 Gf? - : et ie RL : 3 _NVINOSHLUINS Sa Iuvyg II_LIBRARI ES SMITHSONIAN_ INSTITUTION 7 RE GER “ Gy à = = Of =a œ a. [04 ot << =. < Fe = S PS c IE = = = = = 5 (os = m» = m 2 a i = | er 2 3 IBRARIES SMITHSONIAN INSTITUTION NOILNLILSNI NVINOSHLINS S31YvVug17 LIB Z = m Er 2 rm = Pa: 2 à ae © = P > > ZN 5 2 E = FE = E = 4 - a m i = ty ainda : 2 2 | LEN wot Iuvag11 SMITHSONIAN INSTITUTION NOI. = (22) Z iD a = = RE: = < Nu © re) TPS GO. = 5 T À 2 EGY) 3 = 4 5 2 E 7 2: 2 = 23 | 7 2 | a » À = . 2 m IBRARI ES SMITHSONIAN _ INSTITUTION NOLLNLILSNI NVINOSHLIWS $3 ıavaaıı LIEB 2 > au > | u «nn ud = NR Lu = A au “A 2 4 NN = > m <x * — < pr NN = = = = : S KR = = Bae) i = 5 a 5 ay 2 24 > CADRES i OILNLILSNI_NVINOSHLIWS $3 luvug ae LIBRARI ES_ SMITHSONIAN” INSTITUTION” NOI, SU u N, ER a mn Bla: En LOIR + m VW Af 4 NE TR ff fA al Ne NR Ee AT = — ros aul z — SMITHSONIAN INSTITUTION NOILNLILSNI NVINOSHLINS S31YV4811 LIBRARIES = it = EI ae z cs | By ‘Ow = ee \ © wo | 5 2 5 2 = 2 > = = = > 7 = ip 2 m ei DO © z i ie 2 NVINOSHLINS S3IYVYUdEIT LIBRARIES SMITHSONIAN INSTITUTION NOILNLILSNI = [22] Z u... n = NL. on < = .„£ | = < I = = = = aa = a — vat 2 à - = E = = nine N = > = & = = ‘bh > 4 dé ERRITHSONIAN INSTITUTION NOILNLILSNI NVINOSHLIWS S 3 IYV#811 LIBRARIES LJ Er te = < SN 4 > x i NN ead — = \ = a = S31#VY#917 LIBRARIES SMITHSONIAN INSTITUTION NOILNLILSNI INSTITUTION NOILNLILSNI INSTITUTION NOILNLILSNI s3lyvyg11 INSTITUTION SMITHSONIAN NS En | ITUTION NOILNLILSNI _NVINOSHLINS SE NVINOSHLINS S314V48911 LIBRARIE N SAIMVYUGIT LIBRARIES SM NVINOSHLINS" S3INVYGIT-LIBRARI NVINOSHLINS SMITHSONIAN SMITHSONIAN NVINOSHLINS INSTITUTION NOILNLILSN Z < z oO 7) x = = 17) 5 u 5; u fe = œ = x An x = a = < dr < = œ = = = re S = ° > S - z 3 2 ur Z = SMITHSONIAN INSTITUTION NOILNLILSNI NVINOSHLINS S314#V44171 LIBRARIES = = (e] = Oo = = > ra > : = > 14 © = a i = = m = m 2 m = w == (47) Ne ao | NVINOSHLINS S314vV4417 SMITHSONIAN =, ao = (22) z (29) < = / „= = < = = pa ae 4 Z = z = 2 EES 2K EDI 2 = = 4 E <= = 2 = > = ae) oe = > wn za (79) “ me wo = = > SMITHSONIAN _ INSTITUTION NOILNLILSNI NVINOSHLINS S31UVY8917 ui Z u AN 3 4 [22] Ex Q NN X. en oc a AWS œ = <<; pe. < = RW Se Ge = Œ = & YQ. a = = ? O ai 5 ale ° = = = z Toye LS z | NVINOSHLINS SSIYVYEIT LIBRARIES SMITHSONIAN INSTITUTION NOILNLILSN aie z = z = = — wm = ow = TIT; STE SAN, LME, LLIINÈN, LEON, Mi AV Bh. 4 4 "10. in See eae journal should be checked for st rt essary. d for. Nota lepidopterologica Vol. 17 No. 1/2 Basel, 30.X1.1994 ISSN 0342-7536 Editor : Steven E. Whitebread, Maispracherstrasse 51, CH-4312 Magden, Switzerland. FAX : +41-61-8412238. Assistant Editors : Emmanuel de Bros (Binningen, CH) PD Dr. Andreas Erhardt (Binningen, CH) PD Dr. Hansjürg Geiger (Berne, CH) Dr. Alexander Pelzer (Wennigsen, D) Contents — Inhalt — Sommaire FIEDLER, K. & SAAM, C. : Does ant-attendance influence development in 5 European Lycaenidae butterfly species? .…................................... 5 FIEDLER, K. & SCHURIAN, K. G. : Oviposition behaviour in Lycaena Breisgau onl(liyeaenidae) 322.132 2H en. daran. „ann. 25 Hausmann, A. : Morphology and taxonomy of the species belonging to the genus Myinodes Meyrick, 1892 (Geometridae) ........................ 3 Kozıov, M. V. : Geographical variation in wing pattern of Micropterix maschukella Alphéraky, 1876 (Micropterigidae) .............................. 45 KRISTAL, P. M., HIRNEISEN, N. & STEINER, A. : Eine weitere endemische Hepialide aus den Alpen : Pharmacis claudiae sp.n. (Hepialidae) ...... 38 PATOCKA, J. : Die Puppen der Tribus Cyclophorini Mitteleuropas (Geo- MECS) MRC ee nu heine TE EE ote 73 Rıepı, T. : Une nouvelle espèce européenne du genre Pancalia Stephens (@Eosmoplezioidas, Antegueninag)), z... nme 87 Obituary — Nekrolog — Necrologie meee berhiard 9% erenin (1902-1993). RR. in INN Lars 93 Book reviews — Buchbesprechungen — Analyses .................... 30, 44, 100 Notices Rundschreiben'betreffend Dr. Erst PREESNER 4... 2 REA EN TARDE A Bee re ee ER ee ee ee D Nota lepid. 17 (1/2) : 2-4 ; 30.X1.1994 ISSN 0342-7536 Rundschreiben betreffend Dr. Ernst PRIESNER Sehr geehrte Damen und Herren, liebe Kolleginnen und Kollegen falls Sie es nicht schon von Freunden erfahren haben, möchten wir Ihnen mitteilen, daß unser Kollege Dr. Ernst Priesner seit Mitte Juli 1994 verschollen ist. Als Ergebnis intensiver Suchaktionen durch Freunde, Bergrettung, Hundes- taffeln, Hubschrauber und Polizei, die leider erfolglos verliefen, muß davon ausgegangen werden, daß er durch einen Unfall im Gebiet des Pflegersees bei Garmisch in den Bayerischen Alpen ums Leben gekommen ist. Wie Sie alle wissen, hat sich Ernst Priesner seit vielen Jahren intensiv mit der Lockstoffsystematik beschäftigt und in dieser Fachdisziplin eine Weltspit- zenstellung eingenommen. Zahlreiche Fachkollegen hatten mit ihm Kontakt und unzählige gemeinsame Projekte wurden durchgeführt bzw. sind noch im Gange. Das mutmaßliche Ableben von Ernst Priesner ist nicht nur für die Wissenschaft ein unersetzlicher Verlust, es werden sich dadurch auch für viele Kollegen große Schwierigkeiten bei ihren weiteren Arbeiten ergeben. Deshalb gilt es sofort, den umfangreichen und für die Wissenschaft unersetz- lichen Nachlaß zu sichern, ihn sicher und übersichtlich zu lagern und für die Zukunft verfügbar zu halten. Ferner müssen alle Bemühungen dahin gehen, möglichst wenig Information aus nicht abgeschlossenen, laufenden Projekten zu verlieren und wichtige Daten jenen Kollegen zukommen zu lassen, deren Arbeiten beeinträchtigt werden oder ein Ende finden müssen. Das Max-Planck-Institut in Seewiesen sieht sich außerstande, die umfangreiche Bücher, Sonderdrucke, Karteien, handschriftlichen Notizen, Sammlungsteile, Negative, Fotos und Dias sowie die noch in den Labors vorhandenen Duft- stoffe und ihre Komponenten zu lagern und evident zu halten. Vielmehr wird dort dringend Labor- und Büroplatz benötigt. Nach Absprachen zwischen Frau Hedi Priesner, Prof.Dr. Kaissling, dem Ar- beitsgruppenleiter von Dr. Ernst Priesner am Max-Planck-Institut in Seewiesen, Dr. Gerhard Tarmann vom Tiroler Landesmuseum in Innsbruck und nach Genehmigung des Direktors des Institutes, Herrn Prof.Dr. Eberhard Gwinner, ging der gesamte Bestand, mit Ausnahme des chemischen Teils (Pheromone und die Substanzen zur Herstellung), geschlossen an das Tiroler Landesmuseum Ferdinandeum (Naturwissenschaften), da dort im neuerbauten Institut aus- 2 reichend Raum für eine Lagerung und Betreuung vorhanden ist. Die Bear- beitung der Bestände soll durch folgende Maßnahmen gesichert werden : 1. das Max-Planck-Institut leistet einen finanziellen Beitrag ; 2. das Tiroler Landesmuseum Ferdinandeum setzt seine Mitarbeiter zur Be- treuung der umfangreichen Literatur (hunderte Bücher, tausende Sonderdrucke und Kopien), der Sammlungen und des Fotoarchives ein und wird sich be- mühen, weitere Mitarbeiter für die Katalogisierung (EDV-mäßige Aufarbei- tung) zu bekommen ; 3. aus dem Verkauf von Literaturdoubleiten (Bücher und Zeitschriften, die am Ferdinandeum bereits vorhanden sind) sollen weitere Mittel zur Betreuung des Nachlasses, vorallem zur Katalogisierung und Evidenthaltung der zahllosen handschriftlichen Notizen und der Originalprotokolle aus den Gelände- und Laborarbeiten verfügbar gemacht werden ; 4. es wird ein Fond zur Bearbeitung des Nachlasses Ernst Priesner eingerichtet ; alle Kollegen, denen die rasche Aufarbeitung des wissenschaftlichen Nachlasses ein Anliegen ist, da sie Informationen daraus immer wieder brauchen bzw. jene, die durch die Verbindung mit Ernst Priesner wesentliche Impulse für ihre eigenen Forschungen erhalten haben, könne sich durch einen finanziellen Beitrag beteiligen ; 5. für die Aufarbeitung des Nachlasses sollen sachkundige Mitarbeiter stun- denweise engagiert werden. Der chemische Teil des Nachlasses muß in kompetente Hände gehen, das heißt, in eine Institution, die fachlich und personell in der Lage ist, mit den Substanzen zu hantieren und eventuell nach publizierten oder aufgefundenen Rezepturen Pheromone weiter zu produzieren. Diese Möglichkeiten sind am Ferdinandeum in Innsbruck nicht vorhanden. Durch Intervention von Dr. Nils Ryrholm aus Uppsala ist es gelungen, bei Dr. Peter Witzgall in Lund in Schweden eine solche Möglichkeit zu finden. So können die laufenden Projekte vielleicht doch noch einen Abschluß finden. Auch Prof. Clas Naumann in Bonn wird sich bemühen, mit seinen Mitarbeitern Teilgruppen organisatorisch zu betreuen (besonders Sesien- und Zygaenenpheromone). Wir möchten Sie bitten, falls sich für Sie wichtige Informationen in den vielen handschriftlichen Aufzeichnungen befinden, dies mit möglichst genauen Detail- angaben mitzuteilen. Nur so gibt es eine Chance, gezielt zu suchen und Ihnen Informationen rasch zukommen zu lassen. Bitte wenden Sie sich an Dr. Gerhard Tarmann, Tiroler Landesmuseum Ferdinandeum, Naturwissenschaften, Feldstraße 11a, Tel. : +43/512-587286 A-6020 Innsbruck Fax. : +43/512-58728640 Wir hoffen, mit diesen Aktivitäten und Ihrer Hilfe wenigstens die große wissen- schaftliche Lücke, die das mutmaßliche Ableben von Dr. Ernst Priesner für uns alle hinterläßt, so gering wie möglich zu halten. Dr. Gerhard TARMANN Summary Dr. Ernst Priesner, the well-known pheromone specialist, has been missing since the middle of July 1994, when he went to check some pheromone traps inthe Bavarian Alps, near Garmisch. Despite numerous search parties he could not be found and it must be assumed that he met with a fatal accident. Dr. Priesner’s contribution to pheromone research was immense and he led the world in the field of pheromone systematics. His research has resulted in an immense wealth of new faunistic and biological information. Many current scientific projects were dependent on his collaboration and their successful conclusion will now be very difficult. It was therefore considered imperative to save and make available the vast and irreplaceable scientific material he left behind. Apart from the chemicals, the entire contents of Dr. Priesner’s office and laboratory at the Max-Planck- Institute in Seewiesen, Bavaria, have been transferred to the Tiroler Landes- museum Ferdinandeum in Innsbruck, Austria, under the charge of Dr. Ger- hard Tarmann. This material consists of hundreds of books, thousands of scientific papers, card indexes, handwritten notes, photographs etc. A fund will be set up to finance the cataloguing and storage of the material. The Max- Planck-Institute will make a financial contribution and duplicate books will be sold. Donations to the fund are invited. The responsibility for the chemical contents of Dr. Priesner’s laboratory will be assumed by Dr. Peter Witzgall, Lund, Sweden, and Prof. Clas Naumann, Bonn, Germany will try to organise some of the projects (particularly those concerning sesiid and zygaenid pheromones). It is therefore hoped that the running projects can be brought to a conclusion. If anyone requires specific information from Dr. Priesner’s handwritten notes, he is invited to write to Dr. Gerhard Tarmann with as many details of the required information as possible. It is hoped that the measures taken will ensure that the loss caused by Dr. Pries- ner’s disappearance is kept to a minimum. Nota lepid. 17 (1/2) : 5-24 : 30.X1.1994 ISSN 0342-7536 Does ant-attendance influence development in 5 European Lycaenidae butterfly species ? (Lepidoptera) Konrad FIEDLER (!) & Christine SAAM Lehrstuhl für Verhaltensphysiologie, Theodor-Boveri-Biozentrum, Universität Würzburg, Am Hub- land, D-97074 Würzburg, Federal Republic of Germany Summary Caterpillars and pupae of 3 myrmecophilous (Aricia agestis, Polyommatus icarus, P. bellargus) and 2 myrmecoxenous lycaenid butterflies (Lycaena phlaeas, L. tityrus) were reared in the laboratory together with, or without, 2 species of tending Zasius ants (L. flavus, L. niger). Duration of development, mass gain, growth rates, prepupal and adult weights, and the ratio of mass gain per frass production were studied. There was no evidence for significant developmental costs associated with myrmecophily. Rather, we found some marginally beneficial effects of ant-attendance. Males of P icarus and L. phlaeas grew larger in the presence of ants. Mass gain per unit frass was slightly higher with ants in A. agestis (both sexes), P icarus and L. tityrus (males only). We found no consistent differences between the effects of the 2 ant species, nor between rearing treatments involving 2 or 5 L. flavus workers, respectively. Sexual and interspecific differences were documented in most of the parameters. These results show that certain myrmecoxenous and facul- tatively myrmecophilous lycaenid butterflies are able to compensate for their energetic costs associated with myrmecophily. The evolutionary consequences of such low-cost mutualisms are discussed. Zusammenfassung Raupen und Puppen von 3 fakultativ myrmekophilen (Aricia agestis, Polyom- matus icarus und P. bellargus) sowie 2 myrmekoxenen Bläulingsarten (Lycaena phlaeas, L. tityrus) wurden in Gegenwart von Ameisen (Lasius flavus bzw. L. niger) aufgezogen. Entwicklungsdauern, Wachstumsraten, Gewichte und Massenzunahme pro Kotproduktion wurden protokolliert. In keinem Fall ergab sich eine signifikante Beeinträchtigung dieser Parameter durch den Be- such von Ameisen und damit verbundene Sekretabgaben. Schwache positive Sn N correspondence should be addressed ; Tel. : 0931-8884321, Fax : 0931- Effekte konnten vereinzelt beobachtet werden (Gewicht der Männchen von P. icarus und L. phlaeas in Gegenwart von Ameisen größer, Massenzunahme pro Kotproduktion größer bei A. agestis, P. icarus und L. tityrus). Geschlechts- und Artunterschiede in den Entwicklungsparametern traten in den meisten Fällen auf. Die Ergebnisse zeigen, daß die untersuchten Bläulings-Ameisen- Interaktionen für die Schmetterlinge mit geringen, voll kompensierbaren Kosten verbunden sind. Die evolutive Bedeutung solcher Mutualismen mit niedriger Investition wird diskutiert. Resume Chenilles et chrysalides de 3 Lycénides myrmécophiles (Aricia agestis, Polyom- matus icarus et P. bellargus) et de 2 Lycénides myrmécoxènes (Lycaena phlaeas et L. tityrus) ont été élevées en laboratoire avec et sans 2 espèces de fourmis (Lasius flavus et L. niger). Les auteurs ont étudié la durée du développement, augmentation de taille, le taux de croissance, le poids des chenilles adultes et avant la chrysalidation, ainsi que l’augmentation de taille par rapport à la production d’excréments. Dans aucun cas, ils n’ont constaté une modification signicative de ces paramètres due à la visite des fourmis. Dans quelques cas, ils ont observé de faibles effets positifs de la myrmécophilie. Le poids des mâles de P icarus et L. phlaeas augmenta en présence des fourmis ; l’aug- mentation de taille par rapport à la production d’excréments fut plus marquée en présence des fourmis chez À. agestis (dans les deux sexes), P icarus et L. tityrus (mâles seulement). Les auteurs n’ont pas trouvé de différences consistantes entre les effets des deux espèces de fourmis, ni entre les élevages impliquant 2 ou 5 L. flavus (ouvrières). Des différences apparurent dans la plupart des paramètres du développement selon le sexe et l’espèce. Ces résultats montrent que les interactions Lycènes-Fourmis étudiées entraînent de faibles «coûts énergétiques», entièrement compensables. Les auteurs discutent de la signification évolutive de tels mutualismes «à faible coût». Introduction Many species of the butterfly family Lycaenidae live in association with ants throughout part of their larval and/or pupal stage (COTTRELL, 1984 ; PıERCE, 1987 ; FIEDLER, 1991). Ant-associations among lycaenids range from loose and unspecific, facultative interactions to obligatory and species-specific cases of mutualism or, rarely, parasitism. Typically, while feeding on their hostplants, the caterpillars attract ants with the help of chemical stimuli. This ant guard may provide protection against parasitoids or predators (PIERCE & EASTEAL, 1986; Pierce et al. 1987 ; but see PETERSON, 1993). Interactions between lycaenid immatures and ants are mainly mediated by secretions from specialized exocrine epidermal glands (MALIcky, 6 1969 ; CoTTRELL, 1984), although vibratory communication may be important in certain cases (DEVRIES, 1990). Three types of myrme- cophilous organs are known to play major roles. The dorsal nectar organ on the 7th abdominal segment, only present in larvae, secretes droplets of a clear liquid that contain carbohydrates and amino acids (Mascawirz et al., 1975; Pierce, 1983) upon tactile stimulation (TAUTZ & FIEDLER, 1992). So-called pore cupola organs, minute hair- derived glands distributed over the larval or pupal integument, appear to secrete amino acids or, in certain species, mimics of ant-pheromones (PIERCE, 1983). And the tentacle organs on the 8th abdominal segment of various lycaenid caterpillar species emit volatile compounds that cause an alerted behaviour in attendant ants (FIEDLER & MASCHWITZ, 1988 ; BALLMER & Pratt, 1992). All these secretions are produced at some energetic cost by the herbivorous caterpillars. In addition, the innervation and musculature of myrmecophilous organs and vibratory organs also cause metabolic costs. Previous studies on two lycaenid-ant systems demonstrated that ant- attendance may have a negative impact on larval and pupal develop- ment. In the obligatorily myrmecophilous Jalmenus evagoras (Dono- VAN, 1805) from Australia, ant-tended individuals develop faster than untended sisters, but attain lower weights (PIERCE et al., 1987). Larvae of J. evagoras are unable to compensate for nutrient loss to ants (BAYLIS & Pierce, 1992). As a consequence, myrmecophily is associated with significant fitness costs, since male mating success and female fecundity are strongly dependent on adult weight (ELGAR & Pierce, 1988 ; HILL & Prerce, 1989). In the Neotropical Arawacus lincoides (Draudt, [1919]), Rogsıns (1991) observed a slight retardation of larval develop- ment in response to ant-association, but weight was unaffected. Recently, however, beneficial effects of ant-attendance on larval develop- ment have been detected in three additional species. FIEDLER & HÖöLL- DOBLER (1992) found that ant-tended males of the Palearctic Polyom- matus icarus (Rottemburg, 1775) reach higher larval and pupal weights than untended controls. In the Nearctic Hemiargus isola (Reakirt, [1867]), tending by the ant Formica perpilosa enhanced caterpillar growth and thus adult weight, whereas two other ant species did not affect butterfly weight (WAGNER, 1993). These two butterfly species are facultative myrmecophiles, whose larvae associate with a variety of ant taxa and are not dependent on ant-association for survival. Overall, facultative myrmecophiles account for a larger proportion of the species diversity of the Lycaenidae than obligatory myrmecophiles (FIEDLER, 1991). Most recently, CUSHMAN et al. (1994) observed bene- ficial developmental effects in another Australian obligatory myrme- cophile, Paralucia aurifera (Blanchard, 1848). Hence, the prominent developmental costs of myrmecophily as found in J. evagoras might be atypical for the species majority. Therefore, a better understanding of the evolutionary and ecological significance of developmental costs or benefits arising from myrmecophily among Lycaenidae butterflies requires experimental work on a larger set of species from various taxonomic groups and representing all major types of myrmecophily. We here present the results of laboratory experiments with 5 European Lycaenidae species. Three of these possess all three types of myrme- cophilous organs and are facultative myrmecophiles, but show different degrees of myrmecophily. While mature larvae of Polyommatus bel- largus (Rottemburg, 1775) and Aricia agestis ([ Denis & Schiffermiiller], 1775) are rarely found without tending ants in nature, caterpillars of Polyommatus icarus are much less attractive to ants (e.g. THOMAS & LEWINGTON, 1991). P icarus was included to repeat the experiments of FIEDLER & HOLLDOBLER (1992) under a modified rearing regime. We also studied two myrmecoxenous species : Lycaena phlaeas (Lin- naeus, 1761) and L. tityrus (Poda, 1761). Larvae of these species neither possess a dorsal nectar organ nor tentacle organs, but they do have pore cupola organs. In laboratory experiments, weak and unstable ant- associations of these species can be induced and ants then harvest the PCO secretions (FIEDLER, 1991). Neither L. phlaeas nor L. tityrus larvae have hitherto been observed in association with ants in the field. Our aim was to investigate whether or not artificial ant-association involving two ant species and two different numbers of ants per indi- vidual lycaenid has any detectable effects on larval or pupal develop- ment. Furthermore, we wanted to know whether developmental effects differ between lycaenid species according to their degree of myrme- cophily. One might expect more distinct costs in caterpillars that are highly attractive to ants (P bellargus, A. agestis), whereas in myrme- coxenous species developmental costs should be absent or minimal. On the other extreme, secretion rates of obligatorily myrmecophilous caterpillars can be amazingly high (FIEDLER & MascHwirz, 1989 ; Fiedler, unpublished), corresponding to the pronounced developmental costs observed in species like the Australian Jalmenus evagoras (BAYLIS & Pierce, 1992). Finally, the developmental constraints on phyto- predacious lycaenid larvae parasitizing inside ant colonies (alike the Palaearctic Maculinea spp.) are entirely different (e.g. THOMAS et al, 1993, and references therein). Hence, a comparative survey across a larger number of species representing various stages of myrmecophily appears rewarding. 8 Material and methods Butterfly rearing Caterpillars were reared from eggs laid by field-caught (Aricia agestis, Polyommatus bellargus, P. icarus, Lycaena phlaeas, L. tityrus) or laboratory-bred females (A. agestis, P. icarus). Livestock originated from northern Bavarian populations except in ?. bellargus, where part of the experiments was conducted with individuals from southern France. Butterflies were kept in a greenhouse for oviposition or mating. Rearing procedures largely followed those described by SCHURIAN (1989). Eggs were collected every second day and transferred ınto a climatic chamber (25°C, 16:8 h L:D), where the whole development to adult eclosion took place. Groups of first instar larvae were placed in translucent plastic containers (125 ml) lined with moist filter paper. They were fed with cut foliage or inflorescences of appropriate host- plants : Geranium molle L. (Geraniaceae) leaves (A. agestis) ; Coronilla varia L. (Fabaceae) leaves (P. bellargus) ; Medicago sativa L. (Fabaceae) inflorescences (P. icarus); Rumex acetosa L. (Polygonaceae) leaves (Lycaena phlaeas, L. tityrus). Food was exchanged daily. Special care was taken to provide food of approximately equal quality ad libitum throughout the season, since food quality may affect larval myrme- cophily (FIEDLER, 1990 ; Bayis & Pierce, 1991). Ants Two ant species were used. Lasius niger (Linnaeus, 1758) is a common species of open grasslands and is well known to tend a number of lycaenid species in the field (FIEDLER, 1991). L. niger ants are omni- vorous ; they feed on insect prey and collect honeydew or similar energy-rich fluids. Lasius flavus (Fabricius, 1781) is also very common and lives largely subterranean in European grasslands. Its diet consists almost entirely of honeydew produced by root aphids. Because of their subterranean life, L. flavus ants rarely tend lycaenid larvae in nature, but they readily show trophobiotic interactions with lycaenid imma- tures in the laboratory (FIEDLER, 1991). Ant colonies were kept in the laboratory (at 22-26°C) in earth nests (L. flavus), or in artificial nest chambers the bottom of which consisted of plaster of Paris (L. niger). Ants were fed with honey-water and dead insects (mostly cockroaches) as needed. Experiments Experiments started at the beginning of the third larval instar, when the myrmecophilous glands of P icarus, P. bellargus and A. agestis 9 become functional. Experimental caterpillars were reared singly in translucent plastic vials (125 ml) equipped as above and were randomly subjected to one of the 4 following treatments: kept with 5 L. flavus workers ; with 2 L. flavus workers; with 2 L. niger workers ; and controls reared without ants. Food and filter paper were exchanged daily, and the entire larval frass was collected. Ants that had died during the experiment were replaced by nestmates to ensure a constant number of tending ants throughout. The procedure continued during the pupal stage. Daily inspections of every individual larva and pupa confirmed that all immatures of the 3 myrmecophilous species were constantly tended by their ant guard. In the 2 Lycaena species, ant-associations were likewise regularly observed, although occasionally larvae were seen without tending ants for short periods of time. Each experiment lasted until the adult butterfly eclosed from the pupa. All individuals that died prior to eclosion were discarded from the analyses. Mortality rates did not differ between the ant treatments within each species (SAAM, 1993). In total, 358 butterflies were reared to maturity (see SAAM, 1993 for further details). Every individual was weighed 5 times: at the beginning of the ex- periment (freshly moulted L3 : initial larval weight) ; as immobile non- feeding prepupa within a few hours prior to pupation (prepupal weight) ; as freshly moulted pupa (initial pupal weight) ; as fully pigmented pupa within 6 h before eclosion ; and as freshly eclosed adult after emission of the meconium (adult weight). The whole frass production over the third and fourth larval instar was collected, dried in an oven at 65°C to constant weight, and then weighed. All weights were recorded to the nearest 0.1 mg using an electronic Sartorius BA 61 balance. In addition, the durations of the third plus fourth larval instars and of the pupal stage were recorded (in days). From these data the following additional parameters were calculated : PERCENT PUPAL WEIGHT Loss : (Initial pupal weight - final pupal weight) x 100 / (initial pupal weight). Pupal weight losses always occur during development, but could be enhanced by the delivery of pupal secretions (e.g. from pore cupola organs) to ants. RELATIVE GROWTH RATE : RGR = (mass gained in the third plus fourth larval instar) / (prepupal weight X larval duration). The ratio : (larval mass gain) / (total frass production). This is a rough estimate for the efficiency of the conversion of ingested food into biomass. Assuming that the digestibility of food is not affected by the 10 presence of ants (see BayLis & PıErce, 1992), frass production is pro- portional to food consumption, if food quality is kept constant. The data were analysed statistically using ANOVA (with sex and treat- ment as factors) for multiple comparisons, and Mann-Whitney U-tests for comparisons between pairs of samples where normalıty or homo- geneity of variances were not met (SACHS, 1992). Percent data were arcsine-transformed prior to analysis. All P values given refer to 2-tailed tests. Results Duration of larval and pupal development The duration of the third plus fourth instar (Table 1) was largely unaffected by the presence of ants in A. agestis (mean values per experimental series 6-9 d), P. icarus (10-12 d), P bellargus (16-23 d), L. phlaeas (8-10 d) and L. tityrus (11-15 d). Sex differences in the duration of larval development occurred in all 5 species and in almost all treatments. Females generally took 1-2 d longer than males until pupation, but this difference was less distinct in experiments with ? icarus and L. phlaeas. The duration of the pupal stage was likewise not influenced by ant- association in all 5 lycaenid species tested. Sex differences in pupal duration were minimal (males faster in L. tityrus and P. bellargus). The pupal stage of A. agestis, L. phlaeas and L. tityrus took 7-8 d, that of P icarus 9-10 d, and in P bellargus pupal development lasted 11-13 d under the rearing conditions. Prepupal weights (Table 2) Caterpillar mass at the end of the larval stage was independent of ant-association in A. agestis (means 71-82 mg), P bellargus (means 107-126 mg) and L. tityrus (118-127 mg). P icarus males reared in the presence of ants (94.55 + 1.33 mg) were consistently heavier than control males (89.46 + 5.44 mg), although the difference was not signi- ficant. However, ant-association fostered a sex difference in P icarus. Males reared in the presence of ants were significantly heavier than females (U 3.3; = 195 ; Z = 2.825 ; P < 0.01), whereas control males and females reached equal prepupal weights (U7..; = 45; P > 0.5). A similar pattern occurred in L. phlaeas. Male prepupae reared in association with ants (89.6 + 3.05 mg) were significantly heavier than controls (73.8 + 2.34 mg; Ujo..4 = 38; Z = 3.099; P < 0.01), 11 xxx LO EL = A : X9S SU po 0 = 4 : SJUB supe 0 = A BEN su cdc =A : syue *86 9 = 4 DRS su 39°C = : syue su 90°] =H DER SU ÿ6 0 = 4 : syue *+L0 8 = A : X9S SU ÿL0 = 4 : SJuE 19974 (STD IEOF TE II (ZZ) 8t'0 F SI EI (ve) 91 0 + E98 (LZ) 81 0 F €£'6 (91) 89°0 F £9'8I (IT) SIO F 70T (TE) STOF L601 (TI STOFESII (ET) ITO + 6€ L (6€) IT’O + 9€'8 pauiquio9 SJU9U)E9I) JUV kkkCS OT = 4 XS SU 1l/.0=J4 : JU9UIJBII] «II =a WER SU OCT =A : Juaunean *90 7 =A Bros *16€ =A : JUS 91] +C6T =A Bros SU LEO = 4 : Juaumeaı] x*x* 16 CI = A 2 X9S MYC = al : JU90198297) Bela (ID) 80 FLTII (L) 890 F Er El (OL) 67 0 + 08'8 (S) 0r'0 + 0r'8 (2) L9'T + LS'61 (8) ITI F OS TC (L) LEO FLT (EI) SPOF 79'TI (8) SVOFSLL (LI) 0€ 0 + €S'8 syue }nOUJI M (8) I1L'0 F 00 ZI (6) LL'0 + 68 EI (9) LL’O + £8'8 (11) ¥7'0 + LT'6 () 91 I = 00'07 (2) OL'0 FrI'6l (ai) CEO 25 Zi (9) 290 = OS'II MOSES (L) PE 0 F PIE 1961U SNISVT 7 (L) 890 = ET'II (9) 9L'0 + LIYI (L) I£'0 + ££'8 COAAE 76 (p) EC'T F OS'9I (8) 91 I + 05°02 (8) WO F EIOI (3) I€ 0 FSLTI (6) 670 = 9S'L (LI) €€:0 £ 678 SNAD]{ snıspT 7 (SID LG OSE N (L) p80 F LSTI (LL) LTOFTL8 (L) ES'0 + LS'6 (8) €6°0 + 00°61 (AN SEAN (11) p70 $= 0011 (8) WO FOS II (L) 00 + 98°9 (SI) 9€°0 + L0'8 SNAD]{ snıspT S SOIN SO[PUIO SNA] DUAGDIAT se sofewa-] spovjyd puoosAT soe sopews] snd4D]]29 SNIDWL.IWLOAJOT SIM soewsg SNADI1 SNIDWWOAJOT sae sopewag SUS28D DIDI Sa199dS 010 < d SU: 1000 > d xxx: 100 > d xx: 600 > d x ‘ l'0 > d + ‘VAONV ABM-OMI Wry 918 SONJEA | ‘(S194981q ur Sois a[duues) SIOIIS PIepueIs F SULIT 218 UDAIDH ‘SJUE SN1SP7T JO S9U9Sq8 10 JoU9S91d au} ur PaIea1 USM sowads JepıuseaAT ueadomng ddl Jo (sAep ur ‘resur yJınoF snjd pa) JuswdopsAap [eA] Jo uoneinq I SIgeL SU 60 0 = 4 SCH Su C0 0 = 4 : que SU GI T—=4A : X9S kx8T 8 = A : jue xx08°L = 4 > XOS su 66 I =H : SJUB su OT =A : X9S su QL'0 = 4 : sjue x€9 =A : X9S SU T0 0 = 4 : syue Pape (ST) 0€ 7 F S6'07I (TI) Tr € F LL ETI (PT) SOE + 09°68 (LT) 90°€ + LO'E6 (CON Ret SSA (HHI (IT) LL'T + 9S'601 (LE) €€°1 F SS'H6 (EZ) 091 = L0'88 (ST) LP IF 60°€L (Zr) 6S IF PL SL pourquios SJU9U)291) UY su 770 =A : X9S su 860 =H : JusunNean su Cp | =H : X9S +80 € = A : Juauıyean kk80 8 = Æ : X9S suc] [| =H : JUSUIJESJI] xOT 6 = A : X9S SU 6€ 0 = A : Juowgean xxOL'L= A : X9S su C00 = 4 : Juawean IREM (11) 10°S F 9S IZI (DIGS 22 IGG! (OI) ET + OS EL (9) 6£ 8 + cr'E8 (L) tT'S FLONI (8) 9L°S = 69'EII (L) bys = 9768 (EI) €L°7 + 61 68 (6) ILT + S6 EL (61) SLI FLY LL syue noygM (8) 98°€ F EI STI (6) LL'S F 10971 (9) LS'T + OS €6 (LI) 677 = 79°06 (bp) SI'S F ET STI (2) 9+'t = OI CII (TI) STTFELL6 (L) OL'E F IT 98 () CGE EGIL (L) Ov'S + I8'I8 dosıu SNISVT 7 (2) 9L'+t = LEI (9) 86°F + 00'171 (L) 9S'S F 0S'S8 (6) L£'9 + 0L'86 (bp) er 8 + ST SII (8) pp + £9'60I (8) SOT + 07 T6 (8) SET F £I 88 (OI) EOTF Zr IL (SI) 61 7 + 01 08 SNAD]{ SNISDT 7 (EI) IVE F ET 6II (2) LL'9 $= IT SIT (LD) OS'S + rc 88 (L) p8'b + L9'68 (8) SSI FESLIT (9) SS'9 F 7S 901 (LI) £TT + £9 76 (8) IST + £968 (8) 67 T F 9S 91 (LI) 907 + CO 9L SNAD]{ SNISDT S SOIN SOJEUIO SNAÂJ1] DUADIAT SOIN sopewsA spanjyd puovodT sey SEL] SNBADIJAG SNIVWIWUOAJOT soe sopewoJ SNADIL SNIVWWOAJOI ssfeW sopewuag S1S28D DIDI Sa199dS ‘OT'0< d SU : 100 > d xx: S00 > d x VAONYV ÂBM-OM] WOIF ore SOnJeA A ‘(SI9NJEIG UI SozIs a[duIES) SIOIIS PIBPUBJS F SURI 918 U9AIO) ‘SJUE S71SDT JO Duasge JO JOU9S91d ayJ ur PaIesı USM sarmads sepruseaA| uvadoing dA JO (1uS81oM Jom Bur ‘ages pedndaid ur) s}ysioM [VAIL] [EU € SIQEL whereas females showed no effect (mean weights 83.42 + 8.39 mg and 93107 == 3.00:m2 U, =D: Ze ea ale): A general sex difference in prepupal weights emerged only in 2 species. A. agestis males were smaller than females, in P bellargus males were larger. In P. icarus and L. phlaeas, the sex difference was only apparent in ant-tended individuals (males larger than females, see above). Females and males of L. tityrus reached similar prepupal weights. Signi- ficant statistical ant-sex interactions were not observed. Pupal weight loss In all 5 species, ant-association had no detectable influence on pupal weight loss. Mean weight losses accounted for 15-20% in A. agestis, P. icarus and L. phlaeas, but were slıghtly higher in L. tityrus (> 20% in females) and highest in P bellargus (average values per series 23-28%). Sex differences in pupal weight loss were distinct only in L. tityrus (female pupae lost more weight [22-24% on average] than males [average loss 17-19%]). Otherwise, there were neither sex differences nor significant ant-sex interactions. Adult weights (Table 3) Adult weights were not affected by ant-association in A. agestis, P. bellargus and L. tityrus. In P. icarus, ant-tended males (35.07 + 0.76 mg) were ca. 10% heavier than untended controls (31.27 + 1.21 mg; Uz3, = 50 ; Z= 2.20 ; P < 0.02), whereas females showed no effect (ant- sex interaction : Fj.67 = 3.92 ; P < 0.05). A similar effect was observed in L. phlaeas : ant-tended individuals (especially males : 31.41 = 1.70 mg) eclosed from the pupae at higher weights than untended controls (23.65 + 0.85 mg). As a consequence, there was no significant sex dif- ference in adult weights of ant-tended P icarus and L. phlaeas, whereas untended controls of both species showed a distinct size dimorphism (females heavier than males). In A. agestis there was a significant sex difference independent of ant-association, females being heavier than males. No significant size dimorphism occurred in P bellargus and L. tityrus. Growth rates (Table 4) In all 5 species tested, males tended to grow faster than females. This difference was only weakly developed in P. icarus and L. phlaeas, but was pronounced in the remaining 3 species. Only in A. agestis was there a weak trend that caterpillars in association with 5 Lasius flavus ants grew slightly faster than all others. In all, ant-association had no 14 SU 8ÿ I =A DER su00=4 : syue +trt=d : X9S x6P 6 =A : syue SU 89 0 — 4 Ixos su €8 0 = 4 : jue x60 vy =A : X9S SU p8 | =H : sSJUB xt 6€ = A MX AS SU [00 = 4 : sJUB Del COLO == ES Tt iA) CTL SE ee Gy (EI) OL I = IV IE (CO) PEASE RV ASS (OI) PIC + BEI (IT) 161 F 6S 8€ (TE) 9L'0 + LO'SE (OT) $9°0 = ET'SE CORSO E LIEST (Zr) S6 0 F LS'IE pauIquio9 syuawjean JuV SU LOT =H LOS SU CCI =a : 4uauyean sulsc=hH : XOS su cl c =A : uouean su0I=4 : X9S SU ZE O0 = 4 : uouean SU OCT =A axes SU 70 T =A : Juauean} #kkOL'SE = A : XOS su ç90 = 4 : Juaumeal} PISA (LI) PET + SI TV (2) OS'E = 90° ++ (OI) S8'0 + S9'ET (9) 00'€ = LT'0€ (L) 8€°€ FO EP (8) E67 + 09 Ir BIENEN (EI) LSI F +6 SE (6) bl I FISST (61) 871 F V7 TE syue noygM (L) TS'E F LE bP (6) 90'€ F IS’ Lt (9) p17 + SCIE (DIET EOC WTErFETEr (L) 0€°7 + 00'8€ (TI) SEI = O8 VE (S) 68°0 + PE EC (2) SO'l = IO'r7 (L) BEE F 96'SE 4951U SNISVT 7 (L) SVT + EI Ip (9) 09T + cr'9r (9) tt F SO'IE (8) TT + + OF LE (p) STE FET BE (8) 77 € F Sr Ir (BETT = I0'SE (8) +6 0 + 97'SE (OI) 69:0 + IC ST (81) 67 I F 99'0€ snapjf SNISDT 7 (EI) 97 + 686€ (L) SIT + CS OP (LI) ETEFEITE (L) LIT + 60'I€ (8) IS'E + OCP (9) Ly'v Fa'se (ID Or IF Ir'se (L) STI + PS 9€ (8) 160 FEIN (LI) 71 = TL'0€ snapjf SNISVT S ‘010 < 4 SU * 1000 > d xxx: S00>d x: l'O > d + VAONV AEM-OMJ OI ore sanyeA A ‘(S19%981Qq UI SozIs ajdues) SIOIIS PIepueIs F SURO 218 USAIL) SOIN SITE SNAÂ71] DUADIAT soe sopewag spanjyd vuov9dT sofeW sopewuag SNBADIJAG SNIPWWoOAJOT sae sopewag snıvaı SNIDWIWOAJOT SIP SAP, S1JSA8D DIDI sowads ‘SJUE SSD] JO S9U9Sq8 10 JdUaSoId au} ur pareaı uayMm saloads sepıuseaAT ueadoing SAT JO (WYSIOM JOM Zw) SJUSTOM INPY © SIgEL kkx[C VL = A : X9S WEG (0) al : SJuUB xxx LI =A 5 KOS suss0=4 |(ID 700042300] (8) +000 + €800| (2) S000 + 980'0 | (EI) 1000 F +800 : quoUTIvON} (L) +000 + rL0'0 | (6) +000 + ZL0'0| (9) 7000 = 890'0 | (2) $00'0 + 6/00 (8Z) Z00'0 + 80'0 (TT) €00'0 = €L0°0 SOIN SAP snadı DUIDIAT +60 =A OS SU 700 = 4 : Sue TAC 5 XOS su/.01=4 | (OI) £000 + 0110 | (9) 7000 +0110} (2) S000 + 911 0 | (ID) +00 0 + IIT0 : yuauıyeaı} (9) T10°0 + 9010 | (IT) £00°0 = SOI 0 | (6) £00°0 = SOI 0 | (2) S00°0 + TOI 0 (v7) TO0'0 = E110 (LZ) TO0'0 £ +OI 0 SOIN sspewsg spopjyd DuavdAT *8L$ =A : X9S su dec =a : sjue x0£ 9 = A ! XOS SU [81 =A (2) 700'0 + 0S0'0| (+) €00°0 + 80'0 | (+) S00'0 F 650'0 | (8) 7000 + 1S0'0 - JUSUE ON) (8) T00°0 + £r0'0 | (2) Z00'0 + 1500 | (8) 7000 + Lr0'0 | (9) €00°0 + 9r0'0 (91) Z00'0 + 7S0'0 (IZ) 100°0 £ 8+0'0 sofeW spewsg sn810]2Q SNIDWIWUOAJOT SU OCT =A : X9S OÙ GA, || el : syue +08 =H : X9S SU 66 0 = 4 : Juoumean (L) £00'0 + 780'0 | (ZI) €00°0 + 8800 | (8) €00°0 + 1600 | (II) Z00°0 + 6800 (EI) €00°0 + #80 0 | (2) #00'0 F 9800 | (8) Z00'0 F £80'0 | (8) €00°0 + S80'0 soe sopewag SNADIL SnIDwwoAJ]og (TE) 100°0 + 6800 (€Z) Z00'0 + S80'0 **9$ 6 = 4 : X9S SU 9ÿ | =H : SUP kxkxLL' VI =A + X9S LOC TE = (6) S00 0 + LITO | (L) +000 + 8110 | (OI) S00'0 + 611 0 | (8) S00'0 F 0EL'O :quouneon | (61) £00‘0 + 801 0| (2) €000 + HOI 0 | (81) €00°0 Æ III '0 | (ZI) +000 + FI 0 Soe soyeulay SUS28D DIDI pauiquio9 ‘01 0 < d SU: 1000 > d xxx: 100 > d xx ‘ S00 > d x: 1 0 > d + VAONYV ABM-OMJ LOI 918 SONJEA A ‘(S19Y981Q UI SazIs aJduES) SIOII9 PIepue)s F sueaw Je UdAIH ‘SJUE SNISDT JO SOU9SqE JO SOU9SAId 94} ur paıeaı UdYM sotoods sepıuseaA 1 ueodoimA >Aly Jo (p „ Zw /Suu UT) S9IVI YIMOIS SATEIOY ÿ 1921 (SZ) £00'0 + ZI 0 (Zr) O0 0 + IIT'O SU 860 =H . X9S x90 7 = 4 : SJUE SU 98 0 =A Mes SU [60 = 4 : jue su £9T=4 XOS +$8'¢ =A : sque supe CT =A : XOS Su €TO =A : jue Su 770 =A + KOS +69 =A : sjue 199 (80) 0T'0 F ET + (ZZ) 91 OF TIE FOTO F CP (LZ) 61:0 F ETP (91) 60°0 £ 061 (IT) L0'0 + 89°] (1E) 80°0 F SOT (€Z) O1 OF 91 (SZ) S0'0 + Sr] (Zr) WOFTSI p>urqwo) SJUSUNESI] JUV +L LT =A SEN su [oT =A : uaumeaı su (90 = 4 ROS su 660 = 4 : Jusumea} +90 ÿ =A ! X9S *l6¢ =A : JUOW}eOI} «ICV =A ! X9S su 8€ 0 =H : JU9UI)R91] SUOI I =A : XOS SU oT =A : Juaumea1 (11) PT’O FOEE (L) PL OF ESE (OI) 9€°0 F TT + (9) 970 + H9'€ (2) 91:0 + 907 (8) OI'0 + F6 I (2) 90°0 = 681 (EI) 600 I8 1 (6) ILO + vel (61) 900 + HET (8) p€'0 + 80'P (6) 770 + 8S'€ (S) 170 + TEE (11) 0€ 0 + 79V (p) I7OF ILI (L) LO'0 F STI (GD) GIO 42 SINE (L) S70 F CSI (L) 80°0 FI (L) 110 + 691 (L) 6£°0 F ST + (9) P70 F LS'E (9) 990 + L6 + (6) S£'0 F 86'€ (t) 61 0 + 017 (8) 800 = SLI (8) 81.0 + I61 (8) 770 F 8L'I (O1) 800 F IVI (81) 800 +971 (EI) €€0 + IEP (L)IEOF LIE (II) 870 + 96'€ (L)IEOF LEE (8) TI 0 + 061 (9) tI'0 + SSI (ID STOFYOT (8) L0.0 + OL'T (8) 800 + 6r I LDOTOFISI Sole sopewag SNAÂJ1] DUADIAT soe sspeung spanjyd nuavdaT sofeW sopewag sn34D]]2g SNIVWWOAJOT sofeW sopewag SNADI1 SNIDVWIWOAÄJOT SIP sopewag S1JSA8D DIDI O10<d SU‘ S00 > d «+10 > d + VAONY AeM-0m} WoT 97e sonJeA A ‘(S19X981q ur SOZIS Spdues) SIOIIS PIepueIs F SUPAUI 918 UDAIH ‘SJUE SNISDT JO dUaSGe 10 SOU9S9Id 9} UT pores usyM so1oeds sepruseoAT uvadoing day Jo (Iysıam Arp Su /JYsI9M JOM SU) uoronpoid sse.ıj pue UIeS ssew [BAIR] JO ONY S 3192 L significant influence on growth rates. Relative growth rates (daily mass gain divided by prepupal mass) were 0.10-0.13 mg/mg * d in A. agestis and L. phlaeas, 0.07-0.09 mg/mg * d in P icarus and L. tityrus, and 0.045-0.05 mg/mg * d in P bellargus. Frass production and efficiency of food conversion (Table 5) The ratio of larval mass gain (wet weight) and frass production (dry weight) was at most weakly affected by ant-association in all 5 species. In P bellargus, this ratio was lower in the experimental series with 2 L. niger ants. This series was reared later in the season than all others, and the differences most likely indicate a change in hostplant quality rather than any influence of ant-tending. In A. agestis, L. phlaeas and L. tityrus, no consistent effects of ants were found. For P. icarus, ANOVA indicated no influence of ant-association on food utilization. However, ant-tended males had a significantly higher mass gain/frass ratio than ant-tended females (U;,33, = 170; Z = 3.26; P < 0.002), whereas in control experiments this sex difference did not occur (U); = 34; P > 0.2). Overall, sex differences in this parameter were minimal, and average values for the treatments ranged from 1.34- 1.52 mg/mg in A. agestis, 1.70-2.05 mg/mg in both Polyommatus species, and 3.30-4.97 mg/ mg in the two Lycaena species. Discussion Differences in developmental parameters between males and females were observed in all 5 species tested. Generally males developed faster (shorter larval period, higher growth rate). This is in accordance with the protandry of all 5 species in nature. All species examined produce 2 or more generations per year in central or southern Europe. The most rapid development was observed in A. agestis and L. phlaeas. These two species produce 3-4 generations per year in central Europe in favourable seasons. P. icarus was somewhat slower (2-3 generations per year), and L. tityrus as well as P. bellargus took longest to reach maturity. In the latter two species a third generation is a very rare exception north of the Alps (EBERT & RENN- WALD, 1991). Thus, our laboratory results are consistent with pheno- logical observations made in the field. Pupal weight losses were highest in P. bellargus, the species with the longest pupal stage, and were fairly similar among the remaining 4 species. Concerning the ratio of mass gain per frass production, the two myrmecoxenous Lycaena species far surpassed the 3 myrme- cophilous members of the Polyommatus group. Both Lycaena species 18 feed on leaves of Rumex species (Polygonaceae). These leaves appear to have a lower content of undigestible material than the legume hostplants of P icarus or P. bellargus. PıErce (1985) has argued that myrmecophily has selected for the utilization of nutrient-rich hostplants (especially nitrogen-rich legumes and inflorescences), but obviously the 2 Lycaena species perform in a superior manner on the foliage of a non-legume host. According to its biomass/frass ratio, A. agestis utilizes the poorest hostplant material (leaves of Geraniaceae), but nevertheless the caterpillars are highly myrmecophilous. The nutritional constraints on larval myrmecophily are a rewarding field open to further inves- tigation (BAYLIs & PIERCE, 1993). Overall, ants had very little impact on the developmental parameters studied, and the few marginally significant ant-effects which could be detected were mostly beneficial. From facultatively myrmecophilous Lycaenidae butterflies like Polyommatus icarus or Hemiargus isola it is already known that ant-association does not necessarily pose develop- mental costs, but may even allow overcompensation of the investment into myrmecophily (FIEDLER & HÖLLDOBLER, 1992 ; WAGNER, 1993). Our present data fully corroborate that pattern. Aricia agestis and Polyommatus bellargus are both facultative myr- mecophiles whose older larvae are almost never found without tending ants (e.g. THOMAS & LEWINGTON, 1991). In these species, ant- association had very limited effects on developmental parameters, suggesting that larvae of both sexes can fully compensate for the costs of myrmecophily. In recent experiments with A. agestis, we could even demonstrate a beneficial effect of ant-attendance on prepupal weights (tended individuals are ca. 10% heavier, but develop more slowly: Hummel & Fiedler, unpublished). Polyommatus icarus is a facultative myrmecophile with a rather loose relationship to ants (THOMAS & LEWINGTON, 1991). Caterpillars of this species produce less secretion from their nectar organs than P. coridon and its close relatives (FIED- LER, 1991), suggesting a comparatively low energetic investment into myrmecophily. Our experimental data confirm the findings of FIEDLER & HOLLDOBLER (1992) that ant-tended P. icarus males grow larger than untended controls, whereas tended females appear to fully compensate for their costs of myrmecophily. Weak male-limited benefits also occurred in myrmecoxenous Lycaena species (weight in L. phlaeas, food conversion in L. tityrus). Caterpillars of both species lack a dorsal nectar organ. Accordingly, their energetic investment in interactions with ants must be low (only through the ubiquitous pore cupola organs). Ant-associations of these species are 19 unknown in the wild, but under laboratory conditions weak and unstable associations can be induced (FIEDLER, 1991). As in the case of certain ant-tended aphids which show better growth in the presence of ants (Banks & Nıxon, 1958), the physiological mechanisms responsible for overcompensation in some lycaenid cater- pillars remain unclear. Ant-tended caterpillars did not produce more frass than untended controls (SAAM, 1993). Therefore, total food consumption was probably equal between the groups. Rather, the efficiency of food conversion into biomass may be stimulated by tending ants. Circumstantial evidence for this hypothesis was found in A. agestis, P. icarus and L. tityrus, where ant-tended individuals showed enhanced conversion of food. Alternatively, the stimulation of caterpillars could be a predominantly behavioural phenomenon (see discussion in WAGNER, 1993). Perhaps feeding behaviour is less often interrupted in ant-tended individuals, facilitating a more effective food utilization. Caterpillars of P icarus, P. coridon and other myrmecophilous species resume locomotion and feeding more rapidly after experimental disturbance when ants are present (Fiedler, unpublished). In the Nearctic Glaucopsyche lygdamus (Doubleday, 1841), untended caterpillars are much more likely to drop off the hostplant (PIERCE & EASTEAL, 1986). Developmental effects of ant-attendance have not been studied in detail in this latter species, but pupal weights of tended and untended individuals did not differ (PIERCE & EASTEAL, 1986). One could argue that, under the confined conditions of artificial ant- associations, ants do not harvest larval secretions as eagerly as they would do if they could transfer their crop content to their colony. Three lines of evidence contradict this view. Firstly, caterpillars and pupae of all 5 species tested were regularly tended by ants (and the 3 myrmecophilous species constantly so) throughout the whole experi- mental period. Tending levels did not decrease with time. This indicates that the ants, which had no access to alternative food sources, exerted a permanent pressure on the lycaenid immatures to deliver their secretions. Secondly, in P icarus (and in the myrmecoxenes L. phlaeas and L. tityrus) the numbers of ants per larva used in our experiments were comparable to, or even higher than, the average number of tending ants observed so far in nature. Hence, at least in these species com- pensation or overcompensation occurred despite a relatively high level of ant-attendance. Thirdly, the amounts of nectar secretion produced by single lycaenid larvae over the third plus fourth instar are sufficiently small to be 20 sampled completely by a small number of ant workers. In P. icarus and A. agestis, for example, individual lifetime nectar secretion volumes amount to 10 ul or less, equivalent to approximately 1.5 mg carbo- hydrates at most (Fiedler, Burghardt & Hummel, unpublished). How- ever, we cannot rule out the possibility that ın future experiments (e.g. with higher tending levels or involving other ant species) deviating results could be obtained. The actual outcome of potentially mutualistic interspecific interactions can be strongly shaped by environmental conditions such as hostplant quality or density of interacting species (CUSHMAN & WHITHAM, 1991 ; BRETON & ADDICOTT, 1992). There is steadily increasing evidence that compensation or even over- compensation of the energetic costs resulting from ant-association is not uncommon among facultatively myrmecophilous lycaenid butterf- lies. This finding could explain why facultative, and sometimes weak, interactions with ants are so common and taxonomically widespread across the diversity of Lycaenidae butterflies (FIEDLER, 1991). If myr- mecophily were generally a high-cost strategy under severe selective regimes, one would have to expect strong disruptive selection favouring either close and obligatory myrmecophily with high costs, but high rewards for the lycaenids, or favouring the reduction of myrmecophily. In contrast to this expectation, facultative ant-associations appear to be more common in lycaenid butterflies than cases of obligatory myr- mecophily (PIERCE, 1987 ; FIEDLER, 1991). Many examples of facul- tative interactions with ants are probably best described as low-cost mutualisms, where the lifetime energetic investment of individual cater- pillars accounts for a few ul of secretions. It is then not surprising that myrmecophily is an evolutionarily rather stable component in the life-history of so many Lycaenidae species. A broad continuum of cost-benefit relationships in terms of myrme- cophily does exist across the diversity of Lycaenidae butterflies. This is reflected by the variety of developmental effects these ant-association can have in various lycaenid species, ranging from severe costs to substantial benefits. In general, the developmental costs for the larvae appear to parallel the degree of obligateness of the ant-lycaenid asso- ciations : the more dependent the larvae are on ants, the higher the costs the ants can in turn impose. Further comparative studies on species representing various taxonomic groups and different types of myrmecophily will strengthen our ecological and evolutionary under- standing of lycaenid-ant interactions. 