Mth Haile LUTE xa Lt OILL ELE LLLA ESF estate ttle Ld Et tid dd ht M TE tate MAME ae fee en pewrrs x NS e SALLE Woe anne ase ¥ wi a2 7 ‘ & SS % \ SAN are Ss wa N Ay SAN N SILILEELAISESSL CDA E SAX \ SEP IPLIS IIIS IAPS, SLLLLELIIIID LLILPAP SS AGL PLI LPS PILE PELL DDD IPL PEED L IDI LIPID \ { \ NES No \* - @ A . . dy i dads zs “it Zo iH ; . i) H ie ay uy! Witiua Dt i J Me: \ 1 View mir " ai i‘ Ne We Rites i re Me Che Ce ie) shi) i itieiar ay ins Oi ae A ih 1 a ni Ait ieee ‘ ay / hese | vf ‘ Nie ry mY y iy ii + i A ‘yt ha \ i ( i" Ohh | Laie 4} ret if / i r \ 7h ey eet i ; i " Viet rr ey Aga a A ( i ¥ i) A Oy: SMITHSONIAN INSTITUTION UNITED STATES NATIONAL MUSEUM PROCEEDINGS OF THE UNITED STATES NATIONAL MUSEUM VOLUME 73 UNITED STATES GOVERNMENT PRINTING OFFICE WASHINGTON : 1929 ADVERTISEMENT The scientific publications of the National Museum include two series, known, respectively, as Proceedings and Bulletin. The Proceedings, begun in 1878, is intended primarily as a medium for the publication of original papers, based on the collections of the National Museum, that set forth newly acquired facts in biology, anthropology, and geology, with descriptions of new forms and revisions of limited groups. Copies of each paper, in pamphlet form, are distributed as published to libraries and scientific organiza- tions and to specialists and others interested in the different subjects. The dates at which these separate papers are published are recorded in the table of contents of each of the volumes. The present volume is the seventy-third of this series. The Bulletin, the first of which was issued in 1875, consists of a series of separate publications comprising monographs of large zoological groups and other general systematic treatises (occasionally in several volumes), faunal works, reports of expeditions, catalogues of type-specimens, special collections, and other material of similar nature. The majority of the volumes are octavo in size, but a quarto size has been adopted in a few instances in which large plates were regarded as indispensable. In the Bulletin series appear volumes under the heading Contributions from the United States National Herbarium, in octavo form, published by the Nationa! Museum since 1902, which contain papers relating to the botanical collections of the Museum. ALEXANDER WETMORE, Assistant Secretary, Smithsonian Institution. WasuineTon, D. C., July 3, 1929. II TABLE OF CONTENTS AupricH, J. M. A revision of the American parasitic flies belonging to the genus Belvosia. No. 2729, pp. 1-45. ianeAe see ew see Ma ee ey oe Salen: A eee OF Se New species: Belvosia mannt, B. australis, B. semiflava, B. mexi- cana, B. wiedemanni, B. omissa, B. ciliata, B. vittata, B. fron- talis, B. elusa, B. borealis, B. argentifrons, B. townsendi, B. nigrifrons, B. lata, B. smithi, B. spinicoxa, B. williamsi, B. canalis. New varieties: Belvosia recticornis var. ruficornis, B. ciliata var. formosa. BaltLtey, JOHN WENDELL. A revision of the lizards of the genus Ctenosaura. No. 2733, pp. 1-55. September 26, New species: Ctenosaura parkeri, C. clarki. Berry, Epwarp W. Fossil nutlets of the genus Lithosper- imum Nowe 1st pps los. iy, a NOS to. ek ee ee New varieties: Lithospermum fossilium rugosum, L. f. glabrum, L. f. aristatum. Tertiary fossil plants from the Argentine Republic. Ne ia ppa On. October) 17, 1928100 os a a kes New species: Adiantum patagonicum, Pteris nirithuaoensis, Zama australis, Fitzroya tertiaria, Rollinia (?) patagonica, Hydrangea (?) wncerta, Leguminosites calliandraformis, Ana- cardites (?) patagonicus, Sterculia washburnii, Laurelia amaril- lana, Laurophyllum chalianum, Apocynophyllum chalianum, Bignonites chalianus, Phyllites nirihuaoensis, P. mollinedia- formis. Boscuma, H. Two common species of parasitic crustacea (Sacculinidae) of the West Indies. No. 2726, pp. 1-10. May 1oisioggsrimim) iota dole ja rev wWigule J New genus: Lozxothylacus. New species: Drepanorchis occidentalis. Cooxr, C. Wytur. New Vicksburg (Oligocene) mollusks from Mexico. No. 2731, pp. 1-11. April 24, 19281..._____ New genus: Protonema. New species: Gemmula alazana, G. mexa, Pseudotoma alazana, Scobinella prionota, Glyptotoma rhombica, Borsonia agutlae, Ancilla (Ancillina) alazana, Protonema bartschi, Turritella ceibana, Natica alazana, Polynices (Lunatia) lacrimans, P. (Euspira) byramensis, Ampullina vaughani, Dentalium ovale, D. (Dentalium) alazanum, Amussium alazanum, Pecten ceibanus. New variety: Gemmula mexa var. mecxita. 1 Date of publication. 52702—29 Til Article 12 13 22 10 IV TABLE OF CONTENTS Ewine, H. E. The scorpions of the western part of the United States, with notes on those occurring in northern Mexico. No, 2730, pp. 1-24. May. 24 1928 2 22. ee New species: Vaejovis yosemitensis, Centruroides sculpturatus. New variety: Hadrurus hirsutus, var. arizonensis. GILMoRE, CHARLES W. A new fossil reptile from the Triassic of New Jersey. No. 2728, pp. 1-8. March 30, 1928?'..___- New genus: Hypsognathus. New species: Hypsognathus fennert. A new pterosaurian reptile from the marine Cre- taceous of Oregon. No. 2745, pp. 1-5. June 25, 19287____ New species: Pteranodon (?) oregonensis. Gonyer, Forest A. (see D. F. Hewett and Earl V. Shannon)_ Hay, Outver P. Further consideration of the shell of Chelys and of the constitution of the armor of turtles in general. No. 2022 pppoe. WWhareb Zl VOR! 2 a pe Hewett, D. F., Eart V. SHANNON, and Forrst A. GonyER. Zeolites from Ritter Hot Spring, Grant County, Oregon. No. 273/, Ppeaise Sunerlap 1928 See Seg ew ee Hoven, Watrer. Fire-making apparatus in the United States National Museum. No. 2735, pp. 1-72. June 22, Howe, A. Brazier. Contribution to the comparative anatomy of the eared and earless seals (genera Zalophus and Phoca). No. 2736, pp. 1-142. January 26, 19291____ Krircer, Herspert W. A prehistoric pit house village site on the Columbia River at Wahluke, Grant County, Wash- ineton. No, 2732; pp: 1-29. May i7e 1a2s 0s See oe Linton, Epwin. Notes on trematode parasites of birds. No. 2782, pp l—-360Rearch PA 1926) aes eee pee ee New genus: Minuthorchis. New species: Haematotrephus fodiens, Psilostomum lineatum, P. plicitum, P. varium, Petasiger nitidus, Himasthla incisa, A por- chis rugosus, Ascocotyle plana, Minuthorchis sanguineus. LoveripGE, ArtHur. Field notes on vertebrates collected by the Smithsonian-Chrysler East African Expedition of 1926. No: 2738, pp?d-69. June 20<4926.y 20 alent -ulenee Page 24 16 16 14 15 Ly bg 1 Date of publication. TABLE OF CONTENTS Matuocy, J. R. Notes on American two-winged flies of the family Sapromyzidae. No. 2744, pp. 1-18, June 23, TOS te See ee ee eae ey ona ages an New genera: Freyia, Pseudocalliope. New species: Deutominettia approximata, D. bimaculata, Asilos- toma palpalis, A. pallipes, A. atriceps, A. flavifacies, Freyia nigrita, Minettia nigropunctata, M. argentiventris, M. infuscata, M. tucumanensis, M. quadrata, M. brunneicosta, M. verticalis, Sapromyza schwar zi. MERRILL, GreorcE P. Concerning the origin of the metal in meteorites. No. 2742, pp. 1-7, June 20, 1928'____________ Price, Emmett W. New helminth parasites from Central American mammals. No. 2725, pp.1-7. March 30, 1928!_ New genera: Bradypostrongylus, Controrchis. New species: Bradypostrongylus panamensis, Graphidium browni, Controrchis biliophilus. RatuBun, MaryJ. Twonew crabs from the Eocene of Texas. No 2727ppisGoO AprikS, BOQ cr wyo0085. cx9inace “we'd New species: Notosceles bournei, Harpactocarcinus americanus. Scuaus, Wituram. New moths of the family Ceruridae (Notodontidae) in the United States National Museum. No. 2740.cnps 1-90: “dune 27 1988 ~ 25 SF estas sas eer eS New genera: Dugonia, Rhapigia. New species: Nystalea biumbrata, N. eastmant, N. julitha, N. par- sont, N. amatura, Elymiotis corana, E. morana, E. lupicina, E. donatian, E. boisil, Proelymiotis severina, Lysana minasensis, Marthula thoreda, M. cynrica, Eudmoe carrieta, Lepasta branda, Dasylophia blaizea, Farigia liboria, F. luicana, F. thelian, F. alicia, F. sennen, Cerura purusa, Peroara caterina, Psilacron con- galla, P. gordiana, Urgedra oslaca, U. nabora, Dicentria fechima, Schizura salvador, Litodonta centigerna, Notoplusia marchiana, N. talmecana, Misogada brioca, Trumanda schiffi, Disphragis clittusa, D. epimacha, D. marusa, D. arima, D. sabaria, D. hygi- nia, D. druona, D. carantis, D. agapa, D. psalmoida, Hemipec- teros teffeina, Malocampa medommoca, M. mammerta, M. ran- dauta, M. bronacha, Rhuda decepta, Gisara meyeri, G. brewsteri, G. brauni, G. metcalfi, Boriza ignatia, Talmeca dabuisa, T. aga- thosa, Chadisra paragorna, C. finiana, C. ulrica, C. ezrana, C. celsa, C. emeteria, Meragisa salvina, M. vistara, M. simeona, M euthymia, M. mochosema, M. polycarpa, Dugonia eliera, Euxoga amatura, Rifargia haitia, R. auscharia, R. possida, R. everiti, Afilia venadia, A. purulha, Lusura turnina, Lobeza petropolia, L. gilberta, L. huacamaya, L. maronia, L. venica, L. rhenia, L. medina, L. arnoula, L. genebrarda, L. abdjesa, L. gunthierna, Eunotela chacoa, Apela archimma, Dottia boliviata, Pamcaloma abba, Hemiceras ursara, H. liboria, H. turiafa, H. climaca, H. joinvillia, H. chromona, H. teffea, H. taperinha, H. reyburni, H. Page 23 21 19 1 Date of publication. vI TABLE OF CONTENTS hidulpha, H. jovita, H. phocas, H. praxides, H. eustalhia, H. elphega, H. monegonda, H. arbogasta, H. turnina, H. vinvala, H. noctifer, Schausiades almothes, Hapigia duponti, H. hollandia, H. smerinthina, H. apiana, H. eneana, H. beuvea, H. millsi, Rha- pigia deicola, Hapigiodes argentidiscata, Spatalia bronacha, Noto- donta grahami, Cerura nicetia, Somera acasia, Stauropus briachi- sta, S. niteria, S. palladina, Fentonia gualberta, F. erconvalda, F. abraama, F. mangholda, F. cantiana, F. eingana, F. maguila, Chadisra madena, Neopheogia cathana, Phalera ordgara, P. suri- gaona, Besida vinalva, Liparopsis dympna, Pydna adjutrea, P. marconia, P. odrana, P. ubalvia, P. barasamphia, P. ercona, P. goddrica, Turnaca bryantia, T. pantaena, T. suriga, Norraca ordgara, Microphalera styxana, Pygaera hildora. New forms: Symmerista sigea, Rifargia demissa brioca, Rhapigia deicola agnesa. ScHwartz, BensAmMIn. Two new nematodes of the family Strongylidea, parasitic in the intestines of mammals. No. 2725, pps. March: 21 31008) fi. Ae eee Pens eee New genera: Phacochoerostrongylus, Oesophagostomoides. New species: Phacochoerostrongylus pricet, Oesophagostomoides gilinert. SrIWELL, H. R. Two new species of commensal copepods from the Woods Hole region. No. 2739, pp.1-5. June 4, NOG cr eee aa a a a! New species: Amphiascus commensalis, Tisbe wilsoni. SHANNON, Ear V. (see D. F. Hewett and Forest A. Gonyer)- Simpson, CHarLes Torrey. The Florida tree snails of the genus Liguus. No. 2741, pp. 1-44. May 11, 19291__ _ _- 1 Date of publication. Page 18 16 20 LIST OF ILLUSTRATIONS PLATES NOTES ON TREMATODE PARASITES OF BIRDS By Edwin Linton Facing page 1 Monostome: trematode otetheloone ae: Seeees oe ee 36 2. Distomes of herring gull, laughing gull, loon, and grebe__-_________~- 36 3. Distomes:of-grebe and herring-gulle eae OL Bee SIC 36 Any Distomes, Olawhite-wineeduscoter =" = +s. — =e kee Se eee 36 5. Distomes of white-winged scoter and herring gull______-___-----_-- 36 6. Distomes of herring gull, red-necked grebe, loon, and Arctic tern_-_-_- 36 7. Distomes of Arctic tern, green heron, sanderling, and frigate bird-- -- 36 See Distomevorlaughine cule sas as ae ee ee aes ee ee eee 36 Oe Distomes of sur duck andsherring oul = 4" ee see ee ee 36 10. Distomes of herring gull, laughing gull, and ring-billed gull________~- 36 11. Distomes of laughing gull, ring-billed gull, and herring gull________~- 36 Two NEW NEMATODES OF THE FAMILY STRONGYLIDAE, PARASITIC IN THE INTESTINES OF MAMMALS By Benjamin Schwartz 1. Phacochoerostrongylus pricei, new species. ...-..---.----------+---- 6 2. Oesophagostomoides giltnert, new species. —.-....--..~..------------- 6 FURTHER CONSIDERATION OF THE SHELL OF CHELYS AND OF THE CONSTITUTION OF THE ARMOR OF TURTLES IN GENERAL By Oliver P. Hay 16 Carapace of .Chenys fpmoriata__.. saw stead pa icee alt Sa alee ob. 12 2. Carapaces of Chelys and Clemmys insculpta__....------------------ 12 NEW HELMINTH PARASITES FROM CENTRAL AMERICAN MAMMALS By Emmett W. Price 1. Bradypostrongylus panamensis, new species___..._...-.--------------- 8 2. Graphidium browni, new species and Controrchis bibliophilus, new SY OSX Lots ag ns Bh eos ne seth, Sp les atte yA el beh ch = Bh 8 Two NEW CRABS FROM THE EOCENE OF TEXAS By Mary J. Rathbun 1. Notosceles bournei from the Eocene of Texas.___._.________-_-------- 6 2-3. Harpactocarcinus americanus from the Eocene of Texas______------ 6 vVul LIST OF ILLUSTRATIONS A NEW FOSSIL REPTILE FROM THE TRIASSIC OF NEW JERSEY By Charles W. Gilmore Facing page 1~3.. Hynsognathus fenpertg mew SPeCCleS sn sees SU oe a ee THE SCORPIONS OF THE WESTERN PART OF THE UNITED STATEs, WITH NOTES ON THOSE OCCURRING IN NORTHERN MEXIcOo By H. E. Ewing 1-2. Scorpions.of western“United’ States S522 2 0S as eee eee New VIcKsBURG (OLIGOCENE) MOLLUSKS FROM MeExico By C. Wythe Cooke j=2.. Vicksburg'mollusks from. Mexico... "ssc 2 Gece 424 ete ee ee A PREHISTORIC PIT HOUSE VILLAGE SITE ON THE COLUMBIA RIVER AT WAHLUKE, GRANT County, WASH. By Herbert W. Krieger » and pestles Oleiome.: 22 52 sa beta eel ee i ee a Ny pes Of atrowesud spear beams. ice ee ee EWES GhAbTOWREROS yo 5. pe gee gl a Hammerstones;and scaling knives2.2= ou. as) 23 Lae ee eee ee Objectssofgpersonal adormments sf) ese eta Sees ees eee Decorated objects of stone, bone, horn, and wood__-__-_-_------------ White bluffs escarpment at Wahluke, Wash., and the Columbia River at: Wahbluke* Wash= 2 S272 e ioe Wee arnae ere 8 2 Se STS oc Rp hoe A REVISION OF THE LIZARDS OF THE GENUS CTENOSAURA By John Wendell Bailey . Head and body of female of Ctenosaura acanthura____-_------------ . Sacral region and tail of female of Ctenosaura acanthura___--------- Half-grown male of Cienosauraj acanthuras.d_........-.---=-=------ ANG hhE SETENIS Cole (CHOON R TOR GRU LUO = eB eee se epeeere son asee Adult male of Ctenosaura hemilopha___------------ a indy hs capa Stuffed skin of female of Ctenosaura brachylopha___---------------- . Head and body of adult male of Ctenosaura pectinata___----------~-- Aditi nate or CrenosiiunG DECLETULG == en nee ee Ane meee aD ANohible cols Chi OUAVCR NU Os NAR ON ee oe ee a ee LOS Adultiftemale of Clenosaunamectinatws =.= aes) oa ee ee 11. Femoral pores of adult male of Ctenosaura pectinata____------------ 12: Adult male.of Clenesaunaibrevinasinis 5] se ee ee jon Half-crown male or Ctenosaund OnevinOsuits= = ee [4zeAcdultifemslerotaCtenosaunar parce na eee ee ee 15. Heads of Ctenosaura parkeri (left) and C. brevirostris (right)_-------- te, AgGult male af Clenosaina stmilise = oo eee er ee ee eos 17. Half-grown male and young male of Ctenosaura similis_______-_---- 1SawACulbitemalevotClenosaunasstnilise as sees one oe 195 Adult male of:Ctenosaurarstmilts. = PEER: BES PE Se eee 20. Typical habitats of ‘Clenosdura*simiiist 22 SY Nee be eee ai; Adultifemale of Clenosaura bakert— ... 2... 2.5222... ..-.1. ee CONAN RON SE 24 12 LIST OF ILLUSTRATIONS Ix Facing page 22. The dewlap of (a) Ctenosaura bakeri, female; (6) C. palearis, male; (c) C. malearis fenicgle yea). ceed wept See OTe kN eee 23. Adults of Ctenosaura palearis (left) female; (right) male____________ 24. Sacral region of adult male of Ctenosaura quinquecarinata__________- 25. Sacral region of adult male of Ctenosaura quinquecarinata_______-.-- 26. Adult\male of Cienosaiura.quinquecarinata..__ 222222. 2.2--._ = 21. Adultrmalaiel Crenssaura’ clarke 22022 Peo 8) Sates ee 25. Adu male of Clenosauravergikromelas. 2% 2 20.;Adult male of ‘Crencsaura erythromelass+ 2 =... 222. 2.------ a0. Adult male of Cienosaura dejenson22.-..- 28 ee FossIL NUTLETS OF THE GENUS LITHOSPERMUM By Edward W. Berry i... Fossil nutlets of the. genus: ditthospermum=se! uth o. 2 dsc 42- assess FIRE-MAKING APPARATUS IN THE UNITED States NaTIONAL MusEuM By Walter Hough 5) PSKORUREL OVE eo A bitin far aC eg pli ke «ps al ec i ae i ig ae po aS . British Guiana, West Indian, and Mexican drills___--__---_-------- Japanese sacred fire drill, full view and section__...__._.___.__-----.- Bhils; Indialtand Australian: drlis #272 LE eas 2 ees ee ok indutsacred fire: drall\Geplica)ss= 522 ae ee a eee DEBS YSG rang We asa meso Ua D FSi 5 a I ape ots he ee gh Meee A LS pee A African firewalls =e 2e Sue ee 2 SPE SS La eee eines te eee Bambpoosirersaweekhilippines se = ae — ee aaa = ee ety es ee Ee Bamboo strike-a-lights and Battak (Negrito) fire thong__-_---__---- Bamboorstrike-d-lightwivialay slau = oe. Seay Ay ee oe ee Bem OISUOUS welVicitel vista: | metas eee ne ene re ey eet ee = pop Lee oe Fear nee eugene aren eo ho: = pd pk CONTRIBUTION TO THE COMPARATIVE ANATOMY OF THE EARED AND EARLESS SEALS (GENERA ZALOPHUS AND PHOCA) By A. Brazier Howell ih Mounted skeletons of an eared seal (Otariidae, below) and an earless Baie seni (Phocidactabove))- 2. cee 2 tees ces) ee ee ZEOLITES FROM RitteER Hor Sprine, GRANT County, OREGON By D. F. Hewett, Earl V. Shannon, and Forest A. Gonyer 1. a, Chabazite; b, Mesolite and Pseudomesolite; and c, Thomsonite-- --- 2. a, Pseudomesolite and Analcite; and b, Stilbite-___.---------------- FIELD NOTES ON VERTEBRATES COLLECTED BY THE SMITHSONIAN CurysLeR East AFRICAN EXPEDITION OF 1926 By Arthur Loveridge 1.£ Typical W agogo' kraaliin Dodoma’ district. Wagogo cattle sheltering under mimosa trees. Indian shops in Dodoma township--------- 2. Combined § leopard trap and cage. Building stockade at Tulo. A corner, of the birdsroom 206 a8 222202256 <= yeh ee oa taate Shas 72 72 72 72 72 72 72 72 72 72 72 142 18 18 70 70 X LIST OF ILLUSTRATIONS Facing page (vs) . Nest and eggs of finch lark. Catching a spitting cobra. Termite workings where a tree frog was found__-__----------------__-.- 4. Zanzibar galago, the first animal obtained by Expedition. White- bearded gnu after six months in Washington. Soft-shelled tortoise from: Dodommaedy sey sper inet. yeorey numwyetnl LY pak ts pee Taye a eee Two NEW SPECIES OF COMMENSAL COPEPODS FROM THE Woops HOLE REGION By H. R. Seiwell L Amphiascusicommensaiis, new species £02 Lees ell See ee « Lisbenwilsont Me wa SMeCCies ays pe enfants ne MS i) THE FLORIDA TREE SNAILS OF THE GENUS LiGuUS By Charles Torrey Simpson 1-4. Florida tree snails of the genus Iiguus._......---_--=22-2---+--- CONCERNING THE ORIGIN OF THE METAL IN METEORITES By George P. Merrill 1-3. Origin of metal in meteorites__.___________ pds Gwinn ey ene arte 2) TERTIARY FOSSIL PLANTS FROM THE ARGENTINE REPUBLIC By Edward W. Berry 1-5. Argentine Tertiary, plants= 282 82 eo Eee ee eee TEXT FIGURES Two COMMON SPECIES OF PARASITIC CRUSTACEA (SACCU- LINIDAE) OF THE WEsT INDIES By H. Boschma 1. Two specimens of Drepanorchis occidentalis. a, From Mithrazx forceps (A. Milne-Edwards), the surface lying against the thorax of the host, X 3%. 6, The same specimen, the surface lying against the abdomen of the host, * 3%. c, From Macrocoeloma camptocerum (Stimpson), the surface lying against the thorax of the host, < 3. d, The same specimen, the surface lying against the abdomen of the host, X 3. In these figures the mantle opening is found in the upper part; the'stalk inthellower part 22> _ teal 2 fk ee 2. Drepanorchis occidentalis from Mithrax forceps (A. Milne-Edwards), VOT PAU TEE aE SECOLOU: OC a) remem ee sete waren ge oe 3. Drepanorchis occidentalis. a, Part of the external cuticle of a specimen from Pitho anisodon (von Martens), X 440. 06, Part of the external cuticle of a specimen from Mithrar forceps (A. Milne-Edwards), X 440. c, Retinaculum of a specimen from Macrocoeloma campto- cerum (Stimpson), < 440. d, Internal cuticle with retinacula of a specimen from Microphrys bicornutus (Latreille), * 110. e, Retinaculum of a specimen from Mithrazx sculptus Lamarck, X 440. f, Retinaculum of a specimen from Pitho anisodon (von Martens), xX 440. gandh, Retinacula of a specimen from Microphrys bicornu- tus) (Latreille). ><: 44020) se soe cose se SINC LO NS SE 70 70 44 28 Page LIST OF ILLUSTRATIONS XI Page 4. Two specimens of Loxothylacus panopaei (Gissler). a, From Panopeus herbstit Milne-Edwards, the surface lying against the thorax of the host, X 7%. 6b, The same specimen, the surface lying against the abdomen of the host, * 7%. c, From Hurypanopeus depressus (Smith), the surface lying against the thorax of the host, X 7%. d, The same specimen, the surface lying against the abdomen of the host, X 7%. In these figures the mantle opening is found in the upper part, the stalk in the lower part___--..__________---- 8 5. Loxothylacus panopaet (Gissler). From EHurypanopeus depressus (Smith), longitudinal section, ><. SO 273-2 a hey ce eet A atiows a 6. Loxothylacus panopaei (Gissler). a, Part of the external cuticle of a specimen from EHurypanopeus depressus (Smith), X 440. 6, Part of the external cuticle of a specimen from Panopeus herbstiz Milne- Edwards, X 440. cand d, Appendages from two different places on the external cuticle of a specimen from Panopeus herbstiz Milne- Edwards, X 440. e, Cuticular appendages of another specimen from Panopeus herbstit Milne-Edwards, * 440. jf, Cuticular append- ages of a specimen from Panopeus occidentalis (Saussure), X 440. g, Retinaculum of a specimen from Eurypanopeus depressus (Smith), xX 440. handz, Retinacula of two different specimens from Pano- peus herbstiz,Milne-Edwards, < Lalalinos: hea tae a rea 21. Fire-making set and extra hearth. Frobisher Bay. 20. Moss ina leathertr tase. .J00OU7) 2GYS VQ. Bed DINE ee Sh SPOT: eee Boring seti "Cumberland Gulf 1 2 toie 2 he Bo too iareke wilt te Fire-making set. Angmagsalik Eskimo, eastern Greenland _____-_---_ Boring set. Angmagsalik Eskimo, eastern Greenland______--__-__- Fire bag. Eskimo of Holsteinberg, west Greenland______________-_ Lower part of fire-making set (on one end is gum for cement). Mac- kenzie Rivery British Columibiat! ors k 200s iw Peas tO oma. Lower part of fire-making set. Eskimo of Mackenzie River, British Colunibial AAswiiags Poe abs OVER ba AE OMTIOEE a Pup ay: A ee Fire-making set. Eskimo of Anderson River, British Columbia_-_-_- Fire-making set (with mouthpiece of deer’s knucklebone, thong, and tinder of willow catkin). Eskimo of Point Barrow, Alaska______- Fire-making set (hearth showing median groove). Eskimo of Norton RSOWL TYG FeRw CS: MRM Eas 9 he ae. 2d logan ce eR EN Eee NTI oR Lo Lower piece of fire-making set (hearth). Eskimo of Cape Vancouver, Alaska tae co pot serio essed gale 28 bateydysyuegaes OE pera ret teay Fire-making set. Eskimo of Chalitmute, Kuskokwim Region, Alaska_ Fire-making set. Eskimo of Kassianamute, Togiak Region, Alaska_ Fire-making set (hearth with step and five slots). Koggiung, Bristol Bay Aladkaice Lemieews UL hep ee Be een SO ee Fire-making set (hearth with central holes and end step). Koggiung, Bristol, Bay; Alaskac oo 2. 220 2) i ee Bae peemiae epee Fire-making set. Eskimo of Bristol Bay, Alaska__.._.._.....------ Lower piece and spindle of fire-making set. Eskimo of Kodiak Island, Malay fire sticks. Models in bamboo made by Doctor Hough after A. R. Wallace’s description. The Malay Archipelago_-_--------- Hiresmoaking isticks:\) OaINGR > 54a eee eS os Se a, Strike-a-light. Seven Barrows, Berks County, England. b, Strike- a-light. Indians of Fort Simpson, Mackenzie River district, British Columbia a one ee a ne he pak ke i 1. Tinder pocket. 2. Fire bag. Mackenzie River district, British (Caan ee are ate ot OnE. Eee 3. Pyrites. 4. 4a. Flint striker and handle. Mackenzie River district, British: Columblaa2< 329s Be WE oh ee ee . Methodof using ithelstrike=a=li ohn tine eee a ee . English tinder box (with flint, flourish, and bundle of spunks). » Wiheelttinder!box-"Broadaliomn,sINney 22m 2 ees eee eee ee ee . Strike-a-light (briquet). Boulogne-sur-Mer, France___._.._-------- . Flint and steel. Otoe Indians, Kansas and Nebraska __.__-_------- . Strike-a-light (flint, steel, tinder horn, spunk, and pouch). Cheyenne Hiri ieamgy PAT Wari ee ee a ere cata oe ra 49 50 51 52 . LIST OF ILLUSTRATIONS Strike-a-light (pouch for holding flint and steel). Comanche Indians, Flint and steel. Guadalajara Indians, Mexico_------------------- Strke-a-lipht) [@hing @e2 2 te ee oes eheh e tet Smokers’ pipe-lighting outfit (showing flint, steel, pipe pick, and pincers). Koords of Bhotan, Eastern Turkey ------------------ 53-54. 53. Rush fire set pouch. 54. Strike-a-light. Flint, steel, and 55. 56. tinder box) Ainos Of, YezOx, Japan ==. 22 eco 2 See ee eee Tinder box (showing mounted steel, flint, and bundle of shaving matches; box one-third natural size). Japan__._.-------------- Smokers) strike-a-light., “Wokio: Japans =) == ee a ee eee eee CoNTRIBUTION TO THE COMPARATIVE ANATOMY OF THE EARED AND EARLESS SEALS (GENERA ZALOPHUS AND PHOCA) By A. Brazier Howell . Typical terrestrial postures of an eared seal (sea lion or otariid, Zalophus) and earless seal (true seal or Procid, Phoca, above) ---- . Dorsal view of the skull or Zalophus showing areas of muscle attach- ments labeled in capital letters; names of bones in small type---- . Dorsal view of the skull of Phoca hispida, showing areas of muscle attachments and (in small type) names of bones____------------ . Ventral view of the skull of Zalophus: names of bones in small type-- . Ventral view of the skull of Phoca hispida: names of bones in small . Lateral view of the left mandible of Zalophus (Z) and of Phoca hispida (2) showing areas of muscle attachments_7"- 2922-2 --- 2-2-2 . Lateral view of the left scapula of Zalophus (Z) and Phoca hispida (P) with areas of muscle attachments________~_ UY, ah ee Nn el Oe . Left view of the anterior limb bones of Zalophus (Z) and Phoca hispida (P) in approximate positions in which they are usually carried in . Left humerus of Zalophus (Z) and Phoca hispida (P); in lateral view Aabovevandemedialypel Owes ee eee ee ee . Left radius and ulna of Zalophus (Z) and Phoca hispida (P); in lateral viewsabovelmndamedial belowlst 22LE 21 Seaus Nene Be = ese . Left innominate bone of Zalophus (Z) and Phoca hispida (P)-_------- . Left femur of Zalophus (4) and Phoca hispida (P); anterior aspect aboverandeposterion.belows-=-2==S-=45--225-25-2<=--- 2 == 0 . Left tibia and fibula of Zalophus (Z) and Phoca hispida (P); in lateral View above ancsmedlaltnelowee see te ote ee eee . Dorsal view of left tarsus and metatarsus of Zalophus (Z) and Phoca GS ULC (2) a tere ane tenet ay emp ee ey eee a) ein ee a . Left aspect of Zalophus (Z) and Phoco hispida (P), showing platysma- panniculus carnosus sheet of musculature_______.-_-__-_--.----- . Dorsal musculature of Zalophus: superficial layer upon the left and much of the next deeper layer to the right of the medial line__-__- . Dorsal musculature of Phoca hispida: superficial layer upon the left and much of the next deeper layer to the right of the medial line_- . Ventral musculature of Zalophus: superficial layer upon the right and much of the next deeper layer to the left of the medial line--__---_ . Ventral musculature of Phoco hispida: superficial layer upon the right and much of the next deeper layer to the left of the medial line-___~- 70 71 XIV LIST OF ILLUSTRATIONS 20. Superficial musculature of the lateral aspect of the left anterior limb Of Zalophus os 2 f'22 4k: 52 Soe See ee ee ae Sa eee 21. Superficial musculature of the lateral aspect of the left anterior limb ofvPhocethispidass2 sees See ek See ee ee ee 22. Superficial musculature of the medial aspect of the left anterior limb OL Zalophuse kes Sse Le eR. eR eR as Sh La 2 ee 23. Superficial musculature of the medial aspect of the left anterior limb Ole hOCahispidGgmeese os S22 = ee aus Cue ee ek ee 24. Ventral aspect of the muscles extending from the innominate bone (stippled) to the posterior limb of Zalophus (Z) and Phoca his- DG) Bee re ee eS ee es > i ae eye ee ee 25. Cranio-lateral aspect of the superficial muscles of the left posterior LimibofiZalophast 22 eet Re RET AAR ee oe ee eee 26. Cranio-lateral aspect of the superficial muscles of the left posterior ISU 001 OF Gy al EP OLY hel pi Ry ov a Ke aap hae ale Me neta TE EE UE as eee 27. Caudal aspect of the musculature of the left posterior limb of Zalophus_ 28. Caudal aspect of the musculature of the left posterior limb of Phoca Pius Toh Dyes 5 cos hh EN See af nd pa aca I 1 ep aha Bey 29. Position assumed by Mirounga illustrating possible degree of vertebral 30. Diagram illustrating approximate static posture (solid lines) and degree of possible movement (broken lines) in life of each jointed segment of the posterior limbs of Zalophus (Z) and Phoca hispida (P). I, innominate; T, thigh; S, shank; H, heel (Astragalus and Caleaneum) sand yh, memainderof foots 2252) 32 eee o ne ee ee ZEOLITES FROM RirTER Hot Sprine, GRant County, OREGON By D. F. Hewett, Earl V, Shannon, and Forest A. Gonyer 1. Sketch map of Ritter Hot Spring area, Grant County, Oreg_-__----- Tue FLORIDA TREE SNAILS OF THE GENUS LIGUUS By Charles Torrey Simpson 1. Diagram illustrating the distribution and migration of Liguus in Florida. The open spaces are pine woods; those inclosed in lines, hammocks. Dotted lines show migration to and from hammocks. Some bits of forest are reached by several; others are entirely TM SSCCH Nae ARS 2 See hue a eos Se aye ae kee eee ee ele TERTIARY FOSSIL PLANTS FROM THE ARGENTINE REPUBLIC By Edward W. Berry 1. Localities in Lago Nahuel Huapi region, Rio Negro territory; 2. Locali- ty of Mirhoja, Chubut territory; 3. Localities in Rio Chalia refion, Santa Cruz temitory ss 2202 Se Pare ee a _ int Te oe ty dal ART, 1 TREMATODE PARASITES OF BIRDS—-LINTON 29 PROALARIA INDISTINCTA (Guberlet) Figures 64, 65 1922. Hemistomum confusum GUBERLET, Journ. Parasit., vol. 9, pp. 11-12, figs. 4-9, : 1923. Alaria indistincta (Guberlet), GuBERLET, Trans. Amer. Mic. Soc., vol. 41, p. 68. 1926. Proalaria indistincta (Guberlet), LARvgE, Trans. Amer. Mic. Soc., vol. 45, p. 15. Larus delawarensis, intestine. Woods Hole, Mass.: Zarus argentatus, L. atricilla. The following description is based on the specimen from L. atricilla. Dimensions, life: Length 1.30 mm.; anterior portion, length 0.60, breadth, 0.60; posterior portion, length 0.70, diameter 0.40. ‘Dimensions in balsam, compressed, and margins of anterior por- tion folded: Length 1.22 mm.; anterior portion, length 0.56, breadth 0.36; posterior portion, length 0.66, breadth 0.38; pharynx, length 0.045, breadth 0.036; oral sucker, length 0.054, breadth 0.051; ventral sucker, length 0.058, breadth 0.075; diameter of adhesive organ 0.15; ovum 0.092 by 0.061. Anterior portion of body broader and slightly shorter than pos- terior portion; oral sucker and pharynx each longer than broad, ventral sucker broader than long; pharynx adjacent to oral sucker; accessory adhesive organ close to posterior border of ventral sucker. Two conspicuous organs, one on either side of the oral sucker, 0.045 by 0.072 mm. in the two principal diameters, haye a rasplike appear- ance under high magnification. Each is crossed by about 12 trans- verse ridges. Upon focussing up and down, these structures appear to be the roughened edges of a series of plates set on edge and close together. The rami of the intestine originate very close to the pharynx and extend to the posterior end of the body. The two testes lie on the left side, the anterior one near the middle and the posterior one at about the posterior third of the posterior division of the body. They are separated by a fold of the uterus. The ovary lies on the left side at the anterior border of the first testis. The vitellaria extend through- out the entire length of the posterior division of the body, and as far forward as the posterior border of the ventral sucker, and obscure the other structures to a great degree. The ova are relatively large, not numerous, and mainly on the right side, from the posterior edge of the accessory adhesive organ to the posterior end. A few lie between the testes. The dense vitelline glands make it difficult to see details of the anatomy. The genital pore is dorsal, about 0.06 mm. from the posterior end. 30 PROCEEDINGS OF THE NATIONAL MUSEUM VOL. 73 RECORD OF COLLECTIONS Larus argentatus. 1927, August 13. 2, lengths 1.40 and 1.70 mm. L. atricilla. 1913, August 15. 1, length 1.30 mm. (Cat. No. 7949, U.S.N.M.) ALARIA species Figures 66, 67 A single, somewhat damaged specimen from the intestine of a ring- billed gull (Larus delawarensis), January 24, 1914, is here noted. The oral sucker and pharynx are missing, and the specimen is broken at the level of the anterior border of the accessory adhesive organ. Dimensions, balsam: Length 2.17 mm.; anterior division, length 1.40, breadth 0.42; posterior division, length 0.77, breadth 0.42; diameter of ventral sucker 0.06; adhesive organ, length 0.17, breadth 0.10. The bursa is everted, length 0.14, breadth 0.22; cirrus exserted, length 0.14, diameter 0.05. Ova have not yet made their appearance. The anatomy is not satis- factorily shown in the mounted specimen. Vitellaria extend from about the middle of the posterior division forward to the ventral sucker, and mask the anterior testis and ovary. (Cat. No. 7950, U.S.N.M.) This distome suggests Hemistomuwm gavium (Guberlet) .° STRIGEA BURSIGERA (Brandes) Figures 68-72 1890. Holostomum bursigerum G. BranpeEs, Zool. Jahrb., vol. 5, p. 592, figs. 15-18. 1909. Strigea bursigera (Brandes) LUNE, Parasitische Plattwurmer. I. Trema- todes, Brauer’s Suswasserfauna Deuchlands, Heft 17, p. 163. Larus ridibundus. Woods Hole, Mass., Larus argentatus, L. atricilla, L. delawarensis, intestine. Length from 6 to 9 mm.; anterior division of the body more or less pyriform, posterior division subcylindrical, enlarging slightly to the posterior third, where the greatest diameter is attained in the vicinity of the testes. The anterior end is usually reflected dorsally. There are some variations from the above proportions, but they can usually be accounted for by different conditions of contraction. Oral sucker, pharynx, and ventral sucker all small, the ventral a little larger than the oral. The two testes are lobed and situated behind the middle 6 Journ. Parasit., vol. 9, pp. 9-11, figs. 10-13. ART 1 TREMATODE PARASITES OF BIRDS—LINTON 3l of the posterior division of the body, one following the other, but separated by a space in which lie the transverse yolk reservoir, the shell gland and oétype, and the beginning of the uterus. The ovary is situated close to the anterior border of the first testis. As seen in lateral view it is long-oval, the transverse diameter greater than the longitudinal, tapering at the median end, thus becoming some- what pyriform. In a specimen from the laughing gull, mounted in balsam, the longer diameter is 0.28, shorter diameter 0.15. The first testis in this specimen had a maximum breadth of 0.45 and length of 0.24; second testis, breadth 0.59, length 0.36; distance of second testis from posterior end 1.40, or approximately one-fourth the entire length. In whole mounts of specimens from the ring- billed gull the ovary had the same dimensions as those given above; first testis, length 0.56, breadth 0.70; second testis, length 0.49, breadth 0.70; distance of second testis from posterior end -approxi- mately one-fifth the entire length in each. Behind the second testis there is a capacious seminal vesicle and a contorted, thick-walled ejaculatory duct. ‘The copulatory bursa was invaginated in all cases. It is fairly well developed, as shown in Figure 71, which is some- what diagrammatic, and occupies about one-half the distance between the second testis and the posterior end. The vitellaria are dis- tributed mainly on the ventral side from near the constriction be- tween the two divisions of the body to near the posterior end. In the specimens from the ring-billed gull and herring gull they were strongly developed, so much so as to mask much of the anatomy. The yolk reservoir is tubular. It originates on the ventral side and lies along the anterior border of the second testis. There is much variation in the number of ova. For example, in the specimen from the laughing gull there are relatively few ova, about 29, while two from the ring-billed gull contain the one 133 and the other 220, and one from the herring gull contains about 375. In one of the three series of sections the ova are few, in the others there are many. There is not much difference in the size of ova in the different ex- amples. In the balsam mounts they do not vary much from 0.11 by 0.07 in the two principal diameters. The course followed by the uterus is in all cases from its origin between the testes forward towards the anterior end of the posterior division of the body, whence it returns on the ventral side to the posterior end. There is no seminal receptacle, but the early folds of the uterus were seen, in some of the series of sections, to be filled with sperm. In one series of sections the germ duct throughout its somewhat tortuous course, from near the ovary to the shell gland, as well as the earlier folds of the uterus, was filled with sperm. Laurer’s canal opens on the dorsal surface about on a level with the anterior border of the first testis. It is a small, somewhat con- 32 PROCEEDINGS OF THE NATIONAL MUSEUM VOL, 73 torted duct, and proceeds antero-ventrad to the postero-dorsal border of the ovary, where it turns caudad, and appears to join the germ duct. The germ duct lies along the dorsal side of the first testis. It enters the shell gland, where, near the middle of the in- tertesticular space, it becomes the rather thick-walled odtype. The short duct from the yolk reservoir joins it just before it expands into the odtype. The shell gland is relatively large and lies on the left side of the intertesticular space. Beyond the odtype the uterus is at first rather thick-walled and contracted. It then enlarges, becomes thin-walled, and lies in a number of folds between the testes on the right side. These folds were, in most cases, filled with sperm. The uterus passes along the ventral side of the first testis and continues forward to a point near the constriction between the two divisions of the body, where it turns abruptly, and, returning to the posterior end of the body, opens into the ejaculatory duct near the genital pore. RECORD OF COLLECTIONS Larus argentatus. 1915, September 1. 1; length 6.5 mm.; diameter, anterior division of body 1.35, posterior 1; anterior portion orange color, posterior yellowish. (Cat. No. 7951, U.S.N.M.) 1916, January 8. 3; length of longest about 6 mm.; diameter 0.75; anterior yellowish, posterior dark bluish- grey. Dimensions, balsam: Length 9.38 mm.; diameter, anterior 1.36, at level of testes 1.40; ova, longer diameter from 0.096 to 0.108, shorter diameter from 0.063 to 0.075, average 0.10 by 0.07. Measurements made from series of sagittal sections: Oral sucker 0.09; pharynx, length 0.075, breadth 0.09; ait sucker, length 0.135, breadth 0.084; ova, average, 0.111 by 0.063. Larus atricilla. 1911, July 19. 1; white, anterior portion tinged with orange. Dimensions, balsam, lateral view: Length 6.16 mm.; diameter, anterior, 0.65, at level of testes 0.77; oral sucker, length 0.090, breadth 0.075 ; pharynx, length 0.070, breadth 0.075; ventral sucker, length 0.112, breadth 0.089; ova, average of 6, 0.115 by 0.072. (Cat. No. 7952, U.S.N.M.) ART. 1 TREMATODE PARASITES OF BIRDS—-LINTON 33 Larus delawarensis. 1914, January 24. 3; lengths 8, 9, and 11 mm., in formalin. Di- mensions, balsam: (1) Length 6.88; di- ameter, anterior 0.91, at level of testes 1.09; (2) Length 7.8; diameter, anterior 0.91, at level of festes 0.84; largest ova in each about 0.12 by 0.07. Measurements of sagittal sec- tions: Diameter of oral sucker 0.081, pharynx 0.066; ventral sucker, length 0.114, breadth 0.075; ova, average, 0.114 by 0.072. (Cat. No. 7953, U.S.N.M.) EXPLANATION OF PLATES a. ventral sucker. Oo. ovaly. bo. bursa. oe. esophagus. c. cirrus. ph. pharynx. cp. cirrus-pouch. sg. shell gland. ej. ejaculatory duct. sr. seminal receptacle. ex, exeretory vessel. sv. seminal vesicle. g. genital pore. t. testis. gd, germ duct. aw. uterus. h. holdfast organ. vd. vas deferens. 7, intestine. vg. vitelline gland. 1. Laurer’s canal. yd. yolk duct. m. metraterm. yr. yolk reservoir. Unless otherwise stated, sketches were made with the aid of the camera lucida from balsam mounts. PLATE 1 Haematotrephus fodiens, new species, from Gavia immer Fie. 1. Ventral view of free individual; length 11 mm. The dark-brown, opaque material, which filled the intestine, is not represented in the sketch. . Anterior end of same, enlarged; length of pharynx 0.10 mm. Worm from a pedicelled cyst on the serous coat of the pancreas. . Anterior end of same; length of pharynx 0.06 mm. . Portion of body of same in vicinity of genital pore; diameter 0.14 mm. . Posterior end of rami of intestine of same. > oP © PD PLATE 2 Psilostomum plicitum, new species, from Larus argentatus Fic. 7. Ventral view; length 1.61 mm. Psilostomum lineatum, new species, from Larus argentatus 8. Ventral view; length 3.78 mm. Psilostomum varium, new species, from Gavia immer 9. Ventral view; length 1.54 mm. Petasiger nitidus, new species, from Colymbus auritus 10. Ventral view; length 1.75 mm. 70841—28——3 34 Fic. Fic. Fic. Lt. 12. 13. 14. 15. 16. 17. 18. 19. 20. 36. 37. PROCEEDINGS OF THE NATIONAL MUSEUM VOL, 73 PLATE 3 Petasiger nitidus, new species (continued) . Ventral view of head, in glycerine; diameter, including spines, 0.49 mm. The recurved dorsal spines in this specimen seem to be exceptional. Ventral view of another specimen ; diameter, including spines, 0.42 mm. Dorsal view of same. Transverse section at level of ventral sucker; diameter 0.38 mm. pr. prostate gland. Transverse section at level of ovary, diagrammatic. Transverse section near posterior end; diameter 0.20 mm. Hinasthla, elongata (Mehlis), from Larus argentatus Ventral view; length 6.30 mm. Ventral view of head in glycerine; diameter, including spines, 0.36 mm. Postero-lateral spines, more highly magnified ; length of spine 0.054 mm. Female genitalia; diagrammatic, from transverse sections. PLATE 4 Himasthila incisa, new species, from Oidemia deglandi . Ventral view; length 9 mm. . Anterior end of same; diameter at level of ventral sucker 0.6 mm. . Oral spines, foreshortened, sketched from a transverse section; Spencer 6x/4 mm. . Oral spines; longer spines 0.051 by 0.018, shorter 0.039 by 0.012 mm. . Transverse section near anterior border of ventral sucker; breadth 0.43 mm. The section in front of this contained the genital pore. . Transverse section at level of ovary; breadth 0.74 mm. . Transverse section immediately behind ovary; partly diagrammatic, about four consecutive sections used; breadth 0.8 mm. PLATE 5 Himasthla incisa, new species (continued) . Transverse section at level of seminal vesicle; breadth 0.67 mm. . Transverse section near posterior end; breadth 0.88 mm. . Margin of body at anterior end of vitellaria; average length of serra- tions about 0.02 mm. . Margin of body at level of ovary; length of serrations about 0.04 mm. . Transverse section of body wall at level of seminal vesicle; Spencer 6x/4 mm.; cg. cuticular gland; Im. longitudinal muscle. . Transverse section of body wall at level of ovary; Spencer 6x/4 mm. Mesorchis pseudoechinatus (Olsson), from Larus argentatus . Ventral view of body, lateral of head and neck; length 5.82 mm. . Dorsal view of slightly macerated specimen, from which all spines had disappeared; length 4.48 mm. Ventral view of body, dorsal of head and neck; length 3.5 mm. Ventral view of head, in glycerine; diameter, including spines 0.42 mm ART. 1 Fie. 38 39. 40. 41, 42, 43. 44, TREMATODE PARASITES OF BIRDS—LINTON 35 PLATE 6 Mesorchis pseudoechinatus (Olsson) (continued) . Dorsal view of head; diameter 0.82 mm., from Larus argentatus. Dorsal view of median region of body; breadth 0.6 mm. Sagittal section, median region; diameter of ventral sucker 0.22 mm. Ventral view of specimen from Colymbus holbolli; length 2.56 mm. Ventral view of specimen from Gavia immer; length 2.20 mm. Aporchis rugosus, new species, from Sterna paradisea Ventral view of anterior portion of specimen; diameter at ventral sucker 0.35 mm. Posterior portion of same; diameter 0.77 mm. Uterus somewhat diagrammatic in this and the preceding figure. . Ventral view of anterior end; diameter at ventral sucker 0.35 mm. spines diagrammatic. PLATE 7 Aporchis rugosus, new species (continued) Wie. 46. Camera lucida sketch of surface of body back of cirrus-pouch; 47. 48. 49. breadth 0.28 mm. Lateral margin at level of seminal vesicle; length of spines 0.03 mm. Lateral margin at level of testes; average length of serrations about 0.07 mm. Ovum; sketched from an ovum lying isolated from others in the metraterm; dimensions, exclusive of filament, 0.08 by 0.03 mm. Ascocotyle plana, new species, from Butorides virescens . Ventral view; length 0.67 mm. Levinseniella adunca (Linton), from Orocethia alba 51. Memorandum sketch made at time of collecting; tests and vitellaria Ga 52 added from stained specimen; length 0.8 mm. lactosomum cochieariforme (Rudolphi), from Fregata magnificens . Ventral view, uterus diagrammatic. length 2.38 mm. PLATE 8 Minuthorchis sanguineus, new genus, and new species, from Larus artricilla Fic. 53 54 55 56 . Ventral view, uterus somewhat diagrammatic; length 8.8 mm. . Anterior end of same; transverse diameter of oral sucker 0.46 mm. . Ventral view of ovary, etc, partly diagrammatic: ovary 0.56 by 0.44 mm. . Ova; longer diameter 0.09 mm. 36 Fic. 57. 60. 61. 64. 65. 66. 67. Fig. 68. 69. 70. aA 72. PROCEEDINGS OF THE NATIONAL MUSEUM VOL, 73: PLATE 9 Distomum species, from Oidemia perspicillata. (See A, p. 25.) Ventral view; length 0.30 mm. Distomum species, from Oidemia perspicillata. (See B, p. 26.) . Ventral view; length 0.24 mm. Ornithobilharzia species, from Larus argentatus . Anterior end of pair, neck of male contracted; diameter of male at ventral sucker 0.6 mm. Anterior end of female, dorsal view; diameter of oral sucker 0.05 mm. Fragment of male with female; maximum diameter of male 0.52 mm. . Dorsal view of male showing intestine; length 3.9 mm. PLATE 10 Ornithobilharzia species (continued ) . Lateral view of pair; length of male 3 mm. Proalaria indistincta (Guberlet), from Larus atricilla Sketch of specimen in alcohol, slightly compressed ; length 1.383 mm. Ventral view of mounted specimen ; length 1.22 mm. b. Lateral organ. Alaria species, from Larus delawarensis Fragment, anterior, ventral view; length 1.05 mm. Fragment, posterior; length 1.05 mm. PLATE 11 Strigea bursigera (Brandes) Lateral view of specimen from Larus atricilla; length 6.3 mm. Ventral view of anterior end of specimen from Larus delawarensis; breadth 1 mm. Sagittal section of anterior end of specimen from Larus argentatus; diameter 0.56 mm. . Sagittal section, posterior end, somewhat diagrammatic; diameter 0.56 mm. Posterior end of mounted specimen; diameter 1.383 mm. O U.S. NATIONAL MUSEUM PROCEEDINGS, VOL. 73, ART. 1 PL. 1 MONOSTOME TREMATODE OF THE LOON FOR EXPLANATION OF PLATE SEE PAGE 33 70841—28—4 U.S. NATIONAL MUSEUM PROCEEDINGS, VOL. 73, ART. 1 PL. 2 QO ‘y} Ls, 0) DISTOMES OF HERRING GULL. LAUGHING GULL, LOON, AND GREBE FoR EXPLANATION OF PLATE SEE PAGE 33 U.S. NATIONAL MUSEUM PROCEEDINGS, VOL. 73, ART. 1 PL. 3 DISTOMES OF GREBE AND HERRING GULL FOR EXPLANATION OF PLATE SEE PAGE 34 U. S. NATIONAL MUSEUM PROCEEDINGS, VOL. 73, ART. 1 PL. 4 eo Gor ‘i Gris bx if Nel De S > £E35: @ Cont 2 aes Raat XS oe 2S ee eS ase REPT oy 3 Rl tg adet DISTOMES OF WHITE-WINGED SCOTER FOR EXPLANATION OF PLATE SEE PAGE 34 U.S. NATIONAL MUSEUM PROCEEDINGS, VOL. 73, ART. 1 PL. 5 20 oe e “2. ie > a DISTOMES OF WHITE-WINGED SCOTER AND HERRING GULL FOR EXPLANATION OF PLATE SEE PAGE 34 U.S. NATIONAL MUSEUM PROCEEDINGS, VOL. 73, ART. 1 PL. 6 38 DISTOMES OF HERRING GULL, RED-NECKED GREBE, LOON, AND ARCTIC TERN FOR EXPLANATION OF PLATE SEE PAGE 35 U. S. NATIONAL MUSEUM PROCEEDINGS, VOL. 73, ART. 1 PL. 7 DISTOMES OF ARCTIC TERN, GREEN HERON, SANDERLING, AND FRIGATE BIRD FoR EXPLANATION OF PLATE SEE PAGE 3h PROCEEDINGS, VOL. 73, ART. 1 PL. 8 U.S. NATIONAL MUSEUM DISTOME OF LAUGHING GULL FOR EXPLANATION OF PLATE SEE PAGE 36 U. S. NATIONAL MUSEUM PROCEEDINGS, VOL. 73, ART. 1 PL. 9 a me aie DIsTOMES OF SURF DUCK AND HERRING GULL FOR EXPLANATION OF PLATE SEE PAGE 36 U. S. NATIONAL MUSEUM PROCEEDINGS, VOL. 73, ART. 1 PL. 10 DISTOMES OF HERRING GULL, LAUGHING GULL, AND RING- BILLED GULL FOR EXPLANATION OF PLATE SEE PAGE 36 U. S. NATIONAL MUSEUM PROCEEDINGS, VOL. 73, ART. 1 PL. 11 DISTOMES OF LAUGHING GULL, RING-BILLED GULL, AND HERRING GULL FOR EXPLANATION OF PLATE SEE PAGE 36 ates ath TWO NEW NEMATODES OF THE FAMILY STRONGYL- IDAE, PARASITIC IN THE INTESTINES OF MAMMALS By BensamMin ScHwartz Zoologist, Bureau of Animal Industry The first parasite described in this paper was collected by Dr. E. W. Price, of the Bureau of Animal Industry, in the course of a post-mortem examination of a wart hog (Phacochoerus aethiopicus massaieus) which was received at the National Zoological Park, Washington, D. C., on October 26, 1926, and which died June 6, 1927. The worms, which represent a new genus and species, were found in the large intestine in association with three species of the nematode genus Oesophagostomum as follows: O. mwanzae, O. cury- cephalum, and O. yorkei. PHACOCHOEROSTRONGYLUS, new genus SrroneyLmar.—The mouth is directed straight forward. There are four submedian and two lateral cephalic papillae. An external and an internal leaf crown are present. The buccal capsule is rather shallow and cylindrical in shape. The esophagus is club-shaped. Male bursa with two lateral lobes and a well-developed dorsal lobe. Ventro-ventral ‘and latero-ventral rays short, parallel, and close together; externo-lateral, medio-lateral, and postero-lateral rays originate from a common trunk, the last two being parallel and close together, the externo-lateral ray diverging from them. Externo- dorsal and dorsal rays arise from a common trunk. Dorsal ray with a pair of branches anterior to its cleft. Each of the terminal branches of the dorsal ray with one small or rudimentary accessory branch. Spicules slender, filiform, and sheathed. Gubernaculum present. Vulva and anus very close together. The terminal portion of the female bent dorsad. Vagina relatively long, communicating on each side with an ovejector, the ovejectors and uteri being parallel. PHACOCHOEROSTRONGYLUS PRICEI, new species The head (fig. 1) is separated from the rest of the body by a well defined transverse constriction. The buccal capsule is supported by chitinous walls whose appearance in optical section is shown in Figure 1. The external leaf crown consists of 10 elements which: are long and pointed; the internal leaf crown consists of about 24 No. 2723.—PROCEEDINGS. U, S. NATIONAL MUSEUM, VOL. 73, ART. 2. 77067—28 1 a PROCEEDINGS OF THE NATIONAL MUSEUM VOL. 73 small elements. The first part of the esophagus consists of a broad anterior portion and a narrow posterior portion the latter gradually widening out to form the posterior part of the esophagus. The nerve ring is located approximately at the beginning of the second third of the esophagus in the region of the constricted portion of that organ. The cervical papillae are located slightly posterior to the middle of the esophagus. The lining of the walls of the esophageal funnel is modified to form six toothlike structures, triangular in shape with apex directed forwards, Male—The male is from about 8.5 to 10 mm. long by from, 370 to slightly over 400u in maximum width and from about 185 to 200 wide just anterior to the bursa, in the region of the pre-bursal papil- lae. The head is from 117» to 1254 in maximum diameter. The maximum width of the buccal capsule is about 100u. The esophagus is from 756p to 840» long by 202n in maximum width, and 84a im minimum width. The nerve ring is located at a distance of about 270u from the anterior extremity of the esophagus, and. the cervical, papillae are located approximately in the’ middle region, of the esophagus. The ventro-ventral and latero-ventral rays (fig. 6) have a horizontal direction and extend to the margin of the lateral lobes of the bursa. The medio-lateral and postero-lateral rays extend close to the margin of the bursa, whereas the externo-lateral ray which diverges from the former terminates at some distance from the margin of the bursa. The externo-dorsal ray arises at a distance of from 220,» to 250u from the tips of the terminal branches of the dorsal ray. The undivided portion of the dorsal ray has a pair of accessory branches located at a distance of about 50. from the point at which this ray becomes cleft; the accessory branches arise symmetrically (fig. 2) or asymmetrically (fig. 5) from the stem of the dorsal ray which is cleft on its posterior third, the diverging terminal branches being about. 854 long. Hach terminal branch of the dorsal ray has a small accessory branch which may be rudi- mentary (fig. 5) and is located at a distance of about 33. from the tip. The spicules (fig. 5) are slender, alate, from 672 to 840m, jong, and terminate in knoblike tips. The gubernaculum is more or less pear-shaped, from 63~ to 67~ long by about 21y in maximum width. Female-—The female is from 9 to 9.5 mm. long by 470. to 487p wide. The maximum diameter of the head is 134». Diameter of buccal capsule same as in male. The esophagus is from about 873, to 923. long by 2354 to 3194in maximum width. In one specimen the nerve ring is located at a distance of about 2854 from the anterior extremity of the esophagus, being somewhat posterior to the begin- ning of the second third of that organ. In the same specimen the cervical papillae are located at a distance of 436. from the anterior ART. 2 NEW PARASITIC NEMATODES——-SCH WARTZ 3 end of the esophagus which is slightly posterior to the middle of that organ. The distance between the vulva and anus (fig. 7) is 67p. The vagina is about 350u long and each ovejector is about 285. long or somewhat longer. The tail is 84. long and terminates bluntly. The posterior end of the female is turned dorsad and has the appearance of a foot (fig. 4) when the worm is viewed from the side. Host.—Phacochoerus aethiopicus massmeus. Location—Large intestine. Locality.—National Zoological Park, Washington, D. C. Type specimen.No. 27789. U.S.N.M. Paratypes.—No. 27790. U.S.N.M. This species is named after Dr. E. W. Price who collected the nematode specimens from the wart hog and kindly turned them over to the writer for determination. The second lot of specimens, which also represent a new genus and species, were collected by Dr. L. T. Giltner, of the Bureau of Animal Industry, from the intestine of a common wombat (Phascolomys mitchelli) about 10 years ago. The host animal in question was received at the National Zoological Park, Washington, D. C., on December 1, 1916, and died of pneumonia on July 8, 1917. Doctor Giltner made a post-mortem examination of the carcass and dis- covered a number of nematodes in the large intestine which he kindly turned over to the writer for determination. The illustrations accompanying the description of these worms (Plate 2) were made in July, 1917. Unfortunately most of the specimens were allowed to become dry at one time during the intervening perica and it is there- fore difficult to study the material at the preseny time in as great detail as is desirable. However, the salient characters of these nema- todes indicate quite clearly that they represent a new genus and species. OESOPHAGOSTOMOIDES, new genus SrroNeyLipar.—The mouth is directed straight forward. The mouth collar bears four prominent submedian and two lateral papil- lae. (Fig. 2.) The buccal capsule is of moderate depth and cylin- drical in shape. (Fig. 5.) An external leaf crown is present; an internal leaf crown is absent. The esophagus is club-shaped. parallel, well chitinized pieces, each measuring 44» in length. The telamon is composed of two feebly chitinized parts embedded in the wall of the cloaca; the anterior part appears as a relatively wide band showing a deep notch in the antero-dorsal border and a similar notch in the postero-ventral border; the posterior part is composed of a narrow band extending across the dorsal wall of the cloaca and the two ends of this band appear to unite or fuse with a V-shaped structure embedded in the ventral wall. (Fig. 4.) Female 14 mm. long and 220, in diameter. The vulva is a trans- verse slit located about 3.5 mm. from the end of the tail. The tail (fig. 5) terminates in a slender filamentous tip and is also provided with three spike-like processes. The terminal filament is about 23p long and the spines about 16 long. The anus is located about 185y from the tip of the tail. The ovejectors (fig. 6) are strongly muscu- lar and with a combined length, including sphincters, of 585 to 600n. The eggs (fig. 7) are oval, 66 to 69n long and 33 to 40 wide. Host—Three-toed sloth, Bradypus sp. Location.—Small intestine. Locality Central America (Penonomé, Panama). Type specimens.—United States National Museum Helmintho- logical Collections No. 27002. The female of this species closely resembles that of the genus Anoplostrongylus, a genus proposed by Boulenger (1926) for cer- tain trichostrongyles of bats; the male, however, appears to be more closely related to Ornithostrongylus 'Travassos, 1914, and in the key given by Yorke and Maplestone (1926) it would run out at that genus. The dorsal ray, spicules, and gubernaculum appear to be sufficiently different from those of either of the above genera to war- rant the creation of a new genus. GRAPHIDIUM BROWNI, new species Specific diagnosis—Graphidium: Cuticle of the anterior extremity slightly inflated and coarsely striated transversely. (Fig. 8.) The cuticular expansion is about 77» long and 38. in diameter. The body shows numerous fine, wavy striations, and is also finely striated transversely. The oral aperture is surrounded by three small incon- spicuous lips. The esophagus is 650» long in the male and 740 long in the female, slender, slightly enlarged posteriorly, and is 32p wide about the middle and 58, wide at the enlarged posterior por- tion. The nerve ring is situated 237 to 260, from the anterior end. The excretory pore opens ventrally 340 to 390. from the anterior end. Cervical papillae not apparent. Male 8.5 mm. long and with a maximum width of about 130, in front of the bursa. The bursa (fig. 9) is composed of two lateral 4 PROCEEDINGS OF THE NATIONAL MUSEUM VOL, 73 lobes and a smaller inconspicuous dorsal lobe. The rays are well separated and extend to near the edge of the bursa. The ventro- ventral and latero-ventral rays are divergent and about equal in size; the externo-lateral ray is shghtly thicker and longer than the other rays; the medio-lateral and postero-lateral rays are divergent; the externo-dorsal rays arise from the base of the dorsal ray and are curved dorsad near their posterior third; the dorsal ray forms two branches near its tip and each branch is bidigi- tate. Prebursal papillae present. The spicules are equal in length, slender, modified tubular in shape, and 532, long. The tips of the spicules are pointed and incurved, and have a sharp pointed process on the median aspect a short distance from the tip. The shaft of each spicule appears twisted about 156 from its anterior end. The gubernaculum is elongated, curved, well chitinized, and is 128 long. The telamon is composed of two similar, feebly chitinized, retort- shaped structures, embedded in the ventral and lateral walls of the cloaca. (Fig. 10.) ‘The genital cone is small, rounded, and bears two prominent papillae; these papillae are pedunculated and are situ- ated on each side of the cloacal aperture. Female 14 mm. long and with a maximum width of 166y. The vulva is situated about 2.7 mm. from the posterior end of the body. The tail (fig. 11) is slender and pointed. The anus is located about 160» from the end of the tail. The ovejectors (fig. 12) are strongly muscular and have a combined length, including sphincters, of 400,. The eggs are oval, 64n to 70u long by 32n to 384 wide, with shells of uniform thickness, and are not embryonated within the uterus. Host.—Three-toed sloth, Bradypus sp. Location.—Small intestine. Locality —Central America (Penonomé, Panama). Type specimens.—United States National Museum Helmintho- logical Collections No. 27003. This species differs from Graphidium strigosum (Dujardin, 1845), the type of the genus, in the following respects: In G@. strigoswm the spicules, according to Hall (1916), are tubular and measure 1.2 ta 2.4 mm. in length; in G. browni they are modified tubular and their length is only about one-half of the minimum length given for G. strigosum. The gubernaculum in @. brown is long and well chiti- nized; in G. strigoswm it is short and so imperfectly chitinized as to be almost invisible. In G. strigoswm the diameter of the female dimin- ishes abruptly behind the vulva; in G. browni the attenuation is gradual. The cuticular inflation of the cephalic extremity is very distinct and coarsely striated in G. brownd, but this character is not mentioned for G. strigosum. An examination of the specimens of the latter species, donated to the Bureau of Animal Industry by ART. 4 NEW HELMINTH PARASITES—PRICE 5 ‘Professor Railliet, shows that a coarse striation of the anterior end of the body is present but the cuticular inflation is not marked. TREMATODA Family DICROCOELIIDAE CONTRORCHIS, new genus Generic diagnosis.—Dicrocoeliinae: Body oval in outline and with greatest width at the middle of the body. Oral sucker strongly muscular and directed anteriorly. Pharynx well developed; pre- pharynx absent. Esophagus short; intestinal ceca slender and ex- tending to the posterior third of the body. Acetabulum large, situ- ated about one-fourth of the body length from the anterior end. Vitellaria compact, extracecal, and not extending anteriorly beyond the posterior border of the posterior testis. Ovary oval in shape and situated immediately posterior to the posterior testis. Uterus with one ascending and one descending limb, each with numerous trans- verse coils, extending to the posterior end of the body. Testes oval, one lying anterior and the other posterior to the acetabulum. Genital orifice immediately behind the intestinal bifurcation. Excretory pore terminal. Type species.—Controrchis biliophilus, new species. CONTRORCHIS BILIOPHILUS, new species Specific diagnosis.—Controrchis: Length 2.5 to 3 mm.; width 0.85 to 11 mm. In preserved specimens the anterior end is slightly curved ventrally. The anterior third of the body is covered with small scalelike spines. The oral sucker is strongly muscular, 200, to 213» long by 148 to 184, wide, and with the oral aperture terminal. © The pharynx is situated immediately behind the oral sucker and measures 84u to 99 long by 7iu to 844 wide. The esophagus is short, 67» to T1p in length, and bifurcates a short distance in front of the anterior testis to form simple, slender, intestinal ceca which extend to the posterior third of the body. The acetabulum is circular, strongly muscular, 188» to 3554 in diameter, and situated in the median line about 500. from the anterior end of the body. The testes are ovoid, elongated transversely; the anterior testis is situ- ated anterior to the acetabulum and measures 140 to 210 by 280, to 430u; the posterior testis is situated posterior to the acetabulum and measures 140p to 210u by 350u to 430n. The cirrus pouch is pyriform, 142n to 227 long and 65p to 100 wide, and contains a relatively large vesicula seminalis, a small prostate, and a short ejaculatory duct. The genital orifice is situated immediately behind the intestinal 6 PROCEEDINGS OF THE NATIONAL MUSEUM vou, 73 bifurcation. The ovary is oval, 99u to 114 by 127, to 170, and is situated immediately posterior to the posterior testis. The recepta- culum seminis and shell gland are located a short distance behind the ovary. The vitellaria are made up of few compact irregular fol- licles, occupying a space 350 to 400. long, on each side of the body iateral to the intestinal ceca, and not extending anteriorly beyond the level of the posterior edge of the posterior testis. The uterus consists of an ascending and decending branch and of numerous lateral coils, extending posteriorly to the posterior end of the body and anteriorly to the ovary. The eggs are small, oval, brown in color, and are 35» to 38, long and 21, to 24 wide. Host.— Ateles geoffroyi. Location.—Gall bladder. Locality.—National Zoological Park, Washington, D. C. l'ype specimens.—United States National Museum Helmintho- logical Collections No. 27599; paratypes, No. 27369. REFERENCES BouLencer, C. L. 1926. Report on a collection of parasitic nematodes, mainly from Heypt. Part IV. Trichostrongylidae and Strongylinae, Parasitology, Cambridge (Hng.), vol. 18 (1), January 22, pp. 86-100. Hatt, Maurice C. 1916. Nematode parasites of mammals of the orders Rodentia, Lagomor- pha, and Hyracoidea. Proc. U. §. Nat. Mus., vol. 50, pp. 1-258, pl. 1, fig. 290. MILter, Gerrit S. 1924, List of North American recent mammals. U. §. National Museum. Bull. 128, pp. xvi+673. YORKE, WARRINGTON and MAPLESTONE, Pil: 1926. The nematode parasites of vertebrates. With a foreword by C. W. Stiles. xi-+-536 pp. 307 figs. London. ART. 4 NEW HELMINTH PARASITES—PRICE 7 EXPLANATION OF PLATES ABBREVIATIONS ac. acetabulum; c. p. cirrus pouch; d. dorsal ray; e. egg; e. d. externo-dorsal ray; ¢. J. externo-lateral ray; e. p. excretory pore; gb. gubernaculum; ge. genital cone; g. p. genital pore; int. intestine; J. v. latero-ventral ray; m. l. medio-lateral ray; os. oral sucker; ov. ovary; ovj. 1, 2, 3, ovejectors; pgc. papil- lae on the genital cone; ph. pharynx; p. 1. postero-lateral ray; sp. spicules; tf. telamon; ¢. a. anterior testis; ¢. p. posterior testis; wt. uterus; vit. vitellaria; vul. vulva; v. v. ventro-ventral ray. PLATE 1 Fic. 1. Bradypostrongywus panamensis. Anterior end of female. 2. Bradypostrongylus panamensis. Bursa of male; dorsal view. 3. Bradypostrongylus panamensis. Spicules and gubernaculum; ventral view. 2 4. Bradypostrongylus panamensis. Telamon; dorsal view. 5. Bradypostrongylus panamensis. Posterior end of female. 6. Bradypostrongylus panamensis. Ovejectors. 7. Bradypostrongylus panamensis. Egg. PLATE 2 8. Graphidium brown. Anterior end of female. 9. Graphidium browni. Posterior end of male. 10. Graphidium browni. Telamon and genital cone; ventral view. 11. Graphidium browni. Posterior end of female. 12, Graphidium browni. Ovejectors. 13. Controrchis biliophilus. Ventral view. (a Bus as “Ae tte AREY ah outed gi oi! a Pee EK ar Bh) & . impair, Le Tee a Ries 6 HEL i toned : U. S. NATIONAL MUSEUM PROCEEDINGS, VOL. 73, ART. 4 PL. 1 YOM 7) Lol) Z BRADYPOSTRONGYLUS PANAMENSIS, NEW SPECIES FOR EXPLANATION OF PLATE SEE PAGE 7 PROCEEDINGS, VOL. 73, ART. 4 PL. 2 U. S. NATIONAL MUSEUM NEW SPECIES AND CONTRORCHIS BIBLIOPHILUS, NEW GRAPHIDIUM BROWNI, SPECIES FOR EXPLANATION OF PLATE SEE PAGE 7 TWO COMMON SPECIES OF PARASITIC CRUSTACEA (SACCULINIDAE) OF THE WEST INDIES By H. Boscuma Of the University of Leiden, Holland The chief characteristics defining the genera of the Rhizocephala concern the shape and the situation of the internal organs. One of the most important is the manner in which the. visceral mass is attached to the mantle by a mesentery, which may be thin, as in Sacculina and allied genera, or thicker, as in Peltogaster and other parasites of the group. Smith (1906), the author of a monograph on the Rhizocephala, based the taxonomy of the group for a large part on the differences in the size and the place of the mesentery. This author created the new generic name Heterosaccus, the type species of which is Sacculina hians Kossmann, which is characterized by a very short mesentery. In this species it is confined to the imme- diate vicinity of the stalk only. According to Smith the internal structure of the other parasites of the group which live on crabs does not show any morphological differences, and consequently he united them under the name Sacculina carcint. In this respect, however, Smith’s conclusions are wrong, for there are a number of. species which can be distinguished from 4. carcint by constant morphological features. The mesentery of Sacculina carcim is complete; it stretches from the stalk to the mantle opening. In some other parasites, which have been described as species of the genus Sacculina, the mesentery is incomplete, for it terminates at some distance from the mantle opening. For these species of Sacculinidae with incomplete mesen- tery, in a recent paper (Boschma, 1927) I have founded the new genus Drepanorchis, the type species of which is D. neglecta (Saccu- lina neglecta Fraisse). Another constant feature of all the species belonging to this genus is found in the shape of the testes; they are curved, whereas in Sacculina they consist of more or less straight tubes. There are a few species of Sacculinidae which in some respects constitute intermediate forms between the genera Sacculina and No. 2726.—PROCEEDINGS U. S. NATIONAL MUSEUM, VoL. 73, ART. 5 76867—28 1 2 PROCEEDINGS OF THE NATIONAL MUSEUM VOL. 73 Drepanorchis. In these species the mesentery is complete, as in Sac- culina; the testes, however, have a curved shape, as in Drepanorchas. They differ from Sacculina as well as from Drepanorchis in another anatomical respect: The visceral mass is not (as in these two genera) connected with the stalk, but it is attached to the mantle at some dis- tance from the stalk. I unite here the species which show the above- mentioned characteristics under the new generic name Lowothylacus. Accordingly, I consider the family Sacculinidae composed of four genera with the following chief characteristics: Sacculina: Mesentery complete, testes straight, visceral mass united with the stalk. Loxothylacus: Mesentery complete, testes curved, visceral mass united with the mantle at some distance from the stalk. Drepanorchis: Mesentery incomplete, testes curved, visceral mass united with the stalk. Heterosaccus: Mesentery almost wanting (confined to the imme- diate vicinity of the stalk). There are a great number of species belonging to the Sacculinidae, although this number is smaller than Giard (1887, 1888) assumed, who upheld that every species of crab had its own species of rhizo- cephalous parasite. This theory (the theory of the absolute specific- ity of the parasites) has been combated by Smith (1906), who, how- ever, went to the other extreme and expressed as his opinion that all the parasites of the genus Sacculina belong to one and the same valid species, S. carcint Thompson. Smith based this conclusion on the fact that he found no morphological differences between the para- sites on different crabs in the material at his disposal. Now, this conclusion also was too far-reaching, for Giard (1887) had proved already that the parasite called by him Saceulina fraisset (=Drepanorchis neglecta (Fraisse), see Boschma, 1927) differs from 8S. carcint in constant morphological characteristics, the form and situation of the testes in both species being very different. Moreover, Kossmann (1872) had described morphological differences between species of Sacculinidae which were found on different hosts. His diagnoses of new species were based mainly on the structure of the chitinous covering of the mantle, which in many cases bears definitely shaped small excrescences. Kossmann also was convinced that a certain species of Sacculina could live on one species of host only and accordingly described too many forms as new species. The discovery of the appendages of the cuticle, however, was a valuable progress in the study of the Sacculinidae as it furnishes reliable data for the taxonomy of the group. Especially in tropical species (Kossmann’s material had been collected at the Philippine Islands) these small projecting parts of the external cuticle are dis- tinctly visible. In the same way in the collection of the Siboga ART. 5 PARASITIC CRUSTACEA OF WEST INDIES—-BOSCHMA 3 Expedition from the East Indian Archipelago also, the greater part of the Sacculinidae are sufficiently characterized by the peculiar struc- ture of the excrescences on the external cuticle of the mantle (Van Kampen and Boschma, 1925). In the paper cited we were able to prove that a number of morphologically well defined species of Sacculina occur in the East Indies. Two more striking facts were observed, first, that a single species may occur as a parasite of several species of crabs, and, second, that certain species of crabs may be infested with two or more different species of Sacculina. Conse- quently this paper proves conclusively that the opinions of Giard as well as Smith can not be upheld. A comparative study of a great number of European representa: tives of the genus Sacculina (Boschma, 1927) yielded almost the same results: Among the European forms one species of para- site may infest different species of crabs, as in the case of Sacculina carcint. In this region, however, two morphologically different species have never been found as parasites on one species of host. Previously Guérin-Ganivet (1911) published a paper in which notes were given on the anatomy of many European Sacculinids, and this author has already pointed out that a great number of the so-called species of Giard’s have no real systematic value. According to Guérin-Ganivet we may regard a certain form as a definite species only in those cases in which morphological differences from other species can be demonstrated. In the paper cited I based my conclu- sions on the same premises. In some cases a certain parasite infests exclusively crabs belonging to a small systematic group, as Drepanorchis neglecta (Fraisse), which is known to occur on the species of the genera Macropodia and Inachus. Both of these genera belong to the subfamily Inachinae (family Majidae) and the parasite, which is structurally very dif- ferent from all other known European Sacculinids, is the only species of the group which is known to infest these crabs. The parasite of Dorynchus thomsoni, another species of the same subfamily, belongs to quite a different species (Saceulina atlantica), which is not found on any other crab. Another well defined species, Sacculina eriphiae, also seems to occur on one species of crab only, namely, Hriphia spinifrons. On the other hand Sacculina carcini infests a large number of hosts belonging to different families of Brachyura. The West Indian species of Sacculinidae are very imperfectly known. One species has been described as Sacculina panopaei by Gissler (1884) after its host Panopeus herbstii (Milne-Edwards). The external form of this parasite is known, for the description is ac- companied by two text-figures; its internal anatomy and the structure of its cuticle have not been described in the cited paper. Fortunately 4 PROCEEDINGS OF THE NATIONAL MUSEUM VOL. 73 my material includes a number of specimens which undoubtedly belong to the same species and a diagnosis of this species is found in the following pages. The study of its anatomy proves that it belongs to the genus Lowothylacus. In the West Indian region, however, more than one species of Sacculinidae occur. Besides that described by Gissler, this family is represented by many other species, as I demonstrated a few years ago when I examined the material collected by Dr. Van der Horst at Curacao (Boschma, 1925). Three specimens of Sacculina were represented in this collection, each of which constituted the type specimen of a new species. Although there are a great number of West Indian species of Sacculinidae, only two seem to occur as common forms. One of these is Gissler’s species, the other, which is described in the present paper under the name Drepanorchis occidentalis, has not been recorded before. It is an interesting fact that each of these species is found on several species of crabs, but each infests a definite group of systematically related hosts. Drepanorchis occidentalis lives on crabs of the family Majidae, whereas Lowothylacus panopaei is known as a parasite of Panopeus and nearly allied genera, all of which are representatives of the family Xanthidae: Consequently just as in European species the Sacculinidae of the West Indian region are restricted to definite families of crabs within which these two common parasites infest several different species. The descriptions of the two species that follow are entirely based on the material of the United States National Museum, in which both are represented by a number of specimens. DREPANORCHIS OCCIDENTALIS, new species Type.