$ he 3 : rr ae . “e ‘ . ’ water Fe We ba E . oo eve ' : eC on. ee : : ; os 8'6 : % At aoe : , ae meg sche Tray : exe ™ . ak 2 oe Co . ' 3 Pay tacatgia's Vue wor es a ee Vi he.5 U4 Bi eco G&A: ven, er ae EASES BY be, Biter 2 2 Oe Wit ag Stake, RAEN Nee “se Vee eae ah te ee aite vebe vary fire We Coe ett TaN (re ee he ce eer ete wee oe Peer oo 2 whee ee ee eo ee er en ey + A AIR ty Weert Pi heotaa saree’ & a7 ’ tbe When ho me thet h zee ale Lola tin Mae as KEN UV Bet oes OT OR tee & Ww Cau eho ke Petree tre eo “Hah OO ve PVE I (owe ta tem * ee -&% & Aer kk 4 4) OH Rete ee bE at te EN mh Rh ee ee ee Cee : a a Seren ms ae sige eee VS tet ye eee Ste he eke er, wae Ete Be Ree eRe ee +28 werd hi Ya He we Ra ; seas. 0 Ab Eies'b YP ewes ate » Petey & 5 EERE SY OLY Pie ee ee c fae eee «Myre Lon ten ees WEEN Ae eet see tc ey ieee a. +8 es Pelee. Bee ee eee ee (OP ate eer on ae ere PO a Ot OEY re ee ee = a ae ete enero” wren Ser re ee ee en ee Fp en enn gh ot mace Sm ae ee ee SS PEP OP RRM RH we. Poa he os ee eee pre eae ere ee le ane aed we ie ge te re pe ee ROP he ER Pe ne a ee od al dats ee OR OO FOE EM Or E oO & sum svereg 07m =) eet = po reper rip wee Pe reper et woe ee Cel ide eedind Veber eter Hee Le we Oe et ee ated : mee pease eee ee eee re a ee oo he ote ee ae Rs Wank ere de SASS Ree rt Soo ee er Pe ae be - PAS ee Here wwe OT ee a atte sie ae 2 HP mee 9:99 BS Here nas eN Pint be a2 ROP ae Ae oe dae ae pais Peeper pie wee ae Rose Rie ater et 2 ” #60 Bae. oe bp wes «ere peeks HARVARD UNIVERSITY e Library of the Museum of Comparative Zoology hat © €)1+09 91 ae “Vege Y =F: .4 ’ = 7 . 7 -_ - - Le ; - Ey . 7 7 _ 7 7 is e. 7 _ - : ; bet i | 5 _ _ ' vy 7 a ‘ iy Fe , 7 The University of Kansas SCIENCE BULLETIN Vol. 51, nrs. 12-25 12. Comparison of the Crane Flies (Diptera: Tipulidae) of Two Woodlands in Eastern Kansas, with a Key to the Adult Crane Flies of Eastern Kansas by Chen-Wen Young 13. Sexual Size Differences in the Genus Sceloporus by Henry Fitch 14. A Comparative Anatomical Study of Mandibular Structure in Bees by Charles D. Michener and Anne Fraser 15. Ecology and Exploitation of Ctenosaura Similis by Henry S. Fitch and Robert W. Henderson 16. The Classification of Halictine Bees: Tribes and Old World Nonparasitic Genera with Strong Venation by Charles D. Michener 17. A New Genus of Cryptodiran Turtles (Testudinoidea, Chelydridae) from the Upper Cretaceous Hell Creek Formation of Montana by Kenneth N. Whetstone 18. The Effect of Slope-aspect on the Composition and Density of an Oak-Hickory Forest in Eastern Kansas by Rodney Birdsell and J.L. Hamrick 19. Pollen Manipulation and Related Activities and Structures in Bees of the Family Apidae by Charles D. Michener, Mark L. Winston, and R. Jander 20. Summer Crane Flies of Lake Itasca Vicinity, Minnesota by George W. Byers 21. Protozoa from Acid-bog Mosses and Forest Mosses of the Lake Itasca Region (Minnesota, USA) by Eugene C. Bovee 22. The Proboscis of the Long-tongued Bees: a Comparative Study by Mark L. Winston 23. Some Effects of Adenosine Triphosphate on Swimming and Photophobic Response of Euglena gracilis by Mercedes L. Acuna and Eugene C. Bovee »’ « Ni r i] ;) a vr ie - ' tf Pg-65) sen le tov st take 4 ‘anne Yo A ae es) ort To (Coshifdgil soreintG) sik shat) og 7 roePycgtie? .F' 943 of ye) 5 cote ABRAM RTARTA miVehos [tioel | ow anniek ntegens to eo k PR pmesd 20nd yi UOY wearin yal absocesonr bite) or ak ee ores HO esre beimee yl moaks vote uc otusourne Gfudthis’ to vbuste, feninotenk Avi semesie® hy af! . a ee a ee ‘aon i TopeTa ood bas reiuetlon’ A! ce rreHa wo eflinic ervssenes0 AceHofisiiotg¢d baa “soled .2/ GostabaoW ii Jroddh hoe, toda) <2 Aaa! ya Bis ssoint tanohoanivoll sto nmoitegnittergho eT 12! aortens¥ qqotise titw serene ststamrenn b.brabh bt) reirie (try Fu yt a feed yi cohfon!iwice?) relia) anxibetqyid te aanee! wan A) 4X! nose Iho ayowpetan Tete’! ofi norh Cashier alert » Xe Shae Ri BIGUN Tey OTT sar] earn eee WA wet bioa yi has Ae.cgLeogwoo- slid this toéhee-eugih Ie’ Jost A sat #! SGenke Medved ott tmogol we Pl g MOFWOGM vl! Hoy Trotter it yarheh ve Hag, ea ks vt IbA fbiaieiet bine tottafuc tiie fel foi ev f | i” asbinh wiimat old Vo egal mi estudauegg Te0ne!. .f pe «roceth yi eh .veteroit 6 ae lhrentd yt! 5 BIOBs Tart fy Sap oles) i\ Boe ec! } +] 4 Bil tH o£ e i ia f por} “reg lite o! j otarye. at Shey Later vel 126764 Jott, e-eooe aol biok- mor! sugojeyvd 15 Chay qs IORSNNTE) poses! eoens! ots Sevee 20 one gus vil Sy Jai bamod sn Ysqnd ye ORO ISH om) ofa Yo efoeqdath wit so . yhusre ot ene 4 Roti es iti fee ine ederigeode t+? sateouehbé eet eg, emee >. ¢ 2h Ciperws. (ite lou) too aetionae! oftotaoiol® hae SovOt 4d orieier) hes SUNS uh Re bent at vd 24. 2S. Alleviation of the Toxicity of Copper to Tetrahymena Pyriformis by Nontoxic Iron by Eugene C. Bovee, David A. Kegley, David Sternshein, Ellen Wyttenbach and Barton L. Bergquist Population Dynamics and Production of Daphnia Ambigua in a Fish Pond, Kansas by Chi-Hsiang Lei and Kenneth B. Armitage JUL 19 1978 COCO OO EEE OOE OEE OTES OSES TEESE HEHE TEESE OES OOo 8 ow we www eee a een RN I I I SI i a OC SCIENCE BULLETIN COMPARISON OF THE CRANE FLIES (DIPTERA: TIPULIDAE) OF TWO WOODLANDS IN EASTERN KANSAS, WITH A KEY TO THE ADULT CRANE FLIES OF EASTERN KANSAS Tittoaeaanatceteinmemmmmnenanen . et Po . ee < < < « 1. < tote i ‘se me Lm ate* atet atet me tata atat a”, * XP RRR "eee pe e¢.¢.¢. 55 8 a I I ST TI RR MRM By CHEN-WEN YOUNG OA RR I RR RN eatetdreetete oceetece Soberacececececececoceces Vol. 51, No. 12, pp. 407-440 June 30, 1978 eR I OR OKA ANNOUNCEMENT The University of Kansas Science Bulletin (continuation of the Kansas Uni- versity Quarterly) is an outlet for scholarly scientific investigations carried out at the University of Kansas or by University faculty and students. Since its incep- tion, volumes of the Bulletin have been variously issued as single bound volumes, as two or three multi-paper parts or as series of individual papers. Issuance is at irregular intervals, with each volume prior to volume 50 approximately 1000 pages in length. Effective with volume 50, page size has been enlarged, reducing the length of each volume to about 750 pages. The supply of all volumes of the Kansas University Ouarterly is now ex- hausted. However, most volumes of the University of Kansas Science Bulletin are still available and are offered, in exchange for similar publications, to learned societies, colleges and universities and other institutions, or may be purchased at $20.00 per volume. Where some of these volumes were issued in parts, individual parts are priced at the rate of 2 cents per page. Current policy, initiated with volume 46, is to issue individual papers as published. Such separata may be purchased individually at the rate of 3 cents per page, with a minimum charge of $1.00 per separate. Subscriptions for forthcoming volumes may be entered at the rate of $20.00 per volume. All communications regarding exchanges, sales and subscriptions should be addressed to the Excuancre Liprartan, UNIveERsITy oF Kansas Lreraries, LAwreNcrE, Kansas 66045. Reprints of individual papers for personal use by investigators are available gratis for most recent and many older issues of the Bulletin. Such requests should be directed to the author. The International Standard Serial Number of this publication is US ISSN 0022-8850. Editor Eugene C. Bovee Editorial Board William L. Bloom Phihp W. Hedrick Rudolf Jander Harvey Lillywhite Charles D. Michener Norman A. Slade Henry 1D. Stone George W. Byers, Chairman THE UNIVERSITY OF KANSAS SCIENCE BULLETIN Vol. 51, No. 12, pp. 407-440 June 30, 1978 CoMPARISON OF THE CRANE Figs (Diprera: TIPuLiDAe) oF Two WooDLANDs IN EASTERN Kansas, Wirt A Key To THE ADULT CRANE Fires oF EastERN Kansas’ CHEN-WEN YOUNG TABLE OF CONTENTS PAGE ISUsHiRR 1) GaTs Nee ne ema i Bg i ea Ho) te ee 408 (GENERATED EATURES OF “LHESOTUDY “AREAS 2. occc.-cccee0c¢2ctescccecatecesessee-catcusoceesesutece teacveseceeens 408 IGN tava a A BIATG ee ee te eels. De a eh ES ieee Leer ee ee 409 11S TRTEETOUDS, © lee rta end Cate 8 hon oe ae er NINE Ee ES 411 SAGO, TDG UES GN TGIN fees eee ee eee eye reese 2 ee aes eae 41] RELATIVE ABUNDANCE OF SPECIES ........-..--.--:--c-:c-c0-ce-o0es G16 Ticats OR de ee 412 PRN TPG DIMI TSE OE GOD GLE Sty sie tee oss OO a ns te ee ae cuore ee 412 BBS S10 ee ee le Fe Sec oe Hs ae ee 428 PRE ACEUO REG Leimert eee sete bec es oe a ede 0 eee ce 431 Aprenpix—A Key to the Adult Crane Flies of Eastern Kansas ...............------------- 431 1. Contribution no. 1658 from the Department of Entomology, University of Kansas, Lawrence, Kansas 66045. This investigation was supported in part by University of Kansas General Research allocation #3677- x038, Prof. George W. Byers, principal investigator. 408 Tue University oF Kansas SCIENCE BULLETIN ABSTRACT Data on geographical and ecological distribution have been compiled for the 95 species of crane flies found in two natural areas in northeastern Kansas, the Breidenthal Reserve and the Natural History Reservation of the University of Kansas. These distributions correlate well with various major plant communities, or with specific habitats. The ranges of several species are ex- tended significantly westward. New records for species not previously recorded within 500 miles of Lawrence, Kansas, include: Nephrotoma alterna, Tipula (L.) perlongipes, Limonia (D.) haeretica, Dicranoptycha elsa, D. megaphallus, D. septemtrionis, Gonomytia (L.) manca, Ormosia arcuata, O. ingloria, Tasiocera (D.) ursina and Molophilus pubipennis. INTRODUCTION This paper reports the results of inten- sive field studies of the crane fly popula- tions of two natural areas belonging to the University of Kansas, the Natural History Reservation and the Breidenthal Reserve. The primary objective was to determine and compare what species comprise the crane fly faunas of the two reservations, and to ascertain their relative abundance, seasonal distribution, and habitat correla- tions. When species seemed to be excluded from one of the reservations, the supposed- ly appropriate habitat was examined closely in an attempt to determine what ecological differences between the two reservations might account for the absence. The combined fauna was examined to determine the probable origins of eastern Kansas crane flies. Comparative data were obtained from similar studies made by J. Speed Rogers (1942) at the George Reserve in southeastern Michigan, and B. A. Foote (1956) in Delaware County, Ohio. Overall range data were obtained from A Catalog of the Diptera of America North of Mex- ico (Stone, et ali, 1965). Included’ are. a brief account of the habitat, abundance, etc., of each species encountered and a taxonomic key to local species. Until this study, none of the following species had been reported within a 500 mile radius of Lawrence, Kansas: Nephrotoma alterna, Tipula (L.) perlongipes, Limonia (D.) haeretica, Dicranoptycha elsa, D. megaphallus, D. septemtrionis, Gonomyta (L.) manca, Ormosia arcuata, O. ingloria, Tasiocera (D.) ursina, and Molophilus pu- bipennis. GENERAL FEATURES OF THE STUDY AREAS EasTERN Kansas Kansas is at the geographic center of the contiguous continental United States. Ex- cept for its hilly eastern part, it is a plain sloping gradually downward to the east and is usually regarded as a prairie state. However, within its borders there are ap- proximately 1,358,000 acres of forest, mostly in the eastern hills, where edaphic condi- tions and an annual precipitation of ap- proximately 34.75 inches are adequate for upland forests. Some expansion of forests westward has followed agricultural activ- ity and the cessation of prairie fires. East- ern Kansas is in a transition zone between the eastern forests and the central plains. Douctas CounTy Douglas County is in northeastern Kan- sas. Its main topographic features are the Kansas and Wakarusa river valleys, drain- ing eastward, and uplands formed by dif- ferential erosion of the nearly horizontal beds of limestone, shale and sandstone. In the eastern part of the county, plains have Tue Crane Fires in Eastern Kansas 409 developed on glacial deposits. The climate is humid, with cold winters and warm to hot summers, three-fourths of the annual precipitation falling during the warm sea- son from April through September. Most of the county is now agricultural land. THE BrEIDENTHAL RESERVE The Breidenthal Reserve (Baldwin Woods) comprises 110 acres of mixed broad-leaf deciduous forest in southern Douglas County, about 2 miles north of Baldwin City and 15 miles southeast of Lawrence. It includes part of the steep slope along the south side of the Waka- rusa River Valley and is drained by Coal Creek. The area is mostly oak-hickory forest, with variation depending upon the locality and drainage. The creek is nar- towly bordered with flood-plain species of trees. Low oak-hickory forest occurs regu- larly on the moderately mesic hillside and north-facing slopes, while drier, upland, oak-hickory forest grows on the south- facing slopes. Tue UNiversity oF Kansas Natura History RESERVATION The University of Kansas Natural His- tory Reservation is at the north edge of the Kansas River Valley in northern Douglas County, about 5 miles northeast of Law- rence. Two intermittent brooks drain its 590 acres, one in the northwestern part, the other in the southeastern part. A pond, formed by damming of the northwestern brook, has a small swamp around its north- eastern edge. Most of the Reservation sup- ports broaf-leaf deciduous forest, with scattered grassy openings. This area had been heavily grazed before it became the Reservation in 1948, its present vegetation is still in a successional state. CRANE FLY HABITATS Some adult crane flies in the two study areas are restricted to a single type of habi- tat, while others are widespread. Humid- ity and temperature affect flight and other activities, so that distribution of adult flies only approximates that of the larvae; yet there is usually a close correlation. Six, general, crane fly habitats exist in the study areas. Following the description of each, below, is a list of the species of crane flies recorded from it. BoTTOMLAND Forest The Breidenthal Reserve has bottom- land forest along Coal Creek that contains mixed species of oak and hickory (Quercus macrocarpa, Q. alba, O. rubra, Carya ovata, C. cordiformis), sycamore (Platanus occt- dentialis), and American elm (Ulmus americana). The shrub layer is poorly de- veloped. The vernal herbaceous flora is very well developed. Undergrowth in- cludes nettle (Urtica procera), and jewel- weed (Impatiens capensis), with sparse patches of grass. There is sporadic flood- ing in spring and early summer, and a dry season in late summer. This is a mesic area. Crane flies present are: Dolichopeza tridenticulata, Nephrotoma alterna, N. eu- ceroides, N. macrocera, N. polymera, Tip- ula trivittata, T. duplex, T. flavoumbrosa, T. mallochi, T. ultima, T. furca, T. say, T. strepens, Limonia fallax, L. globithorax, L. tristigma, L. divisa, L. pudica, L. domes- tica, Helius flavipes, Dicranoptycha sep- temtrionis, D. tigrina, Epiphragma fascta- pennis, E. solatrix, Pilarta tenuipes, Cla- dura flavoferruginea, Gnophomyta tristis- sima, Gonomyia florens, G. manca, G. sulphurella, Erioptera cana, E. vespertina, E. caloptera, E. needhami, E. armata, Or- mosia romanovichiana, Tasiocera vrsina, Molophilus pubipennis. GRASSLAND Grasslands in which Tipulidae occur in the two study areas are ecotone areas be- tween small grassy fields and the edge of woods. These are too dry for most crane 410 Tue University oF Kansas ScreNcE BULLETIN flies, but some species whose larval stages feed on the roots of certain herbaceous plants growing in the grassland occur as adults in the ecotone of the grassland and woods. Such species are: Nephrotoma fer- ruginea, Tipula bicornis, T. flavibasis, T. triplex, T. paterifera, T. ultima, Erioptera cana, E. septemtrionts. Oax-Hickory Forests The oak-hickory forests of the Breiden- thal Reserve are in stable climax, domi- nated by shagbark hickory (Carya ovata), chestnut oak (Quercus muehlenbergit), red oak (Quercus rubra), and red elm (Ulmus rubra). Understory trees include ironwood (Ostrya virginiana), red mul- berry (Morus rubra), redbud (Cercts cana- densis), and pawpaw (Asimina triloba). The undergrowth is generally open with some growth of poison ivy (Rhus radt- cans), wild gooseberry (Ribes missourt- ense) and buckbrush (Symphoricarpos or- biculatus). Leaf litter and, beneath it, leaf mold are present over most of the level areas and gentler slopes. Fallen trees and decaying limbs there are usually too dry to harbor most crane fly larvae. Mats of dry mosses are scattered on the ground surface. Oak-hickory forests are too dry for most crane flies; few species are present: Nephrotoma ferruginea, Tipula disjuncta, T. duplex, T. fuliginosa, T. perlongipes, T. tuscarora, T. stonet, Limonia triocellata, L. domestica, Dicranoptycha elsa, D. mega- phallus, D. septemtrionis, Cladura flavofer- ruginea, Gonomyta subcinerea, Erioptera cana. STREAMS The small, intermittent brooks on the Natural History Reservation ordinarily flow only in spring and early summer. They dry up quickly by mid-summer, al- though pools may persist in the stream beds for several weeks after flow has ceased. Coal Creek in the Breidenthal Re- serve flows from early spring until August, and has small pools which persist through September. A few, small, ravine tribu- taries feed into it in spring. Several species of crane flies inhabit the banks and beds of these streams. Their larvae probably occur in submerged, rotten, tree branches, and in the algae or mosses covering rocks in and along the streams. These are: Do- lichopeza obscura, D. tridenticulata, D. walleyi, Tipula furca, T. strepens, Limonia rara, L. annulata, L. humidicola, L. bry- ant, L. domestica, L. lecontei, L. commu- ms, Erioptera armata, Ormosia romano- vichiana, Tastocera ursina, Molophilus pu- bipennis. SwaMPp The artificial pond on the Natural His- tory Reservation supports a small swamp of recent origin around its northeastern edge. Tree species common here are willows (Salix spp.), American elm (Ulmus amer- icana), and honey locust (Gleditsta triacan- thos). No shrub stratum is developed. The water rises and covers most of the area during spring and early summer. At the low-water stage the area is almost com- pletely covered by smartweeds (Polygo- num hydropiper). The moist-to-wet soil is the preferred habitat for the larvae of sev- eral species of crane flies. They are: Neph- rotoma alterna, N. eucera, N. euceroides, N. macrocera, N. polymera, Tipula igno- bilis, T. ultima, I. furca, T.strepens, @. tricolor, Limonia tristigma, L. pudica, L. domestica, Ducranoptycha pallida, Epi- phragma fasciapennis, E. solatrix, Pilaria imbecilla, P. quadrata, P. tenuipes, Atarba picticornis, Teucholabis lucida, Erioptera vespertina, E. caliptera, E. parva, E. graph- ica, Tastocera ursina, Molophilus hirtipen- nis, M. pubipennis. Vatiey Hitisipe Woops In the Breidenthal Reserve, low oak- hickory forests grow on well-drained soils Tue Crane Furs 1n Eastern Kansas 411 on the gently sloping, moist to mesic, north-facing hillsides. Ground cover spe- cies include tick trefoil (Desmodium glu- tinosum), Virginia creeper (Parthenocissus quinquefolia), and mayapple (Podophyl- lum peltatum). The valleys in the Natural History Reservation are wooded with American elm (Ulmus americana), honey locust (Gleditsia triacanthos), black walnut (Jug- lans nigra), and osage orange (Maclura pomifera), with dogwood (Cornus drum- mondu) and wild plum (Prunus amert- cana) forming the understory. Shrubs present are poison ivy (Rhus radicans), wild gooseberry (Ribes missouriense), buckbrush (Symphoricarpos orbiculatus), and brambles (Rubus spp.). The herb stratum is sparsely developed, only where the tree canopy has been broken. In the spring, these two previously de- scribed habitats are moderately mesic. Fallen tree trunks in various stages of decomposition, and soil covered by damp and friable humus form a habitat for larvae of several species of crane flies. These in- clude the following: Dolichopeza triden- ticulata, D. walleyi, Nephrotoma alterna, N. eucerotdes, N. macrocera, N. virescens, Tipula trivittata, T. dietziana, T. dorst- macula, T. duplex, T. flavoumbrosa, T. fuliginosa, T. mallochi, T. morrisont, T. translucida, T. integra, T. unimaculata, Limonia cinctipes, L. fallax, L. immatura, L. divisa, L. haeretica, L. immodestotdes, L. liberta, L. pudica, L. domestica, Dicran- optycha elsa, D. megaphallus, D. pallida, D. sobrina, Epiphragma fasciapennis, E. solatrix, Pseudolimnophila contempta, P1- laria tenuipes, Atarba picticornis, Elephan- tomyia westwood, Cladura flavoferru- ginea, Teucholabis complexa, Gonomyia subcinerea, Erioptera cana, Ormosia in- gloria. METHODS This report is based largely upon field observations and collections made in the two study areas described. Collections were made from May 1974 to September 1975 as often as time permitted. Through- out the peak-emergence periods of adult flies, collections were conducted on alter- nate days at the Breidenthal Reserve and the Natural History Reservation. At first all varieties of habitats were searched. Later, searches were concentrated on sev- eral specific sites believed to represent the principal types of crane fly habitats. Net sweeping was the basic method of collecting. All possible hiding places were explored with the net. Specimens were taken either at their resting site, or in the air as they fled. No more than 20 speci- mens per species were captured at each habitat on each trip. All flies caught by net were mounted. In the spring of 1975, a Malaise trap was set in the bottomland along the creek near the mouth of a ravine on the Breiden- thal Reserve; another was set in the swamp woods at the Natural History Reservation. Each was checked at five-day intervals, and all flies trapped were kept in alcohol. Ecological information recorded about the specimens included weather conditions at time of collection, type of habitat with which flies were associated, microhabitat where flies were taken, relative abundance of each species, data on behavior of the flies (when appropriate), and other note- worthy details. Larvae were taken and rearing data were obtained whenever pos- sible. SEASONAL DISTRIBUTION In eastern Kansas adult crane flies usu- ally emerge during wet seasons. At both reservations their appearance begins with Erioptera cana in late March, reaches a peak period in May and early June, and ends with Cladura flavoferruginea in mid- October. Winter crane flies, Chionea sto- neana, appear in January. 412 Tue University oF Kansas SciENCE BULLETIN Many species have single, short, clear- cut adult seasons and can be classified as univoltine, spring, summer, or autumnal forms, according to the flight periods of the adults. However, some species complete two generations per year and have two flight periods. Temporal disjunction be- tween generations is either total or partial. For example, adults of Tipula (Yamato- tipula) furca first emerge in late April, disappear in summer, and recur in early September, while adults of Nephrotoma macrocera can be found on the wing from May until September. Multivoltine species such as Erioptera cana, fly as adults from March to October. Temporary fluctuations in environmental factors markedly affect the number of generations during the sum- mer. Table 1 summarizes seasonal distribu- tion of adults. In the table, each month is divided into ten-day intervals, and the sym- bols ‘X’, ‘x’, and ~’ are used to indicate, respectively, the peak period (common), the intermediate period (numerous), and the period of least abundance (rare) with- in the species. At the right, local occur- rence of the species is summarized; Br.— Breidenthal Reserve; and NH.—Natural History Reservation. RELATIVE ABUNDANCE OF SPECIES Each major habitat was visited for about half an hour on each collecting trip. Categories indicating relative abundance and distributional pattern were modified from Rogers (1942). Abundance is ex- pressed as the number of flies that could be caught per half-hour: abundant—more than 20; common—1l0-20; numerous— fewer than 10; rare—one only. Ecological distribution: widespread—present in four to six habitats; general—continuous in two adjacent habitats; local—restricted to one habitat. ANNOTATED LIST OF SPECIES In this annotated list, the brief account of each species includes its habitat, flight period, relative abundance, and previously known range. Br.—Breidenthal Reserve; NH.—Natural History Reservation. 1. Dolichopeza (Oropeza) obscura (Johnson) Br. 1975; May 22-June 10. Also taken by G. W. Byers on August 30, 1961; bi- voltine. Only three specimens were taken in 1975, found associated with other species of Dolichopeza under a wooden culvert in a woodland ravine. Previously known range: Alberta to Nova Scotia, southward to Arkansas and Florida. 2. Dolichopeza (Oropeza) tridenticulata Alexander Br. NH. 1974-1975; May 23-June 26, and a single male on August 7, 1974; bivoltine. Common, to locally abundant in well- shaded, mesic situations, such as beneath a wooden culvert, under protruding rocks along stream bed, and in the shade of up- turned, shallow, tree roots. Usually the flies were found hanging from the roof of their nesting places with the hind legs pendant. Previously known range: Manitoba to Quebec and Maine, southward to Missouri and Georgia. 3. Dolichopeza (Oropeza) walleyi (Alexander) Br. NH. 1974-1975; May 22-June 17, and July 16-August 7; bivoltine. This species is common in May and June, but rare in August. Most specimens were from wet, well-shaded places. One rich collecting spot was the wooden culvert in Br. Adults were usually taken where tridenticula was also common, or resting in low shrubs and herbs. No. 8. oF 10. Nephrotoma . Nephrotoma . Nephrotoma Tue Crane Fiies in EAsTERN Kansas TABLE IL Summary of Seasonal Distribution of Adults Mar. Species 2 Apr. May June July Aug. Sept. Oct. 413 3 23) 2rd 232 (2) 32 See Ss Bras NE. Dolichopeza (O0.) obscura (Johnson) Dolichopeza (0.) tridenticulata Alex. walleyi (Alex.) a Dolichopeza (0. ) alterna (Walker) Nephrotoma eucera (Loew) euceroides Alex. Nephrotoma ferruginea (Fabr.) Nephrotoma macrocera (Say) Nephrotoma polymera (Loew) virescens (Loew) ll. Tipula (P.) ignobilis Loew 12. Tipula (P.) trivittata Say 13. Tipula (L.) australis Doane 14. Tipula (L.) bicornis Forbes 15. Tipula (L.) dietziana Alex. 16. Tipula (L.) disjuncta Walker 17. Tipula (L.) dorsimacula Walker 18. Tipula (L.) duplex Walker 19. Tipula (L.) flavibasis Alex. 20. Tipula (L.) flavoumbrosa Alex 21. Tipula (L.) fuliginosa (Say) 22. Tipula (1..) mallochi Alex. 23. Tipula (L.) morrisoni Alex. 24. Tipula (L.) perlongipes Johnson 25. Tipula (L.) species near T. perlongipes 26. Tipula (L.) translucida Doane Bie Lipula (L..)\ triplex Walker 28. Tipula (L.) integra Alex. 29 Mipula (iz-) tuscarora Alex- 1). Tipula (P.)-paterifera Alex. 31. Tipula (P.) ultima Alex. 32. Tipula (Tr.) unimaculata Loew) 33, Tipula (Tr.) stonei Alex. 34. Tipula (¥.) furca Walker --xXXx-- = —xxXXx—- =xXXxXxx-— = —-xXXXx- --x- -xx- —-xx-—- =KXXXXX— --x- -xX- ---xXx--- --x--- -xx- x x Tue University oF Kansas ScreNce BULLETIN Maes apes May June July Aug. Sept. ce, No Species LT 2) Be 2 3 2 3) 12) 8 e232 2S BiaNie 35. Tipula (Y.) sayi Alex. - x4 36. Tipula (Y.) strepens Loew ----xXxx---- ---Xx- x x 37. Tipula (Y.) tricolor Fabricius Sam x 38. Tipula (L.) globithrox (O. S.) Sie x 39. Limonia (L.) nara (0. 5.) ==XKKUTS7- 000 OS x x 40. Limonia (L.) tristigma (0. S.) NK — x x 41 Limonia (M.) cinetipes (sav) --- x x 42. Limonia CM.) fallax (Johnson) --XKXX----- SSS x x 43. Limonia (M.) immatura (0. S$.) saae x “aq LAmonia Gs) eriocel Latar Coss S.) anna Tg cia x x 45. Limonia (Dis.) annulata (Linn. ) = x AG wWwimenia CD.) ditisss Vex. 0 NO ati x x 47. Limonia (0.) haeretica (0. 5S.) SS he aS= x 48. Limonia (D.) humidicola (0. 8.) eS x x 49. Limonia ().) immodestoides Alex, ©2222 2 22 2 2 2 2 2 2 2 wee eeneee- eee x 50. Limonia (D.) liberta (0. S.) = =¥XXK==—= x x 51. Limonia (D.) pudica (0. S.) ===xXKx= x x 52. Limonia (R.) bryanti (Johnson) - x 3. Limonta (Ra) domestica tO] S.) 92 ee XXxxXxXX--- eee sye Ss x x 54. Limonia (R.) lecontei Alex. - x 55. Limonia (G.) communis (0. 5.) Sea Ges eee = x x DO. Helaus: (He) flavipes (Macquart) :°;©£2£2£2; 2; 2 2 2 2 aeeeee ee = x x 57. Dicranoptycha elsa Alex. =x- x x 58. Dicranoptycha megaphallus Alex. --xXx- x x 29- Di¢ranoptycha pallida Alex, © 2220202200 nee x 60. Dicranoptycha: Septemtrionis Alex, Seige = S55 x 61. Dicranoptycha sobrina O. S. --- x x 62. Dicramoptycna tigrina Alex, nnn nnn x 63. Epiphragma fasciapennis Alex. == XXX KK==— x x 64. Epiphragma solatrix (0. S.) SS 2 ee x x 65. Pseudolimnophila contempta (0. S$.) Se 00 ee x 66. Pseudolimnophila luteipennis (0. S.) i ——— x x 67. Pilaria imbecilla (0. S.) =x = x 68. Pilaria quadrata (0. S.) --xXx------ x 69. Pilaria tenuipes (Say) 990000 22 wwe ne XXXXXXX---- XX 70. Atarba (A.) picticornis 0. S. SSS eS XX Tue Crane Fiies 1n Eastern Kansas No Species P2 71. Elephantomyia (E.) westwoodi O. S. 72. Cladura (C.) flavoferruginea 0. S. Chionea stoneana Alex. 74. Teucholabis (T.) complexa 0. 75- Teucholabis (T.) lucida Alex. 76. Gnophomyia tristissima 0. 77. Gonomyia (G.) florens Alex. 78. Gonomyia (G.) kansensis Alex. 79. Gonomyia (G.) subcinerea 0. 80. Gonomyia (L.) manca (Ones a) 81. Gonomyia (L.) sulphurella 0. 82. Erioptera (S.) cana (Walker) 83. Erioptera (E.) septemtrionis O. S. Erioptera vespertina 0. S. . Erioptera caliptera (Say) . Erioptera (M.) needhami Alex. Erioptera parva O. S. n 3. Erioptera (H.) armata O. . Erioptera (P.) graphica O. S. - Ormosia arcuata (Doane) 2. Ormosia romanovichiana (0. S.) . Tasiocera (D.) ursina (0. S.) - Molophilus hirtipennis (0. S$.) - Molophilus pubipennis 415 wr. May June July Aug. Sept. Oct. SE 2) SSeS ee Se Sle 2 oes) 2 Br) INA Ss SRE Soe x x =—XXx— x x = x = Se BS See x x —- SS = x x = x = x Noe SeecSSeese as es a x = x aS > x x =SKXKKS SS = S6 = x x aed SS: x = x SOO OCR 0 SSS x x ape =A x = x = x = 32 x =Kx= x x SS SSeS x Sea —— x x Larvae were taken on April 3, 1974, from a thin carpet of moss covering a sod- den, decayed log lying on the forest floor below the dam in NH. Previously known range: Alberta to Nova Scotia, southward to Kansas and Florida. 4. Nephrotoma alterna (Walker) Br. NH. 1974; June 5-July 16, with a single record for August 31; bivoltine. Numerous in the bottomland woods, and in the moist thickets along streams; a few individuals from the low oak-hickory forests. On June 28, 1974, two females were found ovipositing into wet soil at the foot of a slope, about three feet from the margin of Coal Creek. Previously known range: Michigan to Nova Scotia, southward to Florida. 5. Nephrotoma eucera (Loew) NH. 1974; June 13. Rare, one male taken from the tall, 416 Tue University oF Kansas ScrENCE BULLETIN luxuriant herbage around the swamp. Previously known range: Wisconsin to Quebec and Massachusetts, southward to Kansas, Tennessee, and Virginia. 6. Nephrotoma euceroides Alexander Br. NH. 1974-1975; May 23-June 26; uni- voltine. Fairly common in grassy areas around the swamp, and in the bottomland woods where wood nettles and jewelweeds grow luxuriantly, less common in mesic hillside woods. When alarmed, flies up into the trees. This is the most common species of its genus found in the NH; larvae not found at Br. or NH. but were taken at Hole-in-the-Rock, 12 miles south of Law- rence, from soil next to a_ well-rotted stump. Previously known range: Michigan to New Brunswick and Connecticut. 7. Nephrotoma ferruginea (Fabricius) Br. NH. 1974-1975; April 21-June 29 and August 12-September 26; bivoltine (second generation records based on col- lections by G. W. Byers in 1969). The first species of Nephrotoma to emerge in spring, and the only Nephro- toma found in grassland in the study areas. Numerous in the margins of the grassy fields and in adjacent woods. Rare in grassy patches along water courses. This species comes rather freely to light at night. Previously known range: Eastern North America, westward to Colorado, southward to Texas. 8. Nephrotoma macrocera (Say) Br. NH. 1974-1975; May 8-September 21. This species occurs throughout the summer with two peaks of emergence, in June and September; bivoltine. This is the most abundant species of Nephrotoma in Br., fairly common in bot- tomland forests and more mesic parts of the oak-hickory; less common in the swampy area. About a dozen, callow adults were taken on September 13, 1975, in grassy patches along Coal Creek, two days after a week-long rainy period. Previously known range: Wisconsin to Maine, southward to Kansas, Tennessee, and Florida. 9. Nephrotoma polymera (Loew) Br. NH. 1974-1975; May 22-June 28; univoltine. Numerous to common in bottomland forests, and in grassy areas around the swamp. Rare in thickets along creeks. Previously known range: Wisconsin to New Hampshire, southward to Kansas, Tennessee, and South Carolina. 10. Nephrotoma virescens (Loew) Br. 1974-1975; June 11-July 23; univol- tine. Rare; only four specimens taken from wooded margins of Coal Creek, two of them, a mating pair, on June 11, 1974. Previously known range: Michigan to New Hampshire, southward to Illinois and Florida. 11. Tipula (Pterelachisus) ignobilis (Loew) NH. 1975; May 30-June 1; univoltine. Rare and local, only seven specimens taken; three males from a small swarm at a small, vernal stream near the pond; two mating pairs from a moss-covered, tree trunk nearby on June 1, 1975. Adults were reared from larvae found in a mat of wet mosses on a cliff at Hole-in-the-Rock, 12 miles south of Lawrence. Previously known range: Illinois to New Brunswick, southward to Tennessee and North Carolina. 12. Tipula (Pterelachisus) trivittata Say Br. NH. 1975; May 3-19; univoltine. Tue Crane Fiies in EasteRN KAnsAs 417 Rare; from the bottomland forests, or fromm moist to mesic hillside woods. The Malaise trap in the Br. caught four indi- viduals of this species. Previously known range: Iowa to New- foundland, southward to Tennessee and South Carolina. 13. Tipula (Lunatipula) australis Doane NH. May 1958 and May 1960; univol- tine. Records based on collections made by G. W. Byers. A spring species, found in 1958 on the shaded bank of the overflow channel around the dam (NH). In 1960, recorded from the brushy hillside just be- low an old limestone quarry in the NH. Previously known range: ‘Texas to Georgia, northward to Maryland. 14. Tipula (Lunatipula) bicornis Forbes Br. NH. 1974-1975; May 7-June 4; uni- voltine. Abundant and local in grassland, rare or absent elsewhere; adults usually on pink clover or other, taller, herbaceous plants in grassy fields. On the morning of May 8, 1975, about 20 pairs were observed in copu- lation, one foot or less above the ground in a grassy field along the highway at the edge of the Br. In every instance the male was fully matured, the female slightly or very teneral. Pupae were taken on May 4, 1975, in grassland, about 6 mm. beneath the ground surface, which was covered by a thin mat of mosses. Adults emerged on 10 May; several of these and some of the pupae were killed by fungus, Massospora tipulae Porter (identified by Dr. R. W. Lichwardt of the University of Kansas Department of Botany). Previously known range: Wisconsin to New Brunswick, southward to Kansas, Tennessee, and Virginia. 15. Tipula (Lunatipula) dietziana Alexander Br. NH. 1975; April 28-May 6; univol- tine. Numerous and general on moist, north- facing hillsides. In morning and late after- noon many males were seen flying about over the damp leaf mold well up the slopes of the hillside woods, probably in search of emerging females. Previously known range: Kansas to New York, and South Carolina. 16. Tipula (Lunatipula) disjuncta Walker NH. 1975; April 29-May 7; univoltine. Numerous and general in oak-hickory forests, rare on lower mesic hillsides. Males are very active during the day, searching for emerging females. All specimens taken were males; females recorded from G. W. Byers’ collection of May 10, 1960. Previously known range: Iowa to Ver- mont, southward to Illinois and Delaware. 17. Tipula (Lunatipula) dorsimacula Walker Br. NH. 1975; April 20-May 4; univol- tine. The earliest of the subgenus Luna- tipula to emerge in the spring; numerous on mesic hillsides and in vernal seepage areas. In the morning, males were usually found flying low over the damp leaf mold, together with males of T. dietziana. Previously known range: British Co- lumbia to Nova Scotia, southward to Cali- fornia and New Jersey. 18. Tipula (Lunatipula) duplex Walker Br. NH. 1974-1975; May 27-July 26; uni- voltine, with a long adult flight period, at its peak in mid-June. Females rare after mid-July. The long season is probably correlated with local differences in habi- tats. Abundant in oak-hickory and hillside 418 Tue Universiry oF KANsAs SCIENCE BULLETIN woods, common in bottomland forests, and rare in the swamp area, this is the most abundant and conspicuous Tipula of mid-June. Adults range throughout wooded habitats. Previously known range: Kansas to Nova Scotia and Florida. 19. Tipula (Lunatipula) flavibasts Alexander Br. NH. 1974-1975; June 19-July 24; univoltine. The last species of Lunatipula to emerge in summer; numerous in margins of rather dry woodlands, spreading into adjacent, grassy fields. Previously known range: Kansas. 20. Tipula (Lunatipula) flavoumbrosa Alexander Br. NH. 1974-1975; May 8-June 10; uni- voltine. Abundant in the low, damp parts of the oak-hickory and bottomland forests, rare from drier and open parts of hillside woods. Most often found on upper leaves of low shrubs. Previously known range: Kansas to Michigan, South Carolina, and Florida. 21. Tipula (Lunatipula) fuliginosa (Say) Br. NH. 1974-1975; May 17-June 10; univoltine. Numerous in drier parts of hillside woods, a few in the oak-hickory woods. Females usually in damp thickets along the creek and moist, grassy patches in bottom- land woods. Previously known range: Kansas to Ontario and New Hampshire, southward to North Carolina. 22. Tipula (Lunatipula) mallochi Alexander Br. NH. 1974-1975; May 12-June 17; univoltine. Abundant in moist, low oak-hickory, in bottomland forests and damp thickets along the creeks, common in wooded mar- gins of the swamp, widespread in woods in late May. Larvae were taken at Hole- in-the-Rock beneath and in leaf mold of the forest floor. These began to pupate April 26, 1974, and adults started emerg- ing on May 3. Previously known range: Missouri to Maryland and Florida. 23. Tipula (Lunatipula) morrisoni Alexander Br. NH. 1974-1975; May 23-June 17, univoltine. Numerous in open woods and drier hill- side woods; adults usually among lower leaves of trees. Previously known range: Kansas to Rhode Island, southward to Mississippi, and South Carolina. 24. Tipula (Lunatipula) perlongipes Johnson Br. NH. 1974-1975; May 30-June 20; univoltine. Rare, on drier, sparsely wooded, oak- hickory slopes; distinguished from other members of the triplex subgroup by the relatively narrow, yellow wings, and by male genitalial structures. Previously known range: North Carolina to Indiana. Florida, 25. Tipula (Lunatipula) triplex group, species near T. perlongipes Br. 1975; May 31-June 17; univoltine. Rare in bottomland woods and more mesic parts of the low oak-hickory. It dif- fers from perlongipes in the structure of the male hypopygium, which lacks the median depressed lobe on the 8th sternum (Fig. 29). The habitat distribution of this species is more like that of T. flavoum- brosa. Rogers (1942) called this species Tipula species, triplex group near flavoum- brosa. THe Crane Fiies in Eastern Kansas 419 Previously known range: Michigan, westward to Iowa, and Missouri (Rogers, 1942, p. 68). 26. Tipula (Lunatipula) translucida Doane NH. 1974-1975; June 10-June 17; uni- voltine. Rare and local in a narrow thicket along the overflow channel at the pond in the NH. All specimens caught were males, usually among leaves of taller buckbrush and lower branches of trees. Previously known range: Illinois to Pennsylvania, southward to Oklahoma and South Carolina. 27. Tipula (Lunatipula) triplex Walker Br. NH. 1974-1975; May 7-May 31; uni- voltine. Common in the ecotone between grassy fields and edges of woods. Several teneral adults were taken from the grassy field along with T. dicornis on May 8, 1975. Rare or absent in other habitats. Previously known range: Alberta to Newfoundland, southward to Wisconsin and Virginia. 28. Tipula (Lunatipula) integra Alexander Br. NH. 1974-1975; May 8-June 6; uni- voltine. Common in the wetter, low, oak-hickory and hillside woods, absent in grassland, rare in bottomland forests. Resembles T. triplex, {rom which it differs in details of male genitalia. The submedian teeth of the 8th sternum are broad at base with round apex in this species (Fig. 26). It also ap- pears to have a somewhat different habitat. Rogers (1942) believed this form is dis- tinct from either triplex or umbrosa and used the term Tzpula triplex group form C for it. Alexander (1962) described it as Tipula triplex integra, a race of triplex. However, the conspicuous differences in male genitalial structure, habitat correla- tion, and occurrence together with the typi- cal form suggest it is a distinct species. Previously known range: Michigan, Indiana. 29. Tipula (Lunatipula) tuscarora Alexander Br. 1974-1975; May 17-June 2; univol- tine. Rare in open woodland; easily flushed and usually alights on lower leaves of trees. All specimens taken were males. Previously known range: Missouri to Maryland, southward to South Carolina. 30. Tipula (Platytipula) paterifera Alexander Br. NH. 1974-1975; October 10-21; uni- voltine. One female was taken in a grassy patch along Coal Creek in 1974. In 1975 this species was found abundant in the patches of Polygonum which conceal the pond shoreline at NH. It was also common in the grassland where T. bicornis was found in the spring. Adults are active in late afternoon. This species comes frequently to light at the home of the resident nat- uralist at NH. Previously known range: Tennessee, Missouri. 31. Tipula (Platytipula) ultima Alexander Br. NH. 1974-1975; August 29-October 10; univoltine. Abundant in the transition zones be- tween the swamp and mesic hillside woods, this species also spreads into adjacent grass- lands. About 60 larvae were taken on April 3, 1974, in saturated soil on the bank of the overflow channel around the dam at NH. These were reared and observed in the laboratory. They stopped feeding and moved from the saturated soil to drier soil in early July, although pupation did 420 Tue Universiry oF Kansas ScrENCE BULLETIN not occur until about a week before emer- gence, which started on September 3, 1974. Previously known range: Saskatche- wan to Nova Scotia, southward to Wyo- ming, Mississippi, and Florida. 32. Tipula (Trichotipula) unimaculata (Loew) Br. NH. 1974-1975; June 24-July 16; univoltine. Rare in wet areas in hillside woods. Often found resting on algae-covered tree trunks near hillside seepage areas. Previously known range: Michigan to Maine, southward to Illinois, and North Carolina. 33. Tipula (Trichotipula) stonet Alexander Br. 1974-1975; September 5-21; univol- tine. Occasionally numerous on open _hill- sides; adults found resting on leaves of lower branches of trees. Previously known range: Michigan to Ontario, southward to Kansas and Florida. 34. Tipula (Yamatotipula) furca Walker Bro NE M974-1975+ April 21-June:29: and August 31-September 8; bivoltine. Common in the swamp area at NH. and in grassy patches along Coal Creek; rare in damp thickets around the swamp. Females were observed ovipositing into saturated soil at the margin of the pond and along the bank of a brook which feeds into the pond at NH. Previously known range: Kansas to Quebec and Maine, southward to Texas and Florida. 35. Tipula (Yamatotipula) sayi Alexander Br. 1974-1975; September 13-14; univol- tine. Only two females taken, both from a grassy clearing in the flood-plain woods along Coal Creek. Previously known range: Iowa to New- foundland, southward to Louisiana and Florida, Bermuda. 36. Tipula (Yamatotipula) strepens Loew Br. NH. 1974-1975; May 1-June 26, and July 24-August 17; bivoltine. Common in grassy areas around the swamp and in grassy patches along Coal Creek. Numerous in the damp thicket around the swamp, usually found hanging in buck brush. T. strepens and T. furca overlap to some extent in their habitat; strepens ap- parently is able to tolerate lower humidity than furca. Previously known range: Kansas to Newfoundland, southward to New Jersey. 37. Tipula (Yamatotipula) tricolor Fabricius NH. 1975; May 23-June 1; apparently univoltine here, though bivoltine elsewhere in eastern USS. Rare and local in the swamp area; all three specimens were taken from patches of Polygonum hydropiper growing at the edge of the pond. Previously known range: Wisconsin to Quebec and Maine, southward to Arkan- sas and Florida. 38. Limonia (Limonia) globithorax (Osten Sacken) Br. 1974; August 17-September 18; uni- voltine. Numerous and _ local; six specimens were taken from a small swarm above a partially submerged, fallen, tree trunk in Coal Creek. Three records were from damp, decayed wood in a nettle patch along the creek. Previously known range: Wisconsin to Newfoundland, southward to Tennessee and Florida. Tue Crane Fiies in Eastern Kansas 421 39. Limontia (Limonia) rara (Osten Sacken) Br. NH. 1974-1975; June 14-July 23, and August 17-September 18; bivoltine. Numerous and general in its restricted habitats along water courses. Most speci- mens were resting on moist, moss-covered tree trunks about 18 inches above the ground, in mesic flood plain forest. Previously known range: Iowa and Wisconsin to New York, southward to Florida. 40. Limonia (Limonia) tristigma (Osten Sacken) Br. NH. 1974-1975; June 10-July 3; uni- voltine. The most abundant species of subgenus Limonia in two study areas. Abundant in the flood plain forests and swamp, espe- cially in places where nettle and jewelweed are luxuriant, occasionally hanging from stems and leaf margins of tall herbs and shrubs. Previously known range: Alberta to New Brunswick, southward to Tennessee and North Carolina. 41. Limonia (Metalimnobia) cinctipes (Say) Br. NH. 1974; June 11-25; univoltine. Rare, in mesic, low oak-hickory and the thickets between hillside woods and swamp. A female was found resting on the shady side of a tree trunk with wings folded over the back and all legs outspread. Previously known range: Alberta to Newfoundland, southward to Mississippi and Florida. 42. Limontia (Metalimnobia) fallax (Johnson) Bre INE. 197421975; May. 3-June* 25, and September 5-14; bivoltine. Numerous in the low, oak-hickory for- ests, adults usually standing on leaf litter around seepage or wet spots, or on moist, decayed tree stumps. Previously known range: Michigan to New Jersey, southward to Oklahoma and North Carolina. 43. Limontia (Metalimnobia) immatura (Osten Sacken) NH. 1974-1975; June 19-July 2; univol- tine. Rare, in mesic hillside thickets. Two of the three specimens were from the same habitat as L. cinctipes, the other one from jewelweeds in bottomland woods. Previously known range: Maine to British Columbia, southward to Florida. 44. Limonia (Metalimnobia) triocellata (Osten Sacken) Br. NH. 1974-1975; May 24-June 26, and September 5-14; bivoltine. Numerous in upland woods, where the scattered undergrowth is about two feet high. Adults often rest on shrubs. Ap- parently this species can tolerate low hu- midity better than other members of its genus. Previously known range: Alberta to Nova Scotia, southward to Tennessee and Georgia. 45. Limonia (Discobola) annulata (Linnaeus) Br. 1975; July 16. Rare and local, only two males having been taken from dry moss-covered bases of trees in bottomland woods bordering Coal Creek. Previously known range: British Co- lumbia to Newfoundland, southward to Oregon, Tennessee, and Virginia; also found in Eurasia, south to New Guinea. 46. Limonia (Dicranomyia) divisa Alexander Br. NH. 1974-1975; April 21-July 16. Continuous flight period, but a definite 422 Tue Universiry oF Kansas ScrENCE BULLETIN peak of abundance in late May; probably bivoltine. Numerous to common in bottomland woods and mesic parts of oak-hickory woods, especially in shaded ravines having moist to wet banks. Females were observed ovipositing in moist moss on fallen trees. Previously known range: Iowa to Mas- sachusetts, southward to Missouri and Florida and the Greater Antilles. 47. Limonia (Dicranomytia) haeretica (Osten Sacken) Br. 1974-1975; May 3-June 10; univol- tine. Numerous in vernal, seepage areas; most specimens taken in 1975 were under a moss-covered, overhanging rock on a shaded, wet hillside. Rare or absent in all other, drier habitats. Adults were reared from larvae from mosses growing on cliff at Hole-in-the-Rock, 12 miles south of Lawrence. Previously known range: Newfound- land and Rhode Island, westward to Mich- igan. 48. Limonia (Dicranomyia) humidicola (Osten Sacken) Br. NH. 1974; June 11-28; univoltine. A typical stream species, generally dis- tributed in wet, well-shaded spots along streams and in the wet, wooded ravine; usually found in shaded niches where Dolichopeza was common. Previously known range: British Co- lumbia to Nova Scotia, southward to Cali- fornia, Central America and northern Georgia. 49. Limonta (Dicranomyia) immodestoides Alexander Br. 1974-1975; May 6-June 25, and Au- gust 31-September 12; bivoltine. Rare and local; the few records were from hillside woods with luxuriant under- growth. Absent from better drained wood- lands. Previously known range: Oregon to Newfoundland, southward to Indiana. 50. Limonia (Dicranomyia) liberta (Osten Sacken) Br. NH. 1974-1975; May 1-June 20; uni- voltine. Common in vernal seepage in bottom- land woods and in the mesic thicket around the margin of the pond in the NH. Rare to absent on the slopes. Previously known range: Manitoba to Newfoundland, southward to Oklahoma and Florida, and in Bermuda. 51. Limonia (Dicranomyia) pudica (Osten Sacken) Br. NH. 1974-1975; April 29-May 31; univoltine. Common in spring, generally distrib- uted in bottomland woods where the un- derstory is luxuriant, less common in poor- ly drained woodland. Previously known range: Michigan to Maine, southward to Illinois and North Carolina; 52. Limonia (Rhipidia) bryanti (Johnson) NE. 19742 June 17. Rare, only one male taken beneath a protruding rock by one of the brooks which feed into the pond. Previously known range: Colorado to Maine, southward to Arizona and Florida. 53. Limonia (Rhipidia) domestica (Osten Sacken) Br. NH. 1974-1975; May 8-July 16 and September 5-October 10; bivoltine. Abundant and generally distributed in moist creek-margin thickets, beneath shaded banks of ravines, in flood plain woods and near seepages and the swamp; a few spread into drier, upland woods. Previously known range: Kansas and THe Crane Fiiés IN Iowa to New Jersey, southward to Texas and Florida; also in the neotropical region. 54. Limonta (Rhipidia) lecontei Alexander NH. 1974; May 19. One male taken from a decayed, par- tially submerged log in the brook just be- low the dam in the NH. Previously known range: Alaska to Newfoundland, southward to California, and Virginia; also recorded from Eurasia. 55. Limonia (Geranomyia) communis (Osten Sacken) Br. NH. 1974-1975; April 16-June 26 and September 20; bivoltine. Common and local in the continuously- wet to slightly-submerged layer of algae in the stream bed and the submerged film of algae on the margin of the stream where the water is quiet. Adults of both sexes occurred along the brook, and females were observed ovipositing on algae-covered rocks in the brook at NH. Previously known range: Ontario west- ward to Washington and Califonia, south- ward to Florida. This species has fre- quently been confused with L. (G.) canadensis in the literature, as noted by Alexander (1965:49). 56. Helius (Helius) flavipes (Macquart) Br. NH. 1974-1975; May 8-July 16 with a single record for September 14; bivoltine. Rare, only six specimens, from near the swamp and from bottomland woods. Adults usually rest on tall herbs. Previously known range: Alberta to Nova Scotia, southward to Texas and Florida. 57. Dicranoptycha elsa Alexander Br. NH. 1974-1975; May 22-June 3; uni- voltine. Numerous in the same situations where D. megaphallus was found. Adults were mostly from shrubs such as buckbrush, EasTERN Kansas 423 less commonly from poison ivy and other herbs. Previously known range: New York to North Carolina. 58. Dicranoptycha megaphallus Alexander Br. NH. 1974-1975; May 23-June 17; univoltine. This is the most common species of its genus in the two study areas; characteristic of oak-hickory and spreading into nearby habitats. As in all species of the genus, the adults at rest stand upright, high on the tarsi, with body elevated above the upper surfaces of leaves, with the wings folded over the back. Previously known range: North Caro- lina, South Carolina, and Florida. 59. Dicranoptycha pallida Alexander NH. 1974-1975; June 24-July 16; uni- voltine. Rare and apparently very local; all rec- ords are from swamp and low oak-hickory ecotones, where dense thickets are formed, with rich undergrowth. Previously known range: Kansas, In- diana. 60. Dicranoptycha septemtrionis Alexander Br. 1974-1975; July 16-September 7; uni- voltine. Numerous in flood plain woods, from patches of nettles and jewelweed, less com- mon in open woods on hillside slopes. Previously known range: Michigan to Massachusetts, southward to Indiana and North Carolina. 61. Dicranoptycha sobrina Osten Sacken Br. NH. 1974-1975; June 10-17; univol- tine. Rare; three specimens were swept from a rather dry, open, hillside woods, where the undergrowth was largely tick trefoil (Desmodium). Previously known range: Indiana to 424 Tue University oF Kansas ScrENCE BULLETIN New Jersey, southward to Tennessee and Florida. 62. Dicranoptycha tigrina Alexander Br. 1974; August 31-September 21; uni- voltine. Rare; and rather local, all records being from grassy patches along Coal Creek, the flies usually found in the places where D. septemtrionis is common. Previously known range: Indiana, IIli- nois, Tennessee, and North Carolina. 63. Epiphragma fasciapennis (Say) Br. NH. 1974-1975; May 12-June 28; univoltine. Common in the latter half of May and early June, and rather generally distributed in bottomland woods, in the swamp, and in dense thickets between the swamp and low oak-hickory; adults usually found on forest litter or low shrubs, less than two feet above ground. Previously known range: Alberta to Newfoundland, southward to Louisiana and Florida. 64. Epiphragma solatrix (Osten Sacken) Br. NH. 1974-1975; May 14-June 10 and July 15-September 14; bivoltine. Common in the margins of the swamp and in bottomland forests, spreading into moist thickets nearby. Larvae were taken in April, 1974, from both wet, submerged wood in a brook and a decayed log on the forest floor. Adults emerged on May 2, 174. Flies of the summer generation have a smaller body size than those of the spring generation. Previously known range: Missouri to New York, southward to Louisiana and Florida. 65. Pseudolimnophila contempta (Osten Sacken) Br. 1974-1975; May 24-June 3, and Au- gust 3-September 21; bivoltine. Rare, all specimens from the vicinity of shaded, hillside seepage areas. Adults were found resting on the ground rather than on plants. Previously known range: Michigan to Newfoundland, southward to Missouri and Florida. 66. Pseudolimnophila luterpennis (Osten Sacken) Br. NH. 1971-1975; June 1-25, and Au- gust 17-September 21; bivoltine. Numerous in the grassy edges of the swamp at NH. and in grassy patches along Coal Creek, spreading onto lower, more densely shaded hillsides. Adults were found on the wet ground rather than on plants. Previously known range: Quebec west- ward to California and southward to Louisiana and Florida. 67. Pilaria imbecilla (Osten Sacken) NH. 1974-1975; May 15-June 19; uni- voltine. Locally common at margins of the pond, rarely spreading into nearby thickets. Adults usually rest on leaves of shrubs and taller herbage. Previously known range: Illinois to Quebec and Massachusetts, southward to Tennessee and Georgia. 68. Pilaria quadrata (Osten Sacken) NH. 1974-1975; April 29-June 19; uni- voltine. Fairly common, but local, in the vicinity of the pond; adults found in great num- bers on May 3, 1975, on the muddy edges of the pond, on Polygonum hydropiper that grows around the pond, less common in the wet thickets nearby. Previously known range: Iowa to Nova Scotia, southward to Florida. THe Crane FiiEs 1n Eastern Kansas 425 69. Pilaria tenuipes (Say) Br. NH. 1974-1975; May 3-July 16; uni- voltine. The most common species of its genus; common and generally distributed in wet thickets around the swamp and along creeks, also common in the flood plain forests, rare in mesic hillside woods. Adults were found usually standing on leaves of taller shrubs, with wings folded over the back and an elevated stance resembling that of Dicranoptycha. In localities where all three species of Pilaria were found, quadrata occurred in the wettest microhabitats, tenuipes in the less mesic habitat, while zmbecilla was found in between. Previously known range: Wisconsin to New Brunswick, southward to Kansas, Texas, and Florida. 70. Atarba (Atarba) picticornis Osten Sacken Br. NH. 1974-1975; May 22-July 2, and a single record on September 21; bivoltine. Numerous in grassy areas along the overflow channel around the dam in NH., in nettle-jewelweed patches along Coal Creek, and rare in mesic thickets and hill- side habitats. Previously known range: Michigan to New Hampshire, southward to Missouri and Florida. 71. Elephantomyia (Elephantomyia) westwoodi Osten Sacken Br. NH. 1974-1975; June 2-July 23; uni- voltine. Rare, only five specimens taken, all from low, oak-hickory slopes. Adults were found hanging from leaves on low branches of trees. Previously known range: Wisconsin to Newfoundland, southward to Illinois and Florida. 72. Cladura (Cladura) flavoferruginea Osten Sacken Br. NH. 1974-1975; October 2-21; uni- voltine. This species emerges in the dry autumn and is the last tipulid species that occurs in these areas, except for the winter species Chionea stoneana. Adults common and generally distributed in most of the woods. Previously known range: Wisconsin to Quebec and Maine, southward to Missouri and Georgia. 73. Chionea stoneana Alexander Br. 1973; January 20. The only nearly apterous crane fly in the local fauna; uni- voltine. G. W. Byers collected one live male in a pitfall trap, set by an old rubbish dump that had numerous mouse burrows beneath slabs of wood, etc. At other places, he has found C. stoneana in burrows and _ nests of mice. Previously known range: Illinois. 74. Teucholabis (Teucholabis) complexa Osten Sacken INI 1974 June: Two specimens taken from near a ver- nal, seepage area, both resting on leaf litter. Previously known range: Michigan to Connecticut, southward to Oklahoma and Florida. 75. Teucholabis (Teucholabis) luctda Alexander Br. NH. 1974-1975; May 31-June 17 and August 17-October 2; bivoltine. Common and local in swampwoods in early summer when the water level of the pond began to drop and small bodies of water were ponded by tree roots and forest litter. Adults were resting on saturated soil around this water. Also recorded from the shaded bank of the ravine at Br. Previously known range: Missouri to 426 Tue Universiry oF Kansas SCIENCE BULLETIN District of Columbia, southward to Flor- ida. 76. Gnophomyia tristissima Osten Sacken Br. NH. 1974-1975; May 7-July 16; uni- voltine. Common in bottomland woods and wet thickets on herbage among fallen, decaying trees. Previously known range: Northwest Territories to Quebec and Maine, south- ward to Texas and Florida. 77. Gonomyia (Gonomyia) florens Alexander NH. 1974; June 24. One male only, from the grassy margin of a brook at NH. Previously known range: Michigan to Quebec and Maine, southward to Illinois, Tennessee, and North Carolina. 78. Gonomyia (Gonomyia) kansensis Alexander NH. 1956; May 28. This record based on a light trap col- lection made by G. W. Byers in 1956. Previously known range: Oklahoma, Missouri, Illinois, Indiana, and Michigan. 79. Gonomyia (Gonomyia) subcinerea Osten Sacken Br. NH. 1974-1975; May 8-June 16 and August 3-October 10; bivoltine. This is the most abundant species of its genus; common and generally distrib- uted in oak-hickory and mesic bottomland woods. Previously known range: British Co- lumbia to Newfoundland, southward to Utah, Kansas, and Florida. 80. Gonomyia (Lipophleps) manca (Osten Sacken) Br. 1974-1975; June 2-August 17; uni- voltine. Rare, a total of twelve specimens taken in bottomland woods, mainly from grassy margins of Coal Creek. Previously known range: Indiana to Massachusetts, southward to Tennessee, and Florida. 81. Gonomyia (Lipohphleps) sulphurella Osten Sacken Br. 1974-1975; September 5-13; univol- tine. Rare, only two males taken from the same wooded ravine at Br. Previously known range: Kansas to Newfoundland, southward to Texas and Florida. 82. Erioptera (Symplecta) cana (Walker) Br. NH. 1974-1975; March 21-May 30 and September 21-29; multivoltine. The first species to appear in the spring, when most vegetation has not yet started growing; abundant in the spring in nearly all wooded habitats and open grasslands. The fall generation was inconspicuous. Previously known range: Alaska, throughout southern Canada and _ the United States. 83. Erioptera (Erioptera) septemtrionts Osten Sacken Belo; April 7, One female from a juniper tree grow- ing at the ecotone between woods and grassy field. Previously known range: Washington to Newfoundland, southward to Califor- nia, Kansas, and Florida. 84. Erioptera (Erioptera) vespertina Osten Sacken 3r. NH. 1975; May 7-22; univoltine. Numerous; all but two were from the Malaise traps, those two swept from grassy patches near the traps. Previously known range: Iowa to Nova Scotia, southward to Alabama and Florida. THe Crane FLies IN 85. Erioptera (Mesocyphona) caliptera (Say) Br. NH. 1974-1975; May 7-June 24 and July 16-August 21; bivoltine. Abundant where soil is wet or muddy during most of the year, in bottomland forests, swampwoods, and grassy margins of streams; not spreading into the drier oak-hickory woods, but may spread to mesic hillside woods. Previously known range: California to Newfoundland, southward to Florida; also Neotropical. 86. Erioptera (Mesocyphona) needhamt Alexander Br. 1974-1975; May 31-June 11 and Au- gust 3-17; bivoltine. Rare, only four specimens swept from grassy areas along the sandy creek shore, and two from a shaded bank of Coal Creek. Previously known range: Missouri to Nova Scotia, southward to Florida. 87. Erioptera (Mesocyphona) parva Osten Sacken INET. 1975: July 24. One male specimen taken from a grassy area in the swamp woods. Previously known range: Kansas to Michigan and Connecticut, southward to Florida. 88. Erioptera (Hoplolabis) armata Osten Sacken Br. NH. 1974-1975; April 19-June 14 and July 25-August 17; bivoltine. Abundant; adults taken along all water courses, either from shaded banks or from the base and roots of large trees growing by creeks. Large swarms seen on April 28, 1975, along Coal Creek, about five feet above ground. Conspicuous, bimodal variation was found in the dististyles of local males, sug- EasTERN Kansas 427 gesting that two species might have been represented. Previously known range: Colorado to Newfoundland, southward to Oklahoma and Georgia. 89. Erioptera (Psiloconopa) graphica Osten Sacken NH. 1974-1975; April 22-June 26 and September 29; bivoltine. Numerous and local about the margins of the pond, less numerous in the swamp woods and rare elsewhere, all individuals on herbaceous plants. Previously known range: Nebraska to Ontario and Massachusetts, southward to Louisiana and Florida. 90. Ormosia arcuata (Doane) Br. 1975; April 28. One female recorded from the grassy margin of Coal Creek. Previously known range: Alberta to New Brunswick, southward to Tennessee. 91. Ormosia ingloria Alexander Br. 1975; September 7-13; univoltine. Rare; all specimens collected were males, most found resting on tree trunks growing in bottomland and on hillsides. One was taken from a tall shrub. Previously known range: Indiana and Ontario. 92. Ormosia romanovichiana (Osten Sacken) Br. 1975; April 19-28; univoltine, in early spring. Common in spring along the creek and around most vernal seepages, usually rest- ing on wet, moss-covered rocks or on tree bark, where Erioptera (H.) armata was also common. These two species can be easily differentiated in the field by the way they rest on the substrate. This spe- cies holds its body parallel to the substrate, while E. armata always tilts its body, head 428 Tue University oF Kansas ScrENCE BULLETIN downward, to form an angle with the sub- strate. Previously known range: Illinois to Maine, southward to Tennessee and South Carolina. 93. Tastocera (Dasymolophilus) ursina (Osten Sacken) Br. NH. 1975; May 6-14; univoltine. Locally abundant near seepage areas and shaded brooks. On May 7 at 10 a.m. I took about forty males from swarms about two feet above ground along brooks leading to the pond at the NH. One mat- ing pair was found at that time. All other females were from the Malaise trap at the NH. Previously known range: Newfound- land, southward to Tennessee and North Carolina. 94. Molophilus hirtipennis (Osten Sacken) NH. 1974-1975; May 15-June 13; uni- voltine. Rare; only three females were taken from grassy patches in the swamp woods. Previously known range: Ontario to Newfoundland, southward to Illinois, Ten- nessee, and North Carolina. 9. Molophilus pubtpennts (Osten Sacken) Br. NH. 1974-1975; May 6-June 11; uni- voltine. Common to abundant in damp to wet, flood plain forests, and on damp, coarse, sand bars in the bed of Coal Creek. All specimens were females. Rogers (1942) believed that this species as well as M. hirtipennis is chiefly or entirely partheno- genetic. Previously known range: Michigan to Quebec and Newfoundland, southward to Florida. DISCUSSION Crane flies occupy many types of habi- tats in their immature and adult stages. The larval habitat of most species is local and restricted. Certain species live in mud, some in decaying, vegetable detritus, while others occur in rotting wood. Many larvae are scavengers, others are herbivores and still others are predaceous. In most cases, the presence or absence of a suitable, larval habitat determines whether a species can maintain itself in a given area. During this two-year investigation, 95 species of crane flies were recorded from the two study areas, of which 56 species were common to both areas. Another 21 species were found only at the Breidenthal Reserve, the remaining 18 species only at the Natural History Reservation. Other species and subspecies recorded from northeastern Kansas, but not taken in either of the study areas are: Dolichopeza (Oropeza) polita pratti, Tipula (Schum- melia) hermannia, T. (Nippotipula) ab- dominalis, T. (Beringotipula) borealis, T. (Lunatipula) incisa kansensts, Limonia (Geranomyia) rostrata, Dicranoptycha minima, Erioptera (Psiloconopa) armil- laris, and E. (P.) venusta. The crane-fly fauna of eastern Kansas is relatively poor when contrasted with the faunas of the eastern states: New Eng- land (Johnson, 1925, and Alexander, 1925, 1927, 1930, 1936) has 337 species; New York (Alexander, 1919, 1922, 1924;"1929), 318 species; southeastern Michigan (Rog- ers, 1942),.201 species. his can bevex- plained largely by the geographical loca- tion of Kansas. Eastern Kansas is located in a transition zone between the eastern forests and the central plains. The prairie forms a natural barrier, preventing the eastward spread of western species. Conse- quently, there is an abrupt difference in crane fly fauna between eastern Kansas and mountainous central Colorado. Most Kansas crane flies are eastern species, oc- curring also in forests and other appro- priate habitats eastward to the Atlantic Ocean. Since crane flies are more abun- Tue Crane Fires in Eastern Kansas 429 dant in moist woodlands, the relatively small number of tipulid species in eastern Kansas is probably due to the decline of forests from east to west. Climate also affects the distribution of crane flies. The subfamily Limoniinae is more abundantly represented in the humid eastern forests and decreases westward, as average humidity gradually drops. In terms of percentage of the total crane fly fauna, the situation is reversed for the sub- family Tipulinae. This can be seen by comparing the percentage of Tipulinae in New England—34°%, of 337 species (John- son, 1925, Alexander, 1925, 1927, 1930, 1936), in New York—29°% of 318 (Alex- ander, 1919, 1922, 1924, 1929), in Tennessee —30°% of 151 (Rogers, 1930), in Ohio— 34°% of 146 (Foote, 1956), in Michigan— 36% of 201 (Rogers, 1942), and in eastern Kansas—39Y, of %. The significance of the increasing percentage of Tipulinae can be shown by use of the chi-square test (Snedecor, 1956). It is noteworthy that no species of the large genus Limnophila have been found in eastern Kansas. Limnophila comprises 14°% of the crane fly fauna in New England, 17°% in New York, 9% in Tennessee, 8°% in Ohio, 99% in Michigan. Most other aquatic, predaceous genera such as Hexatoma and Pedicia are also absent from Kansas. The absence in Kansas of marshes, bogs, and other permanently wet habitats, other than artificial ponds (all of relatively recent origin) and a few ma- jor streams, usually very muddy, probably accounts for the absence of these genera. The differences in distribution of spe- cies in the two study areas are due more to topographic than to climatic conditions, since there is no significant difference in climatic conditions between the two areas. The general floras of the two areas are obviously quite different, although there are many species common to both. Since the Natural History Reservation was not established until thirty years ago, its vege- tation is still in a successional state, in which climax oaks and hickories do not yet dominate much of the area. In the Breidenthal Reserve, the vegetation has reached a stable, climax state. The two areas differ in humidity and water drainage. Within the Breidenthal Reserve, Coal Creek is a well-developed stream, supplied by numerous, short tribu- taries, so that it flows during most of the year. The slopes around the creek are relatively steep. The hillsides immediately bordering the creek rise 200 feet in less than a quarter of a mile. Apparently much of the rain that falls on the high land sur- rounding the Coal Creek ravine at first becomes groundwater, but later seeps through the steep hillsides to feed moisture into the ravine. Three additional factors contribute to retention of moisture in the Coal Creek ravine. A large proportion of the vegetation at the Breidenthal Reserve has reached the climax stage and retains moisture well. The ravine opens toward the northeast and its outlet is greatly re- stricted by ridges which make it an en- closed basin, as a result of which it is protected from the southwestern summer winds of eastern Kansas. Finally, there are many relatively steep, north-facing slopes at the Breidenthal Reserve, resulting in shade and, consequently, less evapora- tion. The total result is higher humidity at the Breidenthal Reserve than at the Nat- ural History Reservation. On the Natural History Reservation the streams are small and intermittent, being near their headwaters. They dry out shortly after the beginning of summer. While the total relief is comparable to that of the Breidenthal Reserve, the slopes are generally not as steep, and many of the streams follow ravines which open directly to the southwest and so are exposed to the drying effects of summer winds. Further- more, much of the vegetation at the Nat- 430 Tue University oF KANnsAs Sc1ENCE BULLETIN ural History Reservation is still succes- sional and holds less moisture. The investigation made to determine ecological differences between the two res- ervations that would result in the mutually exclusive distribution of the 39 species was narrowed to 7 species, due to the fact that Text Ficure 1. Contour map of the Natural History Reservation. j mile Text Ficure 2. Contour map of the Breidenthal Reserve. the other 32 were collected only infre- quently. The presence at the Natural His- tory Reservation of the artificial pond with its muddy-shore habitat probably accounts for the presence there of Tipula (Y.) tri- color, Pilaria imbecilla, P. quadrata, and Teucholabis complex, which were found only there. No such habitat exists at the Breidenthal Reserve. The small, rapid streams with algae- or moss-covered rocks in and along them may explain why Limonia (R.) bryanti and L. (R.) lecontet were found only at the Natural History Reservation. This type of habitat probably also accounts for the fact that L. (G.) com- munis is abundant in the Natural History Reservation but rare in the Breidenthal Reserve. The emergence of adult flies is influ- enced mostly by local climate. Warm, humid weather in early spring brings ver- nal species out earlier, while a hot, dry summer delays appearance of autumnal species. As compared to the seasonal dis- tribution of crane flies in southern Michi- gan, the eastern Kansas population comes out earlier in the spring and disappears later in autumn, due to the longer duration from the last, spring, killing frost to the first, autumnal, killing frost. Several spe- cies that have but a single adult season in southern Michigan have two distinct ones in eastern Kansas. From the preceding study it seems rea- sonable to draw three conclusions. First, crane flies of eastern Kansas had _ their origins in the eastern part of the continent. Second, geographical and climatic factors have differentiated the crane fly faunas of eastern Kansas and the more eastern states. Third, mainly topographic and historical factors have influenced the differences in crane fly faunas between the Breidenthal Reserve and the Natural History Reserva- tion. THe Crane Fulks LITERATURE CITED ALEXANDER, CHARLES P. 1919. The crane-flies of New York. Part I. Dis- tribution and taxonomy of the adult flies. Cornell Univ. Agr. Exp. Sta., Memoir 25: 766-993, plates XXX-LV. 1920. The crane-flies of New York. Part II. Biol- ogy and phylogeny. Cornell Univ. Agr. Exp. Sta., Memoir 38:691-1133, plates XII-XCVII. 1922. The crane-flies of New York: first supple- mentary list. Bull. Brooklyn Ent. Soc., 17(2) :58-62. 1924. The crane-flies of New York: second sup- plementary list. Bull. Brooklyn Ent. Soc., 19(3) :57-64. 1925. The crane-flies (Tipulidae) of New Eng- land: first supplementary list. Occ. Papers Boston Soc. Nat. Hist., 5:169-174. 1927. The crane-flies (Tipulidae) of New Eng- land: second supplementary list. Occ. Pa- pers Boston Soc. Nat. Hist., 5:223-231. 1929a. The crane-flies of New York: third supple- mentary list. Bull. Brooklyn Ent. Soc., 24 (1)222-29" 1929b. The crane-flies of New York: fourth sup- plementary list. Bull. Brooklyn Ent. Soc., 24(5) :295-302. 1930. The crane-flies (Tipulidae) of New Eng- land: third supplementary list. Occ. Papers Boston Soc. Nat. Hist., 5:267-278. 1936. The crane-flies (Tipulidae) of New Eng- land: fourth supplementary list. Occ. Pa- pers Boston Soc. Nat. Hist., 8:273-292. 1942. Family Tipulidae. Jn The Diptera or true flies of Connecticut. Conn. State Geol. and Nat. Hist. Surv., Bull. 64:196-509. (re- printed 1966) 1962. Undescribed species of nearctic Tipulidae (Diptera). II. Great Basin Nat., 22:1-7. 1965. Family Tipulidae. Iz Stone et al. A Cata- log of the Diptera of America North of Mexico. Agriculture Handbook No. 276. U. S. Dept. Agr. Byers, GEorGE W. 1961. The crane fly genus Dolichopeza in North America. Univ. Kansas Sci. Bull. Vol. XLII, No. 6. Fircu, Henry S. 1965. The University of Kansas Natural History Reservation in 1965. Univ. Kansas Mus. Nat. Hist. Misc. Publ. No. 42:1-60. FootE, BENJAMIN A. 1956. A preliminary survey of the crane-flies of Delaware County, Ohio (Diptera: Tipuli- dae). Ohio Jour. Sci., 56(4) :217-229. JoHNson, CHARLES W. 1925. Fauna of New England: 15. List of the Diptera or two-winged flies. Occ. Papers Boston Soc. Nat. Hist., 7:1-326. O'Connor, Howarp G. 1960. Geology and ground-water resources of 1 EasTERN KaANsAs 43] Douglas County, Kansas. Kansas Geol. Surv. Bull., 148:1-200. Rocers, J. SPEED 1930. The summer crane-fly fauna of the Cum- berland Plateau in Tennessee. Occ. Papers Mus. Zool. Univ. Mich., No. 215:1-50, plates 1-5. 1942. The crane-flies (Tipulidae) of the George Reserve, Michigan. Univ. Mich. Mus. Zool. Misc. Publ. No. 53:1-128, plates 1-8. We LLs, Puivip V. anp G. E. MorRLEY 1964. Composition of Baldwin Woods: an oak- hickory forest in eastern Kansas. Trans. Kansas Acad. Sci., No. 67(1):65-69. APPENDIX A Key To THE ADULT CRANE FLIES OF EasTERN KANSAS (modified from Alexander, 1942) 1. Terminal segment of maxillary palpus elongate; antennae usually with 13 segments; wings with Sci usually atrophied; body size large (Bigs Dice Fs (Tipulinae) 2 Terminal segment of maxillary palpus short; antennae usually Fic. 1. Wing of Tipula (Lunatipula) flavoumbrosa; A—anal veins, C—costa, Cu—cubitus, M—media, m-cu—median-cubital cross-vein, R—radius, Rs— radial sector, Sc—subcosta. Fic. 2. Wing of Limonta (Limonia) cinctipes. Fic. 3. Wing of Nephrotoma euceroides. 1S) Tue Universiry oF KANSAS with 14 or 16 segments; wings with Sci present; body size small or medium (Figs 2). 22.2.2. i) ee (limoniunae)) 41 Legs long and filiform, tarsi as long as femur and tibia together; (Dolichopeza) 3 Legs of normal stoutness; wings wings with vein Ri+2 atrophied Withe vein kiay presents, era 6 Gonapophyses with tips flattened, the apex irregularly toothed (Fig. G1) | Re ee eee . Dolichopeza walley1 Gonapophyses shaped like small knobs, bearing decurved _ stout black spines and bristles (Fig. 5) 4 Tergal arms (9th abdominal ter- gum) widely flared and emargi- nate at tips; teeth of ninth tergum not set close together (Fig. 6) ........ Dolichopeza obscura Tergal arms not flared or emargi- nate at tips; teeth of ninth tergum set close together on common base 5 Three teeth of ninth tergum of nearly equal length (Fig. 7) ........ eee ee Dolichopeza tridenticulata Three teeth of ninth tergum with middle tooth the longest (Fig. 8) Be ee Dolichopeza polita pratt Wings with Rs shorter than m-cu; thoracic dorsum, less often other body surfaces; highly polished (Biggs) Toe e. (Nephrotoma) 7 Wings with Rs elongate, exceed- ing m-cu, body surfaces pruinose or pollinese:.= 272s (Tipula) 13 ‘Thoracic stripesiblick. eae == aS Thoracic stripes not ‘black =... 9 Occiput dull; wing-tip clear ........ AEE ere ete _ Nephrotoma virescens Occiput with a polished triangular ScreNcE BULLETIN 10. ie Ge 15: IG: brand; wing-tip darkened ............ nee Nephrotoma alterna Mesonotum: dull 22 2.2. eee Nephrotoma macrocera Mesonotum polished ..............--.-.----- 10 Flagellar segments unicolorous .... fBireals bie oe _ Nephrotoma ferruginea Flagellar segments bicolorous ........ 11 Stigma dark brown, wing-tips dis- tinctly*darkened) = ene aN meee ee Nephrotoma polymera Stigma yellowish brown, wing-tips not darkened!" ee ees 12 Antennae (male) 19-segmented .... ee eer Nephrotoma eucera Antennae (male) 17-segmented .... Od a a Nephrotoma euceroides Outer cells of wings with macro- trichia ... (subgenus Trichotipula) 14 Cells of wings without macro- friChiar =. tee eee ee eee 15 Body color bright polished yellow, reddish and black; antennae short; macrotrichia of cells restricted to cells Ro and’ My 22... Tipula stonet Body color dull brown and yel- low; antennae of male elongate; macrotrichia in cells Rs to 2nd Vio ee ie ee Tipula unimaculata Rs long, fully twice m-cu; m-cu uniting with Ms+4 some distance before fork of latter _... (subgenus Nippotipula) ...... Tipula abdominalts Wings with m-cu inserted at the fork of Mg +4 or beyond, on base of IN A aecete e Pdy h Baatet R 16 Wings with m-cu long, so that cell Miz is very deep, much wider at base than at margin ... (subgenus Schummelia) ........ Tipula hermannia Wings with m-cu of moderate length, cell Ms only a little wider alylbase them alee an Cin, 2s. ees 17 Tue Crane Fits 1n Eastern Kansas 433 ff | = if A \ \ \ oA C (/| / \ rN \ ys NL Ler Bs 6 4 a a) a Val SEN = /\ (/| Lf WNW (4. \i ——— Vet "1 as \ } Ms nA / —— f\ |.\ Ne jek et See eS \ . == Fic. 4. Gonapophyses of Dolichopeza (Oropeza) walleyi, dorsal aspect. Fic. 5. Gonapophyses of D. (O.) obscura, dorsal aspect. Fic. 6. Ninth tergum of male D. (O.) obscura. Fic. 7. Ninth tergum of male D. (O.) tridenticulata. Fic. 8. Ninth tergum of male D. (O.) polita pratti. Fic. 9. Squama of Tipula (L.) duplex. 17. Male hypopygium with ninth ter- gum and sternum fused into a continuous ring (Fig. 10) .............. 18 Male hypopygium with the ninth tergum and sternum separated by palemembrane. (Fig.11) 2.2.x 23 18. Wings usually patterned with lon- gitudinal bands, mainly along veins; ninth abdominal tergum of male produced backward as me- dian lobe or pair of lobes bearing small black denticles (Fig. 12) .. , fl (subgenus Yamatotipula) 19 Wings not patterned with longi- tudinal bands; ninth abdominal tergum of male shallowly emargi- mate medially; (Eig. 13) 2... Jae ee (subgenus Platytipula) 20 19. Wings unmarked except for stig- mal darkening and dark costal yA Zz, bo Wy border; no dark seam on vein Cu anid m=eul 273 oe eee Tipula sayt Wings longitudinally striped with brown and white; a dark seam on vein Cu and m-cu Wings with the cells beyond cord, including Rs, darkened .................... Seen en ee er eee Tipula tricolor Wings with outer portion of cell Rs white Wings with vein 2A bordered by Drow Some eee Tipula furca Wings with vein 2A not bordered by brown e..is.50) Tipula strepens Wings patterned with brown areas in cell M and at outer end of vein Nie Bk PPB A 8 Ae Tipula ultima Wings without brown areas; costal border of wings dark brown CEs 1) een Tipula patertfera Squamaamakeds 42 12.0 ee 24 Squama with a small group of setae (Higs9)\ eo eee aD Thane (subgenus Lunatipula) 26 Male hypopygium elongated, cy- lindrical, upturned at an angle to remainder of abdomen ....(subge- nus Beringotipula) .... Tipula borealis Male hypopygium not elongated, not upturned at an angle to re- mainder ot abdomen ../ =a. Pe). ee (subgenus Pterelachisus) 25 Wing pattern pale; basal section of M3+4 shorter than basal section OL NAGS ba aa ae Tipula ignobilts Wing pattern heavy; basal section of Ms+4 subequal to basal section Obs Mino. eee Tipula trivittata Wings with cell Ist Mz very small, second section of M:1+2 shorter than or subequal to petiole of cell IMs Se sea Ae Sees aa eee 27 43 i) Oo Us oS) 4 NO Tue Universiry oF Kansas SctENcE BULLETIN Wings with cell 1st Me normally elongated, second section of Mi +2 exceeding petiole of cell Mt ............ 28 Second section of Mi+2 shorter than: pettole oficell) My 280 t. BIE Rec te eon EN . Tipula bicornis Second section of Mi+2 subequal to petiole of cell Mi ee ee eie a i Sirs Tipula morrisont Male hypopygium asymmetrical, right basistyle produced caudad into a conspicuous 2-branched arm Cai 0) ea Tipula fuliginosa Male hypopygium symmetrical ... 29 Cells beyond cord of wing dark- CTIN(CL a eer ee ke CR ne ee 30 Cells beyond cord of wing uni- formly colored Cells basad of cord strongly in- fulimied a A. ..... Ttpula dorsimacula Cells basad of cord uniformly pale IMRCOlOh wees. Oe ar Tipula mallocht Antennae with bases of flagellar segments light yellow, the remain- dem blacks sie7e12 52 Tipula flavibasis Antennae with flagellum, if bicol- with bases of segments darker than remainder 0.0... 32 orous, Antennae (male) elongate, if bent backward extending about to fourth abdominal segment eae ee Le Tipula disjuncta Antennae (male) shorter, not ex- tending caudad beyond base of 1 oY6 (0) ea Yelm easy etic Rarer nea 33 Male hypopygium with caudal margin of ninth tergum having two rounded emarginations, one on either side of double median spinous point; eighth sternum with sclerotized teeth (Fig. 16) .... 34 Male hypopygium without two RiGee: ultima, lateral Ninth segment of male Tipula (Platytipula) Fic. 11. Ninth segment of male Tipula (Lunatipula) bicornis, lateral aspect. Fic. 12. Ninth tergum of male T. (Yamatotipula) furca, dorsal aspect. Fic. 13. Ninth tergum of male T. (P.) ultima, dorsal aspect. Fic. 14. Ninth tergum of male T. (P.) paterifera, dorsal aspect. Fic. 15, Ninth ter- gum of T. (L.) flavoumbrosa, dorsal aspect.. Fic. 16. Ninth tergum of male T. (L.) triplex, dorsal Fic. 17. Ninth tergum of male T. (L.) dietziana, dorsal aspect. aspect. aspect. rounded emarginations on caudal margin of ninth tergum, eighth sternum without sclerotized teeth 37 34. Male hypopygium with median tergal spines long and _ slender, needle-like (Fis, 16). 2 =a 35 Male hypopygium with median tergal spines short (Fig. 15) ........ 36 Go YW Submedian teeth of eighth ster- num slender, parallel-side (Fig. V5) i Gare Oa ay IE BS cl Tipula triplex Submedian teeth of eighth ster- num broad at base, with narrow, Vp 38. 39. 40. al Tue Crane Fiies 1n Eastern Kansas 435 rounded apex (Fig. 26)) ...2...2..5... Tipula integra Submedian teeth of eighth ster- num triangular in outline, broad basally, narrowed apically (Fig. Di] ee Tipula flavoumbrosa Eighth sternum having a median depressed lobe upside down and slightly cephalad of the submedian teeth (Fig. 28) ............ we che A ae Tipula perlongipes arising Ground color of mesonotum gray or grayish, pleura light gray .......... 38 Ground _ color brownish or yellow, pleura yel- lowish ..... Eee pre a Aw) 39 of mesonotum Male hypopygium with eighth sternum armed with four conspic- MoUs Tobes CHIC, 24)'e 28.2 Ge. iw ae earned ee Tipula australis Male hypopygium with eighth sternum bilobed (Fig. 19) .............. Re Sas ee Tipula dietziana Male hypopygium with lateral lobes of ninth tergum produced into long curved horns (Fig. 18) .. eee es ey elvis Tipula tuscarora Male hypopygium without curved fereal horns 9(Fig. 23)... 00.2... 40 Outer appendage of inner disti- style elongate, terminating in an acute spine (Fig. 21) .. Tipula duplex Outer appendage of inner disti- style elongate but broad at apex (Bigs 22 ees J! Tipula translucida Free tip of Scz often present; veins Ri and Rs fused to margin, only two branches of Rs being present; antennae usually with 14 or 16 seg- IMents(EI1Gs2 ee (Limonuni) 42 Free tip of Sco atrophied; veins Ra and Rs separate, with three branches of Rs present (exceptions 44. 46. 48. in Atarba, Elephantomyia, Teu- cholabis); antennae usually 16 Secimemts) (icy 41) teak eee. 68 2. Wings with vein Re lacking ........ eee (Helius) .... Helius flavipes Wings with vein Re present ........ 43 . Antennae 14-segmented; Re basal in position, opposite or not far be- yond the level of r-m .. (Limonia) 44 Antennae 16-segmented; Re far distadofdevel of tm es ne ous) RS (Dicranoptycha) 62 A supernumerary crossvein in cell Ist A, connecting the two anal veins .... (subgenus Discobola) .... ee ..... Limonia annulata No supernumerary crossvein in Cell} NstoAt we At Se ee a 45 . Mouthparts and especially the la- bial palpi lengthened, the rostrum about as long as the combined head and’ thorax (Fig. 32) ...... ee (subgenus Geranomyia) 46 Mouthparts not conspicuously lengthened, shorter than the re- mainder Or dead le es oes 47 Wings heavily patterned with dark brown, including a series of 4°0r > lanee.costal areas<..... .2=- es et Ao eee Limonia rostrata Wangs, animarked: 220.3 s eee Pesan _.... Limonia communis . Antennae of males more or less branched; of females simply ser- fate (h1G)) Sl)! Se eee ee eh EE ION a5 (subgenus Rhipidia) 48 Antennae of both sexes simple .... 50 Wings with abundant pale brown or gray dots inal cells = st, Prive Rade See Limonia lecontet Wings with the markings larger, confined to vicinity of veins .......... 49 436 49, 50. Dit DZ, 53: D4. 2). Tue University oF Kansas ScreENcE BULLETIN Wings with m-cu far before fork of M; antennae with segment 12 andi ls! white in=colot | sine. a Wings with m-cu at fork of M; antennae dark throughout ............ eae ee ee Limonia bryantt Vein Sc shorter, ending opposite basal one-third of Rs or before .... ee (subgenus Dicranomyia) 51 Vein Sc long, ending opposite midlength of Rs or beyond ............ 56 Wing with cell Ist Me open by SEROP MYO afin ke et eee Se Limonia immodestoides Wing with cell Ist Me closed ........ By. Antennae entirely yellow or with basal two segments yellow; body coloration pale yellow or ochre- pic lloyyeeeere eae 53 Antennae dark, brown or black, throughout; body coloration yel- lowish brown, gray or polished Dic iomne see en oe Sane ae 54 Male hypopygium with the rostral prolongation bifid at apex (Fig. a2, gO Nae NR ace ees Limonia divisa Male hypopygium with the rostral prolongation extended into a long blackened point (Fig. 47) ............ Ft Ae et Limonia pudica Femora brown, the tips broadly yellow Geert Limonia humidicola Femora without paler tips ........... 55 General coloration of thorax clear gray, male hypopygium with ros- tral spines originating from en- larged basal tubercles (Fig. 46) .... SOLE MT DUN sa tes! Limonta liberta General coloration of thorax brown, rostral spines without basal bubercles: (bigs435) 53 eee ee = : vA f\ ise J ) ) / {\ en ee Fic. 18. Ninth tergum of male Tipula (Lunatipula) tuscarora. Fic. 19. Eighth sternum of male T. (L.) dietzsiana. Fic. 20. Ninth tergum of male T. (L.) fuliginosa, dorsal aspect. Fic. 21. Right inner disti- style of male T. (L.) duplex, lateral aspect. Fic. 22. Right inner dististyle of male T. (L.) translucida, lateral aspect. Fic. 23. Ninth tergum of male T. (L.) duplex. Fic. 24. Eighth sternum of male T. (L.) australis. Fic. 25. Eighth sternum of male T. (L.) triplex. Fic. 26. Eighth sternum of male T. (L.) integra. Fic. 27. Eighth sternum of male T. (L.) flavoumbrosa. Fic. 28. Eighth sternum of male T. (L.) perlongipes. Fic. 29. Eighth sternum of male T. (L.) triplex group, species near perlongipes. 56. Vein Sci and Scz ending opposite the foro Rs: . 2 3 a ee See (subgenus Metalimnobia) 57 Vein Sci and Sce ending about opposite midlength of Rs ................ Pe Near ere (subgenus Limonia) 60 57. Wings yellow, with three subcir- cular, eye-like brown markings, placed at origin and fork of Rs and at stigma; femora with only one brown band at the tip ............ poo See rs: ee Limonia triocellata Wings without such an ocelliform 9. 60. 6l. 62. 63. 64. 65. Tue Crane FLIEs pattern; femora with more than One Drown Dain! 222... ete eee 58 Knobs. of halteres uniformly brownish black .......... Limonia fallax Knobs of halteres pale at tips ........ 59 Femora with two brown rings, the outer ring narrow and sub- terminallin position 222... a ree Limonia cinctipes Femora with three brown rings, the outer ring nearly terminal in POSICON) 2 sh... Limonia immatura Wings unmarked; free tip of Sce lying markedly basad of Ro ............ ee eee Limonia globithorax Wings patterned; free tip of Sce and Re in transverse alignment .... 61 Ri+e and Re subequal in length; legs uniformly dark brown ............ Pres Pee. Pe 8 FE ee Ns, Limontia rara Ri+2 two or more times as long as Re; wings with three small brown dots along costal border .... Na ae Limonia tristigma Fore femora extensively black- ened, the bases restrictedly pale; remaining femora more narrowly blackened att tipsi1 2.00 aa 63 Femora at most blackened only at extremie tsps = eee ee sit 64 Costal fringe of wings (male) lonpsandiconspicuous 2452-2 .21 ere AR oN Dicranoptycha sobrina Costal fringe of wings (male) Sort and emse wr eee: ee TS Dicranoptycha megaphallus Tips of femora narrowly black- ened or strongly infumed ................ 65 Femora uniformly pale .................. 66 Tips of femora very narrowly dark brown; abdominal terga transversely banded in tigrine IN EasTERN KANSAS 437 31 35 ee 32 a — as > Se ee => r) ————— = SS ——— ~ @ ee ) ( ) Ae Se a — 33 34 7 73 Fic. 30. Head of Pseudolimnophila luteipennis, dor- sal aspect. Fic. 31. Antenna of Limonia (Rhipidia) domestica, dorsal aspect. Fic. 32. Head of Limonia (Geranomyia) communis, dorsal aspect. Fic. 33. Aedeagus of Dicranoptycha elsa, dorsal aspect. Fic. 34. Aedeagus of D. septemtrionis, dorsal aspect. Fic. 35. Wing of Atarba picticornis. Fic. 36. Wing of Cladura flavoferruginea. Fic. 37. Wing of Gonomyia manca. color; male with long gonapo- physes) bind) Nears pe = BE ecb Ra ra Dicranoptycha tigrina Tips of femora broadly blackened; abdominal terga without subter- minal dark, transverse bands; male gonapophyses not bifid at tS ste ee Dicranoptycha minima 66. Body coloration yellow; wing yel- GW etotee Dicranoptycha pallida Body coloration dark brown; wing grayish to, pale brown, --.2- =. 67 67. Male hypopygium with aedeagus bifid) at apex, (E18 733) ies ee Bee Se ie Dicranoptycha elsa Male hypopygium with aedeagus simiple.(Big:: 34) eee Serre: Dicranoptycha septemtrionts 438 68. 69, 70. NI WW THE UNIVERSITY OF Wibialzspurs present .2 2) =e. . (tribe Hexatomini) 69 Tibial spurs lacking . (tribe E Sereno) 2 Rostrum elongate, exceeding one- half length of entire body - (Elephantomyia) _Elephantomyia westwoodi Rostrum short, not exceeding in length remainder of head... 70 Wings with two branches of Rs reaching the margin (Fig. 35) (Atarba) _.. Atarba picticornis Wings with three branches of Rs reaching the margin A supernumerary crossvein in cell (Epiphragma) 72 No supernumerary crossvein in (All Cin eee ee anes ee ae eae : Wings with pale bands; a brown ring at tip of each 73 brown cross- femur _ Epiphragma fasciapennis Wings with an irregular pattern of brown; a brown ring before tip of each femur _.. Epiphragma solatrix Wings with Sc relatively short, Sci ending before level of fork of Rs; antennae with long, conspic- uous verticils; head not conspic- uously narrowed behind (Pilaria) Wings with Sc longer, Sci ending 74 opposite or beyond level of fork of Rs; antennae verticils not conspic- uously long; head strongly nar- rowed and prolonged behind (Fig. 30) (Pseudolimnophila) 76 Wings with cell Mi absent; gen- eral coloration of thorax blackish _ _Pilaria quadrata Wings with cell Mi present .......... 1s Thoracic dorsum dark brown to Kansas SCIENCE 76. SI J 78. i 81. BULLETIN black; antennae of males elongate, one-half length — of . Pilaria tenuipes exceeding body Thoracic dorsum yellow to brown- ish yellow; antennae short in both sexes, not extending caudad_ be- yond base of abdomen en SL 2 es ee Pilaria imbecilla Thoracic pleura gitudinally pale, striped lon- with dark brown, wines unmarked Jo == es UPredoumnoenite borate horace pleura uniform light to dark gray; wings marked with pale brown at stigma and along Nearly apterous species; wings smaller than halteres (Chionea) Fully-winged species Wings with cell Mi present (Fig. 36) (Cladura) Cladura flavoferruginea Wings with cell Mi absent (Fig. i) eA ed PLINY OT More 5 ba) Two branches of Rs reach wing- margin Three ee of Rs reach wing- margin Wings with Sc long, Sci ending beyondorigimof Rs. 25 ee ee (Teucholabis) 81 Wings with Sc short, Sci ending before origin of Rs (Fig. 37) Gonomyia manca Sc long, Sc: ending far beyond midlength Of RSs == eee mus eerie a Teucholabis complexa SIE Se Sci ending slightly be- yond origin of Rs Tue Crane Fires 1n Eastern Kansas 439 —————— a aa ; SS f )) NC a —) \| > SS 7 [hEG 38 Sy SSS A ten =e 44 ——— See i \ — ——~< — a > = 4p 39 SS SSS "Yyia > NS OS Se p (2, iff ° 3mm ° 0.5mm _——— ————————— scale, figs. 38- 41 scale, figs. 42-47 Fic. 38. Wing of Gonomyia sulphurella. Fic. 39. Wing of Ormosia romanovichiana. Fic. 40. Wing of Molophilus pubipennis. Fic. 41. Wing of Erioptera (Erioptera) septemtrionis. Fic. 42. Ventral dististyle of Limonia (Dicranomyia) divisa, dorsal aspect. Fic. 43. Ventral dististyle of L. (D.) haeretica, dorsal as- pect. Fic. 44. Inner dististyle of Gonomyia kansen- sis, dorsal aspect. Fic. 45. Inner dististyle of G. florens, dorsal aspect. Fic. 46. Ventral dististyle of L. (D.) liberta, dorsal aspect. Fic. 47. Ventral disti- style of L (D) pudica, dorsal aspect co bo Wings with vein Rg shorter than petiole-of cell Rg ..... (Gonomyia) 83 Wings with vein Rs longer than PetiolevoreclSi t ie. ce ela 86 83. Cell Rs very small, its extent along costal margin subequal to that of CIMCON CEO 655) 12 ei oes se ke RA a Gonomyta sulphurella Cell R; large, its extent along costal margin much greater than thatolael Roy seeker Cee oes. 84 84. Wings with cell Ist Me closed ........ ME AE Re See Nae Gonomyia subcinerea Wings with cell Ist Mz open, merged with» cell, Mg 2.2... 85 85. 86. 89. Ol. O2; Male hypopygium with inner arm of inner dististyle bifid (Fig. 44) . ee ee Gonomyia kansensis Male hypopygium with inner arm of inner dististyle simple, undi- vided (Fig. 45) ... Gonomyiza florens Wings with distinct macrotrichia in Outer Gelist, 2 ee ee ee 87 Wings with the outer cells gla- DrOUS(..5 8s La kd Ae ee meee 90 Size very small (wings 2.6 mm. or less); cell Rs sessile, without element R2+3+44; body dark brown ae: (Tastocera) ....... Tasiocera ursina Size larger; cell Rs petiolate by presence of Re+a+4 .... (Ormosia) 88 Wings with cell Ist M2 open pee ere ee core Ormosia ingloria Wings with cell Ist Mp closed ........ 89 Wings clouded with dark; anal veins convergent (Fig. 39) ws _...... Ormosia romanovichiana Rs ending in cell Rs, no element Ro+s44 (Fig. 40) vs (Molophilus) ot Rs ending in cell Ru, with short VEln Rog 4 Preseit) 25a 92 Body coloration pale reddish yel- low; fore femora extensively blackened, remaining femora yel- Jowish 2. Molophilus pubipennts Body coloration dark brown ........ rere oo 3 Molophilus hirtipennis Coxae of middle and hind legs only slightly separated by small meral region; knobs of halteres light yellow ........ (Gnophomyia) pmeA ee rete Gnophomyia tristissima Coxae of middle and hind legs widely separated by large meral TE GION 4)-Aak eeeees (Erioptera) 93 440 95. 94, 96. Tue University or Kansas ScreNcE BULLETIN A supernumerary crossvein in cell Rs; vein 2A strongly sinute _.... .. (subgenus Symplecta) ............ Bet tice teak Sex BO Erioptera cana No supernumerary crossvein in cell Rs Senin se ane Cee ee 94 Wings with anal veins convergent, cell Ist A at midlength usually as broad as, or broader than, at mar- SinGBiey 4): ae ee ta ee a peaniaeyeer? (subgenus Erioptera) 95 Wings with anal veins divergent, cell Ist A widest at margin 96 Knobs of halteres dark brown; body coloration brown _.......... ee ee Erioptera septemtrionis Halteres pale throughout; body coloration pale yellow 0... Erioptera vespertina Wings with cell Ist Me open ........ ee a (subgenus Mesocyphona) 97 Wings with cell Ist Mz closed... 99 Wings with a faint brown tinge, the cord and veins at margin with 98. we): 100. 101. small dark brown spots 5 site its ol ee a Erioptera parva Wings with a strong brown tinge, variegated with numerous white spots and dots) =a ee 98 Femora with two brown rings =k pease eran Oy Erioptera caliptera Femora with single brown ring ... ais. ones ee Erioptera needhami Wings with a spur from the angu- lated basal section of vein Mg jut- ting basad into cell Ist Me (subgenus Hoplolabis) 0. Cae oe ee Erioptera armata Wings without such a spur... (subgenus Psiloconopa) 100 Wings with two broad brown crossbands.<2 ea Erioptera venusta Wings with dark pattern broken into small spots or narrow broken DAIS tok ul te tee 101 Femora with two broad black TINS Sgeetra! Ne Erioptera graphica Femora with two narrow brown TUNES: 2 task ole Erioptera armillaris OCT 231978 HARVARD PP ooh eMe tate te tate stele eter eters a tatet et atare te etete ates a ete eter tetatate te ates eteretetetenetete ete steve eters steve ss e704 6 61s 018 6 616.8787: RR SoRSSNO IRIE: oa SR RRR ‘ene ‘oe see me "one ‘ee SEXUAL SIZE DIFFERENCES IN THE GENUS SCELOPORUS sees caeteteceree By HENRY FITCH ee Vol. 51, No. 13, pp. 441-461 September 25, 1978 ANNOUNCEMENT The University of Kansas Science Bulletin (continuation of the Kansas Uni- versity Ouarterly) is an outlet for scholarly scientific investigations carried out at the University of Kansas or by University faculty and students. Since its incep- tion, volumes of the Bulletin have been variously issued as single bound volumes, as two or three multi-paper parts or as series of individual papers. Issuance is at irregular intervals, with each volume prior to volume 50 approximately 1000 pages in length. Effective with volume 50, page size has been enlarged, reducing the length of each volume to about 750 pages. The supply of all volumes of the Kansas University Quarterly is now ex- hausted. However, most volumes of the University of Kansas Science Bulletin are still available and are offered, in exchange for similar publications, to learned societies, colleges and universities and other institutions, or may be purchased at $20.00 per volume. Where some of these volumes were issued in parts, individual parts are priced at the rate of 2 cents per page. Current policy, initiated with volume 46, is to issue individual papers as published. Such separata may be purchased individually at the rate of 3 cents per page, with a minimum charge of $1.00 per separate. Subscriptions for forthcoming volumes may be entered at the rate of $20.00 per volume. All communications regarding exchanges, sales and subscriptions should be addressed to the ExcHance Liprarian, UNIVERSITY oF Kansas Lipraries, LAWRENCE, Kansas 66045. Reprints of individual papers for personal use by investigators are available gratis for most recent and many older issues of the Bulletin. Such requests should be directed to the author. The International Standard Serial Number of this publication is US ISSN 0022-8850. Editor Eugene C. Bovee Editorial Board William L. Bloom Philip W. Hedrick Rudolf Jander Harvey Lillywhite Charles D. Michener Norman A. Slade Henry D. Stone George W. Byers, Chairman THE UNIVERSITY OF KANSAS SCIENCE BULLETIN Vol. 51, No. 13, pp. 441-461 September 25, 1978 SEXUAL SIZE DIPFERENCES.IN THE GENUS CGELOPORUS Henry S. Fitcu CONTENTS TESTE TVA BE iG cea Ma See Se AW dR ge ESE, 44] MRRP GION bieeere entrar went e. Retr eat ae on el nen tens el ee he a 442 A CIRO EISINGID ODDS RS” 9 RE ire ek UR Es a ead SER mM Ho eee ete a aE a ees EE 443 PROD SAND Ie VILAGE REAL Sa, arieh dete ket a Sere ee A oe ey cle Nand eee ee 443 PRES WT ume nash SUE Ey EE ES Cree Or Reet eee ot rowed eo ae ee Sa eee 443 Rata oer miasexUales ZeuCierelce.! aati siee eee 2s ee ee ere yes 443 inyilo ge niye eee ee ee Sede ten cea Ps Se ee eS ee a Se Se ee 447 SIZCROM GC LULCH ROMEO Sas ee Lev cane ee Mente Ie 9 i eee Se ea eee ee 447 Singlesormmuloplexciutches moses 2 ee eS no ee ee 45] Gliismiate pare re ee ied are cn ee Oe eee” EL 2 es Ld La seen ere 453 |BXOXG RYE Se cd Se aN) I PT oy ott ier rg USO re 0 Pe oe ee eRe ore EE 2 453 Ana itay RO lg VAN IUD UL ye es as ater oats ca tee ee nA dy, Sat eet vat als Siva i eee: 453 TESS (Ea ee «Bae AN heen 8 I tel ee ene hr Set Ras ROM SE ln 454 Wi sraleivam atch ese weet teten oe ete Re ci hme eet eed ee he 454 slime: required: to mature Sse. ee ens, ch. Alera sane 454 Geocraphiepvatiation 22.83.02. ls ese elt ak eee eee SN EN AS 455 Comparison of species having small female (varzabilis ) withmone havineslangetenmale s(Olidces)\) © 3 = i) i.e pare ee eee ee 455 NDISGUSSION ......----<--202--- hese siales het ih We Ph Uggs 601k SAMS tel SNC A a SAME AWE RE RN Os hee Re Ee 457 RAOIEU STON Sie ett aan te ea Ow ALL UR tr SRM rl OR OE Te ees 459 EER CUCIR EO UE Ie en eee eters cele Be mee tek, Jo ous Sg eC OAT Tee eee 460 ABSTRACT Measurements were accumulated for 53 populations of Sceloporus, representing most of the well known species in this large, iguanid genus. Males were larger than females in 33 popu- lations, and average male-to-female length (S-V) ratios varied from 123.9%, to 87.7%. Sig- nificant trends toward females being larger than males were found in species that: a) pro- duced a large clutch or litter (vs. small clutch or litter); b) produced a single annual clutch or litter (vs. multiple clutches or litters); c) lived in temperate (vs. tropical) climates; d) were small or medium sized (vs. large, with S-V exceeding 60 mm). Less significant correlation was found with phylogenetic groupings (Group II tending to have relatively larger males than Group III); and with mode of reproduction (viviparous species tending to have relatively larger males than oviparous species). Neither habitat (saxicolous, arboreal or terrestrial), nor development of male or female display colors, nor time of maturity (first to fourth year) showed any signifi- cant correlations with sexual size differences. Intraspecific variation in size ratios of the sexes was found in each of six polytypic species checked and in three of them (sealaris, graciosus, Oc- cidentalis) there was geographic shift from the male being the larger in one area to the female being the larger in another. 