21 Acknowledgements We are indebted to U. Grosch, V. Hummel and T. Baumgarten for their help in butterfly rearing and data collection. P. Seufert kindly provided live stock for part of our rearings, and K. Sommer gave us access to some ant colonies. F. Burghardt and V. Hummel kindly contributed some unpublished data. We are grateful to B. Hölldobler and two anonymous referees for their critical comments on the manuscript. Supported from the Leibniz Prize of the Deut- sche Forschungsgemeinschaft to B. Hölldobler. References BALLMER, G. R. & Pratt, G. F., 1992. Quantification of ant attendance (myrmecophily) of lycaenid larvae. J. Res. Lepid. 30 : 95-112. Banks, C. J. & Nixon, H. L., 1958. Effects of the ant, Lasius niger L., on the feeding and excretion of the bean aphid, Aphis fabae Scop. J. Exp. Biol. 35 : 703-711. Bay is, M. & PIERCE, N. E., 1991. The effect of host plant quality on the survival of larvae and oviposition by adults of an ant-tended lycaenid butterfly, Jalmenus evagoras. Ecol. Ent. 16 : 1-9. BayLis, M. & Prerce, N. E., 1992. Lack of compensation by final instar larvae of the myrmecophilous lycaenid butterfly, Jalmenus evagoras, for the loss of nutrients to ants. Physiol. Ent. 17 : 107-114. BAyLıs, M. & Pierce, N. E., 1993. The effects of ant mutualism on the foraging and diet of lycaenid caterpillars. /n Stamp, N. E. & CASEy, T. M. (Eds.) : Caterpillars : ecological and evolutionary constraints on foraging. Chapman & Hall, New York/ London, pp. 404-421. BRETON, L. M. & ADDICOTT, J. F., 1992. Does host-plant quality mediate aphid-ant mutualism? Oikos 63 : 253-259. COTTRELL, C. B., 1984. Aphytophagy in butterflies: its relationship to myrmecophily. Zool. J. Linn. Soc. 79 : 1-57. CUSHMAN, J. H., RASHBROOK, V. K. & BEATTIE, A. J., 1994. Assessing benefits to both participants in a lycaenid-ant association. Ecology 75 : 1031-1041. CusHMAN, J. H. & WHITHAM, T. G., 1991. Competition mediating the out- come of a mutualism : Protective services of ants as a limiting resource for membracids. Amer. Nat. 138 : 851-865. DeVries, P. J., 1990. Enhancement of symbioses between butterfly caterpillars and ants by vibrational communication. Science 248 : 1104-1106. EBERT, G. & RENNWALD, E. (Eds.), 1991. Die Schmetterlinge Baden- Württembergs, Band 2 : Tagfalter IL. E. Ulmer, Stuttgart. ELGAR, M. A. & Pierce, N. E., 1988. Mating success and fecundity in an ant-tended lycaenid butterfly. /n CLUTron-Brock, T. H. (Ed.) : Repro- ductive success: studies of selection and adaptation in contrasting breeding systems. Chicago Univ. Press, pp. 59-71. FIEDLER, K., 1990. Effects of larval diet on the myrmecophilous qualities of Polyommatus icarus caterpillars (Lepidoptera : Lycaenidae). Oecologia 83 : 284-287. 22 FIEDLER, K., 1991. Systematic, evolutionary, and ecological implications of myrmecophily within the Lycaenidae (Insecta: Lepidoptera: Papilio- noidea). Bonner zool. Monogr. 31 : 1-210. FIEDLER, K. & HÔLLDOBLER, B., 1992. Ants and Polyommatus icarus immatures (Lycaenidae) — sex-related developmental benefits and costs of ant attendance. Oecologia 91 : 468-473. FIEDLER, K. & MaAscuwitz, U., 1988. Functional analysis of the myrme- cophilous relationships between ants (Hymenoptera: Formicidae) and lycaenids (Lepidoptera : Lycaenidae). III. New aspects of the function of the retractile tentacular organs of lycaenid larvae. Zool. Beitr. Berlin N. F. 31 : 409-416. FIEDLER, K. & Mascuwitz, U., 1989. The symbiosis between the weaver ant Oecophylla smaragdina and Anthene emolus, an obligate myrme- cophilous lycaenid butterfly. J. Nat. Hist. 23 : 833-846. Hiri, C. J. & Prerce, N. E., 1989. The effect of adult diet on the biology of butterflies I. The common imperial blue, Jalmenus evagoras. Oecologia 81 : 249-257. Mauıicky, H., 1969. Versuch einer Analyse der ökologischen Beziehungen zwischen Lycaeniden (Lepidoptera) und Formiciden (Hymenoptera). Tijdschr. Ent. 112 : 213-298. MASCHWITZ, U., Wüst, M. & ScHURIAN, K., 1975. Bläulingsraupen als Zuckerlieferanten für Ameisen. Oecologia 18 : 17-21. PETERSON, M. A., 1993. The nature of ant attendance and the survival of larval Icaricia acmon (Lycaenidae). J. Lepid. Soc. 47 : 8-16. Pierce, N. E., 1983. The ecology and evolution of symbioses between lycaenid butterflies and ants. Ph.D. thesis, Harvard Univ., Cambridge/ Massa- chusetts. Pierce, N. E., 1985. Lycaenid butterflies and ants: selection for nitrogen- fixing and other protein-rich food plants. Amer. Nat. 125 : 888-895. Pierce, N. E., 1987. The evolution and biogeography of associations between lycaenid butterflies and ants. Oxford Surv. Evol. Biol. 4 : 89-116. Pierce, N. E. & EASTEAL, S., 1986. The selective advantage of attendant ants for the larvae of a lycaenid butterfly, Glaucopsyche lygdamus. J. Anim. Ecol. 55 : 451-462. Pierce, N. E., Kircuine, KR. L., BUCKLEY, KR. C., TAYLOR, M. F. J. & BENBow, K. F., 1987. The costs and benefits of cooperation between the Australian lycaenid butterfly, Jalmenus evagoras, and its attendant ants. Behav. Ecol. Sociobiol. 21 : 237-248. Rossins, R. K., 1991. Cost and evolution of a facultative mutualism between ants and lycaenid larvae (Lepidoptera). Oikos 62 : 363-369. SAAM, C., 1993. Untersuchungen zum Einfluß der Ameisenassoziation auf die Individualentwicklung einiger europäischer Bläulingsarten. Diplom thesis, Dep. Zool., Univ. Würzburg. Sachs, L., 1992. Angewandte Statistik. 7th edition. Springer, Berlin/ Heidel- berg/ New York 23 SCHURIAN, K. G., 1989. Revision der Lysandra-Gruppe des Genus Polyom- matus Latr. (Lepidoptera : Lycaenidae). Neue ent. Nachr. 24 : 1-181. Tautz, J. & FIEDLER, K., 1992. Mechanoreceptive properties of caterpillar hairs involved in mediation of butterfly-ant symbioses. Naturwissen- schaften 79 : 561-563. Tuomas, J. A., ELMES, G. W. & WARDLAw, J. C., 1993. Contest competition among Maculinea rebeli butterfly larvae in ant nests. Ecol. Ent. 18: 73-79. Tuomas, J. A. & LEWINGTON, R., 1991. The butterflies of Britain and Ireland. Dorling Kindersley, London/ New York/ Stuttgart. WAGNER, D., 1993. Species-specific effects of tending ants on the development of lycaenid butterfly larvae. Oecologia 96 : 276-281. 24 Nota lepid. 17 (1/2) : 25-29 ; 30.X1.1994 ISSN 0342-7536 Oviposition behaviour in Lycaena thetis Klug (Lepidoptera : Lycaenidae) Konrad FIEDLER* & Klaus G. SCHURIAN** * Theodor-Boveri-Biozentrum der Universität, Zoologie II, Am Hubland, D-97074 Würzburg, Germany ** Am Mannstein 13, D-65779 Kelkheim-Fischbach, Germany Summary The oviposition behaviour of Lycaena thetis was observed in the Aladag mountains, southern Turkey. Females drop their eggs singly into the spiny cushions of the larval foodplant (Acantholimon spp., Plumbaginaceae). Zusammenfassung Das Eiablageverhalten von Lycaena thetis wurde im Aladag-Gebirge (Süd- Türkei) beobachtet. Die Weibchen lassen ihre Eier einzeln in die dornigen Kugelpolster ihrer Wirtspflanze (Acantholimon spp., Plumbaginaceae) fallen. The life histories of European or North American species of the Lycaenini (“Copper butterflies”) are, in general, well known. For most Asian Lycaena Fabricius, 1807, species, however, even basic biological information on hostplants, voltinism, or diapause stages is lacking (cf. FIEDLER, 1991). Recently, ToLMAN (1993) published a detailed account of the larval biology of Lycaena thetis Klug, 1834, from southern Greece. Since Tolman based his description on field-collected young, hibernated larvae, the oviposition behaviour of L. thetis remained unknown. Furthermore, Tolman’s paper deals with the westernmost populations of L. thetis. Because the distribution of L. thetis extends throughout Asia Minor to northern Iran (SCHURIAN & HOFMANN, 1982), it remained to be tested whether populations in the heart of the species’ range utilize the same or similar hostplants. On 15.vi.1993, between 11.00-15.00 EEST, we had the opportunity to observe the unusual oviposition behaviour of L. thetis in a population of southern Turkey. The habitat was a south-facing steep slope in a valley of the Aladag mountains (Prov. Nidge), approximately 1800- 25 1900 m above sea level. This slope was mostly covered by limestone boulders and scree. The sparse vegetation contained Juniperus shrubs and single conifer trees. Spiny, cushion-forming, perennial plants (Astragalus and Acantholimon spp.) which are relatively immune to overgrazing by the abundant sheep and goats, formed the lower vegetation between the boulders. # É # ft À mit 4 # Fig. 1. Large contiguous cushions (total diameter > 1 m) of Acantholimon sp. (Plum- baginaceae), the hostplant of Lycaena thetis. When we first walked through the habitat late in the morning, only few territorial males and nectaring females of L. thetis were on the wing. Males preferably basked on barren ground. Around noon, when the air temperatures had reached about 30°C, females became increa- singlv active. They often visited the last flowers of Acantholimon spp. (Plumbaginaceae). This cushion plant with extremely spiny, needle- like leaves (Fig. 1) is the hostplant of L. thetis in Greece (TOLMAN, 1993), although we were unaware of his paper at the time of our observations. Females examined the Acantholimon cushions in the fluttery searching flights that typically precede oviposition in many lycaenid butterflies. After alighting, each female would crawl about on the cushions for several minutes, repeatedly curling its abdomen 26 and probing the plant surface with the antennae and ovipositor. However, despite intensive close examination of the respective plants after the females had flown off, we failed to find any eggs attached to the twigs or leaves. Finally, we succeeded in observing the actual oviposition act. When viewed in contre-jour, we could clearly see that after up to 5 minutes of intensive crawling and probing the hostplant, the female eventually inserts its ovipositor between the needle-like leaves and simply drops a single egg into the cushion. In one case, the egg by chance stuck to a twig deep within an Acantholimon cushion, but unfortunately it fell to the ground during our attempt to secure it. The hemispherical egg was rather large (ca. | mm in diameter) and showed the rough chorionic sculpturing typical for Lycaena eggs. In total, we observed 10 successful oviposition acts, all in the same manner. On at least twice as many occasions, a female left the hostplant without laying an egg. Ovipositions only occurred during the hottest hours around noon. When we left the habitat (15.00), the females tended to bask for long periods on the Acantholimon cushions, and flight activity evidently decreased. We followed various individual females for dozens of metres. These females ignored many potential hostplant cushions and alighted to probe quite a number of Acantholimon plants before an egg was laid. Hence, they appear to be very choosy, but we do not know the factors that finally elicit oviposition. Egg-laying occurred on small as well as on large plants (0 = 30-100 cm) and not invariably in full sun, although high temperatures are clearly required. All eggs were dropped into the central part of a cushion, not at the edges. To obtain oviposition in captivity, we collected a total of 10 females from various habitats in southern Turkey in August 1993. These were caged in a plastic bottle (1.5 1) lined with moist filter paper, twigs of the hostplant Acantholimon as oviposition substrate, and sugar solution as food. However, despite exposing the females to various conditions (direct sunlight, shadow, high temperatures, high or low humidity), not a single egg was laid. The last female died after 14 days in captivity. When earlier attempting to obtain eggs from L. thetis in captivity (1977 and 1984), females had been confined with an erroneously presumed hostplant (Rumex sp., Polygonaceae) without any success. Our failure to induce oviposition in captivity contrasts sharply with successful attempts involving various other Lycaenidae species (Lycaena candens (Herrich-Schäffer, 1844), Agrodiaetus spp., Polyommatus spp.) under similar conditions. Females of L. thersamon (Esper, 1784) (like L. thetis, often assigned to the “subgenus” Thersamonia Verity, 1919), however, 2 laid only few eggs, suggesting that in both species highly xerothermic conditions (and presumably unknown factors) are essential for egg- laying. Dropping the eggs instead of attaching them onto the hostplant is a very rare behaviour in Lycaenidae butterflies, but has been recorded from the Nearctic Lycaena rubidus (Behr, 1866 ; see Funk, 1975). In L. thetis, the females may thereby avoid fatal injuries inflicted by the extremely spiny leaves, and the eggs are probably protected against many enemies within the dense thorny Acantholimon cushions. It re- mains to be proven whether L. thetis hibernates in the egg stage or whether the larvae hatch in late summer to diapause. Both strategies occur within the genus Lycaena. Since TOLMAN (1993) found larvae of 4-7 mm length shortly after hibernation in Greece, diapause as a young caterpillar seems more likely. Our observations confirm that Acantholimon is the hostplant not only for Greek, but also for Turkish L. thetis populations. The plant family Plumbaginaceae is well represented in eremic steppe habitats. In addition, Plumbaginaceae are not too distantly related to the usual Polygonaceae hostplants of most Lycaena species (both plant families belong to the subclass Caryophyllidae). SCHURIAN & HOFMANN (1982) explicitly mentioned the presence of Acantholimon (in part quoted as “ Acantolimnus”) in habitats of L. eberti Forster, 1972, and LUKHTANOV & LUKHTANOV (1994) recorded the hostplant of L. solskyi (Erschoff, 1874) as Acantholimon. Therefore, additional species of Asian Lycaena might also use Acantholimon as hostplant. Fucus (1989) reported that L. thetis (especially females) preferably visited another spiny cushion plant, Drypis spinosa (Caryophyllaceae), in central Greece for nectaring and basking. Whether this plant species could serve as alternative larval hostplant, needs to be demonstrated. References FIEDLER, K., 1991. Systematic, evolutionary, and ecological implications of myrmecophily within the Lycaenidae (Insecta: Lepidoptera: Papilio- noidea). Bonner zool. Monogr. 31 : 1-210. Fucus, J., 1989. Ein Vorkommen von Thersamonia thetis (Klug) in Mittel- griechenland (Lep., Lycaenidae, Lycaeninae). Galathea Ber. Kreises Nürnberger Entomologen 5(1) : 11-18. Funk, R. S., 1975. Association of ants with ovipositing Lycaena rubidus (Lycaenidae). J. Lepid. Soc. 29 : 261-262. LUKHTANOV, V. & LuKHTANov, A., 1994. Die Tagfalter Nordwestasiens. Herbipoliana 3 : 1-440. 28 SCHURIAN, K. G. & HorMmann, P., 1982. Die Thersamonia-Gruppe (Lepi- doptera, Lycaenidae). Nachr. entomol. Ver. Apollo, Frankfurt, Suppl. 2: 1-59. ToLMAN, T., 1993. The natural history, ecology and distribution of Turanana panagaea (Herrich-Schäffer, [1851]) and Lycaena thetis (Klug, 1834) in Greece (Lepidoptera : Lycaenidae). Phegea 21(3) : 81-92. 29 Nota lepid. 17 (1/2) : 30 ; 30.X1.1994 ISSN 0342-7536 Book reviews — Buchbesprechungen — Analyses Oecophorine Genera of Australia. I. The Wingia Group (Lepidoptera : Oecophoridae). Monographs on Australian Lepidoptera, vol. 3. Ian F. B. Common. xvi, 390 pp., 712 Figs. 25.9 X 18.2 cm, hardback. CSIRO Publications, 1994. ISBN 0 643 05524 X. Obtainable from CSIRO Publications, 314 Albert St., East Melbourne, Victoria 3002, Australia, or Apollo Books, Kirkeby Sand 19, DK-5771 Stenstrup, Denmark. Price : $US 100, or $AS 100 in Australia. A discount of 25% is given to subscribers of the series. The third volume of this major work on the Australian Lepidoptera fauna deals with the genera of the subfamily Oecophorinae. With 1886 valid named species and an estimated species total of more than 5000, the Oecophorinae is by far the best represented group of Lepidoptera in Australia. The wealth of species in Australia is disproportionate, being about 70% of the world fauna. With 170 named genera and an estimated total of 275, clearly all cannot be treated in a single volume. The present volume is the first of three parts and covers the 91 genera of the Wingia group. The author is well-known for his excellent book ‘Moths of Australia’ (see review in Nota lepid. 14 (3) : 292). As in the previous volume, introductory chapters on ‘Phylogeny’, including a tentative phylogenetic analysis of the subfamily genera, ‘Morphology’, ‘Biology’ and ‘Diversity and Distribution’ are given. The only synapomorphies of the Wingia group are to be found on the ab- dominal sternum and in the male genitalia. It is not stated however whether an experienced lepidopterist can recognise most species belonging to this group on external characters only. A key to the genera is provided, based on external and internal morphological features. The treatment of each species follows the same format as in the previous volume (for review, see Nota lepid. 16 (3/4) : 265-266). The high scientific quality of the series is maintained in this volume and as such can be highly recommended to students of this family or of the Australian Lepidoptera fauna in general. Unexpectedly, considering the usually high stan- dard of production, in the reviewer’s copy the text was bound to the cover upside down ! Steven WHITEBREAD 30 Nota lepid. 17 (1/2) : 31-43 ; 30.X1.1994 ISSN 0342-7536 Morphology and taxonomy of the species belonging to the genus Myinodes Meyrick, 1892 (Lepidoptera : Geometridae) Axel HAUSMANN Zoologische Staatssammlung, Münchhausenstr. 21, D-81247 München, Germany Summary Two new species and one new subspecies of the genus Myinodes Meyrick, 1892 are described : Myinodes interpunctaria atlantica ssp. n. from Spain, Myinodes constantina sp. n. from Algeria and Myinodes shohami sp. n. from Jordan. Until recently, the genus was considered to be monotypical, with M. interpunctaria (Herrich-Schäffer, 1839) the only known species. Zusammenfassung Myinodes interpunctaria (Herrich-Schaffer, 1839) war in der Literatur bisher stets als einzige Art (Typusart) der Gattung Myinodes betrachtet worden. Eine genauere morphologische Analyse ergab, daß die aus Nordafrika, Südeuropa und Vorderasien bekannten Populationen einen aus mehreren verschiedenen Arten bestehenden Komplex bilden. In der vorliegenden Arbeit werden zwei neue Arten und eine neue Unterart beschrieben : Myinodes interpunctaria atlantica ssp. n. aus Spanien, Myinodes constantina sp. n. aus Algerien und Myinodes shohami sp. n. aus Jordanien. Resume Description de deux nouvelles espèces et d’une nouvelle sous-espèce du genre Myinodes Meyrick, 1892 : Myinodes interpunctaria atlantica ssp. n. d’Espagne, M. constantina sp. n. d’Algerie et M. shohami sp. n. de Jordanie. Jusqu’a tout récemment, ce genre était considéré comme monotypique avec M. interpunc- taria (Herrich-Schäffer, 1839) comme seule espèce connue. Until recently, the genus Myinodes Meyrick, 1892 was considered to comprise only the species interpunctaria (Herrich-Schäffer, 1839). De- tailed morphological studies have revealed, however, that the popu- lations known from northern Africa, southern Europe and the Middle East constitute a complex of different species. In this paper two new Sil species and one new subspecies are described : Myinodes interpunctaria atlantica ssp. n. from Spain, Myinodes constantina sp. n. from Algeria and Myinodes shohami sp. n. from Jordan. Systematic part Abbreviations : BUS : Bet Ussishkin Museum, Tel Dan, N.-Israel. NHMW : Naturhistorisches Museum Wien, Austria. NLK : Naturkundliche Landessammlungen Karlsruhe, Germany. TAU : Tel Aviv University Collection, Israel. ZFMK : Zoologisches Forschungsinstitut und Museum Alexander Koenig, Bonn, Germany. ZSM : Zoologische Staatssammlung Miinchen, Germany. Myinodes Meyrick, 1892 Eusarca Herrich-Schaffer, 1847 (partim) Pseudotagma Staudinger, 1892 Type species : Sterrha interpunctaria Herrich-Schäffer, 1839 : pl. 6 and wrap- per ; by monotypy. VENATION : Hindwing Sc + RI and Rs not fused (as in the subfamily Alsophilinae), M2 tubular. Tongue developed, length about 4 mm. Frons strongly convex. Palpi long. Male antennae with two rows of cilia, female antennae simple, finely ciliate beneath. Male and female hindlegs long and slender, with two pairs of long spurs. In the male genitalia uncus developed, often with a subapical lobe ; juxta with caudal excavation ; costal part of valva strongly sclerotized, harpe prominent ; aedoeagus long, slender, with one cornutus, laterally sclerotized (dif- ferently in each species). In the female genitalia apophyses weak ; ductus bursae comparatively long and stout ; bursa copulatrix longitudinally ribbed (not in M. constantina), without signa, joins ductus bursae laterally on the latter. Ansa of the tympanon apically pointed. In most of these characters very similar to the genus Eumegethes Staudinger, 1898. The systematic position of the genera Myinodes and Eumegethes is not the subject of this paper, They are usually placed in the subfamily “Oenochrominae” (s.1.). However, they are not closely related with this subfamily (s.str.), which is mainly distributed in SE Asia and Australia, or with the subfamily Alsophilinae. Myinodes interpunctaria interpunctaria (Herrich-Schaffer, 1839) Sterrha interpunctaria Herrich-Schäffer, 1839 : pl. 6 and wrapper. Locus typi- cus : Sicily. 32 ‘sqpouldy snus3 au} JO exe} snoLIeA oy} Jo UOTNGIYSIG “| ‘ST "u "ds Tweyoys SepoutAWy "u ‘ds eurt3ue3suo9 SapoutAy "u "dsqns eatyuetg4e etzezoundzaqgut sapoutihw "S-"H erıe3aundagqurt eteqsoundiaequt 33 Sterrha interpunctaria : Herrich-Schäffer, 1840 : 104. Eusarca interpunctaria : Herrich-Schäffer, 1847 : 34. Eusarca interpunctaria : Herrich-Schäffer, 1848 : pl. 64, fig. 390. Fidonia interpunctaria : Heydenreich, 1851 : 54. Phasiane? interpunctaria : Lederer, 1853 : 180. Selidosema? interpunctaria : Guenée, 1857 : 146. Anisopteryx interpunctaria : Gumppenberg, 1893 : 396. Eusarca interpunctaria : Staudinger & Rebel, 1901 : 322. Eusarca interpunctaria : Spuler, 1904 : 86. Eusarca interpunctaria : Spuler, 1907 : pl. 71b, fig. 1. Myinodes interpunctaria : Prout, 1910 : 20, pl. 1, fig. 13. Myinodes interpunctaria : Prout, 1912a : 4, pl. 1b. Myinodes interpunctaria : Prout, 1912b : 32. Eusarca interpunctaria : Culot, 1920 : 49, pl. 45, fig. 932. Eusarca interpunctaria : Oberthür, 1922 : 307. ?Myinodes interpunctaria : Turati, 1925 : 8. ?Myinos interpunctaria : Krüger, 1939 : 352. Myinoides interpunctaria : Mariani, 1943 : 81. Myrinodes interpunctaria : Schmidlin, 1964 : 82. Myrinodes interpunctaria : Parenzan, 1976 : 162, fig. 6a. Myinodes interpunctaria : Fletcher, 1979 : 133. MATERIAL EXAMINED: 16, Sicily, coll. Failla; 1 4, S. Italy, Basilicata, F. Basento, Trivigno Scalo, 28.II1.1977, leg. P. Parenzan, coll. ZSM ; 2 88, S. Italy, Puglia, Mte. Camplo, Laterza (TA), 21.111.1971, leg. P. Parenzan, coll. ZSM; 1666, Tunisia, Tunis distr., El Gouina, 9.11.-4.111.1960, leg. H.P. Müller, coll. ZSM ; 2 99, Algeria s., Algier Distr., El Aziza, 26.111.1989, leg. Kuchler jr., coll. K. Kuchler; 1 9, Algeria, Constantine, leg. Olivier, coll. ZFMK ; 1 9, Algeria, Guelt-es-Stel, 19.1V.1931, leg. Predota, coll. ZFMK. 7 OG, 3 QQ dissected. DISTRIBUTION (Fig. 1): Sicily, S. Italy (Basilicata and Puglia), N. Tunisia, N. Algeria. The local populations of Tripolitania and Cyrenaica have to be preliminarily regarded as belonging to the nominate subspecies. EXTERNAL MORPHOLOGY : Palpi: length 1.25-1.35 mm, scales dark brown, only upperside white. Frons (Fig. 8) with two projections. Length of cilia of male antenna about 0.14 mm, somewhat exceeding thickness of shaft (0.12 mm). Wincs : Forewing length, male : 14.0-15.2 mm ; female (Algeria) : 11.4- 13.7. Postmedial line very slightly dentate, at inner margin not inclined toward wing base. White intervenal line in subterminal area of forewing apex crossing postmedial fascia. Postmedial line and intervenal lines sharply bordered. Small terminal spots black, trianguliform, surrounded by forked white intervenal line. 34 Figs 2-7. Myinodes spp. 2 — M. interpunctaria atlantica ssp. n., 4, Holotypus ; 3 — M. interpunctaria atlantica ssp. n., Q, Paratypus ; 4 — Myinodes constantina sp. n., 6, Holotypus ; 5 — M. constantina sp. n., 9, Paratypus ; 6 — M. shohami sp. n., 6, Holotypus ; 7 — M. shohami sp. n., Q, Paratypus. MALE GENITALIA (Fig. 10a-d): Uncus short, with stout subapical processus. In juxta caudal median notch very deep, basis of juxta forked. Valva costa broad, smoothly edged. Caudal directed spine (harpe) prominent, pointed, slightly curved. Basal lobe of harpe strongly convex, With numerous small spines. Aedoeagus slender and long (mean 1.7 mm). Cornutus weakly sclerotised, situated subterminally. Aedoea- gus bearing a longitudinal row of about four sharp, stoutly sclerotized teeth in terminal part. In the male from Sicily there are six teeth, 35 perhaps an individual aberration. In the Tunisian males these teeth are ısolated from each other. FEMALE GENITALIA (Fig. 14): Females examined from Algeria have a comparatively long and narrow ductus bursae, its left lateral margin concave. Bursa copulatrix broad and large, but narrower than in the spanish subspecies. Caudal edge of lamella postvaginalis convex. HABITAT : Not above 900 m, mostly from 0-300 m. In Southern Italy abundant in a xerothermic locality (Mte. Camplo) with remnants of mediterranean macchia. FLIGHT PERIOD : Beginning of February to mid-April. The Tripolitanian specimen mentioned in KRUGER (/.c.) taken in January. Myinodes interpunctaria atlantica subsp. n. (Figs 2, 3) Myinodes interpunctaria : Exposito, 1978 : 38. Myiniodes interpunctaria : Rungs, 1981 : 223. HoıoTyPE: @, S. Spain, Prov. S. Nevada, Alcolea, 5.1V.1991, leg. Kuchler jr., coll. ZSM, Prep.No. G 6825. PARATYPES : | ®, S. Spain, Prov. S. Nevada, Alcolea, 1.1V.1991, leg. Kuchler jr., coll. ZSM ; 16, id., 5.1V.1991, leg. Kuchler jr., coll. Ky Kuchen a: S. Spain, Fuerte Higuera, Alicante, 2.1V.1993, leg. Kuchler jr., coll. K. Kuchler ; 3 35, S. Spain, Prov. Gador, Beria, 4.1V.1991, leg. Kuchler jr., coll. K. Kuchler ; 18, S. Spain, Prov. Cadiz, Villaluenga, 870m, 13.1V.1986, leg. et coll. A. Exposito ; 1 &, S. Spain, Prov. Malaga, Ronda, Cmo. Carbonera, 15.-28.111.1972, leg. et coll. A. Exposito ; 1 4, C. Spain, Toledo, 24.V.1972, leg. J. Calle, coll. A. Exposito ; 3 @4, SW. Morocco, Marrakesch, O. Tensift, 12.11.1974, leg. Friedel, coll. M. Sommerer ; 1 6, id., coll. ZSM; 1 95 W. Morocco, Zehroun, Mrassine, 1.-15.111.1921, leg. H. Powell, coll. ZFMK. 7 G&, 3 PQ dissected. DISTRIBUTION (Fig. 1) : C. and S. Spain, W. and N. Morocco. EXTERNAL MORPHOLOGY : Palpi: Length in both sexes somewhat variable 1.00-1.25 mm, shorter than in nominate subspecies, dark brown, only upperside white. Frons with two projections. Length of cilia of male antenna about 0.11 mm, not exceeding thickness of shaft (0.11 mm). WINGs : Indistinguishable from M. i. interpunctaria. Forewing length, male : 13.6-16.8 mm ; female : 13.2-13.8 mm. MALE GENITALIA (Fig. 11a-d): Uncus, juxta and costal part of the valva similar to M. i. interpunctaria. Harpe prominent, pointed, some- what more curved than in nominate subspecies. Basal lobe of harpe slightly convex, not so heavily rounded. Aedoeagus slender and long 36 Figs 8, 9. Head of Myinodes spp. 8 — M. interpunctaria H.-S. ; 9 — M. constan- tina sp. n. (mean 1.7 mm). Cornutus as in typical M. i. interpunctaria. Aedoeagus terminally heavily sclerotized, more than in nominate subspecies, bearing row of 3 or 4 sharp, lateral teeth. FEMALE GENITALIA (Fig. 15) : Ductus bursae shorter and broader than in specimens from Algeria. Caudal edge of lamella postvaginalis slightly concave. Bursa copulatrix very similar, somewhat broader. FLIGHT PERIOD : S. Spain mid-March to mid-April ; Morocco first half of March. The very late record from Toledo could indicate a later flight period in C. Spain, but this needs confirmation. Myinodes constantina sp. n. (Figs 4,5) HoLoTypPE : 4, Algeria, Lambese, II-III.1913, leg. Sari Lakhdar ben Laouss, coll. ZFMK, Prep. No. Hausm. 7910. PARATYPES : 4464, Algeria, Lambese, II-IIL.1913, leg. Sari Lakhdar ben Laouës, coll. ZFMK ; 1 @, id., coll. ZSM ; 19, Algeria, Guelt-es-Stel near Boghari, III-IV.1914, leg. Domenech Joseph, coll. ZSM ; 1 ©, id., coll. ZFMK ; 1, Tunisia, Kroumirie, Soudia, 24.V.1941, leg. Chnéour, coll. ZSM. 3 42, 2 QQ dissected. DISTRIBUTION (Fig. 1): N. Algeria: Saharan Atlas and Constantine district. In Guelt-es-Stel sympatric with M. interpunctaria. NW. Tunisia. EXTERNAL MORPHOLOGY : Palpi: Length in both sexes 1.40-1.60 mm, longer than in the other species, dark brown, upperside and basal scales near tongue white. Frons (Fig. 9) with only one central projection. Length of cilia of male antenna about 0.18 mm, exceeding thickness of shaft (0.12 mm). Wins : Forewing length, male : 14.7-15.7 mm ; female : 12.9-13.5 mm. Postmedial line not dentate, at inner margin strongly inclined toward wing base. White intervenal line in forewing apex very short, length about 1/3 of subterminal area. Postmedial line and intervenal lines Su indistinctly bordered. Antemedial line completely lacking. Small ter- minal spots black, punctiform. Intervenal lines near margin not forked, and not encircling terminal spots. Hindwings brighter than in the other species. MALE GENITALIA (Fig. 12a-d) : Uncus long, with very small subapical processus. Caudal median notch of juxta very deep, caudal lobi pointed, basis convex. Costal part of valva sinus shaped. Harpe prominent, less pointed than in M. interpunctaria, strongly curved, without spines. Basal lobe of harpe lacking. Aedoeagus slender and very long (mean 1.9 mm). Cornutus situated subterminally, broader and more sclerotized than in M. interpunctaria. Aedoeagus terminally bearing a long digiti- form and heavily sclerotized processus without teeth. FEMALE GENITALIA (Fig. 16) : Ductus bursae long, straight. Bursa copu- latrix small, irregularily shaped and not longitudinally ribbed as in the other species. FLIGHT PERIOD : No precise data available (February to May”). Myinodes shohami sp. n. (Figs 6, 7) Pseudotagma interpunctaria : Staudinger, 1892 : 168. Pseudotagma interpunctaria : v. Kalchberg, 1897 : 182. Eusarca interpunctaria : Amsel, 1933 : 109. Myinodes interpunctaria : Wehrli, 1934 : 2. Eusarca interpunctaria : Bodenheimer, 1937 : 88. Myinodes interpunctaria : Ellison & Wiltshire, 1939 : 43 Myinodes interpunctaria : Wiltshire, 1957 : 101. Myinodes interpunctaria : Hausmann, 1991 : 115, pl. 10, fig. 63. HoıoTyPE : 4, NE. Jordan, Qasr el Hallabad, 17.11.1958, leg. Klapperich, coll. ZSM PARATYPES : | @Q; C. Israel, En Gedi (Dead Sea), January, leg. G. Müller, coll. ZSM; 18, id., 1111989; 1 © ad; 8.1IL1989; 18, id.; colE PAUSE C Israel, Enot Zugim (Dead Sea), leg. G. Müller, coll. ZSM>; FR N. Israel, N. Ammud, 8.111.-19.111.1991, leg. R. Ortal, coll. ZSM ; 5 88, id., coll. TAU ; 1 @, N. Israel, Hula Reserve, 19.111.1991, leg. R. Ortal, coll. ZSM ; 2 Oo, id., 5.-8.111.1992 ; 18, N. Jordan, Amman, 8.11.1958 ; leg. Klapperich, coll. NLK; 1 ©, id., 28.VIIL.1967 (date probably mislabelled) ; 15 19, N. Jordan, Rumman, 28.11.1968, leg. Klapperich, coll. NLK ; 1 ®, id., 28.11.1965, leg. Klapperich, coll. ZSM ; 3 6&, N. Israel, Sede Nehamya, leg. Shoham, coll. BUS ; 1, id., coll. S. Yathom ; 1 4, N. Israel, Neot Mordehai, leg. Shoham, coll. BUS; 16, N. Israel, Gazith, coll. TAU; 1 @;-C Israel, “"Palaesewa Tel Aviv, leg. Bodenheimer, coll. NHMW; 19, N. Israel, “Syria”, Haifa, coll. NHMW. 8 38,3 QQ dissected. 38 Figs 10-13. @ genitalia of Myinodes spp. 10 — M. interpunctaria interpunctaria H.-S. (topotypical : Sicily); 11 — M. interpunctaria atlantica ssp. n. (Holotypus) ; 12 — M. constantina sp. n. (Paratypus) ; 13 — M. shohami sp. n. (Paratypus, Jordan) ; a — Uncus ; b = Juxta (scale bar = 0,5 mm). c = valva; d = Aedoeagus (scale bar = 1 mm). FURTHER MATERIAL EXAMINED : About 20 more or less damaged specimens from N. and C. Israel, coll. ZSM ; 1 8, “Syria”, coll. NHMW ; 16, id., coll. ZSH; 16, “Syria”, coll. ZFMK ; 446, S. Turkey, Taurus, Marasch, 600- 900m, 111.1930, les. Einh>- Sir, coll. ZFMK; 14, id., coll) ZZHEL Gar Turkey, Amanus, “Syria”, Akbés, 1895, coll. ZFMK. DistTriBuTIion : C. and N. Israel (AMSEL, 1933), N. Jordan (HAUSMANN, 1991), Lebanon (ELLIsoN & WILTSHIRE, 1939: Beirut), S. Turkey (Marasch, Akbes, Mardin ; cf. WEHRLI, 1934) and N. Iraq (WILTSHIRE, 1957). As yet no species of this genus have been found in Egypt. EXTERNAL MORPHOLOGY : Palpi: Length in both sexes 1.15-1.30 mm, much shorter than in Myinodes constantina, dark brown, upperside and basal scales near tongue white. Frons (cf. Fig. 9) with only one central projection. Length of cilia of male antenna about 0.16 mm, exceeding thickness of shaft (0.12 mm). WINGs : Forewing length, male : 12.5-14.6 mm ; female : 10.6-11.4 mm. Postmedial line strongly dentate, more outwardly curved than in the other species, at inner margin not inclined toward base. White inter- venal line in forewing apex longer than in Myinodes constantina, but not crossing postmedial fascia. Postmedial line and intervenal lines sharply bordered. Terminal spots black, punctiform and small, thinly encircled by white forked intervenal line. Forewings of specimens from S. Turkey slightly darker than in those from Jordan and Israel. MALE GENITALIA (Fig. 13a-d) : Uncus very long, subapical processus lacking. Caudal excavation of juxta U-shaped, much less deep than in the other species, basis of juxta convex. Costal part of the valva narrower than in the other species, more convex and distally pointed. Harpe prominent, S-shaped, without spines. Basal lobe of harpe lacking. Aedoeagus comparatively broad and short (mean 1.55 mm). Cornutus terminally located, very stout. Aedoeagus apex laterally bearing heavily sclerotized, distally pointed plate with one or two lateral teeth. FEMALE GENITALIA (Fig. 17): Ductus bursae broad and short. Bursa copulatrix smaller and narrower than in M. interpunctaria. Caudal half of bursa copulatrix more sclerotized than in the other species. HABITAT : From — 400 m (Israel) to 900 m (Taurus). In Israel and Jordan mainly in the swamps and wet areas near Hula Lake, the Dead Sea and some rivers (e.g. Zerga, Nahal Ammud). FLIGHT PERIOD : Israel and S. Turkey : Mid-February to end of March, one specimen in January (C. Israel). Jordan : Throughout February. One Jordan female labelled “28.VIIL” probably a mistake. In Iraq flying in April (WiILTSHIRE, 1957). 40 Figs 14-17. © genitalia of Myinodes spp. 14 — M. interpunctaria interpunctaria H.-S. (NE. Algeria) ; 15 — M. interpunctaria atlantica ssp. n. (Paratypus, S. Spain) ; 16 — M. constantina sp. n. (Paratypus) ; 17 — M. shohami sp. n. (Paratypus, N. Israel) ; scale bar = | mm. 41 REMARKS: Named after the late Mr. Z. Shoham, Israel, for his great merits in the lepidopterological exploration of N. Israel. Key to species 1 Frons with two projections. Basal scales of palpi dark brown. Black terminal spots trianguliform. Intervenal line in forewing apex crossing postmedial line‘... ae. FR as Ma. Ru interpunctaria H.-S. — Frons with only one central projection. Basal scales of palpi white. Black terminal spots punctiform. Intervenai line in forewing apex not crossing postmedial Iime........ ernennen Is ee 2 2 Palpi long (ca. 1.5 mm). Postmedial line not dentate, at inner margin strongly inclined toward wing base. The white intervenal line in forewing apex VERY SOL. Ve Sonera ose ee constantina Sp. n. — Palpi short (ca. 1.2 mm). Postmedial line strongly dentate, at inner margin not inclined toward wing base. The white intervenal line in forewing apex approaching. postmedialkines:1...114 484. Zen shohami sp. n. PARENZAN (1976 : fig. 9) mentions the genus Myinodes from N. and W. Turkey and Rumania, but gives no details. Confirmation of the occurrence of this genus from these areas ıs required. Acknowledgements I wish to express my gratitude to Dr. D. Stüning, Bonn, P. Parenzan, Palermo, G. Ebert, Karlsruhe, A. Exposito, Mostoles, A. Freidberg, Tel Aviv, R. Ortal, Jerusalem, G. Müller, Jerusalem, K. Kuchler, Munich and M. Sommerer, Munich for the loan or donation of material. References AMSEL, H. G., 1933. Die Lepidopteren Palästinas. Zoogeographica 2 (1): 1-146. BODENHEIMER, F. S., 1937. Prodromus Faunae Palestinae. Mem. Inst. d’Egypte 33 : 1-287. Cu1oT, J., 1920. Noctuelles et geomètres d’Europe, II. Rennes. EıLiıson, R. E. & WiiTSHIRE, E. P., 1939. The Lepidoptera of the Lebanon with notes on their season and distribution. Trans. Royal Ent. Soc. London 88 (1) : 1-56. Exposito HERMOSA, A., 1978. Catalogo provisional de la familia Geometridae. Shilap Revta Lepid. 6 : 37-44 ; 125-130. FLETCHER, D. S., 1979. In Nye, I. W. B (Ed.) : The Generic Names of Moths of the World, vol. 3, London, 243 pp. GuENEE, A., 1857. In BoıspuvaAL, J. A. & GuENEE, A. (1854-1858) : Histoire naturelle des insectes (Lepidoptera). Species général des Lépidoptères, 10 : Uranides et Phalénites, Paris. GUMPPENBERG, C. Frhr. v., 1893. Systema Geometrarum zonae temperatioris 42 septentr. Nova Acta Ksl. Leop. Carol. Deutsche Akad. d. Naturforscher, Halle. HAUSMANN, A., 1991. Beitrag zur Geometridenfauna Palästinas : Die Spanner der Klapperich-Ausbeute aus Jordanien (Lepidoptera, Geometridae). Mitt. münch. ent. Ges. 81 : 111-163. HERRICH-SCHAFFER, G. A. W., 1829-1844. Deutschlands Insekten, Fortsetzung von Panzers Fauna insectorum Germanica, 10 Bde. H. 111-190. HERRICH-SCHAFFER, G. A. W., 1843-1856. Systematische Bearbeitung der Schmetterlinge von Europa. Regensburg, 6 Bde. HEYDENREICH, G. H., 1851. Lepidopterorum Europaeorum Catalogus me- thodicus, Edn. 3. Leipzig. KALCHBERG A. von, 1897. Über die Lepidopterenfauna von Haifa in Syrien. Dt. ent. Z. Iris 10 : 161-190. KRUGER, G. C., 1939. Notizie sulla fauna della Sirtica occidentale : Lepidotteri. Annali del Mus. Libico Stor. Nat. 1 : 317-357. LEDERER, J., 1853. Versuch, die europäischen Lepidopteren in möglichst natürliche Reihenfolge zu stellen, nebst Bemerkungen zu einigen Familien und Arten. Verh. Zool.-bot. Ges. Wien 3 : 165-270. MARIANI, M., 1943. Fauna Lepidopterorum Italiae. Parte I. Catalogo ragio- nato dei Lepidotteri d’Italia. Fasc. II e III. Giorn. Sc. Nat. Econ. 42 (3) : 81-227. OBERTHUR, C., 1922. Les Lépidoptères du Maroc. Lepid. Comp. 19 : 1-402. PARENZAN, P., 1976. Contributi alla conoscenza della Lepidotterofauna del- l’Itahia meridionale. II Nuovi reperti di Noctuidae e Geometridae. Ento- mologica, Bari XII : 153-169. Prout, L. B., 1910. Fam. Geometridae, Subfam. Oenochrominae. Gen. ins. 104 : 1-120, 2 pl. Prout, L. B., 1912a. Die spannerartigen Nachtfalter. In Seitz, A. [1912-1916] : Die Gross-Schmetterlinge der Erde, Bd. 4. Verlag A. Kernen, Stuttgart. Prout, L. B., 1912b. Lepidopterorum Catalogus, Pars 8: Geometridae : Brephinae, Oenochrominae, Berlin, 94 pp. Runcs, C. E. E., 1981. Catalogue raisonné des Lépidoptères du Maroc. Inventaire faunistique et observations écologiques. Tome II. Trav. Inst. Sc., Sér. Zool., n. 40, Rabat. SCHMIDLIN, A., 1964. Übersicht über die europäischen Arten der Familie Geometridae (Lep.). Mitt. ent. Ges. Basel, 14 (4,5) : 77-137. SPULER, A., 1904-1908. Die Schmetterlinge Europas. Stuttgart, 4 volumes. STAUDINGER, O., 1892. Neue Arten und Varietäten von Lepidopteren des palaearctischen Faunengebietes. Dr. ent. Z. Iris 4 : 224-339. STAUDINGER, O. & REBEL, H., 1901. Katalog der Lepidopteren des palaeark- tischen Faunengebietes 1. Verlag Friedländer & Sohn, Berlin. Turarı, E., 1925. Missione zoologica del Dott. E. Festa in Cirenaica. XVII. Lepidotteri. Boll. Mus. Zool. Anat. Comp. R. Univ. Torino 39 : 1-9. WEHRLI, E., 1934. Lepidopteren-Fauna von Marasch in türkisch Nordsyrien. Mitt. münch. ent. Ges. 24 : 1-55. WILTSHIRE, E. P., 1957. The Lepidoptera of Iraq. Nicholas Kaye Limited, London & Bagdad. 43 Nota lepid. 17 (1/2) : 44 : 30.X1.1994 ISSN 0342-7536 Book reviews — Buchbesprechungen — Analyses Larger moths of the London area. Colin W. PLANT. xxii, 292 pp. 523 distribution maps 30.5 X 21.5 cm, hardback. London Natural History Society, 1993. ISBN 0 901009 04 0. Available from: The London Natural History Society, Publication sales secretary, 3 Chats- worth Gardens, West Harrow, Middlesex HA2 ORS, UK. Price: 19295 This volume can be considered to be the sequel to ‘The butterflies of the London area’ produced by the same author in 1987. It is however very different in both format and content. The moth volume has a much larger format, there are no coloured photographs and several species are treated per page. It will therefore not be attractive to the general public; the ‘moth hunter’ however will find this book extremely interesting and useful. The London area is defined as that area within 20 miles from St. Paul’s Cathedral. All available records have been collated, but on the distribution maps, only the data obtained between 1980 and 1991 are plotted, although the older records are given in the usual way (open circles) for the rarer species. A transparent overlay with maps showing the built-up areas, woodland and chalk is provided. Over this ten year period, 84% of the 856 tetrads (a tetrad = 2 x 2 km) have been covered, although to varying degrees — a remarkable achievement. A total of 715 macrolepidoptera have been recorded from this area, although 66 were not noted during the ten year recording period. The text for each species is very readable, and reminiscent of that in South’s ‘Moths of the British Isles’. Foodplants known to have been recorded within the area are listed for each species, and also the years of the oldest and most recent records. Appendices give a check list of species for each vice-county in the area, the National Red Data List category, and the number and percentage of tetrads in which each resident species was recorded. This work will undoubtedly be a very valuable tool for London naturalists, conservation officers and lepidopterists. It provides an ideal basis for further studies and the author and record contributors are to be congratulated on their efforts. Their example should be followed. S. WHITEBREAD 44 Nota lepid. 17 (1/2) : 45-52 ; 30.X1.1994 | ISSN 0342-7536 Geographical varıation in wing pattern of Micropterix maschukella Alpheraky, 1876 (Lepidoptera : Micropterigidae) Michail V. Kozıov Laboratory of Ecological Zoology, Department of Biology, University of Turku, FIN-20500 Turku, Finland Summary Five discrete types of forewing pattern can be found within populations of Micropterix maschukella Alphéraky. In Lagodekhi, Eastern Georgia, frequen- cies of wing pattern types were the same for males and females ; no differences were found between the two study years. The frequency of wing pattern type was therefore considered to be a population specific character and was used to study geographical variation. Phenetic resemblances of 15 samples from the Crimea and Caucasus correspond in general to the spatial proximities of the sampling sites. Three geographically consistent units were distinguished : northern (Crimea and Krasnodar district), south-western and eastern. A clear allopatric differentiation within the species was found, but there was no corre- sponding variation in the male genitalia. Resume Parmi les populations de Micropterix maschukella Alphéraky, on trouve cinq types discrets de dessin des ailes antérieures. A Lagodekhi, Georgie orientale, la fréquence des types de dessins des ailes est la même pour les mâles et les femelles ; on n’a pas trouvé de différences entre les deux années de l'étude. La fréquence des types de dessin des ailes a donc été considérée comme carac- téristique des populations et utilisée pour étudier la variation géographique. Les ressemblances phénétiques de 15 échantillons de Crimée et du Caucase correspondent en général aux proximités spatiales des sites des échantillons. Trois unités géographiques consistantes ont été distinguées : nord (Crimée et région de Krasnodar), sud-ouest et est. On a constaté une nette différenciation allopatrique dans cette espèce, mais pas de variation correspondante dans les genitalia mâles. | Zusammenfassung In Populationen von Micropterix maschukella Alpheraky lassen sich fünf Typen der Vorderflügelzeichnung unterscheiden. In Lagodekhi, Ost-Georgien, 45 werden diese Zeichnungsmuster-Typen bei Männchen und Weibchen mit glei- cher relativer Häufigkeit beobachtet ; zwischen den beiden Untersuchungsjahren gab es hierbei keine Unterschiede. Die relativen Häufigkeiten der Zeichnungs- muster-Iypen wurden daher als populationsspezifisch betrachtet und als Maß für geographische Variabilität verwendet. Das Erscheinungsbild von 15 Sammel- proben von der Krim und aus dem Kaukasus läßt sich im allgemeinem mit den räumlichen Abstand der Fundpunkte in Beziehung setzen. Drei geographische Bereiche lassen sich unterscheiden : ein nördlicher (die Krim und die Gegend von Krasnodar), ein südwestlicher und ein östlicher. Innerhalb der Art wurde eine deutliche allopatrische Differenzierung festgestellt, die aber nicht mit einer entsprechenden Variation der männlichen Genitalien verbunden ist. Introduction The contrasting wing pattern is typical for almost all of the approx. 70 species of the Palaearctic genus Micropterix Hübner [1825] (HEATH, 1987). Wing pattern characteristics are widely used in determination keys (Razowskı, 1975; ZAGULAJEV, 1978; Kozıov, 1988; 1989; 1990a ; WHITEBREAD, 1992), and they are of critical importance for the identification of females, whose genitalia are very poor in specific characters. However, variation of wing pattern characteristics has not been studied in this genus and the absence of knowledge of the extent of interpopulation and geographical variation has sometimes caused taxonomic problems. The small (about 8-10 mm wing expanse) day-active irridescent moth Micropterix maschukella Alpheraky is widely distributed and very abundant in the Crimea and Caucasus. In Eastern Georgia the moths emerge at the beginning of May in the valleys ; at the altitudes 1500- 1700 m the last specimens were observed in late July. The moths feed on the pollen of several plant species, usually on elder (Sambucus nigra L.) and Philadelphus caucasicus Koehne. Sometimes they also visit flowers of Rubus spp. and Rosa spp. (pers. obs.). The investigation of wing pattern variation has been prompted by the description of a new species, Micropterix maritimella (Zagulajev, 1983) based on females originating from the population of M. maschukella in Gantiadi, Abkhasia, which I had studied for some years. Material and methods The study consisted of two parts : intrapopulation variation was investi- gated in Lagodekhi Natural Reserve (Georgia, formerly the U.S.S.R. ; 41°50’ N, 46°20’ E) ; geographical variation was studied from specimens 46 collected by the author and those kept in the Zoological Institute, St. Petersburg, Russia. Moths were sampled from inflorescences of host plants by net and im- mediately anaesthetised by chloroform. Each sample was characterized by frequencies of moths with different wing pattern (see Figs 1-5) ; males and females were recorded separately. In total, about 5,500 moths were thus investigated. Samples were compared by chi-square test or, if the sample size was very small, by the non-parametric lambda criterion. Diversity (u) was calculated according to ZHIVOTOVSKY (1982) : u = (yp, where p; is the frequency of the ith type of wing pattern (i = 1...5). The presence of geographical variation was tested by G-statistic for hetero- geneity of proportions (GABRIEL & SOKAL, 1969). Pairwise similarity Ry = LiVPixPit based on the ratio of frequencies (p;) of all the wing pattern types (1) in populations under comparison (k and |) (ZHIvotovsky, 1982) was calculated for all samples involved in the study ; similarity matrix was clustered on the base of mean arithmetic unweighed estimations of the similarity between clades. Results Wing pattern variation The general appearance of the golden pattern in M. maschukella includes two bands (basal and medial) and a subapical spot, sharply distinguished from the cupreous-brownish background. Five discrete types of forewing pattern were found. The first type has a large, almost rectangular subapical gold spot laying along the costal margin of the wing (Fig. 1). The second type differs from the first by having a small dark spot within this gold spot (Fig. 2). In both the third and fourth type there are two spots (small costal and large subapical) on the costal edge. These two types differ in the form of the subapical spot, which has its maximum width either at the costal margin (type 3, Fig. 3), or towards the centre of the wing (type 4, Fig. 4). The 5th type differs from the 4th due to the absence of the costal spot ; the large subapical spot is usually not connected to the costal edge of the wing (Fig. 5), although there are some exceptions. 47 Figs 1-5. The five types (numbers 1-5) of forewing pattern of M. maschukella. The two bands, although variable in form and width, did not show any clearly recognizable types. The right and left wings of the moth usually have the same type of wing pattern. However, 30.6% of specimens collected in Lagodekhi 1989 and 28.7% in 1990 were asymmetrical. But if the pattern of the right and left wing varies independently, the expected number of asym- metrical moths would be significantly (about 2 times) higher than observed (in 1989 : expected 64.0%, G = 799.1, df = 1, P < 0.0001 ; in 1990 : expected 56.9%, G = 291.9, df = 1, P< 0.0001). In spite of the high percentage of asymmetrical specimens, some genetic background of wing pattern types is assumed. But even if the variation is phenotypic only, it does not affect the conclusions of the present study. 