—Cat. No. 60608, U.S.N.M., on Mithraw forceps (A. M.-E.), “Fish Hawk” Sta. 71538. Deadman’s Bay, west coast of Florida. The shape of the animals is roundish or somewhat irregular, often with more or less well-marked angular tips. (Fig. 1.) The size is variable, some specimens have a greater diameter of 4 mm. or less, whilst others are much larger. The largest specimen in the collec- tion (fig. 1c, d@) measures 11.5 mm. from the dorsal to the ventral surface (the diameter at right angles with the axis through the mantle opening and the stalk). As in other species of Sacculinidae the size of the parasites depends at least partially on the size of their hosts: The larger specimens are found on crabs belonging to species which may attain a comparatively large size. The mantle opening, which hes approximately opposite the stalk, is rather wide. It is often surrounded by a wall-shaped projection of the mantle, which may form a tubelike expansion: ART. 5 PARASITIC CRUSTACEA OF WEST INDIES—BOSCHMA 5 The internal anatomy of the species resembles strongly that of the type species of the genus, Drepanorchis neglecta (see Boschma, 1927). The closed end of the testes in D. occidentalis is found in the posterior part of the visceral mass (nearer to the stalk than to the mantle opening) ; in D. neglecta this part of the testes is situated at a com- paratively greater distance from the stalk. The testes of D. occi- dentalis have a fairly large size. (Fig. 2.) The colleteric glands occupy about the center of the lateral sur- faces (in D. neglecta these organs are much nearer to the mantle opening). The colleteric glands contain a comparatively small num- ber of branched tubes. Fie. 1.—Two SPECIMENS OF DREPANORCHIS OCCIDENTALIS. 4@, FROM MITHRAX FORCEPS (A. MILNE-EDWARDS), THE SURFACH LYING AGAINST THE THORAX OF THE HOST, X 334. 0b, THD SAME SPECIMEN, THE SURFACE LYING AGAINST THE ABDOMEN OF THH HOST, X 334. Gy, FROM MACROCOELOMA CAMPTOCERUM (STIMPSON), THH SURFACH LYING AGAINST THE THORAX OF THD HOST, X 3. d, THH SAME SPECIMEN, THE SURFACE LYING AGAINST THE ABDOMEN OF THH HOST, X 3. IN THESE FIG URES THE MANTLE OPENING IS FOUND IN THH UPPER PART, THE STALK IN THH LOWER PART The typical characteristics of D. occidentalis are those of the ex- ternal and internal cuticle of the mantle. The external cuticle is a thin layer of chitin; its thickness does not exceed 10u. It has a smooth surface without any excrescences. Seen from above the sur- face is divided into small areas which have a diameter of 10 to 1ldp and are surrounded by more or less meandering lines. (Fig. 3 a, 0.) On the internal cuticle of the mantle a great number of retinacula are found, more than in any other species of Sacculinid. (Fig. 3, c-h.) The internal cuticle is divided into small areas of approxi- mately 100» diameter; in detached pieces of this cuticle the marginal parts of these areas are more or less wrinkled. Each of these small 6 PROCEEDINGS OF THE NATIONAL MUSEUM VOL. 73 areas bears one or two retinacula. (Fig. 3d.) The retinacula are about 30, long; their broadened basal part is surrounded by a circular _ groove at the place where they are attached to the cuticle. The apical part of the retinacula bears 5 to 10 spindle-shaped excrescences of about 15, length, which are provided with small lateral barbs. (Fig. 3 ch.) In the material of the United States National Museum Drepanor- chis occidentalis occurs on the following hosts: Macrocoeloma camptocerum (Stimpson) ; of Florida. Macrocroeloma diplacanthum (Stimpson) ; Cuba. Microphrys bicor- Fa (UES deferens nutus aiding off ; Florida; Cuba; _ Ba- Y ..,. hamas. o | hese Mithrax forceps A (Herbst) ; Bahamas. Mithrax forceps (A. Milne-Edwards) ; west coast of Florida. (Host of type.) Mithrax sculptus (Lamarck) ; Jamaica. Pitho anisodon (von | maztle Martens); off Florida. a cavity Pitho lherminieri gq with § Schramm; Key West WF oor =~west coast of Florida. Fic. 2.—DREPANORCHIS OCCIDENTALIS FROM MITHRAX Stenocionops furcata FORCEPS (A. MILNE-EDWARDS), LONGITUDINAL SECTION, coelata ( Milne-Ed- oes? wards) ; off Florida. Among these crabs the species of the genera Macrocoeloma, Mi- crophrys, and Mithrax belong to the family Periceride, and the family Majide is represented by the genera Pitho and Stenocionops. a 4 oe ny cee A a si Mo a Dey aa eD Nero At PORT se OS SRN RR Se ee M4, ee fa EMG ha, Oa Se LOXOTHYLACUS, new genus Body laterally compressed, mantle opening opposite the stalk. Visceral mass attached to the mantle at some distance from the stalk. Mesentery thin, extending from the place where the visceral mass is united with the mantle to the mantle opening. Colleteric glands with a number of branched tubes, in or near the central part of the lateral surfaces of the visceral mass. Testes curved, the convex part running along the mesentery. ART. 5 PARASITIC CRUSTACEA OF WEST INDIES—BOSCHMA 7 The type species of this genus is Sacculina corculwum Kossmann. Among my material I have complete series of sections of three specimens belonging to this species. All of these have a complete mesentery and curved testes, whilst the visceral mass is united with the mantle at some distance from the stalk. Kossmann’s figure of the type species does not show distinctly the unusual place of the stalk.2 The other characteristics of the genus (the curved testes and the complete mesentery) are clearly drawn in another figure by the ee eee oe Fig. 3.—DREPANORCHIS OCCIDENTALIS. d, PART OF THE EXTHRNAL CUTICLE OF A SPECI- MEN FROM PITHO ANISODON (VON MARTENS), X 440. 0b, PART OF THE HXTERNAL CUTI- CLE OF A\SPECIMEN FROM MITHRAX FORCEPS (A. MILNE-EDWARDS), X 440. C¢, RET- INACULUM OF A SPECIMEN FROM MACROCOELOMA CAMPTOCERUM (STIMPSON), X 440. d, INTERNAL CUTICLE WITH RETINACULA OF A SPECIMEN FROM MICROPHRYS BICORNUTUS (LATRBILLE), X 110. e¢€, RETINACULUM OF A SPECIMEN FROM MITHRAX SCULPTUS LAMARCK, X 440. jf, RETINACULUM OF A SPECIMEN FROM PITHO ANISODON (VON MARTENS), X 440. g AND h, RETINACULA OF A SPECIMEN FROM MICROPHRYS BICORNU- TUS (LATREILLE), X 440 same author.? Probably the first of Kossmann’s above cited figures. was not altogether correct, for all other peculiarities of my speci- mens closely fit in with this author’s description. As in the type specimen their external cuticle is provided with long and stout spines. On account of the last-named feature the species Lowothylacus corculum. (Kossmann), is one of the best characterized species of the whole family. The West Indian species which is described below differs from L. corculum especially in the smaller size of its cuticular excrescences. 1$ee Van Kampen and Boschma, 1925, pl. 2, fig. 3. 2 See Kossmann, 1872, pl. 2, fig. 5b. %Idem. pl. 2, fig. da. 8 PROCEEDINGS OF THE NATIONAL MUSEUM VOL. 73 LOXOTHYLACUS PANOPAEI (Gissler) Sacculina panopaei GissiEr, 1884 The animal has a roundish shape; in some cases the mantle has more or less distinctly marked tips. (Fig. 4.) The size is variable, though in general the parasites of this species are small; the largest specimen in the collection has a greater diameter of about 6 mm. | The mantle opening lies at the top of a small tube-like expansion -of the mantle, which is usually directed towards the surface which lies against the thorax of the host. In the type specimen (see Vic. 4.—TWwo SPECIMENS OF LOXOTHYLACUS PANOPARI (GISS- LER). Gd, FROM PANOPEUS HERBSTII MILNE-HDWARDS, THE SURFACH LYING AGAINST THE THORAX OF THE HosT, X 77%. 0b, THH SAMH SPECIMEN, THD SURFACH LYING AGAINST THE ABDOMEN OF THH HOST, X 7%. ec, FROM EURYPANOPHUS DEPRESSUS (SMITH), THE SURFACH LYING AGAINST THE THORAX OF THE HOST, X 7%. d, THH SAME SPECIMEN, THE SURFACE LYING AGAINST THE ABDOMEN OF THE HOST, X 7%. IN THESE FIGURES THS MANTLD OPENING IS FOUND IN THD UPPER PART, THE STALK IN THE LOWER PART \ *Gissler, 1884, figs. 1 and 2) this tubular expansion of the mantle also is well developed. In the accompanying figure (fig. 5), a longitudinal section parallel to the dorsal and the ventral surfaces, all the characteristics of the genus Lowothylacus are visible: The stalk is attached to the mantle at some distance from the visceral mass, a section of the closed part -of one of the testes is found in the posterior part of the visceral mass, and in the lower part of the figure (consequently in the anterior part of the animal) the visceral mass is attached to the mantle by the mesentery. The section is from the immediate vicinity of the mantle opening, a part of the sphincter which surrounds the mantle opening is visible in the lower part of the figure. art.5 | PARASITIC CRUSTACEA OF WEST INDIES—BOSCHMA 9 The testes are strongly curved, their extremity lies at a short dis- tance from the vasa deferentia (in LZ. corculwm the closed end of the testes lies much nearer to the mantle opening). The colleteric glands occupy about the central part of the lateral surfaces; they contain a large number of tubes. The external cuticle of the mantle is rather thin, 8-12» approxi- mately. Its upper surface consists of small areas which have an irreg- ular contour; the diameter of these areas is about 9-12. In some of the specimens of the material each of the cuticular areas bears a small spine (fig. 66); in other specimens a much smaller num- ber of these spines occurs, as only about one-fourth of the total number of areas has such an excrescence (fig. 6a). These spines are tapering from the base to the top, which is evenly rounded; they may attain a length of 20n and a thickness (at the base) of 3y. In different parts of the mantle of one speci- men of L, panopaei the spines may be of different sizes. (See fig. 6c, d.) On the whole the differences between the cuticular excrescences of different speci- mens (even those which live on different hosts) are not more striking than those found among the spines of different parts of the mantle of one specimen, ™0,-—esetsnacus ravoratt (isstnn) The internal cuticle of the LONGITUDINAL SECTION, X 380 mantle bears small retinacula (fig. 6, g-2) which consist of a very insignificant basal part and a number (3 to 5) of spindle-shaped excrescences of 6-9 length. Lateral barbs could not be detected on these spindles. The material of the United States National Museum contains specimens of Lowothylacus panopaei on the following hosts: Panopeus occidentalis (Saussure) ; Porto Rico. Eurypanopeus depressug (Smith) ; Florida (?); Texas, Panopeus herbstii Milne-Edwards; Jamaica; Cuba. left testis {s aL oft a ‘Motori land mesentery * 10 PROCEEDINGS OF THE NATIONAL MUSEUM VOL. 73 Gissler’s specimen on which the original description of the species was based was a parasite of Panopeus herbstii, collected on the coast of Florida (Tampa). AV MM oR sie t Fic. 6.—LOxOTHYLACUS PANOPAEI (GISSLER). @, PART OF THE EXTERNAL CUTICLE OF A SPHCIMEN FROM EURYPANOPERUS DEPRESSUS (SMITH), X 440. 0b, PART OF THE PXTHR- NAL CUTICLE OF A SPECIMEN FROM PANOPEUS H®PRBSTII MILNE-EDWarpDS, X 440. € AND d, APPENDAGES FROM TWO DIFFERENT PLACES ON THE EXTERNAL CUTICLE OF A SPECIMEN FROM PANOPEUS HERBSTII MILNE-EDWARDS, X 440. €, CUTICULAR AP- PENDAGES OF ANOTHER SPECIMEN FROM PANOPEUS HERBSTII MILNH-EDWARDS, X 440. f, CUTICULAR APPHPNDAGES OF A SPECIMEN FROM PANOPEUS OCCIDENTALIS (SAuUS- SURE), X 440. g, RETINACULUM OF A SPECIMEN FROM EURYPANOPEUS DHPRESSUS (SMITH), X 440. h, AND i, RETINACULA OF TWO DIFFERENT SPECIMENS FROM PANO- PEUS HERBSTII MILNE-EDWARDS, X 440 LITERATURE H. BoscuMa, 1925. Rhizocephala of Curacao. Bijdragen tot de Dierkunde, Afi. 24. , 1927. Uber europiiische Formen der Gattung Sacculina. Zool. Jahrb., Abt. f. Syst., vol. 54. A. GriarpD, 1887. La Castration Parasitaire et son Influence sur les Caractéres extérieurs du sexe male chez les Crustacés Décapodes. Bull. Scient. du Nord de la France et de la Belgique, ser. 2, vol. 18. 1888. Le Laboratoire de Wimereux en 1888 (Recherches fauniques). Bull. Scient. du Nord de la France et de la Belgique, ser. 3, vol. 19. C. F. Gisster, 1884. The Crab Parasite, Sacculina. American Naturalist, vol. 18. J. Gutrin-Ganivet, 1911. Contribution A l’Etude systématique et biologique des Rhizocéphales. Trav. Sci. du Lab. de Zool. et de Physiol. Maritimes de Conearneau, vol. 3. P. N. VAN Kampen and H. BoscuMa, 1925. Die Rhizocephalen der Siboga- Expedition. Siboga-Expeditie, Monographie, 31bis. R. KossMAnn, 1872. Beitrige zur Anatomie der schmarotzenden Ranken- fiissler. Verh. med. phys. Ges. Wiirzburg, new series, vol. 3 (also in: Arb. zool.-zoot. Inst. Wiirzburg, vol. 1, 1874). ‘G. SmirH, 1906. Rhizocephala. Fauna und Flora des Golfes von Neapel, 29. Monographie. @) TWO NEW CRABS FROM THE EOCENE OF TEXAS By Mary J. Rarusun Associate in Zoology, United States National Museum The species here described come from the same general region in. Texas. The Raninid was taken in a core drill, Lane No. 1, depth. 260 feet, by the Marland Oil Co., a part of the Thomas Jordan Survey of land in Navarro County. The Lane farm is located about 4 miles north of Kerens and 8 or 10 miles west-southwest of Tool in. Henderson County. Until now the genus Notosceles was known: from only one species, a Recent one. The Xanthid was collected by John E. Adams, Bureau of Eco- nomic Geology, Austin, Tex., in the bed of Little Brazos Creek, Brazos County, on both sides of the old Bryan and Brazos Valley Railroad bridge. The type-specimens are in the United States National Museum. Family RANINIDAE Genus NOTOSCELES Bourne Notosceles Bourne, Journ. Linn. Soc., Zool., London, vol. 35, 1922 po 73s type, NV. chimmonis Bourne, a Recent species from the Sulu Sea. NOTOSCELES BOURNEI, new species Plate 1 Raninoides sp. RATHBUN, in Hull, Bull. Amer. Assoc, Petroleum Geologists,. vol. 9, No. 1, 1925, p. 169. Carapace (pl. 1, fig. 4) about two-thirds as wide as long, widest. at the middle of its length, lateral spine in front of middle of antero- lateral margin. Surface minutely punctate, punctae close together ; except on the frontal and orbital regions where the surface is densely and roughly granulate. The granulation begins just in front of the: base of the lateral spine and at the end of a curved (convex forward): ridge which passes obliquely forward to a point just behind the: inner of the orbital fissures and then transversely behind the front: No. 2727.—PROCEEDINGS U. S. NATIONAL MUsEuUM, VOL. 73,, ART... 63. 76793—28 L 2 PROCEEDINGS OF THE NATIONAL MUSEUM VOL. 73 the surface of the ridge as well as the space anterior to it is granu- late. Front with a sharp well advanced median tooth with a small tooth on either side; outline obscure but the accessory tooth points obliquely outward. Orbital slits deep and open, the inner sinus longer than the outer, its outer margin exceeding the inner margin, while the reverse is true of the outer sinus; margin between the sinuses oblique and nearly straight. Outer orbital tooth broad, tipped by a long, but thick and slightly curved spine, convex outside. The antero-lateral spine of the carapace is similar to the orbital spine but longer; it also is curved but directed very slightly outward, but not so far as the line of the widest part of the carapace. The postero-lateral margin is slightly sinuous and marked by a raised line formed by a single row of close granules; it rounds into the truncate posterior margin. Length of carapace 12.6, width 8 mm. The above description is made from a small specimen imbedded in a cross section in such a way that only the dorsal surface of the carapace is exposed. Cat. No. 369608, U.S.N.M. Fortunately another specimen showing sternal as well as dorsal surface was taken previously from the Midway, basal Eocene, near Kerens, Navarro County; it was submitted to Dr. W. L. Stephenson by D. W. Ohern, Borealis Oil Co., Oklahoma City. 'This specimen lacks the frontal and orbital regions; it measures 25.6 mm. in length up to the anterior base of the lateral spine; approximate width 22 mim., estimated length 34.6 mm. Carapace (pl. 1, fig 1) very convex from side to side, high in the middle where it forms a broad smooth ridge; either side of it on the posterior half of the carapace there is a shallow depression; at the widest part of the carapace on the left side (the right side is broken away) there is a large round smooth boss. I am not sure that this is natural to the crab as it looks much like a swelling due to parasitism. Additional occurrence.—Two specimens were taken in the Midway of southwestern Arkansas at Buzzard Bluff, sec. 16, T.14 S.,R 26 W., Miller County, by J. P. D. Hull, and were returned to him. Comparison with type-species.—So far as the carapace and sternum (pl. 1, fig. 2) are concerned this species agrees in all essentials with ‘the Recent or type-species of the genus. The postero-lateral margins converge more rapidly and the posterior margin is correspondingly shorter than in V. chimmonis. The lobe at the middle of the supra- orbital margin is subquadrate instead of triangular. ART. 6 NEW EOCENE CRABS FROM TEXAS—-RATHBUN 3 Family XANTHIDAE Genus HARPACTOCARCINUS A. Milne Edwards Harpactocarcinus A. Mitne Epwarps, Ann. Sci. Nat., Zool., ser. 4, vol. 18, 1862, pp. 46 and 64; type, H. punctulatus (Desmarest) from the Eocene of Priabona. HARPACTOCARCINUS AMERICANUS, new species Plates 2 and 3 The specimens were obtained chiefly from the interior of con- cretions, but the material was so friable that the fossils broke into many pieces. One female and one male were in a layer 10 feet be- low the concretions; they are used as holotype and paratype. Diagnosis—Lateral teeth behind the orbit five, very small. Sur- face of carapace uneven in posterior two-thirds. Chelipeds similar and of moderate size in the female, unequal in male, the major one of enormous size, the fingers greatly elongate. Description—Carapace of female holotype (pl. 2, fig. 3) very little broader than long; chord of antero-lateral margin nearly as long as postero-lateral margin, and amply rounded; postero-lateral margin slightly sinuous, nearly straight. Surface very convex, more so from front to back than from side to side; punctate and very finely granulate. From either side of the widest part of the meso- gastric region a broad, nearly longitudinal furrow runs back to the intestinal region stopping short of the posterior margin. Kraepelin (1911) described a small species of Vaejovis from San Pedro, Calif. In this species the upper side of the body is reddish brown and the under side, as well as the legs, is yellowish. The post- abdomen is very short, each of the first three segments being broader 12 PROCEEDINGS OF THE NATIONAL MUSEUM VOL. 73 than long. There are 10 teeth in the pectinal comb in the male and 9 in that of the female. Nothing is known of the habits or dis- tribution of the species. VAEJOVIS BOREUS (Girard) NORTHERN SCORPION This scorpion (pl. 1, fig. 2), the most northern in its distribution of any American species, is unmarked and dark yellowish brown. In length it varies from 3.5 to 5 cm., and is of slender proportions. The sting has a broad base which expands imperceptibly into the vesicle. Webster (1923) reported the species as occurring in North Dakota in the region known as the “ Bad Lands.” His specimens were identi- fied by the writer. Chamberlin (1924) identified a specimen of this species taken at Medicine Hat, Alberta, Canada. It is not known for sure, however, that V. boreus is actually established in Alberta. Scorpions are constantly found under artificial conditions far north of their natural range, where they have been transported by ship- ments of household goods, fruit, lumber, or other materials. Such specimens have been determined by the writer from Washington and Philadelphia. It is well known, however, that in a state of nature scorpions are not found in the environs of either of these two cities. In the National Museum there are specimens from the fol- lowing States: Arizona, Nebraska, Oregon, Idaho, Wyoming, Mon- tana, South Dakota, and North Dakota. It is believed that this material indicates fairly well the natural distribution of the species. Nothing is known concerning its habits other than the situations in which they are found, which are those common to most scorpions. VAEJOVIS MEXICANUS Koch MEXICAN VAEJOVIS This scorpion is dark, reddish brown and is very granular. The postabdomen is large, the vesicle rather slender and the sting short. There are no ventral keels on the first two segments of the postabdo- men. Length about 4.5 to 5.5 cm. The species is probably the most widely distributed of any in Mexico occurring from Mexico City to the southwestern part of the United States. In the mountainous regions of Mexico it breaks up into two distinct subspecies, dugesi Pocock and smithi Pocock. In the National Museum collection there is but a single specimen which was taken at Eagle Pass, Tex. In the Baerg collection there is a specimen from Durango, Mexico, and two from Guatimape, Mexico. ART. 9 SCORPIONS OF WESTERN UNITED STATES—EWING 13 VAEJOVIS SPINIGERUS (Wood) STRIPED-TAILED SCORPION This species (pl. 2, fig. 4) is easily recognized from the others occur- ring in the region where it is found by its large size and the presence of four longitudinal dark stripes on the under side of the “tail,” or postabdomen. It varies in length from about 5 to 8 cm. and is rather stout-bodied and is exceeded in size in the Southwest only by Hadru- | rus hirsutus (Wood) (pl. 1, fig. 1), a hairy species of a different genus. Its home is the desert region of the Southwest where it has been reported from Texas to California. The type specimens were taken in Texas. This species must range far into northern Mexico, but it is not represented in the Baerg collection from the State of Durango. It is found particularly in rocky waste places where there is some moisture. However, during the spring of 1927 the writer took two adults under a large rock at a street corner in the center of the town of Tempe, Ariz. It is the most common scorpion in the Papago Saguaro National Monument, in the Salt River Valley, Ariz. At Yuma, Ariz., a persistent search for it by the writer failed in locating specimens. However, on May 4, 1927, E. E. Russell, of the Bureau of Entomology, took three specimens here along the reclamation levee and railroad track. A few years ago the writer determined a number of specimens of this species from southern New Mexico, for Dr. W. J. Baerg. Baerg (1924) allowed this species to sting him and reported the results as follows: “The sensation was very much like that of a pin prick, and the resulting pain, which was very slight, lasted scarcely half an hour. _ There was no white area around the punctures and not the slightest swelling or inflammation.” VAEJOVIS FLAVUS Marx This yellowish species is probably our most slender representative of the genus. The pedipalps are particularly weak and slender and the hands smooth and without keels. The fourth and fifth segments of the postabdomen are scarcely as stout as those in front of them. The species occurs in the arid Southwest, but is rarely encountered. Two specimens are in the National Museum, one from Albuquerque, N. Mex., and one from “ Fort Yuma,” Ariz. Records from Mexico appear to be wanting. The species doubtless occurs in that country near the United States border. 14 PROCEEDINGS OF THE NATIONAL MUSEUM you. 73 VAEJOVIS CAROLINIANUS Beauvois | SOUTHERN UNSTRIPED SCORPION Besides the common striped scorpion, Centruroides vittatus (Say) (pl. 2, fig. 5), there is another scorpion species somewhat smaller that is found in many of the Southern States. This species, V. caro- Uinianus, is unmarked, of a dark reddish brown, with a large post- — abdomen and rather slender pedipalps. A museum specimen meas- _ ures slightly over 4 em. in length. The species ranges from South Carolina to Texas. In the Na- tional Museum collection there are specimens from Alabama, Georgia, and Kentucky. Little is known about the habits of this species. VAEJOVIS SUBCRISTATUS Pocock This Mexican scorpion is of medium size (length 5.15 cm. male, | 5.20 female, according to Pocock) and has the body mottled above with reddish yellow and blackish brown. The first two segments of — the abdomen are broader than long. Pocock (1902) reports it from San Andres and Jalapa, Mexico. The only United States record that the writer has found for this species is by Borelli (1909), who | reports two young females from Tucson, Ariz., taken by Professor | Silvestri. VAEJOVIS SILVESTRII Borelli Borelli (1909) described as new a scorpion taken by Silvestri in. the Sierra Madre Mountains, Southern California. This appears to be the only scorpion taken by this collector in a region rich in described species. The species can not be satisfactorily placed, and may be a synonym of one of the several species of Vaejovis known from this region. Genus ANUROCTONUS Pocock In Anuroctonus the middle area of the pectines is broken up, sometimes in an indistinct manner, into seven or less pieces; and these divisions are unequal. The sting is sometimes bulbous near the base. But one species of the genus is known from the United States and northern Mexico. ANUROCTONUS PHAIODACTYLUS (Wood) SHINY-STINGED SCORPION Not only has this scorpion a shiny sting, but a most peculiarly shaped one. Between the tip of the sting and the base, but nearer the latter, there is an inflated, bulbous structure. Probably the term “swollen-stinged scorpion” or “bladder-stinged scorpion ” would be equally appropriate with the one given at the heading of this para- ———— — ART. 9 SCORPIONS OF WESTERN UNITED STATES—EWING 15 graph, which was originally suggested by Essig. This scorpion is from 5 to 6 cm. long, with a reduced postabdomen but a very large vesicle. The pedipalps are stout and the fingers short and strongly curved, These are black, while the remainder of the scorpion is a reddish brown. The shiny-stinged scorpion has a wide distribution. Pocock (1902) reports it from Virginia, Colorado, Utah, California, and Guatemala. The first of these records, however, should be ques- tioned. The species probably does not occur in a state of nature in eastern United States. National Museum specimens are from Utah and California. Genus UROCTONUS Thorell Uroctonus differs from Anuroctonus in the nature of the divisions | of the middle area of the pectines. This area is broken up into over five pieces, the most of which are subequal. Only a single | species is reported from United States and northern Mexico. UROCTONUS MORDAX Thorell MORDANT SCORPION The mordant scorpion has large pedipalps and stout hands, while the “tail” is somewhat reduced. The sting is but slightly curved, except toward its tip. It is a dark-brown species, the carapace and pedipalps being darker than the abdomen, and the abdomen in turn is darker than the legs. U. mordax is a Pacific coast species, being reported from Cali- fornia and Oregon. Banks (1910) reports it from nine localities in California, including one record from Santa Rosa Island. Borelli (1909) reports it from Oregon. In the National Museum collection there are four lots of specimens, all from California. The habits of the species have not been studied. Family BUTHIDAE The family Buthidae is at once distinguished from our other American families of scorpions by the shape of the sternum. The sternum is triangular, the sides being strongly convergent ante- riorly. The membrane at the base of the last tarsal segment of most of the legs has two unbranched spurs. The fixed arm of the chelic- erae is without a ventral tooth. The family Buthidae is the most widely distributed of all the families of scorpions. It is the only family to be represented in each of the six chief zoogeographical regions. Not only is it found in all these regions, but from several to many species are found in 16 PROCEEDINGS OF THE NATIONAL MUSEUM VoL. 73 each of them. Kraepelin (1905) gives the number of genera and species for the zoogeographical regions as follows: Genera Species MDF 0Y C0) VE: ) 0 Wappen a pe Lh ES TS ak Se Uae tA 11 95 Pallacanc tie ay el Sie Se the Act Oe iil Eu fa ge 5 62 5 ag Chin Fs PORN RA eg Sa lM Ds Pap SN ALLE rE a 8 45 Atustraltavn: F274 523242 oT Oh EAEAAE ALAR ERG SR! aS ae 3 8 Wearictie ie) on pias Wha Ei iia os Ed ad he 2 22 INGOUEODIGRE Sr ale 2 us Me eo 4 61 Four genera are now known to occur in the United States. They are keyed out as follows: KEY TO THD GENERA OF BUTHIDAH OCCURRING IN THE UNITED STATHS A’. First tarsal segment of legs III and IV with a distal spur__Uroplectes Peters. A’. First tarsal segment of leg IV without distal spur. \ B*. Oblique rows of teeth on fingers of chelae flanked with supernumerary WV SOL CO Cn oo aa SUP Sa ECO ZL Centruroides Marx. B’. Fingers of chelae without flanking supernumerary rows of teeth. C*. Most of oblique rows of teeth on fingers of chelae overlapping. Tityus Koch. C*. Rows of teeth on fingers of chelae not overlapping. Isometrus Hemprich and Ehrenberg. Genus UROPLECTES Peters ¢ Uroplectes differs from our other genera of Buthidae in having a distal spur on the first tarsal segment of legs III and IV. Itisa genus that belongs to the Old World, our species apparently being an introduction. UROPLECTES MEXICANUS Comstock 2 MEXICAN UROPLECTES Under this name Comstock (1912) describes a species as follows: “This is a pale species. There is no spine under the sting; the teeth on the finger of the palpus are in many oblique rows, with stouter teeth at the end of each and to one side; there are from 30 to 35 teeth in the combs; and the keels on the under side of the last caudal segment are very strongly toothed.” He writes that the species has been found in Texas and California. Kraepelin (1905) is of the opinion that this scorpion probably was introduced from the Old World. Banks (1910) does not include it in his list of California scorpions. There are no specimens of Ameri- can Uroplectes in the National Museum collection. 2 Comstock’s 1912 description is the first for this species. = appears to have used Banks’s manuscript name Uroplectes mexicanus. ART. 9 SCORPIONS OF WESTERN UNITED STATES—EWING Ly Genus CENTRUROIDES Marx _ In Centruroides the first tarsal segment of leg IV is without a distal spur; the teeth of the fingers are arranged in oblique rows with supernumerary rows on each side. This genus is exclusively American, but in this continent has a wide range, being found from Central United States to Chile. Many species are included in it. A key is here given to those of the western part of our country. KEY TO THE CENTRUROIDES OF WESTERN UNITED STATES x. Length over 8 cm.; body unstriped. Large dark species. ' ®B*. Fingers of hand furnished with nine middle rows of teeth, exclusive of SSULCNE Se FUND LAGNA yD OY 2 2 Sa) Sg SRN Sek ee ANF €. nigrescens Pocock. B’. Fingers of hand with only eight middle rows of teeth, exclusive of short CHOICES HIG 20) i EN Ae A Ra Dh ES A a IL ais C. margaritatus (Gervais). | A®. Length less than 7 em. Reddish brown or striped species. : B*. Sting without ventral spine or tubercle near base; fingers one and a half fy times/as Jong asihandli is. 2 oP. 2 8 C. exilicaudata (Wood). | _B’. Sting with a ventral spine or tubercle. Cc’. Abdomen with two more or less interrupted longitudinal stripes Be : above; sting with a spinelike tubercle below. : ? D*. Black stripes irregular but never appearing as a row of dots. io MAStern SPCCieS usr Y Ten” pinup were iy Seale C. vittatus (Say). ss D*. Black stripes partly or entirely interrupted at each body seg- ment so as to appear as a row of spots. Found in Cali- “LOY ot dU 2 gk wok ae Be Re OA Ba C. californicus (Girard). Cc”, Abdomen without longitudinai stripes; sting with a vestigial tubercle below; dorsal surface of abdomen very rough. C. sculpturatus, new species. = anaes e% a Ra CENTRUROIDES NIGRESCENS (Pocock) Sy ee Z BLACK SCORPION aa te ene gee es _ C. nigrescens is a large, long, black scorpion. The color varies from a very dark chestnut brown to a jet black. The appendages are slender, the fingers being about one and a half times as long as the hand. The postabdomen is much longer than the abdomen. The | sting is very long and curved with a sharp subaculear tubercle. A large National Museum specimen measures 10 cm. in length. Pocock (1902) reports this species from Xautipa and Amula in Guerrero, Orizaba, Mexico. Borelli (1909) does not mention it in | is list. Banks (1910) does not give it in his list of California Scorpions. In the National Museum collection there are two speci- |!mens from Eagle Pass, Tex., and one from San Antonio, Tex. H. B. Parks, apiculturist at the state apicultural research labora- Hes =i tory, a few miles south of San Antonio, Tex., told the writer that he is familiar with a large, black scorpion the description of which fits this species. While collecting scorpions at San Antonio the writer did not see any examples of the species. 18 PROCEEDINGS OF THE NATIONAL MUSEUM VOL. 73 CENTRUROIDES MARGARITATUS (Gervais) C. margaritatus is similar in general appearances to C. nigrescens, but has one less row of teeth in the middle series on the fingers, and the basal segments of the postabdomen and the legs are lighter in > color than the body. Male specimens measure as much as 10 cm. in length. : This is a widely distributed species occurring in America from the — southern United States to Brazil and Chile and, according to Pocock, | is found in West Africa. Comstock (1912) reports it from Florida. — Pocock (1902) states that there are specimens in the British Museum | from California. Banks (1910) does not include it in his list of California scorpions. Specimens from the United States are wanting © in the National Museum. According to Baerg (1925) this species is | the one most commonly met in the Panama Canal Zone. | Baerg (1925) experimented with this species, inducing it to sting white mice and himself. The effect on the mice was not serious, and | the bitten area showed no inflammation or other effect. When the scorpion stung Doctor Baerg there was a severe pain at first and the affected area was reddish. No permanent effects were produced. Baerg reports the experiences of two scientists who were stung by this species. In these cases there was considerable swelling pro- | duced and in one case some lameness in the tongue. CENTRUROIDES EXILICAUDATA (Wood) SLENDER-TAILED CENTRUROIDES This scorpion not only has a slender tail, or postabdomen, but also slender appendages. The fingers of the hand are very slender, being | about one and a half times as long as the hand proper. In the male the postabdomen is almost twice as long as the abdomen. There is no vestige of a tooth or tubercle under the sting. The absence of this tooth alone distinguishes this species from all other United States species of its genus. A male specimen in the National Museum collection measures 5.1 cm. in length. This species was orginally described from Lower California, the type specimens being in the National Museum. Also in the Na- tional Museum there are a number of specimens from Lower Cali- fornia, most of them coming from Cape San Lucas, but there are two specimens coming from San Diego, Calif. According to Jackson (1910) the stings of this species may prove | fatal to man. In fact he attributes many deaths to it in the State | of Durango, Mexico. It is possible in this connection, however, that the species involved was not C. exilicaudata, as I do not find it in a large collection of scorpions taken in Durango by Baerg. | ART. 9 SCORPIONS OF WESTERN UNITED STATES—EWING 19 CENTRUROIDES VITTATUS (Say) COMMON STRIPED SCORPION This species (pl. 2, fig. 5) which is frequently referred to as @. _carolinianus (Wood), has two broad, dark longitudinal bands on the } dorsal side of the abdomen. In young specimens these bands are usually entire, but in most of the mature individuals they are partly or even entirely interrupted at the middle of each abdominal tergite. In adults of both sexes the appendages and the postabdomen are of a uniform yellowish brown except for the distal part of the sting, which is black. In young specimens the hands and the fifth sezment _ of postabdomen are black, and the postabdomen has three longitudi- - nal black stripes below. Male specimens have a very long post- ' abdomen, but that of the female is only about a third longer than the abdomen. A male specimen measures 6.7 cm. in length, a fe- _ male specimen, 5.9 cm. C. vittatus probably occupies a greater area of the United States than any other scorpion. Pocock (1902) reports it from Georgia, Florida, Kansas, Texas, and California. The California record, | however, should be ascribed to another species. Specimens in the ' National Museum are from Arkansas, Louisiana, New Mexico, and Texas. The species has also been reported from South Carolina and doubtless occurs in all the Gulf States as well as Kentucky, ' Tennessee, and Missouri. - Concerning the habits and life history of this species considerable ‘isknown. Out of doors it is found very abundantly under the loose bark of large trees and logs, and under logs and stones on the ground. About human habitations it prefers probably above all else the wood- pile and crumbling stone or brick foundations. In some parts of Texas the writer has found it infesting back yards, and reports of ‘its infestation of houses have been fr equent. A number of specimens have been kept in captivity by the writer. One adult received WiDetober 2, 1925, was kept alive until late in August of 1926. In captivity C. sitiatun feeds readily upon small roaches and flies, but "refuses many kinds of insects and nearly all of the larger or iad | ones. While other scorpions have been observed to dig into the !) ‘ground in captivity, such a habit was not observed in this species. ~. vittatus has a habit of clinging to objects lying on the ground, ' so that when the latter are turned over with the hands one is liable ; to press upon the scorpion and get stung. The life history of C. vittatus has been studied by Smith (1927). He found the species to be ovoviviparous. Each young “was born in a very thin and transparent envelope from which it freed itself in about 15 minutes.” The young molted in from 3 to 6 days = 20 PROCEEDINGS OF THE NATIONAL MUSEUM vou. 73 and remained on the back of the mother from 5 to 15 days. Smith estimates the period of growth at from three to four years. The writer has induced this species to sting him and has observed the effects of its sting on others. At the time of the stinging there is a sharp pain, but this soon subsides. A small swollen area, or | wheal, usually develops about the puncture point. This soon disap- | pears. There are no permanent effects of the sting reported for the | species as far as known to the writer. CENTRUROIDES CALIFORNICUS (Girard) CALIFORNIA CENTRUROIDES The striped scorpion of California has been considered as a syn- | onym of (@. vittatus by Pocock (1902), but as pointed out by Wood | it differs from vittaiws in a number of ways. In the National Museum there are three specimens from California determined by Marx. Two of these are from Lake County and are labeled “ Cen- trurus vittatus Say var. californicus W.” and one is from Tule Lake and is labeled “ wittatus Say ” without any variety being given. An examination of these specimens shows that the Lake County specimens are similar to vétiatus but lack any definite dorsal stripes and have the integument much more roughened and in a manner noted by Wood. ‘The one specimen from Tule Lake is without the subaculear tooth or practically so. It should be referred either to C. extlicaudata or a new species now to be described. All dorsal colorations are wanting in this specimen, but this condition may be due to the preservative used. CENTRUROIDES SCULPTURATUS, new species The writer has found an unstriped rough species (pl. 2, fig. 6) with a subaculear tooth in southern Arizona. It is related to extlicaudata but has the subaculear tooth. It is also related to vittatus and cali- fornicus but has neither spots or bands dorsally. It is described as © follows: General color a yellowish brown. There are no dorsal stripes, | spots, or other color markings. Cephalothorax without dorsal color | markings; median groove pronounced; integument granular. Hands © slender, fingers about one and a half times as long as hand. On the | posterior part of the carapace there is a pair of longitudinal, tuber- culate, carinae that extend forward from the posterior margin about one-half the distance to the eyes. ‘They are situated about one-third | the distance from the median groove to the lateral margin of the carapace. Abdomen very coarsely granular above; each tergite with a more or less distinctly elevated tuberculate posterior margin and a — well-developed median carina. Postabdomen longer than abdomen; | i arr. 9 SCORPIONS OF WESTERN UNITED STATES—EWING EL first segment about twice as long as broad; last segment with a long, slender, strongly curved sting, a low pointed tubercle below the sting and a swollen vesicle. About 94 teeth in pectinal comb. Length, 5.2. cm. Type locality.—Tempe, Ariz. Type—Cat. No. 978, U.S.N.M. Description based upon two adult specimens, a part of two lots as follows: One adult and one young from Tempe, Ariz., March 29; 1927, by the writer, under a stone in a rocky place, and one adult and one young from the same place, April 1, 1927, by the writer, under stones near a camp of Mexicans. CENTRUROIDES OF NORTHERN MEXICO In addition to the Centruroides species reported from the western | United States in this paper, all of which probably occur in northern “Mexico, there are others found there, which as far as known do not extend as far north as the United States boundary. ‘These will be ' mentioned. CENTRUROIDES SUFFUSUS Pocock Judging from the material collected by Doctor Baerg in Mexico ‘this is the common scorpion in the State of Durango. It is the Sonoran representative of viitatus, from which it differs in having _ the dorsal black stripes wider than the median yellow one and in hav- ing the interocular triangle but slightly, if at all, darker than the _ area behind the ocular tubercle, as well as in a few other particulars. ‘It is near exilicaudata, but has bright stripes on the abdomen and a minute tubercle under the sting. Pocock described sujfusus as a subspecies of vetiatus. This mate- rial came exclusively from the State of Durango, Mexico. Although compared with Centruroides elegans in the original description, the _ closest affinity of this form is probably with C. ornatus Pocock, of , which it might with some propriety be considered a variety. CENTRUROIDES ORNATUS Pocock ORNATE SCORPION Pocock described this species from J alisco, Mexico. It is almost indistinguishable from his suffusus. In fact, the writer is some- what inclined to consider the common striped scorpion of Durango, if it were to be recognized as only a subspecies, as a subspecies of ornatus rather than vittatus as was done by Pocock. Further collect- ing in Central Mexico will throw much light upon the true affinity of ornatus. One specimen in the Baerg collection from Atotonilco, Mexico, should be referred to this species. 22 PROCEEDINGS OF THE NATIONAL MUSEUM VOL. 73 CENTRUROIDES CHIARAVIGLI Borelli Borelli (1915) described as new a species taken at Dinamita, State of Durango, Mexico. According to his description it is a pale yellow- ish species, with a triangular brown spot on the anterior part of the cephalothorax. The cephalothorax is densely and finely granular. The teeth in the pectinal comb of the male are 28 to 29; in that of female, 26 to 27. Length of male is given as 7 cm.; of female, 5.3 em. Numerous examples of males, females, and young were obtained. Borelli’s description of this species is very suggestive of C. suffusus Pocock. That the males should exceed the females by 1.7 cm. in length is somewhat unusual. Pocock gives lengths of suffusus as follows: Male, 6.2 cm.; female, 4.6 cm. ‘The ratio of Pocock’s meas- urements are, however, similar to those of Borelli. I can find noth- ing in Baerg’s collection of scorpions to correlate with this species except sujffusus, which agrees in many respects. Genus TITYUS Koch In 7%tyus the first tarsal segment of leg IV is without distal spur. The fingers are provided with overlapping oblique rows of teeth but there are no supernumerary flanking rows. This is an American genus. It is best represented in the Tropics but extends from California and Florida to Argentina. Only two United States species are known. They may be separated as follows: KEY TO UNITED STATES SPECIES OF TITYUS A’. Sting with large ventral tooth near base; total length about 8 cm. Found in A) (6) Gr 20 Fz SMe ea DAS API SOE ATOR NI 8 5 BANNAN SEL Op T, floridanus Banks. A’, Sting without a ventral tooth; total length about 5 em. Found in Cali- HO) tg a FRM ele STI ae es LCN EIN Ea release eT T. tenuimanus Banks, TITYUS TENUIMANUS Banks CALIFORNIA TITYUS Our knowledge of this species is limited to the original description of it which was published in 1910. It is described as a yellowish- brown scorpion, 2 inches in length, with a weakly keeled hand and long fingers. It differs markedly from the Florida species, in the absence of the tooth below the sting and in its much smaller size. The only record for it is from Buena Vista Lake, Calif. Genus ISOMETRUS Hemprich and Ehrenberg Isometrus is similar to Tityus, but the rows of teeth on the fingers of the chelae do not overlap. There are no supernumerary rows of flanking teeth. The genus is represented in both the Old and New Worlds. ART. 9 SCORPIONS OF WESTERN UNITED STATES—EWING 23 ISOMETRUS EUROPAEUS Linnacus This long known and widely distributed species is remarkable for the degree of sexual dimorphism shown. The female is of a rather slender body, but in the male the postabdomen and appendages are exceedingly slender. The postabdomen of the male is about twice as long as the body and the hand is about four times as long as broad. The general color is a reddish brown. The subaculear tooth is long, sharp, and conspicuous. The length of the male is from 6 to Z cm. The species is frequently known under the name of J. macu- Jatus De Geer. This scorpion is reported from many parts of the world. It occurs in South America and the West Indies. In the United States it is known from Florida and California. There are two specimens in the National Museum, both from Key West, Fla. LITERATURE CITED Barre, W. J. 1924, The Effect of the Venom of Some Supposedly Poisonous Arthropods. : An. Ent. Soe. Amer., vol. 17, pp. 348-352, 7 figs. 1925. The Effect of the Venom of Some Supposedly Poisonous Arthropods of the Canal Zone. An. Ent. Soc. Amer., vol. 18, pp. 471-478. Banks, N. 1910. The Scorpions of California. Pomona College Journ. Ent., vol. 2, pp. 185-190, figs. 80-81. ' BorkeExti, A. 1909. Scorpioni raccolti dal Prof. F. Silvestri nell’ America settentrionale e alle isole Hawaii. Bol. Lab. Zool. Port., vol. 3, pp. 222-227. 1915. Scorpioni nuovi o poco noti del Messico. Bol. Mus. Zool. and Anat. Comp. Torino, vol. 30, pp. 1-7. CHAMBERLIN, R. V. 1924. The Northern Range of the Scorpion. Science, vol. 59, p. 64. Comstock, J. H. 1912. The Spider Book, 721 pp., 770 figs. Doubleday, Page and Co. DE MELLO CAMpos, O. 1924. Os escorpiones brazileiros. Mem. Inst. Oswaldo Cruz, vol. 17, fase. 2, pp. 237-368, pls. 2-13. Essie, H. O. 1926. Insects of Western North America, 1035 pp., 766 figs. Macmillan Co. JACKSON, H. V. 1910. Interstate Medical Journal, vol. 17, No. 7. KRAEPELIN, K. 1905. Die geographische Verbreitung der Scorpione. Zool. Jahrb. Jena, vol, 22, pp. 321-364. 1911. Neue Beitriige zur Systematik der Gliederspinnen. Jahrb. Hamb. wissensch. Amst., 2 Beih, vol. 28, pp. 60-108, 9 figs., 1 pl. Pocock, R. I. 1902. Arachnida-Scorpiones, Pedipalpi and Solifugae. Biol. Cent.-Amer., 71 pp., 12 pls. SmiruH, F. R. 1927. Observations on Scorpions. Science, vol. 65, p. 64. WEBSTER, R. L. 1923. Scorpion in North Dakota. Science, vol. 58, p. 248. 24 PROCEEDINGS OF THE NATIONAL MUSEUM _ VOL. 73, Arr. $ EXPLANATION OF PLATES All photographs were taken by J. G. Pratt and are natural size. PLATE 1 Fie. 1. Hadrurus hirsutus (Wood). Specimen from Calexico, Calif. 2. Vaejovis boreus (Girard). Specimen from Fort Pierre, 8S. Dak. PLaTE 2 Fig. 3. Vaejovis yosemitensis, new species. 4. Vaejovis spinigerus (Wood). Specimen from Papago Saguaro National Monument, Ariz. 5. Centruroides vititatus (Say). Male specimen from Texas. 6. Centruroides sculpturatus, new species. Specimen from Temple, Ariz. O U. S. NATIONAL MUSEUM PROCEEDINGS, VOL. 73, ART. 9 PL. | SCORPIONS OF WESTERN UNITED STATES FOR DESCRIPTION OF PLATE SEE PAGE 24 U. S. NATIONAL MUSEUM PROCEEDINGS, VOL. 73, ART. 9 PL. 2 SCORPIONS OF WESTERN UNITED STATES FOR DESCRIPTION OF PLATE SEE PAGE 24 NEW VICKSBURG (OLIGOCENE) MOLLUSKS FROM MEXICO By C. Wyre Cookr Geologist, United States Geological Survey The fossil mollusks described in the following pages were obtained in November, 1920, by Dr. T. Wayland Vaughan from the Alazan clay at and near the type locality on Rio Buena Vista, in Vera Cruz, Mexico. It was intended to include the descriptions, which were written in 1921, in a comprehensive account of the stratigraphy and paleontology of the area covered by Doctor Vaughan’s investiga- tions, but as some of the papers by other authors that were to have been included in that report are not yet written this contribution is published in advance of the others. Thanks are due to the officers of the Aguila Oil Co., who presented the collections to the United States National Museum and who have generously given permission to publish all the scientific results of Doctor Vaughan’s expedition, and to the Director of the United States Geological Survey for per- mission to study this collection as part of the writer’s official duties. The illustrations are from photographs made by Mr. W. O. Hazard and retouched by Miss Frances Wieser. The Alazan clay was originally described by Dumble? as follows: Whether the fossiliferous shales at Alazan are an integral part of the lower hard blue shales or are unconformable upon them has not yet been fully determined, but they are probably later and are certainly upper Hocene. The type locality of the Alazan shales is on the Buena Vista River at the crossing of the road between Alazan and Moyutlan. At this place the stream has cut down to the blue shales and exposed that formation along its western bank and in the bed of the river for a distance of more than half a mile. Overlying the shales to the west is a hill of yellowish clay, probably Oligocene. On the east side of the river there is a broad valley covered to a depth of 20 feet or more with recent deposits. The general body of the blue shale seems to have been but little disturbed ; for the most part it is smooth and evenly bedded and has a low dip to the southeast. Three hundred yards below the crossing there is a limited area which shows the surface of the shale more or less disturbed and broken, and it is here that the fossils occur. In places it appears as if small basins or +Dumble, E. T., Geology of the northern end of the Tampico Embayment area: Califor- nia Acad. Sci., Proc., ser, 4, vol. 8, pp. 141-142, 1918. No. 2731.—PROCEEDINGS U. S. NATIONAL MUSEUM, VOL. 73, ART. 10 78959—28 x 2? PROCEEDINGS OF THE NATIONAL MUSEUM VOL. 73 potholes 8 to 10 feet in diameter had been eroded in the underlying shale and the fossil-bearing blue clays laid down in them. At other places the fos- siliferous beds seem broken and piled together in every direction. The entire fossil-bearing area is not more than 200 feet in length, and a few hundred yards below this the main body of shales ends abruptly as though faulted, and the water plunges into a deep pool. The material in which the fossils occur is very Similar to that of the main body of the shales, but the fossils here are entirely confined to the disturbed and eroded area, and not a single fossil was found elsewhere in this exposure, and none at all was found in the main body of shale. The fossils are fragile and, while abundant in this limited locality, are hard to separate from the shale. The fossils from the Alazan shales were submitted to Dr. R. E. Dickerson, who reports that they are of upper Eocene age, containing some forms char- acteristic of the Tejon of California and others of the upper Eocene of the Gulf coast. Although the Alazan clay has heretofore been correlated with upper Hocene deposits, study of the fossil mollusks has convinced the writer that the beds from which they came are of Vicksburg (Oligo- cene) age.” As the overlying Meson formation has been correlated with the Glendon formation, the Alazan doubtless is equivalent to the Mint Spring marl member of the Marianna limestone. No char- acteristic Eocene species were recognized. The species in the fol- lowing list are common to the Alazan clay and the Vicksburg group of Mississippi: Gemmula rotaedens (Conrad) Polynices (Huspira) byramensis Cooke Pleuroliria cochlearis (Conrad) ? Sinum mississippiense Conrad Pleurofusia aff. P. servata (Conrad) Architectonica trilirata vicksburgensis Drillia tantula (Conrad) (Dall) : Drillia ef. D. caseyi Aldrich Cassis caelatura Conrad var.? Latirus protractus (Conrad) ? Pecten poulsoni Morton? Latirus perexilis (Conrad) Macrocallista' (Chionella) sobrina Distorsio crassidens Conrad (Conrad) i Phos mississippiensis (Conrad) Corbula laqueata Casey Turritella mississippiensis Conrad Corbula engonata Conrad The following species are described as new in this paper: Gemmula alazana Cooke Turritella ceibana Cooke Gemmula mexa Cooke Natica alazana Cooke Gemmula mera var. mexita Cooke Polynices (Lunatia) lacrimans Cooke Pseudotoma alazana Cooke Polynices (Huspira) byramensis Cooke © Scobinella prionota Cooke Ampullina vaughani Cooke Glyptotoma rhombica Cooke Dentaliuwm ovale Cooke Borsonia aguilae Cooke Dentalium alazanum. Cooke Ancilla (Ancillina) alazana Cooke Amussium alazanum Cooke Protonema bartschi Cooke Pecten ceibanus Cooke The species named in the preceding lists probably include only a small part of the total number of mollusks in the Alazan clay, Al- 2This correlation was announced Dec. 27, 1923; see Geol. Soe. Amer. Bull., vol. 35, pp. 853, 856, 1924. | : ) arr.10 NEW VICKSBURG MOLLUSKS FROM MEXICO—COOKE 3 most as many more species are represented in the available collec- tions by specimens which, although good enough to identify, are ' too imperfect to serve as'types in view of the probability that more ' and better specimens will be found in the not distant future. Characteristic specimens of Anadara lesuewrit (Dall), a little ark that is very abundant in the Byram marl of Mississippi and rare in the Glendon formation, have been found in the Meson formation at |. Topila, Mexico. The following descriptions of the localities at which the fossils | were obtained are based upon memoranda furnished by Doctor Vaughan. STATIONS M. 47, V., M. 48 V., M. 49 V¥., CROSSING OF THN ALAZAN-MOYUTLA ROAD The exposures are a short distance south of (below) the road crossing over the river west of Alazan. The strata are much dis- turbed by minor crumpling. below the river bench were obtained useful varieties of stone pebbles, float bowlders, and driftwood. | From The Dalles, in Oregon, and Spedis, located on the Washing- © ton side of the Columbia, to the environs of Kettle Falls, near the | Canadian border, the mapping of archeological stations was contin- | ued both along the middle and upper Columbia and tributary river | valleys as the lower Snake, Yakima, Walla Walla, Deschutes, John | Day, Wenatchee, Methow, Okanogan, and others. | Exploration and study previous to this survey of the prehistoric culture types of the middle and upper Columbia River Valley have | been limited in extent or have been supplementary to a project cover- | ing another field. Though limited in their scope, these studies have | produced conclusive archeological data. In volume 1, part 2, of the Memoirs of the American Anthropolog- | ical Association, published in 1906, there appears a monograph by Albert B. Lewis on the ‘‘ Tribes of the Columbia Valley and the Coast | of Washington and Oregon.” This is an excellent summary of source | material concerning the Indian tribes of the Pacific Northwest coast | and those of the Columbia and tributary river valleys in so far as | early historical contacts, travel, and exploration accounts are con- | cerned. Conclusions drawn by Lewis with regard to aboriginal cul- ture areas and subareas, tribal migrations, and trade relationships as they existed at the time contact was first established with the white | race through exploring expeditions such as that of Lewis and Clark are still applicable and are for the most part sound. ‘‘The stone arrowheads, stone mortars and pestles, and carved stone images and animal heads found along the Columbia from the mouth . art. 11 PREHISTORIC PIT HOUSE VILLAGE SITE—KRIEGER 3 > of the Willamette to near the mouth of the Snake show that the simi- ‘larities throughout this region are not of recent origin. Some of the finest and best carvings come from above The Dalles, while very few, ‘if any, have been found below the mouth of the Willamette. It would (hence appear that this stonework, if it did not originate, at least had ‘its highest development in a region where wood was scarce, and thence ‘probably moved down the river. On the lower river wood carving ' probably took its place, as the wood here was soft and easily worked ‘ with stone and shell tools.” / Jncommenting on the archeology of the interior tribes of the Colum- +bia Valley, Lewis refers to the need for further knowledge regarding | culture centers and the various sources of culture diffusion that must have influenced the large area known as the upper plateau and the Great Basin. A study of a limited area of the upper plateau region at Lytton ‘and other sites on the Thompson River in southern British Columbia ‘ with regard to its archeology and early culture connections was made ‘by Harlan I. Smith. A summary of his investigations was published ‘in the Bulletin of the American Geographical Society.’ Smith also carried on archeological investigations for the American ' Museum of Natural History during 1903 in the Yakima Valley in Washington between Clealum of the forested eastern slope of the ‘Cascade Mountains and Kennewick on the Columbia River; also ' between the mouths of the Yakima and Snake Rivers in the treeless ' arid region of the Columbia Valley and in the vicinity of Priest Rapids. } ' finds made during his investigations, but that they antedate the coming of the white man to the valley of the Columbia, as no objects | of European manufacture were included.? Smith finds that the partial identity of the Yakima Valley and _ Thompson River region in the southern interior of British Columbia is supported by definite evidence. ‘‘The preponderance of chipped points over those ground out of stone; cache pits; circular lodges; Tings of stones; and of semisubterranean houses with stones on the encircling ridge; pairs of arrow-shaft smoothers, and bone tubes, were all found to be common to both regions. Tubular pipes, mod- ern copper tubes or beads, incised designs consisting of a circle with a dot in it and engraved dentalium shells, each of a particular kind, besides pictographs in red, rock-slide sepulchres, modern graves walled up with parts of canoes, the marking of recent graves with sticks, and the custom of burying artifacts with the dead were also ———_ OY = ee Ce - tS et 1 Vol. 38, May, 1906. *His larger publication on the ‘‘ Archeology of the Yakima Valley’’ appeared as vol. 6, pt. 1, of the Anthropological Papers of the American Museum of Natural History published in 1910. The results of his studies in British Columbia were published in 1900 as a part of the publications of the Jesup North Pacific Expedition. 4 PROCEEDINGS OF THE NATIONAL MUSEUM VOL, 73 found to be common to both areas. Circles of stones which mark places where cremated human remains were found in this region sometimes indicate graves in the Thompson River region.”’ During the months of August and September, 1924, an archeolog- ical reconnaissance of the lower Columbia River region was conducted for the University of California department of anthropology by W. D. Strong, W. Egbert Schenck, and H. J. Biddle. One of the fea. tures of this reconnaissance was a study of petroglyphs on the Wash- ington or northern side of the Columbia River across from The Dalles, Oreg. Some of the findings of this study were published in the American Anthropologist.2 The authors correlate technique and designs shown in the rock paintings and sculptures, especially those of realistically conceived and executed animal figures among which the buffalo, mountain sheep, elk, and deer, with those described by Spinden from Idaho and by Mallery from Utah, also with recently described petroglyphs from Virginia City, Nev. The authors con- clude ‘‘that while the data are scattered and incomplete The Dalles petroglyphs in question find their closest analogues in the Great Basin area, and would appear to mark the northwesterly limit of that type. * * * The validity of the tentative conclusions arrived at in this paper can only be substantiated by a more thorough study of the petroglyphic and pictographic art of the upper Columbia River and adjacent areas.” An area contiguous to that of the Middle Columbia River Valley on the east was investigated archeologically and ethnologically by Herbert J. Spinden in the summer of 1907 for the Peabody Museum. The results of this and later studies were incorporated in a mono- graph appearing in the Memoirs of the American Anthropological Association and entitled ‘‘The Nez Percé Indians.’’* The area occu- pied by the Nez Percés within historic times extended from the Bit- terroot Mountains on the east to the Blue Mountains on the west, so that the territory claimed by them included what is now a large part of central Idaho, eastern Washington, and eastern Oregon, located within the basin of the Snake River. The Nez Percés are of the same linguistic stock as the Shahaptian tribes of the Columbia. Valley. Their material culture possesses many traits in common, although environmental differences, due in part to the presence of the bison, and a closer proximity to the Indian tribes of the Plains and remoteness from the tribes of British Colum- bia have caused a cultural development along somewhat different lines than that possessed by the ancient sedentary tribes of the mid- dle Columbia before the coming of the horse. Other cultural differ- ences may be noted in the type of house structures; art designs 41Vol. 27, No.1, January, 1925. 4 Vol. 2, pt. 3, 1908. py sere dL PREHISTORIC PIT HOUSE VILLAGE SITE—KRIEGER 5 incorporated on surfaces of horn, wood, bone, and stone; substitution . of the bowl for the tubular stone pipe, and the entire absence of jade | celts and of ornamental objects of copper was noted in some of the ancient cemeteries excavated by the writer along the lower and mid- _ dle portions of the Snake River Valley. Ancient basketry designs also varied in the two areas. Investigations at the better known early centers of Indian popu- lation at The Dalles by the department of anthropology of the Uni- versity of California have led to the discovery of objects and of art designs similar to those of the wooded area of the lower Columbia Basin, although many surface finds of a varied nature have been made there. All of the upper plateau tribes had recourse to the salmon and other varieties of fish in the Columbia and tributary streams and derived in this manner a plentiful food supply. Groups from each of the plateau tribes visited well-known fishing grounds at the several rapids and falls. It is therefore possible to make minor comparisons of the various kinds of abandoned fishing equipment, pestles, bowls, kitchen middens, and other implements and domestic equipment left by these groups at the place of their temporary encampment. Due to the temporary nature of hunting and fishing camps, the large subsurface deposits of charcoal, kitchen refuse, arrow points, canoe parts, and other fragmentary objects found in the vicin- ity of these great trading and fishing centers afford no reliable indi- cation of age. The impermanent and unstratified character of the soil in the shifting beach or low river bench in which such deposits are found is less likely to yield important archeological results than are the permanent winter village sites, with their ruins of abandoned pit houses and adjacent cemeteries. The falls and gorge of the Columbia River where this stream breaks through the Cascade Mountains marks the beginning of the wooded area of the lower river valley. The early aboriginal inhabitants of this region possessed a distinct type of culture which was based prin- cipally on the use of wood in their arts and crafts, while the tribes occupying the middle and upper river were expert in the utilization ‘of stone, horn, and bone. Realistic carvings of human figurines in wood and bone, and curvilinear surface etchings on wood, bone, and stone characterize the lower valley, while a more formal, conventional rectilinear design executed chiefly on antler and stone distinguish the art of the arid middle Columbia Valley. Traces of Indian occupation are in process of rapid obliteration by the plow which is to-day the most productive excavator of antiquities. Of the many sites inspected by the writer, excavation was undertaken ateight. The largest collection of material exhumed, such as ceremo- nial burial offerings and skeletal remains, was obtained from the pre- historic pit house village site and cemetery at Wahluke, Grant County, 6 PROCEEDINGS OF THE NATIONAL MUSEUM VOL. 73 Wash. There was no evidence that burials there had ever been disturbed. Neither was there in the objects recovered from the graves . any indication of Hudson’s Bay Co. influence such as trade beads of glass or of shell beads which in historic times were traded to the Indians as a substitute for the Dentalium indianorum, or of iron knives’ or copper implements. It was likewise impossible to obtain at Wahluke any direct evidence of great antiquity of occupancy or of culture type. Wahluke is in Grant County on the west bank of the Columbia River, which in this part of its course flows north, immediately south of the point where the stream impinges on the precipitous escarpment formed of yellowish gray volcanic débris, white silt, volcanic dust, and ash known as White Bluffs. Wahluke is the site of a former pit- house village consisting of 30 semisubterranean structures erected in: an irregular row extending a distance of 100 rods along the river bench. Each habitation ruin to-day consists of nothing more than a stone-capped rim of earth surrounding a centrally excavated pit of varying depth with a diameter of 30 or more feet. The bench land at Wahluke is broad and high enough to afford protection against the flood waters of the Columbia. The opposite flank of the river is much lower and is subject to seasonal inundations, hence was not occupied by the ancient inhabitants of the region except as a place for procuring game. Low-lying river benches gradually sloping down to the river’s edge were favored watering places for animals which often traveled many miles to reach them, although no evidence was uncovered that the bison ever reached this region. The high river bluff which faces Wahluke at right angles on the north is an exposed section of the Ellensburg formation laid down in the late Tertiary in old fresh water lake beds that at one time extended from the Pacific coast to what is now the upper plateau country east of the Cascade Mountains. In the valley of the Yak- ima River the composition of the Ellensburg formation is coarser than that of the White Bluffs on the Columbia River which con- tains large quantities of volcanic ash and wind-blown dust. At Wahluke the Columbia River is deflected in a general south- easterly direction, where it completes the final sector of its course known as the big bend. The vertical escarpment of the White Bluffs formation lies hard against the northern end of the village site; the cemetery proper is an extension of the village and is directly south of the long and irregular row of semisubterranean pit-house ruins. White Bluffs extend in a line reaching from east to west. From the point where the channel of the Columbia is deflected, the escarpment continues on the west as a range of hills known as Saddle Mountains or, locally, as Sentinel Bluffs, forming a relief feature several hundred feet high. Twenty-five miles farther west this range ART, 11 PREHISTORIC PIT HOUSE VILLAGE SITE—KRIEGER i again lies at right angles to the Columbia, where the channel cuts a gap through it. This gap is just below the confluence of Crab Creek, and about 40 miles by a circuitous course upstream from Wahluke. From the gap through Saddle Mountains to Wahluke the river inscribes a’semicircle to which Saddle Mountains and White Bluffs are tangent. During this section of its course, the Columbia River passes over the 15-mile stretch of Priest Rapids, famous as a gathering place for various tribes during the fishing season. Here, on the west bank, was located the village of Smohalla, a leader in the Ghost Dance cult. Saddle Mountains are of importance ethnologically, as the range forms one of the few natural geographic barriers. Within historic times this range separated the Shahaptian-speaking tribes on the south such as the Wanapim, or Columbia River Indians, the Paliis, the Yakima, the Walla Walla, the Umatilla, and other tribes from the Salish groups, as the Kalispel, Winatshi, Okanagan, Nespilim, and others that occupied the region of central and eastern Washington north of the range. Saddle Mountains are geologically important because of their bear- ing on any attempted interpretation of the antiquity of man in the valley of the Columbia. In the deeply cut gorge of this river, in its escarpment of columnar basalt, is written much of the early Tertiary geologic history of central Washington. South of Saddle Mountains the basaltic lava flow is covered with thick deposits of andesitic materials, volcanic ash and dust, and loess. During the Miocene, sheet after sheet of basaltic lava was poured out over the greater part of Washington, all of eastern Oregon, part of California, and a large area in the Snake River Valley of Idaho. This basalt represents a great number of flows. About 20 of these are exposed in some of the lava bluffs of the Columbia and Snake Rivers. A cross section of the gorge cut by these rivers shows inter- vening beds of varying thickness of soil in which trees grew to a thickness of several inches before they were charred and buried by later flows. Embedded sand, clay, gravel, and soil débris all bear evidence of burning and baking. After the completion of the period of basaltic upheaval and the later depositions of andesitic material in the fresh water lake beds which characterized the Tertiary history of central Washington, there began the gradual uplift of the Cascades in the Pleistocene and the formation of the high plateau region. The invasions of ice sheets from the north and from the northeast date from this period. One of these great ice sheets came down the valley of the Okanogan River from the north and filled the old channel of the Columbia River, causing it to form new and more direct ones, among the more impor- tant of which are the Moses and Grand Coulees. Later, on the 8 PROCEEDINGS OF THE NATIONAL MUSEUM VOL, 73 recedence of the ice sheet, the Columbia River again followed its old channel, which it still occupies. Any cursory study of the geological history of the Columbia Valley must indicate that human occupation of this region during the Pleis- tocene was impossible. Supposedly valid evidence of man’s antiquity in the valley of the Columbia has been found, nevertheless, in the form of crude, unfinished stone implements cached in the vicinity of a glacial morraine in the Lake Chelan country. This cache, however, is entirely superficial or intrusive and was deposited at a much later date. Another cache of unfinished knives or spear points, shaped like- wise from andesite porphyry, was found embedded in a cremation burial on a bench of the Columbia River at a location locally known as Simmons’ graveyard, near Quincy, Grant County, Wash. The bur- ial here is unique in that it lies directly underneath and several feet below a pit-house village erected at a later date. The crude appear- ance of the roughly chipped unfinished leaf-shape thick sectioned stone blades has led to the mistaken assumption of great antiquity. Pleistocene faunal remains which protrude from the vertical escarp- ment of the White Bluffs formation along the Columbia River near Wahluke and are associated with weapon points and implements of chipped stone are no true indication of a living association of Pleisto- cene fauna and ancient man, as has been supposed. The elevation above the country surrounding White Bluffs provided a splendid observation point for the Indian bunter in search for game or on the the lookout for hostile strangers. White Bluffs also was a well-marked trail used by the Indian when he journeyed eastward and northward away from the river on food-gathering expeditions. The chipped stone objects found are clearly intrusive and belong to a much later date. The geologic history of central Washington does offer, however, an explanation in part of the material culture of the early occupants of Wahluke. Environmental factors there have served as a causa- tive agent, likewise as a barrier to the development of a culture complex within other than certain conscribed limits. This basic fact is most strikingly brought out in a consideration of their stone culture. Evidence obtained from the nature of the objects exhumed at Wah- luke would appear to indicate that in prehistoric times up to some as yet undetermined date there existed a close connection of material culture and tribal practices throughout the entire area of the western plateau in what within historic times became known as distinctly Salish, Shahaptian, and Shoshonean cultures. There is definite evi- dence that this culture extended far to the south and formed the nucleus or substratum of the Basketmaker culture of the Southwest. At many places along the banks of the Columbia and tributary streams sedimentary deposits are exposed which were carried down in ART. 11 PREHISTORIC PIT HOUSE VILLAGE SITE—KRIEGER 9 the flood waters caused by the melting of the snow in the Cascades and in the Rocky Mountains. These deposits often cover charred cooking stones, heaps of charcoal, kitchen refuse, and occasionally artifacts shaped from stone, horn, and bone, together with other defi- nite evidence of the location of a camp or burial ground. ‘Temporary fishing camps, where many discarded objects of domestic use and weapons and implements of stone and bone were abandoned and cov- ered with several thick deposits of sediment, were again in later years exposed when the stream formed a new bed or when flood waters eroded the banks. At Pateros, in Okanogan County at the confluence of the Methow River with the Columbia, seven strata showing human occupancy with intervening layers of sedimentary deposits are exposed on the flanks of a small island formed on the Methow side of the chan- nel. At Vantage Ferry, Kittitas County, on the west bank of the Columbia River there are three such strata, and so on almost con- tinuously along either bank of the river wherever local conditions as to geologic formation, elevation of sedimentary river bench land above danger from high water, steepness of banks, and bench location, such as width, accessibility, contour, and other factors warranted. The cemetery at Wahluke contained both primary and secondary burials, but practically no other type than that of ceremonial crema- tion. Burials were placed in graves forming an irregular row along the river bench south of and upstream from the pit-house ruins which at one time made up a village of semisubterranean habitation struc- tures. There is but one site known in the middle Columbia River Valley where a pit-house village had been erected above a habitation site and cemetery of an older date. This site has become known as Simmons’ graveyard and is located about 5 miles downstream from Trinidad, and about 50 miles upstream from Wahluke, and several miles north of Vantage Ferry, where are located the ruins of another pit-house village and cemetery of a later date. Cremation burials at Wahluke are usually three or more feet below the surface when undisturbed. A layer of flat stones was invariably placed in an oblong or circular ring as a protective cover against marauding animals and to prevent erosion of loose sand which forms the bench at this place. A thin covering of soil consisting of wind- blown ash, dust, and calcareous clay over compactly embedded sand makes up the formation of the village site proper. The body to be cremated was placed on a piece of matting of Indian hemp, oriented, sometimes with the head facing upstream, sometimes toward the east, or seemingly haphazard, but always with face down- ward or on the side. Accompanying the burial were ceremonial offer- ings of personal use and ornamentation—the personal property of the deceased. The pyre was built of driftwood logs. The fire must not have been carefully attended, as many of the skeletons are merely 78960—28——2 10 PROCEEDINGS OF THE NATIONAL MUSEUM VoL. 73 charred, while sections of logs, together with burial offerings of wood and objects shaped from bone, are often intact. No indication of burial houses such as were erected by tribes on the lower river were found at Wahluke or elsewhere on the middle Columbia. Several other forms of burial were practiced both at Wahluke and elsewhere along the middle and upper Columbia River. HarlanI. — Smith describes burials in domes of volcanic ash in the arid region locally known as scab land. Low knolls of but a few feet elevation composed of fine volcanic ash have been protected from the erosive action of the wind by grass clumps and sagebrush. Such domes may be the remnants of what was formerly a continuous layer of top soil, or they may have been formed as wind-blown deposits. Scab land or scab rock obtains its name from the flat fragments of basaltic rock embedded in the loess but exposed between the volcanic knolls or domes. Burials in such locations were of individuals and were accompanied with offerings of shell ornaments and of weapons. A protective circle of stones surmounts the dome burial similar to that placed on the cremation burials at Wahluke. This form of burial and other burials which were located in the talus or slide rock were observed by the writer at various locations on the sloping river. cliffs but not at Wahluke. A child’s grave, located on the rim elevation of a pit-house ruin, and several uncremated burials were found in the cemetery outside of the cremation row. The significance of these uncremated burials is not clear. In some of the graves at Wahluke, skeletons were oriented in such positions as to suggest secondary burial; parts of several skeletons were jumbled in a heap and were accompanied by veritable store- houses of burial offerings. Bodies thus buried had apparently been collected from the mamalose or burial islands where they had been exposed before the ceremonial cremation burial in the village ceme- tery. Individual cremation burials at Wahluke usually were primary burials. Skeletal remains from such burials were found to be intact in situ except for the several parts consumed in the cremation. Such individual cremation burials were effected with knees flexed and with skull facing downward or on the side. Incineration was so complete as to prevent recovery of any one entire skeleton. Skeletal fragments, including eight skulls, were recovered from the burn, providing suf- ficient material for reconstruction later at the Museum. In every case the skull showed a degree of frontal-occipital deformation, which was effected by pressure from a wooden cradle-board flap placed over the forehead in infancy, a practice continued by Indian tribes of the lower Columbia Valley within historic times. The cradle-board used by the modern Wanapim, or Columbia River Indians does not have this wooden flap or hinged flange pass- ing over the forehead. There is, however, among these Indians a arr. 11 PREHISTORIC PIT HOUSE VILLAGE SITE—KRIEGER DE certain amount of flattening of the occiput due to contact of the plastic infant skull with the uncovered cradle-board. One of the more pronounced artificially deformed skulls found at Wahluke was from an uncremated burial, although some of the cremated skulls uncovered are quite similar to those of the modern brachycephalic or broad-headed Shahaptian tribes, all of which have a certain amount of occipital flattening but not of the anterior part of the skull. One lesion of a pathologic nature in the skeletal material recovered at Wahluke was noted. This is a fusion of a lower right tibia and fibula, due probably to traumatic origin and occurring probably in sub adult life. Skulls obtained from a cemetery at Vantage Ferry, in Kittitas County, Wash., and from other burial sites farther north which were accompanied by ceremonial burial offerings of a distinctly Hudson’s Bay Co. derivation were in every instance similar to those occurring in the prehistoric burials at Wahluke. Burial offerings found among the burned charcoal and charred bones of the cremation burials at Wahluke were useful and decorative objects constituting the personal belongings of the deceased. Some of the larger pieces, such as hollowed stone bowls and long, polished stone pestles, were intentionally broken or “‘killed.”