De eewanate we + bet mn a badass et bapa erat wens gateererenttrtetantifetet ia " es i rabyeeersicgecemia tenet he Fite hn tat oe ee eee apenweneT™ Sth han aed re * - Ppt aha rer: Se i tebe rea resentite Ph HAS APA ISS ie * “eps ee ivi < Akg heat gaat ow ane maherye vi = ere mi benye a 3 oy eyean poveree aver Canal As — apd calete trai hilt Uke Aegis UO MS Be TTT bathe teed ante po + ae hh vn ip alate. Pa CAE CAPM tr oct a tac Ot ttn a SG an 4 " “ gag ee tt rh A ern me Ps Te €- = q ny _ ye"9 ably ae Le sis ie Binks ats ie WOR OTD ARES eer eeetee ee hited ret Fo fen het bol x Terve Dp etre ES ee — A ° o ‘ + anne eee Niagra | eee tt eben remiss aan aan Ft ‘ pies piperene amare ik ht pane Sabet es « ue ner ia wteserer - é Soe sanlprerranr Coe a epee rena at Ao Paes extagen te Yarra era te rene tueiver aT om en 7 sl we dy. 7] ti n * - aed er ar ta cbt LIE OE Ae en = "i - Ska ifs Dae amare a RR a a elit il Nad ISIN ree ips Pos a ae pehe niet. Hestiir® . por potet ey res comes Bi ee recente Cone par tain VE eNO Sew cabetasoe aac toniat TC TN etait esr ry = Sa ae ETE ATTY Si OTE "TiS ATRL OF HG AI CO ee —_ : et qneere ts ymameeta Nero jen oda ta Dace Ai ple Ae eA ENE rece Oe ere . eg we . ee: ¥ a 27 TEEN eT Se” MPLS TED. re + = . Pet, = Pe a Pte mene LA by Aras, ss 5° Ae Fa Gare a ay Y me rt fe 7¢6s AMERICAN MUSEUM NOVITATES ————___—— NUMBERS 1 TO 36 NEW YORK PUBLISHED BY ORDER OF THE TRUSTEES 1921-1922 THE AMERICAN MUSEUM OF NATURAL HISTORY SEVENTY-SEVENTH STREET AND CENTRAL PARK WEST New York City BOARD OF TRUSTEES (As of September 30, 1922) PRESIDENT HENRY FAIRFIELD OSBORN Fiest Vice-PRESIDENT Srconp Vics-PRESIDENT CLEVELAND H. DODGE J. P. MORGAN TREASURER SECRETARY GEORGE F. BAKER, Jr. PERCY R. PYNE EX-OFFICIO THE MAYOR OF THE CITY OF NEW YORK THE COMPTROLLER OF THE CITY OF NEW YORK THE PRESIDENT OF THE DEPARTMENT OF PARKS ELECTIVE GEORGE F. BAKER ARTHUR CURTISS JAMES FREDERICK F. BREWSTER WALTER B. JAMES THOMAS DeWITT CUYLER OGDEN MILLS WALTER DOUGLAS A. PERRY OSBORN CHILDS FRICK _ GEORGE D. PRATT MADISON GRANT THEODORE ROOSEVELT WILLIAM AVERELL HARRIMAN LEONARD C. SANFORD ARCHER M. HUNTINGTON JOHN B. TREVOR ADRIAN ISELIN FELIX M. WARBURG ADMINISTRATIVE OFFICERS (As of September 30, 1922) DrrEcTOR EXeEcuTIvE SECRETARY FREDERIC A. LUCAS GEORGE H. SHERWOOD ASSISTANT TREASURER THE UNITED STATES TRUST COMPANY OF NEW YORK iii SCIENTIFIC STAFF (As of September 30, 1922) Freperic A. Lucas, Se.D., Director Rosert C. Murpay, Assistant to the Director (in Scientific Correspondence, Exhibition, and Labeling) James L, Cuark, Assistant to the Director (in Full Charge of Preparation) DIVISION OF MINERALOGY AND GEOLOGY W. D. Marruew, F.R.S., Curator-in-Chief GEOLOGY AND INVERTEBRATE PALZONTOLOGY Epmunp Ot1s Hovey, Ph.D., Curator Curster A. Reeps, Ph.D., Associate Curator of Invertebrate Paleontology MINERALOGY Herpert P. Wauittock, C.E., Curator Grorce F. Kunz, Ph.D., Research Associate, Gems VERTEBRATE PALHONTOLOGY Henry Farrrietp Osporn, LL.D., D.Sc., Honorary Curator W: D. Marruew, Ph.D., Curator WaTEeR GRANGER, Associate Curator of Fossil Mammals Barnum Brown, A.B., Associate Curator of Fossil Reptiles Wit1i1am K, Grecory, Ph.D., Associate in Paleontology DIVISION OF ZOOGEOGRAPHY Frank Micuier Cxapman, N.A.S., Curator-in-Chief LOWER INVERTEBRATES Roy W. Miner, A.B., Curator p Wiiiarp G. Van Name, Ph.D., Assistant Curator Frank J. Myprs, Research Associate, Rotifera Horace W. Srunkarp, Ph.D., Research Associate, Parasitology A. L. TREADWELL, Ph.D., Research Associate, Annulata ENTOMOLOGY » Frank E. Lutz, Ph.D., Curator A. J. Mutcuusr, Assistant Curator in Coleoptera Frank E. Watson, B.S., Assistant in Lepidoptera Joseru Bequaert, Ph.D., Assistant in Congo Zodélogy Cuaries W. Lena, B.S., Research Associate, Coleoptera Hersert F, Scuwarz, A,M., Research Associate, Hymenoptera Wiiuiam M. Wueeter, Ph.D., Research Associate, Social Insects iv SCIENTIFIC STAFF v IcHTHYOLOGY Basurorp Dzan, Ph.D., Honorary Curator Joun T. Nicuots, A.B., Associate Curator of Recent Fishes E. W. Gupaer, Ph.D., Associate in Ichthyology HERPETOLOGY G. K. Nostz, A.M., Associate Curator (In Charge) y ARTHUR ORTENBURGER, M.S., Assistant Curator ORNITHOLOGY Frank M. Cuapman, Se.D., Curator W. DeW. Mutter, Associate Curator Rosert CusHMAN Murpay, D.Sc., Associate Curator of Marine Birds JamEs P. Cuapin, A.M., Assistant Curator, African Birds Luptow Griscom, M.A., Assistant Curator JONATHAN Dwiaut, M.D., Research Associate in North American Ornithology Mrs. Etsre M. B. RetcHENBERGER, Research Assistant MAMMALOGY | Roy C. Anprews, A.M., Associate Curator of Mammals of the Eastern Hemisphere H. E. Antuony, A.M., Associate Curator of Mammals of the Western Hemisphere Hersert Lane. Assistant Curator, African Mammals Cart E. AKetry, Associate in Mammalogy CoMPARATIVE ANATOMY Wii K. Gregory, Ph.D., Curator S. H. Cxuss, Assistant in Osteology J. Howarp McGrecor, Ph.D., Research Associate in Human Anatomy DIVISION OF ANTHROPOLOGY Ciarxk WissterR, Ph.D., Curator-in-Chief ANTHROPOLOGY CuarkK WissLeER, Ph.D., Curator Pury E. Gopparp, Ph.D., Curator in Ethnology N. C. Netson, M.L., Associate Curator of North American Archzology Cuar.es W. Mean, Assistant Curator of Peruvian Archzology Louis R. Sutiivan, A.M., Assistant Curator, Physical Anthropology CiarEeNcE L. Hay, A.M., Research Associate in Mexican and Central American Archeology CoMPARATIVE PHYSIOLOGY Rape W. Tower, Ph.D., Curator vi SCIENTIFIC STAFF : CoMPARATIVE ANATOMY Wiuuiam K. Grecory, Ph.D., Curator J. Howarp McGrecor, Ph.D., Research Associate in Human Anatomy DIVISION OF EDUCATION, BOOKS, PUBLICATION, AND PRINTING « George H. Suerwoop, A.M., Curator-in-Chief LIBRARY AND PUBLICATIONS Raupx W. Tower, Ph.D., Curator Ipa Ricwarpson Hoop, A.B., Assistant Librarian Pusiic EpucatTion GrorcE H. Suerwoop, A.M., Curator G. Ciypr Fisuer, Ph.D., Associate Curator Ruts Crossy Nostsz, B.A., Assistant Curator Grace FisHEerR Ramsey, Assistant Curator Pusiic HEALTH CxHarLEs-Epwarp Amory Winstow, D.P.H., Honorary Curator Mary Gree, Assistant CONTENTS OF AMERICAN MUSEUM NOVITATES NumBErRs 1 To 36 PagE IS Renin yaa eral, OU cc cingeaes's «ac og ska else hae aaa ale Muga Ted i RAE MMMM a OOS Ba, 5g wan ow cs eau Re EN eo ee tan tee iii MRI ot FO See ts oP oo ae ean Leben tee ene ak iv is Ne Oa Soa Sas «yup tome ols aeatnie eee COW vii DREN See as. vac & Sao a's wb pine mee na eens wks ix List of New Taxonomic Names....... xi EN Fale a ee Sc cain wb be ceGe ON ecg. ¥ oe ae a cewbee xv No. 1.—‘The Evolution, Phylogeny, and Classification of the Proboscidea.’ By a Henry Fairfield Osborn. 15 pp. (Four text figures.) January 31, 1921. 2.—‘ Descriptions of Apparently New Birds from Bolivia, Brazil, and Vene- zuela.’ By Frank M.Chapman. 8pp. January 31, 1921. 3—A Hawaiian Race of Carangoides gymnostethoides.’ By John Treadwell Nichols. 3 pp. (One text figure.) March 9, 1921. 4.—‘The Bony Structure and Phyletic Relations of Sphzrodactylus and Allied Lacertilian Genera, with the Description of a New Genus.’ By G. K. Noble. 16 pp. (Eight text figures.) March 10, 1921. 5.— Geographic Average, a Suggested Method for the Study of Distribution.’ By Frank E. Lutz. 7 pp. March 14, 1921. 6.—‘ Description of a New Species of Serow from Yiin-Nan Province, China.’ “ By Roy Chapman Andrews. 3 pp. March 24, 1921. 7.— Descriptions of Four New Birds from the Belgian Congo.’ By James P. Chapin. 9pp. (Three text figures.) April 4, 1921. 8.—‘Polychxtous Annelids Collected at St. Paul de Loanda by the American Museum Belgian Congo Expedition.’ By A. L. Treadwell. 3 pp. (Eight text figures.) June 3, 1921. 9.—‘A Remarkable Case of External Hind Limbs in a Humpback Whale.’ By Roy Chapman Andrews. 6 pp. (Four text figures.) June 3, 1921. 10.—‘ First Appearance of the True Mastodon in America.’ By Henry Fairfield - Osborn. 6 pp. (Two text figures.) June 15, 1921. 11.—‘The Geology about Mills Springs, Monticello Quadrangle, Kentucky.’ By Edward J. Foyles. 10 pp. (Four text figures.) June 17, 1921. 12.—‘Notes on North American Blood Flukes.’ By Horace W. Stunkard. 5 pp. July 16, 1921. 13.— New Genera of Paleocene Mammals.’ By W. D. Matthew and Walter Granger. 7 pp. September 6, 1921. 14.—‘ Stehlinius, a New Eocene Insectivore.’ By W. D. Matthew. 5pp. (Two text figures.) September 7, 1921. 15.—‘A New Name for a Subspecies of Uta stansburiana Baird and Girard.’ By Karl Patterson Schmidt. 2 pp. September 8, 1921. 16.—‘A Revision of Atlapetes gutturalis with Descriptions of Three New Races.’ By Jonathan Dwight and Ludlow Griscom. 4 pp. September 9, 1921. vii viii CONTENTS OF AMERICAN MUSEUM NOVITATES “ . 17.—‘ Notes on a New Ox-pecker and Other Little-known Birds of the Congo.’ By James P. Chapin. 16 pp. (Six text figures.) September 16, 1921. 18.—‘ Descriptions of Proposed New Birds from Colombia, Ecuador, Peru, and Brazil.’ By Frank M. Chapman. 12 pp. September 22, 1921. 19.—‘New Mammals from British Guiana and Colombia.’ By H. E. Anhoar 7 pp. October 26, 1921. 20.—‘ Preliminary Report on Ecuadorean Mammals. No.1.’ By H. E. Anthony. 6 pp. November 3, 1921. 21.—‘Some Parasitic Megachilid Bees of the Western United States.’ By T. D. A. Cockerell. 11 pp. December 1, 1921. 22.—'New Species of North American Lizards of the Genera Holbrookia and Uta.’ By Karl Patterson Schmidt. 6 pp. December 1, 1921. 23.—‘The Epeoline Bees of the American Museum Rocky Mountain Expedi- tions.’ By T. D. A. Cockerell. 16 pp. December 5, 1921. 24.—‘ Western Bees Obtained by the American Museum Expeditions.’ By T. D. A. Cockerell. 15 pp. December 7, 1921. 25,—! Descriptions of Proposed New Birds from Central America, with Notes on — | Other Little-known Forms.’ By Waldron DeWitt Miller and Ludlow Griscom. 13 pp. December 7, 1921. 26.—‘A New Pomacentrid and Blenny from the Bahamas.’ By John Treadwell Nichols. 2 pp. (Two text figures.) December 14, 1921. 27.—‘ Descriptions of Proposed New Birds from Brazil, Paraguay, and Argen- tina.’ By George K. Cherrie and (Mrs.) Elsie M. B. Reichenberger. '6pp. December 28, 1921. 28.—‘New Species of Ammonite Opercula from the Mesozoic Rocks of Cuba.’ By Marjorie O’Connell. 15 pp. (Eighteen text figures.) December ‘30, 1921. 29.—‘Five New Species of Salientia from South America.’ By G. K. Noble. 7 pp. (Six text figures.) December 30, 1921. . 30.—‘The Distribution of the Swallows of the Genus Pygochelidon.’ By Frank M. Chapman. 15 pp. (Two text figures.) February 28, 1922. ; 31—‘Descriptions of Apparently New Birds from Colombia, Ecuador, and Argentina.’ By Frank M. Chapman. 8 pp. March 2, 1922. 32.—‘Preliminary Report on Ecuadorean Mammals. No.. 2.” By H. E. Anthony. 7 pp. (One text figure.) March 4, 1922. 33.—‘ Bees of the Genus Perdita from the Western States.’ By T. D. A. Cockerell. 15 pp. March 28, 1922. 34.—‘A Fossil Moth from Florissant, Colorado.’ By T. D. A. Cockerell. 2 pp. — March 29, 1922. 35.—‘A New Fossil Rodent from Ecuador.’ By H. E. Anthony. 4 PP. (Two text figures.) March 30, 1922. 36.—‘ Bees of the Genus Panurginus Obtained by the American Mussa Rocky Mountain Expeditions.’ By T. D. A. Cockerell. 10 pp. April 14, 1922. t The edition of Novitates is 850, of which about 100 are mailed on the date of issue and the others are placed on sale in the Library. LIST OF ILLUSTRATIONS Merritherium, Dinotherium, Rhynchotherium, and Stegodon: fundamental arrangement of the cutting teeth................... cece eee Rhynchotherium euhypodon, R. shepardi, and R. tlascale, new Magi rhynchorostrine types of lower jaw.. en - Palzomastodon wintoni, Trilophodon pirvbilens T. proche: ‘Jonigi- pontvine types Gr aw are Mall eA eee Palzomastodon wintoni, Trilophodon productus, T. giganteus, new species, T'etralophodon, and T. campester: lower jaw and grinding teeth in the longirostrine phylum.........................00005 Carangoides gymnostethoides evermanni, new subspecies. Type and Sphezrodactylus macrolepis and Paragonatodes dickersoni: ventral view Se ne RONNIE Pato) ss a kw 0h Gv ca kaso ORR REE Coleonyzx variegatus and Lathrogecko xanthostigma: hyoid apparatus. .. Gonatodes atricucullaris and Paragonatodes dickersoni: ventral aspect of pectoral girdle... Lepidoblepharis barbouri iad Poleroduciylis aeiblenlee ventral asset! SIMU Serer dces U's sas asus awe cst ages nreveus et Paragonatodes dickersoni and Sphzxrodactylus macrolepis: ventral view of Coleonyx variegatus and Paragonatodes dickersoni: cloacal bones, showing relation to the cloacal slit...................22.0000- Gonatodes atricucullaris, Lathrogecko sxanthostigma, Lepidoblepharis barbourt, and Sphxrodactylus macrolepis: claw sheaths illustrating four stages in a single line of specialization... : Colius nigricollis leucophthalmus, new species: photograph frémn fife; CMRI SINE SUNS iyo ior soc do es sundae Maeks Ue ewe Batis ituriensis, new species: head of adult female.................. Terpsiphone rufocinerea, T. batesi, T. plumbeiceps, and T. viridis: heads of adult males, showing form of crest.. 4 Acholoé orbiculata, new species: anterior wey frat ana posterior elytra, first and later parapodia, anal cirri, and ventral and dorsal sete. . Caudal part of a humpback whale, showing the hind limb in situ... ... Skeleton of the hind limb of a humpback whale; cartilaginous femur and osseous tibia; and cartilaginous tarsus and osseous metatarsal. Mastodon tapiroides americanus: referred inferior molars from the Lower Pliocene of Hungary. Mastodon matthewi: type and para- types from the Lower Pliocene of Nebraska. Mastodon merriami: type molars from the Middle Pliocene of Nevada. Mastodon ameri- canus: referred; from the phosphate beds of South Carolina. . Mastodon merriami, type: internal and external views of two lower grinders; lateral and external views of tusk...................- Sketch map of the northeast quarter of the Monticello Quadrangle, ee MISC, TC GCROIOKY 6200055 3 ow oin'g woe sw Wiel Si os cae hepa Novit. Pace “2 10 3 6 x LIST OF ILLUSTRATIONS Lock at Palace, Cumberland River, showing three cycles of erosion in the background, 06.5 his koin 6k ces Gas ba. la Cae ee Outcrop of the Fort Payne Formation, Mississippian, on endo Creek, showing limestone lens and flaggy shale.. eF Local unconformity in the Warsaw beds, Missiesipnien: on the. tdad between Mill Springs and the University Camp................ Stehlinius uintensis, type specimen: reconstructed skull and lower jaw Stehlinius uintensis, type specimen: reconstructed skull and palate .... Buphagus erythrorhynchus, B. africanus, and B. langi: beaks of three ox-peckers, seen from the side and beneath.................... Neolestes torquatus: head and left foot.............. 00 cece eee eee Nicator chloris: nestling, showing bare face and neck................ Sigmodus rufiventris mentalis: nestling, showing white wing-coverts and parietal areas of rinked skin. /.°5:5:. .s-s«.shes os trintan se vanes Eupomacentrus nepenthe, (9p... «4:20 sic 5 oss x a< weed Rake ota ete s Labrisomus hetinete, type. ois 06 03 oa os cay os 5500's os ks Aptychus profundus: vertical section of the shell, showing the three shell layers. Aptychus lxvis, the aptychus of a species of Aspidoceras.-. Aptychus lamellosus, exterior view, and Aptychus punctatus, vertical section showing three shell layers................0.--02eceeeeee Aptychus didayi. Aptychus of Haploceras..............00.0 cee eee Aptychus cristobalensis, new species: oblique, interior, and top views of Aptychus cubanensis, new species: views of lamellae and valves........ Aptychus pimientensis, new species: views of valves, lamell, and cost in holotype.and PAFAIVNE « ...4 | hob ics Utde donee aes eee Sminthillus peruvianus, new species: type and paratype, showing color variation. . 3 Atelopus rusilenie. Atco Igor: Phyllobates Antanas. ‘Pelmatobius cinereus: types, NOW. SPOS oe occ cases oa ce s0 cas ab Maple ole wee Map of South America showing the known distribution of Pygochelidon . Maps showing the range of distribution of Pygochelidon cyanoleuca, P. patagonica patagonica, and P. p. peruviand.... 2.66... eee eeeees Map showing the route of The American Museum of Natural History’s Ecuadorean Expedition of 1920-1921. .......... ccc eee eee eee Hydriomena (?) protrita, a new fossil moth from Colorado............ Diagram showing the nature of the locality where Drytomomys zxquatorialis, new species, was found.. fh Drytomomys xquatorialis, new species, ine iegecisnceia ‘and Muyooaster COTY DUB. 55.0 i c36s sain jes col y So Wind weap ann ga Raine a tele Novir. Paes 6 7 14 2 3 17 3. 7 10 12 26 1 2 28 3 4 i 8 9 11 29 2 4 30 10 12 32 7 34 1 35 1 2 LIST OF NEW TAXONOMIC NAMES Hicuer Groups Novir. CMM AEEA CMAEIOTIR fos) dose ce ev cua eas eve ene eden. 1 NIN oe ooo Sg os ois sis 85 vs velo oR aeRO 12 manpalotremine Stunkard.. ................0. 00025500. ee pees yr NTS Rian? dpa ag Ee as Sars eA IE SEES INS Ee a SE Geer 4 Ectypodus Matthew and Granger..................-2ceeeeceeceees-s J Eucosmodon Matthew and Granger.................0-ecececeecee? Peradectes Matthew and Granger..................-0ccceeecceeceees Thylacodon Matthew and Granger..................0 cc cceececceeces Lepiacodon Matthew and Granger.................000c ccc eceeeeee Xenacodon Matthew and Granger................. eee ee cece cece ees Acmeodon Matthew and Granger............... 0... ce cece cee eee cers - DLabidolemur Matthew and Granger.................. cece eee eececees wonacius Matthew and Granger. ;... 2... <. 5. eect cewesce Navajoyius Matthew and Granger............0...... cee cece eee eeeees Carpodaptes Matthew and Granger.................-....00e eevee Eoconodon Matthew and Granger............0.02. 000 cece eee eee eeeee Mizxoclenus Matthew and Granger..................0cceecceeeeeeees I MRO a er eek da wees lec vas caddew ss uee 14 TENN Se RNS g ek cc eiccc svc chase cabne an} ne 17 NINN AMRAMANNNI 7 eseutts St is Se co vee ceeucascantsen Dae REESE Gon OR Se a a 23 INNO ATRRIRNINN 5 Slope a es a cea peewee seeeee meee SPECIES AND SUBSPECIES Hthynchothertum tlascadlz Osborn... . 2.6... 0c cee cece wesc ecteese 1 pumennocon mganieus OebOr, ok. cece eee cenvests SEND ORTOGUNS UINOEU Ss c6 Uy bie ccc os oss ec uy e's onepmadeee Capito brunneipectus Chapman. ...........0..0 2.00 cceccacccccccesee & Nonnula amaurocephala Chapman................02 eee ec eeee neces Rhopochares cochabambe Chapman................ 06.00 cece cece eee Wicrorhopias emiliz Chapman. ................. 00s cceccseedenceees Drymophila devillei subochraceus Chapman................22020+++ Hypocnemis hypoxantha ochraceiventris Chapman..................++: Siptornis punensis cuchacanche Chapman................00.+.000- Cistothorus platensis caracasensis Chapman.....................00005 Carangoides gymnostethoides evermanni Nichols....................... 8 Capricornis osbornt Andrews. ........ 00.00... cc ccs cccecsece miges 6 Astur toussenellit canescens Chapin................00ee cece cece eces bj Colius nigricollis leucophthalmus Chapin..................00000000s ood Ls ae va ek b voto bk eat wleaienes meme — NANENNAAATMREWWN NN He —_ ONnNrF kK NOAA rR WN NH OO xil LIST OF NEW TAXONOMIC NAMES Novir. Pacs Terpeiphone batest Caapin i. y0% 0 cs ee a ee Acholoé orbiculata Treadwell.............cccccccceeces ots ee 8 Mastodon matthewt Osborn or eo Sek a oe ee 10 Mastodon merriamé Osborn...o.o55 3 occ la ccc aces Jie ceed Guan camaae Ectypodus musculus Matthew and Granger..................+2..-+++- 18 Peradectes elegans Matthew and Granger.................00002-eeees Thylacodon pusillus Matthew and Granger...............0 0c eee e eee Leptacodon tener Matthew and Granger. .............6. 0 cece e eee ee Xenacodon mutilatus Matthew and Granger.................0.+0000- Acmeodon secans Matthew and Granger.............00ece eee c cece Labidolemur soricoides Matthew and Granger..............2.+02+00+ Ignacius frugivorus Matthew and Granger... ..........00e cee eeeees Navajovius kohlhaase Matthew and Granger...............00000005 Carpodaptes aulacodon Matthew and Granger..................200:- Mixoclenus encinensis Matthew and Granger................-.0-+5: Stehlinius unitensis Matthew .........0.0..cececccccerccctvccccscess AM Uta stansburiana stejnegeri Rohnadtt:: te Sn hin ane neo aia Atlapetes gutturalis parvirostris Dwight and Grisooin. 32ese 6 2.9 ous ee Atlapetes gutturalis fuscipygius Dwight and Griscom................++ Atlapetes gutturalis griseipectus 8 and Griseom .., . 50:55 +06 p> stand Buphagus langi Chapin... isis. ele e's Se 6 6cbsg eine ek tah g a Nothocercus fuscipennis Chante: Yesie-ce sb alg ton ea srele ie op eae Rae Penelope barbata Chapman... Siptornis wyatti equatorialis Chipman: Gdontophorus parumnbec cineicens Clube?’ sig cas 3-0 /y ak 4 ate Nyctibius longicaudatus chocoensis Chapinian.:: Bye ot aan ay Picumnus parvistriatus Chapman. .......... Thamnophilus zarume Chapman.............. Leptasthenura xenothorax Chapman................eeeseceeeeecetes Leptasthenura striata cajabambe Chapman. ............000eeee ee eee Automolus celice Chapman...........ccccecceceeescresccttersesssns Pachysylvia fuscicapilla albigula Chapman............. ce 9 gre ae Bae Basileuterus fraseri ochracetcrista Chapman..........000 +0000 eee neers Sporophila insulata Chapman................4% Sigs vie hue eat ee Tayassu pecart beebei Anthony..........0.. ccc ceceeecccecveccsesees 2 Pecari tajacu macrocephalus Anthony..........s.eceeeseeeeeeeecececes Ceoomys ruttiuse ARHONY. 0566.6 os de cube cece eyes tess sine Echimys longtrostrig Anthony....5: 0. 6. u oes akan 8 eats eee Dinomys gigas Anthony... es es oss «tins sme s tas 6 ee ae Ichthyomys tweedit Anthony........2.cccseccececcccccccrscccesecenss @O Neusticomys monticolus Anthony......+...6..200 cece cert eeeeeeeeeees Blarina montivaga ADtHONY....00. ccc. cece sees eet paeueees on emis ate Anoura geoffroyi antricola Anthony............cesseeec sw enceeeceses Cenolestes caniventer Anthony... ......0 0000s +sed es cceden ss eeu ee mmm Celioxys mendacina Cockerell........cccceccececesccssececeecceseees Oh Calioxys lutzt Cockerell. 0000000 ee Se ee oe cae aie se Un yeeeee Calioxys mesx Cockerell. ).. 52 55 Seniesa ae “aia WA ale meee eae ODAOISE EWE PWWWNNNOTOARWWNHNNH ARH OD AMIwWwaannroauarhnre LIST OF NEW TAXONOMIC NAMES Xlii Novir. Pace Czlioxys lamellicauda Cockerell...............+ \avebisasht bore call eee Holbrookia pulchra Schmidt......... Soha Dane te eran Ot Ea eac al aac lel 22 Holbrookia maculata campi Schmidt....... o/s k ofa che, mia ea ei ae acaaiaieta ie Holbrookia dickersonz Schmidt... .............0eeceeceeeecccecess I I ESE RES De PERE TS) efeieen qatticnss EES ESERIES Got aE SE RR oc Pa OS EPIC NE TN eos 5 sy «3 5 a-5)> 3 nine We aerials I ENANIIR ese 5 sy so ws vaso +o + «hanes Fa ae SII FORMER COOKIE cy cl g b's 5's 0s es es nsec cle smnthipe eanmes 23 Triepeolus schwarzt Cockerell.........- 20... soc c ccc secs cecees ees Triepeolus balteatus Cockerell........ Gack ates & A.c*oim © 0,8 Sokahwe tegen ae Triepeolus rhododontus Cockerell... Wer retrane oe Triepeolus (Synepeolus) insolitus Pickard. Sea's oc k{picln 66 teri mea Triepeolus brunneus Cockerell.... 2.2.22... - cece eee ennerceeceees Triepeolus trilobatus Cockerell..........-.--0ceeecee cece eee cneeees Triepeolus perelegans Cockerell........ 0... -2ece cece cee ceeececcees mysepeotus sequior Cockerell 0... 2.2... 0... oe wees ce crncenes Triepeolus rectangularis Cockerell........ 2.2... cece eee eeeeeeeeeee INIED MUNNMIED © OCKONON 6a cds os sce 2 + 09.0.0, 00100 6 spin en wine's ono IIINRN EME © NIORONONE ets cats Sip 5.0 2 +0 0.0 sos he nemenaeeles ohe$ Triepeotus dichropus Cockerell......5...65 0... 0.00 cseveneccccseecces Triepeolus maculiventris Cockerell...:...........+-- Triepeolus laticaudus Cockerell.............+++++ a as tena aia al A a hal gee Epeolus utahensis Cockerell... Epeolus lutzi Cockerell... Be Sys ‘A Fossil Moth from Florissant, menthe, 21, 9. Colorado,’ 34, 1-2; “Bees of the mes, 21, 6, 8. Genus Panurginus Obtained by modesta, 21, 2, 7, 10. the American Museum Rocky meoesta, 21, 10. Mountain Expeditions,’ 36, 1-10. novomexicana, 21, 1, 7-9. Cockerellia, 33, 1, 5, 13, 15. obtusiventris, 21, 10. 4 INDEX octodentata, 21, 2, 7, 9, 10. Diaphragmus elegans, 11, 8, 10. octodentata @, 21, 3: Dibelodon, 1, 8, 10.: . i piercei, 21, 9. Dibunodon, 1, 11. : porter, 21, 2, 7, 9, 0. Didelphide, 13, 2 pratti, 21, 9. Didelphys, 14, 4. quadridentatus, 21, 9. Diepeolus, 23, 2. quercina, 21, 4, 10. Dinomys, 19, 1. regine, 21, 9. branickii, 19, 6, 7. = ribis, 21, 2, 7-10. - branickii occidentalis, 19, 7. ribis kineaidi, 21, 10. gigas, 19, 6, 7. rufitarsis, 21, 2, 7-10. Dinotheriide, 1, 1. rufitarsis melanopoda, 21, 10. Dinotherium, 1, 3. rufitarsis rhois, 21, 9. Diopatra neapolitana, 8, 1. rufocaudatus, 21, 9. Diplochelidon melanoleuca, 30, 12. sayi, 21, 3,7, 10. Doeringiella, 23, 1. scitula, 21, 9. Doliornis sclateri, 30, 8. sculptifrons, 21, 10. Drosophila, 34, 4. slossoni, 21, 10. Drymophila caudata caudata, 2, 4 sodalis, 21, 9, 10. caudata klagesi, 2, 4. soledadensis, 21, 9. devillei devillei, 2, 4. sonorensis, 21, 9. devillei subochraceus, 2, 4. texana, 21, 2,7,9,10. °° phantatis, 2, 4. texana vegana, 21, 9. ie Dryotriorchis batesi, 17, 15. trigonus, 21, 9. Drytomomys exquatorialis, 35, 2-4. — Coleonyx, 4, 1, 2, 7, 15. Dwight, Jonathan and Griscom, Ludlow: elegans, 4, 3, 6. ‘A Revision of Atlapetes gutturalis, variegatus, 4, 3, 5, 6, 11. with Descriptions of Three a Colius nigricollis, 7, 3, 4. Races,’ 16, 1-4. nigricollis leucophthalmus, 7, 2-5. nigricollis nigricollis, 7, 3—5. Echimys armatus, 19, 5, 6. nigricollis nigriscapalis, 7, 4. castaneus, 19, 5. striatus nigricollis, 7, 3. guiane, 19, 5. striatus striatus, 7,3... : longirostris, 19, 5, 6. Colletes, 23, 2. Ectypodus musculus, 13, 1. marginata, 23, 3. $} Elephantide, 1, 1. succincta, 28, 3. ; Elephantine, 1, 2, 14, 15. Columnaria vacua, 11,.5,9. |! Elephas columbi, 1, 14. Coriphagus, 18,7... |: 0!» hysudricus, 1, 14, 15. Crawfordia, 36, 6. BS er: imperator, 1, 11, 14, 15. Crax globicera, 25; 8. ‘ meridionalis, 1, 14, 15. panamensis, 25, 7, 8. planifrons, 1, 14, 15. Creciscus ruber, 25, 2, 3. primigenius, 1, 14, 15. ruberrimus, 25, 2, 3. trogontherii, 1, 14. Ctenacanthus, 11, 5, 9. Eoconodon, 18, 6, 7. Cyathaxonia cynodon, 11, 5, 9. Epeolus americanus, 23, 6. assamensis, 23, 1. Dasychone, 8, 1. beulahensis, 23, 1. Dasychonopsis bairdii, 8, 1. bifasciatus, 28, 13, 15. INDEX compactus, 23, 1. cruciger, 23, 3. eldoradensis, 23, 1 fervidus, 23, 1. giannellii, 23, 2. hitei, 23, 16. humillimus, 23, 16. interruptus, 23, 1. lutzi, 23, 14-16. lutzi dimissus, 23, 16. mercatus, 23, 15. peregrinus, 23, 1. pusillus, 23, 14, 15. utahensis, 23, 14, 15. ventralis, 23, 1 Epimethea variegata, 36, 1. Estrilda melpoda, 17, 16. Ethmia mortuella, 34, 2. Ethmiide, 34, 2. Eubelodon morrilli, 1, 9, 10. Eublepharide, 4, 1, 3, 15. Eucosmodon, 13, 1. Euelephantine, 1, 1. Euelephas, 1, 1. Eumetria vera, 11, 8, 10. verneuiliana, 11, 7, 8. Eupomacentrus nepenthe, 26, 1. Foyles, Edward J. “The Geology about Mills Springs, Monticello Quad- rangle, Kentucky,’ 11, 1-10. Fraseria cinerascens, 17, 16. ocreata, 17, 16. Furnarius rufus badius, 27, 6. rufus commersoni, 27, 6. rufus paraguay, 27, 5, 6. rufus rufus, 27, 6 Gallinula chloropus, 25, 4. chloropus cachinnans, 25, 3 chloropus centralis, 25, 3. chloropus cerceris, 25, 4 chloropus pauxilla, 25, 3. garmani, 25, 4. ’ sandvicensis, 25, 4. Gampsonyx swainsoni leonz, 25, 13. swainsoni meridensis, 25,13. — Gehyra, 4, 3. Gekko, 4, 3. Gekkonide, 4, 3, 14. Geometride, 34, 1 Girtyella indianensis, 11, 7, 8. Glaucidium brasilianum brasilianum, 31, 5. mice brasilianum phalenoides, 31, 6. brasilianum tucumanum, 31, 5, 6. ferruginea, 31, 6. nanum, 31, 5, 6. pycrafti, 17, 15. — at tephronotum, 17,15. | oo § Glossophaga, 20, 6. Gnathias, 24, 8-10. Gonatodes, 4, 1, 2, 12, 15. annularis, 4, 6. atricucullaris, 4, 6, 8, 13. dickersoni, 4, 6, 7, 14. Grallaricula boliviana, 31, 8. flavirostris brevis, 31, 7. flavirostris costaricensis, 31, 7, 8. flavirostris flavirostris, 31, 7, 8. flavirostris ochraceiventris, 31, 6, 7. flavirostris zarume, 31, 7. | Granger, Walter, see Matthew, W. D. Greeleyella, 36, 1, 2. beardsleyi, 36, 9, 10. Griscom, Ludlow, see Dwight, Jonathan, also Miller, W. DeW. Hapalotrema, 12, 3, 4. Hapalotremine, 12, 4. Haploceras, 28, 1, 6, 10. Haploceratide, 28, 5. Harmostomide, 12, 4, 5. Harmothoé fraser-thomsoni, 8, 1 Hemidactylus, 4, 3. Hemimastodon crepusculi, 1, 5, 8. Hieraaétus africanus, 17, 15. Hipparion, 10, 2. Hirundo cyanoleuca, 30, 1. melampyga, 30, 1. minuta, 30, 1. nigrita, 17, 16. patagonica, 30, 3. puella, 17, 13, 14." Holbrookia dickersone, 22, 2, 3. elegans, 22, 2. lacerata, 22, 3. maculata, 22, 3. maculata approximans, 22, 1-3. maculata campi, 22, 1-3. maculata flavilenta, 22, 3. maculata maculata, 22, 3. propinqua, 22, 2. pulchra, 22, 1, 2. texana, 22, 2. Hydriomena protrita, 34, 1, 2. Hypargos dybowskii, 17, 16. Hypocnemis collinsi, 2, 5. flavescens subflava, 2, 5. hypoxantha hypoxantha, 2, 5. hypoxantha ochraceiventris, 2, 5. Hypomacrotera, 36, 1, 2. callops, 24, 15; 36, 8, 10. callops persimilis, 36, 8, 9. Ichthyomys hydrobates, 20, 4. sonderstromi, 20, 1—4. stolzmanni, 20, 4. tweedii, 20, 1, 2, 4. Ictinia plumbea plumbea, 25, 5-7. plumbea vagans, 25, 5-7. Ignacius, 13, 4. frugivorus, 13, 5. Isepeolus, 23, 1. Jacana jacana intermedia, 31, 3. jacana jacana, 31, 3. niger, 31, 3. scapularis, 31, 3. spinosa, 31, 3. Jupiteria charon, 34, 2. Labidolemur soricoides, 13, 4, 5. Labrisomus heilneri, 26, 2. Lagenorhynchus acutus, 9, 5. Lampribis olivacea olivacea, 17, 15. rara, 17, 14. Laniide, 17, 7, 9, 12. Lathrogecko, 4, 1, 2,7, 11-15. xanthostigma, 4, 3, 5, 6, 13. Lecythoplastes, 17, 14. preussi, 17, 13. Leiopodus, 23, 2. Leodiciqe, 8, 1. INDEX Lepidoblepharis, 4, 1, 2, 7, 11-15. barbouri, 4, 3, 6, 9, 13. festee, 4, 3. Lepidonotus clava, 8, 1. Leptacodon tener, 13, 2, 3. Leptasthenura, 18, 1, andicola andicola, 18, 9. andicola exterior, 18, 9. andicola estima, 18, 9. andicola peruviana, 18, 9. pileata, 18, 8-10. striata cajabambe, 19, 9, 10. striata striata, 18, 9, 10. xenothorax, 18, 8-10. Leptictide, 13, 2, 3. Libytheide, 34, 2. Liolopine, 12, 3, 4. Lithodryas styx, 34, 2. Lithostrotion basaltiforme, 11, 7, 8. harmodites, 11, 7, 8. proliferum, 11, 7, 8. Lobotus oriolinus, 17, 16. Longirostrinz, 1, 1, 3, 4, 6. Loxodonta africanus, 1, 13, 14. antiquus, 1, 13, 14. antiquus namadicus, 1, 13. atlanticus, 1, 13, 14. creticus, 1, 14. melitensis, 1, 14. mnaidriensis, 1, 14. namadicus, 1, 13-15. Loxodontine, 1, 1, 2, 13. Lutz, Frank E. ‘Geographic Average, a Suggested Method for the Study of Distribution,’ 5, 1-7. Lygodactylus, 4, 3. Malacoclemmys geographicus, 12, 1. leseurii, 12, 1,3. Malaconotinz, 17, 8, 9. Malaconotus, 17, 8. Mammontine, 1, 1, 2, 14. Marmosa bombaseare, 32, 5, 6. celica, 32, 4, 5. cinerea, 32, 3. fuseata, 32, 4. klagesi, 32, 3. madescens, $2, 4. ere INDEX 7 musicola, 32, 6. oroensis, 32, 3-5. perplexa, 32, 3. _ phaea, 32, 3. simonsi, 32, 4. sobrina, 32, 4, 5. waterhousii, 32, 3. Mastodon, 1, 7,8; 10, 1-6. americanus, 1, 12; 10, 3 , 4, 6. andium, 1, 4, 10. arvernensis, 1, 10, 11. borsoni, 1, 12. brevirostris, 1, 11. chapmani, 1, 11. hasnoti, 1, 11, 12. humboldtii, 1, 10. matthewi, 10, 2-4, 6. merriami, 10, 3-6. mirificus, 1, 6, 10-12. - pentelici, 1, 11. progenium, 1, 12. pyrenaicus, 1, 12. sivalensis, 1, 10—12. successor, 1, 10. tapiroides, 1, 12; 10, 1, 4. tapiroides americanus, 10, 2-4, 6. tropicus, 1, 10. turicensis, 1, 12. - Mastodonade, 1, 1. Mastodontide, 1, 1, 2, 4, 7. Mastodontine, 1, 12. Matthew, W. D. and Granger, Walter. ‘New Genera of Paleocene Mam- mals,’ 13, 1-7. Matthew, W. D. ‘Stehlinius, a New Eocene Insectivore,’ 14, 1-5. Megabelodon, 1, 7. Megachile addenda, 21, 8. argentata, 21, 9. centuncularis, 21, 9. circumcincta, 21, 9. ericetorum, 21, 9. lanata, 21, 9. latimanus, 21, 8. maritima, 21, 9. melanophza, 21, 8. - rotundata, 21, 9. willughbiella, 21, 9. Megaptera nodosa, 9, 1-6. Meropogon breweri, 17, 15. Merychippus, 10, 2. Mesoblastus glaber, 11, 7, 8. Microclzenodon, 13, 7. Micropus affinis, 17, 14. Microrhopias boucardi bicolor, 2, 4. boucardi consobrina, 2, 4. boucardi virgata, 2, 4. emilie, 2, 3, 4. quixensis, 2, 4. Microsciurus avuneulus, 32, 2, 3. napi, 32, 2, 3. peruvanus, 32, 3. sabanille, 32, 2, 3. Microspingus trifaseiatus, 30, 8. Microsyops, 14, 5. Miller, Waldron DeWitt and Griscom, Ludlow. ‘Descriptions of Pro- posed New Birds from Central America, with Notes on Other Little-known Forms,’ 26, 1-13. Mixoclenus encinensis, 13, 7. Mixodectes, 14, 5. Meeritheriide, 1, 1. Meeritherium, 1, 2, 3, 5. Myocastor, 35, 1, 4. coypus, 35, 3. Nautilus, 11, 7, 8. Navajovius kohlhaase, 13, 5, 6. Necrosorex, 14, 1, 4. Nectarinia congensis, 17, 5, 6. Neolestes, 17, 8, 9. torquatus, 17, 6, 7. Neoplagiaulax americanus, 13, 1. Neoreomys, 365, 1, 4. Nereida, 8, 1. Nereis pelagica, 8, 1. tongatabuensis, 8, 1. Neusticomys monticolus, 20, 2—4. Nicator, 17, 8, 11. chloris, 17, 9, 10. vireo, 17, 9. Nichols, John Treadwell. ‘A Hawaiian Race of Carangoides gymnosteth- oides,’ 3, 1, 2; ‘A New Pomacen- trid and Blenny from the Baha- + mas,’ 26, 1, 2. Noble, G. K. ‘The Bony Structure and Phyletic Relations of Sphzro- dactylus and Allied Lacertilian Genera with the Description of a New Genus,’ 4, 1-16; ‘Five New Species of Salientia from South America,’ 29, 1-7. Nomada accepta, 24, 3, 12. aldrichi, 24, 7. alpha, 24, 3. alpha dialpha, 24, 3, 12. alpha paralpha, 24, 3, 12. bella, 24, 8-10, 13. calloxantha, 24, 4, 11. carinicauda, 24, 7, 12. caroline, 24, 9, 12, 13. citrina, 24, 2, 3, 5, 12. citrina flavomarginata, 24, 4. civilis, 24, 2, 11. clarescens, 24, 10, 13. concinnula, 24, 6, 12. coquilletti, 24, 2. crawfordi lachrymosa, 24, 5, 6, 12. crucis, 24, 1, 11. cymbalarie, 24, 2, 11. edne, 24, 6. edwardsii, 24, 2, 12. gillettei, 24, 6. gutierrezix, 24, 2, 11. heterosticta, 24, 9, 10, 12. illinoénsis, 24, 7, 12. lepida, 24, 9. maculata, 24, 10. melanoptera, 24, 5, 11. morrisoni, 24, 2, 3, 11. morrisoni flagellaris, 24, 4. orophila, 24, 8, 12, 13. packardiella, 24, 11, 12. pecosensis, 24, 3, 12. perplexa, 24, 9, 13. rufula, 24, 4, 5. schwarzi, 24, 8-10. siouxensis, 24, 9, 13. vallesina, 24, 7. vicinalis aldrichi, 24, 7. vicinalis infrarubens, 24, 7, 12. vincta, 24, 1, 11, 12. a a Bath oe INDEX vincta heterochroa, 24, 1, 11. - vulpis, 24, 10, 13. zebrata, 24, 2, 11, 12. Nonnula amaurocephala, 2, 2. ruficapilla, 2, 2. ruficapilla pallida, 27, 4. ruficapilla ruficapilla, 27, 4. Notherodius holostictus, 25, 13. Nothocercus bonapartei bonapartei, 18, 3. bonapartei frantzii, 18, 3. bonapartei intercedens, 18, 3. fuscipennis, 18, 1-3. julius julius, 18, 1, 2. julius salvadorii, 18, 2, 3. julius venezuelensis, 18, 2. nigricapillus, 18, 2. Nothodectes, 13, 2; 14, 5. Numenius vociferus, 25, 13. Nyctibius longicaudatus, 18, 6. longicaudatus chocoensis, 18, 5. longicaudatus longicaudatus, 18, 5. Nymphalide, 34, 2. Nymphalites obscurum, 34, 2. seudderi, 34, 2. Nystactes tamatia hypnaleus, 27, 3, 4. tamatia interior, 27, 3, 4. tamatia pulmentum, 27, 3, 4. tamatia tamatia, 27, 3, 4. O’Connell, Marjorie. “New Species of Ammonpite Opercula from the Mesozoic Rocks of Cuba,’ 28, 1-15. Odontophorus parambz, 18, 5. parambe canescens, 18, 4, 5. parambe parambe, 18, 4. (Ecomys nitedulus, 19, 4. rutilus, 19, 4. Oppelia, 28, 12. bous, 28, 5. euglyptus, 28, 5. lithographica, 28, 5. thoro, 28, 5. Oppeliinz, 28, 5. Oreopelia bourcieri, 31, 2, 3. bourcieri baeza, 31, 2. bourcieri bourcieri, 31, 2, 3. ee i I i ll le a INDEX bourcieri subgrisea, 31, 2, 3. frenata, 31, 2. Ortalis canicollis canicollis, 27, 2, 3. canicollis grisea, 27, 2, 3. canicollis pantanalensis, 27, 2, 3. cinereiceps cinereiceps, 25, 1, 2. cinereiceps saturatus, 25, 1, 2. ruficauda, 27, 3. Osborn, Henry Fairfield. “The Evolu- tion, Phylogeny, and Classifica- tion of the Proboscidea,’ 1, 1-15; ‘First Appearance of the True Mastodon in America,’ 10, 1-6. Osculatia sapphirina, 18, 2; 31, 5. Otus hylerythrus, 17, 15. Oxyclenide, 13, 7. Pachydactylus, 4, 3. maculatus, 4, 11. Pachysylvia fuscicapilla albigula, 18, 11. fuscicapilla fuscicapilla, 18, 11. semibrunnea, 18, 11. Palzomastodon, 1, 5. beadnelli, 10, 1. wintoni, 1, 7-9. Panurgine, 36, 1. ~Panurginus albopilosus, 36, 1, 2. altissimus, 36, 3, 9, 10. atricornis, 36, 5, 6, 8-10. bakeri, 36, 9, 10. bidentis, 36, 2. crawfordi, 36, 2. cressoniellus, 36, 2, 6, 7, 9, 10. cressoniellus calocharti, 36, 6, 7, 10. didirupa, 36, 6, 9, 10. expallidus, 36, 7. flavotinctus, 36, 4. hispanicus, 36, 2. horizontalis, 36, 3, 7. ineptus, 36, 8, 10. irregularis, 36, 2, 9. lactipennis, 36, 2. lutze, 36, 7, 8, 10. montanus, 36, 2. nebrascensis, 36, 2, 5, 9. niger, 36, 1. nigripes, 36, 2. - nitidulus, 36, 2. opaculus, 36, 3, 4, 10. ornatipes, 36, 6. pauper, 36, 3, 4. perlevis, 36, 5, 9, 10. pernitens, 36, 4, 10. picipes, 36, 4. picitarsis, 36, 2. piercei, 36, 2, 5, 7, 9. porter, 36, 4, 9, 10. pulchricornis, 36, 8, 10. renimaculatus, 36, 6, 9. romani, 36, 2. sculpturatus, 36, 2. solani, 36, 2. verus, 36, 7. Panurgus, 36, 1. Paragonatodes, 4, 14. dickersoni, 4, 4, 8, 10, 11. Parmoptila jamesoni, 17, 16. Pecari tajacu macrocephalus, 19, 3, 4. tajacu tajacu, 19, 3, 4. torvus, 19, 3, 4. Pelecanoides garnoti, 30, 14. Pelicinius, 17, 9. geylonus, 17, 8. Penelope argyrotis, 18, 3. barbata, 18, 3. Peradectes elegans, 13, 2. Peratherium, 13, 2. Perdita albipennis, 33, 1-5. albipennis helianthi, 33, 2, 3, 14. affinis, 33, 8, 13-15. bishoppi, 33, 7. bisignata, 33, 11, 14. bruneri, 33, 9, 14. callicerata, 33, 1, 8. calloleuca, 33, 12, 14. canadensis, 33, 2-4. canina, 33, 9. cockerelli, 33, 9, 14. crawtfordi, 33, 7. fallax, 33, 9. fallax fontis, 33, 9, 15. heliophila, 33, 2—4. heterothece, 33, 7. hyalina, 33, 4. ignota, 33, 6, 7, 14, 15. lacteipennis, 33, 2-4. 10 INDEX lepachidis, 33, 4. ’ lingualis, 33, 4. lutzi, 33, 7, 14. melanochlora, 33, 7, 15. mellina, 33, 8. miricornis, 33, 9, 14. miricornis leucorhina, 33, 10, 14. nebrascensis, 33, 8. nolinz, 33, 11, 14. octomaculata, 33, 8. octomaculata terminata, 33, 8. pallidipennis, 33, 2. pallidipennis indianensis, 33, 2, 14. rectangulata, 33, 13. rhois, 33, 12. sexmaculata, 33, 6. snowii, 33, 6. solidaginis, 33, 12-14. solitaria, 33, 1. sparsa, 33, 4. swenki, 33, 9. - verbesine, 33, 2, 4, 5. wickhami, 33, 1, 13, 14. wootone, 33, 1. wunderi, 33, 10, 15. wyomingensis, 33, 13, 15. xanthisme sideranthi, 33, 6. zebrata, 33, 8, 14. zebrata canina, 33, 9. Phelsuma, 4, 1, 6. laticauda, 4, 7. Phenacolemur, 13, 4, 5. Phileremus americanus, 23, 6. Phillipsia, 11, 5, 9. Philydor temporalis, 18, 10, 11. Phiomia, 10, 1. Phocena communis, 9, 5. dalli, 9, 5. Phylledestes vorax, 34, 2. Phyllobates anthonyi, 29, 4-6. Phyllodactylus, 4, 11, 12. siamensis, 4, 3. Phyllotis fruticicolus, 32, 1, 2. haggardi, 32, 1, 2. Piaya cyana mesura, 31, 4. Picumnus parvistriatus, 18, 6. sclateri, 18, 6. Pieride, 34, 2. Pionites melanocephalus pallidus, $1, 5. Pisces, 11, 5, 9. Plagiaulacidz, 13, 14, 1. Platycrinus sculptus, 11, 5, 9. Platystrophia cypha-conradi, 11, 5, 9. Plesiadapide, 13, 4-6; 14, 1, 4, 5. Pliohippus mirabilis, 10, 2. Podica senegalensis senegalensis, 17, 14. Polynoé nalleata, 8, 3. Polynoide, 8, 1. Porzana, 25, 3. Premnornis, 31, 8. Prionopide, 17, 13. Prionops, 17, 12, 13. Prismopora serrulata, 11, 10. Prodryas persephone, 34, 2. Productus magnus, 11, 6, 9. ovatus, 11, 7, 8. Progne chalybea, 30, 7. Prolibythea vagabunda, 34, 2. Proparorchide, 12, 3, 4. Proparorchine, 12, 3, 4 Proparorchis, 12, 4. artericola, 12, 3. Protandrena asclepiadis, 24, 13. bancrofti, 24, 13. Protohippus, 10, 2. Pseudemys elegans, 12, 1, 3. scripta, 12, 1, 3. Pseudomelecta miranda, 24, 15. rociadensis, 24, 15. Pseudopanurgus, 36, 1. zthiops, 24, 15. Pseudospingus xanthophthalmus, 30, 8. Ptilodus, 13, 1. Pustula biseriatus, 11, 6, 8. Pycnonotide, 17, 7-9. Pycnonotine, 17, 9. Pycnonotus, 17, 7, 8. tricolor, 17, 9. Pygochelidon, 31, 15. cyanoleuca, 30, 1-3, 5-14. flavipes, 30, 8, 11. patagonica, 30, 4-7, 13, 14. patagonica patagonica, 30, 1, 3-12. patagonica peruviana, 30, 7-12. : ¥ INDEX 11 Rallus giganteus, 26, 13. Reichenberger, Elsie M. B., see Cherrie, George K. Reticularia pseudolineata, 11, 6, 8. Rheomys, 20, 3. raptor, 20, 4. Rhopochares cochabambe, 2, 2, 3. ruficapilla, 2, 3. Rhynchopora cooperensis, 11, 6, 9. Rhynchorostrine, 1, 1, 3, 4. Rhynchotherium, 1, 3, 8. brevidens, 1, 5, 6. dinotherioides, 1, 6. euhypodon, 1, 5, 6. proavus, 1, 6. shepardi, 1, 5, 6. tlascale, 1, 5, 6. - Riparia rufigula, 17, 14. Rupornis magnirostris insidiatrix, 31, 4. magnirostris magnirostris, 31, 4. magnirostris zamore, 31, 3, 4. Sarcothraustes antiquus, 13, 6. corypheus, 13, 6. Saturniide, 34, 2. Schistosomide, 12, 5. Schmidt, Karl Patterson. “A New Name for a Subspecies of Uta stans- buriana Baird and Girard,’ 15, 1, 2; ‘New Species of North American Lizards of the Genera Holbrookia and Uta,’ 22, 1-6. Scoptelus brunneiceps, 17, 15. Scotopelia bouvieri, 17, 15. Scrapteroides difformis, 36, 1. Sigmodus rufiventris mentalis, 17, 11- 13. Siptornis lilloi, 2, 6. punensis cuchacanche, 2, 5, 6. punensis lilloi, 2, 5. punensis punensis, 2, 5, 6. punensis rufala, 2, 5, 6. wyatti, 18, 4. wyatti equatorialis, 18, 4. wyattii wyatti, 18, 4. Sminthillus limbatus, 29, 1. peruvianus, 29, 1, 2. Spherodactylus, 4, 1, 6, 7, 11, 14, 15. . 5 macrolepis, 4, 2, 4, 9, 10, 12, 13. Spheniscus humboldti, 30, 14. Spirifer bifurcatus, 11, 6-8. biplicoides, 11, 6, 9. Spirorchide, 12, 4, 5. Spirorchine, 12, 4. Spirorchis, 12, 2-5. innominata, 12, 3. Sporophila insulata, 18, 12. minuta, 18, 12. Stegodon bombifrons, 1, 13. cautleyi, 1, 13, 14. cliftii, 1, 13. ganesa, 1, 13. insignis, 1, 13. latidens, 1, 13. stegodontoides, 1, 13. Stegodontine, 1, 1, 2, 12. Stehlinius unitensis, 14, 2-5. Stolopsyche libytheoides, 34, 2. Strix chacoensis, 27, 1, 2. rufipes, 27, 2. Stunkard, Horace W. ‘Notes on North American Blood Flukes,’ 12, 1—5. Synallaxis paramo, 18, 4. Synepeolus, 23, 6. Syrrhopus, 29, 1. Tangara cyaneicollis exruleocephala, 31, 4. Tantalus pictus, 25, 13. ‘ Tapiroides, 1, 3. Tarentola, 4, 3. Tarsiide, 13, 5. Tayassu pecari equatoris, 19, 2. pecari beebei, 19, 1, 2. pecari pecari, 19, 1, 2. Telmatobius cinereus, 29, 4, 6, 7. Telephorus, 17, 8. ; Terpsiphone batesi, 7, 6-9. ignea, 7, 7. plumbeiceps, 7, 7, 8. rufocinerea, 7, 7, 8. viridis, 7, 8. Tetrabelodon, 1, 7, 8. Tetralophodon barbouri, 1, 9, 10. campester, 1, 7-10. corrugatus, 1, 7. 12 INDEX longirostris, 1, 7-9. punjabiensis, 1, 7. Thamnophilus radiatus, 18, 7. radiatus albicans, 18, 7. ruficapillus, 2, 3. zarume, 18, 6-8. Thecadactylus, 4, 3, Thryocrex, 25, 3. Thylacodon pusillus, 13, 2. Tigrornis leucolopha, 17, 15. Tortricid, 34, 2. Tortrix destructus, 34, 2. florissantana, 34, 2. Treadwell, A. L. “Polychzetous Annelids Collected at St. Paul de Loanda by the American Museum Bel- gian Congo Expedition,’ 8, 1-3. Triepelous alpestris, 23, 13, 15. amandus, 23, 10, 14. balteatus, 23, 5, 14. brunneus, 23, 7, 14. callopus, 23, 5. concavus, 23, 2, 13. concinnus, 23, 13. concolor, 23, 2, 8, 14. denverensis, 23, 1. dichropus, 23, 11, 14. fortis, 23, 3, 14. fraser, 23, 9. gabrielis, 23, 1. grindelizw, 23, 1. helianthi, 23, 3, 4, 15. helianthi grandior, 23, 1. helianthi pacificus, 23, 4, 15. hopkinsi, 23, 8. insolitus, 23, 6, 14. isocome, 23, 5. laticaudus, 23, 12, 15. lestes, 23, 11, 14. lunatus, 23, 2, 8, 14. maculiventris, 23, 11, 15. martini, 23, 1. penepectoralis, 23, 2, 3, 15. pectoralis, 23, 2, 15. perelegans, 23, 8, 9, 15. pimarum, 23, 8. rectangularis, 238, 14. rhododontus, 23, 5, 6,9, 14. 3 rhoweri, 23, 1, 5. : 7 schwarzi, 23, 4, 5, 14. sequior, 23, 8, 9, 14. subalpinus, 23, 1. townsendi, 23, 13. trilobatus, 23, 7, 14. wyomingensis, 23, 10. Triisodon biculminatus, 13, 7. heilprinianus, 13, 6, 7. quivirensis, 13, 6, 7. Triisodontidx, 13, 6, 7. Trilophodon angustidens, 1, 6-9; 10, 1. angustidens palsindicus, 1, 7. floridanus, 1, 7, 10. giganteus, 1, 6-10. lulli, 1, 7, 8. macrognathus, 1, 7. ’ obscurus, 1, 7. ; osborni, 1, 7. precursor, 1, 8. productus, 1, 6-9. pygmeus, 1, 8. serridens, 1, 7, 8. serridens cimarronis, 1, 8. _willistoni, 1, 7. Triplophyllum, 11, 6, 8. Trophocleptria, 23, 1. Urotriorchis macrourus, 17, 15. Uta auriculata, 22, 6. bicarinata, 22, 6. clarionensis, 22, 6. concinna, 22, 5. elegans, 15, 1; 22, 5. gadovi, 22, 3, 4, 6. graciosa, 22, 5. irregularis, 22, 6. lateralis, 22, 6. levis, 22, 6. martinensis, 22, 5. mearnsi, 22, 4. microscutata, 22, 5. nelsoni, 22, 4, 6. nigricauda, 22, 5, 6. ornata, 22, 6. ornata linearis, 22, 6. [een ornata ornata, 22, 6. ornata symmetrica, 22, 6. palmeri, 22, 5. squamata, 22, 5. stansburiana, 15, 1. INDEX stansburiana elegans, 15, 1. stansburiana hesperis, 15, 1; 22, 5. stansburiana stansburiana, 15, 1; 22, 5. stansburiana stejnegeri, 15, stellata, 22, 5. thalassina, 22, 4. tuberculata, 22, 4, 6. wrighti, 22, 3, 6. 2, 22, 5. Verreauxia africana, 17, 16. Xanthidium 24, 4, 6, 7. Xantusia vigilis, 4, 15. Xenacodon mutilatus, 13, 3. Xenodacnis parina, 30, 8. Zenaida auriculata, 31, 1. auriculata auriculata, 31, 1, 2. auriculata cauce, 31, 1, 2. auriculata hypoleuca, 31, 1. auriculata pallens, 31, 1. hypoleuca, 31, 1. Zygolophodon, 10, 4. 13 AMERICAN MUSEUM NOVITATES No. 1 THE EVOLUTION, PHYLOGENY AND CLASSIFICATION OF THE PROBOSCIDEA By Henry FAIRFIELD OsBorN Issued January 31, 1921 By ORpDER OF THE TRUSTEES OF THE AMERICAN MUSEUM OF NATURAL HISTORY New Yorx City AMERICAN MUSEUM NOVITATES Number 1 - January 31, 1921 56.9,61 THE EVOLUTION, PHYLOGENY, AND CLASSIFICATION OF THE PROBOSCIDEA By Henry FAIrFIetp potihs The following is the quadruple branching indicated in classifications of the Proboscidea previous to 1910 (Osborn, ‘ Age of Mammals’): ELEPHANTID#—Gray, 1821; Zittel, 1891 Mastopontin2—?Mastodonade Gray, 1821 DINoTHERIIDA—Bonaparte, 1845; Zittel, 1891 Ma:ritHermp#—Andrews, 1906 1.—Primary DIvisions In December 1917, the author presented before the Paleontological Society a paper entitled ‘A Long-jawed Mastodon Skeleton from South Dakota and Phylogeny of the Proboscidea.’! This included a polyphyletic theory of the Proboscidea more or less fully anticipated by previous authors but more radical. This branching, as extended by Osborn in 1910 and 1917, is as follows: Mastropontin2z—Osborn, Age of Mammals, 1910, p. 558 ELEPHANTIN2Z—Osborn, Age of Mammals, 1910, p. 558 BuNoMASTODONTIN2—Osborn, 1918, p. 134. Defined to contain thtee main phyla, originally termed (Osborn, 1918, p. 136): 1, Longirostrine; 2, Rhynchorostrine; 3, Brevirostrine Stecopontin2z—Osborn, 1918, p. 135 =Stegodonts of southern Asia Loxopontin£—Osborn, 1918, p. 135=African and Eurasiatic loxodonts EUELEPHANTIN2*—Osborn, 1918, p. 136=Mammoths of Eurasia and North America Of the above branches it now appears that the Bunomastodontinze is a group rather than a subfamily and must be split up into the three subfamilies provisionally termed (1, 2, 3) above. Adding the Meri- theriide and the Dinotheriidx, this would divide the Proboscidea into ten branches. 1Bull. Geol. Soc. Amer., XXIX, No. 1, March 1918, pp. 133-137. *The term Euelephantine is invalid, because the a Euelephas i is invalid; the term Mammontine (i. e., les mammonts, the mammoths) may be substitu 1 2 AMERICAN MUSEUM NOVITATES [No. 1 In continuing the study of the Proboscidea since 1917, aided by recent observations of Lull, Matthew, Mayer, Schlesinger, Pilgrim, Barbour, and many others, it is the opinion of the author that the poly- phyletic theory of the Proboscidea is not only confirmed but that the phyla are more numerous than the ten branches already named and are geologically far more ancient than appeared in 1917. As regards the rank of the four primary divisions of the Proboscidea which have hitherto been discovered, they are certainly more profound than the four sections of the Rodentia, viz.: I, Sciuromorpha; II, Myomorpha; III, Hystricomorpha; IV, Lagomorpha. They are also more profound than Osborn’s four main divisions of the Perissodactyla, viz.: I, Titanotheroidea; II, Hippoidea; III, Tapiroidea; IV, Rhinocerotoidea. With the reservations, first, that we should not expect to find differ- ent orders of mammals subdivided into branches of equal rank and, second, that the subdivisions of the Proboscidea are either of subordinal or of superfamily value, we may adopt as the four primary divisions: I. McaeritHERIOIDEA typified by the Meritherium in the Oligocene of North Africa II. DinorHertoweEa typified by the Miocene and Pliocene Dinotheres of Eurasia III. Masropontomea to include the Bunomastodontidz, new family, and the Mastodontide IV. E.epxHantorpea to include the Elephantine, Loxodontineg, Stegodontinez, and Mammontine I.—Mceritherioidea Renewed study of Meritherium by Osborn and Matsumoto entirely confirms Andrews’ original opinion that Meritherium belongs in the order Proboscidea, as well as Osborn’s opinion that it stands very far apart from the other proboscideans and is not directly or indirectly ancestral to either of the other three groups. The enlargement of the second upper and lower incisor teeth into mutually abrading tusks, girdled with enamel, presents a firm ground of affinity with a still unknown primitive Lower Eocene proboscidean stem form. There the resemblance ends. These Meeritheres had no proboscis. The face, of brachyopic type, is markedly abbreviated. The cranium is elongated. Thus the facial and cranial proportions are analogous to those of the Sirenia. The upper grinding teeth are bilophodont, pointing to a tetrabunodont ancestry, and differ- ent from the bunomastodont grinders of Palzomastodon, which point toa hexabunodont ancestry. The lower grinders exhibit a rudimentary third crest. Andrews’ opinion that the Meeritheres were amphibious is prob- ably correct. 1921] CLASSIFICATION OF THE PROBOSCIDEA 3 II.—Dinotherioidea All agree that these animals were chiefly fluviatile and amphibious in habit, in this respect resembling Mceritheres but differing in the entire loss of the superior incisor teeth. Early loss of upper tusks released the inferior. In the downturning of the inferior tusks the Dinotheres are paralleled by the Rhynchorostrinz among the Mastodontoidea. In skull Fig. 1. Fundamental arrangement of the cutting teeth in the four superfamilies of the Proboscidea. I.—Meritherium. I1.—Dinotherium, III.—Rhynchotherium. Composition of two species. IV.—Stegodon. S.insignis stage. Scale not uniform. form and in limb and foot structure the Dinotheres parallel the true - proboscideans. They diverge very widely from proboscideans in the evolution of the upper and lower grinding teeth. The primitive Dino- theres present simple bilophodont grinders, similar to those of Meri- therium, and are persistently bilophodont. The upper grinders attain a stage which parallels the molar pattern of the tapir (T'apiroides) among the perissodactyls, but shows no tendency to the trilophodont, tetra- lophodont, or polylophodont structure characteristic of the mastodons and elephants. IIT.—Mastodontoidea The fundamental character of the front teeth in this superfamily is seen in primitive members of the Rhynchorostrine and Longirostrine, namely: _ Second superior incisors enlarged, downturned, divergent, with enamel band on the outer side only. Second inferior incisors downturned (as in Dinotheres) with enamel band on outer side (Rhynchorostrine) or procumbent with no enamel band (Longirostrinz). + AMERICAN MUSEUM NOVITATES © | [No.1 The important functional distinction is that for a very long period of time the upper tusks abraded the outer side of the lower tusks; this probably explains the retention of the superior enamel band. In certain lines (Longirostrine) the procumbent lower incisors persist and the upper incisors retain their primitive downcurved position as in Palzomastodon, In other lines (Mastodontine, Brevirostrine) the lower incisors prac- tically ceased to function; the upper incisors finally turn upward and inward, but may retain the enamel band for a long period (Mastodant ne, fide Schlesinger, and Brevirostrine, vide M. andium). A distinctive character of the grinding teeth of the Mastodonteiden is some evidence of the descent from a hexabunodont ancestral grinder (i. e., with intermediate tubercles or conules) as distinguished from the tetrabunodont ancestral type of Moeritheres and Dinotheres. The rudi- © ments of ancestral conules gave rise to various trefoils or paired median outgrowths or crests, so characteristic of all the Bunomastodontide whether beak-jawed (Rhynchorostrinz), or long-jawed (Longirostrinz), or short-jawed (Brevirostrinz). In each of these subfamilies the grinders independently undergo a more or less closely parallel evolution, evolving single trefoils in Upper eenene and in Miocene time, and double trefoils in Pliocene time. Unlike the Mceritheres and -Dinotheres, the three intermediate molars (i. e., fourth premolar and first and second true molars) invariably become trilophodont, while the third true molars become tetralophodont. At this point there is a divergence into (1) Mastodontide, purely forest living, brachyodont, with simply crested teeth, in which the intermediate molars are persistently trilophodont, with arrested trefoils, and into (2) Bunomastodontide, which pass into tetralophodonty and polylophodonty in some lines, with evolving trefoils. The grinder evolution is adapted to a leaf-browsing habit, in distinction to the prevailing grazing habit developed among the elephantoids. The development of hypsodonty, and chcerodonty (Schlesinger), among these (longirostrine and breviros- trine) browsers is analogous to that in the hippopotami and the hypso- dont suillines. IV.—Elephantoidea One prime distinction in this superfamily is the very early complete loss of the lower incisor teeth, accompanied by the early development of the upper incisors into horizontal or upturned tusks finally devoid of enamel except at the tips in the young stage. Vestigial enamel bands are recorded in early stages of the stegodonts. A second distinctive 1921] CLASSIFICATION OF THE PROBOSCIDEA 5 character is the absence of conule development into trefoils, so charac- teristic of the mastodontoids, and the early tendency to form evenly transverse, more or less mammillate, crests which become in the highest degree hypsodont and polylophodont in adaptation to chiefly grazing habits. 2.—EVOLUTION AND PHYLOGENY OF FAMILIES AND SUBFAMILIES _ The phylogeny of species is now partly known; the Miocene phyla of Europe are being studied by Schlesinger. Systematic classification will follow a full understanding of the evolution and phylogeny. Forty- four generic names have been proposed for these animals and, as yet, an uncounted number of specific names. Generic and specific synonymy awaits (1) a clear separation of the phyla, (2) a determination of the precise geologic levels of types, and (3) a fuller knowledge of all the char- acters of the species. None of these data is complete as yet, hence the present contribution is preliminary to the author’s revision of the syn- onymy.! We await also a restudy by Matsumoto of the characters of the Merritheres, soon to be published in the American Museum Bulletin, also a fuller knowledge of the Dinotheres from unpublished materials in the British Museum. The positive determination of supposed south Asiatic relatives of Meritherium, as well as of Palzomastodon, is very important. Pilgrim (1910, p. 67) provisionally refers to Meritherium a. small, primitive proboscidean molar from the Bugti Hills, Upper Oligo- cene. In the same beds occur Hemimastodon crepusculi, a longirostrine more recent in type than Palzomastodon. Ancestors of the Rhyncho- rostrines and of the true Mastodontines should also be sought in southern Asia. The Rhynchorostrinze The type of this subfamily (Rhynchotherium Falconer, 1868) is a cast (observed in the Genoa Museum by Falconer) of a lower jaw from the Valley of Mexico; the jaw at the t:me had no specific or generic name. The original genotype may be termed Rhynchotherium tlascale, new species, from the locality Tlascala. A similar specimen from Mexico is in the American Museum (Fig. 2 C). The present known range of these animals is Mexico (R. tlascalz), California (R. shepardi Leidy), Colorado (R. brevidens Cope), and Kansas (R. euhypodon Cope). It is note- _ .1There is in preparation an iconographic revision of the known species of Proboscidea to be published in the American Museum Memoirs. 6 AMERICAN MUSEUM NOVITATES [No. 1 worthy that the Rynchorostrines are geologically the earliest forms of proboscideans known in America, i. e., R. brevidens, R. proavus. The presence of a species of this subfamily in Middle Miocene beds was recently confirmed by Loomis. The most distinctive feature of this phylum is the downturning of the symphysis, hence the name Rhyncho- therium, or beak-jawed. A second distinction is the retention of the enamel bands on the lower tusks, to which the specific name euhypodon Cope (i. e., perfect lower teeth) refers. A third distinctive character is the relative simplicity and small size of the third grinders, hence the specific name brevi- dens Cope, applied to the most ancient form discovered in America. It would appear that the retention of effective 4 Fig. 2. Rhynchorostrine types upper and lower tusks relieved the 7 peta tone aides Subba eee grinding teeth in which the evolution B.—R. shepardi Leidy. C.—R. tlascale, is relatively slow. The known geologic new species. All figures 44 natural size. A Seaton succession of species is: Rhynchotlier sume tlascale, new species, 1921. Valley of Mexico dinctheptotdes Andrews, 1909. N.W. Kansas. Pliocene ey _ euhypodon Cope, 1884. Lower Pliocene of Kansas sy shepardi Leidy, 1871. Stanislaus County, California. ?Miocene fa brevidens Cope, 1889. Middle Miocene, Deep River, Oregon | eae proavus Cope, 1873. Middle Miocene, Pawnee Creek, Colorado The maxilla is partly downturned, as well as the mandibular sym- physis. Both the superior and inferior incisors are laterally compressed, bending downward and outward. The grinding teeth remain relatively simple, brachyodont, with posterior grinders not exceeding four and a half crests. The intermediate grinders are trilophodont. The grinding series is reduced to %, + asin M. mirificus. The Longirostrinze , This is the most complete and ancient proboscidean phylum known. The four more or less complete skeletons of 7’. angustidens, T. productus, and 7’. giganteus prove that these were low-bodied animals, with ex- 1921] CLASSIFICATION OF THE PROBOSCIDEA 7 tremely broad pleves and short heavy limbs. In the later phases of their evolution they were probably savanna- and stream-dwellers less closely confined to the forests than the Mastodontine. The extraordinary traveling powers of this family prove that they were well fed and well defended. The geologic and geographic range is indicated in the follow- ing partial list of species. There is an undoubted division of the family into three or four separate phyla, as follows: I II Ill Typical Longirostrines:| Somewhat broader | Without typical tre- Long, narrow inferior | teeth; with typical |foils; Miocene and teeth; with typical tre-| trefoils; Miocene of |Pliocene of America. foils; Oligocene to Plio- | Europe. E. g., T. serridens Cope. cene. E.g., T. angustidens , Cuvier and its successors. Directly ancestral to phylum I appears to be Palzomastodon wintont of the Fayfim Oligocene. The phyla I and II are still to be clearly distinguished in the Miocene of France. Phylum III first appears in the Upper Miocene of America but will probably be found in southern Asia as well as in France. I. T. ANGUSTIDENS PHYLUM ‘ Puiocene Staces, double trefoils, intermediate molars four-crested. Tetralophodon campester Cope, 1878. Republican River, Kansas. - = longirostris Kaup, 1835. Eppelsheim, Germany. s punjabiensis Lydekker, 1886. Middle Siwaliks, India; Dhok Pathan. “y corrugatus Pilgrim, 1913. Lower Pliocene, India. MiocENE AND LowER PLIOCENE STAGEs, intermediate molars three- crested, single trefoils. Trilophodon macrognathus Pilgrim, 1913. Middle Miocene, upper Chinji, India. ee giganteus, new species, 1921. South Dakota. + floridanus Leidy, 1886. Lower Pliocene, Florida. 4 (Megabelodon) lulli Barbour, 1914. Snake River, Nebraska. cis (Tetrabelodon) osborni Barbour, 1916. Near Bristow, Nebraska. s€ (Tetrabelodon) willistoni Barbour, 1913. Nebraska. nin (Mastodon) obscurus Leidy, 1869. Miocene, Maryland. iY (Mastodon) productus, Cope, 1874. Upper Miocene, Clarendon, Texas. be: angustidens palzindicus. Manchhar, Middle Miocene, India. 8 AMERICAN MUSEUM NOVITATES | [No. 1 Trilophodon angustidens Cuvier. ‘Type. Simorre, Middle Miocene, France. pygmzxus. Lower Miocene, Africa. Upper Oricoceng, North Africa, India. Hemimastodon (Tetrabelodon) crepusculi Pilgrim, 1912. Upper Oligo- cene. Bugti, Sind. Palzomastodon wintoni Andrews, 1905. Upper Oligocene, Egypt. — Ill. 7.SERRIDENS PHYLUM Trilophodon (Mastodon) serridens Cope, 1884. Upper Miocene, Clarendon, Texas. “ce * (Tetrabelodon) serridens cimarronis Cope, 1893. Miocene, Clarendon, Texas. / (Dibelodon) precursor Cope, 1893. Blanco, Middle Pliocene, Texas. | tas turicensis Schinz. Middle Miocene, France. In the typical Longirostrines (I) the lower jaws progressively elongate; rapidly attain great length in some of the European and Ameri- can Miocene species, e. g., 7’. lulli, T. giganteus; relatively less elongate in the Pliocene species T'etralophodon longirostris and T. campester, as shown in the accompanying figure (Fig. 4). Throughout, the inferior incisors are without enamel band, spatulate, progressively flattened, horizontally appressed, more or less elongate; as the jaw swings abrading the dentine and inner side of the enamel bands of the superior tusks; the latter are rounded and slightly compressed, never oval as in Rhyn- chotherium. In T. giganteus, new species (Fig. 4 C), the two lower tusks turn toward each other. A characteristic, of all the Mio- Wis. 3.“ Dongieerele. Gueneal cene grinders observed, is the presence jaw and skull, of a single trefoil invariably appear- p_—rAizcPalmomestodon _wintons _Andrews. ing first on the inner side of the upper don productus Cope, Seale not uniform. teeth and on the: outer side of the lower teeth. The double trefoils 1921] CLASSIFICATION OF THE PROBOSCIDEA 9 (inner and outer) on the upper and lower grinders first appear in the uppermost Miocene and Lower Pliocene stages. The intermediate molars in all Miocene species observed are three-crested; hence these animals fall within the genus T'rilophodon Falconer. The transition to the four-crested stage is observed in Upper Miocene types of Europe by Schlesinger. In all Pliocene species observed the intermediate molars are four-crested; hence they fall within the genus Tetralophodon Warren, Falconer. From the Middle Miocene apparently to the close of the Middle Pliocene there was a steady addition of crests to m 3, rising from four and a half crests in the Middle Miocene (7. angustidens) to seven and a half crests in the Middle Pliocene (7. barbouri, new species). At the same time the crests become subhypsodont partly coated with cement. : } py { Samesz7 (J ore Type Fig. 4. Lower jaw and grinding teeth in the longirostrine phylum. A.—Palzomastodon wintoni Andrews. Amer. Mus. No. 13468. B.—Trilophodon productus Cope. Amer. Mus. No. 10582. C.—Trilophodon giganteus, new species. Type, Amer. Mus. No. 17359. D— Tetralophodon longirostris Kaup. Drawn after cast of type. Amer. Mus. No. 10465, Warren Collection. E.—Tetralophodon campester Cope. Type, Amer. Mus. No. 8527. Allfigures }44 natural size. Grinding teeth in C, Trilophodon giganteus, foreshortened in drawing. There is no evidence that any of the typical Longirostrines (I) were transformed into Brevirostrines by jaw-abbreviation. But it would appear that from certain atypical long-jawed forms (perhaps from phylum II of France with broader molar teeth) arose the Eubelodon morrilli type of Barbour, to be described below, without lower tusks, of the Lower Pliocene, Devil’s Gulch, Nebraska. 10 AMERICAN MUSEUM NOVITATES [No. 1 Two New Speciric Stacres.—A new specific stage more recent in character than T. floridanus Leidy appears to be represented in the skeleton and jaws (Amer. Mus. No. 17359) discovered in South Dakota by E. L. Troxell in 1916, to which the name Trilophodon giganteus, new species, may be applied. It exhibits rudiments of double trefoils and the lower incisors are upturned at the sides so as to face partly inward (Fig. 4 C). A second new species is the Tetralophodon barbouri, of which the type (Neb. State Mus. No. 4.22.6.16) is agrinding tooth with double trefoils, seven and a half to eight crests, and cement; thus ad- | vanced much beyond TJ’. campester with six and a half crests on the superior grinders, without cement. The Brevirostrines of South America It now appears that the South American Brevirostrines, M. andium and M. humboldtii, were not derived directly from the Eurasiatic Breviros- trines, typified by the Eurasiatic M. arvernensis and M. sivalensis, nor were the South American forms descended from the M. mirificus Leidy, which appears to be an Eurasiatic migrant. The reason for this opinion is that the molar pattern of M. andium and M. humboldtii is of simple bunomastodont type, with single trefoils (M. andium) and double trefoils (M. humboldtiz), a type familiar in the typical American Longirostrines and Rhynchorostrines only. The Eurasiatic Brevirostrines and the American M. mirificus, on the other hand, have molar grinding teeth which Schlesinger aptly terms cherodont (pig-like, covered with tubercles). Another typical longirostrine character in M. andium is the broad enamel band on the superior tusks. It is not known in any European Brevirostrine. Consequently, it appears probable that the South Ameri- can bunomastodonts independently abbreviated the jaw and that they may possibly be related to the only American form known in which the jaw is abbreviated, namely, Eubelodon morrilli Barbour. The phylum thus appears to be as follows: Mastodon humboldtii Cuvier. Lower Pliocene " andium Cuvier. Upper Pliocene. Andean region “ (Dibelodon) tropicus Cope, 1884. Valley of Mexico 3 successor Cope, 1892. Blanco, Middle Pliocene, Texas ?Eubelodon morrilli Barbour, 1913. Devil’s Gulch, Lower Pliocene, Nebraska Eubelodon morrilli exhibits an abbreviated lower jaw, no lower tusks, superior ‘tusks without enamel band (Barbour, 1920). Thus it differs from M. andium in one important character, namely, in the absence of the enamel band. . 1921] CLASSIFICATION OF THE PROBOSCIDEA 11 The Brevirostrines of Eurasia and North America Suddenly in the Lower Pliocene of the Siwaliks, India, appears M. hasnoti Pilgrim, in the Dhok Pathan horizon, stage of Pikermi, of Eppels- heim. According to Pilgrim this differs from the true M. sivalensis Fal- coner of the Middle Pliocene, Tatrot horizon, in having m# tetralopho- dont and in the slightly marked alternation of the cones. The Middle Pliocene M. sivalensis has a pentalophodont m# and the cones are more alternate. The closely related Brevirostrine (M. arvernensis‘ of southern Europe ranges from the lignites of Casino (Middle Pliocene) to the First Interglacial stage in England (Crag). Similarly, the M. mirificus of Leidy was first described in association with Elephas imperator and de- termined by Hay as of the Aftonian, i. e., First Interglacial stage. The known succession of described forms in these browsing, forest-living elephants of the warm zones is as follows: Mastodon arvernensis. Norwich fluviomarine crag, Norfolk, England vs arvernensis Croizet et Jobert. Typical. Upper Pliocene of Auvergne 8 mirificus Leidy. Typical of EF. imperator-Equus zone of Nebraska chapmani Hays, 1834. Geologic level unrecorded “ sivalensis Falconer. Tatrot horizon, Middle Pliocene, Siwaliks, India 6 brevirostris Gervais, 1846. France “e pentelict Wagner. Pikermi, Lower Pliocene, Greece “ hasnoti Pilgrim. Dhok Pathan horizon, Lower Pliocene, India The unique feature of the members of this phylum is the alternation of the inner and outer cones of the grinding teeth, which are placed obliquely instead of opposite each other. A further distinctive character is the brevirostral, brachycephalic skull, which parallels that in all the Elephantoidea. The jaws rapidly abbreviate and lose the lower tusks in an early stage. The upper tusks, at first elongate and horizontal (M. arvernensis), are upturned and out-turned (M. mirificus). No enamel band has been observed. The grinding action of the teeth, like that of the pigs, peccaries, and hippopotami, explains the early evolution of double trefoils and finally of accessory tubercles; hence the apt term cheerodont, applied by Schlesinger. The intermediate grinders, at first trilophodont, become tetralophodont (M. hasnoti), pentalophodont (M. sivalensis). The posterior grinders m $ evolve six and a half crests (M. arvernensis), the cones becoming subhyposodont. Within the Brevirostrine are at least two phyla: I. Mastodon arvernensis, termed Dibunodon by Schlesinger; and II. Mastodon pentelici, termed Cherolophodon by Schlesinger, with a longer sym- physis, represented in Samos and in Maragha, Persia. III. To a third 12 AMERICAN MUSEUM NOVITATES [No. 1 phylum may belong M. hasnoti, M. sivalensis, India (see Schlesinger,! pp. 224-229), M. mirificus, North America. The Mastodontinz . Forest-living animals are rarely found fossil, e.g., the forest-living Chalicotheres of the entire Tertiary. The massive, low-bodied, low- headed, well-defended Mastodontinze probably evolved chiefly in the north temperate forests of Eurasia. That they were cold-loving animals is shown by their avoidance of southern Eurasia, except for the single appearance of M. borsoni in the Upper Pliocene of the Val d’Arno in northern Italy. Of one of the oldest known forms (M. tapiroides Cuvier) of the Middle Miocene of France, according to Schlesinger, occasional descendants are found in Europe. All are zygolophodont; the cones of the grinders turning into transverse crests, the intermediate conules re- maining rudimentary, no trefoils forming. Superior incisors (Miocene) retain enamel bands (Schlesinger), but, released from apposition with the lower incisors, turn upward and outward; the lower incisors persisting as abbreviated, horizontal, rounded tusks. The jaws and skull abbre- viated, brachycephalic. Intermediate grinders persistently trilophodont. Posterior grinders progressing to a tetralophodont stage only. The vertical chopping motion of the jaws, as distinguished from the grinding motion in the Longirostrines and the Brevirostrines, explains the reten- tion of simple crests and the non-development of trefoils and tubercles. All ancestral stages will doubtless be found in northern Eurasia. The M. borsoni of the Upper Pliocene forests extends into Russia and is directly or indirectly ancestral to the M. americanus, a Pleistocene arrival in North America. The chief known specific stages are: Mastodon americanus. Pleistocene “* progenium Hay. Missouri Valley, Iowa borsoni Hays. Upper Pliocene, Val d’Arno, Italy ? ‘5 turicensis (tapiroides). -Middle Miocene, Simorre, France pyrenaicus. Middle Miocene, France tapiroides Cuvier. Middle Miocene, France These six outstanding species are doubtfully placed in one phylum awaiting further evidence. "> ” ns The Stegodontinz of Southern Eurasia We observe that the Stegodonts are persistent browsers, probably tropical, forest-living proboscideans. According to Pohlig, from the 11917, Denksch. K. K. Naturhist. Hofmuseums, I, Geol.-Paliontol:, Rheihe 1, pp. 1-230. 1921] CLASSIFICATION OF THE PROBOSCIDEA 13 skeleton discovered in Trinil, Java, they have short, massive bodies like those of the Mastodontine of the north temperate forests. The skull and tusks do not lead into either the Elephantinz or the Mammontine types. The phyletic succession of species is clarified by Pilgrim’s geologic sub- divisions of the Siwaliks and by his observations on the succession of specific types, which provisionally may be arranged as follows: Stegodon ganesa (male), S. insignis (female). Upper Pliocene and Lower Pleisto- cene, southeastern Eurasia 3 stegodontoides Pilgrim, 1913. Lehri, Geom Siwaliks. ?Middle Pliocene “« cliftii Falconer. Dhok Pathan horizon, Lower Pliocene, India ‘si bombifrons Falconer. Dhok Pathan, India = cautleyi Lydekker. Perim Island. Upper Miocene “« latidens Clift. Irawadi River, Asia The distinctive feature of the grinding teeth is the rapid multiplica- tion of transverse crests which rise from the formula 4.5.6.6.7-8 in S. cliftit (Lower Pliocene) to 5.9.10.12.13 in S. insignis (Lower Pleistocene). Jaw rapidly abbreviated. Upper tusks straight, parallel, slightly up- curved (adapted to dense forests). Grinders brachyodont to subhypso- dont, crests breaking up into small mammille, valleys filling with cement. The Loxodontinz The late appearance (Upper Pliocene) of these Loxodonts in Italy, the dwarfed evolution in all the Mediterranean islands, the still later geologic appearance (Lower Pleistocene) in Asia, and the existing exclu- sive occupation of Africa by a great variety of Loxodonts, point to Africa as the original center of adaptive radiation of the Loxodontine. This phylum is abundant in the Pleistocene of northern Africa, e.g., L. atlanticus. There is, on the other hand, no trace of these animals in the Pliocene Siwaliks of India; Pilgrim records the first occurrence of L. antiquus namadicus in the Lower Pleistocene of Godavari and Narbada, which also contains Stegodon ganesa and S. in ignis. The gigantic, wide-spreading upper incisors implanted in the maxillo- premaxillary rostrum are quite distinct from those of either the Stego- dontine, Mammontine, or Elephantine. The height attained at the shoulder (L. namadicus) is estimated at sixteen feet (Pilgrim), five feet taller than the existing L. africanus. The affinity of L. namadicus to L. antiquus (Upper Pliocene to Middle Pleistocene, Europe) and to L. africanus has been pointed out by Falconer, Leith Adams, Pohlig, and Pilgrim. Within the subfamily Loxodontine there are a great number 14 AMERICAN MUSEUM NOVITATES [No. 1 and variety of species undoubtedly belonging to more than two phyla, namely: Loxodonta africanus. Recent, Africa, including fifteen species and subspecies antiquus. Upper Pliocene to Middle Pleistocene, Europe namadicus. Lower Pleistocene, India and southern Eurasia creticus melitensis Mediterranean Islands mnaidriensis atlanticus. Pleistocene of northern Africa Another fossil member of this race has recently been recorded in German West Africa. The Mammontinze. The Mammoths It is a striking fact that the oldest geologic appearance of a member of the true Elephantoidea is the Elephas planifrons occurring in the Pinjor horizon, Upper Siwaliks, Middle to Upper Pliocene, India. All the fauna of the great Siwalik deposits underlying this geologic level, according to Pilgrim, contain only Stegodonts, Longirostrines, and Bre- virostrines. This is significant of a north Eurasiatic center of adaptive radiation of both the Mammontine and the Elephantine. The chief distinction between these two subfamilies lies in the flattened forehead of the Mammoths, to which the specific name planifrons refers, a fore- head which becomes increasingly concave and compressed antero- posteriorly until it reaches the high, narrow peak of EH. imperator. Again, the succession of species is probably polyphyletic, awaiting analysis. In descending order the main geologic succession is as follows: Elephas primigenius Blumenbach. Northern Eurasia and North America, Upper Pleistocene columbi Falconer. Middle Pleistocene, North America imperator Leidy. Lower Pleistocene, North America trogontherti Pohlig. Lower Pleistocene, Europe hysudricus Falconer. Uppermost Pliocene, India ‘« meridionalis Nesti. Upper Pliocene, Val d’Arno, Italy “« planifrons Falconer. Pinjor horizon, Middle to Upper Pliocene, India, also Austria and Bessarabia (Russia) The position of E. hysudricus in this phylum is doubtful. The cranium referred to this species by Falconer is not of the mammontine type. In 1913 Pilgrim traced back E. planifrons to the Upper Miocene Stegodon cautleyi, but it would appear at present that none of the known Stegodonts gave rise to the Mammoths. Extreme cranial abbreviation, 1921] CLASSIFICATION OF THE PROBOSCIDEA 15 hyperbrachycephaly, and acrocephaly are great characteristics of all the phyla in this subfamily (excepting possibly that to which EF. hysudricus belongs). There is a wide range of divergence in the thickness and mul- tiplication of the lamelle of the grinders. Elephas imperator may be de- rived from the E. meridionalis type, with very few lamelle, composed of thick enamel bands and with a great coating of cement, or from the E. planifrons Falconer type. The EF. primigenius phylum presents the high- est lamellar formula known, with relatively little cement; this phylum is also distinguished by the loss of a digit in the pes, becoming tetradactyl, a unique character among proboscideans. Very great shoulder height, estimated at thirteen feet, is attained by HE. imperator in the favorable environment of the southern United States and Mexico, as compared with the height of nine feet six inches attained by EF. primigenius in the frigid north. usltcents The Elephantinz Like the Mammoths, these animals suddenly appear in the Upper of India. They are not found in the Lower Pleistocene, where their place in the fauna is taken by the invading loxodont, L. namadicus. In distinction from the Mammoths, the forehead is prominent, convex, in adults highly convex. The upper tusks extend forward and outward, slightly upcurved, not crossing each other in old age as in the more recent _Mammoths E. primigenius and E. imperator. Unlike E. primigenius, — five digits persist on the fore and hind feet. Skull brachycephalic. Jaws abbreviated. Vestigial lower incisors, enamel remaining on the tips only in young stages. Mammille more numerous than in the Loxodonts; less numerous than in E. primigenius. The Unknown Home of the Elephantoidea It appears from the above preliminary studies that the proboscidean phylogeny is still subject to many emendations, transpositions, and cor- rections. It isin the rocks of the great unexplored regions of Eurasia and of Africa—nearly a hundred-fold greater than the regions explored and known—that we must look for the ancestry of the four great branches of the Elephantoidea, namely, the Stegodonts, the Loxodonts, the Mam- moths, and the true Indian elephant type. December 30, 1920. AMERICAN MUSEUM NOVITATES No. 2 DESCRIPTIONS OF APPARENTLY NEW BIRDS FROM BOLIVIA, BRAZIL, AND VENEZUELA By Frank M. CHAPMAN NA ISTO Issued January 31, 1921 By ORpDER OF THE TRUSTEES OF THE AMERICAN MUSEUM OF NATURAL HISTORY New Yorxk City | ——S ; | AMERICAN MUSEUM NOVITATES Number 2 | January 31, 1921 59.82(8) DESCRIPTIONS OF APPARENTLY NEW BIRDS FROM BOLIVIA, BRAZIL, AND VENEZUELA By Frank M. CHAapMAN The material on which this paper is based is contained in the American Museum and in a small collection recently submitted to us for determination by Dr. Emilia Snethlage of the Museu Goeldi at Para. The color terms employed are those of Ridgway’s ‘Color Standards and Color Nomenclature’ (Washington, 1912). Capito brunneipectus, new species Speciric CHaRacTerRs.—Upperparts resembling those of Capito auratus auratus, but male with the throat pale cinnamon-buff, breast ochraceous tawny, abdomen light yellowish olive widely margined with greenish yellow; no wing-bars. Female with the throat and breast richer, the former spotted with black, the greater wing- coverts, except terminally, externally margined with old gold. TypE.—No. 13709, Museu Goeldi, @ ad., June 16, 1917, Villa Braga, Rio Tapajoz, Brazil; Emilia Snethlage. Description OF Mate.—Entire crown shining old gold, browner on the fore- head, yellower on the nape; cheeks, sides of the nape, and back black; inter- scapulars widely margined with lemon-yellow forming two lines from the nape to the lower back, the latter and rump margined with greenish yellow; tail olivaceous, nearly uniform; wings fuscous, all but the outer quills externally margined with dark citrine, the larger part of the outer web of the tertials olive-buff; lesser and median coverts shining blue-black like the back; greater and primary coverts duller; lower coverts and inner margins of quills buffy; throat and sides of the neck pale cinnamon-buff unmarked; breast-band deep buckthorn-brown narrowly tipped with greenish yellow; feathers of the rest of the underparts broadly tipped with greenish yellow not wholly concealing their olivaceous bases, tibize with less yellow, ventral region whiter; lower tail-coverts olive tipped with whitish; feet plumbeous; bill black, a horn-colored band at the base of the mandible. Wing, 81.5; tail, 48; tarsus, 22.5; culmen, 21.5 mm. DEscRIPTION OF FEMALE.—Similar to the male but with the margins of the interscapulars duller, the interscapulum and lesser wing-coverts more or less yellow margined; the greater coverts externally olive-citrine, except at the tip; the throat and sides of the neck with numerous round black spots forming a narrow malar streak, the breast-band deeper, argus-brown; the abdominal region somewhat more olivace- ous. Wing, 79-82; tail, 50-54; tarsus, 21-21.5; culmen, 20.5-21.5 mm. (3 speci- mens). 2 AMERICAN MUSEUM NOVITATES [No. 2 SPECIMENS ExaMIneD.—Capito brunneipectus. Brazil: type locality, 1 7,399. Capito auratus auratus. Colombia: Buena Vista, 2 77,5 29. The discovery of this bird not only adds a very distinct species to the genus Capito but extends the known range of the genus into the southern part of lower Amazonia. It also gives further emphasis to the apparently restricted areas occupied by certain strongly marked species of this group. Thus Capito squamatus is known only from western Ecuador and southwestern Colombia; C. quinticolor from western Colombia; C. hypoleucus from the Cauca-Magdalena Fauna of Colombia; C. dayi from Porto Velho on the Rio Madeira; and C. brunneipectus from Villa Braga on the Rio Tapajoz. In view of the antiquity of the genus to which they belong, can it be possible that the restricted ranges of these birds indicates that they are disappearing species? Nonnula amaurocephala, new species Speciric CHaracters.—Most nearly related to Nonnula ruficapilla (Tsch.), but sides of the head and neck, as well as crown, deep burnt sienna. Typr.—No. 12490, Museu Goeldi, @ ad., July 18, 1916, Manacapurti, Rio Solimoés, Brazil; F. Lima. DeEscRIPTION OF MatE.—Entire head, including nape, lores, eye-ring, auriculars, and sides of the neck rather deep burnt sienna, becoming light Sanford’s brown on the underparts and ochraceous buff on the flanks; the lower abdomen, ventral region and under tail-coverts white; back dull, Brussels-brown; remiges basally the color of the back, blacker apically except on the outer feathers which are uniform; wing- quills like rectrices, the inner quills lightly margined with buckthorn-brown; greater, median and lesser wing-coverts like the back; bend of the wing ochraceous orange; lining of the wing ochraceous buff; feet blackish brown; bill black, the mandible yellowish except at the tip and sides. Wing, 61; tail, 52; tarsus, 14; culmen (broken). DESCRIPTION OF FeMALE.—Similar to the male but underparts somewhat paler. Wing, 63; tail, 55.5; tarsus, 14; culmen, 25 mm. Specimens Examinep.—Nonnula amaurocephala. Brazil: type locality, 1 o, : et Nonnula ruficapilla, Plate xuv1, fig. 1, Sclater, ‘Monograph of the Jacamars and Puffbirds.’ While this form may represent N. ruficapilla, it appears to differ spe- cifically from that bird, in which the sides of the head and neck are gray. Rhopochares cochabambe, new species A Spreciric Cuaracters.—In size and general coloration nearest Rhopochares ruficapillus (Vieill.), but back grayish sharply defined from crown, as in Rhopochares torquatus (Swains.). Type.—No, 139234, Amer. Mus. Nat. Hist., @ ad., Tujima, alt. 8200ft., Prov. — Cochabamba, Bolivia, September 25, 1915; Miller and Boyle. ee ee ee ee 1921] CHAPMAN, NEW SOUTH AMERICAN BIRDS 3 Description oF Mate.