48 1. Krasnolesye 2. Jalta 4. Alushta 6. Crimsk — 3. Gursuf 5. Staryi Crim 7. Gorjatchyi Kljuch 8. Gantiadi 12. Sukhumi 14. Nalchik 13. Besengi 15. Lagodekhi 9. Tkvarcheli 10. Kobuleti 11. Borshomi SS SE OR CRE ee 05 0.6 0.7 0.8 0.9 1.0 Fig. 6. Dendrogram based on the similarity between samples of M. maschukella in wing pattern frequency. In spite of the very low (about 10%) proportion of males in samples obtained from the inflorescences of elder in Lagodekhi, I succeeded in obtaining samples of 15-25 males from six local populations ; dif- ferences between sexes appeared to be non-significant (lambda = 0.04- 1.10). Differences between samples obtained in Lagodekhi in 1989 and 1990 from the same local populations were also not significant. Thus, I concluded that the frequencies of wing pattern types are relatively stable in time, and therefore the samples collected in different years can be compared when studying geographical variation. To increase the sample size, males and females were pooled when counting wing pattern frequencies. Geographical variation The 15 localities in the Crimea and Caucasus, significantly heterogeneous in wing pattern frequencies (n = 5242, G = 1383.6, df = 52, P < 0.0001), were included in the analysis. Clustering of the similarity matrix showed that the phenetic resemblances of samples correspond in general to their spatial proximities (Fig. 6). Three geographically consistent units were distinguished : northern (Crimea and Krasnodar district), south- western and eastern (Fig. 7). 49 20 0 1; 35% t | morphs Nalchik Er / rail / 13 e NZ +4 14 SukhumiX_12 ‚Mo TE 0 100 km Be Ser aC / Fig. 7. Geographical variation in wing pattern frequencies of M. maschukella Alph. in the Crimea and Caucasus. Localities : 1 — Krasnolesye (sample size n = 94, moths collected in 1984) ; 2 — Jalta (n = 54, 1983) ; 3 — Gursuf (n = 60, 1985) ; 4 — Alushta (n = 72, 1983); 5 — Staryi Crim (n = 36, 1913); 6 — Crimsk (n = 36, 1990) ; 7 — Gorjatchyi Kljuch (n = 22, 1988) ; 8 — Gantiadi (n = 42, 1978) ; 9 — Tkvarcheli (n = 14, 1980); 10 — Kobuleti (n = 66, 1973); 11 — Borshomi (n = 46, 1898) ; 12 — Sukhumi (n = 26, 1980) ; 13 — Besengi (n = 70, 1989) ; 14 — Nalchik (n = 54, 1989) ; 15 — Lagodekhi (n = 4550, 1990). Contours correspond to the clusters identified in Fig. 6. Groups of population : A — northern ; B — eastern ; C — south-western. Samples from the northern group demonstrated the lowest observed intrapopulation diversity (coefficient of diversity u < 2.5) ; about 90% of individuals belonged to the fourth type of wing pattern. This group was heterogeneous (n = 416, G = 59.4, df = 28, P < 0.0005) because of the most southern sample (from Gantiadi), which had an intermediate ratio of wing pattern types. If this sample is excluded from the consi- deration, the northern group becomes homogeneous in relation to wing pattern frequencies (n = 374, G = 28.3, df = 24, P< 0.251). Both the south-western and eastern groups are significantly hetero- geneous (n = 126, G = 30.1, df = 8, P < 0.0005, and n = 4700, G = 168.2, df = 12, P < 0.0001, respectively), and more diverse than the northern one (u = 3.5-4.5). In the south-western group the pattern number five was most abundant, in contrast to the eastern group where the second and fourth types had highest frequencies. The small number of localities being compared did not allow investigation of the geo- graphical variation of wing pattern within these groups. 50 Discussion Like leaf miners of the family Nepticulidae (MENKEN, 1990), M. maschu- kella bear characteristics which appear to facilitate rapid speciation : they occur in small isolated populations, the detritophagous cater- pillars and pollen-eating adults have only a few (if any) competitors, and their food resources are highly predictable. But, in contrast to sympatric speciation in Nepticulidae and Yponomeutidae (MENKEN, 1990 ; MENKEN et al., 1992), Micropterigidae demonstrate mostly geo- graphical (allopatric) differentiation. No differences in genitalic structure were found between males of M. maschukella with different wing patterns, or between local populations in Lagodekhi area and between populations from different geographical groups (Kozıov, 1990b and unpublished data). According to the re- cognition concept of species argued by PATERSON (1985), this may be due to visual recognition of the opposite sex in this moth species. In this respect M. maschukella is similar to butterflies, which often de- monstrate strict interspecific differentiation in wing pattern features, while male genitalia are quite similar, i.e. in the genus Erebia Dalm. (WARREN, 1936). Thus, the data obtained showed clear allopatric differentiation within the species, which, however, demonstrated no corresponding variation in male genitalia structure. It is possible that a complex of subspecies or even sibling species may exist under the name Micropterix maschu- kella, but at the present level of knowledge the differentiation in wing pattern is not sufficient to ascribe a taxonomic rank to geographically separated populations. | Acknowledgements I am very grateful to V. Pavliashvili and all the staff of the Lagodekhi reserve for their hospitality during my stay in Georgia. I am greatly indebted to M. Motorkin for his assistance in collecting the moths, E. Zvereva for fruitful discussion, E. Haukioja, N. P. Kristensen, M. R. McClure and T. Vuorisalo for their helpful comments and improvement of the text. The work was sup- ported by the Plant Protection Institute (St. Petersburg, Russia) and the University of Turku Foundation (Finland). References GABRIEL, K. R. & SokAL, R. R., 1969. A new statistical approach to geo- graphic variation analysis. Syst. Zool. 18 : 259-278. SA HEATH, J., 1987. A check list of the genus Micropterix Hübner, [1825] (Lepi- doptera : Zeugloptera, Micropterigidae). Entomologist’s Gaz. 38 : 205-207. Kozıov, M. V., 1988. Short review and key for determination of Micropterix Hbn. (Lepidoptera, Micropterigidae) species of the Palaearctic. 1. Mor- phological description and results of investigation of Dr. H. G.Amsel type material. Vest. Zool. 1988 (4) : 8-14. [In Russian] Kozıov, M. V., 1989. Short review and key for determination of Micropterix Hbn. (Lepidoptera, Micropterigidae) species of the Palaearctic. 2. Key for determination [pt 1]. Vest. Zool. 1989 (6) : 26-31. [In Russian] Kozlov, M.V., 1990. Short review and key for determination of Micropterix Hbn. (Lepidoptera, Micropterigidae) species of the Palaearctic. 3. Key for determination [pt 2]. Vest. Zool. 1990 (2) : 21-26. [In Russian] Kozıov, M. V., 1990b. Variation of Micropterix maschukella Alph. (Lepi- doptera, Micropterigidae) in local populations of Abkhasia and Eastern Georgia. Jn Problemy sovremennoi biologi [Problems of modern biology], pp. 29-33. VINITI, Moscow (Dep. No. 641-B90). [In Russian] MENKEN, S. B. J., 1990. Population structure and evolution in sexual and parthenogenetic leaf mining moths (Lepidoptera, Nepticulidae) : why so little speciation? Symp. Biol. Hung. 39 : 349-353. MENKEN, S. B. J., HERREBOUT, W. M. & Wieses, J. T., 1992. Small ermine moths (Yponomeuta) : their host relations and evolution. A. Rev. Ent. 37 : 41-66. PATERSON, H. E. H., 1985. The recognition concept of species. In VRBA, E. S. (Ed.) : Species and speciation. Transvaal Museum Monograph No. 4, pp. 21-29. Pretoria. RazowskI, J., 1975. Motyle (Lepidoptera) Polski, II. Homoneura. Monografie fauny Polski 5 : 1-96. WARREN, B. C. S., 1936. Monograph of the genus Erebia. 407 p. + 104 pl. London. WHITEBREAD, S.E., 1992. The Micropterigidae of Switzerland, with a key to their identification. Nota lepid. Suppl. 4 : 129-143. ZAGULAJEV, A. K., 1978. Fam. Micropterigidae. Jn MEDVEDEV, G. S. (Ed.) : Key for determination of insects of the European part of the U.S.S.R. 4 (1), pp. 40-43. Leningrad. ZAGULAJEV, A. K. 1983. New and little known species of moth families Tineidae, Micropterigidae and Pterophoridae (Lepidoptera) ofthe USSR and adjacent countries. Ent. Obozr. 62 : 106-122. [In Russian] Zuivotovsky, L. A., 1982. Population characteristics based on polymorphic traits. Jn Fenetica populjatsij [Phenetic of populations], pp. 38-44. Moscow. [In Russian] = Nota lepid. 17 (1/2) : 53-72 ; 30.X1.1994 ISSN 0342-7536 Eine weitere endemische Hepialide aus den Alpen: Pharmacis claudiae sp. n. (Lepidoptera : Hepialidae) Philipp M. Krıstar*, Norbert HIRNEISEN** & Axel STEINER*** * Pankratiusstr. 2, D-68642 Biirstadt ** Hauptstr. 12, D-72149 Remmingsheim *** Bruchsaler Weg 6, D-76327 Wöschbach Summary Another new endemic Hepialid from the Alps: Pharmacis claudiae sp. n. (Hepialidae) — A new species allied to Ph. bertrandi (Le Cerf) and Ph. ansel- minae (Teobaldelli) is described from the Italian Alps (Aosta, Valtournenche) based on a series of males. It differs from the related taxa in wing markings, genital morphology and male activity period. The probably brachypterous female is still unknown. Resume Une espece nouvelle proche de Ph. bertrandi (Le Cerf) et de Ph. anselminae (Teobaldelli), est décrite des Alpes italiennes (Aoste, Valtournanche), fondée sur une série de mâles. L’espéce se distingue des taxa voisins aussi bien par Vhabitus que par les genitalia. En plus, elle diffère de Ph. anselminae et de Ph. carna (Denis & Schiffermüller) par la phase d’activité du mâle. La femelle, probablement brachyptère, reste encore inconnue. Zusammenfassung Pharmacis claudiae sp. n. wird aus den italienischen Alpen (Aosta, Valtour- nenche) anhand einer Serie von männlichen Faltern beschrieben. Die Art unterscheidet sich sowohl habituell als auch genitalmorphologisch von den verwandten Taxa Ph. bertrandi (Le Cerf) und Ph. anselminae (Teobaldelli). Zu Ph. anselminae und Ph. carna (Denis & Schiffermüller) bestehen zudem Unterschiede in der Aktivitätsphase der Männchen. Das vermutlich brachyptere Weibchen ist noch unbekannt. Einleitung Die Entdeckung einer bisher unbekannten Lepidopteren-Spezies in Mitteleuropa ist in heutiger Zeit ein seltenes Ereignis. Nach selbstkri- tischer Beurteilung der vorliegenden Fakten, dem Vergleich von Serien 53 von Genitalpräparaten und nach Einbeziehung genealogischer Über- legungen halten die Verfasser die Aufstellung eines neuen Taxons im Artrang für gerechtfertigt. Es ist wohl daher auch vertretbar, über die nüchterne wissenschaftliche Dokumentation hinaus eine kurze Beschrei- bung der Entdeckung des neuen Taxons beizufügen. Die Entdeckungsgeschichte von Pharmacis claudiae (P. M. Kristal) Am 28. Juli 1992 unternahmen wir (Claudia Kuon, Norbert Hirneisen und Philipp M. Kristal) während unseres Aufenthalts im Aosta-Tal (Italien) eine Tagestour zum Monte Cervinio (Matterhorn) am Ende des Valtournenche. Von der Talstation in Breuil-Cervinia fuhren wir mit der Seilbahn zum Giomein, also zur ersten Station auf ca. 2100 m Höhe. Von dort streiften wir durch das Gelände und fanden neben vielen hochalpinen Tagfalterarten auch einen Wurzelbohrer. Nach an- strengender Jagd auf den in unruhigem Zickzackflug über die alpinen Matten fliegenden Falter wurde diese Pharmacis “carna” von Norbert Hirneisen zur Präparation und Artbestimmung an Philipp Kristal über- geben, der diese “etwas dunkle fusconebulosa” gerne übernahm, da ihm weder die eine noch die andere Art aus seinen 12 Aufenthalten im Aosta-Tal bekannt war. Nach unserem Abstieg stand für uns fest, daß wir am Abend in der Nähe, “irgendwo über 2000 m” Lichtfang betreiben wollten. Da Philipp Kristal durch seine jahrelange Tätigkeit für das naturwissenschaftliche Museum in St. Pierre, Aosta-Tal, eine Genehmigung zum Befahren gesperrter Wege im gesamten Tal besaß, war nur das Problem des An- fahrens eines Leuchtplatzes gegeben. Erfahrungsgemäß erweisen sich auf der Karte verzeichnete Wege oftmals als nur für allradgetriebene Fahrzeuge befahrbar, so daß wir erst nach mehreren Fehlversuchen schließlich einen schönen Platz oberhalb von Antey-Saint-André bei Telınaud auf ca. 2200 m Höhe fanden. Das Terrain dort war spärlich bewachsen und auf den wenigen üppiger bewachsenen Matten grasten die Rinder der nahen Alpe. Nur noch einzelne Lärchen und Föhren umgaben uns, wir befanden uns dort offensichtlich direkt an der Waldgrenze. Dieser Platz erinnerte mich sehr stark an die mir bekannten Fundorte von Pharmacis anselminae im Val di Valeille und im Champorcher-Tal, da auch dort eine relativ starke Beweidung der Ph. anselminae-Biotope stattfindet. Diese Be- weidung scheint den Wurzelbohrern offensichtlich nicht zu schaden, wahrscheinlich ist dieser Umstand für die Ph. anselminae-Populationen sogar förderlich, wenn nicht sogar überlebenswichtig. 54 Nachdem uns Claudia Kuon trotz größter Schnakenplage ein Abend- essen in der “Feldküche” gezaubert hatte, konnten wir gut gestärkt mit dem Aufschlagen unserer beiden Leuchttürme beginnen, die wir mit einer Distanz von ca. 150 m errichteten. Der Anflug war, durch einen vom Tal aufsteigenden Wind begünstigt, sehr gut und wir hatten alle Hände voll zu tun, um alle anfliegenden Arten qualitativ und — soweit möglich — auch quantitativ zu erfasen. Gegen 1 Uhr Sommerzeit begab sıch jeder noch einmal zu seinem Turm, um nach einem letzt- maligen Absuchen des Leuchtplatzes mit dem Abbau zu beginnen. Uns fielen sofort die erst jetzt in Anzahl anfliegenden Wurzelbohrer auf, die wegen ihrer Größe, der sehr dunklen, fast schwarzen Grundfarbe und der grellweißen Zeichnung nicht sofort einer uns bekannten Art zuzuordnen waren. Durch die Distanz zwischen unseren Leuchttürmen konnten wir uns nicht sofort verständigen, und erst vor der Abfahrt konnten wir uns über diesen Bohrer unterhalten. Da Norbert Hirneisen nur ein Belegtier mitgenommen hatte, erwies es sıch als vorteilhaft, daß Philipp Kristal im Hinblick auf die Lokalsammlung des Museums in St. Pierre zwölf Tiere im Fangglas hatte, so daß genügend Anschauungs- material vorhanden war. Wir saßen in dieser Nacht noch lange über dieser Hepialide, die nach den Abbildungen bei FORSTER & WOHLFAHRT (1960) keiner uns bekannten Art zuzuordnen war. Überdies fiel uns auf, daß nach den Angaben dort Pharmacis carna und Ph. fusconebulosa ebenfalls ausscheiden müßten, da carna erst gegen Morgen zum Licht kommt, fusconebulosa jedoch in der späten Abenddämmerung fliegt. Aus eigener Erfahrung wußten wir, daß anselminae tagaktıv ist. “Unser” Bohrer jedoch kam ziemlich genau kurz nach 24 Uhr MEZ zum Licht. Da die Zeit des Paarungsflugs in der Regel genetisch festgelegt ist, kamen wir überein, daß wir uns diesen Bohrer doch noch näher ansehen müßten, denn es sei nicht auszuschließen, daß wir eine weitere, im Aosta-Tal endemische Art vor uns hätten, nachdem der Endemit Pharmacis anselminae auch erst vor ca. 15 Jahren von Teobaldelli im Aosta-Tal entdeckt worden war. Wir opferten unsere restliche Urlaubszeit diesem Unterfangen und konnten in drei Nächten noch weitere 6 Tiere am Licht erbeuten. Die Nachsuche nach Puppen oder Weibchen bei Tage, wie bei anselminae schon erfolgreich praktiziert, blieb jedoch in diesem Falle ohne greif- bares Ergebnis. Beim Anfertigen der Genitalpräparate stellte sich her- aus, daß wir, wie vermutet, nur männliche Falter am Licht erbeutet hatten und daß das Tier von oberhalb Breuil di Cervinio zur gleichen Art gehörte wie die Tiere von Antey-Saint-André. 55 Pharmacis claudiae Kristal & Hirneisen sp. n. Locus Typicus : Italia, Aosta, Valtournenche oberhalb Antey-Saint-Andre (1), 2200 m. Hoıotypus &: Italia, Aosta, Valtournenche oberhalb Antey-Saint-André, 2200 m, 31.7.1992, Lichtfang, leg. Kristal, coll. British Museum (Natural History) London. PARATYPEN : 19 GG gleicher Fundort wie Holotypus, 28.7., 31.7. und 1.8.1992, leg. Kristal & Hirneisen. Coll. British Museum (Natural History) London (1), coll. Staatliches Museum für Naturkunde Karlsruhe (2), coll. Museum Witt, München (2), coll. Museo di Scienze Naturali, St. Pierre, Aosta (2), coll. Kristal (9), coll. Hirneisen (3) ; 1 @ Italia, Aosta, oberhalb Breuil di Cervinio, ca. 2000 m, 28.7.1992, Tagfang, leg. Hirneisen, coll. Kristal. Hasitus (Abb. 1-2): Spannweite 38-42 mm, © 40,5 mm (n = 17). Vorderflügellänge 18-20,5 mm, @ 19 mm (n = 17). Kopf und Thorax dunkelbraun. Vorderflügel dunkelbraun, im postmedianen Bereich mit zwei weißen, fast parallel verlaufenden Fleckenreihen, von denen die äußere nicht durch Wische mit dem Außenrand verbunden ist. Auch die übrige Flügelfläche ist mit mehr oder weniger deutlich ausgebildeten weißen Flecken überdeckt. Alle weißen Zeichnungselemente sind, wie bei Ph. carna, durch doppelte dunkle Linien begrenzt, die hellbraun ausgefüllt sind. Die bei Ph. carna in den Apex ziehende, kettenartig verbundene, hellbraun umsäumte, dunkle Fleckenreihe ist bei Ph. clau- diae als dunkle, hellbraun begrenzte Binde ausgebildet, die ca. 3 mm vom Analwinkel entfernt entspringt und 2 bis 3 mm vor dem Apex auf die Costa trıfft. Die Fransen der Vorderflügel sind deutlich hell und dunkelbraun gescheckt, ebenso die der Hinterflügel bis vor den Analwinkel. MANNLICHER GENITALAPPARAT (2) (Abb. 6-11) : Valven sehr lang und schmal, schwach gebogen. Vinculum auffallend durch die konvexe Ausbuchtung am ventralen Rand und die starke konkave U-förmige Einbuchtung zwischen den zwei stark sklerotisierten, triangulären Vin- culumfortsätzen. Pseudoteguminalplatte (Mesosoma bei NIELSEN & (!) Auf eine genauere Beschreibung der Fundlokalität wird hier mit Rücksicht auf den endemischen Charakter der Art und die damit verbundene Schutzproblematik verzichtet. Die Erfahrungen bei Pharmacis anselminae haben gezeigt, daß es leider genug verantwortungslose sogenannte “Entomologen” gibt, die sich nicht scheuen, an einem Tag an einer Flugstelle von einer nur sehr lokal vorkommenden Art mehrere Dutzend “Belegtiere” einzusammeln. Diesem Sammeltourismus soll kein Vorschub geleistet werden. (2) Die (bei den Hepialiden noch immer uneinheitliche) Genitalterminologie richtet sich nach NIELSEN & KRISTENSEN (1989). 56 Abb. 1, 2. Pharmacis claudiae sp. n. 4, Italien, Aosta, Valtournenche. 1 — Holotypus ; 2 — Paratypus. Rosinson, 1983) im oberen Teil breit, medial nicht verschmolzen, die oberen medialen Spitzen in der Form variabel, teils schmal, teils breiter, gelegentlich papageienschnabelartig geformt, aber immer einheitlich schwach sklerotisiert ; Pseudoteguminalarme verschmolzen und eine stark gefaltete Rinne bildend. Ventraler Rand der Juxta flach konkav gebogen. DIFFERENTIALDIAGNOSE : Im Habitus von Ph. anselminae unter- schieden durch die bedeutendere Größe (Spannweite 38-42 mm gegen- 57 Abb. 3. Pharmacis bertrandi (Le Cerf) €. Frankreich, Hautes-Alpes, Chamonix. hellbraun umrandeten Fleckenzeichnungen (bei anselminae hellocker- braun und ohne irgendwelche Umrandung) und durch die auf beiden Flügeln deutlich hellbraun-dunkelbraun gescheckten Fransen (bei ansel- minae Hinterflügelfransen einfarbig, Vorderflügelfransen manchmal mit schwach angedeuteter Scheckung am Tornus). Von Ph. bertrandi ım Habitus unterschieden durch die dunkelrotbraune bis schwärzlich- braune Grundfarbe (bei bertrandi hell- bis mittelbraun) sowie durch die gescheckten Fransen (bei bertrandi einfarbig). Im männlichen Genital- apparat unterscheidet sich claudiae von beiden Arten durch die sehr langen, schmalen Valven (bei bertrandi und anselminae wesentlich kürzer, nahezu erdnußförmig) und durch die U-förmige Einbuchtung zwischen den triangulären Vinvulumfortsätzen, von bertrandi außerdem durch das Fehlen der sternförmigen Sklerotisierung an der oberen medialen Spitze der Pseudoteguminalplatte und durch die unterschied- liche Form des ventralen Randes der Juxta. Ph. claudiae ist nachtaktiv, Ph. anselminae tagaktiv. WEIBCHEN : noch unbekannt. Die nahe Verwandtschaft zu den Arten mit brachypteren Weibchen, die (soweit bisher bekannt) stark einge- schränkte, endemische Verbreitung sowie die Tatsache, daß alle am Licht anfliegenden Tiere Männchen waren, deuten darauf hin, daß es sich auch bei Ph. claudiae um eine im weiblichen Geschlecht kurz- flügelige Art handeln könnte. AKTIVITÄTSPHASE : Die Aktivitätsphase der Männchen fällt ziemlich genau auf die Zeit zwischen 0 und 1 h MEZ. Ein einzelnes, wohl auf- 58 Abb. 4, 5. Pharmacis anselminae (Teobaldelli), Italien, Aosta, Val Cogne. 4 — G ; 5 — ©. gescheuchtes Tier wurde am Nachmittag gegen 14.30 h MEZ in reißen- dem Flug über die alpinen Rasen fliegend gefangen. Dies dürfte einen Ausnahmefall darstellen, da am betreffenden Fundort trotz Aufmerk- samkeit kein weiteres tagaktives Tier beobachtet wurde. VERBREITUNG: Italienische Alpen, Aosta, Valtournenche zwischen Breuil-Cervinia an den Südhängen des Matterhorns und Antey-St.- Andre. Die bekannten Areale von Ph. anselminae und Ph. claudiae 59 Pseudoteguminalplatte Zwischenplatte Pseudoteguminalarm Valve Juxta triangulärer Fortsatz Tegumen Abb. 6. Pharmacis claudiae sp. n. männliche Genitalien. Italien, Aosta, Valtournenche. Holotypus. Der dorsale Teil der Juxta proximal umgebogen. Maßstab 0,5 mm. sind damit nur um ca. 25 km (Luftlinie) voneinander entfernt, aber durch das tief eingeschnittene Valle d’Aosta getrennt, das für Arten mit flugunfähigen Weibchen eine unüberwindbare Barriere darstellt. Beide Arten müssen das Würmglazial in unvergletscherten Refugien über- dauert haben, Ph. claudiae nördlich des Aostatals und Ph. anselminae südlich davon. VERTIKALVERBREITUNG : Die bisher bekannten Fundstellen liegen in Höhen zwischen 2000 und 2200 m. BıoTor : Alpine Rasen oberhalb der Waldgrenze (Abb. 23). FLuGzeir : Nachgewiesen vom 28. Juli bis zum 1. August. Es ist zu vermuten, daß Ph. claudiae eine ähnlich kurze Flugzeit hat, wie sie für Ph. bertrandi anhand der vorliegenden Daten zu vermuten ist und für Ph. anselminae beobachtet wurde. Eine kurze Flugzeit von einer Woche bei sehr lokaler Verbreitung erklärt sowohl die späte Entdeckung der Arten Ph. claudiae und Ph. anselminae als auch das spärlich vor- handene Material bei Ph. bertrandi. Es sollte an dieser Stelle auch die von DANIEL (1950) für Ph. carna aufgestellte Vermutung einer nur eintägigen Lebendauer der Männchen in die Überlegungen zur Biolo- 60 gie der Arten einfließen, obwohl gerade Ph. carna eine ausgedehnte Flugzeit hat. DERIVATIO NOMINIS : Die Namensgebung erfolgt zu Ehren unserer be- zaubernden Kollegin, Frau Claudia Kuon, die uns in den kalten Hoch- gebirgsnächten immer begleitete, verpflegte und die erschöpften Ento- mologen im Morgengrauen sicher nach Hause fuhr. Sie war trotz des mehr als 60 km weiten Anfahrtsweges durchs Gebirge und der dort oben herrschenden Schnakenplage stets gut aufgelegt und half uns somit in nicht zu unterschätzender Weise. Verwandte Arten Pharmacis bertrandi (Le Cerf, 1936) UNTERSUCHTES MATERIAL: | @ mit 2 Etiketten : “Hepialus Bertrandi Le Cerf 20.V11.1954 [France,] Chamonix H[au]t[es] Alpes” und “coll. Th. Witt München/ Weiden Ht. Alpes Chamonix 20.V11.1954 leg. Rungs”, coll. Museum Witt, München. Hasitus (Abb. 3) : Spannweite 37 mm (n = 1) ; TEOBALDELLI (1977) und FREINA & Wırr (1990) : 35-39 mm. Vorderflügellänge 17 mm (n= 1). Die Zeichnungselemente sind ähnlich wie bei Ph. carna an- gelegt. Die Grundfärbung ist etwas heller, die weißen Flecken sind weit spärlicher vorhanden. Zwei bis drei kleine Fleckchen bilden, deutlicher als bei den verwandten Arten, eine weiße Basalstrieme. Die kettenartige, aus hellbraun umrandeten runden Fleckchen bestehende Binde ent- springt hier ebenfalls im Analwinkel des Vorderflügels, zieht gerade verlaufend zur Costa und trifft ca. 1 mm vor dem Apex auf die Costa. Die Fransen aller Flügel sind einfarbig dunkelbraun. MANNLICHER GENITALAPPARAT (Abb. 12): Valven gebogen, kürzer als bei Ph. claudiae, in der Mitte etwas eingeschnürt, dadurch in der Form an eine Erdnuß erinnernd. Ventraler Rand des Vinculum regel- mäßiger gerundet als bei Ph. claudiae und Ph. anselminae. Einbuchtung zwischen den triangulären Vinculumfortsätzen grob U-förmig bzw. (am ventralen Rand der stark sklerotisierten Zone) abgerundet W-förmig (vgl. auch TEOBALDELLI, 1977, Abb. 3). WEIBCHEN : Die Weibchen sind brachypter. AKTIVITÄTSPHASE : Den Verfassern sind keine Angaben bekannt. VERBREITUNG : Französische Alpen, Alpes-Maritimes (Guil-Tal) und Hautes-Alpes (Abries). Das vorliegende Belegstück ist mit “Chamonix” bezettelt, wobei nicht eindeutig klar ist, ob es sich, wie auf dem Etikett angegeben, um einen Ort in den Hautes-Alpes oder vielleicht um das 61 bekannte Chamonix am Mont Blanc (Haute-Savoie) handelt, wodurch das bekannte Verbreitungsareal der Art beträchtlich erweitert würde. VERTIKALVERBREITUNG : 1900-2400 m (FREINA & Wirt, 1990). BıoToPr : Alpine Kurzrasen (FREINA & Witt, 1990). FLuGzEIT : Zweite Julihälfte bis August (FREINA & Witt, 1990). Vor- liegende Funddaten vom 17. Juli bis 20. Juli. Pharmacis anselminae (Teobaldelli, 1977) UNTERSUCHTES MATERIAL : 16 @@, Italia, Aosta, Valle di Valeille südlich von Lillaz (Cogne-Tal), 1900 m, 12.7.1990, leg. & coll. Kristal. 2 35, Italia, Aosta, Val di Champorcher, Umgebung Rifugio Dondena, 2000 m, 7.7.1991, leg. & coll. Kristal. Hasitus (Abb. 4) : Spannweite 32-33,5 mm, © 32,6 mm (n = 6) ; TEo- BALDELLI (1977) und FREINA & Witt (1990) : 28-33 mm. Vorderflügel- länge 14,5-16,5 mm, © 15,1 mm (n = 18). Im Habitus erscheinen die Tiere untereinander recht einheitlich. Die Grundfärbung aller Flügel ist ein schwärzliches Dunkelbraun, in welchem auf den Vorderflügeln unregelmäßige, nicht scharf begrenzte, hellere, weißlichgraue Fleckchen stehen. Hellbraune Zeichnungselemente als Umrandung der helleren Flecken wie bei den anderen Arten der Gruppe fehlen gänzlich. Die Hinterflügel sind einfarbig dunkelbraun ohne Einmischungen, die Fran- sen aller Flügel sind ebenfalls einfarbig dunkelbraun. MANNLICHER GENITALAPPARAT (Abb. 13-18) : Valven gebogen, kürzer als bei Ph. claudiae, in der Mitte etwas eingeschnürt, dadurch in der Form an eine Erdnuß erinnernd. Ventraler Rand des Vinculum regel- mäßiger gerundet als bei Ph. claudiae aber kantiger als bei Ph. ber- trandi. Vinculum zwischen den triangulären Fortsätzen ın der Regel V-förmig eingebuchtet (zwei der untersuchten Männchen zeigen eine abweichende U-Form, Abb. 17-18) und ventral weniger stark ausge- buchtet. Vinculum stets mit fleckenartigen Sklerotisierungen entlang der Ränder (nur in Abb. 13 zeichnerisch dargestellt, aber in allen Präparaten vorhanden), die bei den anderen Arten fehlen. Pseudo- teguminalplatte insgesamt ähnlich wie bei Ph. claudiae, das obere mediale Ende aber breit abgerundet, nur schwach sklerotisiert. Ventraler Rand der Juxta flach konkav gebogen. In der Vinculumform besteht eine gewisse Variation ; ferner zeigt ein Präparat (hier nicht abgebildet) im Vinculumbereich eine pathologische Deformation. Möglicherweise handelt es sich dabei um Defekte als Folge der Isolation (Inzucht). Wir nutzen die Gelegenheit, die Genitalien von Pharmacis anselminae in Serie abzubilden. Die einzige bisher verfügbare Abbildung, das Foto eines ungünstig 62 Abb. 7-12. Männliche Genitalien von Pharmacis spp. 7-11 — Ph. claudiae sp. n., Paratypen. Italien, Aosta, Valtournenche ; 12 — Ph. bertrandi (Le Cerf). Frankreich, Hautes Alpes, Chamonix. Maßstab 0,5 mm. 63 gelagerten Quetschpräparates zusammen mit einer unzulänglichen Beschreibung (TEOBALDELLI, 1977) hat die Beurteilung dieses Taxons eher erschwert als erleichtert und FREINA & Wirt (1990) sogar zu der Vermutung veranlaßt, daß anselminae und bertrandi als konspezifisch aufzufassen seien. WEIBCHEN : Die Weibchen sind brachypter (Abb. 5). AKTIVITATSPHASE : 9-16 h mit Höhepunkt zwischen 10 und 12 h (Teo- BALDELLI, 1977 ; 1979). VERBREITUNG : Italienische Alpen, Aosta, bisher nur aus der Umgebung des oberen Cogne-Tals südlich des Valle d’Aosta bekannt: Valle di Valeille und Vallone di Urtier (Umg. Peradza) sowie Val di Champorcher (Umg. Dondena). VERTIKALVERBREITUNG : 1800-2500 m (TEOBALDELLI, 1977 und eigene Beobachtungen) (). Biotop : Alpine Rasen oberhalb der Waldgrenze. Die bekannten Bio- tope sind z. T. stark beweidet. FLuGzeir : Anfang bis Mitte Juli von sehr kurzer Dauer. Nach den Angaben bei TEOBALDELLI (1977), den Funddaten der bei FREINA & Witt (1990) abgebildeten Tiere sowie den Belegstücken in der coll. Kristal vom 7. Juli bis 15. Juli. Pharmacis carna (|Denis & Schiffermüller], 1775) UNTERSUCHTES MATERIAL: 6 4, Italia, Trento/ Brescia, Monte Tremalzo, 1800 m, Daten vom 27.7.-3.8.1974, leg. Ströhle, coll. Kristal. Hasitus : Spannweite 32-37 mm, © 34,4 mm (n = 6) ; FREINA & Witt (1990) : && 25-34 mm, 99 32-44 mm. Vorderflügellänge 15,5-17 mm, © 16,3 mm (n = 6). Die Grundfarbe der Flügel ist ein dunkleres Mittel- braun, in dem auf den Vorderfliigeln unregelmäßige helle Zeichnungs- elemente stehen, die wiederum hellockerbraun umrandet sind. Vom Analwinkel ausgehend läuft eine aus ockerfarbenen Ringen bestehende Fleckenreihe direkt in den Apex des Vorderflügels. Die Fransen der Vorderflügel sind kaum erkennbar gescheckt, fast einfarbig mittelbraun, die Hinterflügelfransen sind am Grunde einfarbig dunkelbraun, an den Spitzen etwas heller gefärbt. () TEOBALDELLI (1977) machte verschiedene Aussagen. Die Daten der Typenserie lauten “bei 2000 m” bis “2500 m” ; im Text finden sich außerdem die Angaben “in Höhen zwischen 1800 und 2500 m” (S. 38) sowie “zwischen 1900 und 2200 m ... bei 2500 m” (S. 41). 64 Abb. 13-18. Pharmacis anselminae (Teobaldelli), männliche Genitalien. Italien, Aosta, Val Cogne. Maßstab 0,5 mm. MANNLICHER GENITALAPPARAT (Abb. 19-20): Valven länger und schlanker als bei Ph. bertrandi und Ph. anselminae, jedoch nicht so lang wie bei Ph. claudiae. Ventraler Rand des Vinculum konkav ein- gezogen, trianguläre Vinculumfortsätze schwächer ausgebildet als bei den Vergleichsarten, Einbuchtung zwischen den Fortsätzen flacher, breit U-förmig, wannenartig. Pseudoteguminalarme als zwei klappen- 65 Abb. 19-20. Pharmacis carna (D. & S.), männliche Genitalien. Italien, Trento/ Brescia, M. Tremalzo. Maßstab 0,5 mm. artige Auswüchse ausgebildet, deren Form varüert. Der kleine, schneiden- artige, sklerotisierte Fortsatz in der Mitte des medialen Randes der Pseudoteguminalplatte, der bei den Vergleichsarten vorhanden ist, fehlt bei Ph. carna völlig. WEIBCHEN : Die Weibchen haben voll entwickelte Flügel. AKTIVITATSPHASE : “Hepialus carna hat ein erstes Flugintervall im Morgengrauen, um welche Zeit einzelne dd — so ziemlich als letzte Falter — am Licht erscheinen. ... Die eigentliche Flugzeit setzt jedoch erst um 7.30 Uhr ein... Der Falter fliegt, so lange es ziemlich kühl ist, in der Sonne, etwa von 8.30 Uhr ab nur mehr an schattigen Stellen... Der Flug dauert bis gegen 12 Uhr...” (DANIEL, 1950). Diese Beobach- tungen beziehen sich auf Beobachtungen zwischen 1800 und 2200 m auf apinen Rasen an den Südhängen des Watzmannstockes in den Berchtesgadner Alpen. In tieferen Lagen konnte DANIEL (1950) Ph. carna nur am Licht nachweisen. VERBREITUNG: In Europa in den Alpen und den osteuropäischen Ge- birgen. Ob sich die Angaben vom Ural und aus den asiatischen Ge- birgen sowie aus den subarktischen Regionen Asiens wirklich alle auf Ph. carna beziehen, bedarf noch der Klärung, ebenso die bis heute in der Literatur vertretene Meldung aus der ungarischen Tiefebene (PFITZNER, 1912 ; FREINA & Wirt, 1990). VERTIKALVERBREITUNG : In den Alpen von der montanen bis zur alpinen Höhenstufe. DANIEL (1950) gibt für die Berchtesgadener Alpen Funde zwischen 1150 und 2200 m mit einem Maximum zwischen 1700 und 2000 m an. Die bei FREINA & Witt (1990) abgebildeten alpınen Belegstücke weisen Funddaten zwischen 1300 und 1800 m auf. Ein Tier aus den Karpaten wırd mit 540 m angegeben. 66 pes Abb. 21. Form der Juxta : obere Reihe Pharmacis claudiae, mittlere Reihe Ph. ansel- minae, untere Reihe Ph. bertrandi. Maßstab 0,5 mm. Brorop : Montane und subalpine Populationen finden sich im Bereich üppigster Vegetation (DANIEL, 1950). FREINA & Wirr (1990) verweisen auf kräuterreiche, hanglagige Wiesenmatten. Die alpinen Populationen bewohnen wie die verwandten Arten alpine Rasen. FrLugzeiır : Ende Juni bis in den späten August (FREINA & Wirt, 1990). Diskussion (A. Steiner) Ohne eine Gesamtrevision der Gattung Pharmacis lassen sich keine endgültigen Aussagen über die phylogenetische Stellung der einzelnen Taxa machen. Ph. bertrandi und Ph. anselminae stehen sich ungeachtet der klaren habituellen Unterschiede genitalmorphologisch sehr nahe ; möglicherweise sind sie Schwestertaxa. Es steht zu vermuten, daß die Brachypterie im weiblichen Geschlecht in der Gattung Pharmacis nur einmal aufgetreten ist, so daß alle Arten mit kurzflügeligen Weibchen ein Monophylum bilden und das Merkmal selbst eine Synapomorphie dieser Gruppe darstellt (als weitere Synapomorphien kommen die Merkmale “ventraler Rand des Vinculum ausgebuchtet” und “schneiden- artiger sklerotisierter Fortsatz am medialen Rand der Pseudoteguminal- platte” in Frage). Der weitgehende Verlust der Lokomotionsfähigkeit der weiblichen Imago durch Brachypterie ist ein schwerwiegender Schritt im Evolu- tionsgeschehen einer Art, der gut begründet sein muß. Die Brachypterie kann durch endogene Faktoren bedingt sein: die Weibchen einiger 67 > Een E 39 & © 37 = QO anselminae = 35 = © bertrandi © 33 Q 3 X) © claudiae 31 29 Vorderflügellänge in [mm] Abb. 22. Verhältnis von Vorderflügellänge und Spannweite. Lasiocampiden und Arctiiden, die im Imaginalstadium keine Nahrung aufnehmen können und somit als Adulte nur eine kurze Lebensdauer haben, müssen bereits beim Schlupf den Großteil ihrer Eier fertig entwickelt haben. Durch das schwere Abdomen behindert sind solche Weibchen oft ausgesprochen flugtrage und bewegen sich fast nur zu Fuß, so daß eine Flügelreduktion hier nur die Einsparung ohnehin ungenutzter Körperteile bedeutet. Als Ausgleich für die fehlende Vagi- litat der Weibchen haben dann meist die Raupen Ausbreitungspotentiale erschlossen, so bei Orgyia die Windverdriftung der L,-Raupe, die dafür mit langen Haarbüscheln ausgestattet ist. Brachypterie kann aber auch eine Reaktion auf exogene Faktoren sein, etwa bei Arten, die in ungünstigen Jahreszeiten fliegen (zahlreiche Herbst-, Winter- und Frühjahrsarten, besonders unter den Geometridae) oder ın Regionen leben, in denen die Flugfähigkeit ein Nachteil sein kann, so im Falle von Endemiten auf kleinen Inseln in Gebieten mit starken Winden, die Gefahr laufen, aufs Meer verdriftet zu werden (z. B. Agrotis cunhaensis). Im Hochgebirge hat die Evolution der Bra- chypterie genau die selben Ursachen : Imagines von Arten, die die Glazialzeiten auf inneralpinen Nunatakkern — also ebenfalls in isolierter Lage — überdauerten, waren beim Verdriften von ihrer Refugiumsinsel auf die umliegenden Gletscher zum Tode verurteilt. Geringe Flugak- tivität und in der Folge Flügelreduktion erwies sich in dieser Situation als Selektionsvorteil und wurde von einer ganzen Reihe von Arten genutzt, sowohl in den “massifs de refuge” der Hochgebirge als auch 68 Abb. 23. Habitat von Pharmacis claudiae sp. n. bei Antey-St.-Andre (Aosta). im borealen Bereich (Agrotis fatidica, Xestia (Schoyenia) spp., Elophos spp., Pharmacis spp. u. a.). Doch was in den Glazialia eine Überlebens- strategie war, erweist sich heute als Hindernis für die Dispersion dieser Arten, so daß bei vielen von ihnen davon ausgegangen werden kann, daß ihr rezentes Areal sich seit dem letzten Glazial nur wenig erweitert hat. Ein auffallend geringes Dispersionspotential weisen auch manche flugfähigen Arten auf, z. B. einige Arten der Genera Erebia, Psodos und der Tribus Gnophini. (Daß die Flügelreduktion meist nur das weib- liche Geschlecht betrifft, liegt daran, daß eine rasche Geschlechterfindung 69 nur dann gewährleistet ist, wenn der suchende Partner geflügelt ist, und daß andererseits die Population den Verlust von Männchen eher verkraften kann, denn auch wenige überlebende Männchen sind in der Lage, alle vorhandenen Weibchen zu begatten, während jeder Verlust eines Weibchens den Verlust ihres Eivorrats und somit ihrer gesamten potentiellen Nachkommenschaft bedeutet.) Die Gruppe der gebirgsbewohnenden Pharmacis-Arten mit brachy- pteren Weibchen wird ım folgenden Text nach dem zuerst beschriebenen alpinen Vertreter als bertrandi-Artengruppe bezeichnet, umfaßt aber auch Ph. pyrenaicus (Donzel, 1838) und eventuell einen weiteren Ver- treter im Kaukasus. Möglicherweise existieren noch unentdeckte oder bisher mit Ph. carna verwechselte Taxa in asiatischen Hochgebirgen. Pharmacis carna (eventuell einschließlich nahe verwandter geflügelter Arten) stellt vermutlich die Schwestergruppe der Pharmacis bertrandi- Gruppe dar. Im Laufe mehrerer Eiszeiten wurde das Areal der Phar- macis bertrandi-Gruppe mehrfach zersplittert und zweifellos sind auch manche Populationen, die den Vergletscherungen nicht ausweichen konnten, ausgestorben. Wann sich im alpinen Raum die Aufspaltung in die Arten Ph. bertrandi, Ph. anselminae und Ph. claudiae vollzogen hat, läßt sich beim heutigen Kenntnisstand nur schwer beurteilen ; die deutlichen Unterschiede in Habitus und Genitalmorphologie und die stark reliktäre Verbreitung lassen jedenfalls eine Aufspaltung schon vor dem Würmglazial vermuten. Die rezente Situation stellt sich folgender- maßen dar : Es existieren eine Reihe von isolierten Populationen im Be- reich der Südwestalpen, wobei an jeder einzelnen Fundstelle nur je eines dieser Taxa nachgewiesen ist. Über ihr aktuelles Expansionspotential wissen wir wenig, die Weibchen sind aber flugunfähig und können wohl auch nicht von den Männchen transportiert werden ; als einzige Me- thode der Arealexpansion bleibt die Fortbewegung im Raupenstadium, und die dürfte bei terrestrischen Raupen, die nicht die Möglichkeit haben, sich im Jugendstadium am Faden treiben zu lassen, gering sein. Es sieht also so aus, als ob die einzelnen Populationen noch immer in etwa die Regionen bewohnen, die ihnen während des Höhepunkts des letzten Glazials als Refugien gedient haben. Zur Beurteilung der Frage, ob sich diese Populationen zu Arten differenziert haben, stehen derzeit folgende Informationen zur Verfügung: Habitus und Genital- morphologie der Imagines (bei Ph. claudiae nur der Männchen), zirka- dianer Aktivitätsrhythmus (nicht für alle Populationen), Flugzeit, Bio- tope und Vertikalverbreitung. Die zwei letzten Punkte sind beı allen Populationen so ähnlich, daß zur Beantwortung der Frage nur Mor- phologie, Anatomie, Flugzeit und tageszeitliche Aktivität bleiben. Bei den meisten Hepialidae besteht in großen und zusammenhängenden 70 Populationen in der Regel eine große Variation in der Flügelzeichnung und -färbung. Dies gilt für Ph. fusconebulosa (DE GEER, 1778) und darf wohl auch für den gemeinsamen Vorfahren der Ph. bertrandi-Arten- gruppe vorausgesetzt werden. Wenn bei einem solchen Vorfahren Eng- pässe (bottleneck-Situationen) in mehreren disjunkten Reliktpopula- tionen auftreten, wie dies während der Vereisungs- und Eisrückzugs- phasen des Pleistozäns mehrfach eingetreten sein muß, dann darf ge- radezu damit gerechnet werden, daß sich die einzelnen disjunkten Teil- populationen bezüglich des Merkmals Flügelzeichnung stark ausein- anderentwickelten, so daß selbst nahestehende Taxa (z. B. Schwester- taxa) nicht unbedingt durch ähnliche Flügelzeichnung ausgezeichnet sein müssen und daß die Flügelzeichnung alleine in einer solchen Gruppe kein verläßlicher Indikator für genealogische Verwandtschaft sein kann. Differenzen in der Genitalmorphologie und in der tageszeit- lichen Aktivität können hier eher zur Klärung verwandtschaftlicher Beziehungen herangezogen werden. Selbst in der Flugzeit scheinen sich Ph. anselminae und Ph. claudiae zu unterscheiden (wenn nicht sogar auszuschließen) : Ph. anselminae wurde nur bis zum 15.7., Ph. claudiae erst ab dem 28.7. nachgewiesen, wobei allerdings berücksichtigt werden muß, daß bisher für beide Arten nur wenige Phänologiedaten vorliegen. Bei den geographisch teils sehr eng benachbarten Taxa der Ph. ber- trandi-Gruppe bietet es sich an, durch Raupen- und Puppensuche oder Zucht einerseits die Präimaginalstadien kennenzulernen und deren Mor- phologie, Biologie und Ökologie zu vergleichen, andererseits virgine Weibchen zu erlangen und mit diesen in Biotopen der jeweils anderen Arten Anflugversuche durchzuführen, um das Vorhandensein (oder Fehlen) von Isolationsmechanismen durch Pheromoninkompatibilität, durch anatomische Differenzen in der Genitalmorphologie oder durch sich ausschließende zirkadiane Aktivitätsrhythmen nachzuweisen. Hier wie in ähnlich gelagerten Fällen öffnet sich ein weites Betätigungsfeld auch für den Amateur-Entomologen, der sich nicht auf das Anhäufen toter Imagines beschränken sollte, sondern — mit entsprechender wis- senschaftlicher Untermauerung — wertvolle Beiträge zur Kenntnis der Holomorphe und ihrer Autökologie liefern kann. Schließlich muß mit der Möglichkeit gerechnet werden, daß in den südwestlichen Alpen (besonders auf der entomofaunistisch wenig intensiv durchforschten italienischen Seite) weitere lokale Populationen von Vertretern der Ph. bertrandi-Artengruppe existieren, die noch ihrer Entdeckung harren. 7A Danksagung Wie danken Herrn Thomas Witt (München) für die Ausleihe eines Pharmacis bertrandi-Belegstücks und Herrn Wolfgang Ströhle (Weiden/Opf.) für die Überlassung einer Serie von Pharmacis carna. Literatur DANIEL, F., 1950. Beobachtung über die Lebensweise hochalpiner Vertreter der Gattung Hepialus. Mitt. münch. ent. Ges. 40 : 203-206. FREINA, J. J. DE & Witt, T., 1990. Die Bombyces und Sphinges der West- palaearktis (Insecta, Lepidoptera). Band 2. München (Edition Forschung & Wissenschaft). 134 + 6 unnumerierte S. NIELSEN, E. S. & KRISTENSEN, N. P., 1989. Primitive Ghost Moths. Morpho- logy and taxonomy of the Australian genus Fraus Walker (Lepidoptera : Hepialidae s. lat.). Monographs on Australian Lepidoptera 1 : i-xi, 1-206. NIELSEN, E. S. & Rosinson, G. S., 1983. Ghost moths of southern South America (Lepidoptera: Hepialidae). With a Summary in Spanish prepared by P. Gentili : “Mariposas Fantasmas del sur de Sudamerica”. Scandinavian Science Press, Copenhagen. 192 S. PFITZNER, R., 1912. Hepialidae. In Serrz, A. (Hrsg.) (1909-1913) : Die Gross- Schmetterlinge der Erde. I. Abteilung: Die Gross-Schmetterlinge des Palaearktischen Faunengebietes. Bd. 2: Die Palaearktischen Spinner & Schwärmer : 433-439. TEOBALDELLI, A., 1977. Eine neue Hepialus-Art aus Italien. NachrBl bayer. Ent. 26 : 38-43. TEOBALDELLI, A., 1979. Lépidoptères capturés en Val d’Aoste. Alexanor 11: 98-104 ; 145-152. 12 Nota lepid. 17 (1/2) : 73-86 ; 30.X1.1994 ISSN 0342-7536 Die Puppen der Tribus Cyclophorini Mitteleuropas (Lepidoptera : Geometridae) Jan PATOCKA Institut für Waldökologie der SAW, Stürova 2, SK-96053 Zvolen, Republik Slowakei Summary The pupae of the central European species of the tribe Cyclophorini (Lepi- doptera, Geometridae) — The pupae of 11 species of Central European Cyclo- phorini (Lepidoptera, Geometridae, Sterrhinae) are characterised, described and figured. A key for the determination of pupae of the genus Cyclophora Hübner and some biological data are added. The current system within the genus Cyclophora is discussed on the basis of pupal characters. Zusammenfassung Die Puppen von 11 Arten der mitteleuropäischen Cyclophorini (Lepidoptera, Geometridae, Sterrhinae) werden charakterisiert, beschrieben und abgebildet. Eine Tabelle zur Bestimmung der Arten der einzigen mitteleuropäischen Gat- tung Cyclophora Hübner und einige bionomische Angaben werden beigefügt. Das derzeitige System der Gattung Cyclophora wird anhand der Puppenmerk- male diskutiert. Resume L’auteur caractérise, décrit et figure les chrysalides de 11 espèces de Cyclophorini d’Europe centrale (Lepidoptera, Geometridae, Sterrhinae). Il présente en outre un tableau de détermination pour les espèces du seul genre d'Europe centrale : Cyclophora Hübner, et ajoute quelques données sur la bionomie. Discussion du système actuel de classement dans le genre Cyclophora sur la base des carac- tères des chrysalides. Diese Arbeit knüpft an PATOCKA (im Druck) an und behandelt die Puppen der Cyclophorini (Lepidoptera, Geometridae, Sterrhinae) mit der einzigen mitteleuropäischen Gattung Cyclophora Hübner. Das Sys- tem und die Nomenklatur folgen LERAUT (1980), die morphologische Terminologie (vgl. Abb. 1, 2) MosHer (1916) und McGuFFIN (1967). 73 Abb. 1-10. 1-2 — Cyclophora sp. ; 3-7 — C. pendularia ; 8-9 — C. albiocellaria : 10 — C. annulata. | — Habitusbild in Ventral-; 2 — in Lateralansicht ; 3, 7 — Clypeus, Labrum ; 4,8 — Oculi und Umgebung ; 5, 9, 10 — thorakaler Spiracularhöcker. A = Antennae, Af = Analfeld, An = Analnaht, As = abdominale Spiracula, Bh = Basalhöcker, C] - Clypeus, E = Einsenkung, F = Frons, G = Genae, Gn = Genital- naht, @, Hb = Hinterbeine, Hfl = Hinterflügel, Hn = Häkchen, Kr = Kremaster, Lbr = Labrum, Mb = Mittelbeine, Msn = Mesonotum, Mtn = Metanotum, O = Oculi, Pcl = Postclypeus, Pr = Proboscis, Prn = Pronotum, Sh = Spiracularhöcker, Str = Streifen im Analteil der Vorderflügel, Vb = Vorderbeine, Vfl = Vorderflügel, Vs = Vorderschenkel, 1-10 = 1.-10. Abdominalsegment. Tribus Cyclophorini Diese Tribus enthält in Mitteleuropa nur die Gattung Cyclophora Hüb- ner, 1822, mit folgender Charakteristik : Mittelkleine (ca. 11-14 X 2,5-3,5 mm), ziemlich gleichförmige, in Ven- tralsicht charakteristisch keilförmige Spannerpuppen (Abb. 1), vorn am breitesten und stark abgestumpft. Thorakale Spiracularhöcker ecken- artig vorspringend. Von diesen läuft parallel zum Dorsalrand der Vor- derflügel je eine erhabene Kante. Auch zwischen Clypeus und Frons eine Kante (Abb. 1-2). Färbung fahl (sand-) gelb bis rot- oder grau- 74 bräunlich. Manche Arten weisen außerdem grüne Puppenformen auf. Körper oft gesprenkelt mit Fleckenzeichnungen am Clypeus, an der Vorderflügelbasis, am Thorax und Abdomen. Längs des kantenartigen Dorsalteiles der Vorderflügel oft ein dunkler Streifen, auch die Adern manchmal + verdunkelt. Puppen nur schwach glänzend. Skulptur mittel- grob bis mittelfein, wirr runzelig, zuweilen etwas höckerig, 1.-7. Ab- dominalsegment meist mit — scharfen Punktgrübchen (Abb. 18, 45). Labrum breit, abgerundet bis trapezförmig (Abb. 3, 7, 20). Labium ganz verdeckt. Genae oben beulenartig erhaben. Die Grenze zwischen Proboscis und Genae verläuft eher quer zur Längsachse der Puppe, die zwischen den Vorderbeinen und Antennae mehr schräg. Grenze der Vorderbeine mit den Antennae 4-5 X länger, die mit den Genae etwa 1,5-3 X kürzer als die Grenze zwischen den Vorderbeinen und Oculi (Abb. 4, 8). Vorderschenkel und Enden der Hinterbeine sicht- bar. Proboscis wenig, Vorderbeine deutlich kürzer als die Mittelbeine und Antennae, diese überragen die Vorderflügel nicht (Abb. 1). Basen der Antennae, Frons, der ziemlich gut differenzierte, schmale Vertex und das Pronotum befinden sich an der abgestumpften Frontalseite der Puppe (Abb. 2). Frontolateral (Spiracular-) Höcker des Mesono- tums manchmal mit komplizierten Einschnürungen, zuweilen lappig (Abb. 9, 43). Frontolaterale Ausläufer des Metanotums relatıv spitz (Abb. 19, 39). Thorakale Spiracula undeutlich. Hinterflügel schmal, bis etwa zum Caudalrand des 3. Abdominalsegmentes sichtbar. Spira- cula am 2. Abdominalsegment deutlich, am 3. Segment verdeckt, relativ groß, elliptisch, ihre Höfe breit (Abb. 2). Zwischen dem 9. und 10. Ab- dominalsegment, dorsal und lateral, gibt es weder eine Rinne, noch Einschnitte (Abb. 11, 12). Das 9. Segment ziemlich kurz, insbeson- dere an der Ventralseite. An der Dorsalseite trägt es V- oder trapez- förmige, zuweilen abgerundete Vertiefungen, die kantenartig gesäumt sind (Abb. 11, 12, 27, 31, 33). Analfeld deutlich, meist dreieckig, manch- mal längsgefurcht. Die Analnaht zuweilen mit erhabenen Lateralwällen (Abb. 17, 22). Kremaster + so lang wie basal breit, dorsoventral abgeplattet. In der Dorsalansicht ist er entweder zungenförmig, am Ende relativ spitz (Abb. 33) ; oder am Ende abgestumpft (Abb. 12), bzw. stumpf- oder rechtwinklig (Abb. 31, 42). Ventralseite oft mit Basalhöckern und einer größeren oder kleineren Basalvertiefung (Abb. 6, 17). Von den 4 Paaren der hakenartigen Kremasterborsten entspringt D1 dorsal vor dem Ende des Kremasters und ist kleiner als die übrigen. Diese etwa gleichgroß, dicht in einer Gruppe am Kremasterende konzentriert (Abb. 54). Sonst ist der Kremaster + gerunzelt bzw. gefurcht. 75 Abb. 11-24. 11 — Cyclophora pendularia ; 12-14 — C. albiocellaria ; 15-18 — C. annulata ; 19-23 — C. albipunctata ; 24 — C. puppillaria. 11, 12, 16, 23 — Abdomenende in Dorsal- ; 13, 17, 18 — in Ventralsicht ; 14, 18 — 5. Abdominalsegment in Dorsalsicht ; 15, 20, 24 — Labrum, Clypeus ; 21 — thorakaler Spiracularhöcker. Die Puppe ruht oberirdisch an einem besponnenen Blatt der Futter- pilanze, mit Hilfe der Kremasterhäkchen und eines Gespinstgürtels “tagfalterartig” am Blatt befestigt. Die Überwinterung findet im abge- fallenen Laub bzw. in einer noch belaubten Baumkrone statt. Die Arten leben vorwiegend an Laubhölzern (nur C. suppunctaria an Kräutern) und sind in Mitteleuropa meist bivoltin (selten uni- bzw. trivoltin) ; die Puppen findet man im Sommer und von Herbst zum Frühling. Die einzelnen Arten bewohnen Laub- und Mischwälder, zuweilen die Waldsteppe. C. puppillaria ist als Wanderfalter bekannt. Wirtschaftlich sind sie fast bedeutungslos. 76 Bestimmungstabelle der Arten 2(1) Dorsal am 10. Abdominalsegment eine keilförmige (dreieckige) Vertie- fung (Abb. 11, 23, 38), zuweilen durch die Längsfurchung verdeckt und memes schaut (ADBA DD RUE Men ER emo Ah 2 Dorsal am 10 Abdominalsegment eine trapezförmige (an der Caudalseite absestumpfte)Vertiefungss(Abb 27313 Pam irn; 7 Die Vertiefung dorsal am 10. Abdominalsegment unscharf, durch scharfe Längsfurchung verdeckt (Abb. 11). Puppenfärbung trüb braun- DU Beats. EGR We SARS m. abe Set C. pendularia Die Vertiefung dorsal am 10. Abdominalsegment scharf (Abb. 23, 54). Körperfärbung sandbraun, gelbbraun bis trüb strohgelb, manchmal EMRE GUN ON EM ARABE ISIN IN TOR EN ISH IOUT. MEN... 3 Seiten der Vertiefung dorsal am 10. Abdominalsegment stark geschwun- gen, Spitze spitzwinklig vorgezogen (Abb. 38) ......... C. quercimontaria Seiten der Vertiefung dorsal am 10. Abdominalegment nicht oder kaum geschwungen, Spitze deshalb nicht vorgezogen (Abb. 12,54) ............. + Vorderflügel mit einem dunklen Fleck in der Mitte (Abb. 2). Kremaster in Dorsalsicht meist allmählich zungenförmig zugespitzt (Abb. 23) ....... nee Meee oneness vn C. annulata Vorderfliigel ohne einen dunklen Fleck in der Mitte. Kremaster in Dor- salsicht meist erst im Endteil abgerundet, oder stumpf- bis rechtwinklig ARCS DAC NO PU ODA ie ces rene WER EEE EREERRREER 5 Dunkle Sprenkelung und Flecken an Borstenbasen am Thorax und Ab- domen fehlend (Abb. 19). Labrum abgerundet trapezförmig (Abb. 20) ee AURA Vordleriiiioein Ah. 2). ast le clean C. albipunctata Dunkle Sprenkelung und Flecken am Thorax und Abdomen deutlich (Abb. 14, 55, 56). Labrum trapezförmig (Abb. 7,51) ......................... 