? Just one deco- rated stone bowl wasrecovered. Itisa beautifully symmetrical, pol- ished granitic piece, uniformly hollowed by pecking and crumbling with hammerstones and polished with pumice. A surface design in the form of repeated V-shape bas-relief figures made by pecking and grinding encircles the outer circumference. Paint cups and mortar bowls of stone are for the most part crudely hollowed out, although showing evidence of constant use. Paint cups still contained frag- ments of red and yellow ochers but no trace of a green or other col- ored paint. A green stain covered the surface of elk teeth and certain shell objects. This condition was caused by the penetration of copper salts from near-by copper pendants and beads and was not an intentional paint. A paint cup of Haliotis rufescens shell filled with red ocher used as paint was found. Most of the paint contain- ers exhumed along the Columbia are of granitic stone or of worked pumice. There were no wooden dishes or bowls at Wahluke. A large, flat, circular granitic mortar was picked up at the center of one of the pit- house ruins, at the location of the primitive hearth, as evidenced by the accompanying charred cooking stones of fractured red quartz and charcoal. Pottery was neither made by the ancient occupants of Wahluke nor wasitsuse knowntothem. The lack of a suitable friable potter’s clay _ Inay account for this lack in part, but, as in the case of definitely ' developed culture complexes elsewhere, it is impossible always to _ explain the absence or presence of pottery, agriculture, and the loom 12 PROCEEDINGS OF THE NATIONAL MUSEUM VOL. 73 in terms of environmental factors. Objects recovered from graves in the cemetery and from surface finds at the site of the pit-house village of Wahluke are principally animal, vegetable, and mineral products obtained from regions near by. They consist of objects shaped from stone, bone, horn, the bark of trees, grasses, and various vegetable fibers, human and animal hair, chiefly that of the mountain sheep and of the dog. Many objects shaped from varieties of Denta- lium indianorum and of abalone (haliotis) shell of the varieties Haliotis kamchatkana, Haliotis fulgens, and Haliotis rufescens were exhumed with the burial offerings at Wahluke. Other Pacific coast shells found in quantity-in graves along the middle Columbia, especially at Wahluke, are quite distinct from the unio or fresh-water clamshell and must have been brought to the interior by direct or indirect trade with tribes of the Pacific coast, either by way of the lower Columbia River or across the mountains from Puget Sound. That few objects were found shaped from wood either of a useful or ornamental nature isnoteworthy. Driftwood must have been plentiful if we are to judge from the large amount used in cremation. It is highly probable that artifacts shaped from wood might have been preserved in the burn along with basketry materials, fabrics, and objects shaped from horn and bone had they ever existed. It must therefore be concluded that burials at Wahluke antedate the highly developed technique in wood as practiced by the tribes of the lower Columbia. The more formal and conventional rectilinear art designs of the early occupants of the arid middle Columbia Valley were executed cheifly on antler and stone. Tubular steatite pipes found at Wahluke are of two types. The one, along tubular bowl-shaped object, entirely undecorated, obtained possibly throughintercourse with California tribes; the other, a straight, small-bowled, tubular pipe with long narrow stem, etched as to bowl and stem with rectilinear ornamental designs similar to those executed on other objects from stone, bone, and horn, is undoubtedly native to the middle and upper Columbia Valley. This tubular stone pipe is identical with the native tubular pipe of southern British Columbia and of southern Idaho. Another tubular pipe found rarely is the carved bear figurine type which comes from the northwest Pacific coast tribes. A catlinite bowl pipe was exhumed which indicates influence from the East. Nephrite celts of various dimensions and with highly polished surfaces seem to suggest an important exchange of materials with tribes of British Columbia; it is possible that much of the native copper came originally from the interior of British Columbia. It is impossible to determine to what extent objects of carved stone, such as decorated pestles and tubular pipes, or of copper beads, wristlets, amulets, and bangles, or of neplirite celts, enter prim- itive trade as finished products. This point must be determined by urther investigation. me ART. 11 PREHISTORIC PIT HOUSE VILLAGE SITE—KRIEGER 13 Species of shell from the Pacific coast other than dentalium and haliotis,' perforated either at the apex or lip, were used as objects of personal adornment. Several examples mounted on necklace cords of hemp, cedar bark, and sinew were found among the burial offerings in graves at Wahluke, also at Vantage Ferry; several with fragments of cord still intact. Varieties of shell identified are Diadora aspera, Olivella biplicata, Glycymneris subobsoleta, and a Columbia River species of bivalve belonging to the Protothaca. Among the many objects of personal adornment recovered from the cemetery at Wahluke are rectangular, perforated pendants, scrolled ear bangles, laminated wristlets, and tubular beads hammered and rolled from nuggets of native copper brought from the Cascades or obtained by barter from the coast tribes. Pendant cords of twisted fiber or of sinew were recovered only in part. Ornamental pendants and necklaces of elk and beaver teeth, hawk and eagle claws were still in situ as they were when attached to the body at the time of cremation. Such ornaments like those of horn and bone were incrusted with copper salts and thus preserved more completely than might otherwise have resulted. Etched bone tubes and gam- ing sticks of antler in sets of six, similar to those described by Teit and Smith from British Columbia, were exhumed. Stone ornaments, implements, and weapon points were shaped from semiprecious agatized and petrified woods, opal, chalcedony, and jasper taken from the river bluffs 40 miles to the north beyond Saddle Moun- tains. Ornaments, implements, and weapon points shaped from such materials are expressions of some of the most beautiful examples of the stone-chipping art. Small, narrow-stemmed, and symmetrically worked arrow points of agate, chalcedony, carnelian, jasper, and flint were found with the burials at Wahluke. No weapon points or chipped blades of black obsidian, which is so abundant farther south, were obtained. A chipped elongated, dia- mond-shape ‘‘ceremonial”’ object of mottled black and red obsidian, 8 inches in length, is the only specimen of obsidian obtained from burial offerings at Wahluke. Five similar ‘obsidian ceremonials’’ are represented in individual collections obtained from various sites along the middle and upper Columbia. To convey an idea of the abundance of resources in stone and of the great variety of uses to which such material was put by the early inhabitants of Wahluke, the following list is appended. Materials Uses’ Agatized wood__-___------ Drills; weapon points; scrapers; reamers; knives. GAGs tee tobe a edecpy Drills; weapon points; scrapers; etching tools; knives. MRSA ce cya x Fish knives; net sinkers. Saal aE tle aa ie Knives; weapon points; weaving implements; beads. 14 PROCEEDINGS OF THE NATIONAL MUSEUM VOL. 73 Materials Uses WANA Sac ee ee ee Paddle-shape blades; spindle whorls; abrasives; saws; reamers; knives; wedges; weapon points; mauls; hammers; hammerstones; bowls; paint cups; pestles. Basalt (columnar escarp- Pictographs; petroglyphs; scrapers; spades. ment). @hrysopases 2 eee Kar pendants. Chalcedony 022 uu2- fee Drills; scrapers; weapon points; knives; etching tools. Dilahase. isn ocp ot eee e Pestles; hammerstones. Diatomaceous earth and Abrasives. pumice. Dyorites Sse Sk ee oe Abrasives; whetstones; hammers; knives; net sinkers. j W's) FST RRR A RD Net sinkers. pL ll MS Weapon points; drills; groovers. Gramiece sos ec oes Pestles; rollers; mortar stones; net sinkers; clubs; bowls; hammerstones; grooved hammers and mauls; groovers. Greenstomesi 206 Joo et Drills; hammerstones; abrasives; smoothing stones; pestles. CES SS gi a SS AN a aL ig Flaked and chipped points; scrapers; etching tools; compasses; groovers; reamers; knives. Nephrite: ss o22 22 seks Adzes; celts; chisels. Obsidianie Woo 45 2 eee ““Ceremonials.”’ OC AE ae a a Weaving implements; weapon points; knives; scrapers. Petrifen wood!-- 2. 2228 Drills; scrapers; perforators; weapon points; weav- ing implements; reamers. Ouartzite sa eeeree oes Cooking stones; hammerstones; hammers; net sinkers; mortar stones; anchor stones; mauls; clubs; pestles. Sandsuones i soe Arrowshaft smoothers; bowls; abrasives; ornamen- tal disks. Schist (chlorite and mica)_. Pipes; pendants; weaving implements; beads. Slatel DeRose a sy Knives; weapon points. Steatite or soapstone ----_-- Ear ornaments; pendants; tubular pipes; beads. All forms of the stone ax, whether a hafted hammerstone, a mon- olithic ax, or a grooved and bitted ax, were lacking in the surface finds, and among the grave offerings at Wahluke the aboriginal inhab- itants depended on the hafted discoidal stone war club, the flaked hammerstone, the grooved maul, and the stone wedge in its stead. Utilization of available natural resources by the sedentary, non- agricultural people of Wahluke was indeed thorough. Although the variety of such resources was limited in extent by the practically arid and barren environment, density of the aboriginal population of the middle Columbia Valley may well have been greater than that of the white race occupying the same territory within historic times. To be sure, this utilization depended on native adaptability and a knowledge of the resources afforded by the river which was near at hand and by the distant mountain forests as well. ART. 11 PREHISTORIC PIT HOUSE VILLAGE SITE—KRIEGER 15 Then, too, it must be assumed, as in the Southwest, that all ves- tiges of former habitation do not necessarily presuppose synchronous occupancy. If each permanent pit house village site, and each tem- porary fishing camp, of which traces have been found along the middle and upper Columbia, had been occupied within the period of one decade, the total native population must well have exceeded that of the white race now occupying the same territory. For example, what once were thriving Indian villages of several hundred individ- uals, located at two distinct sites on the Columbia at Priest Rapids, is now a practically uninhabited and for the white man an uninhabit- able region. It is probable, however, that tribal warfare and habita- tion traditions limited the permanent native population, except during the fishing season, to small groups of Wanaptim or Columbia River Indians. The animal resources utilized by members of this tribe as revealed in burial offerings and surface finds at Wahluke are noted in the following: Elk (Cervus Canadensis) _...------ Decorated, geometrically etched, perforate ribs used as fillets; gaff for fishhook from section of antler; weaving implement from antler; bone gouges; digging-stick handle; gaming sticks shaped from sections of antler; wedges for splitting; knives for cutting; wristlets and ornamentally in- cised pendants; teeth perforated for use as pendants; meat dried for food; antler for bow staves. Deer (Odontocoelus Americanus)-.. Horn used as flaking tool, awl, weaving implement; skins used as clothing; sinew for sewing; meat used for food; sinew used to reinforce bow stave; antler used as gouge. 10a TS Se ERR SE ee An extinct variety kept for the use of their shaggy coat of hair in blanket making; kept as watch dogs and used on the hunt but not eaten. Mink (Putorius vison) ....------- Fur as ornaments on headdresses. Beaver_._-___---------.-.-----. Fur; teeth perforated and carved as pend- ants; also as knives, beads. 1G.) 34 6k pall A he i le alia Hunted for its meat; skins used in weaving blankets. Mountain sheep (Ovis cervina)._... Bow stave section; horn spoons. Fresh-water clam or unio (Proto- Used as a food; shell cut and perforated for thaca sp.). use as a pendant. Land otter (Lutra hudsonica) ____- Quiver case. REM DING! S02 oS Le Quills for ornamental display. LSP SRS aM ST Oa Teeth as pendants and weaving implement; claws as ornaments; skins as robes; as food. De ANd coyote oe eo Teeth used as ornaments; skins for covering. 16 PROCEEDINGS OF Sturgeon (Acipenser transmon- tanus). Blue-backed salmon (Oncorhyn- chus sp.). Steel-head salmon (Salmo Gaird- Eel (Lampreta cibaria).._-..-.--~- PSE Ted ee hat gh EPR a Lea CIN Ne END ae Bones of fish, birds, rodents and land animals. THE NATIONAL MUSEUM VOL. 73 Marrow for oil; ornamented tubes; beads; gaming tubes; awls, wedges. Beads; gaming tubes; whistles. Feathers for headdress; for feathering arrow shaft; claw used as pendant. Same use as eagle feathers and claws. As food; dried on racks and stored. As food; dried on racks and stored. As food; vertebrae perforated as beads. As food. Do. Do. No artifacts shaped from its bones uncov- ered. Many undetermined uses. No skeletal material uncovered. No skeletal material uncovered nor arti- facts shaped from bison bones found. Utilization of other natural resources as shells, berries, tubers and other vegetable or plant products, fiber plants, minerals, wood, and stone extends to a large variety of products obtained from the lim- ited materials near at hand and to others obtained at a distance over weil-defined trade or barter routes and on seasonal fishing or hunting trips. Vegetable and plant products are represented in the finds at Wah- luke as follows: Bark used in weaving baskets, mats, and as eord; hearth for fire making; bow stave. Woven fabrics and as twine. Cedar (Thuja gigantea) ..__.._--- Indian hemp (A pocynum cannabi- num). Squaw grass (Herophyllum tenax). Woven fabrics; basketry. Cat-tail (Typha latifolia).__------ Matting and woven fabrics for floors of tipi and sweat house. Matting and as covering for lodges. Whistles; other unknown uses, probably as ferrules, or slip collars, or sockets. Uses unknown. Tule (Scirpus lacustris) ...------- Bilderberry ek. Pets BE Birch-bark rolls (Betula micro- phylla). Sagebrush. \oS99se ie Soe Willow (Salix lastandra)__..-__-- Roughly woven twined fabrics; basketry. Framework for sweat house; basketry; bow stave. Rye grass (Hlymus) ------------- Fabrics and matting. Barley prass.— oS so ee Twine; fabrics; matting. Bear grass (Xerophyllum tenax)_._ Fabrics. Cottonwood (Populus trichocarpa). Bast or inner bark used as bedding and as twine; trunks hollowed by fire into dug- out canoes. arr, 11 PREHISTORIC PIT HOUSE VILLAGE SITE—KRIEGER 17 White pine (pinus monticola) __-~-- Firewood; canoes. Red fir (Pseudotsuga mucronata)_. Firewood. Juniper (Junipera occidentalis)... Berries for food; uses varied. Wild onion (Aliuwm geyeri)_.-.---- Food. Wild carrot (Daucus pusillus)___.- Do. AgmsWOOGE EIU Ae)S Vee Maite iney Used as bow staves. Wild tobacco “kinnikkinnik”? Smoked in pipes as tobacco. (Valeriana edulis). Lichen (Alectoria sp.)..--_------- Used as a food when stripped from bark of pine trees and boiled. Huckleberry (Vaccinium membra- Used as food, both fresh and preserved. naceum). BBGUICR Tou suri Wey al oop s20 ol cae As a twine and in making coiled or twined baskets. Sunflower seed (Helianthus sp.).-. Food. Currant (Ribes aurewm)_.--_.---- Do. Gooseberry (and many other varie- Do. ties of berries gathered in the mountains in midsummer). Salmonberry (Rubus spectabilis).. Used as a food. Wapato (Sagiitaria latifolia)__.._- Roots used as a food. Kouse (Lomatium kaus)_..------- Tubers used as a food in much the same manner as camas. Camas (Camassia esculenta) ._---- Tuberous roots used as food which was roasted in ovens and dried. Habitation structures at Wahluke, as indicated by their ruins, were semisubterranean circular pit houses. They were 30in number. No evidence remained to show the type of superstructure, as practically all of the original framework had rotted away. This condition is in striking contrast with the well-preserved artifacts of wood that were exhumed from cremation burials. Similar conditions of decay in pit- house ruins were noted at Pasco, Vantage Ferry, at the mouth of the Yakima River, and elsewhere. One possible explanation of this condition is to attribute greater age to such structures than to the cremation burials. Another more plausible explanation is that the excavations for pit-house structures are usually on a lower bench than the cemetery and that during an unusually high stage of the river were inundated, water often remaining within these pits for an entire season, Decay of all wood framework and utensils, such as were used, was inevitable. The pit house as a habitation was formerly built by peoples in British Columbia, in Alaska along the Yukon, and on the coastal islands from the Aleutians to Bering Straits, also in Siberia, This habitation structure was formerly known to groups of Plains : Indians, as the Pawnee, Mandan, and others. It survives as a cere- - monial chamber among the Pima, Pueblo, and certain California tribes, and as a sweat house among the Columbia River Indians. Two such structures were observed by the writer—one at the lower q falls of the Yakima River and the other at the mouth of Crab Creek. 18 PROCEEDINGS OF THE NATIONAL MUSEUM VOL. 73 Another form of sudatory like that of the Nez Percés, figured and described by Spinden,’ is on the west bank of the Columbia 6 miles below the confluence of the Methow River. It is also identical with that of the Shuswap, Cree, Tinne, and other northern peoples. The skin-covered conical tent appears never to have been intro- duced into the valley of the Columbia. The framework of the conical mat-covered tipi of the Columbia River tribes is probably a local development and was apparently not introduced from the Shoshoni with the horse. The difficulty encountered by the Columbia River Indians in transporting skins of the buffalo from the east, beyond the Rocky Mountains, prevented any considerable use of the products of the buffalo. Danger of contact with hostile tribes probably also prevented the acquiring of obsidian from the east, deposits of which occur at various points in the volcanic area of the Yellowstone and of the Snake River Valleys. Essential characteristics in the art designs of the ancient Columbia River Indians and in the mode of their appli- cation and execution are those of the historic Indian tribes of the middle and upper Columbia Basin. They are similar, in many in- stances identical, with those of the tribes of southern British Columbia on the north, the Nez Percés on the east, and ina more limited, though none the less definite manner, with those executed on certain plastic materials as sandstone, wood, and born, by the Shoshonean and other southern stocks. Both curvilinear and rectilinear surface designs are applied by etching with bone, stone, or horn points, by rubbing, and by crumbling, though to a lesser extent also by carving out of the solid, by chipping, drilling, and by burning. A small slab of sedimentary sandy limestone was used as a tablet by the ancient dwellers at Wahluke on which to sketch decorative surface designs resembling somewhat the veins and outline form of a lanceolate-shape leaf. No evidence was obtained that such objects had ever been worn as pendants or that they represented more than casual sketches by some artistically inclined aborigine. The etched design is identical with that described by Harlan I. Smith from a sandstone pipe of the Nez Percés. Etched designs are more commonly applied to such media as den- talium shell and elk horn or bone than to other varieties of shell, of stone, orof wood. Rectilinear designs are commonly etched on bowls of tubular pipes of stone, also on stone weaving and plaiting imple- ments. Etching tools are simple yet effective. Some of the more interesting and ingenious devices are two-point compasses and spaced grooving tools. A few beautiful etching tools or points of chipped jasper, chalcedony, and agate are remarkable for the almost micro- scopical dimensions of the working surfaces. Designs executed with ‘ The Nez Percé Indians, vol. 2, pt. 3, p. 199, Memoirs of the American Anthropological Association, | ART. 11 PREHISTORIC PIT HOUSE VILLAGE SITE—KRIEGER 19 such etching tools may be seen to advantage on several dentalia shell where, through the surface etching of curvilinear lines in series and the presence of the surface in relief on intervening panels, beautiful patterns are obtained having much the appearance of an overlay twining basketry weave. The circle and dot design is applied with the three-point compass, or with spaced grooving tools. These decorative designs are applied in series on the surfaces of bone combs, gaming sticks, tubular bone objects, and on pendants, ear ornaments, and wristlets of antler. Many surface designs were found etched or pecked on horn, shell, stone, and bone resembling conventional geometric patterns etched, incised, or pecked on similar materials by the Thompson River Indians of southern British Columbia, as described by Teit and Boas. Other surface designs resemble more those utilized in basketry decoration by the Umatilla, Wasco, and Yakima tribes. Others are entirely unique. A wristlet of elk antler was found with triangular surface etchings built up in series of V-shape paneling entirely encircling the object. Each of the small triangular panels has as one of its sides a portion of the larger V-shape panel or triangle. The same rectilinear design appears on the surface of an overlay twined basket fashioned by the Umatilla (Cat. No. 330551 U.S.N.M.). Etched zigzag lines appear as the decorative effort on bone or stone (schistose) pendants and ear ornaments and are repeated again and again in variations of rec- tilinear surface decorations. Even such rare etchings and carvings of the human figure as do occur appear as rectilinear designs rather than in the more realistic curvilinear style described by J. H. Steward in the American Anthropologist. The carvings from Millers Island and from the mouth of the Deschutes River belong to the coast type. When the horse was introduced from the plateau country on the south and east, the culture of the ancient Columbia River Indians became much altered. The natives were now no longer so much dependent upon the river for food and for transportation as they had been heretofore. The open country between the Snake River, which reaches the Columbia from the southeast, adjoins the territory tray- ersed by the headwater tributaries of the Colorado. The culture of the horse increased rapidly on the Plains, after the nucleus of the herd had been formed by the few horses which escaped from Coronado’s troops. Trade and intercourse with the tribes living to the east of the Columbia increased, and many features of the culture of the Plains tribes were introduced. The absence of many objects of daily use and adornment in the culture of the Plains tribes was noted in the burials at Wahluke. «“ A new type of carving from the Columbia Valley,”” Amer, Anthrop., vol. 29, No.2, N. S., April- June, 1927. 20 PROCEEDINGS OF THE NATIONAL MUSEUM VOL. 73 The lack of chipped obsidian blades and knives has been mentioned. | There were no horse trappings, such as braided ropes of horsehair, nor do any rock sculptures of the figure of the horse appear anywhere except at Rock Island among the petroglyphs pecked into the crudely columnar basaltic cliffs of the Columbia River north of Wahluke, either at Vantage Ferry, Sentinel Bluffs, or Beverly. No objects made from horn of the bison were exhumed, nor were there any other indications of the use of buffalo products. ‘There were no hafted and grooved axes; neither was there any indication in the burials that the Plains type of costume was used. Petroglyphs of the human ficure on the west escarpment of the Columbia at Vantage Ferry, also at Beverly and at Sentinel Bluffs, appear to indicate, on the con- trary, the use of some sort of a feathered headdress. That the prac- tice of wearing feathered headdresses was an early one is further strengthened by the etching of a costumed human figure on a wooden comb which was exhumed at Vantage Ferry, and by a carving on wood of a fully costumed human figurine showing use of a feather headdress, exhumed at Tampico. , EXPLANATIONS OF PLATES Unless otherwise noted, all objects described are from Wahluke, Wash. PLATE 1 Hand Pestles of Stone > No. 1. Large pestle of vesiculate basalt shaped by pecking and crumbling. The pestle is symmetrical in outline and shows tapering from center of shaft upward. Top section has been broken off. Two parallel decorative grooves encircle the pestle at a distance of 11.5 cm. (4.5 in.) from the base. Incised grooves are 1 cm. apart and are each 44cm. deep. Dimensions: 21 cm. (8.3 in.) long; 7 cm. (2.7 in.) basal diameter. Collection of H. T. Harding, Walla Walla, Wash. No. 2. Flat-sectioned pestle of calcareous sandstone. ‘Tapered shaft; bulbous enlargement of basal end. Cap section has shallow cleft at center. Shaped by pecking with celt or hammerstone and smoothing with pumice. Dimensions: 16.5 cm. (6.5 in.) long; 7 cm. (2.8 in.) wide at base. Other stone pestles found at Wahluke have decorative heads showing more deeply incised cleft. Cat. No. 333581, U.S.N.M. No. 3. Hand pestle pecked from native greenstone. Tapered circular walls somewhat flattened at sides due to limitations in size of bowlder from which pestle has beenshaped. Characteristic basal section and irregularly shaped enlargement at top. Found with burial offerings in grave 1. Dimensions: 18 em. (7.1 in.) long; 10 cm. (4 in.) basal diameter; diameter of head cap 5 cm. (2 in.). No. 333578, U.S.N.M. No. 4. Hand pestle of diabase. More symmetricalin shape than No.3. The bulbous basal section is short; tapered walls are slightly convex; cap section is symmetrically rounded and convexly beveled at the top. Dimensions: 16.5 cm. (6.5in.) long; 8.5 cm. (3.3 in.) indiameterat base. Cat. No. 331995, U.S.N.M. No. 5. Decorated head of pestle of worked diabase. Polished surface; carved pestle head, 7 em. (2.8 in.). Cat. No. 3335938, U.S.N.M. image of animal figure forms the cap or head section; the figure has been shaped out of the solid, while features as mouth and eyes are indicated by incised lines; the eye is represented by the circle and dot design. Dimensions: Diameter of No. 6. Cigar-shape hand pestle of smoked granodiorite. Surface symmetri- cally rounded by pecking and beveled to tapering top and basal sections, with greatest diameter near basal end. Similar pestles of greenstone in graves at Wahluke are larger and have polished surfaces, which this pestle does not have. Cat. No. 333587, U.S.N.M. Exhumed from grave 2. Dimensions: 20 cm. (7.9 in.) long; 6 em. (2.4 in.) greatest diameter. an 12.3 cm. (4.9 in.) long; 5 em. (2 in.) wide. 21 No. 7. Small pestle of worked diabase. Pecked into a concavely beveled cylinder with no basal extension or knob at the top. From grave 3. Dimen- sions: 10 cm. (4 in.) long; 5 cm. (2 in.) diameter. Cat. No. 333585, U.S.N.M- No. 8. Small plummet-shape hand pestle of calcareous sandstone. A decora- tive design of beveled encircling rings and intervening pecked grooves forms the head or knob. Side walls are beveled to form a bulbous base. Dimensions: me PROCEEDINGS OF THE NATIONAL MUSEUM VOL. 73 No. 9. Decorative knobbed head of pestle shaped from diabase. The surface is polished and symmetrically rounded. Neither this head nor those of Nos. 5 and 8 are as characteristic of the stone pestles of Wahluke as are Nos. 2-4, and 6. Grave find. Dimensions: 7.5 em. (2.9 in.), diameter of knobbed head; distance from rim of knob to tip, 6 cm. (2.4 m.); diameter of shaft below rim- med knob, 5.5 cm. (2.2 in.). Cat. No. 333594, U.S.N.M. PuatEe 2 Types of Arrow and Spear Heads No. 1. Arrowhead of red jasper. Triangular shape with slightly convex lat- eral edges. Deeply concave base. Stem has straight sides and base chipped to a thin edge with cleft at center. Dimensions: 4.4 em. (1.7 in.) long; 3.6 em: (1.4 in.) wide at base. Grave find. Cat. No. 333947, U.S.N.M. No. 2. Arrowhead of smoky reddish-white translucent agate. Barbed; straight base with cleft; obliquely chipped facets. Found at mouth of the Deschutes River by H, T. Harding. Dimensions: 5.6 cm. (2.2 in.) long; 2.6 em. (1 in.) wide. No. 3. Triangular arrowhead of smoky chalcedony. Deeply concave lateral edges and base; no stem. Dimensions: 4.8 cm. (1.9 in.) long from wing tip to point. Found near Castlerock on the Cowlitz River, Wash. Cat. No. 317360, U.S.N.M. No. 4. Long arrowhead of brownish black and white agatized wood. Lance- olate leaf shape; concave base chipped to thin edge ; nostem. Columbia River beach, near Trinidad, Wash. Dimensions: 6.7 em. (2.5 in.) long; 3.1 em. (h:2 in.) wide. Cat. No. 333941, U.S.N.M. No. 5. Arrowhead of flint. Ovoid; stem has expanding sides and straight thin-edged base; irregularly chipped facets. Dimensions: 6.5 cm. (2.5 in.) long; 3.5 cm. (1.4 in.) wide. Cat. No. 317360, U.S.N.M. No. 6. Arrowhead of igray flinty quartz. Thick sectioned, leaf shape; con- vexly rounded edges slightly constricted at neck, terminating in wide stem with thick sectioned unchipped base. Dimensions: 8.2 em. (3.2 in.) long; 3.2 em. (1.3 in.) wide. Cat. No. 317360, U.S.N.M. No. 7. Arrowhead of gray flint. Triangular in shape; blunt of point with concave base; no stem. Dimensions: 3.7 em. long; 2.9 em. wide at base. Cat. No. 317359, U.S.N.M. No. 8. Arrowhead of dull red jasper. Triangular; concave base chipped to thin edge; no stem. Grave 3. Dimensions: 4.6 cm. (1.8 in.) long; 3 em. (1.2 in.) wide. Cat. No. 333941, U.S.N.M. No. 9. Long, narrow arrowhead of agate. Triangular in shape, with long, acute point and concave base. Stem has expanding edges and straight base. Symmetrically chipped surfaces. Exhumed with burial offerings from grave 4, Dimensions: 7.8 em. (3 in.) long; 2.2 em. (0.9 in.) wide at base. Cat. No. 333944, U.S.N.M. No. 10. Spear point of grayish white flint. Triangular in section; edges parallel near base, which is straight; no stem. In collection of H. T. Harding from beach of Columbia River 2 miles below Vulcan, Wash. No. 11. Leaf-shape flint arrowhead. Convexly rounded edges of point and stem. From the Cowlitz River, near Castlerock. Cat. No. 317360, U.S.N.M. No. 12. Arrowhead of agatized wood. Ovoid, leaf shape; base is concavely rounded and chipped to thin ‘edge. No stem. Collection of H. T. Harding, from Vulcan, Wash. Dimensions: 6.8 cm. (2.7 in.) long, 4.2 em. (1.6 in.) wide ae le a a a poe An, OS Seg ee ART. 11 PREHISTORIC PIT HOUSE VILLAGE SITE—KRIEGER 23 No. 18. Chipped spear point of translucent brown agate. Oval in shape with concave base; no stem; obliquely chipped facets. Dimensions: 10.5 em. (4.1 in.) long; 4 cm. (1.6 in.) wide; 0.9 cm. sectional thickness. Cat. No. 333850, U.S.N.M. No. 14. Chipped spear point of reddish gray flint. Roughly triangular in shape with edges slightly rounded near point; convex base has wide crescentic cleft at center chipped to thin edge, although unchipped and thick sectioned on the obliquely truncated portions near edges. Dimensions: 9.9 cm. (3.8 in.) long; 4.4 cm. (1.7 in.) wide. Cat. No. 333852, U.S.N.M. No. 15. Chipped arrowhead of crystalline quartz with opaline incrustations. Triangular in shape, with edges rounded near weapon point. Crescentic, unchipped, truncated base cut into three sections by two narrow obliquely chip- ped clefts which form a central stem with expanding edges. Dimensions: 7 em. (2.7 in.) long; 3.4 em. (1.3in.) wide. Cat. No. 333947, U.S.N.M. PLATE 3 Types of Arrowheads No. 1 Thick-bodied arrow point of whitish yellow jasper. Triangular, with slightly convex rounded edges; base, straight; stem, long necked with expanding sides. Dimensions: 3.1 cm. (1.2 in.) long; 2.4 cm. (0.9 in.) wide; 0.9 em. thick. Cat. No. 333948, U.S.N.M. No. 2. Thin sectioned arrowhead of black slate. Rough granular surface, irregularly chipped. Bilaterally cleft near base for cross lashing of sinew in hafting; base concave; no stem. Dimensions: 4.1 cm. (1.6 in.) long; 1.9 cm. (0.8 in.) wide. Cat. No. 333949, U.S.N.M. No. 3. Arrowhead of gray flint. Oval shape with acute point; bilaterally cleft for diagonal lashing of sinew seizing. Base, straight. Dimensions: 4.1 em. (1.7 in.) long. Found in grave on Cowlitz River near Castlerock. Cat No. 317360, U.S.N.M. No. 4. Arrowhead of dull red jasper with incrustation of agate at tip. Uni- formly chipped to a beveled section with high mid section. Oval in outline; nocked at the edges near base for diagonal lashing of sinew in hafting. Base, straight, chipped to a thin edge. Dimensions: 3.8e¢m. (1.5in.) long; 2.7 cm. (1.1in.) wide. Cat. No. 333943, U.S.N.M. No. 5. Thick sectioned arrowhead of opaque black obsidian. Oval shape; base, straight. Stem has expanding sides and straight base. Dimensions: 3.7 em, (1.5in.) long; 2cm. (0.8in.) wide. Cat. No. 333943, U.S.N.M. No. 6. Triple bilaterally barbed arrowhead of reddish brown jasper with in- crustations of agate. Narrow cleft base. Dimensions: 5 cm. (2 in.) long; cm. (0.8 in.) wide. Cat. No. 333947-R, U.S.N.M. No. 7. Arrowhead of smoky white chalcedony. Leaf shape, with elongated neck constriction and flaring bilateral wing barbs. Base, concave. Dimensions: 3.8 cm. (1.5 in.) long; 1.4 em. (0.5 in.) wide. Cat. No. 333947-T, U.S.N.M. No. 8. Arrowhead of reddish white mottled chalcedony. Triangular; deeply concave base inset with stem with expanding sides and straight base. Dimen- sions: 4 cm. (1.6 in.) long; 2.4 em. (0.9 in.) wide. Cat. No. 333944, U.S.N.M. No. 9. Long, narrow-bodied arrowhead of chocolate-brown agate. Facets chipped obliquely to high mid section. Base, straight; stem has convex base and expanding sides. Dimensions: 5 cm. (2 in.) long; 1 cm. (0.4 in.) wide. Cat. No. 333948, U.S.N.M. No. 10. Arrowhead of red jasper with opaline incrustations. Oval shape, serrated edges; convexly rounded base. Surface shows evidence of much 24 PROCEEDINGS OF THE NATIONAL MUSEUM VOL. 73 rechipping. Dimensions: 5.5 cm. (2.2 in.) long; 2cm. (0.8 in.) wide. Cat. No 333947-S, U.S.N.M. No. 11. Long, narrow-bodied arrowhead or smoky white agate. Slightly oval outline; base, straight inset with lenticular, diamond-shaped stem. Dimen- sions: 5 cm. (2 in.) long; 1.4 cm. (0.5 in.) wide. Cat. No. 333947-K. No. 12. Arrowhead of gray flint. Triangular; irregularly chipped surfaces; thick-sectioned body with straight base; stem, wide with expanding sides and deeply cleft base for the diagonal cross lashing of seizing sinew in haftings. Dimensions: 5 cm. (2 in.) long; 2 em. (0.8 in.) wide. Cat. No. 333947-A. No. 13. Arrowhead of yellow jasper. Triangular; wide stem, with expanding sides, inset in deeply concave base. Diagonally laid on notches separating base from stem are designed for use in hafting arrow shaft. Dimensions: 3.6 cm. (1.4 in.) long; 2.7 cm. (1.1 in.) wide. Cat. No. 333947-A. No. 14. Chipped arrowhead of glassy moss agate. Triangular; diagonally laid notches at base; stem has expanding sides and straight base. Found by H. T. Harding on beach of Columbia River 2 miles above Sandy Point, Grant County, Wash. Dimensions: 3.5 cm. (1.4 in.) long; 2.2 cm. (0.8 in.) wide. No. 15. Arrowhead of pinkish white jasper. Triangular, with convexly rounded sides. Base, straight; long-necked stem with concavely rounded edges expand- ing at base, where it is chipped thin for hafting. Dimensions: 4.4 em. (1.7 in.) long; 2.2cm.(0.8in.) wide. Exhumed by H.T. Harding at mouth of Deschutes River, Oreg. No. 16. Arrowhead of creamy gray jasper mottled with red. Lanceolate leaf shape; irregularly serrated rechipped edges; thick sectioned. Stem is wide and has expanding sides and straight thick base. Dimensions: 4.6 cm. (1.8 in.) long; 2. cm. (0.8 in.) wide. Found on beach at mouth of Deschutes River. Cat. No. 3339438, U.S.N.M. : No. 17. Arrowhead of opaque creamy white jasper. Triangular, with flaring wing barbs at base. Base is deeply concave and has a stem or nock of even proporticus. Chipped facets are regularly transverse, producing serrations on lateral edges. Dimensions: 4.3 cm. (1.7 in.) long; 2.2 cm. (0.8 in.) wide at base. Cat. No. 333944, U.S.N.M. No. 18. Arrowhead of cream colored jasper. Long, narrow body with evenly chipped serrated edges; convexly rounded base terminating in pointed stem or tang at the center. Goldendale, Wash. Collection of H. T. Harding. Dimen- sions: 4.9 cm. (1.9 in.) long; 1.4 em. (0.5 in.) wide. No. 19. Arrowhead of creamy white jasper. Triangular, flat section obliquely laid notches near edges, and wide stem with expanding edges and rounded concave base at center of base. Dimensions: 3.9 em. (1.5 in.) long; 1.9 em. (0.7 in.) wide. Cat. No. 333947-A, U.S.N.M. No. 20. Arrow point of light brown jasper. Thick-sectioned, evenly chipped, serrated edges terminating in bilaterally chipped barbs at base. Base is con- vexly rounded, terminating in inversely triangular stem or nock for hafting. Dimensions: 4.6 cm. (1.8 in.) long; 1.7 cm. (0.7 in.) wide. Cat. No. 333939, U.S.N.M. No. 21. Chipped arrowhead of red jasper. Narrow body with flaring bilat- erally laid wing barbs at base. Base is deeply concave and is inset with stem of uniform width. Dimensions: 3.9 em. (1.5 in.) long; 2 cm. (0.8 in.) wide at base. Cat. No. 333944, U.S.N.M. No. 22. Arrowhead of slaty gray diabase. Triangular sectioned, with two pairs of bilateral, obliquely laid barbs, separated by wide notches for diagonal lashing of sinew in hafting. Base is concave and is chipped to thin edge. Dimensions: 3 cm. (1.2 in.) long; 2.3 em. (0.9 in.) wide at base. Cat. No. 333947-I, U.S.N.M. aRT, 11 PREHISTORIC PIT HOUSE VILLAGE SITE—-KRIEGER 25 No. 23. Arrowhead of redjasper. Long, narrow, thick section; serrated edges. Base, straight; stem has straight sides and is rounded at base. Dimensions: 4 cm. (1.6 in.) long; 1.38 cm. (0.5 in.) wide at base. Cat. No. 333939, U.S.N.M. No. 24. Arrowhead of glassy, translucent, smoky chalcedony with mottlings of agate at tip. Thin sectioned, finely chipped over all. Transversely deeply chipped notches laid at the edges near base, forming bilateral wing barbs. Base is concave and is chipped to thin edge. This arrowhead and the previously described No. 23 and No. 9 are the more characteristic types at Wahluke. Di- mensions: 3.7 cm.(1.5 in.) long; 1.2 cm. (0.5 in.) wide. Cat. No. 333949, U.S.N.M. No. 25. Arrowhead of mottled agate. Triangular, thin in section; straight base eleft with notch at its center; transverse notches are chipped at sides near the base for lashing seizing of sinew in hafting. Dimensions: 3.1 em. (1.2 in.) long; 2 cm. (0.8 in.) wide at base. Cat. No. 333947-G. PLATE 4 Hammerstones and Scaling Knives No. 1. Hammerstone of andesite, with weathered natura] lateral surfaces. Flakes have been struck off at the edges from either side entirely around the circumference. Found on river beach below site of village located. on the bench above. Dimensions: 14.5 cm. (5.7 in.) diameter, 3.3 cm. (1.3 in.) thick. Cat. No. 333627, U.S.N.M. No. 2. Hammerstone of weathered greenstone. Oblong and roughly rectang- ular in section. The edges are abraded and crudely flaked at one end, also half the distance of one lateral edge. Found on beach below village. Greenstone pestles and celts are uniformly highly polished and symmetrically finished. This object shows no evidence of intentional working either on lateral surfaces or edges. Dimensions: 19.8 cm. (9.6 in.) long; 2.8 em. (1.1 in.) thick; 9.4 cm. (2.5 in.) wide. Cat. No. 333560, U.S.N.M. No. 3. Rectangular worked hammerstone of reddish-brown quartzite. Sur- face is naturally smooth due to weathering except at the edges, sides, and on one end where intentional fracturing and abrasion by use are shown. ‘The sec- tion at center shows less fracturing and is expanded, giving the hammerstone an oblong lenticular diamond-shape appearance. Surface find near cemetery. Dimensions: 15.6 cm. (9.1 in.) long; 5.7 em. (2.2 in.) wide; 3.5 cm. (1.4 in.) | thick. Cat. No. 333557, U.S.N.M. No. 4 Crudely fractured hammerstone of reddish quartzite. One of the many similar unworked tools fractured through use. This object and many similar hammerstone and cooking stones were found on the high bench land above cemetery site. Dimensions: 13.3 cm. (5.3 in.) long; 10 cm. (3.9 in.) wide; 6.5 cm. (2.5 in.) thick. Cat. No. 333564, U.S.N.M. No. 5. Unworked hammerstone of metamorphic greenstone. Smooth sur- _ faces except at ends which show abrasion by use. Found on beach below vil- lage site. Dimensions: 10.1 cm. (4 in.) long; 7.1 em. (2.3 in.) wide; 4.5 em. _ (L7 in.) thick. Cat. No. 333562, U.S.N.M. No. 6. Scaling knife of andesite. A flat, circular, water-worn pebble which _ has been given rectangular form by use and some additional intentional bilat- eral chipping at edges. Removal of facets by fracturing has provided four cutting edges at right angles. Chipping of edges is bilateral and uniform. Sur- _ face find on beach below village site. Dimensions: 9 cm. (3.5 in.) diameter; 1.5 em. (0.6 in.) sectional thickness. Cat. No. 833618, U.S.N.M. No. 7. Scaling knife of basalt. Weathered flat surfaces of pebble bilaterally | q chipped and fractured at two opposite edges only, providing but two cutting 26 PROCEEDINGS OF THE NATIONAL MUSEUM VOL. 73 edges, in contrast with the four cutting edges of No. 6. Other similar flat sur- faced rectangular scaling knives have three cutting edges. Dimensions: 9.7 cm. (3.8 in.) diameter. Cat. No. 333623, U.S.N.M. No. 8. Flaked sandstone knife. This oval form of stone flake is used as a general utility knife and is essentially a flake struck off a larger core by fractur- ing. A large flake has been struck off the mid section and there is rechipping entirely around the circumference on one side only. Beach surface find. Dimen- sions: 10.3 cm. (4 in.) long; 5.9 em. (2.3 in.) wide; 2.4 cm. (0.9 in.) thick. Cat. No. 333625, U.S.N.M. No. 9. Oval-shaped blade or scraper from a core of diorite. Reverse lateral surface is sand weathered with chipped and flaked edges; obverse surface has been shaped by fracturing over all and by rechipping entirely around the circum- ference. Dimensions: 10.7 cm. (4.2 in.) long; 9.9 cm. (2.7 in.) wide; 2.4 cm. (0.9 in.) thick. Cat. No. 333626, U.S.N.M. No. 10. Scaling knife or scraper of brown sandstone. This form of flake knife is similar to No. 8, and is itself a flake from a larger core. A large flake has been struck off its exposed or weathered surface by fracturing from one edge. Surface find. Dimensions: 13 cm. (5.1 in.) long; 5.9 em. (2.3 in.) wide. Cat. No. 333612, U.S.N.M. No. 11. Worked oblong granitic pebble used as scaling knife orscraper. Nat- ural, smooth, weathered lateral surfaces fractured on longitudinal edges but not at the ends by intentional bilateral pecking and further abraded by use. Surface find on beach below village site. Dimensions: 18.9 cm. (7 in.) long; 8.9 cm. (3. Sin.) wide; 5 cm. (2 in.) thick. Cat. No. 333558, U.S.N.M. PuaTE 5 Objects of Personal Adornment No. 1. Rectangular slab of argillite probably used as a mirror. Edges have been worked and corners rounded. No etched figures appear on the lateral sur- faces, nor is there a perforation for suspension. Surface find on cemetery site. Dimensions: 4.6 cm. (1.8 in.) long; 3.1 cm. (1.2 in.) wide; 0.3 cm. sectional thickness. Cat. No. 333688, U.S.N.M. No. 2. Lump of copper carbonate and malachite found with the burial offerings in grave 3. This material is not copper but occurs naturally between layers of mineral rock. Used probably as a paint ingredient. No. 3. Lump of red ocher found in grave 4. There was no paint cup found near the object. The ocher crumbles easily and must have produced a very effective paint. Dimensions: 4.1 cm. (1.6 in.) long; 3.1 cm. (1.2in.) wide. Cat. No. 333669, U.S.N.M. No. 4. Fragment of yellow ocher. Found in grave 5 with other burial offer- ings. Dimensions: 4.8 cm. (1.9in.) long; 2 cm. (0.8in.) wide. Cat. No. 333669, U.S.N.M. No. 5. Pendant of abalone shell (Haliotts rufescens). Perforated for suspen- sion at one end near center by drilling; perforation is of uniform diameter. The edges of the shell have been cut to the form of a rectangle, rubbed at the ends and rounded at the corners. Incised marginal etchings in series of three and five appear at one lateral edge. Dimensions: 12 cm. (4.7 in.) long; 6.3 cm, (2.5 in.) wide; 0.7 cm. sectional thickness. Collected by H. T. Harding at the railroad terminal in Wenatchee, Wash. Probably part of an exposed burial offering. No. 6. Chalcedoniec notched pendant or scraper. Found at Lyons Ferry near Almonta, on the Snake River. Lateral edges have been chipped to thin surfaces and are convexly rounded in shape. There is evidence of rubbing through use. End sections are concave, forming grooves for hafting or for cross lashing of arr. 11 PREHISTORIC PIT HOUSE VILLAGE SITE—KRIEGER 27 sinew cord for suspension. Dimensions: 2.7 cm. (1.1 in.) wide; 2.5 cm. (1 in.) long. No. 7. Hammered nugget of native copper used probably as an amulet. Lateral surfaces have been flattened by cold hammering and the edges are quite irregular. These irregular edges have been used as a hafting hold for the sus- pension cord of twisted fiber, probably Indian hemp. Two strands of this cord appear in crosswise lashing at the center of object. They have become miner- alized through impregnation with copper salts. Found with burial offerings in grave 1. Dimensions: 5.5 cm. (2.2 in.) long; 1 cm. (0.4 in.) sectional diameter. Cat. No. 333700, U.S.N.M. No. 8. Fragment of pendant of abalone (Haliotis rufescens) shell. The frag- ment has a circular beveled perforation drilled bilaterally for suspension. Found in grave 3. Cat. No. 333680, U.S.N.M. No. 9. Ear pendant of abalone (Haliotis kamchatkana) shell. Perforated for suspension near margin and at center. The edge has incised serrations extend- ing around the circumference. ‘There is one incomplete perforation near margin. This variety of Haliotis has a corrugated, convex, reddish tinged outer surface and a typically blue-green concave inner surface; it is an unusual variety among shell offerings in burials. Dimensions: 3.5 cm. (1.4 in.) in diameter. Cat. No. 333681, U.S.N.M. No. 10. Perforated shell bead. (Glycymneris subobsoleta Carpenter.) A flat shell peforated at apex for suspension. Found with burial offerings in grave 7. Dimensions: 2 cm. diameter. Cat. No. 333739, U.S.N.M. No. 11. Bead, perforated; cut from leg bone of a bird. Convexly rounded outer surface. 1 cm. in diameter; 0.8 cm. diameter of perforation. Cat. No. 333690, U.S.N.M. No. 12. Bone bead from perforated wing bone of a bird. Roughly trian- gular in section. Worked on both inner and outer surfaces. Cat. No. 333690, U.S.N.M. No. 13. Perforate shell bead of Glycymneris subsobsoleta. Illustration shows perforation at apex similar to that of No. 10. Diameter of shell, 2.4 cm. (0.9 in.). Cat. No. 333739, U.S.N.M. No. 14. Chipped stone drill or pendant. The object is ovoid and has neck constriction terminating in a three-faceted point. Dimensions: 3.8 cm. long; 1.8 cm. wide. No. 15. Bead or pendant from claw of an eagle or hawk. Stained a light green by contact with oxidizing copper in burial offerings. 4.1 cm. long. Cat. No. 333890, U.S.N.M. ; No. 16. Discoidal bead of steatite. Circular stone bead perforated at center hourglass fashion with a bilateral bevel from center. Irregularly cut outer cir- cumference. Dimensions: 0.8 cm. diameter; 0.3 cm. sectional thickness. No. 17. Small shell (Olivella biplicata) perforate for suspension at basal end for suspension in line with natural opening at the end fold or apex of shell. No. 18. Discoidal shell bead cut from a bivalve species of protothaca or clam- shell. Bilaterally beveled perforation at center. Dimensions: 1 cm. diameter; 0.2 cm. thickness; 0.3 cm. diameter of perforation. Cat. No. 333691, U.S.N.M. No. 19. Shell bead of Diadora aspera, pierced at apex for suspension. Dimen- sions: 2.3 cm. (0.9 in.) greatest diameter; 1.2 cm. thickness. Cat. No. 333740, U.S.N.M. No. 20. Large Olivella biplicata shell bead perforated like No. 17. Cat. No. 333741, U.S.N.M. Dimensions: 2 cm. (0.8 in.) long; 1.3 cm, (0.5 in.) diameter. No. 21. Elk-tooth bead perforate for suspension. No. 22. Bear-tooth bead perforate for suspension at end of root. The pers foration is drilled and is of uniform width throughout. 2a28 PROCEEDINGS OF THE NATIONAL MUSEUM VOL. 73 PLATE 6 Decorated Objects of Stone, Bone, Horn, and Wood No. 1. Section of carved deer horn from which fibrous core has been removed. Exhumed from burial at Vantage Ferry, Kittitas County, Wash. This carved fragment is similar to the carved antler animal figures obtained by H. I. Smith, at Lytton, British Columbia. Use unknown. Dimensions: 16.3 em. (6.4 in.) long; 5 cm. (2 in.) wide; 0.38 cm. thick. Cat. No. 330820, U.S.N.M. No. 2. Decorated pendant of stone (mica schist). Elongated oval shape flat lateral surfaces; thin in section. Perforation for suspension near smaller end is cut hourglass fashion at a bevel from both obverse and reverse sides. Trans- versely incised parallel decorative lines. Dimensions: 9.3 cm. (8.6 in.) long; 6 cm. (2.4 in.) wide. No.3. Weaving heddle of rectangular slab of mica schist. The object shows much evidence of use in the abraded surface near the lower end. Incised on the obverse lateral surface are decorative designs resembling figures of plants and of leaves. Found by H. T. Harding 2 miles below Vulcan in burial offerings; 9 em. (3.5 in.) long. No.4. Tubular bone object. Incised parallel grooved lines encircle circum- ference at either end near margin and one at the center. Dimensions: 4.4 em. long; 1.4 cm. diameter. No. 5. Rectangular slab of carved wood decorated with incised punctated designs. Identical number of punctations appear on each of the four lateral surfaces. Dimensions: 4.7 cm. (1.9 in.) long; 0.9 em. sectional diameter. No. 6. Decorated bone object. Edges are cut in roughly rectangular form, representing a garment. Extensions which had been cut at the sides near the top margin have rotted away or have been broken. The punctated design con- sisting of two parallel rows of incised dots probably represent beads of elk teeth. Found in grave 2. Dimensions: 9.1 em. (8.6 in.) long; 3.6 em. (1.4 in.) wide. Cat. No. 333928, U.S.N.M. No. 7. Carved and decorated comb of wood. The figure consists of a convexly triangular base which is perforated near the apex, and of eight teeth, each of which are broken off. Incised lines appear in parallel combined with the circle and dot design on the reverse side. The obverse has the figure of a woman appar- eled in a fringed garment etched on the surface. The comb, as it appears with teeth broken off, is 3.2 cm. (1.3 in.) wide; 0.5 cm. thick; and 8 cm. (8.2 in.) long. Nos. 8, 10-12. Gaming sticks of bone or horn. These objects were much charred or burned in the cremation fire and have become very brittle. The dec™ orative designs are etched on one lateral surface only, the reverse smooth surfaced. The sticks are oblong and are tapered toward each end. The etched designs are in the form of punctations, each from one to two tenths centimeters in depth, and in series of etched parallel zigzag lines in duplicate, triplicate, and in series of four. Dimensions: 8-9 cm. (3-314 in.) long; 1-114 cm. wide. Cat. Nos. 333661-333665, U.S.N.M. PLATE 7 Western end of White Bluffs escarpment at Wahluke, Grant County, Wash. The point where the Columbia River strikes these bluffs may be seen in the vertical, eroded walls of the escarpment on the right. The vegetation in the foreground is sagebrush; most of the grasses formerly occupying this area have disappeared within historic times. The ancient village of Wahluke is on the right and does not appear in this picture. Se ey art. 11 PREHISTORIC PIT HOUSE VILLAGE SITE—KRIEGER 29 The Columbia River at Wahluke. White Bluffs appear in the background, extending from the extreme left to the distant right. The pit house village ruins are in the foreground and are covered in part with driftwood from the extremely high flood waters of the Columbia River which covered the site in 1894. The cemetery is on the right in the foreground and does not appear in the illustra- tion. I+ is located on higher bench land and has never been covered with the flood waters of the Columbia. The western extension of White Bluffs where it joins with Saddle Mountains is on the left and does not appear in the illustration. O 1) ee th. CCRT ne Cele ' Ue Wat NK [ Aveb tay Ag \ ei aa a Gnd ¥ HDA TA ST es a U, S. NATIONAL MUSEUM PROCEEDINGS, VOL. 73, ART. 11 PL. 1 HAND PESTLES OF STONE FOR EXPLANATION OF PLATE SEE PAGES 2I AND 22 U. S. NATIONAL MUSEUM PROCEEDINGS, VOL. 73, ART. 11 PL. 2 TYpes OF ARROW AND SPEAR HEADS FOR EXPLANATION OF PLATE SEE PAGES 22 AND 23 PROCEEDINGS, VOL. 73, ART. 11 PL. 3 U. S. NATIONAL MUSEUM Types OF ARROWHEADS FoR EXPLANATION OF PLATE SEE PAGES 23 TO 26 U. S. NATIONAL MUSEUM PROCEEDINGS, VOL. 73, ART. 11 PL, 4 HAMMERSTONES AND SCALING KNIVES FOR EXPLANATION OF PLATE SEE PAGES 25 AND 26 U. S. NATIONAL MUSEUM PROCEEDINGS, VOL. 73, ART. 11 PL, 5 OBJECTS OF PERSONAL ADORNMENT FOR EXPLANATION OF PLATE SEE PAGES 26 AND 27 U. S. NATIONAL MUSEUM PROCEEDINGS, VOL. 73, ART. 11 6 * ERT RO ee ay DECORATED OBJECTS OF STONE, BONE, HORN, AND WOOD For EXPLANATION OF PLATE SEE PAGE 28 PL. U. S. NATIONAL MUSEUM PROCEEDINGS, VOL. 73, ART. 11 PL. 7 WHITE BLUFFS ESCARPMENT AT WAHLUKE, WASH. THE COLUMBIA RIVER AT WAHLUKE, WASH. FOR EXPLANATION OF PLATE SEE PAGES 27 AND 28 é * A REVISION OF THE LIZARDS OF THE GENUS CTENOSAURA By Joun WENDELL BAILEY Professor of Biology, Mississippi College, Clinton, Miss. INTRODUCTION A few years ago Dr. Thomas Barbour, of the Museum of Com- parative Zoélogy, in reporting on ‘‘Some Reptiles From Old Provi- dence Island’! made the following statement concerning the genus Ctenosaura: A single young Ctenosaura was obtained, which certainly is closely related to C. completa Bocourt. It is, however, not improbably distinct and undescribed. The genus Ctenosaura, however, is in quite a chaotic condition, but it can not be revised to meet the modern requirements of the discriminating systematist until the types of the early authors can be examined; and in this case the types are widely scattered in various HKuropean museums. ae Later Doctor Barbour took up the question of this genus with Dr. _L. Stejneger, curator of the division of herpetology at the United States National Museum, receiving in response to his inquiries the the following letter: Some time ago you asked my opinion as to Ctenosaura cycluroides and others | but frankly I am as muchin a quandary as you. Many years ago I tried to eet light on the question, but gave up in despair, waiting till I should get more material. In the course of time quite a number of specimens have accumulated, of the group Boulenger calls C. acanthura, certainly over 150—many very large specimens, and of these many not too well preserved. However, I have not had the courage to tackle them again. I had come to a tentative conclusion at the time, based chiefly on the characters of the verticils of the tail, which seem more _ reliable than those of the spines on the vertebral line, but I could not make up my mind which was the real C’. acanthura, which I think can only be ascertained _ from an examination of the type in London. Shortly after the receipt of the above letter by Doctor Barbour the writer matriculated in the graduate school at Harvard Univer- sity, and immediately fell heir to the ‘‘ Ctenosaura problem,” a task that has been difficult and at times discouraging, yet very pleasant because of the friendly interest manifested by coworkers in this country and in Europe. 1Barbour, T. Proc. New Engl. Zool. Club, vol. 7, pp. 81-85, May 6, 1921. No. 2733.—PROCEEDINGS U. S. NATIONAL MUSEUM, VOL. 73, ART. 12 88910—28——_1 1 Z PROCEEDINGS OF THE NATIONAL MUSEUM VOL. 73 This paper is based upon the collections in the British Museum of Natural History, London; des Naturhistorischen Museums, Ham- burg; Zodlogisches Museum, Berlin; Museum d’Histoire Naturelle de Paris; California Academy of Science, San Francisco; American Museum of Natural History, New York City; United States National Museum, Washington, D.C.; and the Museum of Compara- tive Zoblogy, Cambridge, Mass., altogether a large and representative series indeed. Two new species are described, one, Ctenosaura parkeri, from Barranca Iberra, Jalisco, Mexico, is dedicated to H. W. Parker, herpetologist, British Museum of Natural History, through whose kindness the writer was enabled to examine the important types in England. Two visits to the United States National Museum in Washington made it possible to study the types there. The second new species, clarki, is dedicated to Dr. Herbert C. Clark, director of medical research and laboratories, United Fruit Co., through whose interest and efforts the various collections at the Museum of Com- parative Zoélogy have been augmented from time to time. With the exception of Cyclura (Ctenosaura) teres, which was described from a living specimen, by Harlan in 1824, and of which there is no record of its ever having been preserved, and Iguana (Ctenosaura) similis, Gray, which was at one time in the Bell Museum, London, but subse- quently disappeared, the type specimens of every form referred to the genus have been carefully studied. To the following persons the writer wishes to offer his sincere thanks for valuable aid in the preparation of this revision: Mr. H. W. Parker, London; Dr. George Dunker, Hamburg; Dr. Ernest Ahl, Berlin; Dr. F. Angel, Paris; Dr. L. Stejneger and Miss Doris M. Cochran, Washington; Mr. J. R. Slevin, San Francisco; and Dr. Thomas Barbour, Cambridge, Mass. ABBREVIATIONS Mi: Ci Ziye ees ble Museum of Comparative Zoédlogy, Cambridge, Mass. Aino Me (No Hh geo’ American Museum of Natural History, New York, N. Y. TS IN MI Seta United States National Museum, Washington, D. C. Brite Vise se British Museum of Natural History, London, England. OA Se ce ns California Academy of Science, San Francisco, Calif. bi bt See Pe ae 8 ee Male. Bieet 0 oo ele Female. /. Nena © Bir eaee Per ie Adult. WS eee eee she Young. H-prown 2-22-22 Half grown. GENERAL CONSIDERATIONS The genus Ctfenosaura includes 13 species of lizards, the distribu- tion of which is confined to Mexico and Central America. In the West Indian region their place is taken by the genus Cyclura, to | | art, 12 REVISION OF LIZARDS OF GENUS CTENOSAURA—BAILEY > which some of the earlier described species of Ctenosaura were assigned. The two genera are very closcly related, but may be read- ily distinguished from each other by a comparison of the soles of the hind feet; in Cyclura there are peculiar corneous combs or pectina- tions on the under side of the toes; the toes of Ctenosaura are with- out such corneous combs. The species are powerful and very active, and can make a good defense when necessary by the use of their small sharp teeth, and of their spinose tail. This organ is armed with whorls of spinous scales which are very acute and which inflict considerable wounds when driven against the naked surface of the skin. Ctenosaurs are not much valued as food by the natives of Mexico and Central America, except by some Indians, and like other large tree and rock lizards are called iguanas. We know very little if anything of the geologic history of this genus, and one simply gropes in the dark in attempting to treat of this phase of thesubject. However, a few facts relative to the family Iguanidae, to which this genus belongs, will be given merely to throw some light on the possible origin and distribution of the group. All of the Iguanidae are confined to North and South America with the excep- tion of one genus (Brachylophus) which inhabits the Fiji Islands and two others (Chalarodon and Hoplurus) living in Madagascar. A fossil ‘species of iguana (Jguana europaea) has been described from the Eocene deposits of France and England. The Cretaceous genera Iguanavus and Chamops from Wyoming have always been considered as belonging to the Iguanidae, so there is no reason to doubt that the family has originated in America and that it was present during the latter part of the Mesozoic era. Although no very satisfactory conclusions, perhaps, can be reached regarding the main question of the origin of the species of Ctenosaura, the data derived from this study indicate very strongly the close relationship to, and their origin from, a common iguanid stock. Also from the present distribution of the species it seems but logical to believe that they originated at some place in central western Mex- ico, probably Nayarit and Jalisco, and that they have spread thence northward and southward until they cover practically the whole of Mexico and Central America. The transition in morphological char- acters has been gradual, and there is no obvious break in the series, indicating, of course, land migration only. By a strange coincidence the type, Ctenosaura acanthura, is both the most primitive and the most widely distributed species of the genus, and evidently had, at an early date, firmly established itself throughout Mexico, being numerous on both the east and west coasts. Even to-day this species has practically the same distribution. 4 PROCEEDINGS OF THE NATIONAL MUSEUM VoL. 73 The presence of several species within a short radius makes it im- possible to determine the origin of the various species. The order of arrangement of the species in this paper has been made with regard to structural relationship and not according to geographical distribu- tion or to any supposed origin. During the lower and upper Cretaceous, the upper Eocene and the early Oligocene periods what is now the peninsula of Lower California was a part of the mainland of Mexico, the present Gulf of California being dry land. Some of the members of this tribe of lizards migrated northward and westward, away from the foothills of the mountain ranges, finding their way to the semiarid desert regions of the Pacific coast, what is now the Cape St. Lucas region of Lower California. During the late Oligocene period the land between the desert region and the mountain foothills became submerged, creating the present Gulf of California. The ctenosaurs that were then shut off from their kindred on the mainland became adapted to the deserts, under- going of course a few minor changes such as would aid in the preser- vation of the species. The chief changes were in the shortening of the dorsal crest, both in the length of the individual spines and also in the extent of the crest on the back. Color markings were effected to give greater protective resemblance; resemblance to the speckled and splotched habitat of the species. This species is called Ctenosaura hemilopha. Its present range is the entire southern half of Lower ~ California and most of the islands near the peninsula, in the Gulf of California. A few individuals have been collected just across the gulf in Sonora, and as far north as Nogales, Ariz. They were in all probability carried there by man; but it is not impossible that their ancestral stock wandered there before the submergence of the Gulf of California. In the immediate vicinity of the center of distribution of the genus four species have arisen. They probably arose in the following order: brachylopha, brevirostris, pectinata, and parkerv. As the original stock, acanthura, continued its migration southward, other species appeared ; clarki and quinquecarinata. South of the Isthmus of Tehuantepec acanthura is replaced entirely by similis, a very active form which is abundant throughout Central America as far south as Panama. As the lizards continued their southward migration, new conditions in their surroundings led to new structural adaptions. A change in color took place, transverse stripes becoming conspicuous, and these prob- ably serve, as in the case of the tiger, to aid in the concealment of their possessor. This coloration is associated with changes in habits in Central America. Three other smaller species, bakert, palearis, and defensor, each with a very restricted habitat, have also arisen in this territory, all coming perhaps from similis; bakerv is restricted to Utilla Island, Honduras, defensor to northern Yucatan, and palearis to the =e ART. 12 REVISION OF LIZARDS OF GENUS CTENOSAURA—BAILEY 5 semidesert plateau region just south of the Motogua River in Guate- mala. The distribution of erythromelas is unknown. Part of the acanthura group upon reaching the Isthmus of Tehuantepec turned north, following the foothills of the mountain ranges along the east coast of Mexico, where they have been collected as far north as Tamaulipas. Early writers placed members of this genus in various genera. Shaw called his type Lacerta acanthura. Merrem called the same species Uromastyx acanthurus, while most of the early authors placed all these lizards in the genus Cyclura. The genus Ctenosaura was erected by Wiegmann in 1828, based upon Ctenosaura cycluroides (Ctenosaura acanthura), collected in Mexico by F. Deppe. Although the exact locality of Deppe’s specimens is not known, it is thought that they were taken near Vera Cruz, his first landing and collect- ing place in Mexico. In the spring of 1828 Deppe accompanied Doctor Schiede to Mexico, primarily to collect botanical specimens for the museum at Berlin. Zoological material was collected also, and some of the mammals, birds, and reptiles were described by Lichtenstein in 1838 and 1854. Their work was confined chiefly to eastern and southern Mexico, but some collections were made on the west coast by Deppe. Schiede and Deppe landed at Vera Cruz and after spending several weeks in that vicinity proceeded to Jalapa, where they arrived in early August. They left Jalapa November 28 for Papantla and Misantla. While in this vicinity they collected on Orizaba and Cofre de Perote. Writing under date of October 26, 1829, from the City of Mexico,’ Doctor Schiede stated that Deppe left him at Jalapa with the intention of going to California by way of Acapulco, but that he was prevented from carrying out his plans and was in the City of Mexico when he (Schiede) arrived. Schiede died about 1836 and after his death Deppe went on to California, probably by way of Acapulco and thence by vessel to Monterey or San Francisco, as he had originally planned. Early in the following year, 1837, he visited the Sandwich Islands, where he was with J. K. Townsend in Honolulu. The same year he returned to his home at Charlottenburg, about 1 mile from Berlin, where he remained until his death in 1861. The collector and date of collection of Shaw’s Ctenosaura acanthura are unknown, but the collection was made prior to 1802, the time the description was published. It was evidently collected in Mexico, where it is still not uncommon, since its habitat, the coastal region, is very large and contains much unsettled territory. The species hemilopha is common in the cape region of Lower California, while brachylopha is limited to southern Sinaloa, the islands, and the 2 Linnaea, vol. 5, p. 477. 6 PROCEEDINGS OF THE NATIONAL MUSEUM VOL. 73 mainland of Nayarit. It was taken as late as 1913 at San Blas, Nayarit, by J.C. Thompson. The species brevirostris and pectinata range from Nayarit southward to Oaxaca, in company with other _ species of the genus, while parkert, a new species, is described from specimens from Barranca Iberra, Jalisco. However, its distribution extends to Nayarit, specimens having been taken at Tres Marias by M. Forrer about 1885. The species guinquecarinata is known only from Tehuantepec, Oaxaca, while clarki has been collected at only one known locality—Ovopeo, Michoacan. The form defensor is very rare and has been taken only in Yucatan, but similis is very common throughout southern Mexico and Central America, including Panama, and is perhaps the most abundant species of the genus. Slight variations occur throughout the genus, not only in the species but even in the individuals. It is not at all uncommon to find speci- mens having a different number of femoral pores on the two legs. The femoral pores are much larger in the males than in the females. Glands at the base of these pores, in both sexes, produce a brown waxy secretion which hardens and protrudes from the openings. Although its function is unknown it appears to be most conspicuous during the mating period, and it may have some significance in that connection. Furthermore, the femoral pores are not always limited to onerow. Individuals have been examined in which the pores num- bered 7 on each side, 5 in one row and 2 in asecond row, parallel to the first. Another specimen having 7 femoral pores on each side had 6 in one row and 1 in the second-row position. Bothsides were patterned alike. The number of spines or lobes making up the dorsal crest also varies considerably with the species and sex. These dorsal crest spines are larger in the males than in the females. Age also causes a differ- ence in the size of the dorsal crest—the older specimens possessing the tallest crest. Individuals have been examined in which the number of small flat scales separating the whorls of large spinous scales on the upper half of the tail differ on the right and left sides of the ¢entral row of caudal spines. Sometimes the first and second whorls are separated by two rows of flat scales, the second and third whorls by two rows on the right and three on the left; and occasionally one of the spinous whorls is omitted on one side, giving that side twice as many, plus one or two additional, flat scales. This arrangement of the scales does not appear to be due to the loss of any, but merely to their disarrangement, for in the succeeding rows the “omitted” scales are found crowded in; thereby evening the count on both sides of the dorsal row. The greatest variations are to be found in the coloration of the individuals. This question is discussed under the respective species involved, especially in hemilopha and similis, so it is sufficient to say at this point that the young and adults differ very greatly in coloration— ART. 12 REVISION OF LIZARDS OF GENUS CTENOSAURA—BAILEY vi the young as a rule being more or less greenish, while the adults become darker and often marked with black or brown. In very old specimens the color oftentimes becomes a reddish or rusty brown or even black. Both young and old of some species have spots and stripes. The great number of synonyms found in this genus are prob- ably due, at least in part, to the lack of a proper consideration of these variations. At the beginning of this study it was thought that possibly there were some osteological characters upon which this and nearly related genera might be definitely separated. However, a careful examina- tion of Iguana, Ctenosaura, and Cyclura shows only slight differences in the skull, and even these differences can not possibly be called generic differences. As a matter of fact, the only differences are to be found in the general outline of the skull, and these are no greater between genera than between species of the same genus. The skull of Jguana and Cyclura are typically iguaniform in size and shape, while in Ctenosaura pectinata and similis the skull is slightly elongated and flattened dorso-ventrally. Yet in brevirostris the rostrum, as indicated by the specific name, is short; the skull is not flattened but would pass for a true Jguana. It is impossible to distinguish between the genus Ctenosaura and its near allies by means of skeletal characters. The early and most primitive forms of these lizards had very elon- gated tails and bodies—the true reptilian type, so to speak. Thus acanthura, supposedly the most primitive of the living forms of the genus, has a very long tail. It appears that as this form migrated the tail has tended to become shorter. It is interesting to note that along with the reduction in the length of the tail there is a corre- sponding increase in the size of the caudal spines. Also the species possessing the largest spines have the smallest bodies. The large spines on the tail will probably help to protect the species from enemies, while the small size of the body renders it undesirable as food for man, the most relentless enemy of these large lizards. Genus CTENOSAURA Wiegmann Type.