—Crown between auburn and burnt sienna, forming a distinct cap; lores and frontal region (less clearly) buffy; auriculars gray or grayish white; postorbital region and sides of the nape pale smoke-gray, becoming deeper gray on the back and rump, where more or less washed or mixed with cinnamon- rufous; all the tail-feathers, including the central pair, blackish; the central pair more or less margined externally with grayish, in some cases slightly margined or indented with white, usually on the inner web; remaining feathers tipped with white and with white bars on the inner web, not reaching the shaft, the outer feathers usually indented with white on the outer web; wings externally ochraceous tawny, lined with light ochraceous buff; underparts whitish, the throat, ventral region and flanks washed with buffy, the breast barred with black as sharply but less extensively than in Rhopochares ruficapilla. Wing, 68-70; tail, 73-75; tarsus, 25-27; culmen, 15-16 mm. DESCRIPTION OF FEMALE.—Crown cap as in the male, the back averaging more cinnamon-rufous than in the male; tail like the crown, the feathers with paler tips and indistinct blackish bands terminally; underparts much paler than in R. rufi- capilla, buffy white, without black bars. Wing, 64-69; tail, 67-68; tarsus, 26-28; culmen, 14.5-15.5 mm. SpEcIMENS ExamMINED.—Rhopochares cochabambe. Bolivia: type locality, 7 oo,7 29; Valle Grande, 7200 ft.,1 9. Argentina: Perico, 4000 ft., Prov. Jujuy, Ed, Rhopochares ruficapilla. Argentina: Buenos Aires, 2 7; La Plata, 1 9. Uru- guay: Concepcion, 1 &@. Brazil: Rio Grande do Sul, 1 &@; Ypanema (Natterer), ie. This species combines in an interesting way the color characters of R. ruficapilla and R. torquata and, while evidently representing the former, is, in my opinion, specifically distinct. It ranges at least from the Province of Cochabamba, Bolivia, southward to the Province of Jujuy, Argentina, whence we have a male which, aside from having the black breast-bars somewhat wider, agrees with our topotypical series. Possibly specimens recorded from Tucuman and Salta! as Tham- nophilus ruficapillus should be referred to the species here described. Microrhopias emilizs, new species Specitric CHARACTERS.—Male similar in color to male of Microrhopias boucardi virgata but larger, female with the throat and breast chestnut, the belly black. Typr.—No. 10775, Museu Goeldi, 9 ad., May 28, 1914, Alta Mira, Rio Xingu, Brazil; Emilia Snethlage. Description oF MALEe.—Rich velvety black; remiges and rectrices somewhat duller, interscapulars snowy white basally, this color not wholly concealed by their black tips; greater wing-coverts broadly tipped with white, median coverts with rounded white terminal spots, lesser coverts and wing “lining” white; all but median 1Dabbene, 1910, Anales Museo Nacional, Buenos Aires, (3) XI, p. 284. 4 AMERICAN MUSEUM NOVITATES [No. 2 rectrices with white tips 7-8 mm. in length on outer feathers, decreasing abruptly to 2 mm. on inner pair; feet grayish black; bill black. Wing, 55-56; ‘tail, 52-53; tarsus, 16.5; culmen, 14.5-15 mm. (2 spentanthay: DESCRIPTION OF FEMALE.—Similar to the male but slightly duller, particularly on the abdomen, the throat and breast deep, rich eo Wing, 53-54; tail, 51-52; tarsus, 16-17; culmen, 15 mm. SPECIMENS ExaminepD.—Microrhopias emiliz. Brazil: type locality, 1 7,1 9; Rio Tocantins, 1 @,1 @. Microrhopias boucardi virgata. Panama, a large series. Microrhopias boucardi consobrina. Ecuador and Colombia, a large series. Microrhopias boucardi bicolor. Brazil: Rio Tapajoz,1<, 19 ; Rio Roosevelt, 1%; Porto Velho, 1,19; Baron Melgaco, 2? 9. Microrhopias quizensis. Ecuador, 1 9. It seems unusually appropriate to name this new species, in which. the characters are shown only by the female, after its discoverer Dr. Emilia Snethlage, whose energy in the field and zeal in the study have added so greatly to our knowledge of the birds of Amazonia. Drymophila devillei subochraceus, new subspecies SupsPeciric CHARACTERS.—Similar to Drymophila devillei devillei Men. and Hellm., but both sexes with entire underparts ochraceous buff, paler on the throat and center of the abdomen, deeper on the breast and flanks; outer margins of inner wing-quills more ochraceous; female with ochraceous streaking of the upperparts and of wing-coverts somewhat deeper. Typr.—No. 10777, Museu Goeldi, & ad., November 7, 1914, Rio Curud Se tributary of the lower Xingu), Brazil; Emilia Suethlage. SpecimENS EXAMINED. i Pepinieahile devillei subochraceus. Brazil: type locality, 1 #, 1 @. Drymophila devillei devillei. Bolivia: Jatumpampa, 1 &@ (type of Drymophila phantatis Cherrie); Mission San Antonio, Rio Chimoré, 4 oc’, 3 2 @. It is interesting to observe that in this form the male and female are essentially alike below, the former having the breast buff instead of white, as in its allies true devillei and caudata. The female of caudata caudata nearly resembles subochraceus in the ground-color of the under- parts, but is somewhat paler and slightly yellower. The female of caudata klagesi, on the contrary (if our two specimens properly represent it), has the breast white as in the male. In coloration of the underparts klagesi closely approaches true devillei from which it is distinguished chiefly by the lack of white markings on the sides of the central rectrices. Drymophila phantatis Cherrie is apparently not separable from devillet devillei Men. and Hellm. i 1921] CHAPMAN, NEW SOUTH AMERICAN BIRDS 5 Hypocnemis hypoxantha ochraceiventris, new subspecies Susspeciric CHARACTERS.—Similar to Hypocnemis hypoxantha hypoxantha Scl., but somewhat larger, with a larger bill the color throughout browner, the underparts paler yellow, the breast more streaked, the flanks and lower tail-coverts pale ochrace- ous buff, the abdomen slightly washed with this color; wing-coverts, in the female, tipped with warm buff. Male: wing, 55; tail, 46; tarsus, 20; culmen, 14 mm. Female: wing, 53.5; tail, 44; tarsus, 19.5; culmen, 14 mm. Type.—No. 10788, Museu Goeldi, # ad., May 28, 1914, Alto Mira, Rio Xingu, Brazil; Emilia Snethlage. SpecIMENS ExamInED.—Hypocnemis hypoxantha ochraceiventris. Brazil: type locality, 1 #,1 Q. Hypocnemis hypoxantha hypoxantha. Colombia: La Morelia, 1 @. Brazil: Upper Amazon, Pl. xu, P. Z.8., 1868, @ ad. Hypocnemis Seedcane fabAince: Peru: Perené, 1 &, 1 9; Rio Javara, 1 o, 1 9; La Pampa, 1 @. Bolivia: Todos Santos, 7 07,3 9 9. In the coloration of the underparts this species is essentially like H.f. subflava. It is, however, separated from that species and from H. f. flavescens by its yellow instead of white superciliary and more olive, less striped upperparts. The form from Bolivia and Peru (subflava) is evidently a southern race of flavescens, but the specific distinctness of hypoxantha is indicated by its occurrence in the heart of the range of flavescens. . Hypocnemis collinsi Cherrie proves to be inseparable from H. f. subflava Cab. Siptornis punensis cuchacanche, new subspecies Supspeciric CHARACTERS.—Similar to Siptornis punensis punensis Berl. and Stolz., but upperparts more distinctly streaked; underparts, particularly ventral region and lower tail-coverts paler; rufous markings of wing decidedly lighter; tail longer. Resembling Siptornis punensis lilloi Oust., but general tone of the upperparts less rufescent, margins of tertials and greater wing-coverts paler, sayal-brown rather than cinnamon-rufous; rufous band in wing-quills paler and less extensive, particu- larly on outer quills. Typr.—No. 137292, Amer. Mus. Nat. Hist., ad., Cuchacancha, Bolivia, June 13, 1915; Miller and Boyle. Specimens EXaMINED.—Siptornis punensis cuchacanche. Bolivia: type locality, 9 Ho ads.,1 9 ad. Siptornis punensis punensis. Bolivia: Guaqui, 1 @ ad.,1 9. Siptornis punensis lilloi. Argentina: Above Tafi del Valle, alt., 9500 ft.,3 7a ads. (including type of Siptornis punensis rufala Chapman), 1 ed, 43% imm.,2 99 imm. Siptornis punensis punensis was described from Puno, Peru, on the northwest shore of Lake Titicaca. It is probable, therefore, that two specimens from Guaqui, distant 115 miles, on the southern shore of the 6 AMERICAN MUSEUM NOVITATES | [No. 2 lake, recently presented to the museum by Lord William Percy, are essentially typical of that form. They show that specimens from Cuchacancha near Cochabamba, which, in default of topotypical material I had referred to punensis punensis,! are separable from that race as well as from the Argentine race. The latter I described as Siptornis punensis rufala (loc. cit.), but Dr. Hellmayr assures me that this bird is the same as Siptornis lilloi Oust., an opinion which may be accepted as authoritative, though as sited 3 in describing rufala I am unable to make Oustalet’s description conform with our Argentine specimens. Measurements | WING TAIL CULMEN Cuchacancha,” Prov. Coch., Bolivia, 7 76 94 13.5 Cuchacancha,” Prov. Coch., Bolivia, 7 78 97.5 14.5 Cuchacancha,? Prov. Coch., Bolivia, 7 73 93.5 14 Cuchacancha,? Prov. Coch., Bolivia, #7 75 94 13.5 Guaqui,® Bolivia, #7 78 88 13 Cuchacancha,’ Prov. Coch., Bolivia, 71 89.5 13 Guaqui,’ Bolivia, 9 69 87.5 13.3 Cistothorus platensis caracasensis, new subspecies SuBsPEeciIFIC CHARACTERS.—Similar in general tone of color to Cistothorus meri- dx Hellm., but somewhat less rufescent, the crown being uniform in color and olive- brown rather than Prout’s brown; differing also in size, proportions and pattern, the tail being longer and equal to the wing, the tarsi and bill shorter, the latter propor- tionately more decurved; the rump unbarred, the scapulars less striped, the wing- coverts less definitely barred, the bars of the median rectrices broken, the flanks faintly, if at all, barred, the superciliary barely evident. In general pattern of colora- tion much nearer to Cistothorus platensis xquatorialis Lawr., but much less rufescent, the central tail-feathers not regularly barred, the wing, tarsi and bill shorter, the tail nearly as long, and therefore proportionately longer. Male: wing, 44.5; tail, 43.5; tarsus, 16; culmen, 11.5 mm. ' Typr.—No. 150610, Amer. Mus. Nat. Hist., @ ad., Cotiza, Caracas, Venezuela; August 22, 1918; George K. Cherrie. Specimens ExamMINep.—Cistothorus platensis caracasensis. Venezuela: type locality, 2 7; ? Escorial, near Mérida, 1 o. Cistothorus meride. Venezuela: Sierra Nevada, Mérida, alt. 3000 m., 1 @&; Conejos, near Mérida, 3000 m., 1 o. 11919, Bull. Amer. Mus. Nat. Hist., p. 329. 2Siptornis punensis cuchacanche. 3Siptornis punensis punensis. 1921] CHAPMAN, NEW SOUTH AMERICAN BIRDS 7 Cistothorus platensis xquatorialis. Ecuador: Pichincha (type locality), 1 3; Mt. Chimborazo, 3 oc". Colombia: Valle de las Pappas, Cen. Andes, 1 @,1 9; Santa Isabel, Cen. Andes, 3 oo, 5 9 9; Chipaque, near Bogotd, 1 o; Choachf, near Bogota, 4. Cistothorus apolinarit. Colombia: Suba Marshes, Bogoté (type locality), 9. Mr. Cherrie’s discovery of a Marsh Wren in a region so well known as the vicinity of Caracas, extends the known range of this group east- ward from Mérida (a distance of some 300 miles) and, incidentally, is evidence of his skill and energy as a collector. The Caracas race is the least rufescent of any of the described forms of the group, and in genéral tone of coloration closely approaches Tel- matodytes palustris mariane (Scott) of the southeastern United States. I have seen neither Cistothorus platensis tame Cory, from the Paramo of Tam4, on the Venezuelan-Colombian boundary, nor Cisto- thorus xquatorialis fulvescens Todd, from the Paramo of Guerrero, San- tander, Colombia. The first is described as being ‘‘more rufous brown,” the second as “‘more rufescent” than xquatorialis and presumably, therefore, they both differ more from caracasensis than does xquatorialis. Todd (1919) makes no mention of Cory’s (1916) race and, since both birds are from the same general region and appear to differ from x#qua- torialis in much the same manner, their comparison is to be desired. The study of our material discloses the interesting fact that, as at Bogota, two forms of this Wren are found in the Mérida region. The - first, described by Hellmayr as Cistothorus platensis meridz, is, in my opinion, unquestionably a distinct species distinguished from xquatorialis by its short tail (35 mm.), long hind-toe (19 mm.), pronounced white superciliaries, barred wing-coverts, rump, flanks, etc., as more fully noted in the diagnosis of caracasensis. The second, of which we have only one specimen, obviously represents the Caracas race with which it agrees exactly in size and very nearly in color, but has the back more broadly barred with white, the bars on the central tail-feathers com- plete, as in xquatorialis. If these differences are constant they are clearly of subspecific value. In describing this proposed new race as a subspecies of platenszs, I merely follow a convention. As a matter of fact, I have no specimens of platensis and it is possible that the Andean forms of the zquatorialis group may not intergrade with the form of the South Temperate Zone. The known facts in the variation of the Andean races are, however, too contradictory in character to be of predicatory value. Thus, true xquatorialis ranges from at least Chimborazo to Bogota, a distance of 8 AMERICAN MUSEUM NOVIT ATES [No. 2 some 600 miles, without exhibiting racial variations; but in passing from the Péramo of Choachi to the Savanna of Bogota, a distance of 20 miles, we go from the range of xquatorialis to that of the specifically distinct, but representative, C. apolinari. Furthermore, two forms, which I consider as also specifically distinct, are here shown to inhabit the Mérida region. It is obviously, therefore, not always safe to assume that what we believe to be representative forms are also intergrading races. Measurements! Hind-toe Wing Tail Tarsus and Claw Culmen Chimborazo,” Ecuador 48 44 20 1233 12 Pichincha,” Ecaador 48 44 18 14 13 Santa Isabel,? Col. 49 43 20 14 13 Choachi,? Col. 48 44.5 19.5 14.5 12 Choachi,” Col. 47 43 20 14 11.5 Caracas,* Venezuela 43 41.3 16 12.5 11 Caracas,® Venezuela 44.5 43.5 16 13 11.5 Escorial,®? Venezuela 42 42 16 13 ALS Sierra Nevada,* Mérida, Venezuela 47 35.5 19 19 12 Conejos,* Mérida, Venezuela 46 35 18 19 12 1A number of the following specimens are not sexed but there appears to be no appreciable sexual difference in size in this group. ; 2Cistothorus platensis xquatorialis, 8Cistothorus platensis caracasensis. 4Cistothorus meride, - | uae Ms ys AMERICAN MUSEUM NOVITATES No. 3 : A HAWAIIAN RACE OF CARANGOIDES GYMNOSTETHOIDES By Jonn TREADWELL NICHOLs Issued March 9, 1921 By ORDER OF THE TRUSTEES OF THE AMERICAN MUSEUM OF NATURAL HISTORY New York City : ae vy Sega ted) AMERICAN MUSEUM NOVITATES Number 3 March 9, 1921 59.7.58C(96.9) A HAWAIIAN RACE OF CARANGOIDES GY MNO- STETHOIDES By JoHN TREADWELL NICHOLS The American Museum is in receipt of two specimens of Carangoides secured by Dr. B. W. Evermann in the Honolulu market, August 25, 1920, a little over a foot long to base of caudal. In ‘Fishes of the Hawaiian Islands,’ 1905, Jordan and Evermann, two closely related species of this genus are recognized as ferdau (Forskal) and gymnostethgides Bleeker. Our two specimens are referred to the latter as there described, though they agree exactly with neither one. As a matter of fact, they do not agree exactly with each other. One of them has a longer dorsal lobe (1.8 in depth of body, 2.0 in base of fin) and the ventral surface little paler in color than the dorsal. The other has a shorter dorsal lobe (2.0 in depth, 2.2 in base of fin), the ventral surface pale, and also differs somewhat in outline. This may be a sexual difference. It further seems not impossible that ferdau from the Hawaiian Islands, Jordan and Evermann, is a variation of the same fish. True ferdau from the Red Sea and East Indies has appreciably fewer fin-rays. Leaving out of consideration proportional head and depth measure- ments which vary with age in this genus, length of the maxillary sepa- rates gymnostethoides from orthogrammus from off the West Coast of North America, the maxillary not reaching below orbit in the former and extending to nearly opposite front of pupil in the latter. In our two specimens the maxillary just reaches the anterior edge of orbit, but it is described for Hawaiian gymnostethoides by Jordan and Evermann as reaching opposite front of pupil. The head of orthogrammus is given as 2%, which would be large for a fish with the accompanying depth of 3%, and this may be a good char- acter for that species. Also, in Hawaiian gymnostethoides the dorsal lobe is about » as high as the depth of body and base of the soft dorsal, this being higher than described for typical East Indian gymnostethoides or for orthogrammus. 2 AMERICAN MUSEUM NOVITATES | [No. 3 From data which may be gleaned from the literature there seems little more reason for recognizing orthogrammus as distinct than for recognizing the Hawaiian fish. It is certain that we are dealing with variable and very closely related forms. Perhaps the best plan is to provisionally consider orthogrammus conspecific with gymnostethoides and to separate the same into three races. The diagnostic characters of the inclusive species are: teeth on the palate and on both jaws, center of chest with a scaleless area, body mostly sealed, lobe of dorsal fin moderately elevated, not much more than half base of fin; slender, depth more than 2% in length to base of caudal, dorsal soft rays 29-32, anal 25-27, arch of lateral line low, the scutes small, restricted to the posterior portion of the straight part, 25. more or less. The three races would be differentiated as follows. 1. Lobe of soft dorsal about 4 depth of body and base of fin. Naked area on chest TPO FASE Ae ewan a eee C. g. evermanni, new subsp. Honolulu. Lobe of soft dorsal considerably less than 4 depth of body or else base of fin. ...2. 2. Maxillary not reaching below orbit...... C. g. gymnostethoides Bleeker. Batavia. Maxillary to nearly opposite front of pupil.. Scutes about 17. C. g. orthogrammus (Jordan and Gilbert). Clarion Id. 'Carangoides gymnostethoides evermanni, new subspecies The type, No. 7432, American Museum of Natural History, Honolulu saaiienk: August 25, 1920, B. W. Evermann, is 313 mm. long to base of caudal. Depth, 2.7 in this measure; head, 3.5. Eye, 5.2 in head; maxillary, 2.6; dorsal lobe, 1.4; anal lobe, 1.7; pectoral, 1.0 (right side) to 0.8 (left side). Maxillary to under front of orbit, not reaching pupil. Gill-rakers 21 on lower limb of first arch. Height of anterior lobe of soft dorsal 2.0 in base of that fin (not following curve of back), 1.8 in depth of body. Dorsal soft rays 30, anal 26. Chest before the ventrals narrowly naked. Color in alcohol dirty purplish gray, scarcely paler on the belly; fins, except pectoral which is paler, more or less dusky; dorsal, anal, and caudal lobes blackish, Fig 1. Carangoides gymnostethoides evermanni, new subspecies. Type (left) and cotype (right). ea AMERICAN MUSEUM NOVITATES No. 4 THE BONY STRUCTURE AND PHYLETIC RE- LATIONS OF SPHARODACTYLUS AND ALLIED LACERTILIAN GENERA, WITH THE DESCRIPTION OF A NEW GENUS By G. K. Nosre Issued March 10, 1921 By ORDER OF THE TRUSTEES OF THE AMERICAN MUSEUM OF NATURAL HISTORY New Yorx Crtry ee AMERICAN MUSEUM NOVITATES Number4 — March 10, 1921 59.81,15:14.71 THE BONY STRUCTURE AND PHYLETIC RELATIONS OF SPHHRODACTYLUS AND ALLIED LACERTILIAN GENERA, WITH THE DESCRIPTION OF A NEW GENUS By G. K. Nose Sphexrodactylus embraces a group of very small neotropical lizards including one species which is probably the smallest lizard in the world. It is not surprising that the osteology and closest affinities of the genus ' have remained until this time practically unknown. It is not my inten- tion to give here more than a bare outline of the more important struc- tural features of Sphexrodactylus or to discuss points which do not add definite evidence of phyletic relations. The various species of Sphxrodactylus have been exhaustively stu- died by Dr. Thomas Barbour. He has ready for press an elaborate monograph on the group. Dr. Barbour has aided me greatly in my study of the osteology of Spherodactylus and its allies. I am especially in- debted to him for specimens of Aristelliger, Phelsuma, Lathrogecko, Lepidoblepharis, and Gonatodes. These specimens were received in ex- change from the Museum of Comparative Zoology and are now incor- porated in the collections of the American Museum. COMPARISON OF STRUCTURAL CHARACTERS Spherodactylus possesses proccelous vertebre. It would seem that it could no longer be ranged with the gekkonids but should be grouped with Coleonyx in the Eublepharide. A detailed study of the skeleton of Spherodactylus has shown that it is not closely related to Coleonyx. A search for its nearest allies has involved the examination of the skeletons of many genera of gekkonids and eublepharids. Small differences have been observed in the material prepared. It has been difficult to déter- mine the relative value of these differences. Those characters which have seemed the most important are discussed below. Vertebree and Ribs The vertebre of Spherodactylus are proceelous, agreeing in detail with those of the eublepharids Lepidoblepharis and Lathrogecko, and but slightly different from those of Coleonyx. Two views of a single vertebra 1 2 AMERICAN MUSEUM NOVITATES [No. 4 of Sphzrodactylus are shown in Figure 1. It will be noted that the verte- bra is of a very simple type. It was surprising to find that a cartilaginous or fibro-cartilaginous band extends in Sphzxrodactylus from about the middle of the neural arch to the angular portion of the head of the rib. The cartilaginous nature of this band and its position relative to the neural arch and rib strongly suggest that it is the last vestige of the tubercle, a character- Fig. 1.—Vertebra (18th) of Sphzxrodactylus macrolepis Gither, lateral and pos- terior aspect. The former shows the cartilaginous tuberculum (stippled) of the rib. istic structure of the ribs of primitive reptiles but one believed to have been entirely lost in the Lacertilia. Williston (1914, p. 33) states: “By the loss of the tubercle in lizards, the head became truly single-headed, and attached solely to the body; and this condition is characteristic of the order Squamata.”’ A cartilaginous tubercle, if such it may be called, is found not only in Spherodactylus but also in Lathrogecko, Lepidoblepharis, and the neo- tropical species of Gonatodes. In Coleonyz, it is reduced, is more fibrous, and has a more anterior position than in Sphzxrodactylus. In all other lizards which I have examined, gekkonids, iguanids, teiids, xantusids, etc., this structure is represented by a ligament which is sometimes very slender and attached to the body of the vertebra near the articulation of the capitulum. In most iguanids and gekkonids, iti is flattened and some- times very difficult to distinguish. — 1921] SPHARODACTYLUS AND ALLIED GENERA 3 Skull Structure In addition to the proccelous form of the vertebra, one other char- -acter has been used to distinguish the Eublepharide from the Gek- konide. The parietals of the eublepharids are stated to be fused into a single element in contrast to the paired parietals of the latter group. — The parietals of Sphxrodactylus remain perfectly distinct throughout life. It would seem that this was a feature indicating a close affinity to the true gekkonids. An examination of the skulls of the various gek- konids and eublepharids at hand has convinced me that the fusion of the parietals into a single element cannot be considered diagnostic of the eublepharids. Most gekkonids possess paired parietals, but there are exceptions even within a genus. Thus, I find that while all the neotropi- cal species of Phyllodactylus at hand have paired parietals, there is but a single element in P. szamensis. The single parietal is not a constant feature of all eublepharids. It is single in Coleonyx variegatus (Baird) and C. elegans Gray, but double in Lathrogecko xanthostigma Noble. It was described as single in Lepidoblepharis festez Peracca but it is double in Lepidoblepharis barbouri Noble. Cope (1892) pointed out some differences between the skull of Coleonyx and that of Phyllodactylus. I have compared skulls of the same genera but have failed to find any marked differences. There is a re- - duced jugal in Coleonyx as well as in Phyllodactylus. Cope, however, did not consider those differences which he found of great importance, since in a later report (1898, p. 464) he states that the skeleton of the Eublepharide ‘‘is similar’ to that of the Gekkonidz ‘except in the proceelian vertebre and single parietal bone.” Hyoid and Branchial Arches Perhaps no one structure indicates the relationships of Sphero- dactylus better than its hyoid apparatus. As shown in Figure 2A, the arches are very complete. The second epibranchial is well developed and is attached at both ends, a very unusual condition. The distal end is adherent to the exoccipital at the base of the paroccipital process; the proximal end is loosely attached to the second basibranchial some distance from the end. The hyoid arch is a simple bent rod. It is at- tached distally to the paroccipital process. The hyoid apparatus of a number of gekkonids has been figured. I have examined specimens of Phyllodactylus (3 species), Thecadactylus, Hemidactylus, Aristelliger, Gehyra, Lygodactylus, Gekko, Tarentola (2 species), Pachydactylus, e : tts as ; WA-Lprdin q \ " % x + Giinther, in situ to show attachments. B.—Paragonatodes dickersoni (Schmi typical gekkonid hyoid apparatus with a specialized hyoid arch and with the : branchial arch wanting. pt B.br. 1 =basibranchial I; b.br. 2 =basibranchial II; e.br. 1 = hial I; e.br. II; ex.col. =extracolumella ; hy.ar.=hyoid arch; hy. cor. =body of hyoid; pr. lin. =1 1921] SPH#HRODACTYLUS AND ALLIED GENERA — A e. bf. + b.br. 2 hy. cor . -----hy. ar. i B Fig. 3—Hyoid Apparatus. A.—Coleonyx variegatus (Baird), lateral aspect of posterior part of cranium showing attachments of hyoid apparatus to skull. B.— Lathrogecko xanthostigma Noble, ventral view of the hyoid and branchial arches in their normal position. B.br. 1 =basibranchial I; b.br. 2=basibranchial II; e.br. 1 =epibranchial I; e.br. 2=epibranchial II; ex.col. =extracolumella; hy.ar.=hyoid arch; hy.cor.=body of hyoid; par.pr. =paroccipital process; pr.lin. =lingual process. : 6 AMERICAN MUSEUM NOVITATES [No. 4 Phelsuma, and Gonatodes (3 species). In none of these genera, except the neotropical species of Gonatodes, do the hyoid and branchial arches have a form and arrangement approaching the condition in Spherodactylus. The arches of all gekkonids, with the exception just noted, are more or less reduced, especially the second branchial arch. The second epi- branchial is generally present as a short and delicate cartilage lying free in the muscles and considerably removed from any. attachment to either skull or basibranchials. In a number of specimens I could find no indi- cation of such an epibranchial. It may not exist in the African gekkonid described by Schmidt as Gonatodes dickersoni (Figure 2B). Inthe South American Gonatodes atricucullaris Noble and G. annaigne Boulenger, the hyoid is very similar to that of Sphxrodactylus. The chief difference lies in the fact that the second epibranchial, although well developed, is loosely associated with the skull and is free from the basi- branchial. These two species agree with Sphxrodactylus in the long basi- branchials, extensive epibranchials, and simple hyoid arch. No gek- konids, except the neotropical species of Gonatodes, have been found to agree with Sphzxrodactylus in possessing a combination of these three features. It is remarkable that such a distinctive type of hyoid apparatus as that of the neotropical species of Gonatodes should be found in the euble- pharids Lathrogecko and Lepidoblepharis. The arches of these two genera are identical and differ from that of Gonatodes atricucullaris only in the slightly shorter first branchial arch and slightly larger arrow head to the second epibranchial. The distal end of the second epibranchial is not calcified in G. atricucullaris as it is in the several specimens of Lepidoble- pharis barbourt and a specimen of Lathrogecko xanthostigma (Figure 3B) which I have examined. In all three forms, the distal end of the second epibranchial is loosely attached to the paroccipital process and lies closely associated with the endolymphatic sac. It seems obvious that the pres- ence of such a well-developed hyoid in Sphexrodactylus, Lepidoblepharis, Lathrogecko, and the neotropical forms of Gonatodes indicates common ancestry. The most primitive type of lacertilian hyoid apparatus is that found — in Coleonyx. This was not realized until very recently (Fiirbringer, 1919). The figure of Cope (1892, Pl. 1, fig. 8) of the hyoid apparatus of C. variegatus is very incorrect. C. variegatus and C. elegans have similar hyoid and branchial arches. The second epibranchial is continuous with the second basibranchial and there is no suture or break between the two parts. The distal portion of this second branchial arch is attached very 1921) SPH#RODACTYLUS AND ALLIED GENERA 7 loosely to the skull by a ligament. The cartilaginous portions of both hyoid and branchial arches have a characteristic form (Figure 3A). This very primitive type of hyoid apparatus found in Coleonyx seems to indicate that the genus has no close affinity to Sphzxrodactylus. If primary importance were arbitrarily laid on the form of the hyoid and branchial arches in determining relationships, it would follow that Sphexrodactylus is more closely related to the gekkonid Gonatodes than to the eublepharid Coleonyx. Such is probably the correct view. Pectoral Girdle Sphexrodactylus possesses a typical gekkonid shoulder girdle, with subcruciform interclavicle and expanded, perforated clavicle. Its pec- toral girdle differs radically from that of Coleonyx in having four instead of three ribs attached to the sternum. The other two genera of neo- tropical eublepharids agree with Sphxrodactylus as regards the sternal ribs but differ in the form of the clavicle. In neither Lepidoblepharis nor Lathrogecko is the clavicle perforated. Altogether too much emphasis has been laid on form of the clavicle as defining the larger groups of Lacertilia. It is now well known that a . number of iguanids possess expanded and perforated clavicles. The ex- panded, perforated clavicle cannot be considered a diagnostic feature of all gekkonids. The clavicle of the neotropical species of Gonatodes (Figure 4A) is not more expanded than many so-called cylindrical clavicles. If one considers the slightly dilated clavicle of the neotropical species _ of Gonatodes (Figure 4A) as the primitive type, one can readily derive from that the conditions found in the neotropical eublepharids. The clavicle of Lathrogecko is slightly more dilated than that of Gonatodes. In Lepidoblepharis (Figure 5A) it is still more expanded. In Sphzro- dactylus (Figure 5B), the expanded portion has become fenestrated. The series exhibited by Gonatodes, Lathrogecko, Lepidoblepharis, and Sphzro- dactylus illustrates beautifully how the clavicle might have been gradu- ally expanded and in the extreme stage thinned out until a foramen was formed. There is much reason to believe that we have in this series of genera a natural group and that the expanded, perforated clavicle has been evolved from the cylindrical one. It may be well to mention at this point that the subcruciform inter- clavicle is not always present in the gekkonids. I have found that the African Gonatodes dickersoni and the Madagascarian Phelsuma laticauda Fig. 4.—Pectoral Girdles, ventral aspect. A —Gonatodes a B —Paragonatodes dickersoni (Schmidt). : = 1. =clavicle; f.sp.c. : 5 x Faceted eon) ¢ rick sp.c vipa iconata taunt in.cl. = 1921] SPHARODACTYLUS AND ALLIED GENERA 9 Fig. 5.—Pectoral Girdles, ventral aspect. A.—Lepidoblepharis barbouri Noble. B.—Spherodactylus macrolepis Giinther. Cor. =coracoid; cl. =clavicle; f.sp.c.=supracoracoid foramen; in.cl.=interclavicle; sc.= scapula; st.=sternum; st.r.=sternal rib. Fig. 6.—Pelves, ventral view. AParagonatodes dickerson ( Sphzxrodactylus macrolepis Giinther. pee As Ep.p. =epipubis; hyp. =hypo-ischium; pec.p =pectineal process. 1921] SPHARODACTYLUS AND ALLIED GENERA 11 have the transverse arms of the interclavicle reduced or wanting. The former species differs greatly from the neotropical species of Gonatodes in having only three sternal ribs (Figure 4B). Pelvis and Cloacal Bones Gonatodes dickersoni differs from the neotropical species of Gona- todes in the form of its pelvis (Figure 6A) and the presence of cloacal bones in the male. The pubis has a very small pectineal process in G. dickersoni and there is a well-developed hypo-ischium and epipubis. The pubis of the neotropical species of Gonatodes agrees with that of Lathrogecko, Lepidoblepharis, and Spherodactylus in the large pectineal process directed ventrally. The hypo-ischium may be very rudimen- Fig. 7.—Cloacal Bones, showing relation to the cloacal slit. A.—Coleonyzx varie- gatus (Baird). B.—Paragonatodes dickersoni (Schmidt). tary or wanting in Spherodactylus (Figure 6B) and is wanting in the other genera. In Coleonyzx there is no hypo-ischium, but there are two pairs of large cloacal bones (Figure 7A), one pair projecting through the skin. Very little reference appears in the literature in regard to the cloacal bones. Lying free near the hemipenes and below the skin, they have been often overlooked. The hypo-ischium has been often called an os cloace. The hypo-ischium and cloacal bones should not be confused. They are neither homologous nor analogous. I find the greatest development of cloacal bones in Pachydactylus maculatus where, in addition to a broad fenestrated median bone lying transversely across the anterior lip of the cloaca, there is a pair of irregularly shaped bones posterior to either corner of the cloacal slit. The hypo-ischium in this species is very long. In the several species of Phyllodactylus which I have examined, the males are provided with cloacal bones very similar in form to those of Gonatodes dickersoni. — 12 AMERICAN MUSEUM NOVITATES (No. 4 External Characters The obvious external similarity of Gonatedes and Lepidoblepharis has been pointed out by Peracca (1897). Of the three features empha- sized by Ruthven (1916) in distinguishing Lathrogecko from Lepidobie- pharis, only one, that of the form of the digits, can be considered of generic importance. It is apparent from a study of both internal and external structure that Lathrogecko is closely allied to Lepidoblepharis. Spherodactylus agrees with Gonatodes, Lathrogecko, and Lepidoblepharis in the slender form of the body, the narrowness of the head, the arrange- ment of labials, rostral, and nostril, and the shape of the pupil. Some species of Sphexrodactylus agree with some species of Gonatodes in the pronounced sexual dimorphism and general color pattern. Still, it has been very difficult to pick out any definite external characters which demonstrate a closer relation between Spherodactylus and the above genera than between Spherodactylus and any other gekkonoid groups. Cope (1898) seemed prepared to believe that Sphxrodactylus was closely allied to Phyllodactylus. Most other reviewers have considered that the form of the digit tips in Spherodactylus warranted the placing of that genus in an isolated position in any scheme of phylogeny adopted. A careful examination of the digits of Sphxrodactylus will show that their terminal dilations are composed of scales having the same mutual relations as those which make up the claw sheath in Lepidoblepharis. It would seem that an asymmetrical enlargement of one side of the claw sheath of Lepidoblepharis would give exactly the condition found in Spherodactylus. In S. macrolepis and apparently throughout the genus, this enlargement has been the outer scales of the claw sheath in the pes and the outer on all the digits of the manus except the fifth, where it has been the inner side of the original sheath which has become enlarged to form the disk. The homology of these scales becomes much more obvious if the claw sheaths of Gonatodes and Lathrogecko are compared at the same time. It seems fairly certain when these sheaths are arranged in a series that we have before us an actual phylogenetic sequence. The claw sheath of Lathrogecko (Figure 8B) may have been derived directly from that of the neotropical species of Gonatodes by an enlargement of the terminal scales of the digits. The claw sheath of Lepidoblepharis (Figure 8C) could have been developed from the sheath of Lathrogecko by the dropping out of the second median scale. Finally, the disks of Sphzro- dactylus are understandable only if we assume that they were formed from the Lepidoblepharis claw sheath by the asymmetrical enlargement of the 1921] SPHHRODACTYLUS AND ALLIED GENERA 13 Fig. 8.—Claw Sheaths illustrating four stages in a single line of specialization. Homologous scales bear the same letters; original dorsal scales, a—d; laterals, 1 and m. A.—Gonatodes atricucullaris Noble. B.—Lathrogecko xanthostigma Noble. C.—Lepidoblepharis barbouri Noble. D.—Sphzrodactylus macrolepis Giinther. 14 AMERICAN MUSEUM NOVITATES [No. 4 scales on one side of the sheath. The steps assumed in the change from the Gonatodes to the Lathrogecko to the Lepidoblepharis types of sheath are not great; the step from the Lepidoblepharis to the Sphxrodactylus type is less clear but no less admissible. PHYLOGENETIC RELATIONS It follows from the above résumé of distinctive characters that the eublepharids, .Sphxrodactylus, Lepidoblepharis, and Lathrogecko, are closely related to each other and to the neotropical species of the gek- konid Gonatodes. At least one of the Old World species of Gonatodes, and probably all, has no close affinity to the neotropical forms. Since the genus Gonatodes was based on a neotropical species, a new name will have to be proposed for Old World forms, or for at least the one species which we have studied in detail. It is probable that this new genus will embrace all three African species, less probable that it will include the East Indian forms which have until now been referred to Gonatodes. PARAGONATODES, new genus (Gekkonide) Typr.—Gonatodes dickersoni Schmidt. (Type locality, Medje, Belgian Congo.) Dracnosis.—Digits slender, clawed; the distal portion of the digits slightly com- pressed and forming an angle with the claw; these distal portions covered beneath with a single series of scales distally, and with a double series of much smaller ones proximally (see Schmidt, 1919, fig. 6); body slender, with granules and tubercles above, with small scales below; tail cylindrical; pupil circular; eyelid distinct around eye. Hyoid apparatus reduced; no second basibranchiais; no second epibranchials (Figure 2B); interclavicle dagger-form, no transverse arms; clavicle dilated but not fenestrated (Figure 4B); only three sternal ribs; pectineal process of pubis rudimen- tary; a well-developed hypo-ischium; male with a single pair of bow-shaped cloacal bones (Figure 7B); ligamentous tubercle of the ribs much reduced and proximated to. the capitulum. It seems extremely probable that Sphexrodactylus, Lepidoblepharis, and Lathrogecko, with their proccelous vertebre, four sternal ribs, cartilaginous tuberculum, distinctive hyoid, pelvis and cloacal regions, form a natural group of genera. These genera show closer affinity to Gonatodes than to any other gekkonid. Evidence has been brought forth to show that we have in this group a natural series commencing with Gonatodes, and leading through Lath- rogecko and Lepidoblepharis to Spherodactylus. It is believed that this series represents an actual morphogenetic sequence. The more important changes which occurred in this series may be listed. (1) The vertebree Se a PRT D ee 1921] SPHHERODACTYLUS AND ALLIED CENERA 15 changed from amphiccelous to proccelous, and most of the intercentra were lost. (2) The second epibranchial lost its characteristic -arrow- shaped head and became attached to the exoccipital near the base of the paroccipital process. The proximal end of the second epibranchial migrated anteriorly and became loosely attached to the second basi- branchial. (3) The clavicles evolved from narrow but flattened rods to broadly expanded sheets, and finally thinned out in their proximal portions to form median fenestra. (4) The terminal scales of the digits became elongated to form six-scaled claw sheaths. “The posterior dorsal of these six scales dropped out to form five-scaled sheaths. Fin- ally, there was an asymmetrical enlargement of one side of the sheaths to form disks. p ; It is important to emphasize that this series of steps has only been assumed after a study of all the genera of gekkonids and eublepharids available to me; that Gonatodes, Lathrogecko, Lepidoblepharis, and Sphzrodactylus have more in common with each other than can be found between Spherodactylus and Coleonyx, or Gonatodes and any of the ten other genera of gekkonids at hand. In other words, it seems extremely likely that, among other things, the proccelous vertebra have been de- veloped in this series quite independently of similar changes in any other series. It follows that in all probability the Eublepharide had a poly- phyletic origin and, instead of being a very ancient group as hitherto believed, they may be a very recent assemblage, even if a conservative one. It has been suggested that the gekkonids are degenerate forms, their amphiccelous vertebre secondary structures. There is obviously nothing primitive in the highly reduced skull of the gekkonids. Coleonyx with its very primitive hyoid possesses proccelous vertebre. Xantusids with proccelus vertebre also have primitive hyoids, and I have found that Xantusia vigilis retains the intermedium in the carpus as further evi- ‘dence of its ancestral position among primitive Lacertilia. Why, then, should we not reverse our series and evolve Gonatodes from Sphzero- dactylus or at least Lathrogecko? This would necessitate developing intercentra again, evolving cylindrical from expanded clavicles, and changing from specialized to primitive claw sheaths. Altogether too little is known about the osteology of the Lacertilia to be entirely certain. about the directior in which evolution has progressed. The view I have © outlined above seems at the present time the most probable. 16 AMERICAN MUSEUM NOVITATES [No. 4 LITERATURE CITED Corr, E. D. 1892. ‘The Osteology of the Lacertilia.’ Proc. Amer. Philos. Soc., XXX, p. 185-221, Pls. n—v1. 1898 (1900). ‘The Crocodilians, Lizards, and Snakes of North America.’ Annual Report U.S. Nat. Mus. for 1898. FURBRINGER, Max. 1919. ‘Uber das Zungenbein der Reptilien.’ Bijdr. tot de Dierk., Amsterdam, pp. 195-212. Peracca, M. G. 1897. ‘Viaggio del Dr. Enrico Festa nell’ Ecuador e regioni vicine; Rettili.’ Bol. Mus. Zool. Anat. Comp., Torino, XII, No. 300. Rutuven, A.G. 1916. ‘A New Genus and Species of Lizard from Colombia, with Remarks on the Genus Pseudogonatodes.’ Occ. Papers Mus. Zool., Univ. Mich., No. 21. Scumipt, K. P. 1919. ‘Contributions to the Herpetology of the Belgian Congo ' ‘Based on-the Collection of the American Museum Congo Expedition, 1905- 1915. Part I. Turtles, Crocodiles, Lizards, and Chameleons.’ Bull. Amer. Mus. Nat. Hist., XX XIX, pp. 385-624, Pls. vi—xxx11. . Wiuston, S. W. 1914. ‘Water Reptiles of the Past and Present.’ Chicago . (University of Chicago Press). AMERICAN MUSEUM NOVITATES No. 5 GEOGRAPHIC AVERAGE, A SUGGESTED METHOD FOR THE STUDY OF DISTRIBUTION By Frank E. Lurz 7 AMERICAN “A \ MUSEUM \4 NATURAL Issued March 14, 1921 By OrpER OF THE TRUSTEES OF | THE AMERICAN MUSEUM OF NATURAL HISTORY New Yorx City AMERICAN MUSEUM NOVITATES Number 5 Mareh 14, 1921 59.19 GEOGRAPHIC AVERAGE, A SUGGESTED METHOD FOR THE STUDY OF DISTRIBUTION By Frank E. Lutz Students of the distribution of animals and plants have divided North America, especially United States, into sections, within each of which the fauna or flora, or both, is believed to be more or less homogene- ous and more or less distinct from other, corresponding sections. I must confess that these sections seem to me less distinct than they once did. It is a characteristic of human psychology that we classify and make categories where there are no definite classes or categories. We speak of north, east, south, and west. When we go more deeply into the subject we speak of northeast, southeast, and so on. At sea, however, we box the compass by such gradual steps that we have a continuous circle with no separate divisions. This does not mean that “north,” ‘east,’ and so on are not convenient and useful conventions but, as a matter of fact, they are only very general terms. In the same way, there is no definite “‘arctic-alpine,”’ ‘“austral,’’ and so on, and a definiteness can only be maintained for these zones by a special selection of ‘indicator species” or something of that sort. If this procedure be accepted, almost any sort of system of zones may be devised by selecting appropriate ‘“‘indicators.”’ On the other hand, some system is convenient and useful. The care- fully investigated system of Merriam, based on the fundamental studies of Allen, has been widely accepted. Can we take the proposed biotic areas and get some concrete expression for them that will be based on the biota as a whole? This expression should, if possible, be something that will help us to say with a fair degree of assurance that a given limited area or a given species belongs in this section or that. After trying a number of different methods of getting such an expression, it seems to me that what I am calling the ‘‘geographic average”’ gives the best promise. Partly because plants are a large determining factor in the distribu- tion of animals and partly because it was convenient to use the data given in Britton and Brown’s ‘Illustrated Flora of the Northern States and Canada,’ the illustrations given here are based on data secured from the second volume of that work, Portulacacez to Papilionacez, inclusive. There was no reason for taking this portion of the flora rather than some 1 2 AMERICAN MUSEUM NOVITATES [No. 5 other; Imerely wanted arandom sample. The only species intentionally omitted from this lot were those that were said to have been introduced by man and a few in which the distributional range was not clear to me. The limits of the range of a species outline a polygon. If we are given the latitude and longitude of these limits we can average them and get the latitude and longitude of a point which is approximately in the center of the range of a species.!_ This would be the geographic average of that species, and it would indicate rather clearly whether the species tends to be northern or southern, eastern or western. Having calculated the geographic averages of each of an aggregate of species (those in a — particular sphagnum bog, or at a given altitude on a certain mountain, or in a given political division, or what not) one could average these species averages and get a geographic average for the aggregate under consideration. In Britton and Brown’s ‘Flora’ the range of each species is stated by giving the names of the states or other political divisions to which the species extends. I took the approximate center of each of these political divisions as the limit of the range in that direction and calculated — geographic averages for each species. On the basis of the 77 of these that occur in Labrador, I found that the geographic averagein North America, including Greenland,’ of the Labrador flora is at about 51° N., 89° W., with an average range in latitude of about 20° and in longitude of about 79°. Ifthe complete flora were considered, this average might be some- what different but, as a first approximation, it indicates that, in North America, the flora of Labrador is one that centers a little south of Hudson Bay and, on the average, extends southward to about the latitude (41° N.) of southern New York or northern Colorado and westward: to about 130° W., say British Columbia. Britton and Brown’s ‘Flora’ concerns itself only with northeastern North America. Taking up several areas along the northern Atlantic slope and basing the averages on species of plants from Portulacacez to — Papilionacee, inclusive, we get the following. : 1If the range were crescentic in shape, the geographic average might be a point between the horns of the crescent, where the species does not, in fact, exist. 2T did not use limits in Asia or Europe in calculating these averages; they were not given definitely and did not seem important for what I had in mind. Of the 91 North American species which occur in Palearctica and for which I have calculated the geographic averages, 12 that occur in Asia but not in Europe have a geographic average of 48° N. (Range, 22°), 95° W. (Range, 79°), and 14 that occur in Europe but not Asia have a geographic average of 49° N. (Range, 13°), 78° W. (Range, 42°). The numbers on which these averages are based are too small to be entirely significant and they are only on species of northeastern North America. A study based on fuller data is being prepared. i lt tet, i eB =e 1921] GEOGRAPHIC AVERAGE 3 Geographic Average Average Range Latitude Longitude Latitude Longitude Labrador §1° &9° 20° 79° Newfoundland 48 87 20 70 Nova Scotia and New Brunswick 45 87 19 56 New England 42 85 17 . 44 Md., Va., and Del. 39 85 14 26 It is clear, as would be expected, that the average latitude decreases as we take more southern areas but it might not have been so certainly anticipated that the longitude would also decrease, that is, that the geographic average of the areas would steadily move eastward although the areas in question. are more and more western. This seems to be bound up with the further fact that the average longitudinal range! decreases steadily and markedly as we pass from north to south. The similar de- crease in the average latitudinal range, while not so great, is more sur- prising, as one might have expected that Labrador species, being re- stricted by extreme arctic conditions from spreading in a northward direction would have a less average north-and-south range than those in Virginia, for example, which have a great distance, both north and south, to which they can spread. It is true that a study of the average ranges of 386 species of plants native to northeastern North America shows that there is a close posi- tive correlation between the extents of latitudinal and longitudinal ranges but I do not think that this, coupled with the relatively wide longitudinal range of northern species,” is a full explanation of the wide latitudinal range of such species. Perhaps the explanation is that a species which is able to survive difficult northern conditions finds it easy to live in a relatively wide range of other conditions, while the same is not true of the more southern species. Labrador is in what was called the Barren-Ground Fauna of the Arctic Realm by Allen and the Arctic Zone of the Boreal Region by Merriam. There is little or nothing from farther north which’ might 1Longitudinal range as measured by degrees of longitude. It should be remembered that a degree of longitude is not as many miles in higher latitudes asin more equatorial ones. However, this difference is not great enough between Virginia and Labrador to negative the indication that, among these plants of northeastern North America, the more northern ones tend to have a wider east-and-west distribution than the more southern ones. 2Not only is the average longitudinal range in North America of the Labrador species considered here 79°, as compared with 26° for those in Virginia, but other data on the Labrador flora as a whole show that 72% of the Labrador flora occur also in Palearctica. 4 AMERICAN MUSEUM NOVITATES [No. 5 have extended out of its “regular” range into Labrador, but it is con- ceivable that really southern things got into unusually favorable,’ _ ecologically speaking, situations in Labrador and were able to survive there, at least long enough to be collected and recorded by the botanists. These latter plants, classed as Hudsonian, would, for the most part, have a range along eastern North America southward from Labrador. The — material from Labrador can be analyzed as follows. Average Average } , Latitude Longitude (A) 40 species that do not occur far directly southward 54° 90° (B) 37 species that occur directly southward 49 85 The number of species is small for the purpose of computing aver- ages and the grouping is admittedly rough but, as an illustration of method and until we have something better, we can take the geographic average for A as that of the Arctic! and the one for B as that of the Hudsonian. : Newfoundland is considered by Allen to be partly in the Cold Temperate Subregion and partly in the Arctic Realm. Merriam divides it between the Arctic, Hudsonian, and Canadian Zones of the Boreal Region.2. The present material may be analyzed as follows. Average Average Latitude Longitude (C) 57 species which range north of Newfoundland 50° gg° (D) 27 species which occur in Newfoundland but not in Labrador 44 83 Of these, C is doubtless a mixture of Arctic and Hudsonian; and D may be largely Canadian. Nova Scotia and New Brunswick are entirely in the Cold Temperate — Subregion (Canadian Fauna) of Allen but Merriam recognizes Alle- ghanian Zone of the Austral as well as Canadian Zone of his Boreal Region. The geographic averages of 152 plants from there may be analyzed as follows. ‘ 1The geographic average in North America (including Greenland) of 20 Labrador plants that occur in Greenland is 54° N., 81° W. *His map, 1912, Canadian Entomologist, XLIV, opposite p. 128, shows some yellow and blue, indicating Upper Austral and Transition. I do not know whether he meant this or whether it is due to faulty registration of the color-plates in printing. aie - 1921] GEOGRAPHIC AVERAGE | 5 Average Average Latitude Longitude (E) 56 species that occur also in Labrador 50" =*° 90° (F) 26 species that occur also in Newfoundland but not in Labrador 44 85 (G) 70 species that occur in neither Labrador nor Newfound- land 42 86 Of these, E probably represents a mixture of Arctic and Hudsonian species that extend south; F may be largely Cold Temperate (Canadian) species; and G may be Alleghanian. The geographic average for 207 New England plants is stated above to be 42° N., 85° W. Analyzing it in the same way as was done above for the material from Nova Scotia and New Brunswick, we get the following results. Average Average Latitude Longitude (H) 40 species that occur also in Labrador 48° gg° (I) 24 species that occur also in Newfoundland but not in Labrador 44 ~ 84 (J) 69 species that occur also in Nova Scotia or New Brunswick but not northeast of there 42 86 (K) 74 species that do not occur northeast of New England 39 84 Four classes of plants may be considered to be present here: prob- ably largely Hudsonian with a geographic average of 48° N., 89° W.; the Canadian with a geographic average of 44° N., 84° W.; the Alleghanian with a geographic average of 42° N.,86°W.; and an element of warmer- region plants with a geographic average of 39° N., 84° W. This last element is probably to be identified with Allen’s Carolinian Fauna of the Humid Province of the Appalachian Subprovince of the Warm Temper- ate Subregion, or with the Carolinian of Merriam’s Upper Austral. The following is an analysis of the geographic averages of 282 plants that occur in Maryland, Virginia, or Delaware. Average Average Latitude Longitude (L) 12 species that occur also in Labrador 46° g9° (M) 11 species that occur also in Newfoundland but not in Labrador A2 84 (N) 57 species that occur also in Nova Scotia or New Bruns- wick but not northeast of there 41 86 (O) 150 species that do not occur northeast of New England 38 84 (P) 52 species that do not occur northeast of Md. 34 84 6 AMERICAN MUSEUM NOVITATES [No.5 - What is apparently a new element (P) here might be considered to be the coming in of plants that would be classed in Allen’s Austroriparian * Subprovince (Louisianian Fauna) of the Humid Province of the Warm Temperate Subregion or of Merriam’s Austroriparian Fauna of the Lower Austral Zone. As they are only the more northern species, the latitudinal average for this distributional section as a whole is doubtless less. O may be largely Carolinian; N, Alleghanian; L and M each have too few species to make the averages even approximately trustworthy. Collecting the data, we get the following provisional suggestions as to the latitudinal averages for various regions. Arctic More than 52° N. Hudsonian 48 or 49° Nt Canadian 44 or 45° NA ‘Alleghanian 41 or 42° N. Carolinian 38 or 39° N. | Louisianian or Austroriparian Less than 34° N. The tabulation just given indicates, at first sight, that these regions are definite, concrete entities, but, when we bear in mind that they were derived by first grouping the data into disconnected lots, we realize that another grouping might fill in the gaps. Furthermore, they are averages in which we lose sight of the extremes of each group. On the other hand, as stated above, categories, even if only conventional, are often useful aids to thinking and it might be well if we had concrete expressions for our categories—something more definite and tangible than mere names. I believe that the “geographic average”’ is such a concrete expression. I do not mean to intimate that the figures just obtained should be con- sidered final, even for the regions that have been studied here. Although they are based on considerable material, this material includes only a small part of the plants of these regions and it does not include any of the animals. Also, it includes only relatively small areas of each section (“‘zone”’ or whatever we wish to callit). Furthermore, altitude and other ecological conditions have been disregarded. I have been unable to devise, as yet, satisfactory corrections for altitude. Doubtless it could be done in the following way. We could study a limited area containing a wide altitudinal range, for example, 11t will be noted that these are south of the principal boundaries of the “zones”’ in question, as given in Merriam. The finger-like southern extensions of the zones along mountain ranges is one of the causes and the discrepancy will be less when the corrections for altitude have been worked out. However, itisa matter of little moment, since the geographic average need not necessarily fall within the area act covered by the species or the aggregate of species under consideration. 1921] GEOGRAPHIC AVERAGE : 7 North Carolina, Virginia, or the central part of Colorado or of California. We could then calculate the geographic averages for various altitudes, keeping other ecological conditions, such as soil, exposure, etc., as nearly uniform as possible. We might get, in this way, a fairly accurate, con- crete expression for the known fact that going to higher altitudes is, faunistically and floristically, analogous to going to higher latitudes and we might also get a factor for modifying our data so that, instead of working with the exact latitude of a place on a mountain, we could work with a modified figure, the modification being greater as the altitude is greater. In much the same way, the geographic average lends itself to a study of other ecological conditions. Having obtained fairly approximate expressions in average latitude and longitude for each of the zonal names that we care to accept as a matter of convenience, we can then say rather definitely whether a certain animal or plant belongs in a given zone or not. This can be done by a comparison of its geographic average with the geographic averages of the zones. In the same way we can say whether a given small area is in one zone or another—or between two of them. One of our common bumble bees, Bombus americanorum, hasa geographic average of approxi- mately 38° N., 92° W. Its latitudinal average would put it in Carolinian, on the basis of the tabulation given above. Its longitudinal average is greater (farther west) than the longitudinal average of the areas studied here having the same latitudinal average. The matter of longitudinal average is important in connection with the change in humidity that occurs at ‘about 100° W. in the United States. Allen made this the basis upon which he divided the Warm Temperate Subregion into Provinces (Humid and Arid). Merriam laid more stress on temperature than on humidity and considered the latter as causing merely a subdivision of certain of the zones that were based on tempera- ture. I, personally, incline more to the former idea than to the latter, especially where the latitudinal average is less than about 40° N., but I do not yet have enough data to discuss the question further. The present paper is intended to outline a method, rather than to state results obtained by the use of the method. This method of geo- graphic averages is suggested as a better tool for the study of distribution than, for example, indicator species or percentages of species from one area occurring in another. Incidentally, it has been indicated that the named “regions,” “zones,” etc. are far from the definite entities that they often seem to be. At the same time, what are believed to be first approximations to concrete expressions have been obtained for some of them. AMERICAN MUSEUM NOVITATES No. 6 DESCRIPTION OF A NEW SPECIES OF SEROW FROM YUN-NAN PROVINCE, CHINA By Roy CHAPMAN ANDREWS. fri Ly THE Ey Ry ISTORY SCL Issued March 24, 1921 By ORpDER OF THE TRUSTEES OF THE AMERICAN MUSEUM OF NATURAL HISTORY New Yor« City dae way V4 *® =~ a Tr ee AMERICAN MUSEUM NOVITATES Number 6 March 24, 1921 59.9, 735C (51.3) DESCRIPTION OF A NEW SPECIES OF SEROW FROM YUN- NAN PROVINCE, CHINA By Roy CHapMAN ANDREWS Although it is not the purpose to publish extensively upon the collec- tions obtained by the Asiatic Expeditions of the American Museum until the field work has been completed, it is desirable to describe such new species as may come to light from time to time in the preliminary ex- amination of certain groups. The splendid serow which I shot near Teng-yiieh, China, not far from the Burma frontier, is the first animal to be described from the Asiatic collections, and I take pleasure in proposing for it the name Capricornis osborni, in honor of the distinguished President of The American Museum of Natural History, Professor Henry Fairfield Osborn. Capricornis osborni, new species Type No. 43042, 2 juv., Hui-yao (20 miles from Teng-yiieh), Yiin-nan Province, China, May 9, 1917; Roy Chapman Andrews. Forehead, cheeks, neck, breast, and body coal black. The white basal parts of the hairs show through to a certain extent but the general effect is jet black. A nar- row margin of white 6 mm. wide on upper lips from middle of snout to corner of mouth. Above this white band but-below and behind the nostril on each side, is a triangular tawny patch. The lower lip is margined with white which occupies all except the central half of the chin and extends behind the corner of the mouth in along, gradually narrowing streak; this almost reaches the white throat-patch which is about 40 mm. in width. The proximal part of each ear in front is strongly tinged with tawny but on the back this is less pronounced and the ear is largely black. The short mane is intense black like the body, stiff, erect and crest-like; the hairs are about 120 mm. in length. From the mane to the tail, the hair of the mid-dorsal line forms a well-defined ridge. The tail is black in the center with an admixture of tawny hairs; the tip is all black. Inside of fore legs to hoofs, tawny; externally, black to the knees; front of ‘cannon bones”’ black except at. the knees where the black is indistinct and suffused with tawny. Just above the hoofs, the blackish area is thinly sprinkled with light buff and posteriorly, between the dew claws and the hoof, it is all light buff. Buttocks tinged with tawny. Thighs almost: to hocks, black with slight admixture of tawny. Inner side of hind legs to hoofs tawny. From hocks to opposite dew claws, anteriorly, the legs are tawny but with a suggestion of blackish, due to the hairs which are black on the basal half and tawny on the tips. From opposite the dew claws to the hoofs the black is pronounced and thinly interspersed with buff-tipped hairs. The area between the dew claws and the hoofs, posteriorly, is all buff. 1 2 AMERICAN MUSEUM NOVITATES [No. 6 There is no underfur present on any part of the body. Skull badly broken. Measurements of skull: condylo-basal length, 257 mm.; least orbital width, 68; width of palate between first premolars, 37; length of horn on curve, 117; circumference of horn at base, 95. External measurements of type: head and body, 1350; tail, 180; hind foot, 390; ear, 175; height at shoulder, 950. Capricornis osborni is undoubtedly allied to our specimens from Li- chiang, Yiin-nan Province, which I have identified as C. milne-edwardsi. Its chief distinguishing characters are its coal-black body and head, its short black mane and the greater amount of black on the lower part of the legs. Our four specimens of C. milne-edwardsi all have brownish- black bodies and heads, long whitish manes, and little or no black upon the lower legs. In the very heavy mat-like gray mane, my two speci- 'mens of C. argyrochztes from Fukien Province, China, differ strikingly from osborni, although in the amount and disposition of the black on the lower legs the two somewhat resemble each other. C. swettenhami of the Malay Peninsula is distinguished from osborni by the black legs and ee mane, which is a mixture of whitish, black and reddish hairs. In discussing C. milne-edwardsi Mr. R. I. Pocock! has ‘enna ‘“A closely allied form apparently resembling typical milne-edwardsa in color except that the fronts of the cannon bones appear to be black has been recorded by Mr. H. Shaw Dunn from Kyonklongyi and other localities in the North Shan States of Upper Burma where it lives mostly in evergreen forests at altitudes of from 4,500 to 6,000 feet (Field, Jan. 9, 1909.).”’ I have not been able to discover Mr. Dunn’s communication in the ‘Field’ but I have no doubt that the race I am now describing is the one to which he refers. The serow which Lieut. R. C. Beavan? described as inhabiting the vicinity of Moul-mein, Burma, and which Mr. Pocock referred provi- sionally to milne-edwardsi may be this new form. While the affinities of osborni are toward milne-edwardsi, it is interesting as showing an ap- proach toward swettenhami of the Malay Peninsula in the considerable amount of black on the legs and the short black mane. Near Genkang, Yiin-nan Province, we purchased from a native a flat serow skin which lacks the head and lower legs. This specimen was said to have come from the mountains of Keng-ma about 200 miles southeast of Teng-yiieh and not far from the Burma frontier. It is brownish black ‘The Serows, Gorals and Takins of British India and the Straits Settlements.’ By R. I. Pocock, Part II. Journal, Bombay Natural History Society, XXII, pp. 307-308. 21866, Proc. Zool. Soc. London, p. 4. 1921] A NEW SEROW FROM CHINA 3 in general color, has a short crest-like, brownish-black mane, similar in character to that of osbornz, and what remains of the skin shows that both the fore and hind legs were whitish or light buff below the knees and hocks. This specimen may possibly represent the male of C. osborni, for the differences are somewhat similar to those between the male and female of our C. argyrocheztes from Fukien. I shot C. osborni near the village of Hui-yao while hunting monkeys on the precipitous bank of the river. The cliff was almost perpendicular and was covered with a tangled jungle growth. Now and then the rock wall would become less precipitous and the thick cover give place to an open grassy slope. It was when I was about to cross such an opening that the serow dashed out of the bushes where it had evidently been feeding. I fired just before it disappeared over the rim of the gorge and it sank in its tracks, gave a convulsive twist, and plunged into the canyon. It was recovered with considerable difficulty. Although the natives knew that serows lived in this part of the gorge, few of them had ever seen one and it was an object of great curiosity in the village. There.is little change in the country between Hui-yao and the Burma frontier and no reason ‘why C. osborni should not have an unrestricted range into Burma. AMERICAN MUSEUM NOVITATES No. 7 DESCRIPTION OF FOUR NEW BIRDS FROM THE BELGIAN CONGO | By James P. CHAPIN Gu 30 Aug ¥ ine ON Os e 61 oo Mey q RT wy , a Issued April 4, 1921 By ORDER OF THE TRUSTEES OF THE AMERICAN MUSEUM OF NATURAL HISTORY New Yorx City / AMERICAN MUSEUM NOVITATES Number 7 April 4, 1921 59.82 (67.5) DESCRIPTIONS OF FOUR NEW BIRDS FROM THE BELGIAN CONGO! By James P. CHAPIN This is the fifth preliminary paper on the ornithological results of the Congo Expedition of the American Museum and consists of descrip- tions of four new forms from the Ituri and Upper Uele Districts of the Belgian Congo. The special color-names employed are taken from Ridgway’s ‘Color Standards and Color Nomenclature’ (1912). I am obliged to the Carnegie Museum and the courtesy of Mr. W. E. Clyde Todd for the loan of specimens for comparison; and permission has been kindly granted me to examine other material in the U. S. National Museum and the Philadelphia Academy of Natural Sciences. Astur toussenellii canescens, new subspecies SupsPeciric CHARACIERS.—Differs from A. t. toussenellii (Verreaux) in being slightly larger and much paler, especially below. Easily distinguished by its pale gray, instead of rufous, thighs. Type.—No. 157743, A. M. N. H., @ ad., Medje, in northern Ituri District, Belgian Congo, June 28, 1910. ApuLT FEMALE (type).—Crown deep gull-gray, shading gradually to slate- gray on the back; slate-color on the wing-coverts, and dark neutral gray on the outer webs of primaries and secondaries. Primaries beneath light gray with narrow bars of brownish black, on the secondaries these bars disappear, and the color becomes gray fading to white on the basal half of inner webs. Under wing-coverts white, irregularly marked with gray. Rump like the back, upper tail-coverts slightly darker, and median rectrices slate-black, with three large spots of white on the inner web, and a white tip. The outermost tail-feather, about 30 mm. shorter than the middle, is blackish brown, almost uniform, but whitish at the very base on inner web, the remaining feathers have three broad bars of black, with three lighter bars, white on inner web, and a whitish tip. The gray of the crown becomes lighter and bluer on the cheeks, shading to pale gull-gray on the throat; whole of breast and flanks uniform orange-cinnamon, shading to grayish white on abdomen; under tail-coverts white. Thighs pale gull- gray, a few of the feathers with a very slight vinaceous wash. Iris bright orange; eyelids, lores, and cere, yellow; bill horn-blue, with tip black; feet yellow, claws black. Length (skin), 428 mm.; wing, 234; tail, 203; bill (culmen from cere), 20, (including cere), 27; bipthtaioie: 65; middle toe with claw, 54. 1Scientifie Results of the American Museum Congo Expedition. Ornithology, No. 5. 1 2 AMERICAN MUSEUM NOVITATES [No. 7 ApuLt Mate.—No. 157741, A. M. N. H., Avakubi, Ituri, January 20, 1914. Smaller, and more brightly colored on the breast, but thighs likewise pale gray. Iris bright reddish orange. Wing, 199 mm.; tail, 173; bill (from cere), 15, (from base), 22; metatarsus, 57. Immature Birps, from Ituri District, are blackish brown above, white below with dark spots and bars at sides, but unmarked on throat and middle of breast. Specimens Exam*ned—Astur t. canescens. Belgian Congo, Ituri: Avakubi, 1 @# ad.,1 9 ad., 1 Q im.; ‘Medje, 1 9 ad.; Uele: Niangara, 1 9 im. Astur t. toussenellii. Gaboon: Fernand Vaz,1 9 ad. Also Plate vr, Cat. Birds Brit. Mus., I, 1874. Astur sparsimfasciatus. Belgian Congo, Uele District: Aba, 1 9 ad.; Faradje, 1 2 im. Astur castanilius. Belgian Congo, Ituri District: Gamangui, 1 @ ad.; Medje, 2 92 9 im. Astur toussenellit canescens is widely distributed in the Ituri Forest, replacing A. t. toussenellii of the forested regions further west. It ranges northward at least to Niangara in the Uele District. Colius nigricollis leucophthalmus, new subspecies Supspeciric CHARACTERS.—Similar to Colius nigricollis nigricollis, but the hind neck and upper back barred with dark brown, and iris white instead of brown. Typr.—No. 158840, A. M. N. H., o ad., Niangara, Uele District, Belgian Congo, November 16, 1910. ApuLt MALE (type).—Forehead, lores, and chin black; crown, crest and ear- coverts vinaceous-buff (Ridgway); hind-neck and upper back light drab, finely barred with dark brown; scapulars, wing-coverts and outer surface of remiges olive- brown; rump slightly lighter, with a few indistinct bars; upper tail-coverts still lighter brown, with distinct blackish bars. Rectrices olive-brown, the two small outer pairs with broad white exterior margins. Throat black in middle, each feather with a gray central spot; these spots growing larger laterally and posteriorly so that the sides of the throat become light gray barred with dusky, the chest drab-gray barred with dark brown. The color of the underparts shades gradually to a uniform cinnamon-buff on belly and thighs, the breast and sides of body being vinaceous-buft barred with brown. Bases of remiges from below cacao-brown, greater wing-coverts similar, but marginals duller and paler. Iris grayish white; rim of eyelids and skin at base of maxilla black, naked areas behind eye light blue; maxilla black with a spot of bluish gray on basal half of culmen; mandible pale buff, but black at sides of base; feet scarlet, claws black. Length (skin), 310 mm.; wing, 92; tail, 209; exposed culmen, 14; metatarsus, 21.5; middle toe with claw, 26. There.is no appreciable difference between the sexes in color or size. Eighteen males of C. n.leucophthalmus measure as follows: Wing, 87-96 (93.1)'; tail, 179-221 (202.8); exposed culmen, 12.5-14 (13.3); metatarsus, 21-24 (21.9). 1A verages in parentheses. 1921] FOUR NEW BIRDS FROM THE BELGIAN CONGO 3 Six females of C. n. leucophthalmus measure: Wing, 86-97 (93); tail, 184-226 (208.4); exposed culmen, 13-14 (13.1); metatarsus, 21-23 (22). Five specimens of C. n. nigricollis from Leopoldville and Boma, Lower Congo, including both sexes, measure: Wing, 92-99 (95.2); tail, 190-206 (198.6); exposed culmen, 13-14 (13.2); metatarsus, 21-23 (22.2). SpecIMENS EXAMINED.—Colius nigricollis nigricollis. Lower Congo: Boma, 1 #ad.,2 9 9 ad.; Leopoldville,2 7 ad. Cameroon: Bitye, River Ja, 1 7 ad., 1 9 ad.; Lolodorf,7 7 ad.,6 2 2 ad.,1juv. Also Pl. 259, Levaillant, ‘Oiseaux d’ Afrique,’ VI, 1808. Colius nigricollis leucophthalmus.—Belgian Congo, Ituri: Bafwabaka, 4 7 J ad.; Medje,67 0% ad.,3 9 9 ad. 2 io juv., 4 2 2 juv.; Uele: Niangara, 29 ad.; 4 9 9 ad.; Vankerckhovenville,3 7 ad.,2 9.9 ad.; Faradje,3 7 @ad.; Garamba, 1 9 juv. Colius striatus striatus —South Africa: Natal, 5 ad. The type locality of Colius nigricollis Vieillot is Malimbe, in the Portuguese Congo,' on the southern edge of the West African forest. Not only is this black-throated Coly known from the adjoining Lower Congo District, but there are many published records from the Cameroon, mostly north of the forest, from the Shari River, and even from the Uele District, the Upper Ituri, and Lake Kivu. Sclater and Mackworth- Praed? have identified as C. striatus nigricollis specimens from Mt. Baginzi, Meridi, and Kojali in the Bahr-el-Ghazal Province. The range of these Colies extends around rather than through the Congo forests, though they are found in clearings in the Gaboon and Southern Cameroon; and the form occurring to the northeast, in the Upper Uele District, differs markedly from C. n. nigricollis as found near the type locality. Five adult specimens were collected by the American Museum Congo Expedition at Leopoldville and Boma, on the Lower Congo, in 1909 and 1915. In the four cases where the color of the eye was noted it was always dark brown. Reichenow’ and L. Petit‘ likewise describe the eye as brown, but Levaillant stated that he did not know its color. In adult birds from the northern Ituri and the Uele District, of which we collected a much larger number, the iris is invariably white or gray- ish white, and this is well shown in the photograph from life (Fig. 1). Only in young birds is it grayish brown. The colors of the bill and naked skin of the face appear to be identical in the two forms. on — agai 1808, ‘Oiseaux d’ Afrique,’ VI, Pl. ccirx, and Vieillot, 1817, ‘Nouv. Dict. d’Hist. Nat.,’ fon 4 1919, 650. 21903, ‘Vogel Afrikas,’ II, p. 204 41899, Mém. Soc. Zool. France, XII, p. 68. VII, + } AMERICAN MUSEUM NOVITATES [No. 7 Fig. 1. Colius nigricollis leucophthalmus, % ad., Niangara, Uele District, Belgian Congo, December 3, 1910. Photograph by H. Lang, from life, showing white iris. There is also a conspicuous difference in plumage between specimens from the Lower Congo and those from the Ituri and Uele. Typical C. nigricollis has the hind-neck and back of a uniform brown, as is clearly stated by both Levaillant and Vieillot, the original description! reading: “le dessus du cou et le manteau d’un brun uniforme, plus foneé sur les ailes; les e6tés du cou, la poitrine, et les flancs du méme brun, et rayés transversalement ¢’un noir lavé.’”’ There may, indeed, be faint indica- tions of barring, but these are due to the structure of the feather rather than to pigmentation. The birds from the Uele and Ituri are usually more heavily barred on the breast, and this fine barring always extends around on the hind neck and upper back. They are not C. n. nigriscapalis Reichenow, for they agree with C. n. nigricollis in the color of head and under wing-coverts. 11817, ‘Nouv. Dict. d’Hist. Nat.,’ VII, p. 378. 1921] FOUR NEW BIRDS FORM THE BELGIAN CONGO 5 Of Colius n. nigricollis from the Cameroon I have examined two ‘specimens in the Philadelphia Academy from the River Ja (G. L. Bates), and thirteen adult specimens from Lolodorf (J. A. Reis), loaned me by the Carnegie Museum. The color of the iris was noted by Bates as gray- ish brown (c’) and brown (9), and by Reis in one case as chestnut- brown (co). With regard to plumage, these birds from the Cameroon agree exactly with C. n. nigricollis from the Lower Congo. For the present, the range of Colius nigricollis leucophthalmus may be stated as follows: Savannah region of the northeastern Congo Basin, from the Nepoko River northward to the southern border of the Anglo- Egyptian Sudan, and probably extending to the westward along the northern edge of the forest, as well as southward along its eastern border to Lake Kivu. _ Batis ituriensis, new species Speciric CHARACTERS.—Most nearly related to B. minima (Verreaux) from the Gaboon, but differs in having a broad, glossy-black breast-band and a distinct white nuchal spot. Smaller than any species of Batis except B. minima and B. perkeo Neumann!; but the female of the latter has a brown breast-band. Type.—No. 159881, A. M. N. H., 9 ad., Gamangui, on Nepoko River, Ituri District, Belgian Congo, February 4, 1910. - Apu.t Femate.—A large white patch at each side of forehead, but lores, fore part of crown, and postocular region pure black. Posterior half of crown grayish black, bordered laterally by a lighter line of mixed gray and white, which does not extend forward to the eye. Feathers of nape white, tipped with gray; back dark gray, scapulars blacker, rump with oval spots of white; upper tail-coverts black. Wings black, with a conspicuous white stripe running from the outer median coverts, across the greater coverts, and down the outer margins of three inner secondaries; under wing-coverts white save at margin of wing, where they become black. Tail black, the outermost feather widely margined with white, the next two with a narrow white edge, and a small white speck at the tips of other feathers. Under surface white, save for a glossy-black breast-band, 8 mm. wide, and blackish mottling at sides of body. Feathers of tibia black, tipped with white. Iris yellow; bill and feet black. Length (skin), 85 mm.; wing, 48.5; tail, 30; exposed culmen, 11.5; metatarsus, 14. SpecmmEns ExaMInep.—Batis ituriensis. Belgian Congo, Ituri: Gamangui, 1 9 ad. Batis minima. No specimens available, only descriptions by Verreaux, 1855, Revue et Magasin de Zoologie, (2) VIII, p. 219, and by Sharpe, Ibis, 1873, p. 169. Batis minulla. Belgian Congo, Middle Congo: Suata, 1 @ ad. Batis diops ©. Grant, 1910, Trans. Zool. Soc., XTX,Pl. xv, fig. 2. Batis molitor puella. B. E. Afr.: Guasonarok, N. Guaso Nyiro, 1 o ad.; Kijabe, 1 of im. 11907, Journ. f. Ornith., LV, p. 352. 6 AMERICAN MUSEUM NOVITATES [No. 7 Batis bella nyanse. Belgian Congo, Uele: Niangara, 3 7 ad., 2 @o@ juv.; Faradje, 3 oo’ ad., 2 9 9 ad., 2 oo juv. Batis bella congoensis. Rielgian Congo, Middle Congo: ‘eam LO: Our single specimen was taken on the border of a clearing in the forest, where no other species of Batis was ever observed. Verreaux remarked! that Batis minima was found in similar situations in the Gaboon. The type of B. ituriensis was sexed by me, and I believe that the male and female will prove to be similar in coloration, as they are in B. diops. Inasmuch as the male of minima is described by both Ver- reaux and Sharpe as having a greyish-black band across the breast, the present specimen cannot be referred to that species, in spite of its agree- ment in size. Moreover, the type localities are approximately 1200 miles apart and no specimens of B. minima have ever been taken in the inter- vening territory. Fig. 2. Batis ituriensis, 2 ad. Natural size. Terpsiphone batesi, new species Speciric CHARACTERS.—Resembling 7’. rufocinerea Cabanis, with a similar, slightly marked crest but always short-tailed, even in the adult male, where the middle pair of rectrices exceeds the second pair by 14.5 mm. at most, on an average by only 8.9 mm. The head is lighter gray, the back brighter rufous than in rufo- cinerea, Typr.—No. 160095, A. M. N. H., &@ ad., Medje, northern Ituri District, Belgian Congo, March 31, 1910. ApuLt MALE (type).—Back, wing-coverts, inner secondaries, outer edges of all remiges, rump and all feathers of tail bright burnt sienna; under tail-coverts bright orange-rufous. Whole head, throat, and hind neck slate-color, with a noticeably 11855, Revue et Magasin de Zoologie, (2) VIII, p. 219. 1921] FOUR NEW BIRDS FROM THE BELGIAN CONGO ¥ bluer sheen on feathers of crest, which is rather short and rounded, the longest feath- ers measuring 11.5mm. Breast, flanks, and abdomen slate-gray, feathers immediately around vent whitish. Inner webs of remiges brownish black, bordered basally with rufous; under wing-coverts largely gray, but longer ones whitish, and the greater coverts with a slight rufous tinge. Iris dark brown; rim of eyelids slightly expanded, and blue; bill blue with black tip; feet grayish blue. Length (skin), 175 mm.; wing, 74.5; tail (middle feathers), 94.5, (next pair), 80, (outermost), 65.5; iecpowed culmen, 14; metatarsus, 15. Measurements of six adult males are as follows: Wing, 73-77 (74.9)!; tail pa eG feathers), 78-94.5 (88), (next pair), 75-83.5 (79.1); exposed culmen, 12.5-14 (13. 5); metatarsus, 14-15.5 (14.9). _ Apuit Femate.—Somewhat duller and paler, crest without gloss, back San- ford’s brown, middle of abdomen paler gray. Dimensions of three adult females are: Wing, 72-73.5; tail (middle feathers), 73.5-75, (next pair), 71; exposed culmen, 12-13; metatarsus, 14-15.5. SpecIMENS ExaMIneD.