442 Tue University oF KansAs ScrENCE BULLETIN INTRODUCTION In the large iguanid lizard genus Scel- oporus, differences in size between the sexes have been noted by various authors, but no interspecific trends have been shown. In some populations males have been shown to be larger than females while in others this size relationship is reversed. In field studies of Sceloporus occidentalis in Oregon, S. wndulatus in Kansas, S. ma- lachiticus and S. variabilis in Costa Rica, and S. jarrovi and S. virgatus in Arizona, I found strikingly different size ratios be- tween the sexes and have investigated these ratios in other species to seek possible causes and correlations for them. Earlier (Fitch, 1976), I investigated the size relationships of the sexes in 54 main- land populations (representing 45 full spe- cies) of the iguanid genus Anolis and found a virtual continuum in male-to- female length ratios from 73.5% to 125.4%. Size relationships of the sexes in anoles were found to be strongly correlated with climatic conditions. Those kinds having a short and concentrated annual breeding season, enforced by unfavorably cold or dry weather prevailing for part of the year, have consistently large males, whereas those species living in aseasonal climates of rain forests and cloud forests have the sexes approximately equal in size or have females larger than males. The Anolis study led to an intergeneric comparison of sexual size difference in Sceloporus and Anolis. These two success- ful and dominant groups of iguanid lizards often attain high population densities, gen- erating intense competition within and be- tween species. Both are normally territo- rial, with aggressive behavior and spectacu- lar display organs well developed in males. However, Anolis centers its distribution in the tropics and thrives best in humid cli- mates while Sceloporus centers its distribu- tion in the warm, temperate zone and thrives best in arid climates. Anolis is unique among iguanids in consistently pro- ducing a one-egg clutch, laying at brief (and often regular) intervals, with left and right ovaries alternating in production. On the other hand, clutch size varies much among and within Sceloporus species, but with nearly always more than one egg and sometimes more than 20. In the more pro- ductive species, the capacity of the female as an egg container might be an adaptive character, subject to selection, which would alter the size relationships of the sexes. The majority of Sceloporus species are egg- layers, but many of those occurring in montane or northern climates are vivipar- ous and some of the oviparous species have evolved toward viviparity by retaining their eggs until the embryos are partly de- veloped before laying. These diverse repro- ductive strategies might be expected to affect the size relationships of the sexes. The Iguanidae are one of the major families of lizards within which it is a gen- eral rule that males are larger than females. These lizards often live in open places and are visually oriented. Many kinds maintain territories and there are stereotyped species- specific display movements that serve in part as territorial signals warning away po- tential rivals. The display organs, often brightly colored or conspicuously marked, are different in different genera (dewlap, belly patches, underside of tail) and can be presented threateningly to potential rivals, but at most other times are either incon- spicuous or are completely hidden. Species vary tremendously, both within and be- tween genera, in size of display organ and complexity of display. The female’s dis- play organ may be rudimentary or lacking; if present it is nearly always smaller and less conspicuous than the male’s. Relative size of the male in different species and genera seems to be correlated with aggres- siveness and with size and conspicuousness of the display organ. Atypical iguanids in- clude the predatory Crotaphytus (Gam- | | SEXUAL S1zE DIFFERENCES IN THE GENUS SCELOPORUS 443 belia) wislizenti in which the male is markedly smaller than the female, with no special display organ or behavior, and the solitary, cryptic, myrmecophagous Phrynosoma in which the sexes are approx- imately the same size and display organs are not developed. METHODS AND MATERIALS The essential data for this study were the snout- vent measurements of individual adult Sceloporus in substantial series. Thirty such series, from 1,973 specimens, were obtained from the collections in the University of Kansas Museum of Natural History. Measurements were available for 14 other series from published literature: Blair, 1960; Burkholder and Tanner, 1974; Cole, 1963; Crenshaw, 1955; Jackson and Telford, 1974; Mayhew, 1963; Mueller and Moore, 1969; Newlin, 1976; Parker and Pianka, 1973; Tanner and Krogh, 1973; Tinkle, 1973 and 1976; and Webb, 1967. Five series of specimens were examined in the Museum of Vertebrate Zoology, Uni- versity of California collection, and four series of measurements were obtained in the course of my field studies in Kansas, Oklahoma and Costa Rica. Whereas most of the meaurements were based on preserved museum specimens, those from my own field studies and from several published reports were taken from live lizards that were released after capture. Measurements of live material are not strictly comparable to those of preserved material. Hardening and shrinkage of the latter produced shorter measurements, perhaps several per cent less than would have been obtained from the same indi- viduals in life. However, the length ratios of the sexes were not affected, as each series of specimens measured consisted entirely of either living or pre- served animals. The problem of setting the minimum size limits for males and females has been discussed for Anolis (Fitch, 1976) and is similar for Sceloporus. Making the state of the gonads the sole criterion would have eliminated much of the available material, collected at times other than the breeding season. Actually, the criteria were somewhat subjective. In each sub- stantial series the distribution of records tended to approximate a normal curve, but usually was some- what skewed, with more small adults than large adults, and relatively few in the largest size classes. This was due to the fact that size is strongly corre- lated with age, the largest individuals being the oldest survivors, while the smaller adults include (along with some retarded older individuals) many that are newly matured and have been exposed for a relatively short time to normal mortality factors. Obviously, the composition of any local population varies accord- ing to season, depending on the climate. Average adult size is smaller when many newly matured indi- viduals are present and increases as these continue to grow after sexual maturity. Among the series in- cluded here are some that are composites, seasonally or geographically or both, and others that are rela- tively homogeneous. Some authors showing average difference between sizes of the sexes in specific popu- lations may have used different criteria for setting lower limits for adult size. These factors would all tend to increase the variance among populations. To account for species differences in size disparity of the sexes Wilcoxon 2-sample tests (Sokol and Rohlf, 1969) were used, with the 53 populations ranked according to their ratios and divided into two series that might be expected to differ (Table 2). Tests were somewhat limited by lack of eco- logical knowledge concerning the species involved. Size of clutch, frequency of clutch, time required to reach maturity, and even oviparous or viviparous habits are unknown for certain species. ACKNOWLEDGMENTS William E. Duellman. kindly permitted examination of the specimens in the Univer- sity of Kansas Museum of Natural History Collection, and also provided unpublished ecological information on several of the Mexi- can species. Robert Stebbins kindly permitted examination of specimens in the University of California Museum of Vertebrate Zoology. Virginia R. Fitch helped me with the record- ing and summarizing of data from museum specimens. RESULTS Range of Sexual Size Difference. Taste 1 lists the species and populations studied, ranking them in order from the one with the highest male-to-female size ratio (S. variabilis) to the one with the lowest (S. undulatus elongatus). The ratios range from 123.9% to 87.7% in almost a con- tinuum, but males are larger in 57°/ and for all series means combined, males average 104°% of female length. Males are most often larger than females, being territorial, pugnacious and equipped with bright col- ors for display, but it is necessary to explain why the female is larger than the male in 43°/, of the populations and with bulk averaging as much as 1.5 times that of the male. Ten ecological traits, all interrelated, and closely linked with reproductive strate- gies, were statistically tested, as set forth in 444 Tuer University oF KAnsAs SCIENCE BULLETIN TABLE: PopuLaTIon SAMPLES oF Sceloporus RANKED FRoM HuicueEst to Lowerst IN OrpER oF MALE-To-FEMALE- LenctH Ratiot | $ To @ LENGTH MEAN 6 LENGTH MEAN 9 LENGTH GEOGRAPHIC SPECIES AS PER CENT AND RANGE AND RANGE ORIGIN SouRCE variabilts 123958 1652/8 ==246 @/4-57 53.07 .491 (68-44 Costa Rica Fitch field in 97) in 157) Tees clarki 104.0+4.50(138-91 84.1 + 1.58(120-72 Sonora, boulengeri IVES shot 27!) in 36) Sinaloa KU Chihuahua, 116.392.56(130-100 96.95-1.80(116-86 Coahuila, poinsetti 120.04** in 18) in 21) Texas KU S. Calif., ATizs Parker & magister 119.50** 115.5(140-80 in 42) 96.6(120-80 in 33) N.M., Son. Pianka 1973 Michoacan, 53.50£.75(60-49 Colima, pyrocephalus 117.55** 62.891.02(68-58 in 9) in 12) Guerrero KU 60.84 + .456(67-53 52.34 .42(61-48 siniferus L624" sin 32) in 35) Oaxaca KU 60.15+.79( 65-53 Sonora, nelsoni HSS" 126) 52.14 .59(58-48 in 21) Sinaloa KU 50.724 498(60- 45.4845 591(57- cozumelae Ihe es 943 1m) 517) 41 in 33) Yucatan KU Mayhew orcutti 110.87** 102(115-90 in 17) 92(106-85 in 77) Ss Cali: 1963 78.75+1.53(91-61 jarrovt OS 75s" sah S5)) 71.875( 86-57 in 33) Arizona KU 82.6 .60(89-80 Michosean! Webb insignis 108.35** 89.5(99-80 in 10) in 10) Colima 1967 10207221. 11@18-97° 94.89221°32(107-88 Arizona, clarki 103:20%* sm 29) in 21) Sonora KU 65.28 1.13( 72-59 60.36 1.07( 66-54 adlert LOSM6**>in314) in 14) Guerrero KU 67.22+1.50(80- 62.24 1.65(77-55 smaragdinus 108.00** 60 in 14) in 17) Guatemala KU Sinaloa, 64.44 1.75(75-58 59.70+1.51(66-51 Jalisco, utiformis 107-95* 7 sii) 2) in 10) Nayarit KU Tinkle magister 107.87 - = 96 inl 89 in 21 Utah 1976 Veracruz, 55.87 1.03 (64-46 52.04 .668( 62-47 Oaxaca, teapensts LOPS 7. any 4}) in 26) Chiapas KU mucronatus 93.33+1.03(100-85 88.53+1.33(100-81 Veracruz, omiltemanus 105.43** in 21) in 17) Guerrero KU } 4 . 4 4 | SEXUAL S1ZE DIFFERENCES IN THE GENUS SCELOPORUS a4 TABLE 1.—(Continued) l, 6 To Q LENGTH MEAN 6 LENGTH MEAN @ LENGTH GEOGRAPHIC | SPECIES AS PER CENT AND RANGE AND RANGE ORIGIN SouRcE Campeche, 53.95 + .97( 62-45 51.30+.97(61-44 Quintana Roo, | chrysostictus 105.18* in 81) in 82) Yucatan KU | merriami 47.69 + 474(53-42 45,34 284(50-39 Chisos Mts. _annulatus OSS eeerim 96) in 62) Texas KU 52.28 .453( 61-45 49.82 .266(55-44 KU merriamt 104.93** in 60) in 51) S:lexas Fitch 79.12 .59(90-67 75.490 +.44 (86-64 field malachiticus 104.81** in 146) in 208) Costa Rica records | graciosus vandenbur- 60.2 .44(65-55 57.5 47(63-51 | gianus 104.70** in 34) in 26) Sy Galuts MVZ Tanner & Krogh | magister 10459 ** 999540 (115-83) 53) 95.04(107-81 in 57) S. Nev. 1973 | grammicus 51.26 .540(57-42) 49.28 .498(54-44 Coahuila, disparilis 1042022 23)) in 32) Durango KU occidentalis 75.364 .51( 84-65 TDi (89205 Se Calite biseriatus NO325 Oem ine97)) in 46) Baja Calif. MVZ Sinaloa to Webb bulleri 103.07 ~—-100.7(116-95 in 10) 97.7(108-91 in 10) Jalisco 1967 ZAlsi eel II(SNECS 68.65 + 1.49( 82-60 Chiapas, taeniocnemis 102.81 in 19) in 20) Guatemala KU 48.88 443 (51-47 47.864 1.15(52-44 Oaxaca, pictus 102.1 in 8) in 7) Puebla KU Michoacan, | scalaris 46.10 .709( 49-42 45.53 .621(53-41 Morelos, (“aeneus”) 10133 an‘ 10) in 23) Mexico, D.F. KU 88.29 1.67(99-82 OW 25 Ioy(QE17/ spinosus OLEZ3 erin 17) in 18) Oaxaca KU Jal., Mich., 103.54 1.76(118-98 102.671.46(110-97. Mex., D-F., Guan., torquatus 1O0}8555 Sines) in 9) Agua Cal. KU 45.204 1.26(50-42 44. 91+ .720(48-41 Veracruz, megalepidurus 100.65 in 10) in 11) Puebla KU undulatus 60.31 ==.704( 74-55 60.96 + .632(71-55 aexals; | consobrinus OEery sim 45))) in 46) N. Mexico KU Tinkle | graciosus 98.00 49.0 in 25 50.0 in 39 S. Utah 1973 71.58 1.38( 80-64 73.88 1.65 (80-68 formosus 96:90 inl) in 8) Oaxaca KU 446 Tue University oF Kansas ScrENCE BULLETIN TABLE 1—(Concluded) | 6 To 9 | LENGTH MEAN 6 LENGTH MEAN 2 LENGTH GEOGRAPHIC 4 SPECIES AS PER CENT AND RANGE AND RANGE ORIGIN SOURCE graciosus 52.1 .416(61-49 53.9 341(63-48 i “gracilis” 96.66** in 85 in 76) Oregon MVZ } Burkholder — 57.39(63-52 59.91( 69-53 and Tanner — graciosus 95.79** in 106) in 121) Utah 1974 100.733.51(116- KOS Oke UIC) Mexas, i cyanogenys 95.01* 86in 8) 88 in 22) Tamaulipas KU 90.0 1.47(93-86 94.73 +2.66(99-91 “g lundelli 95.01** in 4) in 6) Yucatan KU j occidentalis 66.09 .694(72-61 70.38 1.44(77-68 } occidentalis 94.0** in 23) in 13) W. Oregon MVZ Jackson and woodi 94.28** 47:6 50.5 Florida Telford 1974 scalaris 45.36 .63(50-43 48.82 1.33(55-42 Veracruz “bicanthalis” 94:14** in 14) in 11) Mexico, D.F. KU undulatus 58.58 .72(70-52 62 7e= 5775-57. Arizona tristichus OBL SiS? sol 35))) in 53) N. Mexico KU undulatus 59.82 + 59( 63-57 64.23 .87(67-57 Fitch field hyacinthinus OSes steers) in 11) Oklahoma records occidentalis 73.64+1.18(81-65 82.73 1.43(87-72 E. Oregon, biseriatus 9310024 sans) in 21) Idaho MVZ undulatus 52.222= 37 (69-45 56.25 591 (68-53 Fitch field garmani eter r in (ys) in 44) Kansas records Crenshaw undulatus 90.33** 56.05(65-47 in 59) 62.05(70-53 in 35) Georgia 1955 undulatus 5952== 565-53 66.24 .53(72-60 erythrocheilus 89.86** in 21) in 21) N. Mexico KU Blair olivaceus 89.14** 82.9(93-60 in 34) 93.0(.107-63 in 107) Texas 1960 45.5342 .)7(5)-40 51.25 36( 60-40 Newlin scalaris 88:34" 51m 45)) in 203) S: Galit 1976 Fitch field virgatus 88.43** 52.0(58-48 in 11) 58.8(69-51 in 10) S. Arizona records undulatus 63.104 1.09(71-55 71.95 1.03(83- elongatus 87.70** in 20) 65 in 20) SW Col. KU * Significant dimorphism assumed where P S 0.05 (one asterisk); two asterisks indicate P S 0.01. SEXUAL S1zE DIFFERENCES IN THE GENUS SCELOPORUS 447 the following sections. Taste 2 shows the extent of correlations as revealed by Wil- coxon 2-sample tests. Taste 3 shows the relationships of the species studied, and the occurrence of various ecological traits among them. Phylogeny. Relationships within the genus and to other genera of iguanids are well known through many osteological, mor- phological, karylogical and behavioral stud- ies (Etheridge 1964; Smith 1939; Cole 1963; Purdue and Carpenter 1972). Smith (1939) separated the 95 species and sub- species of Sceloporus which he considered valid into 15 groups of approximately equivalent morphological value. Smith’s arrangement was accepted for 35 years, but eventually was revised and extended by Larsen and Tanner (1974 and 1975). They used over 80 characters, including lepidosis, skull morphology, distribution, behavior and karyology, and applied a statistical treatment with Ward’s cluster analysis to determine degrees of relationship within the genus and construct dendrograms re- flecting them. They divided the genus into three primary groups, each having several subgroups of from one to nine spe- cies. Group I, the smallest of the three, with only three subgroups and seven spe- cies, was considered to be the most primi- tive and the most distinct, and in the 1975 publication it was suggested to comprise a separate genus (Lysoptychus, Cope 1888). Group II with 20 species and Group III with 30 were each found to consist of five subgroups. Only one species in my study, Scel- oporus (Lysoptychus ) merriami (with two populations), was a member of Group I, but 13 species of 14 populations represented all the subgroups of Group II, and 19 spe- cies with 30 populations represented all the subgroups of Group HI. The samples are therefore considered to be representa- tive of the genus, since the species not in- cluded are mostly rare and obscure ones. In a Wilcoxon 2-sample test (Table 2), Group II and Group HI are significantly different at the 95°% level in sexual size differences, with Group III having rela- tively smaller males. However, in each group there are species in which males are larger than females, and vice versa. The subgroups show more significant contrasts. The sexes are approximately equal in size, but with males slightly larger in Group I, Subgroup B (merriami merriami and mer- riamt annulatus) and in Group II, Sub- group A (grammicus, pictus and mega- lepidurus). Males are relatively large in Group II, Subgroups B (pyrocephalus, nel- sont), D (sintferus, utiformis) and E (vart- abilis, cozumelae, teapensis, and chrysostic- tus) and in Group IL, Subgroups A (spi- nosus coeruleopunctatus, orcutti, clarki clarkt, clarki boulengeri and magister—but with the notable exception of olivaceus) and D (jarrov1). Females are generally larger than males in Group III, Subgroup C (undulatus and subspecies, occidentalis except near its southern limits, graciosus except near its southern limits, virgatus, and woodi). In Group III, Subgroups B (lundelli, formosus, adleri, smaragdinus, taeniocnemis, and malachiticus) and D (torquatus, cyanogenys, bullert, insignis, mucronatus omiltemanus, and poinsettt) neither sex was consistently larger. Size of Clutch or Litter. Number of eggs per clutch varied from one (chrysostictus) to 19 (torquatus) in the specimens exam- ined. Blair (1960) recorded a maximum of 30, laid by a large female of olivaceus. Mean clutch size varied from 1.8 in co- zumelae to 14.3 in olivaceus. Clutch sizes of various species and populations are shown in Table 4, some based on published literature, others based on dissections of specimens in the collections of the Univer- sity of Kansas Natural History Museum. Table 5 shows intraspecific variation in clutch size in the wide-ranging species graciosus, occidentalis and undulatus. 448 Tue University oF Kansas SCIENCE BULLETIN TABLE, 2: Witcoxon 2-sAMPLE Tests oF CorrELATIONS IN Sceloporus PopuLATIONs RANKED AccoRDING TO Mavce-FEMALE LENGTH RATIOs DIvIsIon OF NuMBERS IN SAMPLES SAMPLES t-VALUES small brood, mean < 4 33 SOs Vs. large brood, mean > 4 10 single annual clutch or litter 28 3305 VS. multiple clutches 14 tropical 20 282 "Ss Vs. temperate 33 male, less than 60mm S-V 2D 2.60" vs. male, more than 60mm S-V 31 Group II 14 D233" VS. Group III 37 oviparous 36 [97% VS. viviparous 7 saxicolous 12 le. vs. arboreal or terrestrial 35 female display patches developed 12 1.70 vs. female display patches faint or absent 4] male display patches developed 49 ieee! vs. male display patches faint or absent 4 maturity attained in first year 33 334 VS. 18 maturity attained 2nd to 4th year ** Significant at 99%. * Significant at 95°%. Species whose reproductive strategy in- volves producing a large egg-clutch (or litter) may be subject to selective pressure to increase body size of the female as a more capacious egg container. Sexual size difference showed higher correlation with clutch size than with any other factor tested and species producing large clutches or litters tended to have relatively large females. Table 2 shows that 33 populations SEXUAL S1zE DIFFERENCES IN THE GENUS SCELOPORUS 449 ABIL aS Ecovocicar Traits or VArtous Sceloporus SPECIES bp = es sae ea 2 |. ote: 6 Se ee rete te es = de as ma tea eee errr eee ch ae eee Soe en One as a > o ao See Sa aa a ae Be oe eB eee 2 Be YS eee ee YN < 0 KH CO & Or Wak aay 2 ts mH & adlert S Trop. Hl B Vv 7. No Yes | bulleri S) Trop. HIE V S Yes Yes 2 chrysostictus M Trop. IE O at No No 1 clarki M Temp. cA: O A No Yes 1 cozumelae M Trop. je? O ale No No 1 cyanogenys S Temp. Ds V S) No Yes Dats formosus S Trop. Il B Vv A No Yes 1 graciosus S-M Temp: ie O rT No Yes 2 grammicus S Temp. IA Vv A No Yes 1 INSIgNIS S Trop. JH ipl Vv S No Yes z jarrovi S Temp. Hl D Vv S No Yes 1 lundelli ? Trop. Ill B V A No Yes i magister S Temp. Il A O A No Yes 2 malachiticus S ‘nop: Il B Vv A Yes Mes 1 megalepidurus S Trop: II A Vv No No ? merriami S Temp. IB O S Yes Yes 1 mucronatus S Trop. UNE Vv A No Yes 2 nelsoni S Temp. Il B O a No Yes 1 occidentalis S-M Temp. . UIC O A Yes Yes 2 olivaceus M Temp. WA O yen No Yes 1 orcutts S Temp, HA O S No es 2; pictus ? Trop. Ih A Vv ? No Yes 1 poinsetti S Temp. I E Vv S No Yes 2 pyrocephalus ? Trop. Il B O aR No ies I scalarts 5 Temp. ihe O-V aE No Yes 1 siniferus M Trop. II D O ae No Yes 1 smaragdinus S Trop. Hl B Vv A No Yes 1 spinosus P Temp. HI A O A No MES taeniocnem1s S Trop. 00 is: Vv A No Yes 1 teapensis M Trop. 1 Rial ss O ae No Yes 1 torquatus S Trop. TT V S No Yes 25 undulatus S-M Temp. Lie O S-A Some Mes 1-2 utiformis M Trop. litieats) O alk No No 1 variabilis M Trop. II E O a No Yes 1 virgatus S Temp. hee O a No No 1 woodi M Temp. ti-€ O ah Wes Yes 1 450 Tue University of Kansas ScreNcE BULLETIN TABLE 4 Size oF CLurcH or LirrEer in Various Species AND PopuLATIONS OF Sceloporus CouNTSs FROM SPECIES OR MEAN cLUTCH INDIVIDUAL POPULATION OR LITTER RANGE N FEMALES REGION AUTHORITY adleri 3.8 2-6 5 2,3,4,4,6 Guerrero KU Michoacan, Morelos, aeneus (=scalaris) 5.22 .428 4-7 OSE ees Mexico, D.F. KU bicanthalis (—=scalaris) 6.75) ~— 2... al) ee eee Veracruz KU a Campeche, Quintana Roo, chrysostictus 2.43 466 1-4 Gay ieee Yucatan KU clarki boulengeri Se ee 5 4,7,10,10,10 Sinaloa KU Maslin cozumelae 1 a ce TDS ARO oe ce Yucatan 1963 S. Texas, Hunsaker cyanogenys 133 6-18 ene = ee renee Tamaulipas 1959 formosus 2) a ern 4 7,7,9,9 Oaxaca KU Werler grammicus disparilis Df. 4-7 ear ne Veracruz 1951 Tinkle and jarrovi Gy D=a32 My ote Se We teas ete: SeAriz. Hadley, 1973 Goldberg jarrovi i, a aaa CDN aIT 2122 eotes2 S. Ariz. 1971 Ballinger jarrovi 2 ie oN | oe Do Pee ak S, Ariz; 1973 Tanner and magister 6.6 4-10 / Se ee S. Nev. Krogh, 1973 — Parkerand magister 8.4 3-12 Le eee ees S.W. states Pianka, 1973 | Tinkle | magister 6.2 2-9 | 5) I ESRC AD S. Utah 1976 | Fitch | malachiticus cE een eee ee DN eee Costa Rica 1970 Chaney and © merriamti So 25) De ree ce o. S. Texas Gordon, 1954 — Werler TUUCKONAIUS 9 Veracruz 1951 Sonora, nelsoni (so) Sn iy eerie + 46,758 Sinaloa KU Blair olivaceus mee Oe esha gs A eh ence Texas 1960 a. Mayhew | orcutti eo ay Boe ye a eae AR ee S. Calif. 1963 SEXUAL S1zE DIFFERENCES IN THE GENUS SCELOPORUS 451 TABLE 4.—( Continued) CouNTs FROM | SPECIES OR MEAN CLUTCH INDIVIDUAL POPULATION OR LITTER RANGE N FEMALES REGION AUTHORITY | Oaxaca, KU \pictus Soya NEM aroha i atiniee st ie 5) 2,2,4,4,6 Puebla Ballinger (poinsetti 10.45+1.01 6-23 Grae Ti ee ee S.W. Tex. 1973 | Michoacan, \pyrocephalus OOM Ir. ed | ceases 3 Bras. Colima KU 6.22+0.42 Ist yr 2 2 49 | 10.540.58 2nd yr Newlin -scalaris ainel Ole? OQ see 37 SwAmiz: 1976 \ siniferus 5.0 4-6 AN ing emer eran S.Oaxaca KU _ smaragdinus P20 a= 344 3-6 [Okara arabes Guatemala KU | “spinosus 7 coeruleopunctatus 12.66 8-16 Gira Seiad ett Oaxaca KU teapensis Tae an a ne Pe eee 3 PES, Yucatan Pens U; ¥ Werler SEAGIG >< NORDIN SECU SO Mam an ete er 6 Michoacan 1951 i Fitch _ variabilis S00 ae Ue eu eles lic Pa allae espera Costa Rica 1970 i Vinegar _ virgatus 9.45424 4-16 RCA ER ee Saiz: 1976 virgatus 10.2 5-15 | Sy eauaeeteeare secu rae SeAniZ, Cole, 1963 i Jackson and _woodi AS etree 9) Shee in (Waa Oy Bin pe ie ate Ie Florida Telford, 1974 having small clutches < 4 were signif- cantly different, at the 99°% level, in sexual size difference, from 10 populations having large clutches or litters > 4. Single or Multiple Clutches. Some species in this study are not known to produce either single clutches or multiple clutches, and are omitted. So far as I know, all vivip- arous species are single-brooded, since ges- tation extends over several months Ovip- arous species that are also single-brooded include graciosus at high elevations and northern latitudes (Burkholder and Tan- ner, 1974; Mueller and Moore, 1969); mer- riami (Chaney and Gordon, 1954); wir- gatus (Vinegar, 1975); orcutti (Mayhew, 1963) ; magister (Parker and Pianka, 1973) ; and occidentalis (Fitch, 1940; Goldberg, 1974). Species and subspecies known to be multiple brooded include wndulatus undu- latus (Crenshaw, 1955; Tinkle and Bal- linger, 1972); undulatus garmani and un- dulatus hyacinthinus (Fitch, 1970); wood: (Jackson and Telford, 1974); oltvaceus (Blair, 1960); and variabilis (Fitch, 1973). I found that S. clarki boulengert, chrysosti- ctus, cozumelae, teapensis, spinosus coeru- leopunctatus, siniferus and utiformis all had young in various stages of growth at dif- ferent times of year, indicating a long breeding season and multiple clutches. The species known to be single- clutched, when arrayed against those known to have multiple clutches, and sub- jected to a Wilcoxon 2-sample test for correlation with sexual size difference (Ta- Tue University oF Kansas Sc1ENCE BULLETIN TABLE 5 Ciutcu Size 1x Various Popuations oF Sceloporus graciosus, S. occidentalis anv S. undulatus MEAN SPECIES CLUTCH RANGE N REGION AUTHORITY graciosus Oregon, “gracilis” 3:6 meee 32 INECalit, Fitch, 1970 graciosus Burkholder and graciosus 6.03 2-10 143 N. Utah Tanner, 1973 graciosus graciosus oe vp) S. Utah Tinkle, 1973 graciosus S. Calif., vandenburgianus oe ae 25 Baja Calif. Fitch, 1970 occidentalis occidentalis Ee 14 W. Oregon Fitch, 1970 occidentalis Central Sierra, Jameson and occidentalis |G aie 2 ee oe a 1500 m Allison, 1976 occidentalis Central Sierra, | Jameson and occidentalis I= eee Wek 2200 m Allison, 1976 occidentalis Tanner and longipes Tiled 7-14 1D S. Nevada Hopkin, 1972 occidentalis SaCalit,. biseriatus 7.65 3-14 37 Baya Galit: Fitch, 1970 occidentalis Los Angeles Co., biseriatus eT Ne Re 84 California Goldberg, 1973 occidentalis Whittier, biseriatus TDS Fyre a) California Goldberg, 1974 occidentalis San Gabriel biseriatus S/n tenses 41 Mts., Calif. Goldberg, 1974 occidentalis E. Oregon, Fitch unpublished biseriatus BO Ge ee 9 Idaho (MVZ specimens) undulatus undulatus Tae aa ae me ORL Georgia Crenshaw, 1955 undulatus Tinkle and undulatus Wena 2 Ona te Ve S. Carolina Ballinger, 1972 undulatus consobrinus 62 3-8 13 Oklahoma Carpenter, 1959 undulatus elongatus O322 03.) ce) oy tace: Utah Tinkle, 1972 undulatus erythrocheilus 9.0 4-13 6 Oklahoma Carpenter, 1959 undulatus garmani 7.6 5-12 10 Oklahoma Carpenter, 1959 undulatus Tinkle and Ayacinthinus SE OE=ie. eee eee Ohio Ballinger, 1972 SEXUAL SIZE DIFFERENCES IN THE GENUS SCELOPORUS 453 TABLE 6 TREND OF DecREASING MALE-TO-FEMALE S1zE From WarM To CooL CLIMATE MEAN MALE-TO-FEMALE RANGE POPULATIONS SAMPLED CLIMATIC ZONE LENGTH AS PERCENTAGE N Tropical lowlands 2... 109.02 (93.3-130.8) 110 Tropics (both lowland and montane) ..... 106.22 (93.3-130.8) 19 Sliroprcaliimontanes<< 2:2. Aye Se 103.11 (96.9-108.2) a eS: Memperate zone (all samples) 12... 100.20 (87.3-123.7) 34 Memperatcs (USA)! see ie nrc eek 99:15 (94:3-120.1) 29 Temperate (USA excluding SOUbMerm tier Or states) 22.2 2.2.2.27..:. 94.82 (87.7-107.9) 13 ble 2), showed correlation significant at the 99.9°%, level. The single-clutch species have relatively small males and large females. Climate. Sceloporus occurs from about 48° 30’ N near the Canadian border in Washington south through much of the continental United States, Mexico and Central America to about 9°N in Panama. Hence, its local populations are adapted to a wide range of climates from those with long, intensely cold winters and short sum- mers to those with hot, aseasonal climates, or those with extreme, desert conditions. Warm, dry conditions are optimum, how- ever, since many species and individuals occur in the arid, southwestern United States and adjacent Mexico. Seasonal schedules, and reproductive strategies obviously are much altered by climatic factors. Table 6 shows a well- defined trend from relatively small males in cooler climates to relatively large males in the hot climates of tropical lowlands. In Table 2, with 20 mainly tropical species arrayed against DD, 33 species from temperate North America, males tend to be relatively small in the Temperate Zone, with correla- tion significant at the 99°%, level. Body size. Adults of Sceloporus ranged from 39 mm S-V in female S. merriami an- nulatus to 138 in male S. clarki boulengert, with means ranging from 45.2 in male megalepidurus to 105.9 in female cyanog- enys. Most populations studied were in the lower size brackets with decreasing num- bers toward the upper limits. The 22 small- est species (male S-V averaging less than 60 mm) when arrayed against the 31 larg- est than 60 mm) were found to be significantly different (slightly below the 999% level, see Table 2). Species of small body size tend to have With the largest species (those exceeding 90 mm) (averaging more relatively large females. ten arrayed against the remaining 43, differ- ence in sexual size dimorphism was some- what less significant (t=1.78). Oviparity or Viviparity. The oviparous state, primitive for the suborder Sauria and the family Iguanidae, persists in the ma- jority of species of Sceloporus, but many in montane habitats and some that are not Sull others have progressed toward viviparity montane have become viviparous. by retaining eggs during part of their development. Viviparous species include torquatus (Mulaik, 1946), pornsett: (Ballinger, 1973), cyanogenys (Crisp, 1964), mucronatus omiltemanus (Davis and Dixon, 1961), and by inference their near relatives in Group C, Subgroup E, bzdllert and insignis, jarrout’ (Ballinger, 1973), malachiticus (Fitch, 1970; Marion and Sexton, 1971) 454 Tue University oF KANsAs SCIENCE BULLETIN and by inference the near relatives of ma- lachiticus: formosus, smaragdinus, taento- cnemis and lundelli; aeneus in part of its range (Thomas and Dixon, 1976), gram- micus disparilis (Davis and Dixon, 1961; Mulaik, 1946), pictus (Smith and Savit- sky, 1974) and megalepidurus. Other spe- cies are, so far as I know, oviparous. The oviparous species, arrayed against the viviparous, show a tendency to have males smaller than females, but with the difference not significant at the 959% level (Table 2). Habitat. The species of Sceloporus occur in a spectrum of habitats, terrestrial, arbo- real and rocky. However, the genus is gen- eralized, rather than highly specialized for any of these habitat types. There are some euryecic species that occur in a variety of habitats. S. andulatus, especially, has popu- lations adapted to diverse habitats, includ- ing terrestrial (garmani), arboreal (Aya- cinthinus) and saxicolous (elongatus), without conspicuous morphological adapta- tions. In general, the terrestrial species are fine-scaled, with bodies, scales, and limbs slender and tapered, whereas arboreal and, especially, saxicolous species tend to be coarse-scaled with relatively short and thick bodies and appendages. The terrestrial spe- cies are swift runners, but scansorial spe- cies are less active and depend more on cryptic patterns and behavior and on secure hiding places in cavities and crevices. In 22 populations considered mainly arboreal, male-to-female length ratio was most often nearly equal (mean 101.7%), with male superiority greater in 12 popu- lations considered mainly saxicolous (mean 103.894) and 20 populations considered mainly ground-living (mean 105.694). None of these three groups differed statis- tically from the others to a_ significant degree. The kinds considered to be mainly saxicolous (bulleri, cyanogenys, insignis, jarrov1, merriam1, orcutti, poinsett, torqua- tus and the subspecies elongatus and ery- throcheilus of undulatus) were most de- viant from others in sexual size difference, but with difference not significant at the 95+ level: Display Patches. Brightly colored (usu- ally deep blue) display patches are present on the chin and on the sides of the belly in the males of most species. These patches are either lacking in the female, or are barely discernible as slightly darkened areas without bright color, or if they are colorful they are paler than those of the male and less extensive in area. Even though having some display color, the female may lack either the lateral body patch or the chin patch. Female iguanid lizards, including Sceloporus, are known to perform the stereotyped bobbing display of their spe- cies. Even for those that lack colorful dis- play areas, movements may nevertheless serve for territorial assertion, or may func- tion in species-recognition and sex-recogni- tion. Even within one sex in a local popula- tion, development of colored display patches may vary, being present in some, faint or absent in others, so the following groupings are somewhat arbitrary. Females of malachiticus, clarki, mer- riamt (2 populations), bzllert, woodt, oc- cidentalis (3 populations) and in undula- tus the subspecies elongatus, erythrocheilus and tristichus, have colored display patches more or less developed; in other popula- tions female display colors are absent. In males, only cArysostictus, cozumelae, utt- formis and wirgatus lack bright ventral display colors. Table 2 shows that pres- ence or absence of display colors in either sex are not strongly correlated with sexual size difference, but there seems to be some tendency for display colors to develop in females of those species where the females are relatively large. Time Required to Mature. A combination of innate physiological traits and environ- SEXUAL S1zE DIFFERENCES IN THE GENUS SCELOPORUS 455 mental factors affects rates of development and time required to reach maturity. This time varies in different populations, from three months to three years or more. Actual records of individuals, based on mark and recapture, are available for few popula- tions, but well defined age-size cohorts are observable in some. In general the species are either early maturing (at age of one year or less) or late maturing (in second year or later). The late maturing species include those of Group IH, Subgroup E, torquatus, cy- anogenys (Crisp, 1964), potnsett: (Ballin- ger, 1973) and by inference their near rela- tives bullert, insignis and mucronatus; graciosus (Mueller and Moore, 1969; Tinkle, 1973), magister (Tanner and Krogh, 1973; Tinkle, 1976), orcutti (May- hew, 1963), occidentalis, clarki and undu- latus, subspecies elongatus tristichus and erythrocheilus (Fitch, 1970). So far as I know, all others are early maturing, but megalepidurus and spinosus coeruleopunc- tatus, being little known, were not included in the comparison. No correlation between early or late maturity and relative sizes of the sexes is indicated (Table 2). Geographic Variation. Geographic varia- tion in sexual size difference was found in all six species for which intraspecific com- parisons were made. Seven subspecies of Sceloporus undulatus were tested and com- pared. Females were larger in all of them, but the male-to-female length ratio varied from 87.70% (elongatus) to 98.94% (con- sobrinus). Three populations of S. occt- dentalis were compared. In S. occidentalis occidentalis of western Oregon, and S. occi- dentalis biseriatus of the same latitude in eastern Oregon and Idaho, males were smaller than females—89.16°%, and 87.73%. However, in S. 0. diseratus of southern California and Baja California males were slightly larger than females, 103.5694. A parallel trend was found in populations of S. graciosus; in northern S. g. graciosus, from western Oregon and northern Cali- fornia, Yellowstone National Park, and Utah County, Utah, males were smaller than females (96.66, 98.00 and 95.79°/, re- spectively). However, in S. g. vanden- burgianus of southern California, males averaged slightly larger than females (104.69%,). Thus, the intraspecific trends of the wide-ranging occidentalis and gra- ciosus in sexual size differences parallel interspecific trends for the genus as a whole. Comparison of Species Having Small Fe- male (variabilis) with One Having Large Female (olivaceus ). Through Blair’s study (1960) S. oltvaceus of southern Texas is ecologically the best known species of Sceloporus by far. It is near the extreme of species having relatively large females (1.12 times male length). Most other spe- cies that have been subjects of intensive field studies, including several of the sub- species of wndulatus (Crenshaw, 1955; Tinkle, 1972; Tinkle and Ballinger, 1972), occidentalis (Fitch, 1940; Tanner and Hop- kin, 1972), graciosus (Tinkle, 1973; Tanner and Krogh, 1973; Mueller and Moore, 1969), virgatus (Vinegar 1975), and woodi (Jackson and Telford, 1974) are also among those with relatively large females, and available information suggests that, in general, their ecology and social systems are similar to those of olivaceus. There are no comparable studies of the species with relatively large males. For olivaceus, Blair (1960) found no well-defined territories, but each adult male had a home range with a principal station and an average of 8.5 additional stations among which he distributed his time. The stations were on trees, fence posts, or other objects on which the lizards could climb. They used intervening open areas only in crossing from one station to another. Home ranges overlapped exten- sively and the same station might be used by two or more males, but usually both 456 Tue University oF KANSAS SCIENCE BULLETIN TABLE+/ Democrapuic Traits ContrasteD IN Sceloporus olivaceus ANd Sceloporus vartabilis S. OLIVACEUS S. VARIABILIS Sex ratio in numbers of adults, Gr SONG ee She had aes 1 to 1.73 i Isto 1.22 mean 2 home range, ae oe Pe Se 27TS 326 MIME AING Oe MOUNE. LAINE IIS Sencar ee es res coved Seen eres 684 580 RUBt@ne SION OC GAMO! sc. Mi as) we ee a 1 to .89 Ltol24 Oe top natchiimovlengthh rato} aes acee cee essccteetec nto ce sects 1 to: 29 I to: -432 MAYAN piarcl DN el alt 7c & eee ARS es oe [a3 3.0 number of Alatches PVG I ea eae eer eR 3 SS 3 minimum time from hatching to maturity (months) .... 10 4 were not present simultaneously. When two met at the same station, they fought, with one being driven off temporarily. Blair described mating as promiscuous, but his narrative account indicated that “con- sort pair” associations were frequent. The male might spend periods of days in close association with a temporary mate, whether or not she was sexually receptive. How- ever, the male’s range was 2.3 times that of the female’s, and by shifting from one station to another he might associate with a succession of females that overlapped his range. The females were far more tolerant of one another than were the males. Often, two shared a station and Blair witnessed a female-female chase on just one occasion. In olivaceus the male’s display patches on each side of the abdomen are small and narrow. In the female they may be absent, or when present are smaller and paler than the male’s. At the other extreme, vartabilis is the species with greatest sexual size difference and relatively large males (1.24 times fe- male length). Clues concerning the signifi- cance of relatively large males vs. relatively large females could no doubt be obtained by comparing ecological and behavioral data of olivaceus and variabilis. My field study of variabilis in Costa Rica, 1967-1970 (Fitch, 1973) did not include intensive ob- servation of individuals, but more than 1000 lizards were individually marked and 374 were recaptured after substantial inter- vals. Some facts concerning the general ecology and social system of variabilis were obtained. Table 7 contrasts some traits of the species Olivaceus and variabilis. Significant facts revealed concerning variabilis are that: 1) It occurs in extremely high popu- lation densities, especially on the coast along the upper beach. In early December 1967, there were three adult males, six adult females and 52 immatures living within a 10 m radius. Four months later, in the dry season, the same area was occu- pied by six adult males, nine adult females and four immatures. 2) Where density is high, with ranges of many individuals in- cluding adult males, overlapping, there are not discrete territories. 3) Male display, fighting and pursuit is prominent behavior. 4) Male-female associations are common, but ephemeral. 5) The preferred habitat is seral, subject to continual successional change and to gross disturbance. On the beach, especially, favorite stations or look- outs and even the lizards themselves may be swept away in the tide and dropped at new locations (Fitch, 1973). Instability characterizes the habitat and the local pop- ulation. SEXUAL S1zE DIFFERENCES IN THE GENUS SCELOPORUS 457 DISCUSSION The relative body sizes of adult males and females in Sceloporus vary widely, males averaging approximately 249% longer than females at one extreme, and 12.5% shorter than females at the other extreme. In any population, the size relationship of the sexes depends on the interaction of many selective factors. The equilibrium is easily altered and even within a species local populations differ in the size ratios of the sexes. The ecological factors that determine size ratios between the sexes in Scéloporus are probably somewhat different from those acting on any other group of animals. For instance, in birds and mammals and some lower vertebrates parental care of young is an important aspect of behavior, and in predatory kinds the female often protects the young against the potentially cannibalistic male. In such cases as the spotted hyena (Kruuk, 1972), and various raptorial birds (Hill, 1944; Earhart and Johnson, 1970; Mosher and Matraz, 1974) it is therefore advantageous for the female to be larger than the male. In Sceloporus cannibalism is rare, as these lizards are mainly insectivorous. There is no ma- ternal care. Social behavior is primitive and does not involve family ties or group activities. A male and female may inti- mately associate, whether or not the female is sexually receptive. The male may ha- bitually interact with a neighbor along a territorial boundary and he may tolerate in his territory various non-rivals, including females, juveniles and subordinate, but sexually mature, males. Presumably, any departure from parity in the sizes of the sexes permits some partitioning of food resources (Amadon, 1959; Mills, 1976; Rey- nolds, 1972; Selander, 1966; Verner and Willson, 1969) and the greater the size difference the smaller the overlap. Which- ever sex is smaller is subject to competition from immatures. In the more prolific spe- cies, such competition might be severe at certain seasons. Larger size of the female may promote successful reproduction by relieving her of competition, both from the male and from immatures, and by per- mitting her to dominate the male when heterosexual competition does occur. Compared with the diversity in anole dewlaps, the colored, ventral display areas of Sceloporus show remarkably little inter- specific variation. The color patch on the chin, usually blue, may or may not be divided into distinct left and right portions, and may or may not be connected with the belly patches of the same color. In most species the belly patches on the two sides are separated by a paler mid-ventral area, but in old individuals of some kinds, black inner margins of the patches encroach and may fill the intervening space. In most species the dorsal color is cryptic, dull- brown or gray with streaks and spots, and with a paired series of darker transverse blotches on the back which are more prom- inent in juveniles and females than in adult males. In the latter, the ground color is darkened and the original markings be- come obscure, but when the animal is warm and active, a pale bluish or greenish area may show at the base of each scale. In the excitement of courtship or territorial defense, the colored dots expand so much that the male becomes gaudy and conspicu- ous and the entire body is involved in the display. Under most conditions, the belly patches are hidden as the lizard sprawls on the substrate, or raises its body only slightly in crawling and running, although the chin color may be visible from front or lateral view. However, when the lizard displays, it stands high and flattens the body in a vertical plane, puffs out the throat and turns sideways to an opponent, presenting its ventral colors conspicuously. It might be expected that in Sceloporus species having relatively large males these 458 Tue University oF KANsAs SCIENCE BULLETIN would have: 1) well developed display patches and display behavior, 2) aggressive behavior, combat, and maintenance of well- defined territories and 3) mating systems that usually do not involve enduring pair associations, but are promiscuous or polyg- ynous, with maintenance of leks or har- ems. Conversely, in species having the male smaller than the female, or near her size, he might be expected to have less prominent display, less tendency for com- bat and territoriality, less capacity to domi- nate the female, and a mating system that involes more enduring pair associations instead of harems or leks. Actually, too little is known of behavior in free-living Sceloporus to judge the extent of these correlations. The correlations that have been found involve traits of the female, rather than those of the male. In mainland Anolis, those species living in relatively dry climates and having rela- tively short and concentrated breeding sea- sons all had males relatively large com- pared with females (Fitch, 1976). In con- trast, those species living in relatively aseasonal climates of cloud forests or rain forests either have the sexes approximately equal in size, or have the females larger. This trend was explained by the more in- tense competition between males for terri- tories and mates in the stress of a concen- trated breeding season, resulting in selec- tion for large size and aggressiveness. Size relationships of the sexes in Scel- oporus are not determined by the same set of factors that control them in Anolis. In fact, the climatic factors that generate relatively large males in Anolis were found to produce relatively large females in Sceloporus. The key to this difference seems to lie in the consistent single egg of Anolis versus the variable and often large clutch of Sceloporus. In a severely seasonal climate that limits breeding to a brief an- nual interval of optimum conditions, with a single clutch, there is a premium on making the clutch as large as possible. Number of eggs can be increased by mak- ing the individual eggs smaller. This has been accomplished in some instances, but, in general, the smaller the hatchling the poorer would be its chances of survival. The alternative strategy is to increase the female’s capacity to produce and contain eggs by increasing her size. All the corre- lations with sexual size difference that were found seemed to hinge on the female’s egg capacity. There is strong support for the idea that relatively large size of the female is cor- related with single broodedness, and with clutch size, in the intraspecific trends of several wide-ranging species. At the lati- tude of Oregon, S. occidentalis occidentalis and S. 0. bisertatus which produce only one large clutch per season and require two years for the young to mature, have rela- tively large females (male-to-female ratio 94.0% and 93.0% respectively), but in southern California, where smaller clutches and more than one clutch are laid annually (Goldberg, 1974) with some young matur- ing in the first year, disertatus females are relatively smaller (male-to-female ratio 103.669). Jameson and Allison (1976) found that, in the central Sierra Nevada of California, female occidentalis average 25 mm longer at 2200 m than at 1500 m, with mean clutch size of 11.3 eggs at the lower altitude and 13.4 eggs at the higher. (They stated that the size difference between the sexes Was greater at the higher altitude, but did not include figures showing male sizes.) Similarly, the small subspecies of S. undulatus that occur in the southern and eastern states generally produce three or more clutches annually and have males 90°, or more of female length, whereas the large western subspecies, erythrocheilus and elongatus, produce only one or two clutches and have relatively smaller males (89.85 and 87.70% of female length respectively). Variation in sexual size difference is to be SEXUAL S1zE DIFFERENCES IN THE GENUS SCELOPORUS 459 expected in each subspecies that occurs over a wide climatic range, altitudinally or lati- tudinally. In northern, single-clutched populations of S. graciosus females are larger than males (male-female ratio 95.79%, 96.66°% and 98.0%), but in south- ern California the males are larger than females (104.70°,,) and presumably females of this population produce at least two clutches annually as Tinkle (1973) has demonstrated in those of southern Utah. CONCLUSIONS In 53 populations of Sceloporus studied, the size relationships of the sexes varied, in a virtual continuum, from relatively large males at one extreme (123.99 of fe- male length), to relatively small males at the other extreme (87.70% of female length). Males averaged larger in 33 popu- lations and females averaged larger in 20; for the whole group male length averaged 102.89, of female length. Sexual size differ- ence was found to be significant at the 999%, level in 42 of the 53. The size relationships of the sexes are controlled by a complex of factors, includ- ing those of reproductive strategy, and also including social structure. Males are more aggressive than females. In most species they maintain territories by displays of colorful ventral patches and actual fight- ing. However, field studies of behavior under natural conditions are few and it is not yet possible to make interspecific com- parisons in such traits as size and perma- nence of territory, permanence of associa- tion between the sexes, extent of male dominance and prevalence of polygyny. These traits are probably correlated with the relative size of the male. The more significant correlations re- vealed the high female-to-male size ratio in large-brooded and single-brooded spe- cles, as contrasted with those producing small and/or multiple broods. Presumably, in the large-brooded and _single-brooded species, the capacity of the female as an egg container is at a premium and there is selective pressure for her size to increase. The next strongest correlations, both above chew 9977 large females in temperate (as contrasted level, were those of relatively with tropical) climates and relatively large females in small or medium-sized species (as contrasted with large species, of more than 100 mm S-V). In tropical climates, breeding seasons are lengthened and repro- ductive effort is less concentrated than in temperate climates; the trend parallels that of multiple brooded vs. single-brooded pop- ulations. Perhaps in small species, the fe- male’s capacity as an egg container is at an even greater premium. Weaker correlations (significant at the 95°, level but not at 99°%,) were found be- tween sexual size difference and the two major phylogenetic groups, and also be- tween sexual size difference and oviparity vs. viviparity. Possible weak correlations (not significant at the 95°% level) were found between sexual size difference and habitat (rock-living vs. tree- or ground- living) and between sexual size difference with large females and development of female display colors. No indication of correlation with sexual size difference was found for development of male display colors, or for time required to mature (first year vs. second to fourth year). Intraspecific variation in sexual size dif- ference was found in several species, not- ably the wide-ranging polytypic undulatus, occidentalis and graciosus. In each of these, the geographic trend paralleled that for the genus as a whole, with northern popula- tions and those at high altitudes having relatively large females. 460 Tue University OF Kansas SCIENCE BULLETIN EILERA TORE Cll ED Amapon, D. 1959. The significance of sexual differ- ence in size among birds. Proc. Amer. Phil. Soc., 103:531-536. BaLuincer, R. E. 1973. Comparative demography of two viviparous iguanid lizards (Sceloporus jarrovi and Sceloporus poinsetti). Ecology 54(2) :269-283. Brain, W. F. 1960. The rusty lizard, a population study. Univ. of Texas Press, Austin, xvi -+ 185 pp. BURKHOLDER, G. L. AND W. W. Tanner. 1974. Life history and ecology of the Great Basin sage- brush swift, Sceloporus graciosus graciosus. Baird and Girard, 1852. Brigham Young Univ. Scis Bull Biol-ser. 195 no. Se-1'4. CHANEY, A. H. ano R. E. Gorpon. 1954. Notes on a population of Sceloporus merriami. Texas Jour. Sci. 6(1):78-82. Core, C. J. 1963. Variation, distribution and taxo- nomic status of the lizard Sceloporus undulatus virgatus Smith. Copeia 1963 (2) :413-425. CreNsHAW, J. W., Jr. 1955. The life history of the southern spiny lizard, Sceloporus undulatus undulatus Latreille. Amer. Midland Nat., 54 (2) :257-298. Crisp, T. M. 1964. Studies of reproduction in the female ovoviviparous lizard Sceloporus cy- anogenys (Cope). Texas Jour. Sci. 16:481. Davis, W. B. anv J. R. Dixon. 1961. Reptiles (ex- clusive of snakes) of the Chilpancingo region, Mexico. Proc. Biol. Soc. Washington, 74: 37-56. EARHART, C. M. ano N. K. Jounson. 1970. Size dimorphism and food habits of North Ameri- can owls. Condor, 72:251-264. ETHERIDGE, R. E. 1964. The skeletal morphology and systematic relationships of sceloporine liz- ards. Copeia 1964(4):610-631. Fircu, H. S. 1940. A field study of growth and behavior in the fence lizard. Univ. Calif. Publ. Zoos 441 512172. —. 1970. Reproductive cycles of lizards and snakes. Univ. Kansas Mus. Nat. Hist. Misc. Publ. No. 52, pp. 1-247. 1973. A field study of Costa Rican lizards. Univ. Kansas Sci. Bull. 50(2):39-126. ——. 1976. Sexual size differences in the main- land anoles. Occas. Papers Mus. Nat. Hist. Univ. Kansas, 50:1-21. GotpperG, S. R. 1971. Reproductive cycle of the ovoviviparous iguanid lizard, Sceloporus jar- rovt Cope. Herpetologica 27(2):123-131. ——. 1973. Ovarian cycle of the western fence lizard Sceloporus occidentalis. Herpetologica 29 :284-289. ——. 1974. Reproduction in mountain and low- land populations of the lizard Sceloporus oc- cidentalis. Copcia 1974:176-182. ———. 1975. Yearly variations in the cycles of the lizard Sceloporus occidentalis. Jour. Herp. 9(2) 187-189, Hirt, N. P. 1944. Sexual dimorphism in the Fal- coniformes. Auk 61:228-234. Hounsaker, D., Il. 1959. Birth and litter sizes of the blue spiny lizard Sceloporus cyanogenys. Copeia 1959:260-261. Jackson, J. F. ano S. R. TeLrForp, Jr. 1974. Repro- ductive ecology of the Florida scrub lizard, Sceloporus woodi. Copeia 1974(3) :689-694. Jameson, E. W. ano A. ALtison. 1976. Fat and breeding cycles in two montane populations of Sceloporus occidentalis (Reptilia, Lacertilia, Iguanidae). Jour. Herp. 10(3):211-220. Kruux, H. 1972. The spotted hyena: a study of predation and social behavior. Chicago, Uni- versity of Chicago Press, xvi + 335 pp. Larsen, K. R. anp W. W. Tanner. 1974. Numeric analysis of the lizard genus Sceloporus with special reference to the cranial osteology. Great Basin Nat. 34(1):1-41. —. 1975. Evolution of the sceloporine lizards (Iguanidae). Great Basin Nat. 35(1):1-20. Marion, K. R. anv O. J. Sexton. 1971. The repro- ductive cycle of the lizard Sceloporus malachi- ticus in Costa Rica. Copeia 1971:517-526. Mas.in, T. P. 1963. Notes on a collection of herpe- tozoa from the Yucatan Peninsula in Mexico. Univ. Colorado Studies Ser. Biol. No. 9:1-20. MayHew, W. W. 1963. Reproduction in the granite spiny lizard, Sceloporus orcutti. Copeia 1963 Cea lp2e Mitts, G. S. 1976. American kestrel sex ratios and habitat separation. Auk 93(4):740-748. MosHer, J. A. AND P. F. Marraz. 1974. Size di- morphism: a factor in energy savings for Broad-winged Hawks. Auk 91:325-341. MuELLER, F. C. ano R. E. Moore. 1969. Growth of the sagebrush lizard Sceloporus graciosus in Yellowstone National Park. Herpetologica 25: 35-38. Muraik, D. D. 1946. A comparative study of the urogenital systems of an oviparous and two ovoviviparous species of the lizard genus Sceloporus. Bull. Univ. Utah, Biol. Ser. 9(7): 1-24. Newtin, M. E. 1976. Reproduction in the bunch grass lizard, Sceloporus scalaris. Herpetologica 32(2) 3171-184. Parker, W. S. anv E. R. Pranxa. 1973. Notes on the ecology of the iguanid lizard, Sceloporus magister. Herpetologica 29:143-152. PurbueE, J. R. anp C. C. Carpenter. 1972. A com- parative study of the body movements of dis- playing males of the lizard genus Sceloporus (Iguanidae). Behavior, 41:68-81. Reynotps, R. T. 1972. Sexual dimorphism in ac- cipiter hawks: a new hypothesis. Condor 74:191-197, SELANDER, R, K. 1966. Sexual dimorphism and dif- ferential niche utilization in birds. Condor 68:113-151. SEXUAL S1zE DIFFERENCES IN THE GENUS SCELOPORUS 46] SmitrH, H. M. 1939. The Mexican and Central American lizards of the genus Sceloporus. Field Mus. Nat. Hist. Zool. Ser. 26:1-397. SmitTH, H. M. anno A. H. Savitsxy. 1974. Another cryptic associate of the lizard Sceloporus for- mosus in Guerrero, Mexico. Jour. Herp. 8(4): 297-303. Soxoi, R. R. anv F. J. RouHtF. 1969. Biometry. The principles and practice of statistics in biologi- cal research. W. H. Freeman and Co., San Francisco. 776 pp. Sruart, L. C. 1971. Comments on the malachite Sceloporus (Reptilia: Sauria: Iguanidae) of southern Mexico and Guatemala. Herpeto- logica 27(3) :235-259, TANNER, W. W. anv J. M. Hopxin. 1972. Ecology of Sceloporus occidentalis longipes Baird on Ranier Mesa, Nevada Test Site, Nye County, Nevada. Brig. Young Univ. Sci. Bull. Biol. Ser. 15(4) :1-31. TANNER, W. W. anv J. E. Krocu. 1973. Ecology of Sceloporus magister at the Nevada Test Site, Nye County, Nevada. The Great Basin Nat- uralist 33:133-140. THomas, R. A. anp J. R. Dixon. 1976. A re-evalua- tion of the Sceloporus scalaris group (Sauria Iguanidae). The Southwestern Naturalist 20 (4) 523-536. TInKLE, D. W. 1972. The dynamics of a Utah popu- lation of Sceloporus undulatus. Herpetologica 28(4) :351-359. 1973. A population analysis of the sage- brush lizard, Sceloporus graciosus in southern Utah. Copeia 1973:284-295. 1976. Comparative data on the population ecology of the desert spiny lizard, Sceloporus magister. Herpetologica 32:1-6. TINKLE, D. W. anp R. E. Bariincer. 1972. Scel- oporus undulatus: a study in the intraspecific comparative demography of a lizard. Ecology 53:570-584. TINKLE, D. W. anno N. F. Haprey. 1975. Lizard reproductive effort: calorimetric estimates and comments on its evolution. Ecology 56:427- a4. VERNER, J. AND M. F. Wittson. 1969. Mating sys- tems, sexual dimorphism, and the role of North American passerine birds in the nesting cycle. 76 pp. Ornith. Monogr. No. 9, Amer. Ornith. Union. Vinecar, M. B. 1975. Demography of the striped plateau lizard, Sceloporus virgatus. Ecology. 56:172-182. Wess, R. G. 1967. Variation and distribution of the iguanid lizard, Sceloporus bulleri, and the description of a related new species. Copeia 1967:202-213. nH - fa i OCT 23 1978 HARVARD 0.00 0000080666 6685 00888000 COE OOO OOOO OEE EE EOE E888 O OOS Eee OH e688 8 88 eww ww wen mer en eng eo 9% a aa a a i a SCIENCE BULLETIN gs - te ies satel eS eS IS = Bs Sees SanasSeeseabeeeecascheeesaceeeeasecessaccacesssececesesenecesesetosasetesecesecenesstectasstetanetetetanesetenasstetoseseivantetstetatetetenseetetetatstote A COMPARATIVE ANATOMICAL STUDY OF MANDIBULAR STRUCTURE IN BEES iS : BS : weet ms net eM " : OO * Be * : on : = 2° Es ee ee I) EPR EP EPI re By CHARLES D. MICHENER and ANNE FRASER roeeeeeeneeatatetareratanehacheatetavers"areta'e CODE EP IOOI OOS S CREOLE EEE EDIELLEE A AEEEEEE EEE Cn Pees Oa KR SH RS ecstatic tctapctctctattannctasceatstatataetstatetanatone toca eee con eee toee tee eeenes Vol. 51, No. 14, pp. 463-482 September 26, 1978 ANNOUNCEMENT The University of Kansas Science Bulletin (continuation of the Kansas Uni- versity Quarterly) is an outlet for scholarly scientific investigations carried out at the University of Kansas or by University faculty and students. Since its incep- tion, volumes of the Bulletin have been variously issued as single bound volumes, as two or three multi-paper parts or as series of individual papers. Issuance is at irregular intervals, with each volume prior to volume 50 approximately 1000 pages in length. Effective with volume 50, page size has been enlarged, reducing the length of each volume to about 750 pages. The supply of all volumes of the Kansas University Quarterly is now ex- hausted. However, most volumes of the University of Kansas Science Bulletin are still available and are offered, in exchange for similar publications, to learned societies, colleges and universities and other institutions, or may be purchased at $20.00 per volume. Where some of these volumes were issued in parts, individual parts are priced at the rate of 2 cents per page. Current policy, initiated with volume 46, is to issue individual papers as published. Such separata may be purchased individually at the rate of 3 cents per page, with a minimum charge of $1.00 per separate. Subscriptions for forthcoming volumes may be entered at the rate of $20.00 per volume. All communications regarding exchanges, sales and subscriptions should be addressed to the ExcHancEe Liprarian, UNIVERSITY oF Kansas Lipraries, LAWRENCE, Kansas 66045. Reprints of individual papers for personal use by investigators are available gratis for most recent and many older issues of the Bulletin. Such requests should be directed to the author. The International Standard Serial Number of this publication is US ISSN 0022-8850. Editor Eugene C. Bovee Editorial Board William L. Bloom Philip W. Hedrick Rudolf Jander Harvey Lillywhite Charles D. Michener Norman A. Slade Henry D. Stone George W. Byers, Chairman —— THE UNIVERSITY OF KANSAS SCIENCE BULLETIN Vol. 51, No. 14, pp. 463-482 September 26, 1978 A Comparative Anatomical Study of Mandibular Structure in Bees’ (Hymenoptera: Apoidea) Cuares D. MICHENER AND ANNE FRASER TABEE OF CONTENTS ACTIN OTE SS pt ko BUG al ily Beds eel, ARIE SRNR Y ELS TAL ee ee URNS Eee OED Bane 463 |S FUPEX CON BETEIVOR NT, Me skh ee a Se ee eA al a 463 I UAGIER RUAUES epee erent 7 AOD PAE PCOS AAS Sn nee apne tah Lee Ble ot eee 464 BASTCROTRUCTURE AND TILER NINOlUOGYI a cnc ee aces ese eee ness idnate tt 466 Jes Enya Wilon til Hive al OPC’ ae ee Oe ane RE res ES LS lee eon ea ee en le 466 Mancinulanestiriacese ect fa) ot ace Ne LS Ie We ee) 466 @utern surlace: 2 Re foie RRO EET ee Met RED IL gn ONE ORY ek ee 467 Distal wmaaroiinges seer ee A SAIS ed TEN tan BE, A Bre IE De OD Pe ee a 473 SUMMaAnvecthiG Ma DDLEVIatIOMS tc) ser ee ee eek ee eee Be 473 CONGR AR GE GeO LUD eee eet eS ee ee ye ie eRe Ee eee a a 474 Rarasiticn AMENG PNOMGS A oan tier ee Le dene ye ee ci aeee ae ee 476 Fplvelaeitac amelie Mer Omics Stace 20 ee eee 477 Neylocopinae ane wsithuroinae..2 2) vo. Owen eve Se eS ee ee 477 IVE creche ora eee Sees ct SN SEN eee te 2 ee ed 478 Meliponinacmand@Apinae: 2. oi. 2e Se) io ete eee ee ee ee 479 OUD AC amen tome ee IAAL SRL Peed P08 DURE ah we Sieh eins eae e Nee ieee 479 ID ESCUSSIO NAN DG ONCISUSIONS eee ccc eke tree Aer eee le kt ee ol rae eee date a oes EN pried 480 ACKNOWLEDGEMENTS. .......----2-----20----+- TAU es Very ee el Oe. ae ee LIne ANT ey eer nee ay Bone ee 482 [oS TETRIS irre al av a ee ay sa Mo ra le SR Sa A Ret pe PR mh 482 ABSTRACT Mandibular structures, particularly the ridges and grooves of the outer and inner surfaces, were investigated and illustrated for all major groups of Apoidea. A nomenclature is provided for these structures, and homologies among apoid groups are indicated. A basic mandibular type is found among sphecoid wasps, all short tongued families of bees, and also the Antho- phoridae and Fideliidae. Various modifications are found within some of these groups, such as the parasitic anthophorids, the Hylaeinae and Xeromelissinae, and the Xylocopinae. Mark- edly modified mandibles characterize the Megachilidae and the Apidae. INTRODUCTION The taxonomic literature on bees con- emphasizing the distal parts of mandibles tains many descriptions and illustrations —the apical margins and teeth. These are features whose form is easily seen if the C ibuti f e ; : ontribution number 1663 from the Department mandibles are open ainclwitosemcanchions of Entomology, The University of Kansas, Lawrence, k : , Kansas 66045, U.S.A. are sometimes obvious, e.g., cutting edges 464 Tue University oF Kansas SCIENCE BULLETIN on mandibles of females of leaf-cutter bees (Megachile) or multidentate apices of mandibles of those anthidiine bees (e.g., Anthidium) that makes their nests with plant hairs. The present paper is not pri- marily concerned with these features, but deals instead with the surfaces, ridges, and grooves of the mandible, i.e., the body of the mandible. In general, the mandibular body is more conservative than the apical teeth, showing less diversity among species and related genera. A few authors have used the ridges and grooves taxonomically, but in the absence of an overview, no ho- mologies among diverse groups have been apparent and no standard terminology exists. The object of the study here reported was to establish homologies and terminol- ogies for the parts of the mandible, to fa- cilitate their study, and to look for char- acters that might illuminate the relation- ships among major groups of bees. MATERIALS The mandibles of many bees were exam- ined in the course of this study. Generally, mandibular structures are better developed in females than in males; detailed analysis was therefore based on females except as noted. From among the many mandibles studied, those of the species listed in Table 1 were selected as representative of most of the varia- tion in the pattern of ridges and grooves and many of them were diagrammed (outer and inner views) as indicated in the table by references to figures. Species with bizarre swellings or projec- tions are excluded from those illustrated, for they do not contribute to the understanding of the basic structures. However, once the latter are identified, it is often possible to see that a given projection is an elaboration of a particular ridge or interspace. Letters corresponding to the abbreviations in the figures are given after the names of the structures, where the word or expression is defined in the following section and also in a tabular summary at the end of that TABLE SPECIES STUDIED. Famity CoLLeETIDAE Subfamily Colletinae Colletes inaequalis (Fig. 2) Leioproctus herbsti Callomelitta antipodes (Fig. 3) Subfamily Diphaglossinae Diphaglossa gayi (Fig. 4) Ptiloglossa guinnae Subfamily Hylaeinae Hylaeus rugosulus Amphylaeus morosus (Fig. 5) Subfamily Xeromelissinae Chilicola ashmeadi (Fig. 6) Subfamily Stenotritinae Crenocolletes smaragdinus Famity HaticripAE Subfamily Halictinae Halictus ligatus, quadricinctus (Fig. 7) Lasioglossum imitatum, lustrans, texanum, zephyrum Pseudaugochloropsis graminea Augochlora pura Megalopta genalis Subfamily Nomiinae Nomua melanderi Subfamily Dufoureinae Dufourea marginata Systropha curvicornis Famity ANDRENIDAE Subfamily Andreninae Andrena accepta (Fig. 8), illinoiensis Subfamily Panurginae Panurginus occidentalis (Fig. 9) Panurgus calceatus Psaenythia bergi Perdita chihuahua FAMILY OXAEIDAE Protoxaea gloriosa (Fig. 10) Famirty ME viItTTIDAE Subfamily Melittinae Melitta leporina (Fig. 11) Subfamily Dasypodinae Dasypoda panzeri Subfamily Macropidinae Macropis labiata Subfamily Ctenoplectrinae Crenoplectra fuscipes ANATOMICAL STUDY OF MANDIBULAR STRUCTURE IN BEES 465 TABLE 1—(Concluded) Famity ANTHOPHORIDAE Subfamily Anthophorinae Exomalopsis zexmeniae Tapinotaspts coerulea Ancyloscelis apiformis Diadasia enavata Melitoma segmentaria Anthophora occidentalis (Fig. 12) Clisodon terminalis (Fig. 13) Svastra atripes (Fig. 14) Meltssodes agilis Centris poecila (Fig. 15) Ericrocis lata (Fig. 16) Mesoplia garleppi Thyreus ramosa (Fig. 17) Xeromelecta californica Subfamily Nomadinae Nomada annulata (Fig. 18) Triepeolus concavus Biastes brevicornis (Fig. 19) Subfamily Xylocopinae Macrogalea candida (Fig. 20) Allodape mucronata (Fig. 21) Ceratina dupla (Fig. 22) Manuelia gayi (Fig. 23) Xylocopa virginica (Fig. 24) FAMILY FIDELIIDAE Fidelia villosa (Fig. 25) Neofidelia profuga (Fig. 26) Famity MEGACHILIDAE Subfamily Lithurginae Lithurge gibbosus (Fig. 29) Subfamily Megachilinae Paranthidium jugatorium Dianthidium ulkei Callanthidium illustre Anthidiellum notatum (Fig. 27) Anthidium manicatum (Fig. 28) Parevaspis carbonaria Euaspis abdominalis Chelostoma fuliginosum Hertades carinatus Osmia lignaria, subaustralis, subfasciata Hoplitis albifrons (Fig. 30; male, Fig. 31) Creightonella frontalis (Fig. 32) Megachile albitarsis (Fig. 33), frugalis Chalicodoma exilis (Fig. 34) Famity APIDAE Subfamily Meliponinae Meliponula bocandet (Fig. 35) Trigona beccariu, braunst, capitata (Fig. 36), chanchamayoensis, cupira, erythrea, mexicana Dactylurina staudingert Lestrimelitta limao Melipona fasciata Subfamily Apinae Apis mellifera (queen, Fig. 37; worker, Fig. 38) Subfamily Bombinae Euplusia violacea (Fig. 39) Euglossa cordata Eulaema dimidiata Exaerete smaragdina (Fig. 40) Bombus americanorum (Fig. 41) Psithyrus variabilis (Fig. 42) section. Figure | illustrates certain structures not labeled in the remaining figures, but most ridges and grooves are labeled on all the fig- ures where they appear. To avoid crowding the illustrations, structures along the upper and lower mandibular margins are minimally labeled. For example, in most of the illustra- Fic. 1. Outer view of mandible of Diphaglossa gayt, female, with labels for structures not identified in other illustrations. For abbreviations see Tables 2 and 3. tions the distal part of the adductor ridge, where it appears on the outer mandibular sur- face below the condylar groove, is not labeled. Structures on the upper mandibular margin such as the upper carina are ordinarily la- beled on only one rather than both of the drawings of each mandible. Dots on the drawings indicate the areas with hairs. The meanings of the abbrevia- tions are indicated in Tables 2 and 3. 466 Tue University oF Kansas ScrENCE BULLETIN Most of the terminology is new for this paper but (where practical) terms are those used by Michener (1944) and Eick- wort (1969). BASIC STRUCTURE AND TERMINOLOGY For purposes of terminology, the man- dible is considered to lie in a horizontal position (as when it is closed), so that it has upper and lower margins separating the inner and outer surfaces. Eickwort (1969) considered the mandible to project downward and to have anterior and pos- terior margins, but our assumptions seem more practical to us. Mandibular Base: The base of the mandi- ble is attached to the head capsule at the large and irregularly round mandibular socket between the lower end of the eye and the proboscidial fossa. In this socket the mandible can rock only in one plane, because of the double articulation consist- ing of the small anterior mandibular ace- tabulum (Ac), fitting over a condyle near the lower lateral angle of the clypeus, and a large posterior mandibular condyle (C), fitting into an acetabulum in the cranium below the eye. The line between these two articulations is the articular axis, about which the mandible rocks. The outer margin of the mandibular socket is only moderately deviant from the articular axis. A small bulge in the outer margin of the mandibular base, a short distance in front of the mandibular con- dyle, is the abductor swelling (AbS) to which the apodeme of the abductor muscle is attached. Opening of the mandibles requires little power and the abductor swelling, as might be expected, is not far lateral to the articular axis; it provides no great mechanical advantage. The main work of mandibles, i.e., bit- ing, excavating soil or wood, and the like, requires strength in closing the mandibles. The inner margin of the mandibular sock- et is accordingly far mesal to the articular axis and the inner margin of the mandib- ular base is swollen to form the adductor convexity, sometimes called the adductor angle. To this convexity, far mesad from the articular axis and providing consider- able mechanical advantage, is attached the apodeme of the adductor muscle. The whole inner basal mandibular surface is involved in the adductor convexity, while only a small local area forms the abductor swelling. The summit of the convexity is usually flattened and lies between the basal part of the adductor ridge and the trimmal carina (see below). Mandibular Surfaces: Vf an imaginary “plane” through the articular axis were extended toward the mandibular apex, curving and sometimes twisting with the curvature of the mandible so that the max- imum area of the plane is internal, Le., inside the mandible, then the lines of inter- section of the plane with the mandibular surface will separate the owter surface of the mandible from the snner surface. The outer surface is ordinarily gently convex, never far from this imaginary plane. The inner surface, commonly concave distally, is strongly convex basally, forming the ad- ductor convexity. The lines along which the imaginary plane intersects the mandib- ular surface are the ower margin and the upper margin of the mandible. The de- tails of the intersection of the plane with the mandibular surface vary, so that cer- tain structures cannot be unhesitatingly at- tributed to one or the other surface. The adductor convexity is often so large that, for the basal half of the mandi- ble, the upper (or anterior) slope of the convexity could be considered as an upper (or anterior) mandibular surface, com- monly beveled to pass the clypeal and la- bral margins when the mandibles are being ANATOMICAL STuDY OF MANDIBULAR STRUCTURE IN BEES 467 Fics. 2-11. Outer and inner views of mandibles of female Colletidae, Halictidae, Andrenidae, Oxaeidae, and Melittidae. For species names see Table 1; for abbreviations see Table 2 and 3. closed. Likewise, for the basal part of the mandible, the lower or posterior slope of the adductor convexity could be regarded as a lower (or posterior) mandibular sur- face. Because upper and lower surfaces are not evident distally, it seems simplest not to use these terms, but to consider both as parts of the inner surface. Outer Surface: Seen from the outer side, many bee mandibles have a preapical tooth (subapical tooth of Eickwort, 1969; Mich- ener, 1944, etc.) on the upper margin; this tooth and the area basal to it are the pollex (thumb) while the remainder of the man- dible is termed the rutellum (mitten). (For further details see below.) The outer 468 Tue University oF KAnsas SCIENCE BULLETIN surface of the mandible usually has several longitudinal, elevated areas or ridges that are normally smooth and hairless. They fuse toward the apex of the mandible to form a smooth surface, the cap of the rutellum (CRu). Between the ridges are grooves, almost always bearing hairs. Com- monly, one groove (outer) is broadened basally, forming a more or less flat, non- elevated, hairy area that cannot reasonably be called a groove and is therefore called an interspace. Another groove (acetabu- lar) in some bees (especially Megachili- dae) is broadened apically to form an interspace that often contains longitudinal ridges which are probably secondary, not homologous to any of the more universal or primary, named ridges. The lower margin of the mandible basally and sometimes apically is formed by the condylar ridge (CR) which arises near the mandibular condyle and extends toward the apex of the mandible, ultimate- ly joining the cap of the rutellum. Usually, however, in the distal half of the mandi- ble, the adductor ridge from the inner sur- face of the mandible comes into view be- low the condylar ridge, so that the lower margin of the mandible is formed distally by the adductor ridge, which also merges with the cap of the rutellum apically. Oc- casionally, as in Halictus quadricinctus and Bombus, the condylar ridge consists of two (even three) ridges with a hairy groove between them; one is clearly on the outer mandibular surface (CR, 0) while the other is often on the lower margin or almost onto the inner surface (CR, 1). Above the condylar ridge and typically parallel to it is the piliferous outer groove (OG) which usually expands basally to form the broad, hairy outer interspace (OI) (but see below). Rarely, as in Eu- plusia, the distal part of the outer groove is divided, forming an upper and lower outer groove (OG, u; OG, 1). Alterna- tively, elevation of all but marginal areas of the outer interspace may divide the median and basal parts to form upper and lower outer grooves, as in Bombus and Psithyrus. Above the outer groove is the common- ly broad and gently convex outer ridge (OR) (outer diagonal ridge of Eickwort, 1969) which distally is separated by the outer groove from the condylar ridge, and which merges distally into the cap of the rutellum. The outer ridge commonly occu- pies the median part of the outer surface of the mandible. Basally, the outer ridge usu- ally curves upward (hence Eickwort’s term, diagonal ridge) so that when com- plete, its base is near the acetabulum. When there is more than one base, this one is termed the upper root (OR, u). Sometimes, as in Diphaglossa, the upper root of the outer ridge is broad, so that its base occupies the space between the acetab- ulum and the abductor swelling; alterna- tively, the outer ridge may fade away ba- sally into the outer interspace, the base of the upper root being represented, if at all, by an elevation just distad from the ace- tabulum and not connected to the distal part of the ridge. Other roots of the outer ridge are relatively rarely developed. Some- times the ridge has a lower basal ramus or root (OR, 1) extending toward the con- dyle (e.g., Macrogalea, Anthophora) or a median root (OR, m) extending toward the abductor swelling (e.g., Chalicodoma). In these cases, which are very similar to one another and probably represent the same development, the outer interspace lies be- tween the two roots of the outer ridge. The outer ridge is so named because of its position; its basal connections are variable and not appropriate for purposes of naming the ridge. When the outer ridge has lower or median roots, the outer groove is narrow to the base of the mandible and cannot ANATOMICAL STUDY OF MANDIBULAR STRUCTURE IN BEES 469 AcG TE AcC —— AcC TE UC a levels ==. -AGG Fics. 12-22. Outer and inner views of mandibles of female Anthophorinae, Nomadinae, and Xylocopinae in the Anthophoridae. For species names see Table 1; for abbreviations see Tables 2 and 3. expand into the outer interspace. An al- ternative interpretation would be that such roots are really formed by duplication of the condylar ridge. Examination of man- dibles like those of Anthophora or Chiso- don, however, show that the apex of the outer groove is in the same position as in most bees, and the extra ridge arises from 470 Tue UNIvVerSITY OF KANSAS SCIENCE BULLETIN the preapical part of the outer ridge. Man- dibles like those of Allodape and Xylocopa show intermediate stages in development of lower roots. Rarely, the distal part of the outer ridge may be deeply bifid, forming upper and lower branches (OR, d; OR, v) as in Bombus. Above the outer ridge is the piliferous acetabular groove (AcG) (outer anterior groove of Eickwort, 1969), which arises distal to the acetabulum and extends to- ward the distal margin of the mandible. It usually fades away near the base so that the outer ridge (or its upper root) fuses with the acetabular carina basally. It ordi- narily terminates before the distal margin of the mandible, and more or less sepa- rates the pollex, which usually forms the upper preapical mandibular tooth or mar- gin, from the rest of the mandible (as seen from the outer surface), the ratellum. Thus, it does not end in the cap of the rutellum as do the grooves below it on the outer surface. |The rutellum and _ pollex are not separable on the inner, mandibular surface except by the notch on the distal margin in most species. Eickwort (1969) believed that the fimbriate line separates the pollex from the rutellum (our termi- nology) on the inner surface of the mandi- ble, but this is ordinarily not the position of the fimbriate line.| Above the acetab- ular groove is the acetabular carina (AcC), a usually sharp ridge which arises near the acetabulum and extends distally toward the apex of the pollex. On the basal part of the mandible, the acetabular carina and the outer ridge are sometimes completely fused. This appears to be the case, for example, in male Hoplitis and in Mega- chile and Chalicodoma. In some bees (Megachilidae, Apidae) the distal part of the mandible, above the outer ridge, is expanded apically as a broad space between the distal parts of the outer ridge and the acetabular carina. This space constitutes, or is occupied by, the acetabu- lar groove in slender mandibles, but that term is inappropriate for a broad space, especially when there are several secondary ridges or possibly branches of the outer ridge in the area. This area is therefore called the acetabular interspace (Acl1). The acetabular carina may be the upper margin of the mandible. If not, the space and carina commonly narrowly visible above the acetabular carina have their principal connections with the inner man- dibular surface and are described in the next section, In some megachilid bees a striking fea- ture is the outer premarginal fimbria, which arises from the outer premarginal impressed line (OIL). The fimbria or line roughly parallels the distal margin of the mandible across the acetabular interspace, but is absent in the region of the apex proper, i.e., on the cap of the rutellum. In the Meliponinae, there are no ridges on the outer mandibular surface or only the acetabular carina, but an oblique groove (OG) extends from near the acetabulum toward the mandibular apex. It seems likely that, near its base, the oblique groove is the acetabular groove, but distally it must be a secondary groove, for it goes toward the mandibular apex and has a branch going to the lower margin of the mandible. No such pattern occurs in the mandible of other bees. Inner Surface: Arising on the adductor convexity is the adductor ridge (AdR). This ridge, the basal part of which is often not hairless and shining like ridges of the outer surface, follows a _ characteristic course, curving downward in the median part of the mandible toward the lower margin; thereafter it extends distally, com- monly forming or being very near to the lower mandibular margin. This distal part of the adductor ridge was called the outer posterior ridge by Eickwort (1969). The ANATOMICAL STUDY OF MANDIBULAR STRUCTURE IN BEES 471 T Acc TE UDA —— is —— [oes ae SE a Fics. 23-31. Outer and inner views of mandibles of female (except for Fig. 31) Xylocopinae in the Anthophoridae, Fideliidae, and Megachilidae. For species names see Table 1; for abbreviations see Tables 2 and 3. interval above the adductor ridge distally on the outer surface of the mandible, sep- arating it from the condylar ridge, is the condylar groove (CG) (outer posterior groove of Eickwort, 1969). Basally, this interval is on the inner surface of the mandible and, because of the curvature of the adductor ridge, the interval becomes the broad lower surface or slope of the adductor convexity, 1.e., the condylar inter- space (CI). Above the adductor ridge on the inner mandibular surface is the rather broad adductor interspace (Ad1). This interspace is the median zone of the inner mandibu- lar surface and is delimited above by the fimbriate line and trimmal carina. The trimmal carina (TC), sometimes 472 Tue UNIversity oF KANsAs SCIENCE BULLETIN merely a ridge basally, arises on the upper surface of the adductor convexity and sep- arates the basal part of the adductor inter- space from the often flat trimma (T), a special basal mandibular area which is usu- ally smooth and often bears some short, presumably sensory hairs and which slopes toward the upper margin of the mandible and passes the clypeal and labral margins when the mandible is being closed. The trimma forms the upper surface or slope of the adductor convexity and extends to the acetabulum; toward the outer mandib- ular surface it is limited by the acetabular carina if the latter extends far enough ba- sally; if not, it is limited by the upper edge of the basal root of the outer ridge, which is probably fused with the base of the acetabular carina. The trimmal carina commonly extends distally to about the middle of the mandible. From there, it sometimes continues uninterruptedly as the wpper carina (UC) of the distal part of the mandible, which usually forms the upper margin of the pollex. The interval between the acetabular carina and trimmal and upper carinae, basally the broad trim- ma, tapers distally to form a slender trim- mal extension (TE) on the pollex. This is not a piliferous groove but a smooth area, as is the trimma. Sometimes merging with the distal end of the trimmal carina and curving distally, often obliquely across the inner surface of the mandible, is the fimbriate line (FL). This is the line on which the often elevated upper part meets the often depressed me- dian part of the inner mandibular surface. Sometimes the line is a ridge or a carina (the frmbrial carina), more often simply a step between two levels. The fimbriate line is usually marked by a series of hairs along its lower margin, hence its name; these hairs constitute the inner fimbria. The space between the fimbriate line and the upper carina is the fimbrial interspace (FI). It disappears basally where the up- per carina and fimbrial ridge approach one another or unite, but distally it may expand to form the whole upper distal part of the concave inner surface of the mandible. In those bees (especially the Megachilidae) in which the mandibles are broad distally due to the expansion of the condylar interspace on the outer surface, an effect on the inner surface is that the fimbriate line becomes very oblique, even parallel to the distal margin of the mandi- ble. The point where it fuses with or ap- proaches the upper or trimmal carinae moves distad, often quite close to the up- per distal angle of the mandible, thus essentially eliminating the fimbrial inter- space. The inner fimbria sometimes arises from a weak groove, called the inner an- terior groove by Eickwort (1969), but it is ordinarily depressed only relative to the space above it and not relative to the space below it. Hence we have not used the word groove for it. When the fimbriate line is carinate, the carina may overlap the bases of the fimbrial hairs, which therefore emerge from behind the carina. Relations among the trimmal carina, the upper carina, and the fimbriate line vary. Sometimes the three do not meet. Sometimes all three meet, forming a single Y-shaped system. Sometimes the trimmal carina continues uninterruptedly as the upper carina, with no connection to the fimbriate line. Sometimes the trimmal ca- rina continues uninterruptedly as the fim- briate line without connection with the upper carina. Under these circumstances, there seems to be no alternative to three separate terms for the three structures. It may be, however, that the trimmal carina and fimbriate line constitute a single basic structure, sometimes interrupted medially, to which the upper carina is sometimes joined. This suggestion is based on the fact that the trimmal carina, like the fim- ANATOMICAL STUDY OF MANDIBULAR STRUCTURE IN BEEs 473 i T UDA=P. gh ae x _UDA=R, tt “ AcC We 4. 3 OR A \ OR&ACC 5 ao) A lpaaaNG } Ol! Acl- } p Ol OIL \ ) AL Acl R— ey) ) —— R ) ave i ae a GRE : oS 06 Ne — CRS 43 GS OR,u 34 ay UC TC je 2 be a — OP va VSS (7 Fe | 7 FL \ Adl — ; Ad| y y ‘= Ss RAR: Cl a bey Roney Fics. 32-34. Outer and inner views of mandibles of female Megachilidae. For species names see Table 1; for abbreviations see Tables 2 and 3. briate line, typically has a row of hairs along its lower margin. In fact, when the carina is weak or absent, its location is sometimes indicated by the row of hairs; the same is sometimes true of the fimbriate line. Distal Margin: The distal or apical margin of the mandible varies from a single, nar- rowly rounded or pointed tip to a long, simple to multidentate edge. The apex (A), the distal extremity of the cap of the rutellum, is the part of the mandible far- thest from the base and except in most Lithurginae and Xylocopinae and in Clrso- don in the Anthophorinae, it forms the tooth that is a continuation of the lower margin of the mandible, often called the first tooth and here recognized as the first tooth of the rutellum. In some bees (e.g., Panurginus) there are no teeth, i.., the mandible is simple and toothless. In many bees there is a preapical tooth, the apex of the pollex (AP), on the upper margin. In such cases the mandible is often called bidentate, and the apex and preapical tooth constitute the distal margin. In some bees such as Centrts, the distal margin of the rutellum above the apex and below the pollex bears one to several teeth, the upper teeth of the rutellum (Rug, Rus, etc., Ru: being the apex). In some (most Lithurginae and Xylocopinae and Clisodon), there is a rutellar tooth below the apex and formed from the end of the adductor ridge. This is the lower tooth of the rutellum (Ru, 1). When it is pres- ent, the distal margin of the mandible is tridentate, with the median tooth (mandib- ular apex) longest and with the lower tooth of the rutellum below it and the similarly shaped apex of the pollex above it. In most female Megachilinae and to a lesser degree in some aphids, the distal margin of the mandible is elongate, be- cause the region of the pollex is expanded distally to form two to several teeth of the pollex (P1, Pz, etc.) or an edentate margin. The upper end of this margin or the up- permost tooth is the wpper distal angle (UDA) or simply the upper angle (inner angle of much taxonomic work; we have avoided this term because it is not on the inner surface of the mandible). Summary and Abbreviations: To sum- marize and organize the terminology, the longitudinal ridges, grooves, and_inter- spaces are listed in Table 2, starting with the condylar ridge and proceeding upward on the outer surface of the mandible and continuing around the upper margin and downward on the inner surface. Distal structures that are, or may be, continua- tions of basal structures are indented be- low the name of the basal structure. In- 474 Tue University oF Kansas ScrENCE BULLETIN TABLE 2 RipcEs, GROOVES AND INTERSPACES. CONC Yar tid ner et) ot ee nh CR (when compound or additional ridges, outer and inner) ........ Oni [OV TL <1 3 ate) OF (co) A ea OI DULCE SROOVC’ cee ones. ee OG (when compound, upper AM IG WOE ) es. eacca cae secs eeenee u, | OlLemnIG Cem wm een 250065 eh ee OR (upper median, and lower roots Ol OMLer TIPE) ..2 eee u, m, | (upper and lower distal branches) se eens Be ee d,v acefablan Groove; 4). sca ss AcG acetabular interspace -......:2.c..c.:<<: Acl acerab lary CARMA e xa:<.cssascecceent-vs.the AcC (]TS G00 2 Oe ne AO ee ae a trimmial extension: .2...0. 0.3: aE, GMT CANINA oc. tease cceceecasnucadeoctioese TC fimibriate line 228) 8 le, FL Uppeteeariians4 bia. 2G) set UC hinabialeimterspace tant. atl adductor anterspace: socio c teense AdI YalG OYE 20) a9 (6 (21h eee oe ee ee ee AdR condylar interspace 2.0. a. CI CONdylar GrOOVE: 2c. ceccecsecsoeceecns CG dented items in parentheses are subdivi- sions of major structure. Other special terms used are italicized in the preceding account. Those whose ab- breviations are used in the illustrations are listed in Table 3. COMPARATIVE STUDY The Ancestral Type of Mandible: In a large group of ground-nesting bees, the mandibles are rather slender, usually with all or most of the structures listed in the preceding section in readily recognized form. Such mandibles occur in the Col- letinae (Figs. 2, 3), Diphaglossinae (Fig. 1), Stenotritinae, Halictidae (Fig. 7), An- drenidae (Fig. 8), Oxaeidae (Fig. 10), Melittidae (Fig. 11), Anthophorinae (Fig. 12) and Fideliidae (Fig. 26). Mandibles of this type occur also among TABLE 3 OTHER MANDIBULAR STRUCTURES. mandibular acetabulum .......... Ac mandibular condyle .....2.-:..::--=- C abductor swelling?> 20-2. AbS apex (Ist tooth of rutellum) A upper teeth of rutellum:22..- Rup, Rus, etc. lower tooth of rutellum .......... Ru, | cap of-rutellum:) 2e ee CRu recthor pollex 22252 see ee P,, Po, etc. apex of pollex (commonly preapical tooth) .................. AP unnamed “secondary” ridges in certain genera or species R outer premarginal impressed line (Megachilidae) ............ OIL upper distal angle (Mega- chilidae, Apidae) 2.2. UDA oblique groove (Meliponinae) og sphecoid wasps. For this reason we con- sider them as ancestral among bees. Con- siderable variation occurs, however, among such mandibles. Some unusual and pre- sumably derived features of certain mandi- bles in this group of bees are described below. Colletid groups such as the Paracolletini (Leioproctus spp.), Diphaglossinae and Stenotritinae have rather ordinary mandi- bles. In Colletes (Fig. 2) and Ptiloglossa the adductor ridge fades out medially so that the basal part, on the adductor con- vexity and distal to it, is not clearly con- nected to the distal part along the lower margin of the distal half of the mandible. A continuation of the basal part extends distally, however, toward the mandibular apex, as a broad, secondary, longitudinal median ridge in the adductor interspace, a special and doubtless derived ridge not found in most other bees. A similar sec- ondary ridge is weakly developed in Cal- lomelitta. The latter genus is unusual for having the rutellum divided to form two teeth, a modification possibly associated with nesting in wood instead of soil. ANATOMICAL STUDY OF MANDIBULAR STRUCTURE IN BEES 475 = D> UDA=AP TE Wi ae \ os) oR — ) CRO CG Fics. 35-42. Outer and inner views of mandibles of female Apidae. For species names see Table 1; for abbreviations see Tables 2 and 3. Halictids show only limited diversity in mandibular structure. In Halictus quadricinctus, but not in related bees such as H. ligatus and Lasioglossum, an inner condylar ridge lies just beneath the basal part of the lower mandibular margin and extends from near the condyle to the mar- ginal part of the adductor ridge, leaving a pilose groove just behind the usual (outer) condylar ridge. Some wood-nesting halic- tids have slightly modified mandibles. Thus, Augochlora pura has the trimmal extension irregularly shaped and Mega- lopta genalis has the same feature and in addition two large teeth on the inner man- dibular surface. These seem to be enlarge- ments of the fimbrial carina, since there are a few hairs, more or less in line with those along the trimmal carina, above each of the teeth. In the Nomiinae (Nomia) and Dufoureinae (Dufourea, Systropha), the mandibles are ordinary and the fim- briate ridge is a mere change in level (de- pressed below the line), with no hairs in Systropha. 476 Tue University oF Kansas Sc1ENCE BULLETIN Andrena accepta and, to varying de- grees, other species of Andrena are unusual in having the adductor ridge directed downward almost from its base so that the condylar interspace is very short and the condylar groove correspondingly long. In the Panurginae such as Panurginus, Panurgus, Perdita, and Psaenythia, the pollex is reduced and does not form a tooth. Psaenythia has a longitudinal, me- dian secondary ridge in the adductor in- terspace suggestive of that of Colletes, but it is not connected with the adductor ridge and is presumably of independent origin. Protoxaea (Oxaeidae) has an inner con- dylar ridge similar to that of Halictus quadricinctus. It may be a feature of large, slender mandibles. Other striking features of the mandible of Protoxaea are lack of the tooth at the apex of the pollex and the weak condylar and outer ridges that are adjacent to one another distally before merging with the cap of the rutellum. All four subfamilies of Melittidae have quite ordinary mandibles. The most un- usual is Dasypoda, in which the mandible is unusually strongly curved and the cap of the rutellum is enormous, almost half as long as the mandible, as a result of the short grooves on the outer surface. The Fideliidae also falls in the group showing ancestral mandibular structure. In Fidelia the tooth of the pollex is enor- mous, as large as the rutellum. In both Fidelia and Neofidelia the trimmal carina, upper carina, and the brief fimbriate line unite in a Y-shaped pattern. In the nest-making Anthophorinae (ex- amined in Anthophora, Svastra, Melis- sodes, Centris, Diadasia, Melitoma, Exo- malopsis, Tapinotaspis, etc.), the mandibles are of the common ancestral type, with deviations in certain genera as mentioned in subsequent paragraphs. The relations of the trimmal carina, up- per carina, and fimbriate line vary widely even among bees having the generally ancestral mandibular type. In some (e.g., Colletes), all three structures are separated. In Diphaglossa, Fidelia, Neofidelia, Svas- tra, and some Anthophora, all are con- nected to form a Y-shaped complex and this is nearly true, also, in Melitta and Andrena. \n bees such as Anthophora, Panurginus, and Protoxaea, the hairs just below the fimbriate line and trimmal ca- rina form a single continuous row and the point where the trimmal carina stops and the fimbriate line begins is quite arbitrary. This is almost true of Centris, also. In Halictus the trimmal and upper carinae are united with one another and isolated from the fimbriate ridge. In the panurgines and Protoxaea, the fimbriate line is even more longitudinal than in most of the bees here discussed, a common feature among bees with slender mandibles. In Protoxaea the upper carina extends farther basad than usual, parallel to and almost as far as the trimmal carina. A short transverse ridge extends from the fimbriate line to a weak elevation of the upper carina. The outer ridge in mandibles of the general ancestral type ordinarily is rather uniformly wide basally and curves upward to the vicinity of the acetabulum. In Diphaglossa, it is much widened basally, more than half as wide as the mandible, and almost replaces the outer interspace. In Svastra, Anthophora and in Ctenocol- letes, the outer ridge has two bases, the usual upper root near the acetabulum and another, the lower root, directed toward the condyle. The lower root separates the outer groove from the outer interspace; the outer groove continues nearly to the base of the mandible and the interspace lies between the two roots of the outer ridge. Parasitic anthophorids: Mandibles of the Nomadinae, as well as of the Melectini and Ctenioschelini of the Anthophorinae, ice., of the parasitic Anthophoridae, are also ANATOMICAL Stupy OF MANDIBULAR STRUCTURE IN BEES 477 similar to those of the ancestral type but shorter, more like those of many males. Basal parts of the condylar and outer ridges in these parasitic forms fade into the outer interspace, not reaching the base of the mandible or in TAyreus reaching the base, but weakly differentiated. In Nomada and Ericrocts the tooth of the pollex is absent; it is unusually large in Thyreus and Me- lecta. The acetabular carina is absent in Ericrocis and the acetabular groove is rep- resented only by a short row of hairs in an area that is otherwise a continuum from the outer ridge to the upper carina. In Thyreus, Melecta, and Ericrocis, the trim- mal carina seems continuous with the up- per carina; the fimbriate line is absent but the fimbria is present, slanting away from the upper carina and in the melectines con- tinuous with the hairs along the trimmal carina. Such mandibles are variably sim- plified as compared to the ancestral type; those of Biastes (Fig 19) are the least simplified of any parasitic anthophorid studied and would easily be placed among the mandibles of the ancestral type dis- cussed in the preceding section. Hylaeinae and Xeromelissinae: Other somewhat modified mandibles include those of the colletids that use preformed burrows or excavate pith, the Hylaeinae and Xeromelissinae. Amphylaeus, for ex- ample, has short mandibles, broad at the base and tapering apically (Fig. 5), with the cap of the rutellum much reduced by the outer interspace—outer groove which together form a broad triangle. The ace- tabular groove is broad, the acetabular carina absent as a sharp ridge, there being no carina, but only a space or “ridge” be- tween the acetabular groove and the trim- mal extension. The upper carina is on the inner surface of the mandible. A strong, broad secondary ridge, suggestive of that of Colletes, extends from the basal portion of the adductor ridge across the adductor interspace toward the distal part of the mandible. In Hylaeus this ridge is absent. Moreover, the trimmal carina and fim- briate line are absent; there is only a round- ed surface separating the trimma from the adductor interspace. Also in Hylaeus the upper carina is in its usual position on the upper margin of the pollex. Chilicola has more elongate mandibles. An unusual feature is the extraordinarily broad acetabular groove. Xylocopinae and Lithurginae: Bees of these subfamilies, separated according to usual classifications into two, well known families, the Anthophoridae and Megachi- lidae, have short mandibles, wide at the bases and commonly tridentate at the apices, because of the lower tooth of the rutellum which is the enlarged apex of the adductor ridge, in addition to the usual upper preapical tooth which is the apex of the pollex. [Anthophora (or Clisodon) terminalis has similar mandibular dentition although the mandible is more elongate. | Macrogalea, Allodape, and Ceratina have the outer ridge narrow and at the upper margin of the large outer interspace; the ridge is carinate in Macrogalea. In Macrogalea there is a lower root of the outer ridge, a feature present also in Man- uelia. In Manuelia, Xylocopa, and Li- thurge, the outer and condylar ridges are both broad, so that the outer inter- space is small. The mandibles of all these bees tend to have a deep, longitudinal de- pression in the middle of the apical half of the inner surface. The fimbriate line in the Xylocopinae, if recognizable, is longi- tudinal and along the upper margin of this depression, but in Lithurge it is oblique and well above the depressed area. The ground-nesting genus Proxylocopa has more slender mandibles than most Xylocopinae, without the lower tooth of the rutellum, thus resembling the basic ancestral mandibular type. Many wood in- 478 Tue University oF Kansas ScrENCE BULLETIN habiting species of Xylocopa, however, also lack this tooth, as shown for X. virginica (Fig. 24). Megachilinae: The megachilines are re- markable for the tendency of the mandible to be broad, with the apex multidentate, especially in females. This breadth is achieved by great broadening of the area of the acetabular groove (i.., the space between the outer ridge and the acetabular carina) to form an acetabular interspace, itself often with multiple, secondary, longi- tudinal ridges or rami of the outer ridge. This development, which is also seen to a lesser degree in the Apidae, is interpretable on the basis of females alone but verified from examination of males (Fig. 31), which have more slender mandibles. The acetabular groove is not particularly wide in male Hoplitis, for example, and the identity of the condylar and outer ridges seems clear. The variation in the outer ridge makes one question its homology when examin- ing species of certain other genera, espe- cially the females. The two ridges that terminate, as in all bees that have such ridges, on the apical or first mandibular tooth (cap of the rutellum) must be the condylar and outer ridges. It is the more basal parts of the latter ridge that some- times become confusing. In the species il- lustrated, the outer ridge is not complete, but it is strong and continuous in forms such as Osmia subaustralis, subfasciata, and lignaria, which have mandibles otherwise similar to those of Hoplitis. The basal part of the other ridge is united with the acetab- ular carina in all these forms. In No- teriades sp., however, the outer ridge has two continuous roots, the usual upper one directed toward the acetabulum and unit- ing with the base of the acetabular carina, and another, the median root, directed toward the abductor swelling. In Chalh- codoma exilis and Chelostoma fuliginosum the upper root is broken or entirely absent, but the median root is continuous, so that the ridge runs from near the abductor swelling toward the apex of the mandible. In these forms, a secondary ridge, compara- ble to a similar secondary ridge in Anthid- ium, extends from near the acetabulum (where it is probably the base of the upper root of the outer ridge and is united with the acetabular carina) across the acetabular interspace toward the distal margin near the apex of the mandible. Thus there are three ridges on the outer mandibular sur- face leading toward the apex of the mandi- ble instead of two. (There seems to be no justification for regarding the upper of these three ridges as the acetabular carina and all the remaining upper toothed man- dibular margin as an expansion of the trimmal extension; the latter, in most bees including related megachilids, is smooth and hairless, unlike the acetabular inter- space.) In Heriades carinata the distal part of the outer ridge has entirely lost its normal roots and, instead, is elevated and fused near the base with the condylar ridge. The mandible is thus left with a large area lacking ridges. An independent and far more extreme loss of ridges has occurred in Anthidiellum, in which the outer sur- face is hairy almost throughout, the hairy surface curving onto the condylar inter- space without an intervening ridge. The distal parts of the grooves impinging on the small cap of the rutellum are broad, flat, and minutely hairy. The inner sur- face of the mandible is dominated by the broad, minutely hairy concavity of the dis- tal part of the adductor interspace. All this is suggestive of features found in some Apidae and may be related to the fact that nests of Anthidiellum consist entirely of plastic material (resin), like the wax, resin, or mud utilized by Apidae. The acetabular carina and trimmal ex- tension are distinct in most megachilines, ANATOMICAL STUDY OF MANDIBULAR STRUCTURE IN BEES 479 but the carina is absent and the extension is therefore ill defined in Anthidium and Callanthidium. The trimmal carina is ab- sent in female Hoplitis, but is usually pres- ent in megachilines, continuous with the fimbriate line or almost so, and both are margined on the lower side by a band or row of hairs. The fimbriate line is oblique or almost transverse, parallel to the distal margin of the mandible, in most female megachilids. In some forms, like Hoplitis, what appears to be the fimbriate line is angulate. In others, like Crerghtonella, there is a premarginal carina or line distal to the fimbriate line. In Creightonella there is a short ridge branching from the adductor ridge and directed upward and distad. Another sec- ondary ridge, likewise in the adductor in- terspace, is longitudinal and more nearly median in most megachilines. In many megachilines (Dianthidium, Paranthidium, Parevaspis, Euapis, Chali- codoma, Megachile frugalis but not albitar- sis), the distal part of the adductor ridge is provided with a strong fringe of erect hairs, as in the Meliponinae. In the same mega- chilines, the inner fimbria is often well developed, and in Megachile frugalis (not albitarsis) there is a conspicuous, outer pre- marginal fimbria. Meliponinae and Apinae: These groups are remarkable for the lack of ridges and of the cap of the rutellum on the outer sur- face of the mandible. In Meliponinae a groove, perhaps in part the acetabular groove, is strongly evident. In Apis it is absent. When present, this oblique groove slants strongly across the mandible toward the apex (so that at least the distal part is presumably not the acetabular groove) and near the apex it branches, the lower branch going to the lower margin of the mandible, the upper continuing toward the apex. In Meliponula and some Trigona the acetab- ular carina is present, although weak. Ex- cept for the acetabular carina and the oblique groove (when they are present), the outer surface of the mandible is convex and rather uniformly hairy. The inner mandibular surface has a well formed ad- ductor ridge. Distally this ridge is con- spicuously fringed or hairy as in some megachilids and curves downward to form the lower margin of the mandible, but there is no condylar groove separating this margin from the rest of the outer surface of the mandible. The condylar interspace also is continuous with the outer surface of the mandible, with no intervening ridge. In workers, but not queens, of Apis there is a secondary, oblique carina in the ad- ductor interspace, more or less parallel to the distal half of the adductor ridge. Such a carina is absent in Meliponinae. The fim- briate line is well developed in queens of Apis, less conspicuous in workers. In Meli- poninae it is probably present, but its up- per end is deflected distally toward the upper distal angle of the mandible. The trimmal carina is absent; the upper carina is present, forming the upper margin of the mandible distally, but does not approach the base of the mandible. Bombinae: These large bees (see Figs. 39- 42) have much more structure on the outer side of the mandible than do the Meli- ponini and Apinae. Thus in the Euglossini (Eulaema, Euplusia) there is a condylar ridge (double at the base or with an inner ridge behind the outer one as in Halictus guadricinctus) and an outer ridge (weak medially) delimiting an outer interspace and outer groove. Probably the distal half of the outer groove is double, with an up- per and a lower branch, for an extra ridge branching from the median part of the outer ridge joins the upper part of the cap of the rutellum. Between the two branches of the outer groove is a secondary intercallary ridge. The acetabular groove is broad with a considerable space between the bottom of 480 Tue University oF Kansas ScrENCE BULLETIN the groove and the acetabular carina. The acetabular carina is strong and the trimmal extension is broad and with a minutely pubescent area apically. On the inner sur- face, the adductor ridge is normal, the trim- mal carina is absent, but the acetabular carina reaches almost to the mandibular base; it is arcuate inward subbasally and then ascends toward the acetabulum. The upper carina is present, isolated from the oblique fimbriate line. Distal to the latter is an angular transverse premarginal ridge. In Euglossa the mandibles are similar to those of Euplusia, but the ridges and grooves on the outer surface are weak ex- cept for the acetabular groove. Exaerete is also similar, but the outer ridge is only incompletely divided apically by depressed islands; none of the grooves ending on the outer surface of the rutellum contain finely pubescent distal areas, as in the nonpara- sitic genera. In the Bombini the condylar ridge is compound, not only basally [where the uppermost ridge of the condylar ridge sys- tem was called the basal keel by Yarrow (1954) |, but also on the apical part of the mandible where the lower or inner ramus is near the lower mandibular margin, the upper well above it, terminating on the outer groove in Psithyrus and as a process into that groove in Bombus. The distal part of that groove and of the grooves above it on the outer mandibular surface (also of the condylar groove, especially in Psithyrus) are broad, flat, nearly reaching the distal mandibular margin, and are cov- ered with minute pubescence. As in Eu- plusia and most other Euglossini, the outer ridge is divided distally in Bombus, the upper ramus (Yarrow’s main keel) sepa- rating a broad, intercallary depression or groove from the broad distal part of the acetabular groove. The lower ramus 1s Yarrow’s second keel. All the “grooves” in 3ombini are so broad and flat distally as to be better described as depressed areas, and the cap of the rutellum is a mere band on the distal margin of the mandible. DISCUSSIONS AND CONCLUSIONS Recognition of homologous features among the ridges and grooves of mandibu- lar surfaces is possible and these features appear to shed light on relationships among bees. A common or “basic” pattern is found in mandibles of ground-nesting groups of bees of the families Colletidae, Halictidae, Andrenidae, Oxaeidae, Melit- tidae, Anthophoridae, and Fideliidae. These are, for the most part, bees that make their own burrows and cells in the soil. The same pattern is widespread: among sphecoid wasps and it is presumably the primitive pattern for bee mandibles. Members of these groups that nest in wood (e.g., Callomelitta in the Colletidae; Mega- lopta, Augochlora s. str., and Lasioglossum (Dialictus) caeruleum in the Halictidae; and Clisodon, a close relative of Antho- phora, in the Anthophoridae) have similar mandibles, often modified by the acquisi- tion of extra teeth, or not recognizably modified as in the Lasioglossum. Most male bees have mandibles somewhat smaller and more slender than those of females, and often with certain ridges weak or absent. Parasitic Anthophoridae (No- madinae, Melectini, Ctenioschelini) have somewhat simplified mandibles, suggesting those of many male bees (parasitic halictids do not show this tendency). The groups of bees listed above occu- py the whole base as well as some of the top of the phylogenetic tree for the bees as presented by Michener (1944), or in modified form by Michener (1974). There are various derived types of mandibles, however, as discussed below. Thus the colletid subfamilies (Hylaeinae and Xero- melissinae) that nest in preformed cavities in the soil, wood, or stems, or perhaps excavate in pith, have rather different man- ANATOMICAL STUDY OF MANDIBULAR STRUCTURE IN BEES 481 dibles. Such derived mandibular types are not found, however, among bees that make subterranean burrows and earthen cells. A significant finding is the position of the Fideliidae clearly within the basic group. There is no doubt of the similarities of fideliid larvae to those of the Megachili- dae (Rozen, 1973), a relationship supported by the metasomal scopa, but the mandibu- lar structure, independent volsellae, wing venation, and other characters make us hesitant to place the Fideliidae in the Megachilidae. The presence of a lower tooth of the rutellum and of a deep depression on the inner mandibular surface are similarities of the anthophorid subfamily Xylocopinae (most species) and the megachilid sub- family Lithurginae. Perhaps these similari- ties are convergences, as suggested by the different relation of the fimbriate line to the depression in the two groups, nearly all members of which burrow in wood or pith. Other characters will have to be analyzed before the meaning of this simi- larity is clear. The similar dentition of Clisodon, a wood inhabiting genus of an otherwise ground dwelling group, shows that at least this feature can appear by con- vergence and may indicate little about lines of descent. Of much interest is the finding that in the Megachilinae and to a lesser extent in the Apidae, unlike other bees, the distal part of the condylar groove or the area im- mediately above it is expanded to form a condylar interspace. The toothed apical margins of mandibles of these bees are therefore largely derived from the pollex of ancestral groups. It is interesting that these are the two groups which commonly carry foreign materials to the nest for con- struction purposes. The megachilids gen- erally have strongly toothed mandibles for cutting, preparing, and sometimes carrying materials such as leaves, plant hairs, peb- bles, and resin. The apids have weak teeth (or lack them) and more scoop shaped mandibles for handling wax, resin, mud, and the like. In the apid subfamilies Meliponinae and Apinae, the ridges and grooves on the outer mandibular surface are largely or al- most wholly absent, the surface being con- vex and hairy. Loss of mandibular ridges and carinae may be convergent, associated with use of relatively soft materials (wax, resin, mud) in nest making, as suggested by the relatively featureless, hairy mandi- bles of Anthidiellum, a megachilid that makes nests of resin. Absence of ridges may also be associated with minute size, suggesting that the Meliponinae and Apinae arose from minute ancestors, a view supported for the Meliponinae by their reduced wing venation, a common feature of small insects. The Bombinae have mandibular ridges. In the illustrations and discussion we have done our best to identify them with the ridges of other families of bees. However, secondary ridges are present (not the same in the Euglossini, Psithyrus and Bombus) and it is possible that most of the ridges arose newly in Bombinae from ancestors with few ridges like the Meliponinae. In Euglossini the grooves are reasonably simi- lar to those of other families of bees al- though usually minutely pubescent api- cally. In the Bombini, however, the “grooves,” at least apically, are broad, slightly depressed, flat areas covered with short, fine hairs. The above remarks should not be used alone as a basis for reclassification of the Apoidea. However, they should not be ig- nored when such a work is done. The common derived characters of the Lithur- ginae and Xylocopinae, and of the Mega- chilinae and Apidae, may well result from convergence, but if they are supported by other findings, they will require restudy of the familial classification. The functions of the ridges and grooves 482 Tue UNIversity oF KANsAs SCIENCE BULLETIN are not known, but it is reasonable to as- sume that they serve to strengthen the mandibular cuticle. This view is supported by the tendency for weakened and some- times also less numerous ridges and ACKNOWLEDGEMENTS This study supported by National Sci- ence Foundation Grant DEB 73-06815. We wish to thank Mr. Paul G. Decelles for assistance with the drawings and Mrs. Joetta Weaver for editorial assistance and typing. EVPERA TORE, ChlED Erckwort, G. C. 1969. A comparative mor- phological study and generic revision of the augochlorine bees. Univ. Kan- sas Sci. Bull., 48:325-524. Micuener, C, D, 1944. Comparative exter- grooves in males and parasitic female bees, which do not make nests, and in bees that construct nests of soft materials like wax and resin. nal morphology, phylogeny, and a classification of the bees. Bull. Amer. Mus. Nat. Hist., 82:151-326. . 