6 Labrum mit wenig schrägen Seiten, dunkle Flecken an der Basis der Clypealborsten klein, punktförmig (Abb. 7). Kremaster in Ventralsicht mit einer tiefen, breiten Vertiefung im Basalteil, am Ende stumpf abge- RUMI CAibin fl). . sarya! teres eh. cee. nel. C. albiocellaria Labrum mit schrägen Seiten, dunkle Flecken an der Basis der Clypeal- borsten relativ groß (Abb. 51). Kremaster in Ventralsicht mit schwa- cher, eher länglicher Vertiefung, am Ende meist stumpf- bis rechtwinklig CASSER PRIRENT RER C. linearia Vertiefung dorsal am 10. Abdominalsegment im Caudalteil breit (manch- mal u-förmig abgerundet), Caudalrand nicht kürzer als die Lateralränder (ADD Zr) ee ee ae bd te ATEN ar En 8 Vertiefung dorsal am 10. Adominalsegment im Caudalteil schmal, Cau- dalrand viel kürzer als die Lateralränder (Abb. 31, 42, 47) ................ 9 Labrum — halbkreisförmig (Abb. 35). Grenze der Vorderbeine mit den Genae etwa 3 X kürzer als die mit den Oculi (Abb. 4). Kremaster in Ven- tralsicht mit deutlichen Basalhöckern. Analfeld groß, breit, Lateralwälle der Analnaht nicht stark hervortretend (Abb. 29). Vertiefung am 10. Abdominalsegment dorsal mit + scharfen Ecken (Abb. 33) .................. We — _Labrum = abgerundet viereckig (Abb. 24). Grenze der Vorderbeine mit den Genae kaum 2 X kürzer als die mit den Oculi (Abb. 8). Basalhöcker an der Ventralseite des Kremasters kaum angedeutet, Analfeld schmal, Analnaht mit starken Lateralwällen (Abb. 25). Vertiefung am 10. Ab- dominalsegment dorsal mit + abgerundeten Ecken (Abb. 27) ............... ee Kl. serien. aD ee eee à Din C. puppillaria 9(7) Puppe mit gut entwickelter dunkelbrauner Fleckenzeichnung dorsal am Thorax und Abdomen (Abb. 45). Clypeale Borsten auf dunklen Basal- flächen (Abb A): 8 LCL Lads Pee, neuen C. punctaria — Puppe ohne oder mit ganz schwacher Fleckenzeichnung (Abb. 48). Clypealborsten höchstens auf dunklen Punkten (Abb. 46) ............... 10 10(9) Puppe fahlbraungelb oder rötlichgelb gefärbt. Skulptur, besonders an den Vorderflügeln (auch an deren Basis) ziemlich fein. Labrum lateral gewölbt. Caudalseite gerade (Abb. 28). Lateralränder der Dorsalvertie- fung am 10. Abdominalsegment — einwärts gebogen (Abb. 31) ............ lien. sans rd Meer en RE C. ruficiliaria — Puppe fahl braungrau gefärbt. Skulptur gröber, z.T. höckerig. Labrum auch an der Caudalseite gewölbt (Abb. 46). Lateralränder der Dorsal- vertiefung am 10. Abdominalsegment auswärts gebogen (Abb. 47) ....... BEN a SR PR DE PRE. ui ER ee C. suppunctaria Cyclophora pendularia (Clerck, 1759) 5 6,5 © aus der Slowakei und aus Böhmen wurden untersucht. Puppe schlank mit spitzem Abdomenende. Färbung bräunlich hellgrau, dunkle Zeichnung an den Vorderflügeln spärlich, dunkel braungrau. Punkte an der Basis der Clypealborsten etwas größer, sonst dorsal am Thorax und Abdomen klein, unauffällig. Skulptur am Notum und Abdomen wirr gerunzelt, Punktgrübchen an den mittleren Abdominalsegmenten unscharf. Labrum an der Caudalseite — abgerundet (Abb. 3). Grenze der Vorderbeine mit den Oculi etwa 2 X länger als die mit den Genae (Abb. 8). Spiraculare Ausläufer am Thorax stark hervortretend mit feinen Einschnürungen, wenig gelappt (Abb. 9). Vertiefung dorsal am 10. Abdominalsegment keilförmig, hinten in etwa rechtwinklig, unscharf, durch starke Längsfurchung weniger auffällig (Abb. 11). Analfeld schlank dreieckig, Seitenwälle schwach, sonst Umgebung längsgerunzelt (Abb. 6). Kremaster wenig länger als basal breit, zum Ende zungen- förmig verschmälert, am Ende abgerundet, wirr gerunzelt. Basalhöcker an der Ventralseite deutlich, dazwischen fast keine Vertiefung. Häkchen relativ klein, dunkel graubraun (Abb. 6, 11). Raupe an Salix spp., seltener an verschiedenen anderen Laubhölzern. Laubholzbestände an Gewässerufern, mehr in tieferen Lagen, lokal. 78 Cyclophora albiocellaria (Hübner, 1789) 3 6, 4 2 aus der Slowakei wurden untersucht. Puppe ziemlich ge- drungen, hell sandgelb, deutlich und scharf schwarzbraun gezeichnet. Rückenfleckchen dick (Abb. 14). Punktgrübchen an mittleren Abdo- minalsegmenten ziemlich tief, unscharf. Labrum stumpf trapezformig mit steilen Seiten (Abb. 7). Grenze der Vorderbeine mit den Oculi kaum 2 X länger als die mit den Genae (Abb. 8). Thorakale Spiracularaus- läufer relativ stumpf (Abb. 9). Vertiefung dorsal am 10. Abdominal- segment scharf, keilförmig, Seiten wenig geschwungen, Spitze nicht vorgezogen (Abb. 12). Analfeld dreieckig, Seitenwälle der Analnaht niedrig. Sonst Ventralseite des 10. Abdominalsegmentes längsgefurcht. Kremaster meist etwas länger als basal breit, am Ende abgerundet. Ventralseite stark vertieft, Endteil längsgefurcht, sonst Furchung eher wirr. Ba sal höcker deutlich. Häkchen relativ stark und dick, braun (Abb. 12, 13). Raupe an Acer spp. In der Waldsteppe und Randzone der Waldbestände an warmtrockenen Standorten im Süden und Südosten Mitteleuropas. Cyclophora annulata (Schulze, 1775) 5 À, 5 ® aus der Slowakei wurden untersucht. Puppe relativ klein, ziemlich gedrungen, meist unscharf gezeichnet, u.a. ein charakteristischer dunkler Fleck an den Vorderflügeln (Abb. 2), dunkle dorsale Fleckchen am Thorax und Abdomen fließen oft streifenartig zusammen. An den mittleren Abdominalsegmenten tiefe, relativ scharfe Punktgrübchen (Abb. 18). Labium trapezförmig mit schrägen Seiten (Abb. 15). Grenze der Vorderbeine mit den Oculi mehr als 2 X länger als die mit den Genae. Thorakale Spiracularhöcker relativ stumpf, schwächer gewölbt als bei C. albiocellaria (Abb. 10). Dorsalvertiefung am 10. Abdominal- segment derjenigen der vorigen Art ziemlich ähnlich (Abb. 16), ebenso das Analfeld und der Kremaster, dieser ıst jedoch spitzer. Endhäkchen relativ stark, braun (Abb. 16, 17). Raupen an Acer spp., seltener an anderen Laubhölzern. Randzone der Laubwälder, Gebüsch. An mäßig feuchten bis trockenen Standorten wärmerer Lagen, verbreitet. Cyclophora albipunctata (Hufnagel, 1767) 5 4, 5 Puppen aus der Slowakei und aus Böhmen wurden untersucht. Puppe ziemlich gedrungen, sandgelbbraun oder grün, Flügelstreifen oft scharf, schwarzbraun, Fleck im Mittelteil der Vorderflügel nicht vor- 12 Abb. 25-36. 25-27 — Cyclophora puppillaria ; 28-31 — C. ruficiliaria ; 32-35 — C. porata , 36 — C. quercimontaria. 25, 29, 33 — Abdomenende in Ventralsicht ; 26, 30, 34 — thorakaler Spiracularhöcker ; 27, 31, 33 — Abdomenende in Dorsalsicht ; 28, 35, 36 — Labrum, Clypeus. handen. Wurzel-Punkte der Clypealborsten schwach (Abb. 20). Flecken- zeichnungen am Thorax und Abdomen reduziert oder fehlend (Abb. 19). Morphologisch jedoch C. annulata ähnlich. Labrum abgerundet trapez- förmig (Abb. 20), Grenze der Vorderbeine mit den Oculi mehr als 2 X länger als die mit den Genae (Abb. 4). Thorakale Spiracularhöcker mehr hervortretend, der große äußere Teil gewölbt, mit Einschnürungen (Abb. 21). Analfeld mit stärker hervortretenden Wällen. Kremaster an der Ventralseite kaum vertieft, Basalhöcker angedeutet. Häkchen ziem- lich klein, braun (Abb. 22, 23). Raupe an Betula spp., auch an Alnus spp., seltener an anderen Laub- hölzern. In Birkenhainen, Laub- und Mischwäldern an Gewässerufern. 80 Cyclophora puppillaria (Hübner, 1799) 2 © aus Frankreich (Coll. Staudinger, Museum für Naturkunde, Berlin) wurden untersucht. Färbung nach Forster & WOHLFAHRT (1981) “dunkelgrün mit einigen hellen Linien auf dem Rücken und an den Rändern der Flügelscheiden. Kopfspitzen und Kremaster leicht wein- rot”, möglicherweise jedoch auch braungelb. Exuvie trüb strohgelb, Abdomenende mehr bräunlich, Zeichnungen praktisch fehlend. Punkt- grübchen nicht allzu stark. Labrum mit geschwungenen Seiten, Clypeal- borsten ohne dunkle Punkte (Abb. 24). Grenze zwischen den Vorder- beinen und Oculi etwa 2 X länger als die zwischen Vorderbeinen und Genae. Thorakale Spiracularhöcker schwächer hervortretend, lappig (Abb. 26). Das 10. Abdominalsegment dorsal mit einer sehr breiten trapezförmigen Vertiefung, deren Ecken + abgerundet sind. Caudalrand nicht kürzer als die Lateralränder (Abb. 27). Analnaht lang mit + deut- lichen Seitenzweigen, Lateralwälle stark, Analfeld schmal, dreieckig. Kremaster an der Ventralseite ohne deutliche Basalhöcker, Basalteil vertieft, Endteil eher angeschwollen. Dorsalseite wirr gerunzelt. Ende des Kremasters abgerundet (Abb. 25, 27). Raupe in Mitteleuropa angeblich an Quercus spp. An trockenwarmen Standorten im äußersten Süden Mitteleuropas (und im Mediterran- gebiet), als Wanderfalter auch weiter nordwärts vordringend. Cyclophora ruficiliaria (Herrich-Schäffer, 1855) 2 &, 2 ® Puppen aus der Slowakei wurden untersucht. Abdomenende relativ schlank. Färbung fahlgelb, Zeichnung stark reduziert bzw. fast fehlend. Vorderflügel höchstens mit einem schmalen Analstreifen oder mit schwach verdunkelten Adern. Punkte an Borstenbasen (auch die der Clypealborsten, Abb. 28) meist kaum sichtbar. Skulptur an den Vorderflügeln, auch an deren Basis, ziemlich fein. Punktgrübchen an den mittleren Abdominalsegmenten verhältnismäßig scharf. Labrum breit, Seiten gewölbt (Abb. 28). Grenze der Vorderbeine mit den Oculi + zweimal länger als die der Oculi mit den Genae. Thorakaler Spira- cularhöcker mittelstark hervortretend, gewölbt, mit scharfen Querein- schnürungen (Abb. 30). 9. und 10. Abdominalsegment dorsal längs- gefurcht. Dorsalvertiefung am 10. Segment scharf begrenzt, schmal trapezförmig, Seitenränder schräg, viel länger als der Caudalrand und etwas einwärts geschwungen (Abb. 31). Analnaht mit schwachen Seiten- wällen, Analfeld groß. Kremaster relativ kurz, breit, Ventralseite mit angedeuteten Basalhöckern und einer breiten Vertiefung. Kremaster vor dem Ende etwas angeschwollen. Dorsalseite relativ stark, wirr gefurcht. 81 Abb. 37-47. 37-40 — Cyclophora quercimontaria ; 41-45 — C. punctaria ; 46-47 — C. suppunctaria. 37, 41 — Abdomenende in Ventral- ; 38, 42, 47 — in Dorsalsicht ; 39 — Metano- tum, 1. Abdominalsegment, limke Seite; 40, 43 — thorakaler Spiracularhöcker ; 44, 45 — Labrum, Clypeus ; 45 — 5. Abdominalsegment in Dorsalsicht. Das Ende in Dorsalansicht etwas winkelförmig. Häkchen stark, rot- braun (Abb. 31, 32). Raupe an Quercus spp. (in der Slowakei wird Q. cerris bevorzugt). Lokal an trockenwarmen Standorten. Cyclophora porata (Linnaeus, 1767) 1 4, 1 2 Puppe aus dem Naturhistorischen Museum Wien wurden untersucht. Färbung gelbbräunlich, Exuvie trüb strohgelb. Zeichnung reduziert. Vorderflügel nur mit etwas dunkleren Adern. Thorax und Abdomen fast ohne dunkle Sprenkelung. Punkte an der Borstenbasis (auch an den Clypealborsten, Abb. 35) kaum dunkler. Punktgrübchen am Abdomen unscharf, nur in der Umgebung der Spiracula deutlicher. 82 Labrum breit abgerundet, Ränder (auch der Caudalrand) gewölbt (Abb. 35). Grenze der Vorderbeine mit den Genae kurz (fast 3 X kürzer als die mit den Oculi, vgl. Abb. 4). Thorakale Spiracularhöcker mittel- stark erhaben, mäßıg gewölbt, Einschnürungen schwach (Abb. 34). Dorsale Vertiefung am 10. Abdominalsegment breit trapezförmig, eckig, Seiten schräg, Caudalrand kaum kürzer als die Lateralränder (Abb. 33). Analfeld relativ breit, Analnaht mit mäßigen Lateralwällen. Kremaster relativ kurz, vor dem Ende stumpfwinklig oder quer zur Längsachse abgestumpft. Ventralseite mit schwachen Basalhöckern, dazwischen vertieft ; sonst nur mäßig, Dorsalseite deutlicher, skulp- turiert. Häkchen ziemlich kräftig (Abb. 32, 33). Raupe an Quercus spp., Betula spp., vorzugsweise auf sandigen Böden, in Heiden u. dgl. Cyclophora quercimontaria (Bastelberger, 1897) 5 6, 5 2 Puppen aus der Slowakei wurden untersucht. Puppe mäßig gedrungen, oft dunkel braungrau gesprenkelt, Flügeladern + verdunkelt, dunkle Fleckchen an der Basis der Borsten (Abb. 39). Exuvie hell sand- bis strohgelb, Abdomenende oft dunkler. Punktgrübchen mittelscharf. Labium breit trapezförmig, Caudalrand meist konvex (zuweilen auch konkav). An der Basis der Clypealborsten die dunklen Fleckchen deut- lich (Abb. 36). Grenze der Vorderbeine mit den Genae mehr als zwei- mal kürzer als die mit den Oculi (vgl. Abb. 4). Thorakale Spiracular- höcker mittelstark, mäßig eingeschnürt und gelappt (Abb. 40). 10. Ab- dominalsegment dorsal mit einer tiefen, keilartigen Vertiefung, Seiten geschwungen, Spitze spitzwinklig, vorgezogen (Abb. 38). Analfeld mit schwachen Lateralwällen. Kremaster meist länger als basal breit, zum Ende verjüngt, am Ende selbst abgerundet. Ventralseite mit starken Basalhöckern, dazwischen vertieft, Endteil längs-, Dorsalseite eher wirr gefurcht. Häkchen stark, rotbraun (Abb. 37, 38). Raupe an Quercus spp., lokal, an trockenwarmen Standorten (Wald- steppen, Heiden). Bevorzugt buschige Eichen. Cyclophora punctaria (Linnaeus, 1758) 5 À, 5 2 aus der Slowakei wurden untersucht. Puppe ziemlich ge- drungen, sandfarben, seltener grün, ähnlich wie die vorige scharf ge- zeichnet (Abb. 45). Labrum trapezförmig mit schrägen, meist geraden Seiten, am Caudalrand manchmal etwas ausgeschnitten. Fleckchen an der Basis der Clypealborsten deutlich (Abb. 44). Grenze der Vorderbeine mit den Oculi mehr als 2 X länger als die mit den Genae (vgl. Abb. 4). 83 Thorakaler Spiracularhöcker mittelstark erhaben mit deutlichen Ein- schnürungen (Abb. 43). Dorsalvertiefung am 10. Abdominalsegment (im Unterschied zu der sonst ähnlichen vorigen Art bzw. zu C. linea- ria) schmal trapezförmig, Caudalrand viel kürzer als die schrägen, kaum geschwungenen Seitenränder (Abb. 42). Analfeld schlank drei- eckig. Lateralwälle der Analnaht ziemlich erhaben. Kremaster kaum oder wenig länger als basal breit, vor dem Ende in Dorsalsicht zu- gespitzt. Ventralseite mit Basalhöckern und einer meist kleinen, oder fehlenden Vertiefung. Dorsalseite mäßig gerunzelt. Häkchen mittelstark, rotbraun (Abb. 41, 42). Raupe an Quercus spp. In Eichenwäldern aller Art oft häufig. Cyclophora suppunctaria (Zeller, 1847) 5 4, 5 2 aus der Slowakei wurden untersucht. Der C. ruficiliaria morphologisch ähnlich. Unterscheidet sich durch die mehr fahl grau- braune Färbung (C. ruficiliaria ist eher fahlgelb), durch die gröbere und teilweise auch höckerige Skulptur an Thorax, Abdomen- und Vorderfliigelbasis (Abb. 48). Punktgrübchen am Abdomen meist weniger scharf. Abgrenzung des Labrums — lateral und auch caudal — gewölbt (Abb. 46). Das 9. und 10. Abdominalsegment dorsal längsgefurcht. Am 10. ist eine trapezförmige Vertiefung, deren Seiten ein bißchen auswärts gebogen sind. Caudalrand viel kürzer als die Seitenränder (Abb. 47). Das Abdominalsegment ventral und auch das Analfeld längsgefurcht (Abb 50 ; stärker als bei C. ruficiliaria). Kremaster ventral kaum konkav mit + starken Basalhöckern, wirr gefurcht — auch an der Dorsalseite. Kremasterende in Dorsalsicht + stumpf bis rechtwinklig zugespitzt, Häkchen mittelstark, dunkel rotbraun (Abb. 47, 50). Raupe an Kräutern wie Artemisia campestris, Hippocrepis spp., Meli- lotus spp. An trockenwarmen Standorten, z.B. Waldsteppen im süd- östlichen Mitteleuropa, lokal. Cyclophora linearia (Hübner, [1799]) 5 6, 5 2 Puppen aus der Slowakei wurden untersucht. Der C. puncta- ria ähnlich, + rötlich braunsandgelb, Sprenkelung und Flecken am Clypeus, Thorax und Abdomen und Streifen an den Vordeflügeln meist scharf (Abb. 51, 55, 56). Punktgriibchen am Abdomen relativ deutlich. Labrum trapezförmig (Abb. 51). Grenze der Vorderbeine mit den Oculi wenigstens 2 X länger als die mit den Genae (vgl. Abb. 4). Thorakale Spiracularhöcker stark hervortretend, Einschnürungen jedoch kaum angedeutet (Abb. 52). 9. und 10. Abdominalsegment dorsal — gerunzelt. 84 Abb. 48-56. 48-51 — Cyclophora suppunctaria ; 52-56 — C. linearia. 48, 56 — Metanotum, 1. Abdominalsegment ; 49, 52 — thorakaler Spiracularhöcker ; 50, 53 — Abdomenende in Ventralsicht ; 51 — Labrum, Clypeus ; 54 — Abdomenende in Dorsalsicht ‚55 — 3. Abdominalsegment in Dorsalsicht. Am 10. Abdominalsegment eine spitz keilförmige Vertiefung mit wenig oder nicht geschwungenen Seiten (Unterschied gegenüber C. querci- montaria) und kaum vorgezogener Spitze (Abb. 54). Analfeld deutlich, Ränder der Analnaht oft wallartig erhaben. Kremaster meist etwas länger als basal breit ; Ventralseite meist mit deutlichen Basalhöckern und einer Vertiefung. Kremasterendteil in Dorsalsicht etwas zugespitzt. Häkchen ziemlich stark, rotbraun (Abb. 53, 54). Raupe an Laubhölzern, vorzugsweise an Fagus sylvatica und Quercus spp., auch an Vaccinium spp., bevorzugt Eichen-Buchen-Wälder, dort oft häufig. Diskussion Die untersuchten Puppen der Tribus Cyclophorini mit der einzigen mitteleuropäischen Gattung Cyclophora bilden hinsichtlich ihres Baues eine geschlossene, wenig differenzierte Einheit. Auch die einzelnen, auf Grund ihrer bevorzugten Futterpflanzen gebildeten Gruppen — an Salicaceae (C. pendularia), Betulaceae (C. albipunctata), Fagaceae 85 (C. puppillaria, C. ruficiliaria, C. porata, C. quercimontaria, C. punc- taria, C. linearia), Aceraceae (C. albiocellaria, C. annulata), Asteraceae, Fabaceae u.a. (C. suppunctaria) unterscheiden sich puppenmorpho- logisch kaum voneinander. Die Cyclophora-Puppen weichen von den anderen der Familie Geo- metridae stark ab. Ihre Körperform ist der bei den Spannern wenig üblichen Verpuppungsweise an der Oberfäche der Blätter ihrer Nah- rungspflanzen angepaßt und weist dementsprechend weitgehende Adap- tationen auf. Sie entsprechen anderen, sich auf diese Weise verpup- penden (und sonst einander wenig verwandten) Schmetterlingsgruppen. Auffallend ist der lang vorgezogene Kremaster mit der Konzentrierung der hakenförmigen Haftborsten auf sein Caudalende. Eine sonst bei der Familie Geometridae wenig übliche, helle Färbung, nicht selten mit Zeichnungen, fast ohne Glanz, eckige Ausläufer am Vorderteil des Körpers, die Befestigung der Puppe an einem Blatt mittels Kremaster- häkchen und eines Gespinstgürtels erinnern an ähnliche Verhältnisse bei den Familien Papilionidae, Pieridae usw. Während die einzelnen Arten auch als Imagines relativ schwer zu bestimmen sind, ist ihre Unterscheidung anhand der Puppenmerkmale möglich und relativ ver- läßlich. Danksagung Der Verfasser dankt mehreren Fachkollegen, insbesondere Doz. Dr. Z. Lastüvka (Brünn), Ing. J. Skyva (Prag), dem Museum für Naturkunde in Berlin (Dr. W. Mey) und dem Naturhistorischen Museum in Wien (Dr. M. Lödl) für die Überlassung bzw. Ausleihe des Puppenmaterials zur Unter- suchung. Literatur FORSTER, W. & WOHLFAHRT, T. A., 1981. Die Schmetterlinge Mitteleuropas, 5. Spanner (Geometridae). 311 pp. LERAUT, P., 1980. Liste systématique et synonymique des Lépidoptères de France, Belgique et Corse. Supplement a Alexanor et Bull. Soc. ent. Fr., 334 pp. MCGUFFIN, W. C., 1968. Guide to the Geometridae of Canada (Lepidoptera). I. Subfamily Sterrhinae. Mem. ent. Soc. Can. 50 : 1-103. MOsHER, E., 1916. A classification of the Lepidoptera based on characters of the pupa. Bull. Ill. St. Lab. Nat. Hist. 12 : 1-158. PATOCKA, J., 1994. Die Puppen der Spanner Mitteleuropas (Lepidoptera, Geo- metridae). Charakteristik, Bestimmungstabelle der Gattungen. Tijdschr. Ent. 137 : 27-56. 86 Nota lepid. 17 (1/2) : 87-91 ; 30.X1.1994 ISSN 0342-7536 Une nouvelle espèce européenne du genre Pancalia Stephens (Lepidoptera : Cosmopterigidae, Antequerinae) Tadeusz RıEDL Witosa 5/5, 80-809 Gdansk, Pologne Summary A new European species of the genus Pancalia Stephens (Lepidoptera, Cosmo- pterigidae, Antequerinae) — Pancalia baldizzonella sp. n. is described from southern Italy. Adults and genitalia of both sexes are illustrated. The new species is most closely related to P Jatreillella Curtis and P nodosella (Bruand), from which it can be distinguished by the wing markings and genitalic cha- racters. Resume Description de Pancalia baldizzonella sp. n. d’Italie méridionale. Les adultes et les armures génitales des deux sexes de ce taxon sont figurées. Zusammenfassung Pancalia baldizzonella sp. n. wird aus Süditalien beschrieben. Die Imagines und die Genitalien beider Geschlechter werden abgebildet. Die neue Art steht P. latreillella Curtis und P nodosella (Bruand) am nächsten, von denen sie sich durch die Flügelzeichnung und Genitalmerkmale unterscheidet. Introduction Depuis la parution d’un article de GAEDIKE (1967) consacré aux espèces de Pancalia Stephens, 1829, ce genre et ses espèces ont été étudiés à plusieurs réprises (LERAUT, 1984 ; RrEDL, 1984 ; SINEV, 1985). Ces études ont permis d’expliquer la position systématique du genre en question et de confirmer, après deux révisions, l’existence de 9 espèces nomi- nales habitant uniquement la zone paléarctique, à savoir P gaedikei Sinev, 1985, P hexachrysa (Meyrick, 1935), P isshikii Matsumura, 1931, P. latreillella Curtis, 1830, P. leuwenhoekella (Linnaeus, 1761), P nodo- sella (Bruand, [1851]), 2 sichotella Christoph, 1882, P sinense Gaedike, 1967, P swetlanae Sinev, 1985. 87 Fig. 1-2. Pancalia baldizzonella sp. n. 1 — holotype (mâle) ; 2 — paratype (femelle) (phot. G. Baldizzone). Il m’a récemment été possible d’examiner les matériaux récoltés par le Dr. G. Baldizzone en Italie méridionale et de trouver parmi de nom- breux Cosmopterigidae une nouvelle espèce du genre Pancalia qui est donc le dixième taxon spécifique appartenant à ce genre. Pancalia baldizzonella sp. n. Ho.otype (Fig. 1) : Mâle, “Mt. Pollino, 1500 m, Lucania, Piano di Ruggio, 9. VII. 1991, G. Baldizzone leg.”, prép. gén. no. 1415/R. PARATYPE (Fig. 2) : Femelle, prép. gén. no. 1414/R. Portant la même étiquette. L’holotype et le paratype sont conservés dans la collection du Dr. G. Baldiz- zone à Asti. 88 Envergure de l’holotype : 15 mm ; du paratype : 14,5 mm. Tête, thorax, tegulae et palpes labiaux brun foncé, unicolores, brillants. Antennes brun foncé, unicolores chez le mâle et avec la partie subterminale du flagellum blanche chez la femelle (Fig. 1-2). Ailes antérieures foncées ; la base, l’apex, les bords antérieur et posté- rieur sont brun foncé, la partie centrale de l’aile est brun-jaunätre. Le dessin se compose d’une tache costale externe blanche et de tubercules d’écailles réfractives argentées. Le bord antérieur présente trois tubercules dont l’externe est réuni avec la tache costale blanche et l’interne, réuni avec le tubercule postérieur interne, forme avec celui-ci une écharpe transversale basale. Le second tubercule postérieur est rond, petit et situé plus près de la base de l’aile que le tubercule costal intermédiaire. Le troisième est allongé, enfin le quatrième, étroit, peu développé, est situé le long du bord postérieur de l’aile sous le tubercule costal externe et la tache costale blanche. Franges brun foncé. Ailes postérieures et leurs franges brun foncé. La forme des bandes du système de renforcement de la base de l’ab- domen (Fig. 3) est la même que chez P Jatreillella Curtis : la bande transversale est droite, mince, les bandes latérales du premier tergite sont également à peu près droites. ARMURE GÉNITALE MÂLE (Fig. 4-6) : Valves lobiformes, non arquées, leur bord ventral légèrement concave. Juxta développée en tant que deux bras de longeur inégale (Fig. 5), allongés et étroits. Aedoeagus arqué (Fig. 6) ; sa partie proximale possède une saillie ventrale très distincte. Terminaison de l’aedoeagus aiguë. ARMURE GÉNITALE FEMELLE (Fig. 9): Lamelle antévaginale à bords latéraux parallèles, son bord proximal convexe. Ostium bursae étroit, les deux signa peu distincts. Je nomme cette intéressante nouvelle espèce européenne en l’honneur de mon éminent collègue italien, distingué spécialiste des Lépidoptères Coleophoridae, le Dr. Giorgio Baldizzone (Asti). Commentaire P. baldizzonella sp. n. est la plus proche de deux espèces européennes, P latreillella Curtis et P nodosella (Bruand). Elle s’en distingue cepen- dant au premier coup d’ceil par le fond de la partie centrale des ailes antérieures qui est brun-jaunâtre tandis que chez les deux espèces men- tionnées ce fond est ferrugineux foncé. Le dessin des ailes antérieures ne présente aucun caractère particulier. Les différences les plus marquées concernent la forme des armures génitales. Chez le mâle, les deux bras 89 Figs 3-4. Pancalia baldizzonella sp. n. 3 — systeme de renforcement ; 4 — armure genitale mäle. ay q anh 8 Figs 5-8. Pancalia spp., armure génitale mâle : bras de la juxta (5, 7); aedoeagus (6, 8). 5, 6 — P baldizzonella sp. n. ; 7, 8 — P. nodosella (Bruand). de la juxta (Fig. 5) sont relativement étroits et allongés ; chez d’autres espèces de Pancalia, par example chez P nodosella (Bruand) (Fig. 7), ces sclérites sont moins allongés et nettement plus larges. En outre, parmı toutes les especes connues de Pancalia, P. baldizzonella sp. n. presente la saillie ventrale de l’aedoeagus la plus grande (Fig. 6). Cette saillie chez les autres, y compris P nodosella (Bruand) (Fig. 8), est plus petite et moins distincte. Chez la femelle, notre attention doit se porter sur la forme de la lamelle antévaginale, qui est bien caractéristique et diffère sensiblement de celles d’autres espèces de Pancalia. 90 Fig. 9. Pancalia baldizzonella sp. n., armure génitale femelle. En ce qui concerne la localité-type de P. baldizzonella sp. n., il s’agit du Massif Mont Pollino (Lucanie, Italie méridionale) ; Piano di Ruggio se trouve dans une grande prairie située dans le Massif en question. Bibliographie GAEDIKE, R., 1967. Zur systematischen Stellung einiger Gattungen der Helio- dinidae/Schreckensteiniidae sowie Revision der paläarktischen Arten der Gattung Pancalia Curtis, 1830. Beitr. Ent. 17 : 363-374. LERAUT, P., 1984. Quelques données sur le genre Pancalia Stephens en France (Lep. Cosmopterigidae). Ent. gall. 1 : 215-219. Rıepı, T., 1984. La position systématique du genre Pancalia Stephens (Lepi- doptera Cosmopterigidae). Polskie Pismo ent. 53 (1983) : 579-581. SINEV, S. Yu., 1985. A review of the genus Pancalia Stephens (Lepidoptera, Cosmopterigidae) in the fauna of the USSR. Ent. Obozr. 64 : 804-822. 9] Nota lepid. 17 (1/2) : 92 ; 30.X1.1994 ISSN 0342-7536 Aufruf zur Mitarbeit Die Schmetterlinge der Schweiz und ihre Lebensräume Verbreitungskarten (5 X 5 km Netz) der ‘Spinner und Schwärmer’ der Schweiz werden für eine Fortsetzung des inzwischen sehr bekannten und geschätzten Buches “Tagfalter und ihre Lebensräume’ vorbereitet. Um diese Verbreitungskarten zu ergänzen, sind alle Lepidopterologen einge- laden, uns ihre Schweizer Funddaten dieser Falter mitzuteilen. Bearbeitet werden alle Macrolepidopteren Familien, einschliesslich der Psychidae und Hesperiidae, aber ohne die Papilionoidea und (vorläufig noch) die Noctuidae und Geometridae. Gewünscht sind mindestens : Fundort, -datum und -stadien von einwandfrei be- stimmtem Material. Zusatzinformationen wenn möglich : Kanton, CH-Koor- dinaten, Sammler, ob Belegstück vorhanden, Sammlung, Raupenfutterpflanze. Ich werde die Daten sammeln und an die einzelnen Familien-Bearbeiter zur Kontrolle weiterleiten. Daten für die Sesiidae und Thyrididae bearbeite ich selbst. Daten von schwierig zu bestimmenden Arten werden eventuell nicht berücksichtigt, oder nur nach Genitaluntersuchung, wie z.B. bei den meisten Zygaeniden. Die kontrollierten Daten werden in die Databank des Schweize- rischen Zentrum für die kartographische Erfassung der Fauna (SZKF, CSCF) in Neuchätel aufgenommen. Lepidopterologen welche uns Verbreitungsdaten mitteilen werden im Buch erwähnt. Für die lepidopterologische Arbeitsgruppe der Schweiz, S. Whitebread, Maispracherstrasse 51, CH-4312 Magden, Schweiz 92 Nota lepid. 17 (1/2) : 93-99 ; 30.X1.1994 ISSN 0342-7536 Ing. Eberhard JACKH Ÿ (1902-1993) Am 22. August 1993 verstarb Eberhard Jäckh in seinen Haus in Hörmanshofen. Er saß in seinem Sessel, als er schnell und fast schmerzlos zur ewigen Ruhe kam. Er wurde am 5. Dezember 1902 in Kassel geboren, hat also den größten Teil dieses ausgehenden Jahrhunderts erlebt. Zur Zeit seiner Geburt regierte noch der Kaiser in Deutschland und der “Weltkrieg” war noch kein Begriff. Autos standen noch am Anfang der Entwicklung, Rundfunk und Fernsehen waren unbekannt, wie die meisten Geräte der heutigen Technik. Eberhard war das älteste von drei Geschwistern, Sohn von Dr. med. Alexander Jäckh, Chirurg und Frau Erna, geb. Hartdegen, einer talentierten Malerin. 93 Die Eltern starben beide früh. Schon während seiner Studienjahre im Internat entdeckte Eberhard jene Begeisterung für Insekten, die für sein ganzes Leben bestimmend wurde. Die Berufsausbildung begann er in Bremen, mit der Idee, als Schiffsingenieur die Welt kennenzulernen. Nach Abschluß der Ausbildung am Technikum erwies sich jedoch, daß seine immer größer werdende Passion für die Entomologie nicht mit wochenlangen Fahrten auf hoher See zu ver- einbaren waren. So wechselte er über zum Flugzeugbau. Viele Jahre arbeitete er bei der Firma Focke Wulf in Bremen, was ihm die direkte Teilnahme am ganz Europa zerstörenden Krieg ersparte. Auch seine Wohnung in Bremen wurde zerstört. 1937 heiratete er Adele Lakmann. 1940 wurde ein Sohn, auch Eberhard benannt, geboren, der jetzt mit seiner Familie in den USA lebt. Sein Interesse an der Entomologie vertiefte sich. Er konnte seine Kenntnisse erweitern, dank der Zusammenarbeit mit einigen bekannten Insektenforschern. Besonders Prof. E. M. Hering, Berlin, wurde sein Lehrmeister und Freund. Vor allem widmete er sich den Microlepidopteren. Äußerst aktiv beteiligte er sich am Wiederaufbau des Entomologischen Vereines Bremen. Das Bremer Überseemuseum engagierte ihn zunächst als freien Mitarbeiter. Das Hobby wurde zum Beruf, als ihm gleich nach dem Krieg 1945 die Leitung der Entomo- logischen Abteilung des Museums übertragen wurde, die er bis zur Pensio- nierung behielt (1967). Seine Frau Adele verstarb früh in Jahre 1960. Sieben Jahre später heiratete er Ingeborg Hoyer, die ihn als liebevolle Gefährtin bis zu seinem letzten Atem- zuge begleitete. 1974 zogen beide von Bremen zum Alpenrand und erwarben ein Haus in Hörmanshofen im Allgäu. Das Ehepaar ist viel gereist, vor allem nach Italien. Beide sprachen italienisch, liebten die Sonne, das Mittelmeer und die üppige farbenfrohe Natur, die oft Thema ihrer anderen Passion, der Fotografie, wurde. Ihr beliebtestes Reiseziel war immer Ligurien, die Riviera dei Fiori. Dort “entdeckten” sie Conna, eine winzige Ortschaft, eine Gruppe von steinernen Häusern, gelegen an den Ausläufern des Apennin, eingetaucht in Pinien, Ginster und Buschwald. Dort verbrachten sie viele Wochen und sammelten reiche menschliche und natur- wissenschaftliche Erfahrungen. Bedeutsam waren auch ihre Reisen in die USA, wobei die interessanten entomologischen Forschungen verbunden waren mit den Zusammentreffen mit der Familie des Sohnes. Eberhard Jäckh verbrachte seine letzten Lebensjahre zu Hause in Hörmans- hofen gestärkt durch die Liebe seiner Frau und seiner kleinen Enkelin Gian- nina, die viel in seiner Nähe war und viel von ihm gelernt hat. Er genoß ihre zärtliche Liebe, die wohl dazu beigetragen hat ihn bis ins 92. Lebensjahr auf- recht zu halten. 94 In der Entomologie gehört Eberhard Jäckh zu den bedeutenden deutsch- sprachigen Forschern, die die Grundlage für die moderne Lepidopterenfor- schung gebildet haben. Er hat seine während der Studienjahre gewonnenen technologischen Erkenntnisse für die entomologischen Forschungen benutzt und dabei neue Lösungen und hervorragende Ergebnisse erzielt. Unter ande- rem machte er technische Experimente für den Nachtfang. Durch eine Reihe von Versuchen kam es zur Erfindung einer leichten sehr wirksamen tragbaren Lampe, die er auch für Kollegen anfertigte. Als einer der ersten wandte er die Methode der Artenbestimmung durch die Untersuchung der Genitalien an. Mehr als 11 000 mikroskopische Präparate wurden mit den verschiedensten Färbetechniken angefertigt, um besonders scharfe Fotos davon zu erhalten. Denn auch die Mikrofotografie wurde mit Sorgfalt betrieben, bearbeitet im eigenen, sinnvoll mit einfachen Mitteln ein- gerichteten Labor, assistiert von seiner Frau, die professionelle Fotografin ist. So sind seine ganzen Forschungen durch Bilder belegt. Er hinterläßt ein ein- drucksvolles Archiv, das aus über 20 000, nach der systematischen Mikro- lepidopterenordnung klassifizierten Bildern besteht. Dabei sind auch ca. 100 Corodia-Filme, die Schmetterlinge in ihrer natürlichen Umwelt zeigen, sowie ca. 1 000 Schwarzweißfilme. Jede studierte Art ist dokumentiert im Entwick- lungsstadium, wenn möglich in der Natur, mit Genitalfotos beider Geschlechter, in Variationen, sowie manchmal mit Fotos der Larve oder der Spuren an der Futterpflanze. Die Sammlung Jäckh ist von sehr hohem wissenschaftlichen Wert. Sie umfaßt ca. 75 000 perfekt präparierte und etikettierte Exemplare, sorgfältig geordnet in großen, von Jäckh selbst gefertigten Holzkästen. Der Hauptkern besteht aus europäischem Material, sowie einer großen Anzahl in Nordamerika gesammelter Exemplare. Manche Exemplare wurden gezüchtet, um die Biologie der Microlepidopteren zu beobachten, vor allem der “Minierer”, wie Nepticulidae, Lyonetiidae, Gra- cillaridae und Coleophoridae. Dazu gibt es auch eine wichtige Sammlung von sorgfältig katalogisierten minierten Blättern. Die Bibliothek enthält ca. 150 Bände, teils seltene Exemplare, mit Widmungen und Notizen der Autoren sowie ca. 2000 Sonderdrucken von Entomologen der ganzen Welt. Die gesamte Sammlung wurde schon vor Jahren dem Smith- sonian Institution in Washington, USA, übereignet, das somit eine der wich- tigsten und weltbekanntesten Privatsammlungen bekam. Jäckhs wissenschaftliche Veröffentlichungen sind nicht so zahlreich, weil er die große Menge seiner Erkenntnisse immer sehr vorsichtig darstellte. Aber die, die er hinterlassen hat, sind von hohem wissenschaftlichen Wert meistens durch 95 Fotos dokumentiert. Somit ist Jäckh einer der ersten Autoren, der der guten Fotografie, zwecks besserer Objektivität, den Vorzug vor der Zeichnung gab. Im Laufe seiner Tätigkeit hat Jäckh verschiedene Gruppen und Familien der Lepidopteren bearbeitet (Gracillariidae, Lyonetiidae, Gelechiidae, Coleopho- ridae, Tortricidae, Pterophoridae, etc.), was seine umfangreichen Kenntnisse über Microlepidopteren beweist. Unter anderem hat er eine Revision der Gat- tung Batia Stephens und Pseudatemelia Rebel (Oecophoridae) vorgenommen. In den letzten Jahren hat er sich auf die Familie der Scythrididae spezialisiert und einige wichtige Beiträge erbracht mit der Revision verwandter Gruppen und der Beschreibung neuer Arten. Mit großer Bescheidenheit hat er nach 1978 mit der Veröffentlichung seiner wissenschaftlichen Arbeit aufgehört, um “altersbedingte Fehler” zu vermeiden. _ Fortgesetzt hat er seine Arbeiten im Archiv, welches viele neue Arten, sowie Arten der Gattung Scythris enthält, die auf eine Beschreibung warten. Eine der ihn begeisternden Gruppen war die Alucita, über die er die Basis für eine komplette Revision der palearktischen Fauna vorbereitet hat. Glücklicher- weise ist diese Arbeit nicht verloren. Wieder aufgenommen und ergänzt von Jäckhs jungem “Schüler” Axel Scholz, ist sie vor kurzem veröffentlicht worden. Ich habe Eberhard Jäckh 1973 in Conna kennengelernt. Damals entstand zwischen uns eine brüderliche Freundschaft, mit gegenseitigen Besuchen und gemeinsamen wissenschaftlichen Forschungen an der Riviera, auf den Hügeln des Monferrato, auf dem Monte Baldo, der Insel Krk, etc. Wir verbrachten miteinander viele unvergeßliche Stunden. Wir suchten Raupen und minierte Blätter, saßen neben der brennenden Lampe an einem xerothermischen Hang, diskutierten unaufhörlich über die Natur, planten Forschungen, Exkursionen und Publikationen. Aber oft sprachen wir auch über das Leben und die Ge- schichte dieses Jahrhunderts, das er selbst dahinfließen sah. Ich verdanke Eber- hard den größten Teil meiner Kenntnisse über Microlepidopteren. Besonders über alles, was ich über die Technik der Mikroskopie, Mikrofotografie, und der Sammlungs- und Lichtfangmethoden weiß. Alles wurde mir mit großer Geduld und Hingabe erklärt, voller Sympathie und Großherzigkeit. Aber viel verdanke ich ihm auch von einem menschlichen Gesichtspunkt aus. Es war eine echte Lebensschule mit unvergeßlicher Korrektheit, Ehrlichkeits-, Zähigkeits-, Freundlichkeits-, Bescheidenheitslehren und einer riesigen Liebe zur Natur. Giorgio BALDIZZONE 96 Liste der Publikationen 1977: 1933. 1934. 1936. 1940. 1942. 1951 195]: 1951. 1951. 1931: 1951: 1951: 1951. Ose 1951. 1952: 1953. 1953; 1953. 1953. 1953. 1953. Zur Entwicklung von Gracillaria azaleella Brants. Anz. Schädlingsk. 3 : 53-54. Über einige das nordwestdeutsche Faunengebiet kennzeichnende Micro- Lepidopteren. Mitt. ent. Ver. Bremen 21 : 6-10. Zur Überwinterung des Zitronenfalters, Gonepteryx rhamni L. Kosmos 31 : 31-32, 1 Abb. Bemerkungen über Trichoptilus paludum Z. Mitt. ent. Ges. Halle 14: 5-7. Die Insekten des Naturschutzparkes des Lüneburger Heide, III. Die Kleinschmetterlinge (Microlepidoptera). Abh. naturw. Ver. Bremen 31 : 786-806, 4 Abb. Die Microlepidopteren-Fauna des rechtsseitigen Mittelrheintales nebst Beschreibung von Borkhausenia magnatella spec. nov. (Lep. Gelechiidae). Z. wien. Ent Ver. 27 : 137-141, 174-200, 216-221, 230-241, 3 Abb., 1 Taf. Pristophora florella (Mann, 1862) (Pyralidae, Phycitinae) am Mittelrhein. Zepid' 1105. Bemerkenswerte Lepidopterenfunde auf Wangeroog in den Jahren 1947, 1949 u. 1950. Beschreibung der Raupe von Agdistis benneti Curt. Z. Lepid. 1 : 121-122. Praktische Genitalpräparate. Z. Lepid. 1 : 175-180. Die fruchtminierenden Arten der Gattung Nepticula Heyd. (Etainia Beirne) an der vier deutschen Ahornarten (Lep. Nept.). Z. wien. ent. Ges. 36 : 170-178, Taf. 14-16. Rhyacia subrosea Stph. Bombus 64/65 : 275, Nr.497. Agrotis cinerea Schff. Bombus 64/65 : 275, Nr. 498. Für das Gebiet der Niederweser 1950 neu aufgefundene Microlepidop- teren. Bombus 64/65 : 276, Nr. 500. Pseudophia lunaris Schff. (Lepid. Noct.). Bombus 66/67 : 284, Nr. 519. Zur Verbreitung von Pachetra fulminea F. Bombus 66/67 : 284-285, Nr. 520. Eriogaster lanestris L. Bombus 66/67 : 285, Nr. 521. 1. Zanclognatha tarsicrinalis Knoch., 2. Herminia derivalis Hb., 3. Hype- nodes costaestrigalis Stph. Bombus 74/75 : 316-317. Catoptria (Semasia) heringiana n.sp. (Eucosmidae). Z. Lepid. 3 : 43-45, 1 Abb.. Buchbesprechung : HERING, E. Martin: Biology of the Leaf Miners, 420 pp., 2 Tfln, 180 Textabb. Z. Lepid. 3 : 63-64. Aetetis (Elaphria) selini Bsd. (Lep. Noct.) bei Bremen. Bombus 76/77: 322, Nr. 617. Drei hervorragende Kleinschmetterlinge in Nordwestdeutschlands : I. Dio- ryctria faecella Z., 2 Myelois neophanes Durr., 3. Laspeyresia juniperana Mill. Bombus 76/77 : 323- 324, Nr. 621. Polyploca ridens F. wiederfunden. Bombus 78/79 : 333, Nr. 631b. Schutzvorrichtung zum Bau des Verpuppungskokons bei Arten der Gat- 97 1956. 1956. 1957. 19572 1958. 1958. 1958. 1959: 1959 19592 1959 1960. 1960. 1961. 1961. ISI, 1961. 1972: 1972: 1972: 1977. tung Bucculatrix Z. und Lyonetia Hb. Z. wien. ent. Ges. 40 : 118-121, Taf. 6-9, Nachtrag : 40 : 206. Ergänzungen zur Microlepidopteren-Fauna des Bremer Sammelgebietes. Bombus 92/93/94 : 393-395, Nr. 713. Tubuliferodes josephinae Toll in Nordwestdeustchland. Bombus 95/96 : 402, Nr. 719. Eine weitere in Deutschland an Aster linosyris (L.) Bernh. lebende Coleophora-Art. Dt. ent. Z. 4 : 54-60. Auffällig häufiges Auftreten des Totenkopfes, Acherontia atropos L. 1956 im Niederwesergebiet. Bombus 97/98 : 410, Nr. 726. Hipparchia statilinus Hufn. bei Pevesdorf an der Elbe, Kreis Lüchow. Bombus 2 : 10-11, Nr. 15. Polia glauca Hb. im Lande Oldenburg : Bombus 2 : 40, Nr. 47. Wanderfalterbeobachtungen. Bombus 2 : 40, Nr. 48. Beitrag zur Kenntnis der Oecophoridae, die Gattung Tubuliferola Strand, 1917. Dt. ent. Z. 6 : 174-184, 34 figs, Taf. I-IX. Die Gattungsgruppe Stomopteryx Hein. im nordwestdeutschen Tiefland. Bombus 2 : 64-66, Nr. 72. Bemerkenswerde Tortriciden-Funde in nordwestdeutschen Flachland. Bombus 2 : 70-72, Nr. 80. Neue Microlepidopteren der Italienischen Fauna. Boll. Soc. ent. ital. 89 : 85-88, Taf. 1. Neue und bemerkenswerte Funde im Bremer Sammelgebiet. Bombus 2: 86-87, Nr. 108. Eine neue Art aus der Gattung Pseudeucosma Obr., Pseudeucosma sub- tilana nova species. Boll. Zool. agr. Bachic., Serie II : 127-135. Nachtrag zur Microlepidopteren-Fauna des Naturschutz-Parkes der Lüneburg Heide. Bombus 2 : 100-103, Nr. 127. Moderner Lichtfang. Ent. Z. Frankf.a.M. 71 : 93-96. Pterophorus nephelodactylus Eversmann auch in den italienischen Alpen. Boll. Soc. ent. ital. 41 : 158-160, Taf. 1. Buchbeschprechung ; HANNEMANN, Hans Joachim, 1961 : Kleinschmet- terlinge oder Microlepidoptera, I. Die Wickler, 48. Teil, Die Tierwelt Deutschlands “Dahl”. Bonner zool. Beitr. 12 : 354. Caloptilia alchimiella Scop. und Caloptilia robustella spec. nov. Atti Accad. Sci. Torino 106 : 549-560, 10 Abb. Die Gattung Batia Stephens, 1834 s.str., (Lep. Oecophoridae). Redia 53 : 331-345, 5 Taf. Die Gattung Pseudatemelia Rebel, 1910 (Lep. Oecophoridae). Entomo- logica, Bari 8 : 133-140, 4 Abb. Bearbeitung der Gattung Scythris Hübner (Lep. Scythrididae) — 1. Die “grandipennis-Gruppe”. Dt. Ent. Z. 24 : 261-271, 11 Taf. 1977 (1978). Bearbeitung der Gattung Scythris Hübner (Lep. Scythrididae) - 98 2. Eine neue Scythris-Art aus Spanien : Scythris limbelloides n.sp. Z. ArbGem. öst. Ent. 29 : 81-84. 1978. Bearbeitung der Gattung Scythris Hübner (Lep. Scythrididae) — 3. Arten mit einer weissen Längsstrieme. Dr. Ent. Z. 25 : 71-89, 2 Taf. 1978. Bearbeitung der Gattung Scythris Hübner (Lep. Scythrididae) — 4. Un- beschriebene Arten aus Italien. Boll. Mus. civ. Stor. nat. Verona 5: 1-16, 5 Taf. BRINKMANN, A. & JACKH, E., 1926. Ein Jahr Schmetterlingsfang im Königs- moor - Verzeichnis der Kleinschmetterlinge. Jber. ent. Ver. Bremen 14: 11-13. JÄCKH, E. & BALDIZZONE, G., 1977. Sulla sinonimia di Coleophora oriolella Z. e Coleophora mongetella Chrét. Entomologica, Bari 13 : 31-36, 5 Abb. NAUMANN, J. & JACKH, E., 1930. 2. Beitrag zur Kenntnis der Schmetterlings- fauna des Königsmoores (Oyter Moores). Mitt. ent. Ver. Bremen 18: 8-16. SCHOLZ, A. & JACKH, E., 1994. Taxonomie und Verbreitung der westpaläark- tischen Alucita-Arten (Lepidoptera : Alucitidae [Orneodidae]). Alexanor 18 (4) [1993], suppl. : [3]-[63]. 99 Nota lepid. 17 (1/2) : 100-104 ; 30.X1.1994 ISSN 0342-7536 Book reviews — Buchbesprechungen — Analyses Coleophoridae dell’ Area Irano-Anatolica e regioni limitrofe (Lepi- doptera). Contribuzioni alla conoscenza dei Coleophoridae. LXXV. Giorgio BALDIZZONE. 423 pp., 698 photos b/n. Associazione Natura- listica Piemontese, Memorie, III. En vente exclusivement chez Apollo Books, Kirkeby Sand 19, DK-5771 Stenstrup, Danemark, DKK 390, frais d’expedition exclus. Présentation par le Prof. Emilio Balletto, President de la Societas Europaea Lepidopterologica. Vient de paraître le troisième volume des «Memorie» de l’«Associazione Natu- ralistica Piemontese», qui depuis quelques années publie aussi un intéressant Bulletin dédié surtout à l’histoire naturelle subalpine. Ce volume des «Memorie» est monographique et comprend seulement le travail de Giorgio Baldizzone relatif aux Coleophorides de la zone Irano-Anatolienne et des régions limi- trophes. Après l’importante révision des Coleophorides chinois, de l’Inde, de l'Australie et de l’Afrique tropicale, Baldizzone nous présente aujourd’hui en détail la faune du Proche- et Moyen-Orient. Ce travail représente sa 75° contri- bution à la connaissance de cette vaste et complexe famille des Microlépi- doptères Gelechioidea, dont il est depuis longtemps un spécialiste reconnu. Comme le rappelle dans sa préface Emilio Balletto, Président de la «Societas Europaea Lepidopterologica», la région irano-anatolienne est très peu connue, non seulement en ce qui concerne les microlépidoptères ; le travail de Baldiz- zone doit donc être particulièrement apprecié pour l’évidente et vaste contri- bution qu'il apporte au progrès des connaissances relatives à la faune des Lépidoptères de cette importante zone de la Région Paléarctique. Dans cette monographie, dédiée à la mémoire de Eberhard Jäckh récemment disparu, sont traités 310 taxa appartenant aux Coleophorides, dont 84 sont ici décrits pour la première fois sur la base d’un large matériel provenant en majorité des campagnes de recherche de microlépidoptéristes connus, parmi lesquels les plus importants sont sans doute Fritz Kasy, Wolfgang Glaser, Hans Georg Amsel et Eva Vartian. Les nouveaux taxa appartiennent tous au genre Coleo- phora, sauf une espèce du genre /schnophanes. L’auteur a en outre fixé 30 lectotypes et le néotype de C. gypsophilae Christoph, 1862 (= C. vicinella Zeller, 1849). Enfin sept nouvelles synonymies sont reconnues ; ces dernieres, ajoutées aux nombreuses déjà signalées par Baldizzone lui-même ou par d’autres auteurs, permettent de compléter la révision nomenclatoriale des taxa paléarctiques. Comme d’habitude, la très abondante iconographie qui accompagne le travail est très claire et parfaite sur le plan technique, soit en ce qui concerne l’habitus des adultes — il est dommage que les hauts frais d’impression ne permettent 100 pas de les reproduire en couleur comme on l’a fait sur la couverture avec la splendide planche du jeune entomologiste Fabrizio Pensati! — soit en ce qui concerne les genitalia des deux sexes. Ce travail, d’apres une indication tres claire de son auteur, veut seulement mettre a jour la connaissance des Coleophorides de la region irano-anatolienne au sens large, en renvoyant à des contributions ultérieures la révision moderne de la systématique du groupe et toute autre considération zoogéographique. Dans sa contribution en effet, Baldizzone maintient pour l'instant la subdivision classique donnée par Toll, qui divise les Coleophorides en plusieurs groupes pas nécessairement naturels, reconnus sur la base des plans de structure des genitalia des deux sexes. En lisant avec attention la monographie, on peut de toute façon dejà obtenir plusieurs indications zoogéographiques tout à fait intéressantes qui nous permettent de confirmer ou de mieux comprendre le peuplement des régions anatoliennes, arabes et du bassin touranique. De toute façon, ce volume, à tirage limité à 400 copies à cause des frais d’im- pression élevés auxquels l’auteur a dû faire face lui-même pour la plupart, est non seulement à conseiller vivement à tous ceux qui étudient les Lépidoptères, mais aussi à tous ceux qui désirent augmenter leurs connaissances en matière d’entomologie et de zoogéographie. Pietro PASSERIN D’ENTREVES, Dipartimento di Biologia Animale, Università di Torino, Via Accademia Albertina 17, I-10123 Torino (Italie) Index of economically important Lepidoptera. Bin-Cheng Zhang, 1994. 599 Seiten, gebunden. CAB International, Wallingford, Oxon, UK. ISBN 0-85198-903-9. £ 50 ohne, £ 70 mit Diskette. Das Commonwealth Agricultural Bureau International (CABI), wurde 1928 gegründet. Eine der wesentlichen Aufgaben des CAB’s ist die Herausgabe von Referateorganen über die Weltliteratur im Agrarbereich im weitesten Sinne. Unter anderem erscheint das ”Review of Agricultural Entomology” (RAE) (früher : "Review of applied Entomology”) 1994 bereits im Volume 82 und zwar gegründet 1913, also schon vor der Etablierung des CABI. Da jähr- lich ca. 10.000 Arbeiten referiert werden, hat sich im Laufe der Jahre natür- lich ein enormer Datenbestand angesammelt. Verständlich, daß dieser riesige Fundus nach zusätzlichen ‘Ertragsmöglichkeiten’ durchforscht wird. Begrüßens- wert für die ‘User’ der Reviews, wenn ihnen das geballte Wissen, praxisgerecht serviert wird. Praxisgerecht bedeutet in diesem Falle, daß Herr Zhang aus dem RAE-Datenbestand, alle Lepidopteren herausselektiert hat, die seit 1913 in der Literatur eine Erwähnung als “Pflanzenschädlinge’ fanden. Parasiten des Menschen und der Tiere sollen separat behandelt werde. Eine derartige Kompilation bedeutet für jeden angewandten Entomologen eine sehr große Hilfe. Dafür sei Herrn Zhang und dem CABI Dank. Schließlich hat nicht 101 jeder die technischen Einrichtungen und die finanziellen: Möglichkeiten, die leider sehr teure elektronische Datenbank des CABI zu benutzen. Allerdings möchte ich einige ‘Schwachstellen’ auflisten, die ich gerne bei einer Neuauflage beseitigt sehen würde. Daß bei der großen Datenfülle auch mal etwas verloren geht, dürfte verständ- lich sein. Einige wenige Arten habe ich vermißt, so zum Beispiel Duponchelia fovealis (in RAE 1991 ; no. 743, 5542 und 6335). Die Angaben zur geogra- phischen Verbreitung sind immer mit Vorsicht zu handhaben, da viele Länder nicht sehr meldefreudig sind. Allerdings hat Herr Zhang auch nicht alle An- gaben aus den Reviews übernommen, so zum Beispiel bei Opogona sacchari. Ein besonderes Problem sind die Wirtspflanzenangaben. Wenn bei Epichoristo- des acerbella nur drei Wirtspflanzen genannt sind, so ist mir dies unverständ- lich. Die Larve dieses Tortriciden ist zweifellos polyphag und man könnte ohne weiteres 20 bis 30 Wirtspflanzen nennen bzw. der Literatur entnehmen. Beim European corn borer Ostrinia nubilalis hätte er etwa 300 Pflanzenarten aufführen können. Aus Platzgründen (oder auch aus Zeitgründen ?) wurde wohl selektiert. Wäre es, in diesem wie in ähnlichen Fällen, nicht besser gewesen, wirklich nur die Hauptwirtspflanze(n) anzugeben und dann auf die Polyphagie zu verweisen ? Personen, die diese Lepidopteren nicht näher kennen, könnten sonst solche unvollständige Auflistungen für bare Münze nehmen. Die Ver- mischung der ‘common names’ mit den wissenschaftlichen Namen in den Host Records dient auch nicht gerade der besseren Übersicht. Etwas ganz wichtiges für den angewandten Entomologen fehlt jedoch total : ein Wirtspflanzenindex. Es wäre wirklich sehr hilfreich, einen Überblick zu bekommen, welche Lepidopteren an welchen Pflanzen vorkommen können. Da das Buch weitgehend mit Hilfe der EDV geschrieben wurde, dürfte die Erstellung einer derartigen Liste kein Problem darstellen. Der ‘Index of specific and infraspecific epithets’ am Schluß des Buches ist zwar sehr umfangreich, wäre aber mit Seitenangaben besser brauchbar. Vermutlich aus Platzgründen wurde bei den Hinweisen zu den RAE References nur die Bandzahl genannt. Die zusätzliche Angabe der Abstract-Nummer wäre sehr hilfreich und würde das Buch meines Erachtens kaum umfangreicher machen. Insgesamt gesehen, stellt das Buch jedoch auf Grund der ungeheuren Daten- fülle eine äußerst wertvolle Arbeitsgrundlage für alle Biologen, Ökologen und natürlich speziell für Entomologen dar und ist für Entomologen im Bereich Pflanzenschutz eigentlich unverzichtbar. Es wäre sehr zu wünschen, wenn wei- tere derartige Indexe auch für andere Insektenordnungen erscheinen würden. Wolfgang BILLEN Guide pour l’identification des especes francaises du genre Zygaena. Louis Faillie, 1994. 