—Terpsiphone bates. Belgian Congo, Distr. of Stanleyville: Bafwasende; 1 oad.; Ituri: Avakubi, 3 70 ad.; Bafwabaka, 1 @ad.,2 9 9 ad.; Medje, 1 @ ad.,2 9 9 ad. Cameroon: Bitye, 7 lad., 1 2 ad.; Assobam,1 @ ad. Terpsiphone rufocinerea. Belgian Congo, Lower Congo: Boma, 2 @o¥ ad., 2 fo im.,1 9 im. Spanish Guinea: Rio Muni, 2 7 ad. Gaboon: Rio Moondah, 1 fad.,1 9. Terpsiphone plumbeiceps. Belgian Congo, Distr. of Stanleyville: Bengamisa, 1 @ ad.; Uele: Vankerckhovenville, 1 # ad. Nyassaland: Zomba, 1 @. _ Terpsiphone bates is a common and characteristic bird of the Ituri Forest, at least from Bafwasende on the River Lindi and Avakubi on the. Ituri northward to Medje and the River Nava. It associates with T. ignea in the mixed flocks of insectivorous birds that wander through these shady solitudes; but does not venture forth into clearings around vil- lages. Eastward, it extends to Ukaika and between Mawambi and Irumu, whence Sassi reports specimens collected by Grauer? under the name of 7’. rufocinerea. Westward, it reaches at least to Bitye on the River Ja, South Cameroon, as proved by two specimens in the Philadel- phia Academy of Natural Sciences, collected by G. L. Bates and labeled also as 7’. rufocinerea. Among the many names propoesd for species of Terpsiphone, only rufocinerea has ever been applied to this one, and I therefore take pleas- ure in naming it in honor of Mr. G. L. Bates, who has made such inval- uable contributions to African ornithology. T. rufocinerea was described by Cabanis* from Chinchoxo, Loango Coast, and I have collected specimens not very far away, at Boma, on 1Averages in parentheses. 21916, Ann. K. K. Naturhist. Hofmus., XXX, p. 258. 31875, Journ. fiir Ornithologie, p. 236. 8 AMERICAN MUSEUM NOVITATES [No. 7 Fig. 3. Heads of adult males in four species of Terpsiphone, showing form of crest. A.—T. rufocinerea. B.—T. batesi. C.—T. plumbeiceps. D.—T. viridis. X%¥. the Lower Congo. The latter agree with the original description in every respect, and one of Cabanis’ two types is stated to have had the middle rectrices greatly lengthened, as they are in my adult males. In the col- lection of the Philadelphia Academy there are likewise four specimens of T. rufocinerea from the Du Chaillu Collection, taken at Rio Moonda, on the west coast. Two of these have long rufous tail-feathers, and one is remarkable in having several greater wing-coverts on both sides black, bordered with white. Our nine adult specimens of 7’. bates? and the three I have examined from the Cameroon show a striking uniformity in their distinctive char- acters. From 7. plumbeiceps the species here described as new may be known by its shorter crest, bright rufous under tail-coverts, and short tail. It differs from 7. schubotzi, if Prof. Reichenow’s description! is 11911, Orn. Monatsberichte, XIX, p. 82. 1921] FOUR NEW BIRDS FROM THE BELGIAN CONGO 9 exact, in its bluish gray breast, and from the recently described 7’. polio- thorax Reichenow! in having the head grayish, not black, and abdomen pray, not rufous. It bears some resemblance to females and young of 7’. perspicillata and 7. viridis, but is at once separable by its much brighter rufous back and paler head with more rounded crest. The form of the crest in the different species of Terpsiphone is very characteristic, and Figure 3 shows its general outline in four of the species here mentioned. 11916, Journ. fiir Ornithologie, LXIV, p. 161. AMERICAN MUSEUM NOVITATES No. 8 POLYCHATOUS ANNELIDS COLLECTED AT ST. PAUL DE LAONDA BY THE AMERICAN MUSEUM BELGIUM CONGO EXPEDITION By A. L. TREADWELL Ps THE G Issued June 3, 1921 By OrpER OF THE TRUSTEES OF THE AMERICAN MUSEUM OF NATURAL HISTORY New Yorx Crry AMERICAN MUSEUM NOVITATES Number 8 June 3, 1921 59.51,7 (67.5) POLYCHZTOUS ANNELIDS COLLECTED AT ST. PAUL DE LOANDA BY THE AMERICAN MUSEUM BELGIAN CONGO EXPEDITION! By A. L. TREADWELL Several families of polychztous annelids are represented in the col- lection. The Nereide are Nereis pelagica Linneus and N. tongatabuensis MelIntosh. Of the Leodicidze, Diopatra neapolitana Delle Chiaje is represented by a single specimen, and there is a fragment of a tube of some leodicid. There are a few fragments of Chetopterus, too much injured for identifica- tion, and a considerable number of specimens of Dasychonopsis (Dasy- chone) bairdii McIntosh. The appearance of this last species is unex- pected, since it was first described from the West Indies, but it is evident that a sessile species like this might be easily transported on the bottom of ships. - Of the Polynoide there are several specimens of Lepidonotus clava Montagu, and a single mutilated specimen of a Harmothoé, apparently closely related to H. fraser-thomsoni of McIntosh (‘Monograph of the British Annelids,’ Pt. II, ‘Polychetes.’ Ray Society, 1900, p. 337), but the elytra lack the large tubercles found on one side only, which Mc- Intosh describes from that species. The appendages of the head were entirely lost and close comparisons are not possible. A species of Acholoé is new and its description follows. Acholoé orbiculata, new species The animals were all very much coiled in a close spiral. which made counting of the somites and exact determination of the length difficult. Sofaras I could determine, the body is about 50 mm. long, and the average specimen has a prostomial width of 0.5mm. There are 45 or 46 elytra covering the entire dorsal surface of the body to the extreme posterior end. In the preserved material the elytra have a ground color of-a pearly white, with pigment over the elytrophore in a somewhat diffuse patch, while from this patch a narrower and much more sharply defined band extends around the elytron, leaving the margin uncolored. In the entire animal the diffuse part of the pigment is covered in each elytron by the overlapping of the one anterior to it, so that the most striking feature is the succession of ringed elytra. In the first elytron the diffuse patch around the elytrophore does not occur. o — Results of the American Museum Congo Expedition. General Invertebrate Zoology, No. 7. 1 AMERICAN MUSEUM NOVITATES — . . . NOP Ot @> ‘aeeaeage 1921] POLYCHATOUS ANNELIDS 3 The prostomium (Fig. 1) is oval in outline, a little longer than broad, and with the anterior margin on either side rounded instead of being prolonged into peaks. The median tentacle is carried on a large cirrophore, which is inserted in the front of the prostomium, its terminal joint being only about twice as long as the cirrophore, tapering gradually to an acute point. The cirrophores of the lateral tentacles appar- ently arise on the level of the margin of the prostomium, but are really very slightly below it. The cirrophore is about as long as the terminal joint, which is acutely conical. The palps are rather slender, extending beyond thetentacles. The dorsal tentacular cirrus is larger than the ventral and extends about to the apex of the palp. Both palps and tentacular cirri have a few colorless papillae, visible only under a magnification of 75 to 100 diameters. The large first elytrophores bound the pro- stomium postero-laterally and there are two pairs of eyes, one pair at the posterior margin, and the other pair toward the anterior end. From a dorsal view, these appear to be very small, though they are really of a moderate size, but lie far enough under the curve of the prostomium to be partly hidden from the dorsal view. Dorsal cirri (Fig. 5) are long and slender. There is one pair of anal cirri, rather stout processes, arising from the ventral surface of the pygidium (Fig. 6). The first elytron (Fig. 2) differs from later ones (Fig. 3) only in size and in the lack of the diffuse pigmentation around the elytrophore. All elytra have smooth margins and no trace of surface papille. The first parapodium (Fig. 4) has dorsally the elytrophore of the first elytron. The notopodium is smaller than the neuropodium and carries a single acicula and a tuft of setz of the sort shown in Fig. 8. The setal portion of the neuropodium has an anterior and a posterior lip, the former being the larger, with an acicula a trifle larger than the notopodial, and a tuft of sete like those in Fig. 7. The slender ventral cirrus extends beyond the end of the parapodium. A posterior parapodium (Fig. 5, drawn to a scale one half that of Fig. 4) has much the same outline as that of the first, but is much larger, and in the case of cirrus-bearing somites, as is shown in the figure, there is a slender dorsal cirrus, extending beyond the apex of the para- podium. Dorsal to the cirrus, on the body wall, is a fold which was somewhat dis- torted in outline in the specimen figured, but which, on a surface view of the entire animal, has an outline like that of a hammer head. This is apparently a respiratory organ.! The dorsal setze (Fig. 8) are smaller than the ventral, with a bluntly rounded shaft, carrying toward the apex two rows of teeth, each tooth in the form of a narrow plate denticulated on its margin. Only one of these rows of teeth can be seen in the profile view shown in the figure. The ventral setz (Fig. 7) are much larger with their shafts heavier, and enlarged near the apex. Along this enlarged portion, extend two rows of teeth like those on the dorsal sete. As in the latter case only one row appears in profile. 1The presence of processes like these led Grube (1855, ‘Beschreibung neuer oder wenig bekannter Annelidenn.’ Wiegman, ‘ Archiv. f. Naturgesch., I, p. 81) to describe as Polynoé malleata a form from Triest. McIntosh has later (1900, ‘Monograph of the British Annelids,’ Pt. II, Ra Society, p. 397) identified this with Claperede’s Acholoé astericola, a form apparently closely related to A. orbiculata, but differing from it in the character of the elytra. In his diagnosis of the genus Acholoé, McIntosh (loc. cit., p. 396) states that the dorsal cirri occur on every foot, but in his description of A. astericola he says that the T-shaped lobes occur in the cirrigerous feet. In A. orbiculata I find that, as is the rule, elytron-bearing somites alternate with cirrigerous ones. Grube’s original description of Polynoé mal- leata stated that there are 39 elytra, while McIntosh gives 45 as the number in A. astericola, but the _latter author regards the two as synonymous. The arrangement of setz is the same in the posterior as in the anterior somites, and thers is a slender ventral! cirrus, not differing much in outline from .the anterior ones, but very much shorter. AMERICAN MUSEUM NOVITATES No. 9 A REMARKABLE CASE OF EXTERNAL HIND LIMBS IN A HUMPBACK WHALE By Roy CHAPMAN ANDREWS Issued June 3, 1921 By OrpER OF THE TRUSTEES OF THE AMERICAN MUSEUM OF NATURAL HISTORY New Yor« Crty eRe 8 as He Ce at se nD Be al AMERICAN MUSEUM NOVITATES Number 9 June 3, 1921 59.9,51M:14.98,2 A REMARKABLE CASE OF EXTERNAL HIND LIMBS IN A HUMPBACK WHALE By Roy CHAPMAN ANDREWS In July 1919, a female Humpback Whale (Megaptera nodosa) with two remarkable protrusions on the ventral side of the body, pos- teriorly, was captured by a ship operating from the whaling station at Kyuquot, on the west coast of Vancouver Island, British Columbia. Fig. 1. Caudal part ot the whale, showing the hind limb #n situ. One of the protrusions was cut off by the crew of the vessel but the other was photographed in situ by the superintendent of the Station. Mr. Sidney Ruck and Mr. Lawson, officials of the Consolidated Whaling Company, appreciated the importance of the discovery and presented the skeletal remains of the attachment to the Provincial Museum, Victoria, B. C. 2 AMERICAN MUSEUM NOVITATES [No. 9 At my request, Mr. Francis Kermode, Director of the Provincial Museum, very courteously submitted the bones to me with permission to publish upon the result of my examination. Under date of March 4, 1920, Mr. Ruck writes to Mr. Kermode as follows: I enclose herewith three photographs showing the unusual development of the pelvic Rudiments in a whale captured at the Kyuquot Station last July, of which you have the bones. It is to be regretted that better pictures in evidence of this unprece- dented development were not obtained. I have been connected with the Whaling Industry for 22 years and during my time have come in contact with prominent Naturalists such as Professor True of the Smithsonian Institute, Professor Lucas of the Natural History Museum, Brooklyn,} and Professor Andrews of the Natural History Museum, New York, and neither in their experience or mine have the protrusion of the pelvic bones beyond the body ever been seen or heard of. This particular whale was a female humpback of the average length with ele- mentary legs protruding from the body about 4 feet 2 inches, covered with blubber about one-half an inch thick. As shown in the best photograph these legs protruded on either side of the genital opening; the left leg was cut off by the crew of the vessel and lost, and the point at which it was cut off is clearly shown in the photograph. The end of the leg seen in the picture terminated in a kind of round knob like a man’s clenched fist. The two bones of the leg which you have are connected by cartilage which I was informed had shrunk about 10 inches, and possibly more by this time. At any rate the total length of the leg before it was cleaned of the blubber and flesh was, as before stated, about 4 feet, 2 inches, from the body. After studying the material and discussing it with various scientists, I have come to the conclusion that the protrusions actually do represent vestigial hind limbs and show a remarkable reversion to the primitive quadripedal condition. I am well aware that zoologists are inclined to accept reported in- stances of reversion with extreme reluctance and that, at first sight, the tendency will be to consider this a teratological case of no reversionary significance, but the evidence is so strong that I can not interpret it in that way. Mr. Ruck reports that the total length of the leg “before it was cleaned of the blubber and flesh”? was about four feet and two inches. The skeletal remains in my possession consist of two bones and two heavy cartilages. When placed in position as in Fig. 2, the total length is 31 inches. iThen of the U. 8S. National Museum, now of The American Museum of Natural History. Skeleton of the hind limb. Cartilaginous femur and osseous tibia. Cartilaginous tarsus and osseous metatarsal. 3 4 AMERICAN MUSEUM NOVITATES (No. 9 Fremur.—The larger bone is deeply concave proximally and to it is attached a massive cartilage (Fig. 3) which, in its present shrunken condition, is 5% inches in length and 1% inches wide. I estimate that this cartilage was at least 15 inches long and 3 inches wide when fresh. I believe that this cartilage represents the femur. It probably lay entirely within the body, its proximal end being attached to the pelvic vestiges. Such a massive cartilage must necessarily have had a firm support and leads me to believe that the pelvic elements in this individual were of extraordinary size. The pelvic bones, as usually present in the Megaptera, are slender ossifications about 6 or 8 inches in length and would not furnish a firm enough base for the attachment of a cartilage which, in its fresh condition, was as large as a man’s wrist. Since the photograph of the limbs in situ shows that they were directly below the usual location of the pelvic vestiges and since there are no other “ floating”’ bones near this region, the conclusion that they were attached to the pelvic elements is entirely justifiable. Trs1a.—The larger of the two bones I identify as the tibia (Fig. 3). It is 14% inches in greatest length, is well developed, and has a hard, smooth outer surface. At the proximal end its greatest width is 3%, inches, it narrows gradually for three-fourths of its length, and then suddenly expands at the distal extremity, where it is 24% inches wide. Tarsus.—The distal end of the tibia is convex and gives attach- ment to a cartilage which in its shrunken state is 4%; inches long and 1% inches wide (Fig. 4). This cartilage, I believe, represents the tarsus. That it presents no ossifications is by no means surprising as the carpal bones in the fore limbs of cetaceans are sometimes entirely absent and often in a more or less rudimentary condition. Mr. Ruck says “the two bones of the leg which you have are connected by cartilage which I was informed had shrunk about 10 inches and possibly more by this time.’”’ This would give the tarsal cartilage a length of nearly 15 inches. MerTATARSAL.—The distal element in the leg is a hard, well-devel- oped bone which I identify as a metatarsal (Fig. 4). It has the character- . istic shape of the metacarpals in the fore limbs of cetaceans except that it is more slender. It is 6% inches long, 1% inches wide proximally, and 1% inches in distal width; its least width is % of aninch. To the distal end of the metatarsal is attached a heavy cartilage of which only * of an inch remains intact. This cartilage probably formed the extremity of the limb skeleton. 1921] HIND LIMBS IN A WHALE 5 EXTERNAL APPEARANCE OF THE Lims.—In reference to the limb as it appeared in the fresh condition, Mr. Ruck says that the end terminated in a “kind of round knob like a man’s clenched fist,” that the total length was about four feet and two inches, and that it was covered with blubber about one-half inch thick. I infer from Mr. Ruck’s description that the connective tissue and blubber were essentially the same as in the flipper, or fore limb, of cetaceans. The photograph of the limb in situ (Fig. 1) show that there are two prominent, truncated tuberosities on the distal half. The proximal ‘‘bunch”’ evidently indicates the distal end of the tibia and the other is at the extremity of the metatarsal. These tuberosities may very properly be homologized with those on the outer, or anterior, edge of the flipper in the Megaptera which indicate the extremities of the radius and the second digit. This is, I believe, a point which has considerable significance. - Since the stalk-like cartilaginous femur probably lay entirely within the body and the remainder of the limb entirely outside, there was un- doubtedly a certain flexibility at the point of junction with the body. In a paper entitled ‘Untersuchungen an walen,”! Professor W. Kukenthal has described external rudimentary hind limbs in three early embryos of Megaptera. These appear as two more or less caudally directed papille on either side of the genital organ in the same relative position as the hind limbs which I have described in this paper. In Kiikenthal’s Stage I (an embryo 32 mm. in length) the rudiments are best developed and are 12 mm. long. In Stage II (an embryo 28 mm. long) the rudiments are somewhat less distinct, reaching a length of 9 mm. In Stage III (an embryo 30 mm. long) the hind-limb rudiments have still more decreased in size and appear as minute papille. Kiikenthal has also discovered hind-limb rudiments in embryos of Phocena communis and P. dalli, and Guldberg has recorded them in embryos of Lagenorhynchus acutus and Phocena communis. Kiikenthal states that the hind-limb rudiments are found in later embryonic stages of the Mystacoceti than in the Odontoceti and con- cludes that in the evolution of cetaceans the hind limbs lost their func- tional character in the Odontoceti earlier than in the Mystacoceti. Since Kiikenthal’s and Guldberg’s researches have shown that external hind-limb rudiments are still present in some cases in embryonic life, it is by no means impossible that, these vestigial organs should continue their growth and persist until the adult stage. I believe that 1Jenaische Zeitschrift fiir Naturwissenchaft, LI, 1914, pp. 49-52. 6 AMERICAN MUSEUM NOVITATES [No. 9 that is exactly what has occurred in the specimen which I have described above, and that we are confronted with a clear case of partial reversion to a primitive quadripedal condition. . The limbs, according to the statements of the whalers, were sym- metrical; they are in the exact position in which the hind-limb rudiments have been found in embryonic Megaptera; there are strong indications that the cartilaginous femur was attached to the pelvic elements; they are homologous in many respects to the flippers, or fore limbs, and, were this a teratological case, it is doubtful if these homologies would exist. Unwilling as are many evolutionists to accept reported cases of reversion, I can see no other explanation for the facts presented here. That this condition is extremely rare must certainly be true for, so far as I am aware, this is the only recorded case among cetaceans. The pres- ence of rudimentary hind limbs would almost certainly attract the atten- tion ot whalers under any condition and eventually be reported to a scientific institution, as was done in the case under consideration. Although hundreds of thousands of whales have been killed, especially in the last fifty years since the beginning of shore-whaling, no other in- stance has been reported. We are greatly indebted to Mr. Ruck and Mr. Lawson for their quick appreciation of the importance of their discovery and I wish again to express my thanks to Mr. Kermode for giving me the privilege of describing it. be ill ea AMERICAN MUSEUM NOVITATES No. 10 FIRST APPEARANCE OF THE TRUE MASTODON IN AMERICA ° By Henry FAIRFIELD Ossporn rs oe wey * ON at : eet We Issued June 15, 1921 By ORDER OF THE TRUSTEES OF THE AMERICAN MUSEUM OF NATURAL HISTORY New York Crty AMERICAN MUSEUM NOVITATES Number 10 June 15, 1921 56.9,61 M (1183:7) FIRST APPEARANCE OF THE TRUE MASTODON IN AMERICA By Henry FAIRFIELD OSBORN The geologic history of the Mastodon is obscure and the geographic distribution uncertain because of its forest-living habits, but recently great progress has been made in tracing the history of this animal in America, especially through the studies and collections by Matthew and Cook! and Sinelair,? through Schlesinger’s description of the Miocene proboscideans of Europe,’ and finally through Matsumoto’s restudy of the proboscideans of the Oligocene Fayiim deposits of northern Egypt.* From the latter it appears probable that the true Mastodon sprang from the genus Palzomastodon of northern Africa, while the buno- mastodonts (Trilophodon angustidens phylum) sprang from the genus Phiomia of the same deposits. From the American Museum collections hitherto undescribed, Matsumoto has positively separated these genera and the species associated with them, which have been confused ever since the original descriptions by Andrews of Palezomastodon, in 1901, and of Phiomia, in 1902. Matsumoto shows that the name Palzomastodon applies only to the lophodont type of P. beadnelli on which it was founded. This animal has a much broader skull than its contemporary Phiomia. These two animals were profoundly distinct, but, until the remainder of the skull and cutting teeth of P. beadnelli is known, we cannot be sure that this animal is directly ancestral to Mastodon. MIOCENE AND PLIOCENE MASTODONS IN EUROPE To the M. tapiroides of Cuvier, Schlesinger has added a series of forms which are more or less truly lophodont from the Miocene of France- and of Austria, the relations of certain of which to Mastodon are, in our 1‘A Pliocene Fauna from Western Nebraska.’ By W. D. Matthew and Harold J. Cook, 1909, Bull. Amer. Mus. Nat. Hist., XX VI, Art. 27, p . 361-414. ‘Contributions to the Snake Greek ‘auna.’ By W. D. Matthew, 1918, Bull. Amer. Mus. Nat. Hist., XX XVIII, Art. 7, pp. 183-229 2 Additions to the Fauna of the Lower Pliocene Snake Creek Beds(Results of the Princeton Uni- versity 1914 E ition to Nebraska).’ By William J. Sinclair, 1915, Proc. Amer. Phil. Soc., Phila., LIV, No. 217, May-July, pp. 73-95. Die Mastodonten des K. K. Naturhistorischen Hofmuseums.’ By Giinther Schlesinger, 1917, Denksch. K. K. Naturhist. Hofmuseums, I, Geol.-Paliontol., Rheihe 1, pp. 1-230. 4In preparation for the American Museum Bulletin by Dr. Hikoshichiro Matsumoto. 1 2 AMERICAN MUSEUM NOVITATES [No. 10 opinion, doubtful. There cannot be the least doubt, however, as to the affinity of the grinding teeth found in the Lower Pliocene of Hungary, to which Schlesinger applies the name M. tapiroides americanus. These teeth are reproduced herewith (Fig. 1, D,D1) from unpublished photo- graphs, kindly forwarded by the author, to the same scale with corre- sponding grinders (A, A1, A2, A3) from the Lower Pliocene, Snake Creek formation, of western Nebraska, also with lower teeth (B) from the Middle Pliocene, Thousand Creek, Nevada, and with (C) the posterior lower molar of M. americanus from the American Pleistocene. Mastodon tapiroides americanus Schlesinger The upper and lower grinders from the Lower Pliocene, Tasn4d, Usztaté6 Kom., Hungary, embrace a third left superior molar (Fig. 1, D1, see Pl. xi, fig. 5, Schlesinger), also two left inferior molars, m2—. ms (Fig. 1, D, see Pl. xrv, fig. 1, Schlesinger). The linear measurement of the crowns agrees closely with that of the Pleistocene M. americanus, but the vertical measurement is apparently less, i.e., less hypsodont. This indicates that already in the Lower Pliocene the mastodonts had attained the massive proportions of their Pleistocene descendants. The lophs are similarly composed and show no trace of a trefoil ridge. There is nothing to debar these Lower Pliocene mastodonts of Hungary from the true ancestral line of our Pleistocene Mastodon. THe AMERICAN PLIOCENE MASTODONS During the summer of 1908 the American Museum party, under | Dr. W. D. Matthew, first collected in the Snake Creek of western Nebras- ka a fauna subsequently determined as of Lower Pliocene age (Matthew and Cook, 1909, p. 361). The first proboscidean found appeared not to be referable to the true Mastodon (p. 367) but rather to the bunomasto- dont group. Subsequently, in 1918, several distinctive specimens were found in the same beds which may now be named as the type and para- types of a new species of Mastodon (Mastodon matthewi), in honor of Dr. W. D. Matthew, the author who first described this interesting fauna. In 1918! this fauna was divided by Matthew into two life zones, an older zone of Upper Miocene age, in which Merychippus was abundant, and a more recent zone containing Prolohippus, Hipparion, and Pliohippus (cf. mirabilis) of Lower Pliocene age. The latter zone may be known as 1Osborn, H. F., 1918, ‘Equide of the ie org es a and Pliocene of North America. Icono~ aphic Type Revision.’ Mem. Amer. Mus. Nat. Hist., N.S., II, Pt. I; p. 34, ‘‘ Preliminary Key to the Geologic betribution of the Principal Species of Bantdee Type: Colo. /Fus. 92 Drawn from cast of al /7. merriami A./4. 14471 A.M. 18239 Paratypes: A./7. 1/8238 M. matthewr Paratype: M [7217 Mo. matt hens Type: A.M. 18237 M. matthew? ae z fe ZB y = BBs= 21 = Liza = ZA aS oS ed M taptroides Loy re americanus M. ahs les Sh lew. amertcanus Inferior molars referred to Mastodon tapiroides americanus, from the Lower A3. Type and paratypes of Mastodon matthewi Osborn, from the Lower Pliocene uced through the courtesy of Dr. Giinther Schlesinger Fig. 1. Dand D1. Pliocene of Hun . Reprod A, Al, A2, an & molars of Mastodon merriami Osborn, Middle Pliocene of Nevada, Colo. Mus. 92. We Bs erred Mastodon americanus, from the phosphate beds of South Carolina, in the American of Nebraska. B. Ty Muse ‘All! figures one-fourth natural size 3 4 _ AMERICAN MUSEUM NOVITATES [No. 10 Snake Creek B or Procamelus-Hipparion Zone, similar to that of Fort Niobrara Nebraska, of Little White River South Dakota, of Clarendon Texas, and of the Santa Fé Marls B New Mexico. From his collection of 1914, Sinclair also reported (1915, p. 84, Fig. 9) a left last lower molar attributed to ? Mastodon species, collecting locality 1000A. Itis probably in this true early Pliocene of North America, broadly equivalent to Pikermi-Eppelsheim of Europe, that M. matthewi, the first true Mastodon to reach America, occurs. In his second paper, Matthew (1918, p. 199) confirms Sinelair’s division of the Snake Creek probo- scideans into two types and selects for the zygolophodont type, allied presumably to M. americanus, the generic name Zygolophodon Vacek, 1877, type M. tapiroides Cuvier. » Of more recent age, probably Middle Pliocene, is the type of Masto- don merriami from Thousand Creek, Humboldt County, Nevada, dis- covered by Mr. George D. Mathewson in digging one of the excavations along the opal outcrop on a precipitous hill about 500 feet above the level of Thousand Creek and between the main forks of the creek, as described by the geologist, Richard C. Hills... The formation consists of a more or less stratified voleanic ash containing much opalized wood. The type specimens include several bone fragments, portions of the two upper tusks, and parts of the upper molars, in addition to the well- preserved two lower molars here figured as the type. A very important character is the presence of broad enamel bands on the upper tusks (Fig. 2, C, D, E), which are perhaps similar to the enamel bands observed by Schlesinger in the true Miocene and Pliocene mastodons of Hungary. The Thousand Creek fauna of Nevada is regarded by Merriam as of Middle Pliocene age. It is deemed by Osborn as belonging to the Llingoceras-Pliohippus Zone, to which is assigned the temporary number 16 (Osborn, 1918, p. 34). Mastodon matthewi, new species Tyre: the right third superior molar, Amer. Mus. 18237. Para- TyPEs: the right second inferior molar (unworn), Amer. Mus. 17217; the posterior portion of a right third inferior molar (more worn), Amer. Mus. 18238, also of a right second inferior molar, Amer. Mus. 18239. The type and paratypes probably belong to four different individuals. Collected by the American Museum expeditions of 1916 and 1918 under 1The writer is indebted to Dr. Richard C. Hills for a letter, March 9, 1921, which contains a full account of his discovery of this interesting type. Fig. 2. Type of Mastodon merriami, Colo. Mus. 92. See also Fig. 1 B. Two lower grinders: A. Internal view. B. External view. Tusk: C. Lateralview. D. External view. E. External view of opposite tusk. Inch scale as indicated. oO 6 AMERICAN MUSEUM NOVITATES [No. 10 Mr. Albert Thomson. The type and Nos. 18238 and 18239 are from the Snake Creek B. level (Procamelus-Hipparion Zone) of Sioux County, Nebraska; the level of No. 17217 is not recorded. The type (Fig. 1, A) is distinguished by the rapid narrowing of the posterior half of the crown of the third upper molar, including the third and fourth erests; the fourth crest is extremely narrow and bilobed; the rudimentary fifth crest consists of a single cusp. In these features M. matthewi is more primitive than the corresponding tooth of M. tapiroides americanus (Fig. 1, D1) of the Lower Pliocene of Hungary. The association of the lower molars, Amer. Mus. 17217, 18238, 18239, as paratypes is provisional, because the Snake Creek deposition extended over a long period of time and may represent more than two life zones. Of these teeth, me presents three unworn pointed crests with the rudi- ments of a trefoil (Fig. 1, 41); in a second molar (Fig. 1, A2) the tre- foil is less apparent; in the third lower molar (Fig. 1, A3) it is not apparent at all. In the latter tooth, which is probably the posterior half of a third lower molar of the right side, the third and fourth crests are partly preserved; crest five is represented by a broad tuberculate talon. Mastodon merriami, new species Type: Colo. Mus. 92, found in 1909 in the Thousand Creek forma- tion, Humboldt County, Nevada, includes two left inferior molars(Fig. 2, A,B); cast, Amer. Mus. 14471, also portions of two upper tusks. The contours of these grinding teeth, as seen from above (Fig. 1, B), are convexo- (inner side) concave (outer side); the first crest is relatively narrow; the second, third, and fourth crests are relatively broad; the rudimentary fifth crest is little if any more advanced than in M. matthew?; crests two to four exhibit rudimentary intermediate cones and the spurs of a trefoil. The presence of an enamel band on the tusks and the some- what more brachyodont character of the grinding teeth separate this stage from the M. americanus (Fig. 1, C) of the Pleistocene. This species is dedicated to Professor John C. Merriam, in recognition of his pioneer work in describing the fauna of Thousand Creek. AMERICAN MUSEUM NOVITATES No. 11 THE GEOLOGY ABOUT MILLS SPRINGS, MONTICELLO QUADRANGLE, KENTUCKY By Epwarp J. FoyLes Issued June 17, 1921 , By ORpDER OF THE TRUSTEES OF THE AMERICAN MUSEUM OF NATURAL HISTORY New Yor« City AMERICAN MUSEUM NOVITATES Number 11 June 17, 1921 55.1,72(76.9) THE GEOLOGY ABOUT MILLS SPRINGS, MONTICELLO QUADRANGLE, KENTUCKY By Epwarp J. Fortes The Monticello Quadrangle of the U. S. Geol. Survey is bounded by the meridians 84° 45’ W. and 85° W. and by the parallels 36° 45’ N. and 37° N. The northeast quarter of this district may be reached by automobile from Burnside, a station on the main railroad between Cin- cinnati and Cumberland Falls, Ky. Sections were made and fossils collected within the forty-nine square miles of the area. During the three weeks which were spent in studying the stratified rocks enough data were collected to give a comprehensive idea of the physiography, structure, and historical geology of the region. The field work was done under the direction of Professor E. C. Case of the University of Michigan from the summer camp at Mill Springs. The writer is under deep obligations to Dr. Chester A. Reeds of The American Museum of Natural History, at whose suggestion this study was begun and who gave helpful advice and assistance in the writing of the paper. Negotiations for doing the field work and publication were conducted by Dr. E. O. Hovey. Acknowledgements are also extended » to Mr. Charles Butts of the United States Geological Survey for in- formation concerning the stratigraphy of an area nine miles to the southwest of Mill Springs. PHYSIOGRAPHY The area possesses the nature of a late mature to old plain on stratified rocks, slightly uplifted and moderately dissected. The Cumber- land River flows generally from east to west through several prominently incised meanders (Fig. 1). The relief is striking, the highest hills being more than seven hundred feet above the Cumberland River. These hills present beautiful undulating patches of woods and open glades. At their bases broad areas of upland appear dotted with farms. The river flats or “‘bottoms”’ are composed in large part of alluvium which fur- nishes, when cleared of timber, rich soilfor farming. The shifting of the channel in meandering streams to the outside of the bend has left steep 1 2 | AMERICAN MUSEUM NOVITATES — [No. 11 POTTSVILLE Cumberland River CHESTER FT. PAINE HANNAH WiLL | NEW PROVIDENCE ST. GENEVIEVE WARSAW! T ubeuRy S7.LOUIS AICHMOND I L —— A 8B Scale: }mile Legend > F F.. Fossil localities . < 7 Cumberland River edsel (aed CHESTER PORT PAYNE HANNAH HILL RICHMOND | WARSAW ST. GENEVIEVE BURY ST.LOUIS new ES << oe ———— — a PROVIDENC Seat gl c Saas D Fig. 1. Sketch Map of the northeast quarter of the Monticello Quadrangle, Wayne County, Kentucky, with profiles along the lines A-B and C-D. Culture and Topography from U. 8. Geol. Surv., Monticello Sheet. Profiles and Geology by E. J. Foyles. 1921] GEOLOGY ABOUT MILL SPRINGS 3 cliffs along the Cumberland and small creeks. This natural phenomenon has interfered with the establishment of good lines of communication along the stream courses and has obliged the settlers to live in more or less isolated communities. _ Where the Cumberland River crosses the Monticello Quadrangle it presents the characteristics of a rejuvenated stream flowing on nearly horizontal strata. Three cycles of erosion (Fig. 2) are represented in the area. The top of Hoozer Hill and other outlying knobs are the mani- festation of the oldest cycle. The flat upland is representative of the second cycle. The narrow canyons and intervening level stretches along the Cumberland River represent the third cycle. Fig. 2. Lock at Palace, Cumberland River. In the background may be seen the first, second, and third cycles of erosion. This area exhibits not only surface but also underground drainage in a limestone plateau where the climate is moist and the strata are nearly horizontal. The sinks form a special feature of the region. Some of these depressions are over 100 feet deep. Short creeks flow into several of them and disappear, indicating that their waters pass through under- ground channels and issue at the surface elsewhere in the form of springs. Some of these sinks may have been produced by the collapse of the roofs of caves. The field evidence would seem to indicate that the subter- ranean watercourses were developed, in the main, before the third cycle of erosion or post-Tertiary, trenching of the Cumberland valley took place. 4 AMERICAN MUSEUM NOVITATES [No. 11 STRUCTURAL GEOLOGY The rocks of the region are nearly horizontal, a fact which may be seen by observing the sections along the lines A-B or C-D on the map (Fig. 1). The only departures from the level character of the strata are caused by low domes and basins. The mouth of Forbush Creek is the center of a dome whose strata rise high enough to expose the Ordovi- cian rocks above the surface of the Cumberland River. Mill Springs is in the center of a broad basin from which the strata rise toward the north, west, and south. The dip of the beds between Forbush Creek and Mill Springs is 15 feet per mile southeast. A greater dip is exposed at the mouth of Cub Creek. The main section, upon which this report is based, was run from the ferry below the Camp up the steep slope of the hill to Mill Springs and thence to the top of Hannah Hill, a mile and one-half distant to the southwest. This section is shown by a dashed line on the map. Two disconformities were observed. The first is demonstrated by a mass of decomposed material between the Ordovician and Mississippian beds at the mouth of Forbush Creek. The second, which is between the Warsaw and St. Louis formations, may be seen at Mill Springs where the hard gray limestone of the St. Louis rests on the shale at the top of the Warsaw. An unconformity (Fig. 4) was observed in the Warsaw beds near Mill Springs. The red residual soil at Mill Springs is interpreted as being the natural product of weathering of the St. Louis limestone. STRATIGRAPHIC GEOLOGY This area possesses many favorable outcrops for the study of its stratigraphy. Shinbone Cliff on the Cumberland River is one of the finest and largest. Plenty of exposures and some fossils are to be found along the smaller streams. In the deforested uplands rounded boulders appear in the fields and ledges along the roads, while on the slopes of the hills are rocks which have been loosened by frost action. Most of the hills rising above the uplands are densely covered with trees and soil, yet enough outcrops are available for a successful study of the geology. ORDOVICIAN RicHMOND Beps.—At the mouth of Cub Creek the Richmond beds of Ordovician age, some fifty-six feet thick, are represented by thin layers of gray limestone which represent the Arnheim subdivision of this period. The rocks contain sun-cracks, ripple-marks, intraformational mud conglomerate, and miniature cross-bedding, criteria which indicate 1921] GEOLOGY ABOUT MILL SPRINGS 5 that these beds were deposited in shallow water followed by occasional emergence. Aside from dolomite, sphalerite, and calcite crystals in geodes, a few fossils (Columnaria vacua Foerste and Platystrophia cypha- conradi Foerste) were found at the base of these beds. The 23.5 feet of covered rocks, 12 feet of finely-crystalline gray limestone, and 5 feet of decomposed material underlying the Sunbury black shale near the mouth of Forbush Creek may represent the Saluda horizon, as suggested by Mr. Butts. . MIsSISSIPPIAN At the University of Michigan Geological Camp in the vicinity of Mill Springs, the Mississippian is by far the most extensive and fossilif- erous group of rocks, having a total thickness of more than 600 feet. Beginning at the base the formations are the Sunbury shale, the New Providence shale, the Fort Payne shales, cherts and limestones, the War- saw shales and limestones, the St. Louis limestone, the St. Genevieve colitic limestone, and the Chester limestones. _ SuNBuURY SHALE.—One-quarter mile up Forbush Creek occurs an exposure of the black carbonaceous Sunbury shale superimposed dis- conformably on 5 feet of decomposed Saluda material, the highest Richmond exposed in the area. Although the total known thickness of the Sunbury is 28 feet, only 5 feet appear on Forbush Creek. A careful search for fossils revealed Ctenacanthus species, Pisces indeterminable, and Phillipsia species. It is the teeth of Ctenacanthus, a shark, which made it possible to recognize the Sunbury at this place. New ProvipENcE Formation.— Directly overlying the Sunbury on Forbush Creek is the New Providence formation which consists of 29 feet of greenish-gray crumbly shale containing phosphatic concretions the size of marbles and also carbonate of iron concretions. Thisformation yields the fossils Cyathaxonia cynodon Rafinesque and Clifford, Clio- thyridina glenparkensis Weller, Platycrinus sculptus Hall, and Phillipsia species. The crinoid Platycrinus sculptus Hall is represented in great abundance by its stem-plates which are locally known as “fossil buttons”’ and ‘‘ Indian beads.” Fort Payne FormMation.—On the road leading from the Camp to Mill Springs the Fort Payne or Keokuk is exposed vertically for 97 feet at a low-water stage of the Cumberland River. In general, this formation is composed of a stiff dark shale, chert bands, and lenses of coarse, gray and sometimes crinoidal limestone (Fig. 3). The character of these rocks varies greatly in short distances. In the immediate vicinity of the camp 6 AMERICAN MUSEUM NOVITATES [No. 11 22 feet of gray, nodular, impure shale are exposed at the base, followed by 11 feet of limestone with solution cavities, then 25.5 feet of gray geodic shale. On top of this appear a five and one-half foot lens of gray compact limestone, 21 feet of gray shale with two six-inch chert layers near the top, and 12 feet of gray limestone containing the fossils Chonetes species, Spirifer biplicoides Weller, Productus magnus Meek and Worthen, and Rhynchopora cooperensis Shumard. Warsaw Formation.—In the vicinity of Mill Springs the Warsaw, which is well exposed for 92 feet, is divided into six zones. The basal member is composed of 16.5 feet of gray geodic shale containing carbona- Fig. 3. Outerop of the Fort Payne Formation, Mississippian, on Meadow Creek, showing limestone lens above hammer and flaggy shale below. ceous concretions. The next zone consists of 18 feet of brownish-yellow to gray compact limestone. On this lie 16.5 feet of decomposed gray shale which yielded the fossils Pustula biseriatus Hall and Chonetes il- linoisensis (Worthen). At the top of this bed, on the road between the camp and Mill Springs, appears a local unconformity (Fig. 4). Over- lying this are 22.5 feet of yellow, impure limestone containing T'riplophyl- lum species. Then follow 13 feet of limestone in which are found the fossils Spirifer bifurcatus Hall and Reticularia pseudolineata Hall. The last and uppermost member is a gray fissile shalé 5.5 feet thick in which no fossils were found. St. Louis Formation.—Lying disconformably on the Warsaw formation is the St. Louis, a hard gray limestone containing the fossils 1921] GEOLOGY ABOUT MILL SPRINGS 7 Lithostrotion basaltiforme Owen, Lithostrotion proliferum Hall, Spirifer bifurcatus Hall, Mesoblastus glaber Meek and Worthen, and Nautilus species. Although 34.5 feet of the St. Louis are exposed at Mill Springs, it is probable that the overlying covered rock, 80 feet in thickness, is also a part of the St. Louis. Caverns, sink holes and underground drainage characterize this formation. Due to the soil cover, the contact with the overlying St. Genevieve was not observed. The probable line of differ- entiation is shown on the map (Fig. 1). Sr. GENEVIEVE ForMATION.—Superimposed on the St. Louis at Hannah Hill 1.5 miles southwest of Mill Springs is the Fredonia forma- Fig. 4. Local unconformity in.the Warsaw beds, Mississippian, on -the road between Mill Springs and the University Camp. tion of the St. Genevieve. Although no contact between the St. Gene- vieve and the St. Louis was to be seen, it is estimated that the St. Genevieve is 60 feet thick. Itis composed of odlitic limestone containing the fossils Girtyella indianensis Girty, Eumetria verneuiliana Hall and Lithostrotion harmodites Edwards and Haime. CHESTER SerRIEsS.—On the slope of Hannah Hill the Chester is represented by the Gasper and Glen Dean formations totalling a thick- ness of 146 feet. The Cypress and Golconda formations of this series were not observed. Due to soil cover, the upper and lower contacts of the Gasper, which is a fine-grained crystalline limestone, are difficult to determine. The Glen Dean consists of dark coarsely-crystalline lime- stone. These formations yielded the fossils Productus ovatus Hall, : - Bs suorjos009 snosoeu atk GOT ayBys orpoes ST ; ; Qu04SaUtT] qoeduioo = ea 0} MOTTOA-YSTUMOIG 5 BA is oat at Il®@H Snze2rsaszg @)NjSNT ¢'9l UIYWAO A SsuasLoUNp. sajauoy)| AB[D UMOIG YYIM OTeYs AvIs govduIOdUT AytuLIOy uviddississt ¢'2Z soroods wnpphydojdity\-uooun jRoory ‘ouoyseUUT] aANduat MOTTA X ; MBSIE eH snzeounfig safindg €T II®H eyeauyopnasd eisepnayay ou0}soulr'T aU0}SOUTT] OFUI SOpBIN) ¢'¢ a[Bys eyissy ABI UdyyO pee yoo] 1992]6 snjzse]gosa py satseds snjane Ny eH snzeoinfig safiudy Il®H wnsafyoud uoyousoywT ; +¢'FE TOMO auwofiyjeseg UoYousoy]| SUIeA AYIOYD ‘“ouoysouNT] AvIB prey {mefegismoTy 4g aABD qyourure PY AVIn) sisuaueypur epahpey IIR@H eueyinausaa eruawngy O09 oulIe ET BADIA pue spleMpy saypowsey Uwo01y0.1js0YyjVT QUOISOUIVT IPOD} evIUOped,q|-9U9+) “4g ITH snzeao sngonpot auoysouT] ourT[eysA19-ATOULT Jadsery *F OFT u94}}e1g pue poomion suebaa snwbesyderq II®H &00 eLjawng QUOYSOUI'T BSIBOD|) URI] ETD Jaysoyy Ost syoyoog snou : . -1UINJIG, ‘OUOJSPUBS UMOIG-YSIAMOT]AX| I[LAS}Og uviueal[Asuuag 4995 ul ssou -yory, |, S]ISSO i] yooy jo SorstiazoRseyy) aBeys-qng | oRg Sallag ur1aysAg dnory AMOOLNGAM ‘ALNOOO ANAVM ‘ATONVUAGVAD OTIAOLLNOW AHL JO UALUVAS LSVAHLUON AHL AO STISSOA GNV SMOOU AO ATAVL eys100q wpesuoo-eydha erydoushjyeld sapoos us ayoreo = ‘aytu0[oq UIPpeq-sso19 aINyRIUTAy O}BIOUIO[SUOD PHU [BUOT}VULIOJA}UT ut oepuepeyds 9¢ OyS190,J ENIvA VILLEUUNIO) syreur-ojddny . SyoBIO-UNg| uwToquiy auojsourl] ABIS JO S1OAB] UT, puour G*kz yoourg ysnqiog -yory, UBLIAOPIC) ZI euoysoull, Avid QUIT[BySA1O-ATOULT Bpnyeg c [Blieyeul pesoduroseqy @[QBUIULIO}OpUI Sadsty satoeds snyjyweo0uap) yooy c sotoeds visdyyyryd e[Bys Snosovuo0qiBy youjg| Aanqung| -s1epury II®H snydynos snursohje,g saroods vrsdyyryd 62 IPM swsuayseduah eurprliyjoyy SUOT}A19U0d UO ployyg |jo ayeuoqiwo pue oul, jo azeydsogg|) asouapt Bs0y pus oenbssuysy wopoufia eviwuoxpyzehp ayeys Ayquindd Awid-Yystusely|-AOIg MAN] -BANg Ze sorveds snutwohzeq JOANT Pury ' -raquing = ‘eyeys aindum zejnpou ABry Il You MopBesy ‘ jO pod UT SaIPIAGS UOTYNTOg ‘auoysoUNIT] G°Sz areys o1poed Awiry g's sua] sUOJseUNT] yowdUIOD ABD ouAeg dq A[IOAB AA 1z do} 48 Ss1aAB] JAOYO YOUI-xIs OM], “9UOJSOUIT] OJUT Sad..0UL afeYys ABD ITEM saprooydag safrsrd : | paeuinyg ps oe aL Raced ZI| UeyWOM pus yooy snubew snjonpoig aPIYM YBNOL B OF SIOYIVOM ‘sotoeds sayauoy) euoyseUNT] ABIN) 10 AMERICAN MUSEUM NOVITATES (No. 11 Diaphragmus elegans Norwood and Pratten and Eumetria vera Hall. Archimedes laxus Hall and Prismopora serrulata Ulrich are diagnostic fossils of the Glen Dean. PENNSYLVANIAN PoTtTsvILLE FormMatTion.—Succeeding the Glen Dean is the Potts- ville, which consists of a medium-grained yellowish-brown sandstone. On the south side of Hannah Hill there is a bituminous pocket about 8 feet in diameter from which fuel has been dug. This formation, which is 130 feet thick, contains no fossils and caps Hannah Hill. SUMMARY The order of superposition, character, thickness, and names of the fossils of each of the beds which have been discussed are summarized in the preceding table. - BIBLIOGRAPHY Forrste, A. F. 1906. ‘The Silurian, Devonian and Irvine Formations of East-central Kentucky.’ Kentucky Geological Survey. Bull. 7, pp. 10-14. Morssg, W. F., anp Forerste, A. F. 1912. ‘The Waverlian Formations of East- central Kentucky.’ Kentucky Geological Survey, Bull. 16, Serial 19, pp. 1-49. WELLER, Stuart. ‘1914. ‘The Mississippian Brachiopoda,’ Illinois State Geological Survey, Monograph 1, 508 pp. Butts, CHartes. 1918. ‘Mississippian Formations of Western Kentucky.’ Ken- tucky Geological Survey, Frankfort, 1917, 272 pp. Miuuer, A. M. 1919. ‘The Geology of Kentucky.’ Department of Geology and Forestry of Kentucky, Series 5, Bull. 2., 392 pp. Jrtuson, W. R. 1920. ‘Contributions to Kentucky Geology.’ Department of Geology and Forestry of Kentucky, Series 5, Bull. 4., 262 pp. = | AMERICAN MUSEUM NOVITATES © No. 12 $ NOTES ON NORTH AMERICAN BLOOD FLUKES = | ; By Horace W. SrunKARD Issued July 16, 1921 = cf By OrpER OF THE TRUSTEES ‘3 oF x THE AMERICAN MUSEUM OF NATURAL HISTORY New. Yorr Crry %, Ngee kyl ee ee Ye HY eS eo 3" < / AMERICAN MUSEUM NOVITATES Number 12 July 16, 1921 59.51,22 S (7) NOTES ON NORTH AMERICAN BLOOD FLUKES'! By Horace W. STuNKARD ‘The first North American blood flukes discovered by the writer were found in 1913, while he was a graduate student at the University of Illinois. In the autumn of that year an extensive parasitological exami- nation of turtles was begun. Shipments of various species of turtles were obtained for this work from collectors in Havana, Illinois; Walker, Iowa; Newton, Texas; and Raleigh, North Carolina. On November 5, 1913, in the washings of the intestine of Pseudemys elegans collected near Havana, Illinois, a trematode was discovered which attracted attention by its peculiar and. unusual movements. On November 18, 1913, an additional specimen of this particular trematode was found in the cesoph- agus of another specimen of P. elegans caught near Havana, Illinois. On December 15, 1913, a similar trematode was removed from the trachea of Malacoclemmys leseurii collected near Newton, Texas.’ The two specimens removed from P. elegans were stained and mounted in toto; and that taken from M. leseuwrii was cut in cross-sections. No other specimens were found at that time. The following summer turtles were collected in Iowa and Illinois, and in the autumn a second shipment was received from Raleigh, North Carolina. On examination of this material, the peculiar trematode was again encountered and its true nature discovered. On April 10, 1915, three specimens were removed from the heart and six from the large arteries of Pseudemys scripta collected near Raleigh. N. C.. During the spring of 1915, blood flukes were removed from the heart and arteries of Chrysemys marginata col- lected in Iowa, Illinois, Indiana, and Ohio; from Chelydra serpentina collected in Iowa, Illinois, Ohio, Louisiana, and Texas; and from Pseu- demys elegans and Malacoclemmys geographicus collected in LouisiAna. In the fall of 1916, examinations were continued at New York University and blood flukes recovered from the heart and larger arteries of Chry- semys picta and Chelydra serpentina collected in New York, New Jersey, and North Carolina; and also from Chelopus guttatus and Cistudo carolina collected at various points in New York and New Jersey. -This material obviously belonged to a common genus, but it included forms so different that it was impossible to refer them to the same species. 1Contribution from the Biological Laboratory, New York University. 1 2 AMERICAN MUSEUM NOVITATES [No. 12 The study was interrupted from the spring of 1917 to that of 1919, during which time the writer was in the U. 8. army in France. With the resumption of scientific work on release from military service, I found that in a paper dated July, 1918, Dr. G. A. MacCallum had published a description of a trematode from the intestine of Chelopus insculptus, so similar to the blood flukes I had collected that it appeared they must be the same. MacCallum named this parasite Spirorchis, but omitted the specific name. The blood flukes and the form described by Mac- Callum are monostomes of almost the same size and shape; they agree in position and character of oral sucker, position and extent of intestinal ceca, position and extent of vitellaria, vitelline ducts and receptacle, position and shape of ovary, oviduct and uterus, position, character and extent of testes, shape and location of seminal vesicle and vas deferens, as well as the position of the excretory pore. They are alike in character of the intestinal contents, which led MacCallum to describe the form as a hematophagic trematode. The only points of difference are found in the statement of MacCallum that in Spzrorchis a pharynx is present and that the genital pore is median near the posterior end of the body, while in the blood flukes a pharynx is absent and the genital pore is lateral, slightly posterior to the level of the ovary. Conferring with Dr. MacCallum, I learned that the description was made from specimens mounted in toto, but unfortunately the slide could not be found. Dr. MacCallum examined several of my slides and noted the similarity between these worms and that described by him, but was not certain that they were the same. In correspondence with Professor Henry B. Ward, of the University of Illinois, under whose direction my graduate work was done and who was familiar with my studies on blood flukes, I wrote on February 2, 1920, that I was certain that the form described by MacCallum as Spirorchis is not from the intestine but from the mesenteric vessels and that it belongs to the group of blood flukes, and asked whether in his opinion I should describe the blood flukes as a new genus or assign them to the genus Spirorchis. Subsequently, I wrote Dr. C. W. Stiles of the International Commission on Nomen- clature, stating the case and asking for information as to the correct method of procedure in determining a name for the blood flukes. In his reply, dated March 1, 1920, Dr. Stiles gave as his opinion that the name Sptirorchis would be established by ‘finding the original slide or by col- lecting material from the type host and type locality and redescribing the genus.”’ “If it proves that your material is identical with Spirorchis, I believe that Spirorchis would take priority.’”’ ‘‘ According to the rulings 1921] NORTH AMERICAN BLOOD FLUKES 3 of the International Commission, a generic name may be valid even though no specific name is published withit. The first specific name that is published after the generic name hecomes the type of the genus.” In the Journal of Parasitology, March 21, 1921, Ward published certain observations and a description of ‘A New Blood Fluke from Turtles,’ giving to this parasite the name, Proparorchis artericola. He reports that the fluke has been found in several distinct species of turtles and widely separated localities, and records it from Pseudemys elegans at Havana, Illinois; Malacoclemmys leseurii from Newton, Texas; Pseudemys scripta from Raleigh, North Carolina; and Chrysemys marginata from Fairport, lowa. He adds, “The data in my possession are not all referable to the single species which has just been described. In details of structure, in regard to the eggs, in the location in the host in which they have been observed, and in some other details, certain speci- mens differ so distinctly that I can not at present include them under the same heading.” Referring to MacCallum’s paper, Ward accepted his diagnosis as it stands, assigned to the form the specific name innomi- nata, and included the genera Spirorchis and Proparorchis as members of a new subfamily Proparorchine. He removed the genus Hapalotrema Looss 1899, from the subfamily Liolopinse Odhner 1912, and included it with the subfamily Proparorchine in a new family, Proparorchide. Though engaged for several years in the study of blood flukes of turtles, publication has been delayed because of lack of certainty regard- ing two points; first, the question of nomenclature and the relation of the blood flukes to the genus Spirorchis, and second, the difficulty of specific determination of the material at hand. The latter of these questions is still under investigation, and may not be solved until the developmental stages and life history are known; the former has now been answered. The original specimens of Spirorchis, to which Ward assigned the specific name innominata, have been found, and through the kindness of Dr. MacCallum have been loaned to me for examination. After careful study I wish to make certain corrections and additions to the description of the form. As noted previously, the only cardinal differences between the blood flukes and the description of MacCallum are the presence or absence of a pharynx, the position of the genital pore, and the location within the host. The structure described as a pharynx in Spirorchis, though it somewhat resembles such an organ, isin reality the cesophageal commissure of the nervous system and no pharynx is present. : The cesoph- agus is surrounded by unicellular glands which at the posterior region are large and stain deeply. The genital pore is ventral, below the 4 AMERICAN MUSEUM NOVITATES [No. 12 cecum of the left side, a short distance posterior to the level of the ovary. MacCallum traced the cirrus sac to the intestine, but was unable to follow the genital ducts to the structure he regarded as the genital pore, — It is a significant fact that though MacCallum reported the parasite from the intestine, he noted that its intestinal content showed it to be a hematophagic trematode. As Ward pointed out, “It is not unlikely that its presence in the intestine was accidental, due to the opening of some blood vessel during the dissection.’’ The first blood flukes found by the writer were discovered in the washings of the intestine after its dissection, and though I have not as yet been able to secure specimens of Spirorchis from the circulatory system of Chelopus insculptus, I have found eggs of blood flukes in the tissue of that species and I am still of the opinion stated over a year ago that the specimens described by Mac- Callum came originally from mesenteric blood vessels. On the basis of morphological similarity, there can be no doubt that the blood flukes of turtles belong to the genus Spirorchis. I have speci- mens collected from the same hosts and the same localities as those of Ward, and which are certainly specifically identical with his material. These specimens show no generic differences from those discovered by MacCallum. The generic description of Proparorchis as given by Ward agrees with the corrected description of Spirorchis, demonstrating the identity of these forms. With the establishment of their identity, Proparorchis disappears assynonym. With the synonomy of Spirorchis and Proparorchis and the suppression of the latter name, the subfamily and family names Proparorchine and Proparorchide also disappear. Spirorchis remainsthen as theonly known genusand type of the subfamily to whichit belongs, and for which, in conformity with the rules of zoological nomenclature, I propose the name Spirorchine. I agree with Ward that Hapalotrema does not find a natural position in the subfamily Liolopine of the family Harmostomide, and that it must be removed from those groups. It differs from Spirorchis in the location of the ovarian complex and genital pore and also in the possession of an acetabulum. No exist- ing subfamily will contain it, and it may -well be considered as the type of a new subfamily, the Hapalotreminz. Ward included Hapalotrema and the Proparorchine (syn. Spirorchinz) in a new family, for which I propose the name Spirorchide. In position of female organs and genital pore, Spirorchis resembles Aporocotyle.and manifests relationship to that form. The discovery of the American blood flukes of turtles establishes a firmer basis for the conception of the unity and evolution of the blood inhabiting trematodes. 1921] NORTH AMERICAN BLOOD FLUKES 5 In my opinion, the Aporocotylide of fishes, the Spirorchid of turtles, and the Schistosomide of birds and mammals constitute a well-defined group with inherent natural relationships. The Spirorchidx stand in an intermediate position, and the schistosomes are, I believe, derived through them from the Aporocotylide rather than from the Harmo- stomide as maintained by Odhner. Specimens of Spirorchis collected from various species of turtles, on which these notes are based, form part of the collection of blood flukes in the Department of Lower Invertebrates of The American Museum of Natural History. AMERICAN MUSEUM NOVITATES No. 13 NEW GENERA OF PALEOCENE MAMMALS. By W. D. MatrHew AND WALTER GRANGER Issued September 6, 1921 : By ORDER OF THE TRUSTEES OF THE AMERICAN MUSEUM OF NATURAL HISTORY New Yor« City oy by men vy 2 Le™ £ AMERICAN MUSEUM NOVITATES Number 13 Sept. 6, 1921 56.9 (1181:73) NEW GENERA OF PALEOCENE MAMMALS By W. D. MatrHew anpD WALTER GRANGER In the course of the revision of the Puerco and Torrejon faunas of New Mexico and the description of the more recently discovered Tiffany fauna of southwestern Colorado, several new genera were recognized, which it seems advisable to place on record in advance of the full descrip- tion. MULTITUBERCULATA Plagiaulacide Ectypodus' musculus, new genus and species ° Typre.—No. 17373, upper jaw from the Mason pocket, Tiffany beds, Animas Co., Colorado. ToprotrypEs.—A series of seven upper and lower jaws and many isolated teeth and jaw fragments, all from the Mason pocket. _ Generic Cuaracters.—Dentition: 7{j-3-3. Teeth much as in Ptilodus, except that p* is of simple trenchant type instead of molariform. The crown is triangular in cross-section and there are seven tubercles on the interior cutting edge. On the outer face of the crown near its anterior border there are usually two (one in type) minute tubercles; these appear to be a rudiment of the second row of tubercles which in Ptilodus extends the full length of the tooth. P! and P? are tricuspid; p* is quadri- cuspid. The lower jaw (topotypes) is similar to Ptilodus, except that Ps is absent. Speciric Cuaracters.—P!-M?=6.2 mm. Eucosmodon,” new genus Typr.—Neoplagiaulax americanus Cope, of the Puerco formation, Lower Paleocene of San Juan basin, New Mexico. GENERIC CHARACTERS.—Agrees with the true Neoplagiaulax, of the Cernaysian of France, in absence of P* (present in Ptilodus), but differs from both of these genera and from Ectypodus in the large, compressed, fully scalpriform incisor, rootless or nearly so. The species are of considerably larger size than those of the three above- named genera. erenreneved in relief; 6d0ts=tooth; in reference to the peculiar type of sculpture of the lower premolar. : 226=well; xosuos=ornament; édo0¢s=tooth. 2 AMERICAN MUSEUM NOVITATES [No. 13 MARSUPIALIA Didelphide Peradectes' elegans, new genus and species Type.—No. 17376, a pair of lower jaws, nearly perfect, from the Mason pocket, Tiffany beds, Colorado. ParatyPe.—No. 17369, upper jaw with M!~4 Generic CHaArAcTerS.—Dentition: £33. Upper molars tritubercular, paracone and metacone subequal, stylar cusps weak, conules rudimentary. Last upper molar transverse. Lower molars with high trigonids and deep basin heels, paraconid well developed, anterointernal; premolars simple, with high compressed protoconids and small heels on P,-,. Molars increasing somewhat in size from first to third, the fourth slightly smaller with narrower heel. Canines moderately large, 2: as in _ Marmosa. Jaw slender, with inflected angle as in opossums. SpreciFic CHARACTERS.—Size minute, P}-M4=11 mm. About fifteen specimens of upper and lower jaws of this genus were obtained from the Mason pocket, it being, next to Nothodectes, the most abundant form. The generic characters separating Peradectes from Peratherium are found principally in the upper teeth. Thylacodon’ pusillus, new genus and species * Typr.—No. 16414, a fragment of lower jaw, with m2-3, the heel of m; and part of alveolus of m4, from the upper level of the Puerco formation near Ojo Alamo, San Juan basin, New Mexico. Exped. 1913. GENERIC CHARACTERS.—Molar teeth of didelphid type, rather narrow as a whole, the trigonid relatively high, with reduced paraconid. Metaconid and proto- conid high, well separated, acute, the protoconid considerably the higher. Talonid deeply basined with acute marginal cusps, the entoconid internal, hypoconulid well developed on all molars and nearly posterointernal in position (a characteristic didelphid construction), hypoconid postero-external. The high bicuspid trigonid and reduced paraconid serve to distinguish this from other didelphid genera; the tooth is also unusually narrow and the hypoconulid and entoconid more distinct than usual. Speciric CHARACTERS.—Moe-3=5. mm.; My-3=7.2 mm. (approx.). Size of the larger species of Marmosa, about a third larger than M. chapmani. Represented in the collection by the type specimen only. INSECTIVORA Leptictide Leptacodon tener,’ new genus and species Typr.—No. 17179, lower jaws with badly crushed front of skull, from the Mason pocket, Tiffany beds, Colorado. lrjpa=pouch; djxrns =biter; i. e., a carnivorous marsupial. Also suggestive of its relative Peratherium. *0thaxos=pouch; 650(s=tooth, i. e., marsupial tooth. : yi *From \errés = small, delicate; ax = point; é50is=tooth. Intended to be also suggestive of its relationship to Leptictis and Diacodon. Te o! 1921] NEW GENERA OF PALEOCENE MAMMALS 3 Famity CuHaracters (Leptictide).—Jaw long, slender with low condyle, promi- nent narrow angular process curving downward and backward, not inflected, coro- noid process low, broad. Upper molars tritubercular, paracone and metacone sub- equal, external, with strong cingulum around outer base but stylar cusps rudimentary, conules minute, protocone large, subcrescentic, hypocone small but prominently projecting; lower molars with short high trigonids and large basin heels; p* molari- form except that me is smaller than pa and pr and hy smaller than in true molars. ps molariform, but with pa? strong, projecting prominently forward, anterior premolars moderately large, compressed, simple, not crowded. GENERIC CHARACTERS.—Trigonids lower than in Diacodon, paraconids distinct, molars reduced in size from first to third. Protoconid overtopping inner cusp. The least specialized in molars and premolars of any member of the Leptictine. Speciric CHARACTERS.—Size minute, Pep-—M3 =7.3 mm. Xenacodon! mutilatus, new genus and species Type.—No. 17407, a right ramus of the lower jaw with Ps, Mo-3 and alveoli of remaining teeth, from the Mason pocket, Tiffany beds, Colorado. GENERIC CHARACTERS.—Dentition ;;73. Incisors small, canine of mod- erate size, P; two-rooted, Ps with large, well-separated metaconid and proto- conid, rudimentary paraconid and very small heel. Molars as in Diacodon and Palz- ’ olestes with high trigonids of two principal cusps well separated, very small paraconids, large high-cusped heel with deep basin opening inward, high external hypoconid and posterior hypoconulid, entoconid smaller, posterointernal and imperfectly separated from hypoconulid. Distinguished from other genera of Leptictide principally by the peculiar character of its P4. Speciric CHARACTERS.—Size of Palzxolestes puercensis (Mo-3=7. m.) which it resembles in general proportions. Third premolars considerably more robust; the first and second of subequal size, both being two-rooted. Premolars not spaced and only a very small space behind the canine, which is a somewhat larger tooth than i P. puercensis. Represented in the collection by the type only. Acmeodon’ secans, new genus and species Typr.—No. 16599, a part of the lower jaw with Po_M2 and the root of the canine preserved. From the upper level of the Torrejon formation, Torrejon Arroyo, San Juan basin New Mexico. Exped. 1913. ; GENERIC CHARACTERS.—Dentition 77-33. Molars of leptictid type but tri- gonid not so high and paraconid better developed than in Diacodon or Palzolestes. P, of peculiar pattern, the principal cusp (protoconid) much compressed and crested, with strong accessory cusps (paraconid, protostylid) on the anterior and posterior edges and a somewhat weaker posterointernal cusp (metaconid) connected by a prominent crest with the apex of the protoconid; also a well-developed basined talonid with acute posterc-external and posterointernal cusps (hypoconid, entoconid). lgevos=strange; axj=point; ddo0is=tooth. Inreference to the very peculiar ps4, which appears at first glance to be a broken and incomplete tooth. 24xun =a crest or edge; dd0¢s =tooth. 4 AMERICAN MUSEUM NOVITATES [No. 13 P3 large, simple, high, acute and compressed, with anterior postero-external and pos- terointernal crests, and a small, low, simple, acute heel-cusp. P2 much smaller, simple, with anterior and posterior crests and a minute heel-cusp. Canine rather small, oval in cross-section at base. At least one small incisor is present. SpeciFIc CHARACTERS.—P2-M2=19.3 mm.; M;-3=10.6 mm. (approx.). A faint rudiment of anterior basal cusp on Pz and P3; on Pyitis distinct; protostylid a faint rudiment on P3; on Ps this is a strong cusp connected by a crest with the apex of the protoconid. In addition to this type there is a second specimen, paratype No. 16600, a jaw fragment with Ps—-Mi; and alveoli of remaining molars, from the same horizon as the type, Escavada wash. MENOTYPHLA Plesiadapids Labidolemur' soricoides, new genus and species Type.—No. 17400, lower jaws with incisors, My right and left, and alveoli of remaining teeth; from Mason pocket of the Tiffany beds, Colorado. GENERIC CHARACTERS.—Dentition, -ga-3. Jaw short and of moderate depth; incisor greatly enlarged, semiprocumbent, considerably curved upward towards the tip, crown trihedral, with sharp, knife-like, coarsely serrate outer margin, anterior face flattened convex, root elongate. Fourth premolar much reduced, probably two- rooted, crown unknown, diastema much reduced owing to enlargement of incisor. First molar with a moderately high trigonid of two subequal cusps (protoconid and metaconid) well separated, and a low ledge-like paraconid appressed to anterior margin; large rounded basin heel. Third molar with elongated heel. Coronoid process low, narrow, projecting strongly backward, angle flat and of moderate width, with narrow backward projection at tip. Mental foramen below anterior edge of Me. Speciric CHARACTERS.—Mj-3=5.7 mm. (approx.); length of incisor crown (lingual side) =6.7 mm. This form is close to Phenacolemur of the Wasatch beds of Wyoming, but is distinguished from it by the apparently greatly reduced premolar. In Phenacolemur this is a large robust tooth, exceeding in size the first molar. It is possible that Labidolemur is the lower dentition of Ignacius (infra) but there are evidently two forms of small plesiadapids in the Tiffany beds as indicated by lower incisor teeth, one as in the present genus and the other with a much simpler, laterally-compressed and straighter crown. Aside from the type this new form is represented in the collection from the Tiffany beds by a lower incisor, No. 17402, and, somewhat doubtfully, by two other specimens, Nos. 17401, 17405, in each of which *\aBls =pincers and Lemur. Be et ee en el ee ee eT - 1921] NEW GENERA OF PALEOCENE MAMMALS 5 the last lower molay is preserved. This tooth shows the peculiarly broad elongated talonid and is structurally similar to that in Phenacolemur, but compared with this latter genus both Labidolemur and Ignacius are minute. : ? Plesiadapide Ignacius! frugivorus, new genus and species Type.—No. 17368, upper jaw with C, P*-M? and alveoli of remaining cheek teeth. From the Mason pocket of the Tiffany beds, Colorado. Generic CHaracters.—Dentition, “+. Canine small, simple, double- rooted, pointed. P* two-rooted, p* nearly as large as m', submolariform, metacone much smaller than paracone, and no crest connecting metacone with protocone, otherwise this tooth is similar to the true molars. Molars with subequal outer cusps, no external styles, minute protoconules, a broad sloping shelf occupying the postero- internal angle of the tooth. Low crests connect the protocone with the two outer cusps. {ee Cu ARACTERS.—C-M?=8.3mm.; M!-*=6. mm. (approx). Two specimens, Nos. 17377, lower jaw with one molar, and 17408, loose lower incisor, premolar and molar, may pertain to this genus. The incisor is a slender, gently tapering lanceolate tooth; the premolar, presumably P3, is two-rooted, with subconical protoconid and well- developed heel. The molars have a trigonid of two opposite, nearly equal cusps, connected nearly to their tips, and no paraconid; the talonid is broad, basined, and without hypoconulid. A transverse crest connects ' the hypoconid with the metaconid. PRIMATES Tarsiide Navajovius kohlhaase,? new genus and species Type.—No. 17390, upper and lower jaws, probably of the same individual, from the Mason pocket in the Tiffany beds, Colorado. GENERIC CHARACTERS.—Dentition, Ssh Two lower incisors somewhat en- larged, with elongate roots and long, pointed, subspatulate crowns. Canine smaller, one-rooted, premolariform. Third premolar smaller than canine, two-rooted, crown comparatively simple; P, nearly as large as Mi, with trenchant heel and rudimentary metaconid. Lower molars tritubercular, with distinct but small paraconid extended outwardly as a low ledge; protoconid and metaconid equal, submarginal, talonid basined, wide and deep. Upper fourth premolar not as large as M!, with large sub- 1From the town of Ignacio, Colo., about seven miles west of the Mason pocket. _ 2From the Navajo mountains north of the San Juan River. Asa mnemonic convenience the names of our new Tarsioid genera have all been derived from various mountain ranges in the region where they were found, the names chosen being of Indian derivation, and most of them originally applied to Indian tribes of the region; and they are latinized in the same form as Tarsius. : The species name is in honor of Miss Erna Kohlhaase, to whose skilful and patient work is due the preparation of the minute and delicate specimens from the Tiffany beds herein described. 6 AMERICAN MUSEUM NOVITATES. [No. 13 trigonal external cusp and rudimentary internal cusp. Molars tritubercular with rudimentary hypocones on M! and M?, conules minute, no external stylar cusps. Speciric CHARACTERS.—C-—M3 =7.5 mm.; Mj-3=4.7 mm. Represented in the collection by the type only. This is the only true primate thus far described from the Paleocene. The Plesiadapide are primitive types lying on the border between Primates and menotyphlan insectivores. _ INCERT# SEDIS Carpodaptes' aulacodon, new genus and species _ Typse:—No. 17367, lower jaw with Po—Ms3, alveolus of P; and part of canine alveolus. From the Mason pocket of the Tiffany beds, Colorado. GENERIC CHARACTERS.—Dentition, 7-7-3. Canine moderately large; anterior premolars much reduced, single-rooted, knob-like, fourth premolar enlarged, secant with scalloped edge and obscure grooving toward apex of the crown. No diastemata. Molars fundamentally of tarsioid type but considerably specialized in various respects; Me the smallest of the series; trigonid of Mi converted into an antero- posterior crest; protoconids of Me and especially of M3 lower and smaller than metaconids; molar heels large and deeply basined. Jaw short, rather deep anteriorly, mental foramen below P»-3. ; Speciric CHARACTERS:— Po—-M3=8.5 mm.; Mj-3=5. mm. Represented in the collection by the type only. This form cannot be definitely assigned to any family or order; it may be a primate, a menotyphlan insectivore, or neither. CREODONTA Triisodontide Eoconodon, new genus TypE:—Sarcothraustes corypheus Cope, 1888, =Triisodon heilprinianus Cope, 1882. From the Paleocene (Puerco formation) of the San Juan basin, New Mexico. A more careful restudy of the type of the genus Tritsodon, T. quivirensis Cope, in connection with additional and better preserved material of the Triisodontide, shows that it is not congeneric with T. heilprinianus and other species of the Puerco fauna, but that it is con- generic with Sarcothraustes (type S. antiquus) of the Torrejon fauna, which it antedates as a generic name. Moreover, the records of its locality and associated material, with the field data obtained by Mr. Granger in 1913 as to the exposure of the Puerco and Torrejon formation, confirm the suspicion that it is a Torrejon and not a Puerco species. The 1xaprés =fruit; 54r7ns =eater, bloodsucker; in reference to the supposed habits indicated by the character of the teeth, especially the enlarged cutting premolar. The species name refers to the vertical grooving of the tooth, which shows some analogy to the Plagiaulacide and the modern rat kangaroos. 1921] NEW GENERA OF PALEOCENE MAMMALS 7 Puerco genus represented by 7’. heilprinianus (syn. T. biculminatus, ‘Sarcothraustes’ corypheus, etc.), requires a new generic name. The type of T. heilprinianus is very incomplete and its exact provenience unrecorded, so that it appears better to base the genus upon S. corypheus founded upon an excellent type known to have come from the true Puerco, but considered at present as a synonym of 7’. heilprinianus. This genus is nearly related and appears to be directly ancestral to Triisodon of the Torrejon, the type of which genus is 7. quivirensis Cope, 1881. The distinguishing characters separating Eoconodon from Microclenodon and Triisodon, the other two genera of the Triisodonti- dz, are as follows: trigonids low, metaconids and protoconids slightly connate, subequal, paraconid strong, heel large, basined. Oxyclenide Mixoclenus! encinensis, new genus and species Type.—No. 16601, upper and lower jaws with P°-M3, Po-M; left, and roots of M** right, preserved. From the lower level of the Torrejon formation, east fork of Torrejon arroyo, San Juan basin, New Mexico. Exped. 1913. GENERIC CHARACTERS.—The upper molars resemble those of Chriacus, but are wider transversely, and more triangular, external angles more prominent, hypocone less so, M* much reduced, transverse. Premolars with blunt-pointed, somewhat in- flated crowns, more as in Mioclenide, and the more primitive Anisonchine. Lower canine small, partly premolariform. Jaw elongate, shallow, condyle not transverse. Speciric CHARAcTERS.—C-—M3 =26 mm.; Mi-3=10 mm. In addition to the type there is a paratype, No. 17074, a lower jaw with nearly complete dentition, and two other lower jaw fragments. All four specimens are from the lower level of the Torrejon formation. The lower dentition of this genus is very much like that of Cori- phagus Douglass of the Fort Union. Until the upper dentition of Coriphagus is discovered it appears better to hold Mizxoclenus provi- sionally distinct. lyuEos =mingled; (Mio) clenus, (Oxy) Clenus, etc.; in reference to the synthetic character of the dentition. a) AMERICAN MUSEUM NOVITATES No. 14 STEHLINIUS, A NEW EOCENE INSECTIVORE By W. | D. MatrHew Issued September 7, 1921 By OrpER OF THE TRUSTEES OF THE AMERICAN MUSEUM OF NATURAL HISTORY New Yor«k City }. ae AMERICAN MUSEUM NOVITATES Number 14 Sept. 7, 1921 56.9,338(1181:79.2) STEHLINIUS, A NEW EOCENE INSECTIVORE By W. D. MatrHew In 1895 Mr. O. A. Peterson, of The American Museum of Natural History, obtained from the Uinta basin of Utah a block of shale showing numerous fragments of small vertebrates, among which the most interest- ing was an incomplete and badly crushed skull and jaws of a small mammal which Dr. Wortman immediately recognized as a ‘‘new insec- tivore,”’ but which has never been described. I prepared this specimen under the binoculars several years ago but withheld description, hoping to have opportunity to dissect out and reconstruct the displaced frag- ments of the skull. I have concluded, however, that it involved too much risk to be advisable. Miss Erna Kohlhaase worked over the large block from which it had been removed, hoping to find skeleton parts or other fragments, but, although she found a number of small rodent jaws and skeleton bones of small mammals and reptiles, there was nothing that could be confidently referred to the little insectivore. Its singular and highly specialized dentition made its affinities wholly obscure, and it was only in recent years that the researches of Stehlin upon the Plesiadapide and the discovery of several genera of this family in the Eocene and Paleocene of this country cast some light upon its probable affinities. It now appears to be an extremely specialized plesiadapid, possibly related to Necrosorex Filhol. The reduction of the cheek teeth is carried further even than in Apatemys but, unlike that genus, a large and very remarkable cutting premolar is retained in the lower jaw. This tooth, long and knife-like, has no anterior root, the base of the crown resting upon the large front tooth; the posterior root is normal. This _ singular construction is explicable as due to the re-enlargement of a tooth like that of Apatemys, which is small, knife-like and single-rooted; or else to the progressive degeneration of the anterior root due to its being crowded out through the progressive enlargement of the front tooth. iNeither explanation appears entirely satisfactory; the former is in accord with what is actually known of the phyletic record both in this family and in the parallel case of the Plagiaulacide; but will be sternly rejected by certain advocates of ‘‘irreversability in evolution’’ who attach a very different meaning to Dollo’s Law from that explained by its distinguished proponent. The mechanics of the second interpretation appear to be unsound; aslight deepening of the jaw or backward migration of the anterior root would obviate any interference from the root of the front tooth, and the anterior root of a large knife-like tooth of this description would function so importantly in keeping it firm and true that it would be wholly unlikely to degenerate and disappear if there were any way of avoiding such a loss. Once lost, of course, and the tooth dependent for its support solely upon the posterior root and such bracing as it might obtain from its proximity to the socket and root of the large tooth in front, one : ean understand that a re-enlargement might fail to develop a new anterior root through fission of the posterior root, but instead permit it simply to rest against this anterior brace. In any event, the mechanics of this construction is very remarkable. 2 AMERICAN MUSEUM NOVITATES [No. 14 Stehlinius uintensis, new genus and species Typr.—No. 1903. A right lower jaw, with part of the skull from the Upper Eocene (Uinta) of White River, Utah. Exped. 1895. Generic CHarAcTeRS.