1974. The social behavior of the bees. Harvard Univ. Press [Cambridge], 404 p. Rozen, J. G. 1973. Immature stages of lithur- gine bees with descriptions of the Megachilidae and Fideliidae based on mature larvae. Amer. Mus. Novitates, 2527:1-14. Yarrow, I. H. H. 1954. Some observations on the genus Bombus, with special ref- erence to Bombus cullmanus (Kirby). Jour. Soc. British Entom., 5:34-39, SCIENCE BULLETIN MUS. COMP. ZOOL LIBRARY OCT 23 1978 HARVARD UNIVERSITY sn ee ECOLOGY AND EXPLOITATION OF CTENOSAURA SIMILIS BS = % By HENRY 5S. FITCH and ROBERT W. HENDERSON erate eceteascesearanarahataatavate"anetavereanetavone“prehaetstavetetatstotetatecstonstanstatstetonstetstetetatccatenstonseatesstonsn sere Pe gn” = Raa I ST Se re — oe ebatatatatanatocatenatetanetonetoeee en cateteeee RX I KK RRR RIOR Sete ba I RD Vol. 51, No. 15, pp. 483-500 September 27, 1978 seatetetetate: ANNOUNCEMENT The University of Kansas Science Bulletin (continuation of the Kansas Uni- versity Quarterly) is an outlet for scholarly scientific investigations carried out at the University of Kansas or by University faculty and students. Since its incep- tion, volumes of the Bulletin have been variously issued as single bound volumes, as two or three multi-paper parts or as series of individual papers. Issuance is at irregular intervals, with each volume prior to volume 50 approximately 1000 pages in length. Effective with volume 50, page size has been enlarged, reducing the length of each volume to about 750 pages. The supply of all volumes of the Kansas University Quarterly is now ex- hausted. However, most volumes of the University of Kansas Science Bulletin are still available and are offered, in exchange for similar publications, to learned societies, colleges and universities and other institutions, or may be purchased at $20.00 per volume. Where some of these volumes were issued in parts, individual parts are priced at the rate of 2 cents per page. Current policy, initiated with volume 46, is to issue individual papers as published. Such separata may be purchased individually at the rate of 3 cents per page, with a minimum charge of $1.00 per separate. Subscriptions for forthcoming volumes may be entered at the rate of $20.00 per volume. All communications regarding exchanges, sales and subscriptions should be addressed to the ExcHance Lrprarian, UNIVERSITY oF Kansas Lipraries, LAwreNcE, Kansas 66045. Reprints of individual papers for personal use by investigators are available gratis for most recent and many older issues of the Bulletin. Such requests should be directed to the author. The International Standard Serial Number of this publication is US ISSN 0022-8850. Editor Eugene C. Bovee Editorial Board William L. Bloom Philip W. Hedrick Rudolf Jander Harvey Lillywhite Charles D. Michener Norman A. Slade Henry D. Stone George W. Byers, Chairman THE UNIVERSITY OF KANSAS SCIENCE BULLETIN Vol. 51, No. 15, pp. 483-500 September 27, 1978 Ecology and Exploitation of Crenosaura similis Henry S. Fircu! and Rospert W. HENDERSON” CONTENTS PME. SIRIA Cyl Mee 220 he we en Eo enna sere FORE Te see nS eee ee eh oe ahr ee 483 PNET RODE CO Nie. ewer tee ees Cer AE Sree we eh ade EE Pe ys ee 483 IMATERTAESSAND OVI DHODSIee.on2sc0P ease es oie ee Oe ed iee dd Pee ey wae 484 PNCEANONVISEDG EME NGS) Beit 28 Mea Set pon) ee 1 live 2 ua an ree Se EE eee eta he, oe a ie at 484 EC OL OG Youn eee een Ne ATE a. wed, Weer ON are oe wine ere ee ML tt Ae Ae 484 lant tam ne et eee eh ae ee eS Se et ares Sele ye WOR Lee Oo Peat ae 484 IES CROWS ees ee ee eee as eee iat fe aR sera Seem Rae La ee a ees a 485 SOcia PRDCH AVION) Sims Se een a tS ce SOA ee ate te Pete ian ee on ee ee 486 IW CoNWeOnTS TOs Betty A. Ses BRIAN HES vie aaa amy cy ye tetas foes See eeets et aoe ee 487 GOCE al (came ness: eee a Me me Ae eee oe ea, en. tant he tenet ec ht ae 488 Re PROGUCH OME tee en sectors me cel eter CLS ie eae ee eae aE eee 490 (CHO ANG Ua ce Se ee CR Lee NE PRE NE BRED hie e RUS Tus Ste Laine eee Alege 492 FO pUllelRI OM es ACIS ICY ins = seer eka 2 8S cee mek AM Ore sero ete See ae eect 495 Bol ACO ie SUR CLUGs ee: ae eter aes a tore) eer a a a a eae Se eee 496 ESP VOMEATION: pert seek Fehr: 1 Po OEE A Edy Ed a Soe SE Ore et De Ea ee 497 [DY SESE, Fe te geet pI ae Ut Cnr ge SOs pee ER eee Nee eR Ce ee 499 ey ERATOR EM CRED Rete ak ee eee Oe ee, Ce nee oe eae og: Se SET Oe eee 500 ABSTRACT The giant herbivorous iguanid Ctenosaura similis, of southeastern Mexico and Central Amer- ica, Occurs in open woodland or edge habitat in seasonally dry lowlands. Sexual maturity is at- tained late in the second year, ovulation occurs about mid-February, and laying of the eggs (mean 43.4, 12-88) occurs about 5 weeks later. Hatchlings appear about mid-May and have tripled in length at the end of the first year when they are half-grown in length. Females outnumber males 2 to 1; but males are about 1.25 times female length and 1.8 times female weight. Biomass may be several kg per ha (1.67 per ha on a 10 ha sample area in Belize). Each ctenosaur centers its activity at a lookout and shelter; typical foraging radii are from 18.7 m in first-year young to 43.0 m in adult males. Food consists of many kinds of foliage, flowers and fruits, and some animal matter including small rodents, lizards, eggs, and insects. Exploitation of the ctenosaur 1s heavy in some parts of its range, including Nicaragua, where the species is a common article of diet for country people and also is sold by the hundreds in city markets. As a result, numbers have decreased drastically. Conservation is needed, especially protection of reproductive females, to assure a sustained yield of the flesh, a valuable natural resource. INTRODUCTION 1 University of Kansas, Muse E Nz is- ee : nights Be eaten a ee ae eT ene Crenosaura similis, a giant iguanid liz- tory, Lawrence, Kansas 66045. 2 Milwaukee Public Museum, Milwaukee, Wiscon- ard of southeastern Mexico and Central sin 53233. America, is of extraordinary interest eco- 484 Tue Universiry oF KANnsAs Sc1IENCE BULLETIN logically, being important in some areas as a game animal, as a source of high grade protein food for people and, at times, as a pest. Thus, and because it is an abundant and highly conspicuous member of the Middle American herpetofauna, we inde- pendently undertook ecological studies of it in Costa Rica and in Belize, respectively. Subsequently, we combined our efforts in a field study which centered in western Nicaragua and was encouraged and funded by the Banco Central of Managua, Nica- ragua. There, we investigated conservation ori- ented aspects of the ecology not included in our earlier studies, with attention to exploitation, past and present. Although further field work is desirable, the findings presented here will provide some basis for a general understanding of the ctenosaur’s basic ecology, of its actual and potential value as a natural resource and of the meas- ures necessary to preserve thriving popula- tions with a sustained annual yield. We have here combined and integrated diverse data obtained from various localities over a 10-year period, for that purpose. MATERIALS AND METHODS In Costa Rica, from October 1967 to August 1971, intermittently, Fitch studied local populations at Quepos and Boca de Barranca, Puntarenas Province, and at Finca Taboga and Playas del Coco, Guana- caste Province, with casual observations elsewhere. The lizards, chiefly first-year young, were individually marked by toe-clipping. Measurements (S-V, tail) were recorded to the nearest millimeter and exact location was noted in each instance. Length was esti- mated for all those sighted from a car. At Belize City, in 1970-1971, Henderson (1973) also studied ctenosaurs by mark and recapture technique and ob- served the behavior and movements of recognizable individuals over periods of months. Our joint study in Nicaragua in February, March and April, 1976, in- volved examining ctenosaurs at city mercados (N= 827), and purchasing selected individuals (N=114) for examining gonads, and finding and interviewing professional hunters who supplied the city vendors. Those and others who hunted the lizards for home consumption provided information about ctenosaurs’ population trends, local habits and the methods of securing them. Also, we collected series (N=160) for dissection and examination of stomach contents and gonads. ACKNOWLEDGMENTS We are indebted to Sr. Jaime Villa of Managua, Nicaragua, for advice and ma- terial aid of various sorts. Mr. Allen Porter, a U.S. Peace Corps Volunteer, as- sisted us in the field in nearly all phases of the work and carried on with the project in our absence during late March and April, 1976. We thank the Banco Central of Managua and particularly Dr. Jaime Incer and Sr. Roberto Incer of that institu- tion for funding our field study. The Cen- tral Nicaraguan government’s Catastro expedited our field work by making avail- able a jeep and driver. ECOLOGY Habitat: Over its extensive range, in Mex- ico east of the Isthmus of Tehauntepec, and Central America, Ctenosaura similis occurs in varied climates and plant forma- tions, but it is absent from primary rain forest and from cool climates at high alti- tudes. One seen as a road kill, in June 1968, at Rio Itiquis, Costa Rica Highway No. 1, Alajuela Province, approximately 900 m elevation was probably near the upper edge of the species’ occurrence. The species is most prevalent on the relatively arid Pacific versant of Central America, but it intrudes onto the Caribbean versant and even reaches the Caribbean Coast and various small islands of the Continental Shelf in the western Caribbean. The optimal habitat seems to be arid savanna. In Guanacaste Province, north- western Costa Rica, and in Chinandega, Leon and Managua provinces, western Nicaragua, we observed high population densities in the dry season in areas made relatively barren by heavy grazing and/or annual burning of dead vegetation. Broken terrain with scattered, large trees, logs, loose boulders, rock outcrops, gullies with cutbanks, or streams with riparian thickets, provided the required shelter. Lava fields with a sparse vegetation are inhabited. Ecotocy aNp ExpLoirATION oF CTENOSAURA SIMILIS 485 Ctenosaurs have often been found closely associated with cattle at corrals, salt blocks, and pasture edges. Such barren places pro- vide the lizards with enough food from low weedy vegetation missed by the cattle and from foliage and flowers of trees. Rock walls provide a superabundance of shelter and may have unusual concentra- tions of ctenosaurs. Fence posts, especially if large and hollow, are used both for bask- ing and for shelter. Ctenosaurs often live on the margins of cultivated fields feeding on the crops and also on weeds. The disturbed and seral conditions re- sulting from human activities often favor survival and increase of the ctenosaur, some- times found living in close association with people. At the village of Playas del Coco in Guanacaste several were seen in a pig sty where they came to eat garbage. On the northern edge of Belize City and at Boca de Barranca, Puntarenas Province, Costa Rica, colonies of ctenosaurs lived among groups of buildings where vegetation was held in check by artificial trimming and by the trampling of large numbers of peo- ple (schoolyard, beach resort). Their bur- rows were beneath buildings. Many cteno- saurs were observed within the city of Managua, Nicaragua, in small, semi-iso- lated colonies in yards with trees and on roofs. Walled gardens provided an espe- cially secure habitat. Other ctenosaurs lived in vacant lots with trees and grass, or lived among shacks or adjacent to ware- houses or ruins where unused machinery, piled lumber, earthquake debris or trash provided shelter. In the humid Caribbean lowlands, ctenosaurs are relatively scarce and local- ized in dry and open situations. On Great Corn Island, Zelaya Province, Caribbean Nicaragua, they were observed in forest and thick plantations, and appeared to be much more arboreal that their mainland counterparts. At Belize City, Neill and Allen (1959) found ctenosaurs in man- grove swamps, and Henderson (1973) found adults in the open areas about build- ings, but found juveniles in a swampy and brushy adjacent area, often climbing in the bushes. Burrows: Ctenosaurs dig effectively with alternate strokes of the powerful, clawed forefeet. Some find shelter above ground, in hollow limbs or fence posts, but most dig burrows where they are protected above by such objects as buildings, boulders, or tree roots. Vertical banks of road cuts or eroded gullies also provide favored sites and burrows often have well defined trails leading to them. The holes are flat bot- tomed, arched above, and wider than high, with winding and sometimes branched tunnels. Several that we excavated were between 1 and 2 m long. Some burrows have cavernous entrances, enlarged by dig- ging predators or eroded by heavy rains. The tunnels are wide enough to permit the lizard to turn around inside and are hori- zontal or slope gently downward. Various snakes, lizards, rodents and arthropods use ctenosaur burrows for shelter. Ctenosaurs do not emerge at night or on rainy days. Even when weather is fa- vorable, the lizard does not emerge until well after sunshine has reached the burrow entrance. At first, only the head protrudes, and by gradual stages over perhaps half an hour, more of the body. Finally the lizard scrambles to a nearby lookout to bask where the surface is already warmed by the sun. Only when the body temperature reaches 36°-37° C, is the animal fully active and ready to forage. The ctenosaur is most active in the morning, retiring underground to escape the midday heat, and often has been seen in the afternoon lying in burrow entrances or back in the tunnels awaiting more favor- able lower temperatures. Each lizard defends its home burrow against intruders, but may use other out- 486 Tue University oF KAnsAs SCIENCE BULLETIN lying burrows occasionally or habitually. An individual often shifts from one home burrow to another, and collectively, the burrows are the communal property of the local population. Social Behavior: Territoriality in Cteno- saura similis involves exclusion of others from home burrows and basking places, and maintenance of spacing. Intrusions may cause conflict, with fighting mainly between similar-sized first-year young. A threatening approach is followed by spar- ring, lunging, biting, and persistent cling- ing to any part of the opponent that the lizard can seize, while writhing or rolling in an effort to break the other’s grip. No noticeable damage was noted from such encounters. Doubtless the most severe fighting is between large males, but is infrequent be- cause of their relatively small numbers and wide spacing. Over a period of 11 months, Henderson (1973) observed no overt inter- actions between several adult males that lived beneath the same building on the outskirts of Belize City. There seemed to be mutual avoidance of confrontations by maintenance of spacing. However, when an outsider was brought in and tethered, he was promptly attacked by a dominant resident. After a series of struggles with intervening pauses for rest, the resident male gradually abandoned his attempts to oust the tethered intruder and the latter, when released, remained in the vicinity, taking possession of the burrow and bask- ing site of the resident whom he had out- lasted in the territorial confrontation. The adult females seen always main- tained spacing of at least several meters. As in other iguanids, Crotaphytus (Fitch, 1956), Sceloporus (Blair, 1960), and Uta (Tinkle, 1967), there is probably mutual aversion and hostility among females, but with aggressive behavior relatively weak. No female fighting nor pursuits were ob- served. Adult males were often seen in “consort pair” associations with females that seemed to be based on mutual attrac- tion, but at any time the polygynous male might wander away from his consort and find another temporary mate. Formation of harems was probably prevented by mu- tually agonistic responses of females. Each individual, regardless of sex or size, has a territorial center with one or more burrows or basking sites, but terri- torial boundaries are not well defined. The ctenosaur may spend much time basking at one or a few spots, and finally may move on to a different location without ever having used most of the apparent terri- torial area near its burrow and lookout. Often, food in the form of succulent vege- tation is so abundant that the lizard does not need to spend much time in active for- aging nor to venture far. Where food is abundant, but choice re- treats and basking sites are scarce, terri- tories may be partly superimposed, with the same rocks or burrow used by two or more ctenosaurs at different times. Even in such congestion there is not much actual conflict. A smaller, weaker individual au- tomatically avoids a larger, more powerful neighbor and is generally ignored by him. Evans (1951) described social behavior in a colony of Ctenosaura pectinata in central Mexico, living in a rock wall around a bean field. Each adult male controlled a section of the wall and adjacent areas, ordi- narily excluding other adult males. How- ever, perhaps as a result of abnormal over- crowding, there was much overlapping of territories, and a hierarchical system was superimposed on that of territoriality. A “despot” Alpha male could violate the ter- ritory of any other individual with impu- nity, although in such transgressions mutual threat displays were exchanged be- tween the intruder and the resident. Like- wise, other territorial males encroached on EcoLocy AND ExPLoITATION OF CTENOSAURA SIMILIS 487 the territories of lower ranking individuals. Many subordinate individuals were unable to maintain territories, and avoided the large dominant males. The somewhat different social structure in the population that we observed at Be- lize probably resulted from the different situation, rather than specific differences between C. pectinata and C. similis. At the Belize City cemetery, the relative uniform- ity in size and distribution of the grave- stones that provided shelter, dispersed the ctenosaurs, but at Evans’ study area in Mexico, the lack of shelter, except in crev- ices in the rock wall, concentrated them, leading to frequent, stressful interactions. Like most other iguanids (Carpenter, 1965), Ctenosaura similis displays by stereo- typed, species-specific bobbing movements that substitute for actual fighting in most encounters. Henderson (1973) observed and described the display at Belize City consisting of 10-12 vertical bobs of the head with changing rate and amplitude; first, the head was slowly raised to the maximum possible height and then more rapidly lowered; 2 or 3 lower and faster bobs followed, and finally there were 2 or 3 short, quick, upward jerks. The display was often elicited by the sight of a poten- tial rival, and might be given by females and immatures as well as adult males. Also, displays might be elicited by poten- tial threat, such as the approach of a person. At other times the display was given in the absence of any obvious external stimulus, but seemed to reflect excitement or poten- tial danger. It might be given just before or after a shift to a new location. Movements: Taylor (1956), describing the habits of the species in Costa Rica, wrote: “... they may forage a hundred yards or more in varying directions.” However, forays are usually less than 100 yards from the burrow as indicated by the homeward dashes of those disturbed in the open by a pedestrian or vehicle. A male in sexual search, a female in search of a site for egg- laying, or a lizard lured by some preferred food in the season of scarcity may venture farther than usual. More often, the same individual will be seen time after time at approximately the same spot. Henderson (1973), observing a small colony in the vicinity of a school building on the out- skirts of Belize City, found that most ac- tivity is within 1-5 m of the lizard’s bur- row, but he noted one adult male 88 m from its burrow and recorded several other movements, up to 25 m. Fitch (1973) indi- vidually marked many juveniles, mainly in June and July, and recorded 16 recaptures in new locations after periods of weeks. Four relatively long movements were 131, 55, 49 and 46 m; the remaining 12 averaged 10.4 m (4-16). From 2-7 March 1976, we made observa- tions at the Belize City cemetery on a colony of 49 ctenosaurs. The cemetery was approximately 610 x 200 m bordered on each side (north and south) by a mangrove swamp, and on its west end tapered almost to a point, with a crossroad, swamp, and dense vegetation truncating the lizards’ habitat, while at the east end it was bor- dered by buildings on the edge of Belize. Thus, the population was effectively iso- lated. Individual movements were further restricted by a well travelled road that bi- sected the cemetery lengthwise from east to west, and we saw no ctenosaurs venture onto the road. The lizards of this colony, being free from molestation by humans and conditioned to frequent passers-by, were relatively tame and could be easily approached, sometimes to within 1 m, and they could be individually recognized by a combination of size, sex, and peculiarities of color, pattern, tail regeneration, behavior and location. Each lizard had a favorite spot for basking, usually on a tombstone, and had a retreat, usually within or be- 488 Tue University oF Kansas SCIENCE BULLETIN neath the stone. Most of the tombstones were concrete slabs approximately 2 x 1 x 5 m and many were old, weathered and broken with holes or cracks providing ac- cess to the hollow interiors. Individual ctenosaurs were recorded from 1 to 19 times, but on nearly half the occasions (N=120) were seen at the same place. Movements of less than 3 m were disregarded, because the combined retreat and basking site often spanned approxi- mately this distance. For each ctenosaur movements were recorded as radii between the home base and outlying points visited. For 121 records average was 24.4 m (adult males, 43.0 m, N=22; adult females, 22.55 m, N=38; first-year young, 18.7 m, N= 61). Each adult male usually stayed within 1 or 2 m of an adult female. Any ctenosaur, regardless of size or sex, might shift to a different lookout and usually there was a nearby escape shelter. Use of a specific area by an individual ctenosaur does not conform well with the concept of a typical home range: “The area over which an individual animal ha- bitually travels while engaged in his usual daily activities” (Dice 1952). The cteno- saur spends most of its time at one place and may not stray far from that center while food and other requirements are readily accessible. Feeding and other ac- tivities concentrate in the vicinity of the lookout and shelter; degree of utilization is proportional to distance from that home base. The more remote parts of the area that are within the ctenosaur’s foraging radius tend to be little used, but outlying, alternative, secondary lookouts with asso- ciated shelters permit more thorough use of the area. For each of the main groups—adult males, adult females and first-year young— radii tend to form a graduated series from the minimum to the maximum. Relatively short movements were most numerous and longer movements fewer. Several of the movements (4 of 22 for adult males, 3 of 38 for adult females, and 3 of 61 for immatures) were abruptly longer and may represent either exceptionally large home ranges or shifts in range. Excluding these few extra long movements, the maximum radii were 58.0 m for adult males, 38.2 for adult females and 48.2 m for immatures. These radii represent circular areas of 1.051, 457 and .731 hectares, respectively. Food Habits: Although mainly vegetarian, ctenosaurs are known to have taken some animal food including grasshoppers, frogs, young of their own species, a skink, ro- dents, young chicks and various small birds, a bat and even human feces (Alvarez del Toro, 1960; Tamsitt and Valdivieso, 1963; Taylor, 1956; Fitch et’al,, 1971. Alen= derson, 1973). In February, March and April 1976, we examined 146 stomachs of ctenosaurs col- lected in western Nicaragua (Table 1). Green foliage from both herbs and trees comprised most of the diet, being present in 76%, of stomachs. Succulent plants were relatively scarce in the dry season, espe- cially in the open and barren roadside sit- uations where many ctenosaurs were found. Coarse, seral weeds were often eaten, as they were the most available source of plant food. Many of the lizards living along the edges of cotton fields ate leaves of cotton plants (Gossypium). Leaf buds and newly grown tender leaves of trees also made up much of the foliage eaten. A tree about 10 m high in a road- side strip north of Chinandega had lost its foliage when scorched by a grass fire, and new leaf buds were opening at the time of our field work in March 1976. These new leaves seemed especially attractive to cteno- saurs, as we saw several in the tree and shot two, but on the return trip later the same day we saw four, of different sizes, in the tree. Nearly all flower material in the 146 EcoLtocy AND EXPLOITATION OF CTENOSAURA SIMILIS 489 TABLE 1 Foop oF CTENOSAURS; CONTENTS OF 146 STOMACHS FROM WeEsTERN NICARAGUA FesruAry, Marcu anp Aprit 1976 Misc. INvERTE- VERTE- Leaves Flowers Fruits Sreps PLANTs BRATES BRATES FECES Total samples (146) Occurrences ......---- Il 86 D2 11 7. 29 11 1 °% sample weight... 47.1 26.2 9.4 4.0 23 3.0 4.1 ei) Adults (73) Occurrences .......-.. 54 45 9 9 5 8 5 1 °% sample weight .. 45.2 232 10.5 Dye) 3.2 v8 3.8 6.5 Young (73) Occurrences .......... 56 42 13 2 1 21 6 ait % sample weight .. 49.0 31.0 8.0 Ail iS) 4.8 5.0 es stomachs was from trees, of kinds which bloom when they are leafless at the height of the dry season. Most were legumes and were yellow, pink or white in that order of frequency. Trees with either flowers or tender foliage sometimes lured ctenosaurs beyond their usual foraging radii. Miscel- laneous plant material in the stomachs in- cluded stems, and some material too much digested to be recognized in any of the main categories. Vertebrate prey included small cricetid rodents, Oryzomys, Scotinomys (five stom- achs), small lizards probably Sceloporus and/or Cnemidophorus (in two), a partly digested tail of a small ctenosaur (in one) and 9 small lizard eggs probably of Scelop- orus or Cnemidophorus, or possibly Ameiva (in one). One adult ctenosaur had a ctenosaur egg, still undigested and with shell intact in its hind gut and another had eaten three such eggs. The presence of such active prey as mice and lizards, and presence of eggs that would have been deposited in burrows, indicate that some of the prey is obtained by dig- ging. Cannibalism of sorts is indicated by the eating of ctenosaur eggs and tail. Hender- son (1973) noted that habitat separation relieves the young from predation by adults. Newly emerged hatchlings may overlap the adults’ habitat more than do the scansorial juveniles a few weeks later. However, the active hatchlings are so swift and shy that most apparently escape preda- tory adults. Young that are several months old become less scansorial and more like adults in habits. They have doubled or even tripled in length, and are larger than the food objects that adults normally take. First-year young that lived in the areas controlled by dominant males occupied lookouts only a few meters from them, but instantly fled when the adults moved to- ward them. Invertebrates found in stomachs in- cluded 2 lycosid spiders, beetles, butterflies, moths, wasps, bees, dipteran flies, hemip- teran “bug,” leafhoppers and a grasshopper. Some of the smaller insects made up in- significant proportions of the stomach con- tents, and probably were ingested secon- darily on flowers or other vegetation that the ctenosaurs were eating. Invertebrates were less prominent in the food of the adults than in that of the first-year (8-10 months old) young, although these young were much like adults in habitat and be- havior. Younger, smaller ctenosaurs are much more insectivorous. Near Pisté, Yucatan, 490 Tue University oF Kansas Sc1eENCE BULLETIN TABEE, 2 REPRODUCTIVE STATES OF ADULT FEMALE Crenosaura similis IN Fepruary, Marcu AND Aprix, 1976 PERCENTAGES OF FEMALES HAVING: PERCENTAGES OF FEMALES HAVING: OVIDUCAL MEAN SIZES OF OVARIAN EGGS OR FOLLICLES IN INDIVIDUAL OVARIAN OVIDUCAL LARGE RECENTLY FEMALES FOLLICLES VS. EGGS N FOLLICLES VS. LAID N 1-10 Feb. 1.5,4,5,7.5,8,9,10(in 3) 90 10 20 100 0 96 12(in 6),15 17,18 Mle ZO ebsa py) i atetees cee ees = eee ms az 21 Feb-1 Mar. 5,11,14,16,16 27 73 2 100 0 90 2-11 Mar. 153 14 86 7 100 0 22 PED IMiars de ee ee cee 93 7 160 ISSN TG) nn oa eee ne eer 78 22 32 IONS Caen aenamreites ; 6 94 16 ZO A rl as Chace. nchacisss a teeee ee 100 13 JESU dao) o1 ho i BL eeeae Raceeeen cone c an 100 11 Maslin (1963) found stomachs of several juveniles (55-64 mm) filled with insects including fairly large grasshoppers. Allen Porter collected 16 young that probably averaged about 3.5 weeks old (57-80 mm) near Laguna Asososca, Leon Province, Nicaragua, 28 July 1976, and 15 of the stomachs had insect remains: 23 beetles (scarabaeid, chrysomelid, elaterid, coccinel- lid), 6 lepidopteran larvae, 3 leafhoppers, 3 ants, 1 gryllid cricket, 1 butterfly and 1 beetle grub. For 32 of the prey items suf- ficiently intact to be measured or estimated, average length was approximately 10 mm, and prey weight averaged a little less than 09 g. These juvenile ctenosaurs weighed a little less than 10 g on the average, hence prey weight was most often less than 1 per cent of body weight. Although 60° of the items were beetles, some of the prey was of active, swift-moving kinds. Ten of the 16 stomachs had plant material (foliage in 9, yellow flowers in 1, berries in 1) and 59.0% of the food weight was vegetation vs: 41.07, 1msect prey. In gardens and cultivated fields cteno- saurs damage plants by nipping off buds, flowers, fruiting heads and tender leaves. Young bean plants are especially liked. Damage may be substantial, and as a re- sult ctenosaurs are generally considered pests and are killed at every opportunity by agriculturalists. Doubtless at some times and places control measures are justi- fied, but harm done to crops seems fairly trivial, weighed against the benefits accru- ing from utilization of the ctenosaur for food and sport. Reproduction: Like most other iguanids, Ctenosaura similis is oviparous, laying a single clutch of eggs, annually. Early in the dry season in January and February fat bodies are large and ovarian follicles grow rapidly. Follicles are ovulated when they have attained a diameter of 15-18 mm. Table 2 indicates that in the first 10 days of February, 2 of 20 females examined had already ovulated, and 18 contained en- larging follicles, but in 15 of this group follicles were still not mature. In the last 9 days of February (and 1 March) 16 of 27 females had ovulated, 2 others had ma- ture follicles, and three had follicles that were still short of mature size. The last recorded as still having follicles was ex- Ecotocy AND ExPLOITATION OF CTENOSAURA SIMILIS 49] amined on 10 March. In the last 10 days of March, 7 females dissected had recently oviposited and 25 still had mature oviducal eggs, but all of the 16 dissected in early April were already spent. These records indicate that in 1976 in western Nicaragua, at least, the peak of ovulation occurred in mid-February, and the peak of ovipositing was in late March. Hence, eggs were re- tained in the oviducts for approximately 5 weeks. Earliest and latest dates recorded for females with oviducal eggs were 4 February and 2 April, respectively. Indi- viduals vary 3-4 weeks in the timing of their reproduction, but Ctenosaura similis has a much more concentrated breeding season than any other Middle American lizard that has been studied. Number of eggs per clutch, determined from enlarged follicles and from oviducal eggs in unlaid clutches of the females dis- sected in February and March 1976, aver- aged 43.4 (12-88, N=69). The maximum divided into 8 size-classes which are be- lieved to be the most plausible bases for cohorts in successive annual age-classes, taking into account the known growth rates of marked individuals and the fact that each annual cohort in a stable natural population is somewhat less numerous than the next younger group. Within each of the female size-classes in Table 3, the clutch size is much less variable than for the group as a whole, and the mean in- creases from approximately 22 eggs per clutch in the smallest (two-year-old primi- parae) to approximately 70 in the largest adults (8-year olds). The smallest clutch recorded (12) be- longed to the smallest ovigerous female of 191 mm snout-vent (s-v) and the largest clutch (88) belonged to one of the four larger females dissected (338 mm s-y). However, a few females deviated markedly from the average clutch size for their size- group, indicating that factors other than TABLE 3 Aputt FeMAte Crenosaurs THAT Were MeasureD, GROUPED IN S1zE-CLAssEs THAT CORRE- SPOND WITH PossiBLE AGrE-CLAssEs, SHOWING CorRELATION OF CLUTCH SIZE WITH SIZE OF FEMALE S1zE-CLass NuMBER MM S-V E925] aetna eee enna meee POLE 63 DA ASTS ee tN RE nO) pee De 51 BINS eY BAN So eee een ee aE OE 46 in 0 area eres et Ae sl SSUES INOS 25 se care enn ion ee ee 37, Si ES RAC CS aa a eae en ese rece ve 25 SVU 7i8 | wei ee es ce ae Me 17 SCs oY UP Toren eee salen ee eee Oe SA 3 of 88 distinguishes C. similis as one of the more prolific of all lizards, even more so than viviparous species. The wide range of variation between females in numbers of eggs is especially remarkable, and is strong- ly correlated with age and size of the indi- vidual. In Table 3, reproductive females are MEASURED AGE (YEARS) CLUTCH SIZE (MEANS AND EXTREMES) EsTIMATED 2 21.688 1.164 (12-27 in 16) 36.000 + 1.414 (29-41in 8) 40.900 + 3.321 (29-63 in 10) 42.428 + 2.852 (29-51in 7) 51.571 + 2.927 (30-66 in 14) 61.833 + 3.534 (49-74 in 6) 69.500 + 3.359 (59-88 in 8) WO CON WN VU BW body size may strongly influence the num- ber of eggs. For 40 females in which weight of un- laid oviducal eggs as well as total weight was recorded, clutch weight averaged 21.65%, of the total (15.5-36.99,,). For ithe different size-groups of females, averages approximated that of the entire group, with 492 Tue University oF Kansas SCIENCE BULLETIN no discernible trend toward greater repro- ductive effort in larger females. In the ctenosaur, the “reproductive strategy” as defined by Tinkle, Wilbur and Tilley (1970) differs from strategies of most tropi- cal lizards (but resembles that of various Temperate Zone lizards) in delayed ma- turity and in the production of a single annual clutch. It differs from most other lizards, but resembles some other giant iguanids, tetids, and varanids of tropical mainlands in its extremely large clutch, and in low ratio of egg- and hatchling-size to adult. Fat deposits in the posterior ends of the abdomens were bright yellow and were especially conspicuous in females that had not yet ovulated. In 15 of these females abdominal fat bodies ranged in weight from 4 to 54 grams (.5 to 6.5 per cent of the total weight), with a tendency for those with smallest follicles to have the largest fat bodies. Evidently the fat bodies provide much of the sustenance that per- mits rapid enlargement of the follicles as yolk is deposited. In three females with approximately full-sized follicles (16-17 mm in diameter), fat bodies averaged 5 per cent of total weight; in nine females with follicles 12-15 mm in diameter, fat bodies averaged 48 per cent of the follicles’ weight; and in two females having follicles 10-11 mm in diameter, the fat bodies aver- aged 169 per cent of the follicles’ weight. In females that had ovulated, fat bodies were much shrunken, and in most in- stances weighed less than one gram. Fig- ure 1 shows the relative weights of the fat bodies and the ovaries in females with enlarging follicles. Growth: Typical hatchlings are between 55 and 60 mm snout-vent, but some are slightly larger or smaller. As indicated, individual females vary from three to four weeks in the time of laying, so that the earliest hatchlings are as much as 25°% longer than their original length by the time the latest appear. In older young, the size difference increases as some make better growth than others. The largest in each of 18 series of 9 to 51 young (Tables 4 and 5) averaged 138.5°% of the lengths of the smallest. In series that average more than a month old, the largest may be 50°/ or more larger than the smallest. Early growth is best shown by ten series captured at four localities in northwestern Costa Rica in 1968 and early 1969 (Table 4). There were five successive samplings at Playas del Coco and three at Boca de Barranca. Evidently there was some differ- ence between these localities in hatching time, from the first week of May (Boca de Barranca) to the second week (Playas del Coco) or third week (Finca Taboga, Que- pos). Hatchlings were first seen on 1 May at Quepos. A single hatchling was seen on 14 May at Taboga; seemingly hatching was just beginning there. Average length gain s-v is usually approximately 4 mm per week or a little less for the first few weeks, but then slows to somewhat less than 3.5 mm per week. Six young marked in late May and re- captured after intervals of 31 to 36 days, made an average gain of only 2.1 mm per week, probably due to temporary stunting from the adverse effects of toe-clipping. Two other young were recaptured after longer intervals, when they had fully re- covered from the adverse effects of mark- ing. From 1 July 1968 to 12 February 1969 one grew from 86 to 143 mm and from 26 August 1968 to 4 February 1969 the other grew from 95 to 170 mm. These gains in- dicate average growth rates of 1.76 and 3.39 mm per week, respectively, over periods of 32 weeks and 23 weeks. The lizards of the four Costa Rican lo- calities listed in Table 4 range from 9°25’ to 10°35’ North Latitude and have breed- ing schedules several weeks advanced over those occurring in more northern parts of EcoLtocy AND EXPLOITATION OF CTENOSAURA SIMILIS 493 TABLE 4 SIZES OF JUVENILE CTENoSAURS ON VARious DATEs Ar Costa Rican Loca.ities, SHowING EARLY GrowTH MEAN LENGTH S-V EsTIMATED AVERAGE GAIN EsTIMATED (MM.) PER WEEK Date LocaLity (u.) N AGE (WEEKS) SINCE HATCHING 20-23 May Playas del Coco 59.1 (56-63) | 4 27-30 May Boca de Barranca 70.9 (57-82) 5] 5}5) 4.2 coe Quepos 76.1 (60-90) 19 5 3.4 29-30 June Boca de Barranca 83.9 (63-109) 30 8 3.8 7 July Finca Taboga 87.2 (73-107) 19 7 45 9-10 July Playas del Coco 78.1 (63-97) 21 7 33 14-17 Aug. Playas del Coco 95.5 (78-107) 9 12 333 20-30 Aug. Boca de Barranca 103.9 (80-114) 19 14 33 23-26 Aug. Playas del Coco 104.1 (84-127) 15 13 33 4-7Feb. Playas del Coco 147.5 (120-165) 12 35 2.6 the range. Henderson (1973) found that the young first appear in mid-June in Be- lize. Table 5 shows average sizes of series of young collected in various parts of the range, with series arranged in order from latest to earliest calculated hatching dates. The data indicate. that “in; some areas hatching may be delayed as much as two months beyond that characteristic of north- western Costa Rica in lowland areas. The TABLES Sizes oF First-YEAR Ctenosaura similis at VArtous Locavitries AND Dates, SHowING RETARDA- TION OF SCHEDULES NORTHWARD AND AT HIGHER ELEVATIONS. EsTIMATED MEAN AVERAGE LaTITUDE & LENGTH DATE OF HATCHING LocaLity ALTITUDE* S-V Rance N COLLECTION DATE Tenorio, Guanacaste, Bostae Rica ee sees le 10°37’(650m) = (77.25 = (72-87) 4 8-23-52 18 July Tilaran, Guanacaste, WostauRicayis <2. ee hatin NOS285( 5621) meee lcO Geeta 1 8-15-54 11 July Piste, Yucatan; Mex, ...2.... 20°44’ 63.0 (53-71) 22 7-20&21-62 7 July Isla de Ometepe, 7-30: Ieeode Nice. 11°32’ Via 6292), ale pee eae 3 July Lago Asososca, Leén, Nic.11°26’ 70.4 (57-80) 16 7-28-76 3 July listavde Matz, Nic. ..2.22......-. 122077 Sy AS) (60-75) 11 6-29&30-64 19 June 14 km E Rivas, EVASOINIG) tens eee eee! [L267 80.6 (88-75) 11 7-20-64 9 June Managua, INice 82 12°08’ 66.8 (60-80) 9 6-20&21-56 3 June Managua, Nic: 2202 12°08’ 54.2 (49-59) 9 6-3&4-56 1 June S. Antonio, Chinandega, iG. eee cae es 122 EME (63-94) 23 7-5 to 11-64 1 June * Near sea level, except where otherwise indicated. 494 Tue UNIVERSITY OF PERCENTAGE OF BODY WEIGHT on ro) ° me — > a] ™m a) = = Ww ul LHOSISM AGOg 4O S9NVLN39YSd Ol a Pies a. PERCENTAGE OF SAMPLE Kansas ScIENCE BULLETIN BELIZE STUDY AREA N=49 NICARAGUA SHOT SAMPLE NICARAGUA N= 342 SIGHTED SAMPLE 200 300 SNOUT VENT LENGTH MILLIMETERS 100 400 Relative weights of fat bodies (open columns) and complements of ovarian follicles (solid col- umns) in 15 female ctenosaurs having their follicles in different stages of growth. ErG: 2: Size-classes in three population samples of Ctenosaura similis; stippled columns represent males, cross-hatched columns represent females and shaded columns represent individuals of undetermined sex. later dates were for Tenorio and Tilaran, Costa Rica, near the localities of Table 4 but in the relatively cool climates of mon- tane areas. The next were those from Yu- catin 1070 km north of the Costa Rican localities. Isla de Ometepe in Lago de Nicaragua and Isla de Maiz in the Carib- bean, having relatively cool climates mod- erated by surrounding water, are also late. Hatching at various localities in western Nicaragua (Rivas, Isla de Ometepe, Ma- nagua, Lago de Asososca, San Antonio) is 2 to 6 weeks delayed beyond the time of hatching in the Guanacaste lowlands. Although the annual schedule of repro- duction and growth differs by only a few weeks in different parts of the range, the The species thrives best in the relatively hot and sea- differences may be critical. sonally dry climates of western Nicaragua and northwestern Costa Rica. with In regions more precipitation, less insolation, and lower air temperature, there is a cumu- lative retardation; eggs mature later, lay- ing is delayed, incubanon requires more time, and the young make less rapid and consistent growth. Of 16 presumed second-year females, 204-250 mm s-v, collected in western Nica- ragua from 6 February to 22 March 1976, 13 were reproductive, having enlarged fol- licles (4 females), oviducal eggs (7 fe- males), or corpora lutea and enlarged ovi- ducts indicative of recent egg-laying (2 females). The remaining 3, having lengths of 220, 215 and 209 mm s-y, had minute ova and small oviducts. Thus, 18.79% of fe- males in the two-year old size-class were non-reproductive. In other parts of the range where climate is less favorable, post- ponement of sexual maturity until the third year may be the mode, with conse- quent major loss of reproductive potential. In western Nicaragua, 6 February to 5 Eco.tocy AND ExPLoITATION OF CTENOSAURA SIMILIS 495 April 1976, first-year young of 170-189 mm length (females) and 190-197 mm (males) were more numerous than larger or smaller young and probably were modal for their age-group of about 8-10 months. Average growth rates (s-v) from the time of hatch- ing are calculated at about 3.1 mm per week for females and about 3.5 mm for males. However, some young, presumably of approximately the same age, were still as small as 127 mm (female) and 135 mm (male), indicating a wide range among individuals. First-year and second-year young apparently overlap in size mainly in the range 190-200 mm for females and 235- 250 mm for males. If 200 mm is considered the upper limit for first-year females (in our sample of 160), 53 fall in this age-class, with a mean length of 171 mm. For males, a criterion of maturity that may separate second-year adolescents from first-year young is length of spines of the dorsal crest. These spines are long and prominent in adult males, but low and flattened in females and juveniles. Individual males of 238, 241 and 248 mm and all that were smaller had short spines (2.5 mm or less), whereas individuals of 239, 240 and 247 mm, and all but one that were larger, had relatively long spines (5 mm or more). If males up to, but not including, 240 mm are considered first-year young, the sample in- cludes 39 of them which average 188 mm. At the end of the first year, males aver- age aproximately 10 per cent larger than females, and as adolescents and adults they grow about twice as fast as females, judg- ing from maximum lengths s-v of 489 (male) and 347 (female). As in most rep- tiles, growth continues throughout life with allometric changes in proportions. Head shape, especially, is subject to pro- gressive change in males. With advanced age their jaws become elongated and heads are widened posteriorly (Fitch and Hen- derson, 1977). Population Density: In the cemetery at Belize City, covering approximately 9.66 hectares, we recorded 49 ctenosaurs with total biomass calculated at 16.13 kg (1.67 kg per ha), adult females making up 45.2%, adult males 40.89% and first-year young 14.0%. Although the cemetery superficially seemed to provide a fairly uniform habitat, the lizards were distributed with obvious clumping. As there was a surplus of food, the most obvious factor controlling distri- bution was relative age of gravestones in different parts of the cemetery. Where a high proportion of the stones were old, weathered, cracked, or partially collapsed, there were more hiding places for the ctenosaurs than there were where the stones were relatively new. Groups of graves that were enclosed by iron fences also seemed to provide security that made them especially attractive. There were four solitary individuals (subadult male, first- year young and two adult females) and 11 associations of from two to 10 individuals. The largest associations, with 10, eight and seven individuals, each had adults of both sexes and young. In associations of four, all were young. Associations of three had all young in one instance and had a pair of adults and a juvenile in another. Associa- tions with only two lizards consisted of female and young in two instances, both young in two instances, subadult male and young in another. Spacing between neighboring groups, or between groups and solitary individuals, averaged 74.5 m; within groups spacing of individuals averaged 25 m. Except in the case of consort pairs, it is doubtful whether individuals were attracted to others and the clumping observed may have been a result of the unevenly distributed resources, especially suitable shelter. Taylor (1956) wrote that a colony of 10 or more ctenosaurs might be based at one large tree with hollow trunk or limbs. We found such a concentration would be un- 496 Tur University oF Kansas SCIENCE BULLETIN usual and would consist largely of imma- tures. An adult male and female often remain in close proximity for periods of days; otherwise individuals tend to be well spaced. In Nicaragua, where ctenosaurs are ex- tensively hunted, population densities com- parable to that in the Belize City cemetery are unusual, but even higher concentra- tions were observed along roadsides, field edges and gullies in Chinandega Province north of Chinandega and along the Hon- duran border northeast of Palo Grande. In Costa Rica, similarly dense populations were observed at Finca Taboga in Guana- caste Province. Because egg clutches are large, juveniles sometimes occur locally in high population densities. They tend to maintain spacing; aggressive displays and fights hasten their dispersal after hatching. Population Structure: Ctenosaura similts normally requires nearly two years to at- tain sexual maturity, and individuals may survive and continue to grow for many years subsequently. A natural population therefore consists of many discrete annual age-classes, each larger in size and less numerous than the next younger group. Broods are large, but early mortality is heavy. Ratios of juveniles to older individ- uals are distorted by behavioral differences. Young that are several months old have become much like adults in behavior and habitat, and their cohorts have already sus- tained their heaviest losses. Three separate population samples were obtained by different methods in the period February to April 1976, when first-year young were mostly in the age-range 8-10 months and had grown in linear dimen- sions to approximately three times their original size, and to approximately half the size of adults. These three samples (Fig. 2) show important similarities and differences, the latter probably reflecting biases inherent in the sampling techniques. A sample of 160 ctenosaurs was ob- tained by hunting with a .22 rifle from a vehicle on roads and highways of Chinan- dega, Leén, Managua, and Chontales Prov- inces in Nicaragua. We attempted to ob- tain a random sample. However, there is a possibility of bias resulting from greater wariness of old individuals and/or from more persistent late afternoon activity in yearlings and gravid females. Numbers and percentages of different classes in the sample were as follows: first-year young 86 (53.6°%); probable second-year adoles- cents 49 (30.6%); large adult males 7 (4.49%); large adult females 18 (11.3%). There were 32 males (135-220 mm) and 54 females (127-200 mm) in the group of first year-young, and 14 males (221-298 mm) and 35 females (201-250 mm) in the sup- posed second-year group. The adult males ranged from 299-400 mm, adult females from 253-295 mm. It is noteworthy that females constituted two-thirds of the total sample, and were from 63 to 72 per cent in different age-classes. Behavioral differ- ences between adult males and females might cause some bias, but in first-year young sexual differences in behavior are small, if they exist at all, and would not result in significant bias. We therefore conclude that the sex ratio is strongly skewed in favor of females from the time of hatching. A larger sample of 342 ctenosaurs was tallied from a vehicle during drives in the same parts of Nicaragua involved in the preceding sample. These animals were not handled; snout-vent lengths were merely estimated, and usually sex was not deter- mined. Sometimes estimates were made simultaneously by two or three persons and a compromise figure was agreed upon. Often the same animal was first estimated and then shot and measured, providing opportunities for adjustment and _refine- ment of the estimates. However, accuracy EcoLocy AND EXPLOITATION OF CTENOSAURA SIMILIS 497 varied and for those only glimpsed briefly the range of error must have been rela- tively large. Large adults were better rep- resented in this “sighted” sample, and first- year young were only 37.5% (vs. 53.6 in the “shot” sample). A third population sample consisted of the 49 ctenosaurs observed at the Belize City cemetery 2-7 March 1976. Although only one was captured and measured, size estimates were made by observing them at close range and were checked repeatedly for most of them. Sex was readily distin- guished in the full-grown lizards and tenta- tively distinguished in second-year adoles- cents (by head shape proportions, and elongated spine-like scales of the dorsal crest in the males) but could not be dis- tinguished in first-year young. The latter averaged markedly smaller than first-year young in western Nicaragua at the same time of year, and unlike those young they did not overlap the second-year size-class. Of the 49 total, 27 (55.0%) were first-year young. Of the remaining 22, 8 (36.3%) were males and 14 (63.5%) were females. Five of the 8 males were large (320-400 mm) dominant individuals. Figure 2 compares distribution of size- classes (with 25 mm interval) in each of the three samples. It shows that there is a continuum in size from the smallest first- year young to the largest adult males. Main concentrations are those of the first year young (150-200 mm in Nicaragua, but smaller in Belize) and of young adults (mostly females) 226-250 mm. EXPLOITATION The ctenosaur has delicate, tasty, white meat and in México and Central America it is much sought as an article of food. It is normally preferred over the iguana (Iguana iguana) where both occur to- gether. However, in Belize only iguanas are eaten. In that region ctenosaurs are highly localized and because they are much in evidence in the Belize City cemetery, are regarded with a superstitious aversion. Called “wish-willys” by the English-speak- ing black Creoles, these reptiles are be- lieved to feed upon corpses in the graves. Elsewhere the species has probably been used for food since pre-Columbian times, but degree of exploitation varies locally. In parts of western Costa Rica where the ctenosaur is abundant, it is subject to little hunting, but in Nicaragua, El Salvador and Honduras it is intensively exploited and local populations have rapidly dwin- dled. We questioned many campesinos in western Nicaragua to obtain information concerning the ume and amount of reduc- tion. There was almost universal agree- ment that drastic reduction had occurred because of overhunting, but there was much difference of opinion, even in the same general area about the time of reduc- tion. Many of the older campesinos who were questioned said that ctenosaurs for- merly were numerous, but had become so scarce that now they are seldom hunted or eaten. Among 21 informants who had definite impressions as to when reductions had occurred, figures varied from 1 to 30 years, but 3 years was the most frequent estimate (6 instances) and the average was approximately 8 years. Of 87 people questioned, 70 (80°) said they ate ctenosaurs regularly or occasion- ally. Most of these individuals or members of their families obtained the animals by hunting; a few bought their ctenosaurs from the hunter. Forty-nine persons made statements about the number of ctenosaurs eaten per week, which averaged 4.75. How- ever, 17 other persons said they did not eat the meat, and 16 others who said they did (or had in the past) were vague about the amount because their use was occasional, rare or highly seasonal. Although our sample of interviews is small, and is doubt- less biased in various ways it does indicate some general trends and suggests harvest 498 Tue University oF KANSAS SCIENCE BULLETIN on an enormous scale. Over extensive areas of western Nicaragua the majority of cam- pesino families serve ctenosaur meat once to several times weekly, and this flesh is an important protein supplement to diets that otherwise tend to be high in starch and meager in quantity. Hunting ctenosaurs is considered sport and the species qualifies as a game animal. Campesinos were questioned about their methods of capture. In order of frequency mentioned, the usual methods were: with a trained dog (52), with a noose placed at the burrow entrance or manipulated on the end of a long stick (24), with a slingshot (24), with a small bore rifle or pistol (9), and with a digging stick or shovel (6). Most informants mentioned 2 or more of these methods. For those that mentioned only one method, the order and number were: dog (12), slingshot (3), digging (2), gun (2). Sunday hunting is custom- ary, especially for persons who work dur- ing the week, as field hands on fincas; it is often the main outdoor recreation. In some areas, campesinos from early childhood develop a familiarity with cteno- saurs and the techniques of taking them. Certain families and even communities specialize in hunting the lizards, make it their main occupation and supply the mar- ket places. The villages of Palo Grande, Somotillo and Villa Nueva near the Hon- duran border have many hunters who are the main suppliers of the market in Chi- nandega and contribute to the stocks in Leon, Managua, Masaya and Granada. Several other villages in northwestern Nic- aragua are also important suppliers: So- moto (for Chinandega), and Rota and El Sauce (for Leén). Most of the ctenosaurs sold in the Mercado Oriental of Managua were said to have come from the village of San Francisco de Carnicera on the north- eastern shore of Lake Managua. However, the ctenosaurs shipped from San Francisco are assembled by a dealer there from the neighborhood of Cuatros Palos and other outlying villages farther north and east, and represent the combined efforts of ap- proximately 20 hunters. Professional hunt- ers are especially skilled, not only in find- ing and capturing the animals, but in tak- ing them alive and intact. One hunter demonstrated his technique, with a long bamboo pole having a cord noose at the end, and a bait of calf liver suspended in front of it. He told us that with his partner he left home for as much as 3 days at a time on extended hunts to areas several kilometers away, where the lizards were still relatively numerous, and both men returned heavily loaded with sacks of live ctenosaurs. A few dozen skilled hunters are the main suppliers of the market in the larger Nicaraguan cities. They hunt on foot and their activities are concentrated in relatively small areas where ctenosaurs are abundant. Each hunter captures dozens of the animals weekly, but some men limit their hunting to the lizards’ reproductive period, Decem- ber through April, and seek other means of livelihood for the remainder of the year. In February and March 1976, there were usually 1 to 5 vendors selling ctenosaurs at each major mercado in Nacaragua. Occa- sionally a vendor had between 100 and 200 ctenosaurs at one time, but usually the stock was much less— sometimes only one. The vendor normally replenished his stock with a new shipment once or twice a week. Often the shipment consisted of a mixed lot of ctenosaurs and iguanas. Some of the animals were kept in gunny sacks, while others were displayed in large wicker bas- kets, or were strewn over the sidewalk, immobolized by having their feet bound behind them. The live animals attracted considerable attention from passing crowds. When a new lot was displayed, prospective buyers gathered to examine and handle the animals, with brisk trading favoring egg-bearing females and _ large, EcoLocy AND ExPLoITATION OF CTENOSAURA SIMILIS 499 fleshy males. At the two mercados in Leon, live ctenosaurs were usually absent or were displayed in small numbers, but the dressed carcasses were sold at several indoor stalls. Piles of several dozen carcasses were on display on most occasions. We were told that the butchering was done at the home of the vendor, in pre-dawn hours before the mercado opened. At other mercados dressed ctenosaurs were seen in small num- bers, but most of the animals were sold alive. From the numbers of ctenosaurs appear- ing at Nicaraguan mercados, and from the statements of vendors, the aggregate week- ly consumption must total hundreds of ani- mals. Chinandega, Leon, Masaya, Mana- gua, and Granada rank in about that order in numbers consumed. Many other small towns and villages participate in the cteno- saur trade, but on a relatively small scale. Some of the villages in northwestern Nica- ragua that supply major mercados export part of their stock to El Salvador. Trucks with shipments of hundreds cross the bor- der into Honduras from Somotillo and El Espino, Nicaragua, and hundreds more are sent by ferry from Potost directly to El Salvador at Puerta Amapala. Nevertheless, the commercial harvest seems to be much less than the aggregate harvest by individ- uals hunting for family subsistence. Live ctenosaurs were sold for food in 1976 at prices ranging from 1.5 Cordobas to 15 Cordobas (1 C = $0.14 US). The lowest prices prevailed in the villages of north- western Nicaragua. There the hunters usu- ally received 2 or 2.50 C apiece for cteno- saurs and sold them in dozen lots. In other parts of the country where the animals were less common, the hunters usually re- ceived 3 or 4 C apiece for them. The buyer usually paid 5 C at the Chinandega mer- cado, 8 (7-10) at Managua and 8-12 at Masaya and Granada. At the latter two cities, there was often an adjustment of price according to the animal’s size, where- as at Chinandega the price tended to be uniform even though one animal might be as much as 5 times the bulk of another. DISCUSSION Ctenosaura similis is unique among American lizards in several aspects of its ecology: It has become adapted as an adult to feed upon the dominant vegetation, with the result that food supply is not generally limiting. It produces remarkably large egg clutches (mean 43.4, maximum at least 88). It is single-brooded with eggs laid late in the dry season and young appearing early in the rainy season. Its hatchlings are remarkbaly small compared with adults (about 16 per cent of female length and one per cent of female weight). Its fe- males are much smaller than males (80% of male length and 55.6% of male weight) and are more numerous both in samples of adults and in first-year young in a ratio of about 2 to 1. Sexual maturity is attained late in the second year, and the two-year- old primiparae made up 23.2% of the breeding females and produced 10.6% of the eggs in a sample. Although the ctenosaur is “K-selected” (see Pianka, 1970) in having a long life expectancy, postponing maturity tll the second year, and producing a single annual brood, it is “r-selected” in its remarkably large clutch, in relatively small size of its eggs and hatchlings, and in its adaptation to disturbed and seral habitats. All these “~ selected” traits reduce its vulnerability to exploitation by humans and confer po- tential as a successful game animal. Over extensive areas of xeric habitat where the original fauna has been depleted by man- made changes, involving virtual elimina- tion of the birds and mammals that were favorite game animals, the ctenosaur con- tinues to thrive. Heavy grazing, or clearing of the land for cultivation, favor its sur- vival and increase if there are certain essen- 500 Tue University oF Kansas ScIENCE BULLETIN tial habitat features that assure adequate shelter and a year-round food supply. Despite its high reproductive potential and capacity to withstand hunting pres- sure, the ctenosaur has had its populations reduced at an accelerating rate in recent years. Mushrooming human _ populations have resulted in ever-increasing hunting pressure, with hunting intensified in the season of reproduction and concentrated on the gravid females, the least expendable part of the population. The reduction that has already occurred must have involved annual loss of hundreds of tons of high grade protein food in Nicaragua alone, with the prospect of increasing losses until the yield becomes negligible. Management practices that will reverse the trend are acutely needed. It would be easy to suggest measures that would preserve remaining populations and permit their increase, but it is much more difficult to make practical recom- mendations. Exploitation of the ctenosaur is deeply rooted in tradition, whereas the concept of conservation is foreign to the exploiters. In the face of want, campesinos will not readily relinquish their presumed right to hunt ctenosaurs for food or to harvest the gravid females that should be left to replenish the population. Elimina- tion of this potentially valuable species is deplorable and unnecessary. There are slight grounds for optimism in the fact that rural people almost universally recog- nize that ctenosaurs are rapidly becoming scarcer and that this decline results from overhunting. EITERATORE: Chie ALVAREZ DEL Toro, M. 1960. Los reptiles de Chia- pas. Inst. Zool. del Estado Tuxtla Gutierrez, Chiapas. 204 pp. Bratr, W. F. 1960. The rusty lizard, a population study. University of Texas Press, Austin, xvi + 185 pp. CarpPENTER, C. C. 1965. Aggression and social struc- ture in iguanid lizards, pp. 87-105 im Lizard ecology a symposium. University of Missouri Press, Columbia. Dice, L. R. 1952. Natural communities. University of Michigan Press, Ann Arbor, x + 547 pp. Evans, L. T. 1951. Field study of the social behavior of the black lizard, Ctenosaura pectinata. Amer. Mus. Novit., No. 1493:1-26. Fircn, H. S. 1956. An ecological study of the col- lared lizard, Crotaphytus collaris. Univ. Kan- sas Publ. Mus. Nat. Hist., 8:213-274. —. 1970. Reproductive cycles of lizards and snakes. Univ. Kansas Mus. Nat. Hist. Misc. Publ. No. 52, pp. 1-247. . 1973. A field study of Costa Rican Lizards. Univ. Kansas Sci. Bull. 50(2):39-126. Fircu, H. S., A. V. Fircu anp C. W. Fitcu. 1971. Ecological notes on some common lizards of southwestern Mexico and Central America. Southwestern Nat. 15(3):397-399. Fircu, H. S. anp R. W. HENpveERson. 1977. Age and sex differences in the Ctenosaur (Ctenosaura similis). Contrib. Biol. and Geol. Milwaukee Pub. Mus.; 11:1-11. HeEnpERSON, R. W. 1973. Ethoecological observations of Ctenosaura similis (Sauria: Iguanidae) in British Honduras. Jour. Herp., 7(1):27-33. Masuin, T. P. 1963. Notes on a collection of herpe- tozoa from the Yucatan Peninsula of Mexico. Univ. Colorado Studies, Ser. Biol., 9:1-20. NEILL, W. T. anp R. ALLEN. 1959. Studies on the amphibians and reptiles of British Honduras. Publ. Res. Div. Ross Allen’s Reptile Inst. 2(1): 1-76. Pranka, E. R. 1970. On r and K selection. Ameri- can Naturalist, 100:592-597. TamsiTT, J. R. anp D. Vaxpivieso. 1963. The herpe- tofauna of the Caribbean Islands, San Andrés and Providéncia. Rev. Biol. Trop. Costa Rica, 11:131-139. Taytor, E. H. 1956. A review of the lizards of Costa Rica. Univ. Kansas Sci. Bull., 39(1): 3-332. TINKLE, D. W. 1967. The life and demography of the side-blotched lizard, Uta stansburiana. Misc. Publ. Mus. Zool. Univ. Mich., No. 132: 1-182. TinkLE, D. H., H. M. Wixzur anp S. G. TILLEy. 1970. Evolutionary strategies in lizard repro- duction. Evolution 24(1) :55-74. em ti. gf? ijt ts SeLegeteteteteetecesesececesesezesepeneseeteseteteteteteteteletecateiedeteteteteleteteteete ahaa aaa a ae eee case a cates ecetet THE UNIVERSITY OF KANSAS SCIENCE BULLETIN THE CLASSIFICATION OF HALICTINE BEES: TRIBES AND OLD WORLD NONPARASITIC GENERA WITH STRONG VENATION By Charles D. Michener Vol. 51, No. 16, pp. 501-538 November 14, 1978 ANNOUNCEMENT The University of Kansas Science Bulletin (continuation of the Kansas Uni- versity Quarterly) is an outlet for scholarly scientific investigations carried out at the University of Kansas or by University faculty and students. Since its incep- tion, volumes of the Bulletin have been variously issued as single bound volumes, as two or three multi-paper parts or as series of individual papers. Issuance is at irregular intervals, with each volume prior to volume 50 approximately 1000 pages in length. Effective with volume 50, page size has been enlarged, reducing the length of each volume to about 750 pages. The supply of all volumes of the Kansas University Quarterly is now ex- hausted. However, most volumes of the University of Kansas Science Bulletin are still available and are offered, in exchange for similar publications, to learned societies, colleges and universities and other institutions, or may be purchased at $20.00 per volume. Where some of these volumes were issued in parts, individual parts are priced at the rate of 2 cents per page. Current policy, initiated with volume 46, is to issue individual papers as published. Such separata may be purchased individually at the rate of 3 cents per page, with a minimum charge of $1.00 per separate. Subscriptions for forthcoming volumes may be entered at the rate of $20.00 per volume. All communications regarding exchanges, sales and subscriptions should be addressed to the ExcHance Liprarian, UNIVERSITY oF Kansas Lipraries, LAwreNcE, Kansas 66045. Reprints of individual papers for personal use by investigators are available gratis for most recent and many older issues of the Bulletin. Such requests should be directed to the author. The International Standard Serial Number of this publication is US ISSN 0022-8850. Editor Eugene C. Bovee Editorial Board William L. Bloom Philip W. Hedrick Rudolf Jander Harvey Lillywhite Charles D. Michener Norman A. Slade Henry D. Stone George W. Byers, Chairman THE UNIVERSITY OF KANSAS SCIENCE BULLETIN Vol. 51, No. 16, pp. 501-538 November 14, 1978 The Classification of Halictine Bees: Tribes and Old World Nonparasitic Genera with Strong Venation’ Cuar.es D. MIcHENER DEPARTMENTS OF ENTOMOLOGY AND OF SYSTEMATICS AND Eco.oecy, AND SNow ENTOMOLocICcAL MusEUM UNIVERSITY OF KANsas, LawreENce, Kansas 66045, U.S.A. TABLE OF CONTENTS PESTER JOC OVCTSI EG SINDEN Eee ee ape exe, eo eee cae er Pea oy mee et “TP rene STS UNE Coh of O° cea Oe ee ne orn a Key to the Old-World Non-parasitic Genera with Strong Apical Wing Venation .. CSU LEAS IG! skis AUER, SEES Ty SIR ete, eR re yO Ne le eee Mevatouthie: SUDSEnerasOl Rarcllapis <2... 7 ee 2 ee ee DEQ) CEALIGLES MING WARS WSOC INNS oi es cen nme Oe ce ene (RESTA EETION NN SU YX 1 DR ee a edn De RE PUMPS er SUL BSS OOS I UH aT LF 1 et A nD oA A ere RAID Df LADO IS, ATS TU eo at rs Be ce ae, et A Se Sen Sr MG TMUES eG VE GUE CTELG: ee aren we aces I te Te enc ace ssesnnsge te ee Kev stopie sSUDPCMELANOl LGCHYVAGICLUS a n25sc.c.2ctedecresstots cates sasatepe-tieeeze oe SUL RCN SEEAGGH] VN ALICE LEG marmite er, Re Ns ea Fe 3 8 eae Ses ass eck ee WD PCE OM GUCTUSATICW SUID CANIS ee eee ore ee EE Pee eee SENOS LO IC OL ACI TGE TIGA tS COD ans a ea SC are eee Pee RG CUITESEMI E/T ETICROSTO IN amy sa teene eee ato a 4! eran es eee spe ees Keyatothe supcencralor lkrinchostomai=.- 2 a So ee SUD EMUSRE OL NIT COSLONIG: pe ee can orig c eeare cred cestncee erent ree SUBSEMUS ENT INGHO SLOT Ae -ca) et at IE a eee ee eee Se CCM USE) 10 ORO 27 senor fare ss ee Nem ee ee Genus aCe sae eee ere cen ete <0 A DO a Pe ee Key topthie (SUD CMeLa Ol LGC G: eee oc as tann tereass teen far ooo t nace ee Sol Se mnlIGh Sele O Tic Mm cee. Rese e E Lc A Nil ds A ieee UF ee ee ee SUbSeMUSMIVEsHtOnAlictiuse 2) wesc ts ern. (2, SSS ere Eee ee eee Slee eu a (AH G77 ois te Se os oR OE oe Peterkin eer ee er ne ACKNOWLEDGEMENTS ....... ROPE Eh SIE Na de OD ee Ae ee RS es Ae a 502 Tue University oF Kansas SCIENCE BULLETIN ABSTRACT This study segregates and describes the tribes of the subfamily Halictinae. The Nomioidini is the most distinctive, but has not usually been recognized as a tribe. The Old World non-par- asitic Halictini with strong wing venation are revised to the subgeneric level. The recognized species are listed for revised faunas; otherwise trivial names are listed without indications of synonymies. American representatives of the Old World groups are included. The name Patel- lapis is resurrected for a large African group, divided into three subgenera, Lomatalictus n. subg., Chaetalictus n. subg., and Patellapis s. str. A related, large, African genus, Zonalictus n. g., is recognized. The primarily Oriental group Pachyhalictus is raised to the generic level and Dic- tyohalictus n. subg. is described for its African representative. Thrincohalictus is raised to the generic level. Only three subgenera of Thrinchostoma are recognized, Eothrincostoma, T hrin- chostoma s. str., and Diagonozus. For Halictus the usual three subgenera Seladonia, Vestitohalic- tus and Halictus are recognized. INTRODUCTION This work on the classification of sweat bees (Hymenoptera, Apoidea, Halictidae) was begun in order to provide a firmer basis for understanding halictine social evo- lution. The present paper is a segment in a larger study. When all the halictine groups have been included, a comprehen- sive account of the probable lines of descent and of the origins of sociality will be pre- pared. The parasitic halictine groups were treated earlier (Michener, 1978), so that the numerous special features of parasitic genera may be excluded from further con- sideration. The result is important short- ening of the descriptive material for the nonparasitic genera. The three halictine tribes are character- ized below. The genera of one of them, the Augochlorini, were revised by Eickwort (1969). The Nomioidini contains only a single genus and is treated below. The Halictini contains several nonparasitic gen- era. Two of them, possibly to be subdi- vided later, have the third and often the second transverse cubital veins and the sec- ond recurrent vein of the fore wing, at least in females, weakened relative to near- by veins. These genera, Homalictus and 1 Contribution number 1650 from the Department of Entomology, The University of Kansas, Lawrence, Kansas 66045, U.S.A. Lastoglossum (including Evylaeus), are ex- cluded from the present paper and will be treated later. Also excluded from this pa- per are a number of strictly American genera with strong distal wing venation. These are Agapostemon, Caenohalictus, Habralictus, Paragapostemon, Pseudaga- postemon, Rhinetula, and Ruizantheda. These genera are not closely related to those treated below and will be the topic of a later study by R. B. Roberts. The re- maining Halictini, those with strong wing venation found in the Old World, are the principal topic of the present paper. One such genus, Halictus, occurs also in the New World and its variations and species in New World are included. In the descriptive material, noteworthy characters and especially those unique to a group are italicized to facilitate rapid use. In the generic descriptions for the Halic- tini, the various areas or characters are numbered, to facilitate quick comparison of particular features among genera. The lists of species given for the various genera and subgenera are not exhaustive. They contain names of species of which I have seen authentic material, plus names added from the literature when descriptive information is adequate. Many species de- scribed in Halictus s.1., often with no indi- cation of group characters and sometimes compared to unrelated species that are now CLASSIFICATION OF HALICTINE BEES 503 in different genera, can be placed only by re-examination of type material. For areas such as Africa for which no revisional studies exist, all names which I have been able to place as to genus or sub- genus are included in the lists. For areas included in revisional studies or catalogues, synonymous names are excluded, often even when the published synonymy post- dates the revisional studies. Such cata- logues or revisions are those of Sandhouse (1941) and Michener (1951) for North America, Wille and Michener (1971) for the Neotropical region, and Bluthgen (1920, 1921519234. b; 1924) and Ebmer (1969, 1976b) for the Palearctic region. Revisional treatments of Pachyhalictus and Thrinchostoma are indicated in the ac- counts of those genera. Specific names marked by asterisks are placed on the basis of the literature only. Key To THE TRIBEs oF HALICTINAE 1. Anterior tentorial pits in clypeus, sepa- rate from epistomal suture although connected to suture by sulci; fimbria of metasomal tergum V of female not di- vided by longitudinal specialized area Boro Pee et Ra Pee ote, Sts SA Nomioidini ... Anterior tentorial pit in epistomal su- ture; fimbria of tergum V of female in nonparasitic forms divided by longi- tudinal median area of specialized fine, dense pubescence and punctation ........ 2 2. Longitudinal median specialized area of tergum V of female not divided by a cleft; metasomal tergum VII of male with a transverse ridge, usually cari- nate, forming a false apex beneath which the tergum is strongly reflexed to the morphological apical margin, surface above the transverse ridge usu- ally with a recognizable hairless py- Pica laters 225 ews eee Halictini .... Longitudinal median specialized area of tergum V of female divided by a deep cleft in the tergal margin; tergum VII of male without pygidial plate and without transverse premarginal ridge or carina forming a false apex .............. Ee ee Augochlorini TriBE NoMIOIDINI This tribe consists of minute species with dull, metallic, greenish, bluish or brassy, or rarely black, head and thorax and yellow markings in both sexes, usually involving the clypeus, pronotal margin, often the scutellum and metanotum, parts of the antenna and legs, and bands across the metasomal terga. Outer veins of forewings strong. Inner orbit rather strongly, angularly emarginate above middle. Anterior tentorial pit at apex of sharp angle or sulcus deep into general clypeal area, near upturned end of the large preapical transverse clypeal groove. Male: Metasomal tergum VII without recognizable pygidial plate, but margin produced posteriorly; trun- cate or notched. Posterior margin thin, not reflexed anteroventrally as in Halictini. Sternum VIII with well developed spiculum as well as long apical proc- ess. Apex of sternum VI somewhat produced, but entire. Genitalia rather elongate, gonostylus longer than the rest of genital capsule, over twice as long as gonocoxite, without ventral reflexed flap, much ex- ceeding penis valves; plane of dorsal bridge of penis valves vertical. Second tarsomere of hind leg nar- rowed toward base, freely articulated with basitarsus, as third is articulated to second. Female: WLabrum not thickened, apical process minute, with few hairs, not keeled. Metasomal ter- gum V with apical margin and fimbria entire, without median slit or area of specialized texture or vestiture. Genus Nomuotdes Schenck Figures 1, 7-9 This is the only genus of the Nomioi- dini. It consists of minute, usually yellow and greenish black bees. A few characters that vary among genera in other tribes and that are therefore of interest in the present context are as follows: Lower ends of paraocular areas angularly project- ing into clypeus. Inner hind tibial spur of female coarsely pectinate with a very few large teeth. Strigi- lis ending bluntly, with radiating series of apical spines. Costal margin of marginal cell about as long as stigma, shorter than distance from apex of cell to wing tip; apex of marginal cell subtruncate or round- ed. Lateral margins of metasoma with sharp angle separating dorsal from ventral parts of terga, the latter and the sterna often with long scopal hairs (angle less sharp in N. minutissima than in most species). 504 Tue University oF Kansas ScreNcE BULLETIN Fics. i-6. Wings of Nomioidini and Halictini. The scale lines represent 0.5 mm. 1. Nomuioides minutis- sima (Rossi). 2. Thrinchostoma (Thrinchostoma) near sjostedti. 3. Halictus (Halictus) quadrimaculatus. 4. H. (H.) ligatus. 5. H. (H.) maculatus. 6. H. (Seladonia) hesperus. Nomiuoides ranges from southern Eu- rope to southernmost South Africa, west as far as the Canary Islands, eastward to Mad- agascar and across Asia (north to the Cas- pean) to Taiwan and the Philippines, and southeast to Indonesia and Australia. It is common and represented by many spe- cies in arid and semiarid areas, but scarce and local in humid forested regions; it has not been found in New Guinea or islands to the eastward, although present in various Sunda Islands. The genus was revised by Bltthgen (1925) with a supplement in 1937: Nomioides appears to be divisible into two subgenera, as follows: Subgenus Nomioides Schenck, s. str. Nomuioides Schenck, 1866, Berliner Ent. Zeitschr. 10:333. Type species: Andrena pulchella Jurine, 1807=Apis minutissima Rossi, 1790 (monobasic). CLASSIFICATION OF HALICTINE BEES 505 This subgenus contains those species in which the pale, dorsal, integumental bands are on the posterior margins of the meta- somal terga. Fics. 7-9. Nomuordes minutissima (Rossi), male. 7, 8. Dorsal, ventral and lateral views of genitalia. 9. Sev- enth and eighth sterna, the former shown in broken lines. Subgenus Ceylalictus Strand Ceylalictus Strand, 1913, Arch. Naturgesch., 79(A): 137. Type species: Halictus hornt Strand, 1913 (monobasic). Cellaria Friese, 1913 (not Ellis and Solander, 1786), Deutsche Ent. Zeitschr., p. 575. Type species: Nomuoides arnoldi Friese, 1913 (monobasic). Cellariella Strand, 1926, Arch. Naturgesch. 92(A):53 (new name for Cellaria Friese). Type species: Nomuoides arnoldi Friese, 1913 (autobasic). Eunomioides Blithgen, 1937, Commentationes Biol., Soc. Sci. Fennica, 6:3 (no description or biblio- graphical reference). Type species: Andrena variegata Olivier, 1789, by original designation. In this subgenus the pale, metasomal bands are on the median or basal parts of the terga. Ceylalictus is not the name gen- erally applied to this group, because Bliith- gen proposed Eunomuoides for it. How- ever, Ceylalictus has priority; moreover, Eunomioides was never described, nor was a bibliographic reference given. It was therefore not validly proposed. Cellariella easily falls into Ceylalictus as here under- stood, differing by a single venational char- acter. TripE AUGOCHLORINI This tribe is restricted to the Western Hemisphere and most of its species are strongly metallic green, blue, brassy, etc. A few, however, are weakly metallic (as is Seladonmia) and a few almost completely lack metallic tints. Yellow or white mark- ings are usually absent; if present they are limited to appendages, labrum, and lower half of clypeus of males. Outer veins of forewings not weak. Inner orbit usually distinctly emarginate above middle of eye. Male: Metasomal tergum VII without pygidial plate, without transverse premarginal carina and without zone below it reflexed to meet apex of ster- num VI. Sternum VIII with spiculum. Sternum VI usually with median apical notch. Genitalia rather broad, gonostylus usually shorter than gonocoxite, without basal ventral flap; plane of dorsal bridge of penis valves horizontal so that bridge is entirely visible in dorsal view. Second tarsomere narrowly articulated to first, as third is to second. Female: Labrum thick, except in parasitic forms with apical process bearing a strong longitudinal keel. Metasomal tergum V with median longitudinal mi- nutely pilose or roughened area deeply notching into the prepygidial fimbria of long hairs, tergal margin in muddle of this area deeply cleft; or in parasitic forms (Temnosoma), tergum V unmodified with continuous apical hairy area. The genera of this tribe have been treated in detail by Eickwort (1969) and are not further dealt with in this paper. Tripe Haticrini This large tribe contains most Old World Halictinae as well as many of those of the New World. Species vary from minute to large, black to brilliantly me- tallic green or blue, usually without yellow markings in the female (the superficially Nomuoides-like genus Habralictus is an exception) and with yellow if any re- stricted to the clypeus, antennae, and legs in males, less commonly (Agapostemon and related Neotropical genera) forming metasomal bands or present on pronotal lobes. 506 Tue UNIversITY OF KANSAS SCIENCE BULLETIN Outer veins of fore wings often weakened. Inner orbit usually not strongly emarginate. Male: Metasomal tergum VII usually with recog- nizable although often poorly defined pygidial plate margined posteriorly by a transverse ridge or carina which forms the extremity or superficial apex of the tergum, above and behind the morphological posterior margin; area (sometimes only narrow marginal zone) beyond apex of plate reflexed, normally meeting pos- terior margin of sternum VI, occasionally pygidial plate reduced or absent, but even in such cases pos- terior part of tergum VII reflexed as indicated above. Sternum VIII without spiculum. Apex of sternum VI entire. Genitalia rather broad, gonostylus usually shorter than gonocoxite, often with basal ventral flap; plane of dorsal bridge of penis valves vertical. Second tarsomere of hind leg sometimes fused to first, some- times articulated but with articulation broader than that of third to second, sometimes narrow at base like third. Female: Labrum thick, except in parasitic forms with apical process bearing a strong longitudinal keel. Metasomal tergum V with median, longitudinal, mi- nutely pilose area (absent in parasitic genera) deeply notching into the prepygidial fimbria of long hairs, but tergal margin not cleft. Except for Halictus, the genera consid- ered here all belong to a group of genera in which the metasomal sternum IV of the male is armed with coarse and some- times gigantic setae or bristles and fre- quently shortened, mostly or wholly hid- den by HI. Only Thrinchostoma orchi- darum and the subgenus Lomatalictus of Patellapis (perhaps only the one species of that subgenus whose male is known) are exceptions to this feature. This group of genera 1s primarily African although it also ranges across tropical Asia. The presence of a membranous retrorse basal lobe of the male gonostylus in all members of this group suggests a relationship to the Lasio- glossum-Homalictus group, 1.e., to the gen- era of Halictini with weakened distal wing venation. Such a lobe is absent in Halictus, although present in the Neotropical Aga- postemon group. Key to O_tp Wortp Non-Parasitic GENERA WITH StroNc ApicaAL WING VENATION 1. Female with margin of clypeal trunca- tion, distal to preapical fimbria, ex- tended downward at each side of la- brum as a small, rather sharp, impunc- tate projection (except in some minute Asiatic species of the subgenus Vestito- halictus which lack such projections). Fourth sternum of male unmodified or at least without coarse, apical setae. Ventral basal process of male gono- stylus absent or if present directed apically and resembling a second sty- [Misa gas See eee Halictus Female with margin of clypeal trunca- tion, distal to preapical fimbria, extend- ing but little downward at each side of labrum, forming only a low rounded projection (except in some species of Thrinchostoma in which there is a strong projection). Fourth sternum of male usually shortened, commonly hid- den by third sternum, nearly always with apical or subapical coarse setae. Ventral, basal process of male gono- stylus present, directed ventrally or basally, forming a retrorse lobe ............ Apical marginal areas of terga with simple, laterally directed hairs that usu- ally form bands that are conspicuous only in certain lights. Profile of scu- tum in front gently convex, rising but little above level of pronotum. Prono- tum with carina separating dorsal from declivous anterior surface. Recurrent veins both entering third submarginal cell or first recurrent entering extreme apex of second cell Thrinchostoma Apical marginal areas of terga variable, but without simple, laterally directed hairs. Profile of anterior part of scu- tum strongly convex, so that there is a subvertical surface rising well above level of pronotum and then curving strongly or angularly onto dorsum of scutum. Pronotum medially without carina separating dorsal from anterior surface. Recurrent veins entering sec- ond and third submarginal cells, each well. before apex ot its celly@ ieee Malar area about as long (female) to twice as long (male) as diameter of CLASSIFICATION OF HALICTINE BEEs 507 flagellum. Pygidial plate of male not defined. Gonostyli of male not bifur- CALCre ee ASS sere ee Thrincohalictus Malar area usually linear, rarely about half as long as diameter of flagellum. Pygidial plate of male defined at least posteriorly and posterolaterally by a carina. Gonostyli of male bifurcate, one branch sometimes slender and in- GOIS DIGU US ye see ns eee ote es 4 4. Metasomal terga with basal bands of tomentum. Pygidial plate of male rather small. Rami of male gonostylus subequal in thickness. Hind tibia of female with outer surface largely cov- ered with rather short, nearly erect hairs of uniform length .... Pachyhalictus Metasomal terga without basal bands of pale tomentum. Pygidial plate of male large. Outer ramus of male gono- stylus much more slender than inner. Hind tibia of female with hairs on outer surface longer, slanting, as in LSoOK( a Leet hcl td ES ee AAR aC eis oO nar 5 5. First and frequently other metasomal terga usually with colored (blue, green, yellow, white) apical integumental bands. Thoracic pubescence long, plu- mose, usually yellowish; body almost without areas of short, whitish pubes- cence or tomentum as in many halic- tine groups. Clypeus of male often WUC Vell Ope sees ee ee nes Zonalictus Tergal margins not colored. Thoracic pubescence usually shorter, less fully plumose, grayish or whitish; body of- ten with areas of whitish pubescence or tomentum. Clypeus of male without yellowaareas 2) A Patellapis Genus Patellapis Friese Figures 10-44 The hitherto little used name Patellapis is here applied in a much broader sense than previously. It becomes a substantial genus of African Halictini, encompassing species that exhibit much morphological diversity. 1. Nonmetallic black (sometimes with red meta- soma), small to rather large, 5-14 mm long. 2 Punc- tation ordinary, ground between punctures shining to dull. 3. Clypeus neither produced downward nor pro- tuberant forward (except produced and protuberant when head is elongate, as in P. braunsella), less than three times (sometimes only twice) as broad as long, angle at end of truncation not or feebly produced, surface shining between punctures, uniformly gently convex. 4. Line between lower ends of eyes crossing clypeus below or above middle. 5. Malar space linear or about half as long as width of flagellum in P. pastina. 6. Paraocular area not extending down as lobe into clypeus, or forming an obtuse angle, or in P. braunsella a strong lobe. Mouthparts short (as in Pachyhalictus) to quite elongate; glossa in P. braun- sella as long as face, more than twice as long as labial palpi, and postpalpal part of galea over twice as long as broad. 8. Pronotum with subhorizontal, dorsal surface medially about one third as long as flagellar width, often densely tomentose, margined anteriorly by declivity of pronotal surface, not by carina or sharp angle. 9. Dorsolateral angle of pronotum ob- tuse, not lamellate or strongly carinate, a weak carina often extending across posterior lobe of pronotum, a rounded ridge extending downward from dorsolateral angle. 10. Anterior extremity of scutum strongly con- vex in profile, the largely impunctate, vertical, an- terior zone rising well above pronotum and curving onto dorsal surface without sharp line of separation. 11. Pre-episternal groove often short and shallow below scrobal groove. Metanotum not or partly to- mentose. 12. Dorsal surface of propodeum longer than metanotum, shorter than to as long as scutel- lum, striate to granular; triangular area sometimes defined by end of striate or granular zone, broad when recognizable. Lateral and posterior surfaces of propodeum with or without short hairs in addition to long ones, these surfaces not areolate. Posterior surface of propodeum margined by carinae only be- low, laterally. 13. Apical wing veins strong, recurrent veins entering second and third submarginal cells. 14. Third submarginal cell a little elongate, third transverse cubital arcuate, usually straight toward costal margin of wing, or sinuate. 15. Marginal cell rather robust with free part distinctly less than twice as long as part subtended by submarginal cells, apex pointed very near wing margin or separated from it by about a vein width, not appreciably appendiculate. 16. First metasomal tergum much broader than long. 17. Terga without basal areas or bands of tomentum (very sparse tomentum on tergum II basolaterally in P. braunsella). Terga sometimes uniformly sparsely hairy, but more often with hairs denser and more plumose toward posterior margins of terga, often forming strong fasciae of plumose hairs on these mar- gins as in Halictus. 18. Apical margins of terga broadly depressed with punctation about as on more anterior parts, hairs not directed laterally or hairs in fasciae weakly so, only very narrow tergal margins impunctate and hairless, these margins or broader areas translucent and pallid. Discs of terga II and III sometimes with oblique hairs. 508 Tuer University oF Kansas SCIENCE BULLETIN TS —! . | Ps We > : wy : | ge 10 of = 1 4 i We k \ Sx 16 — 12 SSS OH sass0, ; E man : 17 Fics. 10-19. Structures of Patellapis (Lomatalictus). 10-16. P. malachurina. 17-19. P. pallidicinctula. 10, 11. Dorsal, ventral, and lateral views of male geni- talia. 12. Posterior lateral view of male gonostylus. 13. Metasomal sterna VII and VIII of male. 14. Basi- tibial plate of female. 15. Claw of female. 16, 17. Inner hind tibial spurs of females. 18. Claw of fe- male. 19. Basitibial plate of female. Male: 19. Clypeus and legs without yellow or white areas. 20. Body of labrum two (in P. braun- sella) to over three times as wide as long, fringed with bristles, without apical process or with a short triangular process, or in P. schultze: with strong, keeled apical process almost like that of a female. Mandible simple or in Lomatalictus bidentate. 21. Flagellum short to moderate in length, first segment broader than long, second and sometimes third and fourth broader than long to longer than broad, middle segments usually distinctly longer than broad, some- times 1.5 times as long as broad. 22. Basitibial plate present or absent. 23. First two hind tarsal segments apparently articulated, but base of second broader than base of third. 24. Metasoma moderately robust, shaped about as in female, third segment widest or second and third equal. 25. Pygidial plate rather large, defined by strong carina both laterally and apically, smooth area longer than broad to slightly broader than long. 26. Sternum IV often short and largely or wholly hidden by III, usually with a series of bristles. Sternum V unmodified to broadly emar- ginate apically. 27. Sternum VII a transverse band with median apical projection; VII with broadly rounded, truncate or emarginate apical projection, often with hairs. 28. Genitalia broad with somewhat narrow base (broad base in P. schultzer). Gonostylus bifid distally (upper branch sometimes delicate and difficult to see, especially in P. schultzer), more than half as long as gonobase, with retrorse, ventral, basal, membranous lobe which is sometimes bifid (e.g., 1n P. schonlandi). Penis valve rather slender to enlarged medially, inferior basal process slender and parallel sided in P. schultzer or rounded to obliquely truncate. Female: 29. Scape reaching at least to anterior margin of anterior ocellus, sometimes as in P. cinctt- cauda reaching middle of posterior ocellus. Second fla- gellar segment broader than long (about as broad as long in P. montagut), first and even third and others sometimes also broader than long. 30. Labrum with tapering apical process with keel; body of labrum more than twice as broad as long. 31. Hind tibia and its scopa of usual form. 32. Basitibial plate of moderate size, angular or rounded apically, margin elevated, surface dull or shining, with some hairs. 33. Inner hind tibial spur serrate to pectinate. Hind ubia with two apical spines, sometimes short and mere angles, or posterior one commonly absent, so that there is only one spine. 34. Sternal hairs simple to plumose, of moderate length. Patellapis belongs to the group of gen- era with the apical wing venation strong and with the fourth metasomal sternum of males armed with bristles (except in the subgenus Lomatalictus) and nearly always shortened. It differs from Pachyhalictus by the lack of tomentous basal bands on the terga; the ordinary sculpturing and hind tibial shape and scopa of the female, these features being as in most halictines; the pointed marginal cell; the weakly ca- rinate and nonlamellate dorsolateral pro- notal angles; and by the large, well defined pygidial plate of the male with the smooth area usually longer than broad. It differs from Zonalictus by the lack of apical, col- ored tergal bands, the hairy and sometimes fasciate apical tergal margins, the shorter and less fully plumose pubescence of the head and thorax (except for P. malachu- rina and allies which resemble Zonalictus in this respect), the shorter and more ro- bust form, etc. It would not have been illogical, however, to include Zonalictus as a subgenus of Patellapis. Key TO THE SUBGENERA OF Patellapis 1. Claws of female simple or with inner tooth very small, of male with the teeth close together. Fourth sternum of male similar in size and vestiture tothiid: oe Lomatalictus Claws toothed as usual in halictines. Fourth sternum of male usually short- ened, often largely hidden under third, CLASSIFICATION OF HALICTINE BEEs 509 Fic. 20. Top row: Patellapis (Lomatalictus) mala- churina, face of male, face and wing of female. Bot- tom row: Patellapis (Chaetalictus) pearstonensts, wing and face of female, face of male. Scale line = 1.0 mm. with a few to many coarse bristles in AUULATISWEESE TOW soso 2 ts eet 2 2. Terga with conspicuous apical hair bands; basitibial plate margined both in front and behind, apex in female usually rounded .............. Patellapis s. str. Terga without or with weak apical hair bands; basitibial plate not or in- completely defined on anterior margin, apex angulate or pointed .... Chaetalictus The subgeneric classification is not en- tirely satisfactory. When more species are known from both sexes, it should be re- examined. There is great diversity within the genus and even within the subgenera. Lomatalictus new subgenus Figures 10-20 Type species: Halictus malachurinus Cockerell, 1937. Clypeus only weakly convex in profile. Mandible of male bidentate. Claws of female with small inner tooth or in P. pallidicinctula, simple. Basitibial plate slender, narrowly rounded or angulate at apex in female, defined only along posterior margin and at apex in male. Inner hind tibial spur of female finely pectinate-serrate (in P. pallidicinctula) to pectinate. Metasomal tergal apices broadly pallid translucent with strong apical bands of plumose hairs. Fourth sternum of male unmodified. Penis valve without enlarged dorsal crest. The male of pallidicinctula is unknown to me; the above comments on males are based on P. malachurina. Lomatalictus is known only from South Africa. There may be only two species; P. pallidicinctula is clearly different from malachurina, but the other two names may both be synonyms of the latter. Included names, all described in Halic- tus and all new combinations, are as fol- lows: Patellapis (Lomatalictus) levisculpta (Cockerell, 1939) Patellapis (Lomatalictus) malachurina (Cockerell, 1937) Patellapis (Lomatalictus) pallidicinctula (Cockerell, 1939) Patellapis (Lomatalictus) suprafulva (Cockerell, 1946) The name Lomatalictus is based on /o- matos, fringes, plus Halictus, with refer- ence to the apical bands of hairs on the metasomal terga. le = Sn ( \ EN \ ( ( \ | Sy \ | ) re = PtH \) a 21 / Pee ae, 23 RN ae oa amet = | AV Fics. 21-25. Patellapis (Chaetalictus) pearstonensts. 