—Ctenosaura cycluroides Wiegmann, 1828, Oken’s Isis, p. 371 (Ctenosaura acanthura). Ctenosaura WIEGMANN, 1828, Oken’s Isis, p. 371.—Gray, 1845, Cat. Lizards Brit. Mus., p. 191.—Bocourt, 1870, Miss. Sci. Mex., vol. 3, Reptiles, p. 136.—Copsg, 1885, Proc. Acad. Nat. Sci. Philadelphia, vol. 23, p. 262.—BouLENGzER, 1885, Cat. Lizards Brit. Mus., vol. 2, p. 195.—Corn, 1886, Proc. Amer. Philos. Soc., vol. 23, p. 216; 1887, Bull. 32, U. S. Nat. Mus., p. 33.—Gutnrunmr, 1890, Biol. Centr. Amer., Reptiles, Batrachia, p. 50.—Corsn, 1900, Report U.S. Nat. Mus. for 1898, p. 237.— Brown, 1904, Proc. Acad. Nat. Sci. Philadelphia, vol. 56, p. 468.—Dirtmars, 1907, Reptile Book, p. 106; 1910, Reptiles of the World, pp. 140-141.—Barpoor, 1916, Bull. Mus. Comp. Zoél. (Part), vol. 60, No. 4, p. 8 PROCEEDINGS OF THE NATIONAL MUSEUM VoL. 73 140.—SrEsJNEGER and Barsour, 1917, Check list N. Amer. Amph. Rept., ed. 1, p. 44; 1921, Proc. New Eng. Zodél. Club, vol. 7, p. 82.—Van Densurau, 1922, California Acad. Sci. Oc. Papers No. 10, Reptiles of West. N. Amer., vol. 1, p. 64.—STEJNEGER and BarBour, 1923, Check list N. Amer. Amph. Rept., ed. 2, p. 42. Uromastyx MERREM, 1820, Tent. Syst. Amph. (Part), p. 56, 1820.—Gray, 1845, Cat. Lizards Brit. Mus., p. 191. Cyclura Harvan, 1824. Journ. Acad. Nat. Sci. Philadelphia, vol. 4 (Part), p. 250.—Gray, 1827, Philos. Mag., ser. 2, vol. 2, p. 57 (Part).—_Winemann, 1834, Herp. Mex., pp. 15, 41 (Part).—Duménriz et Bisron, 1837, Erpét. Gén., vol. 4, p. 214-244 (Part).—FirzincER, 1843, Syst. Rept., p. 56 (Part).—Gray, 1845, Cat. Lizards Brit. Mus., vol. 2, p. 190 (Part).—Copsr, 1868, Proc. Acad. Nat. Sci. Philadelphia, p. 283 (Part)—Her1Lprin, 1882, Proc. Acad. Nat. Sci. Phila- delphia, p. 333 (Part).—CHapman, 1891, Proc. Acad. Nat. Sci. Philadelphia, p. 366 (Part). Enyaliosaurus Gray, 1845, Cat. Lizards Brit. Mus., p. 192. ; Cachryx Corr, 1866, Proc. Acad. Nat. Sci. Philadelphia, p. 124.—Copz, 1885, Proc. Acad. Nat. Sci. Philadelphia, vol. 23, pp. 262-270.—BouLEneER, 1885, Cat. Lizards Brit. Mus., vol. 2, pp. 195-198. Diagnosis of the genus——The members of this genus have the tail armed with strong spinous scales; tympanum distinct, nearly as large as orbit. The body is scarcely compressed; the scales of the median dorsal row enlarged, forming a dorsal crest. Scales of head and body small, those of the belly being smaller than those of the upper head, and those of the back being smaller than those of the belly. A very strong transverse gular fold, except in two species in which there is a large nondilatable longitudinal gular fold, the dewlap. There is a short series of femoral pores. Mandibular and maxillary teeth pleu- rodont, the lateral teeth only with denticulated crowns; pterygoid teeth present. The tongue is short and thick and slightly notched anteriorly, nonprotractile. Digits compressed, with keeled lamellae inferiorly, but without corneous combs or pectinations on the toes. Of the 27 species that have been described only 13 are valid. They are OC. acanthura (Shaw), 1802; bakeri Stejneger, 1901; brachylopha (Cope), 1866; bremrostris Cope, 1886; clarki Bailey, 1928; defensor (Cope), 1866; erythromelas Boulenger, 1886; hemilopha (Cope), 1863; palearis Stejneger, 1899; parkeri Bailey, 1928; pectinata Wiegmann, 1834; similis (Gray), 1831; quinquecarinata (Gray), 1842. These, may be separated by the use of the following key to the species: KEY TO THE SPECIES OF CTENOSAURA A!.—Median row of dorsal scales enlarged and extending from nape to end of tail, without interruption at the sacrum. These scales are usually large and armed with heavy spines, more pronounced in the males than in the females. Over the sacrum the crest consists of slightly raised and enlarged scales without spines. B!.—Head very short, rostrum conspicuously decurved-_--_---- brevirostris B?.—Head normal, rostrum not conspicuously decurved. C!.—First six whorls of spinous scales of the tail separated from each other by four or more rows of small, flat, smooth scales- - _parkeri ART. 12 REVISION OF LIZARDS OF GENUS CTENOSAURA—BAILEY 9 C?,—First six whorls of spinous scales of the tail separated from each other by fewer than four rows of small, flat scales. D!.—The first and second or the first, second, and third whorls of spinous scales separated from each other by two or three rows of small, flat scales, the next six or eight whorls being separated from each other by two rows of small, flat scales, body marked with black cross bands terminating on belly similis D?.—First five or six whorls of spinous scales separated from each other by three rows of smaller scales, the next five or six whorls being separated from each other by two rows of smaller, flat scales. No such black bands asin D___pectinata A?.—Median row of dorsa! scales low and interrupted at the sacrum, not contin- uous 8s in Al, B!.— Median row of dorsal scales extending only one-fourth to one-half dis- tance to sacrum, and not noticeably raised. C!.—Tail armed with 13 to 20 whorls of heavy spinous scales not int: spaced with whorls of small flat scales___.-_-__-__- defensor C?,—Tail armed with whorls of spinous scales, which are interspaced with one row of small flat scales. D!.—Row of small flat scales very conspicuous throughout length Ofgtailaak reheat SRS E. 2 sie ee a ke ea te BD clarki D?.—Row of small flat scales barely detectable on basal half of tail, but noticeable on distal half._.____.___- erythromelas B?.— Median row of dorsal scales extending to or almost to sacrum, notice- : bly raised, of medium height. C'!.—First two or more whorls of caudal spinous scales separated from each other by one row of small flat scales. D!.—Males and females possessing very pronounced dewlap palearis D?.—Not possessing dewlap but having transverse gular fold quinquecarinata C?.—Proximal whorls of caudal spinous scales separated by two or more rows of small flat scales. D!.— First, second and third worlhs of caudal spinous scales inter- spaced with two rows of small flat scales, the next five or six whorls with one row of small flat scales. E!.—Small dewlap present_.__........_--1-.-.+-2-- bakeri E?.—No dewlap, transverse gular fold present, back marked with prominent black blotches or spots__--hemilopha D?,—First three or more whorls of caudal spinous scales interspaced with three or more rows of small flat scales. E!,.— First and second or first, second and third whorls of spin- ‘ous scales interspaced with three rows of small flat SCAleS a Ae he er Datei. SET sb oe AO acanthura E?,—First, second, third, fourth, and fifth whorls of spinous scales interspaced with three rows of small flat scales brachylopha DISCUSSION OF THE SPECIES CTENOSAURA ACANTHURA (Shaw) Plates 1, 2, 3, 4 Lacerta acanthura Suaw, 1802, General Zoology, vol. 3, p. 1, p. 216.—Gray, 1827, Philos. Mag., ser. 2, vol. 2, p. 57. Uromastyx acanthurus MeRrreEm, 1820, Tent. Syst. Amph., p. 56. 10 PROCEEDINGS OF THE NATIONAL MUSEUM VOL. 73 Cyclura teres HARLAN, 1824, Journ. Acad. Nat. Sci., Philadelphia, vol. 4, pp. 242-251, pl. 26.—Garnan, 1884, Bull. Essex Inst., vol. 16, p. 19. Ctenosaura cycluroides WIEGMANN, 1828, Oken’s Isis., vol. 21, p. 371.—Bocourt, 1874, Miss. Sci. Mex., Reptiles, vol. 3, p. 143.—Sumicurast, 1880, Bull. Soc. Zo6l. France, vol. 5, p. 174.—Ives, 1891, Proc. Acad. Nat. Sci., Philadelphia, vol. 43,,p. 459.—Brown, 1908, Proc. Acad. Nat. Sci., Philadelphia, vol. 60, p. ale fe Cyclura carinata WAGLER, 1830, Nat. Syst. Amph., p. 147. Iguana (Ctenosaura) armata Gray, 1831, Cuv. Griff. Anim. Kingd., vol. 9, Synop- sis, p. 38. Iguana (Ctenosaura) belli Gray, 1831, Cuv. Griff. Anim. Kingd., vol. 9, Synopsis, p. 38. Iguana (Ctenosaura) lanceolata Gray, 1831, Cuv. Griff. Anim. Kingd., vol. 9, Synopsis, p. 38. Cyclura articulata WIEGMANN, 1834, Herp. Mex., pp. 42-43. Cyclura denticulata W1eGMANN, Herp. Mex., pp. 42-43.—HaLLOWELL, 1854. Proc. Acad. Nat. Sci., Philadelphia, vol. 7, p. 103. Cyclura (Ctenosaura) shawii W1EGMANN, Herp. Mex., pp. 42-43.—FITZINGER, 1843, Syst. Rep., p. 56. Cyclura semicristata FITzINGER, 1843, Syst. Rep., p. 56. Cyclura (Ctenosaura) articulata F1TzINGER, 1843, Syst. Rep., p. 56. Cyclura (Ctenosaura) belli FirzincEr, 1843, Syst. Rep., p. 56. Cyclura (Ctenosaura) denticulata F1rzinGER, 1843, Syst. Rep., p. 56. Cyclura denticulata HALLOWELL, 1855. Journ. Acad. Nat. Sci. Philadelphia, ser. 2, vol. 3, p. 36. Cyclura acanthura Sumicurast, 1864, Arch. Sci. Phys. Nat., vol. 19, pp. 49-50; 1864, Ann. Mag. Nat. Hist., vol. 13, p. 500.—Cops, 1871, Proc. Acad. Nat. Sci., Philadelphia, pp. 205-216; 1874, Journ. Acad. Nat. Sci., Philadelphia, ser. 2, vol. 8, pp. 95-124; 1879, Proc. Amer. Philos. Soc., vol. 18, p. 261; 1885, Proc. Amer. Philos. Soc., vol. 22, p. 379. Ctenosaura acanthura Gray, 1845, Cat. Lizards Brit. Mus., p. 191.—Copsn, 1866, Proc. Acad. Nat. Sci., Philadelphia, p. 124.—Sumicurast, 1880, Bull. Soe. Zool. France, vol. 5, p. 175.—BouLEenGER, 1885, Cat. Lizards Brit. Mus., vol. 2, p. 195.—Gtnruer, 1890, Biol. Cent. Amer., Rept. Batr., p. 5.—Ditmars, 1910. Reptiles of the World, p. 141. Cyclura (Ctenosaura) acanthura Copz, 1869, Proc. Amer. Philos. Soc., vol. 6, De Lok Ctenosaura teres Bocourt, 1874, Miss. Sci. Mex., Reptiles, vol. 3, p. 142.—Corn, 1886, Proc. Amer. Philos. Soc., vol. 23, pp. 266-268; 1887, Bull. 32, U. S. Nat. Mus., p. 34.—Van DenBurGgH, 1897, Proc. Acad. Nat. Sci., Philadelphia, vol. 49, p. 461.—Copsr, 1900, Rept. U. S. Nat. Mus. for 1898, p. 238. Cyclura (Ctenosaura) cycluroides GARMAN, 1884, Bull. Essex Inst., vol. 16, No. 1, p. 19. Cyclura (Lacerta) acanthura GARMAN, 1884, Bull. Essex Inst., vol. 16, No. 1, p. 19. Ctenosaura multispinis Core, 1885, Proc. Amer. Philos. Soc., vol. 23, p. 197 (part); 1886, p. 266-267; 1887, Bull. 32, U.S. Nat. Mus., p. 34; 1900, Rep. U.S. Nat. Mus. for 1898, p. 237-240.—DitTmars, 1907, Rept. Book, p. 107.—STEJNEGER and Barsoour, 1917, Check-List N. Amer. Amph. Rept., ed. 1, p. 44.—Van DENBURGH, 1922, Occ. Papers California Acad. Sci., No. 10, vol. 1, Lizards, p. 64-66.—STEJNEGER and BarBour, 1923, Check-List, N. Amer. Amph. Rept., ed. 2, p. 42. Type.—Brit. Mus: Nat. Hist. No. XXII 20-a, Female. Type locality Restricted to Tampico, Tamaulipas, Mexico. ArT. 12 REVISION OF LIZARDS OF GENUS CTENOSAURA—BAILEY 11 Diagnosis.—A conspicuous transverse gular fold; median dorsal scales 65-80, considerably larger than body scales forming a serrated crest of slight elevation extending from the beginning of the neck to the sacrum; dorsal crest not even indicated by a row of carinated scales in sacral region. Tail very long and strongly marked into numerous verticilli (whorls or rmgs), composed of very long and very strongly carinated scales, each terminating in a lengthened point, thereby causing a spiny appearance throughout entire length of tail; whorls of spinous scales separated from each other by rows of smaller flat scales; first and second or first, second, and third whorls separated. by three rows of smaller flat scales; next five or six whorls separated by two rows of smaller flat scales; a few whorls separated by one row of flat scales, these flat scales gradually becoming spinous until at or near middle of tail small flat scales disappear and tail exhibits a spiny appearance to end. (Of course a broken tail that has been regener- ated does not possess the armed scales on the regenerated portion.) Distribution —This ctenosaur has the widest distribution of any member of the genus. It ranges from the States of Sonora and Chihuahua, Mexico, southward to the Isthmus of Tehuantepec, inhabiting sandy beaches and the foothills of the various mountain ranges. Most of the specimens that have been collected have been taken on the coastal slopes of the mountain ranges, very few being recorded from the interior regions. Specimens taken on the islands in the Gulf of California and at Cape St. Lucas, Lower California, were in all probability carried over from the mainland. These lizards are regarded as food by some Indians and are often carried alive from place to place for that purpose. Many specimens in the museums in this country and in Europe bear simply the locality label “‘ Mexico.”” However, enough properly labeled material has been examined to insure accurate distribution charts. Specimens have been taken at Batopilas, Chihuahua, Mex- ico; on the western foothills of the Sierra Tarahumare Mountains; Tampico and Manuel, Tamaulipas; Miramar, Cerro del Gallo, Jalapa, and Panuco, Vera Cruz; Escuinapa and Tres Marias Island, Nayarit; Uruapan, Michoacan; Tlopa, Guerrero; Tetela, Morelos; Tehuantepec, Dominguillo (Domingville); and Cuicatlan, Oaxaca. Description.—Brit. Mus. Nat. Hist. 20a. H. grown female type; Berlin No. 577, H. grown male now M. C. Z. No. 22453, cotype of cycluroides; M. C. Z. No. 16070 adult male. Head elongate, flat above, covered with somewhat small hexagonal scales, and very dis- tinctly marked off, as it were, from body; muzzle narrowed, covered with rather large amoun scales; supraoculars small, flat, aud hexag- onal, externals only about cesHett as large as ehenals, chal separated om each other by three rows of scales; ear opening almost as large as orbit; no dewlap; transverse gular fold present; parietal scales slightly 1? PROCEEDINGS OF THE NATIONAL MUSEUM VOL. 73 smaller than those on muzzle; nostrils large, very near tip of snout, almost tubular, opening obliquely backward; lores flat; 9-11 enlarged supralabials; 8-10 enlarged sublabials. Dorsal scales small, hardly more than half size of ventral scales, gradually increasing in size poste- riorly, smooth; a well-developed dorsal crest composed of from 65-80 carinated scales, beginning just back of head, on neck, and continuing; uninterrupted, to sacrum; in large, old males these spines are cone- like and often reach a height of 8-15 mm. Dorsal crest and caudal crest entirely separate, there being no indication of crest in sacral region. No spines on any scales of fore or hind limbs; femoral pores vary from 4-4 to 9-9. Tail slightly constricted at insertion, rounded posteriorly, at least twice as long as body in unmutilated specimens; caudal scales above and laterally, in whorls, large, spinous; whorls separated by smaller flat scales, of which the median dorsal are spi- nous throughout length of tail; first and second or first, second, and third whorls of spinous scales separated by 3 rows of small flat basal scales; next 10 or 12 whorls of spinous scales separated by 2 rows of small flat scales; other whorls separated by only 1 row of flat scales, which about middle of tail, also become spinous, thereby giving distal half of tail a spiny appearance throughout. At base of tail, ventral scales much smaller, three rows corresponding to each pair above, slightly keeled and pointed posteriorly. After first 3 or 4 rows ventrals and dorsals approach each other in size, 2 rows of ventrals corresponding to a like number of dorsals. Toes rather long, espe- cially those of hind feet; claws strong and sharp. Measurements.— Cotype of Ct. cyclu- Brit. Mus. | roides Ber- |Large acne Type No. 20a, F |/lin No. 577,;,M M.C. Z type M now M. No. 16074 C. Z. No. 22453 Mm. Mm. Mm. Length of Heads-3--2- 3. pete d pea. 1 Opa ea Seah oe 38 40 50 MON Sth Ol DOG Vs see eee a ee eee eee ee 130 130 165 Length of tail_- eens ie fae Ee ae ee oe es eee ee ree Sea 289 310 430 Total denghh ssc ee te eee. ene eee ee een. = None 455 480 653 Width of Head GVeLORDITScee lee eae gels EDL Cain oe 22 22 28 Coloration —Adult: Head, neck, body, tail, and limbs dull brown- ish above; under parts lighter with somewhat indistinct clouds and marblings of a whitish cast. On belly and sides are three or four bands of faded slate or bluish green which extend up and across back, being hardly visible except where dorsal crest spines are involved. A few very large males exhibit blotches of rusty red or cinnamon over body, especially on sides and shoulders. Large adults of this species are often referred to as ‘‘ Black Ctenosaurs.”’ It seems that dried skins lose most of their color, so great weight should not be given to descriptions made from such specimens. ART. 12 REVISION OF LIZARDS OF GENUS CTENOSAURA—BAILEY 13 A few color descriptions representing the observations of different students on various sized specimens, under different conditions, may be of interest, hence the following notes: Shaw in his original description of Ctenosaura (Lacerta) acanthura, based upon an alcoholic specimen, which wasnot more than half grown, says:* “‘Upper part glaucous, variegated with a few small and some- what indistinct clouds and marblings of a whitish cast. The tail and underparts are of a pale or yellowish color.” Harlan, who in 1824 described and figured Ctenosaura (Cyclura) teres* from a living specimen in the Museum of the Philadelphia Academy of Natural Science, gives us this description: ‘‘Color of this species dark green, on some parts of his back brilliant or glistening.’’ Although there is no record of this specimen ever having been pre- served, the description together with the splendid illustration of the specimen leaves no doubt as to its identity with Ctenosaura acanthura. In writing of the color of the young of this species Wiegmann ® says: “The color of the upper parts in this young specimen is a splendid yellowish green intermingled with bluish green and cloudy black-brown cross spots; three brown cross stripes go over the cheeks to the ear; the legs are sprinkled with numerous spots and the tail is ringed with brown. But this uncommon, beautiful coloring seems to disappear with advancing age.’”’ He also says concerning older specimen that ‘“A somewhat larger example shows faded bluish-green, or rather a green and blue glittering gray sprinkled with numberless black dots. Upon them no traces of other marks are left visible.” } Cope,® who described Ctenosaura multispinis, a synonym of acan- thura, from a full-grown male, says: “‘Color above and below black.” The writer examined this type of multispinis, which is a stuffed skin, and found it to be a true acanthura and that the underparts showed indistinct whitish markings, just as do most of the larger specimen. Gunther’ says: The coloration varies and changes with age. The ground color of the young is generally green, marbled with darker on the back, the dark markings forming more or less distinct, irregular cross bands, which are sometimes confluent, some- times spotted with black, and about seven or eight in number on the back. With age the dark color becomes more diffused and irregularly distributed over the body, at places entirely suppressing the ground color, which itself assumes a more olive tinge or changes into yellowish. Specimens from Tampico are uniform black when adult, and of a greenish-olive when young. Remarks.— Originally Ctenosaura acanthura was described by Shaw in 1802 from a half-grown female specimen, the date and place of collection and the name of the collector being unknown. 3Shaw, George. General Zodlogy, vol. 8, part 1, p. 216, 1802. ‘Harlan, R. Journ. Acad. Nat. Sci. Philadelphia, vol. 4. pp. 242-251, pl. 26, 1824. ’ Wiegmann, V. J. Oken’s Isis., p. 371, 1828. ®Cope, E.D. Proc. Amer. Philos. Soc. Phila., vol. 23, p. 267, 1886. *Gunther, A.C. L.G. Biol. Cent. Amer. Rept. Batr., p. 57, 1890. 14 PROCEEDINGS OF THE NATIONAL MUSEUM VOL. 73 In the spring of 1824 a living specimen was brought from Tampico, Tamaulipas, Mexico, by Captain Dallas, and presented to the Acad- emy of Natural History, Philadelphia, where it remained alive for several months. Mr. Harlan of the Academy Museum observed this lizard for several months, and in November, 1824, published a description of the specimen and notes on its habits in captivity. The plate accompanying the description makes it very clear that this species, which he called Cyclura teres, is in reality only an adult of Ctenosaura acanthura. With this evidence in hand and with records of many additional findings of this species in the Tampico district, I hereby restrict the type locality of Ctenosaura acanthura to Tampico, Tamaulipas, Mexico. Wiegmann published in 1828 an account of a new species, Ctenosaura cycluroides,> based upon three specimens collected by Deppe in ‘“Mexico,”’ the same year, and deposited in the Zoologische Museum at Berlin,Germany. He created the genus Ctenosaura at this time. His specimens were kept together as cotypes, No. 577, a male, and Nos. 576 and 578, females. All were the same size and not over one-third grown. No. 577, a cotype, is now in possession of the Museum of Comparative Zoédlogy (M.C. Z. No. 2253), Cambridge, Mass., received in exchange. The three specimens are certainly Cienosaura acanthura. A few years after describing Ctenosaura cycluroides, Wiegmann, for some unknown reason, decided to rede- scribe these same specimens. Accordingly he gave up the genus Ctenosaura, that he had created in 1828, went back to the old genus Oyclura and redescribed them in 1834 as Cyclura denticulata,® using specimen 578 asthetype. Indescribing Cyclura denticulata Wiegmann even lists Ctenosaura cycluroides as a synonym, but he assigns no reason for putting away the original name. Perhaps he liked the new name better. At any rate the types of both species were the same individ- uals, in the same containers and they bore the same accession numbers, locality labels, and collector’s name, all in the handwriting of Wieg- mann himself. How does the writer know these facts? Because in examining the jar containing specimen number 577 the large printed label bearing thename Cyclura denticulata accidently became saturated with alcohol and water and slipped down the side of the jar, thereby exposing the original label. The other jars were treated similarly and yielded like results. A check-up on the specimens, with the curator of the department in question, revealed the fact that the specimens were the same individuals, only the original accession numbers of Ctenosaura cycluroides being listed. In the same publication Wiegmann described a new species, Cyclura articulata, giving as a synonym Iguana (Ctenosaura) armata, which was described by Gray in 1831,! and which is Ctenosaura acanthura of 8 Wiegmann, V.J. Oken’s Isis., p. 371, 1828. 9 Wiegmann, V. J. Herptologica Mexicana, pp. 43-44, 1834. 10 Gray, (Cuvier) Griffith’s Animal Kingdom, vol. 4, p. 38, 1831. art.12 REVISION OF LIZARDS OF GENUS CTENOSAURA—BAILEY 15 Shaw. Again he assigns no reason for changing the name of a species. Fortunately, however, neither of his last two species were ever recorded as distinct, the “law of priority,”’ although not known as such at that time, having taken care of the situation. In 1886 Cope published the description of Ctenosaura multispinis,'!! based upon an adult male dried skin from Dondominguillo (Dondom- ingville), Oaxaca, Mexico. A careful examination of this specimen and comparison with others indicates very conclusively that it is merely a “‘dark phase,’’ or mature individual of acanthura. As a matter of fact the large specimens of acanthura are commonly known as black ctenosaurs. Perhaps the most interesting observations made on this species are recorded by Ditmars.” He says: The old lizards are generally uniform jet black with marblings of olive or even exhibiting reddish blotches. They are surly brutes, immediately showing fight when cornered, not only endeavoring to bite, but dealing ugly blows with the gener- ously spiked tail. From painful experience the writer (Ditmars) can testify that a blow from the spiny tail is capable of producing a severe laceration. If an avenue of escape is open, most specimens prefer flight to combat. If discovered while sunning in their favorite position, on top of a rock in a forest opening, the creature hurls himself into the shrubbery making as much noise as a frightened cow, as it goes away to a considerable distance. This species is not much in the habit of ascending trees; it can, however, climb fairly well. On the ground it is very fleet, running with the body high, the tail slightly elevated. A strong lizard can easily outrun a man as to speed, invariably escaping by darting into a thicket. Very young specimens are uniform, bright emerald green. They are persistently terrestrial, running on their hind legs in kangaroolike fashion when frightened. Observations made in large yards with a number of species of lizards, however, have demonstrated to the writer that the habit is prevalent among many of the long-bodied lacertilians of both the Agamidae and the Iguanidae. He has thus far noted the habit among 10 genera. It seems probable we have here a heredi- tary character, handed down from gigantic reptiles of the past, for several of those creatures, now known only by the ponderous fossils imbedded under mountains of rock, were constructed to stalk about on their powerful hind legs. In its natural environment acanthura is thought of as being strictly vegetarian in its diet, but the dissection of many stomachs shows that it also is very fond of insects. Harlan” observed that a specimen living in the Philadelphia Museum for several months ate nothing of its own accord, but that when raw meat or fruit was placed in its mouth, would swallow it leisurely without chewing. He showed a preference for raw meat, and always rejected cooked meat. During the summer the speci- men subsisted chiefly on fruit and was never observed to drink. During the autumn (November, 1824) he became considerably torpid, remaining in one position for hours, without any disposition to move unless roused, when he displayed considerable activity. He became =. Cope, E. D. Proc. Amer. Philos. Soc. Phila., vol. 23, p. 267, 1886. 2 Ditmars, R. L. The Reptiles of the World, p. 141, 1910. = 13 Harlan, R. Journ. Acad. Nat. Sci. Philadelphia, vol. 4, pp. 242-251, 1824. 16 PROCEEDINGS OF THE NATIONAL MUSEUM VOL. 73 exceedingly tame, and was fond of being washed with a wet sponge.» He showed no disposition to bite, but when teased or tickled on the leg defended himself with his prickly tail, with which he was able to strike in every direction. Material examined.— Specimen Sex Age Locality Date Collector Remarks Brit. Mus.: | XXII 20a_| F. | Half grown_| ‘‘Mexico’’_-..---- Prior top1802-\a(?) ie. Soe Type. oo) AGUb see Tampico, Tamau- | Nov. 13, 1889-| Richardson ----- lipas, Mexico. 63 ye ee doe==— Tres Marias, Na- | Oct. 1, 1881-__| Forrer ---.------ yarit, Mexico. PAN | Eee doe eae 0:53.43 os ee GOf 222222): 2 dole eee 3F. | Half grown. weleley Morelos, | Sept. 30, 1903_| Dr. H. Gadow_- exico. 2M. Aduits=— = a2 o 2 d0s22teeesetes Apr. 11, 1866_| Purchase ------- Berlin Mus.: BiG Lae ee F. | Half grown_| Mexico------------ LS2SeIPEN ss 8 Wa Deppe--ns.=- Cone Cyclu- roides. Dione Mis Sa Goysacno hates (fey eee TS2R eer Se an (6 Ko} en eS Do. b78a22842 1 ie eee doy 2t Sie doli-tecs tis ASD Gets Fane Pater: dons s Cotype Cyclu- roides and type denticu- lata. A.M.N.H.: A540 eee et | Sasce sees do ...._| Guerrero, Mexico-| (?) ----.----.- Nevins= 2° 22225 Us ees Se es eee dom anes Escuinapa, Sina- } (?) ----------- Joe Battysooe- loa, Mexico. g UR ee ee Bes C5 (a eae pega do 3.82 2-28 (Up Sean e ee oe doz. -2s2t3 1595-25-25 Wess Woungessoe|-oaes G0e= ces ee = (G2) SaRe eae | Roe (6 Voy ee 10 specimens. U.S.N.M.: ; 610325 22k Mie pete 4 dojaeas= Mirador, svieran|) (esses. -sse. Dr. C. Sartonius_ Cruz, Mexico. 10234 Soon k |e dose Uruapax, Micho- | 1879.--.--.--- Professor Duges.. acan, Mexico. 20168-72__.| .----- Half grown.| Tehuantepec, | Aug. 29, 1892_) P. L. Jouy.--.-- Oaxaca, Mexico. DAZG IK 2 ate cea et eee 22 = ee IsabelIsland, Tres | Apr. 23, 1897.| E. W. Nelson--- Marias Island, ; Mexico. GSAT REE Ne i ss ee Panuco River, |i@)e- sta e- WsOdellAue about 80 miles above Tampico, Mexico. 30430__.... F. | Half grown.| Tehuantepec, | (?) ----------- F. Sumichrast__- Oaxaca, Mexico. 46835 ------ 1 an |e doe 5 Cuicatlan, Oaxaca, | Oct. 12, 1894-| Nelson and Mexico. Goldman. 46860------ yy | pvounge.-= Tlapa, (Guerrero, | Dec. 2, 1894--|____- GO+sias22h-y Mexico. A194 2-2 223 Miso do aa2e Cuicatlan, Oaxaca, | Oct. 9, 1894--}.....- OCA e Lek Mexico. 8137-2 ee woaetl ao poe Balsass GuerEero.s| LOOl ese seoce Jip EUUTtOr oes Mexico. 68498_____- M. | Half grown_.| Tehauntepec, | 1905--__------|.._.-. GOs -cegcces Oaxaca, Mexico. 1634se eos Mie occas G0 z2s=— Isabel Island, Na- | (?) ----------- JoR. slevin=ssos yarit, Mexico. W163p24=—— = 1a es a 0 1 eee eee don <3-5 can eieban US=,| 19225 ee ose Jee aS do ee and. Brit. Mus-_-_..- M. | Halfgrown-| Lower California, | (?)_-----. ---- M. Botta__---..-| Cotype of Ct. Mexico. interrupta. Miss |zcese (6 ae | oe Se C6 Koester (Gi) ease ee eee (( 0 ernie Se Paris Mus p27 ean eee AOU Gsaee | ee Ors ee See (QRS ee Mi BOvraseen eee Do. ODT SE Se eee eee) GO a eres (oo) eee a ae (be = ees eee doe Do. 2290s ee ae aoeee oes doeeeee MIe@xiGOl2=- ese VSO ose oak (2) see pS eee eee el ae dose Lower California, |) \(?)s2--2-==-—— M.. Botta-_-=2--2 Do. Mexico. 96-120__.__ Bus | ete ees doe see On ae ees (G() Ree Se ee Diguevies---2see— 96-121____- IMT |soe GO seecl seman do === WO) celesae sase|oneo (G10) ea eemesos | OV—438e5 Ja|2.2 2-5 Half grown-!_-.-- doje" 5: Se (@) cine mc tees Co (eyes eee CTENOSAURA BRACHYLOPHA (Cope) Plate 6 Ctenosaura teres brachylopha Cop#, 1886, Proc. Amer. Philos. Soc., vol. 23, p. 269; 1887, Bull. 32, U. S. Nat. Mus., p. 24. Cotypes.—Cat. Nos. 7180, 7181, 7182, 7183, U.S.N.M. Females. Type locality.—M azatlan, Sinaloa, Mexico, 1867; Bischoff, collector. art.12 REVISION OF LIZARDS OF GENUS CTENOSAURA—BAILEY 23 Diagnosis.—This species resembles somewhat Ctenosaura pectinata and brevirostris, but may be distinguished from either of them by the absence of the median dorsal crest over the sacral region. Dorsal crest made up of 65 to 75 short processes and extends only to begin- ning of sacral region. First, second, third, fourth, fifth (and occa- sionally the sixth) caudal whorls of spinous scales separated by three rows of small flat basal scales. Distribution.—This species is found on the low coastal plains of western central Mexico. It has been collected in the States of Nayarit and Sinaloa, and it probably occurs also in Jalisco. The type specimens were collected by Bischoff near Mazatlan, in Sinaloa. Other specimens have been collected at Culiacan, Sinaloa, Tepic, San Blas, Maria Madre Island, Tres Marias Island, and Maria Cleofas Island, Nayarit. Description.—U.S.N.M. Nos. 7180, adult female stuffed skm, 24630 adult male alcoholic specimen. Head normal in length, covered with small hexagonal scales; muzzle slightly decurved. Scales on muzzle larger than other head scales; supraocular small, being separated from each other by three or four rows of scales; nostrils large, much nearer tip of snout than to orbit, almost tubular, opening obliquely backward; rostral larger than mental; lores flat; 10 to 13 enlarged supralabials; 10 to 12 enlarged sublabials; ear opening as large or almost as large as orbit. Dorsal scales small and smooth, hardly more than half the size of ventrals, being almost granular on neck and gradually increasing in size posteriorly. Dorsal crest made up of 65-75 very short processes, which appear as merely elongated com- pressed scales, longer than high, except on interscapular region, where they are as high as long. Three scales on the canthus rostralis, of which the posterior is longer than deep, second deeper than long, and third, adjacent to nares is deeper than long, and divided into a super- ior and an inferior plate; transverse gular fold present; scales on fore and hind legs not spinous. Caudal scales above and laterally in whorls of spinous and flat scales; first seven whorls of spinous scales being separated from each other by three rows of small flat basal scales; next seven whorls by two rows of flat scales and the remainder by one row which itself finally become spinous, giving the distal third of tail a completely spinous appearance. Lower surface of tail covered with transverse series of smaller scales, strongly keeled and pointed posteriorly. Femoral pores 6-6 to 8-8. Measurements.— Cotype U.S.N.M. No. 7180, female. SPE PCAC oat 8 eT oe SORE oe eet ee eile bh ees 45 mm. 2 EVE OY EL 6 b fe eee a es Yn a Se a Nee OER 165 mm. RINSE HRT 2 tea ome tie ered henner eR tee ede Ss oe 310 mm, MERCRTIe See enero yr ae eek ee see nee ae: eC 520 mm. Breadth of head over orbits_.___.___._____-- Ag ca hd Lae ha Phe Re bal 23 mm 94 PROCEEDINGS OF THE NATIONAL MUSEUM VOL. 73 Coloration.—Cope, in his original description, says: ‘‘The color is apparently green in life, punctuated with blackish brown. The punc- tulations arrange themselves into a row of median dorsal spots, and in three of the specimens into transverse bands near the middle of the sides of the abdomen. Tail with broad blackish rings.” Remarks.—This species grows to be as large as any of the Cteno- | saurs, a large male specimen, from Cleofas Island, measuring over 1,000 millimeters from tip of rostrum to tip of tail. The food of this species consists largely of the leaves of trees and smaller plants. Dr. E. W. Nelson,”! who collected extensively in Mexico, found that this species was ‘‘rather common, living mainly in hollow trees and in brushy places.”’ The highest elevation at which he collected the species was 500 feet. Material examined.— Specimen Sex Age Locality Date Collector Remarks | U.S.N.M.: 7180S2E2 S22 WE i) Adult: 22222 Mazatlan, Sina- | 1867-------... Bischoff... -__- Cotype. loa, Mexico. VA8U. 23223 a Ope || Bee Oe 2 S2|ELEEe Goze Loree ee TS 8672-342. ee doi sc2ih sek Do (ilps ee ee Bie eh eee do 222 24/4 2 Gore: oes soe W867 2 = se s| sa d0\22222.5505 Do TABS ae ny 6) SS do Sela dose nse 1672 A Ras doishdam Do 140782252-2 i. || Young =--= Tres Miarias, Ts-l) (22-22. 252-4 AG Morrers 222285 lands, Nayarit, Mexico. 24623 52222 F. | Half grown_| Maria Madre Is- | May 14, 1897_| Nelson and land, Nayarit, Goldman. Mexico. - 24624____.- F, SY(OnIn yen eer Ope ees ee May 15, 1897_|____- 0 e225 24625___._- 1 Et ees dosen Ley doe ea See - pe Cae (See dois As: 24626 sree 1 eee Co Vo preeare | PSEC Oe eee) hd Oe a eS do ewe 24628. .-__- M. | Half grown_!___-- Gite ies ees May 24, 1s97%21272 3 Gowers 24620: 1 Bala NCo Nt Vie ee Se GOS a | ees doess- 22 ae oe GO 222- 2255 PAOSOL2o—oe VEE doz i233 Maria Cleofas Is- | May 29, 1897-.|..._- oo eee LB aE a Very large lands, Nayarit, Mexico. SI28pesse= Mi; |) Younp 22223 (9 a See ee ae (2) cee Boucard .......- 47956_--.-- M A@ulti: 23! Sinaloa, Mexico.__| Apr. 4, 1899_.| Nelson and oar 47957 ------ F. | Half grown_|_-.-- Goes sees s a> AST oe do Lisi 22| 2b doles S1407=— = 1 eee doesce2 | San Blas, Naya- | 1913----.--... J.C. echGn Dae rit, Mexico. DSioa=saees Ly ese ce dowsse2 Tepic, Nayarit, | October, 1897.| J. Hurter__------ : Mexico. 58753-_----- Ete | esOoung = — = Sinaloa, Mexico___|_...- dorzcesse eee GO tess 60988 _ ._.-- Mi +| Half'grownil) (?)22 20 2 SSE (() ene eae R. W. Shufeldt- 65138. =-=-4 Vie Sy Oung ess Tepic, Nayarit, | 1897-.--...... Nelson and Mexico. Goldman. 70665 -- == -2 F. | Halfgrown_) Culiacan, Sina- | Mar. 27,1926... H. Meerschiedt- loa, Mexico. 1 CTENOSAURA PECTINATA (Wiegmann) Plates 7,8, 9, 10, 11 Cyclura pectinata WimGMANN, 1834, Herpt. Mex., pl. 2, p. 42.—DumérIL et Brsron, 1837, Erpét. Gén., vol. 4, pp. 217-221.—Firz1ncEr, 1843, Syst. Rept., p. 56.—Cops, 1886, Proc. Acad. Nat. Sci. Philadelphia, p. 124.—Bocourt, 1870, Miss. Scien. Mex., vol. 3, Reptiles, p. 140.—Coprzr, 1871, Proc. Acad. Nat. Sci. Philadelphia, p. 216.—GarMan, 1884, Bull. Essex Institute, vol. 16, p. 19.— 21Nelson, E. W., Chief of the U. S. Bureau of Biological Survey, supplied his field notes on this genus. pai ART. 12 REVISION OF LIZARDS OF GENUS CTENOSAURA—BAILEY PAR) Cops, 1885, Proc. Amer. Philos. Soc., vol. 22, pp. 379-388. Ctenosaura pectinata GRAY, 1845, Cat. Lizards Brit. Mus., pp. 191.—SumicuHrast, 1880, Bull. Soc. Zool. France, vol. 5, p. 174. Ctenosaura acanthura BOULENGER, 1885, (Part) Cat. Lizards Brit. Mus., vol. 2, p. 197.—GinruER, 1890, Biol. Cent. Amer., pp. 56-57, pl. 30. Type.—Berlin Museum No. 574, male. Type locality.—Restricted to Colima, Colima, Mexico. Diagnosis —This species is very near Ct. brevirostris, sumilis and parkeri, having in common with them the dorsal crest: extending to the base of the tail, uninterrupted in the sacral region, but differing from each of them in one or more important characters. From brevirostris it differs in the length of the rostrum or muzzle; the rostrum of brevi- rostris being short and decurved while in pectinata it is elongate and not decurved in a pronounced manner. The arrangements of the caudal scales in pectinata and brevirostris are essentially the same; the first five whorls of spinous scales being separated from each other by three rows of smal! flat scales; the remaining whorls being separated by two rows of small scales for a short distance, then by one row which gradually becomes spinous and similar to the other caudal scales. But in st¢milis only the first and second (and occasionally the third) whorls of spinous scales are separated from each other by three rows of small flat scales, the subsequent whorls of spinous scales being separated by two rows of flat scales up to about the middle of the length of the tail, then by one row of flat scales which gradually become spinous and similar to the other caudal scales just as in the other related species. In parkeri the first seven whorls of spinous scales are separated from each other by four rows of smaller flat scales. Distribution. —Ctenosaura pectinata occurs on the west coast of Mex- ico from the State of Nayarit southward to Oaxaca. Collections have been made at San Blas, Maria Madre Islands, and Isabel Island, Nayarit; Colima City, and Mount Colima, Colima; Balsas and Acapulco, Guerrero; and San Geronimo, Oaxaca. Description.—Berlin, type, No. 574, adult male; M.C.Z. 2726, adult male and female; 6982, adult female; A.M.N.H. 119, adult female. Head elongate, flat above, covered with small hexagonal scales very distinctly marked off from body. Scales on muzzle smooth and somewhat larger than other head scales; supraoculars small, flat- tened, and hexagonal, externals being only about one-half as large as internals and separated from each other by a row of four scales. Ear opening almost as large as orbit; no dewlap, but a pronounced transverse gular fold present; nostrils large, very near tip of snout, almost tubular, opening obliquely backward; lores flat; supralabials, 26 PROCEEDINGS OF THE NATIONAL MUSEUM VoL. 73 12; sublabials, 14. Dorsal scales small, hardly more than half the size of ventrals, gradually increasing in size posteriorly, smooth; a well-developed dorsal crest composed of from 63 to 75 carinated scales extending from insertion of neck caudad to base of tail, being continuous over sacral region as true carinated scales, but much reduced in height. In old males these dorsal spines reach a height of 10 mm. Leg scales without spines. Femoral pores vary from 5-5 to 7-7. Toes very long, especially those of hind feet; claws strong and sharp. Tail nearly cylindrical, scales on upper side being of two kinds, large and spinous, and small, low, flat scales arranged in whorls. Median dorsal scales are large and heavily armed through- out length of tail. The others are arranged in whorls; some whorls are spinous and some are flat and smooth. First 5 whorls of large spinous scales separated from each other by 3 rows of small flat scales; next 9 or 10 whorls of large spinous scales separated from each other by 2 rows of small flat scales, while on terminal three- fifths of tail all of scales gradually become equally spinous. At base of tail ventral scales are smaller than dorsals, four rows of ventrals corresponding to three above, slightly keeled and pointed posteriorly. Measurements.— Berlin Mu-| 17 .¢.Z.M.| M.C.Z.F. seum, type No. 574 M. No. 2726 No. 6982 Jhenephvotshead a = 2u baie oe: Se oe eee eee 65 85 70 engthofbod ywew2s 2a Sete hae dS A eT ae H 205 0 210 Ger St vOl tail soe eae eo ee ec Fae eee ee ee ee noe | 1405 1180 455 *hotahlengt hot. = 2222 £ th Pe OES a aah SES ee 1660 1485 735 Breadthiof head over orbits...) - 22 eS Re os aoe eased 28 32 30 1 Tail broken off. Coloration.—General body color is brown-olivaceous streaked with yellow. Dorsal spines are yellow wherever the yellow markings cross the mid-back. Upper portion of head is brown, lores yellowish. Except for two small transverse brown bands the lower maxillae is yellowish. Neck brown, with rather long yellow bar running caudad from posterior margin of tympanum, vanishing slightly caudad and above axilla of arm. Abdomen yellowish olive girdled by three brown (sometimes broken) bands. Breast brown; limbs brown with yellow marks and spots. Tail ringed with alternate, wide bands of brown and yellow. Remarks—Wiegmann described this species from a male specimen collected by F. Deppe in ‘‘Mexico.’”’ Many specimens, ranging in age and size from very young to adults, including both sexes, have been examined and found to agree with the type in all essential characteristics. azt.12 REVISION OF LIZARDS OF GENUS CTENOSAURA—BAILEY 27 Material examined.— Specimen Sex Age Locality Date Collector Remarks Berlin: Ey gs eee VE PAG ult 4-2: NVI ORI CO! a BP ae he eee ne ets i. Deppess2-- 22 Type. A.M.N.H 3 ha ht ee ee 1 ey dozer Colima, Mexico___} 1901_-________ L. Diguet_-__..-- 117, 18s Ss ase Cap a ae, GU} Seats iI.) Lee | ate (G\pR eee 4) are 1 IS g| Pees eee ee ed Mount! Colima | 1g0te ss sees ee Goeys: 23h Mexico. U.S.N.M.:: AG) ase! Heeb evoqunges2a Maria Madre Is- | May 15, 1897_| E. W. Nelson___ land, Nayarit, Mexico. PAG33 == 25 3 F. | Half grown_| Isabel Island, Na- | May, 1897____|____- doses Se yarit, Mexico. A (20 Sa so ee) Young.-.___ Prextla, Puebla, | 1894..-.--._.- Nelson and Mexico. Goldman. 47919___..- Mie |e eeeeie cae Rio Balsas, Guer- | June 4, 1903__}_____ Goss fen the rero, Mexico. 47920); 2.2 LOPS (ae eee ee ears ED ae Feral ee GotAs. 2 slo (0.2. 282% a140 222 oo He) Moungs-- = Topic Nayarit, | Dec. 26, 1913-.| J. C. Thompson- exico. 61403_...-- 1 Sel ee dose San Blas, Na-| Dec. 20, 1913_|_..-. (6 (0) Saeeeee 8 yarit, Mexico. “aii LEE Nee VES ee (0 (Sap Soi | ns (6 ORG Renan rey fg ay (Ge Sool ae See Gores = sea== = 51405-.=~-* 1g See! Wot S587 Goyievee. eseer ee oe dope sess 2582 (cK See Ses Z 51406_____- Pe edon eee. doy Seen ee don eee ig eer Aeboo-Osses|| Mog eeeee doz +8 Tres Marias, Na- | 1885____-_---_- Avi orreric 2 3 2 specimens. yarit, Mexico. Cay ee ee 1 OS | (ee as Gomes = seeses dozs eee pet 2 TSS5 sass See heer ady) dose eS M.C.Z.: | 157s see 'M.&! Half grown.) Colima, Mexico-_--| 1914----_---_- G. Gluckert-___- 8 specimens. eae USO Re Paes oe ll Pete eed _ Acapulco, Guer- | 1872_-......_.| L. Agassiz.-..-. Hassler expe- | rero, Mexico. dition. 2726-..-.._| Wt AGiitse2 2 2e2 3 does eee ASTAr ze =e 32 De F. Heidac- | 2 specimens. ein. pee Teel 05 ¢|PaSee- AO zal sek | 2 doses nace? US7A, wire te SBS doz as. ss2: Do 6982. - 1 ham, | er, Woes San Geronimo, | 1902____-__--- (9) es a Oaxaca, Mexico. t CTENOSAURA BREVIROSTRIS (Cope) Plates 12, 13, 15 Ctenosaura brevirostris CopE, 1886, Proc. Amer. Philos. Soc., vol. 23, pp. 266-268; 1887, Bull. 32, U. S. Nat. Mus., p. 34; 1900, Rep. U. S. Nat. Mus. for 1898> . 238. Ee ini acanthura GUNTHER, 1890, Biol. Cent. Amer., Rept. Batr., p. 57, (Part). Type—Cat. No. 24709, U.S.N.M., male. Type locality—Colima, Colima, Mexico, John Xantus, collector. Diagnosis.—This species is very similar to Ct. pectinata, but may be distinguished from it by the very short head with an obtuse muzzle, exhibiting a pronounced decurved profile. Distribution —This species occurs on the Pacific foothills of the mountain ranges from Jalisco southward to Oaxaca. The type was taken at Colima City; others have been taken at San Marcos, Jalisco; Manzanillo, Colima; Sierra Madre, Michoacan; and Guichicovi, Oaxaca. Over 90 specimens were collected at Colima by John Xantus. Description.—U.S.N.M. Nos. 24708, adult female; 24709, half grown male cotype; 47933, adult male. The following description is from Cope’s original, with modifications according to the writer’s observa- tions of the types. 28 PROCEEDINGS OF THE NATIONAL MUSEUM VoL. 73 - Head very short with obtuse muzzle, with pronounced decurved profile. Eyes large, nostril near end of muzzle, in anterior third of distance between end of muzzle and orbit. Scales on top of muzzle and of frontal region subquadrate or subbexagonal, those of temporal regions but little longer than wide. All are more or less convex, temporals more so; rostral plate larger than mental. Six rows of scales between nasal plates, some of which are wider thanlong. Three: canthal scales, of which the anterior is horizontally divided in one specimen. Four rows of wide loral scales above four rows of narrow scales above the supralabials. Supralabials, 11-12; sublabials, 14-15; loreals flat. Scales on muzzle larger than parietals; supraorbitals smaller than other head scales and separated from each other by four rows of scales. Dorsal scales small, hardly more than half size of ventrals, gradually increasing in size posteriorly, smooth. Dorsal crest composed of from 75 to 80 carinated scales, beginning just back of head, on neck, and continuing uninterrupted at sacrum, to base of tail. Crest over the sacrum is very low, but is present as raised carinated scales, thereby maintaining median row of raised dorsal spines or crest. In female specimens entire crest is much lower than that of male; those of adult males reaching a height of 5 to 8 mm. Limbs are without heavy spinous scales. Tail is nearly cylindrical, scales on upper side being of two kinds; large spinous scales and small, low, flat scales, arranged in whorls. Median dorsal scales are large and heavily armed throughout length of tail; others are arranged in whorls; some whorls spinous and some flat. In one specimen the first 5 whorls of large spinous scales are separated from each other by 3 rows of small, flat scales; next 9 or 10 whorls of large spinous scales by 2 rows of small, flat scales; while on the terminal three-fifths of tail all of the scales gradually become equally spinous. At base of tail ventral scales are smaller, four rows corresponding to three above, slightly keeled and pointed posteriorly. Toes very long, especially those of hind feet; claws strong and sharp. In both specimens femoral pores are small, exceedingly so in female, which has six pores on each femur. Male has five pores on each femur. Both have distinct transverse gular fold. Measurements.— ype Cotype U.S.N.M. | U.S.N.M, | (S-N-M. . No M., No 47933 : 24708 24709 Mm Mm. Mm Length of head. -< 2. 3s. ece Aer eka ke pan aU Se eel ee eet aks Se 45 40 55 Hengthof bod ya5 25 5 ek eee aE Se ae oe Es 197 150 195 eneth Of tary. 2S oo oe a ee ee eee ee 1403 42 5 Totailengthe. sh et esate ae ee el os ye eee Ce eee ee ee OSES 1645 610 775 Width‘of head: over orbitsi2.. 22 22ers enero 30 25 3s 1 Part of tail broken off. - ArT. 12 REVISION OF LIZARDS OF GENUS CTENOSAURA—BAILEY 29 Coloration.—General color of head and body blackish brown, being crossed on back between sacral and postscapular regions by five yellow, marks, which are bands posteriorly but become spots anteri- orly. These bands are more pronounced in females. Sides of neck yellow, contrasting strongly with black of throat andnape. This yel- low space is practically divided by a black line, which extends poste- riorly from angle of lower jaw. Under parts are yellowish, streaked slightly with waves of blackish-brown spots; a yellow stripe beginning at posterior border of tympanum extends caudad over shoulder, fading out slightly above and caudad to axilla of arm. Limbs are blackish, and on fore arms are numerous yellow scales; digits and tail are annulated with blackish-brown and yellow rings of about equal width. Remarks.—This species has approximately the same distribution as Ctenosaura pectinata and on account of the great similarity to it may be mistaken for it. Both species frequent trees, but are more often seen scurrying about on the ground and among the rocks. The chief difference between the two species is to be found in the struc- ture of the head. In brevirostris the head is very short and the muzzle is pronouncedly decurved, while in pectinata the head is long and the muzzle only slightly decurved. Material examined.— Specimen Sex Age Locality Date Collector Remarks U.S.N.M.: AQIGSE 2. a|L ane = Young--.__| Colima, Mexico.__| July 1, 1865__| J. Xantus______. 2 specimens, 12230 53> Ven oe d0)22--2 | Sierra Madre Mi- | (?)_--_.-----.- (7a) A ee See ee | choacan, Mexi- li Ico. 18968_---=. Ma ppAdultces’ — San Marcos, Ja- | Mar. 26, 1892_| P. L. Jouy------ | isco, Mexico. 18969_.____ 1 eee dosas2 joaertt (6 Conan ee Mar. 30, 1892_|..__- dos eas 72: 7/1): en Wat d| ees doses | Colima, Mexico_-} (? Jc Se eed Je eManbiS- os soe Cotype. 24709-6352 M Half grown. |--__- GOn 522 $5 ea25- 1) | SASS BA doje. ae Do DEVAN = Sel ear eas Wounies === ee G0.s-22222oe525 Wee ee Necks Oras {22228 fs 10 specimens. BA PO-o | IVEY tak * Goyssses| 2222 (ic) eee ee ee (? Jaan anna soe dO 222. F534 4 specimens, ALD = 5 F. | Half grown_|____- Coo SS eee (70) IRR eS cote = 38 ue A OS oe Sie Y 31484______ F. Oungtoeo4 2.22 GOt ee as ere ( geese. /2ss2sdoet seek} 47933_.-.=. VE) || AGult==s2.- | Guichicovi, Oax- | June 26, 1895- Mulcon and | aca, Mexico. Goldman. fey (| ae F. | Halfgrown_! Colima, Mexico.__| July, 1902____! JeeHurten oa ese 63701----_- | M. | Young--..- Manzanillo, Coli- | February, | J. Xantus______- ma, Mexico. 1863 63702-31...\{ Me |\__.do____.| Colima, Mexico...| (?)-—-------..|-----d0 --.------- 20 specimens. ; 63732- =. ..-= Peat \flvtyss 2 don-5== Ponalas 7Golimayal:(2)s2- 222 ele | (Pee erae, oer 2 specimens. Mexico. 63733.----- IVE, )2=233 dons. ato = dose taseew J (Fee ee eee 0 (0), See epee 63734-68.__{ UF: \__.do See Colima, Menieou. July, 18635222 (222 do ..._---.-_| 35 specimens. CTENOSAURA PARKERI, new species Plates 14, 15 Type.—Cat. No. 18967, U.S.N.M., adult female, Barranca Ibarra, Jalisco, Mexico. April 22, 1892, P. L. Jouy. Paratypes —Cat. No. 18970, U.S.N.M., a half-grown female having same data; Brit. Mus. No. 1, adult male; No. 75, half-grown male; 30 PROCEEDINGS OF THE NATIONAL MUSEUM VoL. 73 No. 76, adult female. From Tres Marias, Nayarit, Mexico, by M. Forrer. Diagnosis.—Dorsal crest very pronounced, not interrupted at sa- crum, the continuity being effected by short lobes or depressed but enlarged scales. Tail ringed on upper half with whorls of strong and very spinous scales. First six whorls of spinous scales are separated from each other by four rows of smaller flat scales; the next three by three distinct rows and a partial fourth which borders the median dorsal spines of the tail. Subsequent whorls up to half the length of the tail are separated from each other by two rows of small scales. Distal half of tail appears spinose throughout, the smaller scales becoming larger and gradually pass into the spinous type. A comparison of this species with its most nearly related species is given under Ct. pectinata. Distribution —This species is known only from Barranca Ibarra, Jalisco, and Tres Marias, Nayarit. Description —U.S.N.M. Nos. 18967, adult female type; 18970, half- grown female paratype; Brit. Mus. Nat. Hist. Nos. 1, adult male; 75, adult male; 76, adult female, paratypes. Head long and narrow, covered with small hexagonal scales and very distinctly marked off from the body. Transverse gular fold present; no dewlap. Scales on muzzle larger than other head scales. Nostrils large, situated in the anterior third of the distance between orbit and tip of muzzle; nostral equal in width to mental, and deeper; lores flat, supralabials, 12; sublabials, 13. The back and sides are covered with small, smooth, subquadrate scales which pass gradually into larger ventrals. Gular region covered with small, smooth scales which become larger posteri- orly. Smallest gulars as large as largest dorsals, but smaller than the ventrals. Scales on limbs without spines. Tympanum nearly as large as orbit. Supraoculars small, flat, and hexagonal, the externals being about one-half as large as the internals, the internals being separated from each other by four rows of scales. The dorsal crest begins imme- diately back of the head and is composed of 73 compressed lobes, being continuous with the caudal crest and not interrupted at the sacral region. The lobes are highest (7 mm.) on the nape, and gradually diminish in height posteriorly until on the sacral region, where they appear merely as enlarged keeled scales. Their size and position makes them conspicuous even in the sacral region. The dorsal crest is much higher in males than in females. The tail is ringed about on the upper half with whorls of strong and very spinous scales. These whorls of spinous scales are separated from each other by rows of smaller, flat scales, the first six by four distinct rows of small scales, the sixth, seventh, and eighth by three distinct rows and a partial art.12 REVISION OF LIZARDS OF GENUS CTENOSAURA—BAILEY 31 fourth row bordering on the row of median caudal spines; the subse- quent whorls of spinous scales, up to half the length of the tail, are separated from each other by two rows of small scales. The distal half of the tail appears spinous throughout, the smaller scales becom- ing larger and gradually pass into the spinous type. The median caudal spines appear to be “‘set into’’ the other scales, so to speak; the margins of the adjacent scales lapping over the edges of the median row. The toes are very long, especially those of the hind feet; the claws are long and sharp. Femoral pores, right side 6, left side 5. Measurements.— U.S.N.M. US. nn ee F., para No. Weer re a Mm. Mm. Length of head_-.._--- 2 Serta See aCe ebae SS ere Cote ceo SS eS 65 40 Mong Linof bod yaess ser es Fb is nh a eS Sa eb Nee 225 155 ene uhvortanlecer a cee weet SSS oe eae eee wee oe oe ee RE SESE S 520 330 PRC emlOTIp Lees = a at RA Se SLSR Mie SAL ey iio SU SSRs Pt a Ad, 810 525 Width of head OVGL/ OLDS: eee ee Ses eS a ee aE ee 35 23 Coloration.—The general color of this species is olive green, lightly washed with vermilion and reticulated with brown and black. The flanks are heavily washed with vermilion. There are eight black blotches on the vertebral line, separated by areas paler than the gen- eral tint. All of the blotched markings are small and are more pro- nounced on the lobes making up the dorsal crest. Laterally they are represented by small blackish brown spots, but as they encircle the body they become conspicuous black bands. The transverse gular fold is heavily marked with black. There is a conspicuous black blotch bordering on the dorso-caudal margin of the tympanum. The tail is ringed with alternate wide bands of brown and yellow. Remarks.—A half-grown female, Cat. No. 18970, U.S.N.M. (same data as above), agrees with the type in all specific characters, but varies slightly in one or two minor details. The first and second whorls of spinous scales on the tail are separated from each other by four rows of small flat scales; the second, third, and fourth by five rows; the fourth, fifth, and sixth by four rows; the sixth, seventh, and eighth by three well-defined rows and a partial fourth row; the eighth, ninth, and tenth by three distinct rows; the remainder of the tail exactly as the type. The femoral pores as 7-7. Dorsal spines 75. Three specimens in the British Museum of Natural History, labeled from Tres Marias, Nayarit, vary slightly in the precise number and arrangement of the caudal scales, but agree with the type in all specific characters. They are designated as paratypes. 32 PROCEEDINGS OF THE NATIONAL MUSEUM VOL. 73 Material eramined.— Specimen Sex Age Locality Date Collector Remarks U.S.N.M.: 18967=—22 = M. | Adult 22 = Barranca Ibarra_-| Apr. 22, 1892 | P. L.Jouy ------ Type. IRO7Oe 2-23 F. | Half grown} Jalisco, Mexico---|__._- (6 (5) Ses ee (6 Pe A et Paratype. Brit. Mus.: of ee Soy ME?) Adult 22222 ‘Tres! Marias) Nay-|2-boi se 252 ase IVE. Rorrers 0 Do. arit, Mexico. | (ope ES M. | Half grown |---__- (0 eae ee AS Oe bees ee ee do Do. TOs eee Be) Adult alae Gon. eee oe eee eee open sce Do. CTENOSAURA SIMILIS (Gas) Plates 16, 17, 18, 19, 20 Iguana (Ctenosaura) similis Gray, 1831, Griff. Cuv. Animal Kingdom, vol. 9, Synopsis, p. 38. Cyclura (Ctenosaura) similis WIEGMANN, 1834, Herpet. Mex., p. 42.—Firzincur, 1843, Syst. Rept., p. 56. Ctenosaura completa Bocourt, 1874, Miss. Sci. Mex., vol. 3, Reptiles, p. 145.— Corer, 1886, Proc. Amer. Philos. Soc., vol. 23, pp. 266—269.—GtntTHER, 1890, Biol. Cent. Amer., Rept. Batr., p. 58, pl. 29-——Coprg, 1900, Rept. U.S. Nat. Mus. for 1898, p. 238.—Barsour, 1921, Proc. New Engl. Zool. Club, vol. 7, p. 82. Ctenosaura acanthura BOULENGER, 1885 (Part Group C), Cat. Lizards Brit. Mus., vol. 2, 197. Type—Museum of Mr. Bell, London, England. Type locality —Restricted to Tela, Honduras, Central America. Diagnosis —Dorsal crest very pronounced, not interrupted at sacrum, the continuity being effected by short lobes in adult male and by compressed scales in females and young. ‘Tail ringed about on upper half with strong and very spinous scales, the first and second (and occasionally the third) of these half rings are separated from each other by three rows of small flat basal scales; subsequent whorls of spinous scales up to the first third of tail’s length by two such rows of small scales; from this point the spinous scales continue without interruption to distal end of tail. Body color chrome tint, with trunk striped by five bands joining over stomach and united by numerous spots of same color. In Ct. pectinata, with which this species is some- times confused, the first five whorls of spinous scales on the tail are separated from each other by three rows of small flat scales. Distribution.—This species occupies the lowlands of Central Amer- ica and southern Mexico, and the sandy beaches of Panama. In Mexico it occurs on the Isthmus of Tehuantepec and the Yucatan Peninsula. The type at the time the description was published was in the personal museum of a Mr. Bell of Londen, but its present whereabouts is unknown. Also the types of Ctenosaura completa, a synonym of similis, collected by Bocourt in 1872, bear no definite f art.12 REVISION OF LIZARDS OF GENUS CTENOSAURA—BAILEY 33 locality label. They are said to have been collected in “Salvador and Guatemala.’’ Such general records are of no real value. Con- sequently I hereby restrict the type locality of similis to Tela, Hon- duras, where the greater number of specimens of this species have been taken. It also has been collected at the following places: In Central America—Belize and Glovers Reef, British Honduras; Ama- pala, Patuca, and San Pedro Sula, Honduras; Panama City, Corozal, and San Miguel Island, Panama; Tirives, Bonilla, Oritina, and Esparta, Costa Rica; Corinto, Chinadega, Polvon, Matagalpa, and Coseguina Volcano, Gulf of Fonseca, Nicaragua; Republic of Salvador; Old Providence Island, off Nicaragua; Bocomon, Cuastotoya, and Hacienda California, Guatemala. In Mexico—Progreso, La Vega, Merida, Chichen Itza, Mujeres Island, and Cozume! Island, Yucatan; Puerto Morelos, Quintana Roo Territory; Chiapas, Tonala, Montecrista, Tabasco, Tehuantepec, Oaxaca, and Suchitepequez on the Los Patos River (14 miles from its mouth). The specimen from Old Providence is a small female, not more than one-fourth grown. It is a true similis and was in all probability car- ried to the island from the neighboring mainland by some fishing or turtling schooner. Description.—Paris Nos. 01-255, adult male; 2252, adult male, type. M.C.Z. No. 22624, adult female; 22088, adult male; 22625, adult male. U.S.N.M. 56782, adult male; 47565, adult female. Head long, triangu- lar, covered with small hexagonal slightly convex scales and very dis- tinctly marked off from the body. Muzzle narrowed; supra oculars small, being separated from each other by four or five rows of scales; ear opening is almost as large as orbit. No dewlap, but the transverse gular fold is very pronounced; parietal scales slightly smaller than those on muzzle; nostrils large, situated in anterior third of distance between orbit and tip of muzzle; rostral larger than mental; lores flat; 13-14 enlarged supralabials; 13-14 enlarged sublabials; dorsal scales much smaller than ventrals, gradually increasing posteriorly in size, and spinousness; well-developed dorsal crest composed of from 60 to 92 spinous scales, constricted and pointing slightly backward. Promi- nent dorsal scales begin just back of head, on neck, and continue unin- terrupted to base of tail; the continuity of this crest is effected by short _lobes in the sacral region of adult males and by compressed scales in females and young. ‘Tail is ringed about on upper half with whorls of strong and very spinous scales, the first and second (and occasion- ally the third) of these whorls of spinous scales being separated from each other by three rows of small flat basal scales; the subsequent - whorls of spinous scales up to the first third of the tail’s length by two 88910—28——3 34 PROCEEDINGS OF THE NATIONAL MUSEUM VOL. 73 such rows of small scales; from this point the spinous scales continue without interruption to the distal end of tail. At base of tail ventral scales are much smaller than dorsals, three rows corresponding to each pair above, slightly keeled and pointed posteriorly. After the first four or five rows, ventrals and dorsals approach each other in size, two rows of ventrals corresponding to a like number of dorsals. Toes very long, especially those of hind feet; claws long and sharp; femoral pores 5-5 to 9-9; tibia without spiny scales. Measurements .— 01-255, | 2252 type, | 22088, M.C.| 56782, U.S. | 47565, U.S. Paris, M. | Paris, M. Z., M. N.M., M. | N.M., F. Mm. Mm. Mm. Mm. Mm. Meneth connexes sce 2a e— se eae 95 70 85 100 60 Length of bodys fo. Bete oe Pe ee 225 205 220 240 190 TON GE OVOMmUaAl ee eae eo oe ee ee 1 330 _ 455 470 580 1410 Total iéengtho 22-25 2 ae 8 se ae eee 1730 1710 775 920 1 660 Width of head over orbits_______---__---_- 37 31 ‘ 35 38 28 1 Tail broken off. Paris No. 2252 is a cotype of Ctenosaura completa Bocourt (Ct. similis). Coloration —General body color chrome tint, trunk being striped by - four or five black bands joining over stomach and united above by numerous spots of the same color. In old specimens the back bands become somewhat narrower and more or less broken up, appearing on the middorsal region of back as two distinct bands. Dorsal crest spines that lie in path of these bands are also colored black. Limbs blackish. Transverse gular fold spotted with black; throat and chin tinted with dark gray. Tail ringed with alternate wide bands of brown and yellow. In the younger specimens the general body color is light olive green, the inferior regions being yellowish, spotted with small brown or blackish dots. Remarks.—These lizards are very common in Central America, the Peninsula of Yucatan, and the Isthmusof Tehuantepec, Mexico. They are most abundant in the lowlands on the sandy flats and beaches. Their chief food consists of tender buds. They also feed on insects, as revealed by the examination of 25 stomachs. Most of the insects were beetles and grasshoppers. In Panama the habits of this species differ slightly from those of thesame species farther north. They occur on both sides of the Isthmus wherever there are sandy beaches, pref- erably with outcroppings of rock. They never appear about muddy shores or mangroves. These habitat associations occur more widely on the dry Pacific than on the moist Altantic side. Even in the dry Panama areas of the Pacific side one rarely sees this species more than 200 yards from the beach. They like the sand banks and rock piles ArT. 12 REVISION OF LIZARDS OF GENUS CTENOSAURA—BAILEY 35 and will lie basking in the hot sunshine. When discovered they scamper away among the rocks, but never take refuge among the shrubs, bushes, or low trees as does the same species farther north- ward. They are common about the old sea wall at Old Panama, about the rip rap falls near La Boca, at the Pacific entrance to the canal, and at Punta Bruja, a few miles to the westward. This species is never quite as numerous in Panama as just back of the beach at Tela, Honduras, where many may be seen at almost any time. Here they may be caught by the dozens in steel traps baited with a hybiscus flower. J. E. Gray described Ctenosaura similis from a dried skin belong- ing to a Mr. Bell of London.” The description, which was published in 1831, is given here ‘‘in toto”’: Allied Iguana, Iguana (Ctenosaura) similis. Grey, black dotted, body with four oblique dark bands; occiput forming a concave band behind; dorsal crest low but continued over the sacrum. Teeth blunt, three lobed; palatines on two raised lines on each side. Head 2 body 9 inches. Mus. Bell. This type, a mounted skin, was formerly in the private museum of a Mr. Bell of London, but subsequently disappeared. A careful search among the specimens and records at the British Museum fail to give any clue as to its whereabouts. Weigmann,” in 1834, and Fitzinger,™ in 1843, listed this species as Cyclura similis. They did not see the specimen but merely adopted Gray’s specific name of similis. Bocourt,”> in 1874, described this species as Ctenosaura completa from two adult male specimens and three young collected in ‘‘ Guate- mala and Salvador,”’ by himself in 1872. The next mention of similis and completa was by Boulenger,”* in 1885, at which time he listed both, along with many others, as syno- nyms of Ctenosaura acanthura. From that time on similis has been overlooked entirely, but its synonym, completa, was recognized as a distinct species by Cope” in 1886, and since then has enjoyed that distinction, being mentioned as such as recently as 1921 by Barbour.”® It is indeed unfortunate that this oversight has existed for so long a time, and I take this opportunity to restore the original name of Ctenosaura similis to this species. 22 Gray, Griff. Cuv. Animal Kingdom, vol. 9, Synopsis p. 38, 1831. * Weigmann, Herpt. Mex., p. 42, 1834. 4 Fitzinger, Syst. Rept., p. 56, 1843. % Bocourt, Miss. Sci. Mex., vol. 3, Reptiles, p.145, 1874. *% Boulenger, Cat. Lizards Brit. Mus., vol. 2, p. 197, 1885. 27 Cope, Proc. Amer. Philos. Soc., vol. 23, pp. 266-269, 1886. “Barbour, Proc. New Eng. Zool. Club, vol. 8, p. 82, 1921. 36 PROCEEDINGS OF THE NATIONAL MUSEUM VOL. 73 Material examined.-— . 7 Le Specimen Sex Age Locality Date Collector Remarks Brit. Mus.: Jef toe Pe bt Ys ie Wo Ub cree Colon (?) Panama_} (?)------_-__. M. Bocourt--_-___ a ee oe Mie gene Se does |aaees GO0ystee eevee (Pee RS a eee do. set se PA a cahs CSE 1 IVE A EI = do:2zties Pacific coast, Cos- | 1909-------__- Ballena_.-_-__._- ta Rica. AGS See heme F. | Halfgrown_| Mujeres Island, | 1889.-.-.-.___ S. F. Gaumer-__- Mexico. AGEs sass F. (2) 22k eee M. Ee a 22a a Cotypes of Ci. pis eae ees M } completa. 8221. =e 2 = M 11002222 - 2. F 01-255_---- M Hamburg Mus.: i 24082 | cree ea | hence Nicaragua____---_- Teo eat Reh Keasenberg..___- DiG2 ae kee [pdocue| eeteen ies tee Corinto, Nicara- | 1901_-.---_____ arisen 2.2 eee gua. 3546. 2002 ee Meee ae Amapala, Hondu- | 1910-.----_--- PT OSSIBR aes eee ras. Yay ty apa a |e | (ae Corinto, Nicara- | 1910.-..__--__|_-__. don gua. S550s-< ee Ns Be eta aes Nicaragua_._..__. L9LOp2 2254275] 2a domes bee. kss FGA TR a es ER ae iSO A Corinto,” UNicara-"|/ 1912250 aaa (6 {o}ee py Pee gua. 7D. sea pC | AR lg |B ad NI Salwadorss---4-- i oh a pees | eee GO) 222 ane Aye EeaINi ede: Ie ne Mi PAG te=- ses Yucatan _-_______- TSG Renee eee F. M. Chapman.. 16398__.--- i F. | Half grown_| Nicaragua.....___- DOG eet Nicaragua expe- dition. 16401 obit AB don See Gop eh e235. 452 GOie oe 1ONG Ao ee wees rea! G0 yb ssa s Cal. A. S.: L072 Sa A eae does Cuastotoya, Jala- | (?).---------- (2) eeeb: we pa, Guatemala. Gey tee ae Bboy Oun gee be! Suchet epequez, | (?)i--f2e22__ (Pe fot ees | Guatemala. AQIAQ See SONI) TeAdiGee Coseguina, Nica- | 1919_----.--.. J.R. Slevin__... Very large. H ragua. MiG war gs i SS10S oases rH Se Yorm pees Palvon, Nicaragua_| 1876_-.---____ Niche asses 5457. ‘ee ARR | ec aoa ehh TS seated dork SE 5 specimens. BQH | F. | Halfgrown_| Corcuera, Nicara- | 1886-..--_--._]----- Go aes 3 specimens. gua. 62702=2 2 x2- IMs |) Moungs2222 Merida, Yucatan_| 1889.-_.-_..._ E. W. Thomp- | 2 specimens. son. (pb preee Se eee dossas4 Progreso, Yucatan-_| 1905______-_.. GL. J.Cole:——-=- 9524. = tae end yee MS dp staee Chinadega, Nica- | 1905-.__..--_- W. B. Richard- | ragua. ‘ son. hfe eer fate sy aa WesCo bb h rece es Metagalpa, Nica- | 1908-_--.-.-..|----_ (3 (pe Sep ems. Se : ragua. 10308-12___; F. | Half grown_ Sean Miguelisiand, 070) dg Deo ees W. W. Brown._-_| 5 specimens. i anama. LOSIGe ee Ween eee doses Panama City, | 1904:-.--2°- | 2s2 dg soto e Panama. 15354-55_..| M. | Young-_-___- Crone, Costa |) 192022222 es EK. R. Dunn. -_-_ 2 specimens. ica. OQ 74a Rees NET al beetoes Gone Esparta, Costa | 1922___.-____. C. T. Under- | 5 specimens. , x Rica. wood. f{. &| Adult and i 21101-15___ F. young. \rela, Honduras =) =), 1925-2222 -- = Dr. H.C. Clark_| 15 specimens. arrapana. 2 HME lh 2 ido et |e - dp. ee a a Se 8 EM donut ere 10 specimens. 22088258 soe NM fe A@ultis oc-2 Glovers Reef, Brit- | 1925_----.--.. L. L. Mowbray-_| Very large. ish Honduras. 22624202 es 1A eee Goveesse Tela, VOndurass |i O2tensee eee T. Barbour. ---. 2AG25ee=2= 1 tr (ea Es dows Ree COS sna e e EOD TEs ee ara ye eS dow esas hihcamoey |r Renee eee eae ee Logi: 9. Liab: Dr. H. C. Clark.| 7 specimens. Nie) nlm=))| 2-2 Youngs Hacienda Cali- | 1926._--___--- A. W. Anthony-| 2 specimens. ber. | fons: Guate- / mala. \ i ART. 12 Material examined.—Conti ued REVISION OF LIZARDS OF GENUS CTENOSAURA—BAILEY Lay f ( 3 Specimen Sex Age Locality Date U.S.N.M : (ifs ees 2 A ee Sa ee Aspinwall) ban-||\¢_22s3e os. ama. LG See Se eee aes IMiexICO™saeeae eee Jaane eeern aan OOS se IVE Adie oe SIVIGXICO?2 25 ee Heli 2 ae sa sa AIOIESER os Brews ee OUIN Ra INIGRTS GU arse 6 aoe Ser iB EGY, eee a ee Cozumel Island, | Jan. 23, 1885__ Yucatan, Mex- ico. ISS7Oe.- es F. {| Halfgrown_| West Indies, Old | 1884__._______ Providence. 13898__..._ M. | Adult_..__- Cozumel Island, | Jan. 1, 1885___ Yucatan, Mex- ico. 17799-03___| M. | Halfgrown_} Honduras_-_-_.-__ LO UIts AseR ES 20290-6____.| M. | Young ----_- Patuca, Honduras_| 1891__________ asi 5—6 22 _| Me. loos. CWe ee Nan! Redrosulay |G )eoseonseeee Honduras. P24 (2) Wucatan, *Mexicos|s 22. --oses soe (277) M. | Half grown_|____- (Ue cea wes © BRET Cone bee ee Ne ee IN oa PRE, doen. Bases Cie pee ts ea ae oe eee p. MES oung= soo" Belize, Hondinase O(n 2p iy Ba eae (5 (ope pol ota is (0) enapeey S See G2) aes 320 ea F. | Halfgrown_ “Gentral EACLE) |i((?)) see bos See ica.’’ Shy es eee Mipsis baka dose! Bonillas, Costa |'(?)-.-.--_--_- Rica. OTOG2. = 25 20}e SoU Young 2222- Tirives, Costa | Apr. 12, 1906 Bice, | a0 Up Rema oy SER TE) eee Gost 4s OF 0 eS ee Pee Here he (5 (oe eae 466852... 1 a (ee doses Tonalla, Chiapas, | Aug. 9, 1895__ Mexico. | 46693______ 1 | here a Oe Mand OR aoeeee = 2-8 ane MO mae hel 47559-62___| M. | Adult______ Nujeres Island, | Mar. 35, 1901 Yucatan, Mex- ico. fc ae Vie |e VOUN e225 |p eee Go Wee eeeee eee Hse 47564______ NES yl eee doy): La Vega, Yuca- Mar. 15, 1901 tan, Mexico. 47565__.-__ Be | Adultiset. - Cozumel Island, | Apr. 11, 1901_ Yucatan, Mex- ico. 41590... MS eYioung eco Palanque, Chia- | May 25, 1900- pas, Mexico. 2. | OTS fg | ete G0es-s2 Puerto Morelos, | Mar. 28, 1901_ Quintana Roo Territory, Mex- ico. 47793__2___ IME AGuI ts tos. Monte Cristo, | May 7, 1900_-_ Tabasco, Mex- ico. CT IM. | Young 2s. Chichen ltza, | February, Yucatan, Mex- 1901. ico. 47802__..__ F. | Halfgrown-| Monte Cristo, | May 7, 1900__! Tabasco, Mex- | ico. 47953-5__.__| M. | Adult___.__ Chichen Itza, | Feb. 10, 1901_ Yucatan, Mex- ico. 54204__..__ M. | Halfgrown- Coronal; Canal | Apr. 11, 1911- one BAiSO.. 2. Mine | Adult 2 Belize, British | May 26, 1914- Honduras. cy) a Nisaerraligrowne|o-. doe ea esas Goplso-5s 56782.__._. ME. il (Adulte Loe) Tehuantepec, | 1905__________ Oaxaca, Mex- : ue 58499______ MeO OUng = ae— 4) eee a gees ee oe 1905 =seees So. 58500___._. M. | Adult__—._ Chinadega, INie=)|f L90t2 22 22 os : aragua. 71375-7___- (4* \ do erat Bocomon, Peten, | (?)----------- ; Guatamala. Collector Bocourt Dr. J. A. Brans- ford. Albatross pedition. ex- C.H.Townsend_ IE Wirherry ese J.C. Ingersoll___ R. Ridgeway --- Am) Alforac ese... Nelson and Goldman. Meeks and Hil- derbrand. J. Hurter Harry Malleis__ Remarks Exchange Paris Mu- seum. Paratype Ct. completa. 6 specimens. 7 Specimens. 2 specimens 3 specimens, Do. Ee ee ee 38 PROCEEDINGS OF THE NATIONAL MUSEUM VOL. 73 CTENOSAURA BAKERI Stejneger Plates 21, 22 Ctenosaura bakeri STEINEGER, 1901, Proc. U.S. Nat. Mus.. vol. 23, pp. 467-468. Type.—Cat. No. 26317. U.S.N.M., male. Type locality—Utilla Island, Honduras, 1900; Dr. J. E. Jarnigan, collector. TNagnosis.—A noticeable dewlap hanging from posterior part of throat; caudal whorls of spines separated by two rows and one row of flat scales. Dorsal crest high over neck and shoulders and gradually becoming shorter caudad, not continuous with caudal crest; upper sides of tibia with somewhat enlarged keeled scales; spines of median caudal crest subequal, much longer than the other caudal spines. - Iistribution.