52 pages, 56 figures au trait, 3 planches couleur hors texte. Format 14,8 x 21 cm. Edition J-M Desse. Prix 90 FFr plus port. En vente chez l’auteur ; 8 rue Polonaise F-72200 La Flèche. Ce guide publié à compte d’auteur le 19-I11-1994 est destiné à permettre l’iden- 102 tification rapide et sûre de toutes les espèces du genre Zygaena actuellement connues en France. L. Faillie aurait donc pu se contenter de la partie centrale du guide, nous en aurions disposé déja depuis plusieurs années. Ce perfec- tionniste ne pouvait se contenter de cet à peu près. Il a donc cherché à mettre en perspective l’objet du guide. Celui-ci débute par quelques considérations sur les notions biologiques qui conduisent à la systématique bien documentée permettant d'approcher la finesse de la pensée des grands spécialistes contem- porains qui guident l'étude de ce groupe d’espéces. Elles illustrent aussi le bon sens qui privilégie les solutions simples aux consi- dérations trop théoriques ou partielles qui ont souvent obscurci l’horizon des études se rapportant à ce groupe. Suit la liste des espèces du genre Zygaena, tirée de la solution proposée par C. M. Naumann & W. G. Tremewan en 1984, suivie de rappels au sujet de la position de Zygaenidae fossiles vis-à-vis de ce genre et des genres voisins. Les 26 espèces françaises qui sont l’objet du guide, sont ensuite énumérées dans le même ordre. Un intéressant tableau de comparaison avec les pays européens limitrophes ou voisins, permet de visualiser la richesse et la variété des espèces de chacun d’eux. Dans les 18 pages suivantes le lecteur trouvera les traits essentiels d’une diagnose de chacune des espèces françaises qui décrit parfaitement les caractères distinc- tifs externes, complétés par un dessin au trait grandeur nature qui met l’accent sur les parties de l’ornementation alaire ou du corps qu’il faut prendre en compte pour déterminer chaque espèce. Ces caractères sont souvent mis en évidence par une flèche qui attire judicieusement l’attention du lecteur. Pour les espèces présentant une variation importante, L. Faillie n’a pas hésité à réaliser plusieurs dessins qui illustrent les principales sous-espèces ou morphes présentes en France. Il y a ajouté souvent des informations très pertinentes et très à Jour sur la répartition, la génétique, la nomenclature ou la taxinomie. Cette partie centrale du guide permet vraiment de réaliser les déterminations correctes dans la plupart des cas. Les pièges et difficultés essentielles sont d’ail- leurs révélés afin de permettre à tous de travailler avec un maximum de sécu- rité. Le guide est complété par un tryptique amovible regroupant sur trois planches photographiques en couleur, les 72 specimens représentant au moins une fois, grandeur nature, chacune des 26 epèces traitées. Cette disposition permet, lors de la détermination, de comparer le dessin au trait avec la figure en couleur et renforce ainsi la qualité de la détermination. Il convient de saluer le soin apporté au choix des exemplaires reproduits, car ils sont tous en parfait état de fraîcheur et correctement préparés, ce qui est rare dans les publications concernant ce groupe. Pour augmenter l'efficacité des déterminations, l’auteur a ensuite judicieusement regroupé les espèces suivant certains critères ornementaux (anneau abdominal et collerette) puis regroupé à la fin du guide, sur 4 pages, tous les dessins au trait représentant les imagos figurant dans la partie centrale. Quelques considé- 103 rations sur la structure interne des genitalia mäle des Zygaena sont ensuite les bienvenues. Elles permettent de bien comprendre les différences entre Z. minos et Z. purpuralis dont la dissection est obligatoire. Une bibliographie adaptée au guide permettra enfin au lecteur d’approfondir les connaissances qu'il aura acquises à la lecture du guide. En résumé, un petit livre intelligent, fruit de décennies d’études dédiées aux zygènes et de rencontres avec les plus grands spécialistes de la question. Marque aussi d’une volonté de se mettre au service des autres pour les aider à com- prendre. Remercions donc chaleureusement Louis Faillie d’avoir consacré beau- coup de son temps à la réalisation de ce petit fascicule qu’il a tenu à financer de ses propres deniers afin de ne dépendre d’aucune obédience. Faisons bon accueil à ce guide qui, je n’en doute pas, deviendra très vite le compagnon indispensable de vos recherches et de vos promenades à la ren- contre des zygenes. Eric DROUET Copyright © Societas Europaea Lepidopterologica, 1994 Printed by Imprimerie Universa Sprl, 24 Hoenderstraat. B-9230 Wetteren, Belgium All rights reserved. No part of this Journal may be reproduced or transmitted in any form or by any means, electronic or mechanical including photocopying, recording or any other information storage and retrieval system, without permission in writing from the Publisher. Authors are responsible for the contents of their articles. 104 toph Häuser, Dr. Pete Tr — patenka, Dr Paul | Waring ‘prs aa oq it a + #48 AT HEN lepidopterologica Vol. 17 No.3-4 1994 ISSN 0342-7536 | ei RE. ae Si ar ic che sind hei Redaktor erhältlich. | si sen francais sont dcr auprès de l'éditeur ay articl au eo RN RUE (max. 2 typed pages) on any we we LR ic lepidopterology will be considered for publication. Full articles will be revie! : reviewed by two referees. ‘Publication languages are English, French and German. F a. ra t should be made t to Sal out area corrections before SE the "AU manuseri a ae be an ae ER dose spacing and wide margins, and | | submitted à a ae In addition. to the original figures, three ee should be fac ee also accepted) shout also es aie Que possible. Full etc must Rs à a summar sa not more than 200 words. A translation of the pee: in at least x ae or. C ies nt issues of se pan: ‘hou: be deckel {ee style a format. is ee words | at the right-hand. ag or Be surnames in capitals, Er = À should be styled as follows : ER VS Br J., 1985. New species of dont Hübner re . Zeuglopera: RAR N ic opterigidae) En Greece and Cyprus. Nota lepid. 8 : 336-340. = 3, Pale eee ee te in in NI = ER | NL: 1g re : esas sunk er d ee ue 2 and all ype deren must Nota lepidopterologica ISSN N249 7694 Schloßgasse 8 + 99448 Kranichfeld + Telefon und Fax: 036450/23 10 + Funktelefon: 0161/3 30 80 43 Im Dienste der Ökologie - Für die Erhaltung der Umwelt Basel. 30.1V.1995 Vol.17 No.3/4 Service: Versand: - Gutachten - Umweltanalytik - Kartierungen - Klimamessung - Landschaftsplanung und -gestaltung - optische Geräte - Pflegearbeiten - Laborgeräte - Auftragsforschung - Freilandgeräte für ökologische Untersuchungen - Beratung - Präparationshilfsmittel UVP / UVS - Literatur (Ökologie, Reise, Entomologie) „Eugen. Deu unsere kostenlosen Kataloge anl Ta FIN eee i ee a ee Nota lepidopterologica Vol.17 No.3/4 Basel, 30.1V.1995 ISSN 0342-7536 Lebensmittel aus ökologischem Anbau u. a.m. Oko-Shop, AlexanderstraBe 33, D-99448 KRANICHFELD Spezialangebot: Entomologiebedarf Netze (Kescher, Streifkescher, Kafersiebe [versch. Größen)) Fallen (Bodenfallen, Lichtfallen, Photoeklektoren u.a.) Sammelschachteln (Raupensammelschachteln, Transportbecher u. a.) Pinzetten (Uhrfederstahl, med. Pinzetten, spitz, rund, schräg) Insektennadeln (versch. Hersteller, versch. Preislagen) Spannbretter (verstellbar, feststehend, Holz, Schaumstoff) Insektenkästen (Doppelglas, Normalausführung, versch. Holzarten) Dublettenkästen (Plaste, Pappe, Holz, mit und ohne Glas) entomologische und ökologische Literatur (antiquarisch und neu, spezielle Literatur auf Anfrage) Etikettenservice (Sie geben uns Text und Anzahl, wir schreiben und drucken für Sie in kürzester Zeit mit Computer) Lichtfangausrüstungen (mit Zigarettenanzünderanschluß für Auto, mit Batterieanschluß, Mischlichtlampen, UV-Röhren u. a.) HONDA-Stromerzeuger (für den netzunabhängigen Lichtfang) Kataloge kostenlos! - Persönliche Beratung auf Wunschl Ecology - the way to understand the nature Nota lepidopterologica Vol.17 No.3/4 Basel, 30.1V.1995 ISSN 0342-7536 Editor : Steven E. Whitebread, Maispracherstrasse 51, CH-4312 Magden, Switzerland. FAX : + 41-61-841.22.38. Assistant Editors : Emmanuel de Bros (Binningen, CH) PD Dr. Andreas Erhardt (Binningen, CH) PD Dr. Hansjürg Geiger (Berne, CH) Dr. Alexander Pelzer (Wennigsen, D) Contents — Inhalt — Sommaire EGROMAIR ate tard Sone ee 5 RT PAPER CO ERA NC 106 Di, Ja M Sn A a 120 FIBIGER, M. : Anumeta arax sp. n. from Turkish Armenia (Noctuidae, (COCA S) dues Res ee EEE... 107 KAıLA, L. & BIESENBAUM, W.: Redescription of Elachista differens arent 1978: (Blachisudae) 2... DA ESEL 113 LARSEN, T. B. : Aricia crassipuncta bassoni Larsen, 1974 from Lebanon Raised sto speciesurank (leyeaenidae) nn. nn NE 121 MUNGUIRA, M. L., MARTIN, J. & PEREZ-VALIENTE, M. : Karyology and distribution as tools in the taxonomy of Iberian Agrodiaetus butterflies By calcmiicl ac) RER ER I ERENTO LERNEN INN FROHES 125 PELZER, A. : Illustrierter Bestimmungsschlüssel für die Präimaginalsta- dien der Schwärmer Europas und Nordafrikas (Sphingidae). Teil II : sia latin ee We Ba ee 141 PORTER, A., SCHNEIDER, R. & PRICE, B. : Wing pattern and allozyme relationships in the Coenonympha arcania group, emphasising the C. gardetta-darwiniana contact area at Bellwald, Switzerland (Saty- EN ee N TEENS 155 SMITH, D. A. S. & Owen, D. F. : Inter-island variation in the butterfly Hipparchia (Pseudotergumia) wyssii (Christ, 1889) in the Canary ISlamelsı(Salyanac) NE En Re ski 175 TENNENT, J. : Danaus chrysippus Linnaeus, 1758 ; a review of records and present status in the Maghreb countries of Morocco, Algeria and ns (Dan aiMae) ears bern been ze 201 TENNENT, J. : Further notes on Berberia de Lesse species in North Africa and confirmation that B. abdelkader Pierret, 1837 and B. lambessanus Staudinger, 1901 are significantly distinct (Satyridae) ........................ 217 105 Short communication — Kurze Mitteilung — En bref DARDENNE, B. & DROUET, E. : Pempeliella ornatella (D. & S.) et Acti- notia hyperici (D. & S.), espèces nouvelles pour la Seine Maritime (Normandie, France) (Pyralidae, Pterophoridae, Noctuidae) ............. 220 Book reviews — Buchbesprechungen — Analyses ................................ 124 Vol. 17 — 1994 Dates of publication — Publikationsdaten — Dates de publication 221 Contents — Inhalt Sommaire Wen er ee 221 New taxa described in Vol. 17 — Neue Taxa in Band 17 beschrieben — Nouveaux taxa décris dans le VOTE RS 224 Editorial The Editor would like to apologise for the late publication of this issue. This has been due to illness in the Editor’s family, on top of increased professional and private commitments during February and March. For the same reason, the SEL “1991-1992 Index of Publications on European Lepidoptera” was also delayed. The situation is again under control and all outstanding correspondence will be dealt with as soon as possible. I am pleased to announce that Dr. Roger Dennis, Wilmslow, GB has joined the Editorial Committee of this Journal. Dr. Dennis, who is well known for his articles and books on ecological lepidopterology, has already been a great help in the processing of a number of manuscripts in English. Such manuscripts from the United Kingdom may now be submitted directly to Dr. Dennis. In addition, M. Yves Gonseth, Neu- chätel, Switzerland has agreed to act as Assistant Editor from June 1995. At the same time, Dr. Hansjürg Geiger will leave us. Dr. Geiger has been an invaluable member of the Editorial Committee since 1986, but wishes to step down due to increased professional commitments. I would like to sincerely thank Dr. Geiger, not only for his editorial contribution, but also for his advice and moral support. Finally, I would like to mention that this Journal ıs now abstracted in BIOSIS and “Current Advances in Ecological and Environmental Sciences”, a CABS review journal. | Steven WHITEBREAD 106 Nota lepid. 17 (3/4) : 107-112 ; 30.1V.1995 ISSN 0342-7536 Anumeta arax Sp. n. from Turkish Armenia (Lepidoptera, Noctuidae, Catocalinae) Michael FIBIGER Molbechs alle 49, 4180 Sore, Denmark Summary During a lepidopterological visit to Turkish Armenia in 1989 a new species of Anumeta Walker, 1858. was found. This ıs described here as Anumeta arax sp. n. In September 1993 a further visit to this area was made. A few notes on collecting in this politically very tense, but entomologically very interesting area are given. Resume Découverte d’une nouvelle espèce d’Anumeta Walker, 1858 au cours d’une expédition lépidoptérologique en Arménie turque en 1989. Description de celle- ci sous le nom d’Anumeta arax sp. n. L’auteur a revisité ces lieux en septembre 1993 et donne quelques indications sur la chasse aux papillons dans cette région, où la situation politique est très tendue, mais qui est très intéressante pour les entomologistes. Introduction Turkish Armenia is one of the most beautiful and lepidopterologically important areas in the western Palaearctic region. However, since 1988 it has not been fully safe to travel, camp or catch insects because of the unstable political situation in the area. The Arax valley is per- haps the most interesting locality of the region. It runs east-west at a low level (at Aralik : 825 m) just north of, and below, the imposing Mt. Ararat. Unfortunately, the Turkish, Armenian and Kurdish peoples all claim sovereignty over the area, which is at present situated ın Turkey. By the end of 1993, the tense situation had become distinctly dangerous for touring lepidopterists who wanted to collect in the area. Together with the Danish lepidopterist, Fritz Schepler, I visited the area at the beginning of September 1993. We collected at night only with “black” lamps (pure ultraviolet 125 watt bulbs and 20 watt tubes), 107 and were as usual met by the very friendly local people, but also by masses of heavily armed, friendly gendarmes and soldiers. Everybody was scared of each other, often with good reason, and this anxiety creates aggression. When I worked in the area in 1989, Nils Esser (a Danish coleopterist) and I were allowed by the gendarmes to camp and catch moths along the asphalt road from Igdir to Aralik, but we had to set up the lamps within 5 metres from the road, and we were not allowed to stay. In 1993 it was impossible — even armed with psychological arguments rehearsed in my mind — to obtain the same permission. We were forced to camp in a storm (only catching with an 8 watt superactinic tube two metres from the car) close to the military camp in the middle of the village of Hasanhan, approx. 30 km west of Aralik. The unexpected, but good records from both visits to the area will be published in a later paper. Here I will only mention the catch in 1989 of a dozen specimens of Drasteria picta (Christoph, 1877), two specimens of Gonospileia munita (Hübner, [1813]), both species new to Turkey, and in 1993 the record of two specimens of Cardiestra vassilinini (A. Bang- Haas, 1927), also new to Turkey and previously only known from the type specimen. In 1989, I also captured one specimen of the genus Anumeta Walker, 1858 which was unknown to me and to everyone to whom I showed it. Not until 1992 in St. Petersburg, visited in connection with the successful SEL Congress in Helsinki, did I find four more specimens of “my” Anumeta in the Zoology Institute, Russian Academy of Science. They were placed under the name caucasica Rjabov, in litt. Dr. Irına L. Sukhareva who kindly helped me during my visit also translated the Cyrillic writing on the labels. The four specimens were all recorded from the Armenian side of the Arax valley (former Armenian S.S.R.) within 10 km from the locality in Turkey where I found it: 10 km north west of Aralik. As the name caucasica to my mind should represent a mountain species further to the north, I have decided to describe it here under the name Anumeta arax sp. n. Anumeta arax Sp. n. HoıoTypE: & (Fig. 1) Turkey, prov. Kars, 10 km NW Aralik, 825 m, 22.vu.1989, genit. prep. 1798, leg. & coll. M. Fibiger. PARATYPES : | @ [Turkey], Aralych [Aralik]. 1 2 [Armenia], Mtschjan Artashantski region (Arax valley), Okt., leg. Arutjunjan. 1 © [Armenia], Dzhuga by Dzhulfa [Arax valley], 4.vu.1932, leg. Rjabov. 1 9 (Fig. 2, ALLo- 108 Figs 1-2. Anumeta arax sp. n. 1 — Allotype, female ; 2 — Holotype, male. TYPE) Armenia, Burastan, Acerjan [Kamarlu region, Arax valley], at light, 26.v1.1948, leg. G. Azaryan, genit. prep. 1799 M. Fibiger. DESCRIPTION (Figs 1,2): Male and female similar in size and wing pattern ; female colouration in fore and hindwing slightly darker. Wingspan: 33-34 mm. All segments of labial palpi porrect ; first segment light grey, as long as second and third together, latter two black. Antenna of male ciliate, of female filiform. Ground colour of head, thorax and forewing blackish grey. Basal and median area of forewing slightly darker. Black costal spots on and between crosslines. Lines weakly defined except for black terminal line on both wings, which is sinuate with white spots between veins on terminal side. Fore- wing fringes dark grey, whitish on hindwing. Stigmata absent, a black spot near reniform. Hindwing with light greyish median band, terminal area darker than basal. Whitish blotches on termen of hindwing (typical for Anumeta and Drasteria) weakly defined in male, hardly visible in female. Underside light greyish powdered with black scales. Terminal area blackish. Median stigma present on hindwing. MALE GENITALIA (Fig. 3): Valva, juxta, and vinculum rather simple, but uncus prominent, heavily sclerotised, with apical hook. Aedeagus straight, no cornutus, but a light sclerotised band. Vesica short, rounded with small diverticula. Note : The preparation of the vesica was not perfect and the paratype male had already been dissected, so it was not possible to evert. FEMALE GENITALIA (Fig. 4): Ovipositor hairy. Apophyses equally broad their whole length. Eighth segment hairy distally. Ductus bursa weakly sclerotised. Corpus bursa unisaccate, cylindrical, two and a half times as long as broad. 109 Fig. 3. Male genitalia of Anumeta arax sp. n. : holotype. Remarks Because of an urgent need for a revision of the genus Anumeta, which also forms the tribe Anumetini, the exact number of species cannot be stated. A little more than a score of Anumeta species are known, all with a desert and semi-desert, central and southwestern Palaearctic distribution. Only three species are recorded from Europe (FiBIGER & HACKER, 1991), in south-east European Russia. The genus was pre- viously unknown from Turkey. Systematically, Anumeta arax sp. n. is most closest related to Anu- meta fricta (Christoph, 1893), A. fractistrigata (Alphéraky, 1882) and A. cestina (Staudinger, 1884). The name of this new species is derived from the type locality : The Arax valley. 110 Fig. 4. Female genitalia of Anumeta arax sp. n. : allotype. 111 Acknowledgements Dr. Irina L. Sukhareva is sincerely thanked for her assistance in the Zoological Institute, St. Petersburg. My wife Mariann and Barry Goater are as usual thanked for checking my English. Literature FIBIGER, M. & HACKER, H., 1991. Systematic List of the Noctuidae of Europe. Esperiana 2 : 1-109. 12 Nota lepid. 17 (3/4) : 113-119 ; 30.1V.1995 ISSN 0342-7536 Redescription of Elachista differens Parenti, 1978 (Lepidoptera, Elachistidae) Lauri KAıLA* & Willy BIESENBAUM** * Finnish Museum of Natural History, Zoological Museum, P.O.Box 17, FIN- 00014 University of Helsinki, Finland ** Feldstrasse 69, D-42555 Velbert-Langenberg, Germany Summary Elachista differens Parenti is redescribed and compared with the other re- cognized West-Palaearctic species of the Elachista gleichenella group on the basis of a series collected in Rhineland, Germany. The female is described for the first time. The species externally closely resembles E. gleichenella (Fabricius) and E. lambeseella Nielsen & Traugott-Olsen. The structure of male genitalia indicates E. lambeseella to be the closest relative of the species. E. differens is also reported from Corfu, Greece, although the male genitalia show some minor differences. Zusammenfassung Elachista differens Parenti, 1978, wird erneut beschrieben und ausserdem verglichen mit den anderen bekannten westpalearktischen Arten der Elachista gleichenella-Gruppe auf der Grundlage einiger im Rheinland, Deutschland, gefundenen Faltern dieser Art. Das bisher unbekannte Weibchen wird zum ersten Mal beschrieben. Die Art gleicht äusserlich sehr der E. gleichenella (Fabricius) und E. lambeseella Nielsen & Traugott-Olsen. Der Bau der männ- lichen Genitalien zeigt, dass E. lambeseella am nächsten verwandt ist mit dieser Art. E. differens wird ebenfalls nachgewiesen von Korfu, Griechenland, ob- gleich die männlichen Genitalien der dort gefundenen Tiere einige kleine Abweichungen zeigen. Resume Redescription d’Elachista differens Parenti et comparaison de celle-ci avec d’autres espèces ouest-paléarctiques connues du groupe d’Elachista gleichenella en se basant sur quelques exemplaires trouvés en Rhénanie (Allemagne). Description de la femelle, jusqu’à présent encore inconnue. Extérieurement, E. differens ressemble beaucoup à E. gleichenella (Fabricius) et a E. lambeseella Nielsen & Traugott-Olsen. La structure des genitalia mâles montre que cette espèce est la plus proche de E. lambeseella. Elle a aussi été signalée de Corfou, 113 Grèce ; les genitalia mâles des exemplaires de ces deux régions présentent toutefois quelques petites differences. Introduction In the Palaearctic region five Elachista species belonging to the gleiche- nella group sensu TRAUGOTT-OLSEN and NIELSEN (1977) have so far been described : E. gleichenella (Fabricius, 1781), E. regificella Sircom, 1849, E. differens Parenti, 1978, E. lambeseella Nielsen & Traugott- Olsen, 1987 and the eastern Palaearctic E. megagnathos Sruoga, 1990. The description of E. differens was based on three male specimens collected from France (type locality) and Italy. The original description (PARENTI, 1978) contained only a short description of the external characters and a schematic figure of the male genitalia; NIELSEN & TRAUGOTT-OLSEN (1987) therefore ignored this species when describing E. lambeseella (type locality Algeria, Lambése), even though these species are closely related. Additional material has shown that E. diffe- rens 1s rather widespread in Europe. In this paper we give a redescription of this species, including a description of the previously unknown female. Elachista differens Parenti, 1978 MATERIAL STUDIED: D NW-Nordeifel Marmagen, Gillesbachtal, 4.7.1992 19, 8.7.1992 1.3 1 ©, 3.7.1993 4.44 6 99, 4.7.1993 5 99 Biesenbaum leg. Coll. Zoological Museum, University of Helsinki (2 & 2 Q), and coll. Biesenbaum. Greece, Corfu, 6.-14.6.1978, Vesa Varis leg., 8 44 in Coll. Zoological Museum, University of Helsinki. DraGnosis : Externally Elachista differens resembles E. gleichenella (Fabricius), but it is on average larger. Costal and tornal spots are usually separated in E. differens, whereas they almost always are joined in E. gleichenella, forming a fascia (Figs 1-2, 4). These species are readily distinguished by the very different shape of valva, uncus lobes and aedeagus in the male genitalia (Figs 5 and 7). In the female genitalia the diagnostic characters of E. differens are the non-sclerotized an- trum and colliculum, narrower signum and pyriform corpus bursae (Figs 11-12, 14). The colliculum of E. gleichenella is dorsally sclerotized (Fig. 15) ; the corpus bursae is constricted just below middle (Fig. 13) [this character is ignored in TRAUGOTT-OLSEN & NIELSEN (1977: Fig. 401]). The male genitalia of E. differens somewhat resemble those of E. regificella Sircom, but the gnathos is much larger, and the aedeagus 114 Figs 1-4. Habitus of Elachista spp. : 1 — E. differens Parenti ¢ (Nordeifel, Germany) ; 2 — E. differens 9 (Nordeifel, Germany) ; 3 — Holotype & of E. lambeseella Nielsen & Traugott-Olsen (Lambése, Algeria) ; 4 — E. gleichenella (Fabricius) 2 (S. Germany). is straight in E. regificella (Fig. 8). The female of E. regificella has a longitudinal, dentate sclerotization in the colliculum (Fig. 16), which is lacking in E. differens. Based on male genitalic characters E. lambeseella Nielsen & Traugott- Olsen, 1987 seems to be the closest relative of E. differens. Externally E. differens can be separated from this species by the broader forewing and larger size (Figs 1-3). In the male genitalia the uncus lobes are longer and more slender, the valva is narrower, the cucullus of the valva is more elongate and slightly bent and no cornuti are present in the aedeagus (Figs 5-6, 9-10). The female of E. lambeseella remains unknown. Redescription (Figs 1-2, 5, 9-10, 11-12, 14) Labial palpi drooping, slightly curved, leaden grey, third segment slightly shorter than second. Head and neck tufts leaden grey with 115 Figs 5-8. Male genitalia of Elachista spp. : 5 — E. differens Parenti (L. Kaila prep. no. 1136) ; 6 — Holotype of E. lambeseella Nielsen & Traugott-Olsen (prep. E. S. Nielsen 1776) ; 7 — E. gleichenella (L. Kaila prep. no. 444) ; 8 — E. regificella (L. Kaila prep. no. 1127). metallic sheen. Antennae unicolorous grey, in distal 2/3rds segments with distally slightly raised scales. Tegulae, thorax and abdomen leaden grey with metallic sheen. Legs grey, underside of tibia creamy white, tarsal segments with white distal rings. Forewing ground colour mottled black with bronzy sheen ; base shining silvery ; fascia from before middle of costa to middle of dorsum, silvery with bluish or greenish reflection ; triangular costal spot creamy white at costa, towards middle of wing silvery just beyond opposite silvery tornal spot ; an irregular third spot between tornal spot and apex formed by some silvery scales ; especially in female the three spots often joined forming fascia with medial angle towards apex. Hindwing and underside of wings grey. Forewing length 3.5-4 mm in male (n=4), 2.5-4 mm in female (n = 13). Most female specimens smaller than males. MALE GENITALIA : Uncus lobes long, gradually tapering toward rounded tip, with row of stout setae. Gnathos large, rounded. Valva rather narrow, tapering towards tip ; costa straight, well sclerotized ; cucullus 116 Figs 9-10. Male genitalia of E. differens : 9 — aedeagus with carina in the margin of distal opening (L. Kaila prep. no. 1133) ; 10 — juxta lobes (L. Kaila prep. no. 1134). Figs 11-13. Female genitalia of Elachista spp. : 11, 12 — E. differens (L. Kaila prep. no. 1137) ; 13 — E. gleichenella (L. Kaila prep. no. 1132). 117 2 ; ; Meg 4 u 77 ; F4; Figs 14-16. Region of ostium bursae in female genitalia of Elachista spp. : 14 — E. differens (L. Kaila prep. no. 1137); 15 — E. gleichenella (L. Kaila prep. no. 1131); 16 — E. regificella (L. Kaila prep. no. 1130). elongate with rounded tip, slightly bent forming an angle with costa. Digitate process rather broad, blunt, with setae. Juxta lobes separated by short incision medially, almost parallel-sided, truncate, apical margin slightly rounded near the incision ; with short, pointed lateral process. Vinculum rounded, with indistinct median ridge. Aedeagus short and broad, S-shaped ; caecum processed, blunt ; distal part oblique, tapering into strongly pointed distal end, margin of distal opening laterally asymmetrical with elongate carina ; without cornuti. FEMALE GENITALIA : Distal margin of eighth tergite with long setae. Apophyses stout, posteriores three times longer than anteriores. Ostium bursae rounded with strongly sclerotized margins. Antrum very short, funnel-shaped, neither antrum nor colliculum sclerotized. Corpus bur- sae rounded with small internal spines ; signum elongate, dentate. Remarks We consider the series from Greece, Corfu, to belong to E. differens, although the uncus lobes seem to be slightly narrower, and carina of aedeagus is smaller in these specimens. These differences are in our opinion minor, and the material available does not allow taxonomic separation of these populations. Further material is needed for an evaluation of the geographic variation of the species. 118 Acknowledgements B. Krutzsch (Berlin) loaned us the holotype of Elachista lambeseella. R. Tyynelä helped with photographs, and K. Mikkola made valuable comments on the manuscript. We thank all these persons for their help. References FABrıcıus, J. C., 1781. Species Insectorum exhibentes eorum Differentias specificas, Synonyma Auctorum, Loca Natalia, Metamorphosin adjectis Observationibus, Descriptionibus 2, 517 pp. Hamburgi et Kilonu. NIELSEN, E. S. & TRAUGOTT-OLSEN, E., 1987. Four new West Palaearctic species of Elachistidae (Lepidoptera). Entomologist’s Gaz. 38 : 103-113. PARENTI, U., 1978. Nuove specie paleartiche del Genere Elachista Treitschke (Lepidoptera, Elachistidae). Boll. Mus. Zool. Univ. Torino N. 4 : 15-26. SIRCOM, J., 1849. Description of three new British Tineidae. Zoologist 7, App. 42. SRUOGA, V., 1990. Seven new species of Elachistidae (Lepidoptera) from the USSR. Tijdschr. Ent. 133 : 75-84. TRAUGOTT-OLSEN, E. & NIELSEN, E. S., 1977. The Elachistidae (Lepidoptera) of Fennoscandia and Denmark. Fauna Ent. Scand. 6. 299 pp. 109) Nota lepid. 17 (3/4) : 120 ; 30.IV.1995 ISSN 0342-7536 Dr. h.c. Karl Burmann + Wir haben die traurige Pflicht Ihnen mitzuteilen, dass der bekannte oesterreichische Lepidopterologe Dr.h.c. Karl Burmann tot ist. Er starb am 26. März 1995 im 87. Lebensjahr. 120 Nota lepid. 17 (3/4) : 121-123 ; 30.1V.1995 ISSN 0342-7536 Aricia crassipuncta bassoni Larsen, 1974 from Lebanon raised to species rank (Lepidoptera, Lycaenidae) Torben B. LARSEN 358 Coldharbour Lane, London SW9 8PL, U.K. Summary A reexamination of material of Aricia crassipuncta bassoni Larsen, 1974 from Lebanon shows that it differs from A. vandarbani Pfeiffer, 1937 from Iran, with which it had been synonymised, and also from A. anteros Freyer, 1838 [Turkey] and A. crassipuncta Christoph, 1893 [Armenia]. The differences are to be found in wing pattern and shape, and in the male genitalia. The taxon is therefore raised to species rank. Résumé Le réexamen de matériel d’Aricia crassipuncta bassoni Larsen, 1974 du Liban, montre qu’il differe de A. vandarbani Pfeiffer, 1937 d’Iran, avec lequel il avait été synonymisé, et également d’A. anteros Freyer, 1838 (Turquie) et d’A. crassipuncta Christoph, 1893 (Arménie). Les différences sont à trouver dans les dessins et les formes des ailes, ainsi que dans les genitalia mâles. Le taxon est en conséquence élevé au rang d’espèce. In 1974, I described the taxon Aricia crassipuncta bassoni from the high mountains of Lebanon. It belongs in the subgenus Ultraaricia Beuret, 1959, a small group of species characterized by the presence of an unusual inferior lobe in the uncus. On the advice of Dr. Burk- hardt Alberti and Dr. Walther Forster, I allied the Lebanese taxon with A. crassipuncta Christoph, 1893 from Armenia, despite their being separated by a distance of more than a thousand kilometres. I was encouraged to do this also because the Lebanese population was traditionally referred to tentatively as ssp. crassipuncta of Aricia anteros Freyer, 1838. In a paper in this journal, NEKRUTENKO (1980) placed the taxon bassoni as a synonym of A. vandarbani Pfeiffer, 1937 from the Iranian Elburs Mountains on morphological grounds, though he also said it might possibly be a valid subspecies thereof. K. Schurian (pers. comm.), to 23 AA N U. bassoni U. anteros Fig. 1. Uncus and valves of Ultraaricia spp. : U. bassoni from Lebanon (Larsen prep. LAI) ; U. anteros from Anatolia (Ankara area) (Larsen prep. LAJ). whom I gave some specimens, did not agree with this decision, and I had for long wanted to re-examine the issue. After many years, I recently regained access to my genitalia mounts from Lebanon. Examination of a male bassoni shows that the specialized inferior lobe of the uncus is exactly like that of a male A. anteros from near Ankara (Fig. 1). Since the main characteristic of A. vandar- bani is a strong reduction of this lobe, the assignment of bassoni to A. vandarbani clearly becomes impossible (no material of bassoni was available to Nekrutenko). On the other hand, Nekrutenko’s redescription of A. crassipuncta, a poorly known species, makes it clear that it differs more from bassoni than I thought in 1974. A. crassipuncta has very pointed forewings, the underside pattern is less strongly developed, the wings almost wholly lack orange marginal lunules, and the overlay of light scales is blueish- grey rather than greenish. To this must now be added that the valve of bassoni is proportionately longer than in the other three species, and the distal spine is long, jutting well beyond the distal end of the valve. The length of the valve is due to elongation of the basal part, the distal half having the usual proportions. I therefore raise Aricia (Ultraaricia) crassipuncta bassoni Larsen, 1974 to species rank (stat. 122 n.). This also seems the most reasonable solution on biogeographical grounds. Males of A. bassoni (illustrated in colour by LARSEN, 1974) are readily recognized by the grey upperside with a strong, greenish sheen, usually with well developed marginal orange lunules. They are very different from the smaller blue Anatolian males. Females are like those of Aricia agestis Denis & Schiffermüller, 1775 with strongly developed marginal lunules. The haploid chromosome number is n = 23 (LARSEN, 1975), which has also been found in nominate A. anteros, and which appears to be typical for the genus. Hiccins (1975) quotes n = 24 for A. agestis, but I found Lebanese males to have n = 23. There are also occasional records of A. crassipuncta from southwestern Turkey (Hicains, 1966 ; NEKRUTENKO, 1980). It 1s thus possible that a population of A. bassoni exists in the Taurus Mountains, with which the Lebanese butterfly fauna has many other affinities, but I have not seen material from there. References Hicains, L. G., 1966. Check-list of Turkish butterflies. Entomologist 99 : 209-222. Hicains, L. G., 1975. The classification of European butterflies. Collins, London. Larsen, T. B., 1974. The butterflies of Lebanon. National Council for Scientific Research, Beirut. LARSEN, T. B., 1975. Chromosome numbers and notes on testicular morpho- logy of some Lebanese Rhopalocera. Entomologica scand. 6 : 218-225. NEKRUTENKO, Y. P., 1980. Revisional notes on the lycaenid butterfly species assigned to Ultraaricia Beuret (Lycaenidae). Nota lepid. 3 : 55-68. 123 Nota lepid. 17 (3/4) : 124 ; 30.1V.1995 ISSN 0342-7536 Book reviews — Buchbesprechungen — Analyses Oekologische Untersuchungen im Unterengadin. Schmetterlinge (Lepi- doptera). W. SAUTER. 137 pp., kartoniert. Ergebnisse der wissenschaft- lichen Untersuchungen im Schweizerischen Nationalpark Band 12, 14. Lieferung, 1993. Bestellungen an: F. Flück-Wirth, Internationale Buchhandlung für Botanik und Naturwissenschaften, CH-9053 Teufen, Schweiz. Preis : 78 Fr. Die vorliegende Arbeit ist Teil einer interdisziplinären, durch geplante Kraft- werksbauten, angeregten Studie im Schweizer Unterengadin. Die Schmetter- lingszönosen unterhalb der Waldgrenze wurden von 1961 bis 1976 mit unter- schiedlicher Methodik erfaßt und durch zusätzliches Datenmaterial bis 1990 ergänzt. Insgesamt konnten 1242 Arten nachgewiesen werden, die alle ein- schließlich der Fundorte aufgelistet sind. In der Schweiz im wesentlichen auf das Unterengadin beschränkte Arten werden zoogeographisch analysiert. Leider konnten hier neuere Ergebnisse über Caryocolum oculatella nicht mehr berücksichtigt werden. Besonders wertvoll erscheint dem Rezensenten die ökologisch orientierte Analyse der Artenbestände im 2. Teil der Studie, die trotz aller vom Autor offen dargelegten Unzulänglichkeiten eine reichhaltige Fundgrube für ähnlich gelagerte Untersuchungen im Mitteleuropa darstellt. Basierend auf den gut dokumentierten pflanzensoziologischen Verhältnissen, sowie den bekannten Daten über Raupenfutterpflanzen wurden die Arten soweit als möglich den unterschiedlichen Vegetationstypen zugeordnet. Die Zönosezugehörigkeit der einzelnen Taxa wurde möglichst exakt vorgenommen und beinhaltet im Sinne von Schwerdtfeger (1975) mehrere Kategorien von zönoseigenen Arten bis zu Irrgästen. Besonders hohe Diversitätsraten wurden im Koelerio-Poetum xerophilae (236 Arten), im Vincetoxico-Festucetum sulcatae (166 Arten) sowie im Violo-Alnetum incanae (116 Arten) nachgewiesen werden. Die Pionier- standorte des Inn sind erwartungsgemäß artenarm, weisen aber einige wichtige Vorkommen auf wie z.b. Merulempista cingillella (Tamariskenzünsler). Die Studie ist trotz des relativ hohen Preises allen ökologisch orientierten Lepidopterologen wärmstens zu empfehlen, und untermauert die Bedeutung von Schmetterlingserhebungen für Aussagen über die Wertigkeit von Lebens- räumen. Peter HUEMER 124 Nota lepid. 17 (3/4) : 125-140 ; 30.1V.1995 ISSN 0342-7536 Karyology and distribution as tools ın the taxonomy of Iberian Agrodiaetus butterflies (Lepidoptera : Lycaenidae) Miguel L. MunGuIRA, José MARTIN and Margarita PEREZ-VALIENTE Departamento de Biologia (Zoologia), Facultad de Ciencias, Universidad Autönoma de Madrid, Cantoblanco, E-28049-Madrid, Spain. Summary A cytotaxonomical study in the main distribution areas of Agrodiaetus ripar- tii, A. fabressei, A. ainsae and A. fulgens revealed differences between these species. In the species with brown males, A. ripartii has n = 90 with two macro- chromosomes and A. fabressei the same chromosome number, but three large chromosomes. In the species with blue males, A. fulgens (considered here a true species) and A. ainsae have n= ca. 103 and n= 108 with two and six macrochromosomes respectively. Macrochromosomes proved to be the best genetic marker to identify the species of Agrodiaetus in the Iberian Peninsula. By combining karyological and morphological data we were able to construct detailed UTM maps for the four species. A. ripartii is found from Catalonia to the Cantabrian Mountains, in the south to the Sistema Ibérico, A. fabressei flies in the Sistema Ibérico and Andalusian Sierras, A. fulgens is a Catalonian species with a very restricted distribution range and A. ainsae lives in the Central and Western Pyrenees and the Cantabrian Mountains. The taxon recently described as A. violetae is considered a subspecies of A. fabressei, based on the similar chromosome morphology and number. The group seems to have evolved very quickly karyologically, but morphological and ecological differences are not so evident. Resumen El estudio citotaxonömico en las principales areas de distribuciön de Agro- diaetus ripartii, A. fabressei, A. ainsae y A. fulgens revelö diferencias entre las especies. Las especies de machos castafios A. ripartii y A. fabressei tienen respectivamente n = 90 y dos macrocromosomas y el mismo numero de cromo- somas, pero con tres cromosomas grandes. En el caso de las especies de machos azules A. fulgens (elevada aqui a la categoria de especie) y A. ainsae tienen n=ca. 103 y n= 108 con dos y seis macrocromosomas respectivamente. Los macrocromosomas resultaron ser el mejor marcador genético para identificar las especies del género en la Peninsula Ibérica. Mediante el analisis conjunto de datos cariolögicos, morfolögicos y de distribuciön (estudiada con mapas 125 detallados en proyecciön UTM), obtnemos una idea clara de la taxonomia del grupo. Asi A. ripartii se encuentra desde Cataluña hasta el Sistema Cantabrico, siempre al norte del Sistema Ibérico, A. fabressei vive en el Sistema Ibérico y sierras andaluzas, A. fulgens esta restringida a unas pocas localidades cata- lanas y À. ainsae vive en los Pirineos Centrales y Occidentales y los Montes Cantäbricos. La recientemente descrita A. violetae se considera una subespe- cie de A. fabressei por su similar numero y morfologia cromosémicas. El grupo parece haber evolucionado muy räpidamente desde el punto de vista cariolögico, mientras que las diferencias morfolögicas y ecolögicas no son tan patentes. Introduction The taxonomy of the genus Agrodiaetus Hübner, [1822] is one of the most complicated among Palaeartic butterflies. The genus lives mainly in the Mediterranean and Middle East, but can also be found in Central Europe and Russia (Hiccins & HARGREAVES, 1983 ; KuDRNA, 1986). The biology is similar in the different species of the group, the larvae use sainfoin (Onobrychis) as foodplants, and overwinter in this stage, feeding during the following spring on the young leaves of the plant (SCHURIAN, 1976; Lepidopterologische Arbeitsgruppe der Schweiz, 1987 ; Manıno et al., 1987; MUNGUIRA, unpublished data). Pupation takes place at the base of the plant and adults begin to fly usually after a month. The flight period ranges from July in hotter places to August in populations living at higher altitudes. Interest in the group lies in the controversial taxonomic position of most species. Before chromosome studies were made, morphological research attempted to provide a clear view of the systematics and taxonomy of the group. Some comprehensive reviews attempted this difficult task (FORSTER, 1961), but the confusion generated by this approach proved the inadequacy of the methods based solely on morphological analysis. Chromosomic studies started by de Lesse in 1952, clearly showed that morphology itself was not enough to under- stand relationship among the species (DE LESSE, 1960a). Unfortunately these studies are not a panacea either and the species’ chromosome numbers also show a complicated pattern, with very different chromo- some numbers in otherwise identical taxa. This shows how hard it is to give simple answers to difficult questions in a group that is under- going a splitting process at the moment we are studying it. For example, in Italy the taxa previously grouped under the name Agrodiaetus ripar- tii (FREYER, 1830) has been split into three different species with different chromosome numbers : A. humedasae Toso & Balletto, 1976 126 with n = 38 ; A. galloi Balletto & Toso, 1979 with n = 66 ; and A. ripar- tii with n = 90 (TRoIANO et al., 1979). AGENJO (1947) lists the Spanish species as Plebejus (Agrodiaetus) damon (Denis & Schiffermüller, 1775), P dolus (Hübner, [1823]), P. admetus (Esper, 1785) and P ripartii. FORSTER (1961) describes ainsae as a subspecies of dolus and considers Agrodiaetus as an independent genus. Following the popular book by GOMEz-BusTILLo & FERNAN- DEZ-RUBIO (1974) the Iberian species of the genus are Agrodiaetus damon, A. dolus, A. fabressei (Oberthür, 1910), A. ripartii and A. ainsae Forster, 1961, to which A. violetae Gomez-Bustillo, Expösito & Martinez, 1979 was added later as a new species. A. fabressei cor- responds to what AGENJo (1947) named P admetus, now known to be restricted to Eastern Europe and Asia Minor. There is a tendency to split the species of the group into new species whose validity has been discussed in several reviews (see for example FERNANDEZ-RUBIO, 1992). Hıscıns (1975) also considers within the genus the species A. amanda (Schneider, 1792), A. thersites (Cantener, 1834) and A. escheri (Hübner, [1823]), but although this inclusion has strong arguments in its favour, the resulting genus is less homogeneous. KuDRNA (1986: 161, 229-231) in the latest European checklist grouped the Agrodiaetus within the genus Polyommatus and listed the following species as present in Spain : P ainsae, P. damon, P. dolus, P. fabressei, P. ripartii, and P. violetae. The first three taxa have blue males and the last three brown males. The inclusion of Agrodiaetus in Polyommatus is not supported by recent research (LELIEVRE, 1992) and we do not assume it for simplicity. We centered our study only in the Agrodiaetus sensu stricto group (following Hiccins & RırLry, 1970), but excluded A. damon which is a fairly distinct species that has never been mistaken with the others anywhere in its range. The objectives of our study were to identify the specific entities of the Agrodiaetus group living in the Iberian Peninsula based on cytotaxonomical and morphological information and to give accurate UTM distributions for every species. Materials and methods Chromosome number studies were made on male testes of at least five specimens for each species and/or geographical area. Testes were dissected in situ under a stereomicroscope in distilled water and fixed with a solution of ethanol : acetic acid (3:1). They were kept at ca. 5°C until analysis was possible. After staining the samples with lacto- propionic orceine they were observed under a phase contrast microscope 127 to count chromosomes using the squash technique (Lorkovic, 1990) with small fragments of the testes. Brown males were studied from Sierra de Cazorla (SW Spain), Sistema Ibérico (Central Spain), Cata- lonia (NE Spain) and the Cantabrian Mountains (N Spain). Blue males were studied from Catalonia and the Pyrenees. This covered all the taxa previously described and the geographic range of the group in the Iberian Peninsula and was considered enough to give an idea of the chromosome numbers for each area and species. Distribution data were collected using faunistic records from the literature and from specimens in the collections of the following Spanish institutions : Museo Nacional de Ciencias Naturales, Sociedad de Ciencias Naturales Aranzadi, Museo de Ciencias Naturales de Vitoria, Museo de Zoologia de Barcelona and Universidad Autönoma de Madrid. Private collections from Fidel Fernandez-Rubio, Jose Luis Yela, Arcadi Cervellé, Jose Maria Font and Jose Luis Nuñez and personal communications from Albert Maso, José Bellavista and Fran- cisco Abös were used for distribution records. Data were also gathered visiting a total of 36 localities from 12 different Provinces in which specimens were taken for morphological analysis. The specimens used for this study are preserved in the zoological collection of the Uni- versidad Autönoma de Madrid (UAMZ). A database was created to process all the faunistic data with information on localities, provinces, dates, UTM coordinates, altitudes and bibliographic references. This is available upon request for anyone interested, but its volume (more than 700 records) made its inclusion in this paper impossible. Maps were produced using a commercial program for automated carto- graphy (CYANUS) for the Iberian Peninsula and Balearic Islands. Each record was assigned to a particular species using karyological data for the localities from which chromosome studies were made. For the rest of the localities identification was based on geographical and morpho- logical affınities with the former records. The morphological analysis of the records for which we had specimens, showed that the individuals identified by this method had the typical morphology of the species to which they were assıgned. The biology of the group was studied at the different localities visited, where we gathered data on foodplants, overwintering stages and im- mature stage morphology. 128 Table 1 Localities, sample size and chromosome numbers of the studied material and all the literature references for Agrodiaetus species in the Iberian Peninsula. “No. males” refers to the number of males that produced metaphase I plates suitable for chromosome counts and the total sample is given in brackets. References are as follows : (1) DE LESSE, 1960b (2) DE LeEsse, 1961a (3) DE LESSE, 1962 (4) DE LESSE, 1968 and (*) our own observations Locality Villanueva Peñahorada Bernués Jaca Taradell Noguera Albarracin Peñalén Cazorla Morella Province Burgos Burgos Huesca Huesca Barcelona Teruel Teruel Guadalajara Jaen Castellön No. males | Chr. No. | Macrochr.| Species 108 9 108-110 103 90 90 90 90 ainsae ainsae ainsae ainsae fulgens fabressei fabressei fabressei fabressei fabressei Teruel 90 Castellön 90 Barcelona 4 90 Barcelona Barcelona Tarragona Villarroya Olocan Amorös Collsuspina Taradell Santa Coloma Penahorada Burgos Gredilla Burgos Jaca Huesca Jaca Huesca fabressei ripartit ripartit ripartit ripartit ripartit ripartit ripartit ripartii ripartit NNNN | BORDER | AD |] sw. NO * * * * x x BON ND * + ke x FEW + LU Results Chromosome numbers in different populations Table 1 gives a summary of the data from our study pooled with data from DE Lesse’s 1960b, 1961a, 1962 and 1968 papers. From these results it is clear that the best genetic marker for the identification of the species of the group in the Iberian Peninsula is the number of macrochromosomes (see WHITE, 1973 for the use of the term, that appears as “gros chromosomes” in DE LESsE, 1960b). These can be defined in Lepidoptera as chromosomes with two to four times the normal size, located in the centre of metaphase I plates and surrounded by normal-sized chromosomes (Figs 1-4). The location of macro- chromosomes contrasts with their peripheral location in Orthoptera, amphibians and reptiles (WHITE, 1973). In lycaenids the number of macrochromosomes is always low while in the other groups it usually outnumbers that of microchromosomes (see WHITE, 1973 for compa- 129 Figs 1-4. Metaphase I plates. 1 — Agrodiaetus fulgens. (Taradell, Barcelona Province, type locality for species) n = 103, six macrochromosomes ; 2 — Agrodiaetus fabres- sei (Penalen, Guadalajara Province), n = 90, three macrochromosomes ; 3 — Agro- diaetus ripartii (Collsuspina, Barcelona Province), n = ca. 88, two macrochromosomes ; 4 — Agrodiaetus fabressei violetae (Sierra de Cazorla, Jaén Province), n = 90, three macrochromosomes. 