—Dentition <°+3. Incisors soricoid, greatly enlarged and root and crown much elongated, trihedral in cross-section, the crown of the upper incisor unknown, the lower incisor curving upward towards tip, the wearing surface on the posterior face obliquely concave, the enamel confined to the anterior face of the tooth. Third upper premolar indicated by a single rather large, oval alveolus, the form of the crown unknown. Fourth upper premolar small, trenchant, pointed. ‘ { 1 ' 1 ' \ \ Fig. 1. Stehlinius uintensis, skull and lower jaw, side view, type specimen, Upper Eocene, Utah. Skull reconstructed from the crushed original, lower jaw reversed: Three times natural size. Upper molars brachyodont, trihedral, scalene to a marked degree, paracone and meta- cone prominent, rounded, subequal, parastyle and metastyle well developed, but no mesostyle; protocone prominent, rounded, posterior wing absent, anterior wing extended into a crest meeting the paracone. Lower premolar enlarged, the crown compressed and elongated, with the anterior part raised into a knife-edge and the posterior part with a strong heel-cusp. The posterior root is normal, the anterior root absent, the anterior part of the crown resting upon the incisor. Lower molars with oblique crested trigonids and large basin talonids, the paraconids being low and not a 1921] A NEW EOCENE INSECTIVORE 3 prominent. Lower jaw of moderate depth, with broad coronoid process, condyle but slightly expanded transversely, angle broad, not inflected, ending posteriorly in a stout hook-like process. Reconstruction of the Skull The anterior part of the skull is preserved, but so broken and dis- torted that its reconstruction as shown in the accompanying figure is partly conjectural, and therefore not included in the generic diagnosis. fed Amer Mus. No.1903 Fig. 2. Stehlinius uintensis, type specimen. A, reconstructed skull, top view; B, palate partly reconstructed. Three times natural size. The reconstruction was drawn by Mr. Erwin Christman and is the final ’ resultant of a series of critical studies and attempts at reconstructing the type skull by Mr. Christman, Dr. Gregory and the author; but it re- mains a tentative and not a positive resultant. The width and character 4 AMERICAN MUSEUM NOVITATES No. 14° of the nasals and premaxille is based upon somewhat doubtful identifica- tion of the displaced fragments. The sutures, which are mostly well shown and afford the best evidence for the identity of these fragments, are indicated wherever we regard them as positively identifiable. The most remarkable features of the skull are the great expansion of the ascending ala of the premaxilla, and the long and wide posterior overlap of the nasal on the frontal bone. This is clearly seen to be a superficial overlap, the frontal extending beneath the nasal plate as far forward as the preorbital line, and showing at the surface in a narrow thread along the median suture. In the marsupials the nasals are greatly extended and expanded posteriorly, but there is little or no overlap on the frontals except laterally in some forms (Didelphys, etc.). The postorbital process of the frontals is indicated only by a slight rugosity, as in most Insectivora. The postorbital crests behind it are obscure, and the constriction not marked. The skull is broken off at about the line between parietals and frontals, but this suture is not certainly recognizable. Owing to the broad ascending premaxillary plates, the width of the muzzle is considerable, but the body of the premaxilla is narrow, the incisors set close together, and above and partly in front of the incisor is a process and crest which may have been extended upward in a bony septum between the anterior nares, but is broken off in the specimen. The palate is narrow at the incisors, but widens rapidly backward, with the maxillo-premaxillary suture crossing it in the middle of a moderately long diastema between the incisor and p*. ‘There is some doubtful indication of another small tooth in this diastema. The posterior border of the palate is somewhat doubtfully recognized just back of m*. Affinities The characters of the teeth place this genus as a specialized member of the Plesiadapide. It may be nearly related to Necrosorea Filhol, of the French Phosphorites, although clearly not identical (Necrosorex: has one too many alveoli, even if one ignores the differences in its ms as due to careless drawing); and, if so, it confirms Dr. Stehlin’s transfer of the Phosphorite genus from the Soricidse, where Filhol very naturally placed it, to the neighborhood of the Plesiadapide; thus affording another instance of the insight of this distinguished paleontologist in recognizing the true affinities of so fragmentary and deceptive atype. It has seemed appropriate on this account that the Uinta genus should be named in Dr. Stehlin’s honor. i nt a ih ae 1921] A NEW EOCENE INSECTIVORE 5 The skull characters ‘are not very close to those of Nothodectes, the only genus of this family in which anything of the skull has been described. Although the skull is badly crushed in Nothodectes and difficult to interpret, I cannot find in it any evidence for the back- ward extension and expansion of the nasals or the relatively enormous ascending plate of the premaxilla that characterize Stehlinius if we have correctly reconstructed this part of the skull. These are marsupial, and especially diprotodont marsupial, characters; but, as noted above, the nasal expansion in this genus is a superficial overlap, quite unlike the marsupial conditions, and probably a secondary specialization from the normal primitive insectivore type; the expanded premaxillary plate is presumably also secondary and indicates parallel adaptation, not rela- tionship to the diprotodont marsupials. Marsupial relationship is, in fact, wholly excluded by the dentition, characters of the jaw, etc. The new genus is referred to the Plesiadapide chiefly upon the evi- dence of the teeth, and without prejudice to possible claims to relation- ship with Mizodectes or Microsyops, which at present are assigned to distinct families. The ordinal position of this whole assemblage of genera is very doubtful. Stehlin regards the Plesiadapide as chiromyoid primates; Gregory and Matthew consider them as tupaioid insectivores (Meno- typhla); but in any event they stand near the boundary line between the two orders. The present genus has no especial suggestion of primate about it; but that is of little significance. ye ee 2h Le ee dele fs = ee aS © tel CA Se Te ee See OL ty NS Sean: | iy att e PEG ps = ca ome a5 be - of iy - : a a ak hn PR ge ee a , . Ce te ee > e ‘ % ¥ os . 2 eat AMERICAN MUSEUM NOVITATES eontee No. 15. 3 ° — Be. "2 -_ ‘A NEW NAME FOR A SUBSPECIES OF UTA: _ STANSBURIANA BAIRD AND GIRARD _ By Karu Parrerson Scumipt ~ eee s : Issued September 8, 1921 ) eS Ses By OrperR OF THE TRUSTEES OF THE AMERICAN MUSEUM OF NATURAL HISTORY New Yorx City Se TROP hp Pe Sr Leap pa! Mh Seay Oi Ma Se Pye 8 . Moat ee «9 \* a Us Pe Sar >, lea, e Fie m BAe oe f : / AMERICAN MUSEUM NOVITATES Number 15 | Sept. 8, 1921 59.81,10 (79.9.) A NEW NAME FOR A SUBSPECIES OF UTA STANSBURIANA BAIRD AND GIRARD By Kart Patrerson ScHMIDT In the ‘Check List of North American Amphibians and Reptiles’ (1917), Stejneger and Barbour have followed Richardson (1915, Proce. U.S. Nat. Mus., XLVIII, p. 412, ff.) in his revision of Uta stansburiana. Richardson divides the species into a northern subspecies, Uta stansburiana stansburiana (Baird and Girard), chiefly in the Great Basin; a southern subspecies, U. s. elegans (Yarrow), ranging from western Texas to Lower California; and a Pacific coast form, U. s. hesperis Richardson, of the coastal region of southern California and northern Lower California (and the San Joaquin Valley). To the second subspecies Richardson applied the name elegans of Yarrow (1882, Proc. U.S. Nat. Mus., V, p. 442), with the type locality La Paz, Lower California. Richardson, however, had only four speci- mens of Uta elegans from the Cape Region of Lower California. The Lower Californian collections of the Albatross Expedition made in 1911, contain an excellent series of Uta elegans from Lower California, numbering twenty-four specimens (seven from the vicinity of La Paz); in addition, through the courtesy of the U. 8. National Museum, I have had an additional six specimens from the Island of Espiritu Santu, opposite La Paz, for examination. The Uta stansburianas from Texas, New Mexico, and Arizona in The American Museum of Natural History agree excellently with Richardson’s definition of U. s. elegans. The Lower Californian speci- mens, however, are at once distinguished in the greater length of the hind leg, which ranges from .74 to .85 of the length from snout to anus in eighteen male specimens, averaging .80, while in twenty-four male Arizonan and New Mexican specimens examined by me, the range is .65 to .79, average .71, and the average of Richardson’s series of twenty- three males is .74. I, therefore, regard the form in the Cape Region of Lower Cali- fornia as specifically distinct, and restrict the name elegans of Yarrow 2°, AMERICAN MUSEUM NOVITATES — , Th vid AMERICAN MUSEUM NOVITATES Sats “Nos £7 NOTES ON A NEW OX-PECKER AND OTHER LITTLE-KNOWN BIRDS OF THE CONGO By James P. CHAPIN 2 4 Issued September 16, 1921 By OrpmR OF THE TRUSTEES OF THE AMERICAN MUSEUM OF NATURAL HISTORY -New Yorx« City | AMERICAN MUSEUM NOVITATES Number 17 . September 16, 1921 59.82(67.5) NOTES ON A NEW OX-PECKER AND OTHER LITTLE-KNOWN BIRDS OF THE CONGO! By James P. CHAPIN In the working up of our collection of Congo birds there appear con- tinually points of interest regarding the distribution and relationships of the rarer birds of this region, which seem to merit immediate publica- tion and a little more space than can be allotted to them in the final report. For this reason, we offer the following remarks on species re- cently subjected to special investigation. A NEW BUPHAGUS FROM THE LOWER CONGO The Tick-birds or Ox-peckers of Africa have been universally con- ceded to belong to two species, both of them widely distributed over the eastern and southern parts of Africa and both extending to Senegal in the northwest. Yet, they invariably shun the forested regions, especially of the Congo basin; and, while Professor Reichenow? states that Buphagus africanus is found locally in the western forest area, he gives in his great work only one such record from Lower Guinea: Gaboon (Marche and Compiégne). In his list of birds known from the Cameroon® we find no mention of either species. This is entirely in accord with our experience during the American Museurn Congo Expedition. Nowhere in the forested districts did we ever see or hear of an Ox-pecker, not even where cattle or horses were being kept, as at Stanleyville. But there are parts.of the Gaboon which are certainly not to be reckoned as forest, since numbers of savanna birds extend their range northward from the Lower Congo along the west coast, and one of these must be a Buphagus. . At Faradje, Upper Uele District, where in 1911 some 700 head of cattle were living on the Government farm and the European traders and administrative agents possessed a few horses and mules, no Ox-peck- ers ever came to visit the domestic animals, although this was well to the north of the forest border. With the big game of the region, however, and even farther south near the Kibali River, there were frequently 1Scientific Results of the American Museum Congo Expedition. Ornithology No. 6. 21903, ‘ Vogel Afrikas,’ II, p. 666. 31911, Mitteilungen aus dem Zool. Mus. Berlin, V, p. 251. 1 2 AMERICAN MUSEUM NOVITATES [No. 17 Tick-birds of the so-called yellow-billed species, whose bill nevertheless has the terminal half bright red. They were found in rather small numbers in attendance upon the white rhinoceros, buffalo, giant eland, and occasionally even the wart-hog. Their well-known habits need no mention here; but the facts that they never approached human habita- tions and were never observed about the cattle are rather surprising. _ Along the eastern border of the Belgian Congo, and especially in the Katanga, Ox-peckers of both the yellow- and red-billed species must doubtless occur; yet the only other place where any Buphagus was observed by us was at Zambi, on the Lower Congo. There Mr. Van Saceghem, a Belgian veterinary, kindly procured two specimens for me in January 1915; and Mr. Lang collected three more in June and July of the same year. They were seen commonly about the domestic cattle, and yet only a few miles away at Boma I sought in vain for them with the herds. I was even told that they did not show themselves on the Island of Mateba, where cattle raising is the principal industry. Accord- ing to’ Mr. Drousie, Directeur de |’ Agriculture at Boma, they were not. seen at Zambi previous to 1908, appearing first in company with a herd installed at some little distance north of the river; and then, when these were brought back to the station, the birds came too. - Both Mr. Lang and I noticed at once that these Ox-peckers were of a much darker color, especially on the rump, than those of the Uele; and later I found that they agreed more or less in coloration with Buphagus erythrorhynchus, even to the dark shade of the rectrices. Yet the bill, instead of being entirely red, had been bicolored exactly as in B. africanus. Comparison with specimens made it clear at once that erythrorhynchus had a bill of quite another shape, yet this difference between the two well-known species, while alluded to in Stark and Sclater’s ‘Birds of South Africa,’ is generally disregarded in favor of color distinctions that are not at all evident in the beaks of old dried skins. The bill of Buphagus africanus is much heavier than that of B. erythrorhynchus, especially on account of the pronounced widening of the basal half of the mandible, the sheath of which, in many adult specimens, is even produced inwardly below so as to encroach upon the feathering of the chin. This character was enough to place our darker birds from Zambi unquestionably in the africanus group, and, furthermore, they lacked all trace of the widened yellow eyelid of erythrorhynchus. Were the species of this curious group of birds more numerous, it would doubtless be fitting to divide them in two genera, and such well- Fig. 1. Beaks of the three species of Ox-peckers, as seen from the side and beneath. A, Buphagus erythrorhynchus; b, B. africanus; c, B. langi. Natural size. 4 AMERICAN MUSEUM NOVITATES [No. 17 \ marked characters would be amply sufficient for the purpose. As matters stand, I propose the subgeneric term Buphagoides to distinguish the species erythrorhynchus. An examination of the specimens of Buphagus africanus in the American Museum, the Academy of Natural Sciences of Philadelphia, and the Museum of Comparative Zoology (eleven specimens from widely - different parts of the continent) fully confirms the color differences between the common Yellow-billed Ox-pecker and that of the Lower Congo. Not only does the latter represent a hitherto undescribed form, but, since there is no evidence whatsoever of intergradation between them, I consider it as specifically distinct from Buphagus africanus and restricted in all probability to the Lower Congo and the adjoining part of the Gaboon. It may well be that the record of Marche and Compiégne ~ refers to this dark form. I propose to name it in honor of Mr. Herbert Lang, leader of the American Museum Congo Expedition, with whom I have worked during five pleasant years in Africa. Buphagus langi, new species Speciric CHaracters.—Related to B. africanus, but smaller and more darkly colored, especially on the breast and on the rump, the latter being grayish instead of rich yellowish buff. No rufous on the rectrices. Type.— 2 ad.; A. M. N. H. No. 163005; Zambi, Lower Congo, January 16, 1915. DESCRIPTION OF ADULT FEMALE (Type).—Whole head and throat dull dark brown (“‘fuscous” of Ridgway); back, wing-coverts and secondaries similar; the primaries and their greater coverts fuscous black; under wing-coverts fuscous. Lower rump and upper tail-coverts ‘‘grayish olive’’; rectrices darker, like the back, both on inner and outer webs. Below, the dark brown of the foreneck shades gradually to a dull cinnamon-buff on the lower breast, abdomen, and under tail-coverts, while the flanks are of a warmer ochraceous buff, tinged with cinnamon. Iris yellow; base of bill bright yellow, its distal half scarlet; feet blackish. MEASUREMENTS OF TyPE.—Wing, 107; tail, 74; culmen, 15; tarsus, 21. Of the four remaining specimens, one is an immature male, one an immature female, and the two others have bills like adults, but were not sexed. One of these from its rather large size is presumed to be an adult male, yet it is slightly darker than the type. The immature male has the whole bill still dark brown, and exhibits an ashy wash over the head, back, and chest, which I believe to be char- — acteristic of the first plumage. Measurements of Five Specimens of Buphagus langi (both sexes).— Wing, 106.5-113 (average, 109.1); tail, 74-81 (77.9); culmen, 13.5 -15 (14.2); tarsus, 20.5-21.5 (21). The dimensions of B. africanus =—-_ 1921] NOTES ON BIRDS OF THE CONGO 5 are given by Reichenow (‘Végel Afrikas,’ II, p. 666) as: wing, 120; tail, 90-105; bill, 16-18; tarsus, 21-23. The wing does vary, I find, from 117 to 124 mm., but Buphagus africanus does seem to be a decided- ly larger bird. — THE NECTARINIA OF THE BANGALA COUNTRY This long-tailed genus of Sunbirds is especially characteristic of the plains regions of Africa and, until a few years ago, seemed to be without a single representative in forested western Africa. Then, in April 1910, van Oort! described both male and female of a new species, Nectarinia congensis, which had been sent in alcohol from Boma on the Lower Congo by A. Greshoff to the Zoological Laboratory in Utrecht, and there had remained unnoticed for twenty-one years. No further reference to the species has appeared since its original description, as a result, [ believe, of the true range of this fine Sunbird along a river where few ornithologists have tarried to make collections. Some nine months before van Oort’s description appeared, we were making the ascent by steamer of the Upper Congo River, profiting by every stop to jump ashore and secure specimens, mostly of birds. The Sunbirds in particular attracted our attention, but no long-tailed species was noticed at Boma, Matadi, or Leopoldville, nor indeed until we had reached Bumba, on July 29. At Barumbu, two days later, we collected an adult male specimen, and thereafter saw no more of the species, even during our stay at Stanleyville. It may seem strange that a Sunbird characteristic of the forested course of the Congo River should not occur on the Upper Aruwimi or the Ituri, yet such appears to be the case. Stranger still is the fact that Nectarinia congensis was not found by us at even Stanleyville during October and November, 1914. I determined therefore to keep a special watch for it on the way down the Congo and, at the end of the very first day’s journey, at Isangi, the acquaintance was renewed. About the government station there some of the roads are lined with the beautiful tree Poinciana regia, known as the “Flamboyant” and introduced, I am told, from Madagascar. ‘To its gaudy red flowers, now open in great numbers, there came a few of the Sunbirds I sought, and a male and female were secured. The next day we stopped for an hour at Barumbu, and here on the same kind of flowering tree I again watched a few of these Sunbirds. On other native trees they were observed later near Lié, near Coquil- 11910, Orn. Monatsb., p. 54. 6 AMERICAN MUSEUM NOVITATES [No. 17 hatville, and at Irebu (December 17), usually near or over the river- bank. All our specimens, with the exception of one not quite fully adult, were found to be in breeding condition; this was even the case with the male taken in July. Near Lié a typical Sunbird nest was hanging from a bush, some six feet above the water, and a male of Nectarinia congensis sitting by it seemed to prove the ownership, though, because of the flooded condi- tion of the stream, it could not be reached during our very short stop. Below Irebu this species of Sunbird disappeared and, though I collected subsequently for several weeks in the neighborhood of Boma, it was never seen there either. I cannot help feeling that Greshoff’s _ specimens really came from much farther up the river. Two days below Irebu by steamer the Congo emerges from the equatorial forest into the southern savanna, with a very distinct fauna; and what forest one sees along the Congo near its mouth is mainly a very heavy growth of man- grove. If our Sunbird inhabited this, we may take it for granted that it would long ago have been collected on the Gaboon coast. So far as my observations go, Nectarinia congensis inhabits the banks of the Upper Congo, from Irebu, near the entrance to Lake Tumba, up to Isangi, at the mouth of the Lomami River. The distance is about 500 miles, but I suspect that the birds keep very much to the banks of the larger streams in this region and never go beyond the limits of the equatorial forest. belt. : Van Oort’s description is very good, especially in view of the long sojourn of his specimens in aleohol. All we need add to it is that the burnished green chest of the male has a bluer posterior edge, sometimes even violet, and that the longer upper tail-coverts are of a like shade. In the case of the female, the upper side of the tail shows a faint green gloss, and the foreneck is heavily spotted with dull blackish. With regard to measurements, those of the eight males in our col- lection are: wing, 63-66 (average, 64.5); middle rectrices, 110—-125.5 (117.3), second longest pair, 49-52.5 (50.7); exposed culmen, 19-20 (19.3); metatarsus, 16-16.5 (16.1). The two females have smaller dimensions: wing, 56, 59; middle rectrices, 44.5, 45; second pair, 41, 41.5; exposed culmen, 16, 18.5; metatarsus, 15.5. NEOLESTES CABANIS AND ITS ALLIES At the confluence of the Kasai and Congo Rivers, on December 19, 1914, I made the acquaintance of a bird subsequently identified as Neolestes torquatus Cabanis, collecting three specimens. Having no 1921] NOTES ON BIRDS OF THE CONGO 7 means of learning its name at the time, I attempted at least to determine the family which it represented and, from the general form of body, limbs, and other details, I soon decided that it belonged with certainty to the Pycnonotide, in spite of the rather unusual color pattern, largely green above, white and gray below, with crown and nape ashy, and a broad black line extending from the lores to behind the eye and down across the breast. The bill reminded me of that of Pycnonotus, but was wider and more arched; the feet did not belie such a relationship, the metatarsi being short and scutellate. The sexes were alike in color. I felt the more confident because all three of my specimens, shot among the bushes in upland savannah, had been eating small fruits; and their — voice was a sort of twitter that suggested a Bulbul. Fig. 2 Fig. 3 Fig. 2. Head of Neolestes torquatus, adult male. Natural size. Fig. 3. Head of Neolestes torquatus, from above. Natural size. Fig. 4. Left foot of Neolestes torquatus. Natural size. Later in the same day, at Kunzulu, somewhat farther down the Congo River, a nest of this same bird was found in a bush, four feet from the ground. It was a frail cup of slender grasses and plant stems, hold- ing two eggs, pinkish white indistinctly speckled with darker pinkish and a faint rufous zone about the larger end. Here, I thought, were additional indications of affinity to the Pycnonotide. During the following month, January, near Boma on the Lower Congo, three more individuals were observed, though none was collected, because of their extreme wariness. When I came to the identification of my specimens at the American Museum, no genus of Pyenonotidz could be found in which they seemed to fit, and my search extended to the Laniide before it bore fruit. There 8 . AMERICAN MUSEUM NOVITATES [No. 17 the monotypic genus Neolestes was placed by Reichenow! in the follow- ing association: ‘“‘Chlorophoneus, Pelicinius, Neolestes, Calicalicus, Nicator. . . .”’ Dr. Sharpe,” too, includes Neolestes in the Malaconotine, and Sclater,’ in Shelley’s ‘ Birds of Africa,’ inserts it between Malaconotus and Telophorus, the latter a group of green-backed Shrikes referred by Reichenow to Pelicinius and Chlorophoneus. This association of Neo- lestes is plainly unnatural, and an extract from Cabanis’ original descrip- tion* will show how the error came about. ‘It belongs to none of the known genera, and its insertion in the classification is rather difficult. For the present it had best be considered, like Calicalicus for example, as an aberrant form of the Malacenotine, and be placed in the neigh- borhood of that genus.” The specimens of Calicalicus madagascariensis | have examined in the Philadelphia Academy collection reminded me much more of the Paride than of any other family; they certainly bore no special re- semblance to Neolestes. Calicalicus is placed by some writers in the Vangids, but this question I shall not attempt to settle. Beyond a doubt, Cabanis’ opinion as to Neolestes belonging in the Malaconotine - was based on a superficial resemblance to certain green-and-yellow Shrikes with black breast-bands, of which I have two species (Chloro- phoneus quadricolor and Pelicinius zeylonus) for comparison. The like- ness is extremely faint, even the black band on the side of the head occupying an entirely different position, since in Neolestes it passes com- pletely above the ear-coverts and in these Shrikes below them. Follow- ing up the question of coloration, we may note that the young of Chloro- phoneus dohertyi is barred on the body’—a shrike character—whereas that of Neolestes is not thus marked but closely resembles the adult, as does the young of Pycnonoius. The bill of these Shrikes is typical of the Malaconotine; that of Neolestes approaches the form seen in Pycnonotus (see Figs. 2 and 3) but is relatively even broader at the base, while the nasal operculum is better developed. The lengthened nuchal filoplumes, sometimes said to characterize the Pycnonotidex, are almost completely lacking in our specimens of Neolestes, but this is of slight importance, for they are found in Malaconotus and even among some Ploceine as well developed as in many Bulbuls. 11914, ‘Die Végel,’ II, p. 289. 21903, ‘Hand-List of Birds,’ IV, p. 299. 31912, ‘Birds of Africa,’ V, = 2, p. 405. 41875, Journal fiir Ornithologie, p. 237. 51902, Rothschild, Nov. Zool., Pl. rx. 1921] NOTES ON BIRDS OF THE CONGO ! 9 The wing helps but little in our decision. Like both Pycnonotidze and Malaconotine, Neolestes has it ten-primaried and rounded. In Pycnonotus the fifth and sixth primaries (counting from inside) are long- est, in Neolestes the sixth, in Pelecinius sixth and seventh. In the adult of Neolestes the tenth primary is relatively shorter than in Pycnonotus tricolor, but this in turn has it shorter than in Pelicinius, The foot of Neolestes is not at all shrike-like. The metatarsus is too short, and its scutellation not at all like that of Pelicinius and most Laniide, but much more similar to that of the Pycnonotide, though not so nearly ‘‘ booted”’ as in Pycnonotus. My first impressions in the field are thus confirmed. Although my judgment is based on external characters, for no anatomical material was preserved, Nevlestes seems undoubtedly a pycnonotid. I .believe, indeed, that its affinities are closer to Pycnonotus than to any other African genus of the family. Now I find that all this has been anticipated by Dr. Gadow, as long ago as 1883, in Vol. VIII of the ‘Catalogue of Birds in the British Museurn.’ Although retaining Neolestes in the Laniide, probably only for convenience, he clearly states in the introduction to that family that Neolestes and Calicalicus are so aberrant as to form links with the Pyenonotinz. Furthermore. he appends to the Key to the Genera of Malaconotine (p. 103) a footnote which subsequent writers seem to have ignored almost completely, though it is well worth quoting here, since my present remarks are simply a confirmation of it. Here may be mentioned Neolestes . . . which has been placed by Cabanis near Calicalicus; it does not appear to be a Bush-Shrike, but to be allied to the Bulbuls or Pycnonotine. Bill not laterally compressed, but considerably broader than high; genys decidedly curved downwards, and not upwards as in all Laniide; nostrils with a well developed coriaceous operculum; strong rictal bristles; tail rounded and slightly shorter than wings. Such a critical examination of one genus of supposed Shrikes might well prompt one to ask ‘‘What of Nicater?’”’ I well remember meeting for the first time in the field two species of this African genus, N. chloris and N. vireo, and my reluctance to place therm among the Laniide has never been quite overcome. Nicator, of course, bears little resemblance to Neolestes and is undoubtedly more shrike-like, with the bill straighter, narrower, and distinctly hooked, the metatarsus long, the toes short. Yet here most of its shrike characters seem to end. I do not feel that the spotting of its wings and tail necessarily indicates a relationship with Malaconotus, but was at first tempted to associate it rather with Bleda among the Pyecnonotidze, which also has a straight compressed 10 AMERICAN MUSEUM NOVITATES [No. 17 bill with more or less of a hook. ‘The feet of Nicator and Bleda are similar in proportions, though the metatarsus is scutellate in the first, practically booted in the second. Both genera have a peculiar gap in the feather tract of the back of the neck, and outer primaries of similar proportions. At all events, I feel that Nicator is as near to the Pyenonotide as to any member of the Malaconotine, though perhaps belonging in neither of these groups. Attention may here be called to one peculiarity of Fig. 5. Nestling of Nicator chloris, to show bare face and neck. Natural size. Nicator which is of interest. The true rictal bristles, in contrast to those of Bleda, are rather poorly developed and they are replaced functionally by a different group of feathers, situated much closer to the eye, the shafts of which are stiffened and prolonged, the barbs being greatly reduced. There is no approach to this in Bleda, and but little resemblance in Malaconotus. 1921] NOTES ON BIRDS OF THE CONGO 11 The feathering of the nestling Nicator is very peculiar and quite different from that of any young Shrike (Fig. 5). At the time when the wings are half grown and the body already well feathered, the feathers of throat and cheeks, as well as those around the eyes, ears, and base of bill, have still failed to put in appearance and there is a broad median apterium running the whole length of the crown, of which no trace can be seen in the adult. The only feathers on the head are thus in two lateral lines on the crown, which join on the nape but are isolated from the spinal tract. The legs are entirely bare of feathers up to the lumbar tract and the appearance of such young birds is unique, the fluffy feathers of the upper breast forming a sort of ruff. I do not think that this is paralleled in Bleda, of which I have, however, seen no specimen quite young enough for comparison. The juvenal plumage of Nicator is similar in color to that of the adult, whereas that of Bleda syndactyla and eximia is strikingly different, for the upperparts, excepting the remiges and rectrices, are mostly of a peculiar rufous or maroon, the underparts whiter; and this first plumage, which recalls that of Turdinus (Timeliide), is molted even before the tail has attained its full length. _ THE JUVENAL DRESS OF SIGMODUS RUFIVENTRIS MENTALIS In glancing over our series of this bird from the Ituri and Uele, it is evident at onee that immature examples are very differently colored about the head from adults, which have smooth bluish gray feathering on the crown and cheeks, set off sharply from the black collar encircling the whole neck. Instead, immature birds with black bills are found te have the lores and rictal bristles black, and a dark band extending from behind the eye to the nape, while the black collar does not encircle the foreneck. A still younger individual, of female sex (A. M. N. H. No. 161114) with wings and tail fully grown, is seen to have a whitish collar extending almost entirely around the hind neck, and the cheeks and throat nearly white. The throat is separated from the purer white chest-patch by a narrow band of rufous crossing the foreneck. Even this specimen, how- ever, has already begun to molt out of the juvenal plumage, for all its secondary coverts are black, and only a white spot at the tip of the second alula-quill gives a hint of their color in the first plumage. Such a nestling as that shown in Figure 6 would be a puzzling bird to identify if we did not have sorne of the transitional stages. Most of the change to the adult plumage seems, nevertheless, to take place in a 12 AMERICAN MUSEUM NOVITATES [No. 17 single post-juvenal molt, which begins with the wing-coverts and is retarded longest about the head. Before the plumage of the head has been entirely renewed the molt of the remiges and rectrices is under way. The young bird we figure (A. M. N. H. No. 161117, o) shows one peculiarity in which it is most unlike the Shrikes, though this genus and Prionops have very often been included in the Laniide. On the back of its crown there are two large bare patches of skin, separated by a narrow Fig. 6. Nestling of Sigmodus rufiventris mentalis, with white wing-coverts and parietal areas of naked skin. Colors of pees largely blackish and white, as in the figure; but throat tinged with cinnamon, and lower breast, belly, and under tail-coverts pale cinnamon-rufous, very much lighter than in the adult. Natural size. median line. Feathers will later grow from this skin, but they are always a little shorter there, a condition which aids in the sharp definition of the blue-gray cap in the adult. What the difference in color-pattern between adult and young may mean we cannot be entirely certain, but it looks like a good case of re- capitulation, an ancestral plumage appearing in the young only. The 1921] NOTES ON BIRDS OF THE CONGO 13 adults of several species of the related genus Prionops have considerable white on the wing-coverts. As- passerine families go, the Prionopide seem to be well marked off from the true Shrikes; but the affinities to the two typical African genera of some of the other forms associated with them, as for instance in Sharpe’s ‘Hand-List,’ seem to me most question- able. THE SECOND KNOWN LOCALITY FOR LECYTHOPLASTES PREUSSI The Cliff Swallow discovered by Preuss at the falls of the Sannaga, near Edea, Cameroon, and named in his honor by Reichenow,! has thus far never been reported elsewhere. It plastered its bottle-shaped nests, wrote Preuss, with flasklike necks pointing obliquely downward, in great. colonies on the vertical clifis, right below the waterfall; and there he caught as many as he liked with a butterfly net. While making a journey from Faradje to Dungu, Upper Uele District, in 1911, I arrived on June 1 at the rest-house overlooking the River Dungu, about midway between the two posts and known as Gangara na Bodjo (= Bodjo’s Hill). Flying about the huts and alighting on the bare ground around them was a flock of a dozen small swallows that looked new to me. I secured three, all adults, and was delighted to find that they resembled small Cliff Swallows, a group that I had not met thus far in the Congo. I recalled that in the preceding February, somewhere along this same part of the road, I had noticed some unusual swallows but thought at the time that they might be Hzrundo puella. I am sure now that they were not. At the time our specimens were taken they must have been breed- ing and were in worn plumage. The sexual organs of one male were noted as “‘much enlarged,” those of the second as somewhat. enlarged, and the ovary of the female slightly so. That they were not simply a wandering flock, far out of their normal range, is likewise indicated by my meeting 2 lone individual, on April 15, 1912, in the same general region, only about twenty-five miles to the south or southeast. To be more exact, it was a four hours’ march west of the village of Gangura, an Azande chief, that this ‘single bird was found, flying about near a strip of woods; but in shooting it I had the ill fortune to mutilate it beyond all usefulness. Though I twice had occasion to visit Gangara na Bodjo again, I never found the swallows there; their great rarity, or extremely local distribution, is attested by the fact that they have never been re- 11898, Orn. Monatsb., p. 115. 14 AMERICAN MUSEUM NOVITATES [No. 17 ported from any other localities than those above-mentioned, which are © separated by a distance of 1300 miles. There can be no doubt of the breeding of Preuss’ Swallow somewhere near the Dungu River, but just where we found it there seem to be no suitable cliffs at all. The only hill of any size I know in that vicinity is at Piagga, one day’s march nearer to Faradje, where a splendid overhanging cliff was found in February 1913 to shelter a nesting colony of Micropus affinis. Yet no Lecythoplastes were observed; and on other higher hills near Aba, Garamba, Gangura’s, Nzoro, and Dungu, only Hirundo puella and Riparia rufiqula were found occupying the cliffs. . In view of the great distance from the type locality, the natural impulse is to look for slight differences in characters, but I can find none whatever. Reichenow’s description fits exactly, even to measurements. Those of our specimens are: wing, o’, 97, 96, 9 96; tail, middle feath- ers, co’, 42, 42.5, 2, 42; outer rectrices, o’, 53, 53.5, 2, 53; exposed culmen, o’, 6.6, 6.2, 2, 6.8; metatarsus, o’, 10.3, 11, 2, 10. The sexes do not differ any more in color than in size; one of our birds lacks the white spots on the outer pair of rectrices, but it is a male. , EASTERN LIMITS OF DISTRIBUTION FOR SOME WEST AFRICAN BIRDS Collections made in recent years in Central Africa have shown re- peatedly how many characteristic West African species extend their range from the Cameroon al! across the Congo forests and even to the Lake Region. The number of such forest birds that have been taken near: Beni, on the eastern border of the Belgian Congo, is surprising; and Dr. V. G. L. van Sonieren has recently made notable additions to the Uganda avifauna of species previously known only from West Africa. A certain number of specimens in our Congo collection, representing West African forms that were not previously known to range so far into the Northeastern Congo or have been perhaps only once recorded from that part of the colony, are worthy of mention here. ‘ Canirallus oculeus ({Temminck] Hartlaub) 4 #,2 9, Gamangui (Ituri); 1 9, Medje (Ituri); 2 #,1 9, Niapu (Bomo- kandi). Podica senegalensis senegalensis (Vieillot) 1 Zim.,1 2, Panga (Aruwimi R.); 1 @ with gray throat, 3 9, Avakubi (Ituri); 1 juv., Niapu (Bomokandi); 1 2 im., Niangara (Uele). Already recorded by Dubois! from Panga. Lampribis rara Rothschild, Hartert and Kleinschmidt 1 2, Avakubi; 1 o, Niapu. 11905, Ann. Mus. Congo, Zool., (4) I, fase. 1, p. 24. ; ‘ t 1921] NOTES ON BIRDS OF THE CONGO 15 Lampribis olivacea olivacea Du Bus 1 &, Avakubi. Tigrornis leucolopha (Jardine) 1 #,1 9, Gamangui; 1 &, Medje; 2 7,1 @ juv., Niapu. Urotriorchis macrourus ({Temminck] Hartlaub) 1 2 im., Avakubi. Dubois has already reported it from Banalia. Astur castanilius (Bonaparte) 1 &, Gamangui; 2 2 im., Medje. The recently described Accipiter beniensis Lénnberg is apparently synonymous. Accipiter sharpei Reichenow 1 &, Bengamisa (R. Lindi); 1 #, 1 ¢@ im., Banalia; 1 2 im., Bafwasende (R. Lindi); 1 @, Avakubi. I consider Aectoiar zenkert and 4: erythropus, which has recently been reported by Sassi from Beni, to be probably the immature stages of A. sharpei and A. hartlaubi. In such a case the name sharpet would be antedated by zenkeri. Hieraaétus africanus (Cassin) 1 9,1 2 juv., Niapu. Dryotriorchis batesi Sharpe 1 o, Stanleyville; 6 #, 3 2, Avakubi; 1 2, Medje; 1 2, Niapu; 1 9, Akenge (Bomokandi). Baza cuculoides (Swainson) 1 f,2 2,2 2 im., Avakubi; 1 2 im., Ngayu (Ituri); 1 &, Rungu (Bomo- kandi). There is one previous record from Semio (N. Uele). Scotopelia bouviert Sharpe 20,1 2,1 2 juv., Niapu; 2 2, Niangara. Bubo poénsis Fraser 2,1 2, Avakubi; 1 o&, Medje. Bubo leucostictus [Temminck] Hartlaub: 1 &, Batama (Distr. Stanley Falls); 1 2,1 2 juv., Medje. Recorded from Popoi (Aruwimi R.) by Dubois. Glaucidium tephronotum Sharpe (=G. pycrafti Bates) 2,1 2, Medje; 1 2, Nala (Bomokandi). Otus holerythrus (Sharpe) 1 2 juv., Medje. Previously known from Banalia. Scoptelus brunneiceps Sharpe 2 o', Avakubi. Meropogon breweri Cassin 1 2, Banalia. Reported from Ubangi R. by Reichenow, and “Province Orientale” by Dubois. Caprimulgus batesi Sharpe Pei eek of juv. 2-9 juv., Medje; 1 9, Avakubi. Bannerman! has reported it ues Poko (Bomokandi). Chetura cassini Sclater 1 #, Bengamisa; 3 #, 1 9, Avakubi; 2 7, 1 2, Ngayu; 1 @, Medje. A single specimen recorded as C. Liiiaiade from Moéra by Sassi, also noted ~ from Aruwimi R. by Reichenow, and from Poko, Uele Distr., by Raemetings: 11919, Bull. Brit. Orn. Cl., XX XIX, p. 96. 16 AMERICAN MUSEUM NOVITATES [No. 17 Centropus anselli Sharpe 1 2, Isangi (mouth of R. Lomami). Verreauxia’ africana (Verreaux) 1 &, Stanleyville; 3 7,3 2,2 #im.,1 9 im., Avakubi. Hirundo nigrita G. R. Gray ‘1. 9, Bengamisa; 1 <, Banalia; 3 &%, 2 9,1 @ juv., Avakubi; 2 9, Gamangui; 2 <”, Bafwabaka (Nepoko R.);1 7,1 9, Rungu;1 2, Nzoro (Upper Kibali R.). Already reported in -‘ Végel’ Afrikas’ from Bafwazabangi, on Ituri R. and from the Aruwimi. Fraseria ocreata (Strickland) - ; 1.3%, Avakubi; 1 &, Ngayu; 1 #,1 9°, Gamangui; 1 2, 2 @ im., Medje. Fraseria cinerascens Hartlaub 1 3, Avakubi. Lobotus oriolinus Bates . 1 &, Medje. Bzopogon clamans (Sjéstedt) 3 0,2 9, Avakubi; 1 3,1 @ im., Ngayu. Camaroptera sipeicilintts (Fraser) 1 &, Avakubi; 1 7, Penge (Ituri); 1 #, Ngayu; 2 a, 19,1 7 im, Medje; 1 2 im., Runes. Chaunonotus sabinei (J. E. Gray) 1 &, Avakubi; 1 @ im., Ngayu. Cinnyris johanne Verreaux 1 &, Dobo (Distr. Bangala); 1 <7, Avelebi Anthreptes aurantium Verreaux 1 9, Stanleyville; 1 %, Panga (Aahewii R. ); 1 &, Bomili (Ituri); 5 &, o0e, Avakeiie 1 7,1 9, Gamangui; 1 &, Gada R. near Niangara (Uele). Ubanal R. Ay Yambuya on lower Aruwimi mentioned in ‘Végel Afrikas.” _ Parmoptila jamesoni (Shelley) 27,1 9,3 @ im., Avakubi; 1 &, Babeyru; 1 #7, 1 9°, Gamangui; 1 9,1 2 im., Medje. Type locality: Yambuya, lower Aruwimi R. : Hypargos dybetoedis (Oustalet) : 1 9, Faradje; 1 &, 2 2, Aba (Upper Uele). Type locality: Kemo, Ubangi R. Estrilda melpoda (Vieillot) 3 9, Stanleyville; 2 7,2 9, Panga. Recorded from Banalia by Dubois. Brachycope chimals (Reichenow) 6 7,1 9, Avakubi. Type locality: Stanley Falls. Extends up the Aruwimi and Ituri risen also down the Congo to Nouvelle Anvers. Recorded by Reiche- now! from Banalia. : : 11910, Vogelf. Mittelafr. Seengebietes, p. 326. > "DESCRIPTIONS OF PROPOSED NEW BIRDS Ly FROM COLOMBIA, ECUADOR, . PERU, ane BRAZIL : aoe PesincM, CHAPMAN ae A os fo SORES : SEP ak 1921 Issued September 22, 1921 ; By Orper or THE TRUSTEES oF - ‘THE AMERICAN MUSEUM OF NATURAL HISTORY ~ New Yorx City AMERICAN MUSEUM NOVITATES Number 18 | September 22, 1921 59.82(8) DESCRIPTIONS OF PROPOSED NEW BIRDS FROM COLOMBIA, ECUADOR, PERU, AND BRAZIL By Frank M. CHapMAN As a result of studies made at the British Museum in parts of May and June of the present year, the author can now propose definite names for a number of birds which, pending comparison with authentic material, had been provisionally identified. There are also here included descriptions of apparently new birds discovered by our recent Anthony-Cherrie Expedition to Ecuador, and of a Leptasthenura from Peru which I am permitted to name by the authorities of the British Museum. I am under deep obligation to these gentlemen, and particularly Dr. Percy R. Lowe, in charge of the Bird Department, for extending to me every facility to prosecute my labors while the guest of their institution. I wish also to thank Mr. Charles Chubb for his invaluable codéperation. Nothocercus fuscipennis, new species Speciric CHaracters.—Throat white, forehead russet, as in N. julius; back and flanks finely vermiculated, as in N. nigricapillus; differing from both julius and nigricapillus in entire absence of markings on wing-quills. Typre.—No. 109,378, Amer. Mus. Nat. Hist.; 9; Andes west of Popayan, Colombia; alt. 10,340 ft.; July 21, 1911; W. B. Richardson. Description oF Type.—Upperparts and wing-coverts between Prout’s and mummy-brown, finely and evenly vermiculated with black, the black markings broader on the rump, upper tail- and greater wing-coverts; anterior half of crown and loral region russet grading into dark clove-brown very finely and faintly marked with russet on the hind head and cervix; ocular and postocular region with the tawny marks more pronounced; tail dark fuscous with a terminal fringe of the color of the back; wing-quills uniform dark fuscous wholly unmarked; throat and chin snowy white sharply defined from the pale sepia foreneck; rest of underparts mainly ochrace- ous tawny, the breast darker, the sides, flanks, tibie and under tail-coverts much like the back and all finely vermiculated with black; feet brownish black; maxilla blackish, mandible Naples-yellow, brownish on its cutting edge and terminally. Wing, 180; tail, 50; tarsus, 61; culmen, 33 mm. Specimens EXAMINED Nothocercus fuscipennis—CotomB1a: Andes west of Popayan, the type; Prov. of Cauca, 1 (in Brit. Mus.). Nothocercus julius julius—CotomsB1a: Bogotd, 3; Laguneta, 1; Almaguer, 1. Ecvapor: Pichincha (Goodfellow), 1 (in Brit. Mus.); ‘Ambato,’ 1. 1 2 AMERICAN MUSEUM NOVITATES [No. 18 Nothocercus julius salvadorii.—The type (in Brit. Mus.) labelled by Verreaux ‘‘Equateur”’ but probably not from Ecuador. Nothocercus nigricapillus—Bouxivia: Locotal, 5800 ft., 1 Q im. Comparison of the single specimen from the Western Andes, provisionally referred to julius,! with a series of that species representing every age from the lately hatched chick to maturity, ‘convinces me of the specific distinctness of the west Colombian bird. Furthermore, a specimen in the British Museum labelled “Prov. of Cauca,” the region whence the supposed new bird comes, shows its characters. Apparently julius always has the upperparts broadly barred with black and the outer web of the secondaries marked with cinnamon at all ages, whereasin — fuscipennis the back is vermiculated, the wing-quills unmarked. It is important to observe that julius julius ranges into Ecuador, a specimen from Pichincha in the British Museum, and one received through Ambato in the American Museum being referable to that species. The Ambato specimen was received in a small collection containing a specimen of Osculatia sapphirina and doubtless came therefore from the Amazonian slopes near Bafios, perhaps thérefore from the same region whence came three females referred by Taczanowski and Berlepsch to julius (Proc. Zool. Soc., 1885, p. 112) with the comment that the upper- parts are darker, more olive and with broader bars, etc., differences which are shown by our Ambato bird and which may be of racial value. Specimen “h” of the ‘Cat. Birds Brit. Mus.’ (X XVII, p. 510) commented upon by Salvadori and later described by Chubb as Notho- cercus julius salvadorii (1914, Bull. B. O. C., XXIII, p. 95) differs from julius in its much more pronounced bars on back and wings, and is certainly not the same as our Ambato bird. It is labelled, evidently by Verreaux, ‘‘Equateur” but quite probably did not come from that country since it does not agree with specimens from either the Pacific or Amazonian side of the Andes. An immature female from Locotal, Prov. Cochabamba, Bolivia, agrees with the description of Nothocercus nigricapillus supposed to have come from Chile. If this identification be correct, it affords us a clue to the range of this species. I append a key to the members of this genus. Throat white. Crown wholly plumbeous.. ............¢.¢e00s0sc0005- N. nigricapillus (Gray). Crown anteriorly russet. Back strongly barred. Secondaries barred only on outer web...N. julius julius Bonaparte. N. julius venezuelensis Cory. LF 11917, Bull. Amer. Mus. Nat. Hist., XXXVI, p. 190. 1921} DESCRIPTIONS OF PROPOSED NEW BIRDS 3 Secondaries barred on both webs........ N. julius salvadorti Chubb. Back finely vermiculated..................... N. fuscipennis Chapman. Throat ochraceous. Throat bright ochraceous. Greater under wing-coverts barred....... N. bonapartei bonapartei (Gray). Greater under wing-coverts not barred. . .N. bonapartei frantzii (Lawrence). Throat pale ochraceous; underparts less rufous. N. bonapartei intercedens Salvadori. Penelope barbata, new species Specrric CHARACTERS.—Similar to Penelope argyrotis Bonap. but much darker throughout, the chin and upper throat feathered, wholly concealing the skin; whit- ish margins to feathers gray instead of silvery white and, both above and below, con- fined to the anterior parts of the body, on the wings appearing only as very slight and inconspicuous markings on the lesser coverts; rump, upper and under tail- coverts and flanks cinnamon-brown instead of Sayal to Mikado-brown. Typre.—No. 156,201, Amer. Mus. Nat. Hist.; 2 ad.; Taraguacocha, Zaruma- Zaraguro trail, Cord. de Chilla, Prov. del Oro, Ecuador; alt. 9750—11,000 ft.; Geo. K. Cherrie. : DescripTIon oF Type.—Crown and anterior parts of the back very dark olive laterally margined with gray, the forehead, superciliary, cheeks and sides of neck largely grayish; whole orbital region to the base of the bill bare, black in dried skin (blue in life ?); center of back and wings externally uniform shining olive-brown un- ~marked; rump and upper tail-coverts bright snuff-brown to warm sepia; tail ex- ternally like the middle back, lateral feathers black, all tipped with light cinnamon- brown decreasing in extent from without inwardly; chin and upper throat covered with black feathers with a slight mixture of gray posteriorly, concealing the skin; lower throat bare, except for a few short hair-like and slightly pinnate feathers; breast blackish, laterally margined with gray; abdomen light cinnamon-brown faintly vermiculated with blackish; lower tail-coverts darker; tibiz olive-brown, the feathers extending to the proximal third of the tarsus; feet reddish, bill black (skin). Wing, 250; tail, 250; tarsus, 59; culmen, 30 mm. : Specimens EXAMINED Penelope barbata—Ecuapor: Taraguacocha, 1, the type; San Lucas, Pacific slope, 2 (Brit. Mus.). Penelope argyrotis VENEZUELA: Merida, 1; Venezuela, 1, type of P. lichten- steini (Brit. Mus.). Cotomsta, 1. The type of this proposed species agrees with the two San Lucas, Ecuador, skins recorded by Mr. Ogilvie-Grant in the ‘Cat. Birds Brit. Mus.,’ XXII, p 502 and believed by him to be probably the young of Penelope argyrotis of Venezuela. Comparison of the three with two Venezuelan and one Colombian example of that species demonstrates, in my opinion, the specific distinctness of the Ecuadorian form. 4 AMERICAN MUSEUM NOVITATES [No. 18 Siptornis wyatti sequatorialis, new subspecies SussPeciric CHARACTERS.—Similar to Siptornis wyatti wyatti but all the dark areas above, including the central tail-feathers, much blacker, the feathers of the back margined with grayish instead of with brownish. Typr.—No. 124,504, Amer. Mus. Nat. Hist.; ad.; Mt. Chimborazo, Ecuador; alt. 13,000 ft.; July 3, 1913; W. B. Richardson. Specimens EXAMINED Siptornis wyatti equatorialis—Ecuapvor: Chimborazo, 3 o, 32 ; Cechce, 1 o. Siptornis wyattt wyatti—CoLomsBiA: Paramo of Pamplona, 1 & (type); Paramo of Chiruqua, 1 <, 1 9 ; Sierra Nevada of Sta. Marta; alt. 10,000-12,000ft.,2 7,19. I have long suspected that the bird recorded as “ Szptornis wyatti”’ from Ecuador probably did not agree with that form but, in the absence of authentic specimens of wyatti, it was not possible to reach a satis- factory conclusion in regard to the status of the Ecuador bird. Compari- son of Ecuador specimens with the type, and apparently only speci- men of wyatti, shows the differences given above and a further comparison with material from the Santa Marta group indicates that the bird from that region is essentially identical with wyatti. That the differences between the Ecuadorian and Colombian birds are not due to any post-mortem change in the color of the plumage is proven by the fact that one of the Ecuador specimens collected by Stolzmann at Cechce, May 18, 1883, is sufficiently old to be comparable with the type of wyatt? collected in 1870, and specimens from Santa Marta collected in 1879 and 1881. On the other hand, specimens of wyatti collected in the Paramo of Chiruqua, Santa Marta, by Carriker in 1914, are comparable both as to condition of plumage and age of skin with our Ecuadorian series of zquatorialis. The Cechce specimen in the British Museum (specimen “ft”? in ‘Cat. Birds Brit. Mus.,’ XV, p. 71) above referred to bears the MS. name “ Synall. paramo sp. n. type de la description pour M. a: but I cannot find that this name was published. Odontophorus parambe canescens, new subspecies SupsPeciric CHARACTERS.—Similar to Odontophorus parambe parambe Roths., but prevailing color of the upperparts grayish rather than rich brownish, the black areas smaller, the markings on wing-coverts and tertials buffy rather than ochraceous; size much larger, the bill longer and notably thicker. Type.—No. 156,205, Amer. Mus. Nat. Hist.; o ad.; Alamor, Prov. Loja, Ecuador; alt. 4500 ft.; October 3, 1920; Geo. K. Cherrie. Specimens EXAMINED Odontophorus parambe canescens.—Ecuapvor: Alamor, 2 o (ine. type). Odontophorus parambe parambe.—Ecuapor: Esmeraldas, 1 &; Naranjo, Prov. Guayas, 2 9 ; near Zaruma, Prov. del Oro, 2 7, 1 9. Cotomsta: Barbacoas, 1 9 ; Baudo, Chocé, 2 o. 1921] DESCRIPTIONS OF PROPOSED NEW BIRDS 5 MEASUREMENTS Depth of Sex Wing Tail Tarsus Culmen Bill at Base Alamor, Ec. fot 153 61 36 20.5 13 “ +4 rot 145 58 38 20 13 Esmeraldas, Ec. ot 130 os 35 18 11.5 Barbacoas, Col. 2 135 49 35 18.5 11.5 Baudo, = roe 130 50 35 18 10.5 Naranjo, Ec. 2 — 134 55 38 19 12 Zaruma, “ rot 144 57 37 20 13 si Be 144 37 19.5 18 Odontophorus parambe occurs in both the Tropical and Subtropical Zones of the Pacific coast of Ecuador and Colombia, from the Peruvian boundary at least to the Chocé. From the last-named region to north- ern Ecuador it shows no apparent geographical variation. Naranjo specimens are. slightly larger while those from near Zaruma exhibit a further approach toward canescens, not only in size but in their grayer color. They are, indeed, so nearly intermediate between the small dark northern race and the large pale southern one that it is difficult to say to which they should be referred. Our series therefore indicates the com- plete intergradation of these two well-marked races. Nyctibius longicaudatus chocoensis, new subspecies Susspeciric CHARACTERS.—Similar to Nyctibius longicaudatus longicaudatus, but general coloration much deeper, the black markings of the upperparts more ex- tensive and more pronounced, the crown largely black, the back and scapulars with sharply defined black shaft-streaks, the brown areas of the upperparts darker, chest- nut rather than ochraceous. — Typre.—No. 111,501, Amer. Mus. Nat. Hist.; o ad.; “testes slightly enlarged”’ ; Névita, Rio San Juan, Choe6é, Colombia; alt. 400 ft.; December 23, 1911; Allen and Miller. SPECIMENS EXAMINED CotomsiA: Névita, 1 @ (the type), 1 9. Braz: 1. Br. Guiana: Bartica Grove, 1 9. Ecuapor: Sarayacu, 3. Comparison of the two specimens of Nyctibius, which, for lack of material, I provisionally referred to longicaudatus in my paper on Colombian birds,! with examples of this species in the British Museum shows, as might be expected, that the bird from the intensely humid Chocé region of Colombia proves to be a well-marked form. The capture 11917, Bull. Amer. Mus. Nat. Hist., XXXVI, p. 272. 6 AMERICAN MUSEUM NOVITATES . |[No. 18 of these specimens extends the known range of Nyctibius longicaudatus west of the Andes to the Colombian-Pacific Fauna. Picumnus parvistriatus, new species Speciric CHARACTERS.—Similar to Picumnus sclateri Tacz., but underparts much less heavily marked with black, the breastbars decidedly narrower than the white ones, the streaks of the abdomen and flanks comparatively obsolete; white crown-spots larger. Typre.—No. 124,368, Amer. Mus. Nat. Hist.; © ad.; Daule, Prov. Guayas, Ecuador; W. B. Richardson. DEscRIPTION OF Type.—Crown black, the feathers with large rounded tips, yellow on the forehead, white on the remainder of the crown; back pale buffy brown, faintly margined with lighter; tail black, the central feathers white on the inner web; outer feathers with a,subapical, diagonal white band; wings fuscous, the inner feathers exteriorly margined with olivaceous; sides of the head white faintly banded with blackish, extending narrowly to the nape; throat and breast banded with black and white, the white bands being pronouncedly broader than the black ones; abdo- men and flanks white, narrowly and obscurely striped with blackish; feet and bill blackish. Wing, 52; tail, 25; tarsus, 12; culmen, 12 mm. DESCRIPTION OF FEMALE.—Similar to male but spots on forehead white instead of yellow. Wing, 51; tail, 24.5; tarsus, 12; culmen, 12 mm. DESCRIPTION OF IMMATURE.—Similar to the adult female but the crown striped instead of spotted. SPECIMENS EXAMINED Picumnus parvistriatus—Ecuapor: Daule1 & (type); Manta, Prov. Manavi, 1 9; Guayaquil, 1 im. Picumnus sclateri—Prrvu: Paletillas, Prov. Piura, 1 &, 2 2. Ecuapor: Alamor, Prov. Loja, 4550 ft., 1 o im.; Portovelo, Prov. del Oro, 2000 ft., 1 2 ; Salvias, Prov. del Oro, 3600 ft., 1 o; Rio Pindo, Prov. del Oro, 1850 ft., 1 o&; Santa Rosa, Prov. del Oro, 1 9°. The form here described is obviously a representative of Picumnus sclateri but the character and extent of its differentiation from that species indicate its specific distinctness. The range of P. sclateri appears to be the Tropical Zone of southwestern Ecuador and extreme north- western Peru. Our eight specimens, which represent the larger part of this area show no racial variation, the most northern being no nearer parvistriatus than the most southern. Similarly, our specimens of the last-named form, which inhabits semiarid Ecuador from Guayaquil at least to Manta and probably Bahia de Caraque, are uniform in color. Thamnophilus zarume, new species Speciric CHARACTERS.—Similar to Thamnophilus radiatus radiatus Vieill., but smaller, the male above and below more narrowly barred,.the bars obsolete on the nape; forehead with more white, lateral crest feathers with white markings; abdomen t + ‘ 1921] DESCRIPTIONS OF PROPOSED NEW BIRDS 7 and flanks buffy and comparatively unbarred; inner webs, of all but one outer and two or three inner wing-quills, with even margins instead of spots; under wing-coverts buffy. Female less clearly rufous above than the female of radiatus, the nuchal region grayish olivaceous, the sides of the head, especially posteriorly, with little or no buffy wash; the underparts much paler. Typre.—No. 129,684, Amer. Mus. Nat. Hist.; @ ad.; Zaruma, Prov. del Oro, Ecuador; September 17, 1913; W. B. Richardson. Description oF Mate.—Cap black, the forehead thickly spotted with white, all but the central feathers of the elongated crown feathers spotted with white on one or both webs; nape blackish, barred or spotted with white, the post-nuchal region grayish, the bars obsolete; back with black and white bars, the former the wider; upper tail-coverts tipped with buffy; tail black with a subterminal white bar, on at least the central feathers, and broken white bars on both webs of all of them; wings black, the outer webs of the feathers with white spots, not reaching to the shaft, the secondaries and tertials tipped with white, the latter marked with white on both webs; the inner margins of secondaries and all but outer primary, with even, well- defined white margins, which, on the inner secondaries, reach the white terminal margin and, basally, the shaft of the feather; upper wing-coverts subterminally barred, and laterally spotted with white; under primary-coverts white narrowly tipped with black, under secondary coverts buffy; sides of the throat striped black and white, lores and auriculars grayish white; throat white faintly streaked with black; breast and sides white narrowly and not continuously barred with black; abdomen centrally white becoming buffy posteriorly and on the flanks and under tail- coverts; the bars on the abdominal region obsolete or wanting; tibie barred with black and white; feet blackish, bill blackish, tomize and mandible horn-color. Wing, _ 66-69; tail, 64-67.5; exposed culmen, 15-17 mm. DEscrRIPTION OF FeMALE.—Crown rufous-chestnut; forehead buffy, chiefly basally; anteorbital region pale buffy; post-orbital region whitish striped with black; back hazel, grayish olivaceous anteriorly with an ill-defined grayish nuchal band; tail deep ferruginous-hazel ; wings blackish, exteriorly like the tail, internally margined with ochraceous-buff, the inner feathers sometimes obsoletely barred; upper wing- coverts like the tail, darker centrally and with a paler subterminal band and narrow black margin; under wing-coverts ochraceous-buff; throat whitish, obscurely streaked; rest of underparts ochraceous-buff, darker on the flanks and ventral region, paler centrally; the breast with a faint suggestion of dusky bars; feet and bill as in the male. Wing, 64-70.5; tail, 62.5-71; exposed culmen, 16-18 mm. Specimens EXAMINED Thamnophilus zarume.—Ecuapor: Zaruma, 6000 ft., 2 o (ine. type), 4 9; Portovelo, 2000-2700 ft.,5 #,3 9; Rio Pindo, 1850 ft., 1 9; Punta Santa Ana, Prov. del Oro, 4000 ft., 1 @, 2 2; Celica, 6900 ft., Prov. Loja, 1 9; Alamor, Prov. Loja, 4350 ft., 1 @,19. Peru: Milagros, 2200 ft., Prov. Piura, 1 7. Thamnophilus radiatus albicans —CotomBia: Bogoté region, 9 7,8 9. - The discovery of a representative of Thamnophilus radiatus on the Pacific coast of southwestern Ecuador is of exceptional faunal interest. No other form of this group is known from either western Ecuador or western Colombia and the species here described falls into 8 AMERICAN MUSEUM NOVITATES [No. 18 the list of species represented at the eastern and western bases of the Ecuadorian Andes, with no connection of range. The new bird also adds one more species to the list of those restricted to southwestern Ecuador and the immediately contiguous part of Peru. The abundance and nature of my material proves beyond question that the characters which distinguish this proposed new form are not attributable to immaturity and the characters themselves are obviously of specific value. THE STRIPE-BACKED MEMBERS OF THE GENUS Leptasthenura Comparison of the specimen of Leptasthenura, provisionally referred to pileata Sel. in Bull. 117 U.S. Nat. Mus., p. 82, with the tpye of that species shows at once that the Torontoy bird is specifically distinct, and I therefore describe it below. Leptasthenura xenothorax, new species Speciric CHARACTERS.—Most nearly related to Leptasthenura pileata Scl., but hazel-rufous of crown extending to interscapulars; a pronounced white postocular stripe; back much blacker, shaft-streaks whiter and narrower; throat conspicuously black and white, the sharply contrasting black margins of the feathers separated by the basal white, reaching along the shaft to the tip of the feather; remainder of under- parts gray, unmarked and clearly defined from the throat area. Typre.—No. 273,010, U. S. Nat. Mus.; @ ad.; Torontoy, Urubamba Valley, Peru; alt. 14,100 ft.; May 14, 1915; E. Heller. Description oF Typr.—Entire crown and nape unifcrm hazel-rufous, un- streaked; lores blackish; postocular stripe white; center of back black with sharply defined white shaft-streaks; back anteriorly brown; the shaft-streaks more rufous and less clearly defined; tail blackish, the three outer feathers with grayish ends decreasing in extent from without inwardly; wings blackish; their coverts, inner quills, median portion of the external web of the inner primaries and basal portion of the external web of the secondaries cinnamon-buff; throat and sides of the neck conspicuously jet black and snowy white, the sharply contrasting black margins of the feathers separated by the arrow-shaped white area, the point of which extends along the shaft to the tip of the feather; remainder of the underparts uniform smoke- gray with a drab tint. Wing, 66; tail, 77; tarsus, 21; culmen, 10 mm. I found also in the British Museum a specimen of what is evidently Leptasthenura striata Ph. and Land., from Iquique not far south of Arica, the type locality of this apparently rare and little known species. A specimen in the American Museum, collected by Beck at Lima, is appar- ently also to be referred to striata though it was not compared with the Iquique example. Compared with the latter, the type of pileata has the crown solid rufous unstriped, the shaft-streaks of the back and throat- spots more pronounced, the breast and belly dusky olive, the former with broad, the latter, with narrow but well-defined central streaks. 1921} DESCRIPTIONS OF PROPOSED NEW BIRDS 9 Four specimens in the British Museum, collected by Baron in the Temperate Zone of northern Peru, apparently represent striata, the coastal form, but are at least subspecifically and perhaps specifically separable from it. I suggest for this apparently undescribed form the name Leptasthenura striata cajabambz, new subspecies Supspreciric CHaracters.—Similar to Leptasthenura striata striata Ph. and Landb. of northern Chile, but rufous of crown deeper, dorsal stripes whiter, throat-spots more numerous and more pronounced. Wing-quills and coverts mar- gined with grayish instead of cinnamon; cinnamon band at base of inner quills much narrower and paler; margins to tail feathers grayer. Typre.—Registry No. British Museum, 99-6, 1, 81; 92; Cajabamba, Peru; alt. 9500 ft.; March 28, 1894; O. T. Baron. SPECIMENS EXAMINED Leptasthenura xenothorar.—Prrv: Torontoy, the type. Leptasthenura pileata.—Prrvu: Andes of Lima (type, Brit. Mus.). Leptasthenura striata striata —CuiLe: Iquique, 1 (Brit. Mus.). Peru: Lima, 1 o& (Amer. Mus.). These two specimens were examined independently. Leptasthenura striata cajabambe.—Prru: Cajabamba, alt. 9500 ft., 1 2 (type, Brit. Mus.); Cajamarca, alt. 10,000 ft., 2 7; Huamachucho, 1 ° . The specimens on which this well-marked form is based were identi- fied by Salvin as ‘“‘Leptasthenura pileata Scl.’’ (1895, Nov. Zool. II, p. 121). They differ, however, from that species in their striped crown and other characters. The members of the genus Lepthsinailta having the back streaked constitute a closely related group confined to the Andean Paramo, or Puna Zone, and to that portion of the Pacific coast washed by the Hum- boldt Current. So far as known, the group includes four species and four subspecies which may be diagnosed as follows: Crown streaked. Breast streaked. Ecuador and Central Andes of Colombia (type locality, Panza, Ec.). L. andicola andicola Sclater. Peru (type locality, La Raya, south of Cuzco). LD. a. peruviana Chap- man.! Colombia (type locality, Paramo of Macotama, Sierra Nevada of Santa Marta). L. a. extima Todd.* Colombia (type locality, Lagunillas, Boyaca). L. a. exterior Todd.* 11919, Bull Amer. Mus. Nat. Hist., XLI, p. 327. 21916, Proc. Biol. Soc. Wash., XXi IX, p. 97. 31919, Proc. Biol. Soc. Wash., XXXIi, p. 115. 10 AMERICAN MUSEUM NOVITATES [No. 18 Breast not streaked Wings externally margined with grayish. Peru (type locality, Cajabamba). JL. striata cajabambe Chapman. Wings externally margined with cinnamon. Coasts of northern Chile, and Peru (type locality, “ Arica). L. striata striata Philippi and Landbeck. Crown not streaked. Lower parts streaked. Peru (type locality, Andes of Lima). L. pileata Sclater.' Lower parts not streaked. Peru (type locality, iced Urubamba Cajfion). L. xenothorax Chapman. Automolus celice, new species Specrric CHARACTERS.—Not closely related to any recognized species of Auto- moius; possibly nearest A. cervinigularis but feathers of the crown less elongate and browner; underparts streaked, etc.; with a general resemblance to Philydor tempo- ralis Sel., but bill much larger. Crown Brussels-brown instead of olive; throat washed with buff, sides of the neck clear ochraceous orange. Typre.—No. 22,115, Amer. Mus. Nat. Hist.; o ad.; Celica, Prov. Loja, Ecuador; alt. 4550 ft.; Siisbanaben 25, 1920; George K. Chemie. Desckaesies or Maue.—Upperparts ochraceous-tawny, whole crown slightly darker, Brussels-brown; a pronounced superciliary ochraceous-buff anteriorly, be- coming ochraceous orange posteriorly where it meets the clear ochraceous orange of the sides of the neck and ill-defined nuchal band; anteorbital and postauricular regions grayish; tail clear, uniform, deep hazel; wings externally like the back; the quills blackish on the inner web margined with rich ochraceous buff; under wing- coverts ochraceous orange; throat buff to antimony-yellow, its sides and a suggestion of a gular band ochraceous orange; remainder of underparts isabella more tawny olive or tawny on the flanks; the breast-streaks like the color of the throat; lower tail-coverts cinnamon-rufous; feet blackish; maxilla blackish; mandible horncolor, except on tip and cutting edge Wing, 85-89; tail, 73-78; culmen, 21.5-23 mm. DeEscrIPTION OF FeMALE.—Resembling the male in color but somewhat smaller in size. Wing, 80-84; tail, 69-75; culmen, 22mm. SPECIMENS EXAMINED Automolus celice.—Ecuapor: Celica, 6 @ (ine. type), 29; Alamor, 2 9; Guachumana, 4050 ft., 1 7. Philydor ieinoratia: —Ecvuapor: Pallatanga, the type; Alamor, 4550 ft., 1 9; El Chiral, 5350 ft., 1 2. The discovery of this distinct species in the Alamor region further emphasizes the faunal characteristics of the southern end of the Pacifie Subtropical Zone of Ecuador. In the character of its bill and feet Automolus celice agrees with other members of this genus, in its shorter, a 11863, Arch. fiir Naturg., I, p. 119. 1921} DESCRIPTIONS OF PROPOSED NEW BIRDS 11 crown-feathers it is like the species of the genus Philydor, while in general coloration it more nearly resembles Philydor temporalis than any other species known to me. Pachysylvia fuscicapilla albigula, new subspecies Susspeciric CHaracters.—Similar to P. f. fuscicapilla ( Sel. and Salv.) but throat dusky white well defined from yellowish of the remaining underparts which are paler and less uniform than in fuscicapilla, the median line being whitish; upper- parts as in fuscicapilla, the bill shorter; with a general resemblance to P. semibrunnea but smaller and with a shorter bill, the crown Saccordo’s umber rather than hazel, this color extending further on the back; the breast without the buffy wash usually present in semibrunnea, the underparts with more yellow, the wing-coverts greener and nearly uniform without the narrow yellowish margin of semibrunnea. Wing, 56; tail, 41.5; culmen, 13 mm. Typre.—No. 11,033, Museum Goeldi; 9; Sta. Julia, Rio Iriri (branch of the Xingu), Brazil; April 17, 1914; Emilia Snethlage. SpecIMENS ExaMINED Pachysylvia fuscicapilla albigula——Braziu: Rio Iriri, the type; Sta. Helena, Rio Jamauchim (branch of the Tapajoz), 1 @. Pachysylvia fuscicapilla fuscicapilla—Ecuapor: Sarayacu, 3 (ine. type; Brit. Mus.). a. semibrunnea.—Co.omBIA: Miraflores, Cen. Andes, 3 o,1 2, 17; Palmira, 1 2 ?; Aguadita, 1 =; ‘Bogotd,’ 1. - This proposed form is based on two birds contained in a small col- lection forwarded by Dr. Snethlage of the Goeldi Museum, identifica- tion of which was deferred pending examination of specimens in the British Museum. Basileuterus fraseri ochraceicrista, new subspecies Suspspeciric CHARACTERS.—Similar to Basileuterus fraseri fraseri Scl., but with the center of the crown ochraceous-orange instead of lemon-chrome slightly tipped with chestnut. Tyre.—No. 120,138, Amer. Mus. Nat. Hist.; o ad.; Chone, Manavi, Ecuador ; December 16, 1912; W. B. Richardson. SPECIMENS EXAMINED Basileuterus fraseri ochraceicrista——Ecuapor: Chone, 2 & (ine. type), 29; Guayaquil,1 =~; Puno ls.,1 9; Balzar, 2; Santa Rita, 1; Babahoyo, 1 (int.); Naranjito, 2 (int.). Basileuterus fraseri fraseri—Ecuapor: Pallatanga, 1 (the type); Prov. del Oro, Santa Rosa, 3 #7, 2 9, 2?; Zaruma, 1 @; Portovelo, 2 7, 5 2,1 ?; El Chiral, 5350 ft., 1 o, 1 9; Salvias, 3600 ft., 1 9 ; Alamor, Prov. Loja, 4550ft., 27,19. Although the differences between the two races of fraseri here recognized might, at first glance, be considered attributable to age or 12 AMERICAN MUSEUM NOVITATES: | [No. 18. individual variation, the large amount of material examined and the fact that apparently a definite range can be given to each form, convinces me of their racial distinctness. The yellow-crowned bird (fraseri) occurs in the humid forested region east of Puna Island and is found in the Subtropical as well as Tropical Zone. : In the first-named zone it extends at least as far north as Pallatanga. The ochraceous-crowned bird (ochraceicrista) occupies the semiarid Tropical Zone from Puna Island and, on the mainland, from Guayaquil north to the Province of Manavi. Two specimens from Naranjito on the Guayaquil and Quito R. R., just west of Bucay where . the continuous forest begins, are intermediate, the yellow bases to the coronal feathers being only partly, instead of wholly concealed by their ochraceous-orange tips. — . Sporophila insulata, new species Speciric CHaRAcTEeRS.—Resembling Sporophila minuta, but rump largely gray, ‘only the most posterior feathers being chestnut, the tail-feathers basally white. ' Type.—No. 118,142, Amer. Mus. Nat. Hist.; < (ad. ?); Tumaco, south- western Colombia; July 28, 1912; W. B. Richardson. DescripTION oF ADULT (?) Mate In Worn PLumMace.—Upperparts, including upper tail-coverts, mouse-gray, only the terminal feathers of the rump rufous-chest- nut; tail black, white at the base, white on the outer feathers much reduced or absent; wings black, secondaries white for basal half, all but two outer primaries basally white, increasing in extent inwardly; underparts rufous-chestnut, the abdomen mixed with whitish (indicating immaturity ?); the lower tail-coverts chestnut; bill and feet blackish. Wing, 50; tail, 36; culmen, 9.3mm. DescRIPTION OF IMMATURE MALE IN Worn PLUMAGE.—Similar to adult, but abdomen and under tail-coverts white. DESCRIPTIONS OF FEMALE IN WORN PLUMAGE.—Resembling female of S. minuta in comparable condition, but somewhat grayer above and paler below and with more white at the bases of the wing-quills. Wing, 48; tail, 36; culmen, 9.3 mm. ( Specimens EXAMINED Sporophila insulata.—Co.omBi1a: Tumaco, 1 ad. @ (type), 2im. @,1 9. Sporophila minuta.—A large series from Nicaragua to Panama, Colombia, Venezuela, Trinidad and Brazil. While apparently a representative of Sporophila minuta, the bird here described evidently deserves full specific rank. It is known as yet only from the island of Tumaco, but whether an island form or not, it appears to be insulated from its nearest relative, since our researches have thus far failed to discover any other representative of Sporophila minuta on the coasts of either Colombia or Ecuador. Be fy a) i ae AMERICAN MUSEUM NOVITATES ety 4 * : sae ~No. 19° _ ‘NEW MAMMALS FROM BRITISH GUIANA AND COLOMBIA | By H. E. ANTHONY {BR ARYS OT. 26 1921. oC hy . AN ERsity oF 108 pe | - Issued October 26, 1921 : By ORDER OF THE TRUSTEES 2 or : THE AMERICAN MUSEUM OF NATURAL HISTORY Se New York City . AMERICAN MUSEUM NOVITATES Number 19 October 26, 1921 59.9(88) NEW MAMMAIS FROM BRITISH GUIANA AND COLOMBIA By H. E. ANtHony Through an arrangement with Mr. William Beebe, Director of the Tropical Research Station of the New York Zoological Society, The ' American Museum of Natural History has been receiving collections of mammals made in British Guiana, where the station is located. A re- port upon the combined collections is in process of preparation, but four new forms have been found and are described in this preliminary paper. In addition the description of a new rodent, of the genus Dinomys, from Colombia is included. Tayassu pecari beebei, new subspecies Type.—No. 42408, Amer. Mus. Nat. Hist.; Kartabo, British Guiana; June 11, 1919; collector Wm. Beebe. The type is an adult female, with skull. GENERAL CHARACTERS.—Closely related to pecari pecari, but differing in the extent of white on the snout and lower jaw. DescripTion.—Coloration about as in p. pecari but white of face and throat markings more yellowish; long hairs of upper parts brownish black; snout, above, only slightly lighter in color than rest of upper parts and not with strongly contrast- ing whitish of p. pecari; chin and throat patch restricted and not in such marked contrast to the surrounding areas; feet dark to hoofs. Skull as in p. pecari. MEASUREMENTS.—Taken from animal in the flesh: total length, 1090 mm.; tail vertebre, 60; hind foot, 224; weight, 80 pounds. There are five specimens in the Kartabo series, two males and three females, and in no one of them is there the extensive face and throat patch seen in p. pecari collected at Porto Campo, Brazil. In all of them this area is noticeably less extensive and markedly more yellowish. The only specimen with the legs entire shows none of the whitish markings seen in p. pecari just above the hoofs. The Kartabo animal gives a strong impression of being unicolor and is not as contrastingly marked as the animals from Brazil. Although the material for comparison is rather inadequate to show detailed differences, there being only three adult specimens of p. pecari from Brazil at hand, the color differences alone appear to be ample for the separation of a new race, which is named in honor of Mr. William Beebe, the director of the Tropical Research Station in British Guiana. The series of p. pecari, collected in Colombia, Santa Marta district, and identified by Dr. J. A. Allen,’ appear to be more or less intermediate 11904, Bull. Amer. Mus. Nat. Hist., XX, p. 427. 1 2 AMERICAN MUSEUM NOVITATES [No. 19 between true pecari and p. beebei, resembling the Brazilian series rather more than the specimens from Guiana. With regard to the subspecies of pecari described from Ecuador by Lénnberg! as equatoris, | am unable to state the characters by which it and the Guiana specimens may be differentiated, since I have not seen ~ zxquatoris, but the two races are separated by almost the width of a con- tinent and the high ranges of the Andes, and it is unlikely that the two will prove to be identical. Lénnberg’s type is an animal too young to reveal characters that may be accepted with certainty as those of an adult, and the restriction of the white snout and throat markings, in which it agrees with beebei, may well be one of the characters of im- maturity. Specimens of pecari obtained from the headwaters of the Rio Napo, in Ecuador, just east of the type locality of xquatoris, but across the Andes, certainly do not resemble beebei very closely in external characters. NuMBER oF SpeciMENs.—LEight, 7 skins, 7 skulls, 2 skeletons, from Kartabo. iy MEASUREMENTS OF SKULLS OF T'ayassu pecari pecari AND T’. p. beebei. Post- Upper Total Basal | Zygomatic} orbital Molar Length Length | Breadth | Breadth Series T. p. beebet No. 42408 9 type 293 245 121 95 85 42409 sex? 300 250 124 100 79 42410 281 241 125 91 81 48366 9° 273 230 120 87 74 T. p. pecari Average of 9 skulls from Santa Marta, Colombia 260 227 111 88.7 77 T. p. pecari—Brazil 36655" 270 226 111 79 76 36656 280 235 114 91 80 36657 287 236 120 93 82 11921, Arkiv. fér Zool. Stockholm, XIV, No. 4, p. 56. 2Last molar just through gum. a 1921] NEW MAMMALS 3 Pecari tajacu macrocephalus, new subspecies Typre.—No. 48366, Amer. Mus. Nat. Hist., old 9, skin and skull; Kartabo, British Guiana; August 26, 1920; collector Wm. Beebe. GENERAL CHARACTERS.—Similar to tajacu but with skull larger and markedly different in structure. DescripTion.—Pelage about as in tajacu, grizzled yellowish and black, with black dorsal area; collar fairly well outlined. Skull larger than that of tajacu, with more massive build, the forward extension of the zygomatic flange continued to canine alveolus and forming a heavy rostrum; outline of entire skull noticeably subtriangular viewed either from above or below, due ‘to extended zygomatic flange; palate throughout anterior portion wider than distance across the molar series of that portion. MEASUREMENTS.—Taken in the flesh: total length, 948 mm.; length of hind foot, 195. g The Guiana collared peccary differs so from the collared peccary of Brazil and from torvus of Colombia that it is necessary to give it a new name. The skins of the new subspecies are not unlike those of either neighbor but the series of skulls shows differences too great and too con- stant to be overlooked. The subtriangular outline of the skull of macro- cephalus is so unlike the outline of the skull of Brazilian tajacu, which is flask-like, that by this character alone the Guiana animal may be easily differentiated. There is an approach to this condition seen in the skulls of torvus from the Santa Marta district of Colombia but it is not so conspicuous. In macrocephalus the rostrum just anterior to the orbits is especially widened and built out. As may be noted from the table of measurements, the skulls of the new subspecies are larger than those of either tajacu or torvus. The character of the anterior lower molar, given by Bangs in his description of torvus,! appears to have little value in separating torvus from tajacu, since in the series of skulls of these two forms now in the Museum’s collection there are individuals of tajacu which have this tooth with the anterior tubercle two-lobed (a condition said by Bangs to be found only in torvus) and there are individuals of torvus having the tooth with the tubercle entire. In the belief that more material from northeastern South America will eventually show torvus to be a subspecies of tajacu, I have made the Guiana animal a subspecies of tajacu, although it resembles torvus fully as much as it does tajacu. NuMBER OF SPEcIMENS.—Eight, as follows: Kartabo, 3 skins, 5 skulls; Kalacoon, 2 skins, 1 skull. 11898, Proc. Biol. Soc. Wash., XII, p. 164. 4 AMERICAN MUSEUM NOVITATES [No. 19 MEASUREMENTS OF SKULLS OF P. t. macrocephalus, P. torvus, AND P. t. tajacu. ee 5 B|8 la BS ~ |. § opie, ae ze g ss o 5 |b go] 8 oO 60 os) — ni $ = Z 4H |fee\ 3 |3 | OC |se | as =f e 13a8| $41] 3a 1/881 6 a\es] se £ eee] 83 | 83 /32| S| S51 23! Ss ® S- a od an) o 2 a ao | & rt & lgSs\ee|é5|82| & | 8$8\85| 8s macrocephalus 41937 245 205 $41 89 40 83 69 176 72 42407 256 208 109 77 38 89 67 177 71 48145 247 | 2077| 109 | 82 39 82° | 67 171 | 71 48223 246 2038 104 1 39 88 64 174 71 483661 250 | 201 | 106| 74 37 85 63 172 | 69 torvus 14677? 240 197 103 75 41 82 67 168 74 tajacu 330? 236 189 108 71 38 85 65 168 70 @comys rutilus, new species Typr.—No. 42910, Amer. Mus. Nat. Hist., 9 ad.; Kartabo, British Guiana; June 27, 1920; collector Wm. Beebe. The type is a skin with skeleton. GENERAL CHARACTERS.—A small, brightly colored species, with very short tail and clear white under parts. Description.—Color above, between amber-brown and hazel (Ridgway), darkest along dorsal area and on crown, the hairs slaty black for basal two-thirds; below, clear white, the hairs white to the base; hands and feet dirty white, almost dusky; dark orbital ring with small dark area at posterior corner of eye; tail brownish, unicolor. Skull small and broad, rostrum very short, zygomata flaring, a low supra- orbital beading. MEASUREMENTS.—Taken in the flesh: total length, 171 mm.; tail vertebra, 94; hind foot, 20. Greatest length of skull, 24.2; sygoins tie breadlite 13.5; length of | nasals, 7.7; interorbital breadth, 4.4; breadth of brain-case, 11; palate, to incisors, 10; palatal foramina, 3.7 2.2; length upper molar series, 3.4. Compared with Gicomys nitedulus, collected at the same place, ; Kartabo, rutilus is somewhat smaller superficially, much brighter in color, with longer, softer, pelage, shorter tail and conspicuously smaller skull. It is possible that its relationships are with rosilla of Thomas, which is said to be a small, richly colored form, with a short tail. How- ever, it is not identical with rosilla since it is even smaller in size and has the under Fiawie: white instead of ochraceous. iT ype. *Selected skull, the largest in the series. ‘Only approximate since condyles are broken. 41904, Ann. and Mag. Nat. Hist., (7) XIV, p. 35, July. 1921] NEW MAMMALS 5 Echimys longirostris, new species Typr.—No. 42886, Amer. Mus. Nat. Hist., sex indet.; Kartabo, British Guiana; July 26, 1920; collector Wm. Beebe. The type is an adult, teeth well worn, skin with skeleton. GENERAL CHARACTERS.—Most like armatus, but differing in characters of pelage and in significant’ details of cranial structure, having much longer nasals and shallow postpalatal notch. Description.—Pelage spiny, but with many unmodified hairs which partially mask the spines; hairs on crown only slightly spinous; color above, a mixture of black, ochraceous and buff, the ochraceous strongest on nose and face and posterior to shoulders along dorsal area; black strongest on neck and shoulders; flanks lighter than dorsal area and merging insensibly into the grayish under parts; hairs of under- parts subspinous, gray at base and tipped with buff; pectoral area more brightly colored than posterior under parts; hands and feet grizzled gray, buff and ochraceous, dirty white distally; tail haired at base for about 50 mm., colored same as rump, sealy for rest of its length, sparsely haired, practically unicolor, ashy in color. Skull elongate with convex superior outline; nasals long, slender, subcylindrical; lateral margins of temporals forming straight lines, not concave; postpalatal notch U-shaped, reaching scarcely beyond posterior margin of last molar; molar pattern typical of the genus. MEASUREMENTS.—Taken from dried skin; total length, 466 m.; tail vertebre, 225; hind foot, 38. Reluctant as I have been to describe a new Echimys from a region already well supplied with names, I have been unable to reconcile the obvious characters of this specimen with those of armatus (=guiane= castaneus). Aside from the differences in the character of the pelage (its apparently less spinous quality and the minor color differences), for which it might. be remotely possible to account by consideration of the state of wear and the season, the cranial differences are too profound to be called individual variation. Eleven skulls of Echimys from the island of Trinidad, castaneus (=guiane=armatus), of varying ages and a Demerara: example of armatus, identified in the British Museum and received in an exchange from there, show complete accord with one another, in cranial characters with only very slight proportional varia- tions. The skull of the new species is radically distinct from all of these skulls in the much longer nasals, the straight and not concave supra- orbital border, and the shallow postpalatal notch. A careful search of the literature showing figures of Echimys skulls from northeastern South America has failed to disclose any examples with the characters above noted. Only the one specimen of longirostris was collected. 6 AMERICAN MUSEUM NOVITATES [No. 19 MEASUREMENTS OF SKULLS OF Echimys longirostris AND E. armatus. g. a 4 £ = | 3 s | 3 is r=| om g S S 4 ee oe eee a) oe ce » S oc) £& rf A ran ra) ; 2D es ae] A, = ® ° a 3 ey = | & | 83le8/34| 2 | & ee ® r= & 5 5 | 938 = & |$D-s 5) § |e8| 821/68) 2 | § (as3 longirostris 42886 type 56.5 | 19.3 | 26.3 | 14.2 | 21.3:| 11.4 | 23.7 ee armatus—Trinidad ‘ 4727 65.7 1 -16.8.} 25.4 | 12.52) 21 10.1:| 21.307 198% 4728 5528.) 16.9} 25:38. 1018 21.3 |.10:6.1: 21.352 eee 4944 51.41 1558: | 24:6-).12 20.3 9.6! 19.8 | 11.5 4946 63 .1..4.156:.9:) 24 11.9 | 19.8 | 10 20.3:} 11:3 6092 56 17.2 | 26.2 | 12.8 | 21.8] 10.3 | 21.8] 11.9 Demerara 36495 §1.6°| 15 24:4) -12.5:| 19:7 9.8 | 21 12:5 Dinomys gigas, new species 1916. Dinomys branickii Allen, Bull. Amer. Mus. Nat. Hist., XX XV, p. 206, May 31, 1916. 4 Typr.—No. 33913, Amer. Mus. Nat. Hist., La Candela, Huila, Colombia; altitude, 6,500 ft. The specimen is a flat hunter’s skin, purchased by Mr. Leo E. Miller, of the Museum’s expedition to Colombia, in 1912. It has no skull and there are no flesh measurements, while the skin lacks the tail and all four feet. Mr. Miller believed that the animal was killed near La Candela, where the hunter himself lived. GENERAL CHARACTERS.— A very large species, heavily striped and with strongly contrasting pattern of black and white. Description.—Color above, black, grizzled and striped with white; head a grizzled gray, neck to shoulders with less white; shoulders to base of tail with four heavy white stripes, continuous except at their extreme posterior end; flanks from fore to hind legs almost clear white, except that the dark bases of the white hairs show through to give a grizzled effect; legs and under parts similar to flanks; hair quite long everywhere. MEASUREMENTS.—Length of flat skin, tip of nose to base of tail, 750 mm, oe 1921] NEW MAMMALS 7 At the time that this specimen was identified by Dr. Allen, loc. cit., as branickii, the Museum had no other specimens of Dinomys, and the identification was based upon the description by Peters.' Although the agreement was not very close, it was thought inadvisable to attempt description of a new form with such inadequate material. Now the Mu- seum has two additional specimens of Dinomys, typical branickii from the Museu Goeldi at Para and branickii occidentalis from the Western Andes of Ecuador, while I have seen other specimens of b. occidentalis in the collection of Mr. Ludovic Soderstrom at Quito. Upon the basis of this material, it appears quite evident to me that the Colombian Dinomys can not be branickii, nor do I believe it to be a subspecies of it. Aside from the difference in color between the Colombian skin and the specimen of branickii, which is dark brown, there is a marked differ- - ence in the color pattern. The specimen from Ecuador is more like the Candela skin in color but here too the difference in pattern exists. The new species has heavy continuous white stripes, averaging 10 mm. wide, of solid white hairs, while the stripes in both branickzz and occidentalis are interrupted and more a series of semi-connected spots. The long hairs of the flanks form a much whiter covering than in the two known forms. The pelage of gigas is noticeably longer than that of the other specimens of Dinomys in the collection but whether. this is seasonal, individual, or of specific character it would be difficult to say. Although the flat skin may have been considerably iveisligil i in preparation, it is so much larger than the skins of branickii and occi- dentalis, which measure a full fifty per cent lessin total length, that the conclusion as to the greater size of gigas is unavoidable. 11873, Monatsb. k. p. Acad. Wissensch. Berlin, p. 552. Sie Me es ee E =S he 4 _ AMERICAN MUSEUM NOVITATES Rise | No. 20 — PRELIMINARY REPORT ON ECUADOREAN MAMMALS. No. 1 —-By-H. FE: ANTHONY - CY ROR ON, Lo TNE Ua ohm ed oF Dal lg By if, ver Ly rida peed « _ Issued November 3, 1921 By ORDER OF THE TRUSTEES | = %, = . or THE AMERICAN MUSEUM OF NATURAL HISTORY ~ eae New Yorxk City - AMERICAN MUSEUM NOVITATES Number 20 November 3, 1921 59.9(86.6) PRELIMINARY REPORT ON ECUADOREAN MAMMALS. NO. 1 By H. E. ANTHONY Field work done by The American Museum of Natural History in Ecuador in 1920 and 1921 has resulted in the acquisition of more than 900 mammals, to which was added, by purchase and as gifts from Mr. Ludovic Soderstrom of Quito, more than 600 mammals, making a collec- tion which aggregates about 1550 specimens. Using this collection as a nucleus, it is planned to send out additional expeditions' and eventually to issue a report in full on the Mammals of Ecuador. From time to time preliminary reports will be published in order to place on record any forms new to science. The following is the first of these reports. Icthyomys tweedii, new species Type.—No. 47798, Amer. Mus. Nat. Hist., <; Portovelo, Prov. del Oro, Ecuador; altitude, 2000 ft.; July 16, 1920; collector, H. E. Anthony. The type isa skin with skeleton. GENERAL CHARACTERS.—A very large species, differing from the known forms of Ichthyomys in size and in cranial characters. ‘DESCRIPTION.— : Color above, a mixture of black, gray, and buff, the general impression being a grizzled brown; below, white, the plumbeous under-fur showing through to a slight extent; pelage throughout of two types of hair, the long, hard hairs and the soft, short under-fur; tail practically unicolor, slate black, rather densely haired. Skull heavily built and strong, much larger than the skull of soderstromi or of stolzmanni; malar apparently absent; nasals tapering to a point posteriorly; rostrum — very broad; a marked interorbital construction; zygomatic root of squamosal strongly developed; dentition essentially as in other species of Ichthyomys. MEASUREMENTS.—Taken in the flesh: total length, 317 mm.; tail vertebre, 150; hind foot, 36. See table on page 4 for measurements of skull. Tweedii is so distinct from the other species of Ichthyomys that it may be advisable to erect for it a new subgenus. As only one specimen was taken, I have been loath to do this, since some of the characters may be, in part, individual. The Portovelo specimen is so much larger that it is not approached by the largest of a series of seven typical soder- stromi from near Quito, while the size of the skull may be readily noted from the table of measurements. 1A party whose members left New York in June and July, 1921, is now in Ecuador. 2 AMERICAN MUSEUM NOVITATES [No. 20 The wide rostrum, and rather flaring zygomatic arches, differentiate tweedii from either soderstromi, stolemanni or hydrobates, while the areas for muscle attachment upon the frontals and parietals are so well marked as to indicate an animal of much greater strength. In the character of the heavy rostrum, tweedii is nearer to stolemanni and hydrobates than it is to soderstromz, since the latter has the slenderest rostrum of the group. The type specimen was brought in to me by a boy who caught it near his mother’s house, near the banks of the Rio Amarillo. Although traps were set out for [chthyomys in every suitable locality throughout the work in southern Ecuador, no additional specimens were secured. It is rather significant to note the low elevation at which this animal was taken, only 2000 feet above sea-level, as compared with the elevations at which the other species have been found, about 9000 feet in Peru and northern Ecuador, about 4000 feet in Merida, Venezuela. I take pleasure in naming this fine species in honor of Mr. A. M. Tweedy, the resident manager of the mine at Portovelo, who extended to the Museum’s expedition all the assistance it lay in his power to give. NEUSTICOMYS, hew genus GEnotyPe.—Neusticomys monticolus, new species. . GENERAL CHaractEeRS.—Allied to Ichthyomys, Rheomys, and Anatomys, from each of which it differs in the greatly reduced hallux and in the character of the dentition. Description.—Size small, fur very soft and close, with scattering of longer, outer hairs; hind foot least specialized of the group for aquatic life, hallux not extend-_ ing beyond tubercle at base of adjacent digit; skull smooth and typically cricetine, ‘without upturned nasals, incisors showing little specialization; infraorbital foramen large; supraorbital foramina laterally placed; palatal foramina very long, extending from molar toothrow almost to incisors; molars essentially as in Ichthyomys but last lower molar with single cusp instead of two. Neusticomys monticolus, new species Typr.—No. 46574, Amer. Mus. Nat. Hist., 9 ad.; Nono Farm, “San Francisco,” near Quito, Ecuador; February 16, 1916; collector, Ludovic Soderstrom. The type is a skin, in excellent condition, and a skull. GENERAL CHARACTERS.—Superficially most like Rheomys in color, character of fur, and size, but with greatly reduced hallux; noticeable external ear, naked median line from nostrils to upper lip, and very soft fur. DESCRIPTION.— Color above, uniform clove-brown (Ridgway), the hairs blackish plumbeous at the base; below, lighter with irregular washing of pale smoke-gray along the median area and a small ivory-yellow pectoral spot; feet soiled whitish; tail above and below like back but with a faint sprinkling of whitish hairs, especially along under surface. 1921] ECUADOREAN MAMMALS 3 Skull compressed, with broad flat braincase, smooth, without ridges on frontals or parietals, a scarcely perceptible depression at base of nasals, but plane of nasals con- tinuous with that of frontal area; rostrum slender; zygomatic arch threadlike or incomplete; infraorbital foramina very large; palatal foramina very large and filling almost entire space between first molars and incisors, broadest medially; conspicu- ous peglike process on maxillary roots of zygomata just external to molar series; incisors of piercing type but not highly specialized as in Ichthyomys; upper molars not differing appreciably from those of Ichthyomys; coronoid process long, slender, falciform; lower molars somewhat different in pattern from those of Ichthyomys and Rheomys, anterior cusp of first molar broader, last molar with only one functional cusp, the posterior cusp being vestigial. MEASUREMENTS.—Taken from the dried skin; total length, 204 mm.; tail vertebre, 111; hind foot, 25. This most interesting specimen is one of a collection of mammals obtained from Mr. Ludovic Soderstrom, who commented especially upon it when he gave it to me, expressing the opinion that it was not Ichthyomys soderstromi. This was quite evident when the specimen was compared with a good series of Ichthyomys, fourteen specimens represent- ing four species, and Rheomys, one specimen. That it should prove to be an unknown genus is rather surprising, but, since it cannot be definitely referred to one of the existing genera, I find it necessary to erect a fourth genus of the Ichthyomys group. From any of the species of Ichthyomys, Neusticomys monticolus may be readily distinguished by its much smaller size and less advanced specialization for swimming. The feet and toes of this new form are only weakly fringed with hairs, while the foot of Ichthyomys is a most obvious swimming structure. From Rheomys, which genus has not been taken south of Panama, to my knowledge, Neusticomys can be told by its greatly reduced hallux and by its single cusped last lower molar. Anatomys leander, described from the Quito region by Thomas,! is much larger, lacks the external ear, which in Neusticomys is proportionally larger than in Ichthyomys and actually as large, has “muzzle set on in a peculiar manner,” while the superior outline of the skull of Neusticomys is almost a straight line. A second specimen of Anatomys recorded by Lénnberg? who confirms the type description, proves that the peculiari- ties of the type specimen are not fortuitous or individual. Although I have not seen Anatomys, I feel Neusticomys can have so little in common with this genus, that I have not hesitated to’describe it as a new. Apparently, Neusticomys is the least specialized of the Ichthyomys group. 11906, Ann. and Mag. Nat. Hist., (7) XVII, p. 86, January. 23921, Archiv. fér Zool., XIV, No. 4, p. 37. “22 “0} ‘SBUIOY,J, WI} OYE, S}UOWIONSvOW; ‘TOD [eure “Ang “Jorg “TW “N “S “0 Aq pouvor Aypuryr P GG ct} ¢'8 cP X99! PF 9 PL | SST | GOT | 8°ZE ; ; PSEHS sazeqosphiy shwofiyzyat v IZ | PPE | 88 6'€ SSK 19-2 9 6 Sieh SE So6 Ee 6010T quuewuzjojs shwofiyzya qT oT G12 | 8'€l |] 24°8 GV 6 IX 9/167 gg G’St | 191 |} 9 IL | 2°18 GELOF cP GIG E1273 cP 6'1X8'¢S | 8'F 9 FI | SST | ST Tg O€LOF wo.siapos shwofiyzyo] Ee GS 9I | €'6 ¢ CXV 918 F GL GI | 8 8T | ¢°ot | €°9e od4} Q6LLF , upaany shiwmohyzyaq GALLEY 99°F 3X6 F | 6E 8°§I | 8'FT 6 | S$ L1Z eapura) shwopeuy UP 9T | € Il | 29 GP 6 TX9F | 8 F €¢ Sol 6 Slit OE S96 9Z06LT ojdes shwmoayy b> | 8ST | 8'Or 9 |8h |SSxLri sr | Fr | SZti| PF 2tl +6 9% adh} $LC9F Ssnjoojuou shwmooysna Ny SI & e9leelo8) 2 188] 3 eieel6 | @ | 8] & Poibdlewt Fleet & thee e Pes Ss 2 1S |S eS cs 5 3 |' 9] § ie Bie ne mea|Sel|eo BS = is w to Ss Z I eels 4 is ae > Ho fa > Zz me Q = R 2 = Bins| # > + Q 3 iS) ee] & e Zz a ror as = o 5 fee} n ww a 2 > : a = Z = q g Se) 4 gc 4 ce) es E Os a |eF] 2 dnoUr) shwohyyo] AHL 0 SUAAWAYY LNTUTAIG] AO STIAMG 40 SLNAWAWASVAT a 1921] ECUADOREAN MAMMALS 5 Blarina montivaga, new species Typr.—No. 47200, Amer. Mus. Nat. Hist., 9 ad.; Bestion, Prov. del Azuay, Ecuador; altitude 10,000 ft; January 15, 1921; collector, H. E. Anthony. The type is a skin with skeleton, the teeth not showing any great amount of wear. GENERAL CHARACTERS.—Resembling equatoris of Thomas,! but differing con- spicuously in color. DESCRIPTION.— Color above, everywhere approximating a dark mouse-gray (Ridgway); below, mouse-gray; no area of demarcation along sides where color of upper parts merges into that of lower parts; tail agreeing in color with body. Skull slightly larger than that of equatoris, with third unicuspid noticeably larger; dentition very lightly pigmented. MEASUREMENTS.—Taken in the flesh: total length, 112 mm.; tail vertebre, 31; hind foot, 15. Skull, greatest length, 22 mm. (20)?; mastoid breadth, 10.5 (9.7); length entire upper toothrow, 9.7 (8.7). Montivaga may be easily distinguished from all other known Blarina from South America on the basis of color alone, being much grayer than the dark brown or blackish pelages, of sgyuamipes, meridensis, thomasi, equatoris, or osgoodi.2 A series of five, all collected at the type locality, are very uniform in coloration. The new species has a hairy foot, showing very little of the squama- tion which characterises squamipes, and, to a certain extent, equatoris. The teeth of montivaga are almost entirely white, displaying but slight pigmentation, a character which appears to be fairly conspicuous in squamipes, thomasi, equatoris, and osgoodi. Anoura geoffroyi antricola, new subspecies Typr.—No. 47282, Amer. Mus, Nat. Hist., 9 ad.; Loja, Ecuador; altitude 9000 ft.; October 30, 1920; collector, H. E. Anthony. The type is a skin with skull. GENERAL CHARACTERS.—Similar to geoffroyi geoffroyi in size but noticeably darker in coloration, less brown. Description.—Above, hairs clove-brown (Ridgway), lighter colored at the base, except on rump where hairs are almost unicolor, the lightest colored basal area being on the shoulders where the color is pale olive-buff; below, hair brown, darker at the base on abdominal region, unicolor on throat; membranes blackish. MeasvrREMENTS.—Taken in the flesh: total length, 81 mm.; hind foot, 13.5. Greatest length of skull, 26 (geoffroyi, 24.5; apolinari, 25.5); breadth of braincase, 10 (10; 9.5); least breadth of rostrum, 4 (4.5; 4); length of palate, to gnathion, 14,5 (13.5; 14); length of upper molar series, 8.5 (7.5; 8.25). 11912, Ann. Mag. Nat. Hist., (8) IX, p. 409. 2Measurements in parentheses are of No. 46683, Blarina equatoris, from the slopes of Pichincha. sDr. Witmer Stone, of the Academy of Natural Sciences of Philadelphia, has kindly loaned me speci- mens of osgoodi, which I rather suspect must stand as a synonym of equatoris, described in 1912 by Thomas, who had specimens from the same slopes whence came the type of osgoodi. More material is needed to settle this point. 6 AMERICAN MUSEUM NOVITATES [No. 20 A large series of this form were taken, both as skins and as speci- mens in alcohol, and twenty-four skins afford an excellent opportunity to note the extent of individual variation. The entire series agrees with the type in dark coloration and no one of them approaches specimens of typical geoffroyi from Trinidad and Merida. In coloration the Ecuador series more nearly resembles Anoura geoffroyi apolinari (Allen), which was described as a Glossophaga' but which upon an examination of the skull I find to be an Anoura subspecifically distinct from geoffroyi. However, they lack the warmer shade of brown seen in apolinari. Cenolestes caniventer, new species: Typr.—No. 47174, Amer. Mus. Nat. Hist., @ ad.; El Chiral, Western Andes; altitude, 5350 ft.; Prov. del Oro, Ecuador; August 2, 1920; collector, H. E. Anthony. The type is a skin and skeleton. GENERAL CHARACTERS.—Resembling fuliginosus but less brownish above and decidedly lighter colored below; larger in size. DeEscrIPTION.— Color, above, fuseus black (Ridgway) in effect, the pelage made up of dark hairs and a sprinkling of buffy tipped hairs, the color of all the hairs plumbeous at the base; below, much lighter than above, the tips of the hairs varying from cream-color to soiled whitish with a darker pectoral area approaching in color the hairs of the upper parts; hands and feet light brown; tail but little lighter below than above, brown. Skull essentially like that of obscurus or fuliginosus but apparently larger. MEASUREMENTS.—Taken in the flesh; total length, 256; tail vertebree, 127; hind foot, 26.5. Greatest length of skull, 33.5; length of nasals, 16; zygomatic breadth, 16.3; mastoid breadth, 11.5; length of upper toothrow, I-M‘, 17.6 Through the kindness of Dr. Stone, I have before me the two speci- mens of Cenolestes fuliginosus collected by Rhoads on Mt. Pichincha, both females, the trunks of which are preserved in alcohol. I have had the skull of one of these carcasses cleaned for examination. I also have for comparison a skin, without skull, from Papallacta, Ecuador, donated by Mr. Soderstrom of Quito, which I have determined to be fuliginosus; and, finally, a specimen of obscurus, male, skin and skull, from the Plains of Bogoté. I have sufficient material to be certain that the south- ern Ecuador Ceznolestes is a distinct species, the most obvious character of separation being that of color, but in addition the new species is apparently larger. Because there is considerable difference in size shown by the series of caniventer, coupled with the fact that the males are noticeably larger than the females, it is not safe at present to be positive that the apparent size difference is the true one. However, caniventer differs more radically in color from either obscurus or fuliginosus than do the latter from one another. MG 11916, J. A. Allen, Bull, Amer. Mus. Nat. Hist., XX XV, p. 84. > AMERICAN MUSEUM NOVITATES | =o No, ot SOME PARASITIC MEGACHILID BEES OF THE WESTERN UNITED STATES By. T. D. A. COocKERELL Issued December 1, 1921 By ORDER OF THE TRUSTEES OF THE. AMERICAN MUSEUM OF NATURAL HISTORY New Yorx City AMERICAN MUSEUM NOVITATES Number 21 December 1, 1921 59.57,99(78) SOME PARASITIC MEGACHILID BEES OF THE WESTERN UNITED STATES By T. D. A. CockERELL The fauna of the Western United States, especially that of the Rocky | Mountain and Pacific Coast Regions, when fully known, will doubtless furnish many valuable clues to aid in solving the perplexing problems of geographic distribution in North America. As still further collecting is desirable before the final report of the American Museum’s expeditions is published, the following notes on the parasitic megachilid bees thus far obtained are presented at this time. The specimens were collected by Dr. Frank E. Lutz, except where otherwise noted, and the field notes are by him. Cetioxys Latreille Celioxys deplanata Cresson eis: 1 9, Huntsville (a few miles east of Ogden), July 26, 1920; 3 9, near Fort Douglas, Salt Lake City, about 5000 ft. alt., July 28, 1920. Cotorapo: 1 ¢, hair of face pure white, Palisades, near Grand Senetion, about 4750 ft. alt., at Melilotus alba, July 18, 1919; 1 9, Wray, about 3700 ft. alt., August 17, 1919; 1 o, large, hair of face creamy white, lower apical spines of abdomen very stout; Regnier, south of Lamar, about 4400 ft. alt., June 8, 1919. All of these specimens were found in quite xerophytic situations. Colioxys novomexicana (Cockerell) Arizona: 1 9, Lowell Ranger Station, about 2700 ft. alt., August -18, 1916; 2 o’, Sabino Basin, about 3800 ft. alt., July 10, 1916; 1 o, Mud Springs, about 6500 ft. alt. All of these localities are in the Santa Catalina Mts. along the Sabino trail from Tucson to Mt. Lemon. Mud Springs is rather high in the oak-pinyon environ- ment; the Sabino Basin presents a variety of conditions within a small area but chiefly xerophytic; the country at Lowell is desert—the lower part of the Upper Bajada (Shreve’s nomenclature)—but the Sabino River and a water hole introduces more mesophytic vegetation. The male is new. It runs in my table (1912, Canadian Ent., XLIV, p. 168) of males to sayi, from which it is known by the two spots of pubes- cence on the anterior part of the mesothorax and by the entirely red femora. The Sabino Basin specimens differ in having the first recurrent nervure received at the extreme base of second submarginal cell, almost meeting the transverse cubital. They sre also a little smaller. 2 AMERICAN MUSEUM NOVITATES [No. 21 Celioxys octodentata Say (=altilis Cresson) Uran: 1 Q, Ogden, about 4350 ft. alt., August 30, 1916. Coztorapo: 2 Q, Wray, about 3700 ft. alt., along Dry Willow Creek, August 18, 1919. INnpraNna: 1 Q, Lafayette, August 16, 1920. Even the western specimens came from moderately mesophytic situations along streams. Ceelioxys modesta Smith Inpiana: 1 9, Lafayette, August 16, 1920. A variety with dark legs and — nervures. Celioxys apacheorum Cockerell CoLtorapo: 1 9, between Boulder and Orodell along the rather mesophytic canyon bottom, about 5600 ft. alt., August 11, 1919. Described from New Mexico. Ceelioxys rufitarsis Smith Uran: 1 9, 3 rather small , Odgen, about 4350 ft. alt., August 30, 1916; 1 3, Huntsville (a few miles east of Ogden), July 26, 1920; 1 9, Provo, about 4500 ft. alt., in an irrigated field, July 31, 1920. Cotorapo: 1 9, Palisades, near Grand Junction, about 4750 ft. alt., July 18, 1919; 1 9,1 7, Rifle, about 5400 ft. alt., July 19, 1919; 1 9, from a vacant lot in Pueblo, August 9, 1920. Celioxys ribis Cockerell Wyomine: 1 9, Jackson, about 6300 ft. alt., among aspens and various plants of a moderately moist pasture-land type, July 15, 1920. Conorapo: 1 &, Ouray, about 8500 ft. alt., among Douglas spruce, aspen, scrub-oak, etc., July 11, 1919; 1 9, August 7, 1920, and 1 <7, August 1, 1919, Tennessee Pass, about 10,400 ft. alt., in the lodge-pole pine area; 1 9,1 o, Leadville, about 10,200 ft. alt., collected in the town by Mr. H. F. Schwarz. Ceelioxys portere Cockerell CoLorapo: 1 9, between Aspen and Highland along Castle Creek, about 8500 ft. alt., oaks, aspens and a few spruce, collected by Mr. H. F. Schwarz, July 25, 1919; 1 9, Glenwood Springs, about 5800 ft. alt., oak, squaw-bush, sunflower, etc., collected by Mrs. F. E. Lutz, August 5, 1920, 1 9, Boulder, about 5500 ft. alt., in town, August 8, 1919. | This was described as a possible variety of lucrosa Cresson but the transverse channel on the second abdominal segment is deep and entire, whereas, according to Sladen, it is widely interrupted in lwerosa. The type locality of lucrosa is New York State. Ceelioxys texana Cresson InDIANA: 1 <, Lafayette, August 16, 1920. 1921] MEGACHILID BEES 3 Ceelioxys edita Cresson CoLorapo: 1 o, Meeker, about 6200 ft. alt., at Grindelia serrulata, July 21, 1919. Crawford has suggested that this is a synonym of deplanata. I am able to recognize some differences, and for the present must regard edita as valid. Ceelioxys sayi Robertson Inprana: 1 o, Lafayette, August 16, 1920. The form from Virginia which I had regarded as sayz has longer axillar spines, the abdomen with larger and sparser punctures, and the last dorsal segment more produced. It is certainly distinct. Robert- son’s name sayi must be considered as based on Say’s octodentata variety a. “Spots and lines of the thorax |i. e., of white pubescence] obsolete; feet, excepting the tarsi, black.’”? The Virginia insect has the spots and lines present, but the clypeus is bilobed. The last dorsal segment is more produced than in octodentata. The femora are black with red knees, tibiz red stained with black, tarsi reddish basally, apically black, the hind pair black from end of basitarsi on. This may accordingly be separated as: # Ceelioxys mendacina, new species Type: 9. Falls Church, Virginia, June 2, (N. Banks). Length a little over 9mm. The male (length, 8.2-9 mm.) is indicated in the table below. . It is quite likely that Say mixed more than one species, even under his variety a; but Cresson’s description of octodentata (1864) which Robertson cites under say, disagrees in saying that the legs are ferrugin- ous, “‘the coxze and sometimes the femora and tibie more or less black- ish.”” The Lafayette sayi agrees with Crawford’s statement that the clypeus is bilobed as viewed from above, not truly emarginate. It seems reasonable to take the Indiana form as true sayz, but the whole matter is perplexing. The male of sayz, according to Robertson, has the legs black, the tibie and tarsi more or less tinged with red. Ceelioxys fragariz Cockerell Cotorapo: 1 <, Meeker, about 6200 ft. alt., along the river bank, July 21, 1919. I give a new description from the Colorado specimen, as the original account was rather brief. o.—Length, about 11 mm., slender; black, including tarsi, mandibles and antennz, but tegule red; face densely covered with pure white hair, but on cheeks it is thinner, not wholly hiding surface, but with no smooth space; eyes grayish, with moderately long hair; third antennal joint shorter than fourth; mesothorax and 4 AMERICAN MUSEUM NOVITATES [No. 21 scutellum densely and very closely punctured, the mesothorax with a median longi- tudinal depression; anterior margin of mesothorax with a pair of conspicuous white hair-spots; bands of white hair behind mesothorax and scutellum; scutellum with very large punctures, no apical projection; axillary teeth long and somewhat curved; wings dusky hyaline, broadly infuscated apically; stigma ferruginous, nervures fuscous; basal nervure falling a little short of transverse median; recurrent nervures joining second submarginal near apex and base; spurs red; abdomen shining, with strong sparse punctures; first segment with basal and apical white bands, band on basal segment interrupted in middle, the others with successively weaker white bands, but strong bands of white hair in the transverse sulci, broadly interrupted on second, succes- sively less interrupted on the following segments, and nearly entire on the fifth; sides of second segment behind sulci with large opaque areas but no fovesx; fifth segment without distinct lateral spines, sixth with slender lateral spines and six apical ones, the upper apical being each divided into two sharp spines; fourth ventral segment with two sharp spines on margin. By the two, spines on the fourth ventral this falls next to C. erysimi Cockerell, which is very closely related. C. erysimi has black tegule. C. quercina Cockerell is of the same group, but the legs are largely bright ferruginous. For other distinctions see Canadian Entomologist, June 1912. Celioxys aperta Cresson Cotorapo: 1 o, Meeker, about 6200 ft. alt., at Grindelia serrulata, July 21, 1919. C. aperta was based on a single female collected by Morrison in Colorado. The male before me may I think be safely referred to the same species. o@'.—Length 10 mm.; black, including antenne, mandibles and tarsi, the tegulze very obscurely reddish in middle; face, front and cheeks densely covered with long pure white hair; cheeks beneath with a small inconspicuous bare area; third antennal joint longer than fourth; eyes pale green, with moderately long hair; mesothorax and scutellum coarsely and densely punctured, scutellum with a median projection; axillary spines very long, sharp, wings brownish, stigma and nervures dark; basal nervure meeting transverse median; first recurrent nervure joining second sub- marginal cell nearer base than second to apex; anterior coxe spined; abdomen with very broad pure white hair-bands on segments 1 to 5; no white hair at bases of seg- ments or in sulci; second segment strongly punctured, without fovexr; sides of fifth with short spines; sixth segment short, with long curved lateral spines and four short apical ones, the upper very broad; venter with broad, dense, white hair-bands. On account of the median projection on scutellum, this falls next to C. germana Cresson and C. totonaca Cresson. For the distinctions see Psyche, October 1905, p. 89. Celioxys lucrosa Cresson Cotorapo: 2 o, Telluride, along the trail near Cornet Creek at about 10,000 ft. alt., chiefly aspen following cut-over spruce; July 9, 1919; 1 &@, Boulder, about 5500 ft. alt., at Grindelia in town, August 8, 1919. — 1921] MEGACHILID BEES 5 Cresson described C. lucrosa from the female, collected by Comstock in New York State, and by Morrison in Colorado. I have never found the female in Colorado. Sladen (Canadian Entom., June 1915) gave characters for both sexes. The males recorded above agree with his account of male lucrosa, and are referred there. I give a description from a Telluride specimen. o'.—Length about 9 mm.; black, including tarsi, mandibles, antenne and tegu- lx; face with long white hair, faintly tinged with creamy; chegks with thin hair, and a large smooth space below; third antennal joint longer than fourth; eyes gray, with moderately long hair; mesothorax and scutellum strongly punctured, but disc of mesothorax shining between punctures; scutellum simple; axillary teeth short, triangular, wings hyaline, broadly dusky apically; stigma ferruginous, nervures _ fuscous; basal nervure falling short of transverse median, second submarginal cell receiving recurrent nervures about equally distant from base and apex respectively; spurs red; abdomen shining, sparsely punctured, with pure white hair-bands on apices of segments 1 to 4, broad at sides but very thin in middle; sulci without hair, interrupted in middle; foves on second segment large and elongate; fifth segment with short stout lateral spines; sixth with long sharp ones; apex of sixth with four spines, the lower ones long and slender; margin of fourth ventral segment in middle smooth and reddish. ; Ceelioxys lutzi,! new species Urau: 1 Q (the type), Ogden, about 4400 ft. alt., in the Ogden canyon near “Pine View,” mesophytic situation along the stream, August 29, 1916. Wyomina: 1 &, Jackson, about 6300 ft. alt., aspens and various plants of a moderately moist pasture-land type, July 15, 1920. Conorapo: 1 , at about 37° 27’N., 106° 54’ W. in Mineral county near Wolfand Fall Creeks along the road across the continental divide, oak, Engelman spruce, etc., June 20, 1919. Q.—Length, 10.5 mm.; black, including tarsi, tegule, antenna and mandibles; head and thorax with dull white hair, abundant on thorax behind; clypeus normal, minutely and densely punctured, with some large punctures interspersed; eyes gray, with short hair; third antennal joint nearly as long as fourth; mesothorax and scutellum very densely and coarsely punctured; scutellum simple; axillary spines short, thornlike, curved; wings dusky hyaline, stigma and nervures dark; first re- current nervure joining second submarginal farther from base than second from apex; basal nervure meeting transverse median; second submarginal cell very broad on marginal cell; spurs black; abdomen shining, sparsely punctured; white hair-bands on apices of segments 1 to 4, broadly interrupted on first; no white hair in sulci; sulcus on second segment entire, the region behind it in middle very sparsely punc- tured; last dorsal sharply pointed, keeled its whole length, not abruptly narrowed at sides; last ventral considerably longer than last dorsal, narrow, minutely notched on each side before apex; third ventral segment strongly punctured, but with a median oval smooth space in which is a small tubercle; last ventral closely punctured. Named after Dr. Frank E. Lutz, the leader of the Rocky Mountain Expeditions, which will enor- or pa gas our knowledge of western entomology when the rich materials have been sorted and recorded. 6 AMERICAN MUSEUM NOVITATES [No. 21 o.—Length about 9 mm.; face with long white hair; third antennal joint quite as long as fourth; large irregular punctures in middle of mesothorax, with a little shining surface showing between; cheeks with a sm20th space below; anterior cox spined; abdomen shining, sparsely punctured, with entire pure white bands on apical margins of segments and white hair at base of fifth and sixth, second segment with small fovex; sides of fifth with very short spines, of sixth with long ones; apex of sixth quadridentate, all the teeth rather slender, the upper ones divergent, the lower forming a U; fifth ventral segment depressed and shining in middle. The female runs to C. mesta Cresson in Crawford’s table, but it~ differs in the color of the tegule and the shining area on under side of abdomen. The male is known from mesta by the character of the fovez on second abdominal segment. Ccelioxys mes, New species CoLorapo: 1 <, on the Chapin mesa at about 37° 12’ N., 108° 29’ W. in the Mesa Verde National Park, pinyon, Sabina, sagebrush, etc., at the flowers of Pent- stemon coloradensis, July 5, 1919. o.—Length about.10 mm., robust; black, including legs, mandibles and anten- ne; tegule dark reddish, head broad, face with long white hair, cheeks little hairy, densely and coarsely punctured, with a smooth space on extreme lower part; third antennal joint slightly longer than fourth; eyes pale green, with moderately long hair; mesothorax and scutellum densely, coarsely punctured; margin of scutellum simple; axillary spines rather long; wings brownish hyaline, stigma clear ferruginous, nervures fuscous; basal nervure meeting transverse median; second submarginal cell receiving first recurrent nervure nearer base than second recurrent nervure from apex; anterior coxee with very stout spines; spurs dark red; abdomen closely punctured, the second and third segments dull and densely punctured, with strong entire transverse sulci, which are not hairy; hind margins of segments 1 to 4 with white hair-bands, and white hair also at bases of 4 and 5; first segment also with a broad basal hair-band, so that not much of its upper surface is exposed; fifth segment with short stout lateral teeth, sixth with long but obtuse ones; apex of sixth with four teeth, the upper two more divergent than the lower; lower rather short; venter closely punctured, segments 2 to 4 with broad white hair-bands. Especially known by the sculpture of the abdomen. Coelioxys lamellicauda, new species CoLorapo: 1 , Meeker, about 6200 ft. alt., collected in the school grounds by Mr. Pearce Baily, Jr., July 21, 1919. o’.—Length about 9 mm.; black, including mandibles, tarsi, antenne and tegule; face and cheeks with dense pure white hair, cheeks with groove below; third antennal joint longer than fourth; eyes gray, with short hair; mesothorax and scutellum densely punctured; scutellum simple; axillary spines long; wings hyaline, dusky apically, stigma dark reddish, nervures fuscous; basal nervure meeting trans- verse median, second submarginal cell receiving first recurrent nervure nearer base than second recurrent nervure from apex; anterior cox with very stout spines; spurs red; hind margins of abdominal segments with broad entire hair-bands; first =P oe Se ee, 2 ee ee, a a Se a eT ee aa pa mT 1921] MEGACHILID BEES 7 segment also with a basal band; no hair in the sulci, which are entire on segments 2 and 3; segment 2 with small shining spaces but no fovex; sides of fifth segment with very short teeth, of sixth with long ones; apex of sixth with four teeth, the upper ones very broad and obtuse, strongly divergent; venter with entire white hair-bands. Especially known by the shining areas (but no fovez) on second ab- dominal segment, and lamelliform upper apical teeth. 7. 11. 12. 13. The above Celioxys may be separated by the following key. UNNI ise spt es Oe a ct io wha ae Fk ee 67 Vane NE CHES et ve 0% 2 MOROM ss lias Soke tt ee RE NUM Ge ROSE R ede. kee eek ob oeio ed SRE RE wR ON « 12. Legs red; eyes with short hair ES Ot ey ELON RG eee OR ORE ose eie 3. Legs mainly or wholly black, or only tarsired....................0..-.4-- 5. - Last ventral segment very broad; margin of clypeus entire. . .deplanata Cresson. Last ventral segment rather narrow. ........5.....00000 cee cece ede ceeees 4. Over 12 mm. long; clypeus deeply notched or bilobed. .novomexicana Cockerell. Under 10 mm. long; clypeus ordinary.............. ae ee octodentata Say. Last ventral broad, entire, without lateral notches.....:...............0.. 6. Rae. weer meetoliod Ob mines 55.0 ih se a eins ta cade cuWe een 7. Axillary spines short; last ventral very broad, with an apical point. modesta Smith. Axillary spines long; last ventral without a salient apical point. apacheorum Cockerell. Last dorsal with salient lateral angles, or abruptly notched. .rufitarsis Smith. De Deen Went: eat BEM: a Ly cc ee EES IO bee 8. Last ventral broad, with strongly convex margins before the long apical pro- jection (see figure in Canadian Entom., July 1915, p. 205); hair on eyes a, aR aegis ei 82 Sas ies Rane es ok SCE Nes ribis Cockerell. Last ventral without strongly convex sides. ..............00..00000. cee 9. Conspicuous white hair-markings on thorax above; clypeus bilobed or broadly emarginate, with dense pure white hair filling space between clypeus and mandibles; angle formed by emargination of clypeus less acute than in ROPSMCHIONNE EE. ea ee Bea mendacina Cockerell. No conspicuous white hair-markings on thorax above..................... 10. Last ventral very narrow; third ventral with a median polished oval impunc- Sale MOS he er Ae aR re A eS lutzi Cockerell. Last ventral not so narrow; third ventral without a specialized area........ 11, Tegule clear bright ferruginous; hair of eyes very short...... sayi Robertson. Tegule dark rufopiceous; hair of eyes much longer.......... portere Cockerell. Lidge atlenet wiaaly Peds ee ee OS eR OSES. 13. Legs red, cox and under side of femora black; second abdominal segment with small fovexr; hair of eyes short; hair of face pure shining white; cheeks with a large bevelled hairless space; apex of sixth abdominal segment quadridentate, the upper teeth strongly diverging. . . .mendacina Cockerell. Legs Wainy OF WHOUY UINOR Sree tks art coe ee ae cece nce cece tees 16. Mesothorax with large well-separated punctures on disc; tegule bright ferru- ginous; end of abdomen multidentate.................. texana Cresson. . 14. 15. 16. 47, 18. 19. 20. 21. 22, AMERICAN MUSEUM NOVITATES [No. 21 Mesothorax very densely punctured; conspicuous light hair in scutello-meso- thoracic Ute: 2. 5 chs cep uc} ewe bee ewe ane kein © Faigle a 14. Second abdominal segment with a pair of small oval foveze on a smooth surface. novomexicana Cockerell. Second abdominal segment without such fovere.................0c0eceeee 15. Larger; abdomen more coarsely and less densely punctured; first recurrent nervure joining second submarginal cell as far from base as second from LOE ae oops ae We nig VO Oe ee deplanata Cresson. Smaller; abdomen more finely and closely punctured; first recurrent nervure joining second submarginal cell very near base, or even meeting first trans- vorse. cubital... civisie tas awes «sboalde 1p dean ee edita Cresson. divergent; tegule almost pure black; hair of eyeslong..... aperta Cresson. Scutellum without a median tubercle... ....... 00... ccc ceecceeeetecceuccs 1; Conspicuous light hair in scutello-mesothoracic suture; apex of sixth abdominal segment with six teeth; fourth ventral with two sharp spines on margin. Sragariz Cockerell. - No conspicuous sutural hair or spots on thorax above; tegule black or dark — POA od Ore ae a Pe el AD Dee eles eee eee eee eenenenees 18. Stigma lecruninoan: first recurrent nervure going only a little beyond frst: transverse cubital; second abdominal segment dull and very densely PUNCH 6s Ps ENO Sy GS Ew al Glee oe mesz Cockerell. First recurrent nervure going considerably beyond first transverse cubital; stigma often pic@Ous.... ...5 0 s)./....0.0,,«5' 95/60 6 o-pia'y obeleieln ate ben een 20. Rather large and robust; second abdominal segment behind the sulcus very densely and finely punctured, without foves; hair of eyes very long. ribis Cockerell. Smaller or more slender; second abdominal segment otherwise... .. ers Second abdominal segment with large transverse fovee; axillary spines very short; apex of fourth ventral abdominal segment produced, smooth, bare and red; transverse sulcus on second abdominal segment interrupted; second submarginal cell on marginal hardly or not longer than first trans- verse cubital 6.6 s+d « capwidsnn sora bes eee eee lucrosa Cresson. Second abdominal segment with fovese minute or absent.................. 22. Axillary spines long; second segment with a pair of shining spaces but no FOVOND Sox's vb voice S Cpl atre WOODEN Cie EERO lamellicauda Cockerell . Axillary spines very short; second segment with small elongate-punctiform foves; second submarginal cell on marginal considerably longer than first transverse cubital; transverse sulcus on second abdominal segment entire. luzti Cockerell. The host-relationships of the American Celioxys are little known. Greenicher found C. rufitarsts Smith parasitic on Megachile melanophxa Smith and M. latimanus Say. He found C. !ucrosa Cresson or a closely ~ related species parasitic on M. addenda Cresson. In Europe there are records of Celioxys parasitic on Anthophora, but they need confirmation. 1921] MEGACHILID BEES 9 According to the statements of Alfken (Die Bienenfauna von Bremen), the species of Celioxys do in some cases live on more than one species of Megachile, but they are by no means indiscriminate. Thus we have: Parasite Host C. aurolimbatus Foérst Megachile ericetorum Lepeletier C. trigonus Schrank M. maritima Kirby C. quadridentatus Linnzeus are M. circumcincta Kirby and M. willugh- biella Kirby C. acuminatus Nylander M. centuncularis Linneus (probably) C. mandibularis Nylander M. argentata Fabricius C. rufocaudatus Smith M., rotundata Fabricius Bingham, in India, observed that C. basalis Smith lived in nests of M. lanata Fabricius. As Celioxys is world-wide, one might expect to find the number of species roughly proportioned to that of Megachile. In Australia, however, the forms of Megachile are excessively numerous and varied, but there are only four Celiorys (albiceps Friese, regine Cockerell, albolineata Cockerell, froggatti Cockerell). It seems probable that the genus originated in the Western Hemisphere; it is especially abundant in the Neotropical Region, with 110 species described up to the present time. - In the Nearctic Region, the species seem to be usually restricted, or nearly restricted, to a single province, as we find with other bees. C. lucrosa Cresson and sodalis Cresson are reported from New York to Colorado; C. rufitarsis Smith is similarly widely distributed in the North- ern part of the continent and the western mountains. C. octodentata Say, mesta Cresson and portere Cockerell also extend from the Atlantic coast region to the Rocky Mountains. Presumably these all infest the wide-spread species of Megachile, whereas the more local ones parasitise the local species of the host-genus. C. fragari# Cockerell was described from an altitude of 6000 ft. on the San Jacinto Mountains of California; it now turns up at about the same altitude in Western Colorado. The Nearctic Celioxrys at the present time number 52 species and six races or varieties. Of these, nine (edita Cresson, insita Cresson, scitula Cresson, texana Cresson, hunteri Crawford, piercei Crawford, arenicola Crawford, asteris Crawford, pratti Crawford), were described from Texas, whence edita and terana extend northward. A race of texana (sonorensis Cockerell) occurs at San José de Guaymas, Mexico. Ten (menthz Cockerell, gilensis Cockerell, portere Cockerell, apacheorum Cockerell, grindeliz Cockerell, ribis Cockerell, soledadensis Cockerell, texana vegana Cockerell, * rufitarsis rhois Cockerell, novomexicana Cockerell) were 10 AMERICAN MUSEUM NOVITATES [No. 21 described from New Mexico. Of these portere extends to Virginia, and ribis has been taken by Grenicher in Wisconsin and by Sladen in Ontario. C. apacheorum is now recorded from Colorado. Six (coquillettt Crawford, angelica Cockerell, fragariz Cockerell, hirsutissima Cockerell, megatricha Cockerell, angulifera Cockerell) have been described from California. Of these, fragariz extends east- ward to Colorado. Four (floridana Cresson, slossoni Viereck, obtusi- ventris Crawford, dolichos Fox), have been described from Florida. C. quercina Cockerell is from Arizona; C. ribis kincaidi Cockerell is from Washington State; C. aperta Cresson, coloradensis Cresson, erysimé Cockerell, deant Cockerell, grindeitx denverensis Cockerell and crassula Cockerell are from Colorado. C. deplanata Cresson occurs from Kansas to Utah and Washington State. The remaining species are from the eastern and northeastern States, viz. rufitarsis Smith, funeraria Smith (Can.) modesta Smith, octodentataSay alternata Say, sayi Robertson, rufitarsis melanopoda Viereck, germana Cresson, lateralis Cresson, lucrosa Cresson, mesta Cresson, sodalis Cres- son (also Colo.), comstockii, Cresson (N. Y.), immaculata Cockerell (Ind.), sculptifrons Crawford (N. Y.), banksi Crawford (Va.), Robertson found seven species in Illinois (sayz, modesta, germana, alternata, texana, rufitarsis, octodentata). There are in addition one new species from Virginia, one from Utah, Colorado and Wyoming, and two from Colorado, described above. It will be seen from the above that most of the records come from a few States, and there are many parts of the country from which the Celioxys are unknown. Cuetynta Provancher Chelynia elegans (Cresson) Cotorapo: 5 9, Ward, about 9300 ft. alt., near town, August 8, 1919. Chelynia monticola (Cresson) Ipano: 1 9, Bear Lake, along Fish Haven Creek at about 6200 ft. alt., July 9, 1920; 1 9, Giveout near Montpelier, about 6700 ft. alt., July 7, 1920; Wyomine: 3 9, Jackson, along Cache Creek from 6300 to 7000 ft. alt., July 15, 1920. Chelynia subemarginata (Cresson) Ipano: 1 9, Bear Lake, along Fish Haven Creek at about 6200 ft. alt., July 9, 1920, collected by Mrs. F. E. Lutz. Wyomina: 1 9,1 6, Jackson, along Cache Creek, at about 6300 ft. alt., July 15,1920; collected by Mrs. F. E. Lutz. Conorapo: 1 &, Telluride, along the trail near Cornet Creek at about 11,000 ft. alt., July 9, 1919; 1 &, South Fork, about 37° 40’ N., 106° 38’ W. and 8200 ft. alt., June 17, 1919; ie, Ps 1921] MEGACHILID BEES 11 2 9, 2very small <, Camp Creek Ranger Station at about 41° N., 106° 12’ W., and 8700 ft. alt., July 19, 1921. All of these places are quite mesophytic, the Idaho one being the nearest approach to desert conditions. The specimens from Telluride and Camp Creek were taken not far from still-remaining snow. The males are very variable in size but appear to represent a single species. ee Fel fe kt as AMERICAN MUSEUM NOVITATES No. 22 : , NEW SPECIES OF NORTH AMERICAN LIZARDS OF THE GENERA HOLBROOKIA AND UTA By Kart PAtrrerson ScHMIDT — Issued December 1, 1921 By ORDER OF THE TRUSTEES . OF THE AMERICAN MUSEUM OF NATURAL HISTORY ; New Yorx Crry tates Bi AMERICAN MUSEUM NOVITATES Number 22 December 1, 1921 © 59.81,1(7) NEW SPECIES OF NORTH AMERICAN LIZARDS OF THE GENERA HOLBROOKIA AND UTA By Karu Parrerson ScHMIptT In the course of an examination of the lizards belonging to the genera Holbrookia Girard and Uta Baird and Girard, I have found it necessary to recognize a number of new forms. In view of the necessary postpone- ment of the publication of more complete reviews of these genera, I have characterized the new forms in the present preliminary paper and have included keys to the two genera which present in abstract the taxonomic conclusions to be discussed more fully in subsequent papers. I am much indebted to the authorities of the United States National Museum, and especially to Dr. Leonhard Stejneger, Head Curator, Department of Biology, for the loan of valuable collections of Hol- brookias and Utas from Mexico and the southwestern United States for study in connection with the collections of The American Museum of Natural History. - Holbrookia pulchra, new species Dracnostic CuHarActers.—A slender, medium-sized species, allied to H. elegans, with tail considerably longer than the body, the hind leg averaging about four- fifths of the body length; dorsal scales flat; ventral scales large, 59 to 70 from collar to anus; femoral pores few, average 11.6; dorsal spots usually sharply defined, the dorsal and lateral series often confluent, enlarged supraoculars and frontals separated by scales not much smaller than either. MEASUREMENTS OF Typr.—Length, 120 mm.; body, 56 mm.; tail, 64 mm.; tail/total length, .53; foreleg, 28 mm.; hind leg, 47 mm. RanGeE.—Huachuca Mountains of southern Arizona, east of Nogales. Typr.—A. M. N. H. No. 14777; Carr Canyon, 5200 ft.; Huachuca Mountains, Arizona; May 14, 1919; R. D. Camp. Holbrookia maculata campi,! new subspecies Dracnostic CHaracters.—Characters of Holbrookia maculata. Distinguished from H. m. maculata by coloration, which resembles that of approximans, and by the fewer, wider and more oblique upper labials, also as in approximans; distinguished from m. flavilenta by the same characters; distinguished from m. approrimans by the longer tail and hind leg, especially of the female; tail/total length varies from .46 to .50 in 9 m. campi, averaging .48, .42 to 48 in Q m. approrimans, averaging .44; length of leg/body length ranges from .77 to .88 in 9 m. campi, averaging .83, .65 to .78in Q m. approximans, averaging .71. 1Named for Mr. Charles Lewis Camp, the collector of the type series, and well known for his con- tributions to North American herpetology. 2 AMERICAN MUSEUM NOVITATES [No. 22 MEASUREMENTS OF Type.—Total length, 101 mm.; body, 53 mm.; tail, 48 mm.; tail/total length, 48; arm, 25 mm.; leg, 42 mm. Ranon, —Probably the southern art of the Colorado Plateau in northern Arizona. Apparently not reaching Utah on the north, possibly entering New Mexico to the east. It is expected that it will be found to intergrade with m. approximans in central Arizona. Typr.—A. M. N. H. No. 7990; o, about 8 miles N. W. of Adamana, Apache County, Arizona; June 21, 1921; Charles L. Camp. Holbrookia dickersonz,! new species Diaenostic CHaracters.—A large, stout bodied species, with a slightly flat- tened tail equal to or slightly shorter than the body; snout very obtuse; dorsal scales small, slightly convex; small granular scales between the enlarged supra- oculars and the frontals; labials very short and at a high angle with the horizontal, strongly keeled, strongly projecting; femoral pores 9-13; coloration of H. m. approxi- mans, but with three lateroventral black spots entirely surrounded by a patch of blue, which is equally distant from axilla and groin and covers more than half the distance between. MmASUREMENTS OF THE TyPr.—Length, 116 mm.; body, 58 mm.; tail, 58 mm.; _ tail/length, .50; foreleg, 30 mm.; hind leg, 46 mm. RanGcE.—Known only from Castanuelas and Alamos de Parras in the state of Coahuila, Mexico. Typr.—U. 8S. N. M. No. 2668 A?; Castanuelas, Coahuila, Mexico; Lieutenant B. Couch, U.S. A. Key To THE Species or Holbrookia 1. Tail flat with broad black ventral bands; lateroventral black marks placed far back, continued dorsally above the lateral fold (Central Texas to south and central Arizona.) 500 ea. 0 i baw eae ya see texana. Tail rounded; no black bands beneath tail (small black spots in one species); lateroventen! marks more anterior, not present dorsally................ 2. 2. , Tail longer than the body in both sexes.............++sses «5455 eee 3. Tail shorter than the body in the female, usually also in the male............ 5. 3. Dorsal scales very small, convex or keeled; a distinct area of supraocular granules between the enlarged supraoculars and the frontals; tail very long; dorsal spots usually indistinct. (Southern Texas.)...... propinqua. Dorsal scales larger, flat; no distinct small supraoculars between the frontals and the central eran dorsal sAsie area a oe wees sharply outlined.. Seca ss os . 4, 4. Size large, habitus robust, ae iar si sceiia 60 mm.; femoral pores usually 12 or more. (Lower altitudes, Tucson, Arizona south along west coast of Mexico through Sinaloa.) ............... cess eteeeees elegans. Size smaller, habitus slender, body less than 60 mm.; femoral pores usually less than 12. (High altitudes, above 5000 ft., Huachuca Mts., to Nogales and Bisbee, Arisona.) «1 . «ss 002 op vedic see beh ie tnsie os kee pulchra. iNamed for Miss Mary C. Dickerson, former curator of the Department of Herpetology, American Museum of Natural History. 2U. 8. N. M. No. 2668 covering four male specimens, I designate the type by means of a lettered tag A. 1 =e eee . , : 1921). HOLBROOKIA AND UTA 3 5. Subcaudal black spots usually present; scales flat, not tubercular in large specimens; dorsal spots very sharply defined, often digitate behind. (Coahuila, Mexico, southern and central Texas, possibly to Kansas.) lacerata. Dorsal spots rarely sharply defined; no subcaudal black spots; scales keeled or PET 1h INPRO NOERIORIS Sos so oi we Fhe vciew nin hese eee ose 6. 6. Three lateroventral black spots entirely surrounded by a blue patch. (Southern oe ae | sR Cail Fla chee ele MR ag Or Ce, Ae dickersonz. Two or three lateroventral black spots, often margined with blue, but no exten- MOO CNS SINNCHE Cod ret re cei pe erat oe ais ok a ca be maculata—7. 7. Snout somewhat pointed, labials narrow, elongate; usually three or four scales between the enlarged nasals; a mid-dorsal light stripe usually and two dorsolateral and two lateral light stripes frequently present. (Wyoming and Nebraska, south to Texas.)..................+..: maculata maculata. Snout more truncate, labials shorter, wider and more oblique to the horizontal; usually two or three scales between the enlarged nasals; no mid-dorsal MN MN ote Nae ee Se en Gi Lich COM cas KEES bax cham ad er os S. 8. Usual dorsal pattern indistinct, replaced by small irregular light and dark spots; ground color very pale. (‘White Sands”’ of southern New Mexico.) m. flavilenta. Large dark dorsal spots present, ground color darker...................... 9. 9. Tail shorter, .42—.50 of total in male, .42-.48 in female, hind leg shorter, .72—.83 _ in male, .65-.78 in female. (Northern Mexico, southern Arizona; ? south- ‘ GEN ORNS eric Ve a sein So SNe bis clea ie a eae p> 6 kaa m. approximans. Tail longer, .48—.51 in male, .46—.50 in female; hind leg longer, .79-.86 in male, .77—.88 in female. (Central and northern Arizona, probably the Colorado EE Dregs eT a Greer Cn, Le Ee eects. w ehSace Kath wind m. campi. Uta wrighti,! new species Diagnostic CHaracTers.—Closely allied to Uta ornata and Uta levis; distin- guished from the former by the small and smooth lateral basal caudal scales, about 32 in the fifth verticil behind the enlarged postanals; from the latter by the well developed dorsolateral line of tubercles, and the more strongly keeled dorsal scales. MEASUREMENTS OF Typr.—Total length, 125 mm.; snout to anus, 43 mm.; tail, 82 mm.; tail/total length, .66; foreleg, 19 mm.; hind leg, 29 mm.; length of head, 11.5mm.; breadth of head, 8.5 mm. Rance.— Western Colorado and southeastern and southern Utah. Typr.—A. M. N. H. No. 18097; &; Grand Gulch, San Juan County, Utah; elevation between 4000 and 5000 ft.; November 9, 1920; B. T. B. Hyde. Uta gadovi,” new species ~ Diacnostic CHaracters.—Frontal entire, four to six rows of enlarged dorsal scales, abruptly larger than the granular scales with no granular scales on the vertebral 1Named for Dr. A. H. Wright of Cornell University, to whom I owe my introduction to vertebrate zoology. Named for Dr. Hans Gadow, the collector, with especial reference to his important zoological explorations in southern Mexico. 4 AMERICAN MUSEUM NOVITATES [No. 22 line; dorsolateral line and lateral fold set with prominent tubercular scales with a -row of tubercles between them; caudal scales strongly keeled, not spinose, in nearly uniform verticils. MEASUREMENTS OF TypE.—Total length, 134 mm.; snout to anus, 53 mm.; tail, 81 mm.; tail/total length, .60; foreleg, 20 mm.; hind leg, 30 mm.; length of head (to anterior border of ear), 11.5 mm.; breadth of head, 9.5 mm. RancGE.—Jalisco and Michoacan, Mexico. Typr.—A. M. N. H. No. 20355; Cofradia, Jalisco, Mexico; 1902-1904; Dr. Hans Gadow. Uta tuberculata, new species Diagnostic CHAaRActTERS.—Allied to Uta bicarinata, with which it has hitherto been confounded; distinguished from bicarinata by: (1) the longer head and less sloping profile; (2) the much less sharply spinose ventrals; (3) smaller size, not reach- ing 50 mm. of body length; (4) more regular series of lateral tubercles; (5) larger — preauricular spines. MEASUREMENTS OF TyPE.—Total length, 105 mm.; snout to anus, 45 mm.; tail, 60 mm.; tail/total length, .57; foreleg, 19 mm.; hind leg, 27 mm.; length of head 12.0 mm.; breadth of head, 9.0 mm. Ranee.—States of Colima and Jalisco, Mexico. Typr.—A. M. N. H. No. 13737; o&; Colima, State of Colima, Mexico; March 28, 1919; Paul D. R. Ruthling. Uta nelsoni,! new species Draenostic CHaracters.—A large species, with a high and short head, long limbs, and a rather elongate body; allied to Uta bicarinata and U. tuberculata. Ventral scales not mucronate; sides not distinctly tuberculate, not at all bristling in appearance; caudal scales irregular in size, in irregular whorls of three verticils each; dorsal series of enlarged scales beginning on the nape, interrupted on the shoulders. MEASUREMENTS OF TyprE.—Total length (tail reproduced), 128 mm.; snout to anus, 58 mm.; foreleg, 23 mm.; hind leg, 33 mm.; length of head, 13.5 mm.; breadth of head, 10.5 mm. RancE.—Known only from the type locality. Typr.—U.8.N. M. No. 46836; &; Cuicatlam, Oaxaca, Mexico; October 9, 1899; E. W. Nelson and A. E. Goldman. Kery To THE Species or Uta 1. Dorsal scales, very small, perfectly smooth; enlarged supraoculars in more than ONC TOW. 5c: cscces se vt acsivh gis e We on hiv © 5.4 bi0re.b00p-0)s hcg 6h 9 OieeRln n 2. Dorsal scales less than 40 in the length of the head, at least faintly keeled pos- teriorly; enlarged supraoculars in one roW......... 000s ee cee eee eeeee 3. 2. Caudal scales small, smooth. (Southern Lower California.) ........ thalassina. Caudal scales large, keeled, spinose. (Northern Lower California.) ... .mearnsi. 3. Dorsal scales nearly uniform, graduated into the smaller laterals.......... A, Dorsal scales with a few median rows abruptly enlarged. .............+.05 14. INamed for Dr. Edward W. Nelson, Chief of the Bureau of Biological Survey U. 8. rote 5 of Agriculture, with especial reference to his contributions to the scientific exploration at Mexi — ee ee a 1921] HOLBROOKIA AND UTA 5 4, 12. 13. 14, 15. 16. Frontal entire.. eg Wedd o5a Phy alk g gh oC eal GB Waele a x cc Frontal tegumveneliy: Meidad:., Lista gAlis ise acre what ee tes ae Guat beatae Wis akin ew 6. Dorsal scales very small, about 35 in head length. (Northern Lower California.) microscutata. Dorsal scales larger, a broad band of enlarged scales down the back (see also 23). (Southern Lower California.) .. 0... 0. cccs ceca deceeseenees nigricauda. Gular scales about 40; femoral pores 17. (San Pedro Martir Island, Gulf of Ss BRR Rel ice RE ee Bal atte Wee a ae uP arene ara ar palmert. Gular scales less than 35; femoral pores RN iio a As puso Goya aA “2 Hind leg short, .71 to .72 of the body length; dorsal scales very weakly keeled. .8. Hind legs more than .72 of the body length; dorsal scales more sharply keeled. 9. Femoral pores average 13; dorsal scales average 103 from occiput to rump. (Utah O00 TUGTROR oo ra tiie colby che ser ces stansburiana stansburiana. Femoral pores average 15; dorsal scales average about 115. (San Benito Islands, off Pacific Coast of Lower California.)................. stellata. Dorsal scales largest in the group, 70-78 from occiput to rump; hind leg .79 of the body length; femoral pores average 15. (Santa Catalina Island, Ca OE CIO Fo Fis oka eat das ccc dines cmos angers squamata . Dorsal scales ee Et RAS ta olan nanan ey eT ren gon Ugh aE Pe 10. Hind leg averages .80 of the body length. (Southern Lower California.) elegans. Hind leg averages .74 to .75 of the body length......................02-5. 11. Dorsal scales average about 86. (Southeastern California to western Texas and adjacent areas in Mexico. Angel de la Guardia Island.). stansburiana stejnegert. Dorsal scales average more than 90......... 02. ccc cece cee cece eee eens 12. Size large, snout to anus 62mm. (San Martin Island, off Pacific Coast of Lower ERD ie aioe ee ePi se 6 him ico eS eeae o Ve Giada martinensis. Smaller, usually not exceeding 50 mm. in body length..................... 13. Dorsal scales strongly keeled, average about 100; posterior femorals strongly keeled. (Southwestern California, San Joaquin Valley and Northwestern RMU CMPMORMIN gr tu ws oo aint nn Waa RS aie De stansburiana hesperis. Dorsal scales weakly keeled, average about 92; posterior femorals weakly keeled. (Cedros and Natividad Islands off Pacific Coast of Lower California.) concinna. DERM ARTA CUM Eis dva Gd Lcini cn eo bea CRs Sees hoe we Bees a 15. PTO I agree aed Mba Pe eae Ob baled CAP OM SRS iene Enlarged dorsal scales nearly uniform, with no series of small scales on vertebral line; tail long, about two-thirds of total. (Southern Nevada, southeastern _ California, and southwestern Arizona.)...................00% graciosa. One or more vertebral series of small scales............. efile ADEE RRND 16. No tubercular scales forming a well defined dorsolateral line; enlarged dorsals WI RE ord ce te ime iiy, one ec c's Css Cis wek eae Rw ie bale ce 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. 27. AMERICAN MUSEUM NOVITATES [No. 22 A few enlarged scales on the dorsolateral line; upper posterior scales on thigh smooth. (Socorro Island, Revilla Gigedo Islands.).......... auriculata. No enlarged scales on the dorsolateral line anteriorly; upper posterior femorals keeled. (Tierra Amarilla, New Mexico.)..............e2eeeeeeee levis. Lateral scales at base of tail small, leaving four abruptly enlarged dorsal rows of caudals; 30-34 scales in the fifth verticil behind the enlarged postanals. (Southeastern. Utah.,).<.- «0:5 sd s.sisinsiovelor'm atesn Biber iee a wrighti. Lateral caudal scales larger, less than 30 in the fifth verticil................ 19. Enlarged dorsals beginning well forward on the nape...................5- 20. Enlarged dorsals beginning on the shoulders..................-- ornata—21. Enlarged dorsals anteriorly in one row on each side; dorsolateral tubercles small. (Tres Marias Islands; Sinaloa and Sonora, Mexico.)..... lateralis. Enlarged dorsals anteriorly in two rows on each side; dorsolateral tubercles very large, close set. (Clarion Island, Revilla Gigedo Islands.) . .clarionensis. Enlarged dorsal scales more or less irregular in size and arrangement; no oblique series of tubercles between the dorsolateral line and the lateral fold. (West- ern Texas, New Mexico and Chihuahua, Mexico.).......... ornata ornata. Enlarged dorsal scales in four very regular rows; oblique series of tubercles on the Bide. : o.oo cin eins ee pe cee sipiecs see sake dus leis 6 sain ats 22. Enlarged dorsals continuous with caudals; tubercular scales and basal caudals spinose; general appearance bristling. (Southeastern Arizona and ad- jacent area in Sonora, Mexico.).............eeceecescees ornata linearis. Tubercular scales and caudals less spinose; enlarged dorsals reduced on rump. (Southeastern California and southwestern Arizona.)..ornata symmetrica. No dorsolateral line of tubercles. (Southern Lower California)... .. -nigricauda. A well defined series of dorsolateral tubercles, at least posteriorly....... 24. No vertebral series of small scales separating the enlarged dorsals........ 25. One or more vertebral series of small scales separating the enlarged dorsals. . 26. Four to six regular series of enlarged dorsals. (Jalisco and Michoacan, Mexico). gadovi. Two or three irregular series of very large dorsals. (Guerrero, Mexico.) trregularis. Ventral scales mucronate. (Oaxaca, Puebla, and Guerrero, Mexico.) . bicarinata. Ventral scales very faintly, if at all, mucronate..............+0e+eeeees 27. Enlarged dorsal scales beginning on the shoulders; lateral tubercles very distinct; scattered tubercles on neck. (Colima and Jalisco, Mexico.)... .twberculata. Series of enlarged dorsal scales beginning on the neck; lateral tubercles in- distinct. _(Cuicatlam, Oaxaca, Mexico.)............++-- oe... .nelsont, 3 AMERICAN MUSEUM NOVITATES No. 23 THE EPEOLINE BEES OF THE AMERICAN MUSEUM ROCKY MOUNTAIN EXPEDITIONS By T. D. A. Cockeretr Issued December 5, 1921 By OrpsER oF THE TRUSTEES OF THE AMERICAN MUSEUM OF NATURAL HISTORY New York Crry Lee) hoes TO * AMERICAN MUSEUM NOVITATES Number 23. December 5, 1921 59.57,99(78) THE EPEOLINE BEES OF THE AMERICAN MUSEUM ROCKY MOUNTAIN EXPEDITIONS! By T. D. A. CocKERELL This collection constitutes the most remarkable series of Epeoline bees which has ever reached my hands in a single consignment. The Triepeolus number 22 species, of which 16 are new; the Epeolus four species, with two species and a variety new. It appears that the Rocky Mountains constitute the greatest center for Epeolines in the worlc. In addition to the species now recorded, the following have been taken in Colorado: Epeolus beulahensis Cockerell. (Type from New Mexico). interruptus Robertson. Type from Illinois). “4 eldoradensis Cockerell. (Type from Eldora, Colorado). + compactus Cresson. (Type from Texas). Triepeotus helianthi grandior Cockerell. (Type from Florissant, Colorado), martini Cockerell. (Type from New Mexico). a grindeliz Cockerell. (Type from Boulder, Colorado). SaaS denverensis Cockerell. (Type from Denver, Colorado) . : rhoweri (Cockerell). (Type from N. Boulder Creek, Colorado). As subalpinus Cockerell. (Type from Eldora, Colorado). “ gabrielis Cockerell. (Type from California). Colorado has in all 32 species of Epeolines. The whole Nearctic Region has 76; the Neotropical over 70. The Old World is not nearly so richly supplied. From Europe we know 15 species; from Palearctic Africa, 7; from the Ethiopian region, 10. From Asia nine species are recorded; from the Australian region none. There are none in the Malay Islands; the Borneo record in Dalla Torre’s Catalogue is an error. Two species occur in India (EZ. peregrinus Cockerell, E. fervidus Smith) and one (E. assamensis Meade-Waldo) in Assam. £. ventralis Meade-Waldo is found in China. In southern South America the genera Isepeolus Cockerell (9 species), Doeringiella Holmberg (6 species), and Tropho- cleptria Holmberg (2 species) represent a considerable diversification of the Epeoline fauna, but Bréthes considers that Holmberg’s two genera should not be separated from Epeolus. Ducke in this matter follows Bréthes, but these authors take E’peolus in a broad sense, not separating 1Unless otherwise stated, the bees reported upon in this paper were collected by Frank E. Lutz and the field notes are by him. ies 2 AMERICAN MUSEUM NOVITATES [No. 23 Triepolus. Leiopodus Smith, with two Neotropical species, is related to Isepeolus. These two genera are not very near to Epeolus. I consider Triepeolus a very good genus, apparently confined to the New World. It appears to live in the nests of Anthophoride, whereas Epeolus lives with Colletes. | Gribodo (1894) deseribed a subgenus Diepeolus for the Algerian E. giannellii Gribodo, having two free joints to the maxillary palpi instead of one. In general, the species resembles E. fallax Morawitz. Argyroselenis Robertson appears to be a synonym of Diepeolus, but the type species of the Algerian and American groups should be compared. Triepeolus concavus (Cresson) CoLorapo: 2 o, 4 9, Wray, about 3700 ft. alt., some at Helianthus, August 17-19, 1919; 4 3%, 2 @; La Junta, about 4100 ft. alt., along the edges of irrigated fields, August 12, 1920, collected by Mrs. F. E. Lutz; 1, Pueblo, in vacant lots in the city, August 9, 1920. This species is new to Colorado. Triepeolus concolor (Robertson) Cororapo: 3 9, La Junta, about 4100 ft. alt., along the edges of irrigated fields, August 12, 1920, collected by Mrs. F. E. Lutz. This is the first record from Colorado. Triepeolus lunatus (Say) Cotorapo: 1 9, Wray, about 3700 ft. alt., in the moist valley of Dry Willow Creek, August 18, 1919; 3 9,1, La Junta, about 4100 ft. alt., along the edges of irrigated fields, August 12, 1920, collected by Mrs. F. E. Lutz. Not previously recorded from Colorado. Triepeolus pectoralis Robertson Cotorapo: 2 7,1 9, Wray, about 3700 ft. alt., in the moist places (1 male taken by evening sweeping along the river), August 19,1919; 1%, Pueblo, in vacant lots in the city, August 9, 1920; 1 o, Boulder, about 5300 ft. alt.,.on the plains between the town and Boulder Lake, August 12, 1919; 1 & (mesothoracic stripes pointed in front, not reaching margin of mesothorax), Grand Junction, about 4500 ft. alt., along an irrigating ditch, August 3, 1920. The first records from Colorado. Triepeolus penepectoralis Viereck Cotorapo: 2 @ (labrum red; clypeus without a median line; longitudinal bands of mesothorax widely separated from lateral patches), Wray, about 3700 ft. alt., dry, sagebrush country, August 17, 1919; 1 & (labrum red; clypeus with a median shining line; longitudinal bands of mesothorax broad and not far distant from the large patches), Fruita, not far from Grand Junction, about 4500 ft. alt., at Helianthus along the road, July 16, 1919. : 1921] EPEOLINE BEES 3 Not hitherto reported from Colorado. ' T. pxnepectoralis from Wash- ington State, as I have recognized it, is variable, and I am not able to separate the Colorado specimens as a distinct species. It is, however, quite possible that more material would justify the segregation of one or even two species. Triepeolus fortis, new species CoLorapo: 1 9, Wray (type locality), about 3700 ft. alt., in the cottonwood area at the head of Dry Willow Creek, August 18, 1919; 2 <7, La Junta, about 4100 ft. alt., at the edge of an irrigated field, August 12, 1920, collected by Mrs. F. E. Lutz. @ (Type). Length about 14mm.; robust, black, with cream-colored markings; first three antennal joints (except on inner side), labrum and mandibles dark red; legs bright chestnut red; tegule dark reddish marked with black, the general effect dark; axilla black. Head broad; eyes purplish, with upper third green; clypeus very — densely and minutely punctured, with some scattered shallow large punctures, and a smooth median line; third antennal joint much shorter than fourth; mesothorax dull and granular, with an ochreous border (except in middle anteriorly) and a pair of dagger-shaped well-separated bands, their broad bases resting on the anterior border and continuous with the ochreous margin laterad; scutellum bigibbous, axille stout and rather long; pleura densely granular and dull, mainly naked, but with pubescence behind the tubercles and a lobe extending backward at level of lower end of tubercles; area of metathorax with two shining spaces, separated by a dull T; wings strongly brownish, stigma and nervures dark brown; spurs black; hair on inner side of hind basitarsi dark, the tarsi also conspicuously darker than tibie; abdominal bands broad and entire, except the basal one on first segment; black area on first segment a transverse band, but its upper side convex, so that it is elongate-semilunar (but the lower side straight) and intermediate between the two types; lateral angles of black on second segment very acute; fifth segment without light hair at sides, its pygidial patch large and triangular; venter without pubescence; last ventral segment neither elongated nor concave in lateral profile. 3. Smaller; clypeus not covered with hair, but dense white hair at sides of face; antenne black with third joint reddish on outer side; femora varying to mainly black; band at apex of first abdominal segment narrowly interrupted; apical plate broad and dark. A very fine species, resembling 7. hetianthi grandior Cockerell but considerably larger and more robust, with longer axilla and broader apical plate of abdomen. ‘The head is broader than in grandior and the pygidial patch is much broader apically. Triepeolus helianthi Robertson The specimens separate into two lots, the larger or typical form, and a smaller variety or race, 8.5-9.5 mm. long. The latter is probably parasitic on a smaller host-species, but it seems to have no special char- acter aside from size. Perkins observed that the British Epeolus cruciger Panzer lived with Colletes succincta and C. marginata, and those with the latter species were found to be of smaller average size. 4 AMERICAN MUSEUM NOVITATES [No. 28 Larger Race Cotorapno: 1 9, Wray, about 3700 ft. alt., at Helianthus, August 18, 1919; 3 Q, La Junta, about 4100 ft. alt., along roadsides, August 12, 1920, collected by Mrs. F. E. Lutz; 1 9, Glenwood Springs, about 5800 ft. alt., among sweet-clover and sun- flowers near town, August 5, 1920. Smaller Race Cotorapo: 1 9, 1 o&%, Wray, about 3700 ft. alt., on dry upland near town, August 17,1919; 2 9, La Junta, about 4100 ft. alt., along roadsides, August 12, 1920, - collected by Mrs. F. E. Lutz; 1 9 (July 29, 1919), 1 @ (August 5, 1920), Glenwood Springs, about 5800 ft. alt., among sweet-clover and sunflowers near town; 1 9, Palisades, not far from Grand Junction, about 4740 ft. alt., July 18, 1919, collected by Pearce Bailey, Jr.; 1 9, Rifle, about 5400 ft. alt., at edge of swamp along R. R., July 20, 1919, collected by Herbert F. Schwarz; 1 9,1 o, Boulder, about 5400 ft. alt., in town (collected by Pearce Bailey, Jr.) and on the plains near Boulder ia Aontint 7-12, 1919. The two races do not seem to have been living in different environ- ments. The males runin my key in Journ. N. Y. Ent. Soc., X XVII, p. 300, to subspecies pacificus Cockerell, except that they are smaller, with more slender antenne. They belong to the smaller race, whereas pacifi- cus type is of the larger. I gather from Robertson’s account that both sexes of his helianthi have the pleura marked alike; I have only a female from him. It remains to be decided whether I should not have separated pacificus, or whether all the Colorado specimens (excluding grandior) should be referred to it, as a distinct western subspecies. _ Triepeolus schwarzi,' new species CoLorapo: 3 ch Meeker, about 6200 ft. alt., at Grindelia serrulata, July 21, 1919. co. Length 8-9 mm.; black, with cream-colored ornaments; labrum and mandibles (except apex) red; face with dense snow-white hair; clypeus densely granular and dull without a smooth line; eyes very pale grayish-green; sides of vertex with large punctures; antenne black, third joint obscurely reddish; meso- thorax with hair-band round margin (except in anterior middle) and two very broad parallel well separated bands (not sharply defined) reaching beyond center, for half their length connected with lateral bands by a thin inconspicuous pubescence; scutellum flattened; axille small, black; pleura densely covered with creamy-white ~ hair, except on area above middle coxe; tegule dull apricot-color; wings hyaline, brownish apically, nervures and stigma piceous; legs clear bright ferruginous; spurs black; abdominal bands broad and entire, but that on first segment, or first three segments, notched in front; black area on first segment a transverse band, obtuse at sides; angles of black at sides of second segment rounded, but at end of a long sinus, the lobes of hair-bands being long and pointed, directed mesad; apical plate !Named after Mr. Herbert F. Schwars, who was a member of the 1919 expedition, and has done much work on the bees collected. Ee ee ee eee es ee ae 1921] EPEOLINE BEES 5 red, very narrow; venter with appressed silvery-white hair, forming a broad triangle on first segment, and very broad bands (not emarginate posteriorly) on second and third. A neat little species, running in my table in Ann. Mag. Nat. Hist., January 1904, to 7. isocomx Cockerell, but the markings are differently colored, the labrum is different, the scutellum different, etc. There is real affinity with the Californian 7’. callopus Cockerell, but the tubercles and clypeus are black, and there are other differences. It differs from T. rohweri Cockerell by the entirely red anterior femora and tibiz (black in rohweri) ete. Triepeolus balteatus, new species Cotorapo: 1 & (the type), Denver, August 28, 1919, collected by Barbara M. and Marjorie D. Schwarz; 1 =, White Rocks, an interesting cretaceous formation in the plains near Boulder, about 5200 ft. alt., at Solidago, August 13, 1919, collected by Mrs. T. D. A. Cockerell. &. Length about 9 mm.; black, including labrum, mandibles (except dark red band in middle), antenne (third joint dark red on outer side), tubercles and axillx; tegule pale testaceous; legs bright ferruginous, with black spurs; eyes entirely pea- . green; face densely covered with silver-white hair, but pubescence in general cream- color; mesothorax covered with not very dense appressed hair, the two bands (parallel and reaching anterior margin) indicated by denser hair, but not very evident, pos- teriorly a dark semicircle (the concavity cephalad) indicates a nearly bare region; scutellum rather feebly bilobed, axilla short; pleura densely covered with silvery- white hair; wings hyaline, faintly dusky, stigma and nervures piceous; hair on inner side of hind tarsi light orange (it is black in 7. isocomex); abdominal bands btoad and entire; first segment with a transverse dark band, narrow, obtuse at ends; lateral angles of black on second segment right angles; apical plate dark reddish, broader than in 7’. schwarzi, its sides not parallel; second and third ventral segments with broad bands of silvery hair, not emarginate posteriorly. Allied to T. denverensis Cockerell but much smaller, with the Tie of the second abdominal segment entirely covered with light hair, the hair on pleura white instead of creamy, etc. Triepeolus rhododontus, new species Cotorapo: 1 o, Wray, about 3700 ft. alt., at the head’ of Dry Willow Creek, August 18, 1919. ; @. Length about 9.5mm.; black, with labrum (except basal middle), mandibles {except apex), tubercles, ends of axille and legs ferruginous, the hind femora with a large blackish patch in front; clypeus bare, dully minutely rugosopunctate, without a median line or ridge; face otherwise densely covered with pure white hair; eyes pale gray; antennz black; mesothorax dull and rough, margined with ochreous hair except in anterior middle, the stripes dagger-shaped but not sharply defined, with the broad base on anterior margin; scutellum moderately bilobed; axille dentiform, curved, acute; pleura with a crescentic transverse band of white hair, the space be- low it nearly bare and strongly punctured, with a shining median area; tegule apricot- 6 AMERICAN MUSEUM NOVITATES [No. 23 color, dull; wings brownish, stigma and nervures piceous, marginal cell rather un- usually short and broad; spurs black; hair on inner side of hind tarsi fulvous; ab- domen with pale ochreous bands, only that at base of first segment interrupted, but that at apex of first notched before and behind, almost interrupted; black area on first segment a transverse band, rounded at sides; lateral angles of black on second acute, the lateral hair-patches pointed mesad; apical plate red, parallel-sided; venter largely reddish, but second segment black except the broad margin, a crescent of thin whitish hair at each side of second and third segments. Allied to T. occidentalis Cresson (a specimen from Cresson’s collec- tion compared) but distinguished by the red labrum, tubercles and axille, the smooth space on pleura, the narrow, parallel-sided apical plate, ete. Triepeolus (Synepeolus, new subgenus) insolitus, new species CoLorapo: 1 <, Pueblo, in a vacant lot in town, August 9, 1920. 3. Length about 12 mm.; robust; black, including antenne, tubercles and axillze, but labrum and median band on mandibles red; legs chestnut red, the anterior femora black except at apex, spurs black, hair on inner side of basitarsi black; tegule piceous, dark rufous posteriorly. Head broad; eyes with lower half purplish, upper half green; clypeus exposed, very densely and minutely punctured, with a smooth median line and a few scattered large punctures; face at each side of antennz with a longitudinal band of white hair, but on front and between antenne it is dark gray- ish; third antennal joint obscurely rufous at end; thorax and abdomen with ochreous ornaments; black area on mesothorax anchor-shaped, the lateral arms or lobes very broad, the bands of ochreous hair pointed posteriorly (style of 7’. fortis, but stem of anchor more slender); scutellum bilobed, axille short and obtuse; wings brownish, stigma and nervures piceous; only two submarginal cells, the second transverse cubital absent, but first recurrent nervure joining second submarginal cell near base, the second submarginal cell not as long as if a typical T'riepeolus had lost a nervure; the first transverse cubital is transverse, not oblique as in typical T’riepeolus; pleura densely punctured, with a pale dull-ochreous L-shaped pubescent mark; abdomen with rather narrow ochreous bands, that at base of first segment broadly interrupted, that on apex of first emarginate anteriorly and posteriorly, that on second slightly emarginate posteriorly; dark area on first segment shaped nearly as in 7. fortis, but not so broad, and truncate at sides, the pale pubescence at sides of segment nearly twice as broad as that at apex; angles of black at sides of second segment slightly acute, the lateral hair-patches broadly rounded; apical plate dark reddish, fairly broad, with concave margins; venter black, with a little white hair in middle of fringes on fourth and fifth segments, and sixth red in middle. Resembles 7’. fortis, but smaller, with red middle and hind femora, much shorter fourth antennal joint, etc. From the position of the first transverse-cubital nervure, the condition of two submarginal cells is evidently normal and the insect falls in a new subgenus, which may be called Synepeolus, from the united submarginal cells. Phileremus americanus Cresson is not related, but is an Epeoline with two submarginal cells. Ducke (1908) called it Hpeolus americanus. 