—This species is confined to Utilla Island, Honduras. This island is only 7 miles long and is situated just off the north coast of Honduras, in the Caribbean Sea. Itislocated within the 100-fathom line of the mainland. It may occur on Bonacca and Ruatan Island also. Description —Cat. Nos. 26317, U.S.N.M., adult female, type; 25324, adult female paratype. Head normal in length, covered with small hexagonal scales having slightly decurved muzzle. Head scales slightly rugose in adults. Supraoculars small, being separated from each other by four rows of scales; parietal scales smaller than those on top of muzzle; nostrils large, much nearer tip of snout than orbit, almost tubular, opening obliquely backward; rostral larger than men- tal; lores flat; 10-12 enlarged supralabials; 9-11 enlarged sublabials; ear opening as large or almost as large as orbit; dorsal scales small and smooth, hardly more than half the size of ventral scales, being almost granular on neck and gradually increasing in size posteriorly; dorsal crest well developed over neck and shoulders, the spines gradually diminishing in size posteriorly until at sacrum they become lost, the dorsal crest not being continuous with caudal crest. Spines and scales of dorsal crest 45 to 50 in number, beginning immediately behind head, the first scale is smallest, while the crest is highest over neck and shoulders, gradually diminishing in size posteriorly, until over small of back (loins) it consists merely of a median dorsal row of enlarged and slightly carinated scales. The spines are very compressed. Maximum height of dorsal crest scales is 3.5 mm., maximum width 2.5 mm. They are falcate in shape; their base is very flexible. A fairly large compressed dewlap hangs from posterior part of throat, 10 mm. from middle of base to top, the base along middle of throat being about 30 mm.; scales on throat and dewlap smaller than ven- tral scales, all smooth; scales of fore limbs normal; those of hind limbs larger, some of those of femur and tibia enlarged and slightly spiniferous. Femoral pores 9-9; tail not constricted at insertion; caudal scales above and laterally in whorls of large spinous scales, < bey etd art.12 REVISION OF LIZARDS OF GENUS CTENOSAURA—BAILEY 39 the central ones being spinous and equal or nearly so throughout length of tail; lateral spines are much less developed, being in fact smaller than median series and being separated by rows of smaller flat scales; first, second, and third whorls of spiniferous scales sepa- rated by two rows of these small flat scales; third, fourth, fifth, sixth, and seventh by one row of flat scales, and the other whorls of spinif- erous scales by two rows of flat scales; caudal ventral scales smaller than dorsals, three rows corresponding to each pair above, strongly keeled and pointed posteriorly. Measurements.— eat i epeteae. Brae cinmernetes is) oes 2 ad. ga So seg eh at athe het 50 mm. LSS OSS) 7076 fa OR OS he pe a el RET ey ee ee Re 160 mm. Lf B oR ER GEE ET So aa i ERR en eee ee 265 mm. UP Gea amie a ie a ed OL i Le ha ca NN le ea, 475 mm. Wiathh oh MeadtOVEeriorbitsL= 2 e243 os Jie hays ae we PvE alee he A 26 mm. Coloration.—Alcoholic specimen, paratype, Cat. No. 25324,U.S.N.M. Dusky brown showing signs of green with yellow variations on neck, throat, dewlap, and abdomen; dorsal crest and back dark brown with occasional outcroppings of yellow or green. Remarks—This species in possessing a very noticeable dewlap shows a close relationship to Otenosaura palearis, from Gualan, Guate- mala, and because of this striking peculiarity needs no comparison with other species of the genus. From palearis, however, this species differs in the less marked differentiation of the enlarged tibial scales and in the scutellation of the tail. In bakeri the first, second, and third whorls of spiniferous scales are separated by two rows of smaller flat scales; the third, fourth, fifth, sixth, and seventh by one row of flat scales, and the others by two rows, while in palearis there is only one row of very small flat basal scales throughout. In palearis, the median dorsal crest consists of alternate large and small spines, while in bakeri the spines of the crest are equal or nearly so. In palearis the lateral spines on the tail are better developed than are the median series, while in baker the scales of the median series are the largest. Then, too, bakeri grows to be larger than palearis, even more so than comparative total length measurements indicate. The head and body of bakeri are very much heavier than that of palearis, but the tail is somewhat shorter. The dewlap of bakeri is not as large as in palearis. This charac- ter, along with the peculiar scutellation of the tail in bakeri, tends somewhat to fill the gap between palearis and the other species of the genus. Material examined.— Specimen Sex Age Locality Date Collector Remarks U.S.N.M.: 2 ee F. | Adult___| Utilla Island, Honduras__| 1898 | Dr.J.E.Jarnigan_| Type. 25324__--_- Dye page 8 C0 cere em CO Ae ee eee TROSs ]eL2 = do) 2323s Paratype. 40 PROCEEDINGS OF THE NATIONAL MUSEUM VOL. 73 CTENOSAURA PALEARIS Stejneger Plates 22, 23 Ctenosaura palearis Strsnecur, 1899, Proc. U. 8. Nat. Mus., vol. 21, pp. 381- 383; 1901, vol. 23, pp. 467-468. Type.—Cat. No. 22703, U.S.N.M., male. Type locality—Gualan, Guatemala, 1899; Mrs. K. I. P. McElroy, collector. Dagnosis.—A large dewlap hanging from posterior part of throat; caudal whorls of spines separated by a single row of flat scales; upper side of tibia covered with large hexagonal scales, each armed with a central spine; dorsal crest high, but composed of 35 to 45 spines only; interrupted in the sacral region. Distribution.—This species inhabits the dry flat sandy zone of Guatemala in the vicinity of Gualan, at an elevation of from 1,000 to 2,000 feet. It has been taken at Gualan, and at Cuastotoya, Jalapa, Guatemala. These two points represent a very restricted area lying just south of the Motagua River. Description —The following description of the type specimen, an adult male, Cat. No. 22703, U.S.N.M., is from Doctor Stejneger’s original paper. Head rather short; muzzle with decurved profile, covered above with rather large and slightly rugose scales; supraoculars small, nearly granular externally, larger, hexagonal and flat internally, separated from each other by three rows of scales; parietal scales slightly smaller than those on top of muzzle, tubercular; nostrils large, much nearer the tip of snout than the orbit, almost tubular, opening obliquely backward; behind nostrils a large, flat scale; one or two canthal scales; lores flat; temporals slightly smaller than the occipitals, tubercular; 10 or 11 enlarged supralabials; 9 enlarged sublabials; ear opening as large as orbit; dorsal scales small, hardly more than half the size of the ventral scales, gradually inereas- ing in size posteriorly, smooth; a well developed dorsal crest, barely indicated on the rump. The spines of the crest, 45 in number, all told, begin almost immedi- ately behind the head; first six spines very small, followed by two somewhat larger ones; ninth is suddenly larger and tenth still larger, equaling the largest; the spines are very compressed, about 8.3 mm. (0.325 inch) high and 3.8 mm. (0.15 inch) wide at base, and faleate in shape; their base is flexible and covered for about one-fourth of their height with two to three rows of minute scales; the last 12 spines decrease gradually in size, the last being equal to the first ones on the nape; about 10 small carinated scales follow until the caudal crest begins; three transverse dermal folds across the throat which, with a similar one behind the ear, join two longitudinal folds on the side of the neck; these extend back- | ward over the shoulder for some distance; between the anterior and posterior transverse gular folds a large compressed dewlap 32 mm. (114 inches) from middle of base to top, the base along the middle of the throat being about 38 mm. (11% inches); scales on throat and dewlap slightly smaller than the ventral scales, all smooth; scales on upper side of arm carinate, those on lower arm slightly larger, more distinctly carinate and somewhat spinous at tip; scales on femur slightly larger than ventral scales, those on the upper surface obtusely keeled and with a small pointed tubercle at tip; scales on upper middle portion of tibia greatly enlarged, more or less regularly hexagonal, each with a falcate art.12 REVISION OF LIZARDS OF GENUS CTENOSAURA—BAILEY 41 spine near center; scales on upper side of hind feet toward toes enlarged, keeled, and spinous; 7 large femoral pores on each side; tail somewhat constricted at insertion, much depressed at base, becoming subcylindrical posteriorly; caudal scales above and laterally in whorls of large spinous scales, separated by a single row of smaller flat scales, the central one being spinous, however; in the spinous row the median scales are the shortest, the lateral one the longest, while in the smaller and smooth row the proportion is reversed, so that the anterior outline of the large row is concave and the posterior outline of the small row is convex; the outlines of each pair or rows perfectly straight; in the spinous row the scale on each side of the central one is without a spine; the lateral spines are straight; the central faleate, the median spines form a caudal crest, in the basal half of which the spines alternate large and small, according to whether they belong to the large or small row; caudal scales below much smaller, three rows corresponding to each pair above, strongly keeled and pointed posteriorly. Adult female: Same as male except that the dewlap is smaller and the dorsal crest is made up of shorter spines. Measurements.— U.S.N.M. No.22703, adult, male DiS mye Aa) aN CO) Coal EEN pene ih ca ego ell hele ace a Smeg es, OZR CR ed 40 mm. Oeste ITO LO Gayza ete meee a UI eS Sh Ble 2 MRR ele 2S ees Sa PLS 140 mm. Lig Sa gece) AY SiC Pace eR er A ape ee PN le ca A Sy ES AN i ge Se 230 mm. TCeetaee ul ea af rho esp haa el rs PO ete eS re Sa ny he ae a fee ee SO 410 mm. Wicitbhnoten cages OvellOroltss= ates oe oe arse Nem OS rat) EN Ea a Te 23 mm. Coloration.—Green with yellow variegations on throat, dewlap, and lateral folds; dorsal crest pale yellowish; on body several ill-defined, chevron-shaped blackish bands, which do not cross the dorsal crest, but the posterior three of which reach the abdomen; tail marked with broad bands of dull blackish brown. Remarks.—Doctor Stejneger says: Another specimen (No. 22704, U.S.N.M.) of the same age and sex, which dif- fers in no essential feature from the one described, except that the dorsal crest contains 36 spines only, and that the interruption between the doral and caudal crests is complete, being not even indicated by a row of carinated scales. Another peculiarity is that one of the small scales at the base in front of each dorsal spine has developed into a very minute spine. A third specimen (No. 24459, U.S.N.M.) is very young, only 198 mm. (7.8 inches) long. The dewlap is already well indicated, being 5 mm. (0.2 inch) deep; all the other diagnostic characteristics are also present and well marked. The dorsal crest is quite pronounced, the spine being triangular, about as high as long; the large ones standing some distance apart, the interval being wider than the basis of the spines; the small ones as the anterior and posterior ends are placed quite close; the number of the spines is 37; the crest perfectly interrupted on the rump. Eight femoral pores. Color essentially as in adults. The material from which the type was described was received at the United States National Museum in 1899 from Mrs. K. I. P. McElroy of Gualan, Guatemala. In May, 1926, Doctor MacPhail, of the United Fruit Co. Hospital at Quirigua, forwarded to Dr. Thomas Barbour, at the Museum of Comparative Zoélogy, 12 very fine specimens. They also were taken 42 PROCEEDINGS OF THE NATIONAL MUSEUM VOL. 73 at Gualan, Guatemala, and agree with the type specimen in every essential characteristic and have been designated as topotypes. This is a remarkable series of a very rare species, nine males and three females. . Two specimens belonging to the California Academy of Science were examined and they also agree perfectly with the type. They were collected at Cuastatoya, Jalapa, Guatemala. In Doctor Stejneger’s original paper on this species he emphasized its close affinity to the other members of the genus, and stated that the presence of the dewlap was not sufficient ground upon which to erect a new genus. His discovery of Ctenosaura bakert three years later, 1901, has bridged the gap between palearis and the other species of the genus and also demonstrated the wisdom of not creat- ing a new genus based upon the dewlap character. Material examined.-— Specimen Sex Age Locality Date Collector Remarks U.S.N.M.: PPO este oe NES Ad lt ss Gualan, Guate- | 1899_.__._-_-__| Noirs. UKE Se: hype mala. McElroy. 7h) eS Se 16 Exes dO) 2 ete ks (GLC) ok Bae eee TRG Que a2 ee ieee doe See 24459 222 Mir 22 oe doe: | eee (Gio eS SS 1 et eee Ne) a oj ee 48065-22252 | AB. }ie24 doe 22 Guatemalatssss S| eth eee Prof. Newton Miller. C.A.S.: bola (eps Sree Mo) | Youngs 225 Cuastatoya, Gua- | (?).---------. (9 Fh See SS ieee - temala. 10GB =e VAG ats =o Guatemala-- 2" =~ (Gd) aati, Se [PW eet era eg M.C.zZ.: 22302 aa 1 ae ie hope ee Gualan,. Guate- | 1926.--.--.___ ' Doctor Mac- | Topotype. mala. Phail. 22394.) 22 ) Og eee doses. vise ORES. ee Se TODS E sop re eta doce ees Do DIOR e wee ee UB: (Zee ee doo sais do 25 eee 19252 seo P Re BAN eB oye eee Tek Do 22893e. = ny (Ao eee Co Ko eee Pigpe e (6 Kenny ie ener 1926262 CoS doritesessese Do 22395 ey MG. |e Gio}2 ame bee Gov gence be fe PO2G 8. 2 Se do sasoeeane Do DOS0G ee Wiss ee Oise | are GO eee NODG eee ae We 5 (oom eB Do 22397 -=--- dy fog ee ee do SN78. OE ee AE NODS ei) Veen ore wees Do 22399: 322 INT See doses [eee ee Oia 1996. 2 Eee! [ere OM ese as < Do. No number_| 3M. |____- (6 Ko} FOR: Gotsc tone eae 1OQ6 sawen ease ee Gone Do. | | CTENOSAURA QUINQUECARINATA (Gray) Plates 24, 25, 26 Cyclura quinquecarinata Gray, 1842, Zool. Misc., p. 59.—Sumicurast, 1873, Arch. Sci. Phys. Nat., vol. 46, p. 259. Enyoliosarus quinquecarinata Gray, 1845, Cat. Lizards Brit. Mus., p. 192. Cyclura (Ctenosaura) quinquecarinata Cop, 1869, Proc. Amer. Philos. Soc., vol. 11, p. 161. Ctenosaura (Enyaliosaurus) quinquecarinata Bocourt, 1874, Miss. Sci. Mex., vol. 3, Reptiles, p. 188; 1876, Journ. Zool., vol. 5, p. 401. Ctenosura quinquecarinata SuMIcHRAST, 1880, Bull. Soc. Zool., vol. 10, p. 175.— Bocourt, 1882, Le Nat., vo]. 2, No. 6, p.47.—BouLENGER, 1885, Cat. Lizards Brit. Mus., vol. 2, p. 198.—Cops, 1886, Proc. Philos. Soe. Amer., vol. 23, pp. 266- 269.—GinruHER, 1890, Biol. Centr. Amer., Rep. and Batr., p. 58.—Dvuazs, 1897, La Naturleza, ser. 2, vol. 2, No. 12, p. 523, pl. 34.—Copn, 1900, Rept. U. 8. Nat. Mus. for 1898, p. 238.—STEJNEGER, 1899, Proc. U.S. Nat. Mus., p. 383. Type—Brit. Mus. Nat. Hist. No. 61, male. Collected March 5, 1841, no locality. art.12 REVISION OF LIZARDS OF GENUS CTENOSAURA—BAILEY 423 Type locality. Restricted to Tehuantepec, Oaxaca, Mexico. Diagnosis —Median dorsal crest made up of low, thin, leaflike scales, extending from nape to sacrum. Upper surface of hind limbs covered with large spinous scales. Upper surface of tail covered with alternate rings of larger and smaller scales, the central, and especially the two or three lateral series of larger scales, being very large and heavily armed with a sharp spine directed backward. Except at base of tail the larger scales forming the three series on either side of the central caudal spines are flat. ; Distribution —The type specimen of quinquecarinata is a stuffed skin, in the British Museum of Natural History. It is without any locality or collector’s label. Of the 23 specimens examined, 19 are designated as being from Tehuantepec, Oaxaca, Mexico; 2 are labeled merely ‘‘ Mexico”; 1 is listed from Oaxaca, Mexico. In all probability all of the specimens came from Tehuantepec, at which place the elevation ranges from 100 to 600 feet above sea level. This species is perhaps confined to the Isthmus of Tehuantepec, and upon the evidence presented above I hereby restrict the type locality of Ctenosaura quinquecarinata to Tehuantepec, Oaxaca, Mexico. Description.—Brit. Mus. Nat. Hist. Nos. 61, adult male, type, stuffed skin; 33, adult male, alcoholic specimen; U.S.N.M. Nos. 30127, female; 30561, male; 30562, male; 30563, male. Head normal in length, covered with small hexagonal scales, with slightly decurved muzzle. Supraoculars very small, being separated from each other by three rows of scales; nostrils large, on canthus rostralis, lateral; loreal region smooth. Supralabials, 8; sublabials, 10; rostral wider than mental; ear opening as large as orbit; strong transverse gular fold; dorsal scales small and smooth, being almost granular on neck and gradually increasing in size posteriorly until over small of the back. On the rump they become a little larger than ventrals, rhom- boidal and obtusely keeled. Outer side of tibia armed with large spinous scales. Dorso-nuchal crest made up of 50 to 60 thin, leaflike scales ranging from 1 to 5 mm. in height, and extending from nape to small of back (loins), or to sacrum, but never continuous over sacrum: more pronounced in males than in females. Tail longer than head and body together, slightly constricted at insertion; depressed in its anterior third, but cylindrical posteriorly; its upper surface covered with alternate whorls of larger and smaller scales; the central, and especially the two (occasionally three) lateral series of the former, very large and spinous; the latter and the three (occasionally four) larger series adjacent to the central spinous row flat. First two or three whorls of large scales at base of tail are all spinous. Lower surface of tail covered with transverse series of smaller scales, strongly keeled and pointed posteriorly. Femoral pores vary from 5-5 to 7-7. 44 PROCEEDINGS OF THE NATIONAL MUSEUM VOL. 73 Measurements — U.S.N.M., No. 30561, adult male Wengthiof head .2502 220 2fUR SA ees DOOR SAS Oe eee 35 mm. Lengthjof body 2.224222. rhs oe ei on pear sale es ope een 95 mm. Wenighh.jOF chau i 2 2g. yer Re a ae Le Wy Ee eee eee oe ae 200 mm. oben Lemeb hr 2 og oe ON ae ge pt nie LE we Ses nee nee 330 mm. Coloration.— Above and on sides pea green, mottled with black or brown; legs also pea green with prominent black splotches, in some instances taking the form of more or less indistinct transverse cross- bands; lower surfaces yellowish, uniform or spotted with brown; throat grayish, mottled; chin dark. The young are green through- out, being somewhat lighter on the under parts. Remarks.—This species is very closely related to clarki, erythromelas, and defensor, and a comparison of it with these species is given under the remarks on erythromelas. Material examined.— Specimen | Sex Age Locality Date Collector Remarks | Bub Mus.: ei ae F. | Adult___| Oaxaca, Mexico__.--._| Nov. 11, 1841 | Purchased from Bocourt collec- tion. DOSS. 8 VE ste Om arte oe GOs iss fate sel Sept. 30, 1903 | H. Gadow-_----.-- 5 specimens. Boe enee. M. |__-do-.._| Tehuantepec city, | Oct. 10, 1890 | A. C. Butler_____ | Oaxaca, Mexico. (7) ene ae HN, el Ro Cs ee nee) a FO ae March 18402 s|5 (2) acest al eee Type. (Psa seg b ee ---d0 ~..-| Oaxaca, Mexico------- (Cp Are ee Se (2) Skate Se Stuffed skin. Paris Mus.: 1322.____ {Mel do... BEN or te AER Ae fey ys F. Sumichrast-_--| 2 specimens. 5016____- Mell ad Zul i laok AG Oh GoM tuaee es eeu WENN den) AS Do Hamburg Mus:.: (ij eae INE |e edo. veers IM pxico tesa oe 2 aw US7So A Se et Mischersces* Le 66222252. VP eee Co Te i a (a ae SES cy (cee meson a Goizs.5222c22 Do. (Gia eee Mieiee sdok ii (2)e_ be ere es ats a WS79e sasese ss Schilling. _.__-._- 1208 3.322125 2u8a |. ke Sass Hugma, ,Perminos-__- 189122. eo Peaersen_-- ---- = ply yee Se ob a ee ee ig avexico}?s --_ --_---- IS BPE Stop Bley GY) GOES D1 GULOS ee ee ee L208 2122 GUCVUOIC CS= eee nea eae oe ee 1, 5, 7, 9, 10, 14, 16 Wlenosauna defensor_=--------_-__=_— 4, 6, 8, 9, 44, 48 Gtensowura denticulata __.-----.--------=--=-- 10 Ctenosaura erythromelas__________- 5, 8, 9, 17, 20, 46 la@iaaull (ojo) ose ee 4, 5, 6, 8, 9, 17, 20 Ctenosauara insulana___-.--.------.--- ff PAO) Pal 2) HUNT UY D oop ee eS err 17, 20, 22 UCOUCES UTS an 10, 13, 15, 16, 21 Page Ctenosaura; paleanis2==2= ase e 4, 8, 9, 39, 40, 42 Parkerlsses-en eee eee sass 2, 4, 6, 8, 25, 29 pectinata_--------- 4, 6, 7, 8, 9, 23, 24, 25, 27, 29, 32 quinquecarinata___----- 4, 6, 8, 9, 42, 43, 44, 48, 50 Gtenosaiina Shani se es ee 10 Ctenosaura similis_______-- 2, 4, 6, 7, 8, 9, 25, 32, 338, 35 GEENOS WUT ORUCT Sears ee ee eee a 10, 13 CGachnyx) anmectens === nan 46, 47, 48 (GBacenta\eacanthima ee 13 Cvclunadp=— = Seals ee Sern ae Dron Owils O20 (CCUM (AUTON = oe = Se SS 9, 10 (Clenosa@una)acanthura ===> = === 2 2 10 ONCE eee oe ee 10 QNLICULOLU = ee en eee 10 DEL ee ee a ee en 10 CUCU OIL CS 2 ee ee eee 10 CEnUCULOO= aaa eee Sp oe 10 eNO DNC en ee oe eee 17 ST ADTs oa ee oe a Ree 10 SUNIL Soe a ee eee 32 LeTESw i eae oe ee er ea eee 2, 10, 13 Centicwlata mesa eee 10, 14, 16 PIC CUUTLCUL Cee nn ee er 24 SETIICHISLOLOs Sen ee ee ee 10 SUTTULLIS ee ee en ek ee = ae ee 35 LET OS eke eae se ee oe Mee ee eee eee 14, 17 CLUE COU LL ae 42 cycluroides, Ctenosaura__--.------- 1, 5, 7, 9, 10, 14. 16 Gefensor, Clenosalras sss se== =e = 4, 6, 8, 9, 44, 48 GCTUICLOLALGAN OU CLUNG see a 10, 14, 16 Distribuiiomlol Lae swecleSa= assesses ee 4,5,6 TSM INCRE: 2 aoe a= = Sec ee Se oeee 8 GrinOUOdles, (HUT ac aca e a oe ee See 6, 48 erythromelas, Ctenosaura _----- 5, 8, 9, 40, 44, 46, 50 Generaliconsidenationse=s== === 2 Geologic history of the genus. --------------- 3,4 hemilopha, Ctenosaura-_----------- 4, 5, 6, 8. 9, 17, 20 1S (Oy VIDOR UC: See ee ees 3 Vigianales ee aes ee ae ee 7 NG ECETUCRCLCCUTLLLL UU Cen ea ee 17, 20 Bellies soe en es ee Se eee 10 CAL ODE tee ee ee eee 3 COLE OSC @ NOT MUO eee 14 NON CCOLOLO een 2 nee me eee 10 SUTULIIS Bo eee oe a See 2 JIGADENOO UR oo 2 et ss See ae ee eee 3 ere Gl ae ee ee ee ee 3 insulana, Ctenosaura_____------------- 17, 20, 21, 22 interrupta, Ctenosaura_-_—--------------=-- 17, 20, 22 Introduction: 2... 2-- === 2-4) -2s.---=222-=5 1 Key onneisPeCleS === === aa =a eae eae 8,9 57 58 INDEX Page Page TEA) CRAM 3 = = ee Ss Out Simm SenCnIStaL aC CLUTT== ene See ame eee 10 Miatenialvexamnine da == see eee bs 2 similis (@tenosaural sss — 2 === 4 6, 7, 8, 9, 32.33, 35 multispinis, Ctenosaura___- -------- 1), 1B}, a5), TG}, |) Gare, (COUN. << = ne ese 32, 35 ING WeSC CLES ane eee ae ae ne ae eee 2, 26, 44 HGOGPO) = = ee eos 2, 4, 6, 7, 8, 9, 29, 32, 35 palearis, Ctenosaura_.-----=-----__- ANSON OFA QHADI IS WECIESICESCILIOC Cl pase ae ee eee 8 parkeri, Ctenosaura___---.-.--.--.-. 2,4, 6, 8, 25, 29 srali@l_ 2 A aE hee ee ee ee 8 Mectkinatas ©benOSaUn aes === snes eee Gree LON QS a CO LETUOS CULM Oe ee 10 6, 7, 8, 9, 23. 24, 25, 27, 29, 30, 32 Gy QluUnds LE SS a ee ae ee Prey Web ayy (AGHIOS (ONQUOUR ao ae se 2 = D4 aothalassina Utdews ss ees 2 ee ees 19 quinquecarinata, Ctenosaura __-_.------------ 4, COTS CLO COTUNLIILS = 5,9 LOL RL CRY AVA) ||) (ine) Wok ibisisibael oe ee Ses = 19 (UB TTC CORUIVLG ges C2 CLLNT Capen eee 49. \\ VianlatlonSs-2 222-28 S-2 oe = a ee = ee 6,7 JOO HMONUTTUD = oo op ee 42 O _ FOSSIL NUTLETS OF THE GENUS LITHOSPERMUM By Epwarp W. Brrry Of the Johns Hopkins University, Baltimore, Maryland There are in the collections of the United States National Museum some hundreds of silicified fruits collected by the late John B. Hatcher from the Loup Fork formation in 1884 which have never been identified or described. More recently similar material has been sent in from the undifferentiated Tertiary of Kit Carson County in eastern Colorado. Among these are numerous specimens representing a new species belonging to Lithospermum, a genus belonging to the family Boraginaceae and not hitherto known in the fossil state. These show a wide range of variation, but after much deliberation I have con- cluded that the best method of treatment would be to consider all of them as varieties of a single botanical species which may be called Lithospermum fossilium. The smallest and most abundant variety may be described first as: LITHOSPERMUM FOSSILIUM RUGOSUM, new variety Plate 1, Figures 1-10 Nutlets relatively small, averaging about 3 millimeters in length, 2 millimeters in width, and 2.5 millimeters in thickness. They are contracted upward to a distal cuspidate but rounded apex, and are asymmetrically inflated, the outside being much more convex than the inside, both the apex and hilum being nearly in the plane of the less inflated inner side. On the inside a rounded keel extends from the apex nearly or quite to the hilum. The hilum is large and circular, from 0.5 to 1 millimeter in diameter. The surface is rugose but varies from nearly smooth to an aerolation of well-marked ridges. This variation in sculpture is not due to abrasion before or after fossilization, I am quite sure, since it would have been equally effec- tive on the apical point or the delicate rim of the hilum, which is not the case. Found in both Kansas (figs. 1-6) and Colorado (figs. 7-10). No. 2734.—PROCEEDINGS U. S. NATIONAL MUSEUM, VOL. 78, ART. 13 86418—28 1 2 PROCEEDINGS OF THE NATIONAL MUSEUM vou. 73 LITHOSPERMUM FOSSILIUM GLABRUM, new variety Plate 1, Figures 11-13 This variety is less abundant than the preceding and is known only from Kansas. In form the nutlets are asymmetrical fusiform, being always larger and relatively longer and narrower than var. rugo- sum. The apex is usually more distinctly cuspidately pointed, the keel on the inner face is less pronounced, and the base is contracted to a narrower hilum. The surface is smooth and polished. It is difficult to determine the extent to which the apex was produced, since it is so readily broken off, and it is usually impossible to detect evidence of fracturing, but there is some evidence of a small amount of elongation. Length, ranging from 3.5 to 7 millimeters. Width, ranging from 1.5 to 3.5 millimeters, and thickness ranging 1.5 to 3.5 millimeters. The seed coat as preserved is one-fourth to one-eighth of a millimeter in thickness. Phillips County, Kans. (Figs. 11 to 13.) LITHOSPERMUM FOSSILIUM ARISTATUM, new variety Plate 1, Figure 14 This variety is rarer than either of the preceding, probably because the seeds are much more fragile. I judge this conclusion to be correct, because among 20 specimens only the one figured has the apical spine preserved for any distance and nearly all have the seed coat more or less broken. This variety is more slender and elongated than either of the others and is more symmetrical in side view and nearly circular in transverse section. The base is contracted to a hilum about the size of that in var. glabrum. The apex is produced as an attenuating spine which may be two-thirds as long as the inflated portion. The keel, though obvious, is less elevated. The surface is smooth and polished. The size varies considerably. The dimensions, exclusive of the spine, which is so rarely preserved, range from lengths of from 5 to 10 millimeters and diameters of from 1.5 to 3 millimeters. The material does not permit of a determination of the constancy or length of the spines. Phillips County, Kans. (Fig. 14.) This is the most remarkable of the variants of this species. I know of no instance among the existing members of the families Boraginaceae, Verbenaceae, and Labiatae where the four-parted ovary develops into four nutlets where these are crowned with any- thing approaching the spines of this fossil variety. That this variety is related to the others is clear by the similarity between it and the shorter pointed variety glabrwm, with which it is associated. 4rtT.13 FOSSTL NUTLETS OF THE GENUS LITHOSPERMUM—BERRY a The genus Lithospermwm comprises about two score existing ‘species of annual or perennial herbs, widely scattered in prevailing dry soil habitats, mostly in the Northern Hemisphere, but sparingly represented in South America and Africa. The four-parted ovary develops into four or fewer hard nutlets of rather characteristic shape and with smooth and polished or wrinkled and pitted bony coats. I have reproduced nutlets of several recent species for comparison with tl fossils. The fossil variety rwgosum is, in both size and form, much like the nutlets of the existing Lithospermum lneari- folium Goldie, shown in side and inside view in Figures 15 and 16. This is a wide-ranging North American form found in dry situations from Manitcba to British Columbia and southward to Illinois, Texas, and Arizona. The species Z. arvense Linnaeus has a similarly roughened surface. The fossil var. glabrum is approached by a considerable number of existing species with hard, smooth surfaces. Of these I reproduce two, namely, Z. ruderale and L. pilosum. (Figs. 17, 18, 19.) Both approach glabrum in size and form, but differ in their greater width and larger hilum. I have seen nothing among the modern forms which closely resem- bles the fossil var. aristatwm. Other species than those reproduced would serve equally well for comparison and illustration, but I have not been able to see fruits of all of the species. EXPLANATION OF PLATE Fies. 1-10. Lithospermum fossilium rugoswm, new species. 1-6. Phillips County, Kans. 7-10. Kit Carson County, Colo. 11-13. Lithospermum fossilium glabrum Berry, new species. Phillips County, Kans. 14. Lithospermum fossilium aristatum Berry, new species. Phillips County, Kans. 15,16. Lithospermum linearifolium Goldie. 15 from side. 16 from inside. 17. Lithospermum ruderale Linnaeus. From side. 18,19. Lithospermum pilosum Nuttall. 18 from side. 19 from outside. All enlarged three times. : an jod- ered ah | o A bolig-aubt: ‘anki i HY) | | oe ey aT PN = tai ian ok cme sit esi nth the ae Oke Bite: 6 Geer oat | a) AORN svi 4 sc toa erate badvent gert0) ne ve ats yew ‘ ‘ogee ‘it i ‘at sine i is ite Hix Me asa olde ee ae bestow nn aiening We patois Loace ea vooahain eigeogie baa Mew ® , Make ims ye.b! Ae eet se vai) re abereatare aA Siw iinet sieuky 2 | iis b Hd RC \ v \ : } fale j : i} ve by 4% "4 h % ghee FS ad. Te ut aT ARLHS | en Sacto ste spill i Sek ‘ * Sot Bady: Rolo a Atte ag A trae. Lat eee I dich HOR EMSS I Ba kh gud Heyy mt Mg Nowe BORE SES, Sh i: Asiaalje oct Ay em bh rit : aod \ j ve . Petes wes, fomiga | canis aA _ i avy ‘ae rant a ? 3 he olite ue niu 2h ma Bt fis US a RY Ligue Ah BH iia dala Biaiadiee a Le Mess Burk NA {alk rvicht Ore wen eaa iN ‘ See sb RS ead 5 1 a Ae Sat RR cay ; Hl Pik : ts Suey ; ") sng U. S. NATIONAL MUSEUM PROCEEDINGS, VOL. 73, ART. 13, FossiL NUTLETS OF THE GENUS LITHOSPERMUM FOR EXPLANATION OF PLATE SEE PAGE 3 PL. 1 pat 3X ae ah ¥ « = 5 y - ace ‘ . e a ‘ 7 7 ven? , 3 2 ‘ if ‘ i - ? ¥ ‘a 7 u , - F Pet ne 3 er ne : ' by > 3 ‘i - . he. % > - =< - FIRE-MAKING APPARATUS IN THE UNITED STATES NATIONAL MUSEUM By Wa.tter Houcu Head Curator of Anthropology, United States National Museum INTRODUCTION In the ethnological collections of the United States National Museum at the period of 1886 there were many fire sticks, chiefly from the North American Indians, and from the Eskimo, with a few from other parts of the world. Prof. Otis T. Mason, then curator of ethnology in the newly organized Museum, observing the interest of his aide in these curious objects, urged him to take up their study. In this oppor- tune time, 42 years ago, the constant stream of ethnological material flowing into the Museum from expeditions brought with it other fire sticks, until a sufficient basis for their scientific description was assem- bled. From the first in the National Museum organized by G. Brown Goode on the basis of arts and industries it was regarded as a necessity to practically work out the method by which the aborigine produced the impedimenta which supplied his needs. Thus before the writer could intelligently handle the subject of aboriginal fire making he should be able to make fire by all the methods known to man. This unexpectedly difficult task accomplished, the first practical monograph on the subject appeared. It was found possible in the study to classify the methods and assign them to races and geographic areas and also to give a synopsis of the technical and developmental status of the methods. The widespread interest in this subject was demon- strated by the demand for the paper, which has been placed out of the reach of collectors for many years. The present paper is a revi- sion and extension of the former publication. The following is a classification of the chief methods of fire making, based upon the presumed order of development: 1Hough, Walter. Fire-Making Apparatus in the U. 8S. National Museum, An. Rep. U.S. Nat. Mus., 1888, pp. 531-587. No. 2735.—PROCEEDINGS U.S. NATIONAL MUSEUM, VOL. 73, ART. 14. 86374—28——1 1 2 PROCEEDINGS OF THE NATIONAL MUSEUM VOL 73 it Simple two-stick apparatus.—Indians of North, Central, and South Amer- | ica; Ainos, Japan; Somalis, Africa; most Australians, etc. The most widespread method. 2. Four-part apparatus; mouth drill and two-hand drill.— Eskimo, some Indi- ans, Siberians, Hindus, and Dyaks. - Compound, weighted drill.—lIroquois and Chukchis. I. On wood (reciprocating motion) by- II. On wood (sawing motion with knife and thong). Malays and Burmese. IIT. OP ORs (plowing or planing ™0- | Polynesians; some Australians. ion). 1. With pyrites (or stone containing tron) and flint—Eskimo and Indians of the North (Algonkian and Atha- IV. Of minerals and bamboo (percus- pascan stocks). sion). 2. Flint and steel— Modern and disused methods and appliances. 3. Flint or other hard substances on bam- boo.—Malay. V. By compressed air. Besides the lens, mirror, and aerophore there are pyrophores, the hydrogen lamp, matches, and various chemical and electrical methods, FIRE MAKING BY ARTIFICIAL MEANS Observations on the customs of the races of mankind, extending in time and area, show that by one or more of several methods all men knew how to make fire artificially. The origin of fire making is evidently lost in the past, but there is no valid reason to put the invention very far back in time. Sound criticism will place it at the period of one of the profound advances marked by new ideas and the beginnings of the great movement of the dissemination of man over the earth. Several deductions are legitimate concerning the nebulous period when the art of fire production was in the making. The first of these is that fire was carefully preserved when no means of lighting it again were at hand and fire occurring in nature from the lightning and volcano was difficult to obtain. Hence the ancients of the early world of man learned all there was to know about the guardianship of fire and were also frugal in the use of this ‘friend of man.” Another deduction is that there is more or less uncertainty in carrying fire about, especially to any distance, under primitive con- ditions. If fire was as important to man as has been imagined, he would not for a long time migrate from his primitive seats. It is not possible to imagine man, undoubtedly a creature of long development and with a long train of acquired experiences, remaining art. 14 FIRE-MAKING APPARATUS—HOUGH 3 in one place like other animal groups. With the use of tools, the ownership of fire, organization based primarily on instinct and more importantly on ideas, man presents himself as a candidate for the conquest of the earth. We know that a very long time ago men in the fire-preserving stage spread over large areas in Europe, lived there for millenniums, and passed on. Other areas in western Asia and northern Africa were also found to show evidence of the presence of early man. The continuing thread that runs through this mass of material culture of man is progress. The sum is progress, irrespec- tive of the involved and doubtful questions concerning the time and the man himself. It is not known whether some groups of early man migrated without fire or whether at times fire was lost and became forgotten. Our theories that fire is indispensable to all humankind under all conditions are subject to modification. It is curious that very many fire myths recount a stage of firelessness and a wresting of fire from those having it. Myths are regarded by scientific men as having a substantial basis of fact and it must be concluded that the fire myths indicate that some groups of men were fireless. No such condition of things has ever been observed among the tribes of modern or historical man. The myths, therefore, may be considered as portraying conditions of considerable antiquity. It may be assumed, therefore, that for a very long period man, pos- sessed of fire and tending it with skill and care, did not know any method of making it at will. How the making of fire artificially came to be developed, for it was a development and not a discovery, can never be known. Which of the methods takes precedence over the others is also uncertain, but from the facts presented it would be argued that wood methods occurred prior to mineral methods. _ The first makers of fire may have been confined to a single social group, clan, or tribe. Among observed tribes instances are many where the art of generating fire by wood friction was jealously intrusted to one or a few individuals. Even among the cultured nations of antiquity there is evidence that fire making was never gen- erally practiced by the common people. Tribal possessions of land, minerals, animals, and plants, as well as tribal secrets, form a phase of observed aboriginal life. They are in the class of village industries or the arts and industries of smaller social units possessing workers of skill. Many of these special arts arise on the coalescence of human units into larger social aggregations rather than by derivation by the methods of acculturation as usually understood. In this sense fire making artificially would originate in larger groups where the contact of minds is more stimulating to invention. Such things were at first kept secret under clan custom, dominance of the priesthood, or for other causes, but in the course of time the technical process became 4 PROCEEDINGS OF THE NATIONAL MUSEUM vou, 73 common property and the esoteric sides remained in the care of reli- gious organizations. The thought on the origin of fire making has tended to vision the human need, and to consider the need supplied by an inventor, much as wonder-working inventions are brought out nowadays. It will be shown that nothing could be farther from the facts. In considering the difficulties which confronted the inventor of the wood-friction fire-making apparatus it is pointed out that the proper wood in proper condition must be found. All advanced Boy Scouts will subscribe to this. Evidently the wood was not selected by the early experi- menters before they knew what was to be done with it. Again, the trap for the fire is a slot or channel cut in the horizontal piece, termed the “‘hearth.’’ This slot is, advisedly, a great discovery. Tinder of a suitable kind in which the spark may be nourished must be found, and this is no small task. Finally the little coal of fire can be brought to a blaze only with great skill and a knowledge of a number of things. These are the chief matters of difficulty connected with the invention of fire making brought out in actual experiment, but there are other minor steps in the process which are seemingly inconsequential yet are vital to efficiency. It is necessary, then, to vision a long preliminary period during which man gained a growing acquaintance with the properties of various substances which were immediately useful to him in various ways. He knew from the first fire that wood was the fuel which burned on his primitive hearth. He may have thought that fire ate wood, but the chief lesson was that wood burned. Other associations of wood with fire may have come from work with this substance. Friction is a common experience and handling wood or working in wood might give to keen perceptions an odor, a vapor of smoke, sug- gesting that there was fire present. Unconsciously perhaps these observations led to more knowledge and gradually to an awakening to a combination of these experiences into something useful, and the fire drill, let us say, followed. Why the drill was the form the fire kindler took is not difficult to imagine. The drill is an ancient and primitive tool supplying the need for piercing holes in various mate- rials. It is the result of diverse means employed to pierce holes, such as scraping, punching, grinding, cutting, and breaking, the processes described being variants of the making of holes. From these experi- ences came finally the drill more especially as a tool for piercing hard substances requiring abrasive processes. Thus when the culmination of the protracted experiments was about to be reached there was a drill which could be adapted to the service of fire making. The Hindu fire-origin myth most practically states that the carpenter _ with his drill first elicited the divine fire. In its present state the Aryan Hindu myth places the origin of artificial fire making at a ART, 14 FIRE-MAKING APPARATUS—HOUGH 3) period of social organization in which occupations as of woodworking were followed, indicating a comparatively late event as suggested. FIRE MAKING WITH THE TWO-PART DRILL Making fire with the two-part drill is not difficult. The apparatus is designed to render fire making easy, yet with two sticks in hand and no knowledge of the details one is placed practically in the posi- tion of primitive man. The drill is a straight, stiff, dry, slender, smooth rod, the diameter of a lead pencil or larger, as shown in the specimens used by various tribes, and up to 20 inches long, but rarely shorter than 12 inches. The hearth admits of far greater variety of form. The large majority, however, are straight-growing sections thicker than the drill and of the same wood. Many of the hearths appear to have been gathered haphazard as a good piece of wood was found, many others are fashioned in a workmanlike manner with flat sides and squared edges, while others are carved and shaped according to the fancy of the native artists. Having the two essentials and provided with tinder and accom- paniments for getting a blaze it is possible to describe in detail the making of fire. We round the lower end of the drill and make a slight holding notch near the border of the hearth in which to start the drill rotating, and we cut a clean, vertical slot from the notch down the edge of the hearth. This slot should be cut deep enough to divide the pit in which the drill operates nearly halfway. Place the hearth on the floor or on firm ground, kneel on one knee, and hold the hearth firmly with the other foot. Take the drill between the extended hands, set the rounded end in the notch, and roll it between the palms, pressing down at the same time. After a few rotations the drill will have bitten into the wood and the dust ground off fallen down the slot, which explains its purpose. Remove the drill and make sure that the slot is central with the hole, also that the drill end is not binding, and if so, whittle it a bit to insure its working freely. Before beginning again it must be understood that the work should be carried on to the finish without displacing the drill from its socket. Also, when rotating the drill the hands will gradually move to the lower part of the shaft. At this juncture grip the drill with one hand, bring the other up to the top, grip the drill with it, and hold it while the other hand can be brought to the first position. By practice this can be done so quickly that no appar- ent cessation of the drilling is observable. To resume the effort to make fire it is better to revolve the drill slowly at first, gradually increasing the pressure till the dust darkens somewhat and smoke arises, then quicken the stroke and pressure to the extreme till the carbonized wood dust pushes down the slot in a coherent roll. If a thin vapor arises from the dust the work has been 6 PROCEEDINGS OF THE NATIONAL MUSEUM VOL. 73 a success. Shortly a little coal of fire will be seen. To get a blaze from this fragile coal is like Langley’s problem in launching the first airplane; the difficult problem which was hardly visioned in the more important effort. There are, of course, a number of minor variations in the procedure. The coal may be very gently fanned where it lies and finely divided material added in right amount. Generally, softened inner bark strips are previously placed under the hearth and shredded grass or bark with perhaps a bit of tinder placed near the slot. The coal emerges in this material, which is taken up on the strips of bark and gently waved in the air. In a little while a flame bursts forth. FIRE MAKING WITH THE FOUR-PART DRILL OR BOW DRILL The explanations referring to the hand drill aiso apply to the bow drill in everything except in the handling of the machine which sup- plants the bare and often excoriated palms of the would-be primitive fire maker. The aboriginal bow is rigid, straight, or slightly curved and from 10 to 20 inches long. The cord of buckskin or hide is tightly fastened to one end and can be adjusted at the other to take up slack. The drill is usually of la1ger diameter and shorter than the hand drill, and tapered toward the ends from the middle. The nut is a block of wood, generally carved in form of an animal among the Eskimo, of convenient size for holding in the hand, and set with a piece of soft, easily polished stone such as marble or soapstone in which a cavity has been made. Make one turn of the cord around the drill, leaving the bow to the right, set the drill in the socket of the hearth, and place the nut. Make a few turns to ascertain conditions, especially whether the cord grips the drill sufficiently to not slip when pressure is applied on the nut. The position of the driller is over the drill, the left hand holding the nut bemg brought around the left flexed knee, which aids pressure on the drill. Begin slowly and increase pressure, nicely balancing the pressure with the grip of the cord. In the concluding rapid work the cord may be tightened by pinching it up between the thumb and first finger of the right hand. When the drill is felt to bite strongly into the wood and throws up a little smoke increase the pressure and rotation until the fire coal is thought to have appeared. Hold the drill in place till this fact is known, as it is much better to continue with a hot drill than a cold one. With the fire plow we have another idea radically different from the drill. In this case a blunt stick is held between the locked fingers, pressed down firmly, and rubbed back and forward on the flat surface of a horizonal hearth, cutting a groove and forcing the dust into a little heap at the end of the groove, and in which the fire rises. As in the drill, this method requires careful and assured movements, calculated to a nicety lest the accumulation of dust be disturbed. At ART. 14 FIRE-MAKING APPARATUS—HOUGH - the culmination of the effort the rubbing stick is raised to a higher angle so that it will bite more strongly. The fire saw offers another curious method which is nearer to the fire plow than to the drill. The fire saw is almost invariably of bamboo, to which the method conveniently adapts itself. A bamboo joint is split into halves; from one is fashioned the sawing part in form of a strip with a sharpened edge. In the concave of the other half a few slivers are raised to hold a piece of tinder. This is set on the ground convex side up and the saw is swiftly rubbed across at right angles and over the the cage holding the tinder. The saw soon cuts through the wall of the bamboo and at its hottest comes into contact with the tinder, which is ignited. A related form is the fire-thong apparatus, which consists of a stick either split and wedged apart, asshownin Plate 9, or havinga horizontal slot cut through the middle. A bit of tinder is stuck in the slot; the cord, which is a rattan length, is passed under the stick at the tinder and pulled up and down, shortly igniting the tinder. QUALITIES OF WOOD It is found by experience that the qualities rendering wood suitable or not for fire making are as follows: Wood with fine grain, or without grain, as in deadwood, and wood decomposed to a certain extent, and stalks of yucca and other vascular flower stalks are to be chosen; new wood, or wood containing certain products of growth, such as gums, resins, starches, sugars, and tannic acid, will not produce a fine, dry, inflammable powder; and, in general, hardwoods are to be rejected. In selecting wood judgment must be exercised after the manner of aboriginal man, who desired wood that was dry, soft, of proper grain, and inflammable, and as a result his selection was nearly always of deadwood. Some woods, however, thoroughly dried and seasoned, will answer very well. In many cases wood must be tested with the drill and discarded if the dust rubbed between the fingers is gritty. Also, one part only of the chosen wood may be good. The following American woods are suggested for fire making. Those queried may be valuable if in proper condition as mentioned. Hemlock, willow, elm, soft maple, sycamore, tulip (?), cedar, cotton- wood, balsam (?); poplar, silver, Lombardy ; white pine (?) and yucca, flowering stalk. There are introduced here the explanation of two plates of a series exhibited in the United States National Museum illustrating the pre- sumed development of the art of fire making.” 2? Extracted from Proceedings, U. S. Nat. Mus., vol. 60, No. 2404, 1922, pls. 1 and 2. Z ° er bo PROCEEDINGS OF THE NATIONAL MUSEUM VOL. 73 . Voleano in action; lava setting fire to forests. Water-color drawing. 178157. . Lightning setting a forest on fire. Water-color drawing. 178158. Camp fire; man borrowing fire. Water color drawing. 178159. Fire saw. Strip of bamboo drawn across a section of bamboo. Dyaks of Borneo and other Malays. 178152. . Fire thong. Rattan thong drawn over a grooved piece of wood. Dyaks of Borneo. 178152. . Fire plow. Blunt stick worked along a groove ina lower stick. Polyne- sians. 178152. . Fire drill. Slender rod twirled between the hands upon a lower stick’ having a cavity with slot. Indians of the United States and widely diffused in the world. 176353. . Fire drill. Rod held in a socket and gyrated by means of a cord. The lower piece of wood has a cavity with slot opening upon a shelf. Eskimo of Alaska. 127644. No. 9. Fire drill. Rod held in a socket and gyrated with a bow and cord. Lower No. 10 Nos No. 12. No. 13. No. 14. No. 15. No. 16. No. 17. No. 18. No. 19. No. 20. No. 21. piece with cavities on a central groove. Eskimo of Alaska. 48078. Fire drill. Pump drill used specially for sacred fire. Iroquois Indians, Canada. Strike-a-light. Flint and iron pyrites struck together as the ordinary flint and steel. Eskimo of Alaska. 178154. Strike-a-light. Flint and steel and box for holding flint, steel, and tinder. Sulphur-tipped splint ignited from the tinder. England. 130486. Strike-a-light. Bamboo tube and striker of pottery used as flint and steel. Two boxesfor tinder. Malay. Tinder pistol. Gunlock adapted for throwing sparks into tinder. Eng- land. 175712. Strike-a-light. Combination of flint, steel, tinder, and extinguisher for carrying in the pocket. Spain. 178155. Fire syringe. Cylinder with closely fitting piston bearing tinder. Driv- ing the piston down smartly kindles the tinder. Siamese and Malays. 176091. Lens. Used for producing fire by focusing sunlight upon tinder. Ancient Greeks. 178151. Hydrogen lamp. Hydrogen gas is made to play upon spongy platinum, causing it to glow. Germany, 1824. 165440. Match light box. Bottle of sulphuric acid, into which splints tipped with potassium chlorate and sugar were dipped. Vienna, 1809. 151711. Matches. Various kinds of phosphorus matches. 178156. Electric gas lighter. Cylinder containing a small dynamo run by pres- sure of the finger, producing sparks between the points at the upper end of the tube. United States, 1882. 200512. The means of applying forces for making fire may be classified as below in the order of their utilization: Frictional, drilling, sawing, plowing on woods. Percussional, striking pyrites, flint and pyrites, and flint and steel. Physical, compression of air, platinum sponge, lens, mirror. Chemical, chemical combination as matches, sodium, pyrophores, etc. Electrical, electrical energy, sparking apparatus. art. 14 FIRE-MAKING APPARATUS—HOUGH 9 ETHNOGRAPHY OF FIRE-MAKING APPARATUS I, FIRE MAKING BY RECIPROCATING MOTION 1. Simple two-stick apparatus.—This method may be said to have a world-wide distribution and to have had no narrow range in time. It is a very interesting study to observe the many different practices that have been superadded to the simple task of twirling two sticks with the design of creating fire. It is also instructive to note how fixed have become tribal characters in so small a thing as the shap- ing of the elements of the fire drill. It has well been said by Doctor Schweinfurth that: A people, as long as they are on the lowest step of their development, are far better characterized by their industrial products than they are either by their habits, which may be purely local, or by their own representations, which (rendered in their rude and unformed language) are often incorrectly interpreted by our- selves. If we possessed more of these tokens we should be in a position to com- prehend better than we do the primitive condition of many a nation that has now reached a high degree of culture.’ This fact holds good with reference to tribes in a higher plane than the learned writer included in this statement, in this way: There are many little things that have not been subject to the modi- fication of time, intercourse, or environment, but coexist with an art. To particularize: Prof. E.S. Morse has shown the value of the sim- ple act of releasing an arrow from a bowstring as a classifier. Close attention to the minor acts and arts will reveal much more than the nice measurements of man’s practically unmodified skeleton. Differences that have become functional in the arts have come down from an early period; when they can be found they are of the greatest value as aids in ethnology. The ethnography of the simple fire drill is studied geographically, beginning in North America with the most northerly tribes that use it, and ranging from north to south in the different sections of the country, among the tribes from which there are specimens in the Museum. Other countries are examined from west to east. The Sitkan fire-drill spindle is unusually long and thick. (Fig. 1.) Both hearth and drills are of the Thwja gigantea, a tree that enters so largely into the life of the Indians along this coast. The wood grinds off very well with much friction; at ordinary speed there is soon a small heap of powder at the bottom of the fire slot. The lat- ter is deeply cut in from the side nearly to the center of the fire hole. The whole hearth has been charred at the fire. This repels moisture, and also renders it easier to ignite the wood, charring being a process somewhat analogous to the decay of wood by rotting. If kept care- fully in a dry place, this apparatus was perfectly adequate for the +The Heart of Africa. New York, 1874, vol. 1, p. 257. 86374—28 2 PROCEEDINGS OF THE NATIONAI MUSEUM von. 73 10 Cat. No. 20644, U.S.N.M. BELLA- 3, FIRE-MAKING SET AND SLOW. MATCHs - No. 127866, U.S.N.M. QUINAIELT INDIANS, QUINAIELT, WASH. COLLECTED BY CHARLES LOUGHBY No. 743879, U.S.N.M. TiinaiT INDIANS, SITKA, ALASKA. r= Q n ar a o< HE san a ZO Hm teers ahi 8 Dw Zz EB a i ay ™ NOAA fo a] aes zr 223 oars) a Su cae BAO Tee mee) Q as : ye no < SOMO fy i! art. 14 FIRE-MAKING APPARATUS—HOUGH 11 purpose of the Sitkan, and in his skillful hands would no doubt give the spark in a minute or so. The long drill would indicate that two worked at it consecutively to keep up a continuous motion. For tinder, the bark of the arbor vitae was used. It is finely frayed, and is much improved by being slightly charred. They also use, preferably, a tinder made from a fungus, because it is “‘quicker,”’ that is, ignites more readily than the frayed bark. The hearth is squared and measures 23 inches; the drill is of equal length. The southern Tlingit drill in the American Museum of Natural History, New York City, has the hearth and drill of equal length. The hearth is a block having the head of a raven at one end and of a bear at the other. The drill is enlarged at the lower end, and the hearth has three fire holes. (Pl. 1, Cat. No. 239100, Alaska; Amer. Mus. Nat. Hist., N. Y.; hearth, 15 inches long (38 cm.); drill, 15.3 inches long (39 cm.).) Going southward from Sitka the next fire-making set in the series is from Bella-Bella, British Columbia. These Indians are of the Sali- shan stock, and are called Bilhulas. The horizontal is a piece of cedar wood dressed square on three faces. It is apparently a piece of an oar - orspearhandle. The fire holes are shallow, and the fire slots are quite narrow. (Fig. 2.) The drills have been scored longitudinally near the rubbing end; this may be a device to cause the wood to wear away more rapidly and furnish fuel to the incipient fire. Fire has evidently been made with this set. Both parts are 114 feet long; the drill is much thinner than that of Sitka. The tinder is a braided length of frayed cedar bark. From a southern family of the Salishan stock, called the Quinaielt Indians, of Washington, the Museum has a complete set collected by the late Charles Willoughby. It consists of a hearth, two drills, and aslow match. The hearth is a rounded piece of cedar wood; opposite the fire holes it is dressed flat, so as to rest firmly on the ground. There are three fire holes with wide notches. The drills taper to each end; that is, are larger in the middle. (Fig. 3.) The powder, a fine brown dust, collects at the junction of the slot and fire hole, and there readily ignites. This side of the hearth is semidecayed. No doubt the slots were cut in that side for the purpose of utilizing this quality. The drills are bulged toward the middle, thereby rendering it possible to give great pressure and at the same time rapid rotation without allowing the hands to slip down too rapidly, a fault in many fire drills. The slow match is of frayed cedar bark, about a yard long, folded squarely together, and used section bysection. Mr. Willoughby says: The stick with three cavities was placed upon the ground, the Indian kneeling and placing a knee uponeach end. He placed one end of the smaller stick in one Fia. 4, — FIRE-MAK- INGSET. Cat. No. 24096, U.S.N.M. KLaAMATH INDI- ANS, OREGON. COLLECTED BY L. 5S. DYar. PROCEEDINGS OF THE NATIONAL MUSEUM VOL. 73 of the cavities, and, holding the other end between the palms of his hands, kept up a rapid half-rotary motion, causing an amount of friction sufficient to produce fire. With this he lighted the end of the braided slow match of cedar bark. This was often carried for weeks thus ignited and held care- fully beneath the blanket to protect it from wind and rain. Fire is easily procured with this set. It takes but a slight effort to cause a wreath of aromatic smoke to cur] up, and the friction easily grinds off a dark powder, which collects between the edges of the slot. When this ignites it drops down the slot in a little pellet, and falls upon the tinder placed below to receive it. Both drill and hearth are 18 inches long. The Klamaths, of Oregon, of the Lutuamian stock, use a fire apparatus that looks very much like that of the Utes. The hearth is a rounded piece of soft- wood thinned down at the ends. (Fig. 4.) The drill is a long, round arrow stick, with a hardwood point set in with resin and served with sinew. (See Ute drill, fig. 7). The holes in this hearth are very small, being less than three-eighths of an inch in diam- eter. They are in the center, and the fire slot being cut into the rounded edge widens out below, so that the coal can drop down and get draught. The wood is quite soft, apparently being sapwood of yew or cedar, while the drill point is of the hardest wood obtainable. It is probable that sand is used on the drill. The hearth is 13 inches long, and the drill 26. The Chinooks, a tribe of Indians of a separate stock, called Chinookan, formerly lived about the mouth of the Columbia River, in Oregon, but are now nearly extinct. James G. Swan, the veteran explorer, inves- tigator, and collector among the northwest coast tribes, says that the Chinooks are the best wet-weather fire makers he ever knew.* To kindle a fire the Chinook twirls rapidly between the palms a cedar stick, the point of which is pressed into a small hollow in a flat piece of the same material, the sparks falling on finely frayed bark. Sticks are commonly carried for the purpose, im- proving with use.° Paul Kane® describes the hearth as a “flat piece of dry cedar, in which a small hollow is cut with a chan- nel for the ignited charcoal to run over. In a short time sparks begin to fall through the channel upon 4 Northwest Coast, p. 248. 5 Bancroft. Native Races, vol. 1, p. 237. 6 Wanderings of an Artist Among the Indians. London, 1859. ART, 14 FIRE-MAKING APPARATUS—-HOUGH finely frayed cedar bark placed underneath, which they soon ignite.’ The Ahts and Haidas also use cedar fire sticks of the usual Indian kind, The Hupa Indians of California are of the Athapas- can stock. Their fire drill is a carefully made piece of apparatus. (Fig. 5.) The hearth is of a reddish, punky piece, made from the roots of a willow (Salix laevigata), or of cottonwood roots (Populus trichocar- pus), The drill is made from the root of the willow mentioned. Fire has been made in one of the holes; the others show the rough, frayed cavities which have been made to start the drill. The notches at each end of the hearth seem to be to facilitate the tying of the pieces together as a precaution to prevent their Joss or separation. They are usually intrusted into the hands of the most skillful fire maker, who wraps them up to keep them from becoming damp. The effec- tiveness of the sticks increases with use and age; a stick and hearth that have been charred by the for- mer making of fire in most cases yields the spark in half the time required for new apparatus. Another advantage is that the drill is softer from incipient decay. That this set is in the highest degree efficient is shown by the fact that the writer repeatedly got a glowing coal, the size of a pea, from it in less than 20 seconds. The hearth is 18 and the drill 21 inches long. The Nokum Indians of Lassen County, Calif., use a small hearth with wedge shape end, probably to steady the piece against a support while making fire. Small block hearths like this are customary among the Ute ‘tribes. The drill is a slender rod worked from cedar, as isthe hearth. (Cat. No.131078; Susanville, Calif., L. L. Frost; hearth, 7 inches long (18 cm.); drill, 11 inches long (28 cm.).) The McCloud River Indians (Copehan stock) make the drill from the buckeye tree. The Indians of Washoe, Nev., from their language , have been classed by the Bureau of Ethnology as a separate stock, the Washoan. Stephen Powers many years ago collected a rather remarkable hearth from these Indians. It has eight rather small holes, in every one of which fire has been made. The wood is soft, well-seasoned pine. Apparently sand has been made use of to get greater friction, as is the Fig. 5.—- FIRE- MAKING SET. Cat. No. 77193, G48. Ne eM Hupa INDIANS. CALIFORNIA. COLLECTED BY LIBUT br .. e, Ray Fig, 6.—FIRE-MAE: INGSET, CAT. NO. 19640. U.S.N.M. W ASHOE INDIANS, NEVADA. COL- LECTED BY STE- PHEN POWERS PROCEEDINGS OF THE NATIONAL MUSEUM VOL. 73 custom of the Zufiis and Apaches. This device, in a measure, obviates the necessity of having tinderlike wood, or wood in a state of partial decay. For the drill any hardwood cylindrical stick might be employed. A strip of buckskin about an inch wide is passed around the hearth over the fireholes to keep them dry. (Fig. 6.) At the end of the hearth is a mass of cement made of the resin of a pine mixed with sand, apparently, a kind of material used by the Indians over a large area in the Great Basin and southward to fix their arrowheads, pitch the water bottles, and for other purposes. It is quite probable that this stick was the property of an arrow maker whose need of fire to melt the some- what intractable cement caused him to combine these functions in one tool. It has a better finish and displays greater skill in its manufacture than the fire tools of the neighboring tribes of Shoshonian (Utes) and Moquelumnian stocks. In fact, it has a close affinity in appearance to those of the very near Athapascan (Hupa, etc.) stock. It is a matter of very great interest to compare with this a stick from the Mackenzie River. (See fig. 26.) The resemblance is striking; it is as though one found a word of familiar sound and import in an unexpected place. The related tribes of the Indians dwelling on the Mackenzie have a wider range than the distance between the localities whence the respective sticks came; in fact, the Athapascans range about 50° in latitude and the southern colonies of this great family are only about 250 miles southeast of the Wash- oans, while, as has been stated, the Hupas are quite near. It would be presumptuous to say at present that this too! is a remnant of the influence of the Atha- pascan wave that swept along the Great Interior Basin, leaving groups here and there in California and other parts to mark its progress, but there is more to its credit than a coincidence of form and function. The museum is in possession of a complete collection of fire-making material from the tribes of the Shos- honian stock. They were collected by Maj. J. W. Powell. The native name for the Ute fire set is whu-tu ni-weap. While the lower member of the set—the * art. 14 FIRE-MAKING APPARATUS—HOUGH hearth—differs among the several tribes in point of material, shape, etc., the spliced drill is characteris- tic of the whole stock. It has never been noticed outside of the southern part of the Great Interior Basin but in one instance—among the Klamaths of Oregon. The main part of the drill is either a reed or a straight sprout, usually the former. At one end a short piece of very hard wood—grease- wood, Sarcobatus vermiculatus—is set in and lashed with sinew. It resembles the Shoshonian arrows, which are foreshafted in this way. They also use sand in common with other neighboring tribes. The Pai-Utes, of southern Utah, make their hearths of a short, rounded piece, usually of the sapwood of juniper. It is tied to the drill with a thong of buckskin when not in use. (Fig.7.) The drill is like the usual one, just described. This is the common form of the Pai-Ute apparatus. The small, two-holed hearth of rounded form and the shortened, spliced drill are for convenience of carry- ing, this kind being used by hunters while away from the lodges. S.J. Hare says that the men do not usually make the fire except when out on a hunting excursion. At the lodge it is the squaw’s duty to make the fire when it is needed. The Pai-Ute is rarely at a loss to get fire; he is master of various devices. Mr. Hare, who was among the Utes for some time, states that when the Indian is in need of a light he uses either the flint and steel, the drill, or, if these are not at hand, he takes two branches and rubs one up and down on the other, soon getting fire. The Australians are said to have practiced fire making by rubbing in the way mentioned. This is the only observa- tion collected of its occurrence in America. It is, in all probability, a difficult, unusual way, only practiced under pressure of necessity among the Utes. They take great pride in their skill; to be a quick fire maker is to achieve fame in the tribe. They are fond of exhibiting their art to white trav- elers in the hope of gain. Another form of hearth (fig. 8) is made of yucca flower stalk, like those of the Apaches and Navaho. The drill is of tule reed, set with a very hard wood 15 Fic. 7.—Fire-M Ak- ING SET. CaT. No. 17230, U.S.N.M, Pal-UTE INDIANS, SOUTHERN UTAH. COLLECTED BY J. W. POWELL . 16 PROCEEDINGS OF THE NATIONAL MUSEUM VOL. 73 Fig. 8.—FIRD-MAK- ING SET. CaT. NiO}4g 2 OTs \6; U.S.N.M. Pat- Utp INDIANS. SouTHERN UTAH. COLLECTED BY J. W. PowrLL head. It is suggested that the reason for splicing the drill is that the hardwood of the kind used for the head (grease-wood) can not be procured in pieces long enough to make the whole drill. This set is apparently one used as a fixture in the Ute domestic economy, the squaws having to light the fire. The duty is mainly relegated to the females in several other Indian tribes and among the Eskimo. Mr. Catlin says that the Sioux objected to letting the squaws have their portraits painted, saying that their women had never taken scalps nor done any- thing better than make fires and dress skins.’ The hearth and drill last figured are, respectively, 20 and 23 inches long, while in the hunting set (fig. 8) the length is 7 and 18 inches. The Wind River Shoshones are also represented. (Fig.9.) The hearth is of hardwood, rudely hacked out and rounded. Upon the slanting edge are eight holes or shallow depressions, prepared for the drill, with notches cut in to meet them from the sides. The drill is a willow branch, 25 inches long, with a hardwood head mortised in, and served with buckskin. It is most probable that sand was used with this set, because, if the parts are not models, it would be necessary to use it on sticks of equal hard- ness like these. I am inclined to believe that they are models, from their appearance and from the difficulty of setting up a pyrogenic friction upon them even with sand. They were collected more than 50 years ago by Maj. J. W. Powell. The Hopi are the most differentiated members of the Shoshonian stock. Mrs. M. C. Stevenson collected the two excellent fire-making sets in the Museum from the Hopi Pueblos. The hearth is a branch of the very best quality of softwood (cotton- wood). Inone hearth an end has been broken off, but there still remain 18 fire holes, showing that it was in use for a long time and highly prized. (Fig. 10.) The drill is a roughly dressed branch of hardwood. It is comparatively easy to make fire on this apparatus. In the set numbered 126694 these conditions are reversed; the hearth is toler- ably hard wood and the drill soft wood. The Hopi fire tools are used now principally in the estufas to light the sacred fire and the new fire 7 Smithsonian Report, 1885, pt. 2. p. 723. ar. 14 FIRE-MAKING APPARATUS—HOUGH 17 Fic. 9.—FIRE-MAKING SET. Cart. No. 22022, U.S.N.M. SHOSHONE INDIANS, WIND RIVER, WYOMING. COL- _ LECTED BY J. W. POWELL 86374—28 as do the Zufiis, and the Aztecs of Mexico did hundreds of years ago. They use tinder of fun- cus or dried grass rubbed between the hands. By their language the Zuiii people belong to a distinct stock of In- dians. Their fire sticks are of the agave stalk, a soft, pithy wood with harder longitudinal fibers, rendering it a good medium for the purpose of making fire. (Fig. 11.) As to the plan pur- sued in grinding out fire, Col. James Stevenson informed the writer that they make a slightly concave place where the burnt holes are seen, cut the notch on the side, sprinkle a little fine sand on the concavity, set the end of the round stick on the sand and roll it rapidly between the palms of the hands, pressing down hard. The “‘sawdust,’’ Colonel Stevenson called it, oozes out of the notch and forms a small mass, which on blowing slightly becomes a burn- ing coal, and the appli- cation of a little tinder creates a blaze. For preserving the fire for any length of time they use a piece of decayed wood. (Fig. 12.) 3 * SET. Cat. No. 128694, U.S.N.M. Hopi In- DIANS, ARIZONA. COLLECTED BY Mrs. M. C. STEVENSON Figs. 11 AND 12,—FIRE-MAKING SET AND SLOW MATCH. CAT. Nos. U.S.N.M. Zuf1 INDIANS, NEW MEXICO. COLLECTED BY JAMES STEVENSON PROCEEDINGS OF THE NATIONAL MUSEUM VOL. 73 Viewed in another aspect than as an implement of necessary or common use, this set is an important cult appa- ratus in the wonderfully complicated religious worship of the Zufis. These people make the sacred fire that was used in their ceremonies by friction of wood that has been wet. New fire is made at the beginning of their new year with great ceremony. ‘The house is swept and everything is moved out of it until the fire is made. Their regard for fire and their customs with reference to it add them to the list of peoples who have held it in similar reverence and have prac- ticed similar customs all over the world, ranging widely in time... The wet- ting of the drill, increasing their labor, may be done to please their gods. This art must have been practiced for a long time in this region, for Henry Metcalf found a hearth (fig. 13) with three fire holes in a cave dwelling at Silver City, N. Mex. It is appar- ently very ancient. The wood is much altered and some salt, probably niter. in all parts of the region. great Athapascan stock, widely in North America. Fic. 13.—LOWER STICK OF FIRE-MAKING SET. OA TON. Boe@08% U.S.N.M. FROM A CAVE AT SILVER City, N. Mex. ColL- LECTED BY HENRY METCALF has become heavy by impregnation with Specimens are found during nearly every exploration in the cliff dwellings. They show entire uniformity The Apache and Navaho belong to the that ranges so Capt. John G. Bourke, United States Army, collected the hearth of yucca wood shown (fig. 14), and says: 127708 AND 69850 With the stick you now see, the Apache Indians in my presence made fire in not quite eight seconds by the watch, and one asserted that he could make ART. 14 FIRE-MAKING APPARATUS—HOUGH it in a number of motions, which, on the watch, occupied exactly two seconds—that is, under most favorable cir- cumstances. The experiments, made under my own ob- servation, ran all the way from 8 to 47 seconds. Sand is generally used, although not essential to success. Captain Bourke’s observation is very interest- ing, as it records the fact that the Apache is the most skillful fire maker in the world. Many other tribes can make fire in less than a minute, I think by far the majority of them, but there is no eight-second record, while if he could prove his ability to do it in two seconds he would arrive at the facility of striking a match. William F. Corbusier has noticed the fire mak- ing of the Apache-Yumas of Arizona (Yuman stock). They use a drill about 2 feet long and one-half inch thick, made of o-oh-kad-je, or ‘‘fire- stick bush.’”’ Its end is dipped in sand and drilled on a soft piece of agave or yucca stalk held down by the feet. They carry a slow torch of dead- wood (spunk) and also use a flint and steel. For tinder they use dry grass or bark fiber. They _ use also a fungus, some species of Polyporus), for the same purpose. Another reference to the fire making of this stock (Yuman) is found in the translation by the late Dr. Charles Rau of the writings of Father Baegert on the Californian Peninsula.’ He says: To light a fire, the Californian makes no use of steel and flint, but obtains it by the friction of two pieces of wood. One of them is cylindrical and pointed at one end, which fits into a round cavity in the other, and by turning the cylindrical piece with great rapidity between their hands, like a twirling stick, they succeed in igniting the lower piece if they continue the process for a sufficient length of time. The Navaho fire set looks very much like a mere makeshift. The hearth is a piece of yucca stalk and the fire holes have but a shallow side notch. The drillisa broken arrow shaft, to which has been rudely lashed with a cotton rag a smaller piece of yucca wood. (Fig. 15.) This careless- ness, which it is rather than lack of skill, is char- acteristic of the Navaho in their minor imple- ments. They resemble the crude Apache in this. ‘American Antiquarian. Mendon, IIl., vol. £, September, 1886, Dp. 283. § Smithsonian Report, 1865, p. 367. 19 Fic. 14.—LOWER PIECE oF FIRE-MAKING SET. CAT, No. 130679, U.S.N.M APACHE INDIANS, ARI- ZONA. COLLECTED BY Capt. JOHN G. BouRKE U.S. ARMY 20 PROCEEDINGS OF THE NATIONAL MUSEUM VOL, 73 Fic. 15.