21. Fourth sternum of male. 22. Gonocoxite and stylus of male in ventral view. 23. Inner hind ubial spur of female. 24. Claw of female. 25. Gonostylus of male in posterior lateral view. Chaetalictus new subgenus Figures 20-26 Type species: Halictus pearstonensis Cameron, L905. Clypeus markedly convex in profile. Mandible of male simple. Claws normal for halictines. Basitibial plate of female rather slender, weak on anterior mar- gin, narrowly rounded to angulate at apex, as in 510 Tuer University oF Kansas ScrENcCE BULLETIN Lomatalictus; of male absent to small, undefined anteriorly, with angulate apex. Inner hind tibial spur of female finely pectinate-serrate (in P. serifera, as figured for P. pallidicinctula) to pectinate. Metasomal terga with translucent margins narrow to broad, apical hair bands weak or absent. Sternum IV of male of normal size to shortened, broadly emarginate, and hidden under III; it bears six enormous bristles or (in P. rubrotibialis) a row of often erect or retrorse bristles, only the lateral ones of which are large, or (in P. pulchrinitens) a row of rather weak bristles. Penis valve without enlarged dorsal crest. This subgenus, known only from south- ern Africa, consists of species mostly small- er than those of Patellapis s. str. Included names, all described in Halic- tus and all new combinations, are as fol- lows: Patellapis (Chaetalictus) atricilla (Cockerell, 1940) Patellapis (Chaetalictus) ausica (Cockerell, 1945) Patellapis (Chaetalictus) calvini (Cockerell, 1937) Patellapis (Chaetalictus) calviniensis (Cockerell, 1934) Patellapis (Chaetalictus) capillipalpus (Cockerell, 1946) Patellapis (Chaetalictus) chubbi (Cockerell, 1937) Patellapis (Chaetalictus) cinctifera (Cockerell, 1946) Patellapis (Chaetalictus) communis (Smith, 1879) Patellapts (Chaetalictus) disposita (Cameron, 1905) Patellapis (Chaetalictus) dispositina (Cockerell, 1934) Patellapis (Chaetalictus) flavorufa (Cockerell, 1937) Patellapts (Chaetalictus) leonis (Cockerell, 1940) Patellapis (Chaetalictus) micropastina (Cockerell, 1940) Patellapis (Chaetalictus) neh (Cockerell, 1937) Patellapis (Chaetalictus) pastina (Cockerell, 1937) Fics. 26-33. Structures of Patellapis. 26, 27. Meta- somal sterna of males, with sparse hairs omitted, but areas of dense pubescence shown, of P. (Chaetalic- tus) pearstonensis and P. (Patellapis) braunsella. 28, 29. Inner hind tibial spurs of females, P. (P.) mon- tagut and braunsella. 30, 31. Dorsal, ventral and lateral views of male genitalia, P. (P.) braunsella. 32. Posterior lateral view of male gonostylus of same. 33. Eighth sternum of same. Patellapis (Chaetalictus) pastinella (Cockerell, 1939) Patellapis (Chaetalictus) pastiniformis (Cockerell, 1939) Patellapis (Chaetalictus) pearstonensis (Cameron, 1905) Patellapts (Chaetalictus) pondoensis (Cockerell, 1937) Patellapts (Chaetalictus) probita (Cockerell, 1933) Patellapts (Chaetalictus) pulchrinitens (Cockerell, 1937) Patellapis (Chaetalictus) rubrotibialis (Cockerell, 1946) Patellapis (Chaetalictus) rufiventris (Friese, 1925) (not Halictus rufiventris Giraud, 1861) Presumably a synonym of pear- stonensis and hence not in need of a new name. Patellapis (Chaetalictus) sanguinibasis (Cockerell, 1939) CLASSIFICATION OF HALICTINE BEEs 5) Ui Patellapis (Chaetalictus) schonlandi (Cameron, 1905) Patellapis (Chaetalictus) semipastina (Cockerell, 1940) Patellapis (Chaetalictus) serrifera (Cockerell, 1937) Patellapts (Chaetalictus) spinulosa (Cockerell, 1941) Patellapis (Chaetalictus) tenuihirta (Cockerell, 1939) Patellapis (Chaetalictus) terminalis (Smith, 1853) Patellapis (Chaetalictus) vambensts (Cockerell, 1940) Patellapts (Chaetalictus) villosicauda (Cockerell, 1937) Patellapis (Chaetalictus) volutatoria (Cameron, 1905) The name Chaetalictus is based on chaetes, bristle or hair, plus Halictus, with reference to the coarse bristles on the fourth sternum of the males. Fics. 34-43. Structures of Patellapis (Patellapis). 34, 35. Dorsal, ventral and lateral views of male geni- talia of Patellapis (P.) schultzei. 36. Posterior lateral view of gonostylus of same. 37-39. Fourth, eighth, and seventh sterna of same. 40, 41. Basitibial plates of female and male, P. (P.) schultzei. 42, 43. Basi- tibial plates of female and male, P. (P.) braunsella. Subgenus Patellapis Friese, s. str. Figures 27-44 Patellapis Friese, 1909, Die Bienen Afrikas, p. 148, in L. Schultze, Zoologische und anthropologische Ergebnisse einer Forschungsreise im westlichen und zentralen Siidafrika, vol. 2, part 2. Type species: Halictus (Patellapis) schultzei Friese, by designation of Cockerell, 1920, Ann. Dur- ban Mus., 2:311. Agrees with description of Chaetalictus except as follows: Mandible of male simple or in minutior bidentate. Basitibial plate of female rather broad, margin well defined throughout, not weak on anterior side, apex rounded or weakly angulate; of male simi- larly broad and well defined in P. schultzei, narrower, pointed apically, but defined by carina at least part way up anterior margin in other species. Inner hind tibial spur of female weakly serrate to pectinate. Metasomal terga with broad translucent pallid margins and strong apical bands of plumose hair. Sternum IV of male unmodified in shape (in P. “minutior’) to shortened, broadly concave apically, and largely hidden by III, in all species with a transverse row of coarse bristles, the lateral ones greatly enlarged in P. braunsella, but not in others. Penis valve with median dorsal carina greatly expanded to form api- cally directed helmet-like crest. This subgenus, known only from Cape Province, South Africa, contains the fol- lowing species: Patellapis (Patellapis) braunsella new species Patellapis (Patellapis) cincticauda (Cockerell, 1946) new comb. Patellapis (Patellapis) minutior (Friese, 1909) Patellapis (Patellapis) montagui (Cockerell, 1941) new comb. Patellapts (Patellapis) schultzei (Friese, 1909) The specimen of P. minutior used for this study was labeled “typus” by Friese and is in the American Museum of Nat- ural History. Other specimens, similarly labeled, in the Berlin museum are a differ- ent species of the same subgenus, with simple mandibles and other features not agreeing with my comments on minutior. Presumably the Berlin specimens are the true minutior. I have therefore placed the name in quotes where reference is to the 512 Tue University oF Kansas SciENCE BULLETIN Fic. 44. Top row: Patellapis (Patellapis) schultze1, face of male and P. (P.) montagui, face and wing of female. Middle row: Patellapis (P.) braunsella, face of male, face and wing of female. Bottom row: Zonalictus albofasciatus, face of male, face and wing of female. Scale line = 1.0 mm. CLASSIFICATION OF HALICTINE BEEs 513 specimen in the American Museum. It does not seem to me appropriate to name the species at present, since I have seen only a single specimen in poor condition. P. cincticauda is placed in this subgenus hesitantly, since I have seen no males and have not examined specimens since recog- nizing the limits of this group. The geni- talia of P. schultzei and P. braunsella are quite different as illustrated, but those of P.“minutior” are intermediate in gonobase width and some other features. Great diversity within as well as be- tween the subgenera of Patellapis some- times confuses the subgeneric limits. The bristles of the fourth sternum of the male in P. braunsella are greatly enlarged lat- erally, suggesting P. (Chaetalictus) rubro- tibialis. The size and shape of the same sternum (not shortened) in P. “minutior” suggest Lomatalictus, as do the bidentate mandibles of the male. The one female specimen of a minute and presumably un- described species of Chaetalictus has claws with only a small inner tooth, like those of P. (Lomatalictus) malachurina. There are very striking differences among species of at least the genera Patellapis and Chaeta- lictus, and the subgenera here recognized may be inadequate to properly reflect this diversity, or the limits of the subgenera recognized may not be ideal. One of the major problems is that for too many spe- cies only one sex is known, so that the full suite of specific characters is not available. The classification presented above is, there- fore, tentative and conservative in that few subgenera are named. When both sexes are known for most of the species, more sub- genera may be desirable. Zonalictus new genus Figures 44-53 Halictus albofasciatus-Gruppe, Bliithgen, 1929, Mitt. Zool. Mus. Berlin, 15:29. Type species: Halictus albofasciatus Smith. 1. Nonmetallic black (sometimes with a partly red metasoma), usually with white, yellow, greenish, or bluish apical, integumental bands on terga; 6.5-14 mm long. 2. Surface of head and thorax commonly dull; frons and usually dorsum of thorax finely reticu- late, the latter sometimes with punctures of two sizes intermixed, rarely shining with scattered punctures. Pubescence long, plumose, commonly yellowish, there being almost no areas of short, whitish, pubes- cence or tomentum as in most halictine groups [but Patellapis (Lomatalictus) malachurina and its relatives are similar in this respect]. 3. Clypeus neither pro- duced downward nor protuberant forward (except strongly produced downward in the long-headed spe- cies Z. zacephalus), two to three times as broad as long, angle at end of truncation not or feebly pro- duced, surface sometimes shining, but usually dull basally, large punctures often limited to distal half of clypeus; clypeal profile sometimes uniformly con- vex, but usually flat basally and strongly convex distally. 4. Line between lower ends of eyes crossing clypeus below or above middle. 5. Malar space linear. 6. Paraocular area not extending down as lobe into clypeus, or forming an obtuse lobe or angle. Mouth- parts short and ordinary or in Z. zacephalus, glossa over half as long as face, in Z. concinnulus glossa longer than face and first two segments of labial palpus much elongated. 8. Pronotum with dorsal surface medially not defined, sloping anteriorly, but elsewhere defined anteriorly by declivity of pronotal surface, not by carina; dorsal surface (collar) often weakly tomentose. 9. Dorsolateral angle of prono- tum obtuse, not lamellate or carinate, a ridge (often weak) extending laterally in some species across pronotal lobe, a rounded or sometimes sharp ridge extending downward from dorsolateral angle. 10. Anterior extremity of scutum as described for Patel- lapis. Preepisternal groove as described for Patellapis. Metanotum not or scarcely tomentose. 12. Dorsal surface of propodeum usually as long as scutellum, sometimes slightly shorter, finely granular or granu- lostriate; triangular area sometimes defined by weak carinae, usually not defined, when recognizable, broad. Lateral and posterior surfaces of propodeum without short hairs in addition to long ones, not areolate. Posterior surface margined by carinae only below laterally. 13. Wing venation as in Patellapis, but marginal cell more slender. 16. First metasomal tergum broader than long or in some males longer than broad. 17. Terga without basal areas or bands of tomentum (rarely very feeble tomentum basolat- erally on tergum II, as in Z. albofilosus), without apical hair bands, hairs smaller and commonly sparser on apical parts of terga 1-4 than elsewhere, these margins usually highly colored, at least on tergum 1, forming white, yellow, greenish, or bluish integu- mental bands. 18. Apical margins of terga broadly and weakly depressed, not punctate, but with hairs similar to but smaller than those on rest of terga, hairs not directed laterally; margins of terga hairless, not translucent when highly colored, but broadly 514 Tue University oF Kansas ScrENCE BULLETIN 45 a Fics. 45-53. Structures of Zonalictus. 45, 46. Dorsal, ventral and lateral views of male genitalia of Z. albo- fasciatus. 47. Posterior lateral view of stylus of same. 48. Seventh tergum of same. 49. Fourth sternum of same. 50-52. Inner hind tibial spurs of females of Z. partitus, albofasciatus, and zacephalus. 53. Ventral view of male gonocoxite and gonostylus of Z. cinc- tulellus. translucent when not colored. Discs of terga II and III without oblique hairs. Male: 19. Clypeus often with apical transverse yellow area; labrum often partly yellow; legs without yellow. 20. Labrum usually over three times as wide as long, without or with a barely evident apical proc- ess, but in Z. concinnulus only about twice as wide as long because of a strong, triangular apical process (not keeled). Mandible simple (bidentate in Z. con- cinnulus). 21. Flagellum elongate, first segment a little broader than long, other segments much longer than broad. 22. Basitibial plate present and well de- fined to nearly absent with only the apex distinct. 23. First two hind tarsal segments articulated as in Patellapis. Metasoma rather elongate, widest at segments 2 and 3. 25. Pygidial plate moderately large, defined by a strong carina both laterally and apically, smooth area usually longer than broad. 26. Sternum IV shortened, largely hidden by III, some- times with a series of coarse bristles becoming pro- gressively larger laterally (as in Patellapis braunsella), but usually with a series of erect or retrorse bristles of uniform size medially and one enormous, isolated, largely hidden, lateral bristle at each side. 27. Ster- num VII a transverse band with median apical projec- tion; VII with produced apex emarginate. 28. Geni- talia broad with somewhat narrow base. Gonostylus bifid distally, more than half as long as gonobase, with retrorse, ventral, basal, membranous lobe which may be large and bifid or may be much reduced in size. Penis valve rather slender, dorsal keel not ex- panded as in Patellapis s. str., inferior basal process broadly rounded or subtruncate at apex. Female: 29. Scape reaching posterior ocellus. Sec- ond flagellar segment as long as broad or usually broader than long, others commonly longer than broad, but mostly broader than long in some species (e.g., Z. zacephalus). 30. Labrum with tapering or sometimes rounded apical process with keel, body of labrum more than twice as broad as long. 31. Hind tibia and its scopa of the usual form. 32. Basitibial plate of moderate size, angular or narrowly rounded apically, margin elevated throughout or anterior mar- gin largely absent so that plate is defined only api- cally and posteriorly; surface of plate with some hairs. 33. Inner hind tibial spur usually coarsely serrate to pectinate with short teeth, but in Z. zacephalus finely ciliate-serrate. Hind tibia with one tibial spine. 34. Sternal hairs short to moderate in length, simple to plumose, not suggestive of a scopa. Zonalictus is closely related to Patellapts and could easily be incorporated into that genus as a subgenus. Because of its rather elongate form, the long, yellowish plumose hairs, and the integumental color bands on the apices of the metasomal terga (very rarely absent but often limited to tergum I or I and II), it has a different appearance from Patellapis or other halictids. Because of this fact, and other characters which are not invariable, such as the yellow on the clypeus of the male, the usually longer male antennae, and the commonly dull integumental surfaces between punctures, I have hesitantly accorded Zonalictus ge- neric status. Zonalictus is found throughout subsa- haran Africa and eastward to Madagascar, Socotra, and Yemen. The specific names known to be included are listed below. Names preceded by an asterisk are placed here on the basis of literature only. Zonalictus aberdaricus (Cockerell, 1945) *Zonalictus abessinicus (Friese, 1916) Zonalictus albofasciatus (Smith, 1879) Zonalictus albofilosus (Cockerell, 1937) Zonalictus albolineolus (Meade-Waldo, 1916) Zonalictus alopex (Cockerell, 1937) Zonalictus andersoni (Cockerell, 1945) *Zonalictus andreniformis (Friese, 1925) *Zonalictus andrenotdes (Friese, 1909) CLASSIFICATION OF HALICTINE BEEs 515 Zonalictus baralongus (Cockerell, 1939) *Zonalictus bilineatus (Friese, 1909) Zonalictus broomi (Meade-Waldo, 1916), nomen nudum Zonalictus burunganus (Cockerell, 1937) *Zonalictus burungensts (Cockerell, 1937) Zonalictus cerealis (Cockerell, 1945) Zonalictus cinctulellus (Cockerell, 1946) Zonalictus concinnulus (Cockerell, 1946) *Zonalictus flavofasciatus (Friese, 1915) Zonalictus flavovittatus (Kirby, 1900) Zonalictus fuliginosus (Cockerell, 1937) Zonalictus gowdeyi (Cockerell, 1937) Zonalictus grandior (Blithgen, 1929) Zonalictus hargreavesi (Cockerell, 1946) Zonalictus heterozonicus (Cockerell, 1937) Zonalictus kabetensts (Cockerell, 1937) *Zonalictus kamerunensis (Friese, 1914) (This is the first of two forms to which Friese gave the same trivial name on the same page.) Zonalictus kavirondicus (Cockerell, 1945) Zonalictus kivuicola (Cockerell, 1937) Zonalictus knysnae (Cockerell, 1945) Zonalictus kristenseni (Friese, 1915) Zonalictus macrozonius (Cockerell, 1937) Zonalictus microzonius (Cockerell, 1937) Zonalictus minor Blithgen, 1929 (not Halictus minor Morawitz, 1876) (Named as a variety of andrent- formis; no replacement name seems needed.) Zonalictus mirandicornis (Cockerell, 1939) Zonalictus moshiensis (Cockerell, 1937) Zonalictus neavei (Cockerell, 1946) Zonalictus nefasiticus (Cockerell, 1935) Zonalictus nomioides (Friese, 1909) Zonalictus pallidicinctus (Cockerell, 1933 Zonalictus partitus (Cockerell, 1933 Zonalictus patriciformis (Cockerell, 1933 Zonalictus pearson (Cockerell, 1933) Zonalictus percinctus (Cockerell, 1937) Zonalictus perlucens (Cockerell, 1933) Zonalictus perpansus (Cockerell, 1933) Zonalictus promitus (Cockerell, 1934) Zonalictus pulchricinctus (Cockerell, 1933 Zonalictus pulchrihirtus (Cockerell, 1933 *Zonalictus rufobasalis (Alfken, 1930) Zonalictus ruwensorensis (Strand, 1911) Zonalictus sidulus (Cockerell, 1937) Zonalictus stanleyi (Cockerell, 1945) *Zonalictus subpatricius Strand, 1911 Zonalictus subvittatus (Cockerell, 1937) Zonalictus tenuimarginatus (Friese, 1925) Zonalictus territus (Cockerell, 1937) Zonalictus tinctulus (Cockerell, 1937) Zonalictus tricolor (Meade-Waldo, 1916, nomen nudum, not Halictus tricolor Lepeletier, 1841 Zonalictus trifilosus (Cockerell, 1945) Zonalictus unifasciatus (Cockerell, 1937) Zonalictus viridifilosus (Cockerell, 1946) Zonalictus vittatus (Smith, 1853) *Zonalictus weist (Friese, 1915) Zonalictus zacephalus (Cockerell, 1937) Zonalictus zaleucus (Cockerell, 1937) The name Zonalictus is based on zone, a belt or girdle, plus Halictus, with refer- ence to the conspicuous integumental meta- somal color bands of many species. Genus Pachyhalictus Cockerell Figures 54-68 1. Nonmetallic black, rather small, robust, 5.5-7 mm long. 2. Frons, scutum and scutellum reticulate 516 Tue UNIversiry oF KAnsAs SCIENCE BULLETIN (more finely on frons), not punctate, or occasionally partly smooth or, as in P. bingham1, with fine wrin- kles and shallow punctures. 3. Clypeus neither pro- duced downward nor protuberant forward, about three times as wide as long or wider, angle at end of truncation not or feebly produced, surface closely punctured or reticulate and dull, uniformly gently convex. 4. Line between lower ends of eyes crossing clypeus below or above middle. 5. Malar space linear. 6. Paraocular area not extending down as lobe into clypeus, lateral clypeal margins only weakly curved and at an obtuse angle to one another. 7. Mouthparts short, the short glossa exceeded by labial palpi, post- palpal part of galea not much longer than broad. Fics. 54-61. Structures of Pachyhalictus (Pachyha- lictus). 54, 55. Dorsal, ventral, and lateral views of male genitalia of P. (P.) species y of Bluthgen, 1926. 56. Posterior lateral view of gonostylus of same. 57. Median part of seventh tergum of same. 58. Fourth sternum of same, with detached giant seta in different view (X indicates a damaged area, reconstructed in drawing). 59. Seventh and eighth sterna of same. 60. Inner hind tibial spur of female of P. (P.) meres- cens. 61. Outer surface of hind tibia of same. 8. Pronotum with horizontal dorsal surface medially one half (in P. retigerus) to one third as long as flagellar width, often densely tomentose, margined anteriorly in middle only by sharp declivity of pro- notal surface, but laterally by carina or lamella. 9. Dorsolateral angle of pronotum obtuse, but formed by strong anteriorly or upward directed carina or lamella which extends across posterior pronotal lobe. 10. Anterior extremity of scutum sharply angular in profile, the largely impunctate vertical or overhanging anterior zone rising well above pronotum and some- times separated from rest of scutal surface by a carina. 11. Preepisternal groove well developed. Meta- plura strongly narrowed below, where narrowest less than one third as wide as at upper end. Metanotum tomentose. 12. Dorsal surface of propodeum slightly shorter than to longer than scutellum, usually coarsely areolate, sometimes irregularly coarsely striate, inter- vals or areolae dull or in P. retigerus shining. Tri- angular area not or scarcely defined. Lateral and pos- terior propodeal surfaces with short hairs in addition to long ones, these surfaces usually areolate. Posterior surface of propodeum usually margined all the way around by carinae, but P. binghami and _ retigerus without carina across summit or on upper parts of sides. 13. Apical wing veins strong. Recurrent veins entering second and third submarginal cells near apices, first sometimes almost interstitial; in P. re- tigerus both recurrents entering cells at apical third or fourth. (One specimen seen lacking first trans- verse cubital, another lacking second; species with only two submarginal cells may exist.) 14. Third submarginal cell short, third transverse cubital vein straight or arcuate, not sinuate or in P. retigerus feebly so. 15. Stigma of moderate size. Marginal cell with free part less than twice as long as part subtended by submarginal cells, apex appendiculate, minutely truncate to pointed, apex separated from wing margin by less than to more than two vein widths. 16. First metasomal tergum much broader than long. 17. Terga II, III, and sometimes IV with strong basal bands of pale tomentum, without apical fasciae. 18. Apical margins of terga broadly, weakly depressed with hairs and punctation usually about as on more anterior parts of same terga; hairs not or scarcely directed laterally; only very narrow tergal margins impunctate, pallid, and hairless (broad mar- ginal zones impunctate when discs of terga are also largely impunctate as for terga I and II of P. re- tigerus). Discs of terga II and III sometimes with oblique hairs. Male: 19. Clypeus and legs without yellow or white areas. 20. Labrum nearly or over four times as wide as long, fringed with bristles, without apical process. Mandible simple. 21. Flagellum short, all but last two or three segments broader than long or middle segments about as broad as long, first segment much broader than long. 22. Basitibial plate absent (present in P. retzgerus). 23. First two hind tarsal segments distinct or fused, but point of union indi- cated by strong constriction. 24. Metasoma robust, shaped about as in female. 25. Pygidial plate small, defined by carina which curves forward laterally, so that it margins the plate both laterally and apically. 26. Sternum IV broadly emarginate posteriorly, me- dian part much shortened, hidden by III, thickened, with an apical series of erect bristles on each side of midline (Blithgen’s species y) and often with an enormous lateral bristle (sometimes hooked as in bihamatus) (armature of sternum IV probably highly variable among species, bristles sometimes entirely hidden by sternum III). Sternum V gently emarginate apically (at least in species y and P. retigerus). 27. Sterna VII and VIII much reduced, membranous, VII a slender transverse strip in species y, with small median apical projection in P. retigerus, VIII dam- aged, but apparently hairless and without significant apical projection. 28. Genitalia broad with somewhat narrow gonobase; gonostylus ornate, main part deeply bifid, as long as gonobase; retrorse, membranous, CLASSIFICATION OF HALICTINE BEEs 517 basal lobe itself bilobed, one part extending distad, the other mesad. Penis valve slender, inferior basal process subtruncate (based on Blithgen’s species + and on P. retigertus). Female: 29. Scape reaching or nearly reaching anterior margin of anterior ocellus. Second, often first, and sometimes other flagellar segments broader than long. 30. Labrum with tapering apical process with keel, body of labrum much more than twice as wide as long. 31. Hind tibia rather robust, lower surface and therefore lower margin as seen from side gently concave, scarcely so in P. retigerus, outer sur- face largely covered with rather short, nearly erect hairs of uniform length, upper surface with short bristles; lower margin on outer surface with long, coarse, hairs, especially those of basal half of the tibia with more numerous, crowded, and_ coarser branches than in most halictids, all directed toward apex of tibia. 32. Basitibial plate of moderate size, triangular to rounded apically, margin elevated, sur- face dull, shining in P. retigerus, with some hairs. 33. Inner hind ubial spur pectinate with a few coarse to many fine teeth. Hind tibia without or with one short tibial spine. 34. Sterna and ventral parts of terga with plumose hairs which in some Indoaus- tralian species are large, richly plumose, and important scopal structures. Pachyhalictus resembles Homalictus in the peculiar shape and vestiture of the hind tibia of the female and some species even have such a large ventral metasomal scopa as to suggest Homalictus. In many ways, however, Pachyhalictus differs from Homalictus, e.g. the robust body, basal bands of tomentum on the terga, and strong second recurrent and third trans- verse cubital veins. In this feature of vena- tion and in the shortened, largely hidden fourth metasomal sternum of the male, Pachyhalictus resembles Patellapis, from which it differs in the structure of the tibia of the female, basal bands of tomentum, etc. The bifid gonostyli also suggest a rela- tionship to Patellapis. The distinctive re- ticulate sculpturing of Pachyhalictus is found in a very few species of Patellapis and Zonalictus. Pachyhalictus is found principally from the Asiatic tropics and nearby islands to New Guinea. One species (P. stirlingi), however, occurs in northern Australia. Another, P. retigerus, morphologically dif- ferentiated but clearly a member of the genus, occurs in southeastern Africa. Key To THE SuBGENERA OF Pachyhalictus 1. Inner hind tibial spur of female coarse- ly pectinate with three to six long teeth; basitibial plate of male absent or nearly so; first two hind tarsal seg- ments of male fused te ee Sas ae, Pachyhalictus s. str. Inner hind tibial spur of female finely pectinate with more than 12 slender teeth; basitibial plate of male present; first two hind tarsal segments of male abeiculated (sete... ose ae Dictyohalictus Subgenus Pachyhalictus Cockerell, s. str. Figures 54-61 Halicti nomuformes Vachal, 1894, Ann. Mus. Civ. Genova, 34:428 (part); Blithgen, 1926, Zool. Jahrb., Abt. Syst., Geogr. Biol. Tiere, 51:400; Bluthgen, 1931, Zool. Jahrb., Abt. Syst., Geogr. Biol. Tiere, 61:286. Pachyhalictus Cockerell, 1929, Ann. Mag. Nat. Hist., (10)4:589. Type species: Halictus merescens Cockerell, 1919, original designation. Recurrent veins entering second and third sub- marginal cells near apices or the first interstitial. Third transverse cubital vein arcuate or nearly straight. Basitibial plate of male absent or indicated only apically. First two hind tarsal segments of male fused, but point of union indicated by strong con- striction. Fourth sternum of male with an enormous lateral bristle hidden by the tergum, at least in species that have been dissected. Inner hind tibial spur of female pectinate with three to six long, coarse teeth. Pachyhalictus s. str. is restricted to the Indoaustralian region, ranging from Cey- lon and India eastward to the Philippines and Taiwan and through Indonesia, New Guinea, to northern Australia. The follow- ing is a list of species names, all new com- binations in Pachyhalictus, based on Bliith- gen (1926, 1928, 1931) plus my examina- tions of type material. Synonymies are those of Blithgen, not re-evaluated here. Pachyhalictus (Pachyhalictus) assamicus (Bluthgen, 1926) Pachyhalictus (Pachyhalictus) bedanus (Bluthgen, 1926) *Pachyhalictus (Pachyhalictus) bihamatus (Bluthgen, 1926) Pachyhalictus (Pachyhalictus) binghami (Kirby, 1900) 518 *Pachyhalictus (Pachyhalictus) burmanus (Blithgen, 1926) Pachyhalictus (Pachyhalictus) buruanus (Blithgen, 1926) Pachyhalictus (Pachyhalictus) celebensis (Bliithgen, 1931) Pachyhalictus (Pachyhalictus) dapanensis (Blithgen, 1926) Pachyhalictus (Pachyhalictus) formosicola (Blithgen, 1926) *Pachyhalictus (Pachyhalictus) interstitialis (Cameron, 1903) *Pachyhalictus (Pachyhalictus) intricatus (Vachal, 1894) (=thoracicus Friese, 1914) Pachyhalictus (Pachyhalictus) javanus (Bluthgen, 1926) *Pachyhalictus (Pachyhalictus) kalutarae (Cockerell, 1911) —=amplicollis Friese, 1918) Pachyhalictus (Pachyhalictus) kocki (Blithgen, 1931) *Pachyhalictus (Pachyhalictus) liodomus (Vachal, 1894) =scopipes Friese, 1918) Pachyhalictus (Pachyhalictus) lombokensis (Bluthgen, 1926) Pachyhalictus (Pachyhalictus) merescens (Cockerell, 1919) Pachyhalictus (Pachyhalictus) murbanus (Blithgen, 1931) Pachyhalictus (Pachyhalictus) negriticus (Blithgen, 1926) Pachyhalictus (Pachyhalictus) penangensis (Bluthgen, 1926) Pachyhalictus (Pachyhalictus) pseudothoracicus (Blithgen, 1926) Pachyhalictus (Pachyhalictus) puangensis (Cockerell, 1937) *Pachyhalictus (Pachyhalictus) reticulosus (Dalla Torre, 1896) (=Halictus reticulatus Vachal, 1894, not Robertson, 1892) *Pachyhalictus (Pachyhalictus) sigirtellus (Cockerell, 1911) Pachyhalictus (Pachyhalictus) stirlingi (Cockerell, 1910) Tue University oF Kansas ScIENCE BULLETIN *Pachyhalictus (Pachyhalictus) sublustrans (Cockerell, 1919) Pachyhalictus (Pachyhalictus) trizonulus (Friese, 1909) *Pachyhalictus (Pachyhalictus) validus (Bingham, 1903) Pachyhalictus (Pachyhalictus) vanajus (Bluthgen, 1926) *Pachyhalictus (Pachyhalictus) vinctus (Walker, 1860) Of these species, P. binghami from Christmas Island in the Indian Ocean, is most distinctive, differing from ordinary Pachyhalictus in the less prominent reticu- late pattern on the head and thorax (scu- tum has shallow punctures and fine wrin- kles, thus intermediate between punctate and reticulate), and in the propodeal sur- face pattern (dorsal surface not areolate, with fine longitudinal wrinkles between which surface is dull; carinae margining posterior surface reaching only three fourths of distance to upper margin of that surface). Dictyohalictus new subgenus Figures 62-68 Type species: Halictus retigerus Cockerell, 1940. Recurrent veins entering second and third sub- marginal cells well before apices. Third transverse cubital vein slightly sinuate. Basitibial plate of male present. First two hind tarsal segments of male dis- tinct, although articulation broader than that between second and third segments. Fourth sternum of male with a slender, posteriorly directed lobe at each side, hidden by the tergum, but without a lateral bristle. Inner hind tibial spur of female finely pectinate with more than a dozen rather short, slender teeth. This subgenus is known only from southeastern Africa. It appears to contain only a single species, Pachyhalictus (Dic- tyohalictus) retigerus (Cockerell), new combination, but there are several syno- nyms as indicated in the Appendix. The name Dictyohalictus is based on diktyon, a net, plus Halictus, with refer- ence to the reticulate sculpturing of the head and thorax. CLASSIFICATION OF HALICTINE BEEs 519 Fics. 62-67. Pachyhalictus (Dictyohalictus) retigerus. 62, 63. Dorsal, ventral and lateral views of male genitalia. 64. Posterior dorsal view of male gono- stylus. 65. Median part of seventh sternum of male. 66. Fourth sternum of male. 67. Inner hind tibial spur of female. Genus Thrincohalictus Blithgen Figures 69-77 Thrincohalictus Bluthgen, 1955, Bull. Research Coun- cil Israel (B, Biol., Geol.), 5:20. Type species: Halictus prognathus Pérez, 1912, by original designation and monotypy. 1. Nonmetallic black, moderately robust, length 9-10 mm. 2. Punctation of the usual sort, rather fine and dense. 3. Clypeus strongly produced downward and protuberant forward, almost twice as broad as long (female) to little broader than long (male), with shining but roughened ground between irregular punctures, upper part flat in profile; angle at end of truncation weakly produced and rounded or absent in male. 4. Line between lower ends of eyes crossing above middle of clypeus (female) or entirely above clypeus (male). 5. Malar area conspicuous, about half as long as basal mandibular width (female) to longer than basal mandibular width (male). 6. Paraocular area extending down as strong lobe into clypeus. 7. Mouthparts long and slender, glossa linear and much exceeding palpi and galea, but galea elongate, post- palpal part about four times as long as wide, and labial palpus about half as long as glossa. 8. Pro- notum with horizontal, dorsal surface less than one half as long medially as flagellar width, tomentose, margined anteriorly by angle where pronotal surface becomes declivous. 9. Dorsolateral angle of pronotum obtuse, a rounded ridge, but no carina extending downward from it. 10. Anterior extremity of scutum strongly convex in profile, the largely impunctate, vertical, anterior zone rising well above pronotum and curving uninterruptedly onto dorsal surface. 11. Pre- episternal groove well developed. Metanotum tomen- tose anteriorly. 12. Dorsal surface of propodeum about as long as metanotum, triangular area ill- defined and rather finely areolate or striate. Area behind triangle and lateral and posterior propodeal surfaces with few short hairs in addition to long ones. Fic. 68. Face and wing of female of Pachyhalictus (Dictyohalictus) retigerus. Scale line = 1.0 mm. Posterior and lateral surfaces of propodeum not areo- late, separated by carinae only below. 13. Apical wing veins strong. Recurrent veins entering second and third submarginal cells at distal third or fourth. 14. Third submarginal cell somewhat elongate, third transverse cubital vein distinctly sinuate, being arcu- ate toward wing apex posteriorly and toward wing base anteriorly. 15. Stigma of moderate size. Mar- ginal cell of the usual shape, free part much less than twice as long as part subtended by marginal cells, apex pointed, separated from wing margin by about a vein width or less. 16. First metasomal ter- gum broader than long. 17. Basal tergal tomentum absent or nearly so. 18. Terga with apical bands of pale plumose hair as in Halictus. Tergal margins broadly depressed, with punctation finer than on discs of terga, both hairs and punctures ending before margins proper, which are smooth and bare; hairs of marginal bands not or scarcely directed laterally; integument of marginal bands translucent brownish. Male: 19. Apex of clypeus and areas on femora, tibiae, and tarsi yellowish white. 20. Labrum slightly over twice as wide as long, with small apical process; long bristles scattered over marginal part of process, not limited to marginal row; no keel. Mandible simple. 21. Flagellum elongate, segments over 1.5 times as long as wide except for first which is wider than long. 22. Basitibial plate absent. 23. First two hind tarsal segments articulated, base of second not or scarcely broader than base of third. 24. Meta- soma rather robust, but nearly parallel sided, seg- ments II and III widest. 25. Pygidial plate not defined by carina, its position occupied by a median projection or tubercle which has hairs like those of adjacent areas, but above which is an ill-defined bare area. 26. Sternum IV shorter than adjacent sterna and largely hidden, its posterior margin fringed with long, curved bristles directed posteriorly, lateral part with- out a special elongate lobe or bristle. Sternum V with apical margin broadly and strongly emarginate, extreme side with a pencile of extremely long, curved, apparently fused hairs hidden under sides of terga. 520 Tur University oF Kansas SCIENCE BULLETIN ‘ | Z a ~ c wi, » XIN \\ Weel 6 Sie) Wee SH 13 \ J a x " aS xX \ \ | | 15 Fics. 69-76. Thrincohalictus prognathus. 69-70. Dor- sal, ventral and lateral views of male genitalia. 71. Posterior lateral view of male gonostylus. 72. Inner hind tibial spur of female. 73. Seventh and eighth sterna of male. 74, 75. Labrum of female and male. 76. Fourth sternum of male. 27. Sternum VII a slender transverse bar with a median apical pointed process. Sternum VIII a trans- verse band, somewhat broadened and sclerotized me- dially. 28. Genitalia rather broad with narrowed base. Gonostylus rather simple, little over half length of gonocoxite, with large dense tuft of hairs near inner margin, with large, hairy “retrorse’’ ventral lobe which, however, is largely directed ventro distally, but with small arm directed mesobasally. Penis valve with inferior basal process broadly rounded apically. Female: 29. Scape reaching to middle of anterior ocellus. First flagellar segment slightly longer than wide. 30. Labrum with keeled apical process narrow; body of labrum rounded apically, about 1.3 times as wide as long. 31. Hind tibia and scopa of the usual form. 32. Basitibial plate rather large, about one fourth as long as tibia, distinctly margined, surface largely hairy. 33. Inner hind tibial spur pectinate with four or five long coarse teeth. Hind tibia with two strong, apical spines. 34. Sterna with rather long hairs, many of them with a few branches. Thrincohalictus resembles Halictus s. str, in its general appearance, apical tergal bands, yellow markings on the clypeus and legs of males, short and rather simple gonostylus of the male, etc. It differs from Halictus in the elongate head, including the malar area, labrum, and proboscis, the igs. Fic. 77. Thrincohalictus prognathus. Face of male, face and wing of female. Scale line = 1.0 mm. short fourth sternum of the male armed with coarse bristles apically, and the large retrorse lobe of the male genitalia. In these features it resembles Thrinchostoma, from which it differs, however, in so many fea- tures that there appears to be no close rela- tionship. The fourth sternum of the male and the reduced angles at the ends of the clypeal truncation of the female suggest a relationship to the Thrinchostoma-Patella- apis-Zonalictus-Pachyhalictus group rather than to Halctus and Lasioglossum. Among these genera, the similarity appears closest to Patellapis, some species of which have apical tergal hair bands and an elon- gate head and mouthparts (P. braunsella), but the simple gonostyli, shorter sternum VIII, almost complete lack of the pygidial plate of the male, yellow markings of the male, as well as the manner of head elon- gation (with long malar area) and other features distinguish Thrincohalictus from Patellapts. Thrincohalictus contains only one known species, T. prognathus (Pérez, 1912), new combination, which ranges from the Armenian S.S.R. and Iran to the Aegean islands (Chios) and south to Israel. Its distinctiveness from relatives in Africa and tropical Asia and its distribution is suggestive of another monotypic genus, Exoneuridia, the only north temperate al- lodapine bee. CLASSIFICATION OF HALICTINE BEEs 521 Genus Thrinchostoma Saussure Figures 2, 78-87 1. Nonmetallic black or part of metasoma and legs, or even whole body, yellowish red; large and rather slender, 8-16 mm long. 2. Punctation of the usual sort, rather fine and often dense. 3. Clypets strongly produced downward and strongly protuber- ant forward, 0.94 to 2.5 times as broad as long, with shining ground between punctures, upper part flat in profile. 4. Line between the lower ends of the eyes crossing clypeus above middle, much above the mid- dle except in T. afasciatum, or even entirely above clypeus. 5. Malar area conspicuous, as long as eye and nearly four times as long as basal mandibular width to less than one third as long as basal mandib- ular width. 6. Paraocular area extending down as a strong lobe into clypeus. 7. Mouthparts long and slender, glossa linear and much exceeding the short galea and palpi. 8. Pronotum with horizontal dorsal Fics. 78-85. Structures of Thrinchostoma. 78, 79. Dorsal, ventral and lateral views of male genitalia of T. (Eothrincostoma) producta. 80. Posterior lateral view of gonostylus of same. 81. Eighth sternum of same. 82. Inferior basal process of penis valve of T. (Thrinchostoma) sjostedi. 83. Posterior lateral view of male gonostylus of same. 84, 85. Inner hind tibial spur of female of T. (T.) afasciatum and T. (E.) producta. surface one half to twice as long medially as flagellar width, densely tomentose, margined anteriorly by an angle or carina, almost on same level as and not overhung by anterior part of scutum. 9. Dorsolateral angle of pronotum obtuse or right angular, a rounded ridge but no carina extending downward from it. 10. Anterior extremity of scutum gently convex, without subvertical zone separated by angle from most of scutal surface. 11. Pre-episternal groove less conspicuous than in most Halictinae, short and shal- low below scrobal groove. Metanotum incompletely tomentose. 12. Dorsal surface of propodeum much longer than metanotum, sometimes as long as scu- tellum. Dorsal, bare, triangular area of propodeum finely to coarsely striate to granular, rather small except in T. afasciatum, usually not extending later- ally as far as transmetanotal suture and pit. Area be- hind triangular area and lateral and posterior propo- deal surfaces with short hairs in addition to scattered long ones. Posterior surface of propodeum margined by carinae only below, rather sharply rounding onto dorsal surface. 13. Apical wing veins strong. Recur- rent veins usually both entering third submarginal cell, but first recurrent sometimes interstitial or enter- ing distal extremity of second submarginal cell. 14. Third submarginal cell of moderate length, third transverse cubital vein straight or in Eothrincostoma, sinuate. 15. Stigma rather small and slender. Mar- ginal cell rather slender, free part less than twice as long as part subtended by submarginal cells, apex minutely truncate and appendiculate. 16. First meta- somal tergum about as long as broad. 17. Basal tergal bands of tomentum as well as apical fasciae of plu- mose hairs absent. 18. Apical margins of terga I-IV to III-IV of females and I-V to III-V of males usually broadly depressed, impunctate, with golden to whitish simple hairs directed laterally, forming bands that are conspicuous in certain lights, except in T. afascta- tum. Male: 19. Commonly apex of clypeus, labrum, tarsi, sometimes tibiae and rest of clypeus, yellowish white. 20. Labrum with strong apical process mar- gined with bristles, without keel. Mandible simple. 21. Flagellum elongate, segments longer than broad, first shorter than the others, apical segment often flattened and curved (not expanded), sometimes pointed. (A few species of Thrinchostoma s. str. have male antennae only 12-segmented, but they are other- wise typical members of their species group, see Blithgen, 1930.) 22. Basitibial plate not or scarcely recognizable. Hind tibia with pallid inner apical enlargement which carries the tibial spurs far from basitarsus and from one another. 23. First two hind tarsal segments fused, a weak constriction indicating point of union. 24. Metasoma rather elongate, parallel sided or widest at third or fourth segment. 25. Py- gidial plate represented by a smooth, shining area, not delimited by a carina either posteriorly or later- ally, this smooth area curving over onto reflexed ven- tral part of tergum VII. 26. Sternum IV broadly emarginate posteriorly, median part much shortened and hidden or nearly so by III (except in T. orchi- darum), but lateral parts extending far posteriorly. Sternum V weakly to strongly emarginate apically, with basal transverse thickening or raised area, some- times spiculate or with large pegs or hooked bristles, this thickening often hidden under IV and absent in T. orchidarum; sternum VI often with basal ele- vation frequently exposed by emargination in V. 27. Sternum VII a slender, transverse bar with median apical pointed process. Sternum VIII large with broad, truncate apical process provided with hairs. 28. Genitalia with general form of a typical halictid, the gonobase somewhat narrow, the gonocoxites broad. Gonostylus large and elaborate, half as long to as long as gonocoxite, with ventral basal mem- branous lobe reflexed mesally or downward and dorsal mesal membranous flap or area. Penis valve 522 Tue University oF KANSAS SCIENCE BULLETIN Fic. 86. Top row: Thrinchostoma (Eothrincostoma) producta, face of male, face and wing of female. Middle row: Thrinchostoma (Thrinchostoma) sjostedi, face of male, face and wing of female. Bottom row: T. (T.) afasciatum, face and wing of holotype female. Scale line = 1.0 mm. with inferior basal process very slender, capitate or bifid. (Genitalia and hidden sterna examined only for Thrinchostoma s. str. and Eothrincostoma.) Female: 29, Scape reaching beyond anterior ocel- lus; first flagellar segment longer than wide. 30. Labrum with moderate to broad tapering apical proc- ess with keel; body of labrum less than twice as wide as long. 31. Outer surface of hind tibia with hairs largely simple, usually some branched hairs along upper margin and large curved branched hairs along basal half of lower margin. (Hairs most branched in Eothrincostoma, intermediate in Thrinchostoma s. str., nearly all simple in Diagonozus). 32. Basitibial plate small, triangular, elevated above surrounding area, but without marginal ridge, surface smooth, with scattered pits, or with large grooves. 33. Inner CLASSIFICATION OF HALICTINE BEEs 523 hind tibial spur rather finely serrate or with broad median tooth beyond which it is coarsely toothed. Hind tibia with one long, slender, tibial spine. 34. Sterna without distinctive scopa. The Asiatic species of Thrinchostoma were reviewed by Bltithgen (1926), the African species, by the same author (1930). Various species have been described since those dates, however. Key To THE SUBGENERA OF TArinchostoma 1. Forewing without an area of dense hairs along second transverse cubital vein, this vein simple and straight; first transverse cubital arising far from base of vein r and margin of stigma; sternum IV of male with a series of enormous simple setae arising from margin both medially under sternum III and laterally, under the lateral ex- tremity of tergum IV, where the setae aiovlargestes sates Eothrincostoma ... Forewing with an area of dense hairs near median part of second transverse cubital vein, in males these hairs form- ing conspicuous dark spot; second transverse cubital vein usually angulate or thickened medially, sometimes in- complete (not reaching marginal cell); first transverse cubital arising very near margin of stigma; sternum IV of male without coarse, specialized setae .......... 2 2. Head extraordinarily produced below eyes, malar area nearly as long as or longer than eye, several times as long as basal width of mandible ....Diagonozus .... Head only moderately produced below eyes, malar area much shorter than eye, three times as long as basal width Ol pmandible Ormlessie ee are a a reese Thrinchostoma s. str. Subgenus Eothrincostoma Blithgen Figures 78-81, 85, 86 Eothrincostoma Bliithgen, 1930, Mitt. Zool. Mus. Ber- lin, 15:501. Type species: Halictus torridus Smith, 1879, des- ignation of Sandhouse, 1943, Proc. U. S. Nat. Mus., 92:548. Clypeus moderately produced below lower ends of eyes; malar area about as long as wide. Fore wing without area of dense hairs on second transverse cubital vein. First transverse cubital arising well away from margin of stigma so that vein r is nearly as long as anterior margin of second submarginal cell; third submarginal cell strongly narrowed anteriorly (1.e., toward costal margin of wing), anterior margin less than half as long as posterior margin; second transverse cubital vein nearly straight, not angulate or thickened, complete. Inner hind tibial spur of female with inner margin rather finely and uniformly serrate, each tooth occupying about as much space as three of the very fine teeth on other margin. Fourth metasomal sternum of male with apical row of enormous bristles, bent near apices, smaller me- dially where they arise under the margin of sternum IJ, larger laterally where the bases are under the lateral margins of tergum IV, and smaller again near apices of lateral processes, also under lateral margins of tergum IV. Sternum V with a pair of similar large discal bristles. Sternum VI with preapical thick- ening which is densely hairy posteriorly, the hairy area narrowly divided by longitudinal hairless band. This subgenus is more like ordinary halictids than the other subgenera in its wing venation and lack of a hair spot on the wings. Moreover, its male antennae and other features do not exhibit the spe- cial features found in many species of Thrinchostoma proper. Eothrincostoma is presumably more primitive than and prob- ably ancestral to Thrinchostoma s. str. Eothrincostoma ranges widely over tropical Africa, southward to Natal. The following is a list of species: malelanum Cockerell, 1937 manyemae Cockerell, 1932 patricium (Strand, 1910) silvaticum Bluthgen, 1930 torridum (Smith, 1879) wellmani Cockerell, 1908 Subgenus Thrinchostoma Saussure s. str. Figures 2, 82-84, 86 Thrinchostoma Saussure, 1890, im A. Grandidier, Histoire Physique, Naturelle et Politique de Mada- gascar, 20(1):52. Type species: Thrinchostoma renitantely Saussure, 1890 (monobasic). Trichostoma Dalla Torre, 1896, Catalogus Hymenop- terorum, 10:381 (unnecessary emendation of Thrinchostoma); Friese, 1909, Die Bienen Afri- 524 Tue University oF Kansas ScIENCE BULLETIN kas, p. 150, 7m L. Schultze, Zoologische und An- thropologische Ergebnisse einer Forschungsreise im westlichen und zentralen Siidafrika, vol. 2, part 2. Thrincostoma Dalla Torre, 1896, Catalogus Hymenop- terorum, 10:641 (unnecessary emendation of Thrinchostoma). Trinchostoma Sladen, 1915, Canad. Ent., 44:214 (un- necessary emendation of Thrinchostoma). Rostratilapis Friese, 1914, Tijd. Ent., 57:26. Type species: Halictus (Rostratilapis) macro- gnathus Friese, 1914, designation of Sandhouse, 1943, Proc. U. S. Nat. Mus., 92:597. Nesothrincostoma Blithgen, 1933, Mitt. Zool. Mus. Berlin, 18:364. Type species: Thrinchostoma serricorne Blithgen, 1933 (monobasic). Clypeus but little (T. afasciatum) to moderately produced below lower ends of eyes; malar area one third to three times as long as wide. Fore wing with area of dense hairs near middle of second transverse cubital vein, these hairs forming conspicuous dark spot (minute in T. serricorne) in males. First trans- verse cubital vein arising very near to stigma so that vein r is short (about twice as long as vein width in T. serricorne) or virtually absent; third submar- ginal cell only moderately narrowed anteriorly (1.e., toward costal margin of wing), anterior margin over half as long as posterior margin; second transverse cubital usually at least slightly angulate medially in area of dense hairs, sometimes absent anterior to that point, usually also thickened medially. Inner margin of inner hind tibial spur markedly widened near middle by broad obtuse tooth, beyond which margin is coarsely toothed (Fig. 84 for afasciatum) to almost smooth and edentate. Sternum IV of male often with long setae on posterior lateral prolongations, but without row of very coarse setae, but such setae often present on basal thickening of sternum V. Sternum VI usually unmodified. T. orchidarum should be restudied and its genitalia examined for possible distinc- tive features (type in British Museum). In this species, which is more hairy than others, the sterna II to IV are similar in form; IV is broadly exposed, but somewhat shorter than the preceding ones. Sterna II to IV each bears a preapical fringe of long hairs, broken medially. Sternum V has no basal thickening, and has a dense continu- ous preapical fringe. Thus, the sterna are less modified than usual in the genus. This subgenus is widespread in tropical Africa southward to Natal, in Madagascar and in tropical Asia from south India and Assam eastward to Vietnam, Kalimantan and Java. The following is a list of included species: aciculatum Blithgen, 1928 afasciatum new species affine Bluthgen, 1928 albitarse Blithgen, 1933 amanicum (Strand, 1910) assamense Sladen, 1915 atrum Benoist, 1962 bequaerti Blithgen, 1930 and form ochropus Blithgen, 1930 bibundicum (Strand, 1910) and form tessmanmt Strand, 1912 bryanti Meade-Waldo, 1914 castaneum Benoist, 1945 conjungens Bluthgen, 1933 emini Bluthgen, 1930 flaviscapus Bluthgen, 1926 fulvipes Blithgen, 1933 fulvum Benoist, 1945 insulare Benoist, 1962 offre: Benoist, 1962 kandti Blithgen, 1930 lemurtae Cockerell, 1910 lualiense Cockerell, 1939 macrognathum Friese, 1914 and form brunneum Blithgen, 1926 michaelts Cockerell, 1932 millart Cockerell, 1916 mwangai Blithgen, 1930 nachtigali Bluthgen, 1930 obscurum Bluthgen, 1933 orchidarum Cockerell, 1908 othonnae Cockerell, 1908 perineti Benoist, 1962 petersi Blithgen, 1930 productum (Smith, 1853) renitantely Saussure, 1891 rugulosum Benoist, 1962 sakalavum Blithgen, 1930 serricorne Bluthgen, 1933 sjostedti (Friese, 1908) and form rufescens (Friese, 1908) sladent Cockerell, 1913 CLASSIFICATION OF HALICTINE BEEs 525 telekui Bluthgen, 1930 tokinense Bluthgen, 1926 ugandae Blithgen, 1930 umtaliense Cockerell, 1936 undulatum Cockerell, 1936 vachali Blithgen, 1930 wissmanni Bluthgen, 1930 Subgenus Diagonozus Enderlein Figure 87 Diagonozus Enderlein, 1903, Berlin Ent. Zeitschr., 48:35. Type species: Diagonozus bicometes En- derlein (monobasic). Lower part of head enormously produced so that head is about as long as thorax; clypeus entirely below lower ends of eyes; malar area about four times as long as basal width of mandible, nearly as long as eye to longer than eye. Fore wing as de- scribed for Thrinchostoma s. str., but third submar- ginal cell with anterior margin about half as long as posterior margin, second transverse cubital vein un- usually strongly angulate and thickened. Inner hind tibial spur as described for Thrinchostoma s. str. Fic. 87. Thrinchostoma (Diagonozus) lettow-vor- becki. Face and side view of head of female. Scale line = 1.0 mm. Males have not been seen by me al- though described by Blithgen (1930). Another distinctive character of the sub- genus is the more elongate pronotum, the dorsal surface of the elevated portion on the midline being considerably longer than the diameter of an ocellus. In other sub- genera this length is little if any longer than an ocellar diameter. The proboscis of Diagonozus is also extremely long, al- though relative to the very long head it is little or any longer than that of Thrincho- stoma s. Str. Diagonozus appears to be a derivative of Thrinchostoma s. str., recog- nizable primarily by the elongate head and proboscis. Diagonozus is known from tropical west Africa. Included species are as fol- lows: bicometes (Enderlein, 1903) ghesquteret Cockerell, 1932 guineense Bluthgen, 1930 lettow-vorbecki Blithgen, 1930 Genus Halictus Latreille Figures 3-5, 88-101 1. Nonmetallic black or with body dull greenish or bluish, metasoma sometimes partly or wholly red; body length 3.5 to 17 mm. 2. Punctation of the usual sort, often fine and dense, but surfaces some- times shining with widely separated punctures. 3. Clypeus of females and some males not much pro- duced or protuberant, but in some males strongly produced downward and protuberant forward, nearly four times as wide as long (e.g., female of H. squa- mosus) to less than twice as wide as long (males of various species), angle at end of truncation strongly produced in female, usually acute as seen from front, absent in male; surface usually with shining inter- spaces among punctures, usually distinctly more con- vex near apex than near base. 4. Line between lower ends of eyes usually crossing clypeus near or above middle, but variable, in H. squamosus near lower margin of clypeus. 5. Malar area linear or in some males nearly half as long as wide, widest near anterior margin. 6. Paraocular area not extending down as lobe into clypeus or at most forming an obtuse or right angular lobe. 7. Distal parts of pro- boscis short, glossa not much longer than labial palpi or postpalpal part of galea. 8. Pronotum with sub- horizontal dorsal surface medially about one third as wide as flagellar diameter, often minutely tomen- tose, margined anteriorly only by declivity of prono- tal surface and sometimes sloping anteriorly so that it merges with the declivous surface. 9. Dorsolateral angle of pronotum usually obtuse, sometimes right angular, not lamellate or strongly carinate, a ridge extending toward posterior lobe of pronotum, but usually not extending across lobe, another ridge (rounded in most Vestitohalictus) extending down from dorsolateral angle. 10. Anterior extremity of scutum strongly convex in profile, the largely impunc- tate vertical anterior zone rising well above pronotum, usually rounding onto dorsal surface, but rarely with a sharp angle between anterior and dorsal surfaces, 526 Tre University oF Kansas ScrENCE BULLETIN anterior zone strongly bilobed in H. squamosus be- cause of deeply impressed longitudinal median line of front part of scutum. 11. Pre-episternal groove often short and shallow below scrobal groove. Metanotum tomentose or not. 12. Dorsal surface of propodeum longer than metanotum, shorter than to as long as scutellum, striate to granular. Triangular area some- times defined by end of striate or granular zone, variable in extent. Lateral and posterior surfaces of propodeum without conspicuous short hairs in addi- tion to long ones, not areolate. Posterior surface of propodeum margined by carinae, if at all, only below, or rarely these carinae reaching upper end of pos- terior surface (as in H. sexcinctellus). 13. Apical wing veins strong. Recurrent veins entering second and third submarginal cells or rarely first recurrent nearly interstitial with second transverse cubital. 14. Third submarginal cell somewhat elongate, third transverse cubital strongly arcuate posteriorly, straight or nearly so anteriorly, the whole vein often sinuate. 15. Marginal cell rather slender to robust, free part equal to or longer than part subtended by submarginal cells, apex narrowly truncate to pointed on wing mar- gin. 16. First metasomal tergum broader than long or in some males longer than broad. 17. Terga with or without basal areas or bands of tomentum. 18. Terga with apical bands of pale plumose hair (some- times only laterally), or rarely entire surface uniformly covered with such hair. Apical margins of terga broadly depressed, with punctation somewhat finer than on more anterior parts of terga, but punctation continuous nearly to posterior borders, at least on more anterior terga, hairs not or not strongly directed laterally, posterior margins of terga dark to trans- parent. Disc of tergum II sometimes with oblique hairs. Male: 19. Clypeus, labrum, and legs usually with yellow areas, tegula sometimes with yellow. 20. Labrum over three times as wide as long, rarely less than twice as wide as long, fringed with bristles, without apical process. Mandible simple. 21. Fla- gellum elongate, middle segments well over 1.5 and often over 2.0 times as long as wide. 22. Basitibial plate absent or sometimes vaguely recognizable as a flattened area of the usual shape, but in no case defined by a carina. 23. First two, hind, tarsal seg- ments apparently articulated, but base of second broader than base of third. 24. Metasoma rather slender, usually parallel sided, segments I to IV al- most equally broad although II and III slightly wider than others. 25. Pygidial plate defined only at apex by a carina, the carina not curving forward and de- fining the plate laterally, smooth area in front of apical carina sometimes small, broader than long, but often extending far toward base of tergum form- ing a longitudinal shiny band, sometimes elevated to form a shining ridge, in H. maculatus narrowed to a longitudinal carina. 26. Sterna IV and V truncate (1.e., unmodified) to broadly emarginate, without coarse bristles. 27. Sterna VII and VIII each with median apical projection, sometimes as in H. macu- latus blunt and short, that of VIII almost absent in Fics. 88-99. Structures of Halictus. 88, 89. Dorsal, ventral and lateral views of male genitalia of Halictus (Halictus) ligatus. 90. Posterior lateral view of gono- stylus of same. 91. Seventh and eighth sterna of same. 92. Posterior lateral view of male gonostylus, H. (Seladonia) confusus. 93-95. Inner hind ubial spurs of females of H. (H.) ligatus, latisignatus, and maculatus. 96-99. Posterior lateral views of male gonostyli of H. (H.) scabiosae, quadricinctus, rubt- cundus, and patellatus. H. parallelus and even emarginate in H. farinosus, the projections usually hairless, but either or both may bear hairs. 28. Genitalia rather broad with somewhat narrow base. Gonostylus rather simple with hairy apical or subapical dorsal lobe or elaborate with one or two clumps of bristles on inner surface, some species (subgenus Seladomia and H. quadricinc- tus) with a “second stylus” arising ventrolateral to major one, this being the homologue of the retrorse lobe of many halictids but directed apically rather than basally; therefore gonostylus without retrorse ventral basal lobe. Penis valve moderately slender, inferior basal process rather slender, rounded apically, inconspicuous. Female: 29. Scape usually reaching posterior ocelli or even beyond, rarely only attaining anterior ocellus or in a few species (H. desertorum, placidulus) not reaching anterior ocellus. Second flagellar segment broader than long to at least as long as broad, first and middle flagellar segments broader than long to much longer than broad. 30. Labrum with tapering apical process with keel; body of labrum usually more than twice as broad as long, rarely less than twice as broad as long, bigibbous, the two gibbosities some- times merged. 31. Hind tibia and its scopa of the usual form. 32. Basitibial plate of moderate size, angular to rounded apically, margin elevated or ab- sent on anterior side in Vestitohalictus, surface dull, CLASSIFICATION OF HALICTINE BEEs 527 Taste 1. Major Differences between Halictus and Lasioglossum sl. Feature distal wing veins strong apical tergal hair bands inferior basal process of penis valve inferior basal retrorse lobe of male gonostylus with some hairs. 33. Inner hind tibial spur coarsely serrate to pectinate. Hind tibia with one or two apical spines. 34. Sternal hairs plumose, of moderate length. Unlike the other genera described in this paper, Halictus is primarily developed in the palearctic region, where it contains many species. One subgenus (Seladonia) is abundant throughout Africa and a very few species reach southern India. In the Western Hemisphere only about ten spe- cies occur in North and Central America and only three reach South America al- though one Se/adonia is found as far south as Brazil. Halictus is closely related to the major group of halictines in which the distal wing veins are weakened, the genus Lasio- glossum s.1. Lasioglossum is not treated in this paper, but it seems important to indi- cate some of the main features by which Halictus and Lasioglossum differ, since some authors still prefer to unite these genera. Distinctive features are indicated in Table 1. Key To THE SUBGENERA OF Halictus 1. Basitibial plate of female slender and not or feebly defined on anterior mar- gin; hairless triangular area of pro- podeum small, not reaching posterior margin of dorsal surface of propodeum even medially. (Pubescence unusually dense and white; integument black or EGCHISH Petes. feet Pee Vestitohalictus Halictus of densely plumose hairs inconspicuous, slender rounded at apex absent or if present, directed apically, not retrorse Lasioglossum weak absent or rarely of weakly plumose hairs broad, truncate or obliquely truncate commonly present, membranous, minutely hairy Basitibial plate of female of the usual shape and defined by a ridge or carina anteriorly as well as posteriorly; hair- less triangular area of propodeum larger, nearly always reaching poste- rior dechvity medially... == 2 bo Integument nonmetallic, black; gono- stylus of male single (double in H. quadricinctus and allies) .......... Halictus Integument, at least of head and tho- rax, metallic greenish or bluish; gono- stylus of male double; i.e., two appar- ently separate gonostyli on apex of gonocoxite, the outer inferior one equivalent to the retrorse lobe of some halictids, but directed apically and in H. hesperus and lutescens much re- GUGCEG ..2 tl eee ee Seladonia Subgenus Seladonia Robertson Figures 6, 92 Seladonia Robertson, 1918, Ent. News, 29:91. Type species: Apis seladonia Fabricius, 1794, by original designation. Pachyceble Moure, 1940, Arq. Zool. Est. Sao Paulo, DAs Type species: Pachyceble lane: Moure, 1940, by original designation and monotypy. Length 4.5 to 10 mm. Integument of body dull greenish, sometimes bluish or brassy, that of meta- soma sometimes nonmetallic black or brownish. Pu- bescence not especially dense or widespread although occasionally as in H. niveocinctulus tending to spread between basal and apical tergal bands and thus sug- gestive of Vestitohalictus. Ridge extending down from lateral angle of pronotum sharply angulate or carinate. Apex of marginal cell pointed. Basitibial plate of female defined by a ridge both anteriorly and posteriorly; inner hind tibial spur of female pectinate 528 Ture University oF Kansas SciENCE BULLETIN with long or short teeth. Triangular area of propo- deum ample in size, reaching posterior declivity me- dially, not margined by densely punctate area. Male gonostylus double (outer, inferior “‘stylus’’ reduced, but present in H. lutescens and H. hesperus of the American tropics); major gonostylus usually with a clump of coarse setae on inner surface. This is the most widespread subgenus of Halictus, being found in the Holarctic region, south in the Western Hemisphere to Brazil, in Africa to the Cape of Good Hope, and into India and Southeast Asia. It is morphologically compact and unified. The following is a list of species names placed in Seladonia: abuensis Cameron, 1908 adolphi-frederict Strand, 1911 aeneobrunneus Pérez, 1895 aerarius Smith, 1873 atroviridis Cameron, 1906 austrovagans Cockerell, 1932 banalianus Strand, 1911 benguellensis Cockerell, 1908 caelestis Ebmer, 1976 candescens Cockerell, 1945 capensis Friese, 1909 centrosus Vachal, 1910 cephalicus Morawitz, 1873, and form neuter Blithgen, 1923 chloropinus Cockerell, 1946 confusus Smith, 1853 and forms arapahonum Cockerell, 1906, alpinus Alfken, 1907, and perkins: Bluthgen, 1925 daturae Cockerell, 1929 diductus Cockerell, 1932 dissensis Cockerell, 1945 *dissidens Pérez, 1903 duplocinctulus Cockerell, 1940 eruditus Cockerell, 1924 expertus Cockerell, 1916 *exquisitus Warncke, 1975 ferripennis Cockerell, 1929 *gaschunicus Blithgen, 1935 *gavarnicus Pérez, 1903 gemmeus Dours, 1872 harmonius Sandhouse, 1941 hesperus Smith, 1862 hotont Vachal, 1903 jucundtformis Cockerell, 1940 jucundus Smith, 1853 kesslert Bramson, 1879 and form nebulosus Warncke, 1975 komensts Cockerell, 1939 lanei (Moure, 1940) laosina Cockerell, 1929 laticinctulus Cockerell, 1946 leucaheneus Ebmer, 1972 and form arenosus Ebmer, 1976 lucidipennis Smith, 1853 lutescens Friese, 1921 medanicus Cockerell, 1945 medaniellus Cockerell, 1945 *meridionalis Morawitz, 1873 mogrenensis Cockerell, 1945 *mondaensis Bluthgen, 1923 *mongolicus Morawitz, 1880 *morinellus Warncke, 1975 *mugodyjaricus Blithgen, 1933 nikkoensis Cockerell, 1911 niloticus Smith, 1879 niveocinctulus Cockerell, 1940 *occipitalis Ebmer, 1972 pervirens Cockerell, 1940 *petraeus Blithgen, 1933 *pjalmensis Strand, 1909 *pontificus Cockerell, 1940 propinquus Smith, 1853 pruinescens Cockerell, 1937 *secundus Dalla Torre, 1896 seladonius (Fabricius, 1794) seminiger Cockerell, 1937 semitectus Morawitz, 1873 silvaticus Bluthgen, 1926 smaragdulus Vachal, 1895 speculiferus Cockerell, 1929 subauratoides Blithgen, 1926 subauratus (Rossi, 1792) and its forms corsa Bluthgen, 1933, and syrius Bluthgen, 1933 subincertus Cockerell, 1940 *subpetraeus Blithgen, 1933 sudanicus Cockerell, 1945 CLASSIFICATION OF HALICTINE BEES 529 tataricus Bluthgen, 1933 tibetanus Bluthgen, 1926 tokarensis Cockerell, 1945 tokartellus Cockerell, 1945 *transbatkalensis Bluthgen, 1933 trichiurus Cockerell, 1940 tripartitus Cockerell, 1895 tumulorum (Linnaeus, 1758) *ymbrosus Cockerell, 1929 vansont Cockerell, 1935 varentzou1 Morawitz, 1895 varipes Morawitz, 1876 vernalis Smith, 1879 *verticalis Bluthgen, 1931 vicinus Vachal, 1895 virgatellus Cockerell, 1901 viridibasis Cockerell, 1946 *wollmanni Blithgen, 1933 Subgenus Vestitohalictus Blithgen Figure 100 Vestitohalictus Blithgen, 1961, Beitr. Naturk. Forsch. SW-Deutschl., 19:287. Type species: Halictus vestitus Lepeletier, 1841, by original designation. (According to Ebmer, 1976b, this was a misidentification; Blithgen actually had H. pulvereus Morawtiz. From the practical viewpoint this is of no importance, since pulvereus and vestitus are similar species of the same subgenus. The type should stand as H. vestitus.) Length 3.5 to 8 mm. Integument of body dull greenish to entirely non-metallic and black, often with metasoma partly or wholly red. Pubescence dense, white or yellowish, commonly covering en- trrely metasomal surface, although denser on posterior margins than elsewhere. Ridge extending down from lateral angle of pronotum rounded. Apex of marginal cell pointed on wing margin or apex separated by a vein width or more from margin. Basitibial plate of female undefined anteriorly, or if defined, plate nar- row and pointed below rather than broad as in other subgenera; inner hind tibial spur of female pectinate with long or short teeth. Triangular area of propo- deum small, short, not reaching posterior declivity, usually margined posteriorly and laterally by densely punctate hairy area. Male gonostylus double as in Seladonia or the outer “‘stylus”’ absent. This subgenus is found in the drier parts of the Palearctic region, from the Canary Islands and the Mediterranean ba- sin to western China. It includes minute as well as moderate sized species, some al- i | u Fic. 100. Halictus (Vestitohalictus) nasica, face and wing of female. Scale line = 1.0 mm. most wholly white because of dense pu- bescence, many of them with the meta- somal integument partly or wholly red. There is much diversity in the subgenus as here delimited. The type species and its close relatives (including forms with non- metallic as well as with greenish integu- ment) have a median apical tuft or longi- tudinal band of dense hair on the fourth sternum of the male, but this is absent in other species. The labrum of the female in some species is quite ordinary, but in others the body of the labrum is much longer than usual (e.g., H. desertorum, in which it is two thirds as long as wide). The labral process is broad and very long, about twice as long as the body of the labrum in the female of H. nasica. In males, also, the labrum is sometimes longer than in other Halictus, only somewhat over twice as wide as long as in H. desertorum. Variation in the male gonostylus and in the basitibial plate of the female is indi- cated in the subgeneric description above. The most conspicuously strange feature of any female Halictus is the clypeus of the minute H. nasica which bears a_ long, downward projecting median process (Fig. 100). A. W. Ebmer (in litt., 1977) questions the placement of H. semiticus and H. pla- 530 Tue University oF KAnsAs SCIENCE BULLETIN cidulus in the following list, and says that their male genitalia are similar to those of Seladonia. It may be that the problem arises in part from misassociation of sexes of H. placidulus, for Ebmer indicates that the female holotype has a small propodeal triangle as in Vestitohalictus while the male has characters suggesting a relation- ship with H. (Seladonia) varentzowt. It well may be that some species intergrade with Seladonia. In general, however, the two subgenera seem quite distinct. The following is a list of species that fall in the subgenus Vestitohalictus. *aenescens (Radoszkowski, 1893) *balearicus Pérez, 1903 *bulbiceps Bluthgen, 1929 *concinnus Brullé, 1840 cupidus Vachal, 1902 cypricus Bluthgen, 1937 desertorum Morawitz, 1876 *fuscicollis Morawitz, 1876 and form transcaspica Bluthgen, 1923 *indefinitus Bluthgen, 1923 *inpilosus Ebmer, 1975 *kuschkensis Ebmer, 1975 *muicrocardia Pérez, 1895 morawitzi Vachal, 1902 and form theseus Ebmer, 1975 mordacellus Blithgen, 1929 *mordax Blithgen, 1923 mucidus Blithgen, 1923 mucoreus (Eversmann, 1852) nasica Morawitz, 1876 *ochropus Blithgen, 1923 persephone Ebmer, 1976 pici Pérez, 1895 placidulus Blithgen, 1923 pollinosus Sichel, 1860 and its forms limissicus Blithgen, 1937 and thevestensis Pérez, 1903 pulvereus Morawitz, 1873, and its form tectus Radoszkowsky, 1876 *pseudomucoreus Ebmer, 1975 pseudovestitus Blithgen, 1925 radoszkovsku Vachal, 1902 *semiticus Bluthgen, 1955 sogdianus Morawitz, 1876 *solitudinis Ebmer, 1975 * surabadensis Ebmer, 1975 *tuberculatus Blithgen, 1925 *vestitus Lepeletier, 1841 Subgenus Halictus Latreille s. str. Figures 3-5, 88-91, 93-99, 101 Halictus Latreille, 1804, Nouv. Dict. Hist. Nat., 24: 182. Type species: Aprs quadricincta Fabricius, 1776, by designation of Richards, 1935 (see below). Odontalictus Robertson, 1918, Ent. News, 29:91. Type species: Halictus ligatus Say, 1837, mono- basic and by original designation. Monilapis Cockerell, 1931, Ann. Mag. Nat. Hist., (10)7:529. Type species: Hylaeus tomentosus Eversmann, 1852, monobasic and by original designation. The type species for the name Halictus has been a subject of much discussion. The following designations and interpretations exis: I. Apis sexcincta Fabricius, 1775, “ejusd.” Andrena rufipes Fabricius, 1793, designation by Latreille, 1810, Considérations générales . . . des insectes. -p. 459. 2. Apis rubicunda Christ, 1791, desig- nation by Curtis, 1833, British Ento- mology, 10:448a. Apis quadricincta Fabricius, 1776, designation by Richards, 1935, Trans. Royal Ent. Soc. London, 83:170. 4. Andrena rufipes auctorum, nec Fa- bricius = Apis sexcincta Fabricius, 1775. This is the interpretation of Latreille’s designation by Warncke (1975). 5. Andrena rufipes Fabricius, 1793. This is the interpretation of La- treille’s designation by Ebmer (1974, 1976a). ios) The only species included by Latreille in 1804 were rufipes, quadricinctus, and flavipes. Designation number 2 is there- CLASSIFICATION OF HALICTINE BEES 551 fore invalid for it clearly involves a species not originally included. The problems center around designa- tion number 1, of which numbers 4 and 5 are interpretations. This designation is in- validated by Opinion number 136 of the International Commission on Zoological Nomenclature (1939), which takes the po- sition that when Latreille in his tabulation of 1810 listed two or more trivial names, there was no type designation. Even if one ignores Opinion 136, the conclusion is the same. The abbreviation “ejusd.” in Designation 1 is for ejsdem or ejusdemmodi, meaning “in the same way.” One might assume that this means “the same species,” and that Latreille was there- fore synonymizing rufipes, an originally included name, with sexcinctus, which was not included by name, but has priority, at the same time that he stated the type species. The International Code of Zoolog- ical Nomenclature [Article 69, (a) (iv) ] states that if an author designates a type species using a name that was not origin- ally included, but at the same time synony- mizes that name with one of the originally included species, the designation of the latter as type species is valid. Thus La- treille’s act would be considered as designa- tion of Apis sexcinctus as the type species. It is irrelevant that the so-called type speci- men of Andrena rufipes is a wasp which does not agree with the original description at all well (Ebmer, 1976a). There is no need to draw the distinction that Warncke (1975) makes between rufipes acutorum (the bee) and rafipes Fabricius (the wasp), for the wasp with the label “rafipes” must be a result of a probably post-Fabrician error. Under the circumstances, it is also irrelevant that rufipes and sexcinctus are not now considered synonymous (Ebmer, 1974, 1976a),. In reality, Latreille (1810) did not use “equsd.” to indicate synonymy. He listed together species that were not at all alike, but that agreed in what he considered as generic characters. For example, for the genus Megachile he lists muraria Fab. equsd. lanata, argentata, and centuncularts. These are extremely different looking spe- cies; he could not have been suggesting specific synonymy. The same is true for Centris where he lists haemorrhoidalis equsd. versicolor, two differently colored and clearly nonsynonymous species. Thus for Halictus, he was evidently saying “the type is sexcinctus, and rufipes also belongs here.” Since sexcinctus was not an orig- inally included species, Latreille’s “designa- tion” is invalid. Ebmer (1976a) has argued that since Latreille, in indicating the type species, listed two species, only one of which was originally included, that one (rufipes) is thereby designated as the type. This view does not appear to be justified by Article 69 of the International Code of Zoological Nomenclature. Moreover, as already indi- cated, in view of Opinion 136, all such con- siderations are irrelevant in any event. Presumably, it was for the reasons out- lined above that Richards made the only valid type designation, number 3 above, the species being Apis quadricincta Fa- bricius. Warncke (1970, see also 1975) desig- nated the same Apis quadricincta Fabricius as the type species of Hylaews, a name pro- posed by Fabricius in 1793. This designa- tion would have the effect of making Hylaeus available as a senior synonym of Halictus. Warncke’s designation is invalid since Latreille in 1810 designated a differ- ent species, Apis annulata Linnaeus, as the type species of Hylaeus. This is a species belonging to the genus known in most parts of the world today as Hylaeus. La- treille’s designation may have been unfor- tunate at the time, for annulata was the only species of its genus included by Fabri- cius under the name Hy/laeus, compared to six species of Halictinae, and the name 532 Tue University oF KANsAs ScIENCE BULLETIN Hylaeus was widely although not unt- formly used at one time for the group now known as Halictini. Nonetheless, Apzs annulata was one of the species originally included in Hylaeus and the designation is valid. There is no legitimacy to Warncke’s argument that Hylaeus of Latreille is a different genus with a different type species from Hylaeus of Fabricius. Length 6 to 17 mm. Integument of body non- metallic, black or brownish, the metasoma rarely partly red. Pubescence not especially dense or wide- spread, metasomal terga usually without basal bands of hair but with apical bands only, in the H. senilis group hair dense, widespread, often white. Ridge extending down from lateral angle of propotum sharply angulate or carinate. Apex of marginal cell minutely truncate to pointed on wing margin. Basi- tibial plate of female defined by a ridge both ante- riorly and posteriorly; inner hind tibial spur of female coarse serrate to short pectinate, or the teeth long in H. latisignatus. Triangular area of propodeum ample in size, reaching posterior declivity medially, not mar- gined by densely punctate area. Male gonostylus usu- ally not double, with or without one or two tufts of coarse setae on inner surface, gonostylus double (i.e., with the equivalent of the retrorse lobe projecting dis- tally) only in H. quadricinctus and its immediate allies such as H. brunnescens. The subgenus Halictus is abundant in the Palearctic region. It does not occur, however, in subsaharan Africa or in south- east Asia and only one species (H. latisig- natus) reaches southern India. Only four species occur in North America. One of them, H. rubicundus, is Holarctic and one, H. ligatus, extends southward into the Neotropical region as far as Colombia and Trinidad. The subgenus Halictus contains several diverse elements, probably as different from one another as they are from Sela- donia. The latter subgenus, however, is easily recognized in both sexes by its green- ish coloration while the groups included in Halctus proper are all non-metallic and the females are difficult to segregate into groups. Since for many species, only fe- males are known or males have not been available for dissection, I have not been able to place numerous names as to group. If the subgenus were divided now, many species would therefore not be assignable to subgenus. For this reason, subdivision has not been formally proposed. The groups, however, are distinguishable by the characters of males listed below, and are numbered 1 to 4. These numbers in front of names in the list of species indicate the groups to which certain species belong. (Since writing the above, A. W. Ebmer of Linz, Austria, the principal specialist on Palearctic halictines, has been kind enough to examine my groupings. In general, he agrees with them and has placed nearly all the species not only in these groups, but in subdivisions thereof. I leave to him the full account of groups or subgenera and placement of the species.) Group 1 Mandible broadened basally. First flagellar seg- ment much broader than long; flagellum somewhat moniliform. Hypostomal area concave. Malar space present. Sternum IV with apical margin broadly concave, sternum longest at extreme sides. Gono- stylus not double, somewhat expanded apically, usu- ally with a clump of long, coarse setae on inner sur- face, but such setae absent in some species such as H. simplex. The name Monilapis is available for Group | and could be used in a subgeneric sense except for the problem of placing species known only from females, as men- tioned above. The name “tetrazonius group” has usually been used. It is a com- pact, Palearctic group characterized by a series of derived features. The clump of coarse setae usually arising from the inner surface of the gonostylus is probably ho- mologous to the coarse setae on the basal extension of the preapical hairy lobe in Group 3. It is not homologous to the clump of specialized, flattened setae found in Group 3 and Seladonia. Group 2 Mandible not broadened basally. First flagellar segment slightly broader than long to longer than broad; flagellum not moniliform. Hypostomal area not concave. Malar space variable. Sternum IV sim- CLASSIFICATION OF HALICTINE BEEs 533 ple or with margin concave, longest sublaterally so that if margin is concave, the posteriormost angles are mesal to the sides of the sternum. Gonostylus not double, relatively simple, with hairy inner apical or preapical lobe, but without a clump of long coarse setae on inner surface. The name Odontalictus is available for this group, which is restricted to the Pale- arctic region except for H. ligatus which is American. The genal tooth of the female of H. ligatus, which led Robertson to pro- vide the name Odontalictus, is not a sub- generic or group character. It is also found in some unrelated Palearctic species such as H. modernus and submodernus and even in H. (Seladonia) wollmanni. H. It- gatus is in the scabiosae subgroup of my Group 2. Group 3 Agrees with Group 2 except as follows: sternum IV, if concave, longest at lateral margins as in Group 1. Gonostylus complex, with preapical, hairy lobe which often projects both basally and apically, and with clump of coarse, flattened setae on inner surface basal to the lobe; sometimes (e.g., in H. rubicundus) with small outer inferior ‘‘stylus,’ this fully devel- oped so that the stylus appears double in H. quad- ricinctus and its allies. The double gonostylus of H. guadri- cinctus and its allies is suggestive of that of Seladonia, as is the clump of specialized setae arising from the inner surface of the gonostylus, but the large, nonmetallic spe- Fic. 101. Halictus (Halictus) farinosus, face of male, face and wing of female. Scale line = 1.0 mm. cies of Group 3 and the small, greenish species of Seladonia do not superficially appear closely related. The name Halictus s. str. is available for this group, which is restricted to the Palearctic region except for three species found in North America. Group 4 Mandible not broadened basally. First flagellar segment longer than broad, flagellum not moniliform. Hypostomal area not concave. Malar area linear. Sternum IV with apical margin concave, sternum widest laterally; V deeply emarginate, VI with large median hairy area. Gonostylus not double, with broad, scarcely hairy, thin lobe extending downward and slender apical process projecting in same direc- tion, with clump of long coarse setae on inner surface, but this displaced basad relative to other species so that it arises basal to apex of gonocoxite. This group contains only the Indian species H. latisignatus, which is distin- guishable in the female by the small me- dian elevation on the apical margin of the clypeus. The distinctive features have been well illustrated by Sakagami and Wain (1966). The following is a list of species of the subgenus Halictus: *acrocephalus Bluthgen, 1923 *adpikenticus Bluthgen, 1923 (3) aegyptiacus Friese, 1916 (1) aegypticola Strand, 1909 *albohispidus Blithgen, 1923 *qlbozonatus Dours, 1872 *alfkenellus Strand, 1909 *gltaicus Pérez, 1903 asperatus Bingham, 1898 asperulus Pérez, 1895 *atripes Morawitz, 1894 *aureipes Dours, 1872 *bagirensis Bluthgen, 1936 (2) *berlandi Blithgen, 1936 (1) *bifidus Warncke, 1975 *brunnescens (Eversmann, 1852) *bucharicus Blithgen, 1936 (1)*carinthiacus Blithgen, 1936 (2)*cedens Blithgen, 1931 (2) cochlearitarsis (Dours, 1872) *consobrinus Pérez, 1895 (2) constantinensis Strand, 1910 (2) constrictus Smith, 1853 (1)*crenicornis Blithgen, 1923 *cyrenaicus Bluthgen, 1930, *determinandus Dalla Torre, 1896 *dschulfensis Bluthgen, 1936 *dunganicus Bluthgen, 1936 (1) *eurygnathopsis Blithgen, 1936 (1) eurygnathus Bluthgen, 1930 (3) farinosus Smith, 1853 (2)*fatsensis Bluthgen, 1936 *fimbriatus Smith, 1853 formosus Dours, 1872 (2) frontalis Smith, 1853 fucosus Morawitz, 1876 (2) fulvipes (Klug, 1817) *fumatipennis Blithgen, 1924 *funerarius Morawitz, 1876 (1)*furcatus Bluthgen, 1925 *georgicus Bluthgen, 1936 *gordius Warncke, 1975 (2)*graecus Blithgen, 1936 (1) *grunwaldti Ebmer, 1975 *gusenleitnert Ebmer, 1975 *hedini Bluthgen, 1935 holomelaenus Bluthgen, 1936 (2)*humkalensis Bluthgen, 1936 (2) *hybridopsis Bluthgen, 1923 intumescens Pérez, 1895 *jaramielicus Blithgen, 1923 kusdasi Ebmer, 1975 (1) langobardicus Blithgen, 1944 (4) latisignatus Cameron, 1908 *libanensis Pérez, 1911 (2) ligatus Say, 1837 (2) *luganicus Blithgen, 1936 *lunatus Warncke, 1975 *lussinicus Bluthgen, 1935 (2) maculatus Smith, 1848 and form priesnert Ebmer, 1975 *marchali Vachal, 1891 maroccanus Bluthgen, 1933 *mediterranellus Strand, 1909 *minor Morawitz, 1876 *modernus Morawitz, 1876 (2) nadigi Blithgen, 1933 nicosiae Bluthgen, 1923 *ochraceovittatus Dours, 1872 *palustris Morawitz, 1876 534 Tue University oF KAnsAs SciENCE BULLETIN (3) parallelus Say, 1837 (1) patellatus Morawitz, 1873 *pentheri Bluthgen, 1924 (1)*ponticus Bluthgen, 1936 *pseudomaculatus Bluthgen, 1925 * pseudotetrazonius Strand, 1921 (1) pyrenaeus Pérez, 1903 (2)*quadricinctoides Blithgen, 1936 (3) quadrictnctus (Fabricius, 1776) (1) *quadripartitus Blithgen, 1923 (3) rubicundus (Christ, 1791) and forms mongolensis Blithgen, 1936, laticinctus Blithgen, 1923, and lerouxu Lepeletier, 1841 (3) rufipes (Fabricius, 1793) sajoi Bluthgen, 1923 (1)*samarensis Bluthgen, 1936 (2) scabtosae (Rossi, 1790) and form powell: Cockerell, 1931 (1) *scardicus Bluthgen, 1936 *sefidicus Bliithgen, 1936 senilis (Eversmann, 1852) sepositus Cockerell, 1921 (2) sexcinctus (Fabricius, 1775) (1) *siculus Bluthgen, 1925 (1) simplex Blithgen, 1923 (=zbex Warncke, 1973) squamosus Lebedev, 1910 *stachu Bluthgen, 1923 (2)*subalfkenellus Blithgen, 1936 submodernus Bluthgen, 1936 (2)*subsenilis Bluthgen, 1955 *takuiricus Bluthgen, 1936 (1) tetrazonianellus Strand, 1909 (1) tetrazonius Klug, 1817 tibialis Walker, 1871 *tomentosus (Eversmann, 1852) *tridivisus Blithgen, 1923 *tsingtouensis Strand, 1910 *turanicola Dalla Torre, 1896 (2) *turkomannus Pérez, 1903 *wagnert Blithgen, 1937 (1) wyernicus Bluthgen, 1936 yarkandensts Strand, 1909 CLASSIFICATION OF HALICTINE BEES 535 ACKNOWLEDGEMENTS I am indebted to numerous museums for the loan of material used in this study, in particular to the authorities of the Brit- ish Museum (Natural History), and to Dr. E. Konigsmann of the Zoologisches Mu- seum, Humboldt-Universitat in Berlin. Other museums that generously lent ma- terial of importance are the Transvaal Museum (Pretoria), the South African Museum (Cape Town), the American Mu- seum of Natural History (New York), the Bishop Museum (Honolulu) and the National Museum of Natural History (Washington). Dr. Y. Hirashima, Ento- mological Laboratory, Kyushu University (Japan), also lent important material. I am especially pleased to acknowledge the help of P. Andreas W. Ebmer of Linz, Austria, with regard to the lists of species of Halictus. Not wishing to detract from his future publications, I have not incorpo- rated all the information which he pro- vided relative to groupings, synonymies, and the like, but I have incorporated many additions and corrections received from him. This study was made possible by grant no. DEB 73-06815 A04 from the National Science Foundation. APPENDIX The following are new species described so that their characters can be incorporated into the descriptions in the body of this paper, plus certain other taxonomic notes that relate to these bees. Patellapts (Patellapis) braunsella new species Figures 27, 29-33, 44 In its elongate bead and associated fea- tures such as the long glossa and the lobe of the paraocular area cutting into the clypeus, this species differs from all other Patellapis. P. (Lomatalictus) pastina has a moderately elongate head and a some- what long glossa, but these features are less extreme than in P. braunsella and must be independently evolved. P. braunsella, as shown by the subgeneric characters, is more closely related to P. schultzi and P. minutior. It is the size of the latter, but differs in many ways including the head shape and associated characters listed above and the more elongate and crenulate an- tennal flagellum. Female: Length 8 mm; forewing length 6.5 mm. Black with dark brown on middle of mandible, under side of flagellum, and small segments of tarsi; apices of metasomal terga and sterna II to IV broadly pallid translucent. Wings clear, veins and stigma rather light brown. Pubescence dull white, moderately abundant and long, plumose but not as heavily so as in P. mala- churina, especially long (much longer than eye width) on genal area; longest hairs on basal half of scape nearly half as long as scape; first metasomal tergum with apical white hair band restricted to sides; base of tergum II, especially laterally, with scattered plu- mose white hairs (the closest approach to basal bands of tomentum found in the genus); terga II-IV with broad, well defined, dense apical hair bands. Hair of fifth tergum orange red, fading to white laterally. Tibial and tarsal hairs yellowish white, yellower on under sides of tarsi; clypeal fringe and hairs of mandi- ble similarly yellow; penicillus orange yellow. Head slightly longer than broad (188:162); upper and lower interorbital distances as 104:100. Clypeus slightly over twice as wide as long (98:42); line between lower ends of eyes crossing clypeus near up- per margin; paraocular area extending down into clypeus as an approximately right angular lobe. Malar space linear. Inner orbits convergent below, except for upper parts, which are convergent above. Antennal sockets separated by less than diameter of a socket. Antennocular:interantennal:antennocellar:interocellar: ocellocular distances as 32:12:52:39:24. Labrum with convex body about twice as wide as long and apical pointed process shorter than body. Frontal carina ending well below level of lower margins of antennal sockets. Upper part of genal area wider than eye, area widest at upper third of eye and narrowing to almost nothing in lateral view at lower end of eye. Glossa as long as head. Scape reaching to level of upper margin of lateral ocellus; first flagellar segment about as broad as long, second broader than long, others longer than broad. Dorsolateral pronotal angles obtuse; a ridge, but no carina, extending across pronotal lobe. Dorsal surface of propodeum shorter than scutellum, separated from posterior surface by a moderately sharp angle, no distinct carina defining posterior surface laterally although several minute ridges mark the lateral limit of that surface below. Basitibial plate 536 Tue University oF KANnsAs ScIENCE BULLETIN rather narrowly rounded apically. Inner margin of inner hind tibial spur minutely serrate-pectinate or ciliate. Clypeus and lower part of paraocular area shining with irregular, large, well separated punctures. Supra- clypeal area minutely roughened, dull, with smaller punctures separated by over a puncture width. Frons and vertex finely and densely strigose-punctate. Genal area more coarsely and shallowly strigose. Hypo- stomal area nearly smooth, shining, flat. Scutum dull, minutely and closely punctured; scutellum and metanotum with much coarser punctures on a mi- nutely roughened but shining ground. Sides of tho- rax and propodeum minutely reticulate or punctate with scattered large, shallow punctures. Dorsal part of propodeum minutely roughened and dull, the tri- angular area with a coarser pattern of fine, radiating striae laterally, medially on basal half or more of tri- angle such striae anastomosing to form irregular small areoleae. Metasomal terga somewhat shining, but surfaces minutely roughened, especially on more pos- terior terga, almost without such roughening on dorsolateral swellings in front of depressed margins of terga I and II; punctation rather fine, coarsest on above mentioned swellings, progressively finer and sparser on marginal areas, where densest punctures separated by about a puncture width. Sterna shining, but minutely roughened, hairs arising from papillae. Male: Length 8 mm; wing length 6.5 mm. Col- oration as in female but mandibular apices, under side of flagellum, and pygidial plate and adjacent areas red brown; all exposed terga and sterna except seventh tergum with broadly pallid, translucent apices. Pubescence as described for female, but all tergal hair bands weak middorsally and even laterally not as dense as in female; terga V and VI without hair bands; base of II without tomentum. No red_ hair at apex of metasoma. Sterna I-IV with apical fringes of hair, sternum V with area of dense hair at each side subapically. Hair of legs, clypeal margin, and mandible nearly as white as that of body. Under sides of all trochanters and femora with particularly long white hairs; under side of hind femur except apex densely covered with such hairs, some nearly half as long as femur, mostly directed basad. Head longer than broad (180:159) upper and lower interorbital distances as 102:82. Clypeus width: length::78:44. Line between lower ends of eyes cross- ing clypeus above middle. Paraocular lobe, malar space, convergence of orbits, and separation of an- tennal sockets as in female. Antennocular: interanten- nal:antennocellar:interocellar:ocellocular distances as 27:12:48:40:26. Labrum with strongly convex, shin- ing body twice as wide as long and small obtuse angle representing apical process. Frontal carina, genal area, glossa as in female. Scape reaching middle of anterior ocellus; first flagellar segment much broader than long, others longer than broad (second over 1.5 times as long as broad), median ones crenulate. Pronotum and propodeum as in female. Basitibial plate defined by strong carina, but plate much more slender than in female and therefore with angulate apex. Sternum IV hidden by third, with row of about 22 bristles arising from premarginal thickening, lateral ones enormous and lying flat, others progressively smaller toward median ones, middle 12 bristles or thereabouts bent at about level of apical sternal margin and Sternum V_ with apical margin broadly emarginate between sublateral lobes. Punctation similar to that of female, but on clypeus and lower part of paraocular area denser; hypostomal area minutely roughened, not smooth and impunc- tate. Propodeum with minutely areolate or reticulate part of triangle extending almost to posterior margin. Tergal punctation somewhat finer than in female, punctures of first two terga separated by about a puncture width, ground shiny and smooth; more posterior terga progressively more roughened and less punctate. Holotype male, Willowmore, Cape Province (Capland on the label), South Africa, February 1, 1905 (Dr. Brauns). Allotype female, same locality and collec- tor, May 15, 1905. Two female paratypes, May 4 and 15, 1905 and two male para- types, August 25, 1906 and October, 1910. The holotype and allotype are in the Transvaal Museum, Pretoria, South Af- rica; a pair of paratypes is in the Snow Entomological Museum, University of Kansas, and the other pair in the British Museum (Natural History). This species is named for the collector, the late Dr. H. Brauns, formerly of Wil- lowmore, Cape Province. thereafter erect. Pachyhalictus (Dictyohalictus) retigerus (Cockerell) Figures 62-68 Halictus retigerus Cockerell, 1940, Ann. Mag. Nat. Mist., (11) 5288; Halictus weenenicus Cockerell, 1941, Ann. Mag. Nat. Hist., (11)8:205 (new synonym). Halictus latifrontosus Cockerell, 1946, Entomologist, 79:43 (new synonym). Halictus crassinervis Cockerell, 1946, Entomologist, 79:183 (new synonym). Examination of types in the British Museum indicated that the specific names listed above are synonymous. The types of the last three were all taken at the same locality in Natal, South Africa, by the same collector. The name crassinervis is based on males, the others on females. The lo- cality for the first name listed is in Rho- desia. CLASSIFICATION OF HALICTINE BEEs 537. The following locality record extends the range to another country: one female, Vipya Plateau, 12 miles northeast of Mzim- ba, Malawi, 5200 feet altitude, 15 April 1967 (C. D. Michener). Thrinchostoma (Thrinchostoma) afasciatum new species Figures 84, 86 This species is described here because it has certain characters not otherwise found in the genus which must therefore be ac- counted for in the generic description. The short malar space, only one third as long as broad or perhaps less, distinguishes this species from all others except T. sladeni Cockerell (see Bliithgen, 1926) from As- sam. The most remarkable feature, how- ever, is the lack of the bands of pale (usu- ally silvery), laterally directed hairs on the posterior marginal areas of the metasomal terga. Such bands characterize all other species of the genus. The incompletely de- scribed T. bryant Meade-Waldo, 1914, also from Borneo, could be the male of T. afas- ciatum. It has black head and thorax, probably lacks radiating striae in the pro- podeal triangle, and thus seems likely to be different although the description says nothing of the apical tergal bands. Female: Length 9.5 mm. Head brownish black; labrum, malar area, clypeus, and lower part of para- ocular area testaceous, this color grading into the dark color of rest of head, supraclypeal area and hypo- stomal area being largely reddish brown. Mandible testaceous except for dark brown apex. Antenna brownish black except base of scape and under sides of segments 7-12 testaceous. Thorax and legs testa- ceous except for mesoscutum which is dusky brown- ish, grading to testaceous posteriorly. Wings yellow- ish, veins and stigma dusky brown, at extreme wing bases testaceous, also veins forming marginal cell beyond stigma and beyond third transverse cubital vein testaceous. First metasomal tergum and narrow basal bands on terga 2-4 testaceous; broad apical bands on terga 1-4 transparent so that basal testaceous bands on terga 2-4 show through; rest of metasomal dorsum brownish black; metasomal venter brown, testaceous basally. Hair of head dull yellowish white, some of long hairs dusky in certain lights; subappressed plumose hairs almost hiding surface of lower part of paraocular area laterally; short subappressed hairs also abundant, but not obscuring surface on rest of paraocular area, frons, vertex and genal areas; long, simple, mostly subserect hairs present on most of head, unusually long, yellowish, and strongly directed forward on supraclypeal area, clypeus, mandible, hypostomal area, and lower genal area. Thoracic hair colored like that of head, short whitish hairs abundant on pleura, sides and posterior face of propodeum, and on metanotum; longer erect hairs mostly simple and dusky in certain lights dorsally, paler and often coarsely plumose lat- erally. Hairs of legs pale testaceous, golden on under sides of tibiae and tarsi. Metasomal hair dull yellow- ish white, long erect dorsal hairs dusky in certain lights; transparent marginal bands of terga 1-4 with only scattered, short, laterally directed hairs. Head broader than thorax, clypeus 2.5 times as broad as long, not much produced downward nor protuberant anteriorly; inner orbits not strongly con- verging below (Fig. 86); line tangent to lower ends of eyes only a little above middle of clypeus; antennal sockets separated by more than diameter of a socket; antennocular distance about twice diameter of an- tennal socket; malar area about three times as wide as shortest length; mandible long, less strongly curved than in the forms with a more produced clypeus; first flagellar segment slightly longer than broad, middle segments markedly so. Interocellar distance much less than ocellocular distance. Genal area about as broad as eye seen from side. Glossa distinctly longer than length of head, apical fourth without long hairs. Inner hind ubial spur as in Figure 84. Scutellum bi- gibbous; dorsum of propodeum longer than scutellum. Forewing with basal vein and m-cu interstitial; sub- marginal cells as in Figure 81; hairs denser around the medially thickened second transverse cubital vein than elsewhere. Clypeus and supraclypeal area shining, with coarse punctures, some of them longitudinally elongate, ir- regularly placed, but mostly about a puncture width apart; rest of head and thorax with minute punctures, widely separated on scutum, the center of which is shining and impunctate, scutellar gibbosities also shin- ing and impunctate; sides of thorax mostly minutely roughened and dull; propodeal triangle large, nearly reaching declivity, with strong, regular, radiating ridges. Metasomal terga grading from the first which is shining with only scattered minute punctures to the fifth which has a dull surface and scattered small punctures; posterior transparent margins of terga I- IV impunctate, shining on tergum I, progressively duller on succeeding terga. Holotype female: Pontianak, Borneo (Kalimantan, Indonesia) (F. Muir) in the collection of the Bishop Museum, Hono- lulu. The specific name is based on a, with- out, plus fasciatus, banded, with reference to the lack of apical tergal bands of later- 538 Tue UNIVERSITY OF KANSAS SCIENCE BULLETIN ally directed silvery or golden hairs, char- acteristic of other species of the genus. Halictus (Seladonia) lutescens Friese, 1921 Halictus ruae Cockerell, 1949, Proc. U. S. Nat. Mus., 98:446 (new synonym). Type of ruae in National Museum of Natural History, Washington, D.C. LITERATURE CITED BLUTHGEN, P. 1920, 1921. Die deutschen Ar- ten der Bienengattung Halictus Latr. Deutsche Ent. Zeitschr., 1920:81-132, 1921:267-302. 1923a. Beitrage zur Kenntnis der Bienengattung Halictus Latr. Arch. Naturg., 89(2) :232-332. ——. 1923b. Beitrage. zur Systematik der Bienengattung Halictus Latr. Kono- wia, 2:65-142. ———. 1924. Contribucién al conocimiento de las especies espaftolas de “Halictus.” Mem. Real Soc. Espaftola Hist. Nat., 11:331-544. 1925. Die Bienengattung Nomioides Schenck. Stettiner Ent. Zeitung, 86: 1-100. ———. 1926. Beitrage zur Kenntnis der indo-malayischen Halictus- und Thrin- costoma-Arten. Zool. Jahrb. (Syst., Geogr. Biol. Tiere), 51:375-698, pls. 4-5. ———. 1928. Beitrage zur Kenntnis der indo-malayischen Halictus- und Thrin- costoma-Arten, 1. Nachtrag. Zool. Jahrb. (Syst., Geogr. Biol. Tiere), 54: 343-406, ———. 1931. Beitrage zur Kenntnis der indo-malayischen Halictus- und Thrin- costoma-Arten. Zool. Jahrb. (Syst., Geogr. Biol. Tiere), 61:285-346. ——. 1934. 1. Nachtrag zur Monographie der Bienengattung Nomioides Schenck. Stettiner Ent. Zeitung, 95:238-283. Esmer, A. W. 1969. Die Bienen des Genus Halictus Latr. s.1. im Grossraum von Linz. Naturkundliches Jahrbuch der Stadt Linz, 133-183. ——. 1974. Von Linné bis Fabricius beschriebene westpalaarktische Arten der Genera Halictus und Lasioglossum. Nachrichtenbl. Bayerischen Ent., 23: Ia aree ———. 1976a. Halictus und Lasioglossum aus Marokko. Linzer Biol. Beitr., 8: 205-266. 1976b. Liste der mitteleuropaischen Halictus- und Lasioglossum-Arten. Linzer Biol. Beitr., 8:393-405. Erckwort, G. C. 1969. A comparative mor- phological study and generic revision of the augochlorine bees. Univ. Kansas Sci. Bull 482325524. LarreiLie, P. A. 1810. Considerations géné rales sur l’ordre natural des animaux composant les classes des crustaces, des arachnides, et des insectes. Schoell, Paris. Micwener, “C. DD. 1951. Halictdae zx C. F. W. Muesebeck, K. V. Krombein and H. K. Townes, Hymenoptera of America North of Mexico—Synoptic Catalog, U. S. Dept. Agric. Monogr. no. 2. —. 1978. The parasitic groups of Halic- tidae (Hymenoptera, Apoidea). Univ. Kansas Sei. Bull, 51:291-339. SAKAGAMI, S. F. ano F. L. Warn. 1966. Halic- tus latisignatus Cameron: a polymor- phic Indian halictine bee with caste differentiation. Jour. Bombay Nat. Fist; Soc, 63:57-73. SanbHousE, G. A. 1941. The American bees of the subgenus Halictus. Ent., Amer- icana, (ns )2 223-39; Warncke, K. 1970. Beitrag zur Systematik und Verbreitung der Bienengattung Prosopis F. in der Westpalaarktis. Bull. Recherches Agronom. Gembloux, (NS )5:754-768. 