1921) EPEOLINE BEES 7 Triepeolus brunneus, new species Cotorapo: 1 9, Lawn Lake, Rocky Mountain National Park, about 10,000 ft. alt., August 22, 1919, collected by Herbert F. Schwarz. @. Length about or a little over 9 mm.; rather robust; black, with cream- colored ornaments, but the usually black areas on mesothorax, scutellum, pleura and abdomen wood-brown from a covering of fine pile, but the area of metathorax ex- posed and black, the hair at sides of metathorax white; a large tuft of pale orange hair proceeding from behind each wing mesad over sides of postscutellum; cheeks with a strong abrupt keel behind. Labrum black; mandibles red in middle; clypeus very minutely and densely punctured, with scattered larger punctures, and no median smooth line; eyes purplish, greenish above; antennz long, black, third joint red in front; prominence between antennz strong; sides of face with white hair, becoming yellowish above; mesothorax with ochreous hair on lateral and posterior margins, and a pair of discal stripes, narrowed anteriorly but almost reaching anterior margin; scutellum faintly bilobed, axillze very small, but pointed; tegule dark rufous, hairy; wings hyaline, slightly brownish apically; stigma and nervures rufofuscous; marginal cell unusually short; upper part of mesopleura with white hair, lower part brown, the two colors not sharply separated; legs bright ferruginous, middle and hind spurs black; hair on inner side of hind tarsi pale orange; abdominal bands broad and entire, including basal one of first segment; brown area on first segment a transverse band with straight margins and obtusely rounded ends; lateral hair-patches rounded; pygidial area rather large, the segment covered with ligkt hair on each side of it; venter brown without markings, the hind margins of the segments appearing very narrowly pallid; last ventral segment normal. A very peculiar species, not close to any other, but similar brown pile may be seen in Argyroselenis minima Robertson, and to some extent in other Epeolines. It tends to be denuded as the specimens grow old. Triepeolus trilobatus, new species Cotorapo: 1 <, White Rocks (see above) near Boulder, about 5200 ft. alt., August 13, 1919. o. Length about or a little over 12 mm., robust; black, including labrum, mandibles (except red band in middle), antennz, tubercles, axillze and legs, the hind femora with a red spot behind near apex, and their tibie with a large red patch on inner side apically; hair on inner side of hind tarsi rather pale chocolate; spurs black; ornaments of head and thorax cream-color, but hair at sides of face (not cover- ing clypeus) pure white; eyes green, lower third purplish; clypeus dull and minutely granular, without a smooth line, its upper corners covered with brown hair; fourth antennal joint longer than fifth; mesothorax dull, margined with ochreous except in middle anteriorly, the black area like a large trilobed leaf, but the middle lobe long and narrow and reaching margin, the bands of pubescence pointed and not very long; scutellum bilobed; axille stout-conical; upper part of pleura with dense ochreous hair, more or less L-shaped (the transverse band very broad), the lower part black and dull, shining between the punctures in middle; tegule piceous, the margin reddish posteriorly; wings brownish, stigma and nervures piceous, first recurrent nervure reaching second submarginal cell a little before middle; middle tibiz with a stripe of shining fulvous hair on outer side; abdomen with broad entire bands, basal one on 8 AMERICAN MUSEUM NOVITATES [No. 23 first segment narrowly interrupted; black area on first segment a broad triangle; lateral corners of black on second segment acute, the edge of the pubescence above (cephalad of) them convex; band on sixth segment white; apical plate broad, very dark reddish; venter black, margins of third segment with silver-white hair laterally. Resembles 7’. concolor Robertson, but differs greatly in ornamenta- tion of mesothorax, which is similar to that of 7. fortis. The apical plate is more narrowed apically than in fortis, but the relationship is very close, much like that of concolor to lunatus. Robertson described con-— color in 1898 as a variety of lunatus, but in 1903 treated it as a distinct species. Triepeolus perelegans, new species Arizona: 1 o, Comobabi Mts., about 32° 1’ N., 111° 42’ W., on the road from Haynes Well to the Indian village of Cobabi, about 3400 ft. alt., mesquite-acacia country, August 9, 1916. o. Length about 10 mm.; slender; black, with first three antennal joints and base of fourth, and all the legs, very bright legraplacen: spurs black (red in 7. hop- kinst Cockerell), hair on inner side of hind tarsi orange; labrum red, mandibles suffusedly reddish; tubercles and axille black; ornaments very pale, with a creamy tint, hair of face, upper part of pleura and sixth abdominal segment white; eyes green, purplish at lowerend. Clypeus with the disc flattened, shining, very minutely punctured, with scattered larger but not strong punctures; mesothorax glistening, black, with a pair of rather short, widely separated stripes, not reaching anterior margin; marginal band only along posterior side, and a patch before each tegula; scutellum bilobed, axilla very inconspicuous; pleura with white hair above, below shining, with well-separated punctures; tegule bright ferruginous; wings hyaline, slightly brownish, stigma ferruginous, nervures fuscous; first recurrent nervure joining second submarginal cell beyond middle; abdomen with the bands very slightly yellowish, approaching pure white, basal and apical ones on first segment with linear interruptions, the others entire; black area on first segment a broad transverse band the ends oblique; lateral angles of black on second acute, the lateral hair-patches rounded; apical plate rather broad, very dark reddish; venter with bands of silver- white hair (successively narrower, and not emarginate posteriorly) on segments 2 to 4, the two fringes of curled hairs pale brown (the second darker than the first), some silver-white hair at sides of margin of fifth segment. A pretty and distinct species, known from 7. hopkinsi Cockerell, from the Grand Canyon of the Colorado, by the color of the spurs, the sparsely punctured lower part of pleura, etc. 7. pimarum Cockerell, also from Arizona, has the clypeus red, and the mesothorax red with a broad median black band. Triepeolus sequior, new species Carietea 1 &, Ridgeway, about 7000 ft. alt., sagebrush country, July 15, 1919, collected by Herbert F. Schwarz. 3. Length about 9.5 mm.; black, including labrum, mandibles (except dark red median band), antenne, tubercles, axille and spurs, but legs bright ferruginous, ee ee 1921] EPEOLINE BEES 9 with anterior femora (except knees) and their tibe in front (except apex) black; hair on inner side of hind tarsi pale orange; ornaments cream-color, but face densely cov- ered with silver-white hair, though the band on sixth abdominal segment is colored like the rest; eyes entirely dull pale green; fourth antennal joint conspicuously longer than fifth; mesothorax bordered with rather long and shaggy ochreous hair, the band very narrow along posterior middle, but not broken in middle anteriorly, the longitudinal bands pointed dentiform projections, the black area anchor-shaped; scutellum bigibbous; axille small; mesopleura densely covered with creamy-tinted hair; tegule piceous, with narrow obscure reddish margins; wings brownish, stigma and nervures (except basally) piceous; abdomen with all the bands broad and entire except that on apex of first, which is narrowly interrupted, the black area on first segment a transverse band; lateral angles of black on second acute (about 50°). The lateral patches of hair pointed, with the side above the angle straight (in T. perelegans it is convex); apical plate dark red, narrow; second and third ventral seg- ments with very broad pure white hair-bands, slightly notched in middle posteriorly; fourth with white hair basally the two long curled fringes dull pale yellowish (dark fuscous at ends in 7’. rhododontus). An ordinary-looking species, easily known from small examples of helianthi by the entirely hairy pleura, in the manner of 7’. cressoni Robert- son. From T. fraserxe (Cockerell) it is easily known by the smaller size, the narrowly (instead of widely) interrupted apical band on first ab- dominal segment, the ochreous band along front of mesothorax not in- terrupted, and the much narrower apical plate. T. frasere also has the anterior tibiz entirely clear red. 7. fraserxe is from Beulah, New Mexico, in the Canadian Zone. T. sequior is from the Transition Zone, about 7000 ft. Triepeolus rectangularis, new species Uran: 1 9,1 o&, Huntsville, near Ogden, July 26, 1920. Q. (Type). Length about 11 mm., robust; black, including labrum, mandibles (except dark red median band), tubercles, axilla and legs, but small joints of tarsi red; third antennal joint (except basal two-thirds of inner side) and base of fourth bright chestnut red; tegulze piceous, the margin partly reddish; wings strongly dusky, stigma and nervures (except basally) piceous; outer transverse cubital angled and more or less appendiculate in middle; eyes dark purplish, the upper two-fifths light green; face at each side of antenne with appressed silvery hair; clypeus dull and minutely granular, with scattered large punctures, and a median ridge descending about three-fifths from the top; fourth antennal joint conspicuously longer than fifth; ornaments of thorax and abdomen cream-color; mesothorax with band along posterior and lateral margins, at anterior corners a quadrate patch of hair, emitting a nearly or quite obsolete bridge of hair to lateral stripes, thus enclosing a triangular black marginal area on each side; longitudinal stripes long and broad, diverging posteriorly, pointed, the base not quite reaching mesothoracic margin; scutellum bilobed, axillze short; pleura with a large L-shaped mark of light hair, but it is irregular, its upper part very broad, and extending beneath wings, while its lower edge is concave; lower part of mesopleura densely punctured but glistening; hair on inner side of hind tarsi 10 AMERICAN MUSEUM NOVITATES [No. 23 orange, on mid-tarsi tinged with coppery; spurs black; abdominal bands broad and entire, that on apex of first with a broken linear interruption; black area on first segment a transverse band, very obtuse laterally; lateral angles of black on second segment rectangles; false pygidium large, the segment on each side covered with pale hair; last ventral segment normal; venter without bands. &. Length about 8.5 mm., slender; red on antennz reduced to apex of third segment and base of fourth; stripes on mesothorax shorter; mesopleura covered with pale hair; both bands on first abdominal segment narrowly interrupted, all the abdominal bands of the same color; apical plate piceous, narrow and parallel sided; second and third ventral segment with rather narrow bands of shining white hair, concave in middle and convex at sides, regarded from behind; the two fringes of curled hairs very dark; eyes entirely pea-green. These look like different species but doubtless belong together. The sexual difference in the clothing of the pleura is similar to that in T. cressont Robertson. This may be compared with 7. wyomingensis Cockerell, a black-legged species with the lateral angles of black on second segment right angles. 7’. wyomingensis has a much larger male, without the red on antennez, with a large part of pleura bare, and the white bands on venter very broad and quite different. It also has much longer, shaggy hair on mesothorax. Triepeolus amandus, new species Cotorapo: 1 o, Meeker, about 6200 ft. alt., at Grindelia serrulata, July 21, 1919. &. Length about or nearly 10 mm., rather slender; black, with very pale creamy-tinted ornaments, the band on sixth abdominal segment clear white; labrum, mandibles (except red median band), antenne (except third joint partly reddened), tubercles, axillee and legs (with spurs) black; face densely covered with silver-white hair; fourth antennal joint very slightly longer than fifth; mesothorax with rather loose ochreous hair forming a complete band all around, the longitudinal stripes hardly differentiated, the spaces laterad of them, nearly to their posterior ends, filled with hair, while the space between them is largely filled, though all this hair is rather thin; scutellum bilobed; axilla small; mesopleura covered with white hair, slightly creamy above; tegule small, chestnut red; wings hyaline, faintly brownish apically, stigma and nervures (except basally) piceous; tibie and basitarsi with brilliant silvery-white appressed hair, middle tibize with a stripe of orange hair; hair on inner side of hind tarsi dusky ferruginous; abdomen with all the bands entire; black area — on first segment a narrow transverse band, rounded at ends and sharply defined; lateral angles of black on second segment right angles; apical plate dark red, pointed; second and third ventral segments with broad white hair-bands; curled fringes very dark brown. Also to be compared with 7’. wyomingensis, differing in the pointed apical plate, the broadly rounded ends of black band an first segment, the red tegule, etc. The upper appendiculation of third transverse cubital is wanting, whereas it is very prominent in wyomingensis. —— | i lt 1921] . EPEOLINE BEES 11 Triepeolus lestes, new species Cotorapo: 1 9, Glenwood Springs, about 5800 ft. alt., at edge of town, July 29, 1919, collected by Pearce Bailey, Jr. @. Length about 10mm.; black, with black legs and cream-colored ornaments, small joints of tarsi obscurely pale reddish; eyes purplish with the upper third green; apex of third antennal joint on outer side dark red; fourth joint conspicuously longer than fifth. Similar to 7. rectangularis, but differing thus: scattered punctures of clypeus fewer and much weaker, and no median ridge; stripes on mesothorax con- sisting of two elongated marks (style of 7. helianthi), entirely isolated from the marginal band, which hardly goes mesad of the anterior corners; scutellum not distinctly bilobed; transverse band on pleura narrow and pointed; both hands on first abdominal segment narrowly interrupted; lateral hair-patches on second segment reduced to triangular projections from the band; transverse black band on first seg- ment slightly bulbous at ends; false pygidium smaller. Triepeolus dichropus, new species Cotorapo: 1 o, Glenwood Springs, about 5800 ft. alt., at edge of town, July 29, 1919, collected by Herbert F. Schwarz. 3. Length about 11.5 mm., robtst; black, including labrum, mandibles (except median dark red band), antennz, tubercles and axille; tegule piceous, dark rufous on outer side; hind tibizw and tarsi bright ferruginous; all the knees, hind femora behind, anterior tibie apically, and their tarsi, more dusky red; spurs black; hair on inner side of hind tarsi orange; ornaments cream-color. A little pure white hair at each side of antennz; clypeus dull and granular, with no median line, the scattered larger punctures few and shallow, mainly confined to the apical region; cheeks with a strong posterior keel; mesothorax glistening in middle, the marginal band narrow and weak, failing in anterior middle, and at sides of middle arching away from margin, the longitudinal stripes indistinct and short, but reaching anterior margin; scutellum bilobed, axilla moderate; mesopleura dull, rugose, with a broad band of pale ochreous-tinted hair down its anterior margin, this sending only a small, curved, tooth-like projection backward; wings hyaline, faintly brownish, stigma piceous, nervures fuscous, paler basally; marginal cell long and unusually parallel- sided; outer transverse cubital angular; abdomen with broad entire bands, all of the same color; black area on first segment a broad transverse band, somewhat pointed at sides; band on second segment having at sides lateral lobes shaped like the last joint of a finger, standing at right angles to the band; apical plate piceous, rather narrow; second and third ventral segments with rather narrow bands of white hair, not emarginate posteriorly; curled fringes ochreous with dark tips. An ordinary form in general aspect, but with many distinctive characters. : Triepeolus maculiventris, new species Cotorapbo: 1 9, Navajo Canyon in Mesa Verde National Park, about 6400 ft. alt., at Helianthus petiolaris, July 5, 1919. Q. Length 10.3 mm.; black, with the labrum and mandibles dark red; tubercles and axille black; legs chestnut red, with the anterior femora black; spurs black; hair on inner side of hind tarsi orange, very short; tegule shining black; wings 12 AMERICAN MUSEUM NOVITATES [No. 23 hyaline, dusky apically, stigma and nervures black, the latter basally fuscous; orna- ments cream-color, sharply defined, but face in region of antennz with appressed white hair; clypeus finely and minutely punctured, but glistening, with scattered weak larger punctures, and a delicate smooth median line; antenne black, third joint dusky reddish apically and on outer side; fourth joint longer than fifth; mesothorax with posterior marginal band broad behind tegule, but narrowing to a line in pos- terior middle; in front of tegule is a broad triangular-cuneiform patch, not joining posterior band or reaching anterior margin; longitudinal stripes distinct and clear- cut, lanceolate, parallel, not reaching anterior margin; scutellum feebly bigibbous, axillze rather large, pointed; pleura with a clear-cut L-shaped mark, but it is exca- vated posteriorly behind tubercles, its lower (transverse) arm is parallel-sided and truncate; lower part of pleura densely punctured, but glistening between the punc- tures, except at upper end; abdomen with very distinct bands, both those on first segment narrowly interrupted, that on second slightly notched postericrly; black area on first segment a broad band, obliquely truncate at ends; lateral angles of black on second segment broadly rounded, but the lateral marks are obliquely directed mesad, their outer face very convex, the inner straight; false pygidium moderate, with a large clear-cut triangular patch on each side; venter black, with a sharply- defined patch of white hair at each side of margin of third segment, and a pair of linear marks on fourth; eyes pale green, the lower third purplish. Distinct by the pattern of the mesothorax, and white spots on venter, together with the red legs and dark tegule. The last ventral segment is normal, not concave in lateral profile. Triepeolus laticaudus, new species Cotorapo: 1 9, Cascade, Ute Pass, August 22, 1914, collected by D. M. Fisk. Q. Length 9 mm., slender; black, including labrum (except a pair of obscure red spots), mandibles (except broad red band), antennz (except reddish apex of third segment), tubercles and axille; tegule lively red, with a large black spot in front, and broad hyaline margins; legs bright red, but anterior femora black except apex and a stripe beneath, and middle femora black above, spurs black; hair on inner side of hind tarsi very pale yellowish; eyes pale green, with a grayish or purplish area below middle, but hind margins broadly green throughout; ornaments cream-color, but white hair around antennz. Clypeus shining, very finely punctuerd, with scattered weak larger punctures, but no median smooth line, the basal two-thirds of middle flattened; fourth antennal joint longer than fifth; mesothorax rugose, with the marginal band feebly developed, the discal stripes long and narrow, parallel and wide apart, very nearly reaching anterior margin; scutellum strongly bigibbous; axille large; pleura with an L-shaped mark, but the broad lower arm curved, the convexity below; the ochreous hair extends down the front of mesopleura below the corner of the L; lower part of pleura densely and finely punctured, not shining; wings brown- ish hyaline, stigma and nervures black; abdomen with bands rather narrow, apical one on first segment rather widely interrupted in middle, that on second with a linear interruption; black area on first segment a very broad band, obliquely truncate at sides, but the outermost ends obtuse; lateral hair-marks on second segment reduced to isolated oblique spots, not united with the band; false pygidium very large; last ventral segment normal; venter with weak hair-bands on segments 2 to 4, and the fifth covered, except at sides, with pale appressed hair. 1921] EPEOLINE BEES 13 Known by the abdominal pattern, and especially the pale hair on last ventral segment; the tegule algo are very distinctive. Triepeolus alpestris, new species Cotorapo: 1 Q, Leadville, about 10,200 ft. alt., in vacant lots in town at Lepidium alyssoides, August 4, 1919, collected by Herbert F. Schwarz. | Q. Length 9 mm.; black, including labrum, mandibles (except obscure red band), antennz (except red outer side and apex of third joint), tubercles and axillx; tegule shining rather dark red; wings brownish hyaline, stigma and nervures black; marginal cell appendiculate at end; eyes with lower third purplish, upper two-thirds pale green; anterior legs black with small joints of tarsi red; middle femora black above and red beneath, their tibie black, except at apex; hind femora red, with a dusky band along anterior side, their tibie black, reddish on inner side; middle and hind basitarsi black, small joints red; spurs black; hair on inner side of hind tarsi appearing copper- ed seen from one direction, pale yellow seen from another; clypeus dully minutely punctured, with scattered very weak large punctures, and no median line; ornaments cream-color, but white hair at sides of face; mesothorax with anchor- shaped black area, the lateral bands in front pointed mesad, and only separated by a line from the broad-lanceolate discal marks, which do not reach anterior margin; scutellum convex, feebly bigibbous; axillze small; pleura covered with hair, except the disc below, which is densely punctured and dull, the exposed area not distinctly defined; abdominal bands entire except the apical one on first segment, which has a linear interruption; black area on first segment a transverse band, evenly rounded at ends; lateral angles of black on second segment approximately right angles; pygidial area moderate; last ventral segment normal; venter with thin pale hair-bands on second and third segments. Known by the pattern of mesothorax and abdomen, combined with the color of legs and character of mesopleure. It is allied to T. townsendi Cockerell, from the White Mountains of New Mexico, but easily separated by the much darker legs and the broad second submarginal cell. In townsendi the longitudinal stripes of mesothorax continue broadly to the anterior margin. Comparison may also be made with 7. concinnus Cockerell from Northern Mexico, but ° that is larger, with yellower markings, and the middle and hind femora and tibiz entirely bright red. The above species of Triepeolus bees may be separated by means of the following table. I include also the few species of Epeolus. CHORE ENNIS ROE Sac 5. ice biol are Ris Ay 9 dns ob air poie oe 0.8 E. bifasciatus Cresson. RU RM Sa eg Naira arg n'a sew yee cg ie alk oes 1. pRB DR 8 Os SB at 6 SERIE GA kDa 2. Legs at least mamly or largely Ted... os. ce ec oe cco cee ecb eee 9. 2. Large species, with mesothorax broadly ochreous-haired anteriorly, without longitudinal stripes; apical ventral segment of 92 strongly concave in MACRRAL SID a se ee ee T. concavus (Cresson). Mesothorax with two longitudinal stripes, or if these are indistinct, species small ; apical ventral segment of known females not concave in lateral profile. . . .3. 14 10. aL: 12. 13. 14, 15. 16. 17. 18. AMERICAN MUSEUM NOVITATES [No. 23 Black area on first abdominal segment triangular or subtriangular.......... 4. Black area on first abdominal segment a transverse band.................5 6. Length under 10 mm.; stripes on mesdthorax not separated from light hair of PAGING scious. oh. so dk 456% Be penises Ft a RS E. utahensis Cockerell. Length 10'mm. ‘or more... 53.660. t ihn tn aed Ok kes Gee 5. Stripes on mesothorax small and entirely isolated... .7. concolor (Robertson). Stripes on mesothorax continuous with marginal pubescence. T. trilobatus Cockerell, #. Tegule dusky red; no separate stripes on mesothorax.. . 7’. amandus Cockerell. Tegule black or almost; distinct stripes on mesothorax................... Te Venter with two very Pen white hair-bands. ..7'. rectangularis Cockerell. #. Venter without white hair-bands. .................003 000000008 Ogee 8. Clypeus with a median ridge, and very large scattered punctures. T. rectangularis Cockerell, 9. Clypeus with no median ridge, and small scattered punctures. T. lestes Cockerell, 9. Dark area on first abdominal segment triangular or subtriangular.......... 10. Dark area on first abdominal segment a transverse band................-. zy; Tegule black or piceous; only two submarginal cells... . . 7’. insolitus Cockerell. Tegule dusky red; dark area on first segment aosnawhat intermediate in type; Ehres en bmiarpingl COUR. sc os bce sce sea aoa sana oe T. fortis Cockerell. Tepnim- clear -Y6d< ...cAwtscsuhGe ave casas cen eek wouitinas T. lunatus (Say). Dark areas of abdomen entirely covered with brown pile; tegule red; labrum PABCK cera Sale tae ie cas a Oe oe Hanae T. brunneus Cockerell. Mesothorax without distinct stripes; small or smallish species; tegule red or testaceous; pleura densely covered with light hair................... 138. Mesothorax with distinct stripes, but they reach anterior margin........... 14. Mesothorax with a pair of entirely isolated stripes. ...........0.00-00 eee 22. Over 8 mm. long; lateral angles of black on second abdominal segment right RNGIes OF LAKES fia eos ee eames Meee T. balteatus Cockerell. Less than 8 mm. long; lateral angles of black on second segment acute. E. lutzi Cockerell. Longitudinal stripes of mesothorax joining marginal band of pale pubescence. 15. Longitudinal stripes of mesothorax not joining marginal band.............. 18. Smaller; tegulse clear red. oo... 2.6645 5c as ees ele vowels +010 6 sig etk eee 16. Larger; tegulse dusky red to piceouS........... 06600000 0e web ais Signe LY. Lower part of pleura not covered with hair; axille red at end. T. rhododontus Cockerell. Lower part of pleura covered with hair; axille short and black. T. schwarzi Cockerell. Lateral angles of black on second abdominal segment right angles; apical plate NATTOWSE 362.55 2 ic clare peek ote ee aie T. dichropus Cockerell, . Lateral angles of black or second abdominal segment acute; apical plate DEORGOR: 5 acct faa snsl oot clon co ea eee sens a cites ee T. fortis, Cockerell, o. Much smaller than fortis, with shorter axillee and longer fourth antennal joint, and whole mesopleura densely covered with hair. .7'. seguior Cockerell, <. PMU POO sacs) ic ctiiaelv as ae REELS FSD eh Lee E. pusillus Cresson, 1921] EPEOLINE BEES 15 rh org Ma At 3g OE AE 2 os ree ees ee ek eee ee 20. Labrum black, with two red spots er eet) T. helianthi pacificus Cockerell, @. EG ES URS SP aL Cape pe tipo c to ei re eee iee OF me a 21. 20. Clypeus with a’ median ridge; mesothoracic bands broad in front (Fruita). T. pxnepectoralis Viereck, variety. Clypeus without a median ridge; mesothoracic bands narrow in front (Wray). T. pxnepectoralis Viereck, variety. 21. Small; stripes on mesothorax very broad, and almost touching. E. lutzi Cockerell, 9. Larger; stripes not thus; tegule dusky reddish to piceous; eyes of male en- MIR BOMUEL «i sod he tind ccs seas T. pectoralis Robertson, o’, 9. SE MMRINSMIE INGE ccd Sie t Pk on a x c's Gata Sits kes SE tte RE IES ep ates es 23. Tabrum: black, oF black with red apote 02 as Ee a nie a's 24. 23. First three joints of antenne clear red............ T. perelegans Cockerell, <. WOMEN DHMDNE 0 ave waa ayeatrns cpa ais ead earn a T. maculiventris Cockerell, 9. 24. Tegule clear red, with a black spot in front.......7. laticawdus Cockerell, 9. Tells GATE PORMIN OF TOBE. gc on os sh oe Co eek ap ae ee ees Meet bares s 25. 25. A band of pubescence on anterior part of mesothorax nearly touching longi- MIRE MORI ge a ean ed nse Bok aN wees T. alpestris Cockerell, 9. No such band.. pt ees . 26. 26. Larger; eyes Sarplinhs below, ‘een ghovee Sry Poe T. helisnthé icbertaon. Q. eM ce ee ce eee eae oS Shes Cay Rae Renae Se eA Ds Mpa OSS cs 27. 27. Eyes purplish below, green above....... T. helianthi Robertson, 9, variety. ae OR iss fais odo ok as oe T. pectoralis Robertson, <, variety. 3 Epeotus Latreille - Epeolus bifasciatus Cresson Cotorapo: 1 o&, Wray, about 3700 ft. alt., dry hills near town, August 17, 1919. New to Colorado. Epeolus pusillus Cresson Wyomine: 1 <, mountains near Sheridan, collected by Dr. C. W. Metz. Robertson (1898) suggested that this might be the true EL. mercatus Fabricius. Epeolus utahensis, new species Uranu: 3 o, Huntsville, near Ogden, July 26, 1920. &. Length about or slightly over 8 mm.; black, with the markings pale, ochreous-tinted; mandibles simple, dusky reddish subapically; labrum rugose, entirely black; eyes entirely pea-green; face with appressed silver-white hair; clypeus minutely rugose-punctate; antennz black, third joint equal with fourth; mesothorax anteriorly broadly covered (except in middle) with hair, with a flame-like extension backward on each side of middle; scutellum hardly bilobed, axille incon- spicuous; mesopleura covered with silvery hair; tegule dull piceous, rufous pos- teriorly; wings brownish hyaline, stigma and nervures black; legs black, with small joints of tarsi, and middle basitarsi at apex, ferruginous; spurs black; hind basi- tarsi with bright orange-ferruginous hair on inner side; abdominal hair-bands broad and entire; black area on first segment very broadly triangular, sharply defined, its lower side bulging a little in middle; band on second segment with large lateral lobes, 16 AMERICAN MUSEUM NOVITATES ._ [No. 23 forming an acute inner angle with band; apical plate narrow; second and third ventral segments with very broad hair-bands which in some lights appear silvery- white (not at all ochreous), that on second strongly emarginate, almost divided, in middle posteriorly. Related to E. hitec Cockerell, but much less robust, and with black legs and antenne. Epeolus lutzi, new species CoLorapo: ‘3 &, Walsenburg (type locality), about 6200 ft. alt., Sebina-piniil country, June 14, 1919; 1 &, Regnier, along the state border south of Lamar, about ’ 4400 ft. alt., pasture land, June 7,1919. Utraun: 1 o, Salt Lake City, about 5000 ft. ' alt., near Fort Douglas, July 28, 1920. &. (Type) Length 6.5-7.5 mm.; black, with pale ochreous markings, the dense hair covering face pure silvery white; tegule apricot-color; mandibles broadly bright chestnut-red in middle; antennz black, third joint obscurely rufous apically; legs bright ferruginous, with appressed white hair, but anterior femora black except apically, and anterior tibie broadly black in- middle; eyes pale green, orbits strongly converging below; labrum black, reddish apically; mesothorax densely punctured, thinly covered with pale ochreous hair, which is denser along the margins, and shows anteriorly two nebulous broad converging bands, not reaching the center; scutellum moderately bigibbous, axillee pointed; pleura covered with light hair; wings hyaline, dusky apically, stigma dark red; nervures fuscous, red basally; spurs clear red; abdomen with broad entire bands, but those on first two segments strongly emarginate anteriorly, their anterior margin on each side of the incision convex; black area on first segment a transverse band, very obliquely truncate at each side; band on second segment with large round lateral lobes, making an acute angle with the band; anterior (basal) margin of bands on third and fourth segments strongly undulate; apical plate very broad, bright red; venter broadly covered with pure white hair in middle, and also on hind margins of first three segments, leaving large black sublateral subquadrate areas. &. Similar, but differing thus: eyes dark purplish, green above; third antennal joint wholly black; mesothorax with the area not covered by the broad converging (basad) longitudinal bands and marginal pubescence black, not covered with pale hair, there is a black triangular area anteriorly, between the bands; fifth segment abdomen with the usual broad silvery lunule. Epeolus lutzi dimissus, new race Cotorapo: 1 9, Leadville, about 10,200 ft. alt., in a vacant lot in town, August 4, 1919. Smaller, length 6 mm.; middle femora black, with the apex broadly red; hind femora somewhat dusky; bands on mesothorax shorter and less distinct; tegule smaller, shining; hair of abdomen whiter, the black area on first segment less sharply defined. A race of high altitudes. JZ. lutzi is less coarsely punctured than #. hitei Cockerell, and differs in the color of the antennz and other charac- ters. It is larger than EZ. humillimus Cockerell, which has interrupted abdominal bands, and widely separated stripes on mesothorax. {) AMERICAN MUSEUM NOVITATES No. 24 ae _ WESTERN BEES OBTAINED BY THE AMERICAN MUSEUM EXPEDITIONS SByY As DAs CocKERELL oO pore Oe a ee N ee a al ah gs} aa & Issued December 7 x 1921 By ORpDER oF THE TRUSTEES OF ‘THE AMERICAN MUSEUM OF NATURAL HISTORY _New Yorx Crry AMERICAN MUSEUM NOVITATES Number 24 December 7, 1921 59.57,99(78) WESTERN BEES OBTAINED BY THE AMERICAN MUSEUM EXPEDITIONS Bet. DA. CocKERELL This is the third report! upon the bees obtained in the course of the American Museum’s entomological survey of the Western States. It includes parasitic bees and Panurgide. Unless otherwise stated, the specimens were collected by Frank E. Lutz and the field notes are by him. Nomapa Scopoli Nomada crucis Cockerell Uran: 19, 2c’, Salt Lake City, about 5000 ft. alt., near Fort Douglas, July 28, 1920, collected by Mrs. F. E. Lutz. Cotorapo: 1 9, Grand Junction, about 4500 ft. alt., in a vacant lot near the Fair Ground, August 3, 1920. Described from the Mesilla Valley of New Mexico, and considered a Middle Sonoran species. The specimens from Utah and Colorado are quite typical. Swenk records it from Arizona. In Texas it intergrades | - with N. terana Cresson. Nomada vincta Say Cotorapno: 59, 7, Wray, about 3700 ft. alt., both sexes at Helianthus, one male taken by evening sweeping in a relatively moist place along the river, some collected by Pearce Bailey, Jr., August 17, 1919; 1 9, Boulder, about 5500 ft. alt., ina vacant lot in town, August 8, 1919. Common in Eastern Colorado, but I did not take it in New Mexico. Nomada vincta heterochroa, new variety Cotorapo: 27, Wray, about 3700 ft. alt., one taken with the typical form in evening sweeping and one in a moist place at the head of Dry Willow Creek, August 17 and 18, 1919. Mesothorax red, with a large cuneiform black mark, having its base on posterior margin; mesopleura red, except for the yellow patch; base of abdomen and bands red instead of black, sometimes with black at bases of‘second and third segments. A color-variety, not a race, but very distinct in appearance. 1For the preceding reports see American Museum Novitates, No. 21, December 1, 1921, and No. 23, December 5, 1921. 2 AMERICAN MUSEUM NOVITATES [No. 24 Nomada zebrata Cresson Cotorabo: 1.7, 19, Wray, about 3700 ft. alt., in more xerophytic places than were vincta, the female at Helianthus, August 18, 1919; 19, Boulder, about 5500 ft. alt., in town, August 8, 1919. Extends southward to the White Mountains of New Mexico. Nomada gutierreziz Cockerell Cotorapbo: 1 9, Pueblo, in a vacant lot in town, August 9, 1920. Previously known from the Mesilla Valley, New Mexico. The anterior coxe have very short but distinct spines. Nomada morrisoni Cresson Cotorapo: 29, South Fork, about 8200 ft. alt., near the junction of the two headwaters of the Rio Grande, June 17, 1919. One differs from the type in having the spots on metathorax en- tirely red and yellow band on first abdominal segment broadly inter- rupted. The other has the spots on metathorax yellow. Nomada civilis Cresson IpaHo: 1, Montpelier, about 6100 ft. alt., near town, July 6, 1920. There are two large yellow spots on scutellum, but in some examples of N. civilis (for example, from Florissant, Colorado) these are absent. - Nomada cymbalarie Cockerell CoLorapbo: 1.7, Mineral County, about 37° 27’ N., 106° 54’ W., near the june- tion of Wolf and Fall Creeks, about 7900 ft. alt., vegetation of oak, Engelmann spruce, etc., June 20, 1919. This falls with the male which I have, without having definite proof, regarded as belonging to cymbalariz. It differs from a Florissant speci- men in lacking the upper (somewhat enlarged) ends of lateral face-marks, in not having distinctly red hair on scutellum, in the smaller and entirely _ yellow scutellar spots, and in the more obscure band on first abdominal segment. These differences might all fall within the limits of variation. Nomada eiwardsii Cresson : Ipano: 19, Victor, about 6300 ft. alt., vegetation of aspens, roses, etc., July 11, 1920. This is a member of the Pacific Coast fauna, extending inland to Idaho. The same is true of the next species, NV: citrina. The Californian N. coquilletti Cockerell also reaches Idaho. 1921] WESTERN BEES 3 Nomada citrina Cresson Ipano: 19, Bear Lake, about 6200 ft. alt., along Fish Haven Creek, July 9, 1920. In my table in 1903, Proc. Acad. Nat. Sci. Phila., LV, p. 582, this runs to citrina var., based on a specimen from Grangeville, Idaho. Nomada pecosensis (Cockerell) CoLorapo: 2 9, Pagosa Springs, about 7700 ft. alt., in the U. S. forest reserva- tion, June 21, 1919. A little larger than the type. Previously known from Pecos, New Mexico, and Palisades, Colorado, taken in May and June. Nomada accepta Cresson Cotorapo: 19, Aspen, about 8300 ft. alt., ee Castle Creek, July 15, 1919, collected by Herbert F. Schwarz. Nomada alpha Cockerell This species differs from morrisoni in the red legs (with hardly any yellow), narrower face, and other characters (see Swenk, Univ. of Nebraska Studies, XII, p. 73), but both are very variable and I now incline to the opinion that they represent diverse forms of a single species. The matter will only be settled when we have larger series, including males. Typical alpha is from Fort Collins, Colorado, 4980 ft. alt. Further confusion is introduced by the discovery of two additional forms of alpha, as follows. Nomada alpha paralpha, new variety CoLtoraDo: 19, Walden, about 8400 ft. alt., among sagebrush on hillside, June 17, 1920. ¢ .—Differs thus: length fully 10 mm.; mandibles not yellow basally; labrum yellowish, but clypeus and other face-markings pale red; red lateral face-marks con- tinuous with stripes along posterior orbits, and with a swelling or lobe opposite the frontal spot; yellow behind eyes indistinct; scape red, with a black stripe behind; third antennal joint hardly longer than fourth; mesothorax with a very broad median black stripe, and narrower lateral ones, inclined to be slightly interrupted in middle, or one may say, black with four red stripes, the outer marginal; middle of pleura red, and a red spot beneath the wings; scutellum entirely red; anterior femora with a large black mark behind; first abdominal segment largely black at base; venter light red. Nomada alpha dialpha, new variety _Cortorapo: 29, Walden, about 8300 ft. alt. on the more mesophytic river- bottom among willows, cottonwood, Iris, etc., June 17, 1920. ¢.—Length, 9.3-10.3 mm.; similar to paralpha, the venter red, or with an obscure yellow band on fourth segment, but mesothorax red with a black triangle on . ae 4 AMERICAN MUSEUM NOVITATES [No. 24 anterior margin. There is more red on front, the larger specimen having a broad red band right across. Red band behind eyes broader, without any yellow. Scape en- tirely red. Mesopleura almost entirely red, contrasting with the yellow tubercles. Anterior femora without black. First abdominal segment red, without black base or yellow spots. These forms have some resemblance to N. collinsiana Cockerell, but are quite distinct. Nomada calloxantha, new species Wyomine: 19, Stewart Ranger Station, in the Jackson Hole country at about 43° 42’ N., 110° 45’ W., about 6700 ft. alt., toneenas pine, Engelmann spruce, etc., July 18, 1920. 9 .—Length about 12 mm.; bright eck tae yellow, marked with red and black; head broad, orbits somewhat eunveriiie below; eyes pale grayish, suffused with reddish, bait on the upper third greenish; hair of head and thorax scanty, dorsum of thorax almost entirely nude; mandibles simple, yellow, black at end; lateral face- marks broad, extending over eyes to a broad stripe down posterior orbits, but inter- rupted by a large red patch on upper part of front; black marginal spots on clypeus, connected by a line with base of antennew; region above and between antennze blackened, but a transverse red band across front; vertex and posterior part of head black, a little red on occiput; atenne stout; Scape thick, yellow, partly red behind; flagellum entirely bright ferruginous; third antennal joint «bout as long as fifth, but conspicuously shorter than fourth; mesothorax coarsely rugose and dull, red, wihtls yel- low stripes over tegule and bahind: and a median black band, narrow and faint in | middle, triangularly expanded posteriorly, and less so in front; prothorax black, with the swollen upper margin and tubercles yellow; a small black area below wings; meso- pleura yellow, with a transverse reddish stain on upper part, and a large red patch below; a broad black area behind mesopleura, bordered with red at sides of meta- thorax; metathorax with a broad median black band, the sides of the basal area hav- ing large yellow patches; scutellum and postscutellum yellow, the former strongly bigibbous; tegulze pale yellowish, semitransparent; wings reddish, stigma and nervures ferruginous; basal nervure going a considerable distance basad of trans- verse median; third submarginal cell broad below; legs yellow; anterior trochanters and marks at base of femora red; middle coxe mainly black, their trochanters red with a yellow spot, and their femora largely red at base; hind coxsee marked with red — and black, their trochanters red, their femora mainly black on inner side, and with a red basal patch above, their tibize red on inner side except at base, their basitarsi with dense short light red hair on inner face; abdomen bright yellow; basal half of first segment red, with a median black mark; four rather narrow dark bands, the first two reddish, on apices of segments and adjoining bases; venter yellow, with two narrow dark bands, failing laterally. A member of the subgenus Xanthidium, running in my tables to N. morrisoni flagellaris Cockerell, but certainly distinct. In Swenk’s table of Nebraska species it runs to the much smaller N. citrina flavo- marginata Swenk, and in his further table of the same group (Univ. of Nebraska Studies, XII, p. 68) it runs nearest to N. rufula (Cockerell), 1921] WESTERN BEES 5 which was described as a variety of citrina. It differs from rufula in the venation and other characters and is, I think, certainly distinct. N. rufula is from Idaho. Nomada melanoptera, new species Cotorapo: 1 9, Wray, about 3700 ft. alt., on a dry hill near town, July 17, 1919. ¢ .—Length about 11 mm., broad and robust; head black and red, thorax and abdomen black and bright lemon-yellow; mandibles simple; anterior coxe without spines; third antennal joint long, much longer than fourth or fifth, but not as long as the two together; basal nervure meeting transverse median; abdomen strongly punctured. Head broad, facial quadrangle broader than long; eyes pale red; mandibles (except apex), labrum, lower margin and broader corners of clypeus, and _ entire sides of face ending obliquely a little above antennz, bright ferruginous; face and front very coarsely punctured, clypeus with a smooth area on lower middle, supraclypeal area with very large sparse punctures; posterior orbits with a narrow . red stripe, flushed with yellow at its upper end; antennz with the first three joints bright red, the next three reddish, the rest black; mesothorax entirely black, very coarsely and densely punctured; prothorax black, with the thick upper border and the tubercles yellow; pleura very coarsely punctured, entirely black; metathorax all black; scutellum bigibbous, with two very large round yellow spots; postscutellum yellow; tegule bright red; wings dark fuliginous, with an irregular hyaline area in the subapical region, but the apical very ‘dark; stigma ferruginous, nervures fuscous; legs bright red; hind tibia with a pointed posterior apical lobe, on the outer margin of which are four equal spines; abdomen broad, black, the first five segments with broad yellow bands, that on second very broad, that on fifth broadly emarginate posteriorly ; sides of apex with dark fuscous hair; venter black, with a red transverse line on first segment, not extending to sides. A remarkable species, which persistently falls with NV. (Holonomada) superba Cresson in the tables, but is actually very different and closely related to N. (Micronomada) arenicola Swenk, but without coxal spines, or yellow on face or pleura. The hind tibie are very distinctive. Nomada crawfordi lachrymosa, new variety Wyominea: 19 (type): Jackson, about 6300 ft. alt., along Cache Creek, among vegetation of moderately moist pasture-land type, July 15, 1920; 1 9, Rawlins, about 6800 ft. alt., among sagebrush on a hill near town, June 25, 1920; 19, Medicine Bow, about 6600 ft. alt., among sagebrush on the ridges having a few Pinus scopu- lorum, June 23, 1920. 9 .—Length about 11 mm.; general color of head, thorax and legs bright ferru- ginous, of abdomen bright yellow; eyes pale reddish gray; mandibles simple, black at end; labrum yellow; lower border of clypeus and sides of face suffusedly yellow; face broad; cheeks with pale red hair; front red; antenne entirely bright ferruginous or darkened apically, third joint long, but shorter than fourth; mesothorax dull and densely punctured, dull red, with a black mark posteriorly; tubercles and upper border of prothorax yellow; pleura red, with a small yellow mark, not always present; 6 AMERICAN MUSEUM NOVITATES [No. 24 scutellum strongly bigibbous, it and the postscutellum red; metathorax with a pair of dull pale yellow spots, and often a black shade in center of basal area; tegule orange, with a yellow spot; wings reddish, stigma bright ferruginous, nervures fus- cous; basal nervure going a little basad of transverse median; anterior tibiz faintly suffused with yellow at base; hind femora with a broad black area behind; hind tibie with an apical outwardly projecting emarginate lobe; hind basitarsi with shining golden hair on inner side; abdomen yellow with four narrow red bands, first segment red at base; venter yellow with bases of segments red, and first segment all red; second segment broadly emarginate posteriorly. A Xanthidium, running in the tables near to zebrata and rufula, but quite distinct. It closely resembles NV. crawfordi Cockerell, differing by the broad continuous yellow band on first abdominal segment, where- by it resembles N. rhodoxantha Cockerell. The proportions of the an- tennal joints are as in crawfordi, and after close comparisons I must regard it as a race or variety of that species, rather than an independent species. Swenk refers to a form of crawfordi from Nebraska, with a complete yellow band on first abdominal segment. Swenk suggests that the males which I described as N. gillettei and N. ednx represent varia- tions of crawfordi. I have no gillettei but, on comparing the type of ednx, it seems to me to be specifically distinct, though very closely allied. : The name of this variety is in allusion to the tear-like yellowish suffusion on anterior orbits. Nomada concinnula, new species Cotorapo: 29, Electra Lake (type locality), near Durango, about 8400 ft. alt., June 29, 1919; 29, Pagosa Springs, about 7400 ft. alt., in U. 8. forest reserva- tion, San Juan valley, June 23, 1919. Both of these regions contained oaks, Pinus scopulorum, etc. : Q.—Length about 9 mm.; head, thorax and legs clear red, almost without yellow; sides of face suffusedly lemon-yellow; tubercles and postscutellum inclining toward orange; mandibles simple, black at end; eyes pale reddish; clypeus closely and finely punctured; antenrze long, bright ferruginous, third joint considerably shorter than fourth, but more than half as long; mesothorax finely granular, entirely red; scutellum strongly bigibbous; lower part of pleura deeper, less yellowish, red than the rest; alittle black about bases of cox, and hind femora variably suffusedly blackened on inner side; tegule yellowish red, shining but punctured; wings red- dened, stigma dull ferruginous, nervures fuscous; basal nervure going a little basad of transverse median; abdomen light red with bright yellow bands, on first segment broken into two spots, on second broad at sides but thin and flexuous in middle, varying to much broader; on third, fourth and fifth broad, the last with a pair of pellucid spots; venter with first segment red, the next three with broad yellow bands the fifth with a pair of large round yellow spots, containing a small reddish spot near margin. 1921] WESTERN BEES 7 A pretty little Xanthidium, perhaps related to rufula. Superficially it looks exactly like N. vallesina Cockerell; but that has the fourth antennal joint much shorter, lacks the yellow at sides of face, ete. Nomada carinicauda, new species » Co orapo: 19, along the South Fork of the Rio Grande at about 37° 36’ N., 106° 43’ W., about 5800 ft. alt., among Pinus scopulorum, Pseudotsuga mucronata, Picea pungens, etc., June 17, 1919. @ .—Length about 9 mm., red, similar to N. depressicauda Cockerell (to which it runs in the table in Entom. News, 1908, p. 323), but with the flattened caudal area much larger (describing about half a circle), with a minutely granular or tessellate bare surface, and three longitudinal keels, the keels and the margin briefly and micro- scopically pubescent. It also differs by the darker red of head and thorax; the longer third antennal joint (still, however, not quite so long as fourth); scutellum more strongly bigibbous. The only yellow markings on abdomen are large spots on second segment, and very small ones on third. Nomada vicinalis Cresson, variety infrarubens Cockerell CoLoraDo: 1.7, Telluride, about 10,000 ft. alt., along Cornet Creek trail, July 9, 1919; 1, South Fork of Rio Grande between Pass and Hope Creeks, about 9300 ft. alt., June 18, 1919; 1.7, Electra Lake near Durango, about 8400 ft. alt., June 29, 1919. All of these localities are in forest regions. These agree with Cresson’s description, except that the venter is _ without yellow, except a suffused spot near apex. The mesothorax appears wholly black, but in the Rio Grande and Telluride specimens it is possible to see a pair of very faint red lines. The scutellum has two red spots, but no yellow. The variety infrarubens was described from Oregon, but the Colorado specimens'‘cannot be separated. On the other hand, I find that N. vicinalis aldricht Cockerell, from Idaho, is distinct. Compared with aldrichi, the present insect differs by the shorter third antennal joint and the deep emargination of the black mark on first ventral segment. I think the Idaho insect must stand as N. aldrichi, a separate species. It is a comparatively large form. Nomada illinoénsis Robertson . Cotorapo: 1 9, Camp Creek Ranger Station in the Medicine Bow Range about 41° N., 106° 12’ W., about 8700 ft. alt., lodgepole pine and sagebrush, June 19, 1920. This is referred to zllinoénsis because it appears to agree with the Nebraska form so referred by Swenk, though it has the third antennal joint considerably longer than in what I had considered to be illinoénsis, from Oklahoma. It appears legitimate, for the present, to interpret ulinoénsis in a rather broad sense, recognizing that when the sexes, habits, and genitalia are known, in all probability several valid species will be segregated. 8 AMERICAN MUSEUM NOVITATES [No. 24 Nomada (Gnathias) orophila, new species CoLorabo: 2 o’, 19, Camp Creek Ranger Station in the Medicine Bow Range about 41° N., 106° 12’ W., about 8700 ft. alt., lodgepole pine and sagebrush, June 19, 1920. & (Type).—Length about 8 mm.; head and thorax black, with coarse long white hair, slightly yellowish dorsally; face broad, orbits converging below; eyes gray; mandibles bidentate, bright yellow with dark apex; labrum, clypeus (except upper border), mark beneath eyes, and narrow lateral face-marks (ending on orbits about level of antennz) all bright yellow; scape yellow in front; flagellum bright red, the first few joints blackened above; third antennal joint almost as long as fourth; upper border of prothorax obscurely marked with yellowish; tubercles yellow, anteriorly reddish; scutellum red, flattened, not bigibbous, densely covered with long hair; pleura with a yellow mark, bordered with reddish, in front; tegul shining ferruginous; wings brownish, stigma ferruginous, nervures fuscous; basal nervure going far basad of transverse median; legs red, anterior knees yellowish, hind femora black behind and beneath, except at apex; abdomen bright red, with narrow black bands, which are in the main on the extreme bases of the segments; very large spots on second segment, smaller ones on third, but no other yellow markings; first dorsal segment black basally, and first ventral black in middle. 9 .—Length about 8 mm.; head and thorax rather dusky red, abdomen bright chestnut red, polished; hair of top of head and scutellum strongly reddish; face red with no yellowish tint; middle of front and region of ocelli black, and a black band passes downward from each antenna, invading sides of clypeus; eyes gray; antennz — entirely bright red, third joint perhaps a little longer than fourth; mesothorax red with a broad median black band, and narrow lateral ones, failing posteriorly; meta- thorax red at sides of middle but broadly black in middle and extreme sides; cheeks black with a red postorbital band; legs bright red, the femora with a black spot at base beneath; abdomen with spots on second segment round and rather small, on third nearly obsolete; first ventral with a large bifid black mark, the lobes of which are very broad and obtuse. In Gnathias, the Rocky Mountain males usually differ from those of the Eastern and Northwestern States in the more or less red meso- thorax; but N. orophila, from high in the mountains, has it black. The male orophila falls near N. cuneata Robertson, but is much more robust in every way. The female also falls next to cwneata, but has less yellow on the abdomen. Nomada bella Cresson Ipano: 19, Victor, about 6300 ft. alt., July 11, 1920; 1 9, Giveout, near Montpelier, about 6700 ft. alt., July 7, 1920. Cotorapo: 1 9 (a variation with basal area of metathorax black), along the South Fork of the Rio Grande at about 37° 36’ N., 106° 43’ W., about 8500 ft. alt. Specimens from Costilla County, Colorado, which Swenk in 1912 placed under bella, were transferred by him to schwarzi in 1915. The type of schwarzi is a male from Veta Pass. The female described by me ) 1921] WESTERN BEES 9 under schwarzi was lepida. At present, I probably have the 2 of schwarzi mixed with Rocky Mountain bella and do not know how to separate it. : Nomada caroline Cockerell Cororapo: 29, Julesburg, at Salix flowers along the river, about 3460 ft. alt., June 7, 1920. ; Compared with a Nebraska specimen, one of these differs by lacking the black band down middle of metathorax, and the somewhat longer third antennal joint. On comparing specimens from Virginia and Texas, I find enough variation to include the Julesburg insect, unless caroline as at present accepted should prove capable of subdivision. The other specimen with exactly the same data has the metathoracic band well developed, and in general agrees well with the Nebraska specimen. Nomada perplexa Cresson Wyromine: 1 9, Jackson, about 6300 ft. alt., among vegetation of moderately moist pasture-land type, July 14, 1920. The form without yellow spots on abdomen. N. perplexa is common in the Northern Atlantic States (Pennsylvania, for example), but I did not expect to see it from Wyoming. ’ Nomada siouxensis Swenk Wyomine: 1 9, Jackson, about 6600 ft. alt., among blue spruce, aspen, and other mesophytic plants along Cache Creek, July 15, 1920. Differs from the original description in having small and inconspicu- ous orange spots at sides of second abdominal segment, but evidently this species. It was described from Sioux County, Nebraska. Nomada (Gnathias) heterosticta, new species Ipano: 19, Victor, about 6300 ft. alt., among aspens, roses, etc., on the hills across the river from the town, July 11, 1920. Q.—Length about or nearly 10 mm.; bright ferruginous red, not dark; man- dibles bidentate, black at end; a dusky shade between antennez, region between (but not around) ocelli blackened, and cheeks black behind, leaving a red band as broad as the black; eyes pale red; antenne entirely clear ferruginous, third joint about as long as fourth; mesothorax with a narrow median black band; f2mora without black, except that the hind femora have a dusky stripe beneath, not conspicuous; tegule dull pale reddish, strongly punctured; wings dilute reddish fuliginous, stigma ferru- ginous, nervures dark fuscous; basal nervure going far basad of transverse median, third submarginal cell unusually broad above; abdomen clear red, without black marks at base above or below; sides of second segment with small yellow spots, but on third, in place of yellow marks, are dusky dots. 10 AMERICAN. MUSEUM NOVITATES [No. 24 In my table this runs nearest to NV. grayi Cockerell, but it differs in a number of characters and is especially recognizable by the markings of. the abdomen. Nomada (Gnathias) clarescens, new species CoLoravo: 19, Walden, about 8300 ft., on the sagebrush hills near town, June 17, 1920. | Q.—Length nearly 10 mm.; clear red, the antenne entirely red, the flagellum with a fine pruinose pubescence; eyes red; mandibles bidentate, black at end; a blackish W-like mark about bases of antenne, and the region between ocelli black- ened; cheeks black behind, leaving a very broad red band; third antennal joint about as long as fourth; mesothorax with a narrow black band; middle of metathorax with an elongate black spot; pleura with abundant long pale hair (short scanty hair in N. heterosticta); scutellum strongly bigibbous (so also in heterosticta); tegule ferruginous, rather shining; wings dusky with the usual hyaline space; stigma dusky reddish (smaller and narrower than in heterosticta); basal nervure going far basad of transverse median, third submarginal cell greatly narrowed above; femora marked with black beneath at base; inner face of hind basitarsi with very pale hair; first abdominal segment with a round black spot on each side near base; second with small yellow spots, the rest without yellow; first ventral with a blackish shade, but no well-defined mark. The first ventral segment and other characters readily distinguish it from N. grayt. : Nomada (Gnathias) vulpis, new species Wyomine: 19, Foxpark, about 9100 ft. alt., in the Medicine Bow Range, lodgepole pine, and sagebrush, June 15, 1920 (snow still lying in patches nearby). ¢.—Length nearly 10 mm.; red, with the aspect of N. clarescens, but a little less robust. It is certainly distinct, by the following characters: black about antennz and ocelli much more extensive, and continuing as sutural lines half-way down sides ' of clypeus; flagellum more slender, and dusky above, toward base strongly blackened; median band of mesothorax broader, and a broad median black band down middle of metathorax, including basal area; extreme sides of metathorax broadly black; red band along posterior orbits much narrower; second submarginal cell not so broad, receiving first recurrent nervure about beginning of last third; second abdominal segment with large clear-cut yellow marks; third wholly without spots; first ventral with a black fish-tail mark. Allied to N. bella, but I think certainly not a variety of it. Compared with a specimen of N. maculata Cresson (9 of bella) from Franklinville, Pa., it is considerably less robust, with smaller head, much more black on face, cheeks mainly black (in the maculata red, with a black patch posteriorly, covered with hair and inconspicuous), yellow on abdomen reduced to a pair of spots, surface of abdomen less shining, ete. It evidently approaches Swenk’s interpretation of female N. schwarzi Cockerell, but the probabilities are against its reference to that species. 1921] WESTERN BEES 11 Nomada packardiella Cockerell Cotorapo: 19, Ouray, about 8500 ft. alt., among oak and Pseudotsuga, July 12, 1919; 19, Tennessee Pass, about 10,500 ft. alt., August 7, 1920; 19, Leadville, about 10,300 ft. alt., August 3, 1919, collected by Pearce Bailey, Jr. The type of packardiella, from Boulder, has the fourth and fifth abdominal segments each with a pair of yellow spots. The Ouray speci- men has the spots faintly indicated on the fifth segment, but in those from above 10,000 ft. they have entirely disappeared, though the lateral spots on second and third segments remain. Contrary to expectation, the antennz of these high-altitude forms are of a clearer red than those of the type. The above species of Nomada may be separated by the following table. Scutellum yellow or with yellow markings....................0.2-000 0005 1- PAT AATEC PON Gg 5) SiGe A ota pa es HS. 14. 1. Mesothorax red, with a median black band or mark....................... 2- Mesothorax black, usually red or yellow on lateral margins................. 6- 2. Scape swollen, largely yellow; face lemon-yellow.....................-4.. 3 Sn Teme: GMO hey A ess acs 6S ie has Sa RA TAG ewe as. os 4. 3. Pleura behind tubercles yellow; fourth antennal joint long. calloxantha Cockerell, 9°. Pleura behind tubercles red, with at most a yellow line; fourth antennal joint pee ae ad A aR an ag a vincta heterochroa - Cockerell, <7. 4. Sides of face with broad cream-colored stripes; a very dark cloud in apical field Of SntORIOR Whigs: seo ae ea en Se gutierreziz Cockerell, 9. Ret OF CARR GE So TORTS: tei Sabo ec beac HO Sa snes 5. 5. Smaller; supraclypeal region black with a quadrate or subtriangular red area. morrisoni Cresson, 9. Larger; supraclypeal region red.....................0-. zebrata Cresson, 9. 6. Clypeus shining black; small species.................. crucis Cockerell, Q. Clypeus black, with lower margin and corners red. .melanoptera Cockerell, 9. Clypeus red; facial quadrangle longer than broad.............. vincta Say, 2. Clypeus yellow... ay £ 7. Region below sarees ‘bleak. the black eiditie’ ina pabink next si divpeal’ nienietins scutellum black with two yellow spots; males.:.................... 8. No such black areas ending in a point at sides of clypeus. ................. Y. S.; : Lareeey aes tasinty yoow 2566s cs Fa ay ods Se civilis Cresson. Smaller; legs red and black.................... cymbalariz Cockerell, var. 9. Smaller; face pale yellow or cream-color; a conspicuous apical dark cloud on TE ee ro ee oe ses wi REELS ERTS EROS crucis Cockerell, <. LOrgers ieee rise OE POY ose oasis LEE es Be hve las. 10. 10. Anterior corners of mesothorax yellow. :...............c ee cee cee eeeetes 11. Anterior corners of mesothorax red or black. ..............000 00 cece eeeee rs; 12 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. AMERICAN MUSEUM NOVITATES [No. 24 Lateral face-marks following anterior orbits (which are parallel) to top of eyes; PONDS BOONE: (a2 55:5: si 9s earch eeaGle Sone ae es zebrata Cresson, o. Lateral face-marks not following orbits to top of eyes...............0.000- 12. Larger; face broader; orbits conspicuously diverging above; lateral face-marks not curved mesad at upper end.................. edwardsit Cresson, 9. Smaller; face not so broad; orbits nearly parallel; lateral face-marks curved nesad at upper en 6 i Se Rea citrina Cresson, 9. Area of metathorax with a short yellow band on each side; third antennal joint OUR 5 SEK Sie sak py oo a ee pecosensis Cockerell, 9. Area of metathoraxiall black: ) 550255. 2 ACA ee vincta Say, #. Mesothorax black (often with a pair of very narrow and obscure red lines in vicinalis injrarubens); scutellum red or marked with red.............. 15. Mesothorax red, or red and black. . 2.) 0.2.20. 05 ive eee ess oe Oe Le Lateral face-marks curving away from orbit at top; mandibles simple. vicinalis infrarubens Cockerell, @. Lateral face-marks not curving away from orbit..................00-0000 16. Mandibles simple; basal nervure going very little basad of transverse median. illinoénsis Robertson, o. Mandibles bidentate; basal nervure going far basad of transverse median. orophila Cockerell, . Abdomen with cream-colored bands, continuous on fourth and fifth segments; chypeus: Fee 2). 55 CA ES Cae eae ree accepta Cresson, 9. Abdomen with continuous lemon-colored bands, at least on one of the first three BORMONS. os a FG 5 el OS Ae ee 18. Abdomen without such bands; clypeus not yellow....................4. 21. First abdominal segment with a broad continuous yellow band. crawfordi lachrymosa Cockerell. First abdominal segment without such a band. ..........0..........00005 19. Mesothorax black with four red bands; front black except sides broadly, and a spot below middle ocellus; tubercles and postscutellum yellow. alpha paralpha Cockerell, 9. Mesothorax Feds 5.000 eis cas wenn enswcne gee dv lp ele ent Larger; region of ocelli broadly black........... alpha dialpha Cockerell, 9. Smaller; region of ocelli red, with at most a little black. concinnula Cockerell, 9. A transverse black patch above or between antennze.................0000- 22. No black patch above or between antennez; mandibles bidentate. caroline Cockerell. Abdomen with narrow black bands; mandibles simple. packardiella Cockerell, 9. Abdomen without distinct black bands, or only on one or two segments. . . . 23. A small black spot at each side of third segment near base; no black spots on fourth segment; mandibles bidentate........... heterosticta Cockerell, 9. No such black spots on third segment... ............ 0. cece ence eee eeee 24, Abdomen with a large highly modified caudal area, with three keels. carinicauda Cockerell, 9. Abdomen without such an area; mandibles bidentate...................- 25. 1921] WESTERN BEES 13 25. Basal area of metathorax black; abdomen with large yellow spots on second NN antes sak cas b akiely on ap dou-i.s ab hed Mek AMOy bella Cresson, var Basal area of metathorax red; second abdominal segment with large yellow spots; third antennal joint considerably shorter than in bella, var. clarescens Cockerell. 26. Small rather slender species, about 7 mm. long; second abdominal segment with large yellow spots; scutellum bigibbous............ caroline Cockerell. INI Sue oil sce Phe anes Sng is wa ven a esache's Che $M atR oink ace wei 27. 27. Hind margins of abdominal segments blackish, contrasting with the bright red color of surface before the margins................... orophila Cockerell. Hind margins not blackish or contrasting...............-.02.02 cece eee 28. 28. Smaller; with dusky abdomen; black mark on first ventral segment of ab- domen like a fish tail, with sharp points............ perplexa Cresson, var. EMewOr, Wier CROAT CR BUGUINON . ee tae aco ie clk s fc aee ae dees ees 29. 29. Flagellum blackened above; clypeus and sides of face dark red, with no yel- NEE BOG yet F oa s eis ede disuse ele scs os eahieboedae vulpis Cockerell. Misgollum-eienr red throughout. isn 6. ss sic hoc ie eh bergen os va eee clowns os 30. 30. Lower part of sides of face yellowish; yellow spots on second abdominal seg- MN RNIN i hed a ws 2k 658s uy vn wee eee siouxensis Swenk, var. Lower part of sides of face not at all yellowish; yellow spots on second abdominal I IIa a aa ta eon apiece. cat ede ie bella Cresson. PROTANDRENA Cockerell Protandrena bancrofti Dunning Cotorapo: 1 9, Wray, about 3700 ft. alt., in moist place near the head of Dry Willow Creek, August 18, 1919. Differs a little from the type in having a very fine supraclypeal line just above the clypeus, and a small spot on each side of clypeus. The clypeal mark is trilobed, the lobes acute, like a leaf. Swenk considered this to be a synonym of P. asclepiadis Cockerell, but it has much darker wings and must, I think, be regarded as distinct. CALLIOPsIs Smith Calliopsis coloradensis Cresson Cotorapo: 2 o’, Denver, August 28, 1919, collected by Barbara M. and Marjorie D. Schwarz; 1 <, Wray, about 3700 ft. alt., in moist place near the head of Dry Willow Creek, August 18, 1919; 2 9,9, Boulder, about 5600 ft. alt., between the town and Orodell, August 11, 1919; 19, 16%, Ward, about 9300 ft. alt., near town, August 9, 1919, collected by Pierce Bailey, Jr.; 59, 7, Meeker, about 6200 ft. alt., in town July 21, 1919, collected by Herbert F. Schwarz and Pearce Bailey, Jr. Wyoming: 1 9, near the Thumb of Yellowstone Lake, Yellowstone National Park, about 7750 ft. alt., July 20, 1920. Uran: 49, 607, Salt Lake City, about 5000 ft. alt., near Fort Douglas, July 28, 1920, collected by Mrs. F. E. Lutz. 14 AMERICAN MUSEUM NOVITATES [No. 24 Calliopsis coloradensis fedorensis (Cockerell) The female has the disc of first abdominal segment beset with fine punctures, but I cannot find good characters for the male. One female (Boulder, Cotorapo, on the plains at about 5300 it. alt., August 12, 1919) had the clypeus black except the lower corners and a T-shaped mark. It is only a variant, as three typical fedorensis females were taken at Boulder, with the same data. One female from Salt Lake City, Urag, at about 5000 ft. alt., July 28, 1920, collected by Mrs. F. E. Lutz, can be referred here, but the punctures on the first segment are much coarser, and it is presumably an independent mutation from coloradensis. Professor O. A. Stevens has taken fedorensis at flowers of Grindelia squarrosa, at Fargo, North Dakota. He obtained, with normal females, a variant in which the clypeus is entirely black, except a slender line across its upper border. Calliopsis chlorops Cockerell, 1899, is not to be separated from coloradensis. The male differs from Cresson’s description in having the tibia brown or piceous posteriorly, but this is not even a good racial character. C. coloratipes (Cockerell) is at least a good subspecies; the male has the face creamy white or very pale yellowish, instead of lemon-yel- low, and the female lacks the black bars on clypeus. C. coloratipes occurs in New Mexico and Arizona, in the Middle and Lower Sonoran Zones. Calliopsis rhodophilus (Cockerell) This is the western representative of C. andreniformis Smith. Two males from Ouray, about 8500 ft. alt., July 11, 1919, and three males from Pagosa Springs, about 7400 ft. alt., June 23, 1919, are typical. One fe- male from Estes Park, August 13, 1919, collected by Herbert F. Schwarz, is a variation with face-marks reduced, the lateral marks reduced to small spots. It thus approaches C. teucrit Cockerell, which may be an ex- treme form of rhodophilus; but the mandibles are entirely dark, not bright ferruginous in middle as in teucrit. The abdominal venter lacks the light reddish bands seen in teucrii. The localities just mentioned are in CoLORADO. Calliopsis verbene nebrascensis Crawford CoLorapo: 1.7, Wray, about 3700 ft. alt., dry hills near town, August 17, 1919. 1921] WESTERN BEES 15 HypomacroTeRA Cockerell and Porter Hypomacrotera callops Cockerell and Porter Cotorapo: 5, Regnier, near the state border south of Lamar, about 4400 ft. alt., at Quincula lobata, June 8, 1919. PsEupopanureus Cockerell Pseudopanurgus xthiops (Cresson) Cotorapo: 89, 50’, Wray, about 3700 ft. alt., some at Helianthus, August 18, 1919; 39, 10.7, La Junta, about 4100 ft. alt., August 12, 1920. Uran: 59, 50, Ogden, August 29-30, 1916. One male is stylopised, and differs from normal males in having the clypeus broadly black at sides, the supraclypeal yellow broader than high, the dog-ear marks much shorter, the lateral face-marks and yellow marks on scape wholly wanting. Also, the anterior and middle tibize have the apical half black, and the sculpture of the abdomen is weaker. The stylopid is undescribed. BoMBOMELECTA Patton Bombomelecta fulvida (Cresson) Cotorapo: 1 9, South Fork (near the headwaters of the Rio Grande), about 8200 ft. alt., near the town, June 17, 1919. Arizona: 1 9, Grand Canyon, May 24, 1918. | Bombomelecta pacifica (Cresson) Cotorapo: 1 <, Cheyenne Pass, near Laramie, about 8500 ft. alt., limber pine- Douglas fir country, June 13, 1920; 1 @, Julesburg, about 3460 ft. alt., near the river, at Pentstemon radicosus, June 7, 1920. PsEUDOMELECTA Radoszkowski Pseudomelecta rociadensis (Cockerell) Cotorapo: 1 9, Gardner, September 1918, -collected by Walter Granger (?); 1 9, Wray, about 3700 ft. alt., at head of Dry Willow Creek, August 18, 1919. Both specimens much worn. ; Pseudomelecta miranda (Fox) Cotorapo: 1 , La Junta, about 3100 ft. alt., along the roadside, August 12, 1920, collected by Mrs. F. E. Lutz. Dien Master Suwa ad rare: le os epee spent! $42 tet 6 « ey AMERICAN MUSEUM NOVITATES | No. 25 3 DESCRIPTIONS OF PROPOSED NEW BIRDS FROM CENTRAL AMERICA, WITH NOTES ON OTHER LITTLE-KNOWN FORMS By Wa.pron DeWirr Miter ann LupLow. Griscom LIQ, ~ Issued December 7, 1921 By Orpmr or THE TRUSTEES OF . THE AMERICAN MUSEUM OF NATURAL HISTORY New Yorr Crty AMERICAN MUSEUM NOVITATES Number 25 December 9, 1921 59.82 (728) DESCRIPTIONS OF PROPOSED NEW BIRDS FROM CENTRAL AMERICA, WITH NOTES ON OTHER LITTLE-KNOWN FORMS By Waupron DeWitt MILLER AND LUDLOW GRISCOM As a result of studies made on the distribution of bird-life in Nicara- gua, the authors here propose definite names for a number of birds, to invite criticism pending the appearance of their final report. There is also included a discussion of the status of other little-known Central American birds, belonging to families that will be treated by Mr. Ridgway in the forthcoming part of his monumental work on the Birds of North and Middle America. We have to thank the authorities of the National Museum, and particularly Dr. Charles W. Richmond, for permission to examine material in their collection bearing on one of these cases. Ortalis cinereiceps saturatus, new subspecies Sussreciric CHARACTERS.—Similar to Ortalis c. cinereiceps but darker through- out and averaging smaller. Crown and nape very slightly darker, deep mouse-gray instead of deep neutral gray; the back, wing-coverts, and rump deep olive-brown, instead of medium olive-brown, the contrast greatest on the wing-coverts; primaries dark rufous rather than bright rufous, the dusky tips and centers of the feathers more extensive and noticeable; tail tips grayish buffy rather than gray with whitish margins, the tips of the three outer feathers averaging 2 mm. less in extent; breast darker olive-gray, shading into light buffy olive-gray on the belly instead of almost pure light gray, the contrast best seen on the thighs; under tail-coverts olive-brown, rather than grayish olive. In each series some specimens are much browner on the breast than others. Type.—No. 101063, Amer. Mus. Nat. Hist.; 7 ad.; near Matagalpa, Nicaragua; March 4, 1907; Wm. B. Richardson. SPECIMENS EXAMINED Ortalis cinereiceps saturatus—NIcAaRAGUA: Matagalpa, the type; Las Canas, Matagalpa, 16’; Savala, 1<7; Muy Muy, 1.7; Rio Grande, 19; Chontales, 1 9; Los Sabalos, 1. Costa Rica: Bonilla, 1; Aquinares, 2 7. Ortalis cinereiceps cinereiceps—PaNnama: Boqueron, Chiriqui, 17, 19; Canal Zone, 17,19; Chepigana, East Panama, 1<7. This form is based on 10 specimens from the humid tropical Atlantic forest region of Nicaragua and Costa Rica, and is compared with a series of what may be regarded as typical material from Panama. We have seen 2 AMERICAN MUSEUM NOVITATES [No. 25 no specimens from southern Costa Rica, but it would be surprising if the bird there did not approach the new form. There are slight differences in size, although the measurements of the two races show a considerable amount of individual variation. MEASUREMENTS O. c. cinereiceps O. c. saturatus Wing 38d 196-216. 8c 187-204 29 193 29 187-192 Tail 30 220-232 87 204—220 29 219-222 ' 29 197 Culmen 3c 24-27 8c 24-26 29 23 .5-25.5 29 22-23 Tarsus 3c 65-68 8a 65-67 29 65-70 29 61-62 Creciscus ruberrimus, new species Speciric Cuaracters.—Closely allied to C. ruber of Mexico and Guatemala, but differing in having the chestnut of the upper back extended over the entire upper- parts including the wing-coverts; the back, rump, upper tail-coverts, and wing-coverts rich deep chestnut instead of chocolate-brown; primaries and tail blackish instead of ashy brown; bill shorter and relatively stouter. Typr.—No. 143692, Amer. Mus. Nat. Hist.; 9 ad.; Jinotega, Nicaragua (alt. about 3000 ft.); April 5, 1917; Miller and Griscom. DecsriPTion oF TypE.—General color above rich deep chestnut, darkest on the rump; primaries and tail-feathers blackish, the secondaries deep blackish brown; crown, nape, and sides of face bright slaty gray, sharply demarcated from the chestnut color of the adjacent parts; throat and breast bright chestnut, becoming rich deep chestnvt on the abdomen, belly, and under tail-coverts; thighs externally deep slaty gray, internally bright chestnut; under wing-coverts and axillars bright chestnut, the greater series ashy, edged with chestnut. SpEcIMENS EXAMINED Creciscus ruberrimus.—NicaraGua: Jinotega, the type. Creciscus ruber.—Mexico, 1; GuATEMALA, 1, MEASUREMENTS Nicaragua Guatemala Mexico Wing 79 78 78.5 Tail 36 — 39 Culmen 18 21 21 Depth of Bill at Base 7.5 7.5 7.5 Tarsus 31 34.5 34 Middle Toe and Claw 37.5 39 40 1921] NEW BIRDS FROM CENTRAL AMERICA 3 This beautiful little Rail was collected in a surprisingly arid habitat, which is separated from that of C. ruber, its close ally, by a wide stretch of such mountainous and broken country that intergradation does not seem at all likely. For the present we prefer to retain the old generic name, especially as Mr. Ridgway, the describer of Thryocrex, has not seen several im- portant species. Until the exact limits of Porzana and Creciscus are determined by a critical examination of all the species involved through- out the world, there seems little point in proposing segregates from either. Our treatment must not, however, be taken as our opinion of the proper generic relationships of the two species here discussed, our material of this group of Rails being entirely inadequate. Gallinula chloropus centralis, new subspecies SuBsPEcIFIC CHARACTERS.—Similar to G. c. cachinnans, but decidedly darker and slightly smaller, the gray of the upper back and breast slightly darker, lower back much darker, “mummy brown,” instead of “argus brown” or “Brussels brown,”’ becoming blackish brown on the rump and upper tail-coveris. Typre.—No. 143693, Amer. Mus. Nat. Hist.; ad. 9; 12 miles south of Metapa, Central Nicaragua; April 25, 1917; Ludlow Griscom. SprecimEns EXAMINED Gallinula chloropus centralis—Nicaracua: Metapa, the type; Tipitapa, 1°. Gallinula chloropus cachinnans.—EAsTERN UNITED Statss, 21 <7, 209. MEASUREMENTS Culmen to Wing. Hind Edge of Tarsus Frontal Shield Nicaragua 2 9 163 .5—168 (165.7) 45-46 (45.5) 50-51 (50.5) Eastern U.S.209 164 -176(170.2) 40-47 (44) 50-57 (54.1) The two adult females listed above are so distinct from a large series of cachinnans that we have no hesitation in describing them. Birds seen at Los Sabalos by Nutting and specimens taken by Holland at Greytown probably belong to this race, but we are unable at this time to state its range more accurately. It is interesting to note that the proposed new race is not in any sense intermediate between cachinnans and pauzilla from western Colombia. The latter is by far the smallest, but is lighter on the back than even cachinnans. There seem to be but single records of the Florida Gallinule from Costa Rica and Panama. It would be interesting to determine accurately the specimens on which they are based. a AMERICAN MUSEUM NOVITATES [No. 25 We follow the latest authorities on the genus, Bangs and Hartert, in treating the American Gallinules as races of the Old World chloropus, but we are not convinced that this is the best course. While our material is not entirely satisfactory for a revision of the genus, we are strongly of the opinion that sandvicensis and garmani are distinct species, and that cer- ceris is a valid race. Asturina plagiata micrus, new subspecies Supspeciric CHaractrers.—Similar to A. p. plagiata, but much smaller, the adults averaging slightly darker, with narrower tail-bands, the immature birds noticeably darker. Typr.—No. 143746, Amer. Mus. Nat. Hist.; ad. o&; 4 miles northeast of Chinandega, Nicaragua; June 12, 1917. SPEecIMENS EXAMINED Asturina plagiata micrus.—NICARAGUA, adults, 4 By Orprr or Tae Trousters | atin THE AMERICAN MUSEUM OF NATURAL HISTORY ~ “New Yorx City ; ok % i el AMERICAN MUSEUM NOVITATES Number 27 December 28, 1921 59.82(8) ; DESCRIPTIONS OF PROPOSED NEW BIRDS FROM BRAZIL, PARAGUAY, AND ARGENTINA By Grorce K. CHEeRRIE AND (Mrs.) Etsre M. B. REICHENBERGER The following descriptions of proposed new birds are based chiefly on birds contained in the Roosevelt Collection made by Mr. Cherrie in the years 1913-1916. They are here published in advance of a complete report on the collection. Strix chacoensis, new species Speciric CHARACTERS.—Similar to Strix rufipes but general color above and bars below darker; blackish, instead of chocclate-brown, the primary coverts distinctly barred with rufous on both webs, and with facial ruff dusky, barred with grayish white, instead of chocolate-brown or fulvous. Type.—No. 149,387, Amer. Mus. Nat. Hist.; o ad.; Fort Wheeler, Paraguay- an Chaco, Paraguay; September 28, 1916; George K. Cherrie. DeEscrIPTION OF Typr.—Above, dusky brownish black, narrowly barred with white and yellowish buff; the white bars, subapical on the feathers and separated from the basal yellowish buff ones, which are mostly concealed by a rather broad brownish black band; the feathers of the hind neck are long and fluffy; the center of the crown is darker than the upperparts, the blackish tips to the feathers being breader and concealing the white bars; median and greater wing-coverts distinctly but irregularly barred with pale tawny buff and notched on the outer webs with white; on the greater coverts these white tips forming a narrow band; the primary coverts dusky brownish distinctly barred with deep ochraceous buff; the white bars on the seapulars very distinct; quills dusky brown barred (broadly on the primaries, nar- rowly on the secondaries) on both webs with ochraceous orange but shading into light ochraceous buff on the edges of the outer webs; tail dusky brown crossed with six bars of pale ochraceous buff that shades into buffy white near the edges of the feathers; Icral plumes silvery white but with the shafts of the plumes developed into black hair-like bristles; remainder of the facial disk dusky grayish white with con- centric bars of dusky brown; ear-coverts grayish white with nearly obsolete dusky brown bars and silvery white shaft-lines; under surface of body barred, blackish brown and white; middle of belly, sides of flanks, thighs, and tarsi ochraceous buff; under tail-coverts light ochraceous buff tipped with white and having a subterminal black band followed by a white one; under wing-coverts and axillaries ochraceous buff, a few of the lowermost outermost dusky brownish; the basal third (more or less) of the inner webs of the primaries and secondaries uniform pale ochraceous buff. <: wing, 251; tail, 146; culmen, 33 mm. “ey ane 2 - AMERICAN MUSEUM NOVITATES [No. 27 SPECIMENS EXAMINED Strix chacoensis.—PaRaGuay: Paraguayan Chaco, Fort Wheeler, 1 @ (type). Strix rufipes—Cui1: Maquehue, Temuco, Cautin, 14,1 ¢. Although we have only one specimen of this form its characters are so well marked that we have no hesitation in describing it as a new species., Strix rufipes has been recorded by Dr. Dabbene from the western part of the Province of Jujuy, Argentina, but whether the record is based on true rufipes or the bird here described we are unable to say. Ortalis canicollis pantanalensis, new subspecies Susspeciric Cuaracrers.—Similar to Ortalis canicollis canicollis of the Para- guayan Chaco, but uniformly darker both above and below; the chestnut color on the outer rectrices more restricted and the head dusky brown; top of the head dusky brown shading into olive-brown on the neck and becoming grayish in the supraloral and supra-auricular regions; bristle-like feathers of the anterior part of the cheeks and chin black, shading into the gray of the sides of face and throat; back of neck, mantle, wings, center of lower back and upper tail-coverts olive-brown with slight bronze cr purplish gloss; under wing-coverts, sides, flanks, and under tail- coverts chestnut; chest and breast brownish or olivaceous gray, washed with chest- nut on the belly; three outer pairs of tail-feathers broadly tipped with chestnut, the distal third of the outer pair being so colored. g: wing, 239; tail, 252; culmen, 25; tarsus, €6 mm. Typr.—No. 127,232, Amer. Mus. Nat. Hist.; 9, ad.; near mouth of Rio San Lorenzo, Matto Grosso, Brazil; December 26, 1913; George K. Cherrie. j Ortalis canicollis grisea, new subspecies Susspeciric CHARACTERS.—Very similar to Ortalis canicollis canicollis of the Paraguayan Chaco but smaller, the throat dark gray, breast and chest gray with an olivaceous wash; tail bluish green, instead of brownish green, the two outer pair only of tail-feathers tipped with chestnut; sides and flanks pale dusky rufous instead of chestnut as in Ortalis c. canicollis; top of head uniform dark gray shading into olive- gray on the nape and becoming brownish olivaceous on the back with a slight bronze gloss; lower back and rump washed with chestnut rather browner down the middle; throat dark gray, breast and chest gray with an olivaceous wash; abdomen paler gray with faint rufescent wash; under tail-coverts chestnut; flanks and sides pale — dusky rufous; the two outer pair only of tail-feathers tipped with chestnut. ¢@: wing, 210; tail, 238.5; tarsus, 58; culmen, 23 mm. Typr.—No. 140,257, Amer. Mus. Nat. Hist.; 9, ad.; Suncho Corral, Santiago del Estero, Argentina; April 22, 1916; Miller and Boyle. , Specimens EXAMINED _ Ortalis canicollis grisea.—ARGENTINA: Suncho Corral, Santiago del Estero, 800 feet, 1 Q (type). Ortalis canicollis canicollis —PAaraGuay: Fort Wheeler, 1<,1 9. ARGENTINA: Embarcacion, Prov. of Salta, 1700 feet, 1 ¢@ (intermediate between grisea and © canicollis). — Oe os — -— = - 1921] DESCRIPTIONS OF PROPOSED NEW BIRDS 3 Ortalis canicollis pantanalensis—Braziu: Palmiras, Rio Taquary, 1 9; Rio San Lorenzo, 1 ¢@. Ortalis ruficauda.—VENEZUELA: Cristobal Colon, ‘Paria Peninsula, 1500 feet, 2 #,2 2; Tucacas, Estado Falcon, 2 <7. This group shows considerable geographic variation. The bird from the Rio San Lorenzo, Matto Grosso (Ortalis c. pantanalensis), is the brown extreme; the bird from Suncho Corral, Santiago del Estero, Argentina, is the gray extreme of the species. Specimens from Fort Wheeler, Paraguay (Ortalis c. canicollis) and a specimen from Embarca- cion, Prov. de Salta, Argentina are intermediate in color. We believe, however, that the material at hand warrants the recognition of three races. These birds are found in forested areas. Ortalis c. pantanalensis was found in the forests which border the watercourses traversing the plains. Ortalis c. grisea and Ortalis c. canicollis (intermediate) were found in a rough or broken highland forested country. In going from the Paraguayan Chaco into the pantanal region of Matto Grosso, there is a marked difference in the brush and tree flora. Key to THE Ortalis ruficauda—canicollis Groupe A. Only middle pair of tail-feathers without rufescent tips... ..Ortalis ruficauda, AA. Six or eight central tail-feathers without rufescent tips. - b. Head gray. ce. Six middle tail-feathers without rufescent tips; sides dark chestnut, this color nearly meeting across the rump, wing larger, 9 , 227.5 CIE ES TRUE ie Pi SEE EM Ortalis canicollis canicollis. ec. Eight middle tail-feathers without rufescent tips; sides pale rufes- cent, wing smaller, 9,210 mm........ Ortalis canicollis grisea. bb. Head brown. e. Six middle tail-feathers without rufescent tips. Ortalis canicollis pantanalensis. Nystactes tamatia interior, new subspecies Susspeciric CHARACTERS.—Similar to Nystactes tamatia tamatia but differing . from all described races of Nystactes tamatia in its longer wing, in being less heavily spotted, and in having no black spots on the center of the abdomen. <: wing, 78; tail, 61.5; culmen, 22.5 mm. Typr.—No. 127,486, Amer. Mus. Nat. Hist.; 1 @; Campos Novos, Cerro do Norte, Matto Grosso, Brazil; February 16, 1914; George K. Cherrie. SPecIMENS EXAMINED Nystactes tamatia interior—Brazit: Campos Novos, Cerro do Norte, Matto Grosso, 1 o,1 9, (inel. type); Tapirapoan, 1 °. Nystactes tamatia tamatia.—BritisH GUIANA: 2 ?. Nystactes tamatia hypnaleus.—Braziu: Santarem, 2?,1 °. Nystactes tamatia pulmentum.—Ecuapor: Napo, 1 ?. 4 AMERICAN MUSEUM NOVITATES [No. 27 MEASUREMENTS Nystactes tamatia tamatia _ Sex Wing Tail Culmen British Guiana ? 72. 64. 24 “Guiana” ? 75. 60.5 23 Nystactes tamatia hypnaleus Brazil: Santarem 2 qi. 66. 26 Hs ig ee 3 74. 62. 24. ‘A “ Q 79. 69. 25.5 Nystactes tamatia pulmentum Ecuador: Napo ? 75. 59. 24 Nystactes tamatia interior Brazil: Matto Grosso, Campos Novos rot 78. 61. 22.5 c &c 73 cc ‘“ fe) 82.5 68.5 25 ef e “ Tapirapoan Q 80. 67.5 23 The discovery of this race extends the known range of its species. We have seen no specimens from Cayenne, the type-locality of N ystactes tamatia tamalia, but we presume. material from British Guiana to be essentially topotypical. Nonnula ruficapilla pallida, new subspecies Suspspeciric CHARAcTERS.—Similar to Nonnula ruficapilla ruficapilla, but generally paler, the gray of the sides of the head and throat not encroaching on the breast and much less extended on the sides; middle of belly pale ochraceous buff, upperparts and tail olive-brown. ¢: wing, 61.5; tail, 63; culmen, 22.5 mm. Typre.—No. 127,126, Amer. Mus. Nat. Hist.; 9 ad.; Tapirapoan, Matto Grosso, Brazil; January 14, 1914; George K. Cherrie. SprciMENs EXAMINED Nonnula ruficapilia pallida.—Braztt: Matto Grosso, Tapirapoan, 2 o, 1 9, (ine!. type). Nonnula ruficapilla ruficapilla—Prrv: Tulumayo, 4000 ft., Prov. Junin, 2 #, LO ies It is obvious that this subspecies is a pale representative of the Peruvian bird. Chloronerpes flavigula magnus, new subspecies SuspsPEciric CHARACTERS.—Similar to Chloronerpes flavigula flavigula, but larger throughout, especially in the length of the wing, and with a heavier bill. Q: wing, 122; tail, 63; culmen, 21 mm. Typr.—No. 127,495, Amer. Mus. Nat. Hist.; 9 ad.; Monte Cristo, Matto Grosso, Brazil; March 18, 1914; Leo E. Miller. SPECIMENS EXAMINED Chloronerpes flavigula magnus.—Brazit: Matto Grosso, Monte Cristo, 1. 3, 1 9, (inel. type); Santarem, 1 9. 2 ( ewe ee ee eee 1921] DESCRIPTIONS OF PROPOSED NEW BIRDS 5 Chloronerpes flavigula flavigula.—Britisu Gurana: 1 o&, 19; Tumatumari, Potaro River, 250 feet, 3 #; Potaro Landing, 1 9; Rockstone, Essequibo River, 150 feet, 2 9. Venezureta: La Union, Caura, 2<7, 2 9; Suapure, 1 9; foot of Mount Duida, 1 ¢. MEASUREMENTS Sex Wing Tail Culmen Chloroner pes flavigula flavigula British Guiana: Tumatumari, Potaro River rot PALS: ee 5 20. British Guiana: Tumatumari, Potaro River fet 113.5 64.1 18.5 British Guiana: Tumatumari, Potaro River of 117.5 61.1 20. British Guiana of 114. 65. AZ: rh s Q 117.5 66. 19. i - Rockstone, Essequibo River Q 113. 64.5 19.5 British Guiana: Rockstone, Essequibo River 2 114.5 63. 19. British Guiana: Potaro Landing Q £3: 60. 18.5 Venezuela: La Union, Caura rot 114. 66. 19. + is = of rot 112: 58. 19. sf bi cp mn Q 114. 65. 19.5 ce “ “c isd Q 109 ; 60 : 18 : 2 : Suapure Q 117.5 66.5 19. ni foot of Mount Duida fe) 117.5 65. 19.- Chloronerpes flavigula magnus Brazil: Monte Cristo ref 121. 62. 21. AE a rd fe} 122. 62. 21. “ Santarem - 9 123. 64. 18.5 We have seen no birds from Cayenne, the type-locality of Chloro- nerpes f. flavigula, but we believe material from British Guiana to be essentially topotypical. We refer the specimen from Santarem to Chloronerpes f. magnus. The wing measurement of this bird is greater than in any other specimen, although the culmen is intermediate in size. Furnarius rufus paraguaye, new subspecies Susspeciric CHARACTERS.—Similar to Furnarius rufus rufus. The upperparts slightly darker, crown browner; and forehead deeper rufescent, size much smaller. o: wing, 87.5; tail, 65.5; culmen, 19 mm. Type.—No. 149,516, Amer. Mus. Nat. Hist.; = ad.; Puerto Pinasco, Para- guay; September 5, 1916; George K. Cherrie. 1Tail moulting. 6 - AMERICAN MUSEUM NOVITATES — Sail EXAMINED | 1d. Furnarius rufus commersoni. See Chipaiiad Matto ‘Gree, Urucum, near Corumb4, Matto Grosso, 3 7, 4 9. ArGEentina: Pericc Jujuy, 1 &*; Embareacion, Prov. of Salta, 2. me MEASUREMENTS ee Furnarius rufus paraguayz se Siege Wing ~— Tail : Paraguay: Puerto Pinasco, Rio Bare: ee guay roe 87.5 ss Trinidad fof 93. Furnarius rufus rufus Argentina: Buenos Aires ; 30 97-113 Uruguay, Montevideo 2 ¥ 101. A series from the mountains of Bolivia, Parataai p Cochabamba and from the vicinity of Sucre, Does (alt. 8600-8700 feet), as well as a series from Argentin Jujuy and Province of Salta (alt. 4000 teet), are interm 3 Furnarius r. rufus and Furnarius rufus commersoni, for nape as rufescent as F. rufus commersoni, and the color of th nearer to F. rufus rufus. een « These birds are found chiefly at the saan’ of the h areas that separate the open country cultivated or otherwi to show a decided preference for building their nests i in or houses, and are quite fearless. a + ya eT ee eee eee se by sen ~ 7 _ AMERICAN MUSEUM NOVITATES No. 28 NEW SPECIES OF AMMONITE OPERCULA | FROM THE MESOZOIC ROCKS OF CUBA By Marjorre O’ ConneLy - By ORDER OF THR TRUSTEES ~ OF THE AMERICAN MUSEUM: OF NATURAL HISTORY ; New Yorx«:Crry AMERICAN MUSEUM NOVITATES Number 28 December 30, 1921 56.4,53 (116:729.1) NEW SPECIES OF AMMONITE OPERCULA FROM THE MESOZOIC ROCKS OF CUBA By MARJORIE O’ CONNELL INTRODUCTION Among the material collected by Mr. Barnum Brown in 1919 in the Province of Pinar del Rio, western Cuba, are seventeen specimens and numerous fragments of the small calcareous, bivalved shells which are known as aptychi. They are the opercula with which the ammonites closed up the aperture of the conch after the animal had withdrawn into the body chamber. So far as I know there has been only one occurrence noted of an Aptychus in American Mesozoic formations. Castillo and Aguilera described Aptychus mexicanus from the Jurassic rocks of Sierra de Catorce, San Luis Potosi, Mexico,! and this seems to. have been a sporadic occurrence. But in Cuba there is a definite horizon which has ~ been traced for several miles and at which nothing but aptychi and an occasional small Haploceras are found. The rocks consist of alternating shales and limestones with the former predominating. They are dark ~ gray but weather light brown or almost white and show a considerable amount of iron stain. The aptychi occur on the surfaces of the shale and are well preserved, although very fragile on account of the iron impregnation. The Aptychus beds were discovered in the heart of the Organ Mountains in outcrops along two stream beds, the Rio San Cristobal and Rio Hondo, northeast of the town of San Cristobal and northwest of Candelaria. From the field relationsand the general stratig- raphy, these strata appear to be of Upper Jurassic age, but the aptychi themselves possess certain characteristics which have heretofore been found only in Cretaceous species. Until further field work can be done it is impossible to state the exact age of the beds but it must be either Tithonian or Neocomian with the paleontological evidence strongly favoring the latter. iCastillo, Antonio del and Aguilera, Jose G. 1895. ‘Fauna Fosil de la Sierra de Catorce, San Luis Potosi.’ Boletin de la Comision Geolégica de México, I. 2 AMERICAN MUSEUM NOVITATES [No. 28 HISTORY OF INVESTIGATIONS AND CLASSIFICATION OF OLD WORLD APTYCHI Hermann von Meyer in his paper ‘Das Genus A ptychus’ (1831), proposed the generic name A ptychus for the paired plates found within the body chamber of many species of ammonites. While he did not recognize the true significance of the plates, he distinguished correctly between two different types which he described as A. levis and A. im- bricatus, each with two varieties. These have since been recognized as representatives of two large groups, the Cellulosi and the Imbricati, respectively. There is no monographie work on Apiychus, the descriptions of species being scattered through the literature in short papers. Since the opercula are often found in formations containing few or no ammonites, it would be exceedingly valuable if all of the data concerning the geo- logical range of every known species of A ptychus, as well as the ammonite genera to which they belong could be brought together. Such a study would have to be undertaken in Europe where A ptychus beds are com- mon, for in America up to the present few representatives of this genus have been found and material is not available for carrying on researches on the subject. On page 12 below is given a short and by no means complete bibliography of the more important references dealing with the morphology and taxonomy of aptychi and the stratigraphic range of Jurassic and Cretaceous species. The systemic position of the bivalved Aptychus plates was formerly a subject for much discussion and a large amount of the early literature was devoted to it. The various theories advanced concerning the type of organisms to which the aptychi belonged may be found in the contribu- tions by Parkinson (1811), Meyer (1831), Voltz (1837), Coquand (1841), and Quenstedt (1849). It is now generally accepted that the aptychi are ammonite opercula, as is indicated by their position in situ in the body chamber in many specimens. Oppel has figured and described, from the Solnhofen beds of Bavaria, a large number of ammonites with aptvchi in the body chamber or at the aperture (see reference on p. 13 for Oppel 1863) and other authors have recorded sporadic occurrences of aptychi in situ (e.g., Meyer, 1831, Retowski, 1891, 1893). Aptychus is not a genus; it is simply a convenient term for referring to opercula when the ammonites to which they belong are unknown. When an aptychus is found in situ in an ammonite it does not receive a separate specific name but is known as the aptychus of that particular genus and species of ammonite. But when aptychi occur alone, as they 1921] NEW SPECIES OF AMMONITE OPERCULA 3 frequently do, with no ammonites in the same stratum, then for con- venience in referring to them they are given specific names. If an aptychus thus named is subsequently found to be the operculum of an ammonite which had previously been de- scribed, the aptychus should be called by the specific name of the ammonite, while the specific name which had been used for the aptychus would become asynonym. Each species of ammonite that possessed an oper- Fe 1s Had eatin culum in all probability possessed a distine- of the shell of Aptychus pro- tive type and, conversely, when different fundus Voltz, showing the types of aptychi are found we may assume _ three shell layers (a, 5, ¢). - ified. U that they belonged to separate ammonite tHe Cue ae Teseghini oe spe cies. Bornemann from Zittel). When aptychi are sectioned they are found to be made up of three shell layers of which the inner and outer are very thin and dense, while the middle is thick and cellular or tubular (Figs. 1, 2, 4, 6). The outer and inner layers are frequently destroyed during fossilization, leaving Fig. 2. Aptychus levis Meyer, the aptychus of a species of Aspidoceras. A. Interior view of a valve showing the growth-lines. B. Exterior of same; A-h, junction line... C. Inner edge of the valve; f, junction surface. D. Vertical section through the valve along the line s-s, of A; 7, growth-lamelle merging ex- ternally (e) but thickening toward the interior (l). Upper Jurassic, Kimmeridgian, Le Havre, France. (After Steinmann.) the more resistent middle one. Aptychi increase in size by the addition of concentric bands of shell around the periphery. The edges of these bands, covered only by the thin outer layer, may show on the convex surface like the ends of off-lapping shingles or they may bend over each other so closely as to show only concentric lines on the external surface. + AMERICAN MUSEUM NOVITATES [No. 28 The internal surface is marked by crowded growth-lines, which are some- times crossed by radial lines from the apex. A classification of aptychi has gradually been evolved and, as added to and emended by Zittel (1885, pp. 402-403), now includes seven groups _ based on the shape and thickness of the shell, the character of the surface layer and of the lamelle. The three most abundantly represented groups in the Upper Jurassic and Lower Cretaceous (Neocomian) are the Cellulosi, Imbricati and Punctati. The Cellulosi are forms with thick curved plates having a fine porous, convex external surface and a concave internal surface covered with fine growth-lines (Fig. 2). The Imbricati have thinner shells and usually the small posterior end is very thick. The convex surface is covered with shingle- like lamelle, the edges of which appear like folds in some cases or like sharp knife-edges in others, the crests being separated by depressions which are rounded or angular (Figs. 1 and 3). The inner sur- face has concentric growth-lines. The aptychi of this group are found in various species of the Oppe- Fig. 3. Apiychus ljinee in the Upper Jurassic, especially at Solnhofen lamellosus Voltz. nq in the Lower Cretaceous and are also found Exterior view. Upper , é ; * Jurassic, Solnhofen, Ba- isolated in the various Aptychus shales and lime- (After Zittel.) stones. The Punctati (Fig. 4) are similar to the Imbricati except that the lamellz overlap more closely and the Fe are bent over until they cover up the intervening hollows; the edges of the crests are marked by lines of punctz on the external surface. The outer shell layer is thicker than in the Imbricati and the distinct- ness of the lines of puncte depends upon the | degrée to which erosion of this layer has pro- Fig. 4 Aptychus puncla- 3 tus Voltz. gressed, the puncte being etched out when “y_ 4:31 section showing thtes : Snes aes H 1 ia shell layers (a, b, c,). _Greatly_en- the amount of erosion is slight, but if this re Nepean a eee continues the lamelle are worn smooth. nemanin from Zittel ) Illustrations of these types maybe found in the works cited below, especially Quenstedt (1849) and Zittel (1885). The Aptychus shale of Cuba contains three species all belonging to the Imbricati and all well represented. No other fossils have been dis- covered in this formation except a few small ammonites belonging to the Se ee 1921] NEW SPECIES OF AMMONITE OPERCULA ‘5 family of the Haploceratide. The determination of the age of the beds therefore depends upon the evidence offered by the aptychi themselves without the aid of the ammonites to which they belonged. In Europe aptychi have been found throughout the Jurassic and Cretaceous of many regions. In certain localities, particularly in Bavaria in the Solnhofen beds, they are usually found zn situ in the body chamber of the ammonite shell. Elsewhere they occur segregated in beds by themselves, as is the case in the Tithonian A ptychus limestones of the Alps and Apennines and in the Eo-Cretaceous or Neocomian Aptychus limestones of the eastern Alps. The Imbricati, those semi-elliptical to quadratic tee with over- lapping lamellze on the convex surface and fine concentric lines on the concave surface, are the most abundant and widespread. They occur in great numbers associated with conchs belonging to the subfamily of the Oppeliine and to the family of the Haploceratide, and also alone. The Jurassic Imbricati for the most part have the lamelle parallel to the curved periphery of the outer edge of the aptychus, as in A. lamellosus - Voltz from the Solnhofen beds (Fig. 3). The lamell, in other words, extend in a continuous curve from the anterior to the posterior end, without bending back toward the apex and forming an angle on the posterior slope. All of the Jurassic Imbricati of this type may be united into the A. lamellosus group, including the following forms as represent- ative species: ' Aptychus crassicauda Quenstedt Middle White Jura A. sparsilamellosus Giimbel Upper Jurassic (Acanthicus beds) A. lamellosus Voltz Upper Jurassic (Solnhofen beds) A. profundus Voltz (=A. imbricatus Meyer) Aptychus of Oppelia bous (Oppel) ~ “ O. euglyptus (Oppel) a e 2 “QO. lithographica (Oppel) ; (a3 cc O. thoro (Oppel) : “cc “cc “cc “é The Cretaceous Imbricati are distinguished from those of the Juras- sic in having the lamelle bent toward the apex as they approach the line of junction (Figs. 5,6). Insomespecies, notably A. didayi Coquand, there is an acute angle formed on the posterior slope. These forms constitute the A. didayi group, including as representative species: Aptychus angulocostatus Peters Lower Cretaceous (Neocomian) A. didayi Coquand = " ro A, lineatus Peters < “ fs A. pusillus Peters A. seranonis Coquand A. undatocostatus Coquand A. insignis Hébert Upper Cretaceous (Meestrichtian) “ “ “ “ce “c “ 6 AMERICAN MUSEUM NOVITATES [No. 28 The same distinction exists between the Jurassic and Cretaceous Punctati, the former having the lamelle essentially parallel to the outer curved periphery, the latter having them bent into an angle along the pos- mentioned: Aptychus punctatus Voltz and A. beyrichi Oppel; of the Cretaceous species: A. rugosus Sharpe from the Mestrichtian of England. Pas ge The species found in the Aptychus shales north of cunadie San Cristobal in Cuba all belong to the Imbrieati and to Natural size. the didayi rather than the lamellosus group. In all of Lower Cretaceous, Neocomian, Central them the lamelle are bent more or less sharply on the rance. é£ er F ice . . . . tet. posterior slope and turn towards the line of junction of the plates, meeting it either at a right angle, as in A. cubanensis, or at an acute angle toward the apex, as in A. pimientensis and ) 1921] NEW SPECIES OF AMMONITE OPERCULA 7 DESCRIPTION OF NEW SPECIES FROM CUBA Group of the Imbricati Aptychus cristobalensis, new species Figures 7 and 8 DESCRIPTION OF THE HoLotyPe (Amer. Mus. No. 19017/3).—The shell is calcare- ous, thin at the apex (anterior) and thick at the narrow distal (posterior) end. The length of the outer junction line is 29 mm., the ‘greatest width 17 mm. and the cord of the peripheral arc from the distal to the proximal ends is 30.7 mm. The broad truncated end is about 10 mm. long. At the apex the three shell layers together are only 0.3 mm, thick; at the distal end the thickness is 6.0 mm. The curved periphery is bounded by a narrow smooth shell band, which widens toward the distal end where it is perpendicular to the convex surface of the plate. Each half of the aptychus is nearly flat in the early growth stages. It becomes gently convex along the short diameter but extremely convex along a radial line at about one-third the distance between the long and short diameters. This convexity is so pronounced as to give rise to a ridge from which the shell slopes down abruptly to the external junction line (i. e., the long diameter). At the apex of each plate there are numerous, almost invisible growth-lines, each of which marks the edge of a lamella. Each additional growth-band is broader than the preceding but the first fifteen lamelle are so narrow as to be scarcely distinguish- able to the naked eye. They are arranged like shingles except that they off-lap instead of overlap away from the apex. The last ten lamell are broad but only a few are continuous from end to end of the shell. The last four or five wedge out along the periphery, each succeeding one being shorter. It is in this manner that the shape of the aptychus is determined, the broadest part being in the region where all the lamel- le are present, the narrow end being occupied by only a few. The lamelle become very thin toward the truncated proximal end and terminate nearly at right angles thereto, there being a narrow smooth band at the edge. In the opposite direction, the lamelle show a slight offset toward the apex as they approach the convex radial ridge, and after they cross it they curve back toward the apex, ending abruptly at the junction line. This flexure is angular until the aptychus is about half-grown, the angle being slightly less than 90°. On the remainder of the shell the flexure looses its sharp angularity, becoming rounded and more open with a divergence of 110—120°. The inner, concave side of the aptychus is covered with fine concentric, evenly- spaced growth-lines about 0.1 mm. apart (Fig. 7). DescriPTION or PARATYPE No. 19017/2.—Length 30 mm., greatest width about 16 mm., cord of the peripheral are 33.2 mm. long. Truncated proximal end about 10 mm. long, thickness of distal end 5.5 mm. Outline of two plates in juxtaposition is semi-elliptical. The junction surface is smooth and at right angles to the convex shell surface. It is triangular in shape with its apex at the point of origin of the plate whence it increases to a width of 3.2 mm. at the distal end. It is faintly concave but the three bounding lines are straight. Between the junction plate and the inside or concave surface of the aptychus there is a bevelled edge (Fig. 8). NUMBER OF SPEcIMENS.—Holotype and 3 paratypes. AssociaATED Sprecires.—Aptychus pimientensis, new species, at locality No. 8. C. 3. 8 AMERICAN MUSEUM NOVITATES [No. 28 Loca.it1eEs.—Northeast of Mt. Pimiento in Rio Hondo, 7 miles northeast of San Cristobal (locality No. 8. C. 3) 3 specimens, Nos. 19017/1, 19017/2, 19017/3 (holotype); on Finca of Rafael Begoa, 9 miles north of San Cristobal in Rio San Cristobal (locality No. S. C. 2) 1 specimen of immature individual, No. 19017/4. Figs. 7 and 8. Aptychus cristobalensis, new species. . Fig. 7. Oblique side view of one valve showing curvature of lamelle on pos- terior slope. The interior of the other valve is shown at the right, the valve being thrust forward. .Natural size. Holotype, A. M. N. H. No. 19017/3. , Fig. 8. Top view of both valves in position. Natural size. Paratype, A. M. N. H. No. 19017,/2. Comparisons.—This species belongs to the group of which A ptychus didayi Coquand is the most widespread European representative. It is, however, relatively narrower than that species and has much finer lamel- le, while the pronounced flexure of the lamelle of the fullgrown shell is rounded and not angular. (See Coquand, 1841, p. 389, Pl. rx, fig. 10, and Quenstedt, 1842, p. 314, Pl. xxu, fig. 21.) It agrees very closely with the description of A. angulocostatus Peters (1854, p. 441) but that author did not give an illustration of his species and the European types are not accessible to me, so that it would be unsafe to identify the Cuban — form with that species. A. cristobalensis probably belongs to a species of Oppelia. Horizon of Related European Species.—A ptychus didayt Coquand is a characteristic Lower Cretaceous (Neocomian) species occurring throughout the Mediterranean province in southern France, in the east- ern Alps, in Austria and northern Africa. (Compare Fig. 7 with Fig. 5.) Aptychus angulocostatus Peters is found associated with A. didayi in the Neocomian Wienerwald sandstone of Austria and Bohemia (Peters, 1854, p. 441; Paul, 1889, pp. 59, 175). : 1921] NEW SPECIES OF AMMONITE OPERCULA 9 Aptychus cubanensis, new species Figures 9 to 14 DescriPTION OF THE HoLotyPE (Amer. Mus. No. 19018/2).—Length of the junction line 17 mm.; greatest width of one valve, 9mm. On the posterior slope the lamellz curve to meet the external line of junction at right angles and assume their maximum size and saliency as they approach this line. The last three or four lamelle to be formed do not extend completely around each half of the aptychus but wedge out Figs. 9 to 14. Ap.ychus cubanensis, new species. Fig. 9. Side view showing lamellae meeting external junction line at right angles. x 2 fisiceyte. A. M.N. H. No. 19018/2. Fig. 10. Tov view of ee aptychus with both valves in juxtaposition. X 2 Paratype, A. M. N. H. No. 19018,’ Fig. 11. Exterior. X 2. Pabatype A. M. N. H. No. 1901873. Fig. 12. View of posterior end of a broken valve showing the lamellie meeting externa! junction line at right angles. x 2. Paratype, A. M. N. H. No. 19018,/6. Fig. 13. Exterior of two valves of nepionic aptychus. XX 2. Paratype, A. M. N. H. No. 19018 /7. Fig. 14. Exterior of immature apty _ showing two valves in juxtaposition along junction planes. XX 2. Paratype, A. M. N. H. No. 19018,’8. 10 AMERICAN. MUSEUM NOVITATES. [No. 28 along the curved periphery. On the long anterior surface the lamelle are thin and sharp but as they increase in size posteriorly their edges become rounded and they appear more like folds separated by channels than like overlapping knife-edges. The external line of junction is thickened and elevated and it serves to form a very definite boundary for the lamelle (Fig. 9). DescrIPTION OF PaRATYPES.—On Specimen 19018 / 7 (Fig. 13), 11 mm. long, 20 lamelle: were counted, this not being the total number since the finer ones near the apex were too small to be distinguished. The two valves are shown in contact along the junction line and when in this position the posterior slopes are horizontal, while the anterior portions are deflected abruptly downward. The outline as a whole is rhombohedral with the distal end squarish and truncated. The largest specimen (No. 19018 /6), which shows only one valve, is approximately 21-22 mm. long and 10 mm. wide (Fig. 12). The nepionic aptychus (specimen No. 19018 /8) is nearly flat, showing only aslight anterior convexity. One of the plates is 3.5 mm. long and 1.8 mm. wide. Thirteen fine, concentric, continuous lamelle could be counted, although at that early stage they appeared merely as growth-lines, the lamellar character not being visible on the surface (Fig. 14). The interior concave side of the aptychus is covered by minute, concentric growth-lines. The middle shell layer is thin near the apex but increases posteriorly. The thickest part of the shell is at the margin on the line of the posterior ridge. NUMBER OF SPECIMENS.—Holotype and seven paratypes. AssOcIATED SPEcIEs.—A. cristobalensis, new species, at locality No.8. C. 2. LocauitiEs.—On Finca of Rafael Begoa, 9 miles north of San Cristobal (locality No.8. C. 2) five specimens, Nos. 19018/1—5; on Finea of Rafael Begoa Kocality No. S. C. 4) 3 specimens, Nos. 19018/6-8. Comparisons.—The distinguishing characteristics of A. culbieilsite are the angle of curvature of the lamellz along the posterior ridge and their abrupt, right-angled termination at the external line of junction. These features readily differentiate it from A. cristobalensis and A. pimientensis, which are found associated with it. The thickening of the lamelle into folds as they reach the external line of junction is also char- acteristic. This species appears to be almost identical with the aptychus of Haploceras figured by Steinmann (1907, fig. 542) from the Neocomian of Chalangon, Département of Dréme, France, (Fig. 6) except that the Cuban specimens are smaller and they are narrower when viewed from the side. The lamellae wedge into each other on the side of the valve, while in the French species they are simple and continuous. Aptychus cubanensis evidently belongs to a species of Haploceras. Aptychus pimientensis, new species E Figures 15 to 18 DeEsCRIPTION OF THE HoLotyPE (Amer. Mus. No. 19019,/1). Thisspeciesissmall, rhombohedral in outline and with delicately etched surface features. The measure- ments for one valve are: length 19 mm., width 10.2 mm., length of the truncated proximal end 8.2 mm (Fig. 15). 1921] NEW SPECIES OF AMMONITE OPERCULA 11 The lamelle are fine and close-set; they increase regularly in size and, beyond the apical portion which is broken away, 26 can be counted. The largest ones wedge out against the smooth peripheral band. There is a pronounced radial ridge on the pos- terior slope; on the posterior side of the ridge the lamelle turn sharply toward the apex at an acute angle of about 30°. The junction surface is smooth, triangular, with a maximum width of 1.7 mm.; it is at right angles to the convex surface of the shell. The surfaces of all the lamelle are finely punctate but this character is most clearly visible on the peripheral surface at the thickest part of the shell. Figs. 15 to 18. Aptychus pimientensis, new species. Fig. 15. Exterior of valve. X 2. Holotype, A. M. N. H. No. 19019/1. Fig. 16. Exterior of both valves in juxtaposition, showing triangular i planes and posterior angulation of lamella. X 2. Paratype, A. M. N. H. No. 19019 Fig. 17. Side view of Tye specimen showing acute S elation. 4P lamelle. xX 2. Paratype, A. M. N. H. No. 1901974 Fig. 18. View of posterior end of one valve, showing costes wedging out against peri- et margin and also along the side, produciUg characteristic en echalon notching. X 2. Paratype, A. M. N. H. No. 19019/5. Description oF ParatyPEs.—Specimen No. 19019/4, which is crushed, shows the sharp angulation of the lamellz on the posterior slope (Fig. 17). Specimen No. 19019/3 shows the two valves in contact along the junction line (Fig. 16). Speci- men No. 19019/5 shows the notching of the lamelle on the anterior slope of the ridge (Fig. 18). NuMBER OF SpectMENs.—Holotype and 4 paratypes. AssocIaATED Species.—A ptychus cristobalensis, new species, at locality No.8. C. 3. 12 AMERICAN MUSEUM NOVITATES [No. 28 Loca.ities.—Mt. Pimiento, 5 miles north of San Cristobal, Province of Pinar del Rio (locality No. S. C. 5), three specimens, Nos. 19019/1-3; northeast of Mt. Pimiento in Rio Hondo, 7 miles northeast of San Cristobal (locality No. S. C. 3), two specimens, Nos. 19019/4, 5. Comparisons.—The most noticeable characteristic of this species is the inflection of the lamelle at an acute angle as they approach the junction margin. This feature serves as a ready means of distinguishing A. pimientensis from A. cristobalensis in which the lamelle are curved, not angular in their inflection. There are also more minute differences: A. pimientensis is relatively longer than A. cristobalensis, has more even and regular lamellz, and is rhombohedral instead of semi-elliptical in shape. This species, like the others from the same beds, bears a general resemblance to A. didayi from which it differs primarily in shape, being rhombohedral instead of triangular, and having finer, more numerous lamellae. It probably belongs to a species of Oppelia. BIBLIOGRAPHY OF SOME OF THE MORE IMPORTANT LITERATURE ON THE GENUS APTYCHUS 1811. Parxinson, JAMES. ‘Organic Remains of a Former World.’ 3 vols; II, pp. 184-187. (Genus Trigonellites (=Aptychus) described p. 186; T. lata, p. 186, Pl. xin, figs. 9, 12; 7’. lamellosa, p. 186, Pl. x11, figs. 10, 11. By the strict rules of priority the name-T’rigonellites should now be used instead of Aptychus, but the latter name is firmly rooted in the litera- ture and there is the further consideration that Parkinson considered the opercula to be pelecypods and the name T'rigonellites is etymologi- cally more appropriate for pelecypods than ammonites.) 1830-33. ZretTEN, C. H. ‘Die Versteinerungen Wiirttembergs.’ (Good illustrations 5 of Aptychus levis latus Meyer (p. 49), Pl. xxxvu, figs. 6 and 7, showing exterior and interior and plates in situ.) 1831. Mryrr, Hermann von. ‘Das Genus Aptychus.’ Nova Acta, XV, pp. 125- 170. (Original description of A. levis, p. 128, Pls. Lyi, ix, figs. 6-9; A. imbricatus, p. 139, Pl. trx, figs. 10-12. Structure of shell, Pl. urx, fig. 13. Aptychi from the Lias, Pl. ux, figs. 1-7. Aptychi regarded as the shells of mollusks devoured by ammonites, p. 156.) 1837. Vourz, PxHitippE Louis. ‘Erste Notiz tiber das Genus Aptychus.’ Neues Jahrb., pp. 304-312. (General resumé of opinions concerning the nature of aptychi. Voltz considered them to be ammonite opercula.) 1837. ‘Zweiter Vortrag tiber das Genus Aptychus.’ Neues Jahrb., pp. 432-438. (Classification of 25 species of Aptychus, 12 new, into Cornei, Imbricati, and Cellulosi. List of ammonite genera with which aptychi are found in situ.) 1841. Coquanp, Henri. ‘Mémoire sur les Aptychus.’ Bull. Soc. Geol. France, (1) XII, pp. 376-391, Pl. rx. (Summary of previous opinions on the sig- nificance of Aptychus (pp. 376-386). Considered by Coquand to be genus intermediate between Loligo and Sepia (p. 387); protographs and 1921] NEW SPECIES OF AMMONITE OPERCULA - 13 protologs (pp. 387-390): A. blainvillei, p. 387, Pl. rx, figs. 8, 9; A. beaumonti, p. 388, Pl. rx, fig. 12; A. radians, p. 389, Pl. rx, figs. 11 and 11 bis; A. didayi, p. 389, Pl. rx, fig. 10; A. seranonis, p. 390, Pl. 1x, fig. 13. List of 33 species of Aptychus classified and with geological horizon, pp. 390, 391.) 1841. Guocxrr, E. F. ‘Ueber den Jurakalk von Kurowitz in Mahren und iiber den darin vorkommenden A ptychus imbricatus.’ Nova Acta, X XIX, Suppl. 2, pp. 275-308, Pls. r-m1. (Mode of occurrence of Apiychus imbricatus in the Jurassic of Kurowitz, pp. 285, 286, 289. Detailed description of examples of this species and of the character of the shell layers, pp. 293-306, Pl. m1, figs. 1-7, 9. The species here described is synonymous with A. punctatus Voltz and is a typical example of the group of sit Punctati.) 1846. Grinitz, H. B. ‘Grundriss der Versteinerungskunde.’ (Aptychi discussed pp. 307-310; five species (not new) described.) 1849. QuenstepT, F. A. ‘Die Cephalopoden.’ (Very full descriptions of numerous previously known species, pp. 306-323. Protolog of A. crassicauda, p. 314, protograph, Pl. xxn, fig. 25. Of especial interest is description of A. didayi Coquand, p. 314, Pl. xxu, fig. 21.) 1854. Peters, K. ‘Die Aptychen der ésterreichischen Neocomien und oberen Juraschichten.’ Jahrb. d. k. k. g. Reichsanstalt, V, pp. 439-444. (Protologs of: A. angulocostatus, A. undatocostatus, A. lineatus, A. pusillus (p. 441); A. rectecostatus, A. reflexus (p. 442); A. aplanatus, A. giganteus (p. 443). Occurrence of A. didayi Coquand in the Neccomian n noted (p. 441). A. latus Voltz (p. 443), A. depressus Voltz and A. profundus Voltz (p. 444) reported from the Upper Jurassic of St. Veit and Lainz, Austria.) 1854. Picret, F—J. ‘Traité de Paléontologie,’ 2nd edit. II. (Numerous previously described species listed (p. 558), some figured Pl. xtvu, figs. 10-17. Excellent illustration of A. didayi Coquand, Pl. xivn, fig. 17.) 1854. Hésert, E. ‘Tableau des Fossiles dela Craie de Meudon.’ Mém. Soe. Geol. France, (2) V, pp. 345-374. (Original descriptions of three new species of Aptychus from the Chalk of Meudon: A. insignis, p. 367, Pl. xxvmt, fig. 6; A. obtusus, p. 367, Pl. xxv, fig. 7; A. crassus, p. 368,- Pl. xxvitl, fig. 8.) 1856. Suarpr, D. ‘Description of the Fossil Remains of Mollusca found in the Chalk of England.’ Pt. III, Cephalopoda. (Aptychi of the Chalk (pp. 53-58). New species described: A. porilocki, p. 56, Pl. xxtv, figs. 2, 3, 4, (6?); A. gollevillensis, p. 56, Pl. xxiv, fig. 5; A. icenicus, p. 57, Pl. xx1v, fig. 7a, b; A. rugosus, p. 57, Pl. xxtv, figs. 8a, b, and 9; A. peramplus, p. 58, Pl. xxrv, fig. 10.) 1858, QueNsrept, F. A. ‘Der Jura.’ (Illustrations and descriptions of a number of previously described species.) 1863, Oprrn, ALBERT. ‘Ueber jurassische Cephalopoden.’ Pal. Mitth. III. (Proto- graphs and protologs of many species of aptychi found in situ in am- monites from the Lithographic Limestone of Solnhofen. Ammonites lithographicus, p. 248, Pl. uxvint, figs. 1-3; A. thoro, p. 250, Pl. txvm, figs. 6, 7; A. steraspis, p. 251, Pl. uxrx; A. bous, p. 252, Pl. uxx, fig. 1; 14 1865. 1866. 1866. 1867. 1868. 1868. 1870. 1875. 1881. 1885. 1885. AMERICAN MUSEUM NOVITATES [No. 28 A. euglyptus, p. 253, Pl. uxx, figs. 1-5; A. hybonotus, p. 254, Pl. Lxxt1, figs. 1-3; A. autharis, p. 255, Pl. Lxx1, figs. 4-6; A. latus, p. 256, Pl. LXxu, fig. 1; A. pipini, p. 257, Pl. txxu, fig. 3; A. aporus, p. 258, Pl. uxxm, figs. 1-3; A. hoplisus, p. 259, Pl. uxximi, figs. 4, 5; Aptychus sp., p. 261, Pl. uxxtv, figs. 3, 4. The Solnhofen beds contain the largest Upper Jurassic Aptychus fauna so far known.) Scuaurotu, Cart von. ‘Verzeichniss der Versteinerungen in Hering: Naturaliencabinet zu Coburg.’ (Protologs and protographs of: A. punctatus, p. 152, Pl. 1v, fig. 13; ‘A. exsculptus, p. 153, Pl. rv, fig. 14. Occurrences given for A. latus Miinster and A. lamellosus Mister, p. 152.) Brenecke, E. W. ‘Ueber Trias und Jura in den Siidalpen.’ Geogn.-Pal. Beitrige, I, pp. 1-204. (A. cf. lamellosus Voltz and A. ef. latus Minster reported in Acanthicus beds, p. 185, A. curvatus Giebel, A. cf. gigantis Quenstedt reported from Diphyakalk (p. 192). A. didayi recorded as characteristic of the Biancone (Neocomian) formation, p. 135.) . OpreL, ALBERT. ‘Ueber die Zone des Ammonites transversarius.’ Geogn.- Pal. Beitrage I, Heft. 2, pp. 207-316. (References to the occurrence of aptychi species, pp. 218, 234, 252, 279. Occurrence of Aptychus didayi in Neocomian of Algeria, p. 273.) WaacEN, WitHELM. ‘Ueber die Ansatzstelle der Haftmuskeln beim Nautilus und den Ammoniden.’ Palzontographica, XVII, pp. 185-210, Pls. Xxxix and xu. (Discussion of the position of the aptychus in the body chamber of ammonites, pp. 192, 193; illustrations on Pl. xu.) Zire, Karu A. ‘Die Cephalopoden der Stramberger Schichten.’ (Discus- sion of characters of Aptychus, pp. 49-52; description of Stramberg species (pp. 52-55); A. punctatus Voltz, p. 52, Pl. 1, figs. 15a, b; A. beyrichi Oppel, p. 54, Pl. 1, figs. 16-19.) Pricret, F.-J. ‘Mélanges Paléontologiques. IV. Fossiles de la Porte-de- France,’ (A. latus Voltz) p. 288, Pl. xu, figs. 1-4) and A. imbricatus Meyer (p. 285, Pl. xu, figs. 5-10) described from the Limestone of Porte-de-France.) ZrrreL, Karu A. ‘Die Fauna der ‘Aeltern Cephalopenfiihrenden Tithon- bildungen.’ (A. punctatus Voltz (p. 31), A. beyrichi Oppel (p. 32), and A. exsculptus Schauroth (p. 32, Pl. 1, fig. 10) and Cellulosi of the latus type (pp. 88, 89) described from the Tithonian of the Alps.) Favre, Ernest. ‘Fossiles du Terrain Jurassique dela Montagne des Voirons.’ Mém. Soc. Pal. Suisse, II, pp. 1-78. Pls. 1-vm. (Four previously known species identified in formations ranging from Transversarium through Acanthicus zones: A. latus (Parkinson), p. 47, Pl. vu, figs. 1-3; A. punctatus Voltz, p. 49, Pl. vu, figs. 4, 5; A. sparsilamellosus Giumbel, p. 50, Pl. vu, figs. 6-9; A. beyrichi Oppel, p. 52, Pl. viz, figs. 10, 11.) Lorton, P. pr. ‘La Zone & Ammonites tenuilobatus D’Oberbuchsitten.’ Idem, VII, pp. 1-60, Pls. 1-x. (Presence of Imbricati recorded, p. 27, Pl. vu, figs. 6, 7, and Cellulosi, p. 27, Pl. vir, figs. 8, 9, no species de- scribed. ) Zrrrer, Karu A. ‘Handbuch der Paleontologie.’ II. (Classification of Aptychi, pp. 400-403, figs. 544-553.) Quenstept, F. A. ‘Die Ammoniten des Schwiibischen Jura.’ (Illustrations and amplified descriptions of numerous previously described species.) 1921] 1890. 1891. 1893. 1893. 1894. 1896. 1898. 1899. 1905. 1907. 1907. NEW SPECIES OF AMMONITE OPERCULA 15 Toucas, A. ‘Etude de la Faune des Couches tithoniques de |’Ardéche.’ Bull. Soe. Geol. France, (3) XVIII, pp. 560-629. (Occurrence of A. punctatus Voltz, pp. 579, 595; A. beyrichi Oppel, p. 580; and A. ef. latus Volta noted in the Tithonian of the Alps, Carpathians, Apennines, etc.) Rerowsk!, O. ‘Die Aptychen sind echte Ammonitendeckel.’ Neues Jahrb., Jahrgang 1891, II, pp. 220-221 and text figure. (Description and illus- tration of A. beyrichi Oppel shown in situ in the aperture of Haploceras elimatwm (Oppel).) RetowskI, O. ‘Die tithonischen Ablagerungen von Theodosia.’ Bull. Soc. Imp. des Nat. Mosc., N.8., VII, pp. 206-301, Pls. rx—xiv. (A. puncta- tus Voltz reported to be very abundant in Tithonian of Theodosia in the Crimea (p. 225); A. beyrichi Oppel described as A. elimati because found in situ in Haploceras elimatum (p. 226); A. ef. exsculptus Schau- roth rare, p. 227, Pl. rx, fig. 2.) Cuorrat, Pau. ‘Faune Jurassique du Portugal.’ (No species described. Rare occurrence of aptychi fragments in limestones of Cabaco series noted; Cellulosi of the latus type frequent in the marl beds of the Monte- junto and Abadia series of Portugal.) Micuak., RicHarp. ‘Ammoniten Brut mit Aptychen in der Wohnkammer von Oppelia steraspis Oppel sp.’ Zeit. d. deutsch. geol. Gesellschaft, XLVI, pp. 697-702, Pl. tiv. (Contains description of embryonic am- monites found in body chamber of Oppelia steraspis, both the young and mature shells having aptychi.) Buiackmore, H. P. ‘Some notes on the Aptychi from the Upper Chalk.’ Geol. Mag., N. S., (4) III, pp. 529-533, Pl. xvi. (Paper deals with Upper Cretaceous aptychi only and they are considered to be opercula of belemnites. A. rugosus, p. 532, Pl. xvi, fig. 16, shows form very similar to Cuban species.) : Crick, G. C. ‘Fossil Cephalopoda in the British Museum.’ (Four Upper Cretaceous (Upper Chalk) species of A ptychus listed (pp. 32, 33). All previously described by Sharpe, g. v.) Pau, C. M. ‘Der Wienerwald. Ein Beitrag zur Kenntniss der nordalpinen Flyschbildungen.’ Jahrbuch der k. k. geol. Reichsanstalt, XLVITI, pp. 53-178, Pl. 1-v1, 26 text figures. (The species of Aptychus occurring in the Wienersandstein are recorded on pp. 58, 59, 60,135, 141, 142, 152, 158, 175.) VetTers, HerMANN. ‘Die Fauna der Juraklippen zwischen Donau und Thaya.’ Beitriige zur Paliontologie und Geologie Osterreich-Ungarns und des Orients, XVII, pp. 223-259, Pls. xxi, xxm. (Occurrence of Aptychus punctatus Voltz, p. 242, Pl. xx1, fig. 4, and A. latus Meyer, p. 243, in the Niederfellabrunn Tithonian noted.) Touta, Franz. ‘Die Acanthicus-Schichten im Randgebirge der Wiener Bucht bei Giesshiibl.’ Abh. d. k. k. geol. Reichsanstalt, XVI, Heft 2 (Aptychi of Acanthicus beds, pp. 80-82, 88. Original description of: A. cellulo- salamellosus, p. 80, Pl. xvut, fig. 4; A. insolidus, p. 81, Pl. x, fig. 2. Occurrence of A. ef. didayi Coquand noted, p. 88, Pl. xm, fig. 6.) STEINMANN, Gustav. ‘Einfiihrung in die Palaontologie.’ Leipzig. (Descrip- tion of aptychi, pp. 319, 320; excellent illustrations, figs. 540-542.) X 8 AMERICAN MUSEUM. NOVITATES | No. 29 | he 3 FIVE NEW SPECIES OF SALIENTIA FROM SOUTH AMERICA - By G.-K. Nosie ~ Issued December 30, 1921 BS ORDER OF THE TRUSTEES OF THE AMERICAN MUSEUM OF NATURAL HISTORY oh New YorE Ciry i ~y)) ms fits — AMERICAN MUSEUM NOVITATES Number 29 December 30, 1921 59.78 (8) FIVE NEW SPECIES OF SALIENTIA FROM SOUTH AMERICA By G. K. NoBLE Among the collections of amphibians from South America which have been acquired in recent years by The American Museum of Natural History, there are included several interesting new forms. Perhaps the most remarkable of these is one which must be referred to the genus Sminthillus, although zoogeographic considerations make it seem highly probable that the form described below has been derived from a different stock from that which gave rise to S. limbatus, the type, and, until this time, sole member of the genus. S. limbatus is confined to Cuba while the species described here is restricted to a limited region in southern Peru. Sminthillus has been only recently defined (Barbour and Noble, 1920, Bull. Mus. Comp. Zool., LXIII, p. 402). Ihave more recently added certain details in regard to the structure of the shoulder girdle (Noble, in press). These structural details are very important - from a phylogenetic viewpoint. But, as I have discussed these features at considerable length in the latter paper, I shall not enter into them here. The genus Atelopus is much in need of revision. A number of diverse stocks have been lumped together under the name A. ignescens. Until the limits of specific variability have been determined for several more forms, it is impossible to discuss the status of these stocks or the relations of a number of the described species. The characters which I have utilized below appear at this time to be diagnostic. Sminthillus peruvianus, new species Draqgnosis.—Readily distinguished from the S. limbatus by its blunter snout, less vertical loreal region, by the presence of a tarsal tubercle, longer tibia and different coloration; from the disk-less species of Syrrhopus, it may be distinguished externally by the more Vertical loreal region and stouter form. RanGE.—Known only from the present series secured near Juliaca, Peru, by H. H. Keays, (no date). Typr.—Aa. M. N. H. No. 14526; from the same locality. DeEscriIPTION OF TyPE.—Snout rounded, equal to the greatest diameter of the orbit; loreal region abrupt, nearly vertical; nostril midway between the tip of thesnout and the anterior corner of the eye; interorbital space a little broader than the upper eyelid; horizontal diameter of the tympanum about one-half, vertical diameter nearly 2 : AMERICAN MUSEUM NOVITATES [No. 29 two-thirds the greatest width of the eye; tympanum about one-half its smallest diameter from the latter. Digits pointed, no terminal disks; a well-developed tarsal and two metatarsal tubercles. Tibio-tarsal articulation reaching just beyond the posterior angle of the eye; when the limbs are held vertical to the axis of the body, the tibio-tarsal articulation overlapping its mate of the opposite side. Skin feebly granular above, smooth below. Fig. 1. Sminthillus peruvianus, new species. Type. Fig. 2. Sminthillus peruvianus, new species. Paratype, showing color variation. Color uniform grayish brown, slightly purplish above; a broad band of dark brown extending on each side from the nostril to the lumbar region; a narrow white line on the posterior face of each thigh joining with a median line which extends anteriorly along the back for more than half its length; posterior surfaces of the lower leg indistinctly barred with dark brown. Lower surfaces of body whitish, the chin and thigh indistinctly suffused with brown. Dimensions Snout to vente sie 142255 lei ee ie 16 mm Greatest width of head.................. Sie ii Wea 5 mm eee iS avela- 3 Fa venkcnl pings rea a Groin to tip of longest toe. . Ate ons, = .21.5 mm. Notes ON Paratyprs.—Four specimens in addition to Ji type are included in the series. These differ greatly in color and slightly in proportions. The two extreme types of coloration are shown in Figures 1 and 2. In the paratype figured, the ground tone is a pale yellowish brown. A sharply defined pattern is formed by blotches of dark brown. The two inguinal spots are nearly black. The ventral surface is uni- formly suffused with a reddish brown which is darker on the throat than on the © thighs. The other three paratypes exhibit stages of intergradation between the coloration of the paratype figured and that of the type specimen. Their browns are 1921] FIVE NEW SPECIES OF SALIENTIA 3 redder than in the type and one has a broad hour-glass shaped pattern of dark reddish brown dorsally. The paratypes exhibit some variation in leg length. In all, the hind limb is longer than in the type. Atelopus bicolor, new species DiaGnosis.—A large species, closely allied to A. boulengeri Peracca, but differing in the webless digits of the manus and the shorter webs of the toes, also in the different proportions, shorter leg and shorter snout. Skin smooth; first toe distinct; tibio- tarsal articulation not reaching eye; generally dark brown above, yellow below, irregularly marked on the sides. RancE.—Known only from the type locality. Type.—A. M. N. H. No. 13132; from Cordillera Kutuku (1800-2000 meters), east of Macas, Ecuador; collected by E. Feyer; 1921. DeEscRIPTION OF Type (Adult male).—Head a trifle wider than long, its length (tip of snout to angle of jaw) contained three and a half times in head and body length; distance from the eye to the end of the snout one-third greater than the long- est diameter of the eye; nostril much nearer the end of the snout than the eye; snout projecting only a little beyond mouth; interorbital space 1.3 or 1.2 times the width of the upper eyelid. Anterior limb stout; distance from axilla to tip of longest digit equals the body length; no webbing between the digits of the manus although the skin between the first and second digits is thick; first digit much shorter than the fourth. Tibio-tarsal articulation nearly reaching the eye; toes webbed to the tips (except the fourth), but the web deeply indented, making the toes appear half webbed; a Single metatarsal tubercle; subarticular tubercles feeble. Skin smooth above and below; a number of feeble transverse furrows on the ventral surface. Color above dark reddish brown; below and on the sides yellow; an irregular dark streak on the sides of the abdomen; yellow of the sides encroaching upon the periphery of the dorsal ground tone, forming a symmetrical yellow margin to the dorsal coloration; limbs broadly crossbarred with yellow; ventral surface of the head and body immaculate; a few dark crossbars on the legs. Dimensions Serena CO VER ho eee tice ee ee eee eek 51 mm. Cieatent width Of NOAd 2.6 Wi ok. fe ec ee 23 15.5 mm. Axilla to tip of longest digit........................37 mm. Groin to tip of longest toe.........................68 mm. Notes oN Paratypes.—Three specimens in addition to the male were collected. Two are males and the third is an immature specimen. They vary considerably in coloration but agree in having the ventral surface yellow and immaculate, the dorsal surface dark brown or reddish with no markings except an irregular border of yellow. Atelopus rugulosus, new species D1AGnostis.—Very similar to A. tricolor Boulenger, from which it differs in the warty upper surfaces, somewhat different proportions and absence of “flash colors.” Inner toe rudimentary; tibio-tafsal articulation extending pee the eye; inter- orbital space twice as broad as the upper eyelid. Atelopus rugulosus, new species. Type. Atelopus bicolor, new species. Type. Phyllobates anthonyi, new species. Type. Telmatobius cinereus, new species. Type. Ventral surface of left foot. 1921] FIVE NEW SPECIES OF SALIENTIA 5 RanGE.—Known only from the type locality. Typr.—a. M. N. H. No. 6097; vicinity of Juliaca, Peru; H. H. Keays. DescriPTIon oF Type (Adult male).—Distance from the angle of the jaw to the tip of snout one-fifth wider than the greatest width of the head, exactly three times as great as the head and body length; snout prominent, overhanging the mouth fora distance equal to the greatest diameter of the eye; nostril nearer the tip of the snout than the eye; interorbital space twice as broad as the upper eyelid. Fingers webbed slightly at the base, first very short; toes entirely webbed, but the web between the third and fourth, and the fourth and fifth digits notched; first toe rudimentary, enclosed within the web; no metatarsal or subarticular tubercles. Tibio-tarsal articulation reaches nearer the tip of the snout than the eye. Skin covered with flat warts above, smooth below, except for the throat which is somewhat granular. Dark brown above spotted with yellow, a broad streak of yellow on each side of the back from eye to groin; upper lip yellow, and a series of yellow spots on the sides of the body; ventral surfaces yellow with a few dark spots on the belly; a tinge of pink on the thighs near their proximal end, but no axillary or inguinal spots of the same color. Dimensions RHOMT 60 VOU art eee ny ek te ha Pree 22 mm. CUGAtERG Width Of DER lS. So 54's rae ora ees 6.8 mm. Axilla to tip of longest digit........................14.5 mm. Groim to tip of longest toe.........................30 mm. Notes on Paratype.—Only two specimens, both males, of this species were secured. The paratype differs from the type in being more spinose, the tubercles being pointed, not flat as in the type. The coloration is essentially the same in both specimens, but the paratype is nearly immaculate below. Phyllobates anthonyi,' new species DiaGnosis.—Similar to P. tricolor (Boulenger), but readily distinguishable by its longer leg, longer first finger, longer snout, and different coloration. The color pattern is somewhat similar to that of P. tricolor but the flash colors are lacking and the stripes are bluish not yellowish. Rance.—Known only from the type locality. Type.—A. M. N. H. No. 13739; from a small stream at Salvias, Prov. del Oro, ° Ecuador; collected August 10, 1920, by H. E. Anthony. Description oF Typr.—Snout depressed, a little longer than the greatest diameter of the eye; nostril much nearer the end of the snout than the eye; loreal region vertical; interorbital space 1.4 times as broad as the upper eyelid; tympanum a little more than half as long as the greatest diameter of the eye, edged above by a prominent supratympanic fold, Disks of the digits small, much smaller than the tympanum; first finger slightly longer than the second; two metatarsal tubercles, the inner prominent; a well-developed tarsal tubercle; tibio-tarsal articulation reaching the anterior corner of the eye. Skin smooth above and on the sides. ‘Named for Mr. H. E. Anthony, Associate Curator of Mammals of the Western Hemisphere in The American Museum of Natural History, and collector of the type series. 6 AMERICAN MUSEUM NOVITATES [No. 29 Ground tone above dark chestnut brown; a broad medial stripe and two dorso- lateral ones of pale bluish gray, the medial stripe widening anteriorly to form the major coloration of the snout, and restricting the ground tone in this region to a number of spots. Ground tone of the sides a little darker than that of the back; in addition to the dorsolateral stripe, a ventrolateral one of the same color; the latter forming anteriorly a prominent stripe on the upper lip, while the former passes along the outer edge of the eyelid and merges into the light tone of thesnout. Thighs cross- barred with pale gray; concealed portion of the thigh blotched with white which may have been yellowish in life. Ventral surface white or slightly tinged with yellow and heavily marbled with dark brown, the marbling most pronounced about the periph- ery of the ventral surface; no dark bands on the chest or throat. Dimensions WHOUE GO VONG. x05) ca ionic eink 5 ee ees poaee ee 21 mm Greatest width of the head. ...2......5< 54.0. cen 6.5 mm Axilla to tip of longest digit.......................-14 mm, Groin to tip of longest toe. ........... 0.5602 eee te) 5 Notes on PARAtyPE.—A single paratype was secured. It differs from the type in having the lateral stripes more whitish and in having the ventral surface marbled with brown on the periphery only. Telmatobius cinereus, new species DraGnosts.—Very similar to 7’. niger Barbour and Noble, from which it differs chiefly in the narrower webbing of the toes (compare Figure 6 with Fig. 1, Barbour and Noble, 1920, Bull. Mus. Comp. Zool., LXIII, p. 414), the narrower interorbital space, the absence of a ventral disk, and the grayish, not chestnut, coloration. RancGre.—Known only from Bestion, Ecuador, the type locality. Typr.—A. M. N. H. No. 13968; from Bestion, Ecuador, caught in a mouse-trap set among low bushes and grass; January 7, 1921, by H. E. Anthony. DescripTION or TypE.—Size moderate; head much broader than long, narrower than the body, its length contained in the total length of the body a trifle more than three times; snout very short and high without canthus rostralis; nostril nearer the orbit than the labial border. Vomerine teeth prominent in two well-defined groups between the choane; tongue about as long as broad. Interorbital space exactly as long as the diameter of the eye; length of the snout 1.4 times that of the eye; tym- — panum hidden; a prominent supratympanic fold. Digits free, stout, not dilated at the tips, the first finger longer than the second, a trifle longer than the fourth; the elbow extended forward reaches nearly to the eye. Toes a trifle less than half webbed, but narrow seams extending to the bases of the terminal phalanges; a well-defined tarsal fold; subarticular tubercles well developed; two metatarsal tubercles of about the same size; heels just in contact when the hind limbs are folded at right angles to the axis of the body; tibio-tarsal articulation reaching to the middle of the orbit or slightly beyond. Skin very glandular above, but not warty; no baggy lateral fold; a few irregular folds on each side of the body; no ventral disk, the skin of the abdomen not marked off from the lateral regions; a slight indication of a pectoral fold formed by a small fold on each side immediately posterior to the humerus; posterior surfaces of the thighs not folded or baggy. 1921] FIVE NEW SPECIES OF SALIENTIA 7 Color above lead gray; ventral surface pinkish, heavily blotched and mottled with dark brown, the spots most abundant on the throat, least on the thighs. Dimensions _ itr On Mae CO NONE cogs eae week nk oe OR TO, Tip of snout to angle of jaw... 35... ec. ee 20 mm. Crea teme Wile OR SOO Orcas rete ca ee ekice as 24.5 mm. Axilla to tip of longest digit........................86 mm. Groin to tip of longest toe........................-88 mm. The species is known only from the type specimen. As might be expected from its distribution, the species is intermediate between 7’. niger Barbour and Noble and 7’. ignavus Barbour and Noble, but more closely related to the former. _ AMERICAN MUSEUM NOVITATES THE DISTRIBUTION OF THE SWALLOWS OF = THE GENUS PYGOCHELIDON ~~ ig By Frank M. CHAPMAN pate : Issued February 28, 1922 > By Orver or THE TRUSTEES “8 . oc OF ae: THE AMERICAN MUSEUM OF NATURAL HISTORY oe ee New York City Dita eS — oe ee a 7 } AMERICAN MUSEUM NOVITATES Number 30 February 28, 1922 59.88.1 P:19(8) THE DISTRIBUTION OF THE SWALLOWS OF THE GENUS ° PYGOCHELIDON By Frank M. CHAPMAN Continued study of our recently acquired collections of South American birds reveals certain facts in regard to the distribution of the blue and white swallows of the genus Pygochelidon which seem worthy of independent record. ° The range of this genus extends from Costa Rica to Tierra del Fuego. It contains four known forms (of which two are herein described for the first time), the general characteristics and distribution of which are set forth below, following which I present some remarks and conclu- sions. I am indebted to Dr. Alexander Wetmore, of the Biological Survey, and Mr. James L. Peters, of the Museum of Comparative Zoology, for information in regard to the distribution of P. cyanoleuca and P. pat- agonica patagonica based on their recent explorations in southern South America. Pygochelidon cyanoleuca (Vieillot) Hirundo cyanoleuca Viertutot, 1817, Nouv. Dict. d’Hist. Nat., XIV, p. 509 (Paraguay. One Paraguay specimen examined). Hirundo minuta Max., 1821, ‘Reise Bras.,’ II, p. 336 (Rio Janeiro, Brazil. Types and freshly collected material examined). H{irundo] melampyga Licut., 1823, ‘Verz. Doubl.,’ p. 57 (Bahia, Brazil. Bahia specimen examined). Atticora cyanoleuca var. montana Barrp, 1865, ‘Rev. Am. Bds.,’ p. 310 (Barranca, Costa Rica. Costa Rica specimens examined). Speciric CHARACTERS.—Sexes alike in color, tail in the male averaging longer. Adult with rectrices and remiges and their shafts, black or blackish; outer margin of outer rectrix never lighter than the rest of the feather; under wing-coverts and axillars fuscous to chetura-drab; lower tail-coverts wholly black with steel-blue reflections; median area of the feathers of the nuchal region whitish (a character more pronounced in northern than in southern specimens); underparts (except sides and flanks) pure white; back varying from a deep indigo-blue to a greenish blue, the variation partly individual, partly seasonal and apparently occurring throughout the range of the species. For example, a Boruca, Costa Rica, male taken May 10 is blue while a San José male taken April lis green. Both blue and green birds were 2 AMERICAN MUSEUM NOVITATES [No. 30 taken at San Antonio, western Colombia, in January; while of nineteen Bolivian tableland birds twelve, taken in May, June, and July, are green and seven, taken in November, December, and February, are blue. Immature birds have the wings and tail fuscous, the lower tail-coverts dusky, more or less tipped with grayish, a trace of cinnamon may suffuse the white under- parts, but the lower wing-coverts are as in the adult and the outer tail-feather is uni- formly colored, the outer margin NEVER being paler than the rest of the feather. S1zz.—There appears to be no marked latitudinal variation in size among birds from apparently the same altitude. Specimens from near sea-level in Ecuador (Rio de Oro) are the smallest in the series while specimens from the Ecuadorian tableland are larger and near the average size. The tail apparently averages longer in the “_ but there is much variation in this respect. Although recorded from Costa Rica to Paraguay! and from western Keuador to eastern Brazil, Pygochelidon cyanoleuca is a bird of the Sub- tropical, rather than the Tropical, Zone and is restricted largely to mountainous regions. In Costa Rica, Carriker does not record it from a lower altitude than 1000 feet, whence it ranges upward to the Irazu district. It is recorded by Bangs from Chiriqui, western Panama, at an altitude of 10,800 feet, and from ‘‘Veragua” by Salvin and Godman. Sharpe and Wyatt (‘Mon. Hirvnd.’) state that Salvin and Godman’s collection contains three specimens secured by McLeannan in “ Panama,” presumably south of Colon where this collector worked. There are no Panama specimens in the American Museum, and I know of no other Panama records. In Colombia we found this swallow to be most abun- dant in the Subtropical Zone, but it ranged from 2000 to 9000 feet. In Ecuador our only record for the coastal region is Rio de Oro, Manavi, whence it ranges upward to the tableland. It has not been found in the Guayaquil region and our most southern Pacific coast region record is Portovelo (alt. 2500 ft.), near Zaruma on the eastern slope of the coast range west of Santa Rosa. It occurs also in eastern Ecuador at Zamora and thence south on eastern, or Amazonian, drainage from Peru (Huan- cabamba, Perico, La Merced, Perené, Sta. Ana, Torontoy, Santo Domingo) to Bolivia, whence all our specimens are from altitudes of from 7700 to 9400 feet (Depts. Cochabamba, Sucre). In Argentina it extends at least to the Prov. of Tucuman, whence we have a wholly typical adult male taken on the Tafi trail at an altitude of 2000 feet, April 12, 1916. East of the Andes this swallow appears to be much less common, It is not, for example, recorded from the Orinoco region by Berlepsch 1Records from Chile are evidently based on erroneous identification or wronely labeled s ens. For example, a skin labeled by Rusby “ Valparaiso’’ (doubtless the one referred to by fs sae in Bull. 50 p. 70; see also Allen, 1889, Bull. A. M. N. H., IT, p. 80) evidently bears an incorrect locality, as unfortunate sly dosome other specimens in this collection. 1922] SWALLOWS OF THE GENUS PYGOCHELIDON 3 and Hartert, or from Cayenne by Berlepsch, and Snethlage gives no definite record from Amazonia, though a specimen from Par4 is listed in the British Museum ‘Catalogue.’ Salvin records it in Guiana only from an altitude of 3500 feet on Mt. Roraima, and a single Guiana specimen is listed from Camacusa. In Venezuela and Brazil it appears to be con- fined largely to the mountainous and coastal areas. It is apparently not uncommon near the coast of extreme southeastern Brazil, and Wet- more secured specimens at Lazcano, in northeastern Uruguay, the most southern record for the species. From the interior of South America there appear to be only two records for this species. Smith secured one specimen September 10, 1885, at Chapada, Matto Grosso, during nearly five years’ collecting, and a specimen in the U. S. National Museum, No. 35040, was secured by C. Wood of the Page Expedition at Bahia Negra, S. lat. 20° on the Paraguay River, in June, 1859. Cherrie secured no specimens of this species during two expeditions in southwestern Brazil, and it is evidently rare or wanting in the interior of South America, at least north of subtropical latitudes. In general, then, Pygochelidon cyanoleuca is not, as has been often stated, distributed “throughout” South America; but is largely confined to mountainous areas and to the Subtropical Zone. Furthermore, while not a bird of the forest, it occurs chiefly in forested regions. Pygochelidon patagonica patagonica (d’Orbigny and Lafresnaye) Hirundo patagonica pv’ORBIGNY AND LAFRESNAYE, 1837, ‘Syn. Av.,’ p. 69 (Patagonia). Atticora hemipyga BuRMEISTER, 1861, ‘Reis. La Plata,’ II, p. 479 (Mendoza, Argentina; one specimen examined). Sprecrric CHARACTERS.—Sexes alike in color (and in size ?); rectrices and remiges and greater wing-coverts fuscous, their shafts brownish; outer margin of outer rectrix narrowly but distinctly EDGED WITH WHITISH; under wing-coverts and axillars mouse-gray; shorter lower tail-coverts white, longer ones sometimes (in more south- ern specimens) basally white; feathers of the nuchal region basally gray; size larger _ than cyanoleuca. Immature birds have the wings and tail fuscous as in the adult. The under wing-coverts and axillars are also mouse-gray and the outer margin of the outer tail-feathers is edged with grayish. The longer tail-coverts are fuscous, tipped with grayish, the shorter ones white in young birds from Mendoza and Prov. Tucuman (4000 ft.), but in three specimens in a corresponding state of plumage from Tilcara (8000 ft.), Prov. Jujuy (February 8-12), the longer lower tail-coverts are fuscous tipped with cinnamon, the shorter ones cinnamon. I do not know whether this char- 4 AMERICAN MUSEUM NOVITATES [No. 30 acter represents a racial or individual variation. An adult female taken February 8, at the same locality, has the under tail-coverts wholly black. In other respects all four Tileara birds are typical patagonica. This is a migratory species. During the summer it is found through- out the South Temperate Zone, south to Tierra del Fuego and north to Peru over the Andes of northwestern Argentina and (presumably) Puna or Temperate Zone in Peru. It winters chiefly north of S. lat. 30° and (presumably) at the eastern base of the Andes as far north as the Marajfion. On the Pacific coast this swallow is doubtless found from Chiloe to near the Peruvian border where it apparently intergrades with the Peru- vian race described below. I have, however, seen specimens only from Temuco to Tofo, sixty miles north of Coquimbo, but it is probable that the birds recorded by Lane! from east of Iquique should be referred to this race. According to this writer the species ranges upward to 12,000 feet, in which event it doubtless crosses the Andes at this, or even a greater, altitude. It is found in southern Chile, and doubtless at high altitudes, only during the summer. I have been unable to ascertain the exact limits of its seasonal range in Chile but according to Barros? it is found in the Valley of Nilahue (S. lat. 34° 30’) only from the end of August to the middle of March. In Argentina it is rarely found in winter south of the latitude of Buenos Aires and Mendoza.’ On September 10, 1916, I observed it in large numbers near the first-named city, where Dr. Dabbene informed me it had just arrived from the north. The return (postbreeding) migration is concluded by March 15. Hartert,* in recording the breeding of this swallow in the Proy. of Tucuman (no altitude stated) on the authority of Venturi, mentions also specimens from Cosnipata, Yurimaguas, and Nauta, Peru. Hell- mayr°’ also refers to the Cosnipata birds, of which one is young and four adult, and confirms Hartert’s identification of them. They were col- lected by H. Whitely, Jr., on September 20, October 5 and 7, 1868, after the date, therefore, on which patagonica reaches its breeding grounds south of the latitude of Buenos Aires, and are consequently not likely to have been winter visitants from the South Temperate Zone. 11897, Ibis, p. 15. 21920, Revista Chilena Hist. Nat., XXIV, p "Reed, 1916, ‘Las Aves de <5 agg ‘de Soca.’ a Ed. Mendoza, p. 39. 41909, Nov. Zool., XVI, p. 51919, Arch. fiir Naturg., et; °. ee a 1922] SWALLOWS OF THE GENUS PYGOCHELIDON 5 In traveling from Paucartambo to Cosnipata, Whitely passed over bare treeless regions at an altitude of 11,900 feet. We have seen that patagonica reaches an altitude of 12,000 feet on northern Chile and it would not be surprising if, like many other South Temperate Zone birds, it should range north on the Andean Temperate, or Puna Zone, whence, assuming that the specimens in question actually came from Cosnipata (alt. about 2350 ft.), it might occasionally visit the lowlands. I frankly confess that the records from Yurimaguas and Nauta seemed to me to require confirmation. Both localities are in the Tropical Zone and the latter is some 250 miles east of the Andes, but the dis- covery in our own collection of a wholly typical adult male in fresh (postnuptial ?) plumage taken November 12, 1919, by H. Watkins at La Merced (alt. 2600 ft.) in the Chanchamayo district of eastern Peru, proves to my complete satisfaction that patagonica occurs in this region. I cannot, however, believe that a species so characteristic of the South Temperate Zone breeds in the Tropical Zone and, in spite of the fact that it has never been recorded from the Peruvian highlands, I conclude that records from Cosnipata, La Merced, Yurimaguas, and Nauta are based on birds which breed in the Temperate or Puna Zones and migrate or wander to the lowlands. So far as I am aware no other swallow of the Temperate or Puna Zone has this habit, but no other swallow of these zones is found also in the South Temperate Zone. Possibly patagonica in extending its range northward on the Andean tableland has retained the migratory habit - which it displays in the South Temperate Zone. Rewations of P. patagonica with P. cyanoleuca Having described the characters which distinguish cyanoleuca and patagonica from each other and outlined their known ranges the question arises, are they specifically or subspecifically related? In other words, do | they or do they not intergrade? While I have no proof that both forms are actually found together during the breeding season, it is evident that in Uruguay, Paraguay, and northern Argentina the limits of their ranges very closely approach, if they do not actually touch, each other.! It is also a fact that throughout its wide range cyanoleuca shows no appreciable geographic or racial variation, specimens from Uruguay, _ ‘Until we know the facts in regard to the seasonal distribution of patagonica in Peru, we cannot discuss its relationships with cyanoleuca in that region. 6 AMERICAN MUSEUM NOVITATES [No. 30 Paraguay, and northern Argentina, where the species most closely approaches the home of patagonica,. agreeing with others from Colombia or Costa Rica. It is further true that, except for a racial differentiation on the coast of Peru and certain variations exhibited by four specimens from Til- cara (8000 ft.) near Tucuman, patagonica is everywhere true to type. The Peruvian variation will be returned to later; the Tileara birds re- quire description here. They were taken February 8-12, 1916. An adult female in worn breeding plumage has the tail, wings, and under wing-coverts wholly black as in cyanoleuca. The remaining three speci- mens are birds of the year in postjuvenal plumage. Like the adult female, their wings, under wing-coverts, and tail are as in patagonica, but the longer under tail-coverts are fuscous tipped with cinnamon, while the shorter ones are cinnamon, a character shown by no immature specimen of cyanoleuca which I have examined. While the adult speci- men might be considered an intermediate between cyanoleuca and pata- gonica, the young birds could not properly be so considered. Moreover, an adult and young taken March 12, 1916, above San Pablo (4000 ft.), Prov. Tucuman, are wholly typical of patagonica, the young bird having the lower half of the under tail-coverts white. While, therefore, we have an ultratypical adult of cyanoleuca taken on the Tafi trail (2000 ft.), Prov. Tucuman, April 12, 1916, it is improbable that intergrades between this species and cyanoleuca would be found at the highest of the three _localities mentioned. If those from the intermediate locality were intergrades, the fusion of the two species might be indicated in this region. Under the cir- cumstances, however, I conclude that the variations shown by the Til- cara specimens are either individual or racial, and that they do not indicate the intergradation of cyanoleuca with patagonica. ‘It is worthy of note that the variation shown by the west Peruvian bird, described below, is similar to that exhibited by the Tilcara adult, that is, it has wholly black under tail-coverts. In the paleness of the under wing-coverts the Peruvian form is even further removed from cyanoleuca than is patagonica, and, as shown beyond, it is quite certain that it does not intergrade with that form. Further evidence of the stability of the characters of patagonica, east of the Andes, is shown by the fact that a specimen from La Merced, eastern Peru, agrees closely with others from Argentina. In view of these facts, I conclude that cyanoleuca and patagonica do not intergrade and hence are specifically distinct. 1922] SWALLOWS OF THE GENUS PYGOCHELIDON 7 Pygochelidon patagonica peruviana, new subspecies Sussreciric CHARACTERS.—Similar to Pygochelidon patagonica patagonica but smaller, the under wing-coverts and axillars paler; the under tail-coverts wholly black with bluish reflections, the lower parts, particularly flanks, tinged with grayish. Type.—No. 152,289, Amer. Mus. Nat. Hist.; @ ad.; December 26, 1918; Huaral, Prov. Lima, Peru; H. Watkins. The Pygochelidon of western Peru has heretofore been referred to cyanoleuca but proves to be a northern representative of Pygochelidon patagonica. - Our collection contains forty-four specimens of this race taken from Moquegua in southern Peru, north to Trujillo. Specimens from Arequipa, listed in the ‘Catalogue of Birds of the British Museum’ under “‘ Atticora cyanoleuca,’”’ are probably to be referred to this form, but it apparently does not reach the tableland. Although our Peruvian collector, H. Watkins, has sent us specimens of this swallow from nearly every station from Trujillo to Moquegua, it is not included in his large collections from the Payta region and the Peruvian-Ecuadorian boundary. This fact, in connection with Stolzmann’s! definite statement that the species does not occur at Tumbez, and its absence from Noble’s? collec- tions made from Payta eastward, indicates that it is not found in ex- treme northwestern Peru. : _ Rewations or P. p. peruviana wits P. cyanoleuca.—The possession by peruviana of the wholly black under tail-coverts of cyanoleuca and the wings and tail of patagonica might be accepted as evidence that it was a connectant between these two species. The fact, however, that in peruviana the under wing-coverts and axillars are even paler than they are in patagonica, shows that in this respect it is less like cyanoleuca than patagonica itself. Moreover, the probability that the range of peru- viana does not reach that of cyanoleuca further indicates their non- intergradation. RELATIONS OF P. p. peruviana witH P. p. patagonica.—Although I have séen no specimens of either of these races from the region between Tofo, Chile, and Moquegua, Peru, just north of the Chilian line, the species has been recorded from east of Iquique and doubtless occurs at suitable localities between these two points, for example, Tacna and Arica. Tofo specimens are typical of patagonica; one from Moquegua is immature, but an adult female of perwviana from Cocachacra, on the coast south of Mollendo, shows an approach to patagonica in its larger 11884, ‘Orn. Per.,’ I, p. 245. In view of Stolzmann’s statement that this swallow was replaced at Tumbez by Progne chalybea, it is interesting to note that the only specimens of this martin sent us by Watkins are from localities in the Payta district where he did not take P. p. peruviana. 21918, Auk, p. 458. 8 AMERICAN MUSEUM NOVITATES ~ [No. 30. size (wing, 99 mm.) and the presence of white in the shorter under tail- coverts. The latter character is shown also by a specimen from Pisco, and these two birds indicate, in my opinion, the intergradation of peru-— viana with patagonica. Pygochelidon flavipes, new species Speciric CHARACTERS.—Resembling Pygochelidon cyanoleuca (Vieillot) but the feet yellow and smaller; the upperparts darker; sides and flanks blackish; throat cinnamon-buff, chin dusky. Tyre.—No. 169,932, Amer. Mus. Nat. Hist.; 9; April 7, 1921; Maraynioe, 10,850 ft., Prov. Junin, Peru; H. Watkins. : Description oF Typre.—Upperparts shining, dusky slate-blue; wings and tail blacker, less blue; throat cinnamon-buff, this color tinging the breast, chin dusky; under wing-coverts, axillars, flanks and tibie fuscous-black; under tail-coverts like wings; tarsi wholly bare, with toes and nails buffy yellow. Wing, 92; tail, 47; tarsus, 9; middle toe without claw, 6.5; depth of tail-fork, 11 mm. This species differs so widely from any specimens in our ie col- lections of allied forms that I venture to describe it on the basis of but one example. It is apparently an Andean Temperate Zone representa- tive of Pygochelidon cyanoleuca and the fact that it comes from Maray- nioc, in the humid Temperate Zone of the Eastern Andes of Peru, is a further proof that exceptionally potent forces have prevailed in the zone in which that locality is situated. The bird’s evident relationships with P. cyanoleuca make it of in- terest to note that, like that species, of which specimens were secured only 3000 feet farther down the Valley, it inhabits a wooded region. If, as seems probable, P. p. patagonica occurs at this or a higher altitude in Peru, it doubtless will be found on the treeless areas of the arid Temperate or Puna Zones. From Maraynioc have been described such distinct generic types as Doliornis sclateri, Xenodacnis parina, Pseudospingus xanthophthalmus, and Microspingus trifasciatus, besides a number of species and races all as yet unknown outside the humid Temperate Zone in Peru. 1922] SWALLOWS OF THE GENUS PYGOCHELIDON MEASUREMENTS OF MALES NUMBER .; cyanoleuca, Costa Rica, Irazu, 9000 ft. 2 > Colombia, Caldas, 2000 ft. 1 . ri o Popayan, 6000 ft. 1 3 3 gs Gallera, 5700 ft. 1 oe bi Venezuela, Barsiquimeto 1 % a Ecuador, Rio de Oro, sea-level 3 ts bhi “ Quito region, 8-9000ft. 3 ze * . Portovelo, 2500 ft. 1 = _ Peru, Perené, 2000 ft. 1 ob 3 ** Uteuyacu, above Perené, 4800 ft. 2 = . Bolivia, Prov. Cochabamba, 7700-9400 ft. 5 ye i Brazil, Matto Grosso, 1 es 3 Argentina, Tafi Trail, 2000 ft. 1 P; P. patagonica, Argentina, Mar del Plata 1 ae ¥ Chile, Temuco 1 ce ks “ce “cc Los Andes 1 Abts " ‘“ Tofo, sea-level 2 P. p. peruviana, Peru, Huaral, sea-level 3 me a3 > cc Lima pe WING 95-96 .5 94 98 97 91 86-90 97-98 94 95 98-102. 98-103 98.5 94 101 101 101 101-101 .5 96-97 95.5-96 MEASUREMENTS OF FEMALES F- cyanoleuca, Costa Rica, Turrialba, 3700 ft. 2 Colombia, Caldas, 2000 ft. ‘ & et San Antonio, 6800 ft. &“ «6 . La Palma, 5500 ft. ‘ ce Ecuador, Portovelo eo " Peru, Ft. Machu Picchu, 5000 ft. ff ef ‘““ Santo Domingo, 6000 ft. ~ : Brazil, Therezopolis, 3200 ft. ” ¥ Bolivia, Prov. Cochabamba, 8800 ft. P. p. patagonica, Argentina, Tileara, Jujuy, 8000 ft. Ses . Chile, Tofo, sea-level P. p. peruviana, Peru, Huaral ce “ce “ Vitarte “ck “cc “ Huacho NUMBER at —_ Re Dh we WING 90 97 96 95 90 96 91-95.5 94 99 101 101 .5-104.5 93 93-94 94 TAIL 51-51.5 45 50 51 42 43-45 .5 48-51 47 44 49-53 50-53 50 46 52 51 51 50-52 49 5-52 51-53 TAIL 44. 5-45 52 50 MAP SHOWING THE KNOWN vi DISTRIBUTION oF tHe SWALLOWS 4 PYGOCHELIDON ochelidon cyanoleuca................ 1 ochelidon patagonica patagonica..2 OF THE GENUS gochelidon patagonica peruviana....3 SRE RIES AS ee ae ee 1922] SWALLOWS OF THE GENUS PYGOCHELIDON 11 SPECIMENS EXAMINED Pygochelidon cyanoleuca——Costa Rica: San José, 1; Boruca, 2; Trazi, 2; Turrialba, 1. CotompBra: Caldas, 4; San Antonio, 2; Popayan, 2; Gallera, 2; Ricuarte, 1; Salento, 2; El Eden, 2; Rio Toché, 2; Sta. Elena, 2; La Palma, 1; El Carmen (near Bogota), 1. VENEZUELA: Barsiquemeto, 2; Mérida region, 1; Caracas, 2; GuAcharo, 2. Brazi: Therezopolis, 1; Rio region, 2; Chapada, Matto Grosso, 1. Ecuapor: Rio de Oro, Prov. Manavi, 3; ‘Quito’ skins, 7; Mocha, 1; Portovelo, 2; Zamora, 1. Peru: Chilpes, Prov. Junin, 1; Tulumayo, Prov. Junin, 2; Uteuyacu, Prov. Junin, 2; La Merced, 1; Perené, 2; Sta. Ana, 1; Ft. Machu Picchu, 1; Torontoy, 1; Santo Domingo, 10. Bortvia: Prov. Cochabamba, Incachaca, 7; Parotani, 5; Vinto, 2; Prov. Sucre, Pulque, 4; Rio Pilcomayo, 1; Rio Cachimayo, 1; California, Prov. Santa Cruz, 1. Arcentina: Tafi Trail, Prov. Tucuman, 2000 ft., 1, April 12. Paraauay: Bahia Negro, 1. Urueuay: Lazeano, Dept. Rocha, February 7, 8 (Wetmore). Pygochelidon patagonica patagonica.—Paracuay: 200 km. west of Puerto Pinasco, September 24 (Wetmore). ARGENTINA: above San Pablo, 4000 ft., Prov. Tucuman, 2, March 12; Tilcara, 8000 ft., Prov. Jujuy, 4, February 8-12; Mar del Plata, 1, October 19; Mendoza, 1, November; Zapala, Neuquen, December 8, 1920 (Wetmore); General Roca, Rio Negro, November 27, 1920 (Wetmore); N. W. Rio Negro, breeding (Peters); Tunyan, Prov. Mendoza (Peters); Protrerillos, 5000 ft., Prov. Mendoza (Peters); Protrerillos, 5000 ft., Prov. Mendoza, March 19, 1921 (Wetmore); Carhué, Buenos Aires, December 18, 1920 (Wetmore); Guamini, Buenos Aires, March 5, 7, 1921 (Wetmore). Cute: Temuco, Cautin, 1, October; Los Andes, 2700 ft., 1, August; Tofo, 60 miles north of Coquimbo, 4, December 3-24. Pygochelidon patagonica peruviana—Prru: Moquega, 1; Coca- chacra, 1; Vitor, 1; Pisco, 1; Lima,3; Chorillos, 1; Vitarte, 6; Huacho, a Teaudias 3; Huaral, 14; Trujillo, 6. Pypadialédon dietoes: aiatae: Maraynioc, the type. GENERAL REMARKS AND CONCLUSIONS From the preceding studies it appears that the west Peruvian form, of a group distributed from Costa Rica to Tierra del Fuego, was de- rived not from the north, as might be expected, but from the south. The geographical origin of the group must remain as much a matter of specu- lation as its ancestry. The extent and geological history of the area it 12 AMERICAN MUSEUM NOVITATES [No. 30 inhabits, however, indicates that cyanoleuca is the older of the two species composing it. This species has been shown to be chiefly mountain- inhabiting, its range being subtropical rather than tropical. If now we look for an ancestral form inhabiting the tropics we find nothing nearer to cyanoleuca than Diplochelidon melanoleuca, which differs from it Gans jae ster wa WiC AL ERAN | eT ; 1 P. cyanoleuca a a aes =A RAT re 7 WENGB)AAle Beet. aeeee | PN Wee Be TY 3 P.patagonica peruviana fon (XS Pe l 2 P.patagonica patagonica 1922} SWALLOWS OF THE GENUS PYGOCHELIDON 13 in the more complete adherence of the basal phalanx of the middle toe to the outer toe, its more deeply forked tail, and bluish black breast- band. The fact, however, that, without regard to locality, specimens of cyanoleuca often have partially black feathers in the breast, suggest that this is a vestigial character, a matter of possiblé significance in this connection. However this may be, the obvious distinctness of cyanoleuca sug- gests the probability of its development in the older mountain areas of Guiana or Brazil; whence, following the coastal mountains, it, on the one hand, reached the Venezuelan Andes and, on the other, subtropical latitudes to the south. Once in the Andean region, it found suitable conditions for range extension in the practically unbroken expanse of the Subtropical Zone, stretching from Venezuela to Bolivia but, although it has reached the Pacific coast just south of the equator, the northward extension of the Temperate Zone on the coast of Peru, under the influence of the Hum- boldt Current, checked its progress on the coast south of Ecuador. Furthermore, since cyanoleuca is found in the vicinity of forests, and probably nests in holes in trees, the absence of wooded areas on the coast of Peru would discourage it from entering that region. Through the mountains of Panama it reached Costa Rica. It appears to be uncommon in the Isthmus at present, a fact readily to be accounted for by the subsidence which has created a hiatus in the ranges of many species common to the Subtropical Zone of Colombia and Panama, but which are wanting in the intervening area.! From the theoretical point of origin in the mountains of eastern South America, the range of cyanoleuca extends southward to subtropical latitudes in Uruguay, Paraguay, and northern Argentina. Here it meets the northern limit in this region of the range of patagonica, with which species it is not known to intergrade. The ranges of the two birds apparently do not overlap, but one replaces the other and they may, therefore, be regarded as representative species. The question now arises: assuming that cyanoleuca is the older of the two, is patagonica an offshoot of it or were they derived from a com- mon ancestor? The fact that their ranges join but that the birds do not intergrade indicates that patagonica is not a geographic derivative of cyanoleuca 14 list of these species is given in the author’ ‘ pope on the ‘ Distribution of Bird-Lifein Colombia,’ 1917, Bull. Amer. Mus. Nat. Hist., XXXVI, p. 14 AMERICAN MUSEUM NOVITATES [No. 30 but that both forms had acquired their distinguishing characteristics before they came into contact with each other. Aside from these suggestions, further discussion of their origin in- volves a more exact knowledge of past physiographic and climatic conditions in South’America than we at present possess. Granted, however, that patagonica had a northern ancestor, we may follow its extension of range southward until it reached the ends of the continent. Once well within the South Temperate Zone, its further southward distribution must have occurred only during the warmer part of the year, and with the coming of winter it evidently retreated toward the north where it now remains at the junction of the South Temperate and Subtropical Zones until returning spring permits it to revisit its breeding ground. Thus has been developed the habit of migration. It is natural to assume that the Pacific coast was reached from Argentina. Thence, west of the Andes, for the facts all indicate the birds’ appearance on the Peruvian coast since the elevation of these mountains, it extended its range toward the equator following that arm of the Tem- perate Zone which, under the influence of the Humboldt Current, passes up the coast to the vicinity of Trujillo, the most northern point at which this swallow is known to occur. Here the counteracting forces of a cold current on the one hand and the approach to equatorial regions on the other, produce what may be termed a Subtemperate Zone, where condi- tions differ sufficiently from those prevailing farther south to bring about the development of numbers of new forms and our swallow now becomes smaller and presents slight but constant color characters which distinguish it from true patagonica, the parent form. The case is an especially interesting one to compare with that of cyanoleuca and patagonica as outlined above. Although in both instances the differentiating characters are slight, in one we have contact of range without intergradation, in the other apparently gradual merging of one form into its representative race. North of Trujillo the effects of a warm, southward flowing, inshore current begin to be apparent, and this vicinity marks the northern known limit on the coast of Peru of Humboldt’s Penguin (Spheniscus humboldti) and the Diving Petrel (Pelecanoides garnoti) the ranges of which extend northward, as Murphy! has shown, to 7°, and 6° 21’ south latitude respectively. 11920, Brooklyn Museum Quarterly, p. 91. 1922} SWALLOWS OF THE GENUS PYGOCHELIDON 15 This study, therefore, shows that, when due consideration is given to relationships and their bearings on geographic as well as physical origin, and to those factors which determine climatic and hence faunal conditions, it is not surprising that the torm of Pygochelidon inhabiting the coast of Peru was derived from the south instead of from the north. Postscript.—In reply to an inquiry, Mr. Harry Watkins, our col- lector in Peru, writes from Lima that he saw no swallows at any of his ~ several stations in the coastal region of Peru north of Trujillo. His ex- perience in regard to these birds agrees, therefore, with that of Stolamann at Tumbez, as quoted above. ‘ AMERICAN MUSEUM NOVITATES No. 31 DESCRIPTIONS OF APPARENTLY NEW BIRDS | FROM COLOMBIA, ECUADOR. AND ARGENTINA : By Frank M. CHAPMAN Issued March 2, 1922 By ORDER OF THE ‘TRUSTEES oF THE AMERICAN MUSEUM OF NATURAL HISTORY NEW Yorx City { AMERICAN MUSEUM NOVITATES Number 31 March 2, 1921 59.82(8) DESCRIPTIONS OF APPARENTLY NEW BIRDS FROM COLOMBIA, ECUADOR, AND ARGENTINA By Frank M. CHAPMAN The following descriptions of apparently new South American birds are chiefly by-products incident to the preparation of a report on the distribution of bird-life in Ecuador. I have to thank Dr. Percy R. Lowe of the Bird Department of the British Museum, Mr. Outram Bangs of the Museum of Comparative Zoology, and Dr. E. W. Nelson of the Biological Survey, for the loan of specimens or for permission to examine the collections under their care. Zenaida auriculata cauce, new subspecies Supspeciric CHARACTERS.—Similar to Zenaida auriculata auriculata (Des Murs) but size slightly smaller, the tail more graduated, its central feathers more pointed, the lower tail-coverts usually more or less vinaceous. Type.—No. 109,386, Amer. Mus. Nat. Hist.; o ad.; Cali, Colombia; August 11, 1911; Richardson and Miller. SPecIMENS EXAMINED Zenaida auriculata cauce.—Coromstia: Cali, 3 3,1 9; Palmira, 1. Zenaida auriculata auriculata.—CuiLE: Temuco, 1 o,1 9; Tofo, 2 ¢. Zenaida auriculata hypoleuca.— Pearl Island,” the type (in Brit. Mus.). Ecvuapor and PErv, a large series. On two former occasions,! I have called attention to the resemblance between specimens of Zenaida auriculata from Chile and the Cauca Valley, but the uncertainty in regard to the proper disposition of Bona- parte’s name Zenaida hypoleuca has discouraged an attempt to determine the true status of the Colombian bird. Examination of Bonaparte’s type in the British Museum shows it to be a specimen of the form for which Bangs and Noble proposed the name Zenaida auriculata pallens and, in satisfactorily disposing of this question, shows also that the Colombian form is unnamed. I have therefore characterized it as above. Its characters appear to present an exceptionally interesting case of parallelism. Briefly stated, the case, so far as it relates to specimens from the Pacific coast region, is as follows. The type form, auriculata auricu- 11917, Bull. Amer. Mus. Nat, Hist., XXXVI, p. 206; 1921, Bull. U. S. Nat. Mus., 117, p. 47. 2 AMERICAN MUSEUM NOVITATES [No. 31 lata, occupies Chile from at least Temuco to Tofo, sixty miles north of Coquimbo. It is characterized chiefly by its almost uniform vinaceous underparts, the abdomen as well as the breast being of this color. In Peru and Ecuador this form is replaced by one in which the abdominal — region is cream-buff, with the under tail-coverts paler than in the Chilian bird. In the Cauca Valley of Colombia, however, a return is made to the characters of the Chilian bird, the bird of that region having the abdomen vinaceous, as in true auriculata, and indeed differing from that race but slightly in size and color. The difference in the graduation of the tail is marked and diagnostic. ’ MEASUREMENTS Extent of — Sex Wing Tail Tail -- Graduation Cali, Cauca, Colombia of 139 86 23 si r- " of 143 85 25 “cc “ ce fot 135 88 31 Temuco, Chile of 152 90 14 Tofo, Chile of 158 92 16 re te ron 151 90 20 Cali, Colombia fo) 133 79 21 Palmira, Colombia fe) 135 75 24 Temuco, Chile Q 138 75 17 . Oreopelia bourcieri subgrisea, new subspecies Susspeciric CHARACTERS.—Similar to Oreopelia bourciert bourciert (Bonap.), but underparts as in O. frenata, the breast pale drab-gray instead of cinnamon-drab or drab; center of abdomen slightly paler than breast, pale smoke-gray rather than drab as in most specimens of bourcieri; lower tail-coverts grayer. Typr.—No. 150,984, Amer. Mus. Nat. Hist.; o ad.; Alamor, 4850 ft., Prov. Loja, Ecuador; July 10, 1919; H. Watkins. SPECIMENS EXAMINED Oreopelia bourcieri subgriseaa—Ecuapor: Alamor, 2 o (ine. type), 1 9; — Celica, Prov. Loja, 1 @. Oreopelia bourciert bourciert—CoLomBiA: 3 o’, 8 9. Ecuapor: Rio Zamora, 1 #,4 9; Zaruma, 5 o',8 9,2 ?. Oreopelia frenata.—Botivia: Vermejo, Prov. Santa Cruz, 1 o@; Yungas, Prov. Cochabamba, 1 o’. This proposed new form is obviously an intermediate between Oreopelia frenata and O. bourcieri, its discovery indicating the inter- gradation of these two forms. In the color of underparts it agrees exactly with a male from Vermejo, Bolivia, while above it as closely resembles true bourciert. Comparison of the type of Oreopelia bourcieri baeza Oa a ee es 1922] NEW SOUTH AMERICAN BIRDS 3 Chubb! with true bourcieri convinces me that it is not separable from that species, an opinion which is supported by the fact that five specimens from the Rio Zamora, in eastern Ecuador, are unquestionably typical bourciert. Specimens from Zaruma in southwestern Ecuador, north of Alamor, are intermediate between bourciert and subgrisea and clearly demon- strate the intergradation of these two races. Jacana scapularis, new species Speciric CHARACTERS.—Similar to Jacana jacana jacana (Linn.), but black of the anterior parts of the body less sharply defined from the back and extending back- wards for an inch or more on the scapulars; light area of the two outer primaries white in both adults and young, untinged with greenish; chestnut areas averaging paler; size, averaging larger. 9: Wing, 134; tail, 40; tarsus, 55; culmen, 31. co: Wing, 136; tail, 40; tarsus, 55; culmen, 32. Typre.—No. 119,626, Amer. Mus. Nat. Hist.; o ad.; Chone, Prov. Manavi, Ecuador; December 16, 1912; W. B. Richardson. SPECIMENS EXAMINED Jacana scapularis —Ecuapor: Chone, 1 &@ ad., 2 @ juv., 1 9 ad.; Puna Is., 1 9 ad.; ‘Quito,’ 1 ad? Jacana jacana jacana.—Brazit, Boutvia, northern ARGENTINA, and PARAGUAY, a large series of adults and ycung. Jacana jacana intermedia.—VENEZUELA: Sacupana, 1 o&@ ad., 2 @ ad.; ; Estado Lara, lad. Cotompia: La Morelia, 1 9. So far as I am aware, no other form of Jacana jacana has been recorded from west of the Andes. In Colombia we found this species only at La Morelia*® in the Amazonian Fauna, while specimens from the Magdalena and Cauca Valleys are referable to Jacana nigra. The range of scapularis is therefore separated from that of jacana by the Andes— an impassable barrier—and this fact, in connection with its pronounced characters, indicates the specific distinctness of the west-Ecuador bird. In this connection it is also pertinent to note that the white outer primaries characterize the immature as well as the adult bird. Rupornis magnirostris zamorz, new subspecies SussPeciric CHARACTERS.—Similar to Rupornis magnirostris magnirostris (Gmel.) but darker, the upperparts deep neutral gray, the bars of the abdominal region and tibiz russet, the bordering gray bars darker and broader, sometimes largely 11917, Bull. Brit. Orn. Club, XX XVIII, p. 33 (Baeza, East Ecuador). *This specimen, from the Lawrence Collection, isa trade skin. It, of course, did not come from Quito, but its black scapulars and white outer primaries indicate that it did come from Ecuador. *Recorded by me (1917, Bull. Amer. Mus Nat. Hist., XXXVI, p. 225) as Jacana.spinosa but, fol- lowing Ridgway’s treatment of this group, should evidentl y be known as Jacana jacana intermedia Scl. 4 AMERICAN MUSEUM NOVITATES [No. 31 replacing the russet bar lying between them; bars on tibie of the same color as those on abdomen; tail, upper tail-coverts, and under wing-coverts without trace of ochra- ceous or cinnamon-rufous. Type.—No. 166,708, Amer. Mus. Nat. Hist.; 9 ad.; Sabanilla, alt. 5700 ft., Rio Zamora, Prov. Loja, Ecuador; November 9, 1920; George K. Cherrie. ‘ SPECIMENS EXAMINED Rupornis magnirostris zamore.—Ecuapor: Sabanilla, 1 (the type); Zamora, 2. Cotomstia: La Palma, 5700 ft., Huila, 1. Rupornis magnirostris magnirostris—SuRINAM: Vicinity of Paramaribo, 9; Wannaweg, 1. VENEZUELA: Sacupana, Lower Orinoco, 1; La Union, Maripa, Lower Orinoco, 1; R. Caura, 2; Cumanacoa, Bermudez, 1. Cotompra: Barrigon, 2; Villavicencio, 3; Honda, 3; Chicoral, 2; Sta. Elena, 1; Barro Blanco, 1; Atrato River, 1; Puerto Valdivia, 1; Dabeiba, 1; Bonda, Sta. Marta, 3!; Sta. Marta, 1.1 The form here described exhibits the distinguishing characters which one would expect to find in a bird inhabiting the humid forests of the Amazonia slope of the Ecuadorian Andes. Its occurrence at La Palma, near the head of the Magdalena Valley, further emphasizes the Amazonian affinities of the avifauna of that locality, where we have already found Piaya cayana mesura and Tangara cyaneicollis caeruleo- cephala.? Ciccaba xquatorialis, new species Speciric CHARACTERS.—Resembling Ciccaba albigularis but throat not white, crown barred instead of spotted, an evident buffy nuchal collar, outer webs of seapu- lars buffy, primaries barred on inner webs; abdominal region whiter. Typr.—No. 35,591, Amer. Mus. Nat. Hist.; “Ambato” (probably East Andean slope below Los Bafios), Ecuador; M.A. Vascomez. — SPECIMENS EXAMINED Ciccaba xquatorialis—Ecuapor: the type. Ciccaba albogularis—Co.LomBia: Choachi (proposed type-locality, see Chap- man, Bull. Amer. Mus. Nat. Hist., XXXVI, 1917, p. 254); 4; Bogoté, 1; Sta. Elena, Antioquia, 1 9; Medellin, 1. Venezunua: Escorial, near Mérida, 1 @,1 9; — Culata, near Mérida, 1 @,1 9. Ecuapor: no locality, 2. Description oF TyPE.—General color above dark cinnamon-brown with buffy and ochraceous markings; band from the base of bill to back of eye white slightly stained with ochraceous; whole top of head finely barred with ochraceous-tawny; a narrow nuchal band largely ochraceous-buff; auriculars ochraceous-tawny barred with black; postauricular region ochraceous-buff barred with blackish; back slightly lighter than crown, the ochraceous-tawny bars wider; outer web of most of the seapu- lars white tinged with buff, and terminally margined with blackish; tail brownish fuscous barred and, laterally and terminally, marbled with ochraceous- buff or ochraceous-tawny; wing-quills like the tail, their outer webs with broadly con- 1\Rupornis magnirostris insidiatrix Bangs and Penard. 21917, See Bull. Amer. Mus. Nat. Hist., XXXVI, pp. 321, 598. ee 1922] NEW SOUTH AMERICAN BIRDS 5 spicuous ochraceous-buff bars which, on the secondaries, have dusky centers; inner webs of all the quills with rather poorly defined but evident ochraceous-buff bars; wing-coverts like the back barred and mottled with ochraceous-buff and ochraceous- tawny; throat and chest ochraceous-tawny finely barred with black; breast and flanks whitish with blackish shaft-streaks, mottlings, fine broken bars and ochraceous- buff bases; lower tail-coverts and tibise ochraceous-buff, the center of the abdomen paler. Wing, 197; tail, 98; tarsus, 24; culmen, 21 mm. The single specimen on which this species is based has remained unidentified for years in the Museum collection in the hope of the re- ceipt of additional material. My work on the birds of Ecuador necessi- tating the determination of its status, the conclusions reached are pre- sented herewith. In the pattern of coloration of the scapulars this species resembles Otus nudipes, but the resemblance ends there, the tarsus in equatorialis being feathered as in Ciccaba, while the upperparts are finely barred instead of spotted with ochraceous-tawny; nor can I detect ear-tufts in the Ecuador bird. The specimen on which this species is based was part of a collection containing examples of Osculatia sapphirina, Pionites melanocephalus pallidus, and other species characteristic of the eastern slope of the Andes, from which region the collection doubtless reached Ambato throug Los Bafios. Glaucidium brasilianum tucumanum, new subspecies Susspeciric CHARACTERS.—Resembling the black and white-barred tail phase of Glaucidium brasilianum brasilianum but upperparts, wings and streaks below fuscous with (in one specimen) a barely perceptible tinge of brown; the crown with small, inconspicuous whitish spots or shaft-streaks; broken nuchal band, white; back with practically no white markings. <; Wing, 90; tail, 65; tarsus, 16 mm. Q: Wing, 95; tail, 65; tarsus, 16 mm. Tyre.—No. 140,625, Amer. Mus. Nat. Hist.; <; Rosario de Lerma, 4800 ft.; Proy. Salta, Argentina; January 10, 1916; Miller and Boyle. SpecIMENS EXAMINED Glaucidium brasilianum tucumanum.—ARGENTINA: Rosario de Lerma, 1 3,2 9+ Glaucidium nanum.—CuiLE: Tierra del Fuego, 1 &@; Punta Arenas, 1 @,1 9; Temuco, 1 @,3 9; Valparaiso, 1; Rio Blanco, 1 @; Tofo, 60m. north of Coquimbo, 1 Q. Peru: Moquegua,! Prov. of Moquegua,1 9. Glaucidium brasilianum brasilianum.—ARGENTINA: La Valle, Santiago del Estero, 1 @, gray phase, 1 <, int. phase, 1 9, int. phase; Suncho Corral, Santiago 1This specimen considerably extends the known range of this species. It is in the grayish brown phase of color with a blackish tail barred with ochraceous-tawny. The tail, however, is tipped with white and the intervening blackish areas are twice as wide as the tawny bars. The spots on the crown are larger and more numerous than in our other specimens, and the bird may represent a northern form of nanum. It measures: Wing, 196; tail, 78 mm. G5 AMERICAN MUSEUM NOVITATES [No. 31 del Estero, 1 @, gray phase. Braz: Urucum near Corumbé, 1 2’, rufous phase, 2 9, gray phase; Descalvaldos, Matto Grosso, 1 o’, rufous phase; Chapada, Matto Grosso, 1 9, rufous phase, 1 9, int. phase, 3 ?, gray phase; eastern Brazil, 3 rufous phase (including two types of ferruginea Wied); Ceara, 1 gray phase, 1 rufous phase; Boa Vista, Maranhao, 1 9, gray phase. Frru: Ollantaytambo, Urubamba Valley, 1 &, gray phase; Pacific coast region, 15, both sexes and phases. Western Ecuapor: 25, both sexes and phases. Glaucidium brasilianum phalaenoides.—TRINiwaD and Caribbean coast region of VENEZUELA and CotomsiA, 24, both sexes and phases. While I have no doubt of the distinctness of the form here described, I do not know whether it should be accorded specific or subspecific rank. The three specimens on which it is based are as nearly alike in color and markings as three specimens of this group can well be and are not to be matched in the large series of other forms examined. The type has a tinge of cinnamon on some of the white areas of the central tail-feathers and the crown markings are minute spots rather than shaft-streaks. All the specimens were taken on January 10 and are in full molt; the in- coming plumage resembles in color that which it is replacing. The four specimens from Santiago del Estero are apparently typical of brasilianum, though I have not enough eastern Brazilian material to determine this point satisfactorily. Two of these birds are in gray plum- age with thickly spotted crowns and more or less white in the back and bear no close resemblance to the form here described. Five “gray” birds from Chapada and Urucum, Matto Grosso, on the other hand, agree with the Rosario de Lerma specimen in the restriction of the crown markings and lack of white on the back, but they are very much browner above than the sooty-fuscous tueumanum. Possibly the absence of crown- markings may be attributed to immaturity, since at least two of these Matto Grosso birds are not fully adult. With Glaucidium nanum also recorded from the Tucuman! region, it is clear that much work remains to be done before we have a thorough knowledge of the distribution and relationships of the forms of this genus occurring in northwestern Argentina. Grallaricula fiavirostris ochraceiventris, new subspecies Susspeciric CHARACTERS.—Similar to Grallaricula flavirostris costaricensis Lawr., but wing longer, bill stouter and averaging longer; ochraceous of underparts more extensive, wholly, or in part covering the abdomen. Differing from Grallaricula flavirostris flavirostris Scl. in its more olive upperparts, more intense color of the 1See Dabbene, 1910, Ann. Mus. Nat. Buenos Aires, III, part 11, p. 255. 1922] NEW SOUTH AMERICAN BIRDS 7 ochraceous areas; comparatively unstreaked breast and ochraceous abdomen. <: Wing, 68; tail, 24.5; tarsus, 21; exposed culmen, 15mm. Q: Wing, 67; tail, 25; tarsus, 23; exposed culmen, 15.5 mm. Tyre.—No. 109,636, Amer. Mus. Nat. Hist.; o ad.; Cocal, alt. 4000 ft., Western Andes, Colombia; June 13, 1911; W. B. Richardson. Grallaricula flavirostris zarumz, new subspecies Supspeciric CHARACTERS.—Similar to Grallaricula flavirostris ochraceiventris but all the ochraceous areas yellower, less orange-ochraceous; that of the forehead, loral and ocular region less pronounced, the auriculars more olivaceous; black ante- orbital crescent stronger; maxilla yellow or olive-yellow, rather than dark brown. Differing from Grallaricula flavirostris flavirostris Scl. in having the back light brown- ish olive rather than Dresden-brown; the breast practically unstreaked; the ab- domen ochraceous; the maxilla yellower. o: Wing, 65; tail, 25; tarsus, 23; ex- posed culmen,15mm. @Q: Wing, 65; tail, 25; tarsus, 22; exposed culmen, 14.5. Type.—No. 129,758, Amer. Mus. Nat. Hist.; near Zaruma, alt. 6000 ft., Prov. del Oro, Ecuador; October 5, 1913; W. B. Richardson. Specimens ExAMINep! Grallaricula flavirostris ochraceiventris——CotompBia: Cocal, the type; San Antonio, 1 9. . Grallaricula flavirostris flavirostris —Ecuapor: Napo, the type; Baeza, 1; Sara- yacu, 2. Cotomspra: Bogotd, 2; “Colombia,” 2 (all in Brit. Mus.). _ Grallaricula flavirostris costaricensis—Costa Rica: Buena Vista, 1; Sarapiqui, 1 9; “Costa Rica,” 2. Panama: Veragua, 5; Chiriqui, 1 9 (type of G. vegeta Bangs). e Grallaricula flavirostris brevis —PANAMA: Mt. Pirri, 2 o&@ (ine. type), 2 9. Grallaricula flavirostris zarumz—Ecuapor: near Zaruma, 6000 ft., Prov. del Oro, 2 ¢@ (ine. type); El Chiral, 5350 ft., Santa Rosa-Zaruma Trail, 3 7,2 9. Examination of the type and other specimens of Grallaricula flavi- rostris flavirostris Scl. in the British Museum enables me for the first time satisfactorily to determine our specimens of this group from the western Andes of Colombia and Ecuador. It appears that the west- Ecuador birds, which I had hitherto provisionally referred to the east- Ecuador form, are quite distinct from it. They not only have an even yellower bill but the black margins to the feathers of the breast are less pronounced than in any other race of the species, being in some speci- mens practically absent, while in true flavirostris they reach their maxi- mum of development and are in strong contrast to the ochraceous or white portion of the feather on which they appear. Furthermore, zarumz, agreeing with the other western races, has the upperparts f 1[t should be stated that the spe*imens listed below were not all examined at the same time and direct comparison in every instance has, therefore, not always Leen rossible. However, all this material has passed through my hands and a large part of it, including the types or topotypical specimens of every form, has been in my possession at the American Museum. 8 AMERICAN MUSEUM NOVITATES _ decidedly more olivaceous than flavirostris. The diffe the two may, in fact, prove to be specific rather than su flavirostris would stand alone and the western forms would races under the specific name of costaricensis. All, however. representative forms and their relationships seem most t by trinomials. The species appears to be confined to the Subtropbotd known distribution of its races is indicated by t the list which specimens have been secured. Maes The genus has not been recorded from Peru; but Bolivia by the recently described Grallaricula t a strongly marked species with a white chest space, “the throat as well as breast margined with black. The margin feathers are broad, sharply defined, and usually enc ge . ochraceous area, producing the sealed effect, which | breast of Premnornis. 11919, Proc. Biol. Soc. Wash., XXXII, p. 257. i AMERICAN MUSEUM NOVITATES— | No. 32 PRELIMINARY REPORT ON ECUADOREAN MAMMAIS. No. 2 By H. E. : ANTHONY Issued March 4, 1922 By Orprr or THE TRUSTEES oF THE AMERICAN MUSEUM OF NATURAL HISTORY New Yorx City AMERICAN MUSEUM NOVITATES Number 32 . March 4, 1922 59.9(86.6) PRELIMINARY REPORT ON ECUADOREAN MAMMALS. No. 2 By H. E. AnrHony This is the second paper based upon the collection of Ecuadorean mammals made in 1920 and 1921.1. The mammals of this expedition, some 1550 specimens, are being studied with the intention of publishing a detailed report at some future time. Additional material is being acquired in the meantime, since it is part of the plan to carry on system- atic field work in this republic. Whenever new species come to light, preliminary reports, such as the present paper, will appear. Phyllotis fruticicolus, new species Typre.—No. 47674, Amer. Mus. Nat. Hist.; 7; Guachanam4, southern Ecuador? altitude, 9050 ft.; October 9, 1920; collector, H. E. Anthony. The type is askin and skull, the latter with the sutures of the parietal region opened, but the elements have not been lost. _ Generat Cuaractrers.—Resembling haggardi superficially, but differing from it in the characters of longer tail, smaller ears, grayer appearance and shorter nasals. DerscripTion.— Color above a mixture of warm buff and blackish, the general effect being de- cidedly grayish, the blackish coming from the dark-colored bases of the hairs, the buff strongest along the sides and about the head; below, soiled white, the hairs slate- colored at base. Ears small for the genus, colored like back without any conspicuous areas of buff or black hairs; hands and feet white; tail bicolor to match body coloring. Skull rather lightly built; nasals very short, with median depression; inter- orbital constriction greatest midway of frontals, margins smooth and unbeaded. MEAsUREMENTs.—Taken in the flesh: total length, 193 mm.; tail vertebra, 102; hind foot, 24. Greatest length of skull, 24.7 (25.3)?; length of nasals, 8.9 (10.3); zygomatic breadth, 12.4 (13.5); interorbital breadth, 3.9 (4.2); breadth of braincase, 11.8 (12.5); length of upper molar series, 4.2 (4). The type of fruticicolus is a young adult, with the molar crowns showing only moderate wear, so it is possible that older specimens might show more highly colored pelage. In color the Guachanamaé specimen may be closely matched by selected specimens from a large series of topo- typical haggardi, but the new species is well characterized by the notice- 1See ‘Preliminary Report on Ecuadorean Mammals. No. 1,’ by H. E. Anthony, American Museum Novitates, No. 20, November 3, 1921. : 2Measurements in parentheses are those of No. 46835, Phyllotis haggardi from Mt. Pichincha, 12,000 ft. altitude. 2 AMERICAN MUSEUM NOVITATES [No. 32 ably shorter nasals and the longer tail. The ear of fruticicolus is smaller and less conspicuous than the ear of haggardi, although the series of the latter would appear to show that there is considerable individual varia- tion in the size of the ear of haggardi. The type is one of a series of three specimens, all collected at Guach- anamé, on a rocky hillside overgrown with tough shrubbery. Microsciurus sabanille, new species Typr.—No. 60464, Amer. Mus. Nat. Hist.; 9 ad.; Sabanilla, Prov. de Loja, Ecuador; altitude, 5700 feet; November 18, 1920; collector, H. E. Anthony. The type is a skin with skeleton. GENERAL CHARACTERS.—A large species, apparently related to avunculus, prac- tically unicolor above, with ochraceous underparts and lacking a postauricular spot. DerscripTion.— Color, above, between dark olive (Ridgway) and mummy brown, depending on how skin is held to the light; below, near ochraceous tawny; top of head lightly washed with ochraceous buff, but not conspicuously so; ears clothed interiorly with short hairs slightly darker in tone than the ochraceous buff of the crown; no post- auricular patches; orbital ring scarcely discernible; forearms and hind limbs like back; hands like back, feet washed with same color as hairs on ears; tail, above, very similar to back, but hairs tipped with clay color, below noticeably darker than color of rest of underparts. Skull large and characterized by long nasals and long palate, the latter extending considerably beyond plane of last molars posteriorly. ME®ASUREMENTS.—Taken in the flesh: total length, 280 mm.; tail vertebra, 146; hind foot, (c.u.), 41. Skuil, greatest length, 37.8; zygomatic breadth, 22.3; length of nasals, 11.7; interorbital breadth, 13.1; breadth of braincase, 19; length of palate, 16.5; length of upper molar series (exclusive of small pm.), 6.1. In pattern of coloration sabanille comes logically within the group of species characterized by similis. The uniform coloration above, without any black dorsal region or crown and with no conspicuous white or buffy areas on or behind the ears, sufficiently distinguishes this new species from the other Microsciurus already known from the Oriente of — Ecuador, napi and avunculus, and from other members of the genus outside of the similis group. It is much larger than any species related to similis and in this respect more nearly resembles avunculus. The type locality of avunculus is given as Gualaquiza, altitude 2500 feet, while sabanillz was taken in forest at an elevation of 5700 feet, which fact may account for the seemingly strange occurrence of two distinet forms so close together, since Sabanilla is on the Rio Zamora not very far distant from Gualaquiza. Thomas,' in the type description of avunculus, com- 11914, Ann. Mag. Nat. Hist., (8) XIII, p. 574. 1922] ECUADOREAN MAMMALS 3 pares it to napi which in turn is “ quite like peruanus.”’ Lacking specimens of both avunculus and napi, I have turned to the type of perwanus, which I find to be very different from sabanillz, not only in the possession of white auricular patches but in cranial characters as well, since the new species has proportionally much longer nasals and longer palate. Marmosa perplexa, new species | Typre.—No. 47188, Amer. Mus. Nat. Hist.; 9 ad.; Punta Santa Ana, Prov. de Loja, Ecuador; altitude 3650 feet; December 21, 1920; collector, H. E. Anthony. The type is a skin and skull, the posterior part of the skull being broken. The type locality is on the trail from Zaruma to Loja and is in the interandean region, on Pacific drainage. GENERAL CHaracters.—Of fairly large size, resembling a small specimen of cinerea, but with fur extending only a short distance onto base of tail; skull short and broad with noticeable frontal depression. DescripTion.— Color above, Saccardo’s umber (Ridgway), the hairs slaty at base; below, chamois, with light ochraceous-buff wash on pectoral region, everywhere the hairs slaty at base; orbital ring blackish, not very extensive; crown slightly lighter than dorsal region, hands and feet proximally dusky, distally soiled whitish; tail brownish above, noticeably lighter in color, faintly mottled with small, indistinct, blotches of white. _ Skull rather short and broad, with heavy zygomata; a faintly developed supra- orbital bead but no supraorbital process; a marked depression in frontal area at nasal suture; first tooth of upper molar series well developed, second noticeably larger than third. MEAsvuREMENTs.—Taken in the flesh: total length, 305 mm.; tail vertebra, 178; hind foot, 22. Skull, zygomatic breadth, 19.4; least interorbital breadth, 6.2; nasals, 14X5; length of palate, to gnathion, 19; length upper toothrow, C-M‘, 14.8; length M'-M‘, 7.7. This murine opossum shows characters that would place its rela- tionships with the cinerea group, were it not for the peculiarly depressed frontal region and the lack of a postorbital process, which condition is not noted in skulls of cinerea, waterhousii and phaea, the only species which perplexa appears to resemble. Skulls of klagesi from Venezuela show a little of the depressed frontal condition, but klagesi has nothing to do with perplexa in any other character. The material on hand for com- parison is rather inadequate and consequently it is difficult to say with certainty to just what group this new species belongs. Marmosa oroensis, new species Type.—No. 47180, Amer. Mus. Nat. Hist.; o ad.; Portovelo, Prov. del Oro, Ecuador; altitude 2,000 feet; September 2, 1920; collector, H. E. Anthony. The type is a skin with skull. : | AMERICAN MUSEUM NOVITATES [No. 32 GENERAL CHaractrers.—A light-colored form, similar externally to simonsi, but with the smooth interorbital region of fuscata or madescens. DESCRIPTION.— Above, between drab and wood-brown, the hairs with slate-colored bases; below, ivory-yellow over entire throat and chest and medially to root of tail, the hairs uni- color to base; along sides of abdomen the light-colored hairs are slaty at base; orbital spot narrow but extending almost to end of nose, area between orbital spots above, slightly lighter in color than back; ears about like back in color; wrists and ankles dusky; hands and feet yellowish white; tail above, very similar to back in color, below distinctly lighter from base to tip. Skull long and slender, with smooth interorbital region, no beading or postorbital processes. : MEAsUREMENTS.—Taken in the flesh: total length, 294 mm.; tail vertebra, 170; hind foot, 20. Skull, greatest length, 35.1; zygomatic breadth, 16.7; interorbital breadth, 6; nasals, 15.94; palate, to gnathion, 19.0; upper toothrow, C-M*‘, 13.8; length, M'-M*, 6.7. M. oroensis is somewhat like stmonsi superficially, although it does not have quite as long pelage nor such light-colored feet and nose. Cranially the two forms have little in common since the skull of stmonsi has marked postorbital processes. It is much lighter in color throughout than madescens although in cranial characters these two species are some- what similar. It may well be that oroensis should stand as a subspecies of sobrina Thomas which is said to be related to fuscata. The Portovelo specimen is quite like fuscata in skull characters but differs from it sufficiently in external details to be distinct, while from’ sobrina it may be differentiated by the greater extent of the light-colored underparts and by the much lighter tone of the upperparts. Unfortunately, there are no available specimens of sobrina for actual comparison. Apparently oroensis is the opossum of the semi-zerophytic tropical zone, fox animals taken in the same general region but at higher eleva- tion and in the forest are the related celicex.! Marmosa celice, new species Typr.—No. 47182, Amer: Mus. Nat. Hist.; < ad.; Celica, Prov. de Loja, Ecuador; altitude, 6900 feet; September 28, 1920; colloatok, H. E. Anthony. The type is a skin with skull which is broken posteriorly. GENERAL CHARACTERS.—Most like sobrina and oroensis but differing, from the former in having tail bicolor for its full length as well as in smaller size, from the latter in darker coloration and slightly different interorbital region. DerscrIPTION.— Color, above, near sepia with hairs slaty at base; below ivory-yellow medially, the widest area at chest, hairs unicolor to base, encroaching hairs of sides and flanks 1See the following description. eS ee ee ee ee 1922] ECUADOREAN MAMMALS 5 light-tipped with slate-colored base; orbital rings extending almost to end of nose; area above included between orbital rings only slightly lighter in color than dorsal region; hands and feet dusky above, distally whitish; tail like back above, distinctly lighter below throughout entire length. : Skull like that of oroensis but having faint postorbital swellings of the frontal elements which appear to foreshadow postorbital processes. MEASUREMENTS.—Taken in the flesh: total length, 278 mm.; tail vertebre, 157; hind foot, 20. Skull,.interorbital breadth across postorbital swellings, 6.8; least interorbital breadth, 5.8; nasals, 15.5 3.6; palate, to gnathion, 18.5; length upper toothrow, C-M*, 13.5; M'-M*, 6.5. ~ M. celicx is very closely related to oroensis but can be distinguished by its darker color above, its much more restricted yellow underparts and its swollen frontals. Skulls of two females show some of this swel- ling but not to the same degree as seen in the skull of the male (type). From sobrina this new form differs in having a distinctly bicolor tail as well as a somewhat smaller skull. The acquisition of more material is necessary to determine the degree of relationship between celice and oroensis which I suspect may prove to be subspecific, or perhaps both may be best treated as subspecies of sobrina. M. celice was taken at Celica and Alamor near the Peruvian border and at Salvias, in the forest near the headwaters of the Rio Amarillo. The altitudinal range as established by these records is from 3500 feet to 6900 feet, all three localities in forest. Marmosa bombascare, new species Typre.—No. 47186, Amer. Mus. Nat. Hist.; 9 ad.; Zamora (junction of Rio Bombascaro with Rio Zamora), Prov. de Loja, Ecuador; November 24, 1920; collector, H. E. Anthony. A skin in fair condition (eaten about tail and feet by ants while in trap), with skull. GENERAL Cuaracters.—Most like musicola, but larger and lacking postorbital processes on the supraorbital border, but with well-developed beading. DEscriIPTION.— Color above, intermediate between cinnamon and russet, the hairs slate-colored at base; below ochraceous buff, the hairs of the median area unicolor to the base, elsewhere slaty; orbital area black, small subtriangular area between orbital rings noticeably lighter than the rest of upperparts; hands dusky, feet soiled whitish; tail brownish, only slightly lighter below, no white tip. Skull heavily built, with conspicuous supraorbital bead, wide flaring zygomata and no postorbital process on the supraorbital border. MEASUREMENTS.—Taken in the flesh: total length, 313 mm., tail vertebre, 187, hind foot, 21. Skull, greatest length, 34.5; zygomatic breadth, 19.3; interorbital breadth, 6.5; nasals 154.3; length of palate, to gnathion, 19.2; length of upper toothrow, C-M’, 13.4; length, M'-M‘, 7. 6 AMERICAN MUSEUM NOVITATES | M. bombascaree appears to be closely related to - length, the strongly yellow underparts, ur the Poe The type skin, that of a female, has the inguinal area suffus ceous as is also the case with a female of musicola. But: ences in cranial characters, I should have made bombascara of musicola. However, the skull of the new species is so di and the detailed structure of the interorbital area, that. the calls for full appane separation. | “it rane a eran MAP OF ECUADOR 81° 48 senterasel eoseeyeemenses, re seeeseesoee sesso reel, Cal 3° y 1° BO" West of Greenwich 7° une 1920 to March 1921, indicated 1. Map to show route of the expedition to Ecuador, J this paper are by dotted line. The principal collecting stations and type localities mentioned in shown. AMERICAN MUSEUM NOVITATES No. 33 ~ BEES OF THE GENUS PERD/TA FROM THE WESTERN STATES | By T. D. A. CockeErRELi Issued March 28, 1922 By ORDER OF THE TRUSTEES OF THE AMERICAN MUSEUM OF NATURAL HISTORY New York City AMERICAN MUSEUM NOVITATES Number 33 March 28, 1922 59.57.99 P (78) BEES OF THE GENUS PERDITA FROM THE WESTERN STATES ‘i By T. D. A. CockERELL The bees recorded below, with the exception of Perdita wickhami, were obtained by Dr. Frank E. Lutz and his associates during their exploration of the Western States for The American Museum of Natural History. They were collected by Dr. Lutz, except where the contrary is stated. The collection adds ten species and three races or varieties to the already long list of western Perdita and greatly extends our knowl- edge of distribution. The material of the Cockerellia group (albipennis, etc.; large forms, with stout, bent mandibles in the female) is particularly interesting and has suggested a discussion of the evolution of the species and races. Perdita wootone Cockerell Cotorano: 5 &, 4 9, Pueblo, vacant lots in town, August 9, 1920; 1 9,2 9, La Junta, August 12, 1920, Mrs. F. E. Lutz, collector; 2, Wray, August 17-19, 1919; 1 &, Tennessee Pass, about 10,300 ft. alt., August 1, 1919, H. F. Schwarz, collector. For the characters of this species, see 1907 Entomological News, XVIII, p. 57. The two Wray males differ greatly in the size of the head. The specimen labelled Tennessee Pass has an unusual amount of black pigment about the lateral ocelli, but the locality seems almost incredible, as the bee is an oligotropic visitor of Nuttallia, which is not likely to occur at such an altitude. Rydberg has recorded N. speciosa (Oster- hout) from 10,000 ft., but this also seems improbable. It may be, how- ever, that Nuttallia and P. wootone do occur at Tennessee Pass in a favor- able exposure. Perdita solitaria Cockerell Arizona: 2 9, Sabino Basin, Santa Catalina Mts., about 3800 ft. alt., August 15-21, 1916. Described from New Mexico. Perdita callicerata Cockerell ARIZONA: 1 <, southwest end of Coyote Mountains, about 3500 ft. alt., at flowers of Baileya multiradiata, August 4, 1916. Described from New Mexico, where it visits Baileya. 2 AMERICAN MUSEUM NOVITATES [No. 33 Perdita heliophila Cockerell CoLtorapo: 2 9, Mesa Verde National Park, about 6000 ft. alt., at Helianthus petiolaris, July 4, 1919. Described from. New Mexico (1916). Perdita albipennis Cresson Cotorapo: 6 9, 1 &, Wray, most of the females at Helianthus, August 17, 1919, one collected by Pearce Bailey, Jr.; 1 9, Grand Junction, at Helianthus, July 17, 1919; 2 9,1 &, Grand Junction, near town, August 3, 1920; 12 9, 10 3, La Junta, August 12, 1920, one @ collected by Mrs. Lutz, one @ ivan very arid hills, four of the males show yellow markings on the dbdomek (var. helianthi Ckll.); 3 &#, Glenwood Springs, at edge of town, July 22-29, 1919 and August. 5, 1920, one collected by Mrs. Lutz is var. helianthi; 1 #, Pueblo, vacant lots in town, August 9; 1920; 1 o, Montrose, July 19, 1919, H. F. Schwarz, collector. Perdita pallidipennis indianensis, new subspecies 3 (Type).—Clypeus and sides of face wholly without light markings; anterior tibize without a pale stripe; margin of stigma and end of marginal cell fuscous. 9 .—Not appreciably differing from the typical form. INDIANA: 1 9,1 el ee i> Soe a ew is - - OE 3 in | _ AMERICAN MUSEUM. NOVITATES No. 34, | ok FOSSIL MOTH FROM FLORISSANT; COLORADO } By T. D. A. CockERELL B26 & Udy Issued March 29, 1922 By ORDER OF THE “[RUSTEES “OF THE AMERICAN MUSEUM OF NATURAL HISTORY New Yorx City oS rn sind DS RC mls ates iat sah i AMERICAN MUSEUM NOVITATES Number 34 March 29, 1922 56.57,81(1182:78.8) A FOSSIL MOTH FROM FLORISSANT, COLORADO By T. D. A. CocKERELL In 1909 a collection of fossils from the Miocene shales of Florissant, Colorado, was obtained by Messrs. Sternberg, Duce, and Rusk, and transmitted to The American Museum of Natural History. The new species were described by the present writer in the Bulletin of the Ameri- can Museum, XXV (1910), with the exception of a moth, which was re- tained with the expectation that it would be described by another. More than ten years have passed and, as the moth has never been recorded, it seems desirable to offer a description. It is represented by a single anterior wing, but this is unusually well preserved. It is a member of the Geometride and, although its generic position, in the absence of antenne, palpi, etc., must remain somewhat uncertain, it appears to fall in the large modern genus Hydriomena of Hiibner. Fig. 1. Fossil Moth, Hydriomena (?) protrita. Enlarged. Hydriomena (?) protrita, new species ‘Anterior wing, length 23 mm.; costa distinctly arched before apex; outer ‘margin 14.5 mm., gently convex; hind margin 17 mm.; basal space somewhat pallid; subbasal line very faint, but apparently arched outward, its lower part directed basad, and sharply angled a short distance above hind margin, in the manner of H. manzanita ea AMERICAN MUSEUM NOVITATES [No. 34 Taylor, except-that the curve is less abrupt and the angle is more pronounced; ante- median band also formed after the manner of H. manzanita, but more distant from base of wing, with the outward curve broader and less abrupt, though the posterior curve or loop and the angle between the two are nearly the same in the two species; the costal region is strongly infuscated, the dark color being broken by the ante- median band, which appears as a pale subvertical line; near the end of the cell is a distinct discal spot or pair of short transverse bars, apparently representing the dark antemedian mark of H. manzanita; the postmedian line is not so far from the ante- median as usual and appears to be formed much as in H; albifasciata Packard, in- wardly bounding the dark apical and outer marginal areas; the fringe on the outer margin is chequered, as in many species. Be, The specimen bears the number 14. So far as it is possible to see, this insect is of an entirely modern type. Itis the first American fossil geometrid. The fossil Lepidoptera of Florissant, so far as known at present, are as follows, extinct genera being marked with an asterisk. BUTTERFLIES PIERIDZ Jupiteria Scudder* Stolopsyche Scudder* charon Scudder a tea libytheoides Scudder Lithodryas Cockerell* — LIBpYTHEIDA styx (Scudder) Barbarothea Scudder* Apanthesis Scudder* florissanti Scudder leuce Scudder Prolibythea Scudder* Chlorippe Boisduval vagabunda Scudder wilmatiz Cockerell NYMPHALID2 Nymphalites Scudder* Prodryas Scudder* obscurum Scudder persephone Scudder scudderi Beutenmiiller and Cockerell MOTHS SATURNIID AD ToRTRICIDA Attacus (?) Tortrix (?) fossilis Cockerell florissantana Cockerell GEOMETRID& destructus Cockerell Hydriomena (?) ETxHMiip® protrita Cockerell Ethmia (?) mortuella Scudder FAMILY UNKNOWN (larva) Phylledestes Cockerell* vorax Cockerell It will be noticed that the butterflies with one exception are assigned to extinct genera. The moths, on the other hand, are doubtfully referred to living genera. It is possible that if we knew the moths better, we could distinguish them from living genera; but it must be confessed thas there is nothing in their appearance to suggest this. . a a 7 | a ak ee a “AMERICAN MUSEUM NOVITATES No. 35 : A NEW FOSSIL RODENT FROM ECUADOR By H. E. AnTHony “7. (xe ai < Issued: March 30, 1922 By OxgpER OF THE TRUSTEES OF THE AMERICAN MUSEUM OF NATURAL HISTORY New Yorx« Crry SA, ttt, AMERICAN MUSEUM NOVITATES Number 35 March 30, 1922 56.9.32 D (86.6) A NEW FOSSIL RODENT FROM ECUADOR By H. E. AN1Hony Mr. A. M. Tweedy, Resident Manager of the mines of the South American Development Company at Portovelo, Ecuador, who has given such great assistance to the expeditions of the American Museum in Ecuador, has added to the collections a most interesting genus of fossil rodents. He gave to Mr. George K. Cherrie, who was in charge of the Museum’s late expedition to Ecuador, July 1921 to January 1922, parts of a skull and skeleton of a large hystricomorph rodent related, among fossil genera, to Neoreomys of the Santa Cruz beds, and among living forms to Myocastor. This material Mr. Tweedy secured from Sefior Carrasco who owns a large hacienda near Nabon, Provincia del Azuay. The accompanying diagram will show the nature of the locality where this rodent was found (Fig. 1). Nabon is in the interandean area, at an elevation of about 9000 feet. The region is open and treeless, the only forest being found in scattered clumps along the higher ridges to the east. Although this section has a Fossil bones are +e ee a I) ce nic 1 ee i Se Ne Z < oak 2, cone oS 2\ Quebrada ss Ges rat a 40 - D2 rye -Ore2-O : A ). rer 2G Z29- 287 ON O20. 00: 55 eee | £8:.0F:Q 2633 Fig. 1. Diagram based upon a sketch made by Mr. A. M. Tweedy, who was given the data by Sefior Carrasco, the original collector of the fossil bones. The entrance tothe cave was unearthed by a landslide. 2 AMERICAN MUSEUM NOVITATES [No. 35 comparatively long dry season and for most of the year is a region of extreme aridity, it is visited by torrential rains which erode the topog- raphy and scour out the ravines. There is little doubt that this hystricomorph lived in the Pleistocene and such an interesting find leads one to hope that additional researches in Ecuador will bring to light more of the members of the preceding epochs. I am indebted to Dr. W. D. Matthew, Curator of the Department of Paleontology of the American Museum, both for permission to describe this material which belongs in his department and for helpful discussions upon the material itself. Drytomomys, new genus A large hystricomorph rodent with very heavy incisors which have a thick cutting edge of enamel; molars rooted but fairly hypsodont, crown pattern made up of a series of deep reéntrant loops which become isolated by wear to form enclosed lakes, the normal number of which is three. There is a very deep fossa in the lateral face of the premaxillary. Genotype: Drytomomys xquatorialis. Drytomomys equatorialis, new species Type: No. 13219, Department of Vertebrate Paleontology, from Nabon, Proy. del Azuay, Ecuador, near the hacienda of Sefior Carrasco, who took the bones from the cave. The type is a fragmentary skull, only the anterior portion of the cranium, carrying four molar teeth, and the anterior portion of the mandible, all of the teeth present. Accompanying these parts of the skull are a few fragments of limb bones, ete. Descrirtion.—General proportions of cranium unknown; a very deep and apparently extensive fossa is indicated as lying in the lateral face of the rostrum, partly in the premaxillary, partly in the maxillary; incisive foramina probably of fair size; mandible robust and heavy, typically hystricomorph; incisors, upper and lower, very large and strong, proportionally enormous, rather deeper than broad and with well-developed cutting edge of heavy enamel; molar teeth rooted but with high crowns, four in each jaw; molar pattern simple, formed by a series of reéntrant loops of enamel which enter from the inside in the case of the upper molars, externally in the lower series; in the state of wear shown by the type most of these loops have been cut off to become completely detached lakes, elliptical in outline, while the tooth is sub-columnar in shape, completely encircled by enamel with scarcely any trace of the entering loop which was ancestral to the lake; true molars with typical pattern of three loops or lakes, lower premolars with two main lakes and three (left) or four (right) additional small lakes. MEASUREMENTS.—Length of upper diastema, approximate, 32 mm.; dimensions of upper premolar, 7.88.7; dimensions of m!, 6.36.8; dimensions of m?, 7.27.5; length of mandibular toothrow, I-M3, 60; crown length of mandibular molar series, 31.5; dimensions of lower premolar, 9.56.7; dimensions of m1, 6.27; dimensions reer \ — Al Ae XK ( S As / 2 Lf y “ye 4 4 44. 4, Y r Fig. 2. A, B, C, D, Drytomomys zxquatorialis, type specimen (A. M. Dept. Vert. Pal. 13219, Nat. Size). E. Myocastor coypus (A. M. Dept. Mam. 42751, Nat. Size). 4 AMERICAN MUSEUM NOVITATES — __ [No. 35 of mg, 7.57.4 ; dimensions of m3, 8.87.5; breadth of upper incisor, 8.4; depth of upper incisor, 10.7; breadth of lower incisor, 8.4; depth of lower incisor, 10.2. Drytomomys may be readily distinguished from all living rodents by its excessively developed incisors and by the molar pattern of com- pletely isolated enamel lakes. While the skull is of about the same size as in the genus Castor, the incisors are actually twice as heavy and the enamel cutting edge is conspicuously heavier. It was because of this highly developed chisel-like incisor that the new genus was named Drytomomys, from the Greek éputoyos, a wood cutter and wus, a mouse. If I am correct and this rodent was a woodcutter, it was even better equipped than the beaver, and bearing out this assumption is the presence of the deep lateral fossa of the rostrum which undoubtedly 3 mark the attachment area of a greatly developed masseter muscle. Apparently its closest affiliations with living rodents are with Myocaster. . The molars of this genus, in the worn stage, show a fairly close approximation to the condition seen in the fossil genus. However, Myocaster molars have an extra reéntrant, internal in the upper series, external in the lower, which the Drytomomys tooth lacks. A specimen of Myocasier (No. 42751) has a molar toothrow equal in length to that of the new genus, but the incisor teeth are about half as heavy. Among fossil genera,.Neoreomys is similar in many details to the Ecuador rodent but differs from it in the same characters as does Myo- castor. In fact Neoreomys resembles M yocaster more closely than it does Drytomomys. No other genus of the Santa Cruz fauna bears as close a relationship to Drytomomys as does Neoreoniys, but Drytomomys is so obviously related to this genus that it would appear perfectly at home in any collection from the Santa Cruz formation. The fragments of limb bones present, the head of a humerus, distal end of an ulna, head of a femur and badly broken distal end of a tibia, all indicate an animal of robust and heavy build. These limb bones are very much larger than those of Myocaster and only a little less robust — than those of Hydrocherus. | AMERICAN MUSEUM NOVITATES = No. 36 ~ ~ < BEES OF THE GENUS PANURGINUS OBTAINED BY THE AMERICAN MUSEUM ROCKY - MOUNTAIN. EXPEDITIONS By T. D. A. CocKERELL Wh DS 8 Ns * Issued April 14, 1922 By Orpir or THE TRUSTEES OF THE AMERICAN MUSEUM OF NATURAL HISTORY ~ New Yorx Crry 7 fi "> An -pae AMERICAN MUSEUM NOVITATES Number 36 April 14, 1922 59. 57, 99 P (79) BEES OF THE GENUS PANURGINUS OBTAINED BY THE AMERICAN MUSEUM ROCKY MOUNTAIN EXPEDITIONS By T. D. A. CockERELL The bees recorded below were collected by Dr. Frank E. Lutz, except when the contrary is stated. I have included certain species of Greeleyella and Hypomacrotera, which are likely to be confused with Panurginus. The genus Panurginus was first distinguished by Nylander in 1848, the type being P. niger Nylander, known from a female, 5 mm. long, collected by Sahlberg in Siberia. Nylander thought that the lack of copious pollen-carrying pubescence indicated a parasitic mode of life and, on this ground, separated the genus from Panurgus. The hair on the legs was, however, more abundant than that on the body. When Friese revised the Palearctic Panurgine in 1901, he had not seen P. niger, nor have any additional specimens been collected, so far as I am aware. The species is not in the British Museum, at Oxford, or in any of the American collections. The precise definition of Panurginus, . therefore, remains somewhat uncertain, though it is probable that P. niger belongs to the genus as we now understand it. F. Morawitz in 1876 founded a genus Epimethea on certain species differing from Panurginus in having yellow tegumentary markings on the abdomen and other parts of the body. LE. variegata F. Morawitz may be taken as the type. It is recorded as occurring in the Caucasus and in Algeria. Probably Epimethea should be recognized as a genus but, although I have seen EZ. variegata in the British Museum, I have never made a close examination of any species of this group. In 1894 Gribodo based his genus Scrapteroides on S. difformis Gribodo, which is now considered identical with Panurginus albopilosus (Lucas), a species of Spain and Algeria. I possess this species and find that it has a black face in the male and that the first recurrent nervure falls considerably basad of the first intercubitus (transversocubital). In these respects it resembles our American P. atriceps (Cresson), but the latter has a much longer flagellum in the male. Panurginus is confined to the Palearctic, Nearctic, and Neotropical Regions. If we exclude Epimethea (6 species), Greeleyella, and Pseudo- panurgus, we find that, as at present known (including the species de- 2 AMERICAN MUSEUM NOVITATES [No. 36 : scribed below), Panurginus has 51 Nearctic, 25 Neotropical, and only 20. : a Palearctic species. In the Swiss Alps and adjacent regions one may a find P. montanus Giraud (1861) visiting flowers of Ranunculus and Hieracium. But the collector who would obtain other species has to travel far—to Spain for P. albopilosus Lucas and P. hispanicus Giraud, — to northern Sweden for P. romani Aurivillius, to Russia for P. lacti-— pennis Friese' and P. sculpturatus Morawitz. Contrast with this the ~ = abundance of species in our Rocky Mountain country, which appears to- ee be the headquarters of the genus. In Asia, species occur in Japan (P. crawfordi Cockerell) and China (P. nitidulus Morawitz, P. nigripes Mora- witz, P. picitarsis Cockerell), but we do not find them in the typical Oriental tropics. In America, on the other hand, they occur in the moist __ tropics; as in the “Tierra caliente” of Mexico (P. bidentis Cockerell) re. and in Brazil (P. solani Ducke). Nevertheless, the strictly tropical species are not very numerous. If we are surprised that the Central European P. montanus has not - produced any segregates, we must note that our closely analogous species, P. cressoniellus, is also without a series of closely related forms. These bees are indiscriminate feeders on a considerable series of flowers, andit appears to be true that oligotropic habits favor the evolution of new species. We know that species of Greeleyella and Hypomacrotera are oligotropic and there are reasons for regarding several species of Panur- _ ginus as such, but the subject requires much more minute study. In general sweeping, it is easy to obtain from a clump of flowers bees which __ were not actually feeding on the flowers cited, and even oligotropic bees Sie wander from their proper plants at times. The collections of Dr. Lutz greatly increase our knowledge of the species and distribution of western Panurginus, but it is evident that, with all that has been done, we still have but an bee knowledge of ig this rich fauna. Panurginus irregularis, new species Cotorapo: 2 <, Boulder, about 5300 ft. alt., on plains, August 12, 1919. Taken with P. piercei Crawford and P. nebrascensis Crawford. Related to P, horizontalis Swenk and Cockerell, but easily known by the dark tubercles and other characters o'.—Length about 6 mm.; slender. Black; mandibles (except apex), labrum, and face up to level of antenns bright yellow, upper level of yellow irregular, denti- form, with a slender process or each side a little way up orbits. Second jointoflabial palpi much longer than third; third and fourth subequal. Process of labrum very broad, truncate. Front dull, vertex shining, with distinct punctures. Scape yellow — 1The British Museum has specimens of P. lactipennis from Persia. a 4 oa rel arly ler UP ee RL Ge ne tee Ret HO ene ec ge - AN ae Pe eT OO, Paes peer ee ee e O nL eee ee et ee ee ee ee age Lee Ree 3 1922] BEES OF THE GENUS PANURGINUS 3 in front; flagellum pale ferruginous beneath except base and last joint, the latter abruptly dark. Thorax above with thin dull white hair; mesothorax and scutellum polished, with sparse distinct punctures; area of metathorax with fine rugee; tubercles black (yellow in P. horizontalis); tegule testaceous. Wingsstrongly dusky; nervures fuscous, stigma dilute reddish; second submarginal cell much shorter than first, re- ceiving the recurrent nervures about equally distant from base and apex. Femora black, yellow at apex; tibie yellow, the anterior ones with a black patch behind, middle and hind ones each with two black marks (tibize thus as in P. citripes Ash- mead, but face-marks different); tarsi yellow, reddish apically; abdomen with fine punctures. The clypeus has a fine median groove. Panurginus altissimus, new species Cotorapo: 1 & (Type), 2 2°, Ouray, about 8000 ft. alt., July 11, 1919, collected by Herbert F. Schwarz; 1 9, Ouray, about 8500 ft. alt., sweeping grass among Douglas fir, aspen, scrub oak, ete., July 11, 1919. & (Type).—Length about 7 mm., rather slender. Black; mandibles black; labrum black, but the process pale yellow, with sloping sides, and the binodose or subemarginate apical margin black; clypeus, irregularly subtriangular supraclypeal mark and large triangular lateral face-marks (almost reaching upper corners of clypeus) sulphur-yellow. Clypeus sparsely punctured, and with a faintly indicated median groove. Scape black; flagellum black, very obscurely reddish beneath. Mesothorax and scutellum polished, sparsely punctured; area of metathorax finely plicatulate; tegule rufous, piceous in front. Wings dusky, especially the broad apical margin; stigma and nervures dark brown; first submarginal cell about as long as second on lower side; first recurrent nervure joining second submarginal cell a long way from base, fully three times as far as second from apex. Legs (including anterior tibie) black, small joints of tarsi becoming brownish apically. Abdomen ordinary, with very fine punctures. Q.—Length about 7 mm. Black without light markings. Mesothorax highly polished, almost impunctate. The female is so like P. piercei Crawford that I should have regarded it as a mountain race with dusky wings, but the male is much more different. Panurginus opaculus, new species Arizona: 1 92 (Type), Mud Springs, Pine Canyon, Sta. Catalina Mts., about 6800 ft. alt., among oak, pinyon, juniper, walnut, etc., July 17-20, 1916; 1 9,same locality, August 19, 1916. Q.—Length about5mm. Black. Clypeus and supraclypeal area shining, rather closely punctured. Antennz dark, the flagellum very obscurely brownish beneath. Mesothorax dull, with a strong median groove; scutellum shining, flattened on disc, finely punctured; tegule rufous. Wings dusky, stigma and nervures very dark brown. ‘Very close to. P. pauper Cresson (specimens collected by Banks. at Ceanothus, Falls Church, Va., compared) and at first sight appearing identical, but distinguished by the dark antennz, granular basal area of metathorax (plicatulate in pauper), and very broadly truncate process of 4 AMERICAN MUSEUM NOVITATES - labrum (rounded in pauper). Itis larger than P. flavotinctus (Coe and has the process of labrum more broadly truncate (flavotinctus hi ing the labrum intermediate between pauper and opaculus). Panurginus pernitens, new species Arizona: 1 9, 8. E. of Kitt’s Peak, Baboquivari Mts., about 4000 ft lt among mesquite with considerable oak, August 1-4, 1916. ss = -—Similar to P. opaculus, and at first glance appearing identical, bu with a few scattered punctures on a highly polished surface. Process of Is narrower than in P. pauper, rounded at end. Area of metathorax minutely, micro scopically Soesracee aes rather dilute peices Ventral abdominal se much larger than P. flavotinctus. The following key separates the females of the pauper erie 1. Pilagellum: entirely dark 2.0538 °). 5 004. xt alge eee Flagellum red beneath except at base 2. Facial quadrangle much longer than broad Facial quadrangle about square.................0000ceceeeeee Bee 3. Larger; process of labrum rounded....................-. Smaller; process of labrum subtruncate.............. Panurginus portere Cockerell Beulah, New Mexico. The Arizona male is smaller, and has the very dark gray instead of green. This male certainly appears to t specific with the form I recorded (Ann. Mag. Nat. Hist., April corner of the truncation a little higher than that on the orbit. — called picipes have the lateral marks pointed on or very close to the separated, but at present I regard them as a form of that species. Oe 1922] BEES OF THE GENUS PANURGINUS 5 Arizona male has the basitarsi black; the hind ones are largely testace- ous in the Pecos specimens. : Panurginus perlevis (Cockerell) Cotorapo: 1 o, 4 9, Wray, about 3700 ft. alt., August 17-19, 1919. One female was collected by Pearce Bailey, Jr. A male and female were at Helianthus. _ This is very close to P. piercei Crawford, but amply distinct. The females may be separated thus: Flagellum beneath bright ferruginous in middle and dark at ends; clypeus not de- pressed in middle and may be faintly carinate; supraclypeal area with many RENE Otc em cP gel ase "as can tea ww a> Daca) as 8 af ep afe ade. e @dlsws 49 ..perlevis. . Flagellum beneath dusky reddish from near base to apex; clypeus depressed in middle; supraclypeal area with very few punctures................. pierce. Male perlzvis differs from piercet by lack of the median groove on clypeus. There is only a small yellow triangle on the supraclypeal area, and between this and the lateral marks are large black areas (the outline of the yellow forming a broad W), while the lateral marks are obliquely truncate above. The flagellum is light yellowish-ferruginous beneath, except the last two joints and extreme base. The hind tibiz have about the basal half yellow (less in piercei), and the wings are not dusky. The process of labrum is yellow with a black apical margin, very broadly truncate, the truncation somewhat concave. The type of P. perlzvis was collected on Helianthus at Las Cruces, New Mexico. Panurginus piercei Crawford Cotorapo: 1 <, Boulder, about 5300 ft. alt., on the plains, August 7-12, 1919. Panurginus nebrascensis Crawford Cotorapo: 2 <, Boulder, about 5300 ft. alt., on the plains, August 7-12, 1919; 1 &, Denver, August 28, 1919, collected by Barbara M. and Marjorie D. Schwarz. Crawford (1912, Canad. Ent., XLIV, p. 368) states that the tibiz are completely annulate with black. They are not so in the Boulder specimens, nor in a cotype received from Crawford. Panurginus atricornis (Cresson) Cotorapo: 2 3,4 9, Estes Park, two females collected by Herbert F. Schwarz, August 13-14, 1919, the remainder by A. E. Butler, July 19, 1916, at about 7800 ft. alt. The males agree with P. atricornis from Beulah, New Mexico, and Viereck recognized this as atricornis after seeing Cresson’s type. Cres- _ son’s description agrees, except that he says the face-marks are white, 6 AMERICAN MUSEUM NOVITATES [No. 36 perhaps in consequence of using artificial light. The females differ from P. portere by having the highly polished mesothorax almost impunctate. Many years ago Mr. Fox sent me a supposed 2 P. ornatipes, from Colorado, out of the Cresson series. It is not ornatipes, but atricornis. Panurginus renimaculatus (Cockerell) Wyomina: 1 9, Sheridan, collected by C. W. Metz. Panurginus didirupa Cockerell Cotorapo: 1 o, Estes Park, August 17, 1919, collected by Herbert F. Schwarz; 2 #, 2 9, Ward, August 8-10, 1919, the females collected by Miss Sara Branham; 1 &, Elbert Creek, near Electric Lake, La Plata Co., about 10,000 ft. alt., June 30, 1919. Ipano: 1 9, near Montpelier, about 6100 ft. alt., July 6, 1920. The female is smaller than P. portere, with the sides of face shin- ing and finely punctured, and the wings strongly brownish. Panurginus bakeri (Cockerell) ARIZONA: 1 9, southwest end of Coyote Mts., about 3500 ft. alt., among mes- quite, palo verde, etc., August 4-7, 1916; 1 9, Bear Wallow, Sta Catalina Mts., about 8100 ft. alt., near Soldier’s Camp, among Douglass firs, ete., at (?) Pseudocy- mopterus montanus, July 13, 1916. Conoravo: 1 9, Ouray, about 10,000 ft. alt. on Summit Road, July 13, 1919; 1 9, in a meadow at Warren Lake, near Aspen, about 10,800 ft. alt., July 26, 1919; 1 <, Electra Lake, near Durango, about 8400 ft. alt., at Potentilla filipes, June 29, 1910; 3 9, 1 &@, Leadville, about 10,200 ft. alt., in the town, August 4, 1919, collected by Bailey, Schwarz, and Lutz; 1 #,1 9, | Ward, August 9, 1919, collected by Sara Branham; 3 oc’, 5 9 (one female with two female stylopids, Crawfordia), Tennessee Pass, about 10,300 ft. alt., August 1-8. For a description of the female, see Cockerell, 1910, Psyche, XVII, p. 245. | Panurginus cressoniellus Cockerell _ Cotoravo: 2 9, Ward, about 9300 ft. alt. near town, August 9, 1919; 2 9, Pagosa Springs, one along west bank of river below town, about 7200 ft. alt., and one — in U.S. Forest, about 7400 ft. alt., June 21—24, 1919. The female flagellum is mainly ferruginous beneath. © Panurginus cressoniellus variety calochorti Cockerell Cotorapo: 39 9, 17 &, Tennessee Pass, about 10,300 ft. alt., July 30-August 8,! various collectors; 5 9, Ouray, about 10,000 ft. alt., on the Summit Road, July 13, 1919, collected by Messrs. Schwarz and Lutz; 1 ¢ (first recurrent vein consider- ably basad of the first transversocubital), Malta, about 10,000 ft. alt., August 4, — 1919; 1 o, Elbert Creek, La Plata County, about 10,200 ft. alt., June 30, 1919; 5 9, Estes Park, about 7700 ft. alt., July 19, 1916, collected by A. E. Butler. ‘Females abounded at the beginning of August, males appeared about a week later. 1922] BEES OF THE GENUS PANURGINUS 7 The female flagellum is dark. The Ward specimens of P. cressoniel- lus cited above come from almost exactly the type locality of variety calochorti. The calochorti form is the prevalent one in Colorado, especially at high altitudes. About half the Tennessee Pass specimens have the first recurrent nervure more or less basad of the first transversocubital, and so should fall with the form named verus. The original type of P. verus Cockerell distinctly differs in some other respects, especially in having the upper apical corner of marginal cell rounded, the cell not sharply truncate as in all these Colorado insects. It is quite certain that the bees here re- ferred to calochorti all belong together, but as among this large series I cannot match the type of P. verus, it remains at least possible that the latter represents a different species, not including the specimens later referred to it. See also Entomological News, XXIII (1912), p. 445, where it was assumed that the later discovered specimens were properly referred to verus. Panurginus lutze, new species Cotoravo: 1 9, La Junta, along the road between irrigated fields, near to wn about 4100 ft. alt., August 12, 1920, collected by Mrs. F. E. Lutz. Q.—Length about 6.4 mm. Black, without light markings, excepting pale yellow dots on anterior and middle knees. Thorax above with abundant, stiff, erect pale brown pubescence, not hiding the surface; vertex with similar pale brownish hair. Head ordinary, shining, facial quadrangle broader than long; tufts of whitish hair at extreme sides of clypeus below; mandibles dark, reddened subapically, simple, somewhat elbowed externally, the apical part of the right one densely rugoso- punctate above; labial palpi rather short, the second joint not much longer than the third; process of labrum very broad, subtruncate, the margin gently rounded; clypeus and adjacent parts highly polished, with sparse punctures; clypeus with a large but shallow circular median depression; facial fover with the upper end diverging from the orbits; flagellum dusky reddish beneath except at base. Meso- thorax and scutellum shining, without evident punctures under a lens, though the microscope shows that the numerous hairs arise from small punctures; metathorax shining, the basal area reduced to a narrow transverse rugulose channel; tegule reddish-testaceous. Wings hyaline, very faintly brownish; stigma large, solid red- dish brown, nervures pale brown; first recurrent joining second submarginal cell at a distance from its base much greater than half length of first transversocubital. Legs ordinary, with pale hair, middle basitarsi comparatively short and broad. Abdomen shining, the apex with white hair. On account of characters of sculpture and wings, it does not seem possible that this is the female of P. expallidus Swenk and Cockerell or P. horizontalis Swenk and Cockerell, described from males collected in Nebraska. In many respects, it resembles P. pierce Crawford, but the latter is considerably larger, with much lighter and redder stigma. I 8 AMERICAN MUSEUM NOVITATES [No. 36 ' take it that P. lutze is a comparatively recent segregate from the prercet type. adapted to a different flower. Compared with P. innuptus Cockerell, the area of metathorax and other characters are quite distinctive. Panurginus pulchricornis, new species Cotorapo: 1 9°, Tennessee Pass, about 10,500 ft. alt., August 6-8, 1920. Q.—Length about 7 mm. Black, similar to P. atricornis Cresson, but the flagellum is ferruginous beneath except at base and apex, and the stigma is piceous, appearing black by reflected light. The process of labrum is broad, truncate, with the. margin gently convex; in atricgrnis it is much narrower. The clypeus is very coarsely punctured though shining, without any smoother space in middle. The mesothorax and tegule are as in atricornis, but the thorax is considerably smaller. The area of metathorax is rather larger, entirely rugulose and opaque. The wings are con- spicuously dusky. The apical plate of the abdomen is long, with straight sides. Panurginus ineptus, new species Cotorapo: 1 9, Tennessee Pass, about 10,500 ft. alt., August 6-8, 1920. Q9.—Length hardly 6 mm. Black, the head and thorax with scanty but long white hair; eyes dark gray; facial quadrangle somewhat broader than long, orbits converging a little below; mandibles reddened about middle; process of labrum broad and truncate; palpi and tongue short. Third antennal joint unusually short, not much longer than broad, and not as long as the next two together, though they are very short; flagellum dusky red beneath, including the last joint. Clypeus highly polished, with scattered punctures; supraclypeal area shining; vertex smooth, with very feeble punctures. Mesothorax shining, with very weak punctures, median groove strong; area of metathorax dull and granular, the heavy posterior rim feebly shining; mesopleura dullish above, more shining below; tegule piceous with a large pallid discal area. Wings brownish hyaline, stigma and nervures dilute reddish brown; stigma robust; marginal cell broadly, obliquely truncate, the angles not rounded; first submarginal cell much longer than second; first recurrent nervure going only just beyond the first transversocubital. Legs black, with pale hair, that on anterior basitarsi thick and pale straw-yellow. Abdomen highly polished, shining throughout, with only scattered very minute piliferous punctures; apex with shining white hair. An isolated species. Hypomacrotera callops Cockerell and Porter CoLtorapo: 2 9, Regnier, Baca County, about 4400 ft. alt., in Gallinas Canyon, at Quincula lobata, June 8, 1919. Hypomacrotera callops persimilis Cockerell ARIZONA: 2 9, 2 oc’, San Xavier Mission, near Tucson, on flood-plains of Santa Cruz, at Physalis angulata, July 24, 1916. - The eyes are dark brown in both sexes; in typical persimilis from Phoenix, Arizona, they are blue-green in both sexes. The supraclypeal mark of the male is well-developed. 1922] BEES OF THE GENUS PANURGINUS 9 Greeleyella beardsleyi Cockerell Co.torapo: 2 o&, 11 9, Regnier, Baca County, about 4400 ft. alt., at Quincula lobata, Sphexralcea coccinea, and Monarda pectinata, June 6-9, 1919. According to my experience, this species is oligotropic on Mal- vastrum coccineum (Spheralcea coccinea of Rydberg). The following key will facilitate the separation of the species recorded above. IS a esac kote REN CO Reo ek EI ec aba Sig eek AH vibcb ig 98 SEES chk DS oD 1 ST ARS BRS Aecee SS GS ee a SO ae para ees at ge Pe gee ia ere eer Men Thane 13 _ gaght eolor of face confined to clypeus.:... 2. 2. Oe ee ee cee ees 2. Light color of face not confined to clypeuS............... 0.6. cece e eee eee 4, 2. Tibie red and yellow; first recurrent nervure meeting first intercubitus. 12. 13. G. beardsleyi Cockerell. Tibiz black, or at most anterior ones pale in front.. ..................20.. 3. First recurrent nervure going far beyond first intercubitus; mesothorax dullish. P. bakeri Cockerell. First recurrent meeting first intercubitus, or falling basad of it; mesothorax OT SI ae ge 2 ae CIR ES rea ame ey beara P. cressoniellus Cockerell. Supraclypeal mark wholly absent; labrum black, flagellum dark............ 5. Supraclypeal mark at least represented by a spot..................-.0004. 6. Larger; clypeus with two black spots; upper corner of lateral face-marks forming G@ Iareer BNMIO.. 66 5s is ic eas P. portere Cockerell. Smaller; clypeus without black spots; upper corner of lateral face-marks form- ing a smaller angle.............. P. porter Cockerell, variety (Arizona). pnieror wings with conspicuous apical cloud; head very broad. H. callops persimilis Cockerell. momen wines without SUCK. COU... bis ide as cee Sse bale des eee 8 a: OR a SEE a AS BOR eee on Oe ete Meme Sap Te aE CELI rea 8. SO OT SME SMS) 5 17 WTO fe Ne NPE UP ar ae PE aoe 9. PA0G Curome-yOuOwW «666 a be Sai fe OR ae ee P. irregularis Cockerell. Face pale primrose-yellow..................... P. nebrascensis Crawford. Flagellum clear reddish orange beneath, except at base and apex. P. perlevis (Cockerell). Miawenuin OST Or: GUBICY DEROALI 665 abe ois sons oe on ce hol een eens + 10. Lateral face-marks forming practically right angles on orbits above. P. piercei Crawford. Lateral face-marks ending above in acute angles.....................05.-. 2h: PRINS TONG os 22 ie, peasg as on BAG OE Sh sew & Kocieve ly esas eas Bath ps 12. Anterior tibiz light in front; eyes green................ P. didirupa Cockerell. pnt SOT CAPA TOO sth ae ccs ete eee es P. atricornis (Cresson). Small joints of tarsi more slender, notred.............. P. altissimus Cockerell. (Females) A light patch in middle of face; tarsi ferruginous. P. renimaculatus Cockerell. 10 14. 15. 21. 23. ' Clypeus and supraclypeal area highly polished; stigma dilute brown. AMERICAN MUSEUM NOVITATES No light patch on ‘faces ou. 6 22215 25S ee eS a es Anterior tibie pale yellowish red in front; flagellum red beneath. G. beardsleyt Anterior tibize dark in front Mandibles bright red in middle; mesothorax highly polished; stigma > Flagellum ‘red. beneath «05 0). ik. 516 Gece ee ae P. cressoniellus Cockerell =e Flagellum entirely dark..............-0.. P. cressoniellus calochorti Cockerell. Mesothorax dull or cullieh, not highly polished ; smallish BpCoees as eee Tegule wifotestanpeen. ic oe ee ae Tegulss piceotin.. io 5. 5565 oeckss sacs es ee atns ene a Clypeus and supraclypeal area punctured, dullish; stigma very dark. P. pernitens Cocker Mesothorax with conspicuous fulvous hair; tegule rufotestaceous........ 22 Mesothorax without such hair ae Larger; mesothorax with conspicuous punctures......... P. porter Cock Smaller; mesothorax without conspicuous punctures....P. lutze Cockerell. hs Flagellum black; mesothorax highly polished, without eonspinlaee punctures te stigma dark brown; wings strongly dusky........ P. atricornis (Cresson) % Flagellum at least largely pale beneath, or if somewhat dusky, oe evidently punctured : 24. Stigma reddish brown, large and broad; flagellum light yellowish-fer co except base and apex; base of metathorax distinctly pligad P. perlexvis (Cocke Stigma more slender, dark or reddish; flagellum dusky beneath, or if p: Stigma piceous; first recurrent nervure pile far beyond first intereubit t Le flagellum red beneath except at ends........... P. pulchricornis Cocker Stigma rather dilute brown. :....05 2 6600 10. Soke oe es First recurrent going only just beyond first intercubitus; flagellum pale beneat Sa rather ‘small’ spécies.e. sks vie ps ok a ens eee P. ineptus Cone First recurrent going considerably beyond first intercubitus.............. +27 Mesothorax hardly at all punctured; sides of face shining, feebly pungts P. altissimus Cocker Mesothorax evidently punctured; sides of face distinctly punctured. a didirupa Cock = 4 Saks 5 > 1 LS ale 7 4 ¥ sn Cake EN a ty pie es loves ci. hE SS 0d StIMy. alee ne ene fa i} Hs 2 ie Maa ee TR eet duor—ty x PLEASE DO NOT REMOVE CARDS OR SLIPS FROM THIS POCKET UNIVERSITY OF TORONTO LIBRARY STORAGE OS rae PRS rT =. a 4 em my “a cptyeee & * fete - reat et ee mane ER Ah, > Myte re teeth BM, oe he SO a del py Ob pS Toh ryt purines sing ur ipl gO OE POONe we TN Sed ang IEA Oat ye mht nN mae ae patvnea ent Oreos ers