—Fir®-MAKING SET. Cat. No. 9555, U.S.N.M. NAavaHo IN- DIANS, NEW MExIco. COLLECTED BY EDWARD PALMER One thinks of the Navaho only with regard to their fine blanket weaving and silver working, so well presented by Dr. Washington Matthews in the reports of the Bureau of Ethnology, and does not consider their arts in other lines.” Thomas C. Battey, a Friend, long missionary among the Indians, kindly gives a description of the Kiowan fire-making process, not now practiced among them but shown to him as a relic of an abandoned art: A piece of very hard and coarse, rough-grained wood, per- haps 8 inches in length, 2 in width, and three-fourths of an inch in thickness is procured. In one side of this and near one edge several holes are made, about one-half an inch in diameter by five-eighths of an inch in depth, rounded at the bottom, but left somewhat rough or very slightly cor- rugated. In the edge nearest these holes a corresponding number of smaller and tapering holes are made, opening by a small orifice into the bottom of each of the larger ones. These are made very smooth. A straight stick, also of hard, rough-grained wood, about 8 or 10 inches in length, about the size they usually make their arrows or larger, is provided. Both ends of this are rounded, but one end is made smooth; the other is left slightly rough. The dried pith of some kind of reed, or more probably of the yucca, some fibers of the same loosely pre- pared like hackled flax, some powdered charcoal, I think formed by charring the yucca, and a piece of hard, thick leather, similar to sole leather, completes the outfit, which is carried in a leather bag made for the purpose. The first- described piece of wood is placed upon the knees of the operator with a quantity of the fibrous‘substance beneath it which has been powdered with charcoal dust; some of the latter is put into one of the holes and the rough end of the stick inserted; the other end is put into an indentation of the leather placed under the chin, so that a gentle pressure may be exerted. The spindle is then rapidly revolved by rolling it one way and the other between the hands. The friction thus produced by the rubbing of the roughened sur- faces ignites the fine coal dust, which, dropping as sparks of fire through the orifice at the bottom of the hole, falls into the dry fibrous preparation, thus igniting that, then by the breath blowing upon it a flame is produced and commu- nicated to some fine, dry wood and a fire is obtained. The whole operation occupies but a few minutes. In Mexico a number of the uncivilized tribes of the mountains continued the fire drill into recent times. It is probably used now as in other parts 10 Doctor Matthews’s mountain chant of the Navahos, in the fifth annual report (1883-84) of the Bureau of Ethnology, gives some very striking cere- monial uses of fire. No ethnologist should fail to read this important con- +ribution to science. ArT. 14 FIRE-MAKING APPARATUS—HOUGH of the world in religious rites. There is abundant data in the pictorial writ- ings of the ancient Mexicans as to the form and use of the simple fire drill, especially in the codex discov- ered by Mrs. Zelia Nuttall. A model of a drill after this codex is shown on Plate 2, figure 2a. One of the rudest fire-making ap- pliances in the Museum was collected by Prof. W. M. Gabb, at Talamanca, Costa Rica. The hearth is a rude billet of charred, black wood, resem- bling mahogany. It has central holes, with no gutter usually, though sometimes a shallow notch is cut on both sides of the fire-hole. The drill is a light branch, rather crooked, but dressed down roughly with a knife. Another hearth is of partly decayed, worm-eaten wood; with this a hard- wood drill can be used, the hearth wasting away instead of the drill. (Fig. 16.) The absence of any fire slot—that is, the use oi the central fire hole—is worthy of notice in this locality. I have only observed its use in various parts of the Eskimo area, from east Greenland to Kodiak; outside of this range I have not noticed it anywhere else among the present tribes of the world. From descriptions given it seems to have been practiced by the Caranchua Indians, a recently extinct tribe in Texas and Mexico. These specimens from Costa Rica are the crudest fire tools, not to be mere makeshifts, that have come to my notice or have been described in the literature examined. The Costa Rican Indians are very interesting in their preservation of several other arts that may justly be classed among the most ancient. One may be men- tioned, that of bark cloth making. 21 Fic. 16—FIRE-MAKING SET. Oat. NO. 15396, U.S.N.M. NATIVES OF TALA- | MANCA, COSTA RICA. W. M. GABB COLLECTED BY 22 PROCEEDINGS OF THE NATIONAL MUSEUM | vou, 73 Professor Gabb made quite a collection from Talamanca, but has not left any notes on these remarkable people, who are well worthy of the careful study of ethnologists. A curious modification of this central-hole plan is figured and described in Oviedo, folio 90, as occurring in Hispaniola—that is, the West Indies, Haiti, San Domingo, etc. He says that ‘‘two dry, light sticks of brown wood were tied firmly together, and the point of the drill of a particular hardwood was inserted between the two and then | worked.” H. Ling Roth ™ thinks that if one can judge from the illus- tration (which is a miserable one) in Benzoni’s work, the natives of Nicaragua also used three sticks in making fire. Benzoni, how- ever, says: ” All over India they light fire with two pieces of wood; although they had a great deal of wax, they knew no use for it, and produced light from pieces of wild pine wood. é From Oviedo’s description I am inclined to believe that the dust in which the fire starts was allowed to fall below on tinder placed beneath the hearth. (PI. 2, fig. 2.) The drill was sufficient for its time for the reason that there was at that period rarely necessity for generating fire; the art of fire preser- vation was at its height. The Cherokees, the most southerly of the Iroquois, James Mooney writes, kept fire buried in the mounds upon which the council houses were built, so that if the house was destroyed by enemies the fire would remain there for a year orso. ‘The Cherokees use the simple rotation apparatus, and, as far as Mr. Mooney can ascertain, never used the the pump drill. They have a tradition that fire originally came out of an old hollow sycamore tree (Platanus occidentalis). Capt. John Smith tells how the Indians of Virginia made fire. He says: Their fire they kindled presently by chafing a dry pointed sticke in a hole of a little square piece of wood, that firing itselfe, will so fire mosse, leaves, or anie such like drie thing that will quickly burn. Writing in the first quarter of the next century, Beverley says: They rubbed Fire out of particular sorts of Wood (as the Ancients did out of the Ivy and Bays) by turning the end of a Piece that is soft and dry, like a Spin- dle on its Inke, by which it heats and at length burns; to this they put some- times also rotten Wood and dry leaves to hasten the Work.'4 Loskiel says of the Delawares: Formerly they kindled fire by turning or twirling a dry stick with great swift- ness on a dry board, using both hands.'® 11 The Aborigines of Hispaniola, Journ. Anthrop. Inst., Gt. Britain and Ireland, vol. 16, p. 282. 12 History of the New World, Hakluyt Society, vol. 21, p. 151. 18 The Natural Inhabitants of Virginia. English Scholars Library, No. 16, p. 68. 14 History of Virginia, 1722, pp. 197, 198. ‘ _ ‘History of the Mission of the United Brethren, p. 54. London, 1794. art. 14 FIRE-MAKING APPARATUS—HOUGH 23 The Cherokees used for a drill the stalk of a composite plant (Senecio) and twirled it on a piece of wood. The art has long been out of common use, but they employed the wooden drill to make fire for the Green Corn Dance into the present century, though flint and steel was then in vogue. Sometimes they passed the bow over the drill. The tinder was of a fungus or dried moss. James Mooney collected this information from some of the older men of the tribe in North Carolina, who have retained the ancient customs and traditions, which the part of the tribe removed to the West has entirely lost. The Creeks (Muskogean stock) had a regularly authorized fire maker who early in the morning made fire for the Green Corn Dance. The apparatus that he made use of was four sticks placed end to end to form a square cross. This was oriented, and at the junction of the sticks new fire was made by friction.® The Choctaws (also Muskogean) of Mississippi, M. F. Berry writes, make fire in the following way: One stick of dry wood that has a hole in it, with a smaller hole at the bottom going through, is placed between the feet. Another piece made round and about 3 feet long is made to revolve rapidly back and forth between the hands in the hole, and the fire drops through the small hole below. When new fire was wanted for the Green Corn Dance or other purposes three -men would place themselves so that each in turn could keep the stick revolving without a stop until fire would drop down through the hole, which was nursed with dry material into a flame. This form of the fire hearth is not represented in the collections of the Museum; the only other description of a process closely like it was given by Thomas C. Battey, who observed it among the Kiowas. It was shown him at that time as a revival of the ancient method. The pierced fire hearth is somewhat impracticable, except in the Malay sawing method. In the rotary drill the small hole would come over the axis of least friction and heat. Unless provision was made for the dust to fall freely underneath by a double cone perforation worked from both sides the dust is likely to become obstructed and smother ‘the fire. It will be seen, too, that it departs very much from the simplicity of the usual fire drill in the fact that a hole must be made through the piece of wood, a matter of some difficulty before the introduction of iron awls. The Seminoles of Florida, the most southern Muskoki, have neglected the art of fire making by simple friction, unless at the starting of the sacred fire for the Green Corn Dance, says Clay MacCauley.” A ‘fire is now kindled either by the common matches, ma-tci, or by steel and flint. 16 Benjamin Hawkins’ Sketch of the Creek Country, 1798-99, pp. 68-72, cited in Pickett’s History of Alabama, vol. 1, p. 108. 1” Fifth Annual Report of the Bureau of Ethnology for 1883-84, p. 518. 24 PROCEEDINGS OF THE NATIONAL MUSEUM VOL. 73 Thus it is seen that wherever in the earlier period of the exploration in this country the observation has been made, the Indian, almost with out exception, was found to be using the friction apparatus con- sisting of two sticks of wood. Some tribes had improved on the working of the invention, while a very few others had perhaps arrived at the use of the higher invention of the flint and pyrites. Returning to the tribes of the wide central plains of our country, we find that the flint and steel soon displaced the fire sticks, except for religious purposes. The Mandans, of the great Siouan stock, were using flint and steel at the time of Mr. Catlin’s visit in 1832. There seems to be a great misapprehension among some of the writers on ethnology as to the general use of the bow drill among the Indians. In mentioning that the Sioux use the bow drill, Schoolcraft is quoted as authority. As a matter of fact the reference is to a ““made-up”’ figure of a bow-drill set, marked ‘‘Dacota.’”’ On the same plate there is a representation of an Iroquois pump drill that is obviously wrong. The lower part of the plate is taken up by a picture of an Indian woman (presumably Californian) pounding acorns in a mortar. To complete the absurdity the whole plate is entitled ‘‘ Methods of obtaining fire by percussion,’’ and is placed in the text of a questionnaire on the Californian Indians, opposite a description of the Californian way of making fire by twirling two sticks.!° Mr. Schoolcraft is not to blame for this state of affairs. In those days illustrations were not ethnological; they were ‘‘padding”’ gotten up by the artist. Nowhere in his great work does Mr. Schoolcraft describe either the Dakota or Iroquois drill. Among the northern Indians in central and northern Canada, however, the bow is used. Sir Daniel Wilson, in his work on Prehistoric Man, notes that the Red Indians of Canada use the drill bow. In August, 1888, at the meeting of the American Association for the Advancement of Science, at Toronto, he gave an account of the facility with which these Indians make fire. He said that at Nipissing, on the north shore of Lake Superior, while he was traveling in a pouring rain, and not having the means wherewith to light a fire, an Indian volunteered to light one. He searched around for a pine knot and for tinder, rub- bed up the soft inner bark of the birch between the hands, got a stick from a sheltered place, made a socket in the knot and another piece of wood for a rest for the drill, tied a thong to a piece of a branch for a bow. He put the tinder in the hole and rested his breast on the drill and revolved it with the bow and quickly made fire. 18 The George Catlin Indian Gallery. Smithsonian Report for 1885, vol. 2, p. 456. 1# Indian Tribes, vol. 3, p].28. 1851-1860. arr, 14 FIRE-MAKING APPARATUS—HOUGH 25 It is perhaps true that some of the Dakotas did use the bow at times, but it is not correct to place it as the customary tool of the whole stock. On the contrary, there is evidence that they used the simple means. Dr. J.Owen Dorsey writes: I was told in 1879 by the late Joseph La Fléche, that the Omahas, prior to the advent of the white men, made fire by using pieces of the “du-4-du-d-hi,” a grass (?) that grows in the Sand Hill region of Nebraska, near the sources of the Elkhorn River. One piece was placed horizontally on the ground, and a slight notch was cut at one end, wherein a few grains of sand were put. The other stick was held between the palms of the hands, with one end in the notch of the horizontal stick, and then rolled first in one direction then in the other till fire was produced. A fresh notch was made in the first stick whenever the old one became useless, and so on until it became necessary to procure a new stick. In the Green Corn Dance of the Minitaries, another Siouan tribe, the “corn is boiled on the fire, which is then put out by removing it with the ashes and burying them. New fire is made by desperate and painful exertion, by three men seated on the ground facing each other and violently drilling the end of a stick into a hard block of wood by rolling it between the hands, each one catching it in turn from the others without allowing the motion to stop until smoke, and at last a spark of fire is seen and caught in a piece of spunk, when there is great rejoicing in the crowd.” ”° The desperate exer- tion was not necessary, except in imitation of the Zufi fashion of wetting the drill to create sacred fire. It will be seen from these references given that the Sioux used the customary Indian method. Later, they may have used the bow to expedite the drill when the wood was intractable. The bow may have been borrowed from more northern tribes, the Algonquians are said to used it;*! Thomas C. Battey says that the Sac-Fox Indians (Algonquian stock) used a soft-wood drill and a hard-wood hearth. “The drill was worked by a bow and the fire caught on the end of the drill and touched to tinder.” Throughout South America the art of fire making with two sticks of wood is found to be as thoroughly diffused as it is in North America. Many of the tribes still use it; we may say that in all tribes the use of flint and steel was preceded by that of the sticks of wood. From Carib-Arawak tribes of British Guiana come simple jungle- fire drills consisting of peeled and dressed rods of soft-yellow wood. A bit of the black bark is left at the upper end of the drill as an ornament. The hearth has a fire pit near the end or in the smaller hearths near the middle. (PI. 2, figs. 1, 1a, Cat. No. 210445, British Guiana, coll. by J. J. Quelch, received from the Field Museum of * Smithsonian Report, vol. 2, p. 315, 1885. Sir Daniel Wilson. Prehistoric Man, vol. 2, p. 375. 86374-—28 4 26 PROCEEDINGS OF THE NATIONAL MUSEUM VOL. 73 Natural History, Chicago; drills 15-19 inches long (88-48.3 cm.) ; hearths 5.5-7 inches long (14-17.7 cm.).) The Guanchos, a mixed tribe of herders on the pampas of Venezuela, practice a peculiar way of fire getting. They select a pliant rod, place one end against the breast and the other against the block forming the hearth, held on a line with the breast. By pressing against the rod it is bent and turned rapidly around like an auger. This impracticable and no doubt very local method is described by Prof. E. B. Tylor.” In Brazil, in the province of Goyaz, the Chavantes, Cayapés, and Angaytés, use the simple fire drill. The Angaytés drill figured looks somewhat likethat ofthe Hopis. Itisusually 28cm. long forthe hearth, and for the drill 20 cm. They use the throat skin of the nandu, Rhea americana, for a tinder sack.. The Lenguas of the same Prov- ince use a strike-a-light consisting of a tinder horn, flint, and steel, which is also figured in the cited report. This set is very interesting, because from it we can say with certainty where the Lengua got it. The steel is the English ‘‘flourish’’ and the flint is the oval, old Eng- lish shape, probably broken somewhat by blows. The Lenguas, being on the line of travel, have adopted the method from English traders. In Rio de Janeiro the Indians had an angular recess at the back of their snuff mills for the purpose of making fire by friction.” The Ainos of Japan formerly used fire sticks, and are said even yet to resort to this method when they have no other means of getting fire. They use also flint and steel, adopted from the Japanese. A specimen (22257) is in the Collections of the Museum. The Japanese formerly used the simple drill; a few are yet pre- served and used in the temples on special occasions. A specimen is exhibited in the Imperial Museum at Tokio. Several years ago Mr. Stewart Culin, after difficult negotiations through Mr. Tsuda of the Tokio Museum, secured a specimen for the Smithsonian Institution from Baron Menge of the Idzumo shrine. The specimen is like that in the Imperial Tokio Museum from the Oyashiro Temple at Idzumo. It is a smooth, most accurately dressed plank 35.5 inches long (90.5 cm.), 4.75 inches wide (12 cm.), and 1.2 inches thick (3 cm.), of Chamaeocyparis obtusa wood. There are 42 fire pits on the two edges, generally 1 inch between centers.. The holes are drilled deeply and several calibers of drill have been used. The drill is a stem of Deutzia scabra with strong walls and large pith. In many of the holes a core is produced as in the tubular dril. This fire drill was used in the Harvest Festival. The inscription in well written characters is, In 22Darwin. Narrative of the Voyage of the Beagle. Vol. 3, p.458. Citedin Early History of Mankind, ane Emil Hassler. In Jahrbuch Mittelschweiz. Commerciel. Gesellsch. Arau, 1888, vol. 2, pp. 114-115. “4 Harper’s Monthly Magazine, vol. 7, p. 745. November, 1853. ART. 14 FIRE-MAKING APPARATUS—HOUGH part, ‘‘fire cuts wood Meiji 35 years November 27,’’ the probable date of the Harvest Festival of thanksgiving and production of new fire. (Pl. 3, mes. 1,)2:) In the Transactions of the Asiatic Society of Japan for 1876 (vol. 2, p. 223) a sacred fire hearth is de- scribed as having a step as observed in some Eskimo forms. This feature has been taken to be a usage required by the environment of the high north. In Japan, however, it may refer to the collecting and saving of the ground-off dust for healing or other esoteric purposes. In reference to the use of the sacred fire drill, the following data have been supplied by Romyn Hitch- cock: The fire drill is used at the festivals of the Oyashiro to pro- duce fire for use in cooking the food offered to the gods. Until the temple was examined officially in 1872 the head priest used it for preparing his private meals at all times. Since then it has been used only at festivals and in the head priest’s house on the eve of festivals, when he purifies himself for their celebration in the Imbidous, or room for preparing holy fire, where he makes the fire and prepares the food. The art of fire making by sticks of wood by the method of rotation is, or has been, as far as we know, universal on the African Continent as it was in the two Americas at the time of the discovery. It is presumable that the ancient Egyptians who had the bow drill used this implement and previous to its invention used the simple drill. The Somalis are a pastoral people of Arab extrac- tion, inhabiting a large maritime country south of the Gulf of Aden. Their fire sticks (fig..17) are pieces of branches of brownish wood of equal tex- ture, in fact the hearth has formerly been used as a drill, as may be seen by its regularly formed and charredend. This is another proof that it is not nec- essary that the sticks should be of different degrees of hardness. The grain of the wood, that of the drill being against and the hearth with the grain, in effect accomplishes what the use of wood of differ- ent qualities results in. The hearth and drill are in the neighborhood of 12 inches long, the former with a diameter of three-eighths of an inch and the latter one-fourth of an inch. They were collected by Dr. Charles Pickering in 1843. It is possible that the Somalis may have car- ried this method with them from Arabia. They Fig. 17.—Firn-MAK- ING SET... “CAT, No. 22) 99 as U.S.N.M. SoMAL- Is, EAST AFRICA, COLLECTED BY DR. CHARLES PICKERING. LENT BY PBHABODY Mo- SEUM THROUGH F. W. PuTNAM 28 PROCEEDINGS OF THE NATIONAL~MUSEUM VOL. 73 conquered this coast, driving back the earlier tribes inhabiting the country in the early part of the fifteenth century. Long since that time, and even now, some Arab tribes practice the drilling of wooden sticks to produce fire. In eastern equatorial Africa the Wataveita, says H. H. Johnston, generate fire in the common African way by rapidly drilling a hard- pointed stick into a small hole in a flat piece of wood. An interest- ing bit of custom comes out in connection with this art among the people. ‘“‘It is the exclusive privilege of the men, and the secret is handed down from father to son, and never under any conditions (as they say) revealed to women.’’ I asked one man why that was. ‘Oh,’ he said, ‘if women knew how to make fire they would become our masters.” The figure (fig. 18) shows how this people of the great Bantu stock make fire; this tribe visited by Mr. John- ston lives on the slopes of the beautiful Kilimanjaro Mountain. Fic, 18.—TAvVEITA AFRICANS MAKING FIRE, AFTER H. H. JOHNSTON. (Sex Jour. Soc. Arts, June 24, 1887) Fire-drill survivals in Asia are now difficult to find. In the ancient writings of India there are many references to the use of the two-stick apparatus. The collection contains a specimen from the Bhilis of the Rajputana, India. It consists of a hearth made from half of a split branch, while the drill is a slender shoot with bark left on. The speci- men was collected by Captain Lovett, of the English Army. (PI. 4, fig. 1, la, Cat. No. 167334; Edward Lovett; hearth, 18 inches long (45.7 cm.); drill, 20.5 inches long (51 cm.).) There was presented to the United States National’Museum by the Natural History Museum of Oxford, England, through Henry Balfour, a replica of a Hindu sacred fire making set. This consists of a squared % Journ. Anthrop. Inst., Great Britain and Ireland, vol. 15, p. 10, 1885. art. 14 FIRE-MAKING APPARATUS—HOUGH 29 block of wood 2% inches thick, a drill in the cutting end of which can be set a cylindrical piece of superior wood or bitt and another piece supplied when it is worn down, a supply of such pieces sawed in a block of wood, a nut to be held with two hands and having an inset of stone, and a cord for rotating the drill. These parts are named, respectively, ‘‘adhararani’”’; the lower, “‘arani’’; ‘‘mantha,”’ the spindle drill; ‘‘sauku,” set or bitt; and ‘“ uttararani,’’ wood for the bitt. In respect to remarks on the necessity of a slot for collect- ing the fire dust, it may be said that this drill is an exception, as it is found that a drill spindle of unusual diameter obviates the necessity of a slot, the tendency of the movement on a large periphery being to roll off masses of the dust which ignite at one or more places. (India, Cat. No. 150887, Natural History Museum, Oxford, England; length of spindle with bitt, 20 inches (51 cm.). (PI. 5, figs. 1-4).) The elaborateness of this fire-making set is an example of the tendency to complexity in cult apparatus. The turned drill and hand rest, the nut of iron, the iron pin, and bands on the drill naturally mark this set as modern in construction. This sacred fire drill is a model of the apparatus used in Brahmanic India by the fire priest, ‘‘agnihotrin,”’ for the daily sacrifices of milk and butter according to the Vedic rituals. The apparatus is set up on an antelope skin. Dr. W. L. Abbott brought from the Jakuns of the Endau River, Johore, a fire-drill set which, on account of the inaccessibility of these natives and the little known of them till lately, may be con- sidered rare. The equipment as carried by the Jakuns consists of a bundle of little rods of about the same diameter, any one of which may be used as a drill or hearth at choice. There is no separation of hearth and drill. This feature is noticed also among the South American jungle tribes. It will be seen that in this case there is no need for a slot, as the working of the drill upon a hearth of equal caliber cuts a slot in the wall of the hearth automatically. (Cat. No. 213441; Dr. W. L. Abbott; 12-20 inches long (31-54 cm.).) Another bundle of fire sticks, native name, ‘‘kooshuk,” from the Jakuns of the Rumpin River, Pahang, consists of rude rods, but having the same features mentioned in the Johore set except that the hearth pieces are slightly larger. (Cat. No. 219931; Dr. W. L. Abbott; hearth, 10 inches long (25 cm.); drill, 15 inches long (38 cm.).) The Malays of the islands of Nias, Pagi, and Simalur, East Indies, have the cord drill. Dr. W. L. Abbott procured several sets from these islands described as follows: The specimen from Sibabo Bay, Simalur Island, consists of a square piece of light yellow wood with used fire cavity in the middle, and adjoining a place with channel down the side of the block for a new working of the drill. The latter is a short, cylindrical, tapering 30 PROCEEDINGS OF THE NATIONAL MUSEUM VOL. 73 piece of the same wood. ‘The cord is twisted brown fiber. The top ofthe drill is smoothed off by wear against the nut, which was a piece of coconut shell. This set is small and compact for carrying on the person. (Pl. 6, fig. 4, 4a, Cat. No. 216340; Dr. W. L. Abbott; 5.5 inches long (14 cm.).) The apparatus from Pulo Simalur is larger than the set described above, and the drill is rotated with a strip of rattan. The wood is yellow, quite firm, and not hard. The hearth is squared and the drill is tapering as in the Pagi specimen. The native name of the fire set is “‘ludang.”’ (Pl. 6, fig. 3, 3a, Cat. No. 221833; Dr. W. L. Abbott; hearth and drill 13 inches long (33 cm.).) A general similarity with the Pagi and Simalur fire sticks is observed in the Nias specimens. It will be seen from the above that the fringe of islands off the south coast of Sumatra may be characterized as an area in which the cord drill is used. The speci- mens brought by Doctor Abbott have been chopped out of light- yellow wood, often showing worm holes. The cord is twisted brown bark. (PI. 6, fig. 1, 1a, Cat. No. 221831, Lafau, Nias; Dr. W. L. Abbott; hearth, 15 inches long (38 cm.); drill, 11 inches (28 cm.).) The north Pagi specimen is cut from very light wood, the hearth is squared. and grooves cut in the regular way, and the drill appears to have been used in the hands. (PI. 6, fig.2,2a. Cat. No. 221830; Dr. W. L. Abbott; drill, 10 inches (25.5 em.); hearth 12 inches (30.5 cm.).) Dr. Jesse R. Harris, United States Army, collected a fire hearth from the river district up the Rio Grande de Mindanao, P. I., pre- sumably of Mandayan origin. Doctor Harris says: “‘The fire drill works with a bow and is a good one.’”’ The hearth is of soft worm- eaten wood and has three rather large cavities with slots. It is like the Malay drills of Simalur, Pagi, and Nias, and much extends the range of the machine drills in these regions. The native name is Ool-in-sung-an. (Cat. No. 247525; 12 inches long (30.5 cm.).) The Museum collection has a specimen from the Battaks of Pala- wan, P. I., which consists of a cleft stick held open at one end by a small stone and deeply sawed where fire has been made. The thong is of rattan one-eighth inch in diameter formed by spiral turns into a ring which is worn as a bracelet by the Battaks when it is not needed for fire making. (Pl. 9, fig. 2, Cat. No. 326012, collected by Mrs. E. Y. Miller.) Mr. R. W. Felkin, * in a study of the Maidu or Moru negroes of Central Africa, 5° north latitude, 30° 20’ east longitude, describes the fire making of that tribe. Hesays that one piece of wood about the size and shape of a large pencil is rotated in a hole in a flat piece of hard wood. One man holds the wood steady whilst two others take 28 Proc. Royal Soc. Edinburgh, Session of 1883-84, p. 309. art. 14 FIRE-MAKING APPARATUS—HOUGH 31 it in turn to rotate the stick. This article of Mr. Felkin’s is com- mended to ethnologists as a model ethnologic study in method and research. That veteran and renowned explorer, Doctor Schweinfurth, gives the following: The method of obtaining fire, practiced alike by the natives of the Nile lands and of the adjacent country in the Welle system, consists simply in rubbing together two hard sticks at right angles to one another till a spark is emitted. The hard twigs of the Anona senegalensis are usually selected for the purpose. Underneath them is placed either a stone or something upon which a little pile of embers has been laid; the friction of the upper piece of wood wears a hole in the lower, and soon a spark is caught by the ashes and is fanned into a flame with dry grass, which is swung to and fro to cause a draught, the whole proceed- ing being a marvel which might well nigh eclipse the magic of my lucifer matches.”* The Gaboon negro fire set is one of the few observed having no dust channel cut on the hearth. The wood, however, is light and apparently first class for fire making with least effort. It resembles the hibiscus wood used by the Hawaiians and other Polynesians, a most admirable material in which fire could be raised without the presence of the usual slot. The hearth is a peeled stem 1 inch in diameter, with large cavity midway. The drills are smaller stems pared down at the end, as is usual. (Pl. 7, Fig. 1, la, Cat. No. 164671; Gaboon River, West Africa; A.C.Good; hearth 23 inches long (58.5 em.), drills, 21.5 and 24.5 inches long (55 cm. and 62 cm.).) Dr. W. L. Abbott collected specimens from the Wa Chaga negroes, Mount Kilimanjaro, East Africa, years ago. The hearth is a small worked-out block carefully shaped or rough, as shown in the figures. The hearth has a cord at one end for tying to the drill for con- venience in carrying. The drill is a straight, slender rod, with neatly cut hole at top for the hearth string. (PI. 7, fig. 4, Cat. No. 161824, Dr. W. L. Abbott; drill, 20.5 inches long (52 cm.); hearth, 5 inches long (13 cm.).) The use of worm-eaten wood is shown in the Wa Chaga hearth (pl. 7, fig. 2) and is evidence that wood is often conditioned for fire making by insects and fungi. Wa Chaga tinder is macerated bark. (Fig. 2b.) The drill is a peeled branch. (Fig. 2a, Cat. No. 151823.) Collected by Dr. W. L. Abbott in 1891. The Somali drill is a workmanlike tool consisting of two smoothed rods of equal length, the drill hole at one end of the rod of larger diameter. It will be noticed that the cut of the drill opens the rod into two V-shape cuts, insuring the perfect collection of dust. The owner of the set pierced the two rods and drew through a slender leather thong to bind them together when not in use. (PI. 7, Fig. 3; 27 The Heart of Africa, vol. 1, pp. 531, 532. New York, 1874. 32 PROCEEDINGS OF THE NATIONAL MUSEUM VOL. 73 Cat. No. 167094, Somalis, East Africa; William Astor Chanler; 24144 inches long (61.5 cm.).) It is an anomaly that the African, to light the fire to smelt the iron out of which he forges his remarkable weapons, should use sticks of wood. An Australian fire set from New South Wales, collected in 1890 by William Villiers Brown, is an example of the careful manner with which the natives prepared and conserved their fire tools. The hearth is cut from soft, worm-eaten wood in a presumably human outline. Three sticks with vascular pith are tied to the hearth. The cut of such sticks leaves a core in the center of the drilled cavity. (PI. 4, fig. 2, Cat. No. 168116; hearth, 1414 inches long, drills, 21 inches long.) 2. Eskimo four-part apparatus.—The arts of the Eskimo yield more satisfactory results to students of comparative ethnology than those of any other people. In all their range the culture is uniform; one finds this fact forced upon his observation who has examined the series of specimens in the National Museum, where they are arranged in order by localities from Labrador to southern Alaska. Prof. Otis T. Mason’s paper on Eskimo throwing sticks ** gave a new interpretation to this fact and powerfully forwarded the study of ethnology by showing the classificatory value of the distribution of an art. Professor Mason points out that though the Eskimo culture is uni- form in general, in particular the arts show the modification wrought by surroundings and isolation—tribal individuality, it may be called— and admit of the arrangement of this people into a number of groups that have been subjected to these influences. The Eskimo fire-making tools in the Museum admit of an ethno- graphic arrangement, but in this paper it is not found necessary to make a close study of this kind. From every locality whence the Museum possesses a complete typical set it has been figured and described. The Eskimo are not singular in using a four-part apparatus, but are singular in the method of using it. The mouthpiece is the peculiar feature that is found nowhere else. The drilling and fire-making set consists of four parts, as follows: The mouthpiece, sometimes a mere block of wood, ivory, or even the simple concave vertebra of a fish or the astragalus of a caribou. More often, they show great skill and care in their workmanship, being carved with truth to resemble bear, seals, whales, and walrus. The seal is the most common subject. The upper part is almost always worked out into a block, forming a grip for the teeth. The extent to which some of these are chewed attests the power of the Eskimo jaw. 2 Throwing sticks in the National Museum. Smithsonian Report, vol. 2, p. 279, 1884. art. 14 FIRE-MAKING APPARATUS—HOUGH 83 Frequently the piece is intended to be held in the hand, or in both hands, hence it has no teeth grip. In the under part is set a piece of stone, in which is hollowed out a cup-shaped cavity to hold the head of the drill. (These stones seem to be selected as much for their appear- ance as for their antifriction qualities. They use beautifully mottled stone, marble, obsidian, and ringed concretions. The drill is always a short spindle, thicker than any other drill in the world. Itis frequently of the same kind of wood as the hearth. The thong is the usual accompaniment of the fire drill. It is raw- hide of seal or other animals. The handles have a primitive appear- ance; they are nearly always made of bears’ teeth, hollow bones, or bits of wood. Sometimes handles are dispensed with. Warren K. Moorhead found some perforated teeth in an Ohio mound that in every respect resemble the Eskimo cord handles. They have also been found in caves in Europe decorated with concentric circles like those on the Eskimo specimens. ; The bows are among the most striking specimens from this people. They are pared down with great waste from the tusks of the walrus, taking the graceful curve of the tusk. The Museum possesses one 241% inches long. It is on their decoration that the Eskimo lavishes his utmost art. The bow does not lend itself well to sculpture as does the mouthpiece, so he covers the smooth ivory with the most graphic and truthful engravings of scenes in the active hunting life in the Arctic, or he tallies on it the pictures of the reindeer, whales, seals and other animals that he has killed. Professor Baird was interested more with these bows than with any other Eskimo products, and desired to have them figured and studied. The distribution of the bow is remarkable. It is not found south of Norton Sound, but extends north and east as far as the Eskimo range. The Chukchis use it,” but the Ostyaks use the ancient breast dril] .°° The bow is used by individuals in boring holes. Itis presumed that its use as a fire-making tools secondary, the cord and handles being the older. The difficulty of making fire is greatly increased when one man attempts to make it with the compound drill; at the critical moment the dust will fail to ignite; besides, there is no need of one man making fire; a thing that isfor the common good will be shared by all. Hence the cord with handles, which usually requires that two men should work at the drill, is as a rule used by the Eskimo. Though the Sioux, and some other North American tribes, made use of the bow to increase the speed of the drill, they did not use the thong with handles, nor was the bow common even in tribes of the * Nordenskiéld. Voyage of the Vega, vol. 2, p. 121, London, 1881. *Seebohm. Siberia in Asia, p. 109. 86374—28 5 s 34 PROCEEDINGS OF THE NATIONAL MUSEUM VOL. 73 Siouan stock that had attained to its use. (See remarks, p. 25.) The bow may be termed a more advanced invention, allowing one man with ease to bore holes. The hearth is made of any suitable wood. It is commonly stepped and has slots. The central hole with groove is also found. These hearths are preserved carefully, and fire has been made on some of them many times. The distribution of the central-hole hearth (see fig. 19) and the slot-and-step hearth (see fig. 32) is rather striking. The central holes are found in the specimens observed from the north coast of Alaska, insular British America, and Greenland, exclusively. The stepped hearth with edge holes and slots is by far the more common in west- ern Alaska, though the other method crops out occasionally; both ways are sometimes used in the same tribe. More often the central holes are bored on a groove (fig. 30), which collects the ground-off particles and facilitates ignition. Rarely fire is made by working the drill on a plane surface, in single, nonconnecting holes. The difference between these features is that it is found to be more difficult to get fire by a single hole without groove or slot than when the latter features are added. The powder forms a ring around the edge of the hole, is liable to be dispersed, and does not get together in sufficient amount to reach the requisite heat for ignition. Of course this is obviated when a second hole is bored connecting with the first, when the latter becomes a receptacle for the powder. It is found that these different ways are due to environmental modification, showing itself as remarkably in fire making as in any other Eskimo art. Both the stepped and central-hole hearth are dif- ferent devices for the same end. The step on the hearth is to keep the pellet of glowing powder from falling off into the snow, so universal in Eskimo land; hence, the simple hearth of primitive times and peoples of warmer climates has received this addition. The same reason caused the Eskimo to bore the holes in the middle of the block. By following the distribution of the center-hole method a clew may perhaps be gotten to the migrations of the Eskimo. From Labrador to Norton Sound, by the collections in the Museum, the center hole is alone used; south of Norton Sound both methods prevail, with a preponderance of the stepped-hearth species. The step seems to be an addition to the Indian hearth; the center is an independent invention. The operation of the drill is well told in the oft-quoted description by Sir E. Belcher. The writer can attest to the additional statement that the teeth of civilized man can scarcely stand the shock. He says: ART. 14 FIRE-MAKING APPARATUS—HOUGH 35 The thong of the drill bow being passed twice around the drill, the upper end is steadied by a mouthpiece of wood, having a piece of the same stone embedded, with a countersunk cavity. This, held firmly between the teeth, directs the tool. Any workman would be astonished at the performance of this tool on ivory; but having once tried it myself, I found the jar or vibration on the jaws, head, and brain quite enough to prevent my repeating it.*! The ethnographical study of the Eskimo fire drill begins with Labrador, including Greenland and following the distribution of the SS ‘| N Ni \ eat 20 Fics, 19-21.—FIRE-MAKING SET AND EXTRA HEARTH, CaArT. No. 10258, U.S.N.M. FROBISHER BAY. COLLECTED BY C. F. Hall. 20, Moss IN A LEATHERN CASE. CAT. No, 10191, U.S.N.M. COLLECTED BY C. F, HALL people among the islands and around the North American coast to Kodiak Island and the Aleutian chain. The following is an interest- ing account from Labrador, showing what a man would do in the exigency: He cut a stout stick from a neighboring larch, and taking out the leather thong with which his moccasins were tied, made a short bow and strung it. He then 4 Trans. Ethnol. Soc., p. 140, London, 1861. 36 PROCEEDINGS OF THE NATIONAL MUSEUM VoL. 73 searched for a piece of dry wood, and having found it, cut it into shape, sharp- ened both ends, and twisted it once around the bowstring; he then took a bit of fungus from his pocket and put it into a little hole which he made in another dry piece of wood with the point of the knife. A third piece of dry wood was fashioned into a handle for his drill.** Eskimo in other localities often use such makeshifts. Cup cavities are often observed in the handles of knives and other bone and ivory tools where they have used them for heads of the fire drill. Cumberland Gulf is the next locality to the northward. There are several specimens in the collection from this part of Baffin Land, procured by the famous explorer, Capt. C. F. Hall, and the less known, but equally indefatigable Kumlein. The fire-making imple- ments from Cumberland Gulf have a markedly different appearance from those of any other locality in the Eskimo area. They have a crude look, and there is a paucity of ornamentation unusual among this people. The drill bow is one of the things which the Eskimo usually decorates, but these bows have not even a scratch. Fic. 22.—Boxinc ser. Cat. No; 84114, U.S.N.M. CUMBDRLAND GULF. CoLLectep’ BY L. KUMLEIN It can be inferred. that in Baffin Land more unfavorable condi- tions prevail than in southern Alaska. It must be this cause, cou- pled with poor food supply, that have conspired to make them the most wretched of the Eskimo. The hearth (fig. 19) is of drift oak. It was collected at Frobisher Bay by Captain Hall. It has central holes, and appears to be very unfavorable wood for fire making. A skin bag of moss (fig. 20) is for starting the fire. The block hearth is also from Frobisher Bay. (Fig.21.) It is an old piece of hemlock, with two central communi- cating holes. The mouthpiece is a block of ivory. Another mouth- piece is a.bit of hardwood soaked in oil; it was used with a bone drill having an iron point. A very small, rude bow goes with this set. (Fig. 22.) #2 Hind. Labrador, vol. 1, p..149. art. 14 ' FIRE-MAKING APPARATUS—HOUGH 37 Our knowledge of eastern Greenland has been very much increased by the explorations of Holm and Garde, who reached a village on the east coast never before visited by a white man. Extensive col- lections were made, both of information and specimens. In refer- ence to fire making, Mr. Holm reports: They make fire by turning a hard stick, of which the socket end is dipped in train oil, very rapidly around by means of a sealskin thong with handles. This stick is fixed at one end into a head set with bone, and the other end is pressed down into a cavity on the lower piece of wood. (Fig. 23.)' Therefore there must be two persons in order to make a fire. One turns the drill with the cord while the other presses it down on the hearth; both support the block with their feet. As soon as the dust begins to burn they fan it with the hand. When it is ignited they take it and put it into dried moss (sphagnum); blow it, and soon geta blaze. In this way they make a fire in an incredibly short time. % In the preliminary report, Mr. Holm gives the time at almost less than half a min- ute. It was made by the Eskimo, Illinguaki, and his wife, who, on being presented with a box of matches, gave up their drill, saying that they had no further use for it. In the same report Mr. Fic. 23,—FIRE-MAKING SET. ANGMAGSALIK EsKIMO, Holm gives an interesting EASTERN GREENLAND. COPIED FROM GQ. HOLM’s note. Hesays: ETHNOQLOGISK AF ANGMAGSALIKERNE, 1887 This fire apparatus is certainly better developed than that which has been described and drawn by Nordenskiéld from the Chukchis.* The principle is the same as the Greenlander’s drill, which they employ for making holes in wood and bone, and which is furnished with a bow and mouthpiece.® (Fig. 24.) The central holes of this hearth are worthy of note, occurring in the farthesti eastern locality of the Eskimo, and in Labrador... ’ Western. Greenland.—The, material, in the Museum from western Greenland is very scanty. The southern coast has been settled for so long a time that the Eskimo and many of their arts have almost become extinct. No view of fire making in Greenland would be com- plete without Davis’s quaint description of it, made 300 years ago, but it was the upper end of the spindle that was wet in trane. A Greenlander ‘‘begaune to kindle a fire in this manner: He tooke a % Danish Umiak Expedition to Eastern Greenland, 1888, p.28. Pl.14 contains the figure. % Voyage of the Vega, vol. 2, p. 126. % Danish Umiak Expedition. Preliminary Report, p 208. This seems scarcely what would beinferred from the development of these inventions. 38 PROCEEDINGS OF THE NATIONAL MUSEUM VOL. 73 piece of boord wherein was a hole half thorow; into that hole he puts the end of a round sticke like unto a bedstaffe, wetting the end thereof in Trane, and in a fashion of a turner with a piece of lether, by his violent motion doeth very speedily produce fire.” * Eskimo graves and village sites yield evidence also that the fire- making tools were not different from those at present used higher north along the coast and on the east coast. Doctor Bessels, speaking of Itah Eskimo of Foulke Fiord in Smith Sound, says: ‘‘The catkins of the Arctic willow are used as tinder to catch the sparks produced by grinding two pieces of stone. Also the widely diffused ‘fire-drill’ is found here; the spindle is held between a piece of bone and a fragment of semi-decayed wood, and is set in Fig. 24—BoORING SET. (ANGMAGSALIK ESKIMO, EASTERN GREENLAND. G. HOLM’s ETHNOLOGICK OF ANGMAGSALIKERNE) motion by the well-known bow, and is turned until the wood begins to ignite. ’”’* The “fire bag” is an accompaniment to all sorts of fire-making apparatus. The fire bag shown (fig. 25) was collected by Captain Hall, at Holsteinberg, western Greenland in 1860. It is made of sealskin, and is a good specimen of the excellent needlework of these Eskimo. It was used to carry, more especially, the fire drill and tinder which require to be kept very dry. There is a wide gap in the collections of the Museum between the locality of the specimen just mentioned and the fire hearth from the Mackenzie River. (Fig. 26.) This specimen is from Fort Simpson presumably, where B. R. Ross collected. It is said to be difficult to discriminate the Eskimo from the Indian on the lower Mackenzie. This hearth may be Indian, as it has that appearance; besides, no % Hakluyt Society, vol. 3, p. 104. 8’ Die amerikanische Nordpol-Expedition, p. 358, Leipzig. an. 14 FIRE-MAKING APPARATUS—HOUGH 39 Eskimo hearth yet observed has side holes and slots like this with- out the step. The Indians of this region are of the great Athapascan stock of the North. The close resemblance of this stick to the one from the Washoans of Nevada has been commented upon. (See fig. 6, p. 14.) There is a very fine old central-hole hearth from the Macken- zie River, collected also by Mr. Ross. It is a rough billet of branch wood, cut apparently with an ax, or hatchet. (Fig. 27.) It is semi- decayed and worm eaten. It has 10 central holes where fire has SSS SSS OS Fig. 25.—Firp BAG. Cat. No. 10128, U.'S.N.M. ESKIMO OF HOLSTEINBERG, WEST GREEN- LAND. COLLECTED BY CAPT. C. F. HALL been made; they are quite deep, forming a gutter in the middle of the hearth. There is, as can be seen, no need of a groove, as the dust falls over into the next hole, collects in a mass, and ignites. The Anderson River set is a very complete and interesting outfit. It was collected many years ago by C. P.Gaudet. The parts are small for convenience of carrying. It is the custom of those who live in snow-covered regions to wrap the drill and hearth together very carefully to keep them dry, as these are the essential parts of the apparatus. It does not matter about the mouth-piece or bow. In this example there is a groove cut along the bottom of the hearth in order to facilitate tying the drill and hearth securely together. The 40 PROCEEDINGS OF THE NATIONAL MUSEUM VOL. 73 hearth is a square block of soft wood with three central holes. (Fig. 28.) The other parts of this set are also worthy of consideration. The mouthpiece is set with a square piece of black stone. The part held in the mouth is very much chewed. One of the wings has a hole for tying, as has the hearth. This is an unusual Eskimo precaution to prevent small objects from being lost in the snow. The drill is short, being only 7 inches long. The bow is the fibula of a deer, pierced at each end for the frayed - thong of sealskin. It has a primitive look, but it admirably serves its purpose. The Point Barrow set was collected by the most successful expedition under charge of Lieut. P. H. Ray, United States Army. The knucklebone of a deer serves as a mouthpiece, the cup cavity and its general shape fitting it for the pur- pose admirably. The drill is regularly made of light pine wood; it is_ slightly smaller in the middle. The hearth is a rudely rounded piece of pine. A fragment has been split off, and on this surface a groove has been cut and three fire holes bored along it. The thong is without handles; it is used to tie the parts together when they are not in use. A bunch of willow twigs, the down of which is used as tinder, is also shown. (Fig. 29.) This set is especially interesting, because it shows the degeneration ofanart. The fire drill is so rarely used at Point Barrow, John Mur- doch says, that it was not possible to get a full set devoted to that ae elie ownn purpose. Those here shown are a maxine ser (on, Makeshift. The method only sur- Fie. 27.—Lower part ONE END IS GUM vives by the conservatism ofafew 2. Gis No FOR CHMBPNT). SET. Cat. No. Car. No. 1978, old men of the tribe, who still cling 5 oe ae U.S.N.M. Mac- SKIMO OF c- Savets Bawa ee old usages. Oneofthese made xpyzimriver,Brit- British Cotum- the drill for Lieutenant Ray, tell- ta CoLuMBraA. BIA. COLLECTED . : 4 C c COLLECTED By B. By B. B. Boss ing him that itwasthe kind usedin _ R. Ross a ArT. 14 FIRE-MAKING APPARATUS—HOUGH Al old times. It seems primitive enough; the knucklebone might well have been the first mouthpiece. The Eskimo farther east sometimes use a fish vertebra for the same purpose; one from the Anderson River hasthis. The cord without handles is undoubtedly the earliest form also. The small wooden and bone mouthpieces of the Eskimo east of Point Barrow to Cumberland Gulf seems to be copies of the deer knucklebone. Another prim- itive adaptation is found in an Anderson River bow, which is made of the fibula of a deer. (See fig. 28.) The fire-making drill collect- ed from the Chukchis by the Vega expedition in the Cape Wan- kerem region, in northeastern Siberia, about the same latitude as Point Barrow, is figured in Nordenskiéld’s_ report.*® It is worked by a bow, and the drill turns in a mouthpiece of a deer astragalus like the Point Barrow specimen. The block has central holes, with short grooves running into each one. Nordenskiéld’s description of the manner of making fires is very detailed. He records that the ‘‘women appear to be more accustomed than the men to the use of this implement.’’ He gives also a most interest- ing observation on the use of a weighted pump drill among the Chukchis. The Chukchis also use % : FIG. 28.—FIRE-MAKING SET. Cat. No. 1327, flint and steel. *® U.S.N.M. EsKIMO OF ANDERSON RIVER, The drilling set from Point Bar- BrItIsH COLUMBIA. COLLECTED BY C. P. GAUDET row shows the appearance of the parts of the fire drill if we substitute the round stick for the flint drill. Some of the old drill stocks are pointed, with finely chipped flint heads. The length of these points varies from 2 to 4 inches; GF Ss. serra ae GH renhae = 8 Voyage of the Vega, London, 1881, vol. 2, pp. 121, 122. .% Idem, vol. 2, pp. 120, 121. 42 PROCEEDINGS OF THE NATIONAL MUSEUM VOL. 73 the transverse section of one would bea parabola. They are in general more finely wrought | las than any of the prehistoric drills 1 FNP found in various localities all Wh MAMMA ii over the world. Prehistoric man was an adept in the art of | drilling stone, bone, and shell; the stone tubes, some of them 18 inches long, bored very truly, ii? are triumphs of the American | Indians. Without doubt the prehistoric drill points were mounted like the Eskimo spec- | imen, and were, perhaps, twirled | between the hands, the almost Ni —— universal method of using the | fire drill. Japanese carpenters drill holes in this way. The winged mouthpiece is also a good example of work- | manship. It is set with a mot- ly \ tled, homogeneous stone that is | tolerably soft, which gives a Nd | minimum friction. This stone i) is much affected by the tribes { over quite an extent of coast for i labrets, etc. It is probably an i | article of trade as are flints. The bow is of walrus tusk, accu- rately made, but poorly engrav- | ‘ ed in comparison with the life- i like art work of the southern | Eskimo. 4 Another drilling set is from | - Sledge Island. The Museum i/ has no fire-making specimen from this locality. The drill | stock is set with a point of jade- iy ite lashed in with sinew cord. The bow is of walrus ivory; it Fic. 29.—FIRE-MAKING SET (WITH MOUTHPIECE OF is rounded on the belly and flat DEER’S KNUCKLEBONE, THONG, AND TINDER OF WIL- ON the back. All Eskimo bows LOW CATKIN). Cat. No. 89822, U.'S.N.M. Eskimo D : PoIntT BARROW, ALASKA. COLLECTED BY P. H, of Ivory have a like curve, no Bay doubt determined by the shape an. 14 FIRE-MAKING APPARATUS—HOUGH 43 of the walrus tusk. In another, the most common form of the bow, its section is nearly an isoceles triangle, one angle coming in the cen- ter of the belly of the bow. The head is intended to be held in one or both hands; it agrees in form with the rude St. Lawrence Island heads. Dr. E. W. Nelson collected at Unalakleet, in Norton Sound, a fire drill, and the native names of the parts. The name of the set is “66- j66-etitat’’; the mouthpiece, “nd-ch66-tuk”’; the drill, “§6-j66-ga-tuk”’; the hearth of tinder wood, “‘athl-uk’’; the bow, ‘‘arshu-low-shuk- pish-ik-sin-uk.”’ This is a complete set (fig. 30) in first-rate order. The hearth hascentral holes along a deep median groove. Its bottom is flat, and it is rounded off on the sides and ends. All the parts are of pine wood, decorated in places with red paint. The drill is quite long, much longer than in any Eskimo set observed. It resembles more the Indian drill for rubbing between the hands. The bow is of wood, which also is quite the exception in other Eskimo regions, where it is of ivory. There are many bows of antler from Norton Sound in the Museum, some of them skillfully and truthfully en- Fic, 30—FIRE-MAKING SET (HEARTH SHOWING MEDIAN GROOVE). CAT. No. 33166, U.S.N.M. Eskimo oF Nor- TON SOUND, ALASKA. COLLECTED BY E. W. NELSON graved. The mouthpiece is plain; not very well made. It is set with a square block of marble. It has the usual hole in one of the wings for the passage of a thong. PROCEEDINGS OF THE NATIONAL MUSEUM VOL. 73 Fig. 31.—LOWER PIECE OF FIRE-MAKING SET (HEARTH). Cart. No. 39601, U.S.N.M. ESKIMO OF CAPE VANCOU- VER, ALASKA. NELSON COLLECTED BY E, W. Cape Vancouver is represented by a fine old hearth. This object has evidently been prized by its owner; it has had two rows of fire holes (fig. 31), one row bored on the step in front of the first holes made; some of the holes are bored clear through. The reason why this was valued is because the wood is so tindery that it is easy to make fire upon it. Chalitmute, in the Kuskokwim region, on the northern side of the bay of that name, opposite Nunivak Island, is the next locality southward to be considered. The parts of this set are exceptionally well fin- ished... The hearth (fig. 32) isstepped. It has four holes prepared for use; on one, fire has been made. The drill is unusually thick. The mouthpiece has no teeth grip, and there is no evidence that it was ever held in the mouth. It is intended to be held in the hand. This mouthpiece is set with an oval socket stone of black obsid- ian, ground down into facets and polished. The cord handles are fine, large teeth of the sea lion. The centers of the circles so characteristic of Eskimo art are inlaid with wood. The holes for the drill cord are narrow; they must have been dug through with a sharp, narrow instrument. As before remarked, this is the region where the hand rest is more used than the mouthpiece, and the bow is not used at all. The fire-making set from the Togiak River was collected in 1886 by Sergt. I. Applegate, of the United States Signal Corps. Kassianamute, from which village it comes, is in the Bristol Bay region, but this set has a different appearance from the former outfits. (Fig.33.) The hearth is a block of wood worked out at one end intoahandle. It is remarkable in having central holes not connecting, and with no connecting grooves. In this it closely resembles the block from east Greenland. (Fig 23.) This hearth is of soft, tindery wood, and doubtless when the holes be- came too deep to allow the powder to mass FIRE-MAKING APPARATUS—HOUGH 45 ART. 14 = SS a ah es EskKIMO OF CHALITMUTE, KUSKOKWIM Nos. 36325 AND 37961. COLLECTED BY E. W. NELSON Cat. Fic. 32.-FIRE-MAKING SET. REGION, ALASKA. 46 PROCEEDINGS OF THE NATIONAL MUSEUM VOL. 73 around the edge the upper part of hearth was scraped down. The mouthpiece is large and is in the form of a seal. It has only a shallow, crescentic teeth grip; from the size of the mouthpiece, its shape, and the absence of a block to fasten between the teeth it must have been = I FIG. 33.—FIRE-MAKING SET. Cat. No. 12750,U.S.N.M. ESKIMO OF KASSIANAMUTE, TOGIAK REGION, ALASKA. COLLECTED BY I. APPLEGATE nearly always held in the hand of one of the operators. It is set with a round pebble, mottled with green. The cord is a thong of rawhide with handles of wood. The next locality is Koggiung, on the southern shore of Bristol Bay, near its head. Two sets are shown from this locality. From the hearths it will be seen that both fire slots on the side and 47 These sets are called ‘‘nt-tshiin.’’ FIRE-MAKING APPARATUS—HOUGH ART. 14 center holes are used here. (Fig. 34.) The apparatus shown in Figure 34 has the stepped hearth. Fic. 34.—FIRE-MAKING SET (HEARTH WITH STEP AND FIVE SLOTS). CAT. No. 127819, U.S.N.M. KoGGIUNG, BRISTOL BAY, ALASKA. COLLECTED BY W. J. FISHER The mouth- Both drill and hearth apparently have been made forsale. piece is a good one, set with a large socket piece of a black stone with green mottlings. It is much used by The teeth grip is This stone is tolerably soft. the Bristol Bay Eskimo for making labrets, etc. 48 PROCEEDINGS OF THE NATIONAL MUSEUM VOL. 73 very shallow. The hearth (fig. 35) is of a very peculiar shape; only one other has been noticed like it. The wood is of the best kind, and fire has been made on it a number of times. In several places the holes have been bored clear through. The mouthpiece bears no Sa SE Fig. 35.—FIRE-MAKING SET (HEARTH WITH CENTRAL HOLES AND END STEP). CaT, No, 1278198, U.S.N.M. Koceiune, Briston Bay, ALASKA, COLLECTED BY W. J. FISHER evidence that it has been held between the teeth. It is highly prob- able that fire was made on these outfits more often by two persons, one holding the mouthpiece, or rest, and fanning the flame, the other pulling the cord. This must be the method in Bristol Bay. Neither ant. 14 FIRE-MAKING APPARATUS—HOUGH 49 the true mouthpiece nor any bow has been procured by the Museum from this interesting region, from whence there are copious collec- tions of ethnological objects. The cords without handles are worthy of notice. Another set from Bristol Bay is said by its collector, Charles McKay, to be used by both Eskimo and Indians. It isa very valuable outfit because of its completeness. (Fig. 36.) The hearth is a rounded piece of wood with four large holes opening by slots onto the step. The drill is a thick, tolerably hard piece of close-grained wood like the hearth. The mouthpiece has no regular block for the teeth grip, but has a crescentic gash on each side instead. It is set with a socket of a rock resembling marble. Nearly all the mouthpieces south of Norton Sound are in the shape of seals or other long animals. Cord handles are used attached to a thick thong of buckskin. Fungus is used for tinder and a blaze is started with cones of the larch. These are kept in the box, the lid of which is tied on with a thong. Kodiak, the lowest limit of the western Eskimo, is as far south as the four-part fire drill extends by specimens in the Museum. (Fig. 37.) The hearth is of cedar wood with three central holes with a connecting groove. Itis neatly finished. The drill is also of cedar and bears the marks of the use of a thong; the top has also been used in the socket of a rest. The drill approaches in length those used for twirling between the hands by the Indians. While the Aleutians use flint and steel, or a stone containing quartz and pyrites, struck against another stone, they still make use of the four-part drill at certain times. Hunting parties, says L. M. Turner, carry the drill to use when their matches run out. It takes two men to work it, one holding the hand rest and the other pulling the thong. The spindle is made of harder wood, so as to wear the light dust which ignites, from the hearth. A moment only is necessary to get fire; this is fed with tinder made of willow catkins and powdered charcoal. Sometimes, in order to get fire, they hold tinder at the mouth of a gun and ignite it by firing off a light charge of loose powder. Possessed of four methods of getting fire, the Aleutian is superior to more fortunately situated people who depend wholly on matches. Pump drill—It appears probable that the pump drill is of Asiatic origin as there are frequent occurrences of this implement in Asia. There is also a pretty uniform distribution of the pump drill across Siberia. Some western Eskimo and Indians use the pump drill for fire making, and it is possible that it was disseminated in Nearctic Canada and the United States at an early period and surviving now in only a few places, as among the Iroquois. 50 PROCEEDINGS OF THE NATIONAL MUSEUM vou. 73 The Iroquois are unique in the United States in making fire with the pump drill. It is well known that several American tribes used the pump drill for drilling beads and for other light, fine work Seg 6 a ‘Gans {) 5 f ) ¥ Hy Ml i iN Hi I i) ih Hl i i av ‘, \ Wit mn Bi i! Mg \ Mn ; J LA 4 (4 : i Ny if , t,t an. By Pi i Ih 1 pil [ ‘J 1 AA, ‘lg ip. ’ ft! ise ‘i bie ol ve Fic. 36.—FIrn-MAKING SET. Cat. No. 55938, U.S.N.M. ESKIMO OF BRISTOL Bay, ALASKA. COLLECTED BY CHARLES McKay requiring little friction and pressure. To render the pump drill effective for fire making it was necessary to increase the size and add a heaver balance wheel. Even then the pump drill is a clumsy fire art. 14 FIRE-MAKING APPARATUS—HOUGH 51 producer and hardly a practical tool for the purpose. How long the Iroquois have had the fire drill is conjectural, but observers as early as 1724 do not mention its use, speaking only of the simple two-stick drill. Il. FIRE MAKING BY SAWING Prof. Alfred Russell Wallace in his work entitled ‘“‘The Malay Archipelago,” (p. 332) has noted the method by sawing two pieces of bamboo; a sharp-edge piece like a knife is rubbed across a convex piece in which a notch is cut, nearly severing the bamboo (fig. 38); after saw- ing across for awhile the bamboo is pierced, and the heated particles fall below and ignite. The Ternate Malays and the Tungaras of British North Borneo ® have improved upon this by striking a piece of china with tinder held with it against the outside of a piece of bamboo, the siliceous coating of the latter yielding a spark like flint. Both of the methods mentioned are in use at different points in the area affected by Malay influence. The Chittagong hill tribes, on the east- ern frontier of British India, use sand on the sawing knife to increase the friction.*! The Karens of Burma, Dr. R. M. Luther informs the writer, hollow out a branch of the Dipterocarpus tree like the lower piece of bamboo spoken of, cut a transverse notch, and saw across in it with a rubber of ironwood. The wood fibers ground off ‘form the tinder; the coal is wrapped up in a dry leaf and swung around the head till it blazes. It takes only two or three minutes to get a blaze this way. Bearing upon the origin of this method of “sawing in these localities, nature 18 ° yc. 37—-Lowsn Piece AND SPINDLE 0 en ee EE eee OF FIRE-MAKING SET. Cat. No. ®D.D. Daly. Proc. Roy. Geog. Soc., p. 10, 1888. 72514, U.S.N.M. Eskimo OF 41 Capt. T. H. Lewis. Hill tribes of Chittagong, p. 83. Cal- Kopiak ISLAND, ALASKA. COL- cutta, 1869. LECTED BY W. J. FISHER 52 PROCEEDINGS OF THE NATIONAL MUSEUM VOL. 73 alleged to suggest the way and to repeat the process that would give to fireless man the hint. Dr. W. T. Hornaday relates that many fires are started in the jungle by bamboo rubbing together in a high windstorm. The creaking is indiscribable; the noise of the Fig. 38.—MALAY FIRE sTICKs. Cat. No. 129775, U.S.N.M. MopDELs IN BAMBOO MADE BY Doc- TOR HOUGH AFTER A. R. WALLACE’S DESCRIP- TION. THE MALAY ARCHIPELAGO, P, 332. rasping and grinding of the horny stems is almost unendurable. In many tribes it is found that often there is more than one method of fire-making practiced. For instance, in Borneo, as we have seen, the Tungaras use the sawing method, the Saribus Dyaks the ‘“‘besiapi,’’ or fire syringe, a most interesting fact,” other Dyaks the rotary drill,* while the Rev. Dr. Taylor says that the Dyaks are acquainted with the use of the bow and string and the upright stick and cord (pump drill). In connection with all these methods probably flint and steel were used. So in Australia, while the rotary drill is the usual way, some tribes have acquired the art of produc- ing fire with knife or rubber—that is, the sawing method presumably under foreign influence. * The specimens of fire saws in the Museum come from the Philippines, collected 25 years ago. They indicate that a node of bamboo from 13 to 15 inches long was sectioned longitudinally for the lower piece and the saw made by splitting off a narrower piece and sharpening one or both edges. In the middle of the hollow of the lower piece fibers are torn up, forming a groove which reduces the thickness of the wall of the bamboo, allowing the saw to cut through to the tinder affixed in the groove and held in place by loose fibers. The saw is worked across the bow of the bamboo hearth at right angles over the spot where the tinder had been previously located. Sometimes this is reversed by holding the saw firmly edge up and rubbing the hearth on it. The use of the fire saw was quite general in the Phil- ippines among all the tribes, while the hand drill or or plow were not used so far is known in the entire archipelago. The specimens shown are from Min- danao andLuzon. (PI.8, figs.,1, 1a, 2,2a,Cat. No, 216,716; Col. F. F. Hilder ; 13.5 inches long and 15 inches long (34.5 cm. and 38 cm.). 42 The American Anthropologist, vol. 1, No. 3, p. 294. Washington, 1888. “&J.G@.Wood. The Natural History of Man, vol. 2, p. 502. “(R. Brough Smyth. The Aborigines of Victoria, vol. 1, p. 393. London, 1878. kn 14 FIRE-MAKING APPARATUS—HOUGH 53 Ill, FIRE MAKING BY PLOWING One of the most marked of fire-making methods in its distribution is that pursued by the Pacific Islanders, confined almost entirely to the Polynesian cultural area. It hasspread to other islands, however, being met with among the Negritos of New Britain: They rub a sharpened piece of hard stick against the inside of a piece of dried split bamboo. This has a natural dust that soon ignites. They use softwood when no bamboo can be procured, but it takes longer to ignite. The flame is fed with grass.* There is a close connection between the Malay sawing method and this, as there is a decided Malay preponderance in the make-up of the pop- ulation of the islands. The fire sticks shown (fig. 39) were procured by Harold M. Sewall, at Samoa, and deposited in the Museum by him. The wood is a light corky variety, characteristic of the Parite tiliaceum, which is used for this pur- pose at Tahiti and many other islands. The rub- ber may be of some hardwood, although fire may be made by means of a rubber of the same kind of wood as that of the hearth, though no doubt it requires a longer time to make fire if this is done. In the Sandwich Islands, Franklin Hale Austin, secretary of the King at that period, says that the rubber is of “koh” or ‘‘ohia’’—that is, hard- wood—and the hearth of ‘‘koh,”’ or softwood, and the friction is always in softwoods; this is true, I believe, everywhere this method is practiced, is in spite of the fact that a soft rubber on hardwood will asnwer as well. Lieut. William I. Moore, United States Navy, gave the writer a complete description of the manipulation of the Samoan fire-getting apparatus. The blunt-pointed stick is taken between the |. 39 ymewaxine clasped hands, somewhat as one takes a pen, and sticks (a suowina projected forward from the body along the groove [30673 Gene at the greatest frictional angle consistent with the SoMoa. Drrostrep By forward motion which has been found to be from) 40° to 45°. Kneeling on the stick the man forces the rubber forward, slowly at first, with a range of perhaps 6 inches, till the wood begins to be ground off and made to go into a little heap at the end of the groove; then he gradually accelerates the speed and moves with a shorter range until, when he pushes the stick with great velocity, the ‘SW. Powell. Wanderings in a Wild Country, p. 206. 54 PROCEEDINGS OF THE NATIONAL MUSEUM VOL. 73 brown dust ignites. This is allowed to glow and if it is required to be transferred to dry leaves or chips of wood it is done by means of a tinder made of frayed or worn tapa cloth. The groove (fig. 39a) is the most characteristic feature of this | apparatus, there being apparently no definite form of implements for this purpose. Fire is made on any billet of dry wood that is avail- able. It is not necessary to cut a slot, or even a groove; the hard- wood rubber will form one, so that there is no more need of apparatus than among the Navahos, where two bits of yucca stalk collected near by form the fire tools. That making fire by this way is difficult to those inexperienced in it is not strange. Mr. Darwin found it quite so, but at last suc- ceeded. The Samoan gets fire in 40 seconds, and so great is the friction and the wood so well adapted that Mr. Austin, before quoted, says it sometimes actually bursts into flame. The Australians in some parts use a method very much like the one described. They rub a knife of wood along “ a groove made in another stick previously filled with tinder.*’ Fire thong—While there is no apparent connection between the fire drill and the fire saw, plow, and thong, there is an approximation in method of operation among the three latter—that is, the fire saw and thong are in close relationship, the plow is related but stands farther away, while the drill is unrelated. Henry Balfour has most successfully monographed the fire thong.* The method has been found in use in southeastern Asia and the Asiatic islands; in New Guinea; West Africa, and western Europe. At first sight it would seem necessary to limit the fire thong method to the area of the distribution of the rattan, whose strong texture admits of the hard usage required in making fire. This is generally the case, as it is difficult in other parts of the world to supply the thong material. Some thongs of bark, however, or strips of flexible bamboo, are used in areas where the rattan does not occur. Matthew W. Stirling, on his expedition to Central New Guinea in conjunction with the Dutch Government, found the fire thong in use among the Pygmies and the fringing Pygmy-Papuan tribes. Curi- ously enough he found the method employed in sawing down trees. This is quite suggestive of a way by which the fire thong may have been discovered. The Battaks of the island of Palawan in the Philippines use the thong fire kindler. The thong of rattan is wound into a wristlet and worn till needed. The stick is cleft and held open by a bit of stone. M. W. Stirling brought from the hitherto unvisited Pygmies of New 48 This is perhaps across the groove. 47R. Brough Smyth. The Aborigines of Victoria, vol. 1, p.394. London, 1878. 48 Frictional fire making with a flexible sawing thong. Journ. Roy. Anthrop. Inst., vol. 44, January- June, 1914, pp. 32-64. art. 14 FIRE-MAKING APPARATUS—HOUGH 55 Guinea specimens of the thong apparatus identical with the Palawan set described and showing an interesting connection-survival. (Pl. 9, fig. 2, Cat. No. 326012; Mrs. E. Y. Miller; 11.5 in. (29:5 em.).) IV. PERCUSSION 1. Flint and pyrites.—The art of fire making by striking two stony substances together was begun in the far past, having originated in experiences connected with the working of stone. Since by striking flint against flint no live spark can be gotten to start a fire, it is nec- essary to infer that by striking two pieces of pyrites together or sub- stituting for one piece a flint, a rather hot spark would be observed to follow the impact. The pyrites strike-a-light was found in use in a number of localities, which seems to indicate a survival of former usage, while in other localities pyrites was used with flint, this arrange- ment being more workmanlike, obviating the breaking of the fragile pyrites. This ancestor of the flint and steel was in use in the Euro_ pean neolithic age and remained current far into the iron age, being used on guns after the invention of gunpowder. Presumably the neolithic equipment was a flint scraper, a lump of iron pyrites, tinder, and a bag to contain them. Many of the scrapers of the sort believed to have been those used in fire making are found in European neolithic deposits, but pyrites rarely, as it tends to decay rapidly. (Fig. 40a.) The working of the flint and pyrites in fire making was different from that pursued with the flint and steel. The steel is struck on the edge of the flint with a sharp scraping blow, while the neolithic scraper was chopped on the surface of the pyrites somewhat as a scraper is ordinarily used, shown in Figure 40. The pyrites lump, therefore, being scraped around the sides assumed a cylindrical form. (See fig. 42.) Dr. Thomas Wilson calls my attention to a discovery of a pyrites nodule by M. Gaillard, in a flint workshop on the island of Guiberon in Brittany. The piece bore traces of use. Doctor Wilson thinks that the curved flakes of flint like the one figured, found so numerously, were used with pyrites as strike-a-lights. The comparative rarity of pyrites is, perhaps, because it is easily decomposed and disintegrates in unfavorable situations in a short time, so that the absence of pyrites does not militate against the theory that it wasused. A subcylindrical nodule of pyrites 214 inches long and bruised at one end was found in the cave of Les Eyzies, in the Valley of Vézére, Perigord, mentioned in Reliquae Aquitanicae (p. 248). This is supposed to have been a strike-a-light. Prof. W. B. Dawkins thinks that: In all probability the cave man obtained fire by the friction of one piece of hard wood upon another, as is now the custom among many savage tribes. Some- times, however, as in the Trou de Chaleux, quoted by M. Dupont (Le Temps 56 PROCEEDINGS OF THE NATIONAL MUSEUM VOL. 73 Prehistorique en Belgique, second edition, p. 153), he may have obtained a light by the friction of a bit of flint against a piece of iron pyrites, as is usual with the Eskimos of the present day.” Professor Dawkins also says that fire was obtained in the bronze age by striking a flint flake against a piece of pyrites, sometimes found together in the tumuli. He figuresastrike-a-light from Seven Barrows, Fic. 40.