1975. Beitrag zur Systematik und Verbreitung der Furchenbienen in der Tirkei. Polskie Pismo Ent., 45:81-128. Witte, A. anp C. D. Micnener. 1971. Ob- servations on the nests of Costa Rican Halictus with taxonomic notes on Neo- tropical species. Rev. Biol. Tropical, 18:17-31. BSS SRR Rs woeara haha ahatee*ehehateMehatoe'ehane ohetehenetehetotetetatetotatetetatete fete tatetatetetetetetecerereceneses 2 o © 2 6 6 8 8 6 6 o- 0. sole a a Be A eR en A RE eseseneeabeetetateeceenceeseteeseeeeeeeeeeee ee etacatatetatatetetetetecatetetenetatetctatetetotacecenerescacecsetetetetetete tema etetehatetetatetetstatetetemetetetete 29.0 ,.9..0 See SASS Rea eee a Teer ee } wi We 4 =, i 4, “7 SCIENCE BULLETIN A NEW GENUS OF CRYPTODIRAN TURTLES (TESTUDINOIDEA, CHELYDRIDAE) FROM THE UPPER CRETACEOUS HELL CREEK FORMATION OF MONTANA By KENNETH N. WHETSTONE Vol. 51, No. 17, pp. 539-563 November 16, 1978 ANNOUNCEMENT The University of Kansas Science Bulletin (continuation of the Kansas Uni- versity Quarterly) is an outlet for scholarly scientific investigations carried out at the University of Kansas or by University faculty and students. Since its incep- tion, volumes of the Bulletin have been variously issued as single bound volumes, as two or three multi-paper parts or as series of individual papers. Issuance is at irregular intervals, with each volume prior to volume 50 approximately 1000 pages in length. Effective with volume 50, page size has been enlarged, reducing the length of each volume to about 750 pages. The supply of all volumes of the Kansas University Quarterly is now ex- hausted. However, most volumes of the University of Kansas Science Bulletin are still available and are offered, in exchange for similar publications, to learned societies, colleges and universities and other institutions, or may be purchased at $20.00 per volume. Where some of these volumes were issued in parts, individual parts are priced at the rate of 2 cents per page. Current policy, initiated with volume 46, is to issue individual papers as published. Such separata may be purchased individually at the rate of 3 cents per page, with a minimum charge of $1.00 per separate. Subscriptions for forthcoming volumes may be entered at the rate of $20.00 per volume. All communications regarding exchanges, sales and subscriptions should be addressed to the ExcHaNnce LiprariaAN, UNIVERSITY oF Kansas Lipraries, LAwrENCE, Kansas 66045. Reprints of individual papers for personal use by investigators are available gratis for most recent and many older issues of the Bulletin. Such requests should be directed to the author. The International Standard Serial Number of this publication is US ISSN 0022-8850. Editor Eugene C. Bovee Editorial Board William L. Bloom Philip W. Hedrick Rudolf Jander Harvey Lillywhite Charles D. Michener Norman A. Slade Henry D. Stone George W. Byers, Chairman THE UNIVERSITY OF KANSAS SCIENCE BULLETIN Vol. 51, No. 17, pp. 539-563 November 16, 1978 A New Genus of Cryptodiran Turtles (Testudinoidea, Chelydridae) From the Upper Cretaceous Hell Creek Formation of Montana KENNETH N. WHETSTONE TABLE OF CONTENTS /NIBRSS TET RSEN GA Bop BOR sag LE GS AE ae OY ea ie A RE eS PRS OSS Oe eT 539 JN TEA RCO) YE COAT EA FO Ne fat ae et ie da ire Den Rear) OL Ry ia nae Sn Sanam PRS Ee ers 539 Sore AMIEL Ome MIE OIMOIL@OGY fers 208 Pe 5 ee ee, eee 541 PUG Rep vole OG. EN NocOR. A WIRGIGES: 22.2 Seer as he eee 542 SISTING DANI ETE CPM DY EES Cha EAS Et (OS eS ee en ee ee eR ee Se ner Uae ee 544 MORPHOLOGIC AND COMPARATIVE DESCRIPTION 225... 544 Sigg eee Te ree ees eM ORS Eo tie a RON. Ge 4 eae 545 (SERVICATEVEDRIGE BRAUER eee ieee eek) A De eo Soh eae ee eee 552 CARAPACE SAND RIP PASTRON were s seem atk ee ee ee 552 IXNTATESIST OE 2 SHIRL Te NIORPHOLOGY.ssss ces cee sed oe ee ee ce ee es 554 BEcioraie MPPRNDAGES¢s = ce ee o hel e OER ee et 2k ee eee 55) IDE GTORAIEN(GIRDIR we tn ee gated A Le A ae AS aed ee eee 556 PETVICMAPPENDAGES Sse ee ee er ee ee ee Sel a. ee 556 REI CeGIRDDE ey weenie eee aL oat Be A lll, 2 oie tek A oe 2 eee Df PEYEROGENY AND GLEASSIFICATION, OF CHELYDRID; TURTLES: 22222 561 /NGUS KENYA G1) Dy GF 100 UH BES se PO CEE a ag eee 562 ILIV BURA TSU BRI onl (8 Ol Baa a ee II ot eR ed oT 562 ABSTRACT Emarginachelys cretacea is described as a new genus and species of cryptodiran turtle and is placed in the Chelydridae on the basis of shared derived characters. The holotype is a nearly complete skeleton from the Hell Creek Formation (Maestrichtian) of Montana. Emarginachelys is the oldest known member of the superfamily Testudinoidea as defined by Gaffney (1975a). The Chelydridae are hypothesized to be a monophyletic group, sharing a cruciform plastron ligamentously attached to the carapace, a reduced entoplastron, long costiform proc- esses on the nuchal bone, an elongate jugal, and the pectineal processes of the pubis not strongly divergent. Emarginachelys is hypothesized to be the most primitive genus of the Chelydridae since it does not have the derived characters shared by advanced chelydrids. Advanced chelydrids have the frontal bones separated from the orbital rim, a premaxillary “hook,” constriction of the otic bridge, a closed incisura columella auris, a ser- rated carapacial margin, and peripheral fontanelles. INTRODUCTION The turtle superfamily Testudinoidea in- pond turtles (Emydinae and Batagurinae), cludes the majority of living turtles, both in the tortoises (Testudininae), and the snap- numbers and diversity. As defined by Gaff- ping turtles (Chelydridae). Previously, the ney (1975a), the testudinoids include the oldest known definitive testudinoids were a 540 Tue UNIversiTy oF KAnsAs SCIENCE BULLETIN chelydrid, Protochelydra zangerli, from the Paleocene of North Dakota (Erickson, 1973), and a pond turtle, Ptychogaster sp., from the Paleocene of the Big Horn Basin in Wyo- ming (Estes, 1975). The family Chelydridae has a sparse fossil record in Tertiary deposits of North America and Europe, and only two species are extant, both restricted to the New World (Fig. 1). Gaffney (1975b) in- cludes the Recent Asiatic genus Platysternon in the Chelydridae, but I do not consider this to be a parsimonious interpretation of the affinities of this genus (see discussion below). Fic. 1. Generalized distribution of recent chelydrid turtles. Star shows location from which the holotype of Emarginachelys was recovered. The species described herein as Emargi- nachelys cretacea n. gen. et sp. is the oldest and most primitive species placed in the Chelydridae. A University of Kansas field party (Don and Stan Rasmussen and John Chorn) collected the holotype in 1971 from the Upper Cretaceous Hell Creek Formation in Montana. The type specimen is a nearly complete skeleton, still mainly articulated when found. Fossil turtles of this quality are extremely rare, especially in Mesozoic strata. The turtle was preserved as a large “clast” in a medium grained sandstone which lies 5.2 m. (27 feet) below the “Z” coal of the Paleocene Ft. Union Formation (Fig. 2). FT. UNION FM. Zz coal HEEL CREEK EM: mudstones and siltstones, locally sandy— 22ft. sandstone with clay pebble conglomerate — QO ft. siltstone to base of measured interval Fic. 2. Geologic section at type locality. The Hell Creek Formation is Upper Cre- taceous (Maestrichtian) in age, its uppermost strata Potassium-Argon dated at roughly 63 million years old (Gill and Cobban, 1973). The vertebrate fossils found with the speci- men included scales of ganoid fishes, croco- dile teeth and bones, and indeterminate dino- saur fragments. In the older literature, the Hell Creek beds were considered part of the Lance Formation, and Hell Creek specimens were often described as “from the Lance,” or “from the Laramie Cretaceous” (e.g., Hay, 1908). Turtles previously known from the A New GENws oF Foss1t TurRTLEs FROM MONTANA 541 Hell Creek Formation include the following baenid genera (Gaffney, 1972a): Hayemys, Plesiobaena, Eubaena, Stygiochelys, Palato- baena, and Neurankylus. Trionychids (soft shelled turtles) have been found in the Hell Creek, Lance, and Judith River Formations. SYSTEMATIC METHODOLOGY In this study, I name, diagnose, and de- scribe a new taxon and formulate a hypothesis of its phylogenetic relationships. A phylogenetic hypothesis must take the form of a three-taxon statement of the sort, “taxa A and B share a common ancestor not shared by C.” Hennig (1950, 1966) shows that only shared, uniquely-derived characters (synapomorphies) can demonstrate the rela- tive recency of common descent. A_phylo- genetic hypothesis is tested against alternate hypotheses and is accepted or rejected based upon the relative parsimony of the alternate hypotheses. Relative parsimony is decided by way of (cf. Nelson, 1970): (1) minimum parallel evolution, (2) anatomical and em- bryological similarity of presumed synapo- morphous characters, (3) minimum parallel evolution of complex characters and char- acters buffered from requirements of similar adaptive modes of the organisms, and (4) minimum reversal of evolutionary trends. Formulation of phylogenetic hypotheses, and testing them using synapomorphous char- acters, are herein called “Hennigian analysis,” instead of “cladistic analysis.” Hennigian analysis provides a corroborated hypothesis of the relative recency of common ancestry, which is then used to nest monophyletic taxa into more inclusive monophyletic groups. Hennigian analysis cannot selectively test the several, alternate, geometric arrangements for only two populations (Fig. 3). Gaffney (1972a) and Tattersall and El- dredge (1977) contend that the geometry of evolution cannot be tested beyond conven- tional Hennigian analysis, such hypotheses as are presented in Figure 3 being untestable conjectures. Martin and Whetstone (Ms.) argue that the following criteria will objec- tively test the various geometrical combina- tions of evolution: 1) except for evolutionary reversals, which must be assumed to be rare, a direct ancestor must be primitive (plesio- morphic) for every character whereby it differs from daughter populations, 2) a pro- posed daughter population cannot predate a hypothesized ancestor, 3) a proposed ancestor which postdates or is contemporaneous with a hypothesized daughter population refutes a hypothesis of a single evolutionary lineage without a cladistic event (Fig. 3-1). Despite its geological age, Emarginachelys possesses Bess AB a, ae: a It GUE Fic. 3. Alternate hypotheses of phylogenetic linkage between two “taxa.” Solid bars represent derived character states. Model II represents the hypothesis that neither “A” nor “B” is ancestral to the other. A many derived features, especially of braincase and plastron, which make unparsimonious any hypothesis of an ancestral position to other known chelydrids. A thoroughly nested classification _ pre- sumes coordinate rank of sister taxa. If this concept is used to name each coordinate sister group pair, classifications rapidly be- come unwieldy and unstable. New names, names of new rank, and a superfluous num- ber of monotypic higher taxa often result (e.g.. McKenna, 1975, and Gaffney, 1977). Gaffney (1977), however, argues that “al- though stability is often considered an im- portant quality of classifications, . . . it is often a spurious and misleading indication of phylogenetic ‘truth’.” These problems in- crease markedly with the incorporation of fossil taxa, not because fossil taxa are in- herently different, but because many higher taxa must be employed to deal with them. Patterson and Rosen (1977) suggest a different approach to the classification of 542 Tue UNIVERSITY OF KANSAS SCIENCE BULLETIN fossils, namely, that “fossil groups or species, sequenced in a classification according to the convention that each group is the (plesio- morph) sister group of all those living and fossil that succeed it, should be called ‘plesions.’ Plesions may be inserted anywhere (at any level) in a classification without alter- ing the rank or name of any other group. They may bear a categorical name represent- ing any conyentional rank, from genus and species upward ... , these ranks being those already existing in the literature, used only for reference and to avoid ambiguity.” I adopt this convention in the classification which follows. HIGHER “PHYEOGENY ‘OF TURTLES Prior to Gaffney (1975a) most Mesozoic turtles were placed in the “wastebasket” taxon Amphichelydia, thought to be inter- mediate in structure between living cry- potodires and pleurodires. Gaffney presents a convincing shared derived character analysis for a basic crypotodire-pleurodire dichotomy and redistributes most of the taxa previously assigned to the Amphichelydia. The more basic differences between crypotodires and pleurodires are in the trochlear system of the skull for the adductor jaw muscu'ature and the akinetic modifications of the braincase and “palatoquadrate.” Cryptodires have a processus trochlearis oticum and a pterygoid brace to the braincase, while pleurodires have a processus trochlearis pterygoidei and a quadrate brace. Within the Cryptodira, Gaffney recognizes four superfamilies, the Baenoidea, Chelo- nioidea, Trionychoidea, and the Testudino- idea. In the Trionychoidea he places the kinosternids, dermatemydids, trionychids, and Carretochelys, based on the reduction of the stapedial artery and the presence of the “caudifibularis” muscle (Zug, 1971), both assumed to be synapomorphous. Triony- choids also have the costal bones meeting behind the neurals, a reduced postorbital bone, and lack a biconvex 4th cervical vertebra. If based solely upon the arterial condition and the musculature, a hypothesis of mono- phyly for these taxa would be questionable. The reduction of the stapedial artery in kinosternids and Dermatemys results from enlargement of the palatine artery, while in trionychids and Carretochelys it is the result of the enlargement of the pseudopalatine artery (McDowell, 1961; Albrecht, 1967). Neither arterial condition seems intermediate to the other and I suggest that they are non- homologous. Walker (1973) has interpreted Zug’s “caudifibularis” as the dorsal head of the M. flexor tibialis externus that has shifted its insertion from the tibia (the primitive insertion found in most turtles) to the fascia overlying the tibia. This muscle shift is found in all trionychoids, but it is not of such complexity that parallel evolution would be unparsimonious. Reduced postorbitals are also found in some testudinids and pleurodires but have been acquired independently in these taxa. An elongate (primitive, but often laterally reduced) postorbital is found in all chely- drids, some testudinids (e.g. Chrysemys scripta), in most other cryptodires, and in some pleurodires. A reduced neural series is characteristic of most pleurodires but is also assumed to be convergent since some primi- tive pleurodires have a complete series of neurals (e.g. Platychelys). The absence of a biconvex 4th cervical centrum is presumed to be primitive for turtles. The Chelonioidea as defined by Gaffney (1975a, p. 418, 428) excludes the more primi- tive members which Gaffney places in this group, the Plesiochelyidae. A more thorough, derived character analysis which attempts to incorporate these turtles is given by Gaffney (1976), who concludes that all chelonioids have a high dorsum sellae, not overhanging the sella turcica as in testudinoids, and which bears a prominent sagittal ridge. An exami- nation of the dorsum sellae of other reptiles reveals that crocodiles, many lizards, some dinosaurs, and Captorhinus also have a high, nen-overlapping dorsum — sellae. | DeBeer (1937, p. 256) notes that the dorsum sellae is high in early embryos of Emys. This struc- ture even bears a sagittal ridge in Captor- hinus (see Fox and Bowman, 1966). My initial reaction to these comparisons was to consider the high dorsum sellae of chelo- A New GENus oF FossiL TurTLES FROM MoNTANA 543 nioids as a primitive feature shared by the reptiles cited, but the presence of the low dorsum sellae in baenoids, testudinoids and pleurodires indicates, in the absence of other evidence, that the common ancestor of cryp- todires and pleurodires possessed a low dor- sum sellae and that the chelonioid structure is a shared, derived character. A discovery that some primitive cryptodires or pleurodires have a high dorsum sellae would refute this hypothesis and support the view that these groups reduced the dorsum sellae independ- ently. Other derived characters in the brain- case and forelimbs are shared by the ad- vanced chelonioid taxa, including the Toxo- chelyidae, to which chelydrids have been allied by some observers (Hay, 1908, p. 27; Zangerl, 1953, p. 267). The testudinoids form a homogenous group for which I can hypothesize only a single, weak synapomorphy, the biconvex 4th cervical vertebra. Characters which may relate them to other turtles include: 1) the biconvex 4th cervical vertebra—absent in primitive baenoids and “trionychoids,” pres- ent in all chelonioids in which cervicals are known; 2) loss of mesoplastra—mesoplastra present in baenoids, lost in all other crypto- dires; 3) loss of nasal bones—nasals present in primitive baenoids and primitive chelo- niods, absent in trionychoids and_ testu- dinoids, 4) emargination of temporal region of skull—skull roof well developed in Pro- ganochelys, baenoids and chelonioids (ex- cept for Corsochelys), emarginate in all testudinoids and trionychoids, except Macro- clemys and Platysternon. I suggest that testudinoids and trionychoids form a mono- phyletic group sharing relatively great tem- poral emargination and loss of nasal bones, and that their most immediate common ancestor lacked a biconvex 4th cervical. A biconvex 4th cervical is hypothesized to have been independently derived by testudinoids, some baenids and chelonioids. This is not yet a convincing argument, since it is based on characters that seem to be “easily” ac- quired, but I propose it as a testable hy- pothesis. Monophyletic groups within the Testu- dinoidea may be proposed as shown in figure 4. The Chelydridae, including Emargina- chelys, Chelydra, Protochelydra and Macro- clemys, are diagnosed by derived characters as discussed below. Testudinids (sensu Romer, 1956) share two, biconvex cervical vertebrae (Williams, 1950), a character not found in other cryptodires, except Neu- rankylus (see Wiman, 1933). Within the Testudinidae, only Platysternon and_ the Emydinae (of McDowell, 1964) have a double articulation between the 5th and 6th cervical centra. The Emydinae are further characterized by the loss of the “batagurine” process and, in more advanced genera, by the reduction of the pterygoid. As a group, the Batagurinae are diagnosed only by primi- tive characters, although monophyletic com- plexes occur within the subfamily as defined by McDowell. Advanced testudines share a number of derived characters including: 1) a high, convex shell; 2) alternate constric- tion and expansion of the costals; 3) thick- ened epiplastra; 4) expanded coracoid; 5) reduction of phalangeal number; 6) fusion of ventral margins of femoral trochanters; 7) contact of quadrate posterior to incisura collumella auris; 8) depression of the palate in ventral view; 9) ventral processes of the prefrontals far apart (Loveridge and Wil- liams, 1957; Auffenberg, 1974). Many of EAG seg en) DERM ATE RS TRIONYCHIDS KINOSTERNIDS TESTUDINOIDS CHELONIOIDEA BAENOIDEA p PLATYSTERNON "BATAGURINAE" EMYDINAE TESTUDININAE CHELYDRIDAE Fic. +. Hypothesized phylogeny of cryptodiran turtles (top) and testudinoids (below). 544 Tue University oF KANsAS ScIENCE BULLETIN these characters are absent in the more primi- tive living and fossil taxa currently assigned to this group and Auffenberg (1974) sug- gests that some of the characters may have evolved several times. In the following discussions, comparisons with baenids and plesiochelyids are based on Gaffney’s (1972a, 1976) descriptions and figures unless otherwise noted. Comparisons with Protochelydra are based on Erickson’s (1973) figures and tables. Terminology of cranial structures follows Parsons and Wil- liams (1961) as illustrated by Gaffney (1972b). SYSTEMATIC DESCRIPTION Superfamily TESTUDINOIDEA Family CHELYDRIDAE Genus EMARGINACHELYS n. gen. Type species—Emarginachelys cretacea n. sp. Diagnosis —Cryptodiran turtle with pro- cessus trochlearis oticum and pterygoid brace to the braincase; foramen stapedio-temporale not reduced; nasal bones absent; prefrontals downturned anteriorly; frontals bordering the orbits; the skull roof narrowed above the orbits; otic bridge broad; supraoccipital crest long and low; jugal and postorbital bones elongate; premaxillary “hook” absent; quad- rate open behind the stapes; cheek region with some lateral emargination; foramen posterior canalis carotici interni not bordered by basisphenoid; foramen carotico-pharyn- geale not enlarged; fossa for attachment of the pterygoideus musculature not extending far anteriorly; triturating surface of maxilla narrow, with a prominent secondary ridge; pterygoid “waist” neither broad, as in Chelydra, nor greatly constricted, as in Macroclemys temmincku; vomer contacting palatines posteriorly; foramen nervi trigemini and foramen cavernosum small, situated an- terior to the dorsum sellae; prootic contacting processus clinoideus laterally; epipterygoid (?) absent; dorsum sellae low, overhanging the sellae turcica; neither foramen caroticum laterale nor foramen anterior canalis carotici interni enlarged; one biconvex cervical ver- tebra; costal bones not meeting behind the neurals; no serration of the carapace pos- teriorly; supramarginal scutes absent; cara- pace weakly keeled medially, costals with parallel ridges; peripherals unsculptured; no carapacial fontanelles; nuchal bone with long, costiform processes; plastron cruciform, liga- mentously attached to the carapace; ento- plastron reduced, but not “T” shaped; right and left sides of plastron in contact, but not sutured together; thecal process on the ilium; pectineal processes of the pubis not laterally expanded; pubis and ischium separated medially. EMARGINACHELYS CRETACEA n. sp. Diagnosis—Same as for the genus. Holotype-—KUVP 23488: carapace; plas- tron; skull lacking lower jaw; right stapes, posterior horn of hyoid; right forelimb and girdle lacking phalanges of digits [V and V; left forelimb and girdle lacking pisiform and distal half of metacarpal V; left hindlimb and girdle lacking most of digits I and V and the distal phalanx of digit IV; right ilium; cervicals 3-7; anterior half of cervical 8; caudals 1-3. Horizon and Type Locality—Hell Creek Formation (Upper Cretaceous); Garfield County, Montana; SW 4 NW 4% S. 35, T. ZION R37 Be MORPHOLOGIC AND COMPARATIVE DESCRIPTION Skull, Dorsal View (Fig. 5, 6).—The skull roof is composed of the frontal, prefrontal, parietal, and postorbital bones. Nasal bones are absent. The prefrontals are strongly downturned anteriorly, unlike Chelydra and Macroclemys. The descending process of the prefrontal forms the anterior wall of the fossa orbitalis and the posterior wall of the fossa nasalis. The dorsal portion of the fissura ethmoidalis is not broadly expanded as in testudines or Adocus, but resembles the condition in Chelydra and emydines. The foramen supraorbitale is preserved on the right side and is indistinguishable from that of Chelydra. A New GENws oF Foss1t TURTLES FROM MONTANA 545 Fic. 5. Emarginachelys cretacea (KUVP 23488), restoration of skull in dorsal view. Skull length (condyle to tip of snout) is 77 mm. 546 Tue UnIversity oF KAnsAs ScIENCE BULLETIN foramen stapedio-temporale processus trochlearis oticum Fic. 6. Key to Figure 5. Taste 1: Skull Measurements for Holotypes of Emarginachelys and Protochelydra. Emargi- Proto- nachelys chelydra Skull length (condyle to tip of snout) 77.1 mm 77.8 Maximum skull width .... 57.0 69.8 Width across quadrates .. BY ins 68 Width of posterior alveolar surface ................ LOST 16.5 Width of snout at anterior end of orbit -..00...0200..-..--- 18.7 20s Width of snout at posterior end of orbit ...... 312 38.22: Anterior snout height .... 135 14 Distance from anterior wall of orbit to nasal notch .... 6.4 6.2 Distance from posterior rim of orbit to anterior margin of temporal emargination .......-.-.-.-..-- 11.4 15.7 The frontals are relatively large and make up a large portion of the skull roof. The prefrontal and postorbital bones do not meet above the orbits. This allows the frontal bones to contact the dorsal margins of the fossa orbitalis. Contact of the frontal bones with the orbits is a primitive feature for cryptodires that is lost in other chelydrids and in some other testudinoids. Dermatemys and Trionyx have the primitive relationship between the frontal bones and the orbit. Compared to those of Chelydra, the parietals of Emarginachelys are much re- duced by the extreme posterior emargination of the temporal region. Gaffney (1975b) theorized that such emargination was primi- tive for the Chelydridae. The fully-roofed condition, is however, almost certainly primi- A New GENws OF FossiL TURTLES FROM MONTANA 547 tive for the Cryptodira as judged by the temporal regions of baenoids, chelonioids, some pleurodires, and Proganochelys. Gaft- ney’s hypothesis is supported by presence of the emarginate condition in the temporal region of the Upper Cretaceous Emargi- nachelys, and I interpret the expanded tem- poral region of Macroclemys as an evolution- ary reversal. The skull roof bones of Emarginachelys are not strongly sculptured as in Chelydra. The frontals extend as far forward on the skull midline as the anterior margin of the orbits. Above and between the orbits, the skull roof is constricted as in Chelydra and Protochelydra. The otic bridge, which covers the otic re- gion dorsally, is longer antero-posteriorly than in any other chelydrid or Platysternon. In this respect, Emarginachelys resembles trionychoids and primitive emydines, and I assume that a broad otic bridge is primitive for testudinoids. The foramen _ stapedio- temporale is situated between the prootic and the quadrate. The sutural contacts between the squamosal, quadrate, and _ opisthotic bones are partially obscured by cracks, but are interpreted to be as in Figure 6. The supraoccipital crest is long, but is lower than in Chelydra or Macroclemys. Crista supraoccipitalis antrum sq postoticum cies : ae qu q) incisura collumella gurisS foramen nervi “a hypoglossi processus articularis foramen interorbitale foramen supraorbitale po aS) < pa Sze mx foramen supramaxillare foramen orbito-nasale Fic. 7, 8. Fig. 7—Restoration of skull in lateral view. Skull length is 77 mm. Fig. 8—Key to Figure 7. 548 Tue University oF KANsAs ScrENCE BULLETIN Fic. 9. Skulls of other chelydrid turtles: A) Protochelydra, B) Chelydra, C) Macroclemys (A after Erickson 1973 and Gaftney 1975b; B, C after Gaffney 1975b). Not to scale. Lateral View (Fig. 7, 8)—The orbital rim is large and prominent, bounded by the prefrontal and maxillary anteriorly and by the jugal and postorbital posteriorly. The infraorbital canal and foramen orbital-nasale are situated approximately as in Chelydra. The orbit is partially walled posteriorly by medial extensions of the jugal and maxillary and by a lateral extension of the palatine. A well-developed, posterior wall for the fossa or- bitalis is not found in Chelydra or Platy- sternon, but is found in some kinosternids and in Macroclemys. In Emarginachelys, the inframaxillary artery entered this wall an- teriorly and exited through the foramen palatinum posterius after passing through a short canal. The foramen palatinum pos- terius is small, unlike that of Chelydra, but like the foramen of Macroclemys. The premaxillae are not produced into a “hook” as they are in other chelydrids, kinosternids, some testudines, and Platy- sternon. The maxillae are not constricted towards the midline. The jugal and post- orbital are relatively long as in all chelydrids. Gaffney (1975b) hypothesized that a rela- tively long jugal was primitive for the Chelydridae. This hypothesis was based on the presence of a long jugal in the geologi- cally old Protochelydra (Fig. 9), since almost all cryptodires have relatively short jugals. The presence of a long jugal in Emargi- nachelys does support this hypothesis, even though this bone is substantially shorter than in Protochelydra. The shorter jugal in Macroclemys temminckii is interpreted as a A New GENws oF Fosstt TurRTLEs FROM MONTANA 549 reversal. The postorbital bone is long in chelonioids, baenoids, and chelydrids and this condition is probably primitive for cryp- todires. In the “cheek” region, Emargt- nachelys is only slightly emarginate, unlike Protochelydra and Chelydra, but similar to Macroclemys. This does not support Gaff ney’s (1975b) hypothesis of a primitively emarginate cheek region for chelydrids. The weakly emarginate condition also occurs in the primitive baenoids, Trimitichelys and Naomichelys, and in Platysternon, Damonia, Proganochelys, Desmatochelys and _ chelo- niids. I suggest that the relatively-great, lateral emargination of Chelydra and Proto- chelydra may be a shared, derived feature for these turtles. The incisura columella auris of the quad- rate is narrowly open posteriorly, unlike the quadrates of all other chelydrids and most testudines. The open condition is probably primitive, since it occurs in most cryptodires (including emydines), and in most sauropsid reptiles. The processus articularis of the quadrate is much shorter in Emarginachelys than in Chelydra. Ventral View (Fig. 10, 11)—The tuber- culum basioccipitale is accentuated by the strong depression of the basioccipital to ac- commodate the insertion of the rectus capitis muscle. There is little postero-lateral expan- sion of the pterygoid, leaving the otic region more open ventrally than in other chelydrids or Dermatemys. The processus interfenes- tralis and the prootic are exposed ventrally in the fenestra postotica. The ventral margin of the foramen posterior canalis carotici in- terni is formed by the pterygoid; the dorsal margin by the prootic. The right stapes is preserved more or less in place. The footplate of the stapes is flattened and the shaft does not extend from the center of the footplate. Both plesio- chelyids and other chelydrids have a similar stapedial morphology, a condition which I interpret as primitive. Dermatemys and kinosternids have conical footplates that are symmetrical about the shaft of the stapeidal rod (McDowell, 1961), an additional synapo- morphy uniting these taxa. The ventral surface of the basisphenoid extends only slightly beyond the mandibular condyles anteriorly, as in Protochelydra and some emydines. Macroclemys, Chelydra, baenids and plesiochelyids generally have the basisphenoid extending farther beyond the mandibular condyles. [I am uncertain which condition is primitive for testudinoids. On the pterygoid brace to the braincase, lateral to the basisphenoid and posterior to the fossa temporalis inferior, are depressions for the attachment of the pterygoideus mus- culature like that of many emydines (e.g. Graptemys), Macroclemys, baenids, and plesiochelyids. I suggest that a posterior de- pression for the pterygoideus musculature is primitive and that the extension of the de- pressed surface anterior to the area of the processus pterygoideus externus is a derived character shared by Chelydra and _ Proto- chelydra (see Erickson’s figure 1). The skull of Emarginachelys is narrower in many proportions than is that of Chelydra (compare Figs. 9 and 10). The distance be- tween the quadrates is relatively less, as is the width of the pterygoid waist, the distance between the processi pterygoidei externi, and the distance between the postero-medial mar- gins of the triturating surfaces of the upper jaws. The processi pterygoidei externi of Emarginachelys are only slightly extended ventrally. The triturating surfaces of the maxillae are relatively narrow and bear a pronounced median ridge on the posterior portion, similar to that of some testudinids. Among cryptodires, this type of ridging is found only in some testudinids (sensu lato), Dermatemys, and Adocus, and seems to be correlated with herbivory. Since ridging is lacking in kinosternids and_ trionychids, which share a number of derived characters with Dermatemys, the presence of midline maxillary ridges cannot be used to unite Emarginachelys with dermatemydids. The foramen palatinum posterius is small, unlike this foramen in Chelydra. It is situ- ated near the postero-medial corner of the maxilla. The labial ridge of the maxilla is straight, unserrated and prominently raised from the level of the triturating surface. The labial ridge is continued across the midline by the premaxillary bone. There is no pre- maxillary “hook,” a derived feature shared by other chelydrids, and no median recess RSITY OF KANSAS SCIENCE BULLETIN Tue UNIvI 10. Restoration of skull in ventral view. Same scale as Figure 5. Fic. A New GENuws oF FossiL TURTLES FROM MONTANA 551 7 —><-apertura narium externa labial gee secondary triturating ridge fossa cartilaginis epipterygoidei \ i 4 = foramen posterior Mea HER canalis carotici interni: qu sq ee processus interfenestralis op ae _foramen praepalatium foramen palatinum posterius processus pterygoideus externus pterygoideus depression ondylus mandibularis tuberculum basioccipitale Fic. 11. Key to Figure 10. exists in the premaxillary. A median recess is present on premaxillae of emydines and batagurines. The triturating surface of the maxilla narrows anteriorly and terminates at the sutural contact between the vomer and the maxilla. The vomer and premaxillae bear the paired foramina praepalatina. As in all chelydrids, the vomer contacts the palatines posteriorly and is not crested ventrally. Braincase (Fig. 12)—The parietals form most of the side wall of the braincase. The foramen nervi trigemini is small and is situated dorsal and anterior to the dorsum sellae. Its margins are formed mostly by the prootic internally and mostly by the parietal externally. There is a distinct fossa cartil- aginis epipterygoidei, but there appears to be no independently ossified epipterygoid. It is also possible that I have incorrectly inter- preted, as cracks, the sutures which delimit this bone. The quadrate is narrowly excluded from the foramen nervi trigemini, but forms the posterior border of the fossa cartilaginis epipterygoidei. The descending processes of the parietals are broad. In this respect and in the dorsal position of the foramen nervi trigemini, Emarginachelys resembles Der- matemys and is unlike Chelydra and Macro- clemys. The structure of the floor of the braincase is difficult to interpret, due to both its unique- ness and to post-mortem damage. The left side of the braincase anterior to the dorsum sellae is considerably crushed and the right side posterior to the dorsum sellae has been damaged by root growth into the skull. The foramen cavernosum is situated anterior to the dorsum sellae, just ventral to the tri- geminal foramen. The anterior placement of the foramen cavernosum is, to my knowl- edge, unique among turtles. The situation is apparently the result of the union of the 1 1 Ne Tue UNIversITY oF KANsAS SCIENCE BULLETIN FORAMEN CAROTICUM LATERALE CRISTA PTERYGOIDEA FORAMEN NERVI_TRIGEMINI Fic. processus clinoideus with the prootic which is anteriorly expanded medial and dorsal to the area normally occupied by the sulcus cavernosus. The dorsum sellae is low and overlaps the floor of the sellae turcica. The anterior internal carotid foramina are situated beneath the shelf of the dorsum sellae and are closer together than in Chelydra. Part of the trabecula of the right side is preserved; beneath it the lateral carotid foramina can be located with probes. They are close in size to the anterior internal carotid foramina, but direct measurement is impossible. The crista pterygoidei is short and bears the sulcus cavernosus as it rises toward the foramen cavernosum. Cervical Vertebrae—Most of the third through eighth cervical vertebrae are pre- The cervical, articulations may be represented by Walther’s (1922) (3(4)5)6 7 8. This is parable to the cervical central pattern of Macroclemys and Chelydra. With only rare served. central formula as com- OSSIFIED TRABECULAE A oN | : FORAMEN ANTERIOR VCANA BASIS TUBERCULI BASAL| 12. Restoration of braincase in dorsal view. exceptions, these turtles have the patterns (2(3(4)5)6)7)8) and (2(3(4)5)6 7 8) (Wil- liams, 1950). The eighth cervical of Emargi- nachelys is not biconvex as in emydines, batagurines, testudines, and Platysternon. A biconvex, fourth cervical is a derived char- acter which occurs only in some eucrypto- dires (sensu Gaffney), with the possible ex- ception of Neurankylus. The cervical centra of Emarginachelys are much shorter than those of recent chelydrids with a comparably sized carapace. Otherwise the centra are morphologically similar to those of Chelydra and Macroclemys. Carapace and Plastron (Fig. 13).—The carapace and plastron of Emarginachelys shows features absent in other chelydrids. The bones of the carapace are thick, es- pecially the neurals and peripherals. There is no emargination anteriorly and_ scalloping posteriorly. There are 11 pairs of peripherals, 8 pairs of costals, 8 neurals, 2 suprapygals, and one nuchal. The neurals are longer than A New GEnus oF Fossit TurtTLEs FROM MONTANA 553 ae . — oT an ve aS vi VE Se 4 \ x \ a > SF Sy” \ = | y % , / | Z : tr ae € f Rey th 2 ns % ~ INI WIA > CAL ro (ae, Fic. 13. Partial restoration of carapace and plastron. wide and narrow sharply anteriorly. The impression of the most anterior vertebral scute cannot be discerned. The impressions of the remaining vertebrals are about as wide as they are long and approximately square in outline. There are no supramarginal scutes as are found in Macroclemys temminckit. A low, median keel extends from the 4th neural posteriorly to the base of the posterior suprapygal. Auxillary ridges extend from the posterior of the 4th through 7th neurals and diverge anteriorly and laterally, term1- nating after a short distance. Costals 5 and 6, on the inferior half, and costals 4 and 5, on the superior half, are strongly sculptured with parallel ridges similar to the costal sculpture of Pseudemys. Costals with similar sculpture from the Upper Cretaceous Lance Formation were described by Estes (1964). These may belong to Emarginachelys, rather than to Neurankylus as suggested by Gaff- ney (1972a). There is no sculpture on the peripherals. Cross sections of several periph- erals are shown in Figure 14. Peripherals 4.5 and 6 are elongated in the ventral direc- tion for the attachment of the plastron. There is a lateral “keel” on these peripherals that merges on peripherals 3 and 7 with the ventral margin of the carapace. Pits in the ventromedial margins of the peripherals are directed inward and downward for reception of the digitate lateral projections of the plastron. The 4th and 7th peripherals are the more deeply pitted while the 5th and 6th are shallowly pitted at the ventro-medial margin. The posterior peripherals are tapered as in Dermatemys. The broad prezygapophyses of the first dorsal vertebra curve downward and out ward to allow vertical flexure of the neck. The ventral surfaces of the succeeding dorsal vertebrae are not flattened as in Chelydra. The 2nd, 3rd, and 4th dorsals are ventrally keeled as in Dermatemys, but the more pos- terior dorsals are ventrally rounded. The 9th and 10th dorsals are procoelous, the 10th 554 Tue UNIversiTy oF KANsAs ScIENCE BULLETIN strongly so. The ribs of the 10th dorsal are free. Ribs of the three sacral vertebrae are modified for a support for the pelvic girdle, as in Chelydra. The distal rib-heads of the costals fit into “V” shaped notches in the peripherals. There appears to have been no ventral closure of the peripherals under the distal rib ends. There are no fontanelles be- tween the costal and peripheral plates as occurs in other chelydrids and in some toxochelyids. The plastron is cruciform, with broad antero-posterior extensions of the hyo- and hypoplastra for the ligamentous attachment with the carapace. A ligamentously attached “buttress” extends from the antero-lateral prong of the hyoplastron along the medial margin of the peripherals and attaches in a pit at the juncture of the 3rd and 4th peripherals, the point of termination of the costiform process of the nuchal bone. The anterior margin of the carapace is thus braced by a ring of bone extending anteriorly from the right 4th peripheral across the anterior margin to the left 4th peripheral. I am not sure of the functional advantages of such a bony support, but I suspect that it is associ- ated with the reduced, ligamentously attached plastron in a heavy bodied, semi-aquatic turtle. The entoplastron is not “T” shaped as in other chelydrids, but is roughly triangular in outline, with a narrowly tapered posterior tip. It is articulated to the surrounding plastral bones by a kinetic squamous. articulation (Fig. 13). Some of the anterior portion of the entoplastron extends laterally over the dorsal surface of the epiplastra and is not visible in ventral view (Fig. 13). The hyo- and hypoplastra are strongly sutured together with most of the plastral bridge being formed by the hyoplastron. The epiplastra are broad compared to those of other chelydrids. The epiplastra articulate with the hyoplastra by convexo-concave “joints” and the right and left epiplastra do not suture together at the midline. The remaining plastral elements articulate with their counterparts of the other side in a loose kinetic articulation. This plastral morphology is unique and_ highly derived, differing from that of sea turtles. Plastral kinesis allowed free movement of the plastron by flexure perpendicular to the antero-posterior axis. The strong sutures be- tween the hyo- and hypoplastra allowed no Kinesis at the midline perpendicular to a lateral axis. The impressions of the three inframarginal scutes and the gular, humeral, and anal scutes are preserved. They are essentially as in Macroclemys. Impressions of the remaining plastral scutes cannot be determined. Analysis of Shell Morphology.—The shell of Emarginachelys differs from that of other chelydrids in the absence of plastral and carapacial fontanelles. In this respect it re- sembles most adult testudinids (sensu Rom- er), dermatemydids, baenids, some kinoster- nids and some Plesiochelys (Bram, 1965). I interpret shell reduction by fontanellization as a shared, derived feature for Protochelydra (lacks plastral fontanelles), Chelydra, and Macroclemys. The Emarginachelys entoplas- tron and xiphiplastra are reduced, relative to the primitive condition of cryptodires, but not as reduced as in Chelydra and Macroclemys. I interpret the “T” shaped entoplastron, re- duced epiplastra, and serrated carapacial margin as further synapomorphies uniting Protochelydra, Chelydra and Macroclemys and the further reduction of the xiphiplastra as derived features shared by Macroclemys and Chelydra. If Protochelydra and Chelydra share a common ancestor not shared by Macroclemys, as suggested here, then reduced xiphiplastra and plastral fontanelles would be hypothesized to be derived in parallel. The cruciform plastron and long costiform processes on the nuchal bone are also found in kinosternids and are particularly well de- veloped in Staurotypus. These features, plus the presence of a single, biconvex cervical vertebra, were used by Williams (1950, whose classification was followed by Romer, 1956) to unite chelydrids and kinosternids into a single family. Since these features are absent in Dermatemys, which shares a derived cra- nial artery pattern with kinosternids (Mc- Dowell, 1961; Albrecht, 1967; Gaffney, 1975a), I suggest that the costiform process and the cruciform plastron were independent- ly derived in kinosternids and chelydrids. The 4th cervical is biconvex in chelydrids while other cervicals are bioconvex in kinosternids and Dermatemys. A New GENws oF Fossit TURTLES FROM MONTANA 555 Fic. 14. Semi-diagrammatic cross sections of periph- eral bones: A) 9th right, B) 5th right, C) 2nd right. Natural size. Plastral “buttresses,” extensions of the plas- tron which suturally contact the costals, occur in many pleurodires, baenoids, dermatemy- dids, and testudinids and are assumed to be primitive for cryptodires. If the long, anterior extensions of the hyoplasta of Emarginachelys represent a reduced buttress, then the com- plete loss of a buttress would be a synapomor- phy uniting other chelydrids. Ligamentous attachment of the plastron to the carapace is a derived feature found in all chelydrids, some emydines, Platysternon, and Claudius among non-chelonioid cryptodires. Since sutural attachment, the plesiomorphous condition, occurs in most testudinids (sensu lato), Dermatemys, and kinosternids, I regard liga- mentous plastral attachment as examples of parallelism for chelydrids and these other taxa. Shell morphology has been used (e.g. Hay, 1908; Zangerl, 1953) to suggest a close rela- tionship for chelydrids and _ toxochelyids. These taxa share the cruciform plastron, “T” shaped entoplastron, ligamentous attachment of carapace to plastron, costo-peripheral fon- tanelles, and plastral fontanelles, all derived characters for cryptodires. Gaffney (1975a, 1976) suggests that this shell reduction is convergent. The toxochelyid braincase indicates clearly the affinities of toxochelyids with the plesio- chelyids and other chelonioids, as proposed by Gaffney. Derived features of the Toxo- chelys braincase which are shared by some, or all, of the chelonioids, but not by chelydrids, include: 1) a high, crested dorsum sellae that does not overlap the sella turcica (see discus- sion above); 2) approximated internal carotid arteries; 3) fusion of the ossified trabeculae with reduction of the sellae turcica. The taenia intertrabecularis (see Nick, 1912) may be present as a keel atop fused trabeculae of toxochelyids (Whetstone and Stewart, Ms.). The shell morphology of Emarginachelys, which I regard as the most primitive chely- drid, supports Gaffney’s hypothesis of con- vergent shell reduction for chelydrids and toxochelyids, since Emarginachelys lacks most of the shell reduction found in later chelydrids or in toxochelyids. Pectoral Appendages (Fig. 15, 16)—The left and right forelimbs are preserved es- sentially as they were articulated in life. The left forefoot lacks only the pisiform bone and the proximal half of the fifth metacarpal. Except for the structure of the intermedium, the morphology of the forelimb compares closely with that of adult chelydrids—the metatarsals and phalanges are relatively short and broad; the phalangeal formula is 2-3-3- 3-3; the centralia are fused. The intermedium is wedge shaped, extending medially onto the distal surface of the radius (Fig. 15). To my knowledge, this carpal morphology is unique to Emarginachelys. The humerus is similar to that of Chelydra. It is strongly “S” shaped, with the neck ex- tending outward from the distal part of the shaft at an angle (angle “alpha” of Zangerl, 1953) of about 90 degrees. Its head is ellip- tical, with a prominent, lateral shoulder. The humeral shoulder of Chelydra and Protochely- dra is poorly-defined while that of Macro- clemys and Emarginachelys is prominent. There is a well-defined, intertrochanteric fossa immediately behind the head. The shaft of the humerus is massive, unlike the slender 556 Tue University oF Kansas SclENCE BULLETIN es FUSED CENTRALIA Vv DISTAL CARPALS HB DY METACARPAL ¥ aw ULNARE Avg | AVG a ME TACARPALS T- Fic. 15. Left forefoot, radius and ulna. Natural size. humeri of emydines and Platysternon. Dis- tally the shaft expands and bears two stout condyles. Pectoral Girdle (Fig. 17)—As in all living turtles, the pectoral girdle is a three-pronged structure with postero-medial, medial and dorsal processes. The pectoral girdle com- pares closely with chelydrids except that the coracoid is less expanded than in Chelydra, and the ventro-medial process of the scapula is more massive. Pelvic Appendages (Fig. 15, 18).—The left A New Genus oF Fossit TurTLEs FROM MONTANA 557 Fic. 16. Left—Lateral and dorsal aspect of left humerus; Right—Ventral and dorsal aspect of left femur. Natural size. hind limb and the left half of the pelvic girdle are well preserved. The left foot lacks most of digits I and V. As far as can be deter- mined, the phalangeal formula is the same as in Chelydra. Of the tarsals, the astragalus and calcaneus are fused; a bump on the distal edge of the astragalus probably represents the fused centrale. The astragalus, calcaneus, and centrale are usually well fused in adult chely- drids, but are sometimes discernible or sepa- rate in juveniles (cf. Zug, 1971). The femur is similar in size and general morphology to that of Chelydra. The femoral trochanters are more massive than those of Chelydra, with constriction of the intertro- chanteric fossa. The fossa is partially enclosed ventrally by a low ridge which connects the distal borders of the trochanters. There is much greater curvature in the shaft than in Macroclemys or Chelydra. The distal con- dyles are strongly produced from the ventral surface. Pelvic Girdle (Fig. 19)—The ilium is strongly inclined posteriorly; at the dorsal end it expands into a rugose surface which served for ligamentous attachment of the pelvis to the carapace. There is a_ well-developed, thecal process on the anterior margin. Among living cryptodires studied by Zug (1971), only kinosternids have the thecal process on the ilium. If the presence of this process is consid- ered synapomorphous, Emarginachelys might be presumed to share a common ancestor with kinosternids and Dermatemys that is not shared by chelydrids. I do not accept this interpretation. As discussed above, the cranial circulation of Dermatemys and kinosternids is unique and presumably synapomorphous. If a close relationship with Emarginachelys exists, the Dermatemys ossified shell and the JI JI © 2) Tue University oF KAnsAs ScIENCE BULLETIN Fic. 17. Left scapulocoracoid in anterior and slightly ventral view. ilium without a thecal process would be hypothesized as evolutionary reversals, or the reduced plastron and ligamentous plastral attachment of Emarginachelys would be hypothesized as convergence with chelydrids. Emarginachelys also has the costals not meet- ing behind the neurals, a biconvex 4th cervi- cal, an elongate jugal, and non-divergent pectineal processes (see below). I prefer to regard the thecal process as independently acquired for kinosternids and Emargin- achelys. A similar process occurs on ilia of some Toxochelys and of Chisternon, a baenid. The pubis is narrower medially than in Chelydra. It is notched antero-medially for the insertion of the epipubic cartilage. Antero- laterally the pubis bears a process, the pec- tineal process, which was attached to the plastron by ligaments. Zug (1961) shows that only Macroclemys and Chelydra among living cryptodires have the pectineal process Fic. 18. Left hindfoot, tibia, and fibula. Natural size. parallel to each other and to the sagittal plane (Fig. 20). I interpret this as a synapomorphy for advanced chelydrids. The pubis of Proto- chelydra is not known. The Emarginachelys pelvis is intermediate between other crypto- dires and chelydrids in having the pectineal processes neither broadly divergent nor par- allel. in most testudinids, including Platysternon, the medial surfaces of the ischium and pubis are approximated, usually with a diamond- shaped foramen between the two bones (Fig. 20). The pubis of living chelydrids (and A New GEnuws oF Fossit TurTLES FROM MONTANA 559 Fic. 19. Above—Left half of pelvis in lateral view. Note thecal process at arrow. Below—Partial restoration of pelvic girdle in dorsal view. Abbreviations as in Figure 20. Natural size. 560 Tue University oF KANsAs ScIENCE BULLETIN DERMATEMYS = fi? es TERRAPENE MALACLEMYS “yuo CHELYDRA PLATYSTERNON Fic. 20. Pelvic girdles of some cryptodiran turtles in dorsal view, after Zug (1971). Abbreviations: P- pubis, p- pectineal process of pubis, F- puboischiadic foramen, I- ischium, m- metischial process. A New GENws oF Fossit TurRTLES FROM MONTANA 561 EMARGINACHELYS PROTOCHELYDRA CHELYDRA MACROCL EMYS Fic. 21. Alternate hypotheses of the relationships of chelydrid turtles. Emarginachelys) is relatively widely separated from the ischium, although the cartilage con- necting them may calcify late in life. There is no diamond-shaped foramen in living chely- drids or Emarginachelys. In Emarginachelys there is a ventrally-directed, metischial proc- ess not present in Chelydra or Macroclemys. PHYLOGENY AND CLASSIFICATION OF CHEEYDRID' TURTLES The four genera of chelydrid turtles rec- ognized here (Emarginachelys, Protochely- dra, Chelydra, and Macroclemys) are hypoth- esized to be a monophyletic group sharing the following derived characters: 1) a cruciform plastron with reduced entoplastron, 2) a long costiform process on the nuchal bone, 3) ligamentous attachment of the plastron to the carapace, 4) an elongate jugal bone, and 5) the pectineal processes of the pubis not broadly divergent. The separation of the abdominal scutes may also be a synapomor- phous character, but the impressions of these scutes are not known for Emarginachelys. Characters (1) and (3) are also found in toxochelyids and Claudius, and character (2) in kinosternids. This distribution is assumed to be the result of convergence since the sister groups of kinosternids and toxochelyids have the primitive condition of these char- acters. Protochelydra, Chelydra, and Macroclemys are hypothesized to share a common ancestor not shared by Emarginachelys, since they share peripheral fontanelles, a closed quad- rate, serrated carapacial margin, frontals not bordering the orbits, constriction of the otic bridge, less emarginate skull roof, and forma- tion of a bony “beak” by the premaxillary bone. Parallel pectineal processes may also be synapomorphous at this level. Gaffney (1975b) proposes a unique common ancestor for Macroclemys and Chelydra, This hypoth- esis is supported by the presence of plastral fontanelles, reduced xiphiplastra, and a less emarginate temporal region in these genera. An alternate hypothesis is that Protochelydra and Chelydra share a unique, common ances- try. This hypothesis is supported by the great lateral emargination of the cheek region and the extension of the depression for the ptery- goideus musculature anterior to the region of the processus pterygoideus externus. I ac- cept Gaffney’s more parsimonious hypothesis and this phylogeny is reflected by the classifi- cation which follows. Gaffney (1975b) placed the Asiatic genus Platysternon in the Chelydridae, on the basis of a presumed sister-group relationship to Macroclemys. Study of additional specimens of the Miocene Macroclemys schmidti indicate that some of the characters assumed by Gaffney to be synapomorphies actually rep- resent convergences (Whetstone, 1978). Also, Platysternon lacks the cruciform plastron, nar- row epiplastra, “T” shaped entoplastron, ser- rated carapacial margin, a long costiform process on the nuchal bone, separated abdom- inal scutes, and parallel pectineal processes of the pubis which are found in Chelydra and Macroclemys. Vf Platysternon shares a com- mon ancestor with Macroclemys not shared by Chelydra, these lost characters must be interpreted as evolutionary reversals. Platy- sternon shares at least one presumed synapo- morphy with testudinids, as discussed above, and an additional synapomorphy with emy- dines alone. A CLASSIFICA TION: OF CHELYRID TURTLES Family CHELypripAeE Swainson, 1839 Plesion EMARGINACHELYs New Name Subfamily Chelydrinae Swainson, 1839 Genus Protochelydra Erickson, 1973 Genus Macroclemys Gray, 1855 Genus Chelydra Schweigger, 1812 562 Tue University oF KANsAs ScIENCE BULLETIN ACKNOWLEDGEMENTS Special thanks are due Don Rasmussen, who collected the specimen described and made his detailed field notes available for my use. John Chorn suggested the present study and L. D. Martin patiently coaxed it along. Orville Bonner prepared the specimen and my wife, Charlene, prepared the illustrations. J. L. Dobie (Auburn University), John Bolt (Field Museum), H. Marx (Field Museum), and Walter Auffenberg (Florida State Mu- seum) kindly loaned comparative material. E. S. Gaffney and E. O. Wiley provided copies of unpublished manuscripts. E. O. Wiley, E. S. Gaffney, R. Zangerl and L. D. Martin read earlier versions of this description and made many helpful suggestions. This study was, in part, supported by a Graduate Sum- mer Fellowship from the Graduate School of the University of Kansas. PIGERATURE“CIVED ALBRECHT, P. W. 1967. The cranial arteries and cranial arterial foramina of the turtle genera Chrysemys, Sternotherus, and Trionyx. Tulane Studies Zool. 14:81-99. AUFFENBERG, W. 1974. Checklist of fossil land tortoises (Testudinidae). Bull. Florida State Museum 18:121-251. Bram, H. 1965. Die schildkroten aus dem oberen Jura (Malm) der Gegend von Solothurn. Schweizerische Palaont. Abhandl. 83:1-190. DeBeer, G. 1937. The development of the verte- brate skull. Clarendon Press, Oxford. 554p. Erickson, B. 1973. A new chelydrid turtle, Proto- chelydra zangerli, from the Late Paleocene of North Dakota. Sci. Publ. Sci. Museum of Minnesota (new series) 2:2:1-16. Estes, R. 1964. Fossil vertebrates from the Late Cretaceous Lance Formation, eastern Wy- oming. Univ. California Publ. Geol. Sci 49:180. ——. 1975. Lower vertebrates from the Fort Union Formation, Late Paleocene, Big Horn Basin, Wyoming. Herpetologica 31:365-385. Fox, R. C. and M. C. Bowman. 1966. Osteology and relationships of Captorhinus aguti (Cope) (Reptilia: Captorhinomorpha). Articles, Univ. Kansas Paleont. Inst. 11:1-79. GaFrNey, E. S. 1972a. The systematics of the North American family Baenidae (Reptilia, Crypto- dira). Bull. Amer. Mus. Nat. Hist. 147:241- 320. —. 1972b. An illustrated glossary of turtle skull nomenclature. Amer. Mus. Novitaties 2486:1-33. . 1975a. A phylogeny and classification of the higher categories of turtles. Bull. Amer. Mus. Nat. Hist. 155:387-436. ———. 1975b. Phylogeny of the chelydrid turtles: a study of shared derived characters in the skull. Fieldiana: Geology 33:157-178. ———. 1976. Cranial morphology of the European Jurassic turtles Portlandemys and Plesiochelys. 3ull. Amer. Mus. Nat. Hist. 157:487-544. . 1977. The side necked turtle family Cheli- dae: a theory of relationships using shared derived characters. Amer. Mus. Novitaties 2620:1-28. Git, J. R. and W. A. Coppan. 1973. Stratigraphy and geologic history of the Montana Group and equivalent rocks, Montana, Wyoming, and North and South Dakota. U. S. Geol. Survey, Professional Papers 776:1-37. Hay, O. P. 1908. The fossil turtles of North Amer- ica. Carnegie Inst. Washington Publ. 75:1- 568. Hennic, W. 1950. Grundzuge einer theorie der phylogenetischen systematik. Deutscher Zen- tralverlag, Berlin. . 1966. Phylogenetic systematics. Univ. Iili- nois Press, Urbana, U.S.A. 263p. LoveripcE, A. and E. E. Witiiams. 1957. Revision of the African tortoises and turtles of the suborder Cryptodira. Bull. Mus. Comp. Zool. 115:163-557, 18 pl: McDowe tt, S. B. 1961. On the major arterial canals in the ear region of testudinoid turtles and the classification of the Testudinoidea. Bull. Mus. Comp. Zool. 125:23-39. . 1964. Partition of the genus Clemmys and related problems in the evolution of the aquatic Testudinidae. Proc. Zool. Soc. London 143:238-279, McKenna, M. 1975. Toward a phylogenetic classifi- cation of the Mammalia. In Phylogeny of the primates; W. P: Luckett and) FovS.Szalay (eds.), Plenum Press, New York. p. 21-46. Netson G. 1970. Outline of a theory of comparative biology. Syst. Zool. 19:373-384. Nick, L. 1912. Das kopfskelett von Dermochelys coriacea. L. Zool. Jahrb., Abt. Anat. 33:1-238. Parsons, T. and E. E. Wittiams. 1961. Two Jurassic turtle skulls: a morphological study. Bull. Mus. Comp. Zool. 125:43-107. Patrerson, C. and D. E. Rosen. 1977. Review of ichthyodectiform and other Mesozoic teleost fishes and the theory and practice of classi- fying fossils. Bull. Amer. Mus. Nat. Hist. 158:81-172. Romer, A. S. 1956. The osteology of the reptiles. Univ. Chicago Press, Chicago. 772p. TATTERSALL, I. and Nites Evprepce. 1977. Fact, theory and fantasy in human_ paleontology. American Scientist 64:204-211. Wacker, W. F. 1973. The locomotor apparatus of testudines. In Biology of the Reptilia, C. Gans (ed.), Academic Press, London 4:1-100. A New GENUws oF Foss1L TuRTLEs FROM MONTANA 563 Wattuer, W. G. 1922. Die Neu Guinea schildkrote Carettochelys insculpta Ramsay. Nova Guinea 13:607-702. Wituiams, E. E. 1950. Variation and selection in the cervical central articulations of living turtles. Bull. Amer. Mus. Nat. Hist. 94: 505-562. Wiman, C. 1933. Uber schildkroten aus der Oberen Kreide in New Mexico. Nova Acta Reginae Societatis Scientiarum Upsaliensis, ser. IV 9: 1-35, 6 pl. Wuetstone, K. N. 1978. Additional record of the fossil snapping turtle Macroclemys schmidti from the Marsland Formation (Miocene) of Nebraska with notes on interspecific skull var- iation within the genus Macroclemys. Copeia 1978:159-162. ZANGERL, R. 1953. The vertebrate fauna of the Selma Formation of Alabama. Pt. 3, The turtles of the family Protostegidae; Pt. 4, The turtles of the family Toxochelydidae. Field- iana: Geology Memoirs 3:61-277. Zuc, G. 1971. Buoyancy, locomotion, morphology of the pelvic girdle and hindlimb, and systematics of cryptodiran turtles. Univ. Mich- igan Mus. Zool. Misc. Publ. 142:1-98. seetetete seesecectatn’ntatete a tetetete ee 0 e's 0's 0 ee ere es 0 ere 8 8 008 68.00 0.0 0.6 88 0.0-0-0-0-0-0.0.0-0.0.0.0.0.9.0.0. 0.9 9.0.0 .0.9.9.0. ata a ne ee eS SCIENCE BULLETIN Seat at EPS EO RRR He RH SK RR seoieeeesees SAF OK DNs OR NS a a I THE EFFECT OF SLOPE-ASPECT ON THE COMPOSITION AND DENSITY OF AN OAK-HICKORY FOREST IN EASTERN KANSAS By Rodney Birdsell and J. L. Hamrick a eA rere be I SR NS SR Pe ON RTI ins SOA Vol. 51, No. 18, pp. 565-573 November 28, 1978 ANNOUNCEMENT The University of Kansas Science Bulletin (continuation of the Kansas Uni- versity Quarterly) is an outlet for scholarly scientific investigations carried out at the University of Kansas or by University faculty and students. Since its incep- tion, volumes of the Bulletin have been variously issued as single bound volumes, as two or three multi-paper parts or as series of individual papers. Issuance is at irregular intervals, with each volume prior to volume 50 approximately 1000 pages in length. Effective with volume 50, page size has been enlarged, reducing the length of each volume to about 750 pages. The supply of all volumes of the Kansas University Quarterly is now ex- hausted. However, most volumes of the University of Kansas Science Bulletin are still available and are offered, in exchange for.similar publications, to learned societies, colleges and universities and other institutions, or may be purchased at $20.00 per volume. Where some of these volumes were issued in parts, individual parts are priced at the rate of 2 cents per page. Current policy, initiated with volume 46, is to issue individual papers as published. Such separata may be purchased individually at the rate of 3 cents per page, with a minimum charge of $1.00 per separate. Subscriptions for forthcoming volumes may be entered at the rate of $20.00 per volume. All communications regarding exchanges, sales and subscriptions should be addressed to the ExcuHaNce Liprarian, UNIVERSITY oF Kansas Lisraries, LAWRENCE, Kansas 66045. Reprints of individual papers for personal use by investigators are available gratis for most recent and many older issues of the Bulletin. Such requests should be directed to the author. The International Standard Serial Number of this publication is US ISSN 0022-8850. Editor Eugene C. Bovee Editorial Board William L. Bloom Philip W. Hedrick Rudolf Jander Harvey Lillywhite Charles D. Michener Norman A. Slade Henry D. Stone George W. Byers, Chairman THE UNIVERSITY OF KANSAS SCIENCE BULLETIN Vol. 51, No. 18, pp. 565-573 November 28, 1978 The Effect of Slope-Aspect on the Composition and Density of an Oak-Hickory Forest In Eastern Kansas Ropney Birpsey’ and J. L. Hamrick DEPARTMENT OF Botany, UNIverRsiTy oF KANnsAS Lawrence, Kansas 66045, U.S.A. TABLE OF \CONTENTS TDSTNLANGUP | gt, co A lh a ee eee PeR Sse ye th: 565 TOLD SCT LOUN Ge as ee ie a ane Se ee SE See ee oe ee aie mene a eS ee Sy 566 BE ORUAT=S ee AON Ca NV NE EMO IDS pares sac co anc cal setae et aan Soe Se cares oct onze seeded antdacecesteta ee 566 SCE STLAIG? Tee ei Ro oe ee ee eee ee PINE Sa, Re, Fils, Eo 567 2S CUSSIOIN ) ie ae eke ek ee ae a ee ES IOI MDE cc Nf 568 -_PPEBIRUN Tf GH ag IR a OD I ENED ORY >/A “SRLLIBS: un Sulghes sOy's as 2 SoM 1m 2,8 etek a a ee cerns EOC MEME OME TeT LAS, SEE: ioe 2 572 ABS TRACY Measures of species-diversity, basal area, Importance-Values of individual species and the presence or absence of individual species of trees were compared for north- and south-facing slopes in northeastern Kansas. Basswood (Tilia americana) was present only on north-facing slopes. Black walnut (Juglans nigra), redbud (Cercis canadensis) and bur oak (Quercus mac- rocarpa) were more frequent on north-facing than on south-facing slopes. Red oak (Quercus rubra), white ash (Fraxinus americana), red elm (Ulmus rubra), black oak (Quercus velun- tina), shagbark hickory (Carya ovata) and hackberry (Celtis occidentalis) were more often present on south-facing slopes. The chestnut oak (Quercus muhlenbergii) was found equally distributed on north-facing and south-facing slopes. The more important species present are ted oak, black walnut, white ash and shagbark hickory. Previous importance of elms, espe- cially American elm, has been reduced, largely because of the Dutch Elm Disease. North- facing slopes differed from south-facing slopes for all measurements, except for average basal area and overall diversities. The hypothesis is presented that slope-differences are due to the microclimatic variations inherent to the slope-aspect. Biogeographical evidence is given and discussed to support this hypothesis. *Present address: Division of Biology, Kansas State University, Manhattan, Kansas 66506, U.S.A. 566 Tue University oF Kansas ScIENCE BULLETIN INTRODUCTION Vegetation varies in response to envi- ronmental parameters (Boughley, 1973). At the margins of plant communities these responses may be dramatic, with only slight environmental changes pro- ducing marked changes in the dominant vegetation. Such areas occur in northeast- ern Kansas where sharp ecotones exist between the oak-hickory and the tall- grass-prairie vegetations. Before the ar- rival of white settlers in the 1850’s, these two types of vegetation overlapped in an interdigitating pattern that was deter- mined by various environmental factors (Fitch and McGregor, 1956). Of these factors, availability of moisture is usually considered to be the most important in limiting the extent of oak-hickory forest in eastern Kansas (Weaver et al., 1925). However, changes in environmental pa- rameters such as slope-aspect or edaphic factors may produce subtle but measure- able changes in the dominant vegetation. In this paper, we describe variation in the dominant forest-canopy within the forest-prairie ecotone region of northeast- ern Kansas and we present the hypothe- sis that dissimilarities found between the vegetation of north- and south-facing slopes are due to differences in the avail- able moisture inherent to the slope-aspect. MATERIALS AND METHODS The study area is a ravine on the John H. Nelson Environmental Study Area (NESA) and is located approximately 15 km. north of Lawrence, Kansas, in south- eastern Jefferson County. The longitude and latitude of the area are 90° 12’ W and 34° 03’ N, respectively, and the elevation is approximately 310 meters (1017 feet). The ravine runs in a westerly direc- tion, forming north- and_ south-facing slopes with a maximum topographic gradient of 16.8 meters (55 feet). To the north and south of the ravine are areas of grassland, whose management prevents the invasion of woody vegetation. Thus, the only established forest vegetation is either in the ravine or directly adjacent to it. Three line-transects were established in a north-south direction perpendicular to the direction of the ravine. The tran- sects consisted of trees tagged at breast height with numbered aluminum tags, and served as guidelines for the location of the sample plots. The lengths of the transects, the width of the forest vegeta- tion, are, from east to west, approximately 110 meters, 140 meters, and 250 meters. Since the transects varied in length, differing numbers of sample plots were systematically placed along each transect. On the eastern transect two 10 x 10 m plots were established, one on the north- facing slope and one on the south-facing slope. The central transect contained a similarly placed pair of plots, plus a third plot located on top of a limestone-outcrop adjacent to the south-facing slope. The western transect consisted of two south- facing and two north-facing plots. Of these four plots, two were located on lime- stone-outcrops and two were located on the slope below. Thus, a total of nine plots were established, four on the north- facing slope, four on the south-facing slope, and one on a flat area adjacent to the south-facing slope. Within each plot, each tree larger than three inches in di- ameter at breast height (DBH) was identified to species, was numbered with an aluminum tag, and its DBH_ was recorded. From the DBH data, basal areas were calculated for each plot. The basal area for the surrounding woodland was ob- tained by the Bitterlich Variable-Radius Technique (Cox, 1967). The two basal area measurements were compared to de- termine if the woodland density for the StopeE Aspect EFFect oN Oaxk-Hickory Forest 567 plot was representative of the area in which it was located. For each species, the relative density, relative frequency, relative dominance, and Importance- Value were calculated for each plot and for the north- and south-facing slopes. The Shannon-Weiner Index of Spe- cies-Diversity was utilized as a measure of species-diversity. The index is given by ~H HI” — > pi Inp; where H” is the index of i=1 species diversity for a group of S species, pi is the relative abundance of ith species, and In pi is the natural logarithm of pi. Species-diversity for each plot was calcu- lated, and north- and south-facing slopes were compared. In addition to species- diversity, species-abundance (J’) was cal- elated by: J —“H’ /InS’ (Tl ramer, 1969), where H’ is the Shannon-Weiner Index of Species-Diversity and S is the number of species present. Data were also obtained from two ex- traneous sources. An analysis of the phys- ical and chemical soil properties for NESA was conducted during the fall of 1975, and aerial photos from 1941 and 1973 were obtained from the Kansas State Geological Survey. RESULTS Six of the nine plots were located on soils mapped either as “Steep and Stony” or as “Detlor Complex,” the primary dif- ference being steepness of slope, with 20-45% and 8-18°% respectively (USDA, 1977); there were no major chemical dif- ferences found for any of the plots. Both soil-mapping units were formed over col- luvial material from the limestone-out- crops. The one plot on top of the lime- stone-outcrop was in the Oska soil series. Except for the two plots located directly on the limestone-outcrops, the eight plots analyzed in association with slope-aspect can be assumed to contain no significant variation in soil quality. Table 1 ranks all species according to their overall Importance-Values. The more important species were red oak (Quercus rubra), black walnut (Juglans nigra), white ash (Fraxinus americana) and shagbark hickory (Carya ovata). The sum of the Importance-Values for these four species was 201.6 (of a possible total of 300). Of these four species, three showed a marked preference for one or the other slope-aspect. The lone excep- tion was red oak, which had only a slight- ly higher Importance-Value on the south- facing slope. Black walnut had a definite perference for the north-facing slope, while white ash and shagbark hickory favored the south-facing slopes. Marked slope preferences were also shown by species with intermediate or low Impor- tance-Values. The most remarkable of these was basswood (Tilia americana) which had an Importance-Value of 46.8 on the north-facing slope, but did not ap- pear on the south-facing slope. Bur Oak (QO. macrocarpa) and redbud (Cercis can- adensis) also had higher Importance- Values on the north-facing slope. Red elm (Ulmus rubra), black oak (QO. velun- tina), and hackberry (Celtis occidentalis) preferred the south-facing slopes. Only chestnut oak (QO. muhlenbergi) failed to demonstrate a slope-preference. Thus, al- though the coefficient of community of the two slopes was relatively high (0.658 as compared to an expected value of 0.850; Cox, 1972), there is a marked difference in species composition between the two slopes. Both numbers of individuals and the total basal area were similar in the ma- jority of plots (Table 2). Two plots (1 and 4) were similar to the other plots in terms of total basal area, but, as a result of having few individuals, the average basal area per individual was. relatively high. The two plots on the limestone- outcrop (6 and 9) contained numbers of 568 Tue Universiry oF Kansas ScrENCE BULLETIN individuals roughly equal to the other plots, but since each individual tree was relatively large, the total basal area was also large. The total basal area measure- ments obtained by the Bitterlich tech- nique were similar to those obtained by actual individual measurements except for the two plots on limestone-outcrops. Both of these plots yielded Bitterlich basal area estimates of roughly half that actu- ally present, primarily due to the narrow- ness of the vegetation directly affected by the Outcrops. Generally, individuals on the north- facing slope were larger than those on the south-facing slope. The ratio between the respective mean basal areas compares fa- vorably with that found in New Jersey by Cantlon in a study on Cushetunk Moun- tain (Kormondy, 1969). Individuals on the north-facing slope of Cushetunk Mountain averaged 1.8 times greater in basal area than those on the south-facing slope. Individuals on the north-facing slope in the NESA ravine averaged 1.4 times greater in size than those on the south-facing slope. In contrast to the basal-area_ differ- ences, the age of the stand appears to be approximately equal for all plots except those on the limestone-outcrops. Aerial photographs from 1941 show little or no forest vegetation within the study area, with the exception of the rock outcrops. The age of the forest can thus be taken to be somewhat less than 40 years. Both species-richness and species-abun- dance were measured by the Shannon- Weiner Index of Species-Diversity. Spe- cies-richness, or the number of species present, ranged from three species per plot to six species per plot. Although neither slope contained substantially greater num- bers of species per plot, certain species (especially basswood) did demonstrate a slope preference. The species-abundance factor (J’) was slightly higher for the north-facing plots than for the corre- sponding south-facing plots, indicating that north-facing plots possess a more nearly equal distribution of species. DISCUSSION The most dramatic effect of slope- aspect is the variation in individual spe- cies’ Importance-Values (Table 1). Of the eleven species found, the three dem- onstrating the more widely variable Im- portance-Values were basswood, white ash, and black walnut. Basswood ap- peared only on the northfacing slope, yet possessed the third Importance-Value for that slope, being surpassed only by black walnut and red oak. Furthermore, only basswood demonstrated a strong slope- preference and a strong preference for the number of plots on which it appeared. Of the three species demonstrating a major slope-preference, basswood has the most limited geographical distribution (Little, 1971). Within the forest-prairie ecotonal region of eastern Kansas, the major environmental parameter limiting the western distribution of basswood is precipitation. It therefore follows that micro-climatic heterogeneity in moisture- availability should be reflected by micro- distributional patterns of basswood more than for species with greater ecological amplitudes. This result is consistent with several studies which reveal that north-facing slopes remain cooler and contain more moisture than corresponding south-facing slopes. In one such study conducted in Michigan during the 1957 growing season (Cooper, 1960), the air-temperature fifty cm. above the ground averaged nearly 5°F higher on the south-facing slope than on the north-facing slope. In addition, soiltemperatures at depths of 2 and 20 cm. produced similar differences, and the percent of moisture (by weight) of the soil at a depth of 2 cm. was as much as Store AspEcT EFFEcT ON Oaxk-Hickory Forest 569 12.7°% higher for the north-facing slope. Thus, the north-facing slope was_ better able to supply moisture to the vegetation during periods of stress by drought. A similar study by Cantlon on Cushe- tunk Mountain in New Jersey lists tem- peratures from 35-6.0°F higher for the south-facing slope than for the north-fac- ing slope (Kormondy, 1969). As a result of the higher temperatures, the south- facing slope has a larger vapor-pressure deficit, and more evaporation. Even great- er extremes in the microclimate have been shown to exist between northeast- and southwest-facing slopes (Benson et al. 1967). Therefore, in eastern Kansas where water availability is one of the more im- portant environmental parameters influ- encing vegetational composition, it is not surprising to find mesophytic species re- stricted to north and east-facing slopes. Moreover, since this forest is only approxi- mately 40 years old, we can assume that it has not yet reached a stable climax. As the forest continues to mature, we might expect to observe a greater heterogeneity in vegetational composition between the slopes (Odum, 1969). Furthermore, the more mesic north-facing slope might be expected to approach a stable climax more rapidly. Further evidence for the more favor- able micro-climate of the north-facing slope is provided by the differences in basal area observed between the two slopes. Since aerial photography indicates that the majority of the existing forest vegetation dates from the late 1930’s and early 1940's, we can assume that these dif- ferences are due to faster growth rates on the north-facing slope. However, with- out a homogeneous species-composition and age-structure between slopes, no defi- nite conclusions can be drawn in regards to absolute growth-rates (Geyer and Naughton, 1970). The fact does remain, however, that the north-facing slope pres- ently supports a larger basal area per individual. Micro-climatic differences due to slope- aspect may not be the only environmental factors that have influenced the present vegetation of this ravine. Drought, which occurs in the Great Plains on approxi- mately 20-year cycles, has had a large effect on the vegetation of eastern Kansas and may provide an additional explana- tion for the restriction of such species as basswood to north-facing slopes. During periods of severe drought, such as that of the 1930's, such species may have been unable to survive on the drier, south-fac- ing slopes (Albertson and Weaver, 1945). Biological factors such as disease may also have had major effects on the com- position of the NESA forest. Dutch Elm Disease was first diagnosed in Jefferson County in 1961 and it had been reported in neighboring counties as much as three years earlier (Kainski et al., 1964). In the early 1950’s, at the University of Kan- sas Natural History Reservation (located approximately 3 km south of NESA), 4 to Y4 of the trees greater than six inches DBH were elms (Fitch and McGregor, 1956). In contrast, of the 74 trees included in this study only five were elms. The decline of the American elms was relatively rapid, following the introduc- tion of Dutch Elm Disease to Kansas. Before the advent of the disease, Fitch and McGregor (1956) stated that Amer- ican elm (Ulmus americana) was much more prevalent as a dominant tree than red elm (Ulmus rubra), with few red elms being over 12 inches (30.5 cm) DBH, although “the saplings of this spe- cies constitute a prominent part of the understory.” A comparison between two studies (Wells and Morley, 1964 and un- published class data, 1975) of Baldwin Woods, 30 km. south of NESA indicates that by the mid-1960’s the population of elms had reached a relatively stable equi- JI librium (Table 3). It should be noted that in both studies, the disease resistant red elm is more important, a reversal of the previous observations of Fitch and McGregor (1956). From the above, one can conclude that the elms, particularly the American elm, have been removed as major dominants in the forest canopy within the last 20 years. Other species may have also been se- lectively removed from the area by the activities of man. Logging of Jefferson County and nearby Douglas County has been widespread for more than 100 years (Fitch and McGregor, 1956) and there is evidence of logging within the NESA area. In addition to the observation of several large, sawed stumps, several in- dividuals of red oak and black walnut have multiple trunks emitting from a common root system. As red oak and black walnut are among the species most heavily used by the lumber-industry in Kansas (Deneke and Funsch, 1970), it is likely that the multiple trunks sprouted after logging. Age measurements of one such multiple trunk-system revealed that the ages of the separate trunks were with- in a range of five years. Other tree-cutting activities, such as the rural practice of heating and cooking with wood, also ac- counted for a share of the woodland dis- turbance until about 1940. Although the sampling methods were not identical, one can compare the results of this study with two others done within a nearby forested area. Both Wells and Morley’s (1964) study and unpublished data from a University of Kansas class (1975) involved Baldwin Woods, an un- glaciated area 30 km. south of NESA. The major finding of the 1975 study was that the topographic position on the slope was as important as slope-aspect in deter- mining the composition of the canopy. However, the topographic gradient of Baldwin Woods is much greater than that 70 Tue Universiry oF Kansas SciENCE BULLETIN of the NESA ravine (circa 100 meters versus 16 meters). While in both in- stances the slope-aspect largely determines canopy composition, conclusions relating to the much larger topographic gradient of Baldwin Woods are not directly ap- plicable “to: the’ site at: NESA. A higher species-diversity was also found in the Baldwin Woods. studies. While some of this increase could be ex- plained by the greater topographic diver- sity of Baldwin Woods, the major factor appears to be the greater edaphic diversity there. Baldwin Woods is unglaciated and has a variety of parent materials while the soils of NESA were formed primarily of colluvium from the limestone-outcrops, with both loess and glacial tll being in- fluential. Other than the actual limestone- ledges (Plots 6 and 9), the soils in the study area have formed from_ virtually identical parent materials. The overall diversity within the forest canopy can also be shown to be dependent upon a set of limiting factors. In both the NESA study and Tramer’s (1969) study of 267 bird populations, diversity was de- pendent upon the number of species (species-richness). For the NESA plots, the population diversity (H’) correlated at significant levels (7 = 0.975 and 0.939) with In S, the natural logarithm of spe- cies-richness. This relationship contrasts with that of phytoplankton in which the species-richness remains stable, and_ spe- cies-diversity is linked to the relative abundance of species (Sager and Hasler, 1969). Certain phytoplankton species are “opportunistic,” and may experience dra- matic fluctuations in population size in re- sponse to changes in availability of re- sources. Thus, although the number of species 1n a given area may remain ap- proximately the same over a period of time, estimates of species-diversity will de- crease due to changes in relative abun- dance. Stope Aspect Errect oN Oak-Hickory Forest bya Conversely, Tramer (1969) notes that birds are “equilibrating,” since the envi- ronmental factors of a given habitat de- termine the number of species which can exist in that habitat. As species-diversity for the NESA tree-canopy was compar- able to Tramer’s bird populations in terms of dependence on species-richness, the diversity of the forest-canopy may also be a result of environmental parameters which regulate the number of species that can exist in a given area. In summary, extensive pressures due to logging, fire, farming practices, Dutch Elm Disease, and periodic droughts have contributed to the present condition of this forest. However, micro-climatic dif- ferences in moisture availability are also of great importance. There is an increase in temperature, evapotranspiration, and water stress on the south-facing slope, re- sulting in a more favorable micro-climate for forest vegetation on the north-facing slope. The more favorable micro-climate is shown by a shift in species composition within the forest-canopy and by larger basal areas per individual for the north- facing slope. LITERATURE CITED ALBERTSON, F. W., AND J. E. Weaver. 1945. Injury and death or recovery of trees in prairie climate. Ecol. Mono, 15: 393-433. Benson, L., E. A. Puicures, P. A. WILDER, ET AL. 1956. Evolutionary sorting of characters in a hybrid swarm. I. Di- rection of slope. Amer. J. Bot. 54(8): 1017-1026. Boucutey, A. S. 1973. Ecology of Popula- tions, 2nd edition. The Macmillan Company, New York. Coorer, A. W. 1960. An example of the role of microclimate in soil genesis. Soil Sei. 9021092120: Cox, W. G. 1972. Laboratory Manual of General Ecology, 2nd edition. Wm. C. Brown Company Publishers, Du- buque, Iowa. DeneEKE, J. R., anp R. W. Funscw. 1970. Early notes on black walnut prove- nance tests in Kansas. Trans. Kans. Acad. Sci. 72:404-405. Fitcu, H. S., anp R. L. McGrecor. 1956. The Forest Habitat of the University of Kansas Natural History Reserva- tion. University of Kansas Publica- tions of the Museum of Natural His- tory 10:77-127. Geyer, W. A., ano C. G. Naucuton. 1970. Growth and management of black walnut (Juglans nigra L.) on. strip- mined lands in southeastern Kansas. Trans. Kans. Acad) Scr732491-502: Karnsk1, J. M., E. D. Hansinc, W. H. Sr1t, Jr., C. L. Kramer, ann O. J. Dicker- son. 1964. Kansas phytopathological notes: 1961. Trans. Kans. Acad. Sci. 67:433-441. Kormonpy, E. J. 1969. Concepts of Ecol- ogy. Prentice-Hall, Inc., Englewood Cliffs, New Jersey. Larme; EL. 1971. Atlassof.U:S. Biecs Volk 1: Conifers and Important Hard- woods. U.S. Dept. Agric. Misc. Publ. 1146. Ovum, E. 1969. The strategy of ecosystem development. Science 164:262-270. Sacer, P. E., anp A. D. Hasrer. 1969. Spe- cies diversity in lacustrine — phyto- plankton. I. The components of the index of diversity from Shannon’s formula. Amer. Natur. 103:51-60. Tram_er, E. J. 1969. Bird species diversity: components of Shannon’s formula. Ecology 50:927-929. U.S.D.A. 1977. Preliminary soil survey for parts of Jefferson County, Kansas. W.S.. Dept. Agric: Weaver, J. E., H. C. Hanson, ano J. M. AIkMAN. 1925. Transect method of studying woodland vegetation along streams. Bot. Gaz. 80:168-187. We ts, P. V., anp G. E. Morey. 1964. Composition of Baldwin Woods: an oak-history forest in eastern Kansas. Trans. Kans. Acad. Sci. 67:65-69. 572 Tue Universiry oF Kansas Scrence BULLETIN TABLE Ms A listing of the eleven tree species appearing in the NESA study plots in order of decreasing Importance-Values. The effect of slope-aspect on numbers, frequency, density, dominance and importance is also given. AVERAGE # ReEvaTivE RevativE RevativE IMPpoRTANCE IMPORTANCE InpivipuaLts Frequency Densiry Dominance VALuE VALUE Rep Oak North-facing 4 17.6 13:3 25.0 55.9 59.6 South-facing 6 17.6 ee 30.3 63.3 Brack WatnutT North-facing 6 17.6 20.0 33.0 70.6 Da2 South-facing 3 11.8 Th 16.2 Sw Wuirte AsH North-facing 3 11.8 10.0 1.8 23.6 443 South-facing if 17.6 28.2 19.2 65.0 SHAGBARK Hickory North-facing 3 11.8 10.0 3.6 25.6 34.5 South-facing 9 11.8 23.1 8.5 43.4 Rep E_m North-facing 1 5.9 33 10.0 19.2 Dall South-facing 4 11.8 12.8 6.3 30.9 Bur Oak North-facing 4 5.9 13.3 1 30.4 25.0 South-facing 1 5.9 26 bet 19.6 Basswoop North-facing 6 17.6 20.0 O2 46.8 134 South-facing 0 0.0 0.0 0.0 0.0 CHESTNUT Oak North-facing 2 5.9 6.7 D2 17.8 19.4 South-facing 2 11.8 S21 4.2 2c Brack Oak North-facing 0 0.0 0.0 0.0 0.0 6.0 South-facing 2 5.9 26 3.4 11.9 Rep Bup North-facing 1 5.9 33 0.9 10.1 5.0 South-facing 0 0.0 0.0 0.0 0.0 HAcKBERRY North-facing 0 0.0 0.0 0.0 0.0 4.6 South-facing 1 5.9 2.6 0.8 9,3 Store Aspect Errect oN Oaxk-Hicxory Forest 573 TABLE 2. Basal area for each sample plot and average basal area per individual. Basa BirTERLICH ASPECT NuMBER OF AREA BasaL AREA BasaL AREA OF INDIVIDUALS Per PLor MEASUREMENT Per TRE (sQ. IN.) Piotr SLOPE Per Pror (sa. 1N/100M7) ~— (sg. 1n/ 100M”) (s.D.) iL North 4 203.29 263.41 50:32 3942 2 South 8 22162 312.19 27) = Med 3 North 8 216.94 282.92 27.12 4.94 > South 12 258.58 243.90 2155, == eels 7 North 10 323.01 292.88 32.200 + 6.69 8 South 10 299.39 253.66 272 =e PAN 6* North 8 866.58 390.24 108.32 + 34.36 9* South 9 746.64 409.75 82.96 + 26.14 em Neither 5 284.70 243.90 56.94 = 23.64 All North Mean 7.50 402.46 307.36 53.70 (Plots 1, 3, 6 and 7) (22252) (314.01) (£56.39) (£65.38) All South Mean 9.75 371.05 304.88 38.08 (Plots 2,5, 8 and 9) (22337) (£250.98) (£76.14) (2547715) Overall Means 8.24 375.41 299.21 45.68 (Plots 1-9) (£2.49) (£249.00) (£61.70) (2255.71) *Plots 6 and 9 are located on rocky outcrops. ** Plot 4 is on a level area adjacent to a south-facing slope. TABLE 3: Comparison of the importance of elm species in three studies located in eastern Kansas. The 1956 study was before the introduction of Dutch Elm Disease. 1956° 19644 19755 Nat. Hist. Res. Baldwin Woods Baldwin Woods Red Elm LV _ 18.3 220 (Ulmus rubra) Preq- * 8.7 6.4 American Elm Jee * Zz 3.0 (Ulmus americana) Freq.” * 0.8 11 Ulmus spp. Vee : is 3 (9 total) Freq? 25.8° to. 57.9% 95 IS) 1Importance Value; * Frequency (in percent); *Fitch and McGregor, 1956; * Wells and Morley, 1964; °Un- published class data, University of Kansas; °For trees greater than one foot DBH on a West Slope; 7 For trees six inches to one foot DBH on a South slope; * Not listed. III PPP I III O,0 90,000.00. 9.0.0, 0,0,0,0,9,9,9,9,9.%.% "5% sree ee 8. EPP re see SCIENCE BULLETIN . COMP. ZOOL. = LIBRARY DEC 18 1978 *. = HARVARD 0 5 RRS IOS OOO weecectctaaceretstatatatetencceteteteeree® UNIVERSITY POLLEN MANIPULATION AND RELATED ACTIVITIES AND STRUCTURES IN BEES OF THE FAMILY APIDAE SR RR eR SH IR RR cetare mPa na? aaa! eeeececeteteatatatetecpeccsesncee POPPER ALEPPO ELE SLI OC LOPE D ISL IL OP POOL OPPO OPPO PEP D Cee Ss a RRR ER AH OS KR RR GS gS By Charles D. Michener, Mark L. Winston, and Rudolf Jander SS A A RR SR Vol. 51, No. 19, pp. 575-601 December 8, 1978 ANNOUNCEMENT The University of Kansas Science Bulletin (continuation of the Kansas Uni- versity Quarterly) is an outlet for scholarly scientific investigations carried out at the University of Kansas or by University facu!ty and students. Since its incep- tion, volumes of the Bulletin have been variously issued as single bound volumes, as two or three multi-paper parts or as series of individual papers. Issuance is at irregular intervals, with each volume prior to volume 50 approximately 1000 pages in length. Effective with volume 50, page size has been enlarged, reducing the length of each volume to about 750 pages. The supply of all volumes of the Kansas University Quarterly is now ex- hausted. However, most volumes of the University of Kansas Science Bulletin are still available and are offered, in exchange for similar publications, to learned societies, colleges and universities and other institutions, or may be purchased at $20.00 per volume. Where some of these volumes were issued in parts, individual parts are priced at the rate of 2 cents per page. Current policy, initiated with ‘volume 46, is to issue individual papers as published. Such separata may be purchased individually at the rate of 3 cents per page, with a minimum charge of $1.00 per separate. Subscriptions for forthcoming volumes may be entered at the rate of $20.00 per volume. All communications regarding exchanges, sales and subscriptions should be addressed to the ExcHance Liprarian, UNIVERSITY oF Kansas Liprarres, LAwreNnce, Kansas 66045. Reprints of individual papers for personal use by investigators are available gratis for most recent and many older issues of the Bulletin. Such requests should be directed to the author. The International Standard Serial Number of this publication is US ISSN 0022-8850. Editor Eugene C. Bovee Editorial Board William L. Bloom Philip W. Hedrick Rudolf Jander Harvey Lillywhite Charles D. Michener Norman A. Slade Henry D. Stone George W. Byers, Chairman THE UNIVERSITY OF KANSAS SCIENCE BULLETIN Vol. 51, No. 19, pp. 575-601 December 8, 1978 Pollen Manipulation and Related Activities and Structures in Bees of the Family Apidae’ Cuartes D. MicHENER™ * *, Mark L. Winston”, aNnp RupotF JANDER?’ 3 CONTENTS BAUS SUINEUA (Cl eee acs sees oe oa are eases c chico naz sacuss1S2setet Tass ezasre-dvedeslevds Seesor dhs pee ee : Jit RS ee Ee Te 575 NPC RG OGL ONS CONS) 0 See erate eee Sia ee ee Sa Ee 1 SE Ae eee : 576 IMATE ETRY CS | i SR NR ee Se ee aN OO eee ere eee Mi IMUEMEIODS) 2:ce2ee2<2--2222- wfc aE se POPE OEE EY MEER EES ER Bice eee cone Zao 578 SIBRIAINSE ORTHROR GIN ESIIN Gas MATERIALS) =o ce ccet asec ences css oot ced xe oe ea b- eoegseciessastadencesceccede. dus Jy ee oe 579 IMANIPUICATION OF SCENTS! BY MALE EUGLOSSING <..:--2:2522c-c.c5ccscesens cocce.ecee-0-oeecceve co sua encase ee seeeeteng ees 580 NETL OTIKEN SIVAN Ue ATION 2222-22 -n22. fesse seeeeesce se oeesazic.cee ee eee fr eee o Sa: ve Se ee 581 SUCRE ROLE] Ep sree carrer eee ce eee RENAL Aes i cl IAAT 8 on a ee ees 581 [Pol heya» IBRYS) SUaVeR cise ee ere ts aN ata ee SO aE a POS, TAS San Se oe 582 “UNyppyes oll Roh etal, ENS ST eas es are ee eee Sods stenetos ot eee 582 Sisy peal Dollemmmbackin oe seat matte ewe reales ester ke ee ee ee 583 Ipsilateral geliyo cul lagPollenrbackin oem ta! Rie 2) 2002 ese. ee eRe 583 Contralateral Pollen Packing in Apis and Bombus .........-...--.. iis ne 587 Contralateral sPollens Rackine inh! Meliponinae 22s 588 piven ConbicularsPollemmlboad Mini Melipomimae . 2... ee se ec ee we coe ase care Senne cnedseactnzcl eee 590 Specializationsmoite Meliponinepeollems (Collecting, 2228-22. a eee 590 glilyeae ENT oil OSS 11 Gaerne entre Pees eee ee es Se 593 EMO UsIONAR Ya GONSIDE RAI ONS isa: 2 8. oe aco 22e lees ses eecs edz scnsseee-de¥laceve secs dede oo skeveeease £. eh SPSS pees .. 994 BvolMtiontomlollens Collecting wand rams pont. <2. 2 2-5. sececces eb econ ee ese ate eens 594 Evolationworacorbiculaesand their shilling Mechanmisims, 2.22222 22:2 222.2ee- coe ee 595 Rollene lac kin ogi ntia MeN [OICla Comeer erent ea hk oa 2 2 Ser ee eee 597 Derivation of Pollen Manipulation from Self-Grooming Movements —............ a 598 FNGKRNOWIEED GE MPN TS emma tee cose teen ENE, Fc. c zone aved oa gae ee eRe We ect 2 ee 600 IBGERATURE MCTGEDy) 2.2 se Se ee Nae oth Se ee eee 2 AiA A Se Yea nen eee 600 ABSTRACT Pollen manipulation is described for all major groups of Apidae, and for comparative purposes, other bees are also considered. In prototypic pollen collecting, pollen is removed from the front legs and head, and carried in the crop. In eutypie pollen collecting, it 1s re- moved from the front and mid legs as well as surfaces of the head and thorax and trans- ferred by the middle legs to the scopa on the hind legs. Different derived or metatypic types of pollen manipulation supplement the eutypic behavior and provide for transfer of pollen to the abdominal scopa in the Megachilidae and from the abdomen to the hind tibiae in the Apidae. In Apidae the brushy scopa of many other bees is modified into a corbicular scopa, which with the hind tibial surface forms the corbicula. Corbicular filling (= pollen packing) can be *Department of Entomology, * Department of ? Contribution number 1659 from the Department Systematics and Ecology, and * Snow Entomological of Entomology, University of Kansas, Lawrence, Kan- Museum, University of Kansas, Lawrence, Kansas sas 66045, U.S.A. 66045, U.S.A. 576 Tue UNiversiry oF Kansas SciIENCE BULLETIN achieved with little or no modification of eutypical behavior; for Apidae this is called Type I pollen packing, in which pollen is put directly into the corbiculae by the middle legs. Nest- ing materials are packed and transported in the same way. On the other hand, the corbicula is usually loaded with pollen from its distal end. This is Type II pollen packing. It may be either (a) ipsilateral, a modified eutypic pattern in which pollen is placed on the outside of the tibiotarsal joint by the middle leg of the same side, and thence pushed basad into the corbicula, or (b) contralateral, a derived pattern in which pollen transferred by the middle legs to the inner surfaces of the hind basitarsi or swept from the abdomen by those basitarsi, is transferred to the corbicula of the opposite tibia. In pollen-manipulating movements and structures, certain Meliponinae appear to be the most primitive apids, apparently lacking Type II pollen-packing behavior as well as structures necessary for it. For other Meliponinae, ipsilateral Type II pollen packing is most important, although some have contralateral Type II behavior as part of their repertoire. The small hind basitarsus without an auricle suggests that contralateral Type II packing is of little im- portance in most Meliponinae, although an auricular area can function like the auricle to push pollen basad into the corbicula. Euglossini are inadequately studied, but probably ex- hibit ipsilateral Type IH behavior. Bombini and Apinae are similar to one another in pollen- handling structure and behavior and for both, contralateral Type II pollen packing is a principal method. Modifications of usual pollen-manipulating movements are seen in scent manipulation by male euglossine bees and in pollen-gleaning activity by certain Meliponinae (Scaura). Most of the pollen-manipulation movements are either the same as self-grooming movements or are opposites of them so that, for example, a structure that is stroked distally for cleaning may be stroked basally to load it with pollen, the cleaning movement proper serving later for unloading. Presumably such loading movements are derived from similar cleaning move- ments. Only the hind basitarsal movements that push pollen basad into the corbicula lack counterparts in cleaning or other known behavior. INTRODUCTION also associated with the transport of ma- terials used in nest construction. The family Apidae is divided into four distinct groups, the subfamily Meliponinae (the sister group to all the rest according to Winston and Michener, 1977), the tribes Euglossini and Bombini (currently united in the subfamily Bombinae), and the sub- family Apinae. In the past, pollen-collect- ing behavior has been well described only for the Apinae; progressively less was known about this behavior in the Bombini, Meliponinae and Euglossini (Maurizio, 1968). The purpose of this paper is to describe movements whereby pollen and other ma- terials are collected and placed for trans- port in the apid “pollen baskets” or cor- biculae, the derivation of these movements from self-grooming behavior, and the meaning of these movements and related structures for apid evolution. Morpho- logical and behavioral features for collect- ing and transporting materials play a cru- cial role in the evolution and adaptive radiation of the bees (superfamily Apoi- dea). These features are therefore im- portant both for bee taxonomy and for evaluating homologies and convergence (Michener, 1944, 1974; Jander, 1976; Win- ston and Michener, 1977). While the adaptations for transport of nonliquid ma- terials in most families of bees relate pri- marily to pollen, in the Apidae they are While some female bees (Euryglossi- nae, Hylaeinae) transport pollen to the nest exclusively in the crop, most carry at least part of their pollen harvest with the help of hairs which are located differently in the various taxonomic groups. Such hairs form the scopa, a term applied to PotLteEN MANIPULATION By Bees (ApIDAE) Bi firh pollen-carrying hairs whether they are on the outer sides of the hind tibiae and basi- tarsi, the under sides of the basal segments of the hind legs, the sides of the propo- deum, or the under surface of the abdo- men (Braue, 1913; Michener, 1944). In the Apidae and the family most similar to it, the Anthophoridae, the scopa is largely restricted to the outer side of the hind tibia, a restriction unusual in other families of bees (but see the Panurginae in the Andrenidae). While the scopa is brushy in the Anthophoridae, it is reduced in the Apidae to a corbicular scopa or fringe sur- rounding a smooth and often concave sur- face on the outer side of the tibia; the fringe and surface together constitute the corbicula or pollen basket. The manner in which the corbicula is filled is the major topic of this paper; the process 1s some- times called pollen packing. In most bees which carry pollen on the scopa of the hind legs, unmoistened pollen is swept off of anthers or off the hairs of the bee’s head by the front basitarsi, then transferred to the middle legs which also typically sweep pollen off the thorax. The middle legs then transfer their pollen to the scopa. These movements have been seen in halictids, andrenids, and in antho- phorids as different as Ceratina, Xylocopa, Melissodes, and Svastra (R.J., personal ob- servations). In many cases (e.g., for halic- tids, Michener and Wille, 1961; Batra, 1966; Roberts, 1969; for Andrena spp., Michener, unpublished) these movements occur while the bee is on the flower, sup- ported by its legs, and only one leg moves at a time; the legs of a pair are not synchro- nized. (The repertoire of these bees may also include movements, perhaps synchro- nous, performed in flight; leg movements during flight have not been investigated.) The movements for pollen handling and transport appear to be combinations and modifications of apoid self-grooming movements (Jander, 1976, and in prepara- tion). The transfer of dirt (in grooming) is consistently from anterior to more pos- terior legs and it is commonly discarded from the hind legs. The same is true in the case of pollen transfer, but cleaning of the posterior legs is delayed until the bee is in its nest where it removes the pollen to be used for larval or adult food. We have been much impressed by the well known fact that bees of the same spe- cies, no doubt often the same individuals, exhibit different pollen-collecting behavior on different kinds of flowers. The reper- toire of each species is probably extensive. Therefore, conclusions about the evolution of the behavioral patterns or the phylogeny of the bees as shown by such patterns are presented with some hesitation. For ex- ample, it is always possible that a be- havioral pattern thought to be restricted to a derived group of bees will be found as an uncommon pattern, or perhaps com- monly with pollen of a particular con- sistency, in a primitive group. Nonetheless, we have interpreted our findings in evolu- tionary terms, we believe with justification, even though more observations will doubt- less extend the known taxonomic range of some behaviors. MATERIALS Observations of pollen packing and re- lated transport behavior have been made by us and by prior authors on the species listed below. Each species name is fol- lowed by the number of critical observa- tions of pollen packing where known (in parentheses), location, and the observers (initials for the authors of the present paper) or literature references. MELIPONINAE: Trigona (Trigonisca) buysoni (6); Rio Anchicaya, Prov. del Valle, Colombia, collecting pollen from Hedychium coronarium (C.D.M.). Trigona (Paratrigona) impunctata (10); 7.5 km southwest of Kourou, French Guiana, collecting pol- len from a large-flowered Cassia (C.D.M., M.L.W.); 6 km southwest of Kourou, French Guiana, collecting 578 Tue Universiry oF Kansas SciENCE BULLETIN pollen from Stylosanthes (C.D.M.); 12 km southwest of Kourou, French Guiana, collecting pollen from melastomaceous shrub (C.D.M.). Trigona (Paratrigona) subnuda; State of Parana, Brazil, collecting pollen from Tvzbouchina spp. (Laroca, 1970). Trigona (Paratrigona) sp.; Valle, Colombia, on flowers of (C.D.M.). Trigona (Scaptotrigona) pectoralis; Pichinde, Prov. del Valle, Colombia, on flowers of Baccharis (?) (C.D.M.). Trigona (Scaptogrigona) postica; Rio Claro, Sao Paulo, Brazil, collecting cerumen from old Trigona nest (Sakagami and Camargo, 1964). Trigona (Cephalotrigona) capitata (10); 6 km southwest of Kourou, French Guiana, collecting pol- len from Stylosanthes (C.D.M.). Trigona (Tetragona) clavipes (8); Kourou, French Guiana, collecting pollen from white Ipomoea (C.D.M., M.L.W.). Trigona (Tetragona) fimbriata; near Kuala Lum- pur, Malaysia, 10 specimens with pollen loads, no behavioral observations (C.D.M.). Trigona (Tetragona) jaty; Costa Rican colony in- troduced to laboratory at Lawrence, Kansas, by E. M. Barrows, collecting pollen from Prunus (R.J.). Trigona (Tetragona) thoracica and itama; Kepong near Kuala Lumpur, Malaysia, collecting pollen from Cassia spectabilis (R.J.). Trigona (Trigona) amalthea (= trinidadensis) (10); highway summit west of Cali, Prov. del Valle, Colombia, collecting pollen from Cucurbita pepo (C.D.M., M. D. Breed, W. J. Bell). Trigona (Trigona) fulviventris fulviventris, fusci- pennis, and silvestriana; Prov. de Guanacaste, Costa Rica, collecting pollen from Cassia biflora (Wille, 1963). Trigona (Trigona) nigerrima (many); Rio Anchi- caya, Proy. del Valle, Colombia, collecting pollen from Hedychium coronarium (C.D.M., M. D. Breed; cinematography by M. D. Breed). Trigona (Trigona) pallida (many); Kourou, French Guiana, collecting pollen from white Ipomoea (C.D.M., M.L.W.); 7.5 km southwest of Kourou, collecting pollen large-flowered Cassia (C.D.M.,. ML.W.): Trigona (Trigona) spinipes and fulviventris guia- nae; State of Parana, Brazil, collecting pollen from Tibouchina spp. (Laroca, 1970). Trigona (Trigona) fulviventris guianae; 12 km southwest of Kourou, French Guiana, collecting pollen from melastomaceous shrub (C.D.M.). Trigona (Scaura) latitarsis; State of Maranhio, 3razil, collecting pollen from Piper and Amaranthus (Laroca and Lauer, 1973). Pichinde, Prov. del Baccharis (?) from a Trigona (Scaura) longula (10); 7.5 km _ south- west of Kourou, French Guiana, collecting pollen from a large flowered Cassia (C.D.M., M.L.W.). Melipona fasciata and favosa; French Guiana, fresh specimens with pollen loads of various sizes (M.L.W.). Loading behavior not observed because of rapid flight. Melpiona fasciata; Osa Peninsula, Costa Rica, col- lecting mud (R.J.). Melipona Kourou, southwest of pollen from pseudocentris; 12 km French Guiana, collecting melastomaceous shrub (C.D.M.). EUGLOSSINI: Euglossa cordata group and E. ignita (20); 49 km south of Cayenne and 19 km_ southwest of Kourou, French Guiana, collecting pollen from Sabicea near cinezea (C.D.M., M.L.W., G. Otis). Euglossa cordata group (2); Kourou, French Guiana, collecting cerumen from an old Trigona nest (C.D.M.). Euglossa cordata group (10); vicinity of Kourou, French Guiana, males collecting cineole and other scents (C.D.M.). Euglossa championi (many); Rio Anchicay4, Prov. del Valle, Colombia, males collecting cineole and other scents (C.D.M., M. D. Breed; cinematogra- phy by M. D. Breed). Eulaema cingulata (1) and Euplusia auripes (2); vicinity of Kourou, French Guiana, males collecting cineole and other scents (C.D.M.). BomBINI: Bombus spp.; Europe, pollen collecting (Hoffer, 1882; Sladen, 1911, 1912a; Buttel-Reepen, 1915). Bombus cayennensis; 45 km southwest of Cay- enne, French Guiana; fresh specimens with pollen loads (C.D.M.). Bombus — pennsylvanicus (= americanorum); Baldwin, Kansas, pollen collecting on Cassia chamae- crista (observations and cinematography by R.J.). APINAE: Apis mellifera; Europe and North America, pollen collecting (Sladen, 1912b; Parker, 1926; Beling, 1931; Ribbands, 1953; Snodgrass, 1956; Legge and Bole, 1975); Lawrence, Kansas, pollen collecting on Cytisus scoparius (R.J.); Europe, Brazil, resin (propolis) collecting (Sladen, 1911, 1912b; Rosch, 1927; Meyer, 1953, 1954, 1956; Sakagami and Camargo, 1964); Lawrence, Kansas, resin (propolis) collecting (ob- servations and cinematography by R.J.). METHODS Behavior while a bee is on a flower is usually easily observed, but much of the pollen manipulation in the Apidae occurs in flight. Certain individuals, especially of the genus Trigona, hover briefly and rather quietly close to the flowers, and can be watched against the background of the flowers when desired. They then often return to the same flower for more pollen, and repeat the hovering. Such individuals are the principal sources of our new data on leg movements. Considerable detail was visible and the behavior pattern was pieced together from observa- tions of many different hovering individuals, mostly of Trigona pallida and nigerrima, from all possible i PotteEN MAnreucation By Bees (Appar) 579 angles. C.D.M. and M. D. Breed observed pollen manipulation by Trigona as well as leg movements of male euglossine bees in southwestern Colombia; Dr. Breed made moving pictures which were later analyzed by C.D.M., R.J. made moving pictures of resin collecting by Apzs mellifera and of pollen col- lecting by Bombus pennsylvanicus. The films were analyzed with the invaluable help of a Super 8 Lafayette Analyzer Projector. Another source of information is pollen on the legs of bees killed while collecting pollen. The anatomical details and the location of pollen ac- cumulation help in determining the packing behavior. These data, collected by C.D.M. and M.L.W. and recorded largely as sketches and notes made at Kourou, French Guiana, usually substantiate the behavioral data. Various behavioral matters that are not directly related to filling the corbiculae are mentioned in passing. To save space, references in such cases are not usually included, but can be found in Michener 974). The terminology used for the movements in- volved in pollen manipulation is that of Jander (1976). In rubbing, two parts move back and forth, one against the other, without losing contact through- out the action. In scraping, strokes in one direction involve contact, but the parts are separated for the return strokes. For hairy structures such as many bees have, these terms may not seem ideal; words like brushing or combing are more descriptive. However, they do not indicate the distinction drawn between rubbing and scraping and are avoided ex- cept when our observations are not good enough to make that distinction. Orientational terms for the legs, especially for movements relative to the hind tibia, can be con- fusing. In the tibia’s usual position, upward might be regarded as toward the base. Because it is move- able, that usage is avoided, and we use instead basad, toward the base or femoral articulation, and apicad, toward the apex. These are simply directional terms; a basad movement can occur near the apex of the tibia. Because the ubia is often extended posteriorly, upward is taken to be toward the upper margin, Le., at right angles to the long axis of the tibia. The same direction relative to the tibia can be called posterior if the long axis of the tibia is considered to be vertical. TRANSPORT OF NESTING MATERIALS Species of Melipona are regularly seen collecting mud and carrying it in the cor- biculae, and Melipona and Trigona both carry cerumen from other nests, as well as gums and resins, in a similar way. Some species of Trigona also carry vertebrate fecal material, mud, or chewed plant ma- terial in the corbiculae. Bassindale (1955) gives an account of propolis packing by T. braunst and Sakagami and Camargo (1964) illustrate and give an account of the similar packing of cerumen for trans- port by T. postica. The latter authors re- port that the mandibles cut out particles of cerumen which are manipulated and pressed together to form a lump by the mandibles and forelegs (probably basitarsi, C.D.M.). “When the lump attains an appropriate size, one of the middle legs reaches forward, and using bristles on its underside (basitarsus? C.D.M.), the bee transfers the lump very rapidly to the corbicular surface of the hind leg of the same side, which is synchronously moved a little forward.” By repetition of these movements, the accumulation on the cor- bicula grows. Often the middle leg is ex- tended back, and gently presses the grow- ing ipsilateral (same side) corbicular deposit. One or both of the hind legs may be raised in a peculiar way above the wings; the function of this movement, if any, is unknown. One of us (R.J.) has observed collecting and transport of mud by Melipona fasciata. Biting with the mandibles and scratching with the forelegs, the bee loosens a bit of moist mud, which is then taken up by the mandibles. With a backward motion, one foreleg takes the bit of mud from the mandibles. The ipsilateral middle leg scrapes the piece of mud from the foreleg by clasping the foreleg from the outside. This movement appears identical to the cleaning of the foreleg by the middle leg during a normal cleaning bout (Jander, 1976). The middle leg then passes the mud backward and presses it from the out- side into the corbicula. This is usually followed by patting movements of the middle leg on to the mud in the corbicula. Only unilateral (one side at a time) trans- fer of mud was observed; mud may be passed backward on one side several times 580 Tue University oF Kansas ScIENCE BULLETIN before the bee uses the other side. After both corbiculae have been filled, but before taking flight, the bee takes a larger lump of mud between the mandibles, first hold- ing it with both forelegs and then pressing it to the mandibles. Euglossa and Euplusia carry resin in the corbicula and Eulaema carries verte- brate fecal material as well as resin. Eu- plusia sometimes carries small pieces of bark stuck to the resin, if museum speci- mens with such bark are meaningful. One female of the Ewglossa cordata group was observed taking cerumen from fragments of an abandoned Trigona nest. Few ob- servations were possible, so that details are not available, but the bee clearly de- tached pieces of cerumen with its mandi- bles, then hovered and while in the air transferred the cerumen to the corbiculae. The middle legs clearly were seen to syn- chronously carry pieces back to the ipsi- lateral hind legs which were brought for- ward, the middle legs then patting the cerumen into the corbiculae. The bee then alit to obtain more cerumen. The cerumen masses on the corbiculae became large and irregular. Because of behavior in other groups as well as in euglossine males described below, we suspect that euglossines may not always hover to trans- fer sticky materials to the corbiculae syn- chronously with both middle legs. Non- flying Meliponinae and Apinae have been seen to transfer sticky materials to the corbiculae asynchronously, with one mid- dle leg at a time. Bombus constructs its nests using a mix- ture of wax and pollen. Wax is secreted by the bees. So far as known, pollen is not transported differently for construction than for food; in fact young larvae may eat some of the wax-pollen mixture. Trans- port of pollen is described below. Apis mellifera, both in Europe and the Africanized bees in Brazil, collects resin (propolis) or cerumen exactly as described above for Trigona postica except that the hind legs are not raised (Sladen, 1912b; Rosch, 1927; Meyer, 1953, 1954, 1956; Saka- gami and Camargo, 1964; see IBRA, 1976). One of us (R.J.) made additional observa- tions of such behavior, summarized as follows: The mandibles gnaw the surface of the resin while at the same time the forelegs and occasionally the middle legs scrape and scratch the surface. Loosened pieces are released by the mandibles to both forelegs. In a very quick movement, one foreleg swings backward and the ipsi- lateral middle leg grasps the foreleg from the outside and scrapes off the propolis. The femorotibial joint of the middle leg is sharply bent in this operation (as when the middle leg cleans the foreleg in groom- ing, Jander, 1976) and the piece of propolis sticks to the inner posterior side of the mid basitarsus, presumably held by the sharp bristles in that area. Immediately the mid- dle leg swings backward and the piece of propolis is pressed by the mid basitarsus onto the corbicula of the ipsilateral hind leg. Then the middle leg is pulled forward while still in contact with the hind leg; the resultant scraping movement leaves the resin in the corbicula. The pressing and scraping may then be repeated a sec- ond time. All these movements were per- formed while the bees were on the ground. MANIPULATION OF SCENTS BY MALE EUGLOSSINI Male euglossine bees collect certain scented substances from orchid flowers and other sources (Vogel, 1966) and can be attracted to scents provided artificially (Dodson et al., 1969). Interestingly, the movements involved are similar to those of pollen collecting by females, although other male bees, so far as known, use similar movements only for self-grooming. Several species were observed; the following ob- servations apply equally to Euglossa, Eu- plusia, and Eulaema, and largely support PotLeN MANIPULATION BY Bees (APIDAE) 581 the detailed accounts and illustrations by Vogel (1966) and Evoy and Jones (1971). (The middle tibia cited by the latter au- thors appears to have been in reality the middle basitarsus.) The front tarsi are rubbed on the ma- terial containing the attractant; sometimes they also scrape the head, especially the eyes, downward and forward as in groom- ing movements. The proboscis is not ex- erted; the tips of the antennae are directed down to or almost to the scent source. After such rubbing, the bee usually hovers and while in the air the middle legs move forward synchronously and apparently scrape the fore tarsi with the under sur- faces of the ipsilateral mid basitarsi. The middle leg is probably flexed as in normal cleaning of the foreleg. The middle legs now move back, and at the same time the hind legs are synchronously flexed and rotated forward. Now the under surfaces of the mid basitarsi scrape synchronously upward, i.e. at right angles to the long axes, across the outer surfaces of the en- larged hind tibiae. The two structures are nearly parallel to one another, and only at the end of each stroke may the mid basi- tarsus contact the hairs of the dorsal meta- tibial groove. The whole sequence is re- peated several times while the bee is hovering, before it alights again at the source of the scent, or departs. The greater part of the contact of the middle tarsus with the hind tibia is with the simple, convex, short-haired, outer, tibial surface and not with the groove which is supposed to absorb the attractant substances. Rarely, instead of hovering, a bee grasps the edge of a leaf with its mandibles after rubbing an attractant with its front tarsi, and then, hanging by the mandibles, it goes through the leg movements described above. It thus frees the middle and hind legs for the movements usually performed while hovering. APID POLLEN MANIPULATION Tue Foretecs. As in most other bees, the front legs (basitarsi) and the proboscis remove pollen from the anthers of flowers; pollen on the proboscis and the head is subsequently scraped off by the forward and downward movements of the front legs. Thus for most species at most kinds of flowers, the front legs are particularly important, being the primary pollen gath- ering structures. Pollen is generally trans- ferred backward directly from the front legs to the middle legs (basitarsi), which also clean pollen from both dorsal and ventral surfaces of the thorax. However, these movements in the meliponine genus Trigona are unusual, as described below. In flowers whose anthers are readily accessible, Trigona species commonly bite an anther repeatedly, loosening pollen, the antennae being bent down, their tips con- tacting the anther or nearly so. If there is already loose pollen, biting is unnecessary. At least T. capitata, pallida, thoracica, and nigerrima, and presumably all species, ex- tend the proboscis, contacting the anthers repeatedly (Fig. 1). The pollen is pre- sumably made sticky with nectar in this way, as is the case with Apis. As these activities continue, bees brush the anthers and especially the proboscis with the front tarsi, presumably accumulating pollen on the hairs of the basitarsi. The proboscis is scraped downward, toward its apex, with both forelegs (basitarsi?) synchronously. The pollen is then transferred by the front tarsi to an area of backward-directed, stiff hairs on the ventral surface of the mese- pisternum in front of and between the middle coxae, and to similar hairs on the middle and hind coxae. These movements are synchronous, left and right forelegs moving simultaneously, the bee being sup- ported by the middle and hind legs (Fig. 2). Moving pictures show that the fore tarsi are scraped forward across the coxal 582 Tue University oF Kansas SciENCE BULLETIN Fics. 1, 2. Trigona nigerrima taking pollen from Hedychium. In Fig. 1, in an interval between biting, the tip of the glossa is in contact with the pollen source and the forelegs are about to scrape the glossa and floral surface. In Fig. 2 the forelegs are trans- ferring pollen to the thoracic venter and mid and hind coxae. These drawings are based on moving picture frames, but much detail has been added since they showed mainly silhouettes. and mesepisternal vestiture to transfer pol- len to the latter. The behavioral repertoire of Trigona species must include somewhat different movements for flowers with loose pollen that does not need to be freed with the mandibles and that may be picked up by parts of the body and appendages other than the front tarsi. When Trigona species are foraging at flowers with abundant loose pollen that adheres to the body, it is no doubt brushed off of the different parts of the body by the inner surfaces of the basitarsi of all the legs. T. pectoralis and T. (Paratrigona) sp. collecting pollen on Baccharis (?) flowers had pollen densely caked on the inner sides of the hind basi- tarsi. Such pollen would never be trans- ferred to the thoracic venter, but must pass directly to the corbicula. This is in contrast to our observations on T. pallida and nigerrima which had little pollen on the hind basitarsi. Potten Packinc. We recognize two basic types of pollen packing in Apidae. Type I resembles the manipulation of nest ma- terials, as described above, in that sticky masses are placed directly onto the corbicu- lae by the middle legs. In Type II, pollen is placed near the distal end of the cor- bicula rather than directly on the corbicular surface, and is then pushed basad into the corbicula. Special morphological features of the distal end of the hind tibia and base of the basitarsus are necessary for Type II pollen manipulation; these fea- tures differ among the groups of Apidae, as do the pollen-packing movements. There exists, therefore, various subtypes of Type II. Tyre I Potten Pacxine. It may be that all female apids retain, as part of their behavioral repertoire, pollen packing in which the middle legs place pollen directly into the corbiculae, as noted below for Trigona (Trigonisca) buyssoni and for T. amalthea on Cucurbita. For the former species this may be the principal method. It was observed on the same flowers where T. nigerrima was transferring pollen in the way more common for the Apidae (Type Il). Most species probably use Type I packing only for large masses of sticky material such as resin, mud, or the large pollen masses of Cucurbita. Potten Manipuration sy Bees (Apipae) 583 Trigona (Trigonisca) buyssont bites the anthers of Hedychium, moistens the loos- ened pollen with nectar (?) from the glossa, and places it on the mesepisternum with the fore tarsi, as described above (under The Forelegs) in greater detail for other Trigona species. Then, while still standing on the anther or petal, supported by front and hind legs, the middle legs move pollen from the thoracic venter to the corbiculae. (The front legs could theoretically play a role in removing pollen from the thoracic venter, but if so, it must be one front leg at a time rather than synchronously since at least one front leg must support the body. Nothing of the sort was seen.) Movement of the middle legs is synchronous, and the pollen is scraped or patted onto the ipsilateral cor- biculae by the middle basitarsi. The abundant, coarse, sticky pollen of Cucurbita probably presents special prob- lems or opportunities for bees collecting it. Our observations were made on Trigona amalthea, a form very close to and perhaps conspecific with T. silvestriana whose Type II handling of Cassia pollen has been ob- served by Wille (1963). But on Cucurbita, after several transfers of pollen (like those described under The Forelegs) from the anthers or from thick accumulations on the corolla beneath the anthers to the mesepisternum and adjacent coxae, the bee, while still in the flower and now supported by front and hind legs, uses the middle basitarsi to remove pollen masses from the mesepisternum and pat them onto the ipsi- lateral corbiculae. The result, after several such movements, is large but loose and 1r- regular pollen masses on the corbiculae. The movements are similar to those of T. postica described by Sakagami and Camargo (1964) for placing cerumen on the hind tibiae for transportation. These movements make no use of the special structures (penicillum, rastellum) at the apices of the tibiae. An old report (Hoffer, 1882) says that pollen is “pressed with the middle legs into the corbicula of the hind leg” by Bombus. This may indicate that Type I behavior is part of the repertoire of Bom- bus, as it is of other groups. Even in Apis mellifera, when the cor- bicular pollen loads become large, the bee may pat them many times with the ipsi- lateral middle legs, probably to smooth and compact them. Some pollen may be added directly to the corbicular pollen masses in this way, although the quantity appears to be small (Parker, 1926). Type II Potten Packinc. This type of pollen packing involves (a) placement of sticky pollen in the region of the hind tibiotarsal joint by movements which de- pend upon the kind of bee and the kind of flower, and (b), most characteristically, pushing the pollen basad up the outer sur- face of the tibia from the region of the tibiotarsal joint. This movement of the pollen places it into the smooth and nearly hairless corbicula. This is unlike the filling of the scopa of other families of bees, where the dense scopal hairs prevent such a process, and it is unlike the Type I apid process in which pollen is placed directly into the corbicula by the middle leg. As described below, in Apis, Bombus, and probably the Euglossini, pollen is pushed basally into the tibial corbicula by the auricle, i.e., the broadened base of the basitarsus. As emphasized by Buttel- Reepen (1915), Maidl (1934), and Win- ston and Michener (1977), Meliponinae have no auricle. A major objective of this study, therefore, was to learn how Meli- poninae fill their corbiculae. Ipsilateral Type II Pollen Packing. Corbicular loading was observed for Trt- gona pallida and nigerrima, and has been briefly described for other species by Wille (1963) and Laroca (1970). After several foreleg movements that place pollen on the thoracic venter (Figs. 1 and 2), the bee 584 Tue University of Kansas Scrence BULLETIN takes wing, thus freeing the middle and hind legs of their supporting function and enabling them to make pollen-transferring motions. While the bee is hovering, the middle legs synchronously rotate far for- ward (Fig. 3) and may scrape backward over the thoracic venter, removing pollen from the hairs of that area. However, in their far forward position they hide the front legs so that it is difficult to see, and photographs do not show, whether or not the front legs first remove the pollen from the mesepisternum and transfer it to the mid legs, as stated by Wille (1963) and Laroca (1970). Our impression from the moving pictures is that both may happen, the forelegs scraping pollen perhaps from median ventral areas for transfer to the mid legs, and the latter removing pollen from the lateroventral areas. In any event the mid legs quickly rotate back (Fig. 4) from a far forward position to a backward position and transfer pollen to the ipsi- lateral hind legs. At first the mid tibia and tarsus come back almost parallel to the hind tibia (Fig. 5), but then the hind leg is flexed forward (Fig. 6) so that the mid basitarsus lies across the outer surface of the tibia or the tibio-basitarsal joint of the hind leg, at right angles to the long axis of the hind tibia. The movement of the hind leg is especially noticeable in hovering T. clavipes and nigerrima be- cause of the long and dark posterior legs, but it occurs in all species and is necessary if the contact is to be at right angles to the hind tibial axis. The position of the mid leg against the outer surface of the apex of the hind tibia shows repeatedly in our moving pictures of T. nigerrima hover- ing while manipulating pollen. Straighten- ing of the hind leg (Fig. 7) and to a minor extent, simultaneous forward movement of the mid leg combine to scrape the pol- len-carrying mid basitarsus across the apex of the hind tibia. The photographs show that sometimes the mid basitarsus is also Fics. 3-7. Trigona nigerrima packing pollen while hovering near Hedychium flowers. Fig. 3, Mid leg has removed pollen from thoracic venter or possibly foreleg. Fig. 4, Mid leg moving quickly backward. Fig. 5, Mid leg over pollen mass (dotted) on hind leg. Fig. 6, Hind leg bent forward so that mid leg lies across tibiotarsal joint. Fig. 7, Hind leg straight- ened and mid leg moved forward, pulling mid tarsus across hind leg. Drawings prepared as for Figs. 1, 2. PoLLEN MANIPULATION BY BEEs (ApipaE) 585 ~N i 4 EF A ay, Ze a. Ss OW Fics. 8-13. Hind leg of worker of Trigona pallida. Fig. 8, Tibia and basitarsus. Fig. 9, Same with large pollen mass. Fig. 10, Basitarsus showing basal structure. Fig. 11, Tibia and basitarsus in anterior view. Fig. 12, Apex of tibia, outer view. Fig. 13, Distal view of apex of tibia. p, penicillum; r, rastellum; a, auricular area; c, corbicula. seemingly pressed against the pollen mass in the corbicula. This could be either ad- justment of the pollen mass or a Type I addition of pollen, but the arrangement of pollen from different sources suggests the former (see section on The Corbicular Pollen Load in Meliponinae). The pertinent hind tibial and basitarsal structures of Trigona pallida are shown in Figures 8 to 13. The penicillum is a row of stiff bristles arising on the lower (or anterior) distal angle of the tibia. They sweep upward (or posteriorly) parallel to the tibial apex, but well separated from it, and then curve distad at their tips. The longest bristles are the outermost while those nearer the corbicular surface are progressively shorter (not true of all spe- cies). The hind basitarsus except at its tibial articulation is offset mesally (Fig. 11), so that there is a gap between its outer surface and the curved apices of the bristles of the penicillum. The structure suggests that through this space the ipsilateral mid- dle basitarsus is drawn at right angles to the long axis of the hind tibia in order to transfer pollen onto the latter. The apices of the penicillar bristles would scrape the 586 Tue University oF Kansas ScrENCE BULLETIN pollen out of the hairs on the outer surface of the mid basitarsus, and the penicillar curvature, during posterior movement of the hind tibial apex relative to the mid basitarsus, would force such pollen basad onto the outer surface of the tibia. The progressive shortening of the penicillar bristles from the outer to the inner ones makes the comb oblique, tending to push the pollen against the apex of the corbicula. Thus the movement of hind leg relative to the middle could alone be responsible for pushing some pollen basad into the cor- bicula. The posterior basal area of the outer surface of the hind basitarsus is provided with hairs which are directed posteroba- sally (Fig. 10), not apically like most other hairs. We speak of this part of the basitar- sus as the auricular area because of its location, comparable to that of the auricle of other subfamilies of Apidae. The hairs of the auricular area often have some pol- len on them and presumably serve to scrape pollen off of the inner hairs of the ipsilateral middle basitarsus. Laterad movement of the hind basitarsus would press the mid tarsus between the penicil- lum and the posterior basitarsus itself (in- cluding the auricular area) as the hind tibial apex is moved backward relative to the mid leg. Posterior flexion of the hind basitarsus would then help to push pollen that comes off onto the auricular area up onto the corbicula, thanks to the direction of the hairs in that area. As repeated passages of the mid basi- tarsus add pollen to the distal end of the ipsilateral outer tibial surface (supple- mented by the pollen from the inner sur- face of the contralateral hind basitarsus, see below), the added pollen must pile up and be pushed basally. The motive forces are presumably the scraping movement of the hind leg along the mid leg as already described, supplemented by back and front (up and down) flexions of the basitarsus. Examination of freshly killed pollen col- lectors of Melipona and Trigona frequently show pollen on the tibial apex in the space between the inner side of the penicillum and the outer sides of the basitarsus and rastellum. In T. pallida a small brush (Figs. 10 and 11) arising along a curved line on the basal part of the auricular area, behind the penicillum, can, with backward flexion of the basitarsus, push such pollen basad onto the corbicular surface. It con- sists of weak hairs hardly able to move the whole pollen mass, but its effectiveness with small amounts of pollen was shown by drawing a detached middle tarsus with pollen on it through the gap described above (between the penicillum and the hind basitarsus) on the hind leg of a freshly killed worker of T. pallida. Part of the pollen was combed off as expected by the penicillum, but the quantity was not enough for the curvature of the penicillum to push much pollen basad onto the cor- bicula. The basitarsus was then flexed backward, with the result that the brush moved the pollen onto the distal end of the corbicula where it remained, held by its stickiness. In Melipona, as in most Trt- gona, there is no defined brush, but rather a hairy auricular area (Fig. 14) with hairs directed posterobasally, as described above. The frequent presence of small amounts of pollen on this area in Melipona and Trigona suggests its importance in push- ing pollen upward. Such movement is possible because of the highly flexible tibio- tarsal joint, activated at least in Apts (Snodgrass, 1956) by three muscles. A casual observer might suspect from the curvature of the penicillar bristles that the penicillum must somehow function as a scoop, accumulating material on its con- cave surface. Actually if our interpretation is correct, the penicillum scrapes pollen onto its convex surface. Like the rastellar bristles in Apis, the penicillar bristles comb in the direction of the apices of the hairs — Potten ManrputaTion sy Bees (Appar) 587 from which they are removing pollen. Such scraping movements can remove in- definite amounts of pollen because new pollen easily pushes away that already present. This would not be the case for a scoop. Contralateral Pollen Packing in Apis and Bombus. Although it largely occurs while the bees are in flight, the process of pollen manipulation and packing onto the corbiculae for transport has been stud- ied repeatedly for Apzs (for references, see Ribbands, 1953; Snodgrass, 1956, and Legge and Bole, 1975). Sladen (1912a) and Buttel-Reepen (1915) found the proc- ess for Bombus to be similar, and the structures involved in Apts and Bombus are remarkably similar. The following comments are based primarily on pub- lished material on Apis mellifera. The bees gather pollen from their hairy bodies or from anthers by movements of the legs. At the same time, they may moisten it with nectar, making it sticky. The basi- tarsi bear the primary brushes involved. The front basitarsi brush both anthers and the proboscis, the latter adding nectar and making the pollen sticky. These basitarsi also brush the head and front of the thorax. R.J. observed Apis workers taking pollen from flowers of Cytisus directly with the middle, rather than the front tarsi, and bees on flowers with loose pollen will have accumulated pollen on most parts of the body and legs. The bee now leaves the flower and hovers. Pollen on the front basitarsi is transferred to the middle ones, which also scrape pollen off of much of the thorax both dorsally and ventrally. The middle basitarsi, one at a time (Beling, 1931), are now scraped between the inner sides of the apposed hind basitarsi. These basitarsi also scrape pollen off of the abdo- men. The hind legs are rapidly rubbed against one another in a pumping motion. In this process the rastellum scrapes distad, removing pollen from the inner surface of the contralateral basitarsus. The pollen ac- cumulates on the posterior basal projection (auricle) of the basitarsus or between the auricle and the rastellum. Then, by pos- terior (upward) flexion of the basitarsus, the auricle forces pollen basad onto the corbicula. Repetition adds more and more pollen at the distal end of the corbicula, forcing the first pollen collected toward the base of the tibia and ultimately filling the entire corbicula with pollen, which may also be patted from the outside by the ipsilateral mid basitarsus, as noted in the section on Type I Apid Pollen Manipu- lation. It seems possible that the latter movement may also add some pollen di- rectly from the mid legs to the corbiculae. In both Apis and Bombus, however, pollen loads composed of different colored pollens from different flowers show that the ma- terial is added from the apex of the tibia (Buttel-Reepen, 1915). Because the published information on Bombus is largely presented only by indi- cating the similarity to Apis, R.J. made observations and moving pictures of work- ers of B. pennsylvanicus collecting pollen from Cassia. While incomplete, these studies generally verify and supplement the observations made early in the century by Sladen and Buttel-Reepen. A Bombus takes a position ventral side up under an inverted flower, usually hanging by its forelegs, sometimes with the midlegs also on the flower. This is clearly a special position for extracting pollen from flowers like those of Cassia with tubular anthers. The forelegs do not perform their usual pollen-collecting function and the mid and hind legs are probably freer than usual for pollen-manipulation movements usu- ally performed in flight. The hanging bee produces rhythmic buzzing sounds—vibra- tions which no doubt release the pollen through the openings in the apices of the anthers (Michener, 1962; Wille, 1963). The pollen falls onto the underside of the bee and especially onto the venter of the abdomen. 588 Tue UNIVERSITY OF While the bee is hanging from the flower, the hind legs (probably under sides of basitarsi but possibly the rastella) sweep back and forth (scraping or rubbing, we cannot say which), transversely, synchro- nously, removing pollen from the abdom- inal sterna. In each such movement, as they approach the midventral line of the abdomen, the distal parts of the tarsi meet and interfere with the process. The result is a longitudinal, midventral line of pollen, not swept up by the basitarsi or tibiae. While still hanging from the flower, the bee performs at least five movements with the middle legs. They were observed to occasionally groom a foreleg by flexing so that the foreleg is scraped simultane- ously by mid femur and basitarsus. (This familiar movement is presumably function- less for pollen collecting in the context of Cassia flowers.) Midlegs more often scrape anteriorly on the dorsum of the thorax and posteriorly on the venter between fore and mid coxae. The latter movement should sweep Cassia pollen from the tho- racic venter. The middle legs also, one at a time, extend backward, are appressed between the hind basitarsi, then pull for- ward as the hind legs are straightened backward. This scrapes pollen from the middle basitarsus onto the under sides of the hind basitarsi. (Basitarsi of pollen- collecting individuals of Bombus cayen- nensis had abundant pollen on their under surfaces—C.D.M.) Finally, midlegs oc- casionally pat the corbicular pollen loads, usually starting at the basal part of a pol- len mass and working toward the distal part. Bombus in flight were photographed while cleaning the forelegs with the mid- legs and the midlegs with the hind legs— movements also seen while bees were hang- ing from flowers. In addition, film analysis revealed pumping movements of the hind legs like those described for Apis. It was verified that the hind basitarsi are not KaAnsAs SCIENCE BULLETIN pressed against one another as during grooming; contact is in the vicinity of the tibiotarsal joint and we believe that the rastellum of each tibia combs pollen from the underside of the opposite basitarsus. This would lead to deposition of pollen on the auricle, which would push it into the distal end of the corbicula, just as in Apis. Contralateral Pollen Packing in Meli- poninae. Although Meliponinae lack auri- cles, most of them have rastella; it was therefore natural to look for contralateral corbicular loading similar to that known for Apis. In some species of Trigona, contralateral pollen packing is of little importance, at least at the flowers where we made our observations. Thus the hind basitarsi of Trigona nigerrima and pallida are brought rather close to one another beneath the body during pollen loading, so that the middle legs can be drawn across the outer surfaces of the hind legs, as de- scribed in the section on Ipsilateral Type II Pollen Packing. The hind legs may touch one another and make some basad-distad alternate pumping movements. Moving pictures of hovering T. nigerrima taken from the rear show occasions when the inner apex) of a hind tibia, bearing the comb or rastellum, combs downward over the inner surface of the opposite or contra- lateral basitarsus, followed by the same movement of the opposite leg. This alter- nating or pumping movement is probably not important for pollen manipulation in the cases most intensively studied, Trigona pallida on Ipomoea and T. nigerrima on Hedychium, for in these cases the pollen was picked up as described above, exclu- sively by the front legs; little or no pollen got onto the abdomen or hind tarsi. More- over, pollen was rarely found on the inner sides of the hind basitarsi in the pollen- collecting bees, and when present, there was but little. The pumping movements in these instances were probably cleaning PoLLEN MANIPULATION BY Bees (APIDAE) 589 or self-grooming activity or stereotyped ac- tivity that may have importance in pollen manipulation at other kinds of flowers. These species, however, like all others in the subgenus Trigona and certain species of the subgenus Tetragona, have a large sericeous area, not covered with bristles, on the inner side of the hind basitarsus. Such a basitarsus must be inefficient, com- pared to that of other Apidae, in brushing pollen from the middle tarsus and there- fore in contralateral pollen packing. The other subgenera of Trigona, like other Meliponinae, have the underside of the hind basitarsus fully bristled and it is among such forms that contralateral pol- len packing is most likely to occur. As noted above, much pollen was found on the inner sides of the hind basitarsi of T. (Scaptotrigona ) pectoralis and T. ( Para- trigona) species collecting pollen on Bac- charis (?). The species of Scaura discussed in the nest section must depend largely on contralateral pollen packing. More sig- nificantly, there was much pollen on the inner sides of the hind basitarsi of T. (Cephalotrigona) capitata and T. (Para- trigona) 1mpunctata collecting pollen from Stylosanthes, even though this is a small- flowered legume whose pollen was being removed from the flowers by the front tarsi only. Thus, unlike Scawra and the bees on Baccharts, those on Stylosanthes were not getting pollen on the body. Pollen must have been actively transferred to the inner sides of the hind basitarsi. Behavioral observations, while not de- cisive, indicated the same conclusion. Pol- len collectors of T. capitata, at least those with large pollen loads, seem to place the mid tarsi between the hind tarsi while hovering, after visits to one to several flowers. Thereafter distad-basad pumping movements of the hind legs were con- spicuous, the inner apices of the tibiae ap- parently scraping the inner surfaces of the basitarsi. Sometimes, however, the mid basitarsi appeared to be outside the hind tibia, which (together with the presence of a strong penicillum) suggests that ipsi- lateral packing also occurs. T. (Paratrt- gona) impunctata is too small for detailed observations while hovering, but it very rarely seems to transfer pollen while rest- ing on a flower. One mid tarsus at a time is extended posteriorly and pulled forward between the two hind basitarsi which are held with their inner surfaces apposed. This movement was seen performed by two pollen-collecting individuals. The only problem in its interpretation is that this is a typical apoid cleaning movement. The observer (C.D.M.) believed that the move- ment was pollen transferral, but recog- nized that it could have been merely clean- ing of the middle tarsus. It was followed by hovering and pumping movements of the hind legs, suggesting pollen packing. Aside from the bristles on the inner side of the hind basitarsus, an essential structure for contralateral pollen packing is the rastellum. This is a row of bristles (Fig. 13) along the inner margin of the apex of the posterior tibia. In Apis and Bombus the rastellum functions to comb pollen off of the inner surface of the con- tralateral hind basitarsus, and it extends more or less the full width of the tibial apex. In Meliponinae the row of bristles is commonly shorter, being largely poste- rior to the basitarsal articulation and peni- cillum, but probably has the same function; the pumping motion of the hind legs in- volves combing of hind basitarsi by the rastella. The result is accumulation of pol- len from the contralateral basitarsus on the apex of the tibia lateral to the rastellum, from which location basitarsal movements combined with the hairs of the auricular area can presumably push pollen basad onto the corbicula, as in the case of ipsi- lateral pollen packing. Probably another significant function of the rastellum is as a fence to prevent 590 Tue University oF Kansas ScreNcE BULLETIN the pollen that 1s transferred to the outer surface of the hind leg from being pushed onto the inner surface in the course of the tarsal movements. Even if, as in the Trigona pallida and nigerrima which we studied, little or no pollen is coming from the inner surfaces of the hind basitarsi, pollen arriving on the outer side of the tibiotarsal area might leak through and be lost on the inner surface of the hind tibia in the absence of a rastellum. The Corbicular Load in Meliponinae. Since in both ipsilateral and contralateral pollen-packing movements, pollen is placed at approximately the hind tibiotarsal joint, there must be a mechanism for forcing pollen basad from that area into the cor- bicula. This mechanism has been described above, and it is reassuring that pollen loads collected from two or three different kinds of flowers by Melipona fasciata, M. favosa, and Trigona fimbriata indicate the ex- istence of such a mechanism. The pollens in all such loads are arranged as though each new kind were added from the distal end of the tibia (Fig. 14), pushing pre- viously acquired pollen basad over the corbicular surface. If successive kinds of pollen were patted onto the surface of the load (Type I), the different kinds would constitute shells one over the other; this is not the case. In other apid subfamilies the auricle of the basitarsus plays an es- sential role in pushing pollen basad into the corbicula but, as noted before, the Meliponinae have no auricle. As the pollen mass on the meliponine corbicula enlarges, it is held and supported by the erect hairs of the lower (or ante- rior) corbicular margin, by the erect “outer penicillum” or parapenicillum of species such as Trigona pallida, by the penicillum proper, and by the few curled hairs at the distal end of the posterior margin of the tibia. The long upper (or posterior) fringe of the corbicula in most Trigona species, however, does not contain more or less Fic. 14. Apex of tibia and base of basitarsus of Melipona fasciata, worker, showing by shading posi- tions of different kinds of pollen. Abbreviations as for Figs. 8-13. Note the pollen on the auricular area behind the penicillum. erect or curled hairs that enclose a corbicu- lar space, as in Apis, Bombus, and Melt- pona, but extends posteriorly from the corbicular surface. The pollen mass moves partly out over these hairs (Fig. 9) and the stickiness of the pollen and the liquid incorporated with it holds the pollen both to the corbicular surface and to the pos- terior fringe of hairs. The pollen mass may be shaped to some degree, or adjusted, by patting movements of the middle legs. Specializations of Meliponine Pollen Collecting. There doubtless exist, within the behavioral repertoire of various meli- ponine bees, many modifications of the patterns described above. Wille (1963) describes how Trigona f. fulviventrts, fuscipennis, and silvestriana cut holes in the tubular anthers of Cassia biflora and extract pollen from them with the glossa. Laroca (1970) indicates that T. spinipes and sabnuda behave smilarly with similar tubular Tzbouchina anthers, while T. fulvi- ventris guianae cuts the tips off of the Tibouchina anthers and then exploits them in the same way. PoLLeN MANIPULATION BY Bees (ApipAr) 59] Similar observations were made by C.D.M. at flowers of another melasto- maceous shrub in French Guiana. The flowers are managed differently by differ- ent meliponine bees, as follows: Melipona pseudocentris curls the body over the group of anthers and buzzes, receiving the pollen from the tubular anthers on the under side of the body as described for other bees by Michener (1962) and Wille (1963). Tri- gona impunctata chews the basal thick parts of the anthers open and extracts pol- len with the glossa and fore tarsi. T. fulviventhis guianae chews off the attenu- ate distal parts (one third to one half) of the anthers, thus providing an entrance much larger than the small apical pore, and reaches in to extract pollen with the glossa. It then scrapes pollen off of the tongue wth simultaneous distad move- ments of the front basitarsi. A single bee often cuts the apices off of most or all of the anthers in a flower before going on to another. Trigona species sometimes are gleaners, picking up pollen from corolla surfaces where it falls following visits to anthers by other insects. Although the pollen on the surfaces is usually invisible, bees collect large pollen loads on the corbiculae from these sources. Wille (1963) described such behavior for T. jaty, nigra, and testacet- cornis on Cassia biflora. On another Cas- sia species C.D.M. and M.L.W. observed that T. ¢mpunctata visited only the anthers, biting them to get pollen, and T. pallida usually did the same thing. The latter species, however, was sometimes also a gleaner, going over the corolla surface be- low the anthers with the antennal tips down to the surface, the glossa slightly excerted, and the mandibles moving; the fore tarsi swept up the pollen, especially by scraping distad on the glossa, to which it probably stuck because of regurgitated nectar. The rest of the manipulation for both T. ¢mpunctata and T. pallida was as described previously. The subgenus Scaura of Trigona, how- ever, appears to consist of specialized gleaners and visitors to inflorescences con- sisting of relatively broad surfaces from which pollen can be swept up. These bees have extraordinarily large and hairy hind basitarsi (also hairy middle basitarsi), the principle subgeneric characteristic. Laroca and Lauer (1973) describe pollen collecting by T. (S.) latitarsis from the cylindrical inflorescences of Prper and from the leaf surfaces beneath the flowers of Amaran- thus. They describe the use of the hind basitarsi for sweeping up pollen and the rubbing of the hind legs against one an- other, as described below for T. longula. T. latitarsts is a minute bee, undoubtedly dificult to observe. From its morphology, we assume that its behavior is similar to that of T. longula, a larger species for which we obtained fuller, although still incomplete, information on pollen manipu- lation. Trigona longula was visiting a large- flowered Cassia species. Only rarely do these bees go to the anthers. When they do, they collect pollen and manipulate it, so far as we could see, like other species of the genus. Nearly all the pollen collect- ing was by gleaning from petal, bud, or leaf surfaces below the anthers. (Flowers were visited by Xylocopa, Eulaema, Eu- glossa, and Centris, whose buzzing releases pollen, as shown by Wille, 1963.) On such surfaces, there was no noticeable deposit of pollen, but individuals of T. longula were able to fill their corbiculae there. In ordinary walking, T. longula moves like any other Trigona species. On pollenifer- ous surfaces, however, the middle and hind legs are splayed out, the inner surfaces of the basitarsi against the substrate, the hind basitarsi bent forward almost at right angles to the body even though the femora and tibiae are directed backward. With the legs in this position, the bee shufiles 592 Tue UNIversiry oF Kansas ScIENCE | along, dragging the distal part of the abdo- men. The tips of the antennae are bent down to the surface; the front legs perform ordinary walking movements. The basi- tarsi and abdomen must pick up pollen from a rather broad swath as the bee moves along. A collecting bee frequently stops, raises its abdomen, and while supported by fore and mid legs, scrapes backward over the abdominal surface synchronously with the inner surfaces of the hind basitarsi. Then, beneath the abdomen, it rubs or scrapes the inner surfaces of these basitarsi against one another or the apices of the tibiae in the course of pumping movements of the hind legs. Then the hind legs are low- ered for support and the middle legs are brought back usually synchronously to the regions of the apices of the hind tibiae, and the mid basitarsal regions are pulled forward across the outer surfaces of the apices of the ipsilateral hind tibiae. This seems to be a shorter movement than in T. pallida, perhaps because the hind legs are not moveable, being used in support, and is oblique, not at right angles to the tibia. Less often the bee hovers and transfers pollen to the corbiculae by means of leg movements that appear to be similar to those of other Trigona species. Details, however, could not be discerned. Microscopic examination showed that the underside of the apical part of the abdomen (metasomal sterna 4 and 5) has hairs which are curved downward at their apices and thus should readily pick up pollen as the abdomen is dragged forward across the substrate. The inner surfaces of the mid and hind basitarsi are unusually hairy. The outer surface of the broad hind basitarsus is swollen, convex except for the broadly concave auricular area (Fig. 15). The rastellum is long for a meliponine bee. From these observations we suppose that pollen collected on the inner surface 3 ULLETIN Fic. 15. Hind tibia and tarsus of worker of Trigona (Scaura) longula (modified from Schwarz, 1948). of the hind basitarsus, both from the sub- strate and from the abdomen, is combed off that surface by the contralateral rastel- lum during the observed pumping move- ments. Pollen should pile up outside of the rastellum; freshly killed pollen col- lectors had pollen between rastellar bristles and on the outer rastellar surface. Then the concave auricular area, functioning like the auricle in other groups of apids, pre- sumably pushes pollen up into the cor- bicula. The unusual width of the hind basitarsi, their use and that of the abdo- men for pollen collecting, and their gen- eral outer convexity so that a concave auricular area can be present, all suggest that contralateral pollen transfer from the inner sides of these basitarsi is more im- portant than in most Trigona_ species. Pollen collected by the front and middle legs is manipulated ipsilaterally, to judge by the observations described above, the penicillum and auricular hairs combing it off and directing it up into the corbicula. We saw no evidence of pollen being re- moved from the hind legs by the middle legs for transfer to the corbicula, as indi- PoLLEN MANIPULATION By Beets (APIDAE) 593 cated by Laroca and Lauer (1973) for T. latitarsis. As this would involve a for- ward movement of material, something not seen in grooming or pollen handling behavior of any other bees (Jander, 1976, and in preparation), we suspect an obser- vational error. This would not be surpris- ing considering the minuteness of T. latitarsts. The Euglossini. Pollen collecting by Euglossa near cordata and ignita was ob- served on small tubular flowers (Sabicea) that served also as a nectar source. Hairs on the basal part of the proboscis pull pollen out of the flower. When a_ bee rears back to withdraw the long proboscis, it places the front tarsi (basitarsi?) on either side of the proboscis, scraping distad several times with the front legs, thus pre- sumably removing pollen from the pro- boscis. While this is going on the epipharynx, which is extraordinarily long in Euglossini, is exerted and probably adds nectar to the pollen. The proboscis, used for this pur- pose in other apids, is so long in Euglos- sini that it could scarcely have this func- tion. The bee then takes wing and hovers, or rarely grasps a leaf edge with its mandi- bles and hangs. In either case, the middle and hind legs are freed of their support function, so as to allow pollen manipula- tion, which is rapid and difficult to observe. The front legs come back synchronously and apparently the tarsi are scraped by the basitarsi of the flexed middle legs, in the usual way. The latter then moved back synchronously to contact the hind legs. At least part of the time the mid leg pats the corbicula and pollen load on the out- side of the ipsilateral hind tibia. Rubbing of the inner surfaces of the hind legs one on the other (pumping motion) was also probably observed; certainly such motions were visible, but contact between the two hind legs could not be verified. Examination of various females of Euglossa, Euplusia, and Eulaema, killed while collecting pollen, provided more in- formation. The front legs often had a little pollen on the inner surfaces of the basitarsi and less on the tibiae, more on the posterior margins of the inner basitar- sal surfaces than elsewhere. The middle legs had similarly distributed pollen, even more predominantly along the posterior margins. Presumably it is the hairs of the posterior parts of the inner surfaces that transfer much of the pollen. On the hind legs, there is little or no pollen on the in- ner surfaces of the tibiae or tarsi, except that when the pollen load is very large, a thin and broken layer may be present on the inner surfaces. Whenever the pollen load is of moder- ate or small size, pollen on the tibiotarsal articular region is limited to the area of the rastellum, and it is on the outer side of the row of bristles, not on the inner side. On the corbicula, any small pollen load is always immediately above the base of the basitarsus, as shown in Figure 16. It must be pushed up to this position by the auricle at the posterior base of the basitarsus. The auricle is present, but unlike that of Apis and Bombus, it is close against the tibial apex, which is so shaped that the base of the basitarsus rides over the convex tibial apex as the tarsus is moved relative to the tibia. When the pollen load is slightly larger, as in most specimens examined, the small fringe at the anterior base of the basitarsus seems to have played a role (Figs. 16 and 17). This fringe must push pollen among the hairs of the anterior corbicular fringe, for they are surrounded by pollen. Thus if the tarsus is flexed backward, the auricle pushes pollen basad, while if it is flexed forward, the anterior basal fringe does so. There is often some pollen on the outer side of the basitarsus, but this is discontinuous with that on the corbicula except when the pollen load is 594 THe University oF Kansas ScIENCE BULLETIN Fics. 16, 17. Structures of hind leg of female of Euglossa ignita. 16, Apex of tibia and base of basi- tarsus, showing by shading location of a small pollen load. 17, Bace of basitarsus of same, showing auricle at right and small fringe at upper left that may play a role in moving pollen. enormous. The movement of the auricle over the bulla or convexity of the tibial apex readily causes this discontinuity. All this supports the observation that the pollen is applied to the outer surface of the hind leg near the tibio-tarsal joint by the middle leg. Apparently the rastel- lum serves primarily as a fence to keep the pollen from “leaking” through onto the inner surface. In any event, the rastel- lum seems to be so positioned that it could comb the contralateral basitarsus only with difficulty. Finally, the huge pollen loads often attained, especially by Euplusia and Eulaema, are made possible partly by the enormous, posterior expansion of the tibia and could hardly be managed except with adjustment and shaping by the middle legs; some pollen is likely to be added di- rected to the pollen load in this way. EVOLUTIONARY CONSIDERATIONS EvoLuTiIon oF PoLLEN COLLECTING AND Transport. In the evolution of pollen- carrying behavior, three successive basic phases can be recognized. If Jander (1976) is correct in considering the crop as the ancestral pollen-carrying structure, the pol- len-collecting behavior of Hylaeus illus- trates the first phase, utilizing structures and behaviors already present in the Sphe- cidae. Hylaews carries pollen in the crop, collecting it by brushing it off the anthers with the forelegs and then scraping it off each foreleg with a comb on the maxillary galea (Jander, 1976). There is also the pos- sibility that Hylaeds eats pollen directly. Moreover, Hylaeus can scrape the head with the forelegs and then eat the pollen that was lodged on the head. There are no known Hylaeus movements for trans- ferring pollen from the thorax or abdomen to the mouth; such pollen is wasted or the small amounts that stick to the body sur- faces are brushed off in a cell, if our knowl- edge of the repertoire of movements is complete (Jander, in preparation). This prototypic pollen collecting of Hylaeus per- sists into the major evolutionary lines of the bees, even though most pollen trans- port in such bees is external. In the evolutionary line that led to the families Anthophoridae and Apidae, the original galeal comb of the primitive bees is replaced by a stipital comb (Schremmer, 1972; Jander, 1976), although in many taxa there is no maxillary comb. Within the Apidae a well developed stipital comb is present only in Bombus and only Bombus is known to have the prototypic pollen- collecting behavior of scraping pollen off of the forelegs with the mouthparts (Jan- der, 1976). Most other Apidae, including Apis, however, have weak stipital combs. The second phase of the evolution of pollen-collecting behavior led to eutypical pollen gathering and transport in a scopa on the hind legs as in the majority of the bees. In eutypical pollen gathering, pollen is transferred from the flower into a scopa Potten MANnriPUuLaTIon By Bees (APIDAE) 59 on the hind legs; as noted in the Introduc- tion, it is passed from the anthers to the scopa via the foreleg and the middle leg. Eutypical pollen manipulation is char- acteristic (with minor modifications) of at least Colletinae, Halictidae, Andrenidae, and Anthophoridae (including Xylocopi- nae), of course with the exception of para- sitic forms in such groups. Most bees of these groups have hairy bodies on which loose pollen lodges as they work in flowers. Since the middle legs can scrape the thorax as well as the front legs, these bees have movements whereby pollen can be trans- ferred from both the head (as in H ylaeus) and the thorax to the scopa. (In many Anthophoridae, scraping movements of the forelegs clean the dorsum of the thorax; Jander, 1976.) Eutypic pollen manipula- tion should be more efhcient than proto- typic. However, the eutypical movements still cannot transfer pollen from the abdo- men to the scopa. All bees scrape the abdo- men with the hind basitarsi to clean it, but pollen on the abdomen must be either lost or merely brushed off in a cell. Type I pollen packing as described above for Api- dae is a modification of eutypical pollen transport behavior. Eutypical pollen gathering has been supplemented by or transformed into a variety of different metatypic or derived pollen gathering and transport methods. For example, in the Megachilidae the scopa on the hind leg is replaced by one on the ventral side of the abdomen. The pertinent pollen-gathering movements are only par- tially known (Michener, 1953), but appear to differ from eutypical movements in that the hind legs transfer pollen to the scopa. As they can also brush the abdomen, mega- chilids should be able to transfer to the scopa for transport, pollen lodging on al- most any part of the body. Within the Apidae pollen transfer movements onto the middle leg are also not noticeably different from those of eutypical pollen gathering Yi (except in the genus Trigona). The de- rived pollen handling features are largely restricted to the interactions between mid- dle and hind legs and pollen packing from the distal end of the tibia. In other words, Type IL pollen packing is among the de- rived methods. The derived or metatypic movements found in most Apidae include scraping of the abdomen with the inner surfaces of the hind basitarsi and transfer of this pollen to the contralateral corbiculae for trans- port. Thus, pollen that lodges from flowers on the abdomen, as well as that on the head and thorax, can be used. Derived movements also include transfer of pollen from the middle legs to the inner surfaces of the hind basitarsi. They clearly include the pumping movements of the two hind legs in which the raste!llum scrapes pollen off of the contralateral hind basitarsus onto the basal surface of the auricle, and move- ments of the basitarsus which push pollen basad into the corbicula, which is thus loaded from its distal end. These matters are described in detail above in the section on Type II pollen packing. EvoLUTION OF CorBICULAE AND THEIR Firrinc Mecuanisms. The Apidae have a corbicula on the hind tibia and for all subfamilies of the Apidae it is known that at least sticky material can be transferred into the corbicula in the eutypical be- havioral sequence, described above in the sections on Type I Pollen Packing and Transport of Nesting Materials. Hence the corbicula and the eutypical behavioral sequence must have existed in a common ancestor of all Apidae. In the Introduc- tion we suggested the development of the corbicular scopa from the brushy tibial scopa of Anthophoridae. We have two theories, not mutually exclusive, to account for the origin of corbicular transport as contrasted to transport among the hairs of a brushy scopa. One is that a single stroke 596 Tue UNIVERSITY OF of the mid leg is sufficient to empty a cor- bicula, while several strokes must be needed to empty a brushy scopa. The former should be more efficient. The other theory is that the corbiculae arose in con- nection with the use of sticky materials for nest construction. It is true that some Megachilidae use resin, mud, and other sticky substances for making cells, but they do not transport such materials in the scopa. It is also true that in the Antho- phoridae there are forms such as some Centris which transport oil (Vogel, 1974) or mud (Michener and Lange, 1958) in the scopa. However, it seems that it would be nearly impossible to remove highly sticky materials like gums and resins from a brush of dense and usually branched scopal hairs (although Roubik and Mich- ener, in press, indicate that this happens in Epicharis). Winston and Michener (1977) therefore suggested that the smooth corbicular surface serves to facilitate re- moval of such material from the hind legs in the nest, after transport. Most Apidae use the corbicula for transport of both con- struction materials and pollen. As the corbiculae, according to this theory, are adaptations for the transport of sticky ma- terial, it is not surprising that these bees make the pollen into a sticky material too, by the addition of nectar as it is collected. Because a corbicula can be filled with sticky material in the eutypical fashion and because it is mechanically impossible for a brushy scopa to be loaded with pollen by the highly derived movements found in Apidae, the evolution of the corbicula with its corbicular scopa presumably preceded that of the highly derived behavior and associated structures (rastellum, penicil- lum) used in pollen packing. This sug- gestion is supported by the structure of certain African Meliponinae, as described below. The progenitors of the Meliponinae would not have had such specialized meli- ponine features as reduced wing venation Kansas SciENCE BULLETIN and stings, but might well have had the ancestral apid pollen-carrying apparatus. There exist in Africa today groups of meliponines in which the corbicula is fully developed, but in which the tibio-tarsal region lacks (primitively or by loss?) one or both of the special structures that relate to corbicular packing in other Meliponinae. Groups which lack the rastellum are T7i- gona subgenera Meliplebeia, Axestotrigona, and Hypotrigona as well as the genus Meliponula, Of these taxa the subgenus Hypotrigona has only a weakly developed penicillus which could not function as de- scribed in the section on Ipsilateral Type II Pollen Packing. In the other groups listed the penicillum bristles are not so nicely graded in length and curviture as illus- trated and described in that section; the penicillum in such cases may well serve only or primarily to support the pollen mass. These same groups have the largest (and flattened) sting sheaths of any Meli- poninae; this must be an ancestral feature and thus strengthens the hypothesis that at least in some cases the lack of rastellum and perhaps the weak penicillum are prim- itive features rather than losses. (These morphological data, but not the interpreta- tions, are from Dr. A. Wille, zm litt.) Pol- len packing in these African groups has not been studied, but it seems almost cer- tain that in Hypotrigona, at least, it is like the packing of nest materials and Type I pollen packing described above, ie., eu- typical; the middle legs presumably place sticky pollen directly into the corbiculae. The American subgenus Trigonisca, a relative of the African Hypotrigona, has relatively feeble penicilla and rastella, and its pollen packing, so far as known, is like that postulated for the ancestral groups, although it may also have in its repertoire the derived pollen-packing methods of most Meliponinae. Potten PackiNG IN THE Aprpaz. Eutypical pollen manipulation as seen in Apidae in- PotteN MANIPULATION By Bees (ApIDAE) 597 volves synchronous movements of the legs of a given pair. Thus the two middle legs even in a Trigonisca resting on a petal simultaneously place pollen on the ipsi- lateral corbiculae. In this respect apids differ from at least many of the non-apid bees, which transfer pollen back to the tibial scopa by movement of one leg at a time. (Observations of non-apids in flight are still needed; their leg movements may then be synchronous.) Movement of both middle and hind legs appears to facilitate pollen transfer to the corbicula, making it easier to get the mid basitarsus to the proper position relative to the hind leg. Simultaneous movement of both mid and hind legs is difficult, however, while the bee uses at least one of these pairs of legs for support. Most apids solve this problem by hovering when employing derived (Type I) pollen-packing methods, so that the legs are freed for the pollen-transfer- ring movements. Such behavior has seem- ingly been transmitted to males of the Euglossini which use similar movements for transferring scents to the hind tibiae while hovering. Presumably a behavioral pattern that evolved among females was activated in males, which in any case must carry the appropriate genes for it. Within the Meliponinae, several de- rived features of pollen manipulation have arisen. Like most Apoidea, meliponines commonly collect pollen with the front tarsi, as well as on other parts of the body when the pollen is loose. Many and per- haps all species of Trigona are unusual among bees in that they transfer the pollen from the forelegs onto the thoracic venter and leg bases for temporary storage. It is later picked up by the middle legs (or possibly again by the front legs). We sus- pect that Melipona has lost this behavior for we find no special accumulations of pollen on the mesepisterna and coxae in this genus, nor are there coarse hairs in these areas like those of Trigona. Melipona workers move so rapidly that details of their pollen manipulation have eluded us. (Our attempts at observation were mostly made at flowers of Mimosa, where the Melipona rushes around through the sta- mens, all parts of its hairy body being dusted with pollen.) In the species of Trigona that we have studied most thoroughly, pollen is trans- ferred to the corbicula by drawing the pollen-bearing middle tarsus across the apex of the ipsilateral hind tibia and base of the basitarsus in such a way that the pollen appears to be picked up and pushed basad onto the corbicula by a structure found only in this subfamily, the penicil- lum. This behavior resembles eutypical movements in that pollen is transferred to the hind leg by the ipsilateral middle leg, but is derived in that it is not placed di- rectly into the corbicula. In the course of this movement pollen will also be scraped off of the mid basitarsus by hairs of the auricular area of the ipsilateral hind basi- tarsus. Pollen also appears to be transferred from inner sides of hind basitarsi to contra- lateral corbiculae in meliponines. They scrape pollen from the abdomen with the hind basitarsi. Meliponines with a penicil- lum ordinarily also have a row of bristles across the inner side of the apex of the tibia, the rastellum, which can comb the inner side of the contralateral basitarsus. When large amounts of pollen are combed off of a basitarsus by a rastellum, the pollen accumulates outside the rastellum and can be pushed basad, onto the corbicula, by basitarsal movements which are effective because of the direction of the hairs of the auricular area and perhaps because of the pollen already on those hairs. However, for most meliponines it is probably more important that such pollen can presumably stick to and be carried basad by other pol- len transferred to the hind leg by the ipsilateral middle leg. 598 Tue University oF Kansas SciENcE BULLETIN For many of the commonest species of Trigona, the importance of the inner side of the hind basitarsus in transferring pollen seems reduced. In the subgenus Trigona s. str. which includes the species that we have studied most carefully and in many species of the subgenus Tetragona, there is a large, hairless, sericeous area occupying up to half of the inner basitarsal surface. Accumulations of pollen on this surface have not been seen and are not common even among the hairs on the rest of the surface. These groups are placed among the derived subgenera of Trigona. Their rather hairless abdomens and perhaps the nature of the flowers usually visited by them may make brushing of the abdomen of little importance. Probably contralateral pollen transfer is more important in other subgenera in which the under surface of the hind basitarsus is fully covered with bristles of stiff hairs. Finally for Meliponinae, the gleaning bees of the subgenus Scaura represent a noteworthy development, with a concave auricular area or false auricle in the swol- len hind basitarsus for pushing pollen up- ward onto the corbicula. The broad basi- tarsi as well as the abdominal hairs are used to brush up pollen from flower and leat: -surtaces. In the remaining Apidae (Apinae and the tribes Bombini and Euglossini of the Bombinae) pollen manipulation is also of derived types, but partially different from those of Meliponinae. In the Apinae and Bombini, whose hind tibial and_ tarsal structure is very similar, pollen from the inner surface of the hind basitarsus is combed off by the contralateral rastellum; it sticks to the tibial apex outside the rastel- lum, and is then pushed upward onto the corbicula by the strongly developed auricle. In the Euglossini there is also an auricle although it appears small because of the enormously expanded hind tibia. The auricle is closely appressed to the swelling or bulla of the tibial apex, so that pollen from the contralateral basitarsus presum- ably could not be pushed between these leg segments onto the auricle, as in Apis and Bombus. Moreover, the rastellum does not project mesally in such a way that it could easily comb the contralateral basi- tarsal hairs, as it does in the other three apid groups. Pollen must therefore be placed on the outside of the hind leg by the middle leg, presumably in the area of the tibiotarsal joint; then it is probably pushed basad into the huge corbicula by the auricle and other basal marginal parts of the basitarsus which slide over the tibial bulla as the basitarsus moves. If Winston and Michener (1977) have correctly presented the cladistic relation- ships among apid groups, the Euglos- sinae must have lost the contralateral pol- len transfer and the relatively large auricle characteristic of Bombus and Apis. The great enlargement of the tibia associated with carrying huge loads of nesting ma- terials and pollen may have had this effect by narrowing the space between the auricle and the swollen tibial apex. Alternatively, the Euglossinae might be, in features of pollen manipulation (as well as in solitary or parasocial behavior), primitive Apidae that, like certain African Meliponinae, have never evolved contralateral pollen transfer. Many more observations of pol- len-collecting Euglossinae are needed to settle this problem. DerivaTION OF PoLLEN MAanIPuLATION FROM SELF-GroomiInG Movements. Pollen transfer from the foreleg to the middle leg is indistinguishable from the correspond- ing cleaning movement (Jander, 1976) and is therefore considered homologous to it. The middle leg, with the pollen on its under and inner side, swings backward in eutypical pollen transfer towards the ipsi- lateral hind leg. The pollen on the middle leg is then pressed into the scopa, and simultaneously the middle leg is pulled PoLtLtEN MANIPULATION BY Bees (APIDAE) 599 forward and basad relative to the hind tibia, so that the distad-directed hairs of the scopa scrape the pollen off the middle leg. This final scopal loading movement is precisely the opposite of, and presumably derived from, normal self-cleaning be- havior, during which bees typically remove dust from the outer side of the hind leg with a distad scraping movement of the ipsilateral middle leg. This cleaning move- ment, however, is used inside the nest for scraping the collected pollen out of the scopa. (Our remarks above ignore the femoral, trochanteral, coxal, propodeal, and sternal scopal areas of many colletids, halic- tids, and andrenids, partly for lack of data, but also because in the context of the Api- dae it is the tibial scopa that is important.) Patting of the corbicular pollen load with the middle basitarsus, presumably to adjust and shape the pollen mass, is seen in most or all apids. This movement dif- fers from normal cleaning of the outside of the hind leg and probably from Type I pollen packing movements mainly by sup- pression of scraping and rubbing compo- nents. It is likely that patting is derived from eutypical or Type I pollen-packing movements, which are themselves prob- ably derived from normal cleaning move- ments. The movements of the mid legs in ipsilateral Type II pollen packing presum- ably have the same origin. The derived pollen-manipulating move- ments of the Apidae also mostly appear to have evolved from cleaning movements. Bees typically clean one middle leg at a time by scraping off foreign particles usu- ally with both but sometimes with only one hind basitarsus. The middle leg is pulled forward, usually between the two hind basitarsi, while the hind legs are pushed backward and downward (Beeken, 1934; Farish, 1972). It is by this very move- ment that the pollen is transferred from the middle leg to the inner side of the hind basitarsus by Bombini and Apinae and at least some Meliponinae. After a bee has cleaned its middle legs between the hind basitarsi, it regularly cleans the latter by pressing them flat against each other and then rubbing them with alternating longitudinal pumping movements. During these movements the tarsi are in continuous contact. Dirt is thereby pushed distad because all bristles and hairs on the inner sides of the hind basitarsi point in that direction. [This cleaning movement was observed in all of the 60 species of bees in seven families listed by Jander (1976) and differs from the corresponding cleaning movement of most other Hymenoptera which clean the hind basitarsus by pulling it past the tibio- tarsal joint of the contralateral hind leg (Farish, 1972; Wagner, 1959). Normally terminal tibial spurs improve the efficiency of this movement and the so-called strigil of the hind leg of sphecids is its morpho- logical concomitant.| In the pollen manip- ulation of Meliponinae, Bombini and Api- nae an almost typical apoid movement pattern is used when pollen is combed off of the inner surface of the hind basitarsus by the contralateral tibial comb or rastel- lum. The only known difference between this pollen manipulation and the true cleaning movement is that during the latter the tarsi are pressed against each other while during the former the tarsi are slightly bent apart so that only the distad- moving rastellum contacts the contralateral hind basitarsus. For the final auricular movement that pushes pollen basad from the rastellum into the corbicula, no homol- ogous self-cleaning movement is known. The peculiar temporary storage of pol- len on the thoracic venter and coxae in the genus Trigona appears to be the reverse of a widespread cleaning movement in which the forelegs scrape backward on the underside of the thorax, as has been ob- served in Nomada vincta, Ceratina (dupla or calcarata), Bombus ruderatus and Pst- thyrus variabilis (R.J.). Further transfer of pollen could easily be mediated by the 600 Tue Universiry oF Kansas ScrENCE BULLETIN middle legs, since backward scraping of the coxae by the middle legs as a com- ponent of self cleaning has been seen in Triepeolus concavus, Anthophora abrupta, and Trigona jaty (R.J.) and has been de- scribed for Apis mellifera by Beeken (1934). ACKNOWLEDGEMENTS This paper was made possible by NSF grant DEB 73-06815. Field work in French Guiana by M.L.W. and C.D.M. was possible thanks to facilities maintained by Gard Otis, David and Gretchen Roubik, and Penelope F. Kukuk for work on the Africanized honeybee under the direction of O. R. Taylor (United States Department of Agriculture, contract 12-14- 7001-363). Field work by C.D.M. in Colombia was possible thanks to facilities and vehicle provided by the Smithsonian Tropical Research Institute and thanks to the courtesy of its representatives in Cali, Colombia, Dr. Reinaldo Diaz and Dr. M. J. West Eberhard. Dr. M. D. Breed photographed pollen- collecting bees in Colombia, using equipment kindly lent by Dr. West Eberhard. The work of Dr. Breed and C.D.M. in Colombia and of C.D.M. in French Guiana was facilitated by funding through the Uni- versity of Kansas Endowment Association. R.J.’s work in Costa Rica was made possible by the Or- ganization for Tropical Studies, and his work in Malaysia by grant BMS 74-1398 from the National Science Foundation. Identifications of some of the Euglossinae were by Dr. R. L. Dressler of the Smithsonian Tropical Research Institute, Balboa, Canal Zone. Anatomical drawings were made by Mr. Paul G. Decelles. 1 PERACLURES Gl PED) BasstnDALE, R. 1955. The biology of the stingless bee Trigona (Hypotrigona) gribodoi Magretti. Proc. Zool. Soc. London, 125:49-62. Batra, S. W. T. 1966. The life cycle and behavior of the primitively social bee, Lasioglossum zephyrum (Halictidae). Univ. Kansas Sci. Bull., 46:359-423. Beeken, W. 1934. Uber die Putz- und Sau- berungshandlungen der Honigbiene (Apis mellifica). Arch. fir Bienen- kunde, 15:213-275. Bevinc, J. 1931. Beobachtungen tber das Pollensammeln der Honigbiene (Apis mellifica). Arch. fur Bienenkunde, 122392-359. Brave, A. 1913. Die Pollensammelapparate der beinsammelnden Bienen. Jenaische Zeitschr. f. Naturwissenschaften, 50: 1-96, pls. 1-4. Butrer-Rerepen, H. von. 1915. Leben und Wesen der Bienen. Braunschweig: Vieweg. Dopson, C. H., R. L. Dresser, H. G. Hurts, R. M. Apams, ano N. H. Witiams. 1969. Biologically active compounds in orchid fragrances. Science, 164:1243- 1249" Evoy, W. H. anp B. P. Jones. 1971. Motor patterns of the male euglossine bees evoked by floral fragrances. Anim. Behav., 19:583-588. Faris, D. J. 1972. The evolutionary impli- cations of qualitative variation in the grooming behavior of the Hymenop- tera. Anim. Behav., 20:662-676. Horrer, E. 1882. Die Hummeln Steier- marks, I. Halfte. 31. Jahresbericht der Landes-Oberrealschule in Graz. 92 pp., piss "Ay. Iovand: 2% IBRA. 1976. Bibliography No. 16. Anno- tated bibliography on propolis. Inter- national Bee Research Association. Janper, R. 1976. Grooming and _ pollen manipulation in bees (Apoidea): the nature and evolution of movements involving the foreleg. Physiol. Entom., Vet 79- 19%. Laroca, S. 1970. Contribuigéo para o con- hecimento das relacdes entre abelhas e flores: coleta de pdlen das anteras tubulares de certas Melastomataceae. Rev. Floresta [Curittba, Brazil], 2: 69-74. Laroca, S. AND S. Laver. 1973. Adaptacao comportamental de Scaura latitarsis para coleta de pdlen. Acta Biol. Parana, 2:147-152. Lecce, A. P. anp B. Bore. 1975. Variations in the single hair of the honeybee corbicula; Jour: -Apicult. “Res., 14: 75-78. Mawr, F. 1934. Die Lebensgewohnheiten und Instinkte der staatenbildenden Insekten. Wien: F. Wagner. Maurizio, A. 1968. La recolte et l’emma- gasinage du pollen par les abeilles, pp. 168-173, In: Traité de Biologie de PoLLEN MANIPULATION lV Abeille, R. Chauvin Ed., Tome III. Paris: Masson. Meyer, W. 1953. Beobachtungen und Ver- suche uber den Transport und die Verwendung von dargebotenem Wachs der Honigbiene. Zeitschr. Bienen- forsch., 2:147-157. ———. 1954. Die “Kittharzbienen” und ihre Tatigkeiten. Zeitschr. Bienen- forsch., 2:185-200. 1956. “Propolis bees” and their ac- tivities. Bee World, 37:25-36. Micuener, C. D. 1944. Comparative exter- nal morphology, phylogeny, and a classification of the bees (Hymenop- tera). Bull. Amer. Mus. Nat. Hist., 82:258-326. 1953. The biology of a leafcutter bee (Megachile brevis) and its asso- jates. Univ. Kansas Sci. Bull., 35: 1659-1784. 1962. An interesting method of pol- len collecting by bees from flowers with tubular anthers. Rev. Biol. Trop- ical, 10:167-175. 1974. The social behavior of the bees. Cambridge: Harvard Univ. Press. anp R. B. Lance. 1958. Observations on the ethology of neotropical antho- phorine bees. Univ. Kansas Sci. Bull., 39:69-96. ano A. Witte. 1961. The bionomics of a primitively social bee, Lasioglos- sum inconspicuum. Univ. Kansas Sci. Bull., 42:1123-1202. Parker, R. L. 1926. The collection and utilization of pollen by the honeybee. Cornell Univ. Agr. Expt. Sta. Mem., 98:1-55. Ripsanps, C. R. 1953. The behavior and social life of honeybees. London: Bee Research Assn. Ltd. Roserts, R. B. 1969. Biology of the genus Agapostemon (Hymenoptera: Halciti- dae). Univ. Kansas Sci. Bull., 48: 689-719. Rosco, G. A. 1927. Beobachtungen an Kittharz sammelnden Bienen. Biol. 7A pees iO lord WA Rovsix, D. W. ann C. D. MicHener. (in press.) The seasonal cycle and nests of Epicharis zonata, a bee whose cells BY Bees (APIDAE) 601 are below the wet season water table. s1otropica. SAKAGAMI, S. F. AND J. M. F. pE CAMARGO. Cerumen collection accompanied by thieving and attacking in a stingless bee Nannotrigona (Scaptorigona) (sic) postica (Latreille), with a considera- tion of territoriality in social insects. Rev. Biol. Tropical, 12:197-207. ScHREMER, F.° 1972. der Nektarraub, das Pollensammeln Der Stechsaugrissel, und der Bliitenbesuch der Holzbienen (Xylocopa). Zeitschr. Morph. Tiere, 72:263-294. Scuwarz, H. F. 1948. Stingless bees (Meli- ponidae) of the western hemisphere. Bull. Amer. Mus. Nat. Hist., 90:i-xvit a 1-40. SLapEN, F. W. L. 1911. How pollen is col- lected by the social bees, and the part played in the process by the auricle. British Bee Jour., 39:491-493. ——., 1912a. The humble-bee. London: McMillan & Co. 1912b. How propolis is collected. Some further notes on pollen-collecting. Gleanings in Bee Culture, 40:335-336. Snopcrass, R. E. 1956. Anatomy of the honey bee. Ithaca: Cornell Univ. Press. Voce, S. 1963. Das sexuelle Anlockungs- prinzip der Catasetinen- und Stan- hopeen-Bliiten und die wahre Funk- tion ihres sogenannten Futtergewebes. Osterreich. Bot. Zeitschr., 110:308-337. 1966. Parfiimsammelnde Bienen als Bestiuber von Orchidaceen und Glox- inia. Osterreich. Bot. Zeitschr., 113: 302-361. 1974. Olblumen und 6lsammelnde Bienen. Tropische Subtrop. Pflanzen- welt [Mainz], 7:1-267. Wacner, M. 1959. Die Putzhandlungen der Hymenopteren. Dissertation: Univer- sity of Munich. Witte, A. 1963. Behavioral adaptations of bees for pollen collecting from Cassia flowers. Rev. Biol. Tropical, 11:205- 210. Winston, M. L. anp C. D. Micuener. 1977. Dual origin of highly social behavior among bees. Proc. Nat. Acad. Sct., ULS-A.,, /4:1135-1137- NV ie gp eee av lewis pale aa (70 00 o> (he eon re | Ree taal 1 a a a a i gan ei © ! i | SC 5 S SU S er : By W. e org Ge ‘ 3 ss BS Se Be ‘et poe * Se By 2 oe *. os pod * . poe 2 oe . Se s, Sq *. . See ee se s oe * Oe se * . cee ss se. * os be) J *s Se se Se se so s -. sa by xe: roe s .. s os BOS eee « os * ‘s Se. See = = ° Sa * -. = = = = 'e ee See = Re * 3 c= ‘ ‘e 79 19 9, h Te a M 13 603-6 p: p 20, Oo. N ji 5 1. Vo ANNOUNCEMENT The University of Kansas Science Bulletin (continuation of the Kansas Uni- versity Quarterly) is an outlet for scholarly scientific investigations carried out at the University of Kansas or by University faculty and students. Since its incep- tion, volumes of the Bulletin have been variously issued as single bound volumes, as two or three multi-paper parts or as series of individual papers. Issuance is at irregular intervals, with each volume prior to volume 50 approximately 1000 pages in length. Effective with volume 50, page size has been enlarged, reducing the length of each volume to about 750 pages. The supply of all volumes of the Kansas University Quarterly is now ex- hausted. However, most volumes of the University of Kansas Science Bulletin are still available and are offered, in exchange for similar publications, to learned societies, colleges and universities and other institutions, or may be purchased at $20.00 per volume. Where some of these volumes were issued in parts, individual parts are priced at the rate of 2 cents per page. Current policy, initiated with volume 46, is to issue individual papers as published. Such separata may be purchased individually at the rate of 3 cents per page, with a minimum charge of $1.00 per separate. Subscriptions for forthcoming volumes may be entered at the rate of $20.00 per volume. All communications regarding exchanges, sales and subscriptions should be addressed to the ExcHaNce LrprariAn, UNIVERSITY oF Kansas Lipraries, LAWRENCE, Kansas 66045. Reprints of individual papers for personal use by investigators are available gratis for most recent and many older issues of the Bulletin. Such requests should be directed to the author. The International Standard Serial Number of this publication is US ISSN 0022-8850. Editor Eugene C. Bovee Editorial Board William L. Bloom Philip W. Hedrick Rudolf Jander Harvey Lillywhite Charles D. Michener Norman A. Slade Henry D. Stone George W. Byers, Chairman THE UNIVERSITY OF KANSAS SCIENCE BULLETIN Vol. 51, No. 20, pp. 603-613 March 9, 1979 Summer Crane Flies of Lake Itasca Vicinity, Minnesota’ GerorcEe W. Byers TABLE OF CONTENTS UA SRGUTRUNCOTE a 5 ee ee ones, Rp isk A 603 “SOTERGIDICS TROIS cess) ent D5 SO tank a eee Ris Soe es 604 POINTED MIB IG TOR OSPECIES? fase hie! alt cee decks cuaivda de sabeveksacicettc dacs eaboce ne cht Seiee on eee eee 605 MINT mTetiy stream remem ee ce ee en Ee Ng Ek aed, cee et SO ORE aad Sita 605 | Dives GTS Yee HTB ETT 10) eee REE Reo eee Coote RTT eink SSN 6 611 MEE RAGHU Re OTE eb Te te ae oceans badaud 20 Moog ze Preach eS ans war ; oo). +a > @anw ae.) aren : = | @& 0 @efaateee sn) wo 4 os Gan a al ‘Tater © ; a (Wa Saree ewe 7 P : - > 7 - _ = 7 SS me Se bo tee ite RR NR se aeses i aptaDb EM eeCeeM te ste ocean eeeetetatatatatatatatatatatatststensistensseteteteneneneceneneneseoececeseceoen RR KR RRR RS POPE oes tatetetee, aS sceceercreee asatatetatetetatonesseensarentecncenatenamehatetahe ered sseteteteres = BS = eo’ a" ee" e's 00 0 6 0 6 6 6's 8 66 6 6 6 6 6 8 6 6 6 6 6 08 66 6 6 6 6 6 6 6 6 oo 8 6 ow oe 8 4 ee ee eee SCIENCE BULLETIN PROTOZOA FROM ACID-BOG MOSSES AND FOREST MOSSES OF THE LAKE ITASCA REGION (MINNESOTA, USA) By Eugene C. Bovee Vol. 51, No. 21, pp. 615-629 March 20, 1979 ANNOUNCEMENT The University of Kansas Science Bulletin (continuation of the Kansas Uni- versity Quarterly) is an outlet for scholarly scientific investigations carried out at the University of Kansas or by University faculty and students. Since its incep- tion, volumes of the Bulletin have been variously issued as single bound volumes, as two or three multi-paper parts or as series of individual papers. Issuance is at irregular intervals, with each volume prior to volume 50 approximately 1000 pages in length. Effective with volume 50, page size has been enlarged, reducing the length of each volume to about 750 pages. The supply of all volumes of the Kansas University Quarterly is now ex- hausted. However, most volumes of the University of Kansas Science Bulletin are still available and are offered, in exchange for similar publications, to learned societies, colleges and universities and other institutions, or may be purchased at $20.00 per volume. Where some of these volumes were issued in parts, individual parts are priced at the rate of 2 cents per page. Current policy, initiated with volume 46, is to issue individual papers as published. Such separata may be purchased individually at the rate of 3 cents per page, with a minimum charge of $1.00 per separate. Subscriptions for forthcoming volumes may be entered at the rate of $20.00 per volume. All communications regarding exchanges, sales and subscriptions should be addressed to the ExcHaNcEe LiprariaN, UNIVERSITY oF Kansas Lipraries, LAwrENCE, Kansas 66045. Reprints of individual papers for personal use by investigators are available gratis for most recent and many older issues of the Bulletin. Such requests should be directed to the author. The International Standard Serial Number of this publication is US ISSN 0022-8850. Editor Eugene C. Bovee Editorial Board William L. Bloom Philip W. Hedrick Rudolf Jander Harvey Lillywhite Charles D. Michener Norman A. Slade Henry D. Stone George W. Byers, Chairman THE UNIVERSITY OF KANSAS SCIENCE BULLETIN Vol. 51, No. 21, pp. 615-629 March 20, 1979 Protozoa from Acid-bog Mosses and Forest Mosses of the Lake Itasca Region (Minnesota, USA) EucENE C. BoveE’ TABLE OF CONTENTS PNB SRA CIM ees eee ee ees Ae , 615 INTRODUCTION 615 MATERIALS AND 616 IRGzs UTS Meee oe eee ene ee 617 DIscussION 617 AR EM PROTOZOA ASSOCIATED WITH (BOG! MOSSES, 262.2200 -2essec-sexc occu csecac-cecasceaccddened nndennacesecaceseeeeceeeeereneese eseamaes 618 ABATE alll PROTOZOAUPASSOCIATED. WITH? FOREST MOSSES 2222.22 2-25 -cecceeses-oenenncecestacennesasesaueresnesuezesenccstansonesseueucas 620 ENGTRINTON LUTE DYGID USING ST EO Dee ee SDE 2 NEO Ot Ee ae 2 eee er oe eee ABR ACTURE@e Ocoee 2 ER ae whe a Ok ee od ap eS Se a oe ee ee ee cee eeee Pirates II] —CitiopHora AssocIATED WITH MossEs ROME Ri OSEaeSESTACID a: ANMEBAS (OF MOSSES) 2<. 2... c.cc0002esoce-2cccce-ctecccndeancnce-odeceacronveacnceeece Pirates I1V,V—Losose Trestacip AMEBAS OF MossEs ABSTRACT In a comparison of protozoa associated with mosses from an acid sphagnum bog of the Lake Itasca (Min- nesota, USA) region with those associated with forest mosses there, these were found: Among bog mosses 145 protozoa were identified to genus, 96 of those to species; 33 were Mastigophora from 2 subphyla, 6 classes, 7 orders, 10 families and 21 genera; 56 were Sarcodina, from 2 subphyla, 3 classes, + orders, 13 families and 23 genera; 56 were Ciliophora, from 3 subphyla, 8 classes, 16 orders, 29 families and 37 genera. Among forest mosses 68 protozoa were identified to genus, 46 of those to species; 9 were Mastigophora from 2 subphyla, 3 classes, 3 orders, 4 families and 7 genera; 39 were Sarcodina from 2 subphyla, 3 classes, 7 orders, 12 families and 23 genera; 30 were Ciliophora from 3 subphyla, 7 classes, 12 orders, 21 families and 24 genera. 23 pro- tozoa characteristically restricted to bog mosses were found in the bog, 8 being testate sarcodines, 15 being ciliates. 33 protozoa characteristically associated with forest soils or forest mosses were found with the forest mosses, 17 being testate sarcodines, 16 being ciliates. Two different associations exist in the two habitats. Only one sarcodine was characteristically found in both habitats. The ciliates of the two associations, though often of the same or similar genera, were more often different species, except for certain cosmopolitan ones. INTRODUCTION Lake Itasca, a large spring-fed lake in Clearwater County, Minnesota, USA, is the source of the Mississippi River. It is sur- rounded by a Pleistocene-glaciated terrain, supporting a mixed hardwood and coniferous forest of red, jack and white pine, ash, maple, oak and other hardwoods, aspen and birch. Other lakes and ponds, with interconnecting meandering streams, are numerous in the region, with intervening wooded swamps, acidic bogs and marshy meadows. Black spruce and tamarack occur near the bogs, 1 Reprint requests should be addressed to the author: Department of Physiology and Cell Biology, University of Kansas, Law- rence, Kansas, USA, 66045. 616 Tue Universiry oF Kansas ScrENcE BULLETIN and willow and alder along the streams. Open marshes adjoin many of the smaller lakes, supporting abundant growth of sedges and grasses. Forest undergrowth includes many herbaceous plants, mosses, ferns and lichens. The region, with its varied habitats and many microenvironments, presents much op- portunity for studying ecological interrela- tionships of the numerous organisms that occur in them. One group of organisms, the Protozoa, have largely been ignored in the ecology of the region. Even though rainfall is not heavy (~57cm/annum), water is abundant in the glacially tilled terrain, the air is humid and the habitats moist. Hence, it is an excellent area for studies of protozoan ecology. This paper presents the results of a short, intensive study of protozoa found associated with mosses of bogs compared to those found with forest mosses in the region. Many ob- servers have reported protozoa among mosses, especially bog mosses and forest mosses, e.g., Penard (1902, 1905, 1909, 1922), Hoogenraad and DeGroot (1952), Heal (1961), Stépanek (1963) and others. Some protozoa, especially certain ciliates deemed characteristically asso- ciated with mosses, have been so named as to reflect that association, e.g., Balantidiopsis muscicola, Condylostoma sphagni, Glaucoma sphagni, Phacodinium muscorum, Spirosto- mum sphagni, Vorticella muralis, and others (Penard, 1922). Other researchers, working with dried mosses from herbaria, have noted that differ- ent species of mosses may harbor different spectra of protozoa, especially of testate amebas. (Hoogenraad and DeGroot, 1952; Stépdnek, 1963). Heal (1961) found that associations of testate amebas in bog mosses differed from those of fen mosses in England, the species of Sphagnum mosses also diffex- ing in bogs versus fens. Bonnet and Thomas (1960) found that associations of testate ame- bas in soils or on mosses differ with the nature of the soil and species of mosses in the eastern Pyrenees Mountains. Stout (1963) found that protozoan associations of ciliates, naked and testate amebas in forest soils and litters differ in samples from mor, calcareous mull, or acid mull substrates from England’s Chiltern Hills. All of these observers found some proto- zoan species to be cosmopolitan, but others restricted more-or-less closely to specific mi- crohabitats related to one or more of several ecological parameters. Those parameters in- clude the kind(s) of mosses, the type(s) of soil, the pH (acid or alkaline) of the soil, the type and quality of humus, the types of plants on the soil, the specific soil horizon, and the relative moistness and drainage of the soil. This study of the protozoa associated with bog mosses (Sphagnum spp.) compared to those of forest mosses of the Lake Itasca re- gion of Minnesota indicates differences worth noting and worth further study. A listing of the protozoa found is given. MATERIALS AND METHODS Two samples of Sphagnum spp. and water around them were collected during early August, 1977, from the Buell Bog= immediately adjoining (on the south) Highway 113 (Minnesota), 1% mi W of the SE entry to the Lake Itasca State Park, latitude N 47° 7’ 45" and longitude W 95° 11’, altitude ~1600 ft. These samples were kept in moist (glass) chambers in a laboratory at 18°-22°C. Another sample of mosses was taken from a rotten, fallen log near the “Ice-House Pond” on the grounds of the Biological Station in the Park, latitude N 47° 13’ 13” and latitude W 95° 11’, altitude ~1590 ft. It contained mainly the mosses Brachythecitum salesdrosum, Mnium cuspidatum and Pylaisiella selwynii.38 These were likewise kept in a moist chamber. About 0.2 ml of water and detritus squeezed from a moss was placed on a clean, glass microscope-slide, then covered with a #1 coverglass sealed in place peripherally with petroleum jelly. The protozoa on the slide were observed at 40, 100, and 430X mag- nifications by phase-contrast lighting. Such a slide could be observed during 48 hrs or more to notice live protozoa thereon and identify them. Two to 8 such slides were prepared each day and were examined, a total of 70 slides being ob- served to identify protozoa present, 38 from the bog-moss samples and 32 from the forest moss sample. A list of protozoa seen in the samples was re- corded, keying each one to genus and, when possi- 2 Named after the late Dr. Murray Buell, algologist, who taught algology many summer sessions at the Lake Itasca Biology Station, often using this bog as a field laboratory and teaching site. 3 Dr. Frank Bowers, Biology Department, Wisconsin State University, Stephens Point, Wisconsin, kindly identified these mosses for me. Protozoa oF LAKE ITasca ArEA MosseEs 617 ble, to species, using a variety of monographic books and papers (Allegre and Jahn, 1943; Corliss, 1977; Jahn and Jahn, 1949; Johnson, 1944; Kahl, 1935; Kudo, 1966; Penard, 1902, 1905, 1909, 1922; Shaw- han and Jahn, 1947; and others). Particular atten- tion was given to the ciliates and testate sarcodines, since they are the Protozoa more often recorded pre- viously as being characteristic of mossy habitats. The Protozoa are classified herein according to the scheme of Jahn, Bovee and Jahn (1979). RESULTS In the Sphagnum spp. from the Buell Bog, 145 species of Protozoa were identified to genus, 97 of those to species. Others seen too briefly to be identified are not included. Of the identified organisms 33 were of the Phylum Mastigophora (29 Phytomastigo- phora, 4 Zoomastigophora) of 2 subphyla, 6 classes, 7 orders, 10 families and 21 genera. Those of the Phylum Sarcodina totalled 56 species identified to genus, 42 of those to species, from 2 subphyla, 3 classes, 4 orders, 13 families and 23 genera. The Phylum Ciliophora was represented by 56 kinds identified to genus, 37 of them to species, from 3 subphyla, 8 classes, 16 orders, 29 families and 37 genera. No protozoa of the Phylum Sporozoa were found among the Sphagnum mosses of the Buell Bog samples. In the forest mosses there were 68 species of Protozoa identified to the genus, 52 of them to species. Of the Subphylum Mastigo- phora 2 subphyla, 3 classes, 3 orders, 4 fam- ilies, 7 genera and 9 species were represented. The Subphylum Sarcodina had 2 subphyla, 3 classes, 7 orders, 12 families, 23 genera and 39 species represented. The Subphylum Cili- ophora was represented by 3 subphyla, 7 classes, 12 orders, 21 families, 24 genera and 30 species. No members of the Phylum Sporozoa were found in the forest moss samples. In the Buell Bog mosses, as well as the many more or less cosmopolitan protozoa seen, there were a number of protozoa listed by other authors as found mainly on or re- stricted to mosses, especially on Sphagnum spp. These were: 1. Sarcodines (Filida); Assulina seminulum, Euglypha cristata, Tri- nema penardi. Sarcodines (Lobeda); Arcella artocrea, Nebela collaris, Heleopora picta, Hyalosphenia papilio, Hyalosphenia elegans. 2. Ciliates; Drepanomonas dentata; Litonotus muscororum; Bryophyllum sphagni; Spathi- dium amphoriforme; Prorodon cinereus; Lep- topharynx sphagnetorum; Holophyra sagi- nata; Colpoda steini; Frontonia depressa; Cylidium elongatum; Cyclidium muscicola; Vorticella lichenicola; Keronopsis helluo; Euplotes muscicola; Blepharisma sphagni. In the forest mosses, similarly, there were a number of Protozoa identified by others as primarily associated with forest mosses, for- est soils and the rhizospheres of forest plants. Those include: 1. Sarcodines (Filida); Eu- glypha ciliata; E. ciliata form glabra; E. um- bilicata; E. laevis; Trinema penard1; Cory- thion dubium; C. pulchellum; Assulina muscororum; Difflugiella oviformis; Crypto- difflugia compressa. Sarcodines (Lobeda); Centropyxis sylvatica, C. s. form minor; C. platystoma; C. elongata; Cyclopyxis puteus; C. kahli; C. ambigua. 2. Ciliates; Spathidium muscicola; S. spatula; S. amphoriforme; Col- poda cucullus, C. inflata, C. irregularis, C. penardi; Tillina magna; Nassula_ protectis- stma; Muicrothorax simulans; Pseudoglau- coma muscorum; Cyclidium muscicola; C. terricola; Gonostomum affine; Gastrostyla muscororum; Rhabdostyla muscorum. It is noteworthy that only one species characteristic of the sphagnum bog was also found in the forest mosses and is considered characteristic of both, namely, Assulina seminulum. Bonnet and Thomas (1960) also found it to be cosmopolitan in their soil and moss samples. Table I shows the Protozoa that were found in the Buell Bog. Table II shows the Protozoa found in the forest mosses. DISCUSSION In comparing the protozoan populations of the bog mosses to those of forest mosses, it is clear, as might be expected, that the more edaphic species occur with the forest mosses. The species associated with bog mosses tend to be aquatic and acidophilous, or cosmo- politan species. The edaphic nature of the forest moss associates is indicated, among ciliates, by (usually) the presence of smaller species with reduced surface area. Plates I 618 THe UNiversiry oF Kansas SCIENCE BULLETIN TABLES Protozoa AssociATED wiTH Sphagnum Mosses or Burt. Boe. Phylum MasTIGOPHORA Subphylum PHyTOMASTIGOPHORA Class CRYPTOMONADEA Order Cryptomonadida Family Cryptomonadidae Chilomonas paramecium Cryptomonas sp. Class CHRYSOMONADEA Order Chromulinida Family Chromulinidae Chrysococcus sp. #1 Chrysococcus sp. #2 Chrysococcus sp. #3 Family Anthophysidae Cephalothamnion sp. Class VoLvocEA Order Volvocida Family Volvocidae Pandorina morum Eudorina elegans Polytoma sp. Class EUGLENEA Order Euglenida Family Euglenidae Euglena klebsi Euglena proxima Euglena terricola Astasia longa Astasia klebsi Astasia sp. #3 Astasia sp. #4 Phacus pyrum Rhabdomonas sp. Menoidium incurvum Menoidium acutissima Family Trachelomonadidae Trachelomonas verrucosa Order Peranemida Family Peranemidae Peranema trichophorum Distigma proteus Family Petalomonadidae Petalomonas trisulcatum Petalomonas sp. #2 Family Anisonemidae Anisonema acinus Anisonema sp. #2 Entosiphon sulcatum Entosiphon sp. #2 Subphylum ZoomasTiGopHoRA Class PRoToMASTIGEA Order Choanomastigida Family Codosigidae Codosiga sp. Order Bodonida Family Bodonidae Bodo sp. Colponema loxodes Class KARYOMASTIGEA Order Dikaryomastigida Family Hexamitidae Hexamita sp. Phylum SarcopINa Subphylum Aurorracta Class ACTINOPODEA Order Heliozoida Family Acanthocystidae Acanthocystis sp. Pinaciophora sp. Class FILoRETICULOSEA Order Filida Family Euglyphidae Assulina seminulum Euglypha cristata Euglypha laevis Euglypha tuberculata Euglypha umbilicata Paraeuglypha sp. Pseudoeuglypha gracilis Tracheleuglypha dentata Trinema lineare Trinema penardi Trinema complanatum Trinema sp. #4 Subphylum HyprauLa Class CycLEa Superorder Lobeda Order Granulopodida Family Arcellidae Arcella artocrea Arcella catinus Arcella discoides Arcella hemisphaerica Arcella mitra Arcella vulgaris Family Centropyxidae Centropyxts aculeata Centropyxis sp. #2 Cyclopyxis arcellotdes Family Cryptodifflugidae Cryptodifflugia oviformis Family Difflugidae Difflugia avellana Difflugia bidens Difflugia elegans Difflugia fallax Difjlugia globulosa Difflugia lebes Difjlugia oblonga Difflugia olla Difflugia pyriformis Difflugia viscidula Difflugia sp. #11 Difflugia sp. #12 Lesquereusia spiralis Protozoa oF Lake Irasca AREA MossEs 619 TABLE 1; (Continued. ) Family Plagiopyxidae Plagiopyxis callida Family Hyalospheniidae Hyalosphenia elegans Hyalosphenia papilio Family Nebelidae Heleopora picta Nebela bursaria Nebela collarts Nebela tuberculata Nebela sp. #4 Family Quadrulellidae Paraquadrula discoides Order Pelobiontida Family Pelomyxidae Pelomyxa palustris Pelomyxa belevsku Order Eruptida Family Hartmannellidae Saccamoeba lucens Saccamoeba sp. #2 Order Conopodida Family Mayorellidae Mayorella sp. #1 Mayorella sp. #2 Mayorella sp. #3 Order Pharopodida Family Vannellidae Vannella minor Vannella sp. #2 Incertae sedis Mastigamoeba sp. Phylum CiLiopHora Subphylum KINETOFRAGMINOPHORA Class HypostoMEa Order Nassulida Family Nassulidae Nassula sp. #1 Nassula sp. #2 Family Microthoracidae Drepanomonas dentata Family Leptopharyngidae Leptopharyx sphagnetorum Order Cyrtophorida Family Chilodonellidae Chiodonella cuculullus Family Chlamydodontidae Chlamydodon sp. Class GyYMNOSTOMEA Order Pleurostomatida Family Amphileptidae Amphileptus claparedei Amphileptus sp. #2 Bryophyllum sphagni Litonotus muscororum Loxophyllum sp. Order Karyorelictida Family Loxodidae Loxodes magnus Loxodes sp. #2 Order Haptorida Family Enchelyidae Lacrymaria sp. Family Spathidiidae Spathidium amphoriforme Family Tracheliidae Dileptus anser Dileptus cygnus Order Prorodontida Family Prorodontidae Prorodon discolor Prorodon cinereus Family Colepidae Coleps bicuspis Coleps inermis Coleps octospinus Coleps sp. #4 Order Prostomatida Family Holophryidae Holophrya saginata Class VESTIBULIFEREA Order Colpodida Family Colpodidae Colpoda steini Subphylum OLIGcoHYMENOPHORA Class HyMENOSTOMEA Order Hymenostomatida Family Glaucomidae Monochilum frontatum Family Frontoniidae Frontonta depressa Family Lembadionidae Lembadion bullinum Order Scuticociliatida Family Uronematidae Uronema sp. Family Loxocephalidae Loxocephalus plagius Family Cyclididae Cyclidium elongatum Cyclidium musccola Class PERITRICHEA Order Sessalida Family Vorticellidae Vorticella campanula Vorticella picta Vorticella lichenicola Vorticella sp. #4 Subphylum PoLyHyMENOPHORA Class HETEROTRICHEA Order Stichotrichida Family Holostichidae Holosticha vernalis Holosticha sp. #2 Keronopsis helluo Uroleptis limnetis Uroleptis sp. #2 Family Keronidae Kerona rubra 620 Tue UNiversiry oF Kansas ScrENCE BULLETIN TABLE I. (Continued.) Spirostomum teres Family Metopidae Metopus intercedens Order Sporadotrichida Family Oxytrichidae Onychodromus grandis Stylonychia sp. Metopus sp. #2 Family Euplotidae Metopus sp. #3 Euplotes ener Family Bursariidae Euplotes sp. #2 Bursaria truncatella Class SPIROTRICHEA ; 4 Bursaria sp. #2 Order Heterotrichida Family Spirostomidae Blepharisma sphagni Blepharisma sp. #2 Spirostomum minus Class OLIGOTRICHEA Order Oligotrichida Family Halteriidae Halteria grandinella EBERT: Protozoa AssociIATED WITH Forest Mosses. Euglypha umbilicata Euglypha vanoyei Placocista spinosa Trinema penardi Family Difflugiellidae Cryptodifflugia compressa Difflugiella minuta Difflugiella oviformis Subphylum HyprauLa Class CycLEa Superorder Lobeda Order Granulopodida Family Amoebidae Amoeba proteoides Family Centropyxidae Centropyxis elongata Centropyxis platystoma Order Bodonida Centropyxis silvatica Family Bodonidae Centropyxis silvatica form minor Bodo sp. #1 Cyclopyxis ambigua Bodo sp. #2 Cyclopyxis kahli Cyclopyxis puteus Family Difflugiidae Cucurbitella sp. Lesquereusia modesta Pontigulasia bigibbosa Order Eruptida Family Vahlkampfidae Ciliophrys sp. Vahlkampfia sp. Class FILORETICULOSEA Family Hartmannellidae Order Filida Saccamoeba sp. #1 Saccamoeba sp. #2 Saccamoeba sp. #3 Order Conopodida Family Mayorellidae Phylum MastTiGoPpHoRA Subphylum PHyromasTIGOPHORA Class VoLVvocEA Order Chlamydomonadida Chlamydomonas sp. Class EUGLENEA Order Euglenida Family Euglenidae Rhabdomonas sp. Order Peranemida Family Anisonemidae Anisonema sp. Family Petalomonadidae Petalomonas mediocanellata Petalomonas sp. #2 Subphylum ZooMAsTIGOPHORA Class PROTOMASTIGEA Cercobodo sp. Rhynchomonas nasutum Phylum SarcopiNna Subphylum Aurorracra Class ACTINOPODEA Order Heliozoida Family Ciliophryidae Family Euglyphidae Assulina muscororum Corythion dubium ry /; + ay Corythion pale hellum Mayorella sp. #1 Euglypha ciliata Order Pharopodida Euglypha ciliata form glabra Family Vannellidae Euglypha cristata Platyamoeba sp. #1 Euglypha laevis Vannella sp. #1 Euglypha strigosa Vannella sp. #2 Protozoa oF LAKE 621 Itasca AREA MosseEs TABLE IT. (Continued.) Order Thecida Family Thecamoebidae Thecamoeba granifera Thecamoeba sp. #2 Family Striamoebidae Striamoeba bradys Striamoeba quadrilineata Phylum CitiopHora Subphylum KINETOFRAGMINOPHORA Class GYMNOSTOMEA Order Haptorida Family Spathidiidae Spathidium amphoriforme Spathidium muscicola Spathidium spatula Family Tracheltidae Dileptus gracilis Family Actinobolinidae Actinobolina vorax Order Pleurostomatida Family Amphileptidae Amphileptus sp. Class VESTIBULIFEREA Order Trichostomatida Family Plagiopylidae Plagiopyla sp. Order Colpodida Family Colpodidae Colpoda cucullus Colpoda inflata Colpoda irregularis Colpoda penardi Tillina magna Class HyposroMeEA Order Nassulida Family Nassulidae Nassula protectissima Family Leptopharyngidae Leptopharynx sp. Family Microthoracidae Microthorax simulans & II shows the comparative size and morpho- logical contours of forest moss ciliates versus bog moss or cosmopolitan species. Plates III, IV & V show characteristic testate amebas from the two associations. Among the sarcodines that form shells, again the forest-moss-associated species are the more edaphic. The numerous aquatic species of Arcella, Hyalosphenia, Nebela and Difflugia associated with the bog mosses are replaced by species of Centropyxis, Cyclo- pyxis and Difflugiella on the forest mosses. Trinema lineare, and Corythion dubium of the bog-mosses are replaced on forest mosses Order Cyrtophorida Family Chilodonellidae Chilodonella wisconsinensis Class SucTOREA Order Exogenida Family Podophryidae Podophyra collini Subphylum OLicoHYMENOPHORA Class HyMENOSTOMEA Order Hymenostomatida Family Tetrahymenidae Colpidium sp. Family Glaucomidae Monochilum sp. Pseudoglaucoma muscorum Order Scuticociliatida Family Cinetochilidae Cinetochilum margaritaceum Family Cyclidiidae Cyclidium musccola Cyclidium terricola Class PERITRICHEA Order Sessalida Family Epistylididae Rhabdostyla muscorum Subphylum PotypHYMENOPHORA Class HETEROTRICHEA Order Stichotrichida Family Holostichidae Gonostomum affine Keronopsis sp. Family Keronidae Kerona sp. Order Sporadotrichida Family Oxytrichidae Gastrostyla muscororum Family Aspidiscidae Aspidisca sp. Class OLIGOTRICHEA Order Oligotrichida Family Halteriidae Halteria grandinella by T. penardi and C. pulchellum. A\l- though Euglypha cristata and E. laevis are found in both associations, they are less which also E. vanoyet not present numerous on the forest mosses, harbor EF. ciliata, E. umbilicata, and E. strigosa, species that were on bog mosses. Among the flagellates found in the two associations, phytoflagellates were found only among the bog mosses (except for a lone Chlamydomonas sp. in the forest mosses sam- ple). Even for the bog mosses the phytofla- gellates were limited to a few, generally Colorless flagellates cosmopolitan — species. 622 Tue University oF Kansas ScrENCE BULLETIN a were largely euglenoids in the bog moss association, but with both euglenoids and bodonids in the forest moss association. For other groups, comparisons are difh- cult to make without further study and will not be attempted until further data can be obtained. Even so, it is evident that the two types of moss-associations are much different so far as the Protozoa they support, the bog-moss Protozoa being largely acidophilous and aquatic, the forest-moss-associated Protozoa being the more edaphic. A comparison of the Protozoa of the Lake Itasca forest mosses to those found by Stout (1963) from soil and litter of the Chiltern Hills shows little similarity for testate sarco- dines, except the presence of cosmopolitan Assulina seminulum, Trinema lineare, Eu- glypha laevis and E. ciliata. Some ciliates noted by Stout (loc. cit.) were found either in our bog moss or forest moss samples, but with no distinct relationship to either as- sociation. There is, however, a distinct resemblance of the testate sarcodines found in the Lake Itasca forest moss sample to those reported by Bonnet and Thomas (1960) from soils and mosses of the eastern Pyrenees Moun- tains. They list Euglypha ciliata and E. c. form glabra, Trinema lineare and Difflugiella oviformis as characteristic of forest mosses. Those were also present in the Lake Itasca forest moss sample. They found Corythion dubium, Euglypha laevis, Centropyxis sil- vatica, Cyclopyxis puteus, C. Rahli, and C. arcelloides to be cosmopolitan in the forest. The Lake Itasca forest moss sample con- tained all but C. arcelloides. That was found on the bog mosses. Although our earlier brief study (Sanford and Bovee, 1974) suggests the probability that some protozoans (e.g., certain species of testate amebas) are more often associated with one moss than another, no attempt was made in this study to determine such a rela- tionship. However, Hoogenraad and DeGroot (1952) noted distributional differences of testate amebas on 5 species of Sphagnum col- lected in New Jersey. Only Assulina mus- corum, A. seminulum and Trinema lineare were found on all 5 species. For the genus Difflugia, they found that no species thereof was to be found on more than 3 of the 5 Sphagnum spp. Amphitrema spp. were seen on only 2 of the Sphagnum spp., the other Lobeda being each limited to a single species of Sphagnum. Hyalosphema spp. were pres- ent on only 2 of the 5 Sphagnum spp. They also noted distributional differences for Ne- bela spp., Centropyxis spp. and Euglypha spp. on the 5 species of Sphagnum from New Jersey. Sanford and Bovee (loc. cit.), examining bog mosses from Twin Lakes near Lake Itasca, compared the testate amebas found on Sphagnum spp. to those on other genera of mosses from the same bog. They found Hyalosphenia papilio, Heleopora petricola, Arcella hemisphaerica, Ouadrulella symmet- rica, Cryptodifflugia oviformis, Assulina seminulum and Euglypha cristata only on Sphagnum spp. They noted Euglypha ciliata, E. filifera, Arcella costata, A. dentata and Centropyxis aculeata on other mosses, but not on Sphagnum spp. Some amebas were on both groups of mosses, but more often on Sphagnum spp., e.g., Euglypha alveolata, or conversely more prevalent on other mosses than on Sphagnum spp., e.g., Euglypha laevis, Trinema lineare, Arcella vulgaris, or were about equally distributed, e.g., Difflugia globulosa and Difflugia binucleata. Bamforth (1969) in examining the moist, generally-acid forest litters and underlying soils in Louisiana for Protozoa, found that certain testate amebas predominated in litters, others in the moist soils. In the litters he found Centropyxis spp. common, but they were infrequent in soils. In this study I found Centropyxis spp. more numerous in the forest mosses than in the bog-mosses. Bamforth (loc. cit.) also remarked that the testate amebas found in his study were gen- erally those of Sphagnum bog habitats. Of 19 species from 10 genera that he found, I found in Buell Bog 7 of the 10 genera, but only 8 of the 19 species that he found, and 4 of those are generally considered to be prob- ably cosmopolitan. Evidently, while the gen- era are similar, there is a likely differential Protozoa oF LAKE ItTAscA ArEA MossEs 623 distribution of species due to some of the ecologic factors cited earlier in this paper. Chardez (1957, 1960, 1965, 1967, and other papers) and with Krizelj (1970) has extensively catalogued the protozoan fauna (especially testacid amebas) of forest soils and of mosses (from sphagnum bogs, of forest floors, rocks and tree trunks). He finds (1967), as I have found, that the different habitats—bogs, moist forest floor, soils, or mosses above ground—exhibit distinct faunas of protozoa, larger species being found in bogs, progressively smaller ones in the less to more dry situations. He (1967) considers Arcella spp. typical of lakes and ponds, Difflugia spp. character- istically found among aquatic mosses, with species differing with the locale and species of mosses, Cyclopyxis spp. typical of mosses above ground (as I have found, also), differ- ing species of Centropyxis in different situa- tions (Centropyxis aculeata in wet locales and C. platystoma in drier ones). Some species he finds (1957) generally distributed on forest mosses, e.g., Corythion dubium, Assulina muscorum, Trinema lineare (I found all three on forest mosses). Some he found (1960) only on sphagnum bog mosses, not on forest mosses, e.g., Nebela collaris, Centropyxis aculeata, Hyalosphenia papilio (this again agrees with findings in my study). Others he found only on forest mosses, not on sphagnum bog mosses, €.g., Corythion pulchellum, Trinema complanatum (again in agreement with my present study). He (1965) classifies the testacid amebas under five ecological categories: moss-dwell- ing, sphagnicolous, aquatic, terricolous or marine. He lists (1960, 1965) most Arcella spp. as aquatic (those I found in the bog must be so considered). Most Difflugia spp. and Centropyxis spp. he calls aquatic, but some are sphagnicolous, e.g., D. pyriformis, D. globularis, other species having aquatic, sphagnicolous and moss-dwelling varieties, e.g., D. oblonga, D. aerophila, and some be- ing moss-dwelling, e.g., D. rubescens, C. sylvatica. Most Nebela spp. and Hyalosphemia spp. he calls sphagnicolous (I also found them so). Some Euglypha spp. he terms aquatic, e.g., E. tuberculata, some sphagnico- lous, e.g., E. cristata, others moss-dwelling, e.g., E. bryophila, others ubiquitous, e.g., E. laevis (again my findings agree). His study (1970, with Krizelj) indicates that certain species are likely to be found on a particular species of moss, e.g., Difflugia lucida only on Eurhynchium striatum, or Centropyxis aerophila only on Ctenidium mol- luscum, but other species on several mosses, e.g., Difflugia oviformis on 4 species of mosses (but not on another four species), most often on Rhitidiadelphus triqueterus. Stull other species of testacid amebas were to be found on all eight species of mosses studied, e.g., Nebela collaris. Clearly, the communities of Protozoa as- sociated with various communities and spe- cies of mosses merit further study. ACKNOWLEDGEMENTS I am pleased to thank the University of Minnesota for the use of research space, microscopic and col- lecting equipment, and housing provided during these studies at the Lake Itasca Biological Station during August 1977 when researches for this paper were conducted there. I especially thank Dr. David F. Parmelee, Director of the Station, and the Department of Ecology at the University of Minnesota for the summer teaching appointment at the Station that made these studies possible. LITERATURE CITED ALLEGRE, C. F. AND JAHN, T. L. 1943. A study of the Genus Phacus Dujardin (Protozoa; Eu- glenoidina). Trans. Amer. Microscop. Soc. 47:233-244. BamrFortu, S. S. 1969. Protozoa and algae of the Mississippi deltaic soils. Proc. Louisiana Acad. Sci. 32:69-77. Bonnet, L. anp THomas, R. 1960. Faune terrestre et d’eau douce des Pyrénées-Orientales. Thé- camebiens du sol. Vie et Milieu 11(4) Suppl.: 1-103. CuarpEz, D. 1957. Thécamoebiens des mousses aériennes. (Privately published and distributed by the author.) 1960. Etude comparee des Thécamoebiens de trois biotypes dans trois milieux differents. Bull. Inst. Agron. Sta. Rech. Gembloux 28: 132-138. 1965. Ecologie générale des Thécamoe- biens. (Rhizopoda, Testacea.) Bull. Inst. Agron. Sta. Rech. Gembloux 33:309-341. ——. 1967. Histoire Naturelle des Protozoaires Thécamoebiens. Les Naturalistes Belge, Brux- elles. N. Boubée et Cie., Paris. 100 pp. 624 Tue Universiry oF Kansas ScrENCE BULLETIN AND Krizevy, S. 1970. Recherches sur |’eco- systeme forét. Serie B. La chénaie mélangée calcicole de Virelles-Blaimont. Contribution #22. Protozoaires thécamoebiens et ciliés du sol. Bull. Inst. roy. Sci. nat. Belg. 46(12): 1-17. Coruiss, J. O. 1977. Annotated Assignment of Fam- ilies and Genera to the Orders and Classes currently comprising the Corlissian scheme of higher classification for the Phylum Cilio- phora. Trans. Amer. Microscop. Soc. 96: 104-140. Heat, O. W. 1961. The distribution of testate amoebae (Khizopoda: Testacea) in some fens and bogs in northern England. J. Linn. Soc. London, Zool. 64:369-382. Hoocenraap, H. R. anp DeGroot, A. A. 1952. Thekamoébe Moosrhizopoden aus Nordamer- ika. Arch. Hydrobiol. 47:229-262. Jaun, T. L.; Bover, E. C. anp Jann, F. F. 1979. How to know the Protozoa, 2nd _ edition. W. C. Brown, Dubuque, Iowa. 279 pp. AND JAHN, F. F. 1949. How to Know the Protozoa. W. C. Brown, Dubuque, Iowa. 356 pp. JouHnson, L. P. 1944. Euglenae of Iowa. Trans. Amer. Microscop. Soc. 48:97-135. Kant, A. 1935. Urtiere oder Protozoa. 1. Wim- pertiere oder Ciliata (Infusoria), in Dahl, F. Die Tierwelt Deutschlands. pp. 1-886. Kupo, R. R. 1966. Protozoology, 5th ed. Chas. C. Thomas, Springfield, Ill. 1175 pp. Penarp, E. 1902. Faune Rhizopodique du Bassin de Léman. Kundig, Genéve. 714 pp. —. 1905. Les Sarcodines des Grand Lacs. Kundig, Genéve. 134 pp. — —. 1909. Sur quelques rhizopodes des Mous- ses. Arch. Protistenk. 17:258-296. 1922. Etudes sur les Infusoires d’eau douce. George, Geneve. 331 pp. SANForp, M. anp Bovesr, E. C. 1974. Differential distribution of testate sarcodines on mosses. Trans. Amer. Microscop. Soc. 93:431-432. SHAWHAN, F. M. anv JaHn, T. L. 1947. A survey of the Genus Petalomonas Stein (Protozoa: Euglenida). Trans. Amer. Microscop. Soc. 46:182-189. StEPANEK, M. 1963. Rhizopoden aus alten, aus- getrockneten Moosproben. Hydrobiologia 21: 304-327, Stout, J. D. 1963. Some observations on the Proto- zoa of some beechwood soils of the Chiltern Hills. J. Anim. Ecol. 32:281-287. Protozoa oF Lake Itasca AREA Mossxs aN ae" iy? ANN | y S. 4 g I Pirate I. Cirioprora AssociaTED WITH MossEs. Figs. 1-3. Dileptus spp.; Fig. 1. Dileptus anser, cosmopolitan, found in sphagnum bog I 2 cy f in sphagnum bog, Fig. 3. D. graciiis, characteristic of wet forest moss. 1-6. Spa m 4 s la politan, found in forest moss; Fig. 5. S$. amphoriforme, characteristic of bog and fo n ( Sac; la ha teristic of mosses, found in forest moss. Figs. 7-8. Frontonia spt Biggs) mbes cue cosmopolitan, not f this st F, depressa, characteristic of mosses, found in sphagnum bog. Figs 9-10. Chilodonella spp Fi EG) nella cucul cosmopolitan, found in sphagnum bog; Fig. 10. C. wisconsinensis, moss-dwelling, found i t mos Figs. 11-12. ¢ spp.; Fig. ll. C. elongatum, found in sphagnum bo Bi ]2 C. muscicola, found in sphagnum 1 forest r se ( (not shown), like C. muscicola, found on forest moss. Figs. 13-15. Colpoda spy Fi ! C. cme cosmopol i of mosses, found on forest ss; Fig. 15. ¢ litan, found in spl int, found on forest n forest moss; Fig. 14. C. inflata, characteristic trregularis and C. penardi (not shown), similar to ( bog; C. 626 Tue University of Kansas ScieENCE BULLETIN SANNY A SE KEE AAAS SS LE 7 ; : 7 SAAN Nw > a mm Qe er AAKTANT SS Pirate II. CitiopHora AssocIaATED WITH Mossks. Fig. 1. Loxophyllum sp., from sphagnum bog. Fig. 2. Bryophyllum sphagni, characteristic of sphagnum bog. Fig. 3. Leptopharynx sphagnetorum, characteristic of sphagnum bog; another species (not shown, similar) on forest moss. Fig. 4. Bryophrya sp., characteristic of mosses, but not found in this study. Fig. 5. Bresslaua bavariensis, typical of forest moss, but not found here. Fig. 6. Kerona rubra, more or less cosmopolitan. found in sphagnum bog. Fig. 7. Keronopsts helluo, characteristic of mosses, found in sphagnum bog. Fig. 8. Euplotes patella, cosmopolitan, not found, shown here for comparison. Fig. 9, Euplotes muscicola, found in sphagnum bog. Fig. 10. Rhabdostyla vernalis, not found in this study, but shown for comparison. Fig. 11. Rhabdostyla muscorum, characteristic of mosses, found in forest moss. Protozoa oF LAKE Itasca AREA MosseEs 627 wR. un a, ry } ae: oo a we NS Arsge=3e SSisSse Puate III. Firose Tesracip AMEBAS OF MossEs (SHELLS ONLY). 7 ] j . ‘ y ? Puy J Fig. 1. Aussulina muscorum, of forest mosses, a, broad view, 6, edge view. Fi bog. Fig. 3. ig ¢v, CSO HS FRET ips LAie ISS! J Prate IV. Lopose Tresractp AMEBAS OF MosstEs (SHELLS ONLY). Figs. 1,2,4,5,7,12. Arcella spp., side views; Fig. 1. A. catinus; Fig. 2. A. hemisphaerica; Fig. 4. A. artocrea; Fig. 5. A. vulga Fig. 7. A. discotdes; Fig. 12. A. mitra; A. vulgaris, cosmopolitan in fresh waters, others typical of sphagnum bog waters Figs. 3,6,16,17. Cyclopyxis spp., side views; Fig. 3. C. kahli; Fig. 6. C. puteus; Fig. 16. C. ambigua, all on forest mosses; Fig. 17. C. arcelloides, cosmopolitan in forest situations, found in sphagnum bog. Figs. 8,9,11,13. Centropyxis spp. Fig. 8. C. elongata, a, long section; Fig. 9. C. platystoma, a, long section; Fig. ll. C. sylvatica, a, long section; all ommon on forest mosses, not on sphagnum; Fig. 13. C. aculeata, cosmopolitan in fresh waters, in sphagnum bog, but zo? in forest mosses. Figs. 10,15. Hyalosphenia spp., typical of sphagnum bog, not of forest mosses; Fig. 10. H. papilio, a, edge view; Fig. 15. H. elegans, a, edge view. Fig. 14. Plagiopyxis callida, a, long section, typical of forest situations, found in Shum bog. Protozoa oF Lake Irasca AREA MossEs 629 BOS ~\ lie gna are! ewe = ind > an Sa RSs 5, 3 = Sl ewe ett PiatEe V. LogposE TEsTacip AMEBAS OF MossEs (SHELLS ONLY). Figs. 1,2. Lesquereusia spp.; Fig. 1. L. modesta, found in forest mosses; Fig. 2. L. spiralis, foun ig bog. 3. Pontigulasia bigibbosa, found in forest mosses. Fig. 4. Same, side view. Fig. 5. Heleopora picta, a, edge view, on sphag num. Figs. 6,14,11. Nebela spp., typically moss-dwelling, found in sphagnum bog; Fig. 6. N. bursaria, a, edge view; Fig. 14. N. tuberculata; Fig. 11. N. collaris, a, edge view. Figs. 7,9,12,13,15-20. Difflugia spp.; Fig. 7. D. viscidula; Fig. 9. D. avellana; Fig. 12. D. oblonga; Fig. 13. D. pyriformis; Fig. 15. D. globulosa; Fig. 16. D. elegans; Fig. 17. D. fallax; Fig. 18. D. lebes; Fig. 19. D. olla; Fig. 20. D. bidens; all found in sphagnum bog. Fig. 8. Dziffiugiella minuta, a, edge view; b, D. oviformis; both filose testacids from forest mosses. Fig. 14. Cucurbitella sp., from forest mosses. Fig. 21. Paraquadrula discoides, from forest mosses. = We" o"e"e"e%e" 0707078088 © 0 8 9 2688S SSS SSS STOO SSS ESO e eee eee wee eseeeseees ees Bes Ss a 5 3 a a 2.8.8.9 0882662096866 5 959066686 62S 8S S88 ow © ow oo we rene ee ae es meesecescatatatatetata tate eMeka etateMete ete otetatetatecnetctatatetatetatetatatetatatatetststatstatatatetetetatsterststeretseraereeen . * . * ? OL SCIENCE BULLETIN The Proboscis of the Long-Tongued Bees: A Comparative Study a Po Aa Pe Bas as SN aa asa SI res DR ELEY By MARK L. WINSTON Vol. 51, No. 22, pp. 631-667 June 1, 1979 ANNOUNCEMENT The University of Kansas Science Bulletin (continuation of the Kansas Unt- versity Quarterly) is an outlet for scholarly scientific investigations carried out at the University of Kansas or by University faculty and students. Since its incep- tion, volumes of the Bulletin have been variously issued as single bound volumes, as two or three multi-paper parts or as series of individual papers. Issuance is at irregular intervals, with each volume prior to volume 50 approximately 1000 pages in length. Effective with volume 50, page size has been enlarged, reducing the length of each volume to about 750 pages. The supply of all volumes of the Kansas University Quarterly is now ex- hausted. However, most volumes of the University of Kansas Science Bulletin are still available and are offered, in exchange for similar publications, to learned societies, colleges and universities and other institutions, or may be purchased at $20.00 per volume. Where some of these volumes were issued in parts, individual parts are priced at the rate of 2 cents per page. Current policy, initiated with volume 46, is to issue individual papers as published. Such separata may be purchased individually at the rate of 3 cents per page, with a minimum charge of $1.00 per separate. Subscriptions for forthcoming volumes may be entered at the rate of $20.00 per volume. All communications regarding exchanges, sales and subscriptions should be addressed to the ExcHaNce LiprariAN, UNIVERSITY oF Kansas Lipraries, LAwreNcE, Kansas 66045. Reprints of individual papers for personal use by investigators are available gratis for most recent and many older issues of the Bulletin. Such requests should be directed to the author. The International Standard Serial Number of this publication is US ISSN 0022-8850. Editor Eugene C. Bovee Editorial Board William L. Bloom Philip W. Hedrick Rudolf Jander Harvey Lillywhite Charles D. Michener Norman A. Slade Henry D. Stone George W. Byers, Chairman THE UNIVERSITY OF KANSAS SCIENCE BULLETIN Vol. 51, No. 22, pp. 631-667 June 1, 1979 The Proboscis of the Long-Tongued Bees: A Comparative Study’ Mark L. WINsTON CONTENTS PATISTERUNGAP ccacceecen s Bathe take ese a ir re er RE Ee hd cee ne Ne AD abel 631 UNGER ODU. Gis] © Near ere eee ee ae eee Le an tee ES a cet es ee 2 Be ee 8 ee Eee 631 MIARERTALSMAN DE VIE THO DS meee ont. seceen. 5 ere et Be Te EN Beis hdd ee Ope, Poe eee ee ee 632 MORPHOL OGYROEMIIE seABIOMARILUARY. (COMPIEEX 6c) 20. ccs ice, ek cc eee nxt noc ee ea gee seca cn dene eee 634 INA 2x |< cme ee mmetnte Piney meee re ISTO eect eee A ee Re 634 LBD LU rr eee mn eB EEE ce tN AN acdc oa dansdeha wen seb t ooeds deca ehacatee cera Ba ee oc eae 641 (Comimnnminas, Sabie. pee eee Bee) i ee Be oe Re ae a et ee anor ne eee Seen ee ee eee 645 Mevachilidacumeseres trae nt eins nities 0t Oi, Ore eat ew ne eS ere Seal can dh catinss ee Gatun ness coer Be 647 FA tho phoroid mer ou pe mess 2s. fei seers oo erste Jee 2 oe Fe ce ee ee 650 AB ISCUSS 1ON eee ere eee ern een eat de UE he als Po Ne ie eee ee eel ee ee ee 653 PANCKSN OUTER DG E MENTS amen eae NN ae sie Aa 8 A eae od ea seerev acces eeeee tee oe 655 NB IGIgERUACTAUORU: © U:1sE 1) anemia ry er ah Sit ee Ree 2 2s fe Sy ce eta Stacie tins ee a 655 ABSTRACT The morphology and variation of the labiomaxillary complex in the long-tongued bees (families Apidae, Anthophoridae, Fideliidae, and Megachilidae) is described, including new characters, and the results applied to apoid taxonomy. Genera have been grouped by cladistic analysis into mouthpart groups, largely according to unique, shared, derived characters. This study supports a monophyletic origin for the long-tongued bees, with the Melittidae, or at least the genus Ctenoplectra, as a sister group. The Megachilidae are distinct from the other families. Among the Megachilidae, the Lithurginae diverge from the Megachilinae in labiomaxillary characteristics, and the Megachilini and Anthidiini are closely related, with the exception of Dioxys, which diverges from all other Megachilinae. Fideliids are grouped with the Anthophoridae and Apidae according to this analysis. Among the apids and anthophorids, close relationship between the Apidae and Xylocopinae is indicated, particularly between the Meliponinae and the Xylocopini. The allodapines form a distinctive group within the Xylocopinae, and elevation of these bees to tribal status may be justified. Triepeolus and Thalestria are distinct from the other Nomadinae, and reclassification of these may also be warranted. The position of Canephorula as a sister-group to the Eucerini is confirmed by mouthpart characteristics. INTRODUCTION Long-tongued bees (families Apidae, An- thophoridae, Fideliidae, and Megachilidae) have the glossa, labial palpi, and galeae as long as or longer than the stipites, and the first two segments of the labial palpi elon- gate, flattened, and sheathlike. Despite the 1 Contribution number 1686 from the Department of En- tomology, University of Kansas, Lawrence, Kansas 66045, U.S.A. functional significance and the many taxo- nomically useful characters of the _labio- maxillary complex of bees, the morphology of and relationships indicated by bee mouth- parts have been neglected in recent litera- ture. The mouthparts of sphecoid wasps were described by Ulrich (1924), and vari- ous authors have described mouthparts of species of long-tongued bees (Michener, 1944, Anthophora edwardsit; Snodgrass, 1956, Apis mellifera; Camargo, 1967, Melipona 632 Tue University oF Kansas SciENCE BULLETIN marginata; luga, 1968, Bombus terrestris, lapidarius, agrorum). Saunders (1890), De- moll (1908), and Correia (1973) compared mouthpart morphologies for certain genera of bees, and many authors have used some characteristics of bee mouthparts for taxo- nomic purposes. The purposes of the present study are: 1) to describe the morphology and variation in the labiomaxillary complex of long-tongued bees, 2) to compare characteristic features of the mouthparts of representative genera, and 3) to use characters of the’ mouthparts for taxonomic purposes. MATERIALS AND METHODS The terms labiomaxillary complex, proboscis, and mouthparts are used interchangeably in this paper to refer to the combined labium and maxillae. The mouthparts of representative genera (Table 1) were removed from specimens, cleared in 10% KOH, washed in acetic acid, and preserved for study in glycerol. Unless otherwise noted, specimens were females. Examination was with a dissecting micro- scope, drawings were made of distinctive structures. Other species were examined in groups whose diver- sity justified such work. For comparative purposes, the classification of Michener (1944, 1974a, Table 1) is used. Since the branching sequences suggested by mouthparts are often similar to those indicated by Michener, taxonomic names are used to identify mouthpart groups. When used in the context of mouthpart group rather than a taxon, the name is followed by the word “group.” All members of a taxon are not necessarily members of the same mouthpart group; such cases are discussed in the text. A similar study being conducted by L. Greenberg is the source of most of the information concerning short-tongued bees discussed here. Genera are placed in mouthpart groups largely according to the principles of cladistic analysis, as modified by Ashlock (1974), Michener (1974b), and Mayr (1976). The mouthpart groups should not be used alone to make the classification of long-tongued bees; they are meant to show only relationships as shown by mouthparts and may or may not be iden- tical to groupings based upon other characters. Dendrograms (Figs. 18-22) are based on mouthparts of the forms examined by me; no doubt other char- acters would improve them, e.g., by reducing the number of multifurcations. Synapomorphous char- acters (Table 2) are indicated by numbers on the stems and parenthetically in the text. Groups lacking synapomorphous characters in the mouthparts are indicated by dotted lines in the cladograms, following the classification of Michener (1944). In the text, the terms similarity, relation- ship, and afhnity are used interchangeably in dis- cussing phenetic closeness. Groups not characterized by synapomorphic characters may be monophyletic, but whether they are holophyletic or paraphyletic (sense of Ashlock, 1971) cannot be determined without consideration of other characters. Table 3 shows the morphological terms used, sources for the terms (major works on apoid struc- ture, not relating to priority), and other terms pre- viously used. The terminology of Michener (1944) and Snodgrass (1956) is used when it agrees with current interpretations. Terms listed without sources but with previous terminology relate to structures previously described, but renamed here. Terms with neither sources nor previous nomenclature refer to structures to the best of my knowledge not pre- viously described or named. In the text, names of structures are italicized where they are most fully described. The labiomaxillary complex consists of the max- illae (laterally) and labium (medially) (Fig. 1). For purposes of description, the proboscis is considered to be extended downward. Thus, “anterior” refers to the surface that is then directed forward, equiva- lent to “dorsal” in papers that consider the pro- boscis to be extended forward. Four representative views of mouthparts (Figs. 23-45) which best reveal important structures are used: a) outer view of the maxilla, b) inner view of the maxilla, c) posterior view of the labium, and d) anterior view of the labium. Stippling indicates membranous areas; dashed lines, sclerotized regions passing behind (in the view presented) others; and dotted lines, weakly sclerotized areas. Below are listed conventions used either for clarity or because certain structures were not examined in detail for all groups. a) Outer view of the maxilla 1) The basistipital process is not generally shown in this view, since it is usually ob- scured by the cardo. It is shown in the inner view. 2) Only the basal end of the galea is usually shown, and no galeal hairs or ribbing are included. 3) The basal end of the suspensory thickening is not shown. 4) Only the apical ends of basally broken car- dines are shown. 5) Only basal segments of long maxillary palpi are shown. b) Inner view of the maxilla 1) The galea has been unfolded to reveal the midrib and basigaleal area. c) Posterior view of the labium 1) Only one half of the lorum, and parts of the associated cardo and_ basistipital process, are shown. 2) The glossa is represented in repose, so that the paraglossae are retracted and are not shown. The glossal rod is drawn only when visible from a posterior view; annulations and hairiness of the glossa are only sche- STON AND MICHENER (1977). MEGACHILIDAE LITHURGINAE Lithurge gibbosus Lithurgommia wagenknechti Trichothurgus dubius MEGACHILINAE ANTHIDINI Anthidiellum notatum robertsoni Anthidium manicatum Aztecanthidium xochipillium Callanthidium illustre Dianthidium ulket Dioxys productus subruber Euaspis abdominalis Heteranthidium bequaerti Immanthidium repetitum Hypanthidium taboganum Nananthidium tamaulipanum Odontostelis bivittata Pachyanthidium bouyssoni Paranthidium jugatorium perpictum Parevaspis carbonaria Spinanthidium wolkmanni Stelis aterrima MEGACHILINI Anthocopa copelandica Ashmeadiella bucconis Chalicodoma (Chelostomoides) angelarum Chalicodoma (Chelostomoides) exilis Chalicodoma cincta combusta Chalicodoma rufipes Chalicodoma torrida Chelostoma fuliginosum Chelostomopsis rubifloris Coelioxys edita Creightonella frontalis Heriades carinata Hoplitis albifrons argentifrons Megachile albitarsis Megachile frugalis pseudofrugalis Noteriades sp. Osmia lignaria Osmia subaustralis Proteriades deserticola FIDELIIDAE Fidelia sp. Neofidelia profuga (male) ANTHOPHORIDAE NoMaDINAE Biastes brevicornis Caenoprosopis crabronina Holcopasites heliopsis Letopodus lacertinus Nomada annulata Thalestria sp. Triepeolus verbesinae ANTHOPHORINAE EUCERINI Eucera chrysopyga Melissodes agilis Peponapis crassidentata Svastra atripes Thygater amaryllis TABLE List oF SPECIMENS EXAMINED, CLASSIFIED AccorDING TO MicHENER (1944, 1974) ann Win- Xenoglossa fulva CANEPHORULINI Canephorula apiformis MELECTINI Melecta californica Thyreus ramosa CENTRIDINI Centris poecila Epicharts elegans ANTHOPHORINI Amegilla comberi Anthophora cockerelli Anthophora occidentalis TETRAPEDIINI Tetrapedia sp. (male) EXOMALOPSINI Ancyloscelis panamensis Caenonomada brunerii Exomalopsis zexmeniae Tapinotapis caerula CTENIOSCHELINI Ericrocis lata Mesocheira bicolor MELITOMINI Diadasia afflicta Melitoma segmentaria XYLOCOPINAE XYLOCOPINI Lestis aeratus Xylocopa brasilianorum varipuncta Xylocopa fimbriata Xylocopa v. virginica CERATININI Allodape stellarum Braunsapts factalis Ceratina (Ceratinidia) sp. Ceratina (Pithitis) sp. Ceratina calcarata Macrogalea candida Manuelia gayi APIDAE APINAE APINI Apis dorsata Apis mellifera BOMBINAE BoMBINI Bombus pennsylvanicus Psithyrus variabilts EUGLOssINI Euglossa cordata (male) Eulaema cingulata Eupusia sp. (male) Euplusia violacea MELIPONINAE Melipona fasciata Melipona marginata Melipona rufiventris Meliponula bocandet Trigona (Hypotrigona) sp. Trigona capitata zexmeniae Trigona chanchamayoensis 634 THe UNIVERSITY oF KANSAS SCIENCE BULLETIN TABLE, 2. SyNAPOMORPHOUS CHARACTERS OF THE MoutTuHparts oF LoNnc-ToncuED Bees. See Figures 18- 22 for dendrograms using these characters; the structures themselves are more fully described in the text. 1) mentum elongated and flared distally, articu- lating distally with the basal process of the prementum and basally with a v-shaped lorum (Figs. 1; 2c, d) ligular arm distinct from prementum, with no region of continuous between them (Fig. 14a, b) basistipital process elongated sclerotization subligular process curved anteriorly (Fig. 13) stipital comb present (Fig. 2a) flabellum present at apex of glossa (Fig. 2c) glossa with sclerotized rod extending its entire length glossa, galea, and labial palpus as long or longer than stipes (Fig. 2) galea with midrib (Fig. 2a) ends of stipital sclerite not expanded (Fig. 6a) stipital comb lost dististipital process present (Fig. 7) labial palpus with brush on cancavity of first segment (Fig. 11) ligular arms % length of prementum or less (Fig. 14a) lacinia with comb (Fig. 8c) ligular arms secondarily fused with prementum (Fig. 14c) inner cardinal process elongated (Fig. 5c) basal process of prementum convexly curved (Fig. 10c) inner and outer cardinal processes elongated (Fig. 3a) subligular process separated from prementum by membranes (Fig. 12a) brush on third segment of maxillary palpus (Fig. 61) matic. The apical portion of the glossa is not represented. 3) Only the basal segment (or a part thereof) of the labial palpus is represented. d) Anterior view of the labium 1) Only one suspensory thickening is shown, and only its distal part. 2) The mentum and lorum are not represented. 3) Only one paraglossal suspensorium, para- glossa, and basal segment of the labial palpus are drawn, and one half of the basiglossal sclerite. These structures, as well as the glossa and ligular arms, are drawn in the extended position, contrary to that in the posterior view of the labium. expanded sclerotized region at junction of stipi- tal and basistipital thickenings (Fig. 5e) paraglossa at least % as long as glossa subligular process expanded into U-shaped proc- ess (Fig. 12f) paraglossa as long as glossa striations in membrane underlying lacinia area between subgalea and stipital sclerite partly sclerotized prementum partly membranous and _ flattened (Fig. 12g) brush on expanded lobe of the palpiger bristles on membranous fold basad to the basi- galeal area subligular process as in Figure 12c sclerotized ridge along outer margin of stipes (Fig. 4)) stipital sclerite expanded apically into knob (Fig. 6d) anterior longitudinal brace robust both ends of stipital sclerite expanded to an- terior edge of membrane lying between stipital sclerite and subgalea (Fig. 61) stipes with strong comb concavity, comb with robust, blunt teeth (Fig. 4h) bipartite stipital thickening fused with stipital sclerite, with sclerotized area at junction of two sections of bipartite thickening (Fig. 5c) basistipital process largely formed by expanded basistipital thickening (Fig. 5b) stipital comb reduced, weak bulla on inner cardinal process lacinia hairless, membranous transverse sclerotized ridge basal process (Fig. 3e) glossa, labial palpus, and galea much longer than stipes to cardinal MORPHOLOGY OF THE LABIOMAXILLARY COMPLEX Maxillae The maxilla of the long-tongued bees re- tains the major structures of an insect max- illa (cardo, stipes, galea, lacinia, and maxil- lary palpus) but in modified form (Fig. 1). The cardo, stipes, and galea are elongated, and the stipes and galea are curved to sheath the labium when extended, features pre- sumably important for nectar uptake. The STRUCTURE Carbo cardinal condyle cardinal macula inner and outer cardinal processes STIPES stipital comb comb concavity basistipital process basistipital and stipital thickenings dististipital process supital sclerite LacINIA MaxILLary PaLpus GALEA blade subgalea basigaleal area lorum MENTUM PREMENTUM basal process subligular process SUSPENSORY THICKENING LasBiaL PaLpus palpiger LicuLarR ARM GLossa flabellum salivary channel BASIGLOSSAL SCLERITE ANTERIOR LONGITUDINAL BRACE PARAGLOSSA paraglossal suspensorium Progoscis oF LONG-TONGUED BEES TABLE 3. NoMENCLATURE. SOURCE PREVIOUS TERMINOLOGY Michener (1944) Snodgrass (1956) Michener (1944) Michener (1944) Michener (1944) Michener (1944) Michener (1944) Snodgrass (1956) Michener (1944) Michener (1944) Michener (1944) Michener (1944) Snodgrass (1956) Michener (1944) Michener (1944) Snodgrass (1956) Eickwort (1969) Michener (1944) hémisternal (luga, 1 968) apophyse cardinale (Iuga, 1968) extensory rod (Snodgrass, 1956) subgaleal sclerite (Winston and Michener, 1977) postpalpal segment of galea (Michener, 1944) prepalpal segment of galea (Michener, 1944) distal plate (Snodgrass, 1956) subligular plate (Michener, 1944) sternal sclerite (Iuga, 1968) suspensory rod (Snodgrass, 1956) anterior conjunctival thickening (Michener, 1944) bonnet-shaped sclerite (Snodgrass, 1956) notal and basiglossal sclerites (Iuga, 1968) basiparaglossa (Iuga, 1968) ligular arm (Snodgrass, 1956) lacinia is anterior to the stipes, near the food canal, and seems to function in closing that opening when the mouthparts are retracted. The cardo (Figs. 2a, d; 3) is the slender, cylindrical suspensory sclerite in the lateral wall of the otherwise membranous basal part of the proboscis; it connects the maxillae and labium to the cranium. The cardo is usu- ally slightly curved, commonly about two thirds as long as the stipes, but as short as half the stipital length in Xylocopa, or as long as the stipes in many genera. The THe UNIVERSITY OF KANSAS SCIENCE BULLETIN 636 MAXILLA it TTT NA LABIUM Fic. 1: Posterior view of generalized labiomaxillary complex. Progoscis oF Lonc-TONGUED BEEs 637 cardo articulates basally with the cranium by the cardinal condyle (Fig. 2a); the basal terminus of the cardo extends beyond this condyle, and on this terminus is inserted the cardinal muscle, reaching to the wall of the head (Snodgrass, 1956), and the elevator muscle of the stipes, extending to the mid- region of the stipes (Iuga, 1968). In many genera, there is a cardinal macula midway along the inner surface of the cardo (Fig. 3b). The distal end of the cardo is expanded into 2 processes, the inner cardinal process and the outer cardinal process (Figs. 2c, d; 3). The inner process curves mesad toward the mentum, perpendicular to it; the outer process curves outward toward the outer margin of the stipes. Generally one eighth to one half as long as the inner process, the outer one occasionally extends beyond the base of the stipes, forming, with the inner process, a bifurcate structure which rests upon the basistipital process (Fig. 3a) (19). The two cardinal processes connect the la- bium and the maxilla (Figs. 2c, d). The inner cardinal process articulates with the lorum, which is contiguous with the pos- terior edge of the inner process, and with the basistipital process which it overlaps. The outer cardinal process provides additional support for the cardinal-stipital articulation. In the Bombini, there is a heavily sclero- tized tranverse ridge where the cardo divides into the two processes; this ridge may strengthen this area (Fig. 3e) (42). In Apis, there is a bulla on the inner cardinal process (Fig. 3f) (40). The stipes (Figs. 2a, b; 4), a boat-shaped sclerite, extends distally along the sides of the prementum, articulating with the labium only through the cardo to the lorum, but connecting to the prementum by membrane basally. It is sclerotized on all but the inner anterior surface, forming a cavity which is closed by membranes and forms the channel in which blood and soft tissues reach the apical part of the maxilla. The stipes is two to five times as long as wide, often with a comb along the distal part of the posterior margin. The shape of the outer surface of the stipes varies considerably; some repre- sentative Outer views are shown in Figure 4. The base is usually narrowed from both the anterior and posterior margins, although some genera have an expanded, antero- proximal, sclerotized flap (Fig. 4a). The apical end may be blunt (Fig. 4a), narrowed (Figs. 4c, e, f), or notched (Fig. 4g). Many genera have a sclerotized ridge medially along the outer surface of the apical third (Fig. 4j) (32). The posterior margin of the stipes is often hairy, particularly proximally. The hairs vary in length (short, medium, long), abundance (absent, scarce, abundant, dense), and type (plumose, non-plumose, bristles) (Fig. 4). Occasionally, the anterior margin of the stipes may also be hairy, or even the entire outer stipital surface. Most anthophorids and apids, and some megachilids (Anthidium, Callanthidium, and Immanthidium), have a stipital comb (Figs. 2a; 4) (5) along a well-sclerotized concave edge of the posterior distal margin of the stipes, the comb concavity. This concavity varies from weak (Fig. 4f) to strong (Fig. 4h), and in some species of meliponines is recessed behind the outer margin of the stipes (Fig. 41). The bristles of the comb are generally robust, but some genera have weaker combs, with bristles attenuated dis- tally and wide gaps between them. In Xy/lo- copa and Lestis the bristles form extremely strong, blunt teeth (Fig. 4h) (36). Generally, stronger concavities contain stronger bristles. Some of the long-tongued bees without combs retain the comb concavity, occasion- ally with hairs in place of the comb. The stipital comb functions in cleaning and _pol- len manipulation (Schremmer, 1972; Jander, 1976). The basistipital process (Figs. 2b, c, d; 5) is at the proximal end of the stipes. De- spite its importance in the cardinal-stipital articulation, it has been neglected in the literature; only Iuga (1968) mentions it as the “apophyse cardinale.” I have renamed it since it is an extension of the stipes. It is formed by merged proximal extensions of the base of the outer margin and the sclero- tized inner anterior edge of the stipes, the basistipital thickening. The basistipital proc- 638 THe UNIVERSITY OF KANSAS SCIENCE BULLETIN : b ~—__basistipital : process basistipital _ thickening _ stipital thickening Stipes lacinia stipital sclerite subgalea— | > comb dististipital process palpus basigaleal area galea ( \—\—_ midrib 0 vane prementum \ suspensory thickening anterior longitudinal /brace N basiglossal A /} sclerite ligular arm subligular process / paraglossa! passer | ie ae ie = — palous—= are glossa od pista eee Fic. 2: Representative views of generalized labiomaxillary complex, showing nomenclature of structures. a) eae view of the maxilla, b) inner view of the maxilla, c) posterior view of the labium, d) anterior view of the labium. Progsoscis OF LONG-TONGUED BEES 639 ——— condyle oan : macula = inner process UP outer process ae e d — ; “6 j f Fic. 3: Cardines of selected genera. a) Neofidelia, b) Anthophora, c) Holcopasites, d) Caenonomada, e) Bombus, £) Apis. ess extends mesad and curves under the inner cardinal process, being mostiy obscured in an outer view of the stipes. In some genera it extends beyond the inner cardinal process, and abuts against the inner edge of the lorum. In many meliponines, the basistipital process is separated from the outer margin of the stipes, and is formed largely by the extension of the basistipital thickening (38). It is also expanded apically as a distinct sclerotized pad which abuts against the lo- rum (Fig. 5b). It is similar to this in the Bombini, but partly formed by the outer margin of the stipes. Membranes loosely con- nect the basistipital process and the inner cardinal process, allowing free longitudinal movement of the maxilla and, through the articulation between the cardo and the lo- rum, of the labium as well. Basally the inner surface of the stipes is narrower than the outer surface so that, in an inner view, the anterior part of the outer stipital wall can be seen (Fig. 2b). The an- terior edge of the inner surface is thickened basally, forming the well sclerotized basi- stipital and stipital thickenings (Figs. 2b; 5). The basistipital thickening forms and_re- inforces the posterior edge of the basistipital process. The stipital thickening extends from the distal end of the basistipital thickening to a point midway along the stipes and is one to three times as long as the basistipital thickening. In Thalestria and Triepeolus the two thickenings meet at an expanded sclerotized area (Fig. 5e) (22). Apically, the stipital thickening extends beyond the edge of the body of the stipes (Fig. 2b), abutting against the proximal end of the stipital sclerite and fusing with it in many genera (Fig. 5c, d). While the stipital thick- ening 1s usually straight or smoothly curved (Fig. 2b), it sometimes has two sections (referred to here as bipartite), as in the Xylocopinae and some Apidae (Fig. 5c) (37). The sclerotization of the stipital thickening probably strengthens the connection between the stipes and the stipital sclerite. The stipital sclerite (Figs. 2b; 6) is a slender sclerite on the inner side of the maxilla close to the inner edge of the stipes, extending from the basigaleal area to the distal end of the stipital thickening. Usu- ally curved, it is separable from the stipes in all but Xylocopa brasilianorum and X. fim- briata, in which it is well-attached to the inner anterior edge of the stipes and over- lapped by membranes (Fig. 38). The sus- pensory thickening is connected by mem- branes to the proximal end of the stipital sclerite and links the prementum and max- illa. In many genera, either the apical end or proximal end of the stipital sclerite, or both, are expanded as triangular or rounded Q o o \ NS e Ww 9) 4 tH, Fic. 4: Stipites of selected genera. a) Eucera, b) Chalicodoma, c) Stelis, d) Anthophora, ¢) Mesocheira, £) Tetrapedia, ¢) Neofidelia, h) Xylocopa, 1) Melipona, 1) Exomalopsis, k) Anthidrellum. 640 THe University oF KANsAs ScrENCE BULLETIN processes (Fig. 6) (33, 35). An oval mem- branous area connects the subgalea, stipital sclerite, and the lacinia; in some anthophor- ids, it is partly sclerotized (27). Snodgrass (1956) calls the stipital sclerite the extensory rod, and it presumably is involved in move- ments of the lacinia and the galea. It has also been called the subgaleal sclerite by Winston and Michener (1977), who thought it to be derived from the inner edge of the subgalea, but L. Greenberg (in prep.) shows it to be fused with the stipes in sphecoid wasps and many short-ttongued bees, sug- gesting a derivation from the inner distal margin of the stipes. All megachilids except Dioxys have a dististipital process (Figs. 2b; 7) (12) per- pendicular to the distal end of the stipes, extending anteriorly. It is a short distal bulge in some genera (Fig. 7a); in others it extends across the galeal-subgaleal junction toward the anterior edge of the galea (Fig. 7b). Its function is not clear. Since the galeal-subgaleal junction rests upon it, it a b __ basistipital \ process ~—_ T~. | “basistipital = thickening | = stipital eee, | thickening f Os // — cw \N Fic. 5: The basistipital process and stipital and basistipital thickenings of selected genera. a) Dioxys, b) Melipona, c) Xylocopa, d) Nomada, e) Triepeolus. may help to move the galea, perhaps as a rod against which the galea can be pulled into the folded resting position. The lacinia (Figs. 2a, b; 8) is a partly sclerotized or sometimes membranous lobe a f es ae b es 7 h Cc St ees d i Fic. 6: Stipital sclerites of selected genera. a) Megachile, b) Euplusia, c) Apis, d) Ceratina, e) Tetrapedia, £) Diadasia, g) Anthophora, h) Bombus, 1) Melipona. midway along the anterior edge of the stipes, basal and mesal to the subgalea. Membranes connect its base to the base of the stipital sclerite and to the suspensory thickening, which passes immediately basal to the la- cinia. Its anterior edge is usually well- sclerotized (Figs. 8a, b), with sclerotization sometimes extending posteriorly along the distal edge as well (Figs. 8c, d). The re- gions supporting the sclerotized edges of the lacinia are membranous; in Apis, the entire lacinia is membranous (Fig. 8e) (41). In most anthidiines and in Coelioxys, there is a lacinial comb along the distal (and some- times anterior) edge, made up of straight, relatively robust bristles (Figs. 8c, d) (15). In other genera the anterior sclerotized areas of the laciniae are unusually hairy, the hairs ranging from sparse (Fig. 8f) to abundant (Fig. 8a). Some genera (such as Apis, Fig. 8e) lack all lacinial hairs (41). The maxillary palpus (Figs. 2a, b) of 1-6 segments arises from a membranous area immediately distal to the apex of the stipes. The basal segment is generally broader than the distal ones. The palpus is often hairy, occasionally with bristles. There is a brush on the third segment in Melitoma and Dia- dasia (Figs. 11b; 43) (21). The galea (Figs. 2a, b) is a long, thin, tapering blade, convex on the outer surface and concave on the inner, posterior surface. It arises from the distal end of the stipes, Prososcis oF Lonc-roNcuep BEES 641 process Fic. 7: Dististipital process of selected genera. a) Lithurge,b) Hypanthidium. Fic. 8: Laciniae of selected genera. a) Xylocopa, b) Stelis, c) Hypanthidium, d) Coelioxys, ¢) Apis, f) Osmia, g) Hoplitis. and is divided into two regions, the post- palpal blade and the much shorter, pre- palpal, triangular subgalea (Fig. 2a). Be- tween the galeal blade and the subgalea, where the galea bends backwards in repose, the galea is narrowed. This narrowed region is strengthened on the inner, concave surface by the dasigaleal area (Figs. 2b; 9), a region of heavier sclerotization generally extending along the basal edge of the blade, more or less transverse to the main axis of the galea. A prominent midrib extends the length of the blade (9), as a fold in the inner galeal wall, often supported basally by the anterior edge of the basigaleal area. Hairs often arise from the midrib, sometimes extending to the edge of the galea. The blade is well- sclerotized basally, often less so distally. The distal area of lighter sclerotization often ap- pears ribbed, probably due to channels through the sclerotic material that connect hairs on the edges of the galea to the region of the midrib. Labium The labium of long-tongued bees can be divided transversely into three regions, the postmentum, prementum, and ligula (glossa, paraglossa, and labial palpus) (Fig. 1). The prementum is between the stipites; the glossa, paraglossa, labial palpus, and asso- ciated sclerites are articulated at its apex. The sclerites of the postmentum (lorum and mentum) connect the base of the prementum to the maxillae. Michener (1944) noted mis- interpretations which confused the mentum with the submentum, and the prementum with the mentum. As there is either one or no postmental plate in other Hymenoptera (Kirkmayer, 1909; Duncan, 1939), the lorum may be a secondarily derived structure not homologous with the primitive insect submentum. I use the term mentum to des- ignate the distal sclerite of the postmentum. The proximal sclerite of the postmentum, the lorum (submentum of some authors) is v-shaped, with its divergent arms articulated to the distal ends of the maxillary cardines as previously described (Figs. 2c, d). Its medial region articulates with the distal sclerite of the postmentum, the mentum, the proximal end of which curves over the lorum (Figs. 2c, d). The mentum is elon- gate, thin, and flared distally where it con- nects with the prementum. The distal mar- gin of the apical expansion of the mentum may be slightly concave (Fig. 10a), concave (Fig. 10b), bifurcated (Fig. 10c), notched (Fig. 10d), or reduced (Fig. 10e), and ar- ticulates with the base of the prementum (Figs. 2c; 10). The connections of the lorum to the maxillary cardines and the prementum through the mentum, allow the labiomaxil- lary complex to be protracted and retracted as a single unit. At least in Apis, the pro- tractor muscles insert on the maxillae, the retractor muscles on the labium, so that movements of the maxillae and the labium 642 THe UNIVERSITY OF KANSAS SCIENCE BULLETIN are completely interdependent (Snodgrass, 1956). The labiomaxillary complex is strength- ened basally by the suspensory thickenings (Figs. 2a, d), a pair of ribbon-like bands in the anterior conjunctiva of the proboscis (the anterior conjunctival thickenings of Mich- ener, 1944). The distal end of each thick- ening connects to the anterior surface of the prementum near the base. From there it extends to the inner edge of the lacinia, then curves anteriorly, supporting the conjunctiva, passing lateral to the mouth before turning toward the paramandibular process of the hypostoma (Michener, 1944). The scleroti- zation of the suspensory thickening is often expanded where it curves anteriorly; this expanded area may represent the fusion of the two segments of the suspensory thicken- ing present in short-tongued bees (except Melittidae, R. McGinley, pers. comm.). Membrane connects this area to the lacinial- stipital junction, further linking the labium with the maxillae. ~) The prementum (Figs. 2c, d) is an elon- gated sclerite, usually slightly wider distally than proximally, located between the two stipites. It is convex posteriorly and concave anteriorly, the concavity being closed by membrane, continuous with the labiomaxil- lary tube, and containing the muscles of the glossa and paraglossae (Michener, 1944). The articulation with the mentum is by means of the basal process of the prementum (Fig. 10), a usually concave expansion of the base of the prementum. In some genera the base is convex (Fig. 10c) (18) or re- duced (Fig. 10e). Distally, the posterior premental surface is trilobed, the outer lobes contiguous with the labial palpi, the central lobe forming the subligular process. In Canephorula, the prementum is partly mem- branous (Fig. 12g) (28). The labial palpus (Figs. 2c, d) articulates with the outer apical lobe of the prementum through the largely membranous palpiger (Fig. 2c), which sometimes is strengthened by a narrow longitudinal sclerotic slip. The lO b \" q C \ area Fic. 9: Basigaleal area of selected genera. a) Heteranthidium,b) Triepeolus, c) Melecta. Fic. 10: Menta of selected genera, showing variation in distal end. a) Eucera, slightly concave, b) Dioxys, concave, c) Nomada, bifurcate, d) Lithurge, notched, e) Exomalopsis, reduced. Progposcis OF LONG-TONGUED BEES 643 Fic. 11: a) Brush on first labial palpal segment of Lithurge, b) Brush on third maxillary palpal segment of Diadasia. palpus consists of four segments separated by membranes. The two basal segments are elongate, flattened, and concave on the inner surfaces so as to sheath the glossa. These two segments are well sclerotized medially, with lighter sclerotization along the lateral margins. The relative lengths of the basal segments vary. The two distal subcylindrical segments arise subapically on the second segment, and project almost perpendicularly to it. The labial palpus is often hairy, some- times bristly. The Lithurginae have a brush on a concavity of the proximal inner edge of the first segment (Fig. lla) (13). There is a small brush on an expanded, sclerotized lobe of the palpiger in Melecta and Thyreus (Fig. 33) (29). ake N »b The base of the glossa is supported pos- teriorly by the subligular process (Figs. 2c; 12), which extends distally from the apex of the prementum, curving anteriorly at its apex, perpendicular to the glossa (Fig. 13) (4). In the Anthophora group, it extends to form a u-shaped process upon which the glossa rests (Fig. 12f) (24). In a few gen- era the subligular process is separated from the apex of the prementum by a narrow membranous area (Fig. 12a) (20). Fig. 12 shows representative configurations of the subligular process. On the anterior surface of the premen- tum, the two /igular arms (Figs. 2d; 14) are located lateral to the base of the glossa. Each is a narrow sclerite, slightly expanded api- cally, extending from midway along the pre- mentum almost to its apex, except in the Lithurginae, where the ligular arm extends nearly to the base of the prementum (Fig. 14b). Each incurved lateral margin of the prementum (or premental fold) has a region of expanded sclerotization at the base of the ligular arm; in the Apidae, Anthophoridae, and Fideliidae, the base of the ligular arm merges with this sclerotized area (Fig. 14c) (16). In the Megachilidae, the ligular thick- ening is not continuous basally with the sides of the prementum, but is the concave anterior surface of the prementum, con- nected to the sides of the prementum by membranes (Fig. 14a). When the glossa is retracted, its base rests between the ligular arms. When protracted, the base of the Vr aS Fic. 12: Subligular processes of selected genera. a) Neofidelia, b) Svastra, c) Tetrapedia, d) Holcopasites, e) Exomalopsis, £) Anthophora (with lateral view), g) Canephorula (whole prementum). 644 THe UNIversiry oF KANsAs SCIENCE BULLETIN paraglossal suspensorium basiglossal prementum, sclerite X a eee \ ; longitudinal subligular brace process flabellum paraglossa Fic. 13: Lateral view of generalized ligular region and glossa. glossa extends beyond the apices of the lig- ular arms. In Megachilidae the ligular arms can move slightly in the same direction as the glossa; this mobility may increase the distance that the glossa can be protracted. The glossa (Figs. 2c, d; 13), arises at the apex of the prementum, as a fusion product of the primitive, paired glossae (Snodgrass, 1956; Michener, 1944). It is usually slightly longer than the prementum, and densely hairy, split posteriorly by a longitudinal me- dial groove, the salivary channel (Snod- grass, 1956). A flexible rod extends the length of the inner wall of the salivary chan- nel (7), although it is often only apparent in a cross-section. Transverse rows of setae alternate with bare areas, giving the glossa a ringed appearance. At its apex the glossa is expanded into the flabellum (Fig. 2c) (6). (Since a systematic study of glossal cross- sections was not done, variation in the sali- l4 he Fic. 14: Ligular arms of selected genera, showing three major types. a) Stel/s (short, distinct from prementum), b) Lithurge (elongate, not fused with prementum), c) Apzs (short, fused with prementum). vary channel, rod, and flabellum is not a part of this study). In the euglossines, the glossa is greatly elongated (as are the labial palpi and the galea), sometimes extending well beyond the tip of the abdomen (43). The basiglossal sclerite (bonnet-shaped sclerite, Snodgrass, 1956; notal and basiglos- sal sclerites, Iuga, 1968) (Figs. 2d; 15) partly encloses the base of the glossa antero-later- ally. Laterally, it forms two thin processes extending posteriorly that appear like the tie strings of a bonnet. Lateral to the basi- glossal sclerite, on the inner side of the paraglossal suspensorium, are two short scler- ites, the anterior longitudinal braces (Fig. 13), present in many but not all genera. Lateral to the glossa are the two para- glossae, elongate lobes each arising on a paraglossal suspensorium at the base of the glossa (Figs. 2c; 13; 16). The paraglossa varies from mostly sclerotized to mostly membranous, commonly largely membran- ous, often concave mesally and fitting snugly against the posterior glossal surface. Usually less than one quarter the length of the glossa, in some genera (Eucerini, Melectini, Cane- phorula) paraglossae extend the length of the glossa (23, 25). They are occasionally hairy. The paraglossal suspensorium (basipara- glossa of Iuga, 1968), a sclerotized base for the paraglossa lateral to the base of the glossa, has a posteriorly directed arm upon which the paraglossa arises. In the Xylo- copinae, Apidae, and some Exomalopsini, Progoscis oF LoNG-TONGUED BEES 645 the paraglossa is broadly attached to the paraglossal suspensorium (Fig. 16a), while in other genera the articulation is narrow (Fig. 16b). Snodgrass (1956) considered the paraglossa to arise from an apical ex- tension of the ligular arm, thinking that the paraglossal suspensorium was part of the ligular arm. However, when the proboscis is protracted, the paraglossal suspensorium and the paraglossa move with the glossa, while the ligular arm remains stationary; thus the paraglossal suspensorium is clearly a sepa- rate sclerite. At rest, the base of the glossa, the paraglossae, and the paraglossal suspen- soria lie between the ligular arms. Movement of the Labiomaxillary Complex The protraction and retraction of the labiomaxillary complex has been described for Anthophora edwards (Michener, 1944) and Apis mellifera (Snodgrass, 1956). When at rest, the proboscis is folded below the head, in three sections, in a z-shaped pattern. The basal section, containing cardines, lo- rum, and mentum, is directed posteriorly, and articulates with the head through the cardinal condyles. The midsection, made up of the stipites and prementum, folds an- terior beneath the cardines. The third sec- tion (glossa, paraglossae, labial palpi, and galeae) rests beneath the stipites and pre- mentum and folds posteriorly towards the neck. As mentioned above, retraction and _pro- traction in Apis depends on maxillary pro- IS b ee y d Ga sau. 671 JPN Yaya) 0a) oy. TRSES YY OSs a A ee = TE AE ARID ecto see 0/2 Effects of ACEP on’ Photophobic Responsé <22..:....2:-.-2-.2+--0ccses--ssneces0a- 4, cits SE ss eee eee OT 1D) TS CUISS1O Nee ee ee oN tec re eae 2 caes uci sola Aaa eee eee ee : 675 LEIRERA TURE Clik Derwent ans ies ee oe. 8 SS ey ae hae od as bee eee ee ee OM ABSTRACT The potassium salt of adenosine triphosphate (KsATP) affects both swimming rate and photophobic re- sponse of Euglena gracilis. Concentrations more than 1 X 10 molar depressed and stopped swimming rate; 1 xX 10% M and 1 X 10° M accelerated it, the latter to more than 200% more than normal rate. Pulsing rate of the flagellum (beats begun per second) as evidenced by high speed cinematography (125 and 200 frames per second) showed stoppage in 15 min in 1 X 10° M ATP; decrease to about 50%, of normal, continued through 30 min in 1 X 10° M ATP; but increase to 80% above normal, rising to 145° above normal in 30 min in 1 x 107* M; and increasing at once to 98% above normal and rising to more than 200% above normal in 30 min in 1 X 10° M. Euglenas untreated with ATP do not react on entering a brighter spot of light, but swerve to a new path on leaving it. Euglenas treated with 1 X 10°* M ATP vio- lently whirl the flagellum, pivoting rapidly on entering the brighter spot of light, then contract violently and become temporarily immobile on leaving it. These results confirm that available ATP regulates the rates of flagellar beating and swimming and, also, that Euglena has two responses to light, 1.¢e., to a sudden increase and to a sudden decrease above or below a critical threshold of intensity. chemicals on its growth and metabolism (see 2 volumes, Buetow, 1968). E. gra- INTRODUCTION Little research has been devoted to the effects of chemicals on swimming and phototaxis of Euglena, despite a plethora of research on its biochemistry and on the effects of pharmacological and other * Supported in part by NSF Grant GB-16616 and Univer- sity of Kansas General Research Grant 3£3590-5038. 1 Based in part on M.A. Thesis, University of Kansas. cilis grown at pH 6.8 swam most rapidly at pH 5.8 and nearly as fast at pH 7.0 (Lee, 1954) and tends to swim toward a region of acid pH (Bowne & Bowne, Present address: Departamento Biologia, Universidad de Oriente, Cumana, Venezuela. 2 Department of Physiology and Cell Biology, University of Kansas, Lawrence, Kansas, 66045, U.S.A.; to whom re- print requests should be sent. 670 THe UNiIversiry oF KANsAs ScIENCE BULLETIN 1967) or to congregate where the pH is near 6.0 (Borgers & Kitching, 1956). Some chemicals which, when present in certain concentrations, inhibit or stop swimming of Euglena include NaCle and CaCle (Schroder, 1927), acetate (Danforth, 1953; Bates & Hurlbert, 1970), glucose (Hofler & Hofler, 1952), heavy metal salts (Jiro- vec, 1935), arsenic (Rubinsky & Zrynkina, 1935), deuterium (D2O) (Mandeville, et al, 1964), the 5-isomer of parathion (Lazaroff, 1968), antibiotics (Goodwin, 1951), dodecyl sodium sulfate (Galdiero & Rossano, 1966), or excesses of gasses, eve. Ny CO2; or Op (Kostir, 1952): “A. lack of Mg** in the medium, or of certain trace-cations, causes immobilization of Euglena gracilis (Wolken, 1967). Certain specific metabolic inhibitors also reduce or stop swimming, e.g., 2,4-dinitrophenol below pH 635 (Mikolajczyk, 1969), iodo- acetate (Danforth & Erve, 1964), urethan and sulfonomid singly or together (Jira & Ottova, 1950), dichlorphenol, indophe- nol, or salicylaldoxime (Diehn & Tollin, 1967). Reported inhibitors of the phototactic response of Euglena gracilis include deu- terium (Pittendrigh, 1960), and various metabolic inhibitors (Diehn & Tollin, 1967). Anything that reduces motion also reduces phototactic response (Jahn & Bo- vee, 1968). Adenosine triphosphate has been indi- cated as the principal energy reservoir for both swimming and_ phototactic move- ments (Diehn & Tollin, 1967; Wolken, 1967; Jahn & Bovee, 1968) and _ glyceri- nated flagella of Euglena gracilis are re- activated and swim in ATP solutions (Mahenda, et al., 1967). Diehn (1969a, 1970) suggests that available ATP levels may alter the critical levels of light-energy needed to cause the motile response by Euglena. In ‘ ‘normal” circumstances (1.., standard culture medium) a change of intensity from 0.2 kW/m? to 0.1 mW/m? produces a positive phototaxis (Diehn, 1969b). A change from 0.2 kW/m* to 05 kW/m* causes a photophobic response. (Diehn, et al., 1975). Alignment perpen- dicular to a plane of polarized light oc- curs at ~ 0.2 kW/m? (Creutz and Diehn, 1976). However, except for our preliminary report (Bovee, et al., 1969; Bovee & Acuna, 1970), no one else appears to have reported the effects of ATP in solution in the surrounding medium on the swim- ming rate of Euglena, or on its responses to light. MATERIALS AND: METEODS A green strain of Euglena gracilis obtained from the Carolina Biological Supply Co. was grown axenically in Chalkley’s solution (Chalkley, 1930) with a few drops of added proteose-peptone solution (Neff, 1959). Before observing normal swimming, one ml of Euglena culture was mixed in 5 ml of Chalkley’s solution and_ buf- fered with KOH to pH 7.0. The euglenas were allowed to adapt for 30 minutes. Swimming rate was then timed electrically to nearest 1/10 sec over a linear course (the length of a calibrated ocular micrometer scale) while observed mciroscopically at 20% and rate was calculated as m/sec. Three timings were made for each of 12 organisms and averaged normal rate determined. The already tested sample was then mixed with K,ATP dissolved in Chalkley’s solution and_ buffered at pH 7.0 with KOH so that the resulting ATP concentration was either 1 X 10-2 M, 1 X 10-3 M, 1 X 10-4 M. Swim- ming was timed after exposures to ATP for 1, 10, 20, or 30) min, and’ at ‘each of /1,. 25.3; 42.95;) 6 and) 24 hrs, 7A least 3 rate-calculations were made for each of 12 organisms at each interval of exposure to ATP, and average rates determined. To make a permanent record of locomotory and_ photo- phobic responses, motion pictures were taken through a Zeiss- Nomarski phase-contrast-interference microscope at 160%, with illumination by a 60-watt incandescent lamp built into the base of the microscope. Light-intensity was controlled via a Zeiss Regel transformer. The motion pic- tures were taken with a Locam 16-mm_ high-speed camera (Model 164-4, Red Lake Labs) mounted at the photo tube of the microscope. Cinematographic speed was 125 pictures/ sec for normal swimming. and 200 p/s for ATP-influenced swimming. Five-second sequences of film (1,000 frames each) were cinephotographed of Euglena immediately after immersion in ATP solution, and after 1, 5, 10, 15, and 30 minutes of immersion. Eastman Kodak tri-X 449 reversal film (type 7278 in 100 ft rolls) was used, then professionally developed. The film was projected and analyzed with an electronically controlled Bell and Howell reversible 16-mm projector adapted for variable speed (1-24 p/s) and_ stop-motion analysis, equipped with a reversible frame counter (‘‘Selectra-frame’’ Mcdel 16N by TRAID Corporation, Glendale, California). Drawings were made by tracing with india ink on clear acetate sheeting taped on glass over a film editor, model S.0.S. (Photo-Cine-Optics, Inc.). Enlargements of tracings were made by proportional transfer to ruled graph paper. Photographic prints were made from the cinefilm with standard photographic enlarger on high contrast photographic paper, using standard photographic darkroom techniques. ATP, SWIMMING AND PHoropHosia oF Englena 671 OBSERVATIONS AND RESULTS ForwaRD SWIMMING Normal rate of swimming for Euglena gracilis in Chalkley’s solution at pH 7.0 ranged from 139 to 145 wm/sec, with an average normal rate of 142 m/sec. Effects of ATP on Forward Swimming The first response of Euglena to addi- tion of the KeATP solutions is the classic “shock reaction” (Jennings, 1906), normal forward swimming being replaced for a few seconds by a pivoting movement. In KeoATP at 1 X 10? M, swimming stopped within 15 min and _ euglenoid movement stopped a few seconds later. No movement was observed for 12 hrs, but normal movement was recovered over- night, sometime after 12 hrs exposure and within 24 hrs in the solution. In 1 X 10° M ATP more than 90% of the euglenas slowed 50°% in forward progress almost immediately, contracted to globular form after 6 hrs and remained inactive for 12 hrs. Sometime within the next 24 hrs, normal motility and swimming rate were rezained) Inv l ><10*.M. ATP a brief, pivoting, shock-reaction was followed by an increase in locomotor rate within 10 min to 13% above normal (160 »m/sec) and then declined almost to normal after 30 min in ATP (144 »m/sec), then again increased within one hr of exposure to 38°, faster than normal (196 p»m/sec) and continued to accelerate during a 24 hr period, being then 99°% faster than nor- mal (282 um/sec). In 1 X 10° M ATP the swimming rate rapidly increased with- in the second and third hrs of exposure to 80°% faster than normal (259 »m/sec), de- clined in rate within a total of 6 hrs ex- posure to 51.3°% faster than normal, and within a total of 24 hrs exposure returned to normal or slightly subnormal swim- ming rate. These data are summarized in Figs. 1 & 2. Effects of ATP on the Pulsing Rate of the Flagellum Analysis of motion pictures taken of normal swimming at 125 f/s and of ATP- influenced swimming at 200 £/s showed acili Euglena er 60 Fic. 1. Effect of 1 X 10-2 M ATP on swimming rate of Euglena gracilis. Se Fic. 2. Effects of 1 X 10-3 M, 1 X 10-* M and l 10-> M ATP on the swimming rate of Euglena gracilis. 672 Tue UNtversiry oF Kansas ScrENCE BULLETIN that the number of helical-waves/sec pro- gressing along the flagellum was altered by exposure to ATP in the surrounding solution. Normal pulsing rate of the flagellum of E. gracilis under the conditions of cine- photography varied from 20 to 22 pulses/ sec, ie., one each 45-50 msec, nearly twice the pulsing similarly recorded for E. w- ridis (Lowndes, 1944; Holwill, 1965). In 1 X 10? M KeATP, flagellar beat- ing was erratic and the number of waves/ sec was much less than normal, failing to propel the body at all after 20 min ex- posure. In 1 XK 10% M ATP, flagellar pulsing was less erratic and the number of waves/sec was 40-50°% fewer than nor- mal (9-11/sec). However, in 1 X 10* M and 1 X 10° M flagellar pulsing was steady and the number of waves/sec in- creaséa am We xX 10" “M ATP’ irom. the average of 20/sec (one each 45-50 msec) to a maximum of at least 65/sec, i.e., one each 14-15 msec, a trebling of the pulsing rate in 30 min. The increase in frequency of wave-origin was coincident with simi- larly greater rate of progress of the helical waves along the flagellum, and not with an increase in the number of waves seen serially progressing along the flagellum at any one time. As in normal swimming, there were no more than 2 waves in prog- ress at any one time. The increase in pulsing rate was al- most immediate, in 1 XK 10° M KsATP rising to about 47/sec (one each 20-22 msec) in less than one minute, a more than doubling of rate. The rate of puls- ing continued to increase, with slight fluc- tuation, to the maximal rate (65/sec) after 30 min eposure to 1 X 10° M KeATP (Fig. 3). PHOTOPHOBIC RESPONSE Euglenas observed directly by phase- contrast microscopy adapted quickly in Chalkley’s solution to the light-intensity FLAGELLAR BEATS/SEC, Euglena gracilis. TIME EXPOSED TO KyjATP/MIN Fic. 3. Effects of several concentrations of ATP on the number of flagellar beats per second by Euglena gracilis. of the microscopic field, swimming through it with no photophobic reaction. However, when the phase-condenser was adjusted so that a slightly, visibly-brighter spot of light 15 to 35 »m diameter was concentrated in the center of the field, the euglenas swerved at an angle of 45° to 60° to right or left of the linear path of swimming on going out of the spot of light. The reaction was an off-response occurring as the anterior end of the eu- glena left the brighter area, and was com- pleted in 0.65 sec (Fig. 4). Each euglena immediately resumed normal swimming along the new line of direction while still in the generally lighted microscopical field. Effects of ATP on Photophobic Response In KeATP solution 1 X 10% M, the same euglenas earlier observed in Chalk- ley’s solution, alone, showed in less than one minute an intense photophobic re- sponse to the same spot of light, although they swam normally in the generally lighted area of the microscopical field. ATP, SWIMMING AND PHOTOPHOBIA OF Euglena 673 UNTREATED WITH ATP SWERVES TO A NEW PATH ON LEAVING AREA OF BRIGHTER LIGHT NO REACTION ON ENTERING AREA OF BRIGHTER LIGHT Fic. 4. Path of Euglena gracilis untreated with ATP on entering and leaving a brighter spot of light. Traced from a motion picture film photographed at 125 frames per second. Each 15th picture in a 200 frame sequence is shown. This response was initially an on-response, beginning before the anterior end of the body had passed through the brighter spot. The flagellum was thrown to the side or forward with undulations increas- ing in amplitude, rate, and thrust so that a vortex of water was created around the flagellum (Fig. 5). The vortex obscured the undulations so that the exact increase of pulsing rate could not then be calcu- lated. This rapid, anterior, flagellar pulsing caused the euglena to pivot rapidly on the fringe of the brighter spot, describing 3 complete turns in 0.35 sec (one complete turn per 117 msec). Within the follow- ing 0.5 sec the euglena, still pivoting, pulled itself forward out of the bright spot. It then contracted to nearly globu- lar form as an off-response on leaving the bright spot and forward swimming was suspended. The pivoting, contraction and cessation of forward swimming occurred within 1.025 sec (Fig. 6). After several minutes these contracted euglenas relaxed, extended and began normal locomotion which increased in rate until it exceeded the normal rate in Chalkley’s solution. This violent photophobic response oc- oT 4 Fic. 5. Eight consecutive pictures from a motion picture taken at 200 frames per second showing the initial reaction of Euglena gracilis, treated with 1 XX 10-4 M ATP, on entering a brighter spot of light. The vortex in the water created by rapid whirling of the flagellum at the upper end of the cell is clearly shown. curred at any time during the first half- hour of exposure to 1 & 10 M KeATP and whenever a euglena entered the brighter spot of light, as evidenced by cinephotography. It also was observed visually at any time during 24 hrs expo- sure to 1 K 10* M KeATP on increasing the light slightly by moving the rheostatic control of the Regel transformer. At- tempts to measure the increase of intensity required to elicit the reaction were incon- clusive due to inadequate sensitivity of equipment available. However, the requi- site change of intensity needed to cause a phototactic or photophobic response may be minimal (Diehn, 1969), so long as it exceeds a critical intensity. 674 THe UNIvERsITy or KANsAs SCIENCE BULLETIN TREATED WITH 1 x 1074 M ATP ) MOVEMENT AWAY FROM BRIGHTER LIGHT, WITH CONTRACTION C/ MOVEMENT OUT OF BRIGHTER LIGHT Me ae in Sos N . KO AREA OF = BRIGHTER LIGHT \ \ \ \ \ MOVEMENT INTO BRIGHTER LIGHT } a / ~— 4 ~~ , . i AS f y PIVOTING, 3 TURNS Fic. 6. Path of Euglena gracilis, treated with 1 XX 10-4 M ATP, on entering and leaving a spot of brighter light. A rapid pivoting occurs on entry and a rapid contraction of the cell on leaving the brighter spot. Traced from a motion picture sequence photographed at 200 frames per second, each picture in a sequence of 28 frames traced to show the reactions. ATP, SWIMMING AND PHOTOPHOBIA OF Euglena 67 DISCUSSION The above results indicate that not only is ATP the energy-source for fla- gellar movements, but that it is involved also in the photophobic response. The flagellum is the only organelle Euglena has which enables it to swim and _per- form the phototactic movements which occur when it alters direction of swim- ming (Jahn & Bovee, 1968). Hence, any circumstances which alter the availability of energy as ATP should and do alter both flagellar motility and beat frequency (Danforth, 1953; Brokaw, 1965; Diehn & Tollin, 1967) and therefore should and do alter phototactic movements (Diehn & Tollin, 1967). The swimming rate of Eu- glena gracilis is greatest at 40 ft. candles, which is the saturating level for the pho- tophosphorylating mechanisms which pro- duce ATP, and both ATP levels and swimming rates rise as the luminar inten- sity increases toward 40 ft. candles (Wol- ken, 1967). The helical undulations along the flagellum are intermittent (“inter- rupted helical waves”; Jahn & Bovee, 1968), probably dependent for origin on periodically initiated waves of chemome- chanical interactions (PM pulses; Thorn- burg, 1967) requiring ATP energy. Our results show that the energy from the extraneous ATP is taken up rapidly from solution and is used immediately and directly by the flagellum. ATP is a rela- tively large molecule. It may or may not be taken up as an intact molecule. It ap- pears perhaps likely that transphosphoro- lase enzymes at the membrane of the fla- gellum may detach the ~Ph and transfer it to nucleotides within the flagellar ma- trix adjacent to inner side of the mem- brane and associated with the fibrils of the axial cylinder. This assumption is similar to that of Siekevitz & Potter (1965) for ~Ph transfer from mitochondria to cyto- plasm without loss of mitochondrial nu- cleotides. Such a system of uptake would 5 explain the speed and intensity with which Euglena responds to the presence of ATP in the surrounding medium. The presence of an actomyosinoid chemomechanical motile machinery as the basis for all kinds of protoplasmic move- ments, including those of cilia and fla- gella, is well supported by a variety of evidence (see reviews by Kamiya, 1959; Gibbons, 1968; Brokaw, 1966: Holwill, 1966; Jahn & Bovee, 1967, 1968, 1969: and others). The flagellum of Euglena, like an actomyosin system, requires ATP as an energy source and both Mg and Ca ions are needed to maintain motility (Wolken, 1967). The detached, glycer- inated flagellum of Euglena swims freely in critical amounts of ATP and those ions (Mahenda, et al., 1967). The facts that available ATP up to a critical level accelerates the flagellar move- ments of Euglena and increases of ATP above that inhibit and stop flagellar move- ments resemble similar effects of ATP on other actomyosinoid systems of cells (Hoffman-Berling, 1960) and on muscle- actomyosin extracts (Weber, 1959). If the mechanochemical pulse which initiates flagellar undulations is dependent on critical cations and especially on ATP levels both in the cell and along the fla- gellum, as appears probable, then variation of ATP concentration should produce all the effects we have noted on swimming rate and photophobic response. Gossel (1957) showed that the parafla- gellar photoreceptor swelling of Euglena conveys the energy of sudden increases of luminar-intensity to the flagellum, al- tering its position and increasing the am- plitude of its undulations, producing the photophobic response. Diehn (1969b) found the normal energy quantum pro- ducing the photophobic response to be above an equilibrium of 2 X 10° ergs/ cm” sec. Our results show that available ATP at certain concentrations augments 676 Tue UNiversiry oF KAnsAs ScIENCE BULLETIN photophobic movements, the duration of capability for such responses, and appar- ently lowers the threshold for the shock reaction to the level of the photophobic response. Since the movements of the flagellum depend on critical concentrations of ca- tions as well as ATP levels (Wolken, 1967), ion fluxes, especially of Ca**, Mg**, K*, and H30*, are required concomitantly and in advance of the ATP-utilization se- quence which produces the undulatory movements of the flagellum. How the photoreceptor stores energy and releases it to the flagellum upon changes of luminar intensity is not known. Enzymes and photosensitive pig- ments in it have been suggested, but not yet found (Pringsheim, 1956; Wolken, 1969). Recent indirect evidence using po- tassium iodide (KI) as a specific inhibitor suggests that certain flavins are involved and may be resident, perhaps as crystals resembling amino-acid oxidase, in the paraflagellar photoreceptor (Diehn and Kint, 1970; Tollin & Robinson, 1969). A recent theory by Bovee and Jahn (1972) suggests that since the photore- ceptor of Euglena is crystalloid (electron- micrographs, Arnot & Walne, 1967; Lee- dale, 1967; polarized-light — sensitivity, Diehn, 1969b) it may be piezoelectric, i.e., generates a current when deformed, or is deformed as it generates a current. Its poised crystalloid flavin molecules, assum- ing it has them, may absorb light-energy, generate a current and discharge a poised oxidation-reduction system (Jahn, 1963). Diehn (1970) suggests that as little as 4 photons of difference in luminary inten- sity above or below the poised level may be enough to excite phototactic move- ments and reorientation of the flagellum and body. Therefore, the photoreceptor may be able to deliver discharge of cur- rent into the flagellum during deforma- tion of its crystalloid structure as light strikes it, i.e. it is a photoelectric body. The same theory, extended (Bovee & Jahn, 1972), assumes that flagellar fibrils are also quasi-crystalloid and are either piezoelectric or semiamorphous conduc- tors, along which piezoelectric currents (resulting from bending the fibrils) trans- locate cations to and from cardinal sites (Ling, 1962) where they are able to set up the ionic state required for ATP-split- ting. These flagellar pulses (PM _ pulses, Thornburg, 1967) originating at the fla- gellar kinetosome as a _mitochondrial- ATP-powered red-ox discharge, precede as an electronic and cationic wave the motile ATP-splitting undulation. Any energy released from the photoreceptor during reaction to increased luminar in- tensity would augment the ion fluxes, the ATP-splitting and the flagellar move- ments (Bovee & Jahn, loc. cit.). The availability of extra ATP-energy to load the energy reservoir to a “hair trigger” level in setting off the red-ox discharges at the flagellar kinetosome should increase the potential number of red-ox discharges/sec, and therefore the number of undulations/sec. The increased ATP-availability and utilization in the flagellum should augment the rate at which the undulations proceed along the flagellum, and depending on cation-asso- ciations, and dissociations, including H- bonds, the amplitude of the wave. The increased number and speed of undula- tions along the flagellum increases the thrust of the flagellum, and therefore the swimming rate/sec. In summary, our results support the contentions of Wolken (1967) that swim- ming rate is linked to available ATP lev- els and of Bound and Tollin (1967) that both motility and phototaxis are related to available ATP levels. There are two separate responses to light indicated: (1) A “light-on” response which is elicited as a photophobic response ATP, SWIMMING AND PHoropHosia oF Euglena 677 with violent whirling of the flagellum, occurring if light intensity is increased suddenly and sufficiently; (2) A “light- off” response, elicited as the Euglena leaves an area of increased light intensity that is too weak to elicit a photophobic response, resulting in a change of direc- tion; or, on leaving a field of intensity sufhcient to have caused the photophobic response, or with added ATP, causing contraction and temporary cessation of movement. These separate responses tend to sup- port Diehn’s contention (Diehn, 1972) that Euglena may have two photoreceptor mechanisms, one for light of lower inten- sities and another for light of higher in- tensities. HIP ERATOURE, CILED AcuNa, M. L. 1970. Some effects of adenosine tri- phosphate on the swimming of Euglena gra- cilis, M.S. Thesis, University of Kansas Li- brary, Lawrence, Ks. Bates, R. C. aND Hurvpert, R. E. 1970. The effect of acetate on Euglena gracilis var. bacillaris as a function of environmental conditions. J. Protozool., 17: 134-38. BorceErs, J. A. AND KiTcuinc, J. A. 1956. Reactions of the flagellate Astasta longa in gradients of dissolved CO: Proc. Roy. Soc. London, 144B: 507-19: Bowne, S. W., Jr., AND Bowne, G. D. 1967. Taxis in Euglena. Exp. Cell. Res., 47: 545-53. Brokaw, C. J. 1966. Mechanics and energetics of cilia. Ann. Rev. Resp. Dis., 93: 32-40. —. 1968. Mechanisms of sperm movement. Symp. Soc., Exp. Biol.,, 22: 101-129. Bruce, V. G. AND PITTENDRIGH, C. S. 1960. An ef- fect of heavy water on the phase and period of the circadian rhythm of Euglena. J. Cell. Comp. Physiol., 56: 25-31. Buetow, D. E., ed. 1968. The Biology of Euglena. Academic Press, N. Y., Vol. I. General Biol- ogy and Ultrastructure. Vol. II. Biochemis- try. CHaLkKLey, H. W. 1930. Stock cultures of amoe- bae. Science, 71: 442. Creutz, C. aNp DiEHN, B. 1976. Motor response to polarized light and gravity sensing in Eu- glena gracilis. J. Protozool., 23: 552-557. DanFortH, W. 1953. Oxidative metabolism of Ex- glena. Arch. Biochem. Biophys., 46: 164-173. DaNForTH, W. anv ErveE, P. 1964. Cited in refer- ence Jahn & Bovee (1968), below. DieEHN, B. 1969a. Action spectra of the phototacti responses in Englena. Acta, 177: 136-143. ——. 1969b. Phototactic response of Euglena to 3iochem. Bioph single and repetitive pulses of actinic light: Orientation time and mechanism. Exp. Cell Res), 56:°375-581. —. 1970. Mechanism and computer simulation of the phototactic accumulation of Euglena in a beam of light. Photochem. Photobiol... 11: 407-418. 1972. The receptor/effector system of pho- totaxis in Euglena. Acta Protozool. 11: 325- 332. DieEnn, B. anp Kint, B. 1970. The flavin nature of the photoreceptor molecule for phototaxis in Euglena. Physiol. Chem. Phys., 2: 483-488. Dienn, B. aND ToLiin, G. 1967. Phototaxis in Eu- glena. IV. Effects of inhibitors of oxidative and photophosphorylation in the rate of pho- totaxis. Arch. Biochem. Biophys., 121: 169- Wie DieEHN, B., Fonseca, J. R. AND JAHN, T. L. 1975. High-speed cinematography of the direct pho- tophobic response of Euglena and the mech- anism of negative phototaxis. J. Protozool., 22: 492-494. GALDIERO, F. aND Rossano, F. 1966. Effeto di una sostanza tensioattiva sull’ Euglena gracilis. Ist. Sieroterap. Ital. Riv., 41: 304-308. Gipzons, I. R. 1967. The organization of cilia and flagella. In: Allen, J. M., ed., Molecular Or- ganization and Biological Function. Harper and Row, New York. ——. 1968. Biochemistry of Motility. Ann. Rev. Biochem., 37: 521-46. Goopwin, C. M. 1951. A study of the effect of penicillin and dihydrostreptomycin on proto- zoa with special reference to the problems of bacteria-free culture and speciation in En- glena. Proc. Iowa Acad. Sci., 58: 451-67. HoFrMAN-Beruinc, H. 1962. Die Innere Mechanik der Geisselbewegung. In: Handbuch der Pflanzenphysiologie (W. Ruhland, ed.) Springer, Berlin., Vol. 17. HOFLer, K. anp H6Fier, L. 1952. Osmoseverhalten and Nekroseformen von Euglena. Proto- plasma, 41: 76-102. Horwitt, M. E. J. 1966. The motion of Euglena viridis: The role of flagella. J. Exp. Biol., 44: 579-88. 1967. Contractile mechanisms in cilia and flagella. Curr. Topics Bioenerg., 2: 287-333. JaHN, T. L. anp Bover, E. C. 1967. Motile be- havior of protozoa. pp. 41-200. In: Research in Protozoology, Vol. I (T. T. Chen, ed.) Pergamon Press, Oxford. and . 1968. Locomotive motile re- sponse in Euglena. pp. 45-108. In: Buetow, D. E., ed. The Biology of Euglena, Vol. 1. Academic Press, New York. Jenninos, H. S. 1906. Behavior of the Lower Or- 678 Tue UNIversiry oF KANSAS SCIENCE BULLETIN ganisms. 1962; Reprinted, Indiana University Press, Bloomington. 366 p. Jira, J. anv Orrova, L. 1950. Ovlivre ethylure- thanie na Protozoa. Biol. Listy, 31: 82-93. Jirovec, O. 1935. Predbezna zprava o ucincich nekterych kovu na kmeny Euglena gracilis. Cseck. Zool. Spolec (Vest.), 2: 101-16. Kamiya, N. 1959. Protoplasmic streaming. Proto- plasmatologia, 8 (3a): 1-199. Springer, Wien. Kostir, W. J. 1952. Factors which induce motility or passivity in Euglena. Proc. Soc. Proto- zool., 3: 5. LazarorF, N. 1968. Motility and flagellar behavior of algal flagella treated with phosphorothiote compounds. J. Physiol., 4 (S): 8. Ler, J. W. 1954. The effect of pH on forward swimming in Euglena and Chilomonas. Physiol. Zool., 27: 272-75. LeEepaLE, G. F. 1967. The Euglenoid Flagellates. Prentice-Hall, Englewood Cliffs, N.J. 242 pp. Linc, G. N. 1962. A Physical Theory of the Living State: The Association Induction Hypothesis. Blaisdell, N. Y. 680 pp. MaHENpRA, Z.; Vora, R.; WoLKEN, J. J. AND AHN, K. S. 1967. Biochemical studies of the fla- gella of Euglena gracilis. J. Protozool., 14 (S) 7: MANDEVILLE, S. L.; Cresp1, H. L. anp Katz, J. J. 1964. Fully deuterated Euglena gracilis. Sci- ence, 146: 769-70. MikoLajczyk, E. 1969. Reversible immobilization of metabolic movement of Euglena gracilis evoked by 2,4-dinitrophenol. J. Protozool., 116;:(S)23. PrincsHEIm, E. G. 1956. Contributions toward a monograph of the genus Euglena. Nova Acta Leopoldina N. F., 18: 1-168. Reyinsky, S. B. anp Zrynkina, L. M. 1935. Uber Kernveranderungen bei Euglena — gracilis (Ehrbg.) bei chronischer Arsenvergiftung. Arch. Protistenk., 85: 334-40. ScHropeErR, V. N. 1927. Influence of electrolytes up- on the difference of potential of the cell wall in Euglena ehrenberghi. Proc. 2nd Congr. Zool. Anat. USSR, 167-68. SIEKEVITZ, P. anD Porrer, V. R. 1965. Biochemi- cal structure of mitochondria. II. Radioactive labeling of intramitochondrial nucleotides dur- ing oxidative phosphorylation. Biol. Chem., 2153 237-5): Tuornsurc, W. 1967. Mechanisms of biological motility. In: A. Cole, ed., Theoretical and Experimental Biophysics Vol. I. Decker, N.Y. Totuin, G. 1969. Energy transduction in algal pho- totaxis. Curr. Topics Bioenerg., 3: 417-46. ToLiin, G. anv Rosinson, M. I. 1969. Phototaxis in Euglena. V. Photosuppression of photo- tactic activity by blue light. Photochem. Photobiol., 9: 411-16. Weser, H. H. 1955. The link between the metab- olism and motility of cells and muscles. Symp. Soc. Exp. Biol., 9: 271-81. WoLkEN, J. J. 1967. Euglena. 2nd ed. Appleton- Century-Crofts. N.Y. 1969. The photoreceptor molecule and the phototactic responses of Euglena. J. Cell Biol., 43: 159a. OS RN NAA ed SCIENCE BULLETIN ALLEVIATION OF THE TOXICITY OF COPPER TO TETRAHYMENA PYRIFORMIS BY NONTOXIC IRON ee 5x RR RK na By Eugene C. Bovee, David A. Kegley, David Sternshein, Ellen Wyttenbach and Barton L. Bergquist Vol. 51, No. 24, pp. 679-684 July 10, 1979 ANNOUNCEMENT The University of Kansas Science Bulletin (continuation of the Kansas Uni- versity Quarterly) is an outlet for scholarly scientific investigations carried out at the University of Kansas or by University faculty and students. Since its incep- tion, volumes of the Bulletin have been variously issued as single bound volumes, as two or three multi-paper parts or as series of individual papers. Issuance is at irregular intervals, with each volume prior to volume 50 approximately 1000 pages in length. Effective with volume 50, page size has been enlarged, reducing the length of each volume to about 750 pages. The supply of all volumes of the Kansas University Quarterly is now ex- hausted. However, most volumes of the University of Kansas Science Bulletin are still available and are offered, in exchange for similar publications, to learned societies, colleges and universities and other institutions, or may be purchased at $20.00 per volume. Where some of these volumes were issued in parts, individual parts are priced at the rate of 2 cents per page. Current policy, initiated with volume 46, is to issue individual papers as published. Such separata may be purchased individually at the rate of 3 cents per page, with a minimum charge of $1.00 per separate. Subscriptions for forthcoming volumes may be entered at the rate of $20.00 per volume. All communications regarding exchanges, sales and subscriptions should be addressed to the Excnance Liprarian, UNIVERSITY oF Kansas Lripraries, LAwrENCE, Kansas 66045. Reprints of individual papers for personal use by investigators are available gratis for most recent and many older issues of the Bulletin. Such requests should be directed to the author. The International Standard Serial Number of this publication is US ISSN 0022-8850. Editor William L. Bloom Editorial Board William L. Bloom Philip W. Hedrick Rudolf Jander Harvey Lillywhite Charles D. Michener Norman A. Slade Henry D. Stone George W. Byers, Chairman THE UNIVERSITY OF KANSAS SCIENCE BULLETIN Vol. 51, No. 24, pp. 679-684 July 10, 1979 Alleviation of the Toxicity of Copper to Tetrahymena pyriformis by Nontoxic Iron.’ * EucENE C. Boveg, Davin A. KecLey, Davin STERNSHEIN, ELLEN WyTTENBACH AND Barton L. Bercouist” TABLE OF CONTENTS ENESSIISRIA Cymer er ee ek es a a ee hod PS ls cco cece he Re ee eee eee 679 ENTER ODU GiB 1O Nipeees ses ee een ok AE ek ES eas 8 cnc icccnsdeezcsebegence cased See dem ean cce ee ee 679 INUACERTAU SIAN IDO NVUE TH ODS Gece. <2 20 eote. 28 2.c2 ots 2. ses Bie asst Sdawedec eva baeeeanscdecevacttacdecks ee eee a ee 680 REES Uc 1S eee a eae, ERE eID, 1a SO ehiasecccdactcstetceeneaeei le te eee ee see seot 7h SO BOB RPE CtSMOME COD DERM tac ae gee ND ee ee ee ee Oe eee 681 BAT CCESHOLMITO Teter ne ee a tea Ah ere A eee ee _ 681 Eitects sola Copper and wlron Pogethen® -222222222.2-.-<2ceddense<-nengcdsucsaceack poseucesseeese eee cast eed cee ee Ook BqualepAtmounts@mor ach cot ee eee ee see oe sa deeecendacteunccs Mase ceeaceoeee ee eee BoA 2. tots eee . 681 WinequalaAimounts Of: ah cc ce cd As :fcc0c. bee ca = eee, Sonus adosctbe cutest ce oos be cheve So chastesdeeees Sotenee peeesee eee eee een 682 IB) ISGUSS TONG ne ates uk ccd spgbi a ue dta dn duc cttcgatecatlenceete tancwosee ee ce coon 10 So oe nee oR Renee eka ea ne ee 673 ABSTRACT Copper (as CuCle) is toxic to and retards growth of Tetrahymena pyriformis at 3 to 5 parts per million in a proteose-peptone growth medium. Iron (as FeCl2* 4H2O) is stimulatory to growth in 3 to 15 ppm. Equal amounts of iron and copper together stimulate growth, most so at 15 ppm Cu with 15 ppm Fe. Un- equal amounts of them are also stimulatory, the most so being 15 ppm Cu with 10 ppm Fe. Copper apparently inhibits mitochondrial synthesis of adenosine triphosphate, acting on flavinoid pigments and cytochromes by preempting sites normally occupied thereon by iron. Iron, in turn, lifts the block by displacing the copper at those sites. Stimulation of growth by the combined copper and iron may be due to the role of copper in the enzyme systems that form the iron porphyrins of the cytochrome system and its presence in polyphenol oxi- dases and tyrosinase. INTRODUCTION biodegradation and the danger to larger animals, including man, that may be poi- soned by the metals as they are concen- trated by progression through the food- webs of nature. Toxicity of heavier metals to animals, protozoan to human, is well-documented (reviews: Bremer, 1974; Hartung, 1972; Louria, 1972). Most heavy metals are toxic, often lethal, to animals in amounts of 20 parts per million or less in water. * Department of Physiology & Cell Biology, University of : : Kansas, Lawrence, Kansas; to whom reprint requests should The presence of toxic heavy metals in eeren Es 1 Supported by Kansas Water Resources Institute Matching tenes Rcaloes concert) both because Grants A-058 and B-040 and, in part, by NSF Grant GB- of their immediate danger to the survival 16616. 0 A 5 F 2 Present Address: Department of Biology, University of of certain small organisms involved in Northern Iowa, Cedar Falls, Iowa 50613. 680 THe Universiry oF Kansas SCIENCE BULLETIN Copper is toxic to humans (Louria, loc. cit.), livestock (Bremer, loc. cit.), poultry and game birds (Henderson, 1975), fist (Christensen et “ali, 1972: Stephenson and Taylor, 1975), annelids (Hartung, 1972), crustacea (Chaisemartin, 1973), insects and their larvae (Hartung, 1972) and protozoa (Cairns and Dickson, 1970; Roth et al., 1968), and to phyto- plankton (Rice, 1973). The well-known toxicity of copper and other heavy metals to many fungi and bacteria is cited in most general textbooks of microbiology (e:eq) Salles 31948): Nonetheless, copper is sometimes a necessary trace element, required for blood-pigments of certain invertebrates (Elvehejm, 1935), some enzyme systems of vertebrates (Schulze, 1941) and inverte- brates, e.g., Tetrahymena pyriformis (Hill, 1972). It is also involved in the produc- tion of iron-porphyrins in vertebrate ani- mals (Schulze, 1941). While much effort has been directed toward determining toxicity of heavier metals, little has been done to determine if nontoxic chemicals normally present in aquatic organisms and their environments may alleviate or block the toxicity of the heavier metals. Some studies show that small amounts of one toxic metal may alleviate the effects of another one, e.g., the well-known copper-molybdenum an- tagonism (e.g., Suttle, 1974). A few stud- ies show that organic chelaters may reduce ions of heavier metals from water to non- toxic levels, e.g., chelation of copper by ethylene-diamine-tetracetate (EDTA) (Ste- phenson and Taylor, 1975). The subject of naturally occurring ions as counters to toxicity appears to be relatively uninvesti- gated, although it is generally conceded that “hard” waters can carry a greater burden of toxic metals than “soft” waters. We have tested the effects of iron (Fe**) as a counter-ion against the toxicity of copper (Cu**) for the growth of the ciliated protozoon, Tetrahymena pyrt- formis, to determine to what degree the iron alleviates the toxicity of the copper. MATERIALS AND METHODS Tetrahymena pyriformis (Clonally de- rived from Strain HSM) were grown axenically and synchronously dividing (Scherbaum and Zeuthen, 1955) at 29°C + (2° in 250 ml of proteose-peptone me- dium (Elliott and Hayes, 1955), lacking the yeast-extract fraction, in 500 ml screw- capped, sidearm Ehrlenmeyer flasks at 29°C = 0.2 degree. About 30,000 Tetra- hymena were used as each inoculum for control and experimental cultures. Counts of cultures were made by hemocytometer and Whipple cell methods to determine at what spectrophometric, light-transmit- tance reading a midway-logarithmic growth of 200,000 organisms/ml is indi- cated. For experiments, Tetrahymena were grown in media containing: (A) Cu" at 359, 10 "or ss ppma (BPS wl0 ror tS ppm Cu** with, also, equivalent Fe** (.e., 3-ppm_ Cu plus’ 3 ppmrke arctc)caS) Cu 1I>\ppm-plus 5:ppm Fer (D)Peuy 15 ppm plus 10 ppm Fe**. Solutions were made with Cu** as CuCle and Fe*™ as FeCl2 * 4H2O, since Cl in the dilutions used is non-noxic to Tetrahymena. All experiments were run three times, in triplicate each time, and mean growth curves plotted. Light transmittance was measured twice per day, in most experi- ments, after inoculation of cultures (or once per day in some experiments) for 5-7 days, until peak growth had been reached and decline of the population began. RESULTS 1. CONTROLS—Rates of growth be- gan with a brief lag phase of 15 to 20 hours, then entered the logarithmic phase ALLEVIATION OF THE ToxiIciTy of CopPpER TO Tetrahymena pyriformis 681 of growth which continued through the next 24 to 30 hours. Control populations continued to increase non-logarithmically through another 20 to 40 hours to asymp- tote before beginning to decline (Fig. 1). Populations reached about 450,000 to 500,- 000 per ml before decline began. 80 re} (=) Log. % Light Transmittance 100 is) i 2 3 DAYS OF GROWTH Fic. 1. Growth of Tetrahymena pyriformis in proteose- peptone medium at 29° C. 2. EFFECTS OF COPPER—Copper, alone, retarded the growth of Tetrahy- mena in as little as 3 ppm, growth always being slower and peak populations less numerous than the controls. The lag phase of the growth was lengthened, the logarithmic phase shortened and/or de- pressed and the asymptote was reached at a lower population density than for con- trols. Growth decreased as the concen- tration of Cu** was increased with 20 ppm Cu** almost completely eliminating growth (Fig-2): 3. EFFECTS OF IRON—In concen- trations of 3 to 15 ppm, Fe** was stimula- tory to growth, most significantly at 5 ppm, less so at 3, 10 or 15 ppm (Fig. 3). Poteet SOF EQUAL AMOUNTS OF IRON AND COPPER TOGETHER—At equal concentrations of both Cu** and Fe**, the latter eliminated ight Transmittanc l DAYS OF GROWTH Fic. 2. Growth of Tetrahymena pyriformis with 3 to 20 parts per million of copper as CuCl.,. the toxic effects of the former. At 3 ppm of each, growth exceeded both that with Cu** alone (3 ppm) or of the controls grown at the same time, but not that with Fe** alone at 3 ppm. At 5 ppm each of Cu** and Fe**, growth exceeded that with Cu** at 5 ppm, or of the controls, and parallelled and ultimately exceeded the stimulated growth with Fe** alone at 5 ppm. At 10 ppm each of the Cu** and Fe**, growth significantly exceeded both % Light Transmittance © e Log. DAYS OF GROWTH Fic. 3. Growth of Tetrahymena pyriformis with 3 to 10 ppm of iron as FeCl, - 4 H,O 682 Tue UNIVERSITY OF Kansas SciENCE BULLETIN growth with 10 ppm Cu‘ alone, or the controls, and also exceeded growth with 10 ppm Fe** alone. At 15 ppm each of Cu** and Fe**, growth significantly ex- ceeded the growth with 10 ppm Cu” alone, or 10 ppm Fe** alone, or the con- trols (Figs. 4-7). » EFRECIS ~OF. UNEQUAL AMOUNTS OF Cu* AND Fe" —At 15 ppm Cu** and 5 ppm Fe", the latter only temporarily alleviated the toxic effects of the former, growth being initially faster, but not reaching as high a peak, and de- clining more quickly than with 15 ppm 80 Cu 3 ppm -Fe 3 apne 90 >, Cu 3 ppm % Light Transmittance Log. 100 DAYS OF GROWTH Fic. 4. Growth of Tetrahymena pyriformis with copper and iron in equal amounts of 3 ppm. 80 od we tes 10 ppm AGS Fe 10 ppm .* = ee 10 ppm ‘+ % Light Transmittance g- DAYS OF GROWTH Fic. 6. Growth of Tetrahymena pyriformis with copper and iron in equal amounts of 10 ppm compared to growth with 10 ppm of copper alone, to growth with 10 ppm of iron alone, and to the control rate. % Light Transmittance Cu™ alone. At 15 ppm Cu** and 10 ppm Fe, growth ultimately and significantly exceeded that in other combinations tested, except for early growth where it was tem- porarily slower than the controls, or that with 15 ppm Fe**, alone, or that with 15 ppm Cu™ plus 15 ppm Fe**. The growth in the latter exceeded growth of all other combinations in early stages and was ex- ceeded in later stages only by the growth with 15 ppm Cu plus 10 ppm Fe** (Figs. BO), 80 CONTROL 90 Log. 100 0 1 2 3 4 5 6 DAYS OF GROWTH Fic. 5. Growth of Tetrahymena pyriformis with copper and iron in equal amounts of 5 ppm. 80 es ais —~.Cu 15 ppm yak Sire 15 =| ti CONTROL 90 : an y % Light Transmittance Ze Lo; DAYS OF GROWTH Fic. 7. Growth of Tetrahymena pyriformis with copper and iron in equal amounts at 15 ppm compared to growth with 15 ppm of copper alone, to growth with 15 ppm of iron alone and to the control rate. ALLEVIATION OF THE Toxicity oF Copper To Tetrahymena pyriformis 683 80 Cu 19 ppm £Fe 10 reas Cu 5 ppm aRCne Fe Sippa sea es ai aieke Pa Sa -“ Cu 3 ppm -- fe z Fe 3 coat : 4 CONTROL vee ee bu 15 ppm Fe 15 cont 90 Log. % Light Transmittance 100 DAYS OF GROWTH Fic. 8. Growth of Tetrahymena pyriformis with copper and iron together in equal amounts at various concentrations. 70 pr é N. / \ 1 *\ Cu 15 ppm 3 Fe 10 ppd / / 8 } Cu 15 : _ Cu ppm | pero Fe 15 tet Log. % of light transmittance DAYS OF GROWTH Fic. 9. Growth of Tetrahymena pyriformis with copper and iron together in unequal amounts at various concentrations. DISCUSSION Iron obliterates the toxicity of copper at concentrations of the latter which are detrimental to growth and survival of small aquatic organisms, such as Tetra- hymena pyriformis. Some data indicates that other biologically non-toxic metals may reduce the toxicity of heavy metals, eg. calcium vs. cadmium (Bergquist, 1974) or potassium, sodium or magnesium vs. cadmium (Bovee, unpublished). Iron, calcium, potassium and magne- sium, as ions, are critical triggers for ma- jor, cellular, enzyme-systems, e.g., those involved in making and using adenosine- triphosphate (ATP). Evidence from elec- tronmicroscopy and biochemistry indicates that toxic, heavy metals poison these en- zyme-systems, especially at cellular and mitochondrial membranes (Organ, 1972; Bergquist, 1974) by displacing requisite ions, thereby inhibiting energy-utilization and storage. Added Fe** evidently prevents the at- tachment of Cu** to the sites where iron is required, particularly at the cytochromes and flavinoid electron acceptors. Why the combination of copper and iron is stimulatory (except where copper is relatively high and iron much lower in concentration) we have not determined. Iron alone is stimulatory to growth of Tetrahymena pyriformis (Shug et al., 1969) and traces of copper are necessary for growth of Tetrahymena (Kidder and Dewey, 1951) and other organisms, in- vertebrate (Elvehejm, 1935) or vertebrate (Heilbrunn, 1952). For vertebrates, copper and iron are both involved in the formation of hemo- globin and the concomitant release of iron from the liver (Schulze, 1941). For cer- tain invertebrates, e.g., crustacea, copper is a component of the blood pigment, hemocyanin. It is also present in certain polyphenal oxidases and tyrosinase (West and Todd, 1957). Despite a minimal requirement for copper by many animals, excesses are toxic, blocking enzymes critical in the tricar- boxylic acid cycle, e.g., succinic dehydro- genase (Heilbrunn, 1952), and ribonucle- ases (Roth, 1959). Copper is also highly toxic for other small invertebrates in unde- termined manners, e.g., heliozoa (Roth, et al., 1968). 684 Tue University oF Kansas ScrENcE BULLETIN Neither the toxic effect of the copper, nor its minimal requirement are fully explained. We assume that excesses of it, in addition to poisoning sites in enzymes normally occupied by iron, also block the succinic dehydrogenase of the tricarboxylic acid cycle, thereby doubly blocking that cycle, both at its “bridge” from glycolysis and by way of the flavinoid-cytochrome chain, where the copper displaces iron. The iron can presumably displace or prevent attachment of the copper at sites where iron normally occurs, protecting those sites. The copper then, presumably, acts normally at and is readily available only to sites on enzyme systems where it is required, resulting in the combined stimulatory effect to growth, since both more iron and more copper are available to enzymes systems which require them. In the environment, excesses of copper are less likely to be toxic where iron is also present, if the data found here may apply to other small, aquatic invertebrates and vertebrates as well as to Tetrahymena. Since the general, nutritional require- ments of Tetrahymena are similar to those of many other animals (Hill, 1974), the extrapolation is reasonable. This would further suggest that indus- trial effluents containing copper could more safely be introduced into waters with a relatively high content of iron than into waters lacking iron. Industries unavoidably releasing some copper into local waters might better be located, therefore, in areas where waters naturally contain iron than in areas where iron content of the water is minimal to absent. LITERATURE, CITED 3ERGQUIST, B. L. 1974. The effects of cadmium upon growth, locomotion and _ ultrastructure of Tetrahymena pyriformis. PhD. Daisserta- tion. Univ. Kansas, Lawrence. 147 pp. Berzer, S. B. 1975. Copper toxicity in Busycon canaliculatum L. Biol. Bull. 148:16-25. BreMErR, I. 1974. Heavy metal toxicities. Quart. Rev. Biophys. 7:75-124. Cairns, J., JR.. AND Dickson, K. L. 1970. Reduc- tion and restoration of the numbers of fresh- water protozoan species following acute ex- posure to copper and zinc. Trans. Kansas Acad. Sci. 73:1-10. CHAISEMARTIN, C. 1973. Comparative analysis of Cu** toxicity in Astacus leptodactylus and Oronectes limosus. Depression of ion-regu- latory functions of a Na-K-dependent ATP- ase activity in the gills. Compt. Rend. Soc. Biol. (Paris). 167:324-328. CHRISTENSEN, G. M., McKim, J. R., Brunocs, W. A. AND Hunt, E. P. 1972. Changes in the blood of the brown bullhead, Ictalurus nebu- losus (LeSueur) following short and long term exposure to copper (II). Toxicol. Appl. Pharmacol. 23:417-27. Eviiott, A. M. anp Hayes, R. E. 1955. Tetrahy- mena from Mexico, Panama and Columbia, with special reference to sexuality. J. Proto- zool. 2:75-81. E_veHeyM, C. A. 1935. Copper content of animals and blood pigments. Physiol. Rev. 15:471- 507. Hartunc, R. 1972. Biological effects of heavy metal pollutants in water. Adv. Exp. Biol. Med. 40:161-172. HeiLprunn, L. V. 1952. General Physiology, 3rd ed. W. B. Saunders Co., Philadelphia. 818 pp. Henperson, B. M. 1975. Acute copper toxicosis in the Canada Goose. Avian Dis. 19:385-87. Hitt, D. L. 1972. The Biochemistry and Physiology of Tetrahymena. Academic Press, New York. 230 pp. Kipper, K. W. ano Dewey, V. C. 1951. The bio- chemistry of ciliates in pure culture. In: Lwoff, A., ed., Biochemistry and Physiology of Protozoa, Vol. 1. Academic Press, New York. pp. 323-400. Louria, D. B. 1972. The human toxicity of certain trace elements. Ann. Intern. Med. 76:1307- 1319. Mitts, W. L. 1973. Some toxicity studies on se- lected protozoa in certain toxicities of mer- cury. PhD. Dissertation. University of Kan- sas, Lawrence, Kansas. 148 pp. Orcan, A. E. 1972. Water flux in Tetrahymena. Are osmotic differentials important? PhD. Dissertation, University of Kansas Library, Lawrence, Ks. 128 pp. Rice, H. V. 1973. The effects of some trace metals on marine phytoplankton. CRC Crit. Rev. Microbiol. 3:27-48. Rorn, J. S. 1959. Comparative studies on tissue ribonucleases. Ann. N.Y. Acad. Sci. 81:611- 617. Rotn, L. E., Pravaya, D. J. AND SHIGENAKA, Y. 1968. Degradation of microtubule-containing helio- zoan axopodium by cuprous and_nickelous ions. J. Cell Biol. 39, 115a. Sate, A. J. 1948. Fundamental Principles of Bac- teriology, 3rd ed., McGraw-Hill, New York. 730 pp. ALLEVIATION OF THE Toxicity oF Copper to Tetrahymena pyriformis 685 SCHERBAUM, O. AND ZEUTHEN, E. 1955. Tempera- ture-induced synchronous division in_ the ciliate protozoon Tetrahymena _ pyriformis growing in synthetic and _proteose-peptone media. Exp. Cell Res. Suppl. 3:312-325. ScuutzE, M. O. 1941. The relation of copper to cytochrome oxidase and the haemopoietic activity of the bone marrow of rats. J. Biol. Chem. 138:219-237. SHuc, A. L., Etson, C. aND SHraco, E. 1969. Ef- fect of iron on growth, cytochromes, glycogen and fatty acid of Tetrahymena pyriformis. J. Nutr. 99:379-86. SkipMoRE, J. F. 1974. Factors affecting toxicity of pollutants to fish. Vet. Rec. 94:456-58. STEPHENSON, R. R. aNp TayLor, D. 1975. The in- fluence of EDTA on the mortality of the clam (Venerupis decussata) exposed to sublethal concentrations of copper. Bull. Environm. Contam. Toxicol. 14:304-8. SuTTLe, N. F. 1974. Recent studies of the copper- molybdenum antagonism. Proc. Nutr. Soc. 33:299-305. West, E. S. anp Topp, W. R. 1957. Textbook of Biochemistry, 2nd ed. MacMillan Co., New York. 1356 pp. aa 1 on ons wert iit inhi cana) © irre oe , a 7 ne : ca SCIENCE BULLETIN Re ote me = ss Rn an oe, ‘ete be) ‘ne oot Bs ee ete oe Om tone .*. ‘ete oe ee See ° « OG Ss S = POPULATION DYNAMICS AND a = s PRODUCTION OF DAPHNIA AMBIGUA a IN A FISH POND, KANSAS By SR Sf: s Chi-Hsiang Lei es and s Kenneth B. Armitage srereecesececesmcecesPcstecareraraneratane & Vol. 51, No. 25, pp. 687-715 January 11, 1980 ANNOUNCEMENT The University of Kansas Science Bulletin (continuation of the Kansas Uni- versity Quarterly) is an outlet for scholarly scientific investigations carried out at the University of Kansas or by University faculty and students. Since its incep- tion, volumes of the Bulletin have been variously issued as single bound volumes, as two or three multi-paper parts or as series of individual papers. Issuance is at irregular intervals, with each volume prior to volume 50 approximately 1000 pages in length. Effective with volume 50, page size has been enlarged, reducing the length of each volume to about 750 pages. The supply of all volumes of the Kansas University Quarterly is now ex- hausted. However, most volumes of the University of Kansas Science Bulletin are still available and are offered, in exchange for similar publications, to learned societies, colleges and universities and other institutions, or may be purchased at $20.00 per volume. Where some of these volumes were issued in parts, individual parts are priced at the rate of 2 cents per page. Current policy, initiated with volume 46, is to issue individual papers as published. Such separata may be purchased individually at the rate of 3 cents per page, with a minimum charge of $1.00 per separate. Subscriptions for forthcoming volumes may be entered at the rate of $20.00 per volume. All communications regarding exchanges, sales and subscriptions should be addressed to the ExcHance LrprariAn, UNIVERSITY oF Kansas Lisraries, LAwrENcE, Kansas 66045. Reprints of individual papers for personal use by investigators are available gratis for most recent and many older issues of the Bulletin. Such requests should be directed to the author. The International Standard Serial Number of this publication is US ISSN 0022-8850. Editor William L. Bloom Editorial Board Philip W. Hedrick Rudolf Jander Harvey Lillywhite Charles D. Michener Norman A. Slade Henry D. Stone George W. Byers, Chairman THE UNIVERSITY OF KANSAS SCIENCE BULLETIN Vol. 51, No. 25, pp. 687-715 January 11, 1980 POPULATION DYNAMICS AND PRODUCTION OF DAPHNIA AMBIGUA IN A FISH POND, KANSAS Cui-Hsianc Ler’ anp KENNETH B. ARMITAGE” TABLE OF CONTENTS PASE SrIGRUN Ge Seee ce cence fase Se ee eased Sn Ss cicelin ies A a 687 HINTIER OD UW Con O Ni resener cet c sate ete Bsc d ee JETS 65 heen dees a blah, Nn ee SR Le i oe, 688 DESCRIP ONMO EE OND ia saan = ih. eee tes Ae eee A 688 NUATSERICATES IRAN D WUE ODS joo este ical crave se Nhe ee ew ost See ees 2 ee Se a 689 Zooplanktonmsamplingtand. /Aimalysis: gts eee ce ee 689 physico-chemuica leeAtialiysishes 22" ..8- 2S eels ea ee a er 689 SeStOnmpArialiycisyeen se wees C eke wuiae eee taelGusly hy eres ae ey ee Reese ee ee 689 Dear ei O ITE Mts eee ew cee a eee emer Ae t/a, A eee A adeee edict ies NY ee ee 690 Gi ii ateyge DTOG UCLIVIEYs 22828 wee ee ee Oe ee ee 690 (al oniticu Vial wesw eee ae ee a een ean ene eT Pd RE os UE NC ee ee ... 690 SES EO Mimi me Sorat SU Nees Sees en SL Rare nates acti yar eek ee ce 690 DD OPA RUGATUOICUG wc. Sane ec te na eee LR ec ast a Sate ates nash 2 once ge 690 RopulationsParameters and: Production’ Estimate: 2222 ce 691 [BYoyobileNoyosoy) ete ee VanVe(vss toh | ee Py pee I Oe eee ee 691 PTO CAUCE O Te sp tesess ease SOE SN 8 ee oc Ae Ye Oe 691 DRIES Ue 1S meee eee ee eS ne tee ie, ese eet ec eg Se ee ee ee 692 Somer environmental Parameters, 2. ss.ssvece cece eee one eek oe we ac ees ee eee 692 @alonifics Valuic Of SESton 2. eee as cen ec cece che ot ean etc ee ee eee ce ene 694 @alonitieavialuie ofeDaphnia and: Hogg: .. eee ee ce ee cece ele a ses cae te eee 694 Seasonalwbatterm: of Zooplankton Abundance: sice.c.sesc: cece ease 695 Population Datarand Production (Estimates: s<.ccscsc.cesqs28ecce sees eoeeaclecs tase ce etc cee ee eee ce 697 Searles ee se ast Ss pretties ES RE se 697 Populations Compositiom and oize> Distribution. 222.2. cccce <= oseere = nc cce ce eoenc nea eee cee 697 Seasonal Variation, Egg Production, Female Size ....... behest .3. We Naa a ee 698 Population \Parannieters, eerc.ceesccccecsseccstt coe -neceec dete ecto ca cg sect cea teams te oe oc See acces eae 701 Biomass and, Proc uctio nn. yes se ke As 5 ER ea ae ee te ee 701 DISCUSSION MPM re A nee re Rg Arica es ed Sects sh eee tas emi eens See 702 INGIGN OWITFEDGENUENIT'S te eee dco cn RE Na ose POUR ea ee 705 TB LARE ACURA GUTS eee See Nao ED, Tran 2 a lh i RT i 705 SIV ACE TEES umn SS RM Pr Ee er a ea Ete ee ak ee eee 708 ABSTRACT High population and egg density of Daphnia ambigua Scourfield occurred in winter and spring when tem- peratures were low and food concentrations were high. Mean clutch size ranged from | to 5.7 with an average of 2.01 + 1.12 (S.D., N= 86) while the number of eggs per adult female ranged from 0 to 5.1 with an average of 1.09 + 0.98 (S.D., N91). Highest values usually occurred during population increase and high food concentrations. There was no consistent relationship between clutch size and size of the reproductive females. When mean clutch size was low, clutch size did not increase with increasing size of females carrying the eggs. The size of reproductive (parthenogenetic) females was generally larger during the winter and early spring 1 Based on research submitted as a Ph.D. Dissertation, University of Kansas. 2To whom reprint requests should be sent: Division of Biological Sciences, University of Kansas, Lawrence, KS, 66045, USA. 688 Tue UnNrversiry oF KAnsAs SciENCE BULLETIN when water temperatures were low and food concentrations were high. Population density, egg density, clutch size, and egg number per adult female were all positively correlated with the estimates of food concentration (seston and chlorophyll a) but were negatively related to temperature. Sexual reproduction occurred in late March and April and lasted for about one month. Ephippial females did not appear when mean clutch size of parthenogenetic females was 1.0. Ephippial females appeared nearly two months later than males did. Instantaneous birth rate (b’) fluctuated between 0 and 0.501 with an aver- age of 0.067 for the period June 1972 to June 1973. Instantaneous birth rates were positively correlated with temperature but not the estimates of food concentration. The highest calculated instantaneous rate of population change (r’) was 1.943 and the highest instantaneous rate of death rate was 1.589. Standing crop biomass ranged from 0 to 1.6 g dry wt/m* (0 to 8.4 kcal/m*) with a mean of 251.1 mg dry wt/m* (1.3 kcal/m*) for the period June 1972 to June 1973. Production (somatic growth, Ps, and egg production, P;) was estimated by egg ratio method and Winberg’s method. Estimates of daily and monthly production calculated by the two methods were different but the estimates for the entire study period were similar, 6.8 g dry wt/m® (34.7 kcal/m*) and 6.6 g dry wt/m* (33.1 kcal/m®*), respectively. The total production of resting eggs, 66.4 mg dry wt/m* (390.8 cal/m*) was about 1% of the total production (Pz + P; + resting eggs) for the entire study period. The P/B coefficients estimated from dry weight and from caloric content were similar. The monthly P/B ratios ranged from 1.1 to 7.7 and were generally higher in the warmer months than in colder months. The annual P/B ratio was 25. The average daily P/B for the period June 1972 to June 1973 was 0.068 and that for the period July 1972 to June 1973 was 0.066. Therefore, the average biomass turnover time was 14.7 and 15.2 days, respectively. Daily P/B ratio was related more closely to temperature than to food. The monthly production of exuvia (Pex) varied from 18.3 to 2752.8 cal/m*, and the total Pex for the period July 1972 to June 1973 was 11.2 kcal/m‘*. Pex contributed 9% ~ 49% to monthly Pz + P; + Pex and accounted for 259% of annual Pz + Pr + Pex. INTRODUCTION Detailed biological and ecological stud- ies of member species are essential for the understanding of the structure and dy- namics of ecosystems. In order to compile an energy budget for a natural zooplank- ton population and to understand how populations respond to environmental changes, data on population dynamics and production of the species concerned are required. Production estimates can be used to assess the relative importance of various species in the economy of the zooplankton community (Petrovich et al., 1961; Patalas, 1970; Comita, 1972; Schindler, 1972). Pro- duction estimates also yield turnover rate information which can be used to evaluate the effect of environmental factors on tivity. Information on population dynam- ics and production of planktonic crus- taceans in Kansas is still very limited although several ecological studies on planktonic crustaceans are available (e.g. Armitage, 1961; Armitage and Davis, 1967; Armitage and Smith, 1968; Prophet, 1957; Prather and Prophet, 1969; Angino et al., 1973; Prophet and Waite, 19745 Palavanchuk, 1977). The purpose of the present study was to analyze the popula- tion dynamics and estimate the production of a wild population of Daphnia ambigua in a fish pond in Kansas with the hopes of stimulating further research in zoo- plankton production in this state. DESCRIPTION OF THE POND zooplankton and to compare aquatic sys- tems (Stross et al., 1961; Winberg et al., 1965; Patalas, 1970; Pederson et al., 1976). Daphnia ambigua Scourfield is a com- mon and widespread zooplankter in aquatic communities of Kansas (Prophet and Waite, 1974). Because Daphnia are utilized directly or indirectly as food by young and adult fish, this organism could have a large impact on aquatic produc- The pond chosen for the study is lo- cated on Campus West, University of Kansas, Lawrence. The pond when full had an area of about one hectare and a maximum depth of about five meters. Most of the drainage basin was covered by grasses and scattered deciduous trees. The pond was sometimes used as a source of water for other experimental ponds. Thus its water level, and consequently its PorpuLaTION Dynamics AND PRODUCTION OF Daphnia 689 area and depth, varied as a function of precipitation, runoff, removal from near the bottom, and evaporation. MATERIALS AND METHODS Zooplankton Sampling and Analysis Zooplankton were sampled at least twice weekly from June 1972 to October 1972 and less frequently (once every week or once every two weeks) from November 1972 to February 1973. From March 1973 to July 1973 zooplankton samples were collected once every second day. During the ice-free season, plankton samples were taken with a #10 silk bolting net fitted on a Clarke-Bumpus plankton sampler which was towed at a depth of 0.5 meter for a period of five minutes in the central, deeper portion of the pond. Samples were collected at night, usually 1 to 2 hours after sunset. One revolution of the Clarke- Bumpus sampler was assumed to equal four liters of water. When there was ice on the pond, a hole was cut through the ice at the center of the pond and a vertical column of water was taken during the day by lowering a Wisconsin plankton net with #20 mesh size to the bottom of the pond, leaving it there for 5 minutes, and then hauling it to the surface. All samples were fixed immediately with 95°% ethyl alcohol and then preserved in a mixture of 95% ethyl alcohol and 2% formalin. In the laboratory each sample was diluted with distilled water to a known volume (usually 100 to 200 ml) depending upon the number of plankters present in the sample. At least three one ml sub- samples from each 100 or 200 ml of diluted sample were counted on a Sedgwich-Rafter cell under low power of a stereo-microscope. When the number of plankters in the sample was very low, the entire sample was counted. To facilitate the analysis of the population compo- sition of Daphnia ambigua, individuals in each sam- ple were grouped into five categories similar to those of Lei and Clifford (1974); except females with carapace length equal to or larger than 0.721 mm without eggs or embryos in their brood chamber were considered adults. The average number of individuals per ml of each species of Cladocera and Copepoda was calcu- lated from the three one ml subsample counts. This average number was multiplied by 100 or 200, de- pending on the final volume of diluted sample, and then divided by the total number of liters of water from which the sample was obtained, to calculate the number of individuals per liter. For undiluted sam- ples, the number of individuals in the whole sample was divided by the total number of liters of water from which the sample was obtained, to calculate the number of individuals per liter. The carapace of animals belonging to category 1 was opened with fine needles, and the numbers ot eggs or embryos in the brood chambers were counted under a dissecting microscope (32x). The term “egg number” is used irrespective of whether eggs or embryos were counted. The mean egg number per brood was calculated from a subsample of at least 25 females having eggs or embryos in the brood chambers. To determine the egg density (No. of eggs or embryos per ml), all of the eggs or embryos in the 1 ml subsample were counted: these included eggs in brood chambers and those free from the brood chamber due to preservation. Length measurements of 200 individuals (if available) from each sample were made from the top of the head to the base of the shell spine (total body length), and from the anterior margin of the carapace to the base of the shell spine (carapace length). Physico-chemical Analysis All water samples were taken with a 2.]-liter Kemmerer Water Sampler at 0.5 m intervals from the surface down to within 10 cm of the bottom of the pond, with the center of the water sampler at the depth being sampled. Samples were combined for the determination of plant pigments, dissolved solids, orthophosphate, nitrogen, and seston. All water samples were filtered through a #10 silk bolting net (mesh size 153 um) to remove larger organisms and debris before any further analysis was performed. The concentrations of orthophosphate, ammonium- N, nitrate-N, and nitrite-N were determined colori- metrically (Hach Chemical Company Manual 1969) with a Bausch and Lomb Spectronic 20 Spectropho- tometer. Alkalinity (Hach Catalog No. 9:5) and_ total dissolved solids (American Public Health Associa- tion, 1965) were determined using standard pro- cedures. The pH of the pond water was measured with a IL 175 PORTO-matic pH meter; water tem- peratures were measured with a thermistor ther- mometer coupled to a galvanic cell oxygen analyzer (Precision Scientific Co., Chicago). Light penetra- tion was estimated with a Secchi disc having a diameter of 20 cm. Seston Analysis The dry weight of the seston was determined from a 500 ml or 1000 ml sample, depending upon the concentration of seston. The sample was filtered through a dry, tared Whatman GF/C_ fiber-glass filter which had been pre-ignited at 500°C for 24 h. The filter with seston was dried at 60°C for 24 h and reweighed to determine the dry weight of seston; it was then ashed at 500°C in a muffle furnace for 30 minutes and reweighed to determine the ash content of the seston. During the period between March 1973 and July 1973, the size distribution of seston particles was measured weekly by means of a Model Zn Coulter- counter with a 140 wm-diameter aperture tube. Be- cause pond water must be made saline to conduct electric current, an appropriate amount of sodium 690 Tue UNIversity oF Kansas ScIENCE BULLETIN chloride was added directly to the sample to get a salinity of 0.59% before counting (Mulligan and Kingsbury, 1968). When the concentration of seston was high (>36000 particles/ml), the sample was diluted with a 0.594 sodium chloride solution which had been filtered through a 0.45 wm Millipore mem- brane filter. Plant Pigments Phytoplankton for pigment analysis was concen- trated from 500 to 1000 ml of sample (depending on the density of phytoplankton) onto a Whatman GF/C fiber-glass filter (5.5 cm diameter) placed in a Buchner type porcelain filter funnel fitted on a suction flask attached to a suction pump. The filter with phytoplankton was ground up in a porcelain mortar in 10 ml of 95°4 acetone. The whole was carefully pipetted into a 15 ml capped bottle which was wrapped with aluminum foil and placed in a refrigerator for 24 h. After 24 h extraction, the content of the bottles was centrifuged for 10 minutes and the optical densities of the supernatant were measured with a Zeiss PMQ II Spectrophotometer. The amount of chlorophyll a, 4, ¢ and carotenoid presenf in each sample was calculated (Strickland and Parsons, 1968). Primary Productivity Primary productivity was measured by the oxygen light and dark-bottle technique of Gaarder and Gran (1927). Dissolved oxygen was measured by azide modification of the Winkler technique (Amer. Public Health Assoc., 1965). In calculations, both PQ and RQ were assumed to be unity. The water samples were first strained through #10 silk bolting net to remove zooplankton and large debris. Calorific Values Seston Dry weight of the seston was determined as described above except a 0.45 um Millipore mem- brane instead of Whatman GEF/C fiber-glass filter was used to concentrate seston. The filter with seston was carefully cut into small pieces, pressed into a pellet with a Parr pellet press, and burned in a Parr 1411 Semimicro Oxygen-bomb Calorimeter to determine the total calorific value of seston and filter. By subtracting the calorific value contributed by the filter, the calorific value of seston was thus obtained. The calorific value of 0.45 um Millipore membrane filter was determined by burning a pellet of membrane filter. Four determinations of 0.45 um membrane filter yielded a mean calorific value of 3053.7 + 83.37 (standard deviation) cal/g with a coefhicient of variation of 2.73°%%. This value is very close to the values obtained by Comita and Schindler (1963) and Moshiri (1968). Daphnia ambigua Calorific values were determined on Daphnia cultured in 15 gallon aquaria and fed abundantly with Chlamydomonas. Before collecting Daphnia from stock cultures for calorific determinations, they were starved for at least 12 days to avoid gut con- tamination and to produce egg-free females. The contents of the aquarium were first siphoned through a #00 silk bolting net (mesh size 0.752 mm) to separate adult females (total body length >0.87 mm, Group II animals), and were then strained through a #1 silk bolting net (0.417 mm mesh size) to separate medium size females (BL between 0.62 mm and 0.87 mm, Group JI animals). Finally, the contents were strained through a #16 net to con- centrate young females (BL < 0.62 mm, Group 1). Size group I and II consisted of immature daphnids; size group III daphnids were all adults. Each group of daphnids collected on the net was flushed several times with distilled water and washed into a beaker. Animals in the beaker were further sorted with an eyedropper. After the complete separation, each group of animals was rinsed several times with dis- tilled water and dried in an oven at 60°C for 24 h. The dry samples of each group of daphnids were then combined and mixed with an agate mortar, dried for another 24 h at 60°C, and stored in a crucible in a desiccator until the calorific determina- tion was made. Calorific value of D. ambigua was determined either by direct calorimetry or by the wet combustion method with potassium dichromate in concentrated sulfuric acid as oxidizing agent. In direct calorimetry, the sample (dry powder of Daphnia) ranging in weight from 4.5 to 10.5 mg was combined with 100 to 150 mg of benzoic acid and pressed into a pellet. Owing to the difficulty of obtaining a large sample of Daphnia for bombing, the use of benzoic acid as filler was necessary. The wet combustion method was a modification of the procedure of Ostapenya (1965, described in Winberg, 1971). A dry, ground sample of Daphnia (weight 0.84 to 3.69 mg) was placed together with 100 mg of AgsSOx into a 125 ml combustion flask, and 10 ml of 0.1 N solution of potassium dichromate in concentrated sulfuric acid was then carefully added into the flask by allowing the solution to run slowly down the inner surface of the neck of the flask which was slowly rotated. The flask was swirled gently to disperse the sample in the dichro-: mate-acid solution, and was then covered with a 50 ml beaker and heated for 60 min. at 140°C in a thermally regulated drying oven. After being heated, the flask was cooled to room temperature and the neck and wall of the flask were washed with 10 ml of distilled water. The 50 ml beaker, serving as a cover of the flask, was also washed with 5 ml of distilled water which was added to the flask. After adding 3 drops of phenyl-antranilic acid as an indicator, the excess dichromate remaining in the flask was titrated with a 0.02 N solution of ferrous ammonium sulfate (Fe(NHs)2(SOs)2*6H2O). Two control blanks (each containing 100 mg of AgeSOz and 10 ml of 0.1 N dichromate-acid solu- tion, excluding daphnid sample) were treated identi- cally to the sample flasks and titrated with 0.02 N Poputation Dynamics AND Propuction or Daphnia 691 ferrous ammonium sulfate solution. The amount of oxygen required to oxidize the sample was calculated from the amount of dichromate used and expressed as mg Os/mg sample. This value was then multi- plied by 3.4 (oxycalorific coefficient) to obtain the calorific value (cal/mg) of the sample. In order to calculate the calorific value in terms of ash-free dry weight (cal/ash-free mg), the frac- tion of ash content in the sample was determined by two methods; direct combustion of samples of dry ground Daphnia in a muffle furnace at 500°C for 2 h (direct ignition method) and by further com- bustion for 2 h in a muffle furnace at 500°C after being burned in an oxygen bomb calorimeter for calorific determination (oxygen bomb method). Eggs at stage I (Lei and Clifford, 1974) were dissected from adult females of laboratory cultured Daphnia, washed with distilled water, and dried for 24 h at 60°C. The dried eggs were stored in a crucible in a desiccator until the calorific determina- tion was made by the wet combustion method. Population Parameters and Production Estimate Population Parameters The instantaneous birth rate, b’ was calculated from the formula (Paloheimo, 1974) be =In (1 bs /No)/D (1) Where Eo and No are the number of eggs or embryos (egg density) and the number of animals (popula- tion density), respectively; D is the embryonic (or egg) developmental time in. days at the respective temperature. The egg developmental time (D in days) at the respective temperature (T in °C) was estimated from the equation, In D = 3.48977 — 0.17960T + 0.00244T°? (Lei, 1979). The finite birth rate (number of newborn per animal per day) was calculated from the formula (Caswell, 1972) B=D’ (e”-1)/1’ (2) where b’ is instantaneous birth rate, and r’ is the instantaneous rate of population change. r’ was cal- culated from successive pairs of population counts from the formula (Hall, 1964) r’= (In Ne-In No)/t (3) Where No is the initial population size and Ne the population size after an interval of t days. This equation is based on the assumptions that eggs were produced and hatched continuously, and that the population age distribution approximated the stable form for the period between observations. Knowing b’ and 1’, the instantaneous death rate, d’, of the populations was calculated as: d’ = b’ = r’ Production Production is defined as the increase in biomass which occurs in a given period of time (or as the total quantity of biomass formed over a stated pe- riod), whether or not all of it survives to the end of that period (Mann, 1969). Production estimates were calculated by dividing the standing crop biomass by turnover time. Turn- over time was estimated from finite birth rate (Cummins et al., 1969) as T=1/B where T is the turnover time in days, and B, the finite birth rate. Thus production rate, P (mg/m*/ day, or cal/m*/day) can be calculated as: P = (Standing crop biomass) /T which is equivalent to P = (Standing crop biomass) - (B) This method of production calculation will be re- ferred to as “egg ratio method.” Length-frequency data and length-weight rela- tionships (Lei, 1979) were used to convert standing crop numbers to standing crop biomass. They were calculated as: Biomass (mg dry weight/m’*) = ZNiWi a NeWe — NepWep or Biomass (cal/m*) = ZNiWiC; -+ NeWeCe “= NepWepCep where .Ni is the number of individuals in different size groups; Ne, the parthenogenetic egg density; Nep, the density of resting eggs; W:, the dry body weight of different size groups; We and Wep the dry weight of individual parthenogenetic and resting eggs, respectively; C; the calorific value of D. ambigua (4.475 cal/mg dry weight for juveniles and 4.9236 cal/mg dry weight for adults); Ce the calorific value of parthenogenetic eggs (5.883 cal/mg dry weight) and Cep the calorific value of resting eggs. For comparison, production was also estimated from the formula of Winberg et al. (1965): P (mg dry weight/m*/day) = (NeWe)/Te + (NjAW;)/T; + (NaAWa)/Ta and P (cal/m*) = (NeW-Ce)/Te + (NsAWjC;)/T; + (NaAW2Ca)/Ts where P is the production; Ne, Nj and Na the num- ber of eggs (or embryos), juveniles, and adults, re- spectively; We the weight of an egg; AW; and AWa the weight increment of juvenile and adult, respec- tively (i.e. the difference between the initial weight of given stage and the initial weight of the following stage); Ce, Cj, Ca the calorific value of eggs, juveniles, and adults, respectively; Te, Tj, Ta the duration of development of eggs, juveniles and adults respectively. The duration of juvenile (i.e. postembryonic de- velopmental time, T; in days) at the respective tem- perature (T in °C) was estimated from the equation, In Ty; — 4.00664 —0.24266 (In T)* (Lei, 1979). Ta (average life span of adult) was calculated as- suming the growth rate of body length in adults is approximately 3 times lower than in juveniles (Patalas, 1970; Weglenska, 1971), by Ta = [3T) (La — Loa) ]/ (Lon — Los) 692 Tue Universiry oF Kansas Sc1ENCE BULLETIN where Loy is the initial body length of the juvenile (i.e. the body length of neonata), Loa the body length of primiparia, and La the average body length of an adult individual in the investigated population. This second method of production estimation will be referred to as “Winberg’s method.” The total production for the entire season was obtained by plotting daily production for the entire season and determining the area under the curve planimetrically. Molting is a frequent occurrence in the life of crustaceans and each exuvium represents a significant fraction of dry weight of an individual. Wissing and Hasler (1968) reported a calorific value of 5.062 cal/mg dry weight (5.753 cal/mg_ ash-free dry weight) for the chitin of Daphnia which repre- sented 15-20°4 of the calories of an_ individual. According to our calculation, based on a_ calorific value of 5.062 cal/mg dry weight for chitin and assuming that the chitin represents 15°% of the calories of an individual Daphnia, the chitin repre- sents on the average, 13.8% (13.0 to 14.5%) of the dry weight of an individual D. ambigua. Be- cause growth of Daphnia is accomplished by regular molting, a significant amount of energy contained in the chitin is lost together with casted exuvia. Therefore, in the calculation of the energy budget for a considerable period of time, the amount of energy expended in exuvia must be considered. To estimate the daily energy expenditure in exuvia by the Daphnia population, the amount of energy in the chitin was assumed to be 15°% of the total calories of an individual Daphnia, and was calculated as: Molt (cal/m*/day) = (0.15Ny;WjC;)/D, 4- (0.15NaWaCa) /Da when Nj; and Na are the density (No./m*) of juvenile and adult Daphnia; W; and Wa, the mean dry weight of one individual of juvenile and adult Daphnia; Dy and Da the duration of juvenile and adult instars (in days), temperature adjusted; C; and Ca the calorific value (cal/mg dry weight) of juvenile and adult. The duration of juvenile (Dj) and adult instar (Da) at the respettive temperature (T in °C) was estimated from the equation: In Dj = 2.78332 — 0.15814T + 0.00214T? and In Da = 3.41135 — 0.17707T +. 0.00256T”, respectively (Lei, 1979). The total energy expenditure in exuvia for the entire season was obtained by plotting daily expenditure for the entire season and determining the area under the curve planimetrically. RESULTS Some Environmental Parameters Detailed data of physico-chemical parameters are recorded in Lei (1979). Concentrations of orthophosphate were generally higher in 1972 than in 1973. Lowest values occurred in May and June 1973; highest values occurred in November 1972 and October 1973. The concentra- tions generally are in the high end of the range of concentrations typical of surface waters of lakes of humid-temperate re- gions (Hutchinson, 1957). The mean for the period of study was 0.045 mg/l (==. 01060 SD): Nitrogen occurred as ammonia, nitrate, and nitrite. Nitrite-N concentrations al- ways were less than 0.01 mg/l. Nitrate-N concentrations varied from 0.007 mg/l to 0.141 mg/l with a mean of 0.0735 mg/l. Highest values occurred in early spring and in autumn; lowest values occurred in late spring and in the summer. The values of nitrate-N are higher than would be expected in natural waters, and may represent either an influx from run-off water from the pond basin or the nitrifica- tion of ammonia in situ. The concentra- tion of ammonia-N exceeded the concen- tration of nitrate-N throughout the period of study. Lowest values occurred during the winter months. The concentrations and seasonal distribution of ammonia-N are not unusual for temperate freshwaters (Hutchinson, 1957). Dissolved solids tended to be highest in the spring and lowest in the fall. The range of concentration was 145 to 309 mg/l with a mean of 202.3 mg/l. Turbidity, by reducing light penetra- tion, may affect the photosynthetic capacity of the primary producers. Secchi disc values were generally lower than would be expected for a typical North American lake. There was no clear seasonal pattern of transparency. The values varied be- tween 16.8 cm and 260 cm with a mean of 15789: cm: Chlorophyll a is a measurement of the concentration of algae which could be food for Daphnia. Values ranged from 0.46 to 28.4 mg/m® (Fig. 1) with a mean Poputation Dynamics AND Propuction or Daphnia 6 35 30 CHLOROPYLL @ (yg PER LITER) JUNE AUG. OcT. DEC. FEB. I972 \o Ww 70 e Chlorophyll @ O------ o Seston 60 SESTON CONCENTRATION (CAL. PER LITER) APRIL JUNE AUG. OCT. I973 Fic. 1. Seasonal variation in chlorophyll @ and seston in the Fish Laboratory Pond. of 6.43 mg/m*. The suspended matter, measured as seston, may also be a source of food. The range of seston concentra- tions, expressed as mg dry weight/l, was 1.1 to 73.5 mg/l with a mean of 10.3 mg/l. The highest concentrations of seston oc- curred in fall or winter. Most of the seston particles had a diameter between 2 and 8 pum. Particles of this size are readily selected as food by zooplankters (Edmondson, 1957; Brooks and Dodson, 1965; Gliwicz, 1969). Temperature, pH, alkalinity and pri- mary production were measured at more than one depth. However, for conven- ience, values of these parameters are re- = - nN nN Ww {e) oO (oe) on [o) WATER TEMPERATURE (°C) fe} uo JoJv A S$ ON DJF M AM J J A S ON 1972 I973 Fic. 2. Mean water temperature of the Fish Labora- tory Pond. ported as the average of all depths meas- ured. Temperature varied from 4.2 to 29.9°C with a mean of 22.1°C. Tempera- ture was relatively high by early June and peaked in late July and mid August. By late August water temperature began to decline but did not drop below 20°C until October (Fig. 2). Water temperature reached its minimum (4.2°C) by mid November 1972, and was maintained at near 5°C until February 1973. The water warmed rapidly in April, and reached 20°C in early June 1973. pH varied from 6.7 to 9.5 with a mean of 8.3. The value gradually increased from near 8 in June to values higher than 9 in mid August through October. By early November pH dropped to less than 8 and pH remained between 7 and 8 (except for one value below 7) until June 1973 when the values again exceeded 8 (Lei, 1979). The annual pattern of change of con- centration of total alkalinity was similar to the pattern of pH. Values ranged from 52.8 to 147.7 ppm with a mean of 98.7 ppm 694 Tue UNiversiry oF KANsAs ScrENCE BULLETIN for the period studied. Total alkalinity exceeded 100 ppm through July 1972. Values decreased markedly in August and reached 50 ppm in September. Total al- kalinity increased rapidly in November and remained above 100 ppm for the duration of the field work. Phenolphtha- lein alkalinity ranged from 0 to 17.2 ppm with a mean of 5.39 ppm for the period studied. Rates of gross primary production were generally high throughout the summer of 1972. Rates during the winter were low, often one third or less the summer rates. Although rates increased in the spring of 1973, they remained much lower than they had been in 1972. The range of gross primary production of the period studied was 17.9 to 330.8 mg carbon/m*/day with a mean of 146.3 mg carbon/m?*/day. Calorific Value of Seston The calorific value of seston collected from the Fish Laboratory Pond varied from 1320 to 3631 cal/g dry weight with a mean value of 2220.2 + 648.0 cal/g dry wt (Lei, 1979). The calorific values of seston found in this study were lower than those reported for two Thames Val- ley reservoirs, England (Kibby, 1971), but similar to those reported for Coon Lake, Ontario, Canada (Schindler et al., 1971). The calorific value of dry matter in planktonic and benthic organisms is di- rectly related to the ratio between the or- ganic and mineral fraction in the dry matter of organisms (Ostapenya and Sergeev, 1963). The relationship between the calorific value and the percent of or- ganic matter in aquatic organisms was described by either a linear (Sherstyuk, 1971, Thayer et al., 1973) or a curvilinear (Platt et al., 1969) regression equation. The relationship between the calorific value and the percent organic content of seston in this study was described by a linear regression equation as NS eo where Y is the calorific value of seston (cal/g dry wt) and X is the percent or- ganic matter (ie. 100-°%ash content). Both the correlation coefficient (r= 0.9, df = 15, P < 0.001) and regression coef- ficient “(b==43.9- (FP = 107.0. «di = 1 15- P < 0.001) were highly significant. Calorific Values of Daphnia and Eggs The mean calorific value obtained by the wet combustion method for group I, group II, group II, and eggs was 3828, 4041, 4369, and 5188 cal/g dry weight, respectively (Table 1). They were all sig- nificantly different from one another (P < 0.05; Student-Newman-Keuls test, Sokal and Rohlf, 1969) indicating that the calorific content of D. ambigua changes with developmental stages. The change of calorific value with the developmental stages was reported for other animals by Richman (1958), Comita and Schindler (1963), Khmeleva (1968), Klekowski et al. (1967), and Kibby (1971). The mean calorific values obtained by direct cal- orimetry for group I, 4369 cal/g dry wt and for group III, 4924 cal/g dry wt were also significantly different from each other (G==3 07 di— 5) P << 00>) The completeness of oxidation of or- ganic matters by the wet combustion method (bichromate oxidation) was tested on several pure organic materials. The completeness of oxidation varied from 81.2% to 106.6% with a mean of 935 a9 97, (civ 1979) 5 comparison on tie results obtained by the wet combustion method with those derived from direct calorimetry suggests that the completeness of oxidation of D. ambigua was 87.6% for group I and 88.7% for group HI. The mean completeness of oxidation for these PopuLaTION DyNAMIcs AND PRODUCTION OF Daphnia 695 two groups considered together, 88.2°, is close to the value of 90°% obtained by Ostapenya (1965). Because calorific values obtained by the wet combustion method were lower than those obtained by direct calorimetry, the calorific values of group II and eggs obtained by wet combustion method were corrected for 11.8% (ie. 100% -88.2°%,) incomplete oxidation. The mean calorific values after correction were 4583 cal/g dry wt for group II and 5883 cal/g dry wt for eggs. There was a significant difference be- tween the values of ash content obtained by direct ashing of samples in a mufile furnace and by oxygen bomb combustion (Table 2). The t value for group I was 853 (df =4; P< 0.01), and that for group III was 3.47 (df = 7; P < 0.02). The ash content determined from weight loss on ignition in oxygen bomb calorim- etry was underestimated (Paine, 1964). Bomb combustion produced a systematic error of underestimate and a random er- ror of variation between trials (Reiners and Reiners, 1972). These errors varied among the different types of material tested, but in general both types of error increased with the ash content of the ma- terial. The average systematic error of underestimate of ash content for all sam- ples they tested was 1.46%, which led to an error of 1.56°% when adjusting calorific coefficients to an ash-free basis. Oxygen bomb combustion produced an under- estimate of ash content in the present study (Table 2). Therefore, the ash values obtained by direct ashing in muffle-furnace were used in calculating calories per gram ash-free dry weight for D. ambigua. The mean ash content of group I was sig- nificantly larger than the mean ash content of group II and group II. The ash con- tent of group II and group III animals did not differ significantly. Because of the difficulty of obtaining enough sample for ashing, the ash content of the eggs of D. ambigua was not determined. If the ash value of 4.0°% reported for the eggs of D. galeata mendotae (Moshiri and Cummins, 1969) is used, the cal/g ash- free dry wt of the eggs of D. ambigua would be 6128. Seasonal Pattern of Zooplankton Abundance Two species of Calanoida (Diaptomus siciloides and D. clavipes), three species of Cyclopoida (Mesocyclops edax, Cyclops bicuspidatus thomas, C. vernalis) and one species of Harpacticoida (Canthocamptus sp.) occurred in the Fish Laboratory Pond. Detailed population data are recorded in Bet (1979); Diaptomus (mostly D. siciloides) ap- peared throughout the study period; the population density (copepodids and adults combined) ranged from 1.2 (March 22, 1973) to 92.1 per liter (Aug. 7, 1972). The numbers were most abundant (above 22 individuals per liter) during the summer months of 1972 (June through mid Sep- tember) and decreased in the following months and remained below 20/1 through the fall and winter of 1972 and early spring 1973. The population increased to 65.4/l on April 24 of 1973 and then declined again. Population densities were com- paratively lower in June and July of 1973 than in the same months of 1972. Cyclops bicuspidatus thomasi was the most abundant cyclopoid. Mesocyclops edax and C. vernalis occurred sporadically. The population of all cyclopoids (cope- podids and adults combined) was high during the summer months of 1972 and late spring and early summer of 1973. The highest density (60.4/1) occurred on May 18, 1973. Nauplii of all copepods combined were most abundant during the early summer (late June and early July of 1972; late May and June of 1973). The numbers of 696 Tue Universiry oF Kansas ScrIENCE BULLETIN nauplii were low during the winter (1972) and early spring (1973). The highest density (50.8/1) appeared on June 8, 1973. Fourteen species of Cladocera were identified in the pond with Daphnia am- bigua, Daphnia parvula and Ceriodaphnia lacustris being the most abundant. Dza- phanosoma leuchtenbergianum appeared in late June through mid October of 1972 with two population maxima, one on July 31 (6.2/1) and other on August 30 (8.3/l). Diaphanosoma was completely absent from the pond in the following months and had not appeared again by the end of July 1973. Scapholebris kingi was present occa- sionally during June, July and October of 1972; density never exceeded 1.5/l. Moina sp. appeared in late June of 1972 but was not found in the early part of July 1972. It appeared again in late July and remained in the pond through August and early part of September 1972. The peak popula- tion (44/1) was found on August 22, 1972. Chydorus sphaericus, the most abun- dant chydorid in the pond, was collected in most of the samples except those col- lected in late July and early August of 1972. The highest population density (43.5/l1) occurred on November 1, 1972. Other chydorids (Alona sp., Kurzia latis- sima, Leydigia quadrangularis and Pleu- roxus sp.) were collected occasionally in the samples and the numbers were insig- nificant. Simocephalus sp. was present in spring of 1973, and early summer of 1972 and 1973, but the density was below 1.0/1. Two species of Ceriodaphnia (C. la- custris and C. reticulata) were present in the pond, C. lacustris was predominant most of the time. The 1972 population of Ceriodaphnia had three peaks; the first on June 29 (227.1/l), the second on August 30 (145.7/1) and the last on September 21 (162.0/1). The population was completely absent from the plankton from November 26, 1972 until April 18, 1973. Brooks (1959) mentioned that the male of C. lacustris was not known from North America. However, some males of C. lacustris were found in the samples collected from the study pond. They appeared in late June, September, October and early November of 1972 along with ephippial females. Males and ephippial females also occurred in the population of C. reticulata. Three species of Daphnia (D. ambigua, D. parvula and D. pulex) occurred in the pond. D. ambigua predominated during the colder season from November of 1972 through early June of 1973. D. parvula predominated in the warmer season from early July through early October of 1972 and completely replaced D. ambigua from mid-June through July of 1973. The high- est numbers of D. parvula (173/l) were collected’.on August. 22, 1972. 1D) pulex coexisted with each of the other species but the numbers generally were much lower because the size of D. pulex is about twice that of D. ambigua and D. parvula, the low numbers of D. pulex were prob- ably the result of size-selective predation of fish (bass and bullheads) present in the pond (Hrbacek e¢ al., 1961; Brooks and Dodson, 1965; Galbraith, 1967; Brooks, 1968; Hall et al., 1970; Nilsson and Pejler, 1973). ‘The highest. density. (27/1) of pulex occurred on May 24, 1973. All three species of Daphnia produced males and ephippial females in this pond. Six genera (Asplanchna, Brachionus, Conochilus, Keratella, Lacane, and Pla- tyzas) and one unidentifiable species of rotifers occurred in the study pond. Only populations of Asplanchna were enumer- ated. Asplanchna occurred from July 16 through October 3 of 1972, then completely disappeared until April 22, 1973. This spe- cies was most abundant in early and mid May of 1973. The highest population density (266.3/l1) occurred on May 12, 1973. In 1973, a population pulse of an PopuLaATION DYNAMICS AND PRODUCTION OF Daphnia 697 unidentifiable species of rotifer occurred in March and a population pulse of Conochi- lus, lasted from May 26 through June 18. Chaoborus larvae were also present in the plankton from June through October of 1972. They were most abundant from mid August through mid October of 1972 but were completely absent between early November of 1972 and mid May of 1973. Population Data and Production Estimates Standing Crop The numbers of D. ambigua in the pond were very low from June through early August of 1972 (Fig. 3). The population 1000 500 100 NUMBER PER LITER 0.05 0.0! JUNE AucG. OCT. DEC. 1972 1973 Fes. APRIL JUNE Fic. 3. Standing crop of Daphnia ambigua in the Fish Laboratory Pond. Arrows indicate when the population was zero. disappeared on the 7th of June, reappeared on the 11th of June, and reached a density of 5.0/1 on the 3rd of July. The popula- tion declined rapidly, disappeared again after the 16th of July and reappeared again on July 27, and numbers increased rapidly to 45.8/1 by August 22. The numbers fluctuated considerably during September and October and then increased steadily / to 168.4/l on January 10, 1973. Chlorophyll a concentrations also increased at this time (Fig. 1). The population declined in late January when the concentration of food decreased, then increased again to a high population level on February 22. A rapid decline in population density during late March and early April was associated with a bloom of an unidentifiable species of rotifer. Following the end of the rotifer bloom in late March, the numbers of D. ambigua increased rapidly at the same time food concentration increased and reached a maximum population of 319.6/1 on April 26. A bloom of the colonial rotifer Conochilus in late May and early June, a low food concentration (Fig. 1), and rapidly rising temperatures (Fig. 2) were associated with the rapid decline of the D. ambigua population in early June. D. ambigua was not found after June 12 and was replaced by a population of D. parvula. Population Composition and Size Distribution The mean carapace length of females in the first adult instar, as determined from growth studies, was 0.721 mm (Lei, 1979). Although the smallest parthenoge- netic female with eggs found in the field samples was 0.649 mm (Fig. 6), individu- als equal to or larger than 0.721 mm with- out eggs or embryos were considered adults and those which were smaller were considered immature. On June 3 and June 5 no adults were ovigerous (Lei, 1979). The absence of recruitment was as- sociated with the disappearance of D. ambigua after June 5. Daphnia present in mid June were all immature. These animals probably hatched from ephippia. This ex-ephippial population died out completely after July 16. After about 10 days a new population, consisting entirely of immature animals, probably hatched from ephippia. However, as the number 698 Tue Universiry oF Kansas ScrENCE BULLETIN of reproductive females increased in the population, the population grew in the following months. Males were present in late June and early July, and again on November 26 of 1972 but no ephippial females were found. All males present in this period were im- mature. In 1973, males first appeared on January 15, and were present until the population disappeared in June. The per- centage of males present during the study period ranged from 0 to 29.6%. The high- est percentage of males occurred on April 18 (Lei, 1979). Ephippial females first ap- peared on March 24 nearly two months later than males did. The highest per- centage of ephippial females, 11%, ap- peared of April 22. The period of sexual reproduction lasted for about one month. Females and males were grouped sepa- rately into 32 size groups (Lei, 1979). The size range of individuals and the maxi- 100 a uJ = + 75 (og uJ a O Or50 uJ 25 JUNE AUG. OCT i972 I644 DEC Bee: mum size an individual attained varied considerably throughout the study period. Generally, the largest size range and the largest animals were found at lower tem- peratures (December through early April, Fig. 6). When water temperature in- creased after early April, the maximum size of individuals tended to decrease. Seasonal Variation in Egg Production and Body Size of Reproductive Females All references to clutch size and num- ber of eggs per adult female refer only to parthenogenetic eggs; ephippial eggs are excluded. Egg densities of D. ambigua during the summer and early fall of 1972 never exceeded 15.9/l. After mid October, egg densities increased rapidly to 471.1/l in January 1973 (Fig. 4). The increase of egg densities at this period was associated with a high mean clutch size (Fig. 5), a high standing crop (Fig. 3), and a large JURA) UNE Fic. 4. Seasonal variation in the egg density of Daphnia ambigua in the Fish Laboratory Pond. PopuLaTion Dynamics AND PRropuctTIoN oF Daphnia 699 e@ reproductive 9 ol O------0 adult 9 4 e e 2 Saw en oh O— g M4 cee NUMBER OF EGGS PER FEMALE ow JUNE OCT. ISf2 DEC. FEB. ? Q pe) sy e . ° : ee 3 as | SS zO e sts, 2a,” Be 3 APRIL JUNE ae) Fic. 5. Seasonal variation in the number of eggs per reproductive female and per adult female in the Fish Laboratory Pond. proportion of reproductive females in the population (Lei, 1979). The subsequent decline in egg numbers was associated with a decrease in the number of repro- ductive females; mean clutch size re- mained high. Egg numbers increased to a second peak in late February. A low, third peak of egg density occurred in mid April. This peak was associated with an increase in clutch size (Fig. 5); the number of repro- ductive females was low (Lei, 1979; Fig. 3). During late May, although mean clutch size was low, the combination of high population density and large propor- tion of reproductive females in the popula- tion produced a final peak of egg density (Bigs): Mean clutch size ranged from 1 to 5.7 with, an average of 201 (£1.12 S.D, N= 86), while the number of eggs per adult female ranged from 0 to 5.1 with an average of 1.09 (+ 0.98 S.D., N= 91). Highest values usually were associated with periods of population increase and high food concentrations (compare Figs. 1&5). The highest number of eggs per adult female and largest clutch size oc- curred in December 1972 and January 1973 (Fig. 5) when water temperature was low (Fig. 2) and food concentration was high (Fig. 1). Ephippial females did not appear when mean clutch size of females was 1.0. Dur- ing the month when ephippial females were present, the food concentration was higher than the period from May 31 to June 6 when mean clutch size was 1.0 and the number of eggs per adult female was low (Figs. 1 & 5). Because ephippial females of D. ambigua in the pond usually carried two resting eggs (occasionally three), food levels during the period when mean clutch size was 1.0 probably did not provide enough energy for the production of resting eggs. 700 Tue UNIversiry oF Kansas ScrENCE BULLETIN 1.30 4—~A largest female e mean 1.20 Oose o smallest female fo) 1.00 0.90 0.80 SIZE OF OVIGEROUS FEMALES (mm) 0.70 0.60 JUNE AUG. OCT: [SiGe DEC. FEB. APRIL JUNE I973 Fic. 6. Seasonal change in the carapace length of reproductive female Daphnia ambigua in the Fish Laboratory Correlation coefficients were deter- mined between body size and clutch size for 49 sampling dates; 34 values of r were significant (P = .05). All significant val- ues of r were positive (Lei, 1979). When mean clutch size was low, clutch size did not increase with increasing size of females carrying the eggs. Furthermore, during early and late September of 1972 and early June of 1973, all size groups of repro- ductive females carried only one egg in the brood chamber (Fig. 5). To test the relationship between food concentration and parthenogenetic egg production of D. ambigua, we calculated simple correlation coefficients between clutch size and available food and between egg number per adult female and avail- able food. Index of available food was expressed as chlorophyll @ (ug) or seston (cal) per liter at time t-n when the eggs were laid (where n is the time taken for eggs to develop at the respective tempera- ture). Chlorophyll @ or seston per liter was divided by the total number of Daph- nia, Ceriodaphnia, and Diaptomus present per liter at time t-n to get chlorophyll a or seston per animal. For comparison, the index of available food was also expressed as chlorophyll a or seston per liter and simple correlation coefficients were calcu- lated. There was significant, positive cor- relation between the parthenogenetic egg production of D. ambigua and the esti- mate of food concentration in the pond (lable 3): Parthenogenetic egg production of D. ambigua was negatively correlated with temperature (‘Table 3). The size of reproductive (parthenoge- netic) females in the pond fluctuated con- siderably throughout the season, and was Poputation Dynamics AND Propuction or Daphnia 701 generally larger during the winter and early spring when water temperatures were low and food concentrations were high (Figs. 1, 2, 6). Population Parameters Instantaneous birth rates (b’) fluctuated between 0 and 0.501 during the study pe- riod; highest values occurred in summer (Table 4). The instantaneous birth rate was significantly and positively correlated with temperature (r=0.38; df=101; P < 0.001) but was not correlated with estimates of food concentration (r = 0.14; df=101; P>0.1). Turnover time (of the number of individuals) ranged from 1.3 to 404.3 days. The highest calculated rate of popula- tion increase (r’) was 1.943 and the high- est rate of population decline was -1589 for the period studied. Estimates of the instantaneous death rate (d’) were ob- tained by subtracting values of r’ from b’. Because the value of d’ depends on the difference between two quantities already calculated with error, it is the least re- liable statistic associated with the egg ratio method (Edmondson, 1960). Thirty- nine negative death rates occurred among 100 estimates for the population of D. ambigua. Biomass and Production Population biomass ranged from 0 to 1.6 g dry wt/m?® (0 to 8.4 kcal/m*). Bio- mass was largest during winter and spring but was low in summer (Tables 4, 5). The mean biomass for the study period was 251.2 mg dry wt/m*® (1.3 kcal/m’*) (Table 5). Daily production was separated into growth and reproduction (Pg+ Pr) and exuvia (Pex). Daily Pg -+ Pr estimates cal- culated by the egg ratio method (ERM) ranged from 0 to 1575 mg dry wt/m* (0 to 783.8 cal/m*), and those calculated by Winberg’s method (WM) ranged from 0 to 129.3 mg dry wt/m? (0 to 6733 cal/ m*) (Lei, 1979). The maximum daily Pg+P, calculated by each method oc- curred on May 24, 1973. When egg density and percentage of females carrying par- thenogenetic eggs were high, the estimates of daily P, + P, calculated by ERM were generally larger than those calculated by WM. When egg density and percentages of females carrying parthenogenetic eggs were low, the estimates derived from ERM were smaller. Monthly P,-+P, estimated by ERM ranged from 1.3 mg dry wt/m* (65 cal/ m”*) in June 1972 to 1.8 g dry wt/m® (9.6 kcal/m*) in December 1972 while monthly P.-+ P, estimated by WM ranged from 12.2 mg dry wt/m® (56.3 cal/m*) in July 1972 to 15 g dry wt/m® (7.4 kcal/m?) in May 1973 (Table 6). The monthly values calculated by ERM were smaller than those calculated by WM for most months but were larger during the colder months (November 1972 through February 1973) when egg production was high. Although monthly estimates of Pg+P; differed considerably, estimates for the entire study period were very similar, 6.8 g dry wt/m* (34.7 kcal/m*) for ERM, and 6.6 g dry wt/m* (33.1 kcal/m*) for WM (Table 6). The P,+P,; estimates calculated by the two methods were averaged (Table 5). The daily production of resting eggs was calculated as (NepWep)/Tep for mg dry wt/m*/day and (NepWepCep)/Tep for cal/m*/day; where Nep is the number of resting eggs per cubic meter, Wep the mean dry weight of a resting egg and Tep the time period required for the for- mation of an ephippium (equivalent to the duration of the adult instar). Because dry weight and calorific value of resting eggs were not determined, mean dry weight and calorific value of parthenoge- netic eggs were used. The total produc- tion of resting eggs was obtained by plot- ting daily production for the period of 702 Tue University oF Kansas SctENCE BULLETIN sexual reproduction and determining the area under the curve planimetrically. Sex- ual reproduction of D. ambigua occurred in March, April and May of 1973; the total production of resting eggs was 66.4 mg dry wt/m® (390.8 cal/m*) which was about 1°% of the total production for the entire study period (Table 5). The P/B coefficients estimated from dry weight and from caloric content are similar (Table 5). The monthly P/B ratios ranged from 1.1 in February 1973 to 7.7 in August 1972, and were generally higher in the warmer months than in the colder months. The average daily P/B from June 1972 through June 1973 was 0.068 and from July 1972 through June 1973 was 0.066. Therefore, the average bio- mass turnover time was 14.7 and 15.2 days, respectively (Table 7). Daily production of exuvia (Pex) ranged from 0 to 233.8 cal/m* (Lei, 1979). Monthly Pex varied from 18.3 cal/m* in July 1972 to 2.8 kcal/m* in May 1973, and the total Pex from June 1972 to June 1973 and from July 1972 to June 1973 was 11.2 kcal/m*® (Table 6). Pex contributed from 9°% in December 1972 to 49% in June 1973 to monthly Pg + P; + Pex, and accounted for 25°4 of annual P. DISCUSSION The ash content of D. ambigua was within the range of ash content reported by other workers (Table 8). The ash values reported by Schindler et al. (1971), and by Wissing and Hasler (1968, 1971) were obtained by the loss of weight fol- lowing oxygen bomb combustion; there- fore, their ash values are underestimates. The ash contents reported by Moshiri and Cummins (1969) were determined by di- rect ashing in a muffle furnace. However, their values were considerably lower than the ash values of D. ambigua (Table 8). The calorific values of D. ambigua (cal/g dry wt and cal/g ash free dry wt), are in the range of values reported by other workers for other species of Daphnia (Table 8). The significant correlations between both egg density and population density of D. ambigua and food concentration suggest that the seasonal trends in these population parameters generally were de- termined by food. In this study high food concentrations occurred during low tem- peratures. At low temperatures the time of egg development increased which de- creased birth rate even though food con- centration was high. This relationship probably explains why food concentration and instantaneous birth rates were not significantly related. The disappearance of D. ambigua from the pond after June 12, 1973 could prob- ably be attributed to the high respiratory expenditure of daphnids in response to high temperatures (Armitage and Lei, 1979) and to decreasing food. During this period the concentrations of chlorophyll a and seston were low (Fig. 1) and the population densities of Ceriodaphnia and Diaptomus were high (Lei, 1979). A bloom of Aphanizomenon sp. also oc- curred during this period; blue-green algae are a poor food source for zooplankters (Schindler, 1971, Arnold, 1971). Competi- tion with colonial rotifers probably was significant. On several occasions we lost laboratory cultures of Daphnia, including D. ambigua, when a rotifer bloom devel- oped in the culture. High temperature and competition for food prevented daphnids from obtaining sufficient energy for both maintenance and reproduction and led to the complete dis- appearance of the population. However, the ability of D. ambigua to produce a high population and egg density in the cold season can be attributed to physio- logical adaptations. The filtering and res- piratory rate-curves of cold-acclimatized PopuLaTiIon Dynamics AND PRODUCTION OF Daphnia 703 field daphnids showed classical translation when compared to the rate-curves of warm-acclimatized animals (Armitage and Lei, 1979). Acclimatization enables D. ambigua to maintain high activity and utilize the high food concentrations dur- ing the colder season. Seasonal mean clutch size (2.01) of field D. ambigua obtained in this study is similar to the value reported for other populations of D. ambigua (1.97, Angino et al., 1973; 1.67, Kwick and Carter, 1975) but is much lower than the value obtained for a laboratory population (Lei, 1979). Because the laboratory population had much higher food concentrations, the low mean clutch size of field animals probably resulted from the low food concentration in the field. The brood size of Daphnia is positively correlated with the size of ovigerous fe- males both in laboratory cultures (Ander- son et al., 1937; Anderson and Jenkin, 1942; Green, 1954; Richman, 1958; Bui- kema, 1973; Lei and Clifford, 1974; Vijver- berg, 1976) and in field populations (Green, 1956; George and Edwards, 1974; Lei and Clifford, 1974). However, for the field population of D. ambigua, there was no consistent relationship between clutch size and size of females. Seemingly, the significant positive relationship between egg number and size of reproductive fe- males occurs only under favorable nutritive conditions. Clutch size and mean length of par- thenogenetic females fluctuated together for populations of D. magna (Green, 1956) and D. hyalina (George and Edwards, 1974). But for populations of D. schgdlert (Lei, 1968) and D. ambigua (Figs. 5 & 6), clutch size and the mean length of the parthenogenetic female did not fluctuate together. When environmental conditions, especially food concentration, change, a given-sized female may produce more eggs at one time than at another. Furthermore, under favorable food conditions, smaller females could produce more eggs than could larger females under poor food con- ditions. Therefore, the positive relationship between clutch size and the size of oviger- ous females might not hold when con- sidered over an entire season. The positive correlation between par- thenogenetic egg production and the esti- mate of food concentration is contrary to the negative correlation reported for D. ambigua in another pond in eastern Kan- sas (Angino et al., 1973). They attributed the negative relationship to the grazing effect of reproducing females on the algal population. Parthenogenetic egg produc- tion of D. ambigua was negatively cor- related with temperature (Table 3). Respiration increases at high temperatures (Armitage and Lei, 1979); probably less energy is available for egg production. However, in this study higher food con- centration during the colder season might contribute to the negative relationship be- tween egg production and temperature. A very low average number of eggs per adult female is expected in an equi- librium population which has reached the carrying capacity of its environment in terms of the food supply. At equilibrium only one egg, on the average, will be pro- duced by each animal; because the life expectancy of daphnids includes more than one reproductive instar, equilibrium brood size (eggs per adult female) must be less than one (Hall, 1964). The average brood size in an equilibrium, food-limited labora- tory population of Daphnia obtusa was less than 0.5 eggs/adult (Slobodkin, 1954). In this study, the population of D. am- bigua during most parts of June, July and September of 1972, and also most parts of March, May and part of April and June of 1973 probably was food limited and might have reached an equilibrium. Dur- ing these time periods, the mean number 704 THe Universiry oF Kansas ScrENCE BULLETIN of eggs per adult female was less than one and on certain dates approached zero. The larger size of the smallest repro- ductive females occurring during the cool season can be attributed to high food con- centrations and low temperatures. Size of primiparia and maximum size of Daphnia are larger at high food concentration (Green, 1954; Richman, 1958; Hall, 1964; Weglenska, 1971) and low temperature (MacArthur and Baillie, 1929; Lei, 1968, 1979). The larger mean size and maxi- mum size of reproductive females of D. ambigua occurring when temperatures were low resulted from the increased sur- vival and continued growth of adult fe- males. Larger size also may be associated with high food concentrations (Hall, 1964), low temperatures (MacArthur and Baillie, 1929), and low fish predation (Ap- plegate and Mullan, 1969). Respiratory expenditure might be a key factor deter- mining how large an animal can grow in a food-limited natural environment. Be- cause respiration of D. ambigua increases with increasing body size and temperature (Armitage and Lei, 1979), the high res- piratory rate of large individuals at high temperature could prevent large daphnids from obtaining enough energy for both growth and survival. Theoretically, death rate (d’) can not be negative because a negative death rate means that the observed rate of population growth (r’) exceeds the potential rate of population growth (b’). Therefore, in real populations, negative death rates re- sult from sampling error, immigration, or hatching of resting eggs (Edmondson, 1972). In this study, immigration may be eliminated because there was no source of immigrants. Therefore, the most prob- able cause is sampling error though hatch- ing of resting eggs may have contributed to negative death rates. Negative death rates occurred frequently in a population of Leptodora kindtu and other prey pop- ulations (Cummins ef al., 1969). The common negative death rate in the popu- lation of Daphnia retrocurva in Sunfish Lake, Ontario was attributed, in part, to the hatching of ephippia (Clark and Car- ter 1974): More zero values occurred among the daily Pg-+ Py estimates calculated by the egg ratio method (ERM) than among those calculated by Winberg’s method (WM) "Cer 1979). Becausey in “ERIVE production was calculated by multiplying the finite birth rate by standing crop bio- mass, the Pg + P; estimate was zero when birth rate was zero, even when the stand- ing crop biomass was large. Because birth rate was estimated from egg numbers, the birth rate was zero when eggs were absent. The estimates of zero production on dates when eggs were absent but daph- nids were present is probably unrealistic. Although there is no egg production (Pr), somatic production (Pg) probably results from the growth of individuals. There- fore, during the period when egg produc- tion was absent or extremely low, ERM probably underestimated production (e.g., June and July, 1972; June 1973). On the contrary, when egg production was high, ERM could overestimate production if the mortality of eggs was not considered. If eggs that are non-viable and/or lost with the death of ovigerous females are in- cluded in the calculation of birth rate, both birth rate and production will be overestimated. We do not know if the similarity of the results obtained for the whole study period by the two methods was fortuitous. If the similarity is real, then in a long term study (one growing season or one year) either of the two methods could be used to estimate produc- tion. However, in a short term investiga- tion (e.g., one week or one month), pro- duction estimated by the two methods would be very different. Depending on the intensity of egg production in the Poputation Dynamics AND Propuction or Daphnia 705 period studied, the values obtained by ERM would be either larger or smaller than the values obtained by WM. The P/B coefficient of 25 for the pe- riod July 1972 to June 1973 is very close to the annual P/B coefficient reported for Daphnia hyalina (20.8 to 25.9, George and Edwards, 1974) and for a Daphnia am- bigua-parvula complex (20.1 to 24.8) in Lake B (Geiling, 1969) but is lower than that reported for Daphnia cucullata (52.2, Petrovich et al., 1961) and for the D. ambigua-parvula complex (58.6 to 68.0) in Lake D (Geiling, 1969). A much shorter biomass turnover time of 3.8 days was obtained for the period July through August 1972. This short turnover time is close to that of 4 days for Daphnia galeata mendotae for the same months in Base Line Lake, Michigan (Hall, 1964). Mean daily P/B coefficient and turn- over time vary considerably among the same and/or different species of Daphnia in the same or different water bodies (Table 7). The daily P/B coefficient is equivalent to the turnover rate of biomass and reflects the reproductive capacity of a species under the conditions of its water body (Bekman and Menshutkin, 1964). Daily P/B coefficients for herbivorous planktonic crustaceans from lakes of dif- ferent trophic types tended to increase in proportion to the productivity of a lake (Patalas, 1970); the lowest value for lakes in Byelorussia was 0.06 for mesotrophic Lake Naroch. Daily P/B for herbivorous planktonic crustaceans in a thermally pol- luted lake was twice as high as that of a similar but unpolluted lake (Patalas, 1970). However, Pederson et al. (1976) reported that the daily P/B coefficients of the zoo- plankton were not related to the trophic status of several lakes in Washington. Turnover time, depending on the method of estimation, is the reciprocal of turnover rate of number of individuals or of bio- mass, and is also indicative of conditions at which a species lives rather than the inherent characteristics of a species (Stross et al., 1961). Therefore, the difference in the P/B coefficient and in the turnover time for populations of the same species may reflect differences in the temperature regimes (Hall, 1964) although nutrition (Weglenska, 1971) and other factors may also be involved. The differences in mean daily P/B coefficient and turnover time for D. ambigua seem to be related more to temperature than to nutrition (Table 7). ACKNOWLEDGEMENTS This research was supported by Grant A-043-KAN from the Kansas Water Resources Research Institute and by grants from The University of Kansas Compu- tation Center. The authors thank Dr. Harvey B. Lillywhite and Dr. W. John O'Brien for their critical comments. LITERATURE CITED AMERICAN Pusiic HEALTH AssociaTION. 1965. Standard methods for the examination of water and wastewater. American Public Health Association, Inc., New York. ANDERSON, B. G. AnD J. C. Jenkins. 1942. A time study of events in the life span of Daphnia magna. Biol. Bull. 83:260-272. , H. Lumer anv L. J. Zupancic, Jr. 1937. Growth and variability in Daphnia pulex. Biol. Bull. 73:444-463. AnciIno, E. E., K. B. ARMITAGE AND B. SAXENA. 1973. Population dynamics of pond zoo- plankton. II. Daphnia ambigua Scourfield. Hydrobiologia 42:491-507. AppLEGATE, R. L. anp J. W. Mutran. 1969. Ecol- ogy of Daphnia in Bull Shoals Reservoir. U.S. Dept. of Interior, Fish and Wildlife Service, Bureau of Sport Fisheries and Wild- life, Research Report 74. ArmitacE, K. B. 1961. Species composition and seasonal distribution of limnetic crustacean zooplankton of northeastern Kansas. Trans. Kan. Acad. Sci. 64:27-35. anp M. Davis. 1967. Population structure of some pond microcrustacea. Hydrobiologia 29 :205-225. —— anv C.-H. Ler. 1979. Temperature accli- matization in the filtering rates and oxygen consumption of Daphnia ambigua Scourfield. Comp. Biochem. Physiol. 62A:807-812. anp B. B. SmirH. 1968. Population studies of pond zooplankton. Hydrobiologia 32: 384-416. 706 Tue University oF Kansas ScrENcE BULLETIN ArNoLp, D. E. 1971. Ingestion, assimilation, sur- vival, and reproduction by Daphnia pulex fed seven species of blue-green algae. Limnol. Oceanogr. 16:906-920. BEKMAN, M. Yu. AND V. V. MENsHUTKIN. 1964. An analysis of the production process in simple population (in Russian). Zh. obshch. Biol. 25:177-187. (Cited in Winberg 1971.) Brooks, J. L. 1959. Cladocera, p. 587-656. In W. T. Edmondson (ed.), Freshwater Biology. John Wiley & Sons, New York. 1968. The effect of prey size selection by lake planktivores. Syst. Zool. 17:272-291. AND S. I. Dopson. 1965. Predation, body size, and composition of the plankton. Science 150:28-35. Burkema, A. L., Jr. 1973. Some effects of light on the growth, molting, reproduction and survival of the cladoceran, Daphnia pulex. Hydrobiologia 41:391-418. CasweLtt, H. 1972. On instantaneous and finite birth rates. Limnol. Oceanogr. 17:787-791. Ciark, A. S. anv J. C. H. Carrer. 1974. Popula- tion dynamics of cladocerans in Sunfish Lake, Ontario. Can. J. Zool. 52:1235-1242. Comira, G. W. 1972. The seasonal zooplankton cycles, production and _ transformations of energy in Severson Lake, Minnesota. Arch. Hydrobiol. 70:14-66. aND D. W. ScHINDLER. 1963. Calorific val- ues of micro-crustacea. Science 140:1394-1396. Cummins, K. W., R. R. Cosra, R. E. Rowe, G. A. Mosuir1, R. M. ScANLON AND R. K. ZAJDEL. 1969. Ecological energetics of a natural pop- ulation of the predaceous zooplankter Lepto- dora kindtu (Focke) (Crustacea: Cladocera). Oikos 20:189-223. Epmonpson, W. T. 1957. Trophic relations of the zooplankton. Trans. Amer. Microscop. Soc. 76:225-245. —. 1960. Reproductive rates of rotifers in natural populations. Mem. Inst. Ital. Idrobiol. L227 7, —. 1972. Instantaneous birth rates of zoo- plankton. Limnol. Oceanogr. 17:792-795. GaarDER, T. anD H. H. Gran. 1927. Investigations of the production of plankton on the Oslo Fjord. Rapp. Proces-Verbaux Reunions, Conseil Perm. Intern. Exploration Mer. 42:1-48. GasraitH, M. G., Jr. 1967. Size selective preda- tion on Daphnia by rainbow trout and yellow perch. Trans. Amer. Fish. Soc. 96:1-10. GeiLinc, W. T. 1969. Zooplankton production in a series of strip-mine lakes. Ph.D. Thesis, University of Missouri. GeorcE, D. G. anp R. W. Epwarps. 1974. Popula- ton dynamics and production of Daphnia hyalina in a eutrophic reservoir. Freshwat. Biol. 4:445-465. Guiwicz, Z. M. 1969. The food sources of lake zooplankton. Ekol. Pol. Ser. B 15:205-223. (Fish. Res. Board Can. Transl. Ser. No. 2716.) GREEN, J. 1954. Size and reproduction in Daphnia magna (Crustacea: Cladocera). Proc. Zool. Soc. Lond. 124:535-545. ——. 1956. Growth, size and reproduction in Daphnia (Crustacea: Cladocera). Proc. Zool. Soc. Lond. 126:173-204. Hacn Caratoc No. 9. Water and sewage analysis procedures. Hach Chemical Co., Ames, Iowa. HacH CHEMICAL ComMpANy MANuAL. 1969. Col- orimetric procedures and chemicals for water and wastewater analysis. Hach Chemical Co., Ames, Iowa. Harti, D. J. 1964. An experimental approach to the dynamics of a natural population of Daphnia galeata mendotae. Ecology 45: 94-112. ———, W. E. Cooper anp E. E. Werner. 1970. An experimental approach to the production dynamics and structure of freshwater animal communities. Limnol. Oceanogr. 15:839-928. Hicvsricut-ILKowska, A. AND T. WEGLENsKA. 1970. The effect of sampling frequency and_ the method of assessment on the production val- ues obtained for several zooplankton species. Ekol. Pol. 18:539-557. Hresacek, J.. M. Dvoraxova, V. KoriNEK AND L. ProcHazKova. 1961. Demonstrations of the effect of the fish stock on the species composi- tion of zooplankton and the intensity of metabolism of the whole plankton association. Verh. Internat. Verein. Limnol. 14:192-195. Hurcuinson, G. E. 1957. A Treatise in Limnology. I. Geography, physics and chemistry. John Wiley & Sons, New York. IvLev, V. S. 1934. Eine Mikromethode zur Bestim- mung des Kaloriengehalts von Nahrstoffen. Biochem. Z. 275:49-55. (Cited in Prus, 1970.) KuMeELevA, N. N. 1968. The expenditure of energy on respiration, growth, and reproduction in Artemia salina (L). (Transl. from Russian.) Biologiya Morya. Fiziologicheskie Osnovy Ekologi Vodnykh Zhivotnykh — 15:71-97. (Fish. Res. Board Can. Transl. Ser. No. 1226.) Kissy, H. V. 1971. Energetics and population dy- namics of Diaptomus gracilis. Ecol. Monogr. 41:3i1-32 7. KiekowsklI, R. Z., T. Prus anp H. ZyRoMSKA- Rupzka. 1967. Elements of energy budget of Tribolium castaneum (Hbst) in its develop- mental cycle, p. 859-879. In K. Petrusewicz (ed.), Secondary productivity of terrestrial ecosystems, Vol. 2. Warszawa-Krakow. Kwick, J. K. ano J. C. H. Carter. 1975. Popula- tion dynamics of limnetic Cladocera in a beaver pond. J. Fish. Res. Board Can. 32: 341-346. Ler, C. 1968. Field and laboratory studies of Daph- nia schddlert Sars from Big Island Lake, Alberta. M.Sc. Thesis, Univ. of Alberta. ——. 1979. Energetics of Daphnia ambigua Scourfield (Crustacea: Cladocera). Ph.D. Thesis, Univ. of Kansas. AND H. F. Criirrorp. 1974. Field and lab- Poputation Dynamics AND Propuction oF Daphnia 707 oratory studies of Daphnia schédleri Sars from a winterkill lake of Alberta. Publications in Zoology, No. 9. National Museums of Canada, Ottawa. Lewxowicz, M. 1971. Biomass of zooplankton and production of some species of Rotatoria and Daphnia longispina in carp ponds. Pol. Arch. Hydrobiol. 18:215-223. MacArtTuHuR, J. W. ano W. H. T. Baritire. 1929. Metabolic activity and duration of life. I. Influence of temperature on longevity in Daphnia magna. J. Exp. Zool. 53:221-242. Mann, F. H. 1969. The dynamics of aquatic eco- systems. Adv. Ecol. Res. 6:1-81. Mosuiri, G. A. 1968. Energetics of the predaceous zooplankter Leptodora kindtii (Focke) and selected prey species. Ph.D. Thesis, Univ. of Pittsburgh. AND K. W. Cummins. 1969. Calorific values for Leptodora kindtii (Focke) (Crustacea: Cladocera) and_ selected organisms. Arch. Hydrobiol. 66:91-99. Mutuican, H. F. anv J. M. Kincspury. 1968. Ap- plication of an electronic particle counter in analyzing natural populations of phytoplank- ton. Limnol. Oceanogr. 13:499-506. Nivsson, N. anv B. Peyter. 1973. On the relation between fish fauna and zooplankton composi- tion in North Swedish lakes. Rep. Inst. Fresh- water Res. Drottningholm 53:51-77. OstapENyaA, A. P. 1965. Polonota okisleniya or- ganicheskogo veshchestya vodnykh bespoz- vonochnykh metodom bikhromatnogo okis- leniya. Dokl. Akad. Nauk Belorussk. SSR, 9, no. 4, 273-276. (Cited in Winberg 1971.) AND A. I. SerceEv. 1963. The caloric con- tent of the dry substance of water invertebrates used as food by fish. (Transl. from Russian.) Voprosy Ikhtiologii, Vol. 3, No. 1, 177-183. (Fish. Res. Board Can. Transl. Ser. No. 874.) Paine, R. T. 1964. Ash and caloric determination of sponge and opisthobranch tissue. Ecology 45 :384-387. Patavancuuk, D. J. 1977. Winter fauna of Clado- cera and Copepoda in ponds and ditches, Lyon County, Kansas. Emporia State Res. Stud. 26:37-50. PALoHEIMo, J. E. 1974. Calculation of instantaneous birth rate. Limnol. Oceanogr. 19:692-694. Paratas, K. 1970. Primary and secondary produc- tion in a lake heated by thermal power plant. Proc. Inst. Environ. Sci. 16th Annual Techn. Meeting: 267-271. PECHEN, G. A. anp A. P. KUZNETSOVA. 1966. Con- cumption and utilization of food by Daphnia pulex (De Geer). Dokl. Akad. Nauk Belo- russk. SSR, 10:344-347. (Fish Res. Board Can., Transl. Ser. No. 872.) Peperson, G. L., E. B. Wetcu ann A. H. Linn: 1976. Plankton secondary productivity and biomass: their relation to lake trophic state. Hydrobiologia 50:129-144. PetrovicH, P. G., E. A. SHUSHKINA AND Gz A. PecHen. 1961. Evaluating zooplankton pro- duction. Doklady Biological Sciences (trans- lation of biological section of Doklady Akad- emii Nauk SSSR 139:1235-1238) 139:640-642. Pratt, T., V. M. Brawn anv B. Irwin. 1969. Caloric and carbon equivalents of zooplankton biomass. J. Fish. Res. Board Can. 26:2345- 2349. PRATHER, J. AND C. W. Propuer. 1969. Zooplankton species diversity in John Redmond, Marion and Council Grove Reservoirs, Kansas, Sum- mer, 1968. Emporia State Res. Stud. 18:5-16. PropHet, C. W. 1957. Seasonal variations and abun- dance of Cladocera and Copepoda and some physical-chemical conditions of the Fall and Verdigris Rivers in Wilson and Montgomery Counties, Kansas. Emporia State Res. Stud. 535-29. ——— anv S. Waite. 1974. A_ species list of Cladocera and Copepoda in Kansas. Trans. Ks. Acad. Sci. 77:42-47. Prus, T. 1970. Calorific value of animals as an element of bioenergetical investigations. Pol. Arch. Hydrobiol. 17:183-199. Reiners, W. A. AnD N. M. Reiners. 1972. Com- parison of oxygen-bomb combustion with standard ignition techniques for determining total ash. Ecology 53:132-136. RicuMan, S. 1958. The transformation of energy by Daphnia pulex. Ecol. Monogr. 28:273-291. ScHINDLER, D. W. 1968. Feeding, assimilation and respiration rates of Daphnia magna under various environmental conditions and_ their relation to production estimates. J. Animal Ecol. 37:369-385. —. 1972. Production of phytoplankton and zooplankton in Canadian Shield Lakes. pp. 311-331. In Z. Kajak and A. Hillbricht- Ilkowska (eds.), Productivity problems of freshwaters. Warszawa-Krakow. _ A. S. CrarkK anv J. R. Gray. 1971. Sea- sonal calorific values of freshwater zooplank- ton, as determined with Phillipson bomb calorimeter modified for small samples. J. Fish. Res. Board Can. 28:559-564. ScHINDLER, J. E. 1971. Food quality and zooplank- ton nutrition. J. Animal Ecol. 40:589-595. SHERsTYUK, V. V. 1971. Calorific value of food organisms in the Kremenchug Reservoir. Hy- drobiol. J. 7:85-88. SLopopKIN, L. B. 1954. Population dynamics in Daphnia obtusa Kurz. Ecol. Mongr. 24:69-88. Snow, N. B. 1972. The effect of season and animal size on the caloric content of Daphnia puli- caria Forbes. Limnol. Oceanogr. 17:909-913. SoxaL, R. R. ano F. J. Rowtr. 1969. Biometry. W. H. Freeman and Company, San Francisco. STRICKLAND, J. D. H. ano T. R. Parsons. 1968. A practical handbook of seawater analysis. Bull. Fish. Res. Board Can. No. 167. Stross, R. G., J. C. Negss AnD A. D. HAsLer. 1961. Turnover time and production of the plank- 708 Tue Universiry oF Kansas ScreNcE BULLETIN tonic Crustacea in limed and reference portion in the Soviet Union. Soviet J. Ecol. 3:295- of a bog lake. Ecology 42:237-245. 304. (English translation of Ekologiya 3: TuHayEerR, G. W., W. E. Scuaar, J. W. ANGELOVIC 5-18.) AND M. W. LaCrorx. 1973. Caloric measure- , G. A. PecHEN anv E. A. SuusHKina. 1965. ments of some estuarine organisms. Fish. Bull. The production of planktonic crustaceans in 71:289-296. three different types of lake. Zool. Zh. 44: VIJVERBERG, J. 1976. The effect of food quantity 676-688. (Transl. NIL RTS 6019.) ie ae Pe ee iii Sa es on Wissinc, T. E. anp A. D. Haster. 1968. Calorific SS etd A AR ee ec aa a aa values of some invertebrates in Lake Mendota, biologia 51:99-108. ; . Wisconsin. J. Fish. Res. Board Can. 25: Wec.LenskaA, T. 1971. The influence of various 2515-2518 concentrations of natural food on the develop- ; ; ment, fecundity and production of planktonic 1971. Intraseasonal change in caloric con- crustacean filtrators. Ekol. Pol. 19:427-473. tent of some freshwater invertebrates. Ecol- Winserc, G. G. 1971. Methods for the estimation Bey DB Veo Te: of production of aquatic animals. Academic Press, London and New York. . 1972. Investigations of the biological energy balance and the biological productivity of lakes Wricut, J. C. 1965. The population dynamics and in Canyon Ferry Limnol. Oceanogr. 10: production of Daphnia Reservoir, Montana. 583-590. TABLE. 1 CaLoriFIc VALUE oF Daphnia ambigua 1X CALORIES PER GRAM OF Dry WEIGHT (OR CALORIES PER GRAM ASH FREE DRY WEIGHT) AS DETERMINED BY OxyGEN BomMB CALORIMETRY AND WeET ComBusTION METHOD. OxyGEN Boms METHOD Wet ComeusTIon METHOD Analysis Size I Size Ill Eggs Size I Size II Size Ill (1) 1 4381 (5.97) 4833 (10.51) 5227 (0.79) 3632 (1.05) Se) ake) 4041 (0.84) 2 Ae 5589 2 4356 (5.20) 4533 (6.70) D3 51291) 3845 (2.38) 4001 (1.92) 4388 (1.67) 5561 2242 3 5067 (7.85) 5203 (1.77) 3830 (1.95) 3944 (1.80) 4242 (3.69) 5860 4 5130 (6.80) 3844 (2.00) 4079 (1.87) 4570 (1.20) 2935) 5 5055 (5.98) 3988 (1.54) 4278 (2.48) 4348 (2.02) 5846 6 3943) (1:23) 4627 (1.70) 7 4094 (2.20) (1) Mean 4369 4924 5188 3828 4041 4369 Std. dev. 17.68 245.56 47.72 127.05 122.13 215.16 Coef. of Var. 0.4% 5.0% 09%, B23, 310% 49% (1) Calories per gram dry weight; the number in the parentheses is weight of sample (mg) used _ for determination. (2) Calories per gram ash-free dry weight calculated on the basis of muffle furnace percent ash (see Table 2). AsH CONTENT TABLE 2 (PERCENT OF DRY Daphnia ambigua. Poputation Dynamics AND PropucTIon oF Daphnia WEIGHT) OF SIZE Exp. CLass No I 1 2 S x SD; II 1 Z 3 4 x tS} Deo Ill 1 2 3 4 5 6 x Sp Direcr OXYGEN IGNITION Boma (% Asx) (% Asx) 21.31 (2:91) 16.26 ( 4.49) 21.63 (2.50) 17.28 ( 5.34) 22.08 (2.60) 15.22 ( 5.20) 21.67 16.25 0.39 1.03 14.33 (3.49) - 18.47 (3.14) - 15.71 (3.50) - 9.30 (2.58) _ 14.45 - 3.84 - 14104 (4.70) 11.60 (10.51) 13.10 (4.20) 8.87 ( 5.98) 1346)(3.79) 11.08 ( 6.70) - 11.68 ( 9.02) - 11.59 ( 7.85) — 11.84 ( 6.80) 13.53 js | 0.47 iis Number in Parentheses = weight of sample used (mg). TABLE 3 THE CORRELATION BETWEEN PARTHENOGENETIC Ecc PRoDUCTION AND Foop CONCENTRATION (SESTON IN CALORIES, CHLOROPHYLL @ IN “G) AND TEMPERATURE. Variables r n Clutch size vs. seston/animal 0:345** — 86 Number of eggs per adult 2 vs. seston/animal 0.489*#* 9] Clutch size vs. seston/] 0.332** 86 Number of eggs per adult 9 vs. seston/] 0.452*** 91 Clutch size vs. chl. a/animal 0.601*** 86 Number of eggs per adult 2 vs. chl. @/animal 0.356** 91 Clutch size vs. chl. a/1 0.606*** 86 Number of eggs per adult 2 vs. chl. a/1 03214". (91 Clutch size vs. temperature —0.432** 86 Number of eggs per adult vs. temperature —0.282* 91 n == sample size; “p = 0:05;-**p = 0:01; +> < 0.001. 709 710 Tue Universiry oF Kansas Scr—ENcE BULLETIN TABLE 4 PoruLaAtion Data For Daphnia ambigua 1N THE FisH LABoratory Ponp. DEVEL- INsTAN- Popu- Turn- INSTAN- Ecos Eccs OPMENT FINITE TANEOUS LATION OVER TANEOUS MEAN STANDING CROP MEAN PER PER TIME BIRTH BIRTH CHANGE TIME DEATH reMp. No. PER MG PER CALORIES CLUTCH ADULT LITER (pays) RATE RATE RATE (DAYS) RATE DATE (C) LITER M? PER M® SIZE e) (E) (D) (B) (b’) (r’) (T) (d’) 1972 Jun 3 25.0 2.48 9.46 44.0 - 0 0 - 0 0 —1.467 - 1.467 25.2 0.13 0.60 2.8 - 0 0 - 0 0 = = 7 26.0 0 0 0 - - 0 - - - - - = OR 26:4 0 0 0 - - 0 -- - - - - = Pl 26:2 0.05 0.07 0.3 - - 0 - 0 0 - - = Sea 2o29 0 0 0 - - 0 - - - - - = 15 25.0 0.03 0.04 0.2 —- - 0 - 0 0 0.187 — -(.187 8223-5 0.06 0.06 0.3 - -- 0 - 0 0 1.065 — -1.065 2 235.2 37: 3.81 18.6 3.30 1.86 0.44 1.89 0.1243 0.149 —0.374 821102523 23 24.5 0.65 1.58 7 226d. 1033 0.03 1.74 0.0270 0.028 -0.076 37.0 0.104 28 25.3 OM4 0.65 29 ~ 0 0 0 a3 —- -1.943 29 25) 3.09 4.78 21.4 - - 0 - 0 0 0.210 — -0.210 Jul 12438 4.70 8.21 37.0 - 0 0 - 0 0 0.036 — -0.036 Sy PAS) 5.04 13.35 62.8 2.0 0.40 0:35 A74" 0:03:83 40:038, 0 26.1 0.038 pe 2422 0 0 0 - - 0 - - - - - = 9 24.6 0.03 0.12 0.6 — 0 0 - 0 0 —0.137 - 0.137 11 25.4 0.02 0.09 0.5 - 0 0 - 0 0 0.385 — 0.385 sy 52) 0.04 0.05 0.2 - - 0 - 0 0 —0.115 - 0.115 16 26.2 0.03 0.11 0.5 = - 0 - 0 0 - - = OM 277, 0 0 0 - - 0 - - _ - - = 23° 28:7 0 0 0 - - 0 - - - - - - 25) 28:2 0 0 0 - = 0 — — - - ~ = 27 e279 0.03 0.07 0.3 - - 0 = 0 0 1.246 — -1.246 29 2755 0.30 ls) a4 2108 2.08 0:09 1:49" 0514 0:175°—0:296 6.6 0.471 Sil 27-2 OL. 0.33 1S = - 0 - 0 0 0.418 —- -0.418 Aug 2 26.4 0.39 1.90 OES a 2e5 Ui 2250) 0.44 1.57 0.5797 0.489 0.332 eye MO) si7/ (eS) 2.02 9193 AOE Or e265) 6 a lL62 1.57 1.61 0.3208 0.357 —0.220 3:1 0577 8 26.4 1.62 915 46.7 2.86 2.39 1.94 1.57 0.4863 0.501 —0.061 2.1 0563 10 28.0 eae 7.07 SD 2e 223) le) 0.87 1.45 0.2433 0.327 -—0.623 4.1 0:950 2983 0.12 O15 0.7 - - 0 - 0 0 0.647 — -0.647 165295 0.43 este, 7.0 1:66 1-66 0.24 1.37 0.4493 0.322 0.635 2.2 -0.313 e297 155 6.69 53.67 72.20 2.20 T.2> 1536 1015830) 0436 0555 1.7 -0.118 20 29.8 4.68 16.79 titre) 9 Sines SalI 3.57 1.36 0.7790 0.418 1.140 1.3 -0.723 22 27M. W458 Mid INA ZEA 65 15.2 oly 0723; 0197-0274 5.8 0.471 AAS, “Zap ANS) 6041 295.4 3:19 1.66 7:34 Ves) 02213 0:224 —050211 Ao) OAS 28 24.4 14.71 51.39 2469 1.00 0.86 3.33 IW75 01531) (0817 1024 6.5 —0.407 30 24.4 41.94 144.19 689.6 1.00 0.92 8.60 1.75 0.0967 0.107 -0.197 10.3 0.304 Sep 2 24:2 23:23 67.9 318.6 1.00 0.20 0.90 1.77 0.0163 0.022 -0.583 61.4 0.605 >) 23:9 4.04 153517 74.7 1.00 0.06 0.11 1.81 0.0147 0.015 -0.045 67.9 0.060 1! 23.6 3.08 15.67 739 91-38 © 1-38 2.27 1.84 0.4106 0.300 0.600 2.4 —0.300 23.7 Nid:62 67.55 326.7 W47 106 5239) 83< O15 01s 08139. —08133 La 0272 15 23:7 16.29 61.64 298.4 1:47 1.05 472 ‘1.83: (0:1284 05139 —0.163 7.8 0.302 2328, oe EG 45.95) © 22516 142 30:9) 4.28) les2s OM876: FON 715 SOR83 5.3 —0.012 21 23.4 24.46 84.63 404.3 1.00 0.36 2.58 1.86 0.0478 0.054 -0.243 20.9 0.297 23° 22:8 15:04 63°59 306:2 1.00 0:53 2:88 1.94 030903) 01090 0/002 1 02089 25 219 15.08 64.04 306.6 1.00 0.22 139 2.07 0.0461 0.043 0.151 21.7 -0.108 27 21.0 20.40 85.19 407.3 1.00 0.04 0.36 2.21 0.0072 0.008 -0.178 138.1 0.186 S30 197, 1297 49-650 28:2" 1200) 10:23 0.82 2.46 0.0259 0.027 -0.083 38.6 0.110 Octes, Sa w8.5 934 4307 9 2227 9» Fic008 029 1.43 2.72 0.0510 0.052 -0.051 19.6 0.103 Se AVA 8.44 40.11 1965)" 12007 052 2.19 3.13 0.0710 0.074 -0.075 14.1 0.148 Se 13 6.75 33:24 164.0 1.14 0:49 2:00 3: 72- OA72 101070 0961 8.5 —0.891 10" 14:0 ~ 46:12 193.59 945.3 1291 054 11.05 4.28 0.0383 0.050 -0.567 26.1 0.617 12 11.6 14.84 68.90 338.6 1.42 0.95 6.16 5.67 0.0617 0.061 0.015 16.2 0.046 Mar Apr May Jun . No. PER LITER 16.0 31.07 2217 22.10 26.39 27.10 49°59 168.42 60.01 52.64 34.53 58.86 168.68 166.37 100.79 162.90 74.60 68.98 71.68 102.81 75.60 Sea 72.04 50254 60.03 S112 84.38 39.83 41.83 102.35 13%./3 204.84 150.39 156.54 29778 319.64 125.24 85.29 63.01 116.45 100.22 66.32 91.09 103.61 63.18 20.97 66.90 38.24 168.97 129.08 To.a4 71.30 DA 95.95 Poputation Dynamics AND Propuction or Daphnia STANDING CROP MG PER Mo 73.65 150.91 103.74 100.94 14474 144.48 483.78 1598.07 185.90 196.31 209.977 285.95 795.96 825.73 458.88 74.19 272.0 315.07 287.33 448.18 343.30 299°37 365.67 22994 33447 LIL 2S 48234 255.24 223.99 486.50 481.40 797.58 636.14 661.94 996.98 1111.87 42457 265.07 220.32 439.51 418.70 273.24 |29°45 344.07 210.53 104.02 300.07 177.84 808.14 474.64 295.71 252.88 366.11 383,94 CALORIES PER M® 365.0 15a 518.1 498.2 731.7 738.4 2574.9 8399.8 903.7 959.3 1063.6 1416.0 3940.6 4076.2 2249.0 2750.1 1297.2 15196 1364.8 2153.9 1642.8 1452.0 LAS e2 1115.6 1662.7 1734 2401.8 1292.8 2 2410.4 2300.9 3862.9 SIS 3244.5 4763.8 5265.5 199155 1231.8 1030.9 2068.1 2002.6 1335.0 1609.4 1653.1 095%) 501.8 1475.5 894.6 4020.5 2306.9 145.2 1183.4 Laer, MEAN CLUTCH SIZE 1B 1.50 1.57 1.75 2.30 3.30 Deh2 5.03 Sra 4.80 a79 4.39 3.76 3.96 3.21 2.19 es I) 1.46 es. 1.63 LOY, 1.85 1.68 1.69 1.55 eA 2.32 2.36 2.88 3.03 3.50 3:33 293 1239 1.08 1.06 1.20 1.00 1:35 V2 1:25 1.38 1.32 1.23 1.40 1.24 1.61 bye) 1.26 1.05 1.00 1.01 711 TABLE 4 (Continued) Deve.- INSTAN- Porvu- Tuan- INSTAN- Eccs Eccs OPMENT FINITE TANEOUS LATION OVER TANEOUS PER PER TIME BIRTH BIRTH CHANGE TIME DEATH ADULT LITER (pays) RATE RATE RATE (pays) RATE (E) (D) (B) (b’) (r’) (T) (d‘) 1.37 8.93 6.68 0.0690 0.066 0.074 14.5 -0.007 1.05 14.52 6.59 0.0567 0.058 -0.052 17.6 0.110 1.54 14.17 7.72 0.0625 0.063 -0.005 16.0 0.068 141 11.58 11.48 0.0372 0.037 0.025 26.9 0.011 1.98 24.27 15.11 0.0432 0.043 0.002 23.1 0.041 3.30 33.65 15.59 0.0526 0.052 0.032 19.0 0.020 5.13 164.36 14.64 0.1022 0.100 0.047 9.8 0.053 5.03 471.10 13.98 0.0863 0.096 -0.206 11.6 0.302 3.05 23.52 13.98 0.0235 0.024 -0.013 42.5 0.037 1.65 22.35 14.19 0.0245 0.025 -0.038 40.9 0.063 2.21 36.01 14.19 0.0519 0.050 0.059 19.3 -0.009 1.68 34.03 14.19 0.0344 0.032 0.132 29.1 -0.010 1.63 98.87 12.94 0.0356 0.036 -0.002 28.1 0.037 1.86 98.11 11.31 0.0390 0.041 -0.100 25.6 0.141 1.58 46.06 10.21 0.0387 0.037 0.096 25.8 -0.059 1.12 32.80 9.37 0.0162 0.020 -0.391 61.1 0.410 0.74 12.22 9.11 0.0163 0.017 -0.039 61.2 0.056 0.52 11.76 8.86 0.0180 0.018 0.019 55.7 -0.001 0.37 6.54 8.49 0.0113 0.010 0.180 88.8 -0.170 0.39 13.57 8.26 0.0139 0.015 -0.154 71.9 0.169 0.59 13.07 8.04 0.0187 0.020 -0.122 53.6 0.142 0.71 15.61 7.62 0.0322 0.031 0.098 31.0 —0.067 0.59 20.55 7.52 0.0306 0.033 -0.177 32.7 0.211 0.46 9.69 7.62 0.0237 0.023 0.057 42.2 -0.034 0.80 28.13 7.83 0.0478 0.049 -0.054 20.9 0.103 0.71 20.67 8.15 0.0473 0.042 0.251 21.1 -0.209 0.93 44.85 8.38 0.0424 0.051 -0.375 23.6 0.426 1.76 36.71 8.61 0.0768 0.076 0.025 13.0 0.051 1.77 27.50 8.86 0.0720 0.057 0.447 13.9 -0.390 2.17 65.96 8.26 0.0649 0.060 0.149 15.4 -0.088 144 34.07 7.22 0.0327 0.031 0.132 30.6 -0.102 1.22 63.55 6.51 0.0385 0.042 -0.155 26.0 0.196 2.19 76.38 5.81 0.0711 0.071 0.010 14.1 0.061 0.90 31.54 4.91 0.0441 0.037 0.322 22.7 -0.284 0.60 24.21 4.28 0.0186 0.018 0.035 53.8 -0.017 0.22 1845 3.96 0.0113 0.014 -0.469 88.4 0.483 0.14 4.48 3.84 0.0083 0.009 -0.192 120.0 0.201 0.07 1.10 3.76 0.0030 0.003 -0.303 338.9 0.306 0.10 1.44 3.68 0.0068 0.006 0.205 146.9 -0.199 0.38 10.51 3.49 0.0238 06.025 -0.075 42.0 0.100 0.78 23.32 3.32 0.0569 0.063 -0.207 17.6 0.270 0.78 19.44 3.23 0.0841 0.080 0.106 11.9 -0.026 1.12 34.14 3.16 0.1074 0.101 0.129 9.3 -0.028 0.79 21.67 3.07 0.0548 0.062 -0.247 183 0.309 0.38 5.23 3.01 0.0213 0.026 -0.445 46.9 0.471 0.71 6.59 3.04 0.0950 0.074 0.473 10.5 -0.399 0.97 30.09 2.82 0.1147 0.132 -0.280 8.7 0.411 141 26.83 2.56 0.2512 0.207 0.372 4.0 -0.164 1.53 117.92 2.54 0.1950 0.208 —0.135 5.1 0.343 1.05 41.69 2.54 0.1017 0.110 -0.163 9.8 0.273 0.32 8.56 2.48 0.0404 0.041 -0.053 24.8 0.094 0.11 1.85 2.40 0.0108 0.011 0.028 92.3 -0.017 0.49 17.72 2.34 0.0954 0.090 0.110 10.5 -0.020 0.27 9.42 2.26 0.0410 0.042 -0.055 24.4 0.097 1839.2 1.00 712 Tue Unrversiry oF Kansas ScreENcE BULLETIN TABLE 4 (Continued) DEVEL- INSTAN- Popu- Turn- INSTAN- Eccs Eccs oPpMENT FINITE TANEOUS LATION OVER TANEOUS MEAN STANDING CROP MEAN PER PER TIME BIRTH BIRTH CHANGE TIME DEATH Temp. No. PER MG PER CALORIES CLUTCH ADULT LITER (pays) RATE RATE RATE (DAYS) RATE Dare (C) LITER M3 PER M® SIZE 2 (E) (D) (B) (b’) (r’) (T) (d’) (Oy @7Ales) 84.18 331.82 1579.1 1.00 0.01 0.44 2.16 0.0025 0.002 0.068 404.3 —0.065 82109 96.35 423.49 2036.8 - 0 0 - 0 0 —1.273 - 1.273 10 22.9 12> 23.49 1095 — 0 0 - 0 0 —1.589 - 1.589 12 23.6 0.32 0.68 310) = - 0 - 0 0 - - - 14 23.5 0 0 0 - - 0 - - - - - - TABLE 5 MonTHLY AND ANNUAL MEAN BrioMass AND PRopucTION (P=P,-+P;) For D. ambigua. VALUES OF P ARE THE MEANS OF THE EsTIMATES PRovIDED BY THE EGG Ratio AND WINBERG METHODS. MONTHLY VALUES OF P oF REstING Eccs SHOWN SEPARATELY IN PARENTHESES. PRODUCTION OF REsTING Eccs Is INCLUDED IN THE AN- NUAL VALUES OF P, MEAN Biomass (B) PropucTion (P) P/B CoEFFICIENT mg dry wt/m*® —cal/m® mg dry wt/m*® —cal/m* mg/mg cal/cal 1972 June 1.8 8.2 8.1 37.4 4.6 45 July 1.8 8.4 6.9 32.1 3.8 3.8 Aug. 38.1 184.7 291.5 1428.7 thy Tell Sep. 56.1 269.6 202.8 997.8 3.6 57 | Oct. 86.3 423.8 184.3 914.8 2a 22 Nov. 123.5 621.6 183.9 946.7 15 eS) Dec. 483.8 2574.5 1273.8 6845.8 2.6 fH | 1973 Jan. 660.1 3420.9 1165.5 6179.8 1.8 1.8 Feb. 430.6 2140.1 469.3 2355.0 i 1.1 Mar. (0.29) (Gle7)) 396.1 1922.3 466.7 2280.0 12 12 Apr. (65.7) (386.6) 546.2 2643.5 903.3 4370.1 1.8 1.8 May (0.43) (25) 352.3 1606.6 1296.4 6395.1 3.9 4.0 June 109.3 524.7 233.9 1105.5 21 21 1972 June to 1973 2512 1257.6 6753.0 34279.9 26.9 273 June 1972 July to 1973 272.0 1361.7 6744.8 34242.6 24.8 25.1 June MonTHLY PRopUCTION AND ToraL PropucTion For D. ambigua. Date 1972 June July August September October November December 1973 January February March April May June June 1972-June 1973 July 1972-June 1973 CoMPARISON OF MEAN Biomass, MEAN Darty Propuction, MEAN DaILy SEVERAL SpEcIEsS OF Dap/nia. Poputation Dynamics AND Propuction or Daphnia 713 TABLE 6 RESTING EGGS NOT INCLUDED. ERM = Ecc RATIO METHOD, WM == WINBERG'S METHOD. Buopnenon Pe a Py Pex ERM WM mg/m* cal/m* mg/m* cal/m* cal/m* 163 6.5 14.9 68.2 19.9 1.6 7.8 12.2 56.3 18.3 226.1 1106.8 356.9 1750.6 504.2 126.0 659.3 279.7 1336.2 576.9 1717 847.3 197.0 982.2 382.8 214.4 1092.8 153.4 800.5 200.0 1808.5 9566.0 739.0 4125.5 641.6 1551.6 8125.9 779.3 4233.7 794.3 495.3 2460.1 443.4 2251-1 729.2 380.6 1860.7 552.7 2699.3 1147.9 6273 3071.0 1179.4 5669.3 2384.9 1101.3 5419.5 1491.5 7370.6 2752.8 95.0 458.8 372.8 L522 1058.8 6800.8 34682.5 6572.3 33095.7 11211.8 6799.4 34676.0 6557.4 33027.4 11191.8 TABLE 7 MEAN TURN- MEAN MEAN MEAN OVER PERIOD Biomass DAILY DAILY TIME OF SPECIES (B) PRODUCTION P/B_ (pays) StuDy D. sch¢dleri 1.568 g/m? 0.227g/m* 0.145 69 Apr~Sep 1958 D. galeata Tale 0.114 0.10 10.0 Apr~Sep 1958 mendotae D. cucullataand 0.950 g/m? 0.128g/m° 0.135 7.41 Jul~Aug 1966 D. longispina hyalina DoT, 1547 0.281 3.62 Jul~Aug 1966 D. hyalina 0.570 g/m? 0.027 g/m* 0.047 21.3 Jan~Dec 1970 0.320 0.020 0.063 15.9 Jan~Dec 1971 D. cucullata 1.690 g/m* 0.337g/m*> 0.199 5.0 Jul~Aug 1964 D. ambigua- 0.478 g/m? 0.048 g/m? 0.100 10.08 Mar~Nov 1965 parvula complex 0.367 0.037 0.100 10.0 Mar~Nov 1966 0.064 0.016 0.250 4.0 Mar~Nov 1965 0.117 0.028 0.239 4.24 Mar~Nov 1966 D. galeata 0.407 g/m* May~Nov 1966 mendotae 0.030 May~Nov 1967 MEAN MEAN WATER SESTON TEMP. 21-6 27.4 19.5 15.0 14.0 CONC. (C°) (caL/L) Wright ( Wright ( P/B, AND MEAN TURNOVER TIME FOR SouRCE 1965) 1965) Patalas (1970) Patalas (1970) George and Edwards (1974) Hillbricht-Ilkowska & Weglenska (1970) Geiling ( 1969) Cummins et al. (1969) 714 Tue University oF Kansas ScrENCE BULLETIN TABLE 7 (Continued) D. cucullata D. ambigua D. ambigua D. galeata mendotae D. galeata mendotae . longispina . pulex Sv D. longispina D. cucullata D. cristata 0.064 ¢/m* 0.019 ¢/m® 0.038 0.016 g/m* 0.005 ¢/m* 0.004 0.018 0.017 0.025 0.009 g/m* 0.005 g/m* 0.360 g/m? 0.48 053 0.057 g/m* 0.064 0.009 0.25 0.263 0.105 0.066 0.068 0.069 0.15 0.19 0.21 0.095 0.091 0.087 0.31 0.29 0.23 0.16 4.0 5-6 3.8 5) 15.2 Gi 1h 4.0 147 32:2 6.7% 5.38 apt 10.58 11.0°® 11.510 3.211 See 4,312 6.212 Apr~Nov May~Nov Jul~ Aug Jun~Nov Jul 1972~ Jun 1973 Jun 1972~ Jun 1973 Jan~Jun Jul~ Aug Jul~ Aug Jul~ Aug Jun~Sep Jun~Sep Jun~Oct May~Sep May~Sep May~Sep 1960 1974 1972 1972 1973 1966 1967 1961 1956 1956 1967 1962 1962 1960 1968 1969 1968 1969 135 25.8 Petrovich et al. (1961) Kwick and Carter (1975) Present study Cummins et al. (1969) Hall (1964) Stross et al. (1961) Lewkowicz (1971) Winberg etal. (1965) Winberg (1972) 1 Lake Mikorzynskie (no thermal effluents). 2 Lake Lickenskie (receives thermal effluents). 3 Lake B of a series of stripmine lakes in Missouri. 4 Lake D of a series of stripmine lakes in Missouri. 5 Carp pond (unfertilized). 6 Carp pond (fertilized with phosphorus fertilizer). 7 Carp pond (fertilized with phosphorus and nitrogen). 8 Lake Batorin (Eutrophic). 9 Lake Myastro (Mesotrophic). 10 Lake Naroch (Oligotrophic). 11 Lake Krivoe. 12 Lake Krugloe. SPECIES Daphnia magna Daphnia magna Daphnia magna Daphnia magna Daphnia pulex Daphnia pulex Daphnia pulex Daphnia pulex Daphnia pulex Daphnia galeata mendotae Daphnia dubia Daphnia spp. Daphnia spp. Daphnia pulicaria Daphnia ambigua Poputation Dynamics AND Propuction or Daphnia 715 SPECIFICATION Mix eggless and with egg 1.0~2.5mm juvenile 2 2 with eggs 0.7, 1.3, 1.8mm juvenile reproductive animal 1.92mm eggless with eggs eggs only females with eggs mean of all size groups (1.1~3.1mm) <0.62mm (carapace length) 0.62~0.87mm (carapace length) >0.87mm (carapace length) (adult) eggs only B-Bomb Calorimetry. C-Chemical Composition. W-Wet Combustion. *-Mean of all size groups in Table 2 of Snow (1972). TABLE 8 CALorIFic VALUES oF Daphnia Species. Cat/c Ca./c AsH- METHOD DRY WT °% ASH FREE DRY WT REFERENCE Cc 3694 33.17 4988 Ivlev (1934) Cc 3700 —_— 4967 Fischer, unpublished (after Prus, 1970) B 5640+60 TA=02 5898+63 Moshiri & Cummins (1969) B 4852~4925 — ~- Schindler (1968) WwW 4330 = _- Pecken & Kuznetsova (1966) W 4640 = — Pecken & Kuznetsova (1966) CG 4365 18.25 5674 Ivlev (1934) (after Prus, 1970) B 4419 “= 5362 Richman (1958) C 4141 17.6 —_ Birge and Juday (1920) (after Richman, 1958) ce 5350 Birge and Juday (1920) (after Richman, 1958) B 4478+372 - -- Comita and Schindler (1963) B 511889 770.2 5511485 Moshiri and Cummins (1969) B 537254 $3222 5817458 Moshiri and Cummins (1969) B 5850 4.0 6098 Moshiri and Cummins (1969) B 4767 8.36 331 Schindler, et al. (1971) B 4630 8.99 5030 Wissing & Hasler (4170~5115) (5.9~13.0) (4648~5395) (1971) B -- 9.78 5006 Wissing & Hasler (5.5~13.4) (4132~5643) (1969) B 4532 6.28 Wissing & Hasler (3776~5652) (3.8~9.0) (1969) B 399" _ — Snow (1972) B 4369+18 21.67 5D /8==23 Present Study W 45832139 14.45 53574162 Present Study B 49244246 13525 5694+285 Present Study W 588354 “= — Present Study Date Due ae aay - Acme Bookbinding Co., Inc. 100 Cambridge St. | Charlestown, MA 02129 wn nonwunniy 093 362 3 CF te Te IE oe Me Lar i fa ae Seater ee Oe ose ‘ » ! . . re eee eee sega ¢ epee it bis 5 i er Bee fae nlf oo pM te Foe Seog ey rays ve rk Shik BB?! . oe sade se é is verte was as vhs z i ' eke . ' ve . , . oret,’ 4 ‘4 Me : . ¥ ee eovue ‘ : . os Ce rere woe t ’ ‘ ; rte F ) ot mee ; ‘ Far wee i E SR OOM ’ v ae ae * ee of r J ; apne at oO none « ’ per de® ; ; . wae oF Pa cine We eM) fe , TF. ; ‘ * ¢ 5 ( : jee 4 4 . " , , . Beg Gt Galt tind ‘ 7 re : t A Py ROC ee ace é : fe ‘ i of ‘ ¢ ; fi 5 ; YI : A t er ee et gre « a ; Fy obeh ter + 9 Pew) tre ae bd ie FOR oe, é te : podem acta ts 4 r ‘ , , ‘i a AP nN ‘ 8 Se vet 2 . oy. i ieee j rs ree ye pao sdree fa nie ® ; Bert Mt oS Le hve A ie 2 DER ELMAR L. Pe ¢ A he Pee 29 VAS a 4 VARA AARC EME rs af : A ee "4 , i { \ LAF AOL MEBY ‘ seat teats : i : ie ofa : - y. - io ar. “ 2 , ‘ ee : ' ’ we Orne : : KF i F 1 } b, As y Seay er tina 4 _ : ' y , eed p a n , ; . < r A . . Spa as 2 ; J i A, i ueaos ~ f re 5 F foe . f Wei area e x 2 eae 4 24 : A ‘ ‘ p