130 rison), and it is probable that these configurations depend on mechanical aspects of the cell division process. The number of chromosomes is also distinctive, but the metaphase I plates of Agrodiaetus have large chromosome numbers and accurate counts are difficult in most cases. The results from Table 1 support the following species’ arrangements : A. ainsae with n — ca. 108 lives in the Pyrenees and a wide area sur- rounding the Cantabrian Mountains, whereas in Catalonıa the speci- mens previously assigned to A. dolus should now be considered a distinct species. The name A. fulgens (Sagarra, 1925) is valid for this species, because it was given to what was thought a subspecies of A. dolus in the same localities where the samples for chromosome studies were taken. The reason for considering fulgens a distinct species is the different chromosome number (n = ca. 103, while dolus has n = 124 and ainsae n = 108) and above all, the different number of macro- chromosomes of these specimens (six for fulgens (Fig. 1) as opposed to four in dolus, DE LESSE, 1961b). de Lesse (1962, 1966) only gives the chromosome number for A. ainsae and does not provide figures for it, but our specimen from Peñahorada (Burgos Province, northern Spain) had two macrochromosomes although the metaphase I plates we obtained were not suitable for accurate chromosome counts. The number of macrochromosomes is therefore six in A. fulgens, four in A. dolus and two in A. ainsae. Although our sample for fulgens 1s very small (only 3 specimens with metaphase I plates), the separation of ainsae from dolus was made with similar samples (five specimens, DE LESSE, 1962). Moreover it is not chromosome number, but the number of macrochromosomes that we consider to be a strong argument to split fulgens from ainsae, because it is highly improbable that specimens with such a different chromosome morphology can belong to the same species. A. ripartii and A. fabressei share the same chromosome number (n = 90) but clearly differ in the morphology of chromosomes, fabressei having three (Fig. 2) and ripartii two (Fig. 3) macrochromosomes. DE LESSE (1960a, Table 1) states that fabressei has four macrochromosomes, but after examining our photographs and drawings we have concluded that the species clearly has three large chromosomes constantly seen in all the metaphase I plates. The difference between de Lesse’s results and ours is due to the existence of one or two medium sized chromo- somes that in some plates resemble macrochromosomes, but are driven out of the centre of the spindle in a majority of our figures from fabressei. Large chromosomes are always located in the centre of meta- phase I plates, and they are easily spotted when comparing long series of plates. This has been possible with the material from the Sierra 131 de Cazorla, previously regarded as A. violetae, where the presence of three macrochromosomes is clear (Fig. 4). The latter race has therefore a true fabressei karyotype, which supports the idea outlined in FER- NANDEZ-RUBIO (1992) that it is a subspecies of fabressei and not of ripartii, as has been proposed (BALLETTO, pers. comm.). The distribution of both karyotypes seems to be clearcut from the data in Table 1. A. ripartii 1s a species living from the Cantabrian Mountains to the Catalonian Pyrenees whereas A. fabressei lives in central Spain and in some southern mountain ranges. Morphology There is not a single character clearly separating all ripartii specimens from fabressei, and ainsae from the Catalonian species fulgens. In the first case a group of characters can identify most of the specimens, but in the latter this proves to be very difficult. Generally speaking fabressei lacks the white band along the v4 on the hind wing’s underside, but there is a great variability on this character. Traditionally brown males without the white band have been regarded as fabressei, and this has produced records of the species in the Pyrenees (GöMEZ, 1988) and Catalonia (GOMEz-BusTILLO & FERNANDEZ-RUBIO, 1974). The book by MANLEY & ALLCARD (1970) is unusual because it illustrates an individual with white band taken at Penahorada (Burgos Province) under the name fabressei and a very similar specimen (with white band) from Albarracin (Teruel) as ripartii. These identifications are in dis- agreement with our karyological results. On the other hand a white band is also present in some fabressei specimens leading some authors to consider these as ripartii (GOMEZ-BUSTILLO & FERNANDEZ-RUBIO, 1974 ; MANLEY & ALLCARD, 1970 for the male collected in Albarracin, Teruel). The Catalonian race agenjoi Forster, 1965 has been considered a different species (Hıscıns & HARGREAVES, 1983), a subspecies of fabressei (MANLEY & ALLCARD, 1970 ; GOMEz-BuUSTILLO & FERNAN- DEZ-RUBIO, 1974) or a subspecies of ripartii (AGENJO, 1964 ; DE LESSE, 1968, PÉREZ, 1979). Black spots on the underside are larger in fabressei than in the normal ripartii, but again the Catalonian specimens are an exception to this and resemble typical fabressei. Another distinctive feature of fabressei has traditionally been the enlargement of the black spot between v2 and v3 in the forewing’s underside (DE LESSE, 1968), but our material from the whole geographic range of both species does not support the distinctness of this feature. The apical angle of the forewing is smaller in fabressei than in ripartii. Althoug this seems to be a constant character, it is hard to measure and by itself is not use- ful enough to identify all the specimens clearly. The third supposed 132 species with brown males, A. violetae, has a mixture of morphological characters from ripartii and fabressei with a white band in the underside of the hindwing (absent in six of our sample of 22 butterflies) and a small angle in the apex of the forewing. As a result of ıts chromo- some morphology we consider it a subspecies of A. fabressei. Although there are exceptions to all the characters mentioned above, most of the specimens can be identified by a mixture of characters. Therefore specimens without white band in the underside, large spots ın the underside and small apical angle can be considered fabressei, provided they are in the distribution range of the species. Individuals outside the normal species’ range need chromosome study to be sure of their identity. As far as the species with blue males is concerned identification is also difficult. A. fulgens has traditionally been regarded as a race with brighter blue and lack of the white band ın the hindwing’s under- side. These two features are useful for most specimens, but again a substantial morphological variation takes place in ainsae for both characters, making individual diagnosis difficult in some cases. Distribution Given that it is very difficult to identify the species of the group based on morphological features, we considered the possibility to use both chromosome numbers and distribution data to assign a given individual to a species of the group. In almost all the studied northern localities, three Agrodiaetus species live together : damon, which is a well cha- racterized distinct species, a species with blue males and a species with brown males. We assumed that each area has a sıngle brown and/or blue species (this was confirmed by the chromosome study) and by mapping the species tried to define areas isolated by barriers separating species. In the localities of the Sistema Iberico (Central Spain) only a species with brown males flies together with damon and the chromo- some studies revealed this to be fabressei (DE LESSE, 1960b, and our observations, Table 1). A. fabressei (Fig. 5) is therefore restricted to the Sistema Iberico, a mountain range running from north to south in Central Spain and to the Sierras of Andalusia. The species lives in the Serranias of Cuenca and Albarracin, the high plains of La Alcarria, some calcareous out- crops in Segovia Province and north of Madrid, and reaches the Province of Soria in the north, being isolated from ripartii by the Picos de Urbiön, whose northern slopes flow down to the Ebro Valley. A. fa- bressei violetae is found in a total of twelve UTM squares (10 X 10 km) 138 Fig. 5. Distribution of Agrodiaetus ripartü (circle) and A. fabressei (star) in the Iberian Peninsula. Each symbol represents the presence of the species in a 10 X 10 km UTM square. Open symbols represent localities where chromosome studies were available. from four mountain ranges in Southern Spain (Sierra Tejeda, Sierra de Almijara, Sierra de Cazorla and Sierra de Alcaraz). A. ripartii (Fig. 5) lives over a wide area, from Catalonia through the Pyrenees to the Cantabrian Mountains, where it lives mainly on the southern slopes. Two subspecies can be distinguised: the Catalonıan agenjoi and the Pyrenean and Cantabrian ripartii, separated by an area that more or less matches the boundary between the Pyrenees and a mountain range running from north to south in eastern Catalonia (Serralada Vertical) in which the species becomes scarce (Fig. 5). DE LESSE (1961a) describes the presence of specimens with a typical ripartii 134 Fig. 6. Distribution of Agrodiaetus ainsae (circle) and A. fulgens (star) in the Iberian Peninsula. Each symbol represents the presence of the species in a 10 X 10 km UTM square. Open symbols represent localities where chromosome studies were available. karyotype in Olocau del Rey (Castellön Province), an area where fa- bressei is widespread. If this finding is confirmed it may be possible that ripartii has some populations more or less mixed with fabressei colonies in this contact zone. A. ainsae (Fig. 6) is the most widespread species with blue males in this area. Specimens from the western Pyrenees and the Cantabrian mountains belong to this species, but not a single locality has been found south to the Picos de Urbiön, suggesting that the southern limit for ripartii is also valid for the blue species of the group. 185 A. fulgens (Fig. 6) has until now been considered a subspecies of dolus (GOMEZ-BuSTILLO & FERNANDEZ-RUBIO, 1974, Hıccıns, 1975). It has only been cited from a handful of localities in Catalonia and between the strongholds of this species and the previous one there is a wide area with scarcity of records that can be seen when the distribution of both species is plotted together (Fig. 6). It should be noted that this lack of records is not a consequence of lack of information, for this is a well explored area. Biology Allthe studied species use Onobrychis as larval foodplants. In the Cata- lonian localities, the Pyrenees, the Sistema Ibérico and the Cantabrian Mountains the foodplant is always O. viciifolia Scop. The only popu- lation that uses a different foodplant is A. fabressei violetae in the Sierra de Cazorla. Its foodplant is O. argentea Boiss., but we do not consider this difference to be relevant with respect to its taxonomic status, because the two plants are ecologically similar, and O. argentea is the only species of this plant genus living in Sierra de Cazorla. In all other aspects the biology of the studied species is very similar. They all overwinter as third instar larvae (we studied this for fabressei, ripartii and ainsae) and pupate during the spring at the base of the foodplant. The fullgrown larva is very similar in species living in the same area. Thus we were unable to distinguish between A. fabressei and and the very different A. damon in Albarracin (Teruel Province), and between A. ripartii and A. ainsae in Jaca (Huesca Province). All the larvae have tentacles and dorsal nectary organs (Newcomer’s gland), and the coloured band in the lateral zone of the larva can have different colours, being yellow in the specimens from Jaca in the Pyrenees and pink in the larvae from Albarracin. Wether this can be a distinctive feature between ripartii and fabressei or just a character related to the ecological peculiarities of the site requires a more comprehensive study. Egg morphology is being studied under the SEM microscope by two of us (MuncuirA & MARTIN, in preparation) and all the species exhibit a very similar chorionic pattern, with no distinctive features among them. Discussion Previous results have shown how complicated a taxonomic study of this difficult butterfly group can be. Although butterfly taxonomy in Europe is often regarded as being settled, there are some groups in which a great deal of research is still needed. Two of the main 136 taxonomic groups within which the boundaries between species are not yet well defined are found in Agrodiaetus, and Polyommatus (Lysandra) of the coridon (Poda, 1761) group (DE BAST, 1985 ; Mensı et al., 1988 ; LELIEVRE, 1992). Our approach to the taxonomy of the group was to begin to identify the different karyotypes found in the Iberian Peninsula. Then we trıed to study as many areas as possible to assign karyotypes to relevant areas from the biogeographical point of view. Plotting this information with distribution maps, we tried to draw accurate maps of every species. The separation of species in contact areas such as the boundaries between A. ainsae and A. fulgens is still tentative and a karyological study of almost every locality is needed to be completely sure. The limit between A. ripartii and A. fabressei also needs some further study, but in this case we think that our proposal is closer to reality because the morphology of specimens from Abejar (Soria) resembles the typical fabressei, with most butterflies lacking the white band on the hind- wing’s underside (MANLEY & ALLCARD, 1970). On the other hand the specimens from the northern slopes of the Sistema Ibérico (e. g. Castañares de las Cuevas in La Rioja) are morphologically true ripartii with the white band present in all the studied material. Some research needs to be done on the identity of some populations close to the town of Morella in central eastern Spain, where DE LEssE (1961a) identified karyotypes belonging to fabressei and ripartii in nearby populations, but never in the same one. This area may prove to be a contact zone for the two species, and the segregation of populations may not be as clearcut as depicted in the distribution maps. In a locality between the towns of Olocan and Forcall, de Lesse (1961a) found males with a typical ripartii karyotype and a single male with fabressei morphology, for which a chromosome study was not possible. Speaking of this fabressei specimen he writes that “on a vu qu’aucun doute ne peut exister sur son identité”. Therefore a contact area between the two species might exist in Morella, but this statement has to be proven through more chromosomic studies. On the other hand, although some authors have reported the two species from the same locality, nobody has yet identified specimens of the two different karyotypes living to- gether. Populations having different chromosome numbers are usually thought to belong to different species. Although DE LEssE (1960a ; 1966) is very cautious to split species by their distinct chromosome numbers, his data have always been used as a proof to make new species’ arrange- ments (GOMEZ-BUSTILLO & FERNANDEZ-RuBIO, 1974 ; Hicains, 1975). Some intraspecific variation in chromosome numbers and chromosome 15% morphology can not be plainly rejected (WHITE, 1973), particularly in such similarly looking and difficult groups as the Agrodiaetus. Never- theless, before a more thorough study is done on the significance of chromosome number variability in the evolution of lycaenids, we assume it is safer to assign populations with different chromosome numbers to different species. This approach was made in most chromo- somic studies in butterflies, as for example to support the separation of A. humedasae (TROIANO, et al., 1979), to split a species complex in several species previously regarded as identical (SOUMALANIEN & Brown, 1984) or as a general rule to explain chromosome number variability (see WHITE, 1973). On the other hand a karyological feature of great taxonomic importance has emerged during our study: the variability of the number of macrochromosomes observed in the centre of metaphase I plates. This character was previously used by DE LESSE (1960b), but our data prove that it is the best genetic marker to identify the species of Agrodiaetus in the Iberian Peninsula. We can therefore recognize fulgens by its six macrochromosomes and separate it from ainsae that has only two, whereas the species with brown males differ in having two (ripartii) or three (fabressei) of these large chromosomes. As a conclusion to our study, the Spanish Agrodiaetus with their chromosome numbers and distribution are: A. ripartii (n = 90, two macrochromosomes) living from Catalonia to the Cantabrian Moun- tains, A. fabressei (n= 90, three macrochromosomes) living in the Sistema Ibérico (South to the Picos de Urbiön), the limestone out- crops of Central Spain and some southern mountain ranges (Sierra Almijara, Cazorla, Alcaraz and Tejeda), A. ainsae (n = 108, two macro- chromosomes) from the Pyrenees and Northern Spain and A. fulgens (n = ca. 103, six macrochromosomes) living in Catalonia. Acknowledgements We are very grateful to Fidel Fernandez-Rubio, Arcadi Cervellö, Jose Maria Font, José Luis Nufiez and José Luis Yela for kindly giving access to their private collections. Isabel Izquierdo, Ibön de Olano and Jesus Aldaba gave facilities for the revision of public collections and Albert Masö and Francisco Abös provided unpublished distribution records. José Serrano and Jaime Gosalvez helped us in many ways with cytogenetical analysis, discussion of data and giving access to their laboratories. Enrique Garcia-Barros made valuable suggestions to a previous draft of the manuscript. 138 References AGENJO, R., 1964. Secciôn de capturas. IV. Graellsia 20 : 191-202. BAST, B. DE, 1985. La notion d’espèce dans le genre Lysandra Hemming, 1933 (Lepidoptera Lycaenidae) (Première partie). Linn. Belgica 10 : 98-110. FERNANDEZ-RUBIO, F., 1992. Guia de mariposas diurnas de la Peninsula Ibérica, Baleares, Canarias, Azores y Madeira. Piramide, Madrid. Forster, W., 1961. Bausteine zur kenntnis der Gattung Agrodiaetus Scudd. (Lep. Lycaen.). II. Z. wien. ent. Ges. 46 : 74-80. Gomez, C., 1988. Atlas provisional de los lepidöpteros de la zona norte. Tomo III. Serv. Publicaciones Gobierno Vasco, Vitoria. GöMEZ-BusTiL1o, M. R. & FERNANDEZ-RUBIO, F., 1974. Las mariposas de la Peninsula Ibérica. Ropalöceros. Tomo II. 258 pp. ICONA, Madrid. Hicains, L. G., 1975. The classification of European butterflies. 320 pp. Collins, London. Hıscıns, L. G. & HARGREAVES, B., 1983. The butterflies of Britain and Europe. 256 pp. Collins, London. Hicains, L. G. & Rire, N. D., 1970. A field guide to the butterflies of Britain and Europe. 380 pp. Collins, London. Kuprna, O., 1986. Aspects of the conservation of butterflies in Europe. Butter- flies of Europe (O. Kudrna, Ed.). Vol. 8. 323 pp. Aula-Verlag, Wiesbaden. LELIEVRE, T., 1992. Phylogenie des Polyommatinae et structure genetique de six espèces du genre Lysandra, Hemming (Lépidoptères Lycaenidae). PhD Thesis, Université de Provence. Lepidopterologische Arbeitsgruppe der Schweiz, 1987. Tagfalter und ihre Lebensräume. 516 pp. Schweizerische Bund für Naturschutz, Basel. LESSE, H. DE, 1952. Quelques formules chromosomiques chez les Lycaenidae (Lépidoptères, Rhopalocères). C. R. Acad. Sci. Paris 235 : 1692-1694. LESSE, H. DE, 1960a. Spéciation et variation chromosomique chez les Lé- pidoptères Rhopalocères. Ann. Sci. Nat. Zool. Biol. Anim. 12 Ser. 2: 1-223. LESSE, H. DE, 1960b. Les nombres de chromosomes dans la classification du groupe d’Agrodiaetus ripartü Freyer. Revue fr. Ent. 27 : 240-264. LESSE, H. DE, 196la. Cohabitation en Espagne d’Agrodiaetus ripartii Freyer et A. fabressei Oberthür. Revue fr. Ent. 28 : 50-53. LESSE, H. DE, 1961b. Les nombres de chromosomes chez Agrodiaetus dolus Hübner et les espèces voisines. Alexanor 2 : 57-63. LESSE, H. DE, 1962. Variation chromosomique chez Agrodiaetus dolus Hb. (Lep. Lycaenidae). Alexanor 2 : 283-286. LESSE, H. DE, 1966. Variation chromosomique chez Agrodiaetus dolus Hübner (Lepidoptera Lycaenidae). Annls Soc. ent. Fr. 2 : 209-214. LESSE, H. DE, 1968. Agrodiaetus ripartii Frey. dans la region de Barcelone (Lycaenidae). Alexanor 5 : 203-235. Lorkovic, Z., 1990. The butterfly chromosomes and their application in systematics and phylogeny. pp. 332-96, in ©. Kudrna (Ed.) : Butterflies of Europe. vol. 2. Introduction to lepidopterology. 557 pp. Aula-Verlag, Wiesbaden. 159 MANINO, Z., LEIGHEB, G., CAMERON-CuRRY, P. & CAMERON-CURRY, V., 1987. Descrizione degli stadi preimaginali di Agrodiaetus humedasae Toso & Balletto, 1976 (Lepidoptera, Lycaenidae). Boll. Mus. reg. Sci. nat. Torino 5 : 97-101. MANLEY, W. B. L. & ALLCARD, H. G., 1970. A field guide to the butterflies and burnets of Spain. 192 pp. E.W. Classey, Hampton. Mensı, P., LATTES, A., SALVIDIO, S. & BALLETTO, E., 1988. Taxonomy, evolutionary biology and biogeography of South West European Polyom- matus coridon (Lepidoptera : Lycaenidae). Zool. J. Linn. Soc. 93: 259-271. PEREZ, J. J., 1979. Notas sobre les especes catalanas del gener Agrodiaetus (Hbn., 1822) (Lep. Lycaenidae). Butll. Soc. cat. Lepid. 22 : 7-9. SCHURIAN, K., 1976. Beitrage zur Biologie der Gattung Agrodiaetus, I Agro- diaetus ripartii Frey. (Lep., Lycaenidae). Ent. Z. Frankf. a. M. 86: 196-199. SOUMALANIEN, E. & BROWN, K. S. Jr., 1984. Chromosome number variation within Philaethria butterilies (Lepidoptera : Nymphalidae, Heliconiini). Chromosoma 90 : 170-176. TROIANO, G., BALLETTO, E. & Toso, G. G., 1979. The karyotype of Agro- diaetus humedasae Toso & Balletto, 1976 (Lepidoptera, Lycaenidae). Boll. Soc. ent. ital. 111 : 141-143. WHITE, M. J. D., 1973. Animal cytology and evolution. 961 pp. Cambridge University Press, Cambridge. 140 Nota lepid. 17 (3/4) : 141-154 ; 30.1V.1995 ISSN 0342-7536 Illustrierter Bestimmungsschlüssel für die Präimaginalstadien der Schwärmer Europas und Nordafrikas (Lepidoptera, Sphingidae) Teil IT : Eilarven Alexander PELZER DorfstraBe 20, D-30974 Wennigsen, Bundesrepublik Deutschland Summary Keys to the preimaginal instars of the hawkmoths of Europe and North Africa (Lepidoptera, Sphingidae). Part II : First instar larvae. This is the second of a small series of keys dealing with mature larvae, first instar larvae, and pupae. They aim at the identification of living specimens. Therefore, all characters that are not visible in the intact animal are omitted. It is hoped that the keys will be useful for faunistic purposes and for research in the ecology of hawkmoths. Zusammenfassung Dieser Schliissel ist der zweite aus einer kleinen Reihe, die sich mit den erwach- senen Raupen, den Eilarven und den Puppen befaßt. Sein Ziel ist das Be- stimmen lebender Tiere. Daher wird auf alle Merkmale, die am lebenden Tier nicht erkennbar sind (z.B. Mandibelformen), bewußt verzichtet. Die Schlüssel könnten vor allem in der Faunistik und für die ökologische Forschung von Nutzen sein. Resume Cette clé est la deuxième d’une petite série qui traite des chenilles — adultes et au stade LI — et des chrysalides. Elle a pour objectif de permettre la détermination des espèces vivantes à ces différents stades. Par conséquent ne sont pas pris en considération les caractères invisibles chez l’animal vivant (p. ex. la forme des mandibules). Ces clés pourraient rendre service en faunistique et pour les recherches sur l'écologie des Sphingidae. 141 Einleitung Eilarven verschiedener Schmetterlingsgruppen sind bisher vorwiegend zur Klärung der Systematik herangezogen worden (z.B. WASSERTHAL, 1970). Die hier wichtigen Merkmale sind jedoch in der Regel erst nach dem Töten des Tieres erkennbar und daher für eine einfache Artbe- stimmung vielfach ungeeignet. Eilarven sind bereits in der Frühzeit der wissenschaftlichen Entomologie ein beliebtes Studien- und Mikroskopierobjekt gewesen (WEISMANN, 1876 ; PouLton, 1885 ; 1886 ; 1888 ; GILLMER, 1904 ; DENso, 1906a,b). Auch ın den letzten Jahren sind Eilarven etlicher Schwärmer beschrieben worden (z.B. HEINIG, 1976; 1978 ; 1981 ; HArBıcH, 1978 ; Pirraway, 1979 ; PELZER, 1982 ; HEINIG & HASLER, 1986 ; FREINA, 1994). Auf die Merkmale, durch die sich verwandte Arten unterscheiden, gehen die Beschreibungen jedoch nur in Ausnahmefällen ein (PELZER, 1988). Von einfachen Bestimmungsschlüsseln für die Präimaginalstadien, mit denen sich bereits ein lebendes Tier bestimmen läßt, könnte neben der Systematik besonders die Faunistik profitieren ; im faunistischen Schrift- tum tauchen Eilarven als Artnachweis bisher nicht auf. Dabei sind Eier oder Eilarven manchmal nicht schwieriger zu finden als die Falter (z.B. PELZER, 1982). Könnte man bereits die Eilarve bestimmen, bräuchte man das fragliche Tier nicht bis zur erwachsenen Raupe oder gar bis zum Falter aufzuziehen — was auf Exkursionen in der Regel nur selten möglich sein dürfte. Die Bestimmung von Eilarven ist ın vielen Fällen erstaunlich einfach. Im Gegensatz zu erwachsenen Raupen weisen sie nämlich nur eine sehr geringe Variabilität auf. Andererseits wirkt die insgesamt nur kleine Zahl verwertbarer Merkmale jedoch limitierend. Systematik, geographische Abgrenzung, Material und Methoden Die verwendete Systematik wird im 1. Teil der Reihe erläutert (PELZER, 1991) ; sie folgt weitgehend der Auffassung von ROTHSCHILD & JORDAN (1903). Der Name Laothoe tremulae (Fischer von Waldheim, 1830) wird durch L. amurensis (Staudinger, 1892) ersetzt (Pirraway, 1993). —— Abb. 1-6. Eilarven europäischer und nordafrikanischer Schwärmer. In Klammern hinter dem Artnamen der jeweilige Abbildungsmaßstab. 142 Abb. 1. a) A. atropos (14 X); b) A. convolvuli (21%); c) S. ligustri (12 X) ; d) H. pinastri (19 X). abv Schwarmerarten Europas und Nordafrikas Acherontia atropos (Linné, 1758) Totenkopf Agrius convolvuli (Linné, 1758) Windenschwärmer Sphinx ligustri Linné, 1758 Ligusterschwärmer Hyloicus pinastri (Linné, 1758) Kiefernschwärmer Dolbina elegans A. Bang-Haas, 1912 Marumba quercus (Denis & Schiffermüller, 1776) Eichenschwärmer Mimas tiliae (Linné, 1758) Lindenschwärmer Smerinthus caecus Ménétriés, 1857 Smerinthus ocellatus (Linné, 1758) Abendpfauenauge Laothoe populi (Linné, 1758) Pappelschwärmer Laothoe amurensis (Staudinger, 1892) Hemaris fuciformis (Linné, 1758) Hummelschwärmer Hemaris tityus (Linné, 1758) Skabiosenschwärmer Hemaris croatica (Esper, 1779) Daphnis nerii (Linné, 1758) Oleanderschwärmer Macroglossum stellatarum (Linné, 1758) Taubenschwanz Proserpinus proserpina (Pallas, 1772) Nachtkerzenschwärmer Rethera komarovi (Christoph, 1885) Sphingonaepiopsis gorgoniades (Hübner, 1819) Hyles lineata (Fabricius, 1775) Linienschwärmer Hyles gallii (Rottemburg, 1775) Labkrautschwärmer Hyles euphorbiae (Linné, 1758) Wolfsmilchschwärmer Hyles nicaea (Prunner, 1798) Nizzaschwärmer Hyles centralasiae (Staudinger, 1887) Hyles zygophylli (Ochsenheimer, 1808) Hyles hippophaes (Esper, 1789) Sanddornschwärmer Hyles vespertilio (Esper, 1779) Fledermausschwärmer Deilephila elpenor (Linné, 1758) Mittlerer Weinschwärmer Deilephila porcellus (Linné, 1758) Kleiner Weinschwärmer Hippotion celerio (Linné, 1758) Großer Weinschwärmer Hippotion osiris (Dalman, 1823) Theretra alecto (Linné, 1758) Das im Schlüssel behandelte Gebiet umfaßt Europa bis zum Ural sowie Nordafrika nördlich der Sahara. Nach dem gegenwärtigen Kenntnis- stand treten 32 Schwärmerarten mehr oder weniger regelmäßig in diesem Areal auf (Tab. 1). Hinzu kommen einige Irrgäste aus Nord- amerika, Afrika und Asien, die hier nicht behandelt werden (Listen in MEERMAN, 1987 und Pitraway, 1993). Die Merkmale der wenigen Hybriden, die aus dem Freiland bekannt sind, vermitteln stets zwischen denen der Elternarten. Die Schlüssel basieren ausschließlich auf eigenen Daten. Sie beruhen auf zahlreichen Zuchten und Freilandbeobachtungen der meisten hier behandelten Arten. Von fünf Arten lag mir allerdings kein Material vor. Es handelt sich dabei um Arten, für die Europa auf dem äußersten Rand ihres Verbreitungsgebiets liegt. 144 IE: ’ is (11 X) c) L. amurens ’ . tiliae (12 X) ; b) tus (14 X). lla S. oce 5) ad) Abb. 2. a) M. quercus (8 X) popul e) Die Aufnahme von R. komarovi in den Schlüssel erfolgt nach einem Farbdia, das mir Prof. Dr. L.T. Wasserthal (Erlangen) dankenswer- terweise überlassen hat. Die noch unbeschriebene Eilarve von S. caecus dürfte der von S. ocellatus sehr ähneln. Aus der einzigen, mir vor- liegenden Beschreibung der Eiraupe von D. elegans geht lediglich her- vor, daß sie ein Horn besitzt (SOFFNER, 1959). Hier und bei den Arten S. gorgoniades und H. osiris kann — anders als bei den erwachsenen Raupen — nicht abgeschätzt werden, wo im Bestimmungsschlüssel man vermutlich ankäme. Zur Bestimmung ist eine Handlupe mit wenigstens 10-, besser 20facher Vergrößerung notwendig. Zur Ermittlung des Horn-/ Afterklappen- längen-Verhältnisses (H/ A, s.u.) besonders geeignet sind Meßlupen mit eingearbeiteter Millimeterskala, wie sie von Briefmarkensammlern ver- wendet werden. Einfache Längenangaben, wie z.B. die Gesamt-Körperlänge, sind als Bestimmungsmerkmal gewöhnlich ebenso unbrauchbar wie simple Farbangaben. Bei einigen Arten werden einige Zeit nach Beginn der Nahrungsaufnahme Zeichnungselemente des zweiten Stadiums mehr oder weniger deutlich sichtbar. Solche variablen Merkmale werden nicht berücksichtigt. Die Färbung der Hartteile ändert sich dagegen während eines Stadiums nicht und ist daher für die Bestimmung geeignet. Auch die Beweglichkeit des Horns ist ein gutes Merkmal : manche Arten können ihr Horn weit nach vorn neigen (Abb. 3d), bei anderen ist es praktisch unbeweg- lich. Wo die Futterpflanzen der Raupen bei der Bestimmung helfen können, werden sie mit aufgeführt. Ein „mathematisches“ Merkmal hat sich als besonders einfach und brauchbar erwiesen : das Verhältnis der Hornlänge relativ zur Länge der Afterklappe (Horn-/ Afterklappen-Längenverhältnis, H/A). Dieses Längenverhältnis läßt sich schon rein optisch gut abschätzen. Bei der Kiefernschwärmerraupe auf Abb. Id beträgt es etwa 1, d.h. Horn und Afterklappe sind etwa gleich lang. Das Schwanzhorn ist bei Eilarven — sofern es nicht beim Schlupf verbogen worden ist — immer gerade. Zur Ermittlung von H/A ist stets die Gesamtlänge des gestreckten Horns zu verwenden. Die Anordnung der einzelnen Arten auf den Tafeln weicht z.T. von der in Teil I ab, um ähnliche Raupen direkt nebeneinander zu stellen. Da die lebenden Eilarven einiger Arten bei Lupenvergrößerung nur schwer zu bestimmen sind, gibt ein Hilfsschlüssel zusätzlich Unterschei- dungsmerkmale des zweiten Raupenstadiums an. 146 Abb. 3. a) H. fuciformis (20 X); b) H. croatica (22 X) ; c) A. tityus (22 X) ; d) D. nerü (16 X). Hauptschlüssel Die Zahl in Klammern hinter dem Artnamen ist die jeweils zugehörige Abbildungsnummer. 1. Horn kurz (Verhältnis Horn- zu Afterklappenlänge [H/A]< 1), höchstens so lang wie in Abb. [dt ae... ee RC SES 2 1°. Horn lang (H/ A > 1), mindestens so lang wie in Abb. 2c ................. 14 2. Kopf, Brustbeine und Horn verdunkelt, oft schwarz (vergl. Abb. 5c) 3 2. zumindest'derKoptichell-r...2.. al en ee 6 3. frisch geschlüpft einfarbig schwarz ; nach Nahrungsaufnahme oder im Durchlichtdunke bis mittelonin 22 H. euphorbiae (5e) 3”... anders... ihnen ee ei SO E 4 4. Kopf wie lackglänzend, klar in schwarze und bräunliche Partien geglie- dert ; Horn an der Spitze deutlich gegabelt ; auf Nadelhölzern. ............... ds dea bide ee RE EEE HA. pinastri (1d) 4°. anders ; nicht auf Nadelhölzern 2... ee 5 5. Kopf und Nackenschild einfarbig schwarzbraun ..... H. centralasiae (5d) 5. Kopf scheckig, nicht einfarbig ; Nackenschild hell .......... H. lineata (Sc) 6... Horn hell oder reduziert 7... er fl 6. Horn dunkel, normal ausgebildet 2.220 Re 11 7. Horn klein, aber normal ausgebildet (vergl. Abb. 4f) .......................... 8 7”. Horn fast oder ganz zurückgebildet "en 9 8. mit langen, schwarzen, Y-förmigen Borsten ...................... H. tityus (3c) 8... mit hellen kurzen-Borstene cess ec eeee eee ee H. gallii (4f) 9. Borsten: Del a ae we P. proserpina (4c) 9”: Borsten SCHWATZz........... en ee ee. 10 10. Basalfelder der Borsten ebenfalls schwarz ................... H. vespertilio (4e) 10% BasalteldergerBorstenhelen. ee D. porcellus (4d) 11. Borsten lang, etwa halb so lang wie das Horn .…............................... 12 11°. ‚Borsten viel’kürzer 2.0. on oe ee Se EL ER 13 12. Brustbeine dunkel, Horn nur mit den üblichen 2 kräftigen Endborsten.... I Eee M. stellatarum (4a) 12°. Brustbeine hell, Horn mit 4 kräftigen Borsten ............ R. komarovi (4b) 13. Kopf und stärker sklerotisierte Teile einfarbig hell ; auf Euphorbiaceae ... DEN a AR ae MS RS cleus cetera tree Se Se H. nicaea (5a) 13. Kopf mit bräunlicher Scheckung ; stärker sklerotisierte Teile (z.B. Bauch- beinschienen) oft dunkel gerandet ; auf Zygophyllaceae ......................... ER EN a oe ore ea 0 ae lec lsondenoes H. zygophylli (Sb) 14. Nachschieber nach hinten spitz ausgezogen (vergl. Abb. 2a) ............. 15 14°. Nachschieberhintenaabgerundet ne RER cone eee eee eee eee 18 148 Abb. 4. a) M. stellatarum (24 X); b) R. komarovi (15 X, Foto Wasserthal) ; c) P proserpina (26 X) ; d) D. porcellus (19 X) ; e) H. vespertilio (23 X) ; f) A. gallii (23 X). 15. 15% 16. 16°. 17. 17. 18. 18’. 19. 19°. 20. 20°. 21. 217 22. 22% 23. 237. Horn wie mit schwarzbraunem Hagelzucker bestreut ; Prothorax ragt kapuzenartig über den Kopf hinaus ............................. M. quercus (2a) Ua Le) bs Bn eR ree Mere Mr See NEE 16 Horn über den größten Teil der Länge mittel- bis dunkelbraun. .............. Be ere eae Beer Ome eas eh aay ut Re S. ocellatus (2e) Horn von heller Grundfarbe... na ee 17 Horn relativ kurz (H/ A < 1,5), fast glatt ................... L. amurensis (2c) Horn lang (H/A > 1,5), durch viele kleine Borsten deutlich aufgerauht... ee Get TRG Orta dara een dr Sac dane ec ee aes N L. populi (2d) Körper mit langen, hellen, Y-förmigen Borsten (Abb. 3a-b) .............. 19 ohne sölche.Borsten 2a rn ee ee 20 Borsten auch auf Nackenschild und Kopf tief gegabelt ; auf Dipsaca- EDER en CC DNA H. croatica (3b) Nackenschild und Kopf mit einfachen oder nur ganz leicht gegabelten Borsten au Eapnifoliaeeae 2.2.0.2. ae H. fuciformis (3a) Körper mit dunklen Borsten auf ebenfalls dunklen Basalfeldern ........ 21 Basalfelder der.Borsten hell... mn. ee ee eee 22 auch Kopf mit schwarzen Borsten .......................... H. hippophaes (6a) Kopf mitzhellien’Borstener ee me A. convolvuli (1b) Horn extrem lang (H/A = 6 !), sehr beweglich ................ H. celerio (6c) Horn kürzer (H/A = 4) 2... ee RE 23 Raupe durch zusätzliche Kleinborsten (Sekundärborsten ; besonders im Gegenlicht gut zu sehen) sehr rauh wirkend ..................... M. tiliae (2b) nur mit einem einfachen Borstenmuster (vergl. Abb. Ib) ........................ ER AN EEE NE Bern Sane ene 24 und Hilfsschlüssel Anmerkung : Lebende Eilarven der folgenden 5 Arten sind bei Lupenvergröße- rung nur schwer zu bestimmen ; die Futterpflanze der Raupe gibt jedoch oft wichtige Hinweise. 24. 24’. 25. 257 26. 26’. 27: 272 28. 28°. 150 Horn relativ kurz und dick (H/ A = 2,5), unbeweglich ... D. elpenor (6b) Horn anger CH A= 4) ne ee nn EEE 25 Kopfborsten dunkler als Koplt 2.2... 2. see 00 eee eee eee 26 Kopfborsten so hell wie oder heller als Kopf 1... 2er 27 Homdunkelbraun auf Oleaceeass er re ee S. ligustri (1c) Horn schwarz ; auf Vitaceae (u.a.), nicht auf Oleaceae ...... T. alecto (6d) auf Apocynaceae ; Horn äußerst beweglich ....................... D. nerii (3d) nicht auf Apocynaceae ; Horn praktisch unbeweglich ......................- 28 auf Solanaceaeı Olcaceae (ua) er Fr en A. atropos (la) auf Vitaceae, Onageraceaelua Te ne ee 2 T. alecto (6d) Abb. 5. a) A. nicaea (16 X) ; b) H. zygophylli (18 X); c) H. lineata (24 X); d) H. centralasiae (19 X) ; e) H. euphorbiae (15 X). Hilfsschlüssel : 2. Raupenstadium Lebende Eilarven der Arten A. atropos, S. ligustri, D. nerü, D. elpenor und 7: alecto sind bei Lupenvergrößerung nur schwer zu bestimmen. Dieser Hilfsschlüssel verwendet daher zusätzlich Merkmale des zweiten Raupenstadiums. 1. Körper glatt, mit subdorsalen, paarigen Augenflecken ........................ 2 1°. Körper durch Chitinzapfen aufgerauht, ohne Augenflecken. ................ d 2. mit einem Augentlecks(Metathorax) RP D. nerii 2”. mit mindestens zwei subdorsalen Augenflecken .................................. 3 3. mit ZWELSUDIOTSAlEN Ansentlecken re... eee D. elpenor 3”. mit mindestens drei subdorsalen Augenflecken ....................... T. alecto 4. lateral mit deutlichen Diagonalstreifen (von vorn/unten nach hinten/ oben) 4.....0n. a ee. EEE S. ligustri 4. ohne. Diagonalstreifen 2... 0 A. atropos Dank Mein Dank gilt wiederum Herrn Prof. Dr. L.T. Wasserthal (Erlangen) für technische und wissenschaftliche Anregungen, Herrn Dr. E.A. Loeliger (Oegst- geest [NL]) für Lebendmaterial zahlreicher Arten, Frau I. Paas (Bochum) für die fotografischen Abzüge und Herrn E. de Bros (Binningen [CH]) für die französische Übersetzung der Zusammenfassung. Literatur Denso, P., 1906a. Lepidopterologisches von der Umgebung des Genfer Sees. Ent. Z. Guben 20 : 329-340. Denso, P., 1906b. Beiträge zur Kenntnis der Ontogenese europäischer Sphin- gidenraupen. Ent. Z. Guben 20 : 411-418, 463-470, 523-527 + Taf. 9-12. FREINA, J. J. DE, 1994. Über Biologie, Morphologie und Taxonomie von Hyles tithymali deserticola (Bartel), mit vergleichenden Studien zu Hyles euphorbiae mauretanica (Staudinger) (Lepidoptera : Sphingidae). Ent. Z. 104 : 33-41. GILLMER, M., 1904. Ein Beitrag zur Entwickelungsgeschichte von Phryxus livornica, Esp.. Ent. Z. Guben 18 : 116-123. Harsicu, H., 1978. Zur Biologie von Acherontia atropos (Lep. : Sphingidae). 1. Teil. Ent. Z. 88 : 29-36. Hernia, S., 1976. Nachzucht von Daphnis nerü (Lep., Sphingidae). Ent. Z. 86 : 25-30. Hernic, S., 1978. Zur Biologie des Großen Weinschwärmers (Hippotion celerio) (Lep., Sphingidae). Ent. Z. 88 : 53-62. Hermie, S., 1981. Einige Beobachtungen zur Biologie des Linienschwärmers (Hyles livornica Esper) (Lep. : Sphingidae). Ent. Z. 91 : 241247. 152 Abb. 6. a) H. hippophaes (17 X); b) D. elpenor (18 X); c) H. celerio (15 X); d) T. alecto (14 X). Heınıg, S. & HÄsLer, G., 1986. Angaben zur Biologie von Hemaris croatica (Lep. : Sphingidae). Ent. Z. 96 : 193-199. LEDERER, G., 1949. Ein Beitrag zur Biologie von Celerio hippophaes hippo- phaes (Esper 1789) mit einigen Bemerkungen über Celerio nicaea nicaea (Prunner 1798). Ent. Z. 59 : 65-70, 75-78, 87-88, 101-102. MEERMAN, J. C., 1987. De Nederlandse Pijlstaartvlinders (Lepidoptera : Sphin- gidae). Wetensch. Meded. Kon. Ned. natuurhist. Ver. 180 : 60 pp. PELZER, A., 1982. Zur Kenntnis der frühen Stände von Hyles centralasiae siehei (Püngeler) (Sphingidae). Nota lepid. 5 : 134-140. PELZER, A., 1988. Die Präimaginalstadien von Laothoe amurensis — ein Ver- gleich mit L. populi (Lepidoptera : Sphingidae). Nota lepid. 11 : 274-278. PELZER, A., 1991. Illustrierter Bestimmungsschlüssel für die Präimaginalstadien der Schwärmer Europas und Nordafrikas (Lepidoptera : Sphingidae). Teil I: Erwachsene Raupen. Nota lepid. 14 : 220-233. PITTAWAY, A. R., 1979. On Rethera komarovi manifica [sic] (Brandt) (Lepi- doptera : Sphingidae). Entomologist’s Gaz. 30 : 3-6 + Taf. 1. Pitraway, A. R., 1993. The hawkmoths of the Western Palaearctic. Harley Books, Colchester. Pourrton, E. B., 1885. Further notes upon the markings and attitudes of lepidopterous larvae, together with a complete account of the life-history of Sphinx ligustri and Selenia illunaria (larvae). Trans. R. Ent. Soc. Lond. 1885 : 281-329 + Taf. 7. Pourrton, E. B., 1886. Notes in 1885 upon lepidopterous larvæ and pupæ, including an account of the loss of weight in the freshly-formed lepido- pterous pupae, & c. Trans. R. Ent. Soc. Lond. 1886 : 137-179. Pourron, E. B., 1888. Notes in 1887 upon lepidopterous larvæ, & c., including a complete account of the life-history of the larvæ of Sphinx convolvuli and Aglia tau. Trans. R. Ent. Soc. Lond. 1888 : 515-606 + Taf. 15-17. ROTHSCHILD, W. & JoRDAN, K., 1903. A revision of the lepidopterous family Sphingidae. Novit. Zool. 9, Suppl. SOFFNER, J., 1959. Dolbina elegans O. [sic] Bang-Haas in Europa (Lep. Sphin- gidae). Ent. Z. 69 : 269-270. Voss, H. von, 1911. Die Entwicklung der Raupenzeichnung bei einigen Sphin- giden. Zool. Jb. Syst. 30 : 573-642 + Taf. 16-19. WASSERTHAL, L. T., 1970. Generalisierende und metrische Analyse des pri- mären Borstenmusters der Pterophoriden-Raupen (Lepidoptera). Z. Morph. Tiere 68 : 177-254. WEISMANN, A., 1876. Studien zur Descendenz-Theorie. II. Ueber die letzten Ursachen der Transmutationen. Leipzig : Engelmann. 154 Nota lepid. 17 (3/4) : 155-174 ; 30.1V.1995 ISSN 0342-7536 Wing pattern and allozyme relationships in the Coenonympha arcania group, emphasising the C. gardetta-darwiniana contact area at Bellwald, Switzerland (Lepidoptera, Satyrıdae) Adam H. PORTER (!), Robert W. SCHNEIDER & Brad A. PRICE Biology Department, Bowling Green University, Bowling Green, OH 43403-0212, USA Summary The butterflies Coenonympha gardetta, C. darwiniana and C. arcania are closely related and have parapatric distributions. We studied wing pattern and allozyme variation in three sample sites near Bellwald in Canton Valais, Switzerland in an area where C. gardetta and C. darwiniana meet, and in a C. arcania population from northern Italy. Principal component analysis identified traits that separated the C. arcania population, but separate taxo- nomic groups could not be distinguished in the Bellwald region even when C. arcania was dropped from the analysıs. Allozyme data showed high poly- morphism characteristic of other Coenonympha populations, and also sepa- rated the C. arcania population. F-statistics revealed that the sampled popu- lations at Bellwald, even though separated by up to 2 km, are probably sub- sites within a single large, demographic population. We believe C. gardetta and C. darwiniana should be considered conspecific ; C. gardetta is the older name. C. arcania should provisionally be kept distinct taxonomically, but closer study of contact regions between C. gardetta and C. arcania are required to rule out mere isolation by distance as the reason for the observed level of differentiation. Resume Coenonympha gardetta, C. darwiniana et C. arcania sont des espèces proches parentes dont la répartition est parapatrique. Les auteurs ont étudié le dessin des ailes et la variation allozyme de trois lots provenant des environs de Bellwald (Valais, Suisse), région où C. gardetta et C. darwiniana sont en contact, ainsi que chez une population de C. arcania d’Italie septentrionale. L'analyse des principaux éléments a révélé des caractères qui séparaient la population de C. arcania, mais on n’a pas du distinguer de groupes taxo- nomiques séparés dans les biotopes de Bellwald, même lorsque C. arcania (‘) Corresponding author : Adam Porter at the above address, or E-mail : aporter@opie.bgsu.edu ; Tel. : (419) 372-2691 ; Fax : (419) 372-0224. 153 était exclu de l’analyse. Les données allozymes ont révélé un polymorphisme considérable caractéristique d’autres populations de Coenonympha ; elles ont également séparé la population de C. arcania. Les statistiques F ont prouvé que les populations-échantillons de Bellwald, même séparées par une distance de 2 km, sont probablement des sous-stations à l’intérieur d’une seule et même grande population. Les auteurs pensent que C. gardetta et C. darwiniana devraient être considérées comme co-spécifiques ; C. gardetta est le plus ancien nom. C. arcania devrait être provisoirement conservé comme espèce taxono- miquement distincte. Mais on devrait étudier de plus près les zones de contact potentiel entre C. gardetta et C. arcania afin de pouvoir exclure une simple isolation par la distance pour expliquer le degré de différenciation qu’on a constaté. Introduction The butterfly taxa Coenonympha gardetta (de Prunner, 1798), C. dar- winiana Staudinger, 1871 and C. arcania (Linnaeus, 1761) are parapatric in the Alps, respectively occupying high (= 1800 m), middle (800- 2000 m), and low (= 1000 m) elevational bands. These taxa have long been seen as closely related (e.g. DAVENPORT, 1941). C. gardetta and C. darwiniana have contact areas where they are reported to intergrade in the southern Alps (Lepidopterologische Arbeitsgruppe der Schweiz, 1987), leading to questions about their taxonomic status as separate species. Because C. arcania is also parapatric, and darwiniana is some- times listed in its synonomy (e.g. FORSTER & WOHLFAHRT, 1976), it is legitimate to question the relationships among all three taxa. In this study, we examine three populations at the contact zone be- tween Coenonympha gardetta and C. darwiniana, and a geographically distant population of C. arcania. If there is indeed partial (or complete) genetic isolation between these taxa, as the current taxonomy suggests, then our contact populations should contain an excess of “pure” forms of each taxon, and few intermediates. Upon closer statistical exami- nation, suites of diagnostic wing pattern and/or allozyme traits would appear to be correlated within individuals in these populations. How- ever, if the major diagnostic differences between these taxa are produced by environmental conditions operating on a common genotypic array rather than by genetic differences maintained by “reproductive” barriers, then there should be no such correlations within contact populations. A similar lack of correlation would result in contact areas if hybrids between immigrants from taxonomically differentiated regional popu- lations were not at a selective disadvantage relative to “pure” individuals. Furthermore, for traits inherited in a co-dominant Mendelian fashion 156 (allozymes in this study), such correlations can also be expressed in the form of F-statistics (WRIGHT, 1969). These have the advantage of permitting us to enlist the analytical power of evolutionary theory to make inferences about underlying populational processes, in particular gene exchange among populations. Of course, it is gene exchange be- tween putative taxonomic groupings that we wish to infer in the process of making taxonomic decisions. Taxonomists have traditionally done this “by eye” and therefore less reliably, especially for the traits whose genetic bases are unknown. Methods Mixed populations of Coenonympha gardetta and C. darwiniana were sampled in July 1991 in the vicinity of Bellwald, Switzerland, on the north slope of Canton Valais in the Rhone Valley. Population I was collected on a steep, SW-facing slope at 1700m in a meadow under an open-canopy fir forest on 13.vu.1991. Population II was sampled also at 1700 m, but 1 km N on a NW-facing, colder slope, locally above treeline, on 20.vii.1991. Population III was at 2000 m at the top of the ski lift, above the treeline in open meadow, on 21.vii.1991. The C. arcania population was collected at 1400 m from Monte Motta- rone, near Streza, Italy, on 23.vu.1991. Individuals were haphazardly netted and stored alive under refrigeration until they could be frozen at -80C. Wing pattern morphometrics The taxonomic literature (DAVENPORT, 1941 ; HicGins & HARGREAVES, 1991; Lepidopterologische Arbeitsgruppe der Schweiz, 1987) was consulted to determine the wing pattern elements previously used to distinguish the three taxa, especially between the more similar taxa, gardetta and darwiniana. The characters proved to be mainly the size and location (relative to the wing margin) of the eyespots in the distal wing cells of the ventral hindwing, and the extent and location of the white band proximal to these spots. Spots near the apex of the ventral forewing have also been used. Wings were removed and stored separately when specimens were pre- pared for electrophoresis (described below). We measured the following characters for each of eight ventral hindwing cells, along the axis of the cell : the diameter of the black center of the eyespot (absent = 0), the distance from the center of the eyespot to the wing margin (absent = unscored), the width of the white band measured to the edge 157 of the eyespot halo (absent = 0), and the edge of the white band to the edge of the wing (absent = unscored). We also measured the diameter of the forewing eyespots (absent = 0). We did not record from the outer wing spot rings because yellow outer rings could not be consistently distinguished when the ground colour was pale ; our scoring system thus regarded any all-yellow spots as being absent. Wing length was measured as an index of body size, and gender was recorded. The left wings were used except when one or more characters was missing due to damage, whereupon the right wings were used. Measure- ments were made at 20x magnification on a colour video monitor using a computerised image-analysıs system. A Wild® microscope was fitted with a Sony® video camera ; this was connected to IBM PC® computer operated using the image-analysis program Optimas® (v. 3.01, BioScan, Inc.). Data collection was mechanised using a macro written in the Optimas procedural language, and measurements were saved directly to a file. We avoided characters involving colour because they were not amenable to accurate measurement using this software. We analyzed the wing patterns using principal component analysis. This method condenses the large number of measurements per individual into a more manageable number of statistically independent characters, and is justifiable both statistically and biologically. The premise, statistically, is that some characters are likely to be correlated, where- upon they carry redundant information and should be weighted to take this into account. For example, large individuals are likely have larger measurements, and we should factor out the body size differences before we attempt to consider relative eyespot size differences. Bio- logically, the premise is that if the taxa are genetically isolated, then their wing patterns will have evolved independently in the separate lineages, and different pattern elements will be correlated, within lineages. The elements that are correlated would form the set of dia- gnostic characters useful for distinguishing the lineages. If genetic isolation were indeed involved, a small number of principal components would contain all the correlated diagnostic characters and describe most of the overall wing pattern variation among individuals, even within contact areas. But if the taxa were freely interbreeding, then characters in contact areas would tend to be assorted independently among individuals ; they would be uncorrelated. Statistically, this would be indicated if the overall wing pattern variation were spread among a larger number of principal components, evidence of independence among the wing pattern elements, and by unimodal variation along principal component axes. 158 Principal components analysis is, in its philosophy, what the experienced taxonomists of older generations did “by eye”. The advantages of the statistical approach are three. Firstly, it is explicit, thus repeatable by others and carries known data limitations. Secondly, it is grounded in statistical theory so it takes sample size into account in a way that cannot be done properly otherwise. Finally, it can pick out much more subtle patterns than can the eye of an experienced worker, and con- versely, it can demonstrate that some patterns perceived by less ex- perienced eyes are fantasy. However, both approaches depend on the ability to pick out the “right” characters to measure, those that will give the best discrimination ; this is why, for characters in this study, we relied on the literature for the acknowledged expertise of previous workers. We used Systat® (v.5.1, Systat Inc.) software on a Macintosh computer for the principal components analysis. We used a Pearson’s r correlation matrix of the 28 primary measurements/individual to generate all principal components (PCs) ; gender and body size were included in these analyses as controls. The statistics require data sets without missing entries, so we were forced to omit the measures of spot or white line location for some wing cells if any individuals were absent that trait (the rejected alternative was to eliminate those individuals). We examined the loadings of characters onto each PC for suites of diagnostic characters. The proportion of the total wing pattern variance explained by each PC was used as an estimate of the overall in- dependence of traits. We then analyzed the corresponding PC scores of each individual (1.e., the “measurement” of the individual along the PC axis, produced as a weighted combination of the original measure- ments in that individual) using one-way ANOVAs for to find significant variation among populations, using SuperANOVA® (v4.0; Abacus Concepts, Inc.) software on a Macintosh computer. The entire analysis was repeated without the C. arcania population to look more closely at differentiation in the Bellwald region ; here we could use a larger data set because fewer individuals were absent the eyespots or white lines in cells. Electrophoresis Horizontal starch gel electrophoresis was performed on head and thorax tissue using standard methods described elsewhere (PORTER & GEIGER, 1988; PORTER & MATOON, 1989). We scored 19 putative genetic loci: alcohol dehydrogenase (ADH; enzyme commission number 1.1.1.1), adenylate kinase (AK-1 ; 2.7.4.7), aldolase (ALDO ; 159 4.1.2.13), esterase (EST-1 ; 3.1.1.1), fumarase (FUM ; 4.2.1.2), glutamic- oxaloacetic transaminase (GOT-1, GOT-2 ; 2.6.1.1), glyceraldehyde-3- phosphate dehydrogenase (GAPDH; 1.2.1.12), a-glycerophosphate dehydrogenase (aGPDH ; 1.1.1.8), hexokinase (HK ; 2.7.1.1), isocitric dehydrogenase (IDH-1; 1.1.1.42), lactic dehydrogenase (LDH; 1.1.1.27), malic dehydrogenase (MDH-1, MDH-2; 1.1.1.37) , malic enzyme (ME-1, ME-2 ; 1.1.1.40), peptidase (PEP-1 ; 3.4.1.1), phospho- glucomutase (PGM ; 2.7.5.1) and phosphoglucose isomerase (PGI ; 38.1.9)! We calculated standard statistics describing the extent of allozyme varia- bility within populations. These include the mean number of alleles observed per locus, the mean heterozygosity observed per locus (H,,,) and that expected based on Hardy-Weinberg expectation (H,,,), and the percent of sampled loci that were polymorphic in the population (%P). We describe variation among populations in two ways. First, we cal- culated Nei’s unbiased genetic distance and produced a summary phenogram using UPGMA. This method is probably most familiar and permits comparison across a wide range of taxa. We also estimated Fsr (WRIGHT, 1969) among the Bellwald populations using WEIR & COCKERHAM’S (1984) method which accounts for sampling variation. We used weighted averaging over alleles and loci, and jackknifed over loci for the error estimates. Our estimates were interpreted using the relationship M = (1/Fsr-1)/4, where M is a gene flow parameter de- scribing the effective number of individuals moving among popula- tions each generation (COCKERHAM & WEIR, 1993). A fundamental result in theoretical population genetics is that when M > 0.5, then gene flow produces substantial genetic similarity among populations at neutral loci (WRIGHT, 1931). We will not provide the theoretical and statistical details here ; interested readers may consult population genetic texts (e.g. HARTL & CLARK, 1989) for introductory concepts, CocKERHAM & WEIR (1993) and references therein for current statistical theory, and PORTER (1990) and PORTER & GEIGER (1988 ; 1995) for examples of applications to butterfly populations. Results Wing pattern morphometrics Coefficients of variation (c.v.) for each trait are shown for each popu- lation and for all populations combined (Table 1). A c.v. of 0.1 means that the standard deviation is 10% of the mean for the trait, a reasonably 160 Table 1 Coefficients of variation for the ventral wing pattern characters used in this study. Missing values indicate the trait was not present in the population character Mn HR Bellwald Mt. total III Mottarone 17 87 reine length forewing spot diameter l 2 hindwing spot diameter NnBWN— ite line width 5 W spot location l 2 3 4 5 6 7 8 P l 2 3 4 5 6 white line location O0 I Un BB © D = high level of variability. Variability is high in the wing pattern traits taxonomists have identified as important, being generally higher than c.v. — 0.2, and is comparable among populations. Some traits, par- ticularly those present in only a few individuals, showed c.v. > 1, an extreme level of variability. This included pattern elements in hindwing cells 7 and 8, and in the forewing spots. 161 First consider variation in all populations. Eigenvalues and the propor- tion of total variance explained by the first ten PCs are in Table 2. The character loadings of the first six PCs are given in Table 3. PC 1 loaded highly for most characters, especially for forewing length, and we interpret it as a general body size character. Note that several traits did not load here, indicating that these varied relatively independently of body size (this points out the hazards of an alternative approach : dividing all measurements by body size for standardisation before statistical analysis). These include the spot diameter in cell 6 and the white line width in cell 1. PC 2 describes an inverse ratio of spot diameter and white line width, and is largely independent of body size. PC 3 mainly describes the shape of the white line as it traverses the cells, with loadings being positive in the first cells and negative in the later cells ; the spot diameters in cells 2 and 6 and the width in cell 2 also load here. This PC captures a previously reported taxonomic difference between arcania, with a line that narrows posteriorly, and the others, with a line of relatively constant width. PCs 4 and 5 describe subtle relationships between spot diameters, white line widths and their locations in several cells. PC 6 describes sexual dimorphism, and characters that also load here include the spot diameter of cell 6 and the white line width in cell 8. Table 2 Principal component analysis with all population included. Eigenvalues of the first ten PCs and the percent of the total variance explained PC Eigenvalue % variance explained © \O 00 I ON Un BR © D — Figure 1 shows differentiation among populations in PC scores. We found significant interpopulational differences in PC 1 (ANOVA; F37 = 14.958; P < 0.0001), with two groups segregating in the followup test (Duncan’s New Multiple Range Test, P < 0.05): the 162 Table 3 Character loadings on the first six principle components using all populations. Numbers for traits refer to wing cells. Loadings are the extent to which a character is correlated with the PC gender forewing length forewing spot diameter 2 hindwing spot diameter OL PRWND = white line width l 2 3 4 5 6 7 8 indwing spot location 3 white line location D 3 Mottarone and Bellwald II population both had larger mean body sizes than the other populations (Figure 1a). The Mottarone population (arcania) differed from the others in PC 3 (ANOVA ; F3 7) = 17.164 ; P < 0.0001), indicating a difference in white line shape between these groups (Figure 1b). PC 6 also showed significance (ANOVA ; F37 = 3.233 ; P= 0.027), but we attribute this to a difference in sex ratios among populations. These differences demonstrate that the arcania population is phenotypically different from the remaining po- pulations, and by themselves, suggest a possibility of genetic isolation. Means of PCs 2, 4 & 5 were not significantly different among po- pulations, and we consider them to represent patterns of variation common to all populations. The differences between arcania vs. Bellwald populations could poten- tially obscure more subtle differentiation between gardetta and dar- winiana in the contact zone, so it is appropriate to reanalyze the data dropping the Mottarone population. 163 1.5 PC 1 -1 Bellwald | Bellwald Il Bellwald Ill Mt. Mottarone 1.5 Bellwald | Bellwald Il Bellwald Ill Mt. Mottarone Fig. 1. Principal component means (95% c.1.) showing differences among populations. (a) PC 1, describing body size ; (b) PC 3, describing the shape of the white line. 164 Table 4 Principal component analysis with only Bellwald populations. Eigenvalues of the first ten PCs and the percent of the total variance explained PC Eigenvalue % variance explained © \O 00 I A Un BR © ND — Here we concentrate on the gardetta/darwiniana sampling sites at Bellwald, excluding arcania. Eigenvalues and the proportion of total variance explained by the first ten PCs are in Table 4. The character loadings of the first six PCs are given in Table 5; it is important to remember that these PCs describe different combinations of characters than those in the previous analysis. PC 1 again loaded highly for most characters, especially for forewing length, and we interpret it again as a general body size character. The forewing spot diameter and the several of the white line widths vary little with body size, nor is there sexual size dimorphism. PC 2 shows strong negative loadings for white line width and weak positive loadings for spot diameters. We interpret it as a white line width parameter that is in relative agreement with a diagnostic character often used to separate gardetta and darwiniana. PC 3 describes a differentiation pattern in inter-cell ratios of spot and white line sizes — at one extreme are darwiniana-like individuals with the middle cells having relatively smaller spots and extreme cells having wider lines, at the other are individuals with the reverse. PC 4 describes sexual dimorphism, with hindwing spot 6 being larger in males and the white band slightly more centrally located in females. PCs 5 & 6 describe mainly relationships among spot sizes. We found significant differences among populations only in PC 4 (ANOVA ; F357 = 5.093 ; P = 0.009), and this was attributed to the different sex ratios of these population samples. The difference in body size from the previous analysis was not quite significant here (ANOVA ; F; 57 = 2.736 ; P = 0.073). Distributions for all three Bellwald popu- lations along PC axes 2 & 3 are shown in Figure 2 ; these axes describe 165 Table 5 Loadings for the first six PCs for the gardetta-darwiniana contact area only gender forewing length forewing spot diameter 2 hindwing spot diameter OU B © N = white line width O0 I ON Un BB © ND = 5 = ndwing spot location U 8 © ND white line location Un © ND most of the gardetta-darwiniana diagnostic characters. Despite the numerous individuals that appear to have gardetta and darwiniana characteristics, there is unimodal variation and no evidence from wing pattern that these taxa are genetically isolated. Allozyme differentiation Allelic frequencies for variable loci are given in Table 6. The loci EST-1, GAPDH and aGPDH were monomorphic. Summary statistics de- scribing genetic variability within populations is given in Table 7. Varia- bility is remarkably high, a common phenomenon in butterfly popu- lations, including Coenonympha (PORTER & GEIGER, 1988 ; PORTER & MAToon, 1989), and indicates that these populations are quite large and have not been through population bottlenecks in their recent past. 166 darwiniana om O jae 3 D QG O 5 -4 -3 -2 -1 0 1 2 ardetta darwiniana @ PC 2 Fig. 2. Distributions of individuals from the Bellwald populations along PC axes 2 & 3 are unimodal. 1 — from Bellwald I; 2 — Bellwald II; 3 — Bellwald III. PCs are standardised to have their means at zero and scale in units of standard deviations. The populations do not differ significantly along either axis. Differentiation among populations is described graphically using a distance Wagner procedure (FARRIS, 1972) on a matrix of pairwise RoGers’ (1972) genetic distances (Figure 3a), and using UPGMA cluster analysis on a matrix of Ner’s (1978) unbiased genetic identities (Figure 3b). The Mottarone arcania population segregates from the others, at a level suggesting that it may be genetically isolated. The remainder cluster closely, at levels commonly observed among con- specific populations in other butterfly taxa. The high similarity within the Bellwald group permits closer popu- lational analyses using F-statistics. Within the Bellwald population group, we obtained an estimate of 9 (= sr) = - 0.00223 + 0.00127 (s.d.), averaged over alleles and loci. This slightly negative value is a result of sampling error and shows that 0 is not significantly different from zero, indicating that there is no appreciable genetic differentiation among sites. This is perhaps more easily understood when translated to a gene flow estimate (M) using M = (1/0 — 1)/4, yielding a value of 942 < M S panmictic (95% c.i.). This large number of individuals exchanged among sites each generation suggests that the sampled “populations” are effectively sub-sites within a single, larger panmictic population in the Bellwald area. 167 Table 6 Allele frequencies for variable loci, by population. Sample sizes for each locus in brackets Monte Mottarone Bellwald I Bellwald II Bellwald III Aw>a =, a ®) m w>S-mangaw = N >r ed O > 0% m = œ en Su>S Sr>Suw> ! — A B ME A B ME-2 A B Ow> 5 de — < amyUmerarcrmammyaws Mt. Mottarone Bellwald I Bellwald Il Bellwald Ill D Mt. Mottarone = Bellwald | Bellwald Ill Bellwald II Te el .875 9 .925 .95 975 1 Nei's 1978 unbiased genetic identity Fig. 3. Genetic differentiation among populations, described by (a) a distance Wagner procedure (Farris, 1972) using RoGERs’ (1972) genetic distance, rooted using the mid- point of the longest path ; (b) UPGMA using Nei’s unbiased genetic identity. Table 7 Genetic variability within populations (s.e.). A : mean alleles per locus ; %P : percent of loci that were polymorphic ; H,,; : the observed proportion of heterozygotes, averaged over loci ; H.x, : the proportion of heterozygotes expected from Hardy-Weinberg genotypic proportions, averaged over loci. Standard errors calculated using a jackknife procedure, over loci population Monte 2.27 (0.02) 60.0 (0.9) 0.229 (0.005) | 0.199 (0.004) Mottarone Bellwald I 2.33 (0.03) 60.0 (0.9) | 0.187 (0.004) | 0.207 (0.004) Bellwald II 2.00 (0.02) 53.3 (0.9) | 0.213 (0.004) | 0.212 (0.004) Bellwald II | 2.47 (0.03) 60.0 (0.9) | 0.192 (0.004) | 0.213 (0.004) 169 Discussion Both morphological and allozyme analyses show that the Monte Mot- tarone C. arcania population is different from the others, but that continuous variation exists between C. gardetta and C. darwiniana at Bellwald. The morphometric analyses isolated essentially the same suites of “diagnostic” character states that taxonomists have used, showing that the results are not spurious as has been suggested in similar studies (HAMMOND, 1985). Neither is the high genetic similarity a spurious result of low variability : genetic variability is high enough that if two species have been coexisting at Bellwald, then genetic drift should have already differentiated the populations (see PORTER & GEIGER, 1995 for a discussion of this effect). There is no evidence to suggest that these are distinct species at Bellwald. The C. gardetta-darwiniana contact region is similar to contact regions between C. tullia (Müller, 1764) taxa in western North America. C. tullia-group taxa in California, Nevada and Oregon differ in eyespot size and number, dorsal and ventral ground colour and the shape and placement of the white lines, and were distributed among several no- minal species. However, populations are highly variable, and taxono- mists, with the notable exception of DAVENPORT (1941), had tended concentrate on widely separated localities, ignoring contact areas and intrapopulational variability. When contact areas were examined, and allozyme data used as corroborating evidence, it was found that wing patterns intergraded, gene flow was high, and no genetic isolation was apparent (PORTER & GEIGER, 1988). Instead, the geographic variation in wing pattern was attributed to unknown selective and/or historical factors, being maintained in the face of strong gene flow between the taxonomically recognised forms. In one remarkable example, a popu- lation on coastal dunes was sharply differentiated in wing pattern from one only 8 km away on a hilltop, yet the allozyme data could only be reasonably interpreted as showing evidence of strong gene flow between them (PORTER & MATOON, 1989). The allozyme data indicate that the C. gardetta-darwiniana population at Bellwald is probably quite large and demographically continuous over an area of several km?. This area encompasses the respectively “typical” subalpine and alpine habitats of C. darwiniana and C. gar- detta, and is consistent with the interpretation that no genetic isola- tion exists between them. The implication is that individuals are quite mobile, readily moving distances of 2 km or more in their lifetimes. Whereas this should obviously be verified with a demographic study, we point out that marked C. tullia california Westwood, 1851 in Cali- 170 fornia have been recaptured at distances >1 km (WEISSMAN, 1972), and stray individuals have been seen in the Central Valley (Davis area) at least 40 km from potential source populations (SHAPIRO, 1982 ; PorTER, pers. obs.). We believe C. gardetta and C. darwiniana should be considered con- specific and the younger name, darwiniana, be placed in synonymy. This decision is based on the high genetic and phenotypic variability in the contact zone without apparent isolation, and remains subjective pending verification by closer demographic studies there. However, the patterns are strongly suggestive, and the onus now properly belongs on a splitter to demonstrate that C. gardetta and C. darwiniana are genetically isolated, rather than being extremes of a cline as we be- lieve they are. Indeed, the high gene flow estimate would argue that darwiniana not even be used as a subspecies name, because indivi- duals from populations of “darwiniana” wing phenotype are likely to be much more closely related in pedigree to those from “gardetta” populations directly upslope than to butterflies from populations with “darwiniana” phenotypes at more distant localities. A minor difference between C. gardetta and darwiniana in the male genitalia has been identified (Hıscıns, 1975), but DAVENPORT (1941) found the genitalia to be quite variable. Regardless of the validity of this putative difference, small differences in the genitalia per se have little value as evidence of reproductive isolation between parapatric taxa (PORTER & SHAPIRO, 1990). The possibility also remains that C. arcania and C. gardetta (+ dar- winiana) are conspecific, as we have only studied widely separated populations and these taxa too may intergrade in their contact areas. However, both the allozyme and wing pattern differentiation in our limited data are consistent with differences between closely related sympatric taxa, and we believe their continued designation as species — the status quo — is appropriate pending publication of studies done in contact areas. Anticipated results from a current study of the C. arcania group, on a larger geographic scale than ours (WIEMERS, 1994), will bring us much closer to the solution of this problem. Systematists of Coenonympha have long relied heavily on qualitatively described wing pattern dimensions as characters, even while lamenting their intra- and interpopulational variability (DAVENPORT, 1941). Species-level taxonomic decisions in Coenonympha could be greatly illuminated if we understood more about how wing pattern is develop- mentally, genetically and ecologically controlled, because it would help us determine the extent to which character variation could be used 171 as markers for more profound genetic differentiation. Studies of the wing patterns of Coenonympha indicate that several taxa have clines in spot pattern, with spot size diminishing with increasing elevation (BRUNTON et al., 1991) or latitude (K. PORTER, 1980 ; DENNIS et al., 1986) ; these patterns also appear to occur in North American C. tullia but have not yet been studied quantitatively. Though the eyespots of the Nymphalidae are probably serially homologous (NisHout, 1991), multivariate statistical analyses have demonstrated that the pattern elements in different wing cells are largely developmentally independent (NuHouT, 1985 ; Dennis et al., 1984). Dennis et al., (1986) proposed that geographic variation in spot size in C. tullia is attributable to parallel variation in selection pressures imposed by different suites of predators or predator abilities relative to gross habitat type. The alter- native that one might infer from clinal variation, that spot pattern is not genetically controlled at all, seems unlikely given the weaknes- ses of correlations of spot sizes within individuals and the heritability of spot pattern in other Satyrinae (BRAKEFIELD & NOoORDWIK, 1985). It would also be difficult to credibly explain, without invoking adap- tive genetic differences, why spot size gets smaller with increasing ele- vation in C. gardetta, but larger with increasing elevation in C. corinna and North American C. tullia, even though the ecological causes of these putative adaptations remain elusive. This is not to exclude an important role for environmental effects : polyphenism is apparent in several satyrine taxa, e.g. Bicyclus (BRAKEFIELD & REITSMA, 1991) and C. tullia (WEISSMAN, 1972), indicating the presence of genotype by environment interactions in the control of wing pattern. Though C. arcania and C. gardetta usually have only a single annual generation, similar underlying genetic mechanisms could well be influencing their wing patterns. Unfortunately, there are substantial technical difficulties to surmount before these issues can be quantitatively studied in Coeno- nympha and, especially, before they can be used to truly test assumptions underlying species-level taxonomic decisions. Acknowledgements Special thanks go to J. Buggmann, in whose house AP stayed in Bellwald, and H., I. & M. Geiger-Buggmann, the gracious hosts. The electrophoresis was supported by the Zoologisches Museum in Ziirich. M. Wiemers provided a copy of his excellent thesis ; he and an anonymous referee provided useful comments on a previous draft of this paper. 11762 Literature BRAKEFIELD, P. M. & RerrsMA, N., 1991. Phenotypic plasticity, seasonal climate, and the population biology of Bicyclus butterflies (Satyridae) in Malawi. Ecol. Ent. 16 : 291-303. BRAKEFIELD, P. M. & Noorpwuk, A. J. van, 1985. The genetics of spot pattern characters in the meadow brown butterfly Maniola jurtina (Lepidoptera : Satyrinae). Heredity 54 : 275-284. Brunton, C. F. A., BAXTER, J. D., QUARTON, J. A. S. & PANCHEN, A. L., 1991. Altitude-dependent variation in wing pattern in the Corsican butterfly Coenonympha corinna Hiibner (Satyridae). Biol. J. Linn. Soc. 42 : 367-378. COCKERHAM, C. C. & Weir, B. S., 1993. Estimation of gene flow from F-statistics. Evolution 47 : 855-863. Davenport, D., 1941. The butterflies of the satyrid genus Coenonympha. Bull. Mus. Comp. Zool., Harvard 87 : 215-349, 10 plates. Dennis, R. L. H., PORTER, K. & WırLıams, W. R., 1984. Ocellation in Coenonympha tullia (Müller) (Lepidoptera : Satyridae) I. Structures in correlation matrices. Nota lepid. 7 : 199-219. Dennis, R. L. H., PORTER, K. & WiLLıams, W. R., 1986. Ocellation in Coeno- nympha tullia (Müller) (Lepidoptera : Satyridae). II. Population diffe- rentiation and clinal variation in the context of climatically induced anti- predator defence strategies. Entomologist’s Gaz. 37 : 133-172. Farris, J. S., 1972. Estimating phylogenetic trees from distance data. Am. Nat. 106 : 645-668. FORSTER, W. & WOHLFAHRT, T. A., 1976. Die Schmetterlinge Mitteleuropas. Tagfalter Diurna (Rhopalocera und Hesperiidae). Franckh’sche Verlags- handlung. Stuttgart. Hammonpn, P. C., 1985. Rebuttal to the Arnold classification of Speyeria callippe (Nymphalidae) and defense of the subspecies concept. J. Res. Lep. 24 : 197-208. HARTL, D. L. & CLARK, A. G., 1989. Principles of Population Genetics. Sinhauer, Sunderland, Massachusetts. H1GGins, L. G., 1975. The Classification of European Butterflies. Collins, London. HiGGins, L. G. & HARGREAVES, B., 1991. The Butterflies of Britain and Europe. 3rd Edn. Collins, London. Lepidopterologische Arbeitsgruppe der Schweiz, 1987. Tagfalter und ihre Lebensräume. Schweitzerischer Bund für Naturschutz, Basel und Foto- rotar AG, Egg/ Zürich 516 pp. Net, M., 1978. Estimation of average heterozygosity and genetic distance from a small number of individuals. Genetics 89 : 583-590. NuHoUT, H. F., 1985. Independent development of homologous pattern elements in the wing patterns of butterflies. Dev. Bio. 108 : 146-151. NuHoOUT, F. H., 1991. The Development and Evolution of Butterfly Wing Patterns. Smithsonian Institution Press, Washington. 173 PORTER, A. H., 1990. Testing nominal species boundaries using gene flow statistics : the taxonomy of two hybridizing Admiral Butterflies (Lime- nitis : Nymphalidae). Syst. Zool. 39 : 131-148. PORTER, A. H. & GEIGER, H., 1988. Genetic and phenotypic population structure of the Coenonympha tullia complex (Lepidoptera : Nympha- lidae : Satyrinae) in California : no evidence for species boundaries. Can. J. Zool. 66 : 2751-2765. PORTER, A. H. & GEIGER, H., 1995. Limitations to the inference of gene flow at regional geographic scales — an example from the Pieris napi group (Lepidoptera : Pieridae) in Europe. Biol. J. Linn. Soc. (in press). PORTER, A. H. & Marroon, S. O., 1989. A new subspecies of Coenonympha tullia (Müller) (Nymphalidae : Satyrinae) confined to the coastal dunes of northern California. J. lepidopt. Soc. 43 : 229-238. PORTER, A. H. & SHAPIRo, A. M., 1990. Lock and key hypothesis : lack of mechanical isolation in a butterfly hybrid zone. Ann. entomol. Soc. Amer. 83 : 107-114. Porter, K., 1980. A quantitative treatment of clinal variation in Coenonympha tullia (Müller) (Lep., Satyridae). Entomol. Mon. Mag. 11: 71-82. ROGERS, J. S., 1972. Measures of genetic similarity and genetic distance. Studies in Genetics, Univ. of Texas Publ. No. 7213 7 : 145-153. SHAPIRO, A. M., 1982. Within-range butterfly dispersal : an urban garden as a detector. Atala 8 : 46-49. Weir, B. S. & CockERHAM, C. C., 1984. Estimating F-statistics for the analysis of population structure. Evolution 38 : 1358-1370 WEISSMAN, K. G., 1972. The population biology of Coenonympha tullia on Jasper Ridge : strategies of a bivoltine butterfly. Ph. D. Thesis, Stanford University. WIEMERS, M., 1994. Differenzierungsmuster bei Artbildungsprozessen : Mor- phologisch-biometrische und enzymelektrophoretische Untersuchungen am Coenonympha arcania (Linnaeus, 1761) Superspezies-Komplex. Diplomarbeit, Universitat Bonn. WRIGHT, S., 1931. Evolution in Mendelian populations. Genetics 16 : 97-159. WRIGHT, S., 1969. Evolution and the Genetics of Populations. Vol. 2. The Theory of Gene Frequencies. University of Chicago Press, Chicago. 174 Nota lepid. 17 (3/4) : 175-200 ; 30.1V.1995 ISSN 0342-7536 Inter-island variation in the butterfly Hipparchia (Pseudotergumia) wyssii (Christ, 1889) (Lepidoptera, Satyrinae) in the Canary Islands David A. S. Smirx* & Denis F. OWEN** * Natural History Museum, Eton College, Windsor, Berkshire SL4 6EW, England ** School of Biological and Molecular Sciences, Oxford Brookes University, Headington, Oxford OX3 OBP, England (!) Summary Samples of the endemic Canary grayling butterfly, Hipparchia (Pseudoter- gumia) wyssü (Christ, 1889), were obtained from all five of the Canary Islands where it occurs. Each island population comprises a distinct subspecies but the differences between them are quantitative rather than qualitative ; hence a system 1s devised by which elements of the wing pattern are scored to permit quantitative analysis. The results demonstrate significant inter-island differences in Wing size and wing pattern. The underside of the hindwing shows the greatest degree of inter-island variation. This is the only wing surface that is always visible in a resting butterfly ; its coloration is highly cryptic and it is suggested that the pattern was evolved in response to selection by predators long before H. wyssü or its ancestors reached the Canaries. Subsequent evolution of the details of the wing pattern differed from island to island because each island population was probably founded by few individuals with only a fraction of the genetic diversity of the species. It is postulated that the basic “grayling” wing pattern is determined by natural selection, but the precise expression of this pattern on each island is circumscribed by the limited gene pool of the original founders. Resume Des exemplaires du Satyride Hipparchia (Pseudotergumia) wyssii (Christ, 1889), endémique des Iles Canaries, ont été récoltés dans les cinq îles de cet archipel où ils existent. Les populations de chaque île forment une sous-espèce distincte, mais les différences entre celles-ci sont plutôt quantitatives que qualitatives. On a donc établi un système basé sur certains éléments précis des ailes permettant une analyse quantitative. Les résultats ainsi obtenus prou- vent l’existence de différences significatives entre les populations de chaque (1) Address for correspondence. 13 île en ce qui concerne la taille et le dessin des ailes. Le dessous des ailes posté- rieures est l'élément qui présente le plus fort degré de variation d’une île a l’autre. C’est en fait la seule partie de la surface des ailes qui est visible lorsque le papillon se pose. Son homochromie extrême est un élément de camouflage par mimétisme. On peut envisager que ce dessin a évolué du fait de la sélection par les prédateurs longtemps avant que A. wyssii ou ses ancêtres aient atteint les Canaries. L'évolution ultérieure des détails dans ce dessin des ailes fut différente d’une île à l’autre parce que la population de chaque île avait pro- bablement été fondée par un petit nombre d’individus qui n'étaient chacun porteurs que d’une fraction de la diversité génétique de l’espèce. On peut donc penser que si le dessin typique «Hipparchia» du dessous des ailes postérieures est déterminé par sélection naturelle, la réalisation concrète de ce dessin sur chaque île se trouve différenciée du fait du stock de gènes incomplet des premiers arrivés «fondateurs». Zusammenfassung Der Augenfalter Hipparchia (Pseudotergumia) wyssii (Christ, 1889) ist auf den Kanaren endemisch. Untersuchungen an Material von allen fünf Inseln des Archipels, auf denen die Art vorkommt, ergaben, daß jede Inselpopulation eine eigenständige Unterart darstellt, die sich aber eher quantitativ als qualitativ von den anderen unterscheidet. Daher wurde ein System entwickelt, das durch die Untersuchung von Flügel-Zeichnungselementen eine quantitative Analyse erlaubt. Die Ergebnisse zeigen signifikante Unterschiede in Flügelgröße und Flügelzeichnung zwischen den Inselpopulationen. Die größte Variabilität weist die Unterseite des Hinterflügels auf. Dies ist die einzige Flügelfläche, die an einem ruhenden Falter stets sichtbar ist. Sie zeigt eine ausgeprägte Tarnfarbe, deren Muster vermutlich durch Feinddruck selektiert worden ist, bevor A. wyssii oder ihre Vorfahren die Kanaren erreichten. Die spätere Entwicklung der Details der Flügelzeichnung variiert von Insel zu Insel, weil jede Inselpopulation wahrscheinlich nur von wenigen Individuen -mit nur einen Bruchteil der gene- tischen Vielfalt der Art -begründet wurde. Es wird postuliert, daß das grund- legende Hipparchia-Flügelmuster durch natürliche Selektion bestimmt wird, aber die genaue Ausprägung dieses Musters auf jeder Insel vom begrenzten Genpool der ursprünglichen Gründer abhängt. Introduction The Canary grayling butterfly, Hipparchia (Pseudotergumia) wyssii (Christ, 1889) (Lepidoptera, Satyrinae) is endemic to the five Canary Islands of Tenerife, La Gomera, Gran Canaria, La Palma and El Hierro (Fig. 1). Until recently little was known of this species and even its presence on one island, La Palma, was in doubt (LEESTMANS, 1975) ; however a comprehensive review which followed several collecting trips (WIEMERS, 1991) has established that the butterfly is relatively common 176 0 Lanzarote 3 Canary Islands Tenerife Ga wyssii La Palma“ = Fuerteventura tilosi La gomera Gran Canaria gomera tamadabae El Hierro bacchus Fig. 1. The Canary Islands showing the distribution of the five subspecies of Hipparchia wyssü. on Tenerife, La Gomera, La Palma and El Hierro (Herr Wiemers did not collect on Gran Canaria), and absent from the two eastern islands of Fuerteventura and Lanzarote, both of which are too dry for grass- feeding satyrine butterflies. This paper presents the results of our own sampling and field work on all five islands over the period 1987-91. Grayling butterflies have a complex wing pattern, particularly on the underside, and the variation is difficult to quantify. Here we identify scorable and quantifiable phenotypic variation in wing pattern and coloration and use this to estimate the extent to which each island population is differentiated. Our paper is complementary to two previous analyses of variation in Canary Island satyrine butterflies : Maniola jurtina (Linnaeus, 1758) (OWEN & SMITH, 1990) and Pararge xiphioides Staudinger, 1871 (OWEN & SMITH, 1993). The island populations Each island population is distinctive and we treat each as a subspecies ; in contrast WIEMERS (1991) recognizes four species : A. wyssii (Tenerife and Gran Canaria), H. gomera (La Gomera), H. bacchus (El Hierro) jig! and A. tilosi (La Palma). We recognize five subspecies, one on each island while agreeing that some or all of them may have diverged sufficiently to be designated as species ; as we discuss later, there is rarely conclusive evidence either way (from crossing) when dealing with allopatric populations and, in its absence, splitting should be avoided (CoRBET & PENDLEBURY, 1992). Plates 1 and 2 show the complex but distinctive underside patterns of males and females of each of the five subspecies. H. wyssii wyssii (Christ, 1890). Tenerife. Mostly confined to Pinus canariensis forest at altitudes of 1300 — 1700 m and also found above the pine forest zone to 2300 m at Las Cañadas. H. wyssii gomera (Higgins, 1967). La Gomera. Widespread in open country and light woodland at 300 — 1200 m. (There is no natural pine forest on La Gomera.) H. wyssii tamadabae Owen & Smith, 1992. Gran Canaria. Mainly in P. canariensis forest at 300 — 500 m but extending to 1800 m and especially associated with shallow barrancos. Known chiefly from the north-west of the island. H. wyssii bacchus (Higgins, 1967). El Hierro. Strongly associated with vineyards in barrancos at 300 — 700 m and extending in small numbers to 1500 m in laurel forest, but apparently rare or absent in P canariensis forest. H. wyssii tilosi Manil, 1984. La Palma. Apparently widespread in deep and steep-sided barrancos above 500 m in mixed laurel/ pine forest on the east and north-east side of the island, but probably absent from the extensive P canariensis forest around the Caldera de Taburiente. This subspecies may be much more split up into relatively isolated populations than the others. On all islands the butterflies are on the wing in July and August but can occur as early as May and as late as September. The early stages from four islands are described in WIEMERS (1991). The larvae are grass-feeders but precisely which species of grass are utilized in the wild has not been determined. There is one breeding generation a year. Adult butterflies occasionally visit flowers but are more often seen visiting damp patches and ripe fruit, especially grapes; on Gran Canaria, and probably also on Tenerife, they probe cracks in the bark for pine resin. On Tenerife and Gran Canaria in particular there is a marked downhill movement in the evening and an uphill movement in the morning. Most individuals involved are females. This behaviour is spectacular, especially in places where movement is funnelled by a 178 dry barranco ; we have no explanation for it but suggest it may be associated with locating egg-laying sites or with diurnal temperature changes. The cryptic coloration of the underside, especially of the hindwing, is superb. When at rest on a pine trunk with the forewing lowered, a butterfly is almost impossible to see. When a resting butterfly is approached it raises its forewings and exposes the eyespots and at first walks around the tree trunk, sometimes circling the trunk several times. This is presumed to be anti-predator behaviour but we have no observations of successful predation, only an attempted catch by a blue tit, Parus caeruleus Linnaeus, 1758, and two unsuccessful strikes by a large asiliid fly. Some of the butterflies in our samples have wing damage suggestive of unsuccessful attacks by lızards. The existence of H. wyssii on four of the Canary Islands has been known since 1888-9 (review in WIEMERS, 1991) but until 1966 it was well-known only from Tenerife, where it has been taken repeatedly. The rediscovery of populations on La Gomera and El Hierro in 1966 by Guichard and Ward (GuICHARD, 1967) was considered “unexpected and exciting...No butterflies of such distinctive appearance have been discovered in the Western Palaearctic Region for very many years” (Hıscıns, 1967). (Higgins was apparently unaware of earlier records for both islands dating from 1889 (REBEL & ROGENHOFER, 1894) and of NORDMAN (1935) for La Gomera.) The La Palma population re- mained unknown until 1983 (MANIL, 1984), possibly because its habitat is sO inaccessible, while the true habitat and status of the Gran Canaria population seems to have been unknown until we discovered it in 1990 (OWEN & SMITH, 1992 and this paper). Source of samples and methods of scoring In 1987-91 we visited all the islands where A. wyssii occurs and collected random samples, from single locations in the case of four islands and from six different and isolated populations, four of which are new records for the species, on La Palma. Samples were deliberately limited because of uncertainty as to exactly how common the species is on each island ; in the event we discovered that no island population is endangered except by habitat destruction resulting from increased cultivation. The butterflies were pinned and set for more detailed examination. Sample sizes for each island are given in Table 1. Fig. 2 is a diagram of grayling wings showing the numbering system for space(s) where spots (border ocelli) may be present or absent ; it 179 Table 1 Inter-island variation for forewing length (mm) in 7. wyssii in the Canary Islands Island abbreviations : T = Tenerife, G = La Gomera, C = Gran Canaria, P = La Palma, H = El Hierro. * One very small and otherwise aberrant female (22.8 mm), much the smallest individual of either sex recorded, substantially reduces the mean of this sample, without which it is 30.3 mm (n = 8). basal discal postdiscal submarginal Fig. 2. Diagram of the wings of Hipparchia wyssii showing the nomenclature of spaces (s) and the location of other wing pattern elements mentioned in the text. 180 also shows the position of other elements of wing structure mentioned in the text. Each butterfly was examined by eye using a X 2 lens. All the scoring was done by one of us (DASS) to ensure consistency. The characters scored are listed in Appendix I. Forewing length was measured (with vernier calipers to an accuracy of 0.1 mm) from the apex to the point where the basal area joins the thorax. It was possible to measure forewing length in every specimen collected but for some other characters not all specimens could be scored because of damage. Analysis Size Table 1 shows the mean forewing lengths for males and females. The rank order of forewing length for males is El Hierro > La Palma > La Gomera > Tenerife > Gran Canaria. If the single aberrant female from La Palma is omitted (Table 1), the rank order of forewing length for females is similar : El Hierro > La Gomera > La Palma > Tenerife > Gran Canaria. Table 2 shows the results of a one-way ANOVA for forewing length. The overall values of variance ratios (F) are significant for both males and females at the 0.1 per cent level, indicating that forewing length (which reflects body size) has diverged significantly among the island populations. Individual island comparisons are shown in Table 2 only where they are significant. Of the seven significant comparisons, five Table 2 One-way analysis of variance for forewing length in H. wyssii from the Canary Islands Source Degrees Variance of variation of freedom ratio (F) Probability ne u All islands CvG CvP CvH | TES Female BEAT All ıslands TvH CvG CvH PvH Island abbreviations as in Table 1. Comparisons for individual island pairs are included only if statistically significant. 181 Plate 1. Hipparchia wyssii, undersides of males (actual size). 1 — H. w. wyssü (Tenerife) ; 2 — H. w. gomera (La Gomera) ; 3 — H. w. tamadabae (Gran Canaria) ; 4 — H. w. bacchus (El Hierro) ; 5 — H. w. tilosi (La Palma). involve Gran Canaria (the smallest subspecies) and four El Hierro (the largest). Therefore, size may be a diagnostic character for at least these two islands but it is possible that larger samples would reveal that all five populations are statistically different in size. The marked difference in size between Gran Canaria and El Hierro butterflies can be seen in Plates 1 and 2. Hindwing border ocellus (spotting) phenotypes A list of hindwing spotting phenotypes is given in Appendix III. The spots, which belong to the border ocellus system (NıJHour, 1991), are 182 Plate 2. Hipparchia wyssii, undersides of females (actual size). 1 — H. w. wyssü (Tenerife) ; 2 — H. w. gomera (La Gomera) ; 3 — H. w. tamadabae (Gran Canaria) ; 4 — H. w. bacchus (El Hierro) ; 5 — H. w. tilosi (La Palma). numbered antero-posterially from the costal side as follows : 1 (space 6), 2 (s5), 3 (s4), 4 (s3), 5 (s2) (the position of the spaces is shown in Fig. 2). Spots 1 — 4 are pale ochreous or white on the upperside but invariably white on the underside. Spot 5 is always black and may be white-pupilled (underlined) or ringed with a yellowish halo (super- script bar). Table 3 shows the distribution of hindwing spotting phenotypes on each of the five islands. As shown, we recorded 18 upperside (UPH) pheno- types (A-R), and 22 underside hindwing (UNH) phenotypes (1-22). 183 Table 3 Distribution of the hindwing upper and underside border ocellus phenotypes of H. wyssii (see Appendix III for codes) on the five Canary Islands where it occurs | Phenotype |T|G|C|H| P| Phenotype |T|G|C|H|P| Phenotype |T|G|C|H|P| T | Phenotype |T|G|C|H| P| Phenotype |T|G|C|H] P| Phenotype |T|G|C|H] P| C|H| P| Phenotype | T|G|C|H| P| Phenotype | T|G|C|H|P + + - + + + + + - -- + + ar r - + -- + + + +|+]+]+ + + 4 +\+] |+/+ +) + + + + +) [+ + - + +++ + ++ + + + + + A a § recorded in males only (19), * recorded in females only (28), # recorded in both sexes (11). Island abbreviations as in Table 1. 58 different combinations were recorded ; the theoretical maximum is 396 though the actual number is probably far less. Considering the small size of some samples, however, many more phenotypes would be expected in larger samples, especially in males from Tenerife (n = 3) and females from La Gomera (n = 3). Nineteen phenotypes occur in males only, 28 in females only and 11 in both sexes : females are thus more variable than males. Table 4 summarizes inter-island differences in the distribution of hind- wing phenotypes and compares hindwing diversity on each of the five islands. Of the 58 hindwing phenotypes recorded only 12 (= 20%) occur on more than one island, none is found on all five islands and only three (= 5%) occur on four islands. This alone suggests considerable inter-island diversification. The distance coefficients for phenotypes shared with other islands (Table 6D) show that La Palma has the most distinct population, followed by Gran Canaria, Tenerife and La Gomera/El Hierro. La 184 Table 4 Analysis of the inter-island distributions (Table 3) of the hindwing phenotypes of H. wyssii (Appendix III) in the Canary Islands, with the phenotypic diversity on each island Islands Number of phenotypes (both sexes) Sample size Number phenotypes shared between 4 islands Number phenotypes shared between 3 islands Number phenotypes shared between 2 islands Total shared phenotypes Mean % phenotypes shared with other islands Mean phenetic distance from other islands Total unique phenotypes Per cent unique phenotypes Margalef Diversity Index (D,4) Berger-Parker Diversity Index (//d) Island abbreviations as in Table 1. Palma is also the most distinctive population in terms of the percentage of unique phenotypes, followed again by Gran Canaria and Tenerife. Small sample size in one or the other sex from some ıslands necessitates combining sexes to estimate phenotypic diversity. Although this pro- cedure inevitably involves the loss of some resolving power, the results are nevertheless unambiguous. We use two contrasting measures of diversity, the Margalef Index and the Berger-Parker Index (MAGGURAN, 1988). The former is primarily a measure of richness (number of pheno- types) (CLIFFORD & STEPHENSON, 1975) and has the disadvantage that it is strongly influenced by sample size (SOUTHwoo»p, 1978) ; the latter emphasises evenness (or dominance) and is little affected by sample size (MAY, 1975). In this case, where all samples are fairly small, the Berger-Parker Index is preferred although encouragingly the rankings for the two indices are in substantial agreement. The La Palma (Berger- Parker) and Gran Canaria (Margalef) populations have the highest diversity, El Hierro and La Gomera (especially) the lowest, with Tene- rife intermediate (both indices agree on the ranking of the latter three islands). In general the larger islands (Tenerife, Gran Canaria and La Palma) are more diverse than the smaller (El Hierro and La Gomera), as could be predicted from biogeographical theory (MACARTHUR & WILSON, 1967). The high diversity on La Palma, in particular, is enigmatic : it may result from independent evolution of sub-populations in relative isolation within deep barrancos, a situation which probably does not prevail on the other islands. 185 Table 5 Mean values for spots, pupils and halos comprising the hindwing border ocellus system in H. wyssii from the Canary Islands land i Pool PS al 2 iP ni won UPH spot average (max. 5) UNH spot average (max. 5) Pupil average (max. 6) Halo average (max. 6) N (N*) UPH spot average (max. 5) UNH spot average (max. 5) Pupil average (max. 6) Halo average (max. 6) N (NX) * Numbers in parenthesis are sample sizes for spot averages reduced due to hindwing damage in some specimens. § Sample means significantly different (in ¢ tests) from others in the same row. Island abbreviations as in Table 1. Table 5 shows the UPH and UNH spot averages for males and fe- males for each island. Correlation for UPH and UNH spot number in individuals is low and not significant in either males (r = 0.070, n = 68) or females (r = 0.073, n = 55) ; NısHouT (1991) has emphasised that dorsal and ventral wing coloration patterns develop from epithelial monolayers within separate compartments and that correlation between them is by no means inevitable. As shown, La Palma has the highest values for spot average for both UPH and UNH. El Hierro and La Gomera have the lowest values with Tenerife and Gran Canaria inter- mediate. The frequency of white pupils in the six black spots (border ocelli) (UPF/UNF s6 and s3 and UPH/UNH s2) and of the yellow halos around the same spots, provides a set of characters additional to hind- wing spotting which can be used to differentiate populations. Table 5 gives pupil and halo averages for males and females for each island. The sum of pupils (maximum 6) and halos (maximum 6) for an 186 individual butterfly gives a measure which we call spot enhancement (maximum 12); both pupils and halos render the black spots more distinctive and conspicuous. Males and females from La Palma and Gran Canaria have significantly higher enhancement values than the other populations but it is obtained by different means : mainly by pupillation on Gran Canaria (a consistently diagnostic feature of ssp. tamadabae) and by the development of halos in ssp. tilosi from La Palma. Both features are recorded from some of the other islands but less distinctly and much less frequently. On La Palma and Gran Ca- naria, the black spots tend to be larger and, when combined with white pupils and yellow halos, they are rendered much more apparent. The ssp. bacchus phenotype is also highly distinctive in a quite different way from ssp. tilosi or ssp. tamadabae, because of the intense contrast between the postdiscal black line (element f of the central symmetry system (NıJHoUT, 1991)) and the band of white background immediately distal to it, especially on the UNH (Plates 1 and 2); ssp. bacchus has low values for spot average and much the lowest development of spot enhancement (Table 5). As the background of the bacchus wing is extensively melanised, several other black pattern elements (d, e and i (NısHout, 1991)) are also relatively obscured. Subspecies gomera is intermediate between ssp. bacchus and ssp. wyssii in all the above respects (Plates 1-2). Table 6 Matrices showing mean Euclidean distances (MED) for five H. wyssii OTUs in the Canary Islands : (A) males, (B) females, (C) combined sexes, (D) hindwing spotting phenotypes (both sexes) (Table 3, Appendix III). A-C are based on 34 non-metric characters and D on numbers of phenotypes shared between islands (Table 4) T G C H P T G C H 12 Kendall’s rank correlation for A and B is not significant (t = 0.4, P — 0.11) ; C and D are significantly correlated (7 = 0.822, P = 0.04). Island abbreviations as in Table 1. 187 Phenetic distance coefficients The phenotypes of all members of the genus Hipparchia Fabricius, 1807 are difficult to measure as much of the variation is quantitative and therefore judgment of it subjective. Hence for the calculation of phenetic distance coefficients (Table 6), we have chosen 34 non-metric characters (Appendix I) which are discrete (i.e. can be consistently scored as present or absent) and variable in frequency within or between islands. This statement does not imply that these characters are the only ones by which the five island populations can be compared and differentiated, nor that they are necessarily even the most distinctive. Many characters are manifested only or mainly in males (9) or females (7) ; those which are variable in both sexes (18) nevertheless differ in frequency and so they are scored separately for males and females. Therefore the matrices (Table 6) are based on 27 characters for males, 25 for females and 52 for the combined sexes. As shown in the character matrix (Appendix II), few characters are absolutely diagnostic, most differing only quantitatively between islands. Hence each character score is weighted (0-100%) for its comparative frequency. The data in Table 6 represent the Euclidean distance MED (SNEATH & SOKAL, 1973) between pairs of island populations (OTUs). ED is calculated for each OTU (subspecies) pair using the formula : EDy =[2(X, - Xu)" where X; and X; are the scores of OTUs (subspecies) j and K for character 1. Mean Euclidean distances (WED) over all characters are calculated as : MED, = \[2(X, - Xx)21/n where n is the number of characters. Comparing the distance values (WED) between sexes (Table 6A, B, Fig. 3A, B) it is clear that females have in general diverged further than males. In the case of La Palma this applies to the female MED values compared to all other islands ; for Gran Canaria it is confined mainly to the comparison with Tenerife. Furthermore, the differences between the male and female phenograms (Fig. 3), and the absence of any significant rank correlation between their inter-island phenetic distances (Table 6), strongly suggest that there has been a degree of independence in the evolution of the phenotype of the two sexes, the male being the more conservative . 188 PG, (AY Th 0.2 0.7 3 5 20.3 {1} is 0.8 T _ 5 0.4 D © 2 w 0.9 5 8 0.5 ® > < 0.6 1.0 Fig. 3. Phenograms, derived by UPGMA, showing the mean Euclidean distances (MED) (Table 6) between the five subspecies of Hipparchia wyssü: A, males ; B, females ; C, combined sexes ; D, hindwing border ocellus phenotypes (both sexes) shared between islands (Table 3). Island abbreviations as in Table 1. Considering combined sexes (Table 6C), La Palma (mean distan- ce — 0.523) ıs the most distinctive followed by Gran Canaria (0.468). El Hierro (0.450) is intermediate, with Tenerife and La Gomera (both 0.422) the least distinctive on average. The two most similar populations are El Hierro and La Gomera. However, there is no rank correlation (SIEGEL & CASTELLAN, 1988) between the geographical distances separating islands and phenetic distances (Kendall’s 7 (tau) = 0.422; P=0.11). In Table 6D we compare the islands by numbers of shared hindwing phenotypes. The isolation of La Palma and the closeness of La Gomera and El Hierro are again clear and, as above, there is no rank correlation between the physical distance separating islands and the number of shared phenotypes (t = 0.244 ; P= 0.38). Encouragingly, however, the two methods of inter-island comparison (Table 6C, D) are themselves significantly correlated by rank (t = 0.822 ; P= 0.04). Interpretation of inter-island variation The phenograms in Fig. 3 are derived by the UPGMA method (SNEATH & SoKAL, 1973) from the data in Table 6. The relative isolation of 189 the La Palma subspecies is evident in all the phenograms ; La Gomera and El Hierro form a cluster ; Tenerife and Gran Canaria may also form a cluster, possibly for male phenotypes only (Table 6A, 6C), but 6B and 6D suggest that Tenerife is closer to La Gomera/El Hierro with Gran Canaria more isolated. The raw data for males (Table 6A) indicate that ssp. tilosi (La Palma) is closer to ssp. gomera (La Gomera) than to any other but this is not the case for females. The phenetic distance of ssp. tilosi females from all other subspecies would suggest either a comparatively early dichotomy from the ancestral stock or more recent but rapid evolution. To explain this anomaly it is neces- sary to postulate different rates of evolution in males and females, the former being relatively conservative and the latter evolving rather rapidly, especially on La Palma. All data sets (Table 6) show that ssp. tamadabae (Gran Canaria) is closest to ssp. wyssü (Tenerife). In this case too, however, there is a sex difference, the males being much closer than the females. The El Hierro subspecies (ssp. bacchus) is un- doubtedly closest to ssp. gomera and is equally distinctive, mainly due to the dark background (unscored) (Plates 1 & 2), in both sexes. It should be noted, however, that WIEMERS (1991) observed that the eggs of these two subspecies are quite dissimilar. The origin and relationships of A. wyssü There are three other species belonging to the subgenus Pseudotergu- mia Agenjo, 1947 (KUDRNA, 1977). If examined comparatively in the manner advocated by HARVEY & PAGEL (1991), they are the outgroup for reconstructing ancestral character states. 7. fidia (Linnaeus, 1767) is a west Mediterranean species ranging from Morocco, Algeria and Tunisia through Portugal, Spain and S. France to Italy; A. pisidice Klug, 1832 occurs in the Middle East from Sinai through Jordan, Palestine, Lebanon and Syria to south Turkey ; A. tewfiki (Wiltshire, 1949) is a relict endemic restricted to Yemen and southwest Arabia. All three species share with H. wyssii tamadabae the white pupilling of the two forewing ocelli in s6 and s3 (UPF and UNF). These cha- racters are likely to be primitive for the subgenus and their absence from most individuals of the other subspecies of H. wyssii is probably due to loss. There are several other similarities between ssp. tamadabae and the other Pseudotergumia species (the nomenclature of the fol- lowing pattern elements follows NiJHouT (1991) : first, the bold black proximal band of the central symmetry system (element d), which bisects the discal cell (UNF); second, bold discal spots (element e) on the discoidal vein (UNF) and, third, bold proximal (medial) and 190 distal (postmedial) bands (elements d and f) of the central symmetry system (UNH), given in our original description of ssp. tamadabae (OwEN & SMITH, 1992). All things considered, it seems likely that A. wyssii tamadabae is closest to the ancestral Pseudotergumia (and Hipparchia) phenotype. In other words, the states of some of the above characters in most individuals of H. wyssii ssp. wyssii, gomera, bacchus and rilosi are derived. The origin of H. wyssii itself is a matter for conjecture. On the evidence of present distributions, an origin for H. wyssii from H. fidia stock, which inhabited the western Mediterranean area, seems the most plausible. Indeed, while admitting that Hipparchia phenotypes are in general conservative, the adult phenotype of H. fidia is in many respects close to H. wyssii tamadabae. However, from a comparison of the male genitalia and androconia, KUDRNA (1977) believed H. wyssii to be closer to H. tewfiki than to either A. fidia or H. pisidice. Judging from his excellent figures of the very similar genitalia and androconia of all the Pseudotergumia species, his conclusion requires exceeding- ly fine distinctions to be made, mainly on grounds of size. As is com- monly the case when genitalia characters are used in insect taxonomy, no allowance was made for allometry (GouLD, 1966): H. tewfiki is considerably smaller than all other species of Pseudotergumia but, of the subspecies of H. wyssii examined by Kudrna, is closest in body size to ssp. wyssii (which he believed inhabited both Tenerife and Gran Canaria). The above remarks notwithstanding, as the only worker to have examined all known museum specimens of the four Pseudoter- gumia species, we tentatively accept Kudrna’s conclusion that 7. wyssii is closer to A. tewfiki than to A. fidia. In view of the conservatism of Hipparchia adult phenotypes and the considerable continuous variation between individuals within taxa, a protein or DNA study is needed to reach more robust conclusions. If A. wyssii and A. tewfiki are indeed closely related then their present distributions, in the Canaries and southwest Arabia, respectively, are highly disjunct. To account for the presence of H. wyssii in the Canary Islands, it is therefore necessary to invoke both vicariance and several island-hopping events. We argue that neither is improbable as both are prevalent features of the flora and fauna of the Canary Islands. First, during the Miocene, southern Europe and the present Sahara Desert area were covered with subtropical to tropical evergreen forests (AXELROD, 1986). As drier climates spread through the Upper Miocene and Pliocene, disjunctions in range developed, which are well-docu- mented for the flora (BRAMWELL, 1976) resulting in highly disjunct Fall distributions between, on the one hand, the Canary Islands and, on the other, far-distant locations such as Arabia, Socotra, Mauritius, the East African highlands, the Himalayas and south Asia. Two of these disjunctions are particularly relevant to a possible scenario for the origin of H. wyssii: Pinus canariensis may be most closely related to P rox- burghii of the western Himalaya, with fossil relatives in southwest Asia (AXELROD, 1986) ; the genus Appollonias (Lauraceae), one of the four lauraceous species found in the /aurisilva, has only two living species, A. barbajuna in the Canaries and A. arnotti in south India (SUNDING, 1979). Moreover, there are parallel examples among butterflies : the distribution of Vanessa indica (Herbst, 1794) (Nymphalidae), with distinct subspecies in Macaronesia (the Canaries and Madeira) and south Asia, has attracted recent attention (LEESTMANS, 1978 ; SHAPIRO, 1992a, 1992b) ; the genus Cyclyrius Butler, 1897 (Lycaenidae), which has only two living species, C. webbianus (Brullé, 1840) in the Canaries and C. mandersi (Druce, 1907) from Mauritius, presents a similar case (OWEN & SMITH, 1993). There is considerable evidence that many genera of butterflies from several families and subfamilies, including some satyrınes (PORTER & GEIGER, 1988 ; PORTER & SHAPIRO, 1991), are slow-evolving or brady- telic (Simpson, 1944). We believe that Hipparchia, in particular the ancestor of tewfiki-wyssii is probably a bradytelic taxon and that A. wyssii has a relatively ancient origin (LARSEN, 1984) in the Miocene, say 10 — 15 Ma. The first dispersal event, presumably from North Africa, was probably to the eastern islands of Gran Canaria and Fuerteventura, possibly at the same time as the invasion of the /auri- silva and P canariensis forest and before these islands became arid. The second dispersal event may have been from Gran Canaria to La Gomera. However, the low phenotypic diversity on La Gomera, compared to Gran Canaria, and indeed all other ıslands, suggests a bottleneck effect at introduction which is still reflected in the modern population. Subsequent island hops would follow, first to Tenerife, not earlier than 6 Ma (SCHMINKE, 1976) and then to El Hierro and La Palma during the Pleistocene (2 — 0 Ma). The increasing aridity of the Sahara region during the Pliocene would then produce the vicariance event which split the zewfiki-wyssii ancestor into two relict species. Comparative morphology and high phenotypic diversity, as well as the antiquity of Gran Canaria (SCHMINKE, 1976), support the proposed ancestral status of ssp. tamadabae. Whether ssp. wyssii on Tenerife originated from La Gomera or Gran Canaria (or both) remains an open 192 question, each being almost equally likely on grounds of comparative morphology (Table 6). The data indicate that ssp. bacchus (El Hierro) originated from La Gomera, the nearest source, at a date which must have been = 2 Ma ago. The case of ssp. tilosi, however, appears anomalous and defies simple interpretation. While it is undoubtedly closest to ssp. gomera (Table 6), it is the most distinct of all five subspecies (Fig. 3) and shares the smallest number of hindwing phenotypes with other subspecies ; ıt also shows the highest phenotypic diversity (Table 4). We suggest above that the fragmented distribution of ssp. tilosi into many small isolated populations, a situation which does not apply to the other subspecies, may have resulted in numerous founder effects with unusual and dif- fering starting sets of alleles ; sub-population differences might further intensify under subsequent genetic drift. Our collection is much too small to hope to find quantitative differences between the six sub- populations from La Palma represented, but the high phenotypic diversity in the bulked sample may reflect heterogeneity arising from the fragmented population structure of this subspecies. However, the fact that ssp. rilosi is the most distant (phenetically) from all other subspecies and occurs on the island known to be the youngest (= 2 Ma) (SCHMINKE, 1976) and furthest from a continental source of recruitment, suggest two other possible interpretations of the evolutionary history of H. wyssii. First, the species may have colonised the Canary archipelago only within the last two million years, in which case the diversification we describe occurred within the Pleistocene or later. If this is the case, then rapid evolution has occurred on El Hierro and (especially) La Palma. It follows from this that the pre-Pleistocene geological history of the islands would be irrelevant to any attempt to reconstruct the evolution of H. wyssii. Alternatively, the species may have reached the eastern islands a long time ago but dispersed to La Palma and El Hierro, where it has undergone rapid evolution, comparatively recently. This last interpretation best fits the facts. Discussion Graylings are essentially “underside” butterflies. Only rarely is the upperside visible and it is the underside that has the intricate and characteristic patterning. When a butterfly comes to rest, the underside of the forewing is well-exposed and its spots are conspicuous. After a short interval, the forewing is lowered and hidden by the hindwing, and the butterfly is beautifully camouflaged against the background 193 of rock or tree trunk. If alerted, it raises the forewing and the eyespots are again visible. This behaviour is widespread among members of the Satyrinae but is particularly well-developed in Hipparchia and it is usually interpreted as anti-predator strategy. The forewing eyespots catch the attention of a predator which may then strike at a non- vulnerable part of the wing (BRAKEFIELD, 1984), enabling the butterfly to escape by flying away. If there is no strike, the butterfly assumes the resting position with the forewing concealed. Our observations on all subspecies of A. wyssii suggest that this presumed anti-predator behaviour is the norm. The rarely exposed upperside is rather uniformly dark and may function in thermoregulation, though only in flight, as Hipparchia group species are exclusively lateral baskers at rest (SHREEVE & Dennis, 1992): it is probably never involved in anti-predator behaviour. Hence, we view the ground plan of the underside pattern of all five subspecies as having evolved in response to selection by predators ; most of this evolution must have occurred before what is now H. wyssii reached the Canary Islands. Assuming a mainland origin for A. wyssii, the initial colonisation of any one of the islands may have involved few individuals ; indeed a single fertilised female is all that is required. Subsequent colonisation of other islands might similarly involve few individuals. Each colonisation has produced relatively large but isolated populations which could easily have been founded with only a fraction of the genetic diversity of its source population(s) leading to selective and epigenetic constraint. The restricted gene pool of each founder population would thus be unique to each island and differ stochastically from all other islands : this alone could account for most of the inter- island differentiation we have described. Genetic drift in the initially small populations of founders would be expected to lead to a further loss of genetic diversity and to the fixation of inter-island differences, which are mainly so minor that their influence on survival may be trivial. NiyHoutT (1991) emphasises that small changes in the source or steep- ness of diffusion gradients for theoretical morphogens, thresholds for the expression of alternative pigment genes and pattern elements and the timing of the sequence of morphogenetic events, can produce quite profound changes to the visible phenotype. We believe that develop- mental constraint, probably involving very few gene loci with alter- native alleles, has produced the high proportion of unique character combinations that now occur on each island. The overall result is that the distinct grayling phenotype is retained but its precise expression 194 on each island is different and dependent on the genetic history of each population. The hindwing underside is the most complex of a grayling’s wing surfaces and it is this which is most often exposed to potential predators : it is a cryptic wing surface but the precise means by which crypsis is attained differs from island to island. Of the 30 species of butterfly found on the Canary Islands, four (in- cluding H. wyssii) exhibit inter-island differentiation in wing pattern morphology, while others probably do so but have not been subjected to detailed analysis. In all four cases, the island differences are quan- titative rather than qualitative. In two species, Maniola jurtina and Pararge xiphioides, subspecific names have not been given, while in the other two, Gonepteryx cleopatra (Linnaeus, 1767) and H. wyssii, subspecific (or specific) status is based on small differences in adult size, Wing pattern and the structure of genitalia (KUDRNA, 1975, 1977), and in the latter case, also on egg and larval morphology (WIEMERS, 1991). While these features differ slightly between islands, they vary only to an extent, as in comparisons between Palaearctic and Nearctic Aglais Dalman, 1816 and Nymphalis Kluk, 1802, that either species or subspecies might be expected to differ (MILLER & MILLER, 1990). To demonstrate speciation more positively would require cross-breeding or better, analysis of mtDNA. However, our view is that, although it hardly matters whether the island populations of H. wyssii are de- signated as species or subspecies, we have found no compelling evidence for full speciation and are thus persuaded to leave them as well- differentiated subspecies. Acknowledgements DASS thanks the administrators of the Educational Fund, Eton College, for a research grant to work in the Canary Islands in 1990. Mr Derek Whiteley drew the figures and he and Mrs Barbara Southall took the photographs for Plates 1 and 2. References AXELROD, D. I, 1986. Analysis of some paleogeographic and paleoclimatic problems of paleobotany : The 33rd Sir Albert Charles Seward Memorial Lecture. Paleobotanist 35 : 115-129. BRAKEFIELD, P. M., 1984. The ecological genetics of quantitative characters of Maniola jurtina and other butterflies. Jn R. I. VANE-WRIGHT and P. R. ACKERY (Eds.), The biology of butterflies, 167-190. Symposium of the Royal Entomological Society of London, No. 11, Academic Press, London. 195 BRAMWELL, D., 1976. The endemic flora of the Canary Islands. In KuNKEL, G. (Ed.), Biogeography and ecology in the Canary Islands, pp. 207-240. W. Junk, The Hague. CLIFFORD, H. T. & STEPHENSON, W., 1975. An introduction to numerical classification. Academic Press, London. CoRBET, A. S. & PENDLEBURY, H. M., 1992. The butterflies of the Malay Peninsula. 4th. edition revised ELioT, J. N. United Selangor Press, Kuala Lumpur. GouLD, S. J., 1966. Allometry and size in ontogeny and phylogeny. Biological Reviews 41 : 587-640. GUICHARD, K. M., 1967. Butterflies in the Canary Islands. Entomologist 100 : 293-299. Harvey, P. H. & PAGEL, M. D., 1991. The comparative method in evo- lutionary biology. Oxford University Press, Oxford. Hicains, L. G., 1967. Hipparchia (Pseudotergumia) wyssii CHRISTOPH, with descriptions of two subspecies. Entomologist 100 : 169-171. KUDRNA, O., 1975. A revision of the genus Gonepteryx Leach (Lep. Pieridae). Entomologist’s Gaz 26 : 3-37. Kuprna, O., 1977. A revision of the genus Hipparchia Fabricius. E. W. Clas- sey, Faringdon. LARSEN, T. B., 1984. Butterflies of Saudi Arabia and its neighbours. Stacey International, London. LEESTMANS, R., 1975. Etude biogéographique et écologique des Lepidopteres des iles Canaries (Insecta Lepidoptera). Vieraea 4 : 9-116. LEESTMANS, R., 1978. Problèmes de spéciation dans le genre Vanessa. Vanessa vulcania Godart stat. nov. et Vanessa buana stat. nov. : bonae species. (Lepidoptera : Nymphalidae). Linneana Belgica 7 : 130-156. MACARTHUR, R. H. & Witson, E. O., 1967. The theory of island biogeo- graphy. Princeton University Press, Princeton, NJ MAGURRAN, A. E., 1988. Ecological diversity and its measurement. Croom Helm, London. Mani, L., 1984. Découverte de Hipparchia (Pseudotergumia) wyssii CHRIST dans l’île de la Palma (Canaries) et description d’une nouvelle sous- espece : Hipparchia wyssii tilosi nova ssp. (Lepidoptera Satyridae). Linn. belg. 9 : 359-367. May, R. M., 1975. Patterns of species abundance and diversity. In Copy, M. L. & Diamonpn, J. M. (Eds.), Ecology and evolution of communities, pp. 81-120. Harvard University Press, Cambridge, Mass. Mizer, L. D. & Mie, J. Y., 1990. Nearctic Aglais and Nymphalis (Lepi- doptera: Nymphalidae): Laurasia revisited? Entomologist 109: 106-115. NısHout, H. F., 1991. The development and evolution of butterfly wing patterns. Smithsonian Institution Press, Washington. NORDMAN, A. F., 1935. Verzeichnis der von Richard Frey und Ragmar Stora auf den Kanarischen Inseln gesammelten Lepidopteren. Commentationes biological 6 : 1-20. Helsinki. 196 Owen, D. F. & SMITH, D. A. S., 1990. Interpopulation variation and selective predation in the meadow brown butterfly Maniola jurtina (L.) (Lepido- ptera : Satyridae) in the Canary Islands. Biol. J. Linn. Soc. 39 : 251-267. Owen, D. F. & SmiTH, D. A. S., 1992. A new subspecies of Hipparchia wyssii (Christ) (Lepidoptera : Satyrinae) from Gran Canaria, Canary Islands. Entomologist’s Gaz 43 : 253-256. Owen, D. FE & SmitH, D. A. S., 1993. Interisland phenotypic diversity in Pararge xiphioides (Staudinger) (Lepidoptera : Satyrinae) in the Canary Islands. Biodiversity Letters 1 : 23-30. Owen, D. F. & SMITH, D. A. S., 1993. The origin and history of the butterfly fauna of the North Atlantic islands. Bolm Mus. Mun. Funchal Suppl. 2: 211-241. PORTER, A. H. & GEIGER, H. J., 1988. Genetic and phenotypic population structure of the Coenonympha tullia complex (Lepidoptera : Nympha- lidae : Satyrinae) in California : no evidence for species boundaries. Can. J. Zool. 66 : 2751-2765. Porter, A. H. & SHAPIRO, A. M., 1991. Genetics and biogeography of the Oeneis chryxus complex (Satyrinae) in California. J. Res. Lepid. 28: 263-276. REBEL, H. & ROGENHOFER, A., 1894. Zur Lepidopterenfauna der Canaren. Annin. naturh. Mus. Wien 9 : 1-96. SCHMINKE, H.-U., 1976. The geology of the Canary Islands. /n KUNKEL, G. (Ed.), Biogeography and ecology in the Canary Islands, pp. 67-184. W. Junk, The Hague. SHREEVE, T. G. & Dennis, R. L. H., 1992. The development of butterfly settling posture : the role of predators, climate, hostplant, habitat and phylogeny. Biol. J. Linn. Soc. 45 : 57-69. SHAPIRO, A. M., 1992a. How did Vanessa indica (Herbst) (Lepidoptera : Nymphalidae) get to the Canary Islands and Madeira ? Entomologist 111: 10-21. SHAPIRO, A. M., 1992b. A postscript on Vanessa indica (Herbst) (Lepidoptera : Nymphalidae). Entomologist 111 : 162-163. SIEGEL, S. & CASTELLAN, N. J. Jr., 1988. Nonparametric statistics for the behavioral sciences, 2nd. ed. McGraw-Hill, New York. SIMPSON, G. G., 1944. Tempo and mode in evolution. Columbia University Press, New York. SNEATH, P. H. A. & SOKAL, R. R., 1973. Numerical taxonomy : The principles and practice of numerical classification. W. H. Freeman, San Francisco. SOUTHWOOD, T. R. E., 1978. Ecological methods. Chapman and Hall, London. SUNDING, P., 1979. Origins of the Macaronesian flora. /n BRAMWELL, D. (Ed.), Plants and islands, pp. 13-40. W. Junk, The Hague. WIEMERS, M., 1991. Hipparchia wyssii (Christ, 1889) Komplex : Beitrag zur Morphologie, Biologie, Ökologie und Verbreitung auf den Kanarischen Inseln (Lepidoptera, Satyridae). Nota lepid. 14 : 255-278. 12/7 APPENDIX I Characters used for analysis of inter-island differences in Hipparchia wyssii in the Canary Islands WEF Small ochreous border ocellus (h) in s7. # Black border ocellus in s6 (h) with white pupil. Black border ocellus in s6 (2) ringed with yellowish halo. Pale submarginal spot (h) in s5. * Pale spot in s5 (4) white (+) or brown. * Pale submarginal spot (h) in s4. * Pale spot in s4 (6) white (+) or brown. * Black border ocellus (h) in s3. * Black border ocellus (8) in s3 with white pupil. # 10. Black border ocellus (8) in s3 ringed with yellowish halo. 11. Pale postdiscal spot in s6. * 12. | Above spot (11) in s6 white (+) or pale brown. # 13. Pale postdiscal spot in s3. * en 14. Pale ochreous or white border ocellus (h) in s6 (spot 1). 15. Similar ocellus (h) in s5 (spot 2). 16. Similar ocellus (h) in s4 (spot 3). # 17. Similar ocellus (h) in s3 (spot 4). 18. Small submarginal black border ocellus (h) in s2 (spot 5). * 19. Spot 5 (18) with minute white pupil. # 20. Spot 5 (18) ringed with narrow yellowish halo. # 21. Large black border ocellus (h) in s6 with white pupil. 22. Spot in s5 fused with halo of border ocellus in space 6. 23. Black border ocellus (h) in s3 with white pupil. # 24. Black border ocellus (h) in s3 ringed with yellowish halo. 25. s3 between elements f (postmedial line) and h ochreous (+) or brown. 26. Postdiscal area of s2 ochreous (+) or brown. 27. Diffused ochreous spot proximal to element d in discal cell. * 28. Diffuse white submarginal border ocellus (h) in s6 (spot 1). 29. Similar spot (h) in s5 (spot 2). 30. Similar spot (h) in s4 (spot 3). 3l. Similar spot (h) in s3 (spot 4). 32. Black spot in s2 with minute white pupil. 33. Pale patches immediately lateral to postmedial line (f) flecked white (+) or grey. 34. Above patches (33) fused to form irregular but uninterrupted stripe. Letters in bold type identify the probable homologies of pattern elements according to the nymphalid ground plan (NisHout, 1991). s = space (Fig. 2) UPF = upperside forewing ; UPH = upperside hindwing ; UNF = underside forewing ; UNH = underside hindwing ; + = present ; characters expressed or variable only in one sex are marked * for males (9) and # for females (7) ; all other characters (18) are scored in both sexes. 198 APPENDIX II Character matrix for A. wyssii from the Canary Islands. The data are expressed as percent occurrence [es Sel bl ie BW REIN SW D m = A G0 90 DH — CO CO 00 \O CoO RK = WO WwW © © & & W © © iS) | Se EN | N al NS | | su no Soo-maa ON — © Un © ON W SAS SSD) mi = O0 ~) W \O © © O \O SL2SHRUSO | © © © 00 In 00 00 = OO ~) ~) OV SONAAAN | ee — oo 0 8 8 5 8 8 0 7 Y 3 0 0 0 0 0 3 0 0 a 5 0 0 © © OO OS IS © © © — SS WBSANWBIARSD oo OWN Un © © © W 00 199 APPENDIX III Classification of the hindwing border ocellus (spotting) phenotypes of A. wyssü in the Canary Islands UPH UNH | Spots Code Spots Code Spots Code Spots Code B © D B F G H I J Spots are numbered antero-posterially from the costal side as follows : 1 (s6), 2 (s5), 3 (s4), 4 (s3), 5 (s2). White pupils are indicated as e.g. 5 and yellow halos as e.g. 5. 200 Nota lepid. 17 (3/4) : 201-216 ; 30.1V.1995 ISSN 0342-7536 Danaus chrysippus Linnaeus, 1758 ; a review of records and present status in the Maghreb countries of Morocco, Algeria and Tunisia (Lepidoptera, Danainae) John TENNENT 1 Middlewood Close, Fylingthorpe, Whitby, N. Yorkshire YO22 4UD, England Summary Danaus chrysippus Linnaeus, 1758, was first reliably noted from Algeria in 1912, from Morocco in 1943 and from Tunisia in 1952; it has since been recorded there sporadically. Sightings have increased in recent years and some have wrongly been claimed as new country records, due possibly to the erroneous claim in early editions (and English language editions/reprints up to 1993) of Higgins & Riley’s “Field guide to the butterflies of Britain and Europe” that the species had not been recorded from Algeria or Tunisia. Known records of Danaus chrysippus in the Maghreb countries of Morocco, Algeria and Tunisia are reviewed here. It is shown that the butterfly has established breeding populations in all three countries. Hostplants are discussed. Resume Danaus chrysippus Linnaeus, 1758, fut signalé pour la première fois d'Algérie en 1912, du Maroc en 1943 et de Tunisie en 1952 ; il a fait l’objet, depuis, de citations sporadiques. Les observations se sont multipliées ces dernières années, dont certaines indûment revendiquées comme nouvelles, censément en raison de l’assertion erronée des premières éditions (et éditions en langue anglaise, réimprimées jusqu’en 1993) du «Guide des Papillons de Grande-Bretagne et d'Europe» de Higgins & Riley, donnant l’espèce comme inconnue d'Algérie ou de Tunisie. Les citations de Danaus chrysippus des pays maghrébins du Maroc, d'Algérie et de Tunisie sont ici rassemblées. Il est démontré que le Petit Monarque a établi des colonies stables dans ces trois pays. Les plantes- hôtes sont étudiées. Zusammenfassung Danaus chrysippus Linnaeus, 1758, wurde aus Algerien erstmals 1912 gemeldet, aus Marokko 1943 und aus Tunesien 1952 ; seitdem wurde er dort sporadisch gesichtet. In den letzten Jahren häuften sich die Beobachtungen ; dabei wurden 201 einige fälschlich als Erstnachweise bezeichnet. Dies ist vermutlich darauf zurückzuführen, daß in früheren Ausgaben des Feldführers „Die Tagfalter Europas und Nordwestafrikas“ von Higgins und Riley irrtümlich behauptet wird, die Art sei aus Algerien oder Tunesien nicht bekannt. Die bisher bekann- ten Nachweise von Danaus chrysippus in den Maghreb-Ländern Marokko, Algerien und Tunesien werden zusammengestellt. Es wird gezeigt, daß der Falter in allen drei Ländern bodenständige Populationen besitzt. Die Futter- pflanzen werden diskutiert. Introduction There has been some confusion in recent years over the occurrence and distribution of the butterfly Danaus chrysippus Linnaeus, 1758, in the Maghreb states of Morocco, Algeria and Tunisia. Hiccins & RILEY, in all English language editions to date of their Field Guide to the Butterflies of Britain and Europe, currently the only guide to the region, stated that the butterfly is a rare migrant in Morocco and has not been reported from Algeria or Tunisia. The most recent French language edition is more up to date. A mass of further information has been published, in a number of languages in a variety of entomological journals, on the distribution of D. chrysippus in the Maghreb and in Europe with some recent papers claiming new records for Algeria and Tunisia. There has also been speculation and discussion on the probable hostplants of D. chry- sippus in North Africa. This paper sets out to collate published records to date ; to assess the present status of the butterfly in the Maghreb and to identify host- plants in each country. All localities can be found on the map (Fig. 1). Whilst the occurrence of D. chrysippus in Europe is outside the scope of this paper, it has been seen more frequently there in recent years and published records noted are included in the form of a bibliography. Records to date (M) = Morocco ; (A) = Algeria ; (T) = Tunisia 1905 — Joannis (1908 : 83), in his list of Saharan Lepidoptera recorded by Chudeau on a voyage from Algiers to Timbouctou in 1905-6, noted D. chrysippus ‘var.’ alcippus Cramer, 1777 taken at Oued Kadamellet on 21 September 1905 and as this has been quoted (including by the present author !) as the first published record of Danaus chrysippus in Algeria, it is worth taking a moment to establish where Kadamellet lies. It transpires that the locality is in Niger. 202 ¢ jesspJupwp| ® uejsob| DIDHEND uUyg e DAW peno® unowiwwiL e DIPP1DU9 4anoBBno] © vısınn x yonoybo7 e psy.” DAysig era eg e ““e $ auljuDjsUu0D e S#3191v "eIsıun], PUR BIIOS[Y ‘0990J0W UI "1 snddisduyo snoung Jo 39u211N990 UMOUY “| “LA N N N SUMO}[) >; suoljpjndod Buipee1qO SS sBuiyybis snddisksyo @ N Aa» DIUD/8/NDW (D1DyDS u1e/sem) POISSON e —. — —. 'sı Aiouo9 yas JUUDPNO1D] © NDUSZD| © U90YDJIDW : 0929080 D9UD|qDSDY 203 The author was unable to find reference to Oued Kadamellet on any modern map. However, Mont Kadamellet was eventually identified on a very old map at 19° 34’ N, 8° 36’ E, some 45 km NNE of Iferouane, Air, the position loosely described by Joannis (1908 : 82). In June 1905, an agreement between the Commandant of what was then Haut Sénégal et Niger and the military commander of the Depart- ment de l’oasıs (part of Territories du Sud Algerien), determined a boundary between Algeria and French West Africa which was completed in principle by the Niamey Conventions of 1909 (International Boundary Study No. 99). Thus ’Kadamellet’ was and is in Niger, about 200 km SE of the Algerian border. 1907 — Seitz (1907 : 75-76), noted that [form] chrysippus was to be found on the Canary Islands “and the opposite districts of Morocco ... The absence of the insect from Algiers [Algeria] is very remarkable ... ab. alcippoides... in the Palearctic region only on the Canaries and in the opposite districts of Morocco ...”. The source of these comments is not clear ; there are no D. chrysippus from N Africa in the Seitz collection, now in the Senckenberg Museum, Frankfurt am Main (H. Schroeder, pers. comm.) 1912 — RoruscnuiL_p (1913: 114), recorded a male specimen from South Oued Mya (A) [30 April 1912] (another was seen but not captured) ; a female north of Ain Guettara (A) [8-11 April 1912] and a male from Igosten, Tidikelt (A) [15 April 1912] among the butterflies collected by Hartert in an expedition to the central western Sahara. He correctly noted these specimens as the first records for Algerian territory. The specimens remain extant in the Rothschild collection at the BM(NH). 1923 — Demaison (1923 : 134) reported his brother finding a single f. alcippus at Ghardaia (A) on 22 April 1923 and seeing several more on subsequent days in the same place. His brother also apparently found it commonly at Touggourt oasis (A) on 1 May 1923. In April 1927 Demaison’s brother returned to Ghardaia but saw no further specimens (Demaison, 1932 : 93), prompting Demaison to suggest that its occurrence in large numbers in the area in 1923 was due to an “accidental migration”. Apparently all individuals seen then were f. alcippus. He mentioned Gomphocarpus fruticosus (Asclepiadaceae), a known hostplant of D. chrysippus (ACKERY & VANE-WRIGHT, 1984 : 210), as a possible hostplant there. 1943 — Runcs (1945 : 15) noted 6 examples of f. “kanariensis” Fruh- storfer, 1898, [a doubtfully distinct race essentially of the “chrysippus” form (ACKERY & VANE-WRIGHT, 1984 : 115)], and one f. alcippus at 204 Messeied, Rio de Oro in SW Morocco (Western Sahara), taken on 20 January 1943 by Morales Agacino who observed further adults and larvae associated with Calotropis procera (Asclepiadaceae). 1947 — Runcs (1950 : 144-5) stated that despite much travelling in southern Morocco, he failed to find D. chrysippus until December 1947 when he found all stages of the insect abundantly some kilometres west of Taroudannt (M) ; of 38 specimens taken, only 6 were f. alcippus, the remainder being “f. kanariensis” |chrysippus] or alcippoides. The hostplant was Asclepias curassavica f. atropurpurea (Asclepiadaceae) ; RunGs expressed surprise at the apparent rarity of D. chrysippus in view of the availability of a number of different and presumably quite suitable Asclepiad hostplants. 1952 — CHnéoUR (1953) recorded Bede telling him that it was to be found frequently in the Biskra and Ghardaia areas of Algeria and that Kruglik had observed it in some numbers at Nefta (T). He recorded the capture, by Demoflys, of a single f. alcippoides at Gabés (T) on 4 July 1952. He added that since the time of this first observation, other specimens had been seen at Tozeur (T), Ain Draham (T) and Mégrine (T) (by Chpakowsky) and at Barrage de l’Oued Kebir (T) (by Massal and Arnould). This record was repeated (Chnèour, 1954 : 222) and the butterfly was depicted on the front cover of volume 7 of Bull. Soc. Sci. nat. Tunis. 1956 — Wyatt (1956: 220) recorded the species in the Sous valley (Taroudannt) (M) and noted that approximately 30% of individuals seen were f. alcippus. 1970 — Hıccıns & RILEY, in their Field Guide to the Butterflies of Britain and Europe, stated that D. chrysippus was a rare migrant in Morocco and not reported from Algeria or Tunisia ; claims reiterated in subsequent English language editions (1973, 1975, 1980, 1983) and reprints up to 1993. The French language edition of 1988, the only one the author has examined, went some way to correcting these mistakes. 1971 — Owen (1971 : 138 [Fig. 9.2]), included all of Tunisia and most of Algeria and Morocco well into the Sahara desert in the distribution of f. chrysippus in Africa. He showed f. alcippus occupying most of Morocco and part of western Algeria, far removed from the extensive populations of western Africa and went on to say (1971: 140) that maps were prepared after examination of the large collections in the British Museum [Natural History] and in the Hope Department of Entomology at Oxford. This is puzzling since the author was unable to find any D. chrysippus from N Africa other than from Egypt, Libya 205 and Mauretania and those recorded by Rothschild in 1912 in the BM(NH) collections, or in the Hope Department of Entomology at Oxford University Museum. A number of specimens from Morocco and Tunisia have been deposited by the present author in the BM(NH) collection. | 1971 — Kuprna (1972: 268) recorded the capture by Gawadi and Wilson on 1 September 1971 of 1 male and 3 females at Taroudannt (M) where it was common and fresh, having apparently bred locally. 1974 — PıERRE (1974) provided comprehensive discussion on the distribution of D. chrysippus throughout Africa and included parts of Algeria in the distribution of f. alcippus, as well as including W Moroccan and S Algerian records of f. chrysippus. This was based on a number of specimens in the collections of the Museum National d’Histoire Naturelle in Paris, namely Alger, Maison Carée (1, ex. Balachowski — alcippus) ; Touggourt (A) (5, ex. Babault — alcippus) ; Ghardaia (A) (l, ex. Demaison — alcippus) and Timmimoun (A) (1 chrysippus) (J. PIERRE, pers. comm.) 1975 — SAMRAOUI (1993 : 69) noted a specimen captured in Tebessa (A) in 1975, in the INVP collection in Algiers. 1979 — DEVARENNE (1981 : 171) recorded it in March and September/ October 1979 around Ghardaia (A) and figured an extreme male aberration taken in October of that year. 1979 — DE FRIENA (1981) recorded Menrad taking a series at Sousse (T) on 16 December 1979, some of which were f. alcippoides ; he also discussed several possible hostplants. 1979/80 — SCHUURMANS (1981) found D. chrysippus at Sousse (T) in 1979 and recorded Myncke finding it commonly from the begin- ning of August until mid-October at Monastir (T), Sousse (T) and Jendouba (T). 1980 — LEMPKE (1981 : 35) recorded a specimen at Sousse (T) on 2 March 1980. 1980 — van CAPPELLEN (1981) illustrated a female specimen found by Bolland on 11 March 1980 at Sousse (T). 1980 — LttrceEn (1981 : 55) found two examples of the nominate form at Tozeur (T) on 4 October 1980. 1980/1981 — There are specimens in the Museum National d’ Histoire Naturelle, Paris, from near Sbeitla (T) (7, Ex. Barbery & Aubertin, 1980/81 — chrysippus) ; R. Lheureux apparently also observed 3 spe- 206 cimens at Béthioua (A) and Mostaganem (A) in 1980 (J. PIERRE, pers. comm.) 1982 — BURTON (1982) mentioned several hundred D. chrysippus seen by Tombs in the region of Monastir (T) and Sousse (T) from 6 to 10 June 1982. 1983 — Cassar (1983) noted a total of 19 specimens on 10 and 13 July 1983 near Kairouan (T), La Kesra (T) and El Kef (T). 1985 — Between 12 and 24 June 1985, LEGLER (1986) saw a number of specimens in the area of Hammamet (T). 1985 — DEVARENNE (1990 : 154) reported it as common in various localities in Tunisia ; all stages of the butterfly were found in May 1985 between Sousse (T) and Port-el-Kataoui (T) feeding on Pergularia tomentosa (Asclepiadaceae). The butterfly was seen ın “almost every part of Tunisia, including Tunis”. 1985/6 — Cassar (1989) reported a single specimen seen by Balzan at Tunis (T) on 9 April 1985 and scores seen at Douz (T) in early October 1986. He went on to report a male specimen on 9 October 1986 at Touggourt (A) ; this last record was believed to be the first for Algeria. 1988 — Hünı (1988) saw the species near El Djem (T) and Sidi Bou Ali (T). 1988 — STEINIGER & EITSCHBERGER (1990 : 169), noted a specimen seen to the east of Tangier (M) on 25 October 1988. 1990 — Samraour & BENYAcouB (1991) observed a build up of D. chrysippus apparently migrating from east to west in the region of Annaba (A) and El-Kelaa (A) in May 1990; the last specimen seen in Annaba was on 17 October. On 27 August 1990 about 20 specimens were seen at Touggourt (A), also apparently moving in an cast/ west direction. 1990 — Observed to be widespread in small numbers beween Casa- blanca and Marrakech (M) on 3 October 1990; also seen near Tazenakht, 60km SW of Ouarzazate (M) on 11 October 1990 (Martin Jacoby, pers. comm.). 1991 — Samraout et al. (1992) noted the butterfly at Lac Bleu (A) (2) on 28 May 1991, El-Chatt (A) (1) on 14 June 1991, Lac des Oiseaux (A) (1) on 21 June 1991, Biskra (A) (numerous) on 19 July 1991, Laghouat (A) (numerous) on 26 July 1991, Ghardaia (A) (numerous) on 28 July 1991, Touggourt (A) (3) on 5 August 1991 and Taman- rasset (A) (1) on 10 October 1991. Calotropis procera (Asclepiadaceae) was suggested as a probable hostplant in central and southern Algeria. 207 1991 — 1 example was seen at the Gorges de Zeghzel, Berkane (M) on 28 September 1991 and about 15 specimens at the Moulouya delta (M) on 9 October 1991 by G. Chavanon (H.-J. Falkenhahn, pers. comm.). Author’s observations 1991-1993 MOROCCO: On 11 September 1991, a few individuals were seen in different localities on the eastern and northern outskirts of Taroudannt. The following day quite large numbers (40-60) were seen west of the town where there was a large stand of Asclepias curassavica growing in an irrigation ditch (TENNENT, 1993 : 27). On 3 March 1992 the area was revisited when it was found that the irrigation ditches had been cleared of all “debris”, including curassavica plants ; only a single D. chrysippus was seen to the east of Taroudannt. On 11 May, four specimens were seen in close proximity to curassavica plants at Ait Iazza, east of Taroudannt and on 4 June, a single example was seen flying across the main road 24 km east of the town. A further specimen was seen just south of Taroudannt on 2 September, the last visit of that year. In 1993, the locality west of Taroudannt where both the butterfly and hostplant were common in the autumn of 1991,was visited on 26 May. Although curissavica had begun to repopulate the ditch, individual plants were very small and no D. chrysippus were seen. However, at Ait lazza, both the plant and the butterfly were quite common ; with about 20 butterflies observed in a half hour period. Of D. chrysippus butterflies seen in Morocco, some 60-70% were of the nominotypical form, with the remainder being f. alcippus or alcippoides ; the latter were very variable, displaying a white hindwing patch of variable size or merely a few white scales around the inner margin. The author travelled extensively in Morocco in 1992 and 1993 but never saw D. chrysippus other than around Taroudannt and the Sous river valley. It was seen on each visit to the area, even when merely “passing through” ; individuals appeared generally to be freshly emerged. ALGERIA : From 15-17 March 1992, small numbers of D. chrysippus (about 10 in total) were seen close to the town of Ghardaia in central Algeria. Some individuals were possibly in transıt, but others appeared settled in an area west of the town where Pergularia tomentosa (Asclepiadaceae) grew commonly. A few Asclepias curassavica plants seen incorporated in a display along the central reservation of a street on the outskirts of Berriane, 40km north of Ghardaia, were thickly 208 covered in dust ; no D. chrysippus were seen. It was not seen at Toug- gourt some days later, although only a very cursory search was carried out. Two specimens were seen at Ghardaia on 3 October. In 1993, 2 individuals were seen flying aimlessly in the centre of Taman- rasset town on 24 April and a third (or one of the previous two again ?) in the same place the following day. A single example was seen flying across the road 34km south of Ghar- daia on | May. The very small entomological collection at the Institut National de la Protection desVegetaux in Ghardaia contains only three butterflies — 2 male D. chrysippus and 1 male Pieris rapae Linnaeus, 1758 (identified as P napi !) ; according to an official of the Institute, D. chrysippus is a regular visitor to the Institute gardens. Without exception, butterflies seen by the author ın Algeria have been of the nominotypical form. TUNISIA : On 4 October 1992, single specimens were seen at Degache and Tozeur ; the butterfly was very common in an oasis on the southern outskirts of Douz, about 150 individuals were seen in one afternoon including seven pairs in copula. Two larvae were noted on (?) Cynan- chum sp. (Asclepiadaceae). One freshly emerged male f. alcippus was seen ; the remainder of those seen at Douz and subsequently elsewhere in Tunisia, were f. chrysippus. On 5 October it was also common (in- cluding one pair seen in copula) in cultivated areas on the western outskirts of Gabes ; the following day two were seen flying in Gabes town centre and singletons seen at Oudre ; in the desert 42km west of Gabes ; El Guettar and Gafsa. On 7 October it was flying quite commonly in a dry river bed on the outskirts of Moulares and 3 days later, three were seen ca. 30km north of Sbeitla. In 1993, a year when the Spring season was delayed, one D. chrysippus was seen on the eastern outskirts of Gafsa and another near El Guettar on 30 March ; on 5 April a single specimen was seen at Douz. In October 1992, at the Institut National Agronomique de Tunisie (INAT) in Tunis, a display case containing preserved stages and host- plant (Pergularia tomentosa) of Danaus chrysippus was seen in the foyer of the zoology department, apparently prepared by Mon- sieur Hedi Smiri, an amateur lepidopterist working at the Institut National de la Recherche Agronomique de Tunisie (INRAT) in Sfax. On 23 March 1993 the author met Monsieur Smiri who confirmed that he had been breeding D. chrysippus regularly during the previous ten years from early stages found locally ; a search failed to find any 209 pupae, the only stage we might have expected to see since the butterfly had not yet emerged that year. Present Status Danaus chrysippus ıs a strongly migratory butterfly, with individuals often wandering far from established colonies. However, there is strong evidence that the species has established breeding colonies in Morocco, Algeria and Tunisia within the last 50 years. MOROCCO : Danaus chrysippus is clearly well established in the Sous valley, where it has almost certainly been resident since 1947 or earlier. It seems likely that the Morocco populations originated from western Africa ; forms alcippus and alcippoides form a significant percentage of the population. Although other Asclepiads are available, particulary in the south of the country, the only confirmed hostplant utilised by D. chrysippus in Morocco is Asclepias curassavica, introduced into Africa as a garden plant from Tropical America (Owen, 1971 : 34). The plant, which grows as a weed on the edges of fields and irrigation ditches around Tarou- dannt, seems to be irregularly but routinely destroyed by the local farmers and this probably has a direct bearing on the fluctuation of the D. chrysippus population. The botanist J. Gattefossé appears to have been the first to find curassavica in the Taroudannt area of Morocco in 1940 (Gattefossé, 1941 : 214). ALGERIA : It is interesting that the “invasion” of D. che to the Ghardaia and Touggourt areas recorded by Demaison in 1923 consisted exclusively of f. alcippus, whereas those seen in recent years have all been (where recorded) of the nominate form. Clearly, the butterfly has established for some