—a STRIKE-A-LIGHT. SEVEN BARROWS, BERKS COUNTY, ENGLAND. From DAWKINS EARLY MAN IN BRITAIN, P. 358. (SEE Dr. JOHN Evans ANCIENT STONE IMPLEMENT, PP. 284, 288, FOR A SIMILAR FIGURE); b STRIKE-A-LIGHT. Cat. No. 1861, U.S.N.M. INDIANS OF ForRT SIMPSON, MACKENZIE RIVER DISTRICT, BRITISH COLUMBIA. COLLECTED BY B. R. Ross Lambourne, Berks, England, an outline of which is reproduced here for comparison with the one from Fort Simpson, British Columbia. | (Fig. 40a and 6.) Pyrites has been found in a kitchen midden at Ventnor, in connection with Roman pottery ® Chambers’s Encyclo- “® Barly Man in Britain, p. 210. London. 50 Tdem, p. 258. art, 14 FIRE-MAKING APPARATUS-—HOUGH 57 paedia, article, ‘‘ Pyrites,’’®! is authority for the statement that pyrites was used in kindling powder in the pans of muskets before the gun flint was introduced. : It is thus seen that this art has a high antiquity and that on its ancient areas its use comes down nearly to the present day, the flint and steel being its modern or allied form. In North America this art is distributed among the more northerly ranging Indian tribes and the Eskimo of some parts. Its use was and is yet quite prevalent among the Indians of the Athapascan (for- merly Tinné) stock of the north. By specimens in the Museum and notes of explorers it is found to range from north of Dixon’s Sound to Labrador, the following localities being represented: Stikine River, Sitka, Aleutian Islands, Kotzebue Sound, Point Barrow, the Mackenzie River district, at Fort Simpson, and probably Hershel Island, Pelly Bay, Melville Peninsula, Smith Sound, and Labrador. The Canadian and Algonquins strike two pieces of pyrites (pierres de mine) together over an eagle’s thigh, dried with its down, and serving instead of tin- der.” From other sources we know that the extinct Beothucs of Newfoundland did the same.* As far as can be ascertained, the Eskimo and Indians both use the method, so that it is not characteristic of either, as the four-part drill is of the Eskimo, as contrasted with the simple rotation sticks of the Indians. A description of a flint and pyrites outfit, as at present used, will give a general idea of the status of the invention. In dif- ferent localities the manipulation differs somewhat, as will be noted farther on. The strike-a-light (No. 128405) was collected by Capt. E. P. Heren- deen from natives who told him that it came from Cape Bathurst, hence he assigned the specimen to this locality on the evidence. John Murdoch has, with a great deal of probability, questioned this and thinks that it came from Herschel Island with the rest of Mr. Herendeen’s collections and did not come from as far east as Cape Bathurst. While there is no improbability that this method is prac- ticed at Cape Bathurst, yet the specimen has the appearance of the Mackenzie River strike-a-lights, hence it is deemed advisable to locate in the Mackenzie River district at Herschel Island. The essential parts of the apparatus are a piece of pyrites, a piece of flint, and tinder. In the more northern parts of the Eskimo area tinder is made from the down from the stems and catkins of various species of dwarf arctic willows. At present the natives often soak the tinder in a strong solution of gunpowder and water to make it quick; an older way was to mix powdered charcoal with it. This ‘1 Journ. Anthrop. Inst., Great Britain and Ireland, vol. 7, p. 83. 52 Lafitau. Moeurs des Sauvages Ameriquains, p. 272. An earlier account is found in Le Jeune, Rela- tion de 1634, p. 24. Quebec, 1858, 8Journ. Anthrop. Inst., Great{Britain and Ireland, vol. 5, p. 225. 58 PROCEEDINGS OF THE NATIONAL MUSEUM VOL. 73 plan is like the charring of the linen rags used in the old-fashioned tinder boxes of 40 years ago. The Eskimo then puts the tinder into a little round, flat pouch, with a flap in the middle. (Fig. 41, 1.) The pyrites (fig. 42, 3) looks like a short pestle, to much of which appearance the repeated scraping has no doubt given rise. The up- per end is concave, while the lower end has the original smooth sur- face of the concretion. Pyrite is found at Point Barrow in spherical masses of various sizes up to several pounds in weight. These spheres are nearly always cracked in two and scraped on the plane surface for very obvious reasons. This gives the shape seen in Fort Simpson and Long Barrows specimen. Mr. Murdoch says that the Eskimo think that pyrites comes down from above inmeteors. They call it “ fire- stone.”” A native related that in old times they did not use flint, but two pieces of pyrites, and got “big fire.” The flint (fig. 42, 4) is an ob- long piece of chert, square at the base and rounded at the forward end. It is more elab- orately made than the flakes so numerous in Europe, one of which was found with the piece of pyrites in the English Bar- rows. The Mackenzie River Fic. 41.—1. TINDER POCKET, 2. FIRE BAG. : : (PART OF STRIKE-A-LIGHT SET.) CAT. No. 128405, scraper 1s more like the curved U.S.N.M. MAcKENZIE RIveR District, BRITISH gncientone. In most cases the COLUMBIA. COLLECTED BY E. P. HERENDEEN é ° flints used are not mounted in a handle; this specimen, however, is fixed in a handle made of two pieces of wood held together by a thong of seal skin. (Fig. 42, 4a.) The bag (fig. 41, 2) is made of reindeer skin. The little bag that hangs from the larger has a double use; it is a receptacle for reserve tinder, but its chief use is for a toggle; being passed under the belt it prevents the loss of the outfit, which is said to be carried by the women. An oblong pad, stuffed with deer hair, is sewed to the mouth of the firebag to protect the hand from sparks and blows of the flint. NGS Whe | ) 5 Ae Mh \\y )) | ; Si MG ¥ ies Sy ie arr. 14 FIRE-MAKING APPARATUS—HOUGH 59 To get a spark, the Eskimo places (fig. 43) the piece of pyrites on the pad held in the left hand over the curved forefinger, the large end down and the thumb set in the cup-shaped cavity in the top. The flap of the tinder pocket is turned back and held on the forefinger under the protecting pad. The flint is held in the right hand and by a scraping motion little pieces of pyrites at a dull red heat fall down into the tinder. The pellet that glows is transferred to the pipe or fire, and the flap of the tinder pocket is turned down, serving to keep the tinder dry and to extinguish it if necessary.” There comes in here appro- priately a note of B. R. Ross on the burial customs of the Kutchin Indians of the eastern Athapascan stock. He says: They bury with the dead a flint fastened to a stick, a stone to strike it on (pyrites) to make fire, and a piece of the fungus that grows on the birch tree for tinder and some touchwood also. There is no mention of this process of firemaking by the older writers of Greenland, Cranz and Egede, though they carefully note and describe the plan by wood boring. Later explorers going higher north in western Greenland have found Fic. 42—3. Pyrires. 4, 4a. FLINT STRIKER AND : : ra HANDLE. (PART OF STRIKE-A-LIGHT SET SHOWN IN it. Dr. Emil Bessels, writing FIG. 41.) Cat. No. 128405, U.S.N.M. MACKENZIE about the Itah Eskimo of Smith © River Disraicr, BritisH CoLuMSIA. COLLECTED ed BY E. P. HERENDEEN Sound, says: The catkins of the Arctic willow are used as tinder to catch the sparks which have been produced through the grinding of two pieces of stone.* Dr. E. K. Kane gives a more complete account from nearly the same locality, the Arctic Highlands of northwest Greenland. He says that the Eskimo of Anoatok struck fire from two stones, one a plain piece of angular milky quartz, held in the right hand, the other & Extracted from an article by the author in Smithsonian Report, vol 2, 1888, pp. 181-184. 565Smithsonian Report for 1866, p. 326. 8 Die amerikanische Nordpol-Expedition, p. 358. Leipzig, 1879. 60 PROCEEDINGS OF THE NATIONAL MUSEUM VOL, 73 apparently an oxide of iron [pyrites or iron ore?]. They were struck together after the true tinder-box fashion, throwing a scanty supply of sparks on a tinder composed of the silky down of the willow cat- kins (Salix lanata) which he held on a lump of dried moss.” Very much farther west on Melville Peninsula Parry gives a com- plete and interesting description of the primitive way. This account gives us a link between the western and eastern Eskimo. . He writes: For the purpose of obtaining fire the Eskimo use two lumps of common pyrites, from which sparks are struck into a little leathern case (see fig. 25, pl. LX XIV) containing moss well dried and rubbed between the hands. If this tinder does not readily catch, a small quantity of the white floss of the seed of the ground willow is laid above the moss. As soon as a spark has caught it is gently blown till the fire has spread an inch around, when the pointed end of a piece of wick being applied, it soon bursts into a flame, the whole process having occupied perhaps two or three minutes.® The Museum was in possession of a specimen catalogued, ‘‘Moss bag and lumps of pyrites used by Innuit for getting fire,” collected by Capt. C. F. Hall at Pelly Bay, in latitude 69°, longitude 90°, several degrees west of Melville Peninsula. The only other record of the process under consideration among the Es- Fic. 43-—Mrrtop. of vane tae imo isfound in the Aleutian Islands. STRIKE-A-LiGHT. Cat. No. 128405, There is absolutely no evidence had ak Fs Drawing By W. H- by the writer that the Eskimo south of Kotzebue Sound (western Eskimo) use the pyrites and flint for making fire. The latest information about the Aleutian Islanders is given in a manuscript of the careful explorer, Lucien M. Turner. His observation will serve to explain the description of striking a light by earlier travelers. They use the four part drill but they also use pyrites. A stone containing quartz and pyrites is struck against another similar one, or a beach pebble, into a mass of sea-bird down sprinkled with powdered sulphur. This ignites and is quickly caught on finely shredded blades of grass or beaten stalks of wild par- snips. This method prevails to this day on the islands west of Unalashka. The people told Mr. Turner that this was the ancient way. There is a doubt in the writer’s mind that Sauer’s (Billing’s Expedition, p. 59), and Campbell’s (Voyage, p. 59,) observations, brought 8 Arctic Explorations, vol. 1, p. 379. 58 Second Voyage, p. 504. London, 1824. ART. 14 FIRE-MAKING APPARATUS—HOUGH 61 together by Bancroft,®® were accurate with regard to the stones used. All the other details are correct, but they say they took two pieces of quartz, rubbed them with sulphur, and struck them together. It is well known that pieces of quartz even when rubbed with sulphur will not strike a spark of sufficient heat to cause ignition. The pieces used must have been pyritiferous quartz as noticed by L. M. Turner. To summarize, the following facts arise out of the foregoing con- siderations of the flint and pyrites method: (1) It is very ancient, inferring from the few reliable finds of pyrites and flint in juxtaposition. (2) Its distribution is among high northern tribes, both Eskimo and Indian. (3) As far as known, its range is limited to this area, only one other instance coming to our notice, that of the Fuegians. 2. Flint and steel—The flint and pyrites method is the ancestor of the flint and steel. The latter method came in with the iron age. It is found in the early settlements of that period. A steel for strik- ing fire was found in the pile dwellings of the Ueberlingen See.” The Archeological Department of the Museum has a specimen of a strike-a-light of the early age of iron in Scandinavia. It is a flat, oval quartz stone with a groove around the edge; it is thought to be for holding a strap by which it could be held up and struck along the flat surface with the steel. It is scored on these surfaces. The spec- imen in the Smithsonian is from the national museum at Stockholm. In Egypt it is believed to have been used for a long period, though there is no data at hand to support the conclusion. In China it has been in use for many centuries. Chinese history, however, goes back to the use of sticks of wood. The briquet must have been car- ried nearly everywhere by early commerce from the ancient countries around the Mediterranean, as it was into new lands by later commerce. Many persons remember the tinder box that was taken from its warm nook beside the fireplace whenever a light was wanted; the matches tipped with sulphur used to start a blaze from the glowing tinder are also familiar to the older generation. The tinder boxes in use in this country were just like those in England from time imme- morial down to 50 yearsago. (Fig. 44.) Edward Lovett, of Croydon, England, who has studied this matter thoroughly, calls attention to the resemblance of the old English tinder flints to the neolithic scrap- ers. These scrapers, picked up at Brandon, can scarcely be discrim- inated from those made at the present time at that place, and there 5 Native Races of the Pacific States, vol. 1, p. 91. 6 Keller. Swiss Lake Dwellings, pl. 28, fig. 29. 61 Sir J. W. Dawson gives an interesting account of the strike-a-light flints used in Egypt in 1844,in Modern Science in Bible Lands, p. 30. 62 PROCEEDINGS OF THE NATIONAL MUSEUM VOL. 73 is a suspicion that the present tinder flint has come down directly from neolithic times. The old English steel, or ‘‘flourish”’ (fig. 44) is the characteristic shape, and has been carried by English commerce into many places. A picture of a strike-a-light used by the Lenguas of Brazil seen lately, shows the unmistakable old “ flourish.” SSSSSD LODE Ln EN OOTP eee “FLOURISH,’? AND BUNDLE OF SPUNKS), ENGLAND. COLLECTED BY LOUIS AND MAURICE Fig. 44.—-ENGLISH TINDER BOX (WITH FLINT, Car. No. 75516, U.S.N.M. FARMER The tinder boxes had also a damper to extinguish the tinder of burnt linen and to keep it dry. The lids were furnished often with a candle socket. This feature, says Mr. Lovett, has led to their preservation as candlesticks long after they were superseded by matches. Fig. 45.— WHEEL TINDER BOX. CAT. No. 130431, U.S.N.M. BROADALBIN, N. Y. PRESENTED BY F.S. HAWLEY Many devices were invented in order to improve on the crude way of holding the flint and steel in the hands to strike the spark into the tinder box. One of these was the wheel tinder box. (Fig. 45.) The compartment near the wheel held the tinder. The flint was placed in a socket on the sliding lid and the wheel was turned by unwinding ART. 14 FIRE-MAKING APPARATUS—HOUGH 63 a string from off its axle with a sharp pull as in spinning atop. The flint was pressed against the rapidly revolving wheel, and a shower of sparks fell into the tinder. The tinder pistol, whose name suggests its use, was another device. Other devices were intended to be carried in the pocket and were probably brought out by the introduction of tobacco and the need of smokers for a convenient light. The pocket strike-s-light is still used. The one shown (fig. 46) was bought in 1888 by E. Lovett, at Boulogne-sur-Mer. They are still used by the peasants and workpeople of France. An old specimen in the Museum of this character is from Lima, Peru. The roll of tinder, or ‘‘match,” is made of the felt lining of an ant’s nest (Polyrachus bispinosus). Among many of our North American tribes the flint and steel super- seded the wooden drills as effectually as did the iron points the stone arrowheads. Fic. 46.—STRIKE-A-LIGHT (BRIQUET). (CAT. No. 129693, U.S.N.M. BOULOGNE-SUR-MER. FRANCE. COLLECTED BY EDWARD LOVETT) Some of these tribes were ripe for the introduction of many modern contrivances. Civilized methods of fire lighting appealed to them at once. Among the Chukchis, Nordenskiéld says, matches had the honor of being the first of the inventions of the civilized races that have been recognized as superior to their own.® It was so among our Indian tribes; the Mandan chief ‘‘ Four Bears” lighted his pipe by means of a flint and steel taken from his pouch when George Catlin visited him in 1832.® The Otoes (Siouan stock) made use of the flint and steel shown in Figure 47. The flint is a chipped piece of gray chert, probably an ancient implement picked up from the surface. The steel is a very neatly made oval, resembling those of the Alban- ian strike-a-lights,® or the Koordish pattern. (Fig.52.) Here arises one of the perplexities of modern intercourse; perhaps both of these steels were derived from the same commercial center. 62 See figure in D. Bruce Peebles’s address on Illumination, in Trans. Roy. Scottish Society of Arts Edinburgh. vol. 12, pt. 1, p. 96. 83 Voyage of the Vega, vol. 2, p. 122. *t The George Catlin Indian Gallery. Smithsonian Report for 1885, vol. 2, p. 456 *5See figure in Journ. Anthrop. Inst., Great Britain, vol. 16, 1886, p. 67. 64 PROCEEDINGS OF THE NATIONAL MUSEUM VOL. 73 The flint, steel, and tinder were always carried in a pouch, usually suspended from a belt as in specimen No. 8481 from the Assiniboins (Siouan stock) of Dakota. This is a buckskin waist belt, beaded and fringed, ornamented with bells of tm. It supports a flapped pouch for the flint, etc. The tinder used was fungus. The pouch of the Cheyennes (Algonquian stock) is compact, and neatly made of leather. (Fig.48.) The equipment is complete and of a superior order. The bone cup is used to hold the tinder while striking a spark into it. It is the tinder horn of early days, a cow’s horn which was used to hold tinder before sheet-iron boxes came into . ‘ : N x. s Fic. 47.— FLINT AND STEEL. CAT. No. 22431, U.S.N.M. OTOE INDIANS, KANSAS AND NEBRASKA. COLLECTED BY J. W. GRIEST use. The Lenguas of Brazil use a horn for the same purpose.® In the Aino set (fig. 54) can be seen this feature. The tinder with this set is rotten wood. Nearly all Indians know the value of fungus tinder. The Comanche Indian strike-a-light is a similar pouch to the one described, but much poorer in equipment. (Fig. 49.) A broken rasp, a piece of chert, and a piece of spunk is enough for the purpose, and a bag made from a saddle skirt to hold them completes the outfit. The flint and steel is still used nearly all over Mexico, Doctor Palmer informs me. There is at present a manufacture of gun and strike-a-light flints at Brandon, England, whence they are shipped to Spain, Mexico, Italy, and other civilized countries. Doubtless this 66 See figure in Jahrbuch Mittelschweiz. Commercial. Gesellsch. Arau, vol. 2, 1888, pp. 114-115. 65 FIRE-MAKING APPARATUS—-HOUGH ART. 14 flint from Guadalajara (fig. 50) came from Brandon. It is real cal- careous flint, such as does not exist in this country. The steel is the “swallowtail”? pattern. The tinder is of prepared fungus sold in little packets. quit ii AUN \\s v A ( \\ nh NNN \ AANA \ MM : SAE sy t) ANN \\ A SS a A.A) a A \ i WAN ae N NE \ N \ Ny N N yt RY NE N NES ue yy, N mi y j : tiga . Vip G 7) SOPs Ly ELISE: WH TINDER HORN, SPUNK, AND poucH). Cat. No. 22104, U.S.N.M. COLLECTED BY Dr. J. H. BARRY Fig. 48.—STRIKE-A-LIGHT (FLINT, STEEL, CHEYENNE INDIANS, ARKANSAS. The Koords of Bhotan, eastern Turkey, carry a pipe pouch con- pick and a pair of pin- taining besides flint, steel, and tinder, a pipe (Fig. 52.) Thetinder cers, to transfer the lighted tinder to the pipe. The is prepared from a fungus, probably a species of polyporus. VOL. 73 PROCEEDINGS OF THE NATIONAL MUSEUM 66 steel, shaped like an old-fashioned bell pull, is a very good form for hold . the hand. ing in “s os] ‘aS = eer = 2 o = ; ‘ AWAIARAR Re Sp LON M ND WVVONNOLVUNT NURSE INE: 22 ASSES RN! = AAS XX XONAR AK KKK NANAK VOX KY UA RD (eB) fas] Nm eS a S) 2 Ss HS fo} Ay Cat. No. 6972, COLLECTED By EDWARD PALMER CH FOR HOLDING FLINT AND STEEL.) (Pou COMANCHE INDIANS, TEXAS. Fig. 49.—STRIKE-A-LIGHT. U.S.N.M. In Tibet One owned by (Fig. 51.) they are made very large and are finely decorated. pouch in which to keep the flint and tinder. The pouch was trimmed with encrusted silver set Mr. W. W. Rockhill has a curving steel between 5 and 6 inches long, finely carved. with jewels. ART. 14 FIRE-MAKING APPARATUS—HOUGH 67 The Ainos of Japan use fiint and steel for striking a light, this method having supplanted the generation of fire by sticks (p. 26.) This outfit shown (fig. 54) is complete. The shoe-shaped steel Fic. 51,—STRIKE-A-LIGHT. Cat. No. 130311, Fic. 50.—FLINT AND STEEL. Cat. No. 126576, U.S.N.M. CHINA. Girt ofr GEORGE G. FRYER U.S.N.M. GUADALAJARA, INDIANS, MEXICO. COLLECTED BY EDWARD PALMER Fig, 52.—SMOKERS’ PIPE-LIGHTING OUTFIT SHOWING FLINT, STEEL, PIPE PICK, AND PINCERS). Cat. No. 130607, U.S.N.M. Koorps OF BHOTAN, EASTERN PURKEY. COLLECTED BY ReEv. A. N. ANDRUS. is attacned by a piece of sinew to the cork of a small wooden bottle containing the soft charcoal used as tinder. The flint is a small piece of ferruginous silex. With this set is a piece of stick which 68 PROCEEDINGS OF THE NATIONAL MUSEUM VOL. 73 54 Fias, 53-54.—53, RUsH FIRE SET POUCH. 54, STRIKE-A-LIGHT. FLINT, STEEL, AND TINDER BOX. CAT. No 22257, U.S.N.M. Arnos oF YEzO, JAPAN. COLLECTED By B.S. LYMAN art. 14 FIRE-MAKING APPARATUS—HOUGH 69 retains fire for a long time. It is the root of the Ulmus campestris, or laevis, formerly used for the fire drill, but has come into a second- ary place since the introduction of the flint and steel. To strike a light the Aino takes out the cork with the steel attached and stirs up the tinder with the sharp point. He then holds up the flint in his hand over the box and strikes a spark down into it. He then transfers the coal to his pipe, or material for fire, or fire stick, with the point of the steel. These articles are kept in a rush pouch of twined weaving. (Fig. 53.) A much ruder pouch of fishskin is in the Museum. The Japanese tinder box has two compartments, one with a damper for the tinder and the other larger one for the flint and steel. This box is a familiar object in Japanese kitchens. The mounting of the steel in wood is an improvement on holding it between the fingers. (Fig. 55.) No one, it seems, ever thought of so mounting the steel in western countries. The matches are broad shavings tipped at both ends with sulphur, and are the Japanese rendering of the ““spunks”’ used with our tinder box. Smokers in Japan carry a very small strike-a-light. (Fig. 56.) The cloth pouch with a long flap that can be rolled around several times and tied contains the three essentials, flint, steel, and tinder, the latter of burnt cotton. 3. On bamboo—Under percussion is classed the bamboo and porce- lain strike-a-light first described by Sir Alfred Russell Wallace as used by the natives of Ternate, Malay Archipelago. Sir Alfred remarks that the Ternate people make great use of bamboo in their daily life and describes a particular method by which fire is struck from the flinty surface of the bamboo with a small piece of broken china, producing a spark which is caught on tinder. This apparatus vies with the fire piston as fire-making curiosities. Necessarily the method is confined to the bamboo area strictly, but has never been found in the Western Hemisphere. The bamboo selected and from which the specimens in the United States National Museum are made has a rough surface layer feeling like fine sandpaper. This coating is in the form of a flinty layer about one-half millimeter thick which is chipped away in small bits under the stroke of the china. The material of the layer is probably a combination of silica with some organic substance rendering it capable of taking up the force of the blow and converting it into heat sufficient to ignite tinder. ‘The specimens in the Museum consist of a joint of bamboo or cane, a tinder box of bamboo with cap lid, and hooped with braided rattan. A cord passes through holes at the bottom, through holes in the cap, and forms a loop to pass around the bamboo joint. Some of the Battak tinder boxes are decorated with incised patterns. The Battak specimens usually have two tinder boxes. 70 PROCEEDINGS OF THE NATIONAL MUSKUM voL 73 a SN SS hen Fig. 55.—TINDER BOX (SHOWING MOUNTED STEEL, FLINT, AND BUNDLE OF SHAVING MATCHES; BOX ONE-THIRD NATURAL SIZE). Cat. No. 127137, U.S.N.M. Japan. GIFT OF THE JAPANESE DEPARTMENT OF EDUCATION, TOKIO ART. 14 FIRE-MAKING APPARATUS—HOUGH “1 The specimens are as follows: Cat. No. 326011, a set from the Battaks of Palawan, collected by Capt. E. Y. Miller. The bamboo tube is 14 inches long (35.5 cm.), three-fourths inch deep (2 cm.); the boxes are 514 inches long (13.5 em. and 3.1 inches deep (5.25 cm.). (PI. 10, fig. 1.) One of the boxes contains tinder and a bit of flint. Cat. No. 326012, Battaks, Palawan, P. I. Mrs. F. G. Miller has a bamboo tube 17.6 inches long (44.5 em.), 1 inch deep (2.6 em.). One tinder box is 5.5 inches long (14 em.) and 1.7 inches deep (4.5 cm.).. The box contains tinder and a piece of flint. (PI. 10, fig. 4.) Cat. No. 232283, Malays of Balabac, an island south of Palawan, P. J.; collected by Capt. E. Y. Miller. Captain Miller has two tinder boxes not matched as. to size. The bamboo tube is 17.45 inches long (44 em.), 1.1 inches deep (2.5 cm.). The larger tinder box is 7 inches long (18 cm.), 2.4 inches deep (6 cm.). (PI. 10, fig. 2.) The tinder is brown and appears to be the scurf of a palm. The striker is a gunflint. Two well-dec- orated tinder boxes brought from Palawan by Captain Miller are Figure 3, Plate 10, and measure 5.5 inches long (14 em.), 2.4 inches deep. (6 cm.); 5.5 inches long (14 cm.), 2 inches deep (5 cm.), Cat. No. Fic. 56.—SMOKERS’ STRIKE-A-LIGHT. CAT. Nc. 128138, 326013. U.S.N.M. Tokio, JAPAN. GIFT OF THE JAPANESE Plate 9, Figures i, ae show DEPARTMENT OF EDUCATION ° A boxes open exposing tinder. Above Figure 3 is a pistol flint found in one of the boxes. Ordinarily a bit of broken dish is employed, since flint is not local in a vast Pacific area. There is evidence that the bamboo strike-a-light had a considerable range in Malaysia, and notices of it have come from Cochin China, southern Philippines, Ternate, and Waigiou, an island off the north- West point of New Guinea, all on a line running southeast from Cochin China. 72 PROCEEDINGS OF THE NATIONAL MUSEUM _ VoL. 73, arr. 14 Vv. BY COMPRESSION OF AIR The fire syringe, as it is called, consists of a piston and plunger. Generally the piston is a smooth circular canal drilled in hardwood or horn. The plunger fits the cavity with exactness. In practice a bit of tinder is placed in a slight cavity at the end of the plunger; the latter is set in the orifice and driven down with asharp blow. Quickly withdrawing the plunger the tinder is found alight. The principle is that in being compressed to a smaller volume air gives up heat. In the case of the fire syringe this is enough to ignite tinder. This is a method which has been employed by many tribes of men in Malaysia, and it appears to be a native invention. Plate 11 shows three specimens from various parts of the Philippines, Figures 1,3, and 4, Cat. No. 235261, Mindoro, Philippine Commission; 5 inches long (12.5 em.); Cat. No. 215659, Luzon, Dr. Charles E. Woodruff, United States Army, 3% inches long (10.75 em.); Cat. No. 216736, Luzon, Col. F. F. Hilder, 5 inches long (13 cm.). Figure 2 is of horn, Cat. No. 176007, Lower Siam, Dr. W. L. Abbott, 3.5 inches long (9 cm.). Figure 5 is of hard palm. Cat. No. 175270, Java, M. F. Savage, 84% inches long (21 cm.). VI. TINDER It is no doubt true that acquaintance with tinderlike substances was forced on man by the behavior of the camp fire in consuming at different rates such material. Tinder is also implied in preparing and arranging the fuel for starting a new fire. From these considerations it seems probable that this feature involved in the invention of the fire drill had been prepared in a meas- ure long previously. The collection of tinder in the Museum is almost exclusively of vegetal substances, but in many cases these have been improved by the addition of charcoal, gunpowder, and niter. Animal substances are necessarily rare and so far as observed consist of the down of birds and the nest lining spun by an ant (Polyrachis bispinosus), the latter from South America. Vegetal substances used as tinder are classified as follows: (a) Bark, especially the outside spent layers of trees with stringy bark in the first stages of decay; (b) scrapings of inflammable wood; (c) scurf down from leaves and about the flower- ing areas of certains plants; (d) downy catkins or down from seed heads out of bloom; (e) dry leaves rubbed fine or grass treated in the same manner; (f) rotten wood also used for retaining fire; (g) fungi, either natural as in the sheet fungi or worked into condition for use as by boiling in solution of potassium nitrate or saltpeter; (A) imper- fectly charred cotton or linen cloth or thick, soft cords impregnated with a chemical. The Chinese use soft paper prepared in a similar manner. The Japanese so far as known are unique in using mixed tinder composed of several of the substances mentioned above. O | U. S. NATIONAL MUSEUM PROCEEDINGS, VOL. 73, ART. 14 PL. 1 SOUTHERN TLINKIT DRILL FOR DESCRIPTION OF PLATE SEE PAGE II U. S. NATIONAL MUSEUM PROCEEDINGS, VOL. 73, ART. 14 PL. 2 BRITISH GUIANA, WEST INDIAN, AND MEXICAN DRILLS FOR DESCRIPTION OF PLATE SEE PAGES 21, 22, AND 25) PROCEEDINGS, VOL. 73, ART. 14 PL. 3 U. S. NATIONAL MUSEUM eae - 8 cscs RE a 2 Ee ou bean a, is spe ; pe tay ~ wna nina ET, puts ane oe Oe eee ee FOR DESCRIPTION OF PLATE SEE PAGE 27 JAPANESE SACRED FIRE DRILL, FULL VIEW AND SECTION U. S. NATIONAL MUSEUM PROCEEDINGS, VOL. 73, ART. 14 PL. 4 BHILS, INDIA, AND AUSTRALIAN DRILLS FOR DESCRIPTION OF PLATE SEE PAGE 28 AND 32 U. S. NATIONAL MUSEUM PROCEEDINGS, VOL. 73, ART. 14 PL. 5 HINDU SACRED FIRE DRILL (REPLICA) FOR DESCRIPTION OF PLATE SEE PAGE 29 U. S. NATIONAL MUSEUM PROCEEDINGS, VOL. 73, ART. 14 PL. 6 ieee ae RT po Lz EAST INDIAN FIRE DRILLS FOR DESCRIPTION OF PLATE SEE PAGE 30 U. S. NATIONAL MUSEUM PROCEEDINGS, VOL. 78, ART. 14 PL. 7 AFRICAN FIRE DRILLS FOR DESCRIPTION OF PLATE SEE PAGE 31 U. S. NATIONAL MUSEUM PROCEEDINGS, VOL. 73, ART. 14 PL. 8 BAMBOO FIRE SAW, PHILIPPINES FOR DESCRIPTION OF PLATE SEE PAGE 62 U. S. NATIONAL MUSEUM PROCEEDINGS, VOL. 73, ART. 14 PL. 9 BAMBOO STRIKE-A-LIGHTS AND BATTAK (NEGRITO) FIRE THONG FOR DESCRIPTION OF PLATE SEE PAGES 30, 55, 71 U. S. NATIONAL MUSEUM PROCEEDINGS, VOL. 73, ART. 14 PL. 10 BAMBOO STRIKE-A-LIGHTS, MALAYSIA FOR DESCRIPTION OF PLATE SEE PAGE 71 U. S. NATIONAL MUSEUM PROCEEDING, VOL. 73 ART. 14 PL. 11 FIRE PISTONS, MALAYSIA FOR DESCRIPTION OF PLATE SEE PAGE 72 Oe a ee ee a ee CONTRIBUTION TO THE COMPARATIVE ANATOMY OF THE EARED AND EARLESS SEALS (GENERA ZALO- PHUS AND PHOCA) By A. Brazier Howe, Collaborator, United States National Museum FOREWORD The interest of anatomists has long been intrigued by the Pinni- pedia and it is probable that no order of a comparable size has been more often investigated from this standpoint. The pinnipeds have a very important place in the program of the author relative to his investigation of aquatic adaptations in mammals and he at first thought that this work with the order would be rendered relatively easy by the apparently full reports upon both the myology and osteology, illustrated in some cases by handsome plates, with which he was casually acquainted. Only a little investigation was needed, however, to establish the fact that these reports were not of great aid, for they are chiefly descriptive, and many discrepancies were apparent. Of the earlier dissections of pinnipeds those by Duvernoy (1822), Humphrey (1868), and Lucae (1873) are all important, although some of their details are to be viewed with suspicion and many of their conclusions are extremely unlikely. But scant attention need be paid them in the present report, however, for their details are well incorporated in the paper of W. C. S. Miller (1887), who discusses them with really unnecessary fullness, and their inclusion here would not only constitute repetition, but would be otherwise unde- sirable as befogging the report to a bewildering degree. Compari- sons have therefore been made only with the findings of Miller and Murie, where these investigators differ from conditions as en- countered by me. Miller was an accomplished anatomist who dis- sected a variety of pinnipeds, presumably with great skill. His text treats fully of a Phoca vitulina and an Arctocephalus, although com- parisons are made where desirable with several other phocids and an Otaria (=Eumetopias). A serious defect, however, is that his re- port is unillustrated as far as concerns the musculature, and his de- No. 2786.—PROCEEDINGS U. S. NATIONAL Museum, VoL. 73, ART. 15. 86377—28——1 f! 2 PROCEEDINGS OF THE NATIONAL MUSEUM VOL. 73 scriptions are often so involved and bristling with details, unimpor- tant save when a study is being made of individual variation, that one is often at a loss to fathom his exact meaning. He made some mistakes in interpreting what he saw, and doubtless others which it is difficult to discover, but one gathers the impression from working with his paper that he was a capable, trustworthy man, doing work of a high order of merit. Murie’s reports upon the myology of H'wmetopias (his Otaria) (1872) and Odebenus (his Trichechus) (1870) are descriptive rather than comparative. They are accompanied by beautiful plates some of which are far more satisfactory than any I could execute, but others are vague and misleading. He misnames and misinterprets a number of muscles, although to but a slightly greater extent than did Miller. It is, of course, beyond question that Murie was an able and brilliant human anatomist, but it is perhaps not out of place for me to say that after working with his sea lion and Globiocephala reports line by line I have received a definite impression that implicit reliance can not always be placed upon the myological details which he presents. One might, therefore, justifiably enquire regarding the value of an additional report upon the anatomy of the Pinnipedia. The reason is that the others are largely descriptive or compare individual muscles, but no one has heretofore analyzed the differences occuring in the otariids and phocids, the significance of these from a functional aspect, the reasons for the osteological peculiarities, and the organi- zation of the pinniped as a dynamic machine built for aquatic loco- motion. My myological report is but a necessary part of the whole. The conclusions to which the anatomical evidence points has not been discussed in entirety, however. Most of the myological discussion is presented with the muscles, some of the osteological with the bones, and still more under the general discussion. Yet additional facts and theories are being reserved which are considered to belong more properly with a comparison of the Pinnipedia with other aquatic mammals. I have placed those interpretations upon the anatomical peculiarities of the Pinnipedia which to me seem most logical, but it can not be claimed that all of these are correct, or that some of them will not need modification when additional facts are brought to light. In the drawings of muscles no especial system of reduction is used, the proportions being merely such as will fit conveniently upon a page. In the bone drawings, however, comparative details are presented to represent relative difference in size, and because the trunk length of the Phoca skeleton used in comparisons was seven-tenths that of the Zalophus, the reduction of the latter’s bones is but seven-tenths of those of the former. In this way one may more readily compare osteological details. ———— SS 4 x ant.15 ANATOMY OF THE EARED AND EARLESS SEALS—HOWELL 3 HABITS For a proper understanding of the pages that follow it will be necessary briefly to discuss the habits of the eared and earless seals, and to mention certain acts which their form allows them to perform or prevents them from doing. Zialophus californianus, as representing the sea lions or eared seals. is fundamentally a long, rather slender animal, save when very fat or in the case of mature bulls, which spend sassene ai time on land but seldom venture farther than a few yards away from the sea. The fore and hind limbs are both highly modified into paddles and considering the highly specialized condition in this respect, terrestrial locomotion is accomplished with more agility than one would imagine eas eee Fig. 1.— TYPICAL TERRESTRIAL POSTURES OF AN EARED SEAL (SEA LION OR OTARIID, ZALOPHUS) AND EARLESS SEAL (TRUE SEAL OR PROCID, PHOCA, ABOVE) _to be possible, and the animal travels on land much as would a fissiped carnivore with legs of equal length, galloping about with considerable speed. The neck is extraordinarily mobile, enabling a trained sea lion to perform surprising feats of balancing with its nose, probably to a more perfect degree than could any terrestrial mammal. Marked flexibility of the lumbar region also exists, and otariids—especially fur seals—can contort this part of the back in a striking manner. The forelimb is used as the primary and almost exclusive means of aquatic locomotion, and as the chief support for the body when on land. In the latter situation the manus is bent at the wrist and extended laterad. The pes is also bent at a right angle to the shank, but as the shin is bound down to the pelvis, so as 4 PROCEEDINGS OF THE NATIONAL MUSEUM ” you. 73 to be held within the body contour and virtually immovable, the only way that the foot can assume the plantigrade position used in walk- ing is for the sacral vertebrae to be forced into a position that is practically vertical to the ground, which is done with ease. The hind limbs apparently play a much less important part in swimming than one would infer from the degree of specialization which they exhibit. The Phoca hispida, as typifying the earless seals or phocids, is really a very different animal. It, too, is stream line in form but in a somewhat different manner from the otariid. Usually fat, it is of greater circumference than the otariid of the same mass, excepting adult bulls of the latter, and in most forms at least is relatively shorter. The shortness of the neck seems especially marked and to a greater degree than is actually the case, for it is very wide, tapering to the broad thorax. The neck is not markedly flexible, as is that of the otariid—in fact, it is probably less so than in the average fissiped, and one gains the impression that the entire trunk is less agile than in the sea hon. The forelimbs are weak, are not used as a primary means of propulsion through the water but in lateral, water-treading movements, and their use on land is limited to such acts as aiding in the surmounting of a low obstruction. Swimming is accomplished by a rhythmic, transverse movement of the hind feet, presented palm to palm, the movements being on the whole comparable to those of a fish in swimming. Both otariids and phocids may swim for con- siderable distances on the back, but the former assumes this position evidently for brief periods only, while the latter is more prone to do so, at least in captivity. For several anatomical reasons, as dis- cussed later, the Phocidae can not place the hind foot to the ground in a plantigrade manner and apparently never attempt to do any- thing with the feet while on land save keep them, palm to palm, elevated well out of harm’s way. Terrestrial progression is accom- plished by a caterpillarlike wriggling in the sagittal plane—not in the horizontal or transverse one—with the forefeet close to the sides and the hind ones elevated above the ground. MATERIAL The material assembled for the present work was not all that was desired but was the best that could be procured. As representing the Phocidae an embalmed subadult female of Phoca hispida taken by H. C. Raven, Ponds Inlet, Baffin Island, August 30, 1926, was obtained by exchange from the American Museum of Natural His- tory. But a single skeleton, partially articulated, of this species could be located and that was borrowed from the Museum of Com- parative Zodlogy through the courtesy of G. M. Allen. It is of a j \ ~_——— ART. 15 ANATOMY OF THE EARED AND EARLESS SEALS—-HOWELL 5 subadult, unsexed, and bears the number 6297, from Cumberland Gulf, April, 1878, collected by L. Kumlien. As representing the Otariidae, the National Museum secured through the interest of the Johns Hopkins Medical School an embalmed juvenal female Zalophus californianus that died September 6, 1925, after being at the Baltimore Zoological Park for only a short time. Sundry an- atomical observations were also made upon a large adult female of this species that died at the National Zoological Park during June, 1927. For osteological comparison the most suitable skeleton avail- able was that of a subadult male, disarticulated, No. 200847 of the National collection, that died at the National Zoological Park De: cember 19, 1915. There was also at hand some less satisfactory skeletal material of this same species, some of Phoca, mounted skele: tons of both families, and an extensive collection of skulls. It should be understood that in the following pages the above speci- mens, upon which the present study is primarily based, are referred to not by number, but by such terms of designation as “ the otariid ” (the embalmed specimen for the muscles, and the skeleton for osteo: logical details), or “my phocid.” My study of the prepared speci- mens has been supplemented by observation, as often as possible, of both wild and captive specimens of Zalophus and Phoca (of the vitulina and richardi sorts). EXTERNAL FEATURES The length from nose to tip of tail was 978 in the Zalophus and 1,019 mm. in the Phoca, so these embalmed specimens were as nearly comparable in size as one could reasonably wish. The length of tail in the former was 60, and in the latter 72 mm., and the circumference of the thorax respectively 430 and 780 mm. ‘The sea lion was exces- sively emaciated and not only was there no fat but most of the muscles where somewhat shrunken. The seal, on the contrary, was very fat, this being, tender and free from fibrous tissue. Over the shoulder it was about 80 mm. in thickness, thinning toward the head, caudad in the region of the hind flippers, and upon the forelimbs. In a state of nature females and immature males of Zalophus are usually sleek and of slender appearance, although captive specimens and old males may become fat and logy; but it is normal for at least most of the Phocidae to have an extensive blubber layer. All pinnipeds have a form that is markedly “stream-line ” but which animal is the more efficient in this respect we do not know. Both are covered with short stiff hairs, the pelage of the phocid being the thicker. In the otariid the mystacial pad had a width of 40 mm. and ap- peared rather narrow. The vibrissae were directed chiefly caudad 6 a PROCEEDINGS OF THE NATIONAL MUSEUM VOL. 73 and the nostrils were directed at an angle of about 15° or possibly 20° rostro-dorsad of the cranial axis. The mystacial pad of the Phoca had a width of 75 mm. and appeared very wide and as though inflated. The vibrissae were directed mostly laterad, but also down- ward and forward, and the whole mystacial area was more walrus- like than that of the otariid. The nostrils were directed rostro- dorsad at an angle of about 45° to the cranial axis, thus being situated more dorsad than in the sea lion. The direction of the eyes in the otarlid was at an angle of about 50° with the vertical, and about 15° in the phocid. ‘The eye was larger in the former, the width between the canthi being 70 mm., and in the phocid 40 mm. The latter had a few supraorbital vibrissae, which were entirely lacking in the former. In the Zalophus the pinna of the ear is slender and with a length of 28 mm., while the Phoca has no pinna. In both animals the auditory tube is of considerable length, but because of the more arched cranium of the otariid, especially in old males, the auditai orifice is located relatively less dorsad than in the phocid. Thus, in the latter the nostrils, eyes, and ears occupy a position more decidedly dorsad. The neck of a small otariid is very mobile, and as it is relatively slender, it appears longer than the very broad neck of a phocid, tapering, as it does, to the head. In the eared seals the necks of bulls develop to an astonishing extent, however, this being partly muscular for combat and partly fatty. In this family the base of the neck and _ thorax are cylindrical, or even slightly flattened transversely in form, while in Phoca the tendency has been farther away from the typical terrestrial carnivore and there is an appearance of slight flattening dorso-ventrad. The whole body appears definitely longer in the sea lion, but this is difficult of proof. In this animal the lumbar region is exceedingly limber, because of the elasticity of the inter- vertebral disks and the form of the vertebrae themselves, this being so largely as an aid to terrestrial locomotion; but-such is not the case in the earless seals, in which there is apparently no marked ability to bend the lumbar region ventrad. In the otariid the tail was vir- tually conical with a length of 60 mm., but in the phocid this mem- ber, 72 mm. long, was flattened dorso-ventrad, measuring 43 mm. in width by 25 mm., and fit nicely into the angle formed by the adpressed hind flippers. In the otariid the axillat was at a point a trifle proximad of the center of the ulna, but in the Phoca it was opposite the ulnare. In the former the visible portion of the fore leg had a length of 300 mm. from the axilla and was very highly modified as a paddle, being 1The term axilla as herein used is employed to designate the ventral and caudal junc- ture of the fore limb with the body, not in its more precise meaning of the region beneath the shoulder joint. ART, 15 ANATOMY OF THE EARED AND EARLESS SEALS—-HOWELL i thicker and longer upon the cranial border and thinner and shorter upon the caudal one. Each digit had a minute circular nail set within a pit in the integument, located approximately at the end of the terminal phalanx, and a cartilagenous prolongation of the digits extending considerably farther. The distance from the metatarso- phalangeal joint of the first digit to the nail was 85 mm., and from the nail to the termination of the digit, 65 mm. The entire manus was rather stiff, without free movement of the digits, and there was a minimum of possible abduction or adduction. The interdigital membrane reached to the ends of the digits, and the sole was naked and wrinkled. In the Phoca the external part of the foreleg meas- ured but 100 mm. from the axilla and was not highly modified into a paddle. The manus was short and broad and hairy upon the palm, it was abducted more than in most mammals, and the articulations of the carpal bones were loose to the touch. As in the otariid the first digit was the longest, and there was sequential diminution to the fifth, but in that animal the difference in length between these two was very marked and in the phocid very slight. There are broad, heavy nails upon all the digits of the latter, and there are no cartilagenous extensions. The crotch, between the hind limb and tail, was in the otariid at the most caudal of the muscles between the innominate and the leg, but in the Phoca conditions differed, for there was a considerable distance between the crotch and the last muscles, this being occupied by tough fatty tissue. In the former animal the crotch was opposite the middle of the calcaneal tip (heel), while in the phocid it was located about 20 mm. farther caudad, a distance not sufficient to make much differ- ence in the mobility of these members. As discussed more fully elsewhere the hind foot of the otariid readily assumes a plantigrade position at an angle of 90° with the shank, while in the Phoca the angle so formed, without undue forcing, does not exceed 60°. From the crotch the hind limb of the latter measured 220 mm. in length, and of the otariid, 260. When viewed from the rear with joints ag relaxed as it was possible to get them in the preserved specimens the plantar planes of the feet of the otariid presented the appearance of a very steep-sided V, and in the Phoca, of an equally steep-sided A. Pronation and supination corresponded with these positions, but sa little of the leg projected from the body that one could not determine the precise amounts. In both animals the first and fifth digits are considerably the most robust and the longest. In the Zalophus only there is also a cartilagenous extension to each digit. The distance from the metatarso-phalangeal joint to the nail of the first digit measured 77, and from the nail to the tip of the toe, 78 mm. The cartilages of the middle three digits were relatively a bit shorter, and that of each toe projected beyond the interdigital membrane, as shown 8 PROCEEDINGS OF THE NATIONAL MUSEUM VOL. 73 in Figure 24. The extended breadth of the hind foot at the nails was 133 mm., but the difference between the collapsed and extended width was slight. Upon the first and fifth digits there were little more than nail pits, as in the case of the fore foot, but the nails of the middle digits were long, slender, and almost straight. Unlike the case of the fore foot, the digits of the pes in this genus are capable of con- siderable flexure, even to the cartilages, and the latter may be flexed out of the way of the nails, permitting the use of these in scratching. The sole is naked, but is covered with hair in the phocid. In the latter the first and fifth toes are relatively more robust than in the otariid and possible abduction of these digits is much greater, partly because of the more generous width of the interdigital membrane, Thus the greatest width of the foot at the base of the nails was 190 mm. in this specimen. Whereas in the otariid the static posture assumed by the hind feet in the water is usually somewhat trailing and relaxed, in the phocid they are usually adpressed and, especially on land, carried straight out behind and unsupported. All five digits have exceedingly slender, almost straight nails. OSTEOLOGY. It is intended here to present not a complete description of the skeleton of two species of pinnipeds as such, but rather to compare critically two skeletons which are considered to be representative of the families Otariidae and Phocidae. Minor differences in the interrelationship of the bones of the skull are not dwelt upon, but rather is it intended to enumerate and attempt to evaluate those differences which are belived to be functional, as well as the phylo- genetic ones, to discover why and to what degree pinnipeds differ from fissipeds, and in just what manner otariids differ from phocids. The osteology of the pinnipeds has been described by a number of others but no one heretofore has investigated the mechanical and myological reasons for their osetological details. It has been impossible in the present instance to accumulate skeletons of all, or even satisfactory material representing the ma- jority of, pinnipeds, but comparisons of skulls have been made, family characters as based upon cranial details have been checked over, and where the characters of the otariid or phocid differ from those common to their respective families, these are mentioned. Other than of juveniles but one skeleton of Phoca hispida (No. 6297, Mus. Comp. Zoodl., Cumberland Gulf, April, 1878, sex un- recorded, by L. Kumlein) could be located, and this is compared with one of a subadult male Zalopus californianus (No. 200847, U.S.N.M., from National Zoological Park, December 19, 1915) which is nearer the same age than any other at hand. For comparison certain meas- ART. 15 ANATOMY OF THE EARED AND EARLESS SEALS—HOWELL 9 urements and percentages are also presented of the skeleton of an adult cat (Felis catus). Some vertebral measurement was needed as a standard of com- parison but it was thought advisable to exclude from this the cervical series. The sum of the length of the thoracic, lumbar and sacral vertebrae has therefore been taken as a yardstick. It was found that this measurement in the Phoca skeleton was almost precisely seven-tenths that of the Zalophus. In drawings of single bones of the latter the scale is, therefore, as though the osteological details of this individual were precisely seven-tenths of their true size. At- tention should be called to the fact that classification of the osteo- logical characters as being of myological or phylogenetic derivation is at times arbitrary and purely for convenience. Any character doubt- less become phylogenetic if present for a sufficient length of time. By “transverse process” is meant any: vertebral process situated laterad without reference to its homology, as of the atlas or of a sacral vertebra. SKULL The illustrations give a better idea of the general form of the skull than can a description. That of the Zalophus is long and narrow, and in the Phoca short and broad, most of the difference in length oc- curring anteriorly. In the Zalophus the skull is 27 and in the Phoca 23 per eent of the body length. In the same order the glenoid-rostral measurement is respectively 70 and 61 per cent of the total length of the skull, and that for breadth to length is 53 and 66 per cent respectively. Beginning rostrad it is seen that there has been a slightly greater recession caudad of the anterior nares, relative to total length of skull, in the Phoca. In the Zalophus there is a well defined process formed by the premaxillary tips, which is absent in the Phoca. The reason for this is apparently either muscular or possibly cartilagenous, but nothing to account for it was met during dissection. Similarly the anteorbital processes of the maxillae, pres- ent in Zalophus only, should be due chiefly to details of the orbicu- laris oculi, and possibly also the frontalis, but as mentioned later I am not reporting upon the facial musculature. The absence of supra- orbital processes in the Phoca and their presence in Zalophus is cor- related first with the lack in the former of a distinct “interorbital ” extension of the temporalis, with the greater size in that animal of the eyes and the true orbits (as distinguished from the anterior temporal fossae), and their more dorsal position, or rather, the more pronounced ability of the eye to look straight up. This more dorsal direction of sight can be accomplished in two ways—(a) by a bowing out and broadening of the zygomatic arches, accompanying which change there must be either a decided increase in the strength of the 10 PROCEEDINGS OF THE NATIONAL MUSEUM VOL. 73 masseter muscles or a corresponding increase in the breadth of the lower jaw, accompanied by certain rather complicated changes, both muscular and osseous, in the region of the palate; or (6) by a decrease in width of the interorbital septum, to this extent allowing the eyes to roll upward and inward. The interrorbital septum of all pinni- CLEIDOMAST.~ TRACHELOMAST: \ j TEMPORALIS HY STERNOMAST. i SPLENIUS PART 7 BIWENT. CERV. LP” Cea ; << RECT. CAP. POST, MAT. RHOMB. ANTICUS/ | —_ J-— OBLI@ CAP. SUPER. | eas COMPLEXUS loscuntian Fic. 2.—DoOrRSAL VIPW OF THE SKULL OF ZALOPHUS SHOWING AREAS OF MUSCLE ATTACHMENTS LABELED IN CAPITAL LETTERS ; NAMES OF BONES IN SMALL TYPH peds is proportionately thinner than in existing terrestrial carniv- ores, indicating that even in the Zalophus in which interorbital breadth is 11 per cent of total skull length, dorsal vision is used con- siderably. In the Phoca, however, this is excessively thin, its width dorsad being but 3 per cent of the length of the skull, and more ART. 15 ANATOMY OF THE EARED AND EARLESS SEALS—-HOWELL 11 ventrad almost paper thin, allowing the animal to look straight up with ease. In some of the Phocidae (as Stenorhynchus) the inter- orbital septum is not more reduced than in the sea lion, the skull is long and narrow and the orbit is not especially large. In the Phoca hispida, as previously mentioned, the zygomatic width is relatively great and the orbit proper, large. The size of the eye has resulted ——__ WS TEMPORALIS V/ RAOMB. ANTIC: CEPHALOHUMERAL SEMISPIN. CAP AUMEROTRAPEZIUS OBLIA. SUPER — RECT. CAP POST ACCESS Z7 RECT. CAP. POST, MAT. RECT, CAP. POST. MINOR aS Fic. 8.—DorRSAL VIEW OF THE SKULL OF PHOCA HISPIDA, SHOWING AREAS OF MUSCLE ATTACHMENTS AND (IN SMALL TYPH) NAMES OF BONES | occipital in the forcing caudad of the postorbital process of the zygoma to a point immediately caudad of the malar-squamosal suture, while in Zalophus (and some phocids, as Stenorhynchus) this process is located considerably craniad of this suture. Within the orbit it is seen that the excessive thinness of the inter- orbital septum of Phoca hispida has crowded the ethmoturbinals to “ 12 PROCEEDINGS OF THE NATIONAL MUSEUM vou. 73 such a degree that these have actually forced their way here and there through the fragile frontals which overlie them. An occasional Zalophus skull has what appears to be an imperforate lachrymal be- tween the maxilla and frontal just within the orbit. In the search for this bone in the Phocidae one needs be careful to identify the maxilloturbinal, a smooth part of which is often to be seen on the border of the orbital vacuity. Certainly the bone is entirely absent in the vast majority of phocids, having either dwindled and disappeared from the maxillo-frontal suture, or what seems more logical, has been forced by the enlargement of the orbit relatively farther caudad, and now represented perhaps by cartilage within the confines of the maxillo-frontal vacuity, which cartilage would dis- appear during cleaning of the skull. In Stenorhynchus only among phocids have I been able to find a bone which might represent the lachrymal and the homology of this is uncertain. In a single skull of this genus there is a small bone bounded by the vacuity, maxilla and palatal which may be the lachrymal. The maxillo-frontal vacuities of the orbit are usually much larger in the otariids than phocids, although the smallest in the former may be no larger than the greatest in the latter. The infraorbital foramen is relatively larger in the Zalophus than in P. hispida, but there is much varia- tion in this item within the two families. Its size in the pinnipeds is an index to the development of the infraorbital nerve, which serves the mystacial pad. The maxillo-naso-labialis muscle, which is the chief mechanism for opening the anterior nares, arises from the maxilla directly caudo-ventrad of the infraorbital foramen. The point of origin is not indicated upon the skull of Zalophus nor of most phocids, but it is marked by a relatively deep fossa in P. hispida. The only noteworthy feature of the zygoma that has not been mentioned is the apparent fact that in the Otariidae the malar extends to the glenoid fossa, which it does not quite do in the Phocidae. One not infrequently encounters the statement that in the Otariidae the dorsal process of the zygomatic arch occurs definitely craniad of the jugal-squamosal suture, while in the Phocidae it occurs either at this point or a bit caudad. This is a secondary character and the position of the process is attributable to the relative size of the orbit, and hence, of the eye. As usual the zygoma tells little in regard to the masseter muscles. The molars of the Otariidae are simple and conical, evidently hay- ing assumed their present form because of the slight use to which they are put in simply helping to tear fish and similar food, rather than in shearing tough meat and gnawing bone, as is the habit of the fissiped. The molars of the Phocidae are of a more complicated form and have at least two cusps—often more. Presumably the food predilections of all pinnipeds are very much the same, and it is not ArT. 15 ANATOMY OF THE EARED AND EARLESS SEALS—HOWELL 13 only impossible to explain the dental differences between the two families at the present time but in the absence of ancestral remains it is unsafe to theorize on the probable development of the tooth pat- terns. The dental arch is different in the two families, being narrow in the Zalophus, with alveoli almost parallel, while these diverge to a greater extent in phocids, with Stenorhynchus occupying an inter- mediate position in this as in so many other respects. The palatal region is narrow in Zalophus and broad in the Phoca, while the hamular processes are located far caudad in the former of their position in the latter animal. In classifications of the Pinnipedia attention is usually called to the presence in otariids and the absence in phocids of the alisphenoid canal, mediad of the glenoid fossa, for the passage of the external carotid artery. This is a convenient character for classification but is not necessarily a precise criterion, for in the skull of a fur seal (Callorhinus alascanus, No. 237266 U.S.N.M.) this canal is present upon the left side but absent on the right. In Zalophus the foramen rotundum (for the maxillary nerve) is just craniad of this canal, and in the Phoca, mediad to the anterior part of the glenoid fossa. Next caudad is the foramen ovale (for the mandibular nerve), and near this, the Eustachian canal. The apparent position of the stylo- mastoid foramen (for the facial nerve) differs in the two animals, but in both it is located as usual between the audital bulla and the mas- toid. Ventrad of the basioccipital level upon the medial side of the bulla in the Phoca is the carotid canal (for the external carotid artery) and the direction of this is latero-dorsad. In the Zalophus this foramen is mostly dorsad of the lateral margin of the basioc- cipital, and its direction is directly craniad. A probe introduced from the caudal end emerges into sight next to the Eustachian canal. The jugular, or posterior lacerated foramen is next caudad, this being a large fenestration in both, but somewhat more extensive in the Phoca. Between this and the condyle is the condyloid foramen (for the hypoglossal nerve), larger in the Zalophus, but this difference is not found to be uniform in the other pinnipeds. The pterygoid fossa is much broader in Phoca hispida than in Zalophus but the significance of this is not readily interpreted. The irregularity of the surface of the basioccipital in the latter indicates stronger anterior rectus capitis muscles in this animal. In this as in certain other species of phocids (as Cystophora) there is a large “medial vacuity in the basioccipital, which seems never to occur in otariids. In all eared seals apparently the occipital condyles are relatively narrow, while in the Phocidae they are much more flaring, this being especially pronounced in Phoca hispida. The reason for this is evidently that the articulation of the head with the neck has 14 PROCEEDINGS OF THE NATIONAL MUSEUM VOL. 73 greater need for mobility in otariids, which is accomplished partly by a reduction in the width of the joint. The paroccipital processes vary greatly in the degree of their distinctness in pinnipeds. In otariids they are apparently always well defined and continuous, by a crest, with the mastoid processes. 1a, | i { F. INFRAORB., , W) | | 3 Ua | f J a = } ' f, ZG SH /, 4 Li 4 aie A SEZ rT MASS. \ branes i \LISPH. CANAL is MASS. PROF | : NIPTERYG INTERN | presphen: [| > : pterygoid. Ly ; alisphen. — SE Na > al * Bi HAE & sor CUE AM | ALK bastsphen WSs hi S A i ‘ ‘o PTERYG, EXT. eT, my \ . ele \ eis t f FL STYLOMAST 7 _ Ve : J PSDIGASTRIC. F. CAROTID . +6 mi ~ i STYLOGLOS. FLACERUM POST.“ Se la \C YRECT. CAP. ANT. MINOR Fconpye/ Sy ! | AECT. CAP. ANT: MAT. Fig. 4.—VENTRAL VIEW OF THE SKULL OF ZALOPHUS ; NAMES OF BONES IN SMALL TYPE In those phocids in which they are distinct they are never continuous with the latter, but they may be either sharp and projecting (Monachus, Stenorhynchus), or absent as true processes (Mtrounga), their position being indicated merely by slight swellings in that part art.15 ANATOMY OF THE EARED AND EARLESS SEALS—HOWELL 15 , of the bone. The latter description fits the condition in Phoca hispida. In the Zalophus the paroccipital-mastoid process begins at the posterior lacerated foramen and extends caudo-laterad. The crest then turns and extends craniad, broadening abruptly and ending as the caudal boundary of the external auditory meatus. Even in the subadult skull there is no indication whatever of a distinct mastoid, it having to all intents disappeared in the parietal-occipital MAX. NASO-LAB. «& RASA: SUPERF. aan PROF. vomer-}-@ EST. TUBE, Gall presphen-h pee a pterygoid alisphen-@& MEATUS AUD: squamosal-— /SPLENIUS bastoccip XS ING Q\ \ WSTRACHELOMAST. F. STY LOMAST. STERNOMAST. F. CAROTID DIGASTRIC. F. LACERUM POST: a we RECT. CAP. ANT. MINOR F CONDYL RECT. CAP. ANT. MAT. Fig. 5.—VENTRAL VIEW OF THE SKULL OF PHOCA HISPIDA; NAMES OF BONES IN SMALL TYPE fusion. In Phoca hispida there is the slight swelling denoting the position of the paroccipital process, and from this, a lateral bulge of the somewhat inflated mastoid, nonmuscular in character, to the moderately developed mastoid process located caudad of the bony lip of the auditory meatus. As previously inferred, in those phocids which have prominent paroccipital processes there is no continuous prominence from this to the mastoid process. In a fetal Phoca vitu- 16 PROCEEDINGS OF THE NATIONAL MUSEUM VOL, 73 lina the paroccipital process is much better developed than even a postnatal individual of Callorhinus. If the paroccipital-mastoid crest of the Zalophus be viewed as an extension ventrad of the bone, there is little muscular stimulus to be seen—nothing save that of the digastric. If it be viewed as chiefiy a lateral development, then there is more reason. In Phoca hispida none of the long muscles is attached to the lateral part of the oc- cipital crest, the three muscles of the mastoid process are narrowly tendinous, and the digastric arises from a pit between the mastoid and bulla. This accounts at least partially for the suppression of the paroccipital process as such in the Phoca—for there are no muscles attached thereto—and for the moderate size of the mastoid process, with its three muscles. In Zalophus, however, the powerful cephalo- humeral, the splenius-trachelomastoid insertion, and the sternomas- toid—which always has a significant effect upon its insertion—are at- tached to the latero-ventral part of the occipital crest, and in ad- dition, the digastric arises all along the paroccipital-mastoid crest. There is no attachment at all confined to the mastoid process in its restricted sense save the weak cleidomastoid. The audital bulla in otariids is small, shrunken-looking and often rugose, extending to form a projecting lip to the auditory meatus directly ventrad of its orifice. There are certain phocids with an intermediate type of bullae, as through Monachus to Stenorhynchus and then Phoca, the culmination being in Cystophora with its great globose bullae: and in phocids the projecting lip of the meatus is situated more caudad. A fundamental difference in the bullae of the two families is to be seen in the fetal state. In a Callorhinus skull (length 96 mm.) the bullae are very small, noninflated, with tympanic ring very distinct, and border of the auditory meatus regular and subcircular. In one of Phoca vitulina (length 113 mm.) the bullae are perhaps 10 times as large, roundly inflated and' with the ecto-entotympanic suture almost obliterated, the two parts com- bining to form a single evenly rounded surface. —-—+4 -—- —SEMITEND:——_- ——_- -—--— > SOLEUS TIB. POSTA_ FLEX. DIG. LONGA_ | | ea PERON. DIG. 5~__ Fic. 18—LEFr TIBIA AND FIBULA OF ZALOPHUS (Z) AND PHOCA HISPIDA (P); IN LATERAL VIEW ABOVE AND MEDIAL BELOW VOL. 73 art. 15 ANATOMY OF THE EARED AND EARLESS SEALS—-HOWELL 41 drical, while in the latter it tapers from both ends toward the center, and there are two well defined fossae—one cranio-laterad, from which arises the peroneus brevis, and the other caudad, giving origin to the flexor hallucis longus. In the otariid the origins of both these muscles have a somewhat different relationship in regard to the fibula. The distal head of the otariid fibula is rather short, being definitely shorter than is the medial malleolus, relatively smooth and without grooves, and with but slight articulation with the ankle. In the Phoca the fibula projects distad of the medial malleolus, its surface for articulation with the ankle is almost as extensive as that of the tibia, and there are two very deep grooves laterad. The more caudal of these is for the peroneus digiti quinti, and the more cranial for the peroneus longus. Foot.—If the osteological length of the foot for the present purpose be considered as comprising the distance from the tip of the terminal phalanx of the second digit to the caudal part of the condyle of the astragalus, then for the Zalophus and the Phoca, respectively, it con- stitutes 28 and 34 per cent of the body length, and 45 per cent of the leg length in both. The tarsal elements of these two pinnipeds comprise astragalus, calcaneum, centrale, first, second, and third tarsales, cuboid, and a medial sesamoid. The astragalus is especially interesting and ex- hibits differences of much significance. In the Zalophus the position of this bone is slightly more dorsad of the calcaneum, the tibial facet is much larger than the fibular and its slope is more sharply ventrad, in part to compensate for the more distal position in this animal of the medial malleolus as compared to the lateral. This facet extends craniad in Zalophus only just onto the neck, and caudad in the Phoca only onto the posterior extension of the bone. From the fibular facet there is a broad process extending cranio-laterad, absent in the Phoca, and in the Zalophus, a constricted neck and expanded head, with an extensive, rounded facet for articulation with the centrale. In the Phoca the neck is of greater diameter than the head, and ventrad the articular facet is more extensive, a result of the greater degree of movement possible in this animal. In the Zalophus no part of the bone extends caudad of the tibial facet for a greater distance than a couple of millimeters. In the Phoca the astragalus is prolonged caudad in a truly remarkable process which extends beyond the termination of the calcaneum and is grooved for the pas- sage of the flexor hallucis longus tendon. It is the tendon of this muscle only and its tension operating on the process of the astragalus, that prevents the foot of Phoca from assuming a position at a right angle to the shank. The calcaneum of Zalophus is markedly constricted in the middle, being moderately expanded caudad and greatly so craniad, but that 42 PROCEEDINGS OF THE NATIONAL MUSEUM VOL, 73 of the Phoca lacks any such constriction. As might be expected from the great caudad development of the astragalus in the phocid the caudal process of the calcaneum is correspondingly reduced in robustness, which is not the case in the otariid. In the latter the anterior and posterior facets (the anterior and middle facets of man are fused in the Pinnipedia) are less extensive than in man, and still smaller in the phocid. Dorsad in the Phoca only there are two deep grooves, the more medial for the tendons of the peronei brevis and digiti quinti, and the more lateral for the peroneus longus. In the otariid only the expanded distal portion allows slight contact with the centrale. The centrale of Zalophus is flat in a cranio-caudal direction but broad transversely, while in Phoca it is more irregular, is relatively narrow transversely and deep dorso-ventrad, making the narrowest a SCT \ ASTRAGALUS——-— ; CENT RALE~_ _ —-——CUBOID SESAMOID—--Q TARS ALES—- f } Fic. 14.—DorsAL VIEW OF LEFT TARSUS AND METATARSUS OF ZALOPHUS (Z) AND:-PHOCA HISPIDA (P) part of the tarsus through the centrale-cuboid, while in the otariid it is across the neck of the astragalus and through the calcaneum. The lateral part of the cuboid is deeply grooved in Phoca only, for the passage of the peroneus longus tendon, and the whole bone is not only more irregular in shape, but relatively a bit larger than in the Zalophus. Tarsale 1 is very large, corresponding to the size of the hallux, but les considerably more laterad (less dorsad) in the Phoca. In both animals tarsale 2 is slightly smaller than tarsale 3. Mediad to and fairly between tarsale 1 and the centrale of Zalophus is a well- defined sesamoid bone. It is mostly embedded in the tendon of the tibialis posticus but in adults its position is osteologically indicated by small but distinct articular facets upon both bones. In the Phoca dissected there was apparently no sesamoid occurring as a real bone, - =eo~ =e Bs ART. 15 ANATOMY OF THE EARED AND EARLESS SEALS—--HOWELL 43 but in the skeleton studied there was one, relatively as large as in the Zalophus, but in contact only with tarsale 1—not with the centrale. In the otariid the four lateral] metatarsals have responded to a transverse crowding and proximad are flattened in this direction, while the large hallux is very much flattened and transversely ex- panded. The fifth is slightly more robust than the three middle metatarsals. In the Phoca there has been equal or greater crowding proximad but the metatarsals have been less responsive, interlock to a much greater extent, and send a number of processes here and there. Another result of this crowding of the tarsal elements is the slight forcing in a plantar direction of the first, and more decidedly of the fifth, metatarsals of the phocid. It is apparent that in the latter there is some decrease in the mobility of the tarso-metatarsal articu- lations. As with the otariid the fifth metatarsal is slightly more robust than the three middle ones, and the first still heavier, but the difference in size is not so pronounced. In profile the dorsal outline of the metatarsals in this genus. is markedly concave, to a conspicu- ously greater degree in this animal than the otariid. In both animals the first and fifth digits are longer than the middle three and in the otariid there is a tendency toward flattening of the _ phalanges. Several investigators (as Kukenthal, 1890) have stated that all except the terminal phalanges of the pedal digits of the pinni- peds show distal as well as proximal epiphyses, in this respect ap- proaching conditions in the Cetacea. As far as my own experience is concerned, there is but one small juvenal Callorhmus available which seems to have nodules of bone in the cartilage of the toe joints repre- senting distal epiphyses. All other individuals at hand are suffi- ciently old so that no line of fusion at the distal ends of the phalanges can with certainty be traced. The middle three digits of the Zalophus have well-formed but slender nails, while those. of the first and fifth are rudimentary. In otariids there is a terminal cartilage projecting in each pedal digit beyond the nail, and rather scanty material leads me to believe, for the present at least, that these cartilages are relatively longer in juveniles than in the adult. In the Phoca the nails are better formed, those of the first and fifth being larger than the other three. They project beyond the tips of the toes, as there are no terminal cartilages in this family. MYOLOGY In the following pages the comparisons are based upon the mus- culature of the Zalophus dissected, notations being made upon the muscles of the Phoca only when these showed details that differed. Attention is called to points wherein Murie’s H'wmetopias and Odo-. benus, or Miller’s Arctocephalus and Phoca differed from the condi- tions in the respective families as I found them to be represented by 44 PROCEEDINGS OF THE NATIONAL MUSEUM VOL. 73 the animals dissected by me. But in some instances the above authorities—especially the former—omitted reference to certain muscles, or groups of muscles, and in other cases the descriptions were so involved as to be obscure. All details of differences are not given, of course, but only those of sufficient degree or quality to be deemed of some importance. The differences in origin and insertion of a muscle in the two animals are not always as great as indicated in the osteological drawings. Thus one investigator might consider that the vastus profundus (fig. 12) arises in Zalophus from only the proximal part of the femur, and in Phoca, in which animal there is slightly more tenacious attachment, that the origin is from the entire shaft, although in reality the muscular difference is very slight indeed. As previously mentioned the Zalophus dissected was a juvenile and excessively lean, but its condition of preservation was excellent. The Phoca, on the contrary, was very fat and its condition poor, due in part to rupture of some of the biood vessels, especially about the shoulders, and consequent infiltration ‘between the muscles of blood, the caked condition of which made dissection difficult at times. The coloration of the muscular tissue in these preserved specimens was about the same as in the usual fissiped, but it is common knowledge that the flesh of a freshly-killed pinniped, as well as of the Catacea, is unusually dark. As the flesh of the horse is of a similar color, however, this is not necessarily connected with aquatic specialization. It may not be out of place here to mention that the musculature of the Pinnipedia does not differ as greatly from that typical of the terrestrial Carnivora as one might expect from the osteology, and there were but few times when I experienced any difficulty in readily homologizing the musculature to my satisfaction. MUSCLES OF THE HEAD : 1. SUPERFICIAL FACIAL MUSCULATURE For the reason that Ernst Huber is making a separate study of the facial musculature of both specimens which I dissected, there will be but four of this group mentioned in the present report, this for the reason that they are intimately concerned with functions which it is wished to discuss. M. platysma (fig. 15) was extensive in both Zalophus and Phoca but was not otherwise peculiar. M. naso-labialis arose mediad of the anterior orbit and near the middorsum of the rostrum, diverging slightly fanwise, and inserting into the mystacial pad. M. maxillo-naso-labialis (figs. 4,5) arose from the zygomatic process of the maxilla caudad of the infraorbital foramen. In the Zalophus there was no osteological indication of origin, but in the Phoca the ———— ART, 15 ANATOMY OF THE EARED AND EARLESS SEALS—-HOWELL 45 muscle was considerably heavier and the point of origin is marked by a relatively deep fossa, but this is not present in all the Phocidae. Insertion was into the mystacial pad deep to the naso-labialis. M. mandibulo-auricularis was a complex of several minute and slen- der muscles which were not separated. Origin was from the dorsal surface of the zygomatic root of the squamosal craniad of the audi- tory meatus, while insertion, seemingly more intricate in Zalophus, was about the auditory tube where this reaches the body surface. 2, MASTICATORY MUSCULATURE M. masseter (figs. 4, 5, 6) was partially divisible in the Zalophus into two portions. ‘The more superficial arose from the cranial third of the zygomatic arch and was inserted along the border of the mandible ventrad of the masseteric fossa, from just rostrad of this to the tip of the angular process. The deeper portion was insepara- ble craniad from the more superficial, and origin extended caudad as far as the tip of the jugal. Insertion was into the entire masseteric fossa of the mandible ventrad of the base of the coronoid process. In the Phoca this muscle was completely divisible, the more super- ficial arising from the full extent of the jugal, with insertion along the caudo-ventral border of the mandible from just rostrad of the angular process almost to the condyle. The deeper part arose simi- larly from the rostral end of the jugal caudad to the capsular liga- ment of the glenoid fossa. Insertion was upon the whole caudo- ventral half or more of the masseteric fossa of the mandible. M. temporalis (figs. 2, 3, 6) of the Zalophus was divisible along its posterior portion into a superficial and a deep part. The former arose by aponeurosis from the mediai border of the temporal fossa, rapidly thickened, and was inserted upon the lateral surface of the coronoid process, its more superficial fibers beneath the zygoma blending with the adjoining masseter. The deeper part arose from the entire temporal fossa and inserted upon the medial surface of the whole coronoid process. In the young animal dissected the tem- porals did not yet approach the middorsal line. In the Phoca the temporal was not divisible and was very much weaker and less exten- sive. Insertion was for a short distance upon the lateral, and upon the entire medial surface of the coronoid process. M. pterygoideus externus (figs. 4, 5) arose in the Zalophus from the bony bridge over the alisphenoid canal, with insertion upon the roughened area directed cranio-mediad upon the medial condyle of the mandible. In the Phoca the origin was analogous, from the bridge of bone separating the diner ovale and scarier Inser- tion was more robust, rostro-ventrad of the condyle upon the medial mandible. Miller found conditions similar in P. vitulina and Arcto- 46 PROCEEDINGS OF THE NATIONAL MUSEUM vou, 73 cephalus, but the muscle was not mentioned by Murie for Humetopias or Odobenus. M. pterygoideus internus (figs. 3, 4) in the Zalophus was much larger than the externus, arising from the border of the bony wall of the interpterygoid fossa and practically coinciding with the extent of the palatal bone along this border. Insertion was below the condyle of the mandible. In the Phoca origin was similar save that it extended as far rostrad as the maxillary root of the zygoma. Inser- tion was along a ridge upon the medial side of the caudal border of the mandibular ramus that extended from the angular process to within a short distance of the condyle. Substantially the same for P. vitulina and Arctocephalus but not mentioned by Murie for Hwme- topias and Odobenus. 3. INTERRAMAL MUSCULATURE M. digastricus (figs. 4, 5, 6, 18, 19) was monogastric in both. In the Zalophus origin was from the whole lateral border of the paroccipital process, with insertion upon the caudal three-fifths of the ventro- medial border of the mandible, extending also just laterad upon a shght prominence ventrad of the masseteric fossa. In the Phoca no tendinous division was noted, such as mentioned by Miller for P. vitulina, but origin was evidently the same, being from the depression upon the mastoid directly caudad of the center of the bulla. The muscle then spread so as to invest the entire audital bulla. The insertional end was much smaller than in the Zalophus, the attach- ment being to the medial side of the inferior border of the mandible, from the angular process craniad for several centimeters. Maurie found a slightly tendinous division of the muscle in Odobenus, but not in Hwmetopias. M. stylohyoideus (figs. 18, 19) in the Zalophus was represented by but a few fibers deep to the digastric, extending from the region of the stylohyal, though not definitely from any part of the temporal bone, with insertion onto the basihyal. In the Phoca there was a very thin slip from below the auditory meatus to the basihyal, but superficial to the digastric, as in man. Miller gives this normal origin and insertion for this muscle in his phocids but fails to state its relation to the digastric. In my Phoca there was another slip upon the left side deep to the digastric, as in the Zalophus, but it seemed to pass entirely dorsad of the basihyal and disappeared in the neighboring tissue. It was noted as questionable in homology and when an attempt was made to verify it upon the right side it could not be found. Maurie did not discuss the muscles of this region in his Otariidae. a ART. 15 ANATOMY OF THE EARED AND EARLESS SEALS—HOWELL 47 M. mylohyoideus (figs. 18, 19) was normal for both Zalophus and Phoca. Origin was from the inner border of the lower jaw, with fibers passing caudo-mediad, and insertion was into the medial raphe and upon the basihyal. 4. MUSCLES OF THE TONGUE M. mylohyeideus (figs. 18, 19) was normal for both Zalophus and arose from the connective tissue about the stylohyal and from a - slight prominence upon the audital bulla. (Fig. 4.) It passed closely mediad of the diagastric and thence into the tongue. In the Phoca it was similar save that origin was from the inferior lip of the auditory meatus. M. hyoglossus (figs. 18, 19) arose in the Zalophus from the thyro- and basi-hyal, deep to the mylohyoid, passing deep to the styloglossus and extending to the tongue tip. In the Phoca it was similar save that the origin was chiefly from the thyrohyal. M. genioglossus was heavy in the Zalophus and arose chiefly from the basi- and cerato-hyals, with relatively few fibers from the lower jaw. Insertion was into the tongue as usual. In the Phoca origin was from the lower jaw only, with no direct connection with the hyoid. In this animal there also seemed to be elements of a chondro- glossus muscle, represented by fibers passing from the thyrohyal into the tongue. For P. vitulina and Arctocephalus Miller reported the usual attachments to lower jaw, hyoid and tongue. MUSCLES OF THE BODY MUSCLES OF THE NECK Superficial group.—HWM. sternomastoideus (figs. 2, 5, 18, 19) in the Zalophus arose narrowly from the presternal tip and by fisciculi from its antimere (the juncture being free from the muscles be- neath) as far craniad as the thyroid cartilage. Insertion was fas- cial and along the lateral third of the occipital crest adjoining the _ cephalohumeral, and more tendinous upon the mastoid process. In the Phoca the origin was located more caudad, it being from the border of the presternum, and insertion was limited to the mastoid process. Miller described for Arctocephalus a part of the sterno- mastoid which was utterly unlike anything which I encountered. In addition to the sternal origin, he stated that a part arose from the deltoid ridge of the humerus and from the fascial slip representing the clavicle. The latter part may be homologous with what I am terming the cleidomastoid; but in another place Miller said that it constituted the omohyoid, which, because of its insertion, is hardly 48 PROCEEDINGS OF THE NATIONAL MUSEUM you, 73 likely. From Murie’s text I am unable to judge of the precise degree of resemblance with H’wmetopias and Odobenus. M. cleidomastoideus (figs. 2, 4, 18) was found in the otariids only. In two it was slender and had origin by fasciculi from the cranial border of the broad or posterior division of the cephalohumeral— this representing the clavicle. Insertion was chiefly tendinous upon the mastoid process adjoining that of the trachelomastoid. No men- tion of this muscle in Hwmetopias was made, but, Murie found it in Odobenus. Supra- and infra-hyoid group.—M. omohyoideus (fig. 18) was fairly well developed in the Zalophus. Its origin was inseparable from the deep half of the cleidomastoid, thus arising from the border of the cephalonumeral, this representing the clavicle. It passed be- neath the sternomastoid and was inserted upon the basihyal. I do not consider that it was present in the Phoca, but Miller says that in P. vitulina it was a part of the sterno-thyro-hyoid; and in Arctoceph- alus the outer margin of the sternomastoid, which latter is not to be considered seriously. M. sternohyoideus et sternothyroideus. (Figs. 18, 19.) In the Zalo- phus the former had almost disappeared. There was a single thin muscle arising from the manubrium deep to the presternum with insertion chiefly upon the thyroid cartilage but a separate slip ex- tended also to the hyoid superficial to the thyreohyoid. This was separable from the main muscle no farther caudad than the thyroid cartilage. In the Phoca these two muscles were inseparable at their origin from the lateral two thirds of the first costal cartilage. A single thin muscle band then extended rostrad separable into two parts, a superficial slip inserting upon the thyro- and basi-hyals, and a deeper to the thyroid cartilage. Miller said that in Arctocephalus the common origin was from the tip of the presternum. M. thyreohyoideus (figs. 18, 19) was normal, stretching from the thyroid cartilage to the thyrohyal. M. geniohyoideus (figs. 18, 19) in the Zalophus arose from a re- stricted area near the symphysis menti, and broadened to an inser- tion upon the entire basihyal. In the Phoca this was just reversed, for the origin was broader than the insertion, which latter extended but a few millimeters from the midventral line. Deep lateral and subvertebral group.—=M. scalenas (figs. 18, 19). I regard the homology of the different divisions of the scalenus in the pinnipeds as too uncertain to render wise any attempt at present to call them anticus, medius, and posticus, and prefer to refer to them by number. In the Zalophus there were two divisions that were almost vestigeal. (Z) arose from the third rib partly deep to the serratus magnus. It extended as a slender wisp of muscle to a weak ART. 15 ANATOMY OF THE EARED AND EARLESS SEALS—HOWELL 49 insertion upon the anapophysis of the fourth cervical vertebra. (2) was even weaker, extending from the first rib to the anapophysis of the fifth cervical, possibly with a few fibers to the fourth as well. in the Phoca there were three divisions. (7) arose by digitations from the third to fifth ribs. The one from the fifth arose deep to the most cranical slip of the serratus magnus. There was a dorsal slip superior to the most caudal digitation of the depressor scapulae and another ventral one, both arising from the fourth rib, besides a single ventral slip arising from the third rib. The fibers of all converge to form a tapering, cylindrical muscle which was strongly inserted by tendon upon the anapophysis of the third cervical only. (2) arose from the first rib and was inserted upon the anapophyses of the fourth, fifth, and sixth cervical vertebrae. (3) also arose Fic. 15.—LEFT ASPECT OF ZALOPHUS (Z) AND PHOCA HISPIDA (P) SHOWING PLATYSMA- PANNICULUS CARNOSUS SHEET OF MUSCULATURE from the first rib but dorsad of the second division. Insertion was by fibrous digitations upon the anapophyses of the third, fourth, fifth, and sixth cervicals. In Hwmetopias Murie found that inser- tion of the two divisions was upon the sixth and seventh cervicals; in Odobenus that the muscle was single, from the third and fourth ribs to the atlas. Miller gave two divisions for Arctocephalus and three for P. vitulina. The latter was much as [I found it in my specimen, save that insertion was said to be on the parapophyses of the cervical vetebrae. Both origins and insertions of Arctocephalus were given as very much more extensive than in my Zalophus. M. longus colli in the Zalophus may be said to occur in three por- tions, but the anterior or atlantic one was exceedingly complex. Its cranial part lay directly laterad of the rectus capitis anterior major 86377 —28——4 5H PROCEEDINGS OF THE NATIONAL MUSEUM VoL. 73 and arose as follows: A ventral slp from the parapophysial plates of the fourth and fifth cervicals. A slip next mediad (separated from the last by a digitation from the posterior oblique division and by the two scaleni) arose from the anapophyses of the fifth and sixth and the parapophysis of the seventh cervical. The third slip arose from the anapophyses of the fourth, fifth, and sixth cervicals. The insertion of the depressor scapulae and slips of the transversalis cervicis then intervened dorsad. A fourth slip arose from the postzygapophyses of the fourth and fifth cervicals and a fifth slip from those of the third and fourth. Between the caudal ends of these two intervened a small part of the biventer cervicis. These five slips then converged to an insertion, largely tendinous, upon the transverse process of the atlas laterad to the inferior oblique. There was no anterior oblique and it is inferred that this division of the longus colli took its place. The second division was as usual, connecting the centra of the cervical as well as an indeterminate number of thoracic vertebrae. The third division or posterior oblique arose within the thoracic cavity from the first few thoracic vertebrae. It inserted by partially tendinous slips upon the para- pophysial plates of the fourth, fifth, and sixth cervicals. In the Phoca there were but two divisions. Origins of the first, or rather second, was from a number of the thoracic vertebrae (Miller says seven for P. vitulina). It then passed laterad of the rectus capitis anterior major and inserted by four tendinous slips upon the parapophyses of the third, fourth, fifth, and sixth cervicals. Dis- section of the more anterior division was unsatisfactory because of caked blood, but origin was evidently from the medial part of the parapophyses of the third to sixth, inclusive, cervical vertebrae. Insertion of each bundle was upon the centrum of the vertebra next rostrad from the fifth cervical at least to the axis and probably to the atlas as well. There were three parts to this muscle in Arctocephalus, and but two were mentioned for Phoca vitulina, E'wmetopias, and Odobenus. In all of these the divisions were relatively simple. M. rectus capitis anterior major (figs. 4, 5) (longus capitis) in the Zalophus arose not only from the parapophyses of the third, fourth, and fifth cervicals, but from the centra of the axis and atlas as well. In the Phoca it was larger and origin was from the third, fourth, fifth, and sixth cervicals. Insertion in both was upon the mediocranial part of the basioccipital, mediad of the rectus minor. Miller’s Phoca and Arctocephalus dissections agree with my Phoca. M. rectus capitis anterior minor (figs. 4, 5) was very small, arising from the cranial base of the transverse process of the atlas. In both the Zalophus and Phoca insertion was upon the latero-cranial part art.15 ANATOMY OF THE EARED AND EARLESS SEALS—HOWELL 5] of the basioccipital, laterad of the rectus major. Miller found that in Arctocephalus origin was also from the axis. MUSCLES OF THE TRUNK Muscles of the thorax.—M. panniculus carnosus (figs. 15, 18) was inseparable from the platysma. In the Zalophus the fibers of the latter extended directly ventrad over the scapula, and of the pan- niculus, converged progressively from the dorsal fascia toward the axilla, clear to the root of the tail. In the extended specimen the ventral border of this muscle passed over the hip, and therefore the knee, with a slight ventral sag, joined the dorsal border of the pectoralis, from which it was separable with difficulty, and inserted into the connective tissue distad of the medial elbow. In the Phoca a very different state of affairs obtained. There was no converging of the fibers to the axilla, but origin was somewhat laterad of the middorsal line and all fibers were uniformly directed ventro-craniad at an angle from the vertical of about 35°, covering the knee and extending almost to the base of the tail, in which vicinity the line of origin extended somewhat more dorsad. Insertion of the portion over the scapula was onto the fascia of the middle forearm, and of the remainder upon the fascia of the ventral surface slightly laterad of the midventral line. Miller found the true panniculus very similar in P. vitulina but failed to indicate the degree of convergence of the fibers toward the axilla—a most fundamental point. The platysmal part of this sheet he divided in an unnatural manner, terming the sphincter colli profundus the pectoral panniculus, and the like. Similarly, Murie without a doubt confused his dissection of the panniculus of Ewmetopias, and not only illustrated the true panniculus as extend- ing in several different directions, due possibly to the “set” posi- tion of his specimen, but the parts of the facial musculature which extend over the neck (platysma, spincters colli superficialis and profundus) are shown in incorrect relationship. M. pectoralis (figs. 9, 10, 18, 19, 20, 22) in the Zalophus was imper- fectly separable into three parts. The superficial sheet occurred as a pars anticus, arising midventrad from as far craniad as the pre- sternal tip and caudad practically to the fifth costal cartilage; and a pars posticus, arising midventrad from 50 mm. caudad of pars anticus to the xiphoid cartilage. Both of these divisions united and were inserted upon the tough fascia covering the distal palmar aspect of the forearm continuous with the fibrous sheet of tissue immediately beneath the skin of the palm. Muscle fibers ceased about at the wrist. The pars profundus had origin extending from the presternal tip to the xiphoid and converged to a tough aponeurotic insertion along the deltoid ridge of the humerus and the slight ridge extending there- 52 PROCEEDINGS OF THE NATIONAL MUSEUM VOL. 73 from to the medial trochlea and thence stretching onto the forearm. (See p. 79.) The part over the deltoid was again divisible at inser- tion, one fascial sheet extending deep to the other. It should be emphasized that in this genus the border of the panniculus carnosus is practically fused with the border of the pectoralis, forming a powerful sheet of muscle inclosing the trunk, which is of the greatest importance in the economy of the animal. This is very different from the condition in Phoca. In this Phoca the pectoral sheet was also partially separable into three divisions but in a different manner, and only toward insertion. Origin was from the midvental line from the tough tissue dorsad of the entire presternal cartilage and then ventrad of the whole bony part of the sternal complex. Thence caudad origin passed laterad of the midventral line gradually to the side near the hind limb. From these points the fibers converged to the axillary region. Here it was found that the sheet from the pre- sternum and sternum had become the most superficial, with insertion upon the medial lip of the deltoid ridge, thence by fascia into the axilla, and in the opposite direction, as far as the forearm. There was a second, lateral part, narrow and with the fibers coming from the region it the flanks, and in between these two a third division, the filers passing deep to the other two and inserting with the second division and with the thicker, ventral part of the latissimus dorsi by a single tough aponeurosis upon the medial lip of the deltoid ridge, just deep to the first division. The sternal part of the pectoral complex was robust, while the abdominal part was thinner. For EHumetopias Murie reported a single superficial pectoral (manubrium to fifth rib) inserting upon the proximal half of the humerus and to the axilla, and a second layer (first rib to xiphoid) with fascial insertion over the arm. In Odobenus he found a super- ficial division from the presternal tip to the xiphoid and ending in the fascia of the forearm, and a deeper, from the fourth costal cartilage to 6 inches caudad of the xiphoid, with insertion upon the whole length of the humeral shaft. In both of Murie’s animals there was a narrow third division, not differentiated by me in Zalophus, from the manubrium, which merged with the first division. The differences in the pectorals reported by Miller are of a relatively minor nature. Suffice it to say that in this order the pectorals are exceedingly powerful and of great extent. There is some question regarding the proper treatment in many mammals of the sheet of muscle of which the serratus magnus is a part. In the human the serratus is usually considered as compris: ing the entire muscle, but reference to the lower Mammalia indicates that it would be wiser to treat it as two muscles. This has at times been done, as by Reighard and Jennings (1901), who term the ante- rior division the levator scapulae, and I have therefore followed the art.15 ANATOMY OF THE EARED AND EARLESS SEALS—HOWELL 53 same course. This name now seems to me unfortunate, as the levator angulae scapulae is often so called and the two are entirely distinct. It therefore is preferable to refer to it according to its function which is a depressor of the scapula. In various animals this part of the sheet is variable, and purely for convenience it will be divided into a cervical and a thoracic part. M. depressor scapulae (figs. 7, 18, 19) (levator scapulae part). In the Zalophus the cervical part arose by digitations from the anap- ophyses of the last five cervical vertebrae, directly adjoining the atlantoscapularis superior. It was continuous with the costal part, arising not by distinct slips but practically continuously from the fascia over the angles of the first four ribs. This part passed deep to the costal portion of the scalenus and together with pars cervicis the whole sheet was inserted along the entire vertebral border of the scapula save for about 25 mm. of its most cranial part. The caudal 25 mm. joined the broad tendon of the serratus magnus. In the Phoca origin of the cervical portion was the same and of the costal portion, from the ventral terminations of the bony parts of the first two ribs. Insertion was upon the medial aspect of the vertebral border of the scapula from opposite the spine caudo-ventrad around the glenovertebral cartilage. .In H’umetepias insertion of this, as a part of the serratus magnus, included the atlas, this possibly being my atlantoscapularis superior. M. serratus magnus (figs. 7, 16, 17, 18, 19) in the Zalophus was exceedingly powerful and was placed to best advantage for a back- ward pull of the scapula. Origin was by fascia from as far craniad as the second rib, thus overlying the depressor scapulae costalis for the distance of two intercostal spaces. Origin continued fascial to the fourth rib and thence to the tenth it was from the bony termina- tion of each rib and the caudal border of its cartilage. The more ventral portion of the muscle was partly aponeurotic but it rapidly increased in thickness until near insertion it was some 10 mm. deep. The more dorsal fibers had an inclination directly craniad to the insertion, 25 mm. in length, upon the glenovertebral angle of the scapula. The anterior half of this was broadly tendinous, and being laterad of a part of the depressor scapulae, fibers of the latter also joined it. In the Phoca origin was from as far caudad as the twelfth rib. The digitation from the fifth passed over, and from the third and fourth under, the scalenus. It is doubtful if origin should be considered as reaching farther craniad than the third rib, as anterior to that the muscle sheet was of a much finer texture, but there was no discernible division between this muscle and the de- pressor scapulae. Insertion was as in Zalophus. Miller stated that in P. vitulina origin was from as far caudad as the tenth and in 54 PROCEEDINGS OF THE NATIONAL MUSEUM CEPHALOHUM Y PLENIUS HUMEROTRAP y NK RHOMB. ANT. gZ \ f \S SS WATLANTOSCAP. INFER. SS RSS 7 SERRAT. MAG. ae OBLIG. ABDOM. EXTERN: Y OARTORWS 14+2 TENS. FASC. FEM- GLUT. MIN. RECT. THEM: PYRIFORMAS GLUT. MED: OBTUR. INTERN. GEMEL. INFER. GLUT. MAX: QUAD. FEM. BICEPS 1—— BICERS 2 \ \P Ns emivTenD. Fic. 16.——DoRSAL MUSCULATURE OF ZALOPHUS; SUPERFICIAL LAYER UPON THE LEFT, AND MUCH OF THE NEXT DEEPER LAYER TO THE RIGHT OF THE MEDIAL LINE von, 73 me ART. 15 Se GPE eR ANATOMY OF THE EARED AND EARLESS SEALS—HOWELL BIVENTER CER. SPLENIUS COMPLEXUS ATLANTOSCAP. SUPER. HRY SPINOTRAP, | i (\ ATLANTOSCAP. (NFER. i B \\ SUPRASPIN. TRICEPS LAT: ZZ WZ . < ZA WY SEN y Atha 4 LATIS DORS\ 1-F LATIS DORSI 2B EA GLUT. MED. SARTORIUS GLUT. MAX: ADDUCT. POST. SEMITEND. @ Fig. 17—DoRSAL MUSCULATURE OF PHOCA HISPIDA}; SUPHR- FICIAL LAYER UPON THE LEFT, AND MUCH OF THE NEXT DEEPER LAYER TO THE RIGHT OF THE MEDIAL LINE 56 PROCEEDINGS OF THE NATIONAL MUSEUM VOL. 73 Arctocephalus, the eleventh rib. The description for Odobenus is obscure. Mm. intercostales externi (figs. 18,19) in the Zalophus were extensive and covered all the intercostal spaces except a part of these that lay deep to the sternocostalis. In the Phoca the external intercostals between the first three ribs did not reach much beyond the bony _ part of the costae, but caudad therefrom they gradually approached nearer the sternum. Mm. intercostales interni exhibited no peculiarity save that the lateral fibers almost met over the ribs. M. sternocostalis (figs. 18, 19) (supracostalis Murie, transversus costarum latus Lucae) in the Zalophus arose deep to the pectoral mass from the sternum between the first and fourth costae. Inser- tion was by two slips passing to the lateral (dorsal) part of the first and second costal cartilages. In the Phoca it arose from the manubrium and first two sternebrae, and insertion was upon the first costal cartilage. Muller reported this shghtly more extensive in the animals which he dissected, but Murie found it essentially the same as I did. No sternocostalis posterior, such as Miller found in Phoca vitulina between the third and seventh ribs, was encoun- tered by me. Levatores costarum muscles were noted in both genera, but were not followed, nor were any of the muscles within the thorax sought. MUSCLES OF THE ABDOMEN M. rectus abdominis (figs. 11, 18, 19) in the Zalophus was rather weak. It arose from the symphysis pubis and adjoined the mid- ventral line only as far craniad as the ninth rib. Thereafter slips inserted progressively upon the eighth, seventh, sixth, and fifth ribs, none of these being in contact with the sternum. In the Phoca it was relatively much broader and stronger. Discernible muscle fibers disappeared from its medial part at the fifth and from the lateral border at the third rib, from which point a broad tendinous sheet extended with a medial inclination to the sternum as far craniad as the first rib, but not to the presternum. In E'umetopias insertion was upon the sixth rib. Minor differences of insertion occurred in the specimens dissected by Miller, save that in Phoca vitulina the broad tendon of insertion also extended laterad to the humerus. Because of excessive contraction of the thorax in the Zalophus, the other abdominal muscles were much wrinkled. They were thus very difficult to dissect, and as the preservation of this region in the Phoca was poor, all parts, especially craniad, were not completely investigated. ART. 15 ANATOMY OF THE EARED AND EARLESS SEALS—HOWELL 57 M. obliquus abdominis externus (figs. 16, 19) in the Zalophus origi- nated from the last 10 ribs. From the sixth, seventh, and eighth the respective digitations were well marked, but caudad therefrom the slips were distinguishable only with difficulty, they being very tenuous and the fibers precisely parallel with those of the external intercostals. The digitations were not in direct conjunction with those of the serratus, but arose farther dorsad and deep to the latter. Origin was also from the deep dorsal fascia for a short distance. There was no Poupart’s ligament, as there was in H'umetopias and hence no direct connection with the iliac crest, but a sheet of tough fascia running caudo-ventrad formed a part of the border of the inguinal canal and inserted upon the tuber of the pubis. Insertion was ectad of the rectus almost to the midventral line. In the Phoca this sheet arose from the last 13 ribs, the first few digitations being rather involved in the relationship of their fibers. They were in contact with part of the scalenus, depressor scapulae, and serratus magnus. Caudad of the ribs the fibers of the dorsal border were parallel and in contact with the ilicostalis lumborum or lateral mass of the long system and connected to.it by tough fascia. The innominate border developed a Poupart’s ligament. Insertion was upon the midventral line ectad of the rectus. In Arctocephalus this muscle seems quite similar to Zalophus save that insertion is upon the brim of the pelvis, while that of Phoca vitulina is essentially similar to my Phoca. M. obliquus abdominis internus in the Zalophus arose from the cartila- genous border of the thorax, from the deep dorsal fascia, from the crural arch, extending from caudad of the ventral crest of the ilium to the pectineal process, and then from the border of the pubis to within several centimeters of the symphysis. Insertion was upon the sheath of the rectus. That of the PAoca had a similar origin save that the attachment was to Poupart’s ligament rather than the brim of the pubis direct. The fibers descended practically no farther ventrad than the border of the rectus, and inserted upon the aponeu- rotic sheet common to this and the transversalis. In H’umetopias the only innominate connection mentioned was with the crest of the ilium. M. transversalis abdominis in the Zalophus arose by interdigitations with the fibers of the diaphragm, ostensibly as usual from the carti- lagenous border of the thorax, the deep dorsal fascia, and with the internal oblique from the crural arch. Insertion was upon the deep part of the rectus sheath. In the Phoca this muscle did not extend quite to the innominate but the fibers stretched almost to the mid- ventral line. 58 PROCEEDINGS OF THE NATIONAL MUSEUM you, 73 LUMBAR MUSCLUS M. quadratus lamboram (figs. 11,24) in the Zalophus arose by slender bundles from the last three thoracic and all the lumbar vertebrae. The bundle coming chiefly from the anterior centrum of the penulti- mate lumbar extended upon the left side of the base of the femoral process of the ilium and its fibers fused with those of the iliacus. Upon the right side this slip of the quadratus inserted fully 2 centi- meters farther craniad and was much more slender. In the Phoca origin did not extend craniad of the penultimate thoracic vertebra and I could not demonstrate that any of its fibers reached the ilium, but only the sacrum. The whole muscle was very much more robust than in the otariid. MUSCLES OF THE BACK Superficial, secondary back muscles.—M. cephalohumeral (figs. 2, 8, 9, 16, 17, 18, 19) should perhaps be placed here. It is formed by the fusion of the clavotrapezius and clavobrachialis as in many carnivores lacking a clavicle. In the Zalophus it was rather complex and arose by three. heads—a narrow one from the connective tissue just laterad to the tip of the presternum; next a second narrow one from the cranial margin of the deep pectoral, and third a broad head that extended rostrad from the fascia above the entire deltoid ridge of the humerus. The three heads joined and the broad, rather thin sheet of muscle resulting inserted along the medial two-thirds of the occipital crest and thence from the middorsal line as far as the humerotrapezius. Its chief action is probably in certain move- ments of the head, but in the Phoca this was reversed, and it operates rather to move the forearm. In this animal it arose as a tenuous sheet from the middorsal line as far rostrad as the interorbital con- striction and caudad beyond the occiput. The fibers passed ventro- caudad over the side of the neck and converged to insert upon the lesser tuberosity and the adjoining portion of the greater as well. Miller stated that in Arctocephalus the more caudal attachment was to the humerus only. The condition of this muscle in his Phoca was the same as in mine, save that insertion was chiefly upon the greater tuberosity. In Odobenus caudal attachment was to the humerus only, but as near as I can tell the condition in Hwmetopias was more comparable with that of Zalophus. M. hamerotrapezius (figs. 3, 7, 9, 16, 17) is homologous with the usual acromiotrapezius. In the Zalophus it arose from the middorsum ad- joining and in the same plane with the cephalohumeral, the origin extending caudad over a part of the spinotrapezius. The fibers were directed latero-ventrad and it was free from the underlying spine of the scapula along the dorsal two thirds of the latter, but to the ven- ART. 15 ANATOMY OF THE EARED AND EARLESS SEALS—-HOWELL 59 tral third it was intimately attached by fasciculi. Near the insertion its cranial border was overlapped by the cephalohumeral. It in- serted, mostly deep to the latter, along the entire deltoid ridge of the humerus and its distal border extended even as far as the forearm, the fibers not ending in fascia but seeming to be incorporated with the supinator longus in company with the deltoid. In the Phoca this muscle did not overlie the spine. It arose from the middorsum from mediodorsad of the spine almost to the occiput, this being partially deep to the caudal border of the cephalohumeral. Inser- _ tion was robustly along the entire spine of the scapula and also by lot ~ ce TR oe gn tii be hee wee re ee ee Ts aes ee A= Bara eae fascia along the caudal border of the deltoid ridge as far distad as a crater-like fossa, into which inserted a stout tendon which diverged from the humerotrapezius farther dorsad. For Arctocephalus Miller’s description is not entirely clear, but the muscle seems to resemble that of my Phoca; and H'wmetopias also conforms largely to this pattern—certainly not to that of Zalophus. On the other hand conditions in the latter genus were very similar to what Murie encountered in Odobenus. M. spinotrapezius (figs. 7, 16, 17) was long and slender, arising from the middorsum toward the caudal thorax. In the case of the Zalophus there was a deep, tough aponeurosis upon which most of the fibers inserted; but more ectad there was a fiat tendon developed and this ran cranio-ventrad to the spine near the ventral border of the muscle. That of the Phoca was similar save that the muscle fibers inserted directly upon the spine. In Arctocephlus insertion is said to be by the fibers blending with the dorsal surface of the deltoid, M. latissimas dorsi (figs. 9, 16, 17) in the Zalophus arose from the dorsal fascia, extending from a few centimeters craniad of the glenovertebral angle of the scapula to the vicinity of the last rib. The anterior border passed superficial to the glenovertebral angle and the fibers of the whole sheet converged to two partially separable insertions—the more dorso-cranial one, representing perhaps two- ‘thirds of the muscle, to a fascial insertion along the border of the teres major, and the other, to a fleshy insertion along the dorsal border of the insertional end of the pectoralis. For H’wmetopias this muscle was reported as single, with insertion upon the bicipital ridge. Its insertion was not given for Odobenus. In the Phoca the dorso-cranial part was quite thin, with origin gradually from the dorsal fascia and fibers converging to the axillary tissue. The ven- tral part was rather abruptly 10 times as thick as the remainder, with insertion upon the deltoid ridge with the second and third divisions of the pectoralis. M. rhomboideus anticus (figs. 2, 3, 7, 16, 17) in the Zalophus arose from the medial third of the occipital crest deep to the cephalo- 60 PROCEEDINGS OF THE NATIONAL MUSEUM VOL. 73 HYOCLOSS MYLOHYOID. GENIOHYOID- DIGAST- STYLOGLOS; STERNO-HYOID.+ -THYRO!D- STYLOWYOID OMOHY OID. THYREOHY OIDs STERNOMAST. TRACHEO-VERT. MUS CLE\DOMAST,. SCALEN) CEPHALOHUM. “STERNOCOST. FECTUS ABDONY]\\\ x Fic. 18.—VENTRAL MUSCULATURE OF ZALOPHUS; SUPERFICIAL LAYER UPON THE RIGHT, AND MUCH OF THE NEXT DEEPER LAYER TO THE LEFT OF THE MEDIAL LINE ART. 15 ANATOMY OF THE EARED AND EARLESS SEALS—HOWELL 61 GENIONNOIDA /ZT\ HYOGLOS. LX STYLOGLOSA /\# STYLOWYOID. 2 ‘a ; Ms: nS ATLANTOSCAP. ae i i ATLANTOSCAP. INFERS \ig | SCALENUS 1 vi YY SCALENUS 2. \Ws Fifi STERNOCOST. ANT. Veh INTERCOST. EXTERN: B E MYLONYOID. : AR STERNOMASTON. I fh Ai TRACHEO-VERT. MUSCLES H Np =) | : SV CEPHALOHUM. ie CANO Bl SSA ZA Te we \ BN } WANS jf ees 1 iy ny? \giainyt ys DEPRESSOR. SC al FECT. SUPERF. SERRAT. MAG. PECT. PROF. RECTUS ABDOM.~ fh PECT. ABDOM. LAT. OBLIQ. ABDOM. EXT. Fic. 19.—VENTRAL MUSCULATURE OF PHOCA HISPIDA ; SUPERFICIAL LAYER UPON THE RIGHT, AND MUCH OF THE NEXT DEEPER LAYER TO THE LEFT OF THE MEDIAL LINE 62 PROCEEDINGS OF THE NATIONAL MUSEUM Vou. 73 humeral and inserted along the dorsal three-fifths of the spine of the scapula. In Arctocephalus this was called rhomboideus capitis by Miller and was similar to my Zalophus. According to Murie conditions were quite different in Humetopias. He termed it rhom- boideus capitis, with origin from the middorsum of the cervical and anterior thoracic region and insertion upon the dorsal part of the spine and the caudal half of the vertebral border of the scapula. In Odobenus it did not extend caudad of the spine along the vertebral border. In my Phoca origin was from the nuchal ligament as far craniad as the occiput. It passed deep to the vertebral border of the scapula and inserted along the caudal border of the glenovertebral cartilage, its fibers fasciculating with those of the serratus magnus. For his Phoca Miller described two anterior rhomboids—a capitis and a cervicis—but from his text I can not tell whether these were entirely separate muscles (one corresponding to an occipitoscapu- laris) or two divisions of what might be considered as a single sheet. Attachment to the whole vertebral angle of the scapula is indicated, but otherwise conditions were similar to my Phoca. M. rhomboideus dorsi (figs. 7, 17) in the Zalophus arose from mid- dorsad by interdigitations of its fibers with those of its antimere. It was a rather decadent muscle, and its coarse, loosely connected fibers ran laterad to insert upon the vertebral border of the scapula, from slightly craniad of the spine to the glenovertebral angle. Murie considered that this represents the major and minor divisions. In the Phoca it was similar, save that insertion was only by fasciculi into a part of the insertion of the serratus magnus along the ventral border of the glenovertebral cartilage. M. atlantoscapularis superior (figs. 7, 16, 17, 19) in the Zalophus had origin from the transverse proceess of the atlas just entad of the in- ferior division of this muscle. It was slender and extended to aponeurotic insertion upon the vertebral border of the scapula for a short distance just craniad of the spine. Near the insertional end it was parallel to the fibers of the depressor scapulae and one can readily see that the two muscles might fuse, the sheet then being folded over the coracovertebral, angle. Authors have expressed doubt, from time to time, that either division of this muscle as here termed is homologous with the human levator (anguli) scapulae. The innervation in a large number of diverse mammals will have to be investigated before this point is settled, but at any rate it would be theoretically easy for this superior slip to migrate ventrad along the anterior border of the scapula, or for the inferior slip to migrate cn to the head of the humerus, as is actually the condition in Phoca, and then dorsad to the more usual levator anguli scapulae position. Conditions were the same in my PAoca save that insertion was dor- Binet a Scat" oo eg. ei cata cel eee "OA a art.15 ANATOMY OF THE EARED AND EARLESS SEALS—HOWELL 63 sad of the spine. Not mentioned by Murie as a distinct muscle in Eumetopias but possibly included as a part of his serratus magnus. M. atlantoscapularis inferior (figs. 7, 9, 16, 17, 19) also arises from the transverse process of the atlas, but superficial to the origin of the superior division. In the Zalophus insertion was by fascia upon the ventral two-fifths of the spine of the scapula adjacent to the anterior rhomboid and upon the neighboring part of the humerus. In the Phoca the insertion had migrated more distad and was by fascia upon the greater tuberosity and deltoid ridge of the humerus ad- joining (ventrad of) the humerotrapezius. Miller called this muscle atlantohumeral in Phoca and atlantoscapular in Arctocephalus, ac- cording to the insertion. They are undoubtedly homologous and I prefer to employ the same term for both, even though it is slightly ambiguous in the case of the seal. It is the levator anguli scapulae of Murie. In addition the latter mentions, rather vaguely, an acces- sory slip in Odobenus which he considers to be the homologue of a levator claviculae. M. serratus posticus. Deep to the rhomboid layer of the Zalophus and a couple of centimeters caudad to any part of the scapula was a tenuous, vestigeal bit of muscle which arose from fascia and ended likewise beneath the caudal border of the larger rhomboid. I did not distinguish it in my Phoca, possibly because of the presence of clotted blood in this region, but Miller did in his. It has not been reported from any of the other eared seals proper, but Murie found it fairly well developed in Odobenus. Deep, intrinsic back muscles.—WM. splenius (figs. 2, 5, 16, 17) arose from the middorsum, extending in the Zalophus from the occiput to about the second thoracic spine, and inserting along the entire occipital crest, from the vertex to the mastoid process, at the latter point being continuous with the trachelomastoid. In the Phoca origin extended craniad only as far as the spine of the axis, and insertion was limited to the mastoid process. This is Murie’s splenius capitis. M. erector spinae, sacrospinalis, extensor dorsi communis or long system of the back was but moderately developed in the Zalophus, and the pair of muscles together in the specimen dissected measured but 100 mm. in width at the widest part in the posterior thoracic region. It was indivisible in the lumbar region, the whole muscle arising from the vertebrae, sacrum, and inner surface of the ilium. Be- tween the ninth and sixth ribs it was partly separable into an iliocostalis dorsi and longissimus portion, but craniad to the sixth rib these once more to all intents constitute a single muscle. This fused part sent a slip to each of the first 10 ribs and to the anapo- physes of each of the last five cervical vertebrae, this part at least 64 PROCEEDINGS OF THE NATIONAL MUSEUM VoL. 73 representing a longissimus cervicis. Mediad, and craniad of the fourth thoracic vertebra, the biventer cervicis intervened, and adjoin- ing the spines was the spinalis dorsi, distinguishable with clarity eraniad of the eighth rib. The spinalis cervicis portion of this inserted upon the cervical spines to and including the axis. In the Phoca this was an astonishing mass of muscle, the two sides together measuring 210 mm. in width, and a definite impression was gained that although enormously stronger than in the eared seal dissected, its normal uses were for less complicated and more re- stricted in movements. Caudad it was clearly divisible into a medial longissimus dorsi and an iliocostalis lumborum of slightly lesser width. These two, however, were not separable with any degree of east throughout their deeper portions. The superficial fibers of the iliocostal part were seen to arise from the glistening aponeurosis covering the longissimus, but the chief origin was fleshy and from almost the entire cranially-directed surface of the “ medial” border of the ilium, forming an attachment of enormous strength. Within 50 mm. of this bone the iliocostal had attained a width of 40 and a depth of 45 mm. More craniad there was fibrous connection with the ribs over which it passed until a tendinous slip was given off to join the seventh, but apparently no other ribs in that vicinity. The muscle then broadened and developed tendon bundles, five of these passing cranio-dorsad to join the transverse processes of the first five thocacic vertebrae and four others cranio-ventrad to join the first four ribs. Thus, the anterior part of this muscle was simple, doing little but acting as an anchor for the posterior part. In the pos- terior thoracic region the longissimus had rather obscure spinalis dorsi elements, there being tendons extending cranio-mediad to join each spine, but no distinct muscle could be separated. In its more lateral part the longissimus had slight fibrous attachment to the ribs and mediad there were tendons gradually developing from the trans- verse processes extending craniad to be lost in the muscle mass. These increased in size in cranial sequence from the last thoracic and the series culminated in a very strong tendon from the tenth after which these ceased. Thus, these tendons seemed to be largely instru- mental in the development, upon these thoracic vertebrae only, of distinct metapophyses and constituted a semispinalis dorsi element. There were also smaller longissimus tendons of opposite inclination attached to the anapophysis of each vertebra. At about the seventh thoracic the longissimus split into two parts, both rapidly becoming slender. The more medial was the spinalis dorsi, becoming the spinalis cervicis. Its attachments were by fasciculi to the spines as far as the fourth cervical, although a few fibers may have extended still farther craniad. The more lateral division was the longissimus art. 15 ANATOMY OF THE EARED AND EARLESS SEALS—-HOWELL 65 (transversalis) cervicis. It sent tendinous slips to the transverse processes of the first four thoracics and was tucked in laterad of the trachelomastoid, inserting upon the dorsal part of the anapophyses of the last four cervicals. M. trachelomastoideus* (figs. 2,5), or longissimus capitis, although a part of the long system morphologically, had better be considered a separate muscle. In the Zalophus it arose from the postzygapoph- yses of the last four cervical and first two thoracic vertebrae, and in- sertion was strongly upon the mastoid process continuous with that of the splenius. In the Phoca origin was from the third to seventh (inclusive) cervicals. Miller reported that in his Phoca the muscle was partially divisible, a second slp inserting upon the transverse process of the axis; but nothing like this occurred in my specimen. In Arctocephalus origin was from the last five cervicals only; in Eumetopias from the fourth and fifth and in Odobenus, the fifth and sixth thoracic spines, according to Murie, who terms this the splenius colli. M. biventer cervicis (figs. 2,17) in the Zalophus had origin by partly tendinous slips from the postzygapophyses (or their vicinity) of the first four thoracic and last two cervical vertebrae. Insertion was upon the middorsal line for several centimeters caudad of the ver iex, and for about two centimeters along the medial occipital border. In Arctocephalus origin is said to be from the second, third, and fourth thoracic vertebrae: In Odobenus “from the seventh dorsal”; in Eumetopias “ from the fifth, second, and first dorsal spines.” M. complexus (figs. 2, 17) in the Zalophus arose by slips from the vicinity of the postzygapophyses of the middle five corvicals, the one to the sixth lying mediad of the cranial border of the biventer cervicis. Insertion was rather narrow in a tendinous sheet for a couple of centimeters along the occipital border laterad to the insertion of the biventer cervicis. In Arctocephalus and EHwmetopias origin is said to be from the third to seventh cervicals; in Odobenus “ from the fifth anterior dorsal.” In my Phoca the biventer cervicis and complexus were indivisible, forming the following: M. semispinalis capitis (fig. 3), arising in this Phoca from the vicinity of the postzygapophyses of the last six cervical and first two thoracic vertebrae. It was a heavy muscle at origin, the fibers converging to a thin, aponeurotic insertion upon the medial third of the occipital crest. Miller, however, apparently had no trouble in separating this muscle in his Phoca vitulina into a biventer and complexus, but the attachments were very similar to those of mine. 4It is believed that in the case of the wood rat (Howell, 1926) the two divisions of the biyenter cervicis as stated should rather have been considered to comprise both that muscle and the complexus, while the complexus, so termed, is in reality the tracehlomastoid. 86377—28 5 66 PROCEEDINGS OF THE NATIONAL MUSEUM VOL 73 M. rectus capitis posterior major (figs. 2,3) in the Zalophus arose not only from the whole length of the spine of the axis but also from the periosteum over the spine of the third cervical. Insertion was upon the supraoccipital laterad to the border of the biventer cervicis. In the Phoca it was larger and covered both the minor division and the inferior oblique as well. Origin was from practically the entire Jength of the axial spine and insertion was upon the occiput from the vertex almost to the lateral margin of the semispinalis capitis insertion. In Arctocephalus and Humetopias origin was confined to the axis. In Odobenus it also “ has attachments to the five posterior cervical zygapophyses.” M. rectus capitis posterior major acessorius (fig. 8), so termed by Miller and Murie for the animals which they dissected was not distin- guished in the Zalophus, but in the Phoca arose from the cranial border of the axial spine, with insertion upon the occipital between the superior oblique and minor rectus. IM. rectus capitis posterior miner (figs. 2, 3) lay deep to the last. It took origin from the dorsal arch of the atlas and inserted upon the supraoccipital deep to the last. In the Phoca it was similar save that the insertion was situated mediad to the accessory part of the major division. M. obliquus capitis superior (figs. 2, 3) arose from the transverse process of the atlas, with insertion strongly upon the lateral part of the occipital plane. M. obliquus inferior in the Zalophus arose from the spine of the axis deep to the rectus major, with insertion upon the transverse process of the atlas. In the Phoca it seemed to be easily divisible into two similar parts. Miller found it single in his Phoca, however. The perineal muscles were not investigated but it was noted that the levator ani, apparently occurring in two divisions, was poorly developed, or rather that the fibers were coarse and separated by much fatty tissue. MUSCLES OF THE ANTERIOR LIMB MUSCLES OF THE SHOULDER GIRDLB M. supraspinatus (figs. 7, 9, 16, 19, 20,21) was complex in the Zalo- phus, as is often the case. Origin was from the entire supraspinous fossa and anterior surface of the spine. Near the insertion it was separable into two slips, the more dorsal inserting upon the tip of the greater tuberosity with a few fibers passing also to the lesser. The second slip inserted upon the greater tuberosity also, just distad of the tip. In the Phoca origin was the same, and the muscle was partially divisible ectad, as indicated in Figure 21. Insertion was upon the greater tuberosity and upon the ligament extending from art.15 ANATOMY OF THE EARED AND EARLESS SEALS—-HOWELL 67 the greater to the lesser. In the animals dissected by Murie and Miller this muscle was essentially the same, with minor variations. M. infraspinatus (figs. 7, 9, 20,21) in the Zalophus probably had not more than one-fifth the strength of the supraspinatus. It arose over the spinal two-thirds of the infraspinous space and converged to a slip that passed over the lateral head of the femur to insert upon a slight ridge upon the lateral base of the greater tuberosity. In the Phoca this muscle was even much smaller—to a surprising extent. Origin was from the deep part of the spine and about one-half of the narrow space which constituted the infraspinous part of the scapula. It passed craniad over the acromial notch and inserted upon the capsule of the shoulder joint adjoining the supraspinatus and over the greater tuberosity. The ventral part of the cephalohumeral is homologous with a part of the deltoid. The only remaining part of the latter was the deltoid. M. deltoidedts (figs. 7,9, 17, 20,21). In both my specimens this was relatively very powerful and overlay almost all of the infra- spinous part of the scapula. In the Zalophus it arose by tendi- nous and fleshy fibers from the entire posterior border of the spine of the scapula and by tough aponeurosis from the dorsal border of the infraspinous fossa, although muscle fibers were not found so far dorsad near the spine. Insertion was aponeurotic, chiefly along the lateral border of the deltoid ridge, but a tendinous slip also passed to the ectepicondyle of the humerus, and the distal border passed inseparably into the substance of the supinator longus. In the Phoca origin was only from the caudal border of the spine, with insertion partly tendinous upon the caudal border of the deltoid crest. In Arctocephalus no mention was made of any junction with the supinator longus, but it was stated that a “tendinous slip goes to the fibro-cellular bar lying upon the anterior border of the radius.” In Odobenus there was no connection with the supinator longus. M. teres minor (fig. 7) in the Zalophus was exceedingly thin but broad at origin, which was from the “teres major fossa,” or the pos- terior third of the infraspinous space. It apparently fused with the subscapularis over the axillary border and the insertion was insep- arable. In the Phoca it was as broad distad as, but much thinner than, the infraspinatus. It arose not as in Zalophus but aponeuroti- cally from the distal half of the ridge between the two concavities of the infraspinous space, and was located between the latter muscle and the triceps. Near insertion the fibers joined the tendinous deep belly of the deltoid, and were inseparable. Miller stated that in Phoca vitulina insertion was upon the greater tuberosity and the capsule of the joint; but he did not find it in Arctocephalus. In H'umetopias 68 PROCEEDINGS OF THE NATIONAL MUSEUM EXTEN. METAC. POL-#H LAT. RAD. LIG-T/ (CUT) P ECT. PROF-+ Fic. 20.—SUPERFICIAL MUSCULATURE OF THE LATERAL ASPECT OF THE LEFL ANTERIOR LIMB OF ZALOPHUS vou, 73 a is ANATOMY OF THE EARED AND EARLESS SEALS—HOWELL 69 : : BX \ é \\Wr EXT. MET. POL. 4 a { TRICEPS LONG. Ww d SN S 'e S ~“ oe : y 4 ined ~ t a RS . Oe ~ dp LL? NS SSR ¥ Fic. 21.—SUPERFICIAL MUSCULATURE OF THE LATERAL ASPECT OF THE LEFT ANTE- 4 RIOR LIMB OF PHOCA HISPIDA 70 PROCEEDINGS OF THE NATIONAL MUSEUM VOL. 73 Murie reported it in close union with the teres major, and in Odo- benus he did not succeed in segregating it. IM. teres major (figs. 7, 9, 21, 22, 23) in the Zalophus arose from the more dorsal part of the axillary border of the subscapularis and by fasciculi from the dorsal quarter or third of the axillary border of the bone. The insertion was broadly tendinous at about the middle of the humeral shaft and under cover of the biceps. In the Phoca it arose over practically the entire axillary half of the infraspinous space, including the glenovertebral cartilage. It passed between the triceps and the subscapularis to insert upon the well-defined rugosity in the bicipital groove of the humerus, slightly proximad to the middle. M. subscapularis (figs. 9, 21, 22, 23) exhibited complexity of its fibers, as usual. In the Zalophus it arose from the entire medial sur- face of the scapula and even extended over the axillary border, while insertion was upon the lesser tuberosity deep to that of the episubscapularis. In the Phoca it was exceedingly massive, and overhung both the cranial and axillary borders. The latter part was partly separable, originating from the ventral border of the gleno- vertebral angle, and might almost be considered as constituting a separate muscle. Insertion of the whole was upon the lesser tuber- osity, in this case larger and more prominent than the greater, a fact due chiefly to the power of this muscle. M. episubscapularis (figs. 7, 22), in the nomenclature of which I follow Miller for convenience, is undoubtedly a division of the sub- scapularis and was folded over upon a part of the latter in the Zalophus only. It arose from the medial coracovertebral angle and border of the scapula and passed over the cranial edge of the sub- scapularis. It developed a tough tendon within the substance of the muscle which branched, one slip going to the lesser tuberosity in company with muscle fibers. The other was attached to a faintly defined groove upon the medial part of the ridge extending distad from this tuberosity and beneath the insertional end of the teres major. Upon this tendon and the adjoining dorsal border of the teres major the remaining muscle fibers inserted. This muscle has been found in all of the eared seals so far dissected, and the walrus also. Murie seems to have considered it to be a derivative of the supraspinatus in his report on L'umetopias, but listed it as a part of the subscapular for Odobenus. M. subscapulo-capsularis (fig. 9) is found in the Phoca only, and was represented by a few fibers arising from the extreme distal part | of the subscapular surface of the scapula, with insertion upon the capsule of the joint and the adjoining base of the lesser tuberosity. Miller stated that in Phoca vitulina it arose from the base of the axillary border. AKT. 15 ANATOMY OF THE EARED AND EARLESS SEALS—-HOWELL 71 MUSCLES OF THE UPPER ARM The flexors comprise two single muscles: M. biceps brachii (figs. 7, 10, 22, 23) was single and took origin, largely by tendon, from the rudimentary coracoid process upon the cranial margin of the glenoid cavity. In both genera it passed through the bicipital groove, between the greater and lesser tuber- osities and was inserted by tendon onto the bicipital rugosity of the radius. M. brachialis (figs. 9, 10, 20, 21, 22) was also single, with origin in the Zalophus from the cranial half of the proximo-lateral shaft of the humerus and from the caudo-lateral border of the deltoid ridge. A few of its fibers fused with the supinator longus. In the Phoca origin was from practically the whole of the lateral shaft of the humerus caudad of the deltoid ridge. In both animals the muscle passed deep between the supinator longus and pronator teres to insert by a strong tendon upon the rugosity just distad of the lesser sigmoid cavity of the ulna. Maurie and Miller wrote that in H'wme- topias and Arctocephalus this muscle arose by two heads, the two together being very similar to the one in my Zalophus. Murie reported a very weak coraco-brachialis in Odobenus, but I am not at all convinced by his text that he was not mistaken in this. In contrast to the paucity of flexors, the extensors of the brachium are powerful and specialized. M. epitrochlearis (fig. 22) (as of Reighard and Jennings, not the dorsoepitrochlear of some authors) was found in the Zalophus only, but neither Murie nor Miller seem to have encountered a similar muscle during their dissections of eared seals, they, perhaps, having considered it as integral with the triceps. It has, however, all the characteristics of a normal epitrochlear as found in so many mam- mals, save that the great specialization of the long triceps has caused the latter to curve around partly superficial to it. It arose from the fascia and connective tissue investing the medial part of the triceps longus and latissimus dorsi. The fibers ceased in the fascia of the ulnar border of the forearm, and thus the insertion had mi- grated somewhat distad, as have so many other muscles of the anterior extremity. The true triceps had best be considered as consisting of three parts as usual. M. triceps longas (figs. 7, 10, 16, 17, 20, 21, 22, 23) in the Zalophus was divisible into two portions. The more caudal part arose from the dorsal third of the slight ridge that bisects (roughly) the in- fraspinous space of the scapula, and by an aponeurosis covering the teres major. It passed to the olecranon, twisted in a peculiar man- ner, and was attached rather lightly at this point. It was then ap- 72 PROCEEDINGS. OF THE NATIONAL MUSEUM SUBSCAP.—E SS ———F % uw (=, wo uw S J A Y Y/ és gs Yj => 4 his 2 == Hy i Wi if 22 Z a H}}» Y, L Ss eS Meh Jy i SES a Why) WN LZ ee = heed: WM Page = = Yiyy == ti SS A —=———S—— << = | , a |) ko = SN TERES MAS- Mh EP\TROCH.« ui H FLEX. DIG.COM.4 FLEX. DIG.COM. 1 PALM. LONG: \ FLEX. CARP! ULN/ WORRY AVN \ sy \\ FLEX. DIG.COM.1+2 Fic. 22.—SuUPERFICIAL MUSCULATURE OF THE MEDIAL ASPECT OF THE LEFT ANTERIOR LIMB OF ZALOPHUS 3 5 ant. 15 ANATOMY OF THE EARED AND EARLESS SEALS—HOWELL 73 y "i SUPRASEIN. SA | XN 1] = TRICEPS LONG. TRICERS) CA ANCON. \NTERN-| PALM LONG. PRONATOR TERES ABD. DIG. 5 LONG. FLEX.CARP\ RAD. FLEX. CARP\ ULN. FLEX. DIG. COMMUN Fig. 23.—SUPERFICIAL MUSCULATURE OF THE MEDIAL ASPECT OF THE LEFI ANTE- RIOR LIMB OF PHOCA HISPIDA + ae 74 PROCEEDINGS OF THE NATIONAL MUSEUM VOU, 73 parently inseparable from the adjoining part of the lateral triceps. It disappeared in fascia over the ulnar border of the forearm. The more cranial division did not really constitute a distinct muscle but was easily separable. It arose from the distal half of the central ridge of the infraspinous space and was strongly inserted upon the medial olecranon and (by fascia) onto the medial epicondyle of the femur. In the Phoca I consider this muscle to have been single. It arose from the dorsal third of the ridge upon the infraspinous space and thence the line of origin curved about the adjoining part of the vertebral border of the scapula as far as the spine, in contrast to this part of the origin in the Zalophus which was directed away from the spine. It-became very thick and passed over the elbow to a rather abrupt insertion upon the superficial fascia of the forearm. M. triceps lateralis (figs. 7, 9, 10, 17, 20, 21, 23) in the Zalophus arose partly from the aponeurosis covering the cranial portion of the long head, but chiefly by narrow fasciculi from the latero-caudal neck of the humerus. The borders of this muscle were distinct but the deep fibers blended with those of the medial head. With the latter there was also some fibrous attachment to the olecranon but the muscle finally passed over this and together with the adjoining part of the long head was directed down the ulnar border of the forearm and disappeared in fascia above the wrist (in fig. 20 the edge of this part of the muscle is shown trimmed away). In the Phoca I list this muscle as occurring in two divisions. The more lateral and the one shown in figure 21 arose from an area that really constitutes the latero-caudo-proximal border of the deltoid ridge, quite to the head of the humerus. There is considerable question in my mind as to whether the second division does not really belong with the triceps longus—a point difficult to prove by the innervation. But the origin was separated from that of the true long triceps by a considerable interval and it seems more natural to place it with the lateral division. It arose from the distal (cranial) third of the central ridge of the infraspinous space, passed down the back of the brachium and joined the first division, both inserting upon the more dorsal process of the olecranon. M. triceps medialis (figs. 9, 10, 23) arose from the entire caudal part of the shaft of the humerus and was continuous with the anconeus externus. Insertion in both genera was upon the expanded radial aspect of the olecranon, and slightly from the caudal or dorsal aspect as well. Miller reported that in his Phoca and Arctocephalus conditions were on the whole very similar to what I found in my seal and sea lion, respectively, save that he divided the triceps into four parts—a dorsi-epitrochlear corresponding to my first division of the long head in Zalophus and long head in Phoca. In his Phoca origin was from ART. 15 ANATOMY OF THE EARED AND EARLESS SEALS—-HOWELL Ff) the dorsal border of the scapula only. His long head corresponds to the lower division of the long head in Zalophus and upper division of the lateral head in my Phoca. His external head corresponds to my lateral head in Zalophus and lower division of the lateral head in Phoca. And his internal head corresponds to my medial one. Murie listed four similar divisions for Hwumetopias and Odobenus. Humphry reported that in Phoca the long head reached the paddle, which has not been confirmed by anyone since. M. anconeus externus (figs. 9, 10, 20,21). It is only because Murie and Miller listed this as a muscle separate from the medial triceps that I follow the same course, for in both my specimens the two were fused. In both genera it may be said to originate from the lateral epicondyloid ridge, with insertion along the dorsal two-thirds of the external border of the olecranon. M. anconeus internus (figs. 9, 10, 22, 23) has origin from the same situation in both Zalophus and Phoca—the entepicondylar ridge— save that in the former it was from the most prominent part of the bone, and in the latter, caudad of the entepicondylar foramen, rather than from the more prominent part craniad thereto. Insertion of both was upon the tuberosity upon the medio-dorsal part of the olecranon. MUSCLES OF THD FOREARM The flexors consist of the following muscles: M. palmaris longus (figs. 10, 20, 22, 23) was of phenomenal power in the Zalophus and was fused with the flexor carpi wlnaris. The two together arose from the entire medial surface of the broad proximal half of the ulna, including the olecranon. Certain of the more su- perficial fibers decussated with those of the border of the triceps longus. It rapidly became broadly tendinous (25 mm. wide) as it passed over the wrist, was attached (the flexor ulnaris portien) to the pisiform, and then spread as in Figure 22, a powerful branch going to digit 1, and a weaker to digit 5, deep to its flexor sublimis tendon. It is perhaps the strongest flexor of the manus, working upon the two borders in an extraordinarily powerful cupping action. In the Phoca it was rather small and arose from the olecranon be- tween the long abductor of the fifth digit and the internal anconeus. Its tendon expanded to cover the three medial digits and adjoining part of the carpus. It is difficult to see why Murie and Miller both made the mistake they did in their treatment of this muscle. For Arctocephalus, Hume- topias, and Obobenus, the palmaris longus of Miller and primus of Murie was really a superficial division of the digitorum profundus, tentatively termed by me a flexor pollicis longus because of its posi- tion. Innervation is by the median nerve, and although this fact \ 76 PROCEEDINGS OF THE NATIONAL MUSEUM vou. 73 was inferentially mentioned by Miller, he did not take into account that such innervation prevents affinity with the palmaris. There was intimate attachment of its tendon to the palmar fascia, indubi- tably, and ultimately the fusion will probably be complete, but care- ful dissection proves that the main portion of the tendon extends to the pollex. Miller’s palmaris superficialis and Murie’s secundus division were merely the antibrachial extension of the pectoralis, which also had intimate connection with the palmar fascia. I found nothing whatever in the Phoca that could correspond with Miller’s second head of the palmaris for the same genus. His first head is very similar to the muscle as I found it. M. pronator teres (figs. 9, 10, 22, 23) lay mediad of the biceps and brachialis. In the Zalophus it arose in intimate relation with the flexor carpi radialis from the proximo-caudal margin of the medial epicondyle of the humerus. In the Phoca it was from practically the same spot. Insertion in both was upon the radius from near the bicipital rugosity to just distad of the process or angle near the center of its radial border. The ulnaris and radialis muscles of the Mammalia are usually rather uniform in their locations, but the digitorum muscles—espe- cially the flexors—are annoyingly in the habit of occurring in a great number of combinations. In many cases these are of such a nature that it is unwise to place too much reliance upon the homology as indicated by the nomenclature employed, for in this case the innerva- tion proves to be of little or no aid. Not only does origin vary, but the tendons of insertion are often unreliable as criteria for nomen- elature—a muscle homologous with a pollicis may not reach the thumb, or a digiti quinti the fifth digit. The pinnipeds have these flexor muscles of such a sort, and I deem it eminently wise not to refer to the muscular divisions by name but by number, for they are not now to be homologized with any certainty. M. flexor digitorum communis (figs. 9, 10, 22, 23) in the Zalophus consisted of four separable elements, while in the Phoca there were three which were separable only at origin. Caput 1 (figs: 10, 22) in the Zalophus was a part of Murie’s flexor sublimis digitorum in EH'wmetopias. It arose upon the ulna from caudo-mediad of the sigmoid cavity, passed distad directly super- ficial to the second division and split into three branches which went to the three lateral digits. The one to the third proved to be double, however, and there seemed to be a broadening of the tendon in the direction of digit 2, possibly indicating the relic of a former branch at this point. These three tendons were the most superficial of any going to their respective digits, the one to the fifth ultimately passing superficial even to the lateral palmaris longus branch. (See fig. 22.) ART. 15 ANATOMY OF THE EARED AND EARLESS SEALS—-HOWELL 77 Caput 2 (figs. 9, 10, 28) in the Zalophus was the second part of the flexor sublimis of Murie for Zumetopias. It arose from the medial epicondyle of the humerus and from. the medio-radial border of the - ulna from just distad of the coronoid to within a short distance of the end of the shaft. Its short tendon partly fused with the deep surface of that of the first division and when dissected free it was found to go to digit 4 only. In Humetopias the two heads of the flexor sublimis joined and the tendon sent branches to the four lateral digits. In this as in Zalophus the tendons of the first two divisions were entirely separable from those of the third division. In Arctocephalus this part of the flexor complex was single and there was fusion of the tendons with those of the third division. Caput 3 (figs. 10, 22) in the Zalophus and Murie’s flexor pro- fundus for EKumetopias, arose over the medial radius between the biceps, pronator teres, interosseous membrane, and to within a short distance of the distal end of the shaft. Its tendon sent branches to the first three digits. In Humetopias origin was from both the ulna and radius, and two tendons extended to the pollex, two to the index, and one to digit 3. Caput 4 (figs. 9, 22) in the Zalophus was a flexor pollicis longus, innervated by a branch of the median nerve and homologically a division of the flexor profundus, though whether morphologically the same division of this muscle as its analogue in human anatomy is not known. For Lumetopias, with origin the same, Murie termed it palmaris longus primus, he believing that its tendon ended in the palmar fascia. Of course, such may actually have been the case in that genus, but the condition in Zalophus proves that it is a part of the flexor communis. Origin was from the medial epicondyle ad- joining the flexor carpi radialis. Its tendon passed in the super- ficial flexor layer and near the base of metatarsus one it divided into two, both branches inserting upon the pollex—presumably upon its first phalanx. In Arctocephalus this complex was considerably different. It arose by three heads which Miller termed as follows: A flexor sublimis, from the medial epicondyle with tendon fusing with that of the next; a flexor profundus, from the medial surface of the proximal ulna; and a flexor pollicis longus, from the medial surface of the shaft of the radius and slightly from the ulna, which joined the common tendon of the flexor profundus, and this went to all five digits. For Odobenus Murie reported it still different, with the sublimis and profundus bellies partially fused, and a flexor pollicis radiad. All of these were conjoined to a broad tendinous sheet with five branches to the digits. The three divisions named are analogous 78 PROCEEDINGS OF THE NATIONAL MUSEUM vou. 73 to those of man, but there seems to be some doubt as to whether they are homologous. In the Phoca the flexor digitorum communis arose by three heads. Caput a had origin from the medial epicondyle below the flexor — carpi radialis; caput b had origin from the whole of the medial sur- face of the olecranon and from two-thirds of the shaft of the ulna; while caput c originated from the radius, upon the ulnar side of the pronator teres insertion. The three heads joined to form a very broad tendinous band which split into five branches, one going to each digit. This was substantially the same as conditions in the Phoca vitulina of Miller, who called caput a the flexor sublimis, caput b the flexor profundus, and caput ¢ a flexor pollicis longus. His nomenclature may possibly be correct and the divisions seem to be homologous with those of Arctocephalus, but they certainly are not with those bearing the same terms for H’'wmetopias, as designated by Murie. M. flexor carpi radialis (figs. 9, 22, 23) lay upon the ulnar side of the pronator teres. It arose in both animals from the medial epicon- dyle. In the Zalophus its tendon inserted chiefly upon metacarpus one deep to that of the supinator longus, but a tenuous branch went also to metacarpus 2. In the Phoca insertion was also upon meta- carpus 1 and 2, and probably 3 as well. Miller reported three branches to the three metacarpals in Phoca and three in Arciocepha- Zus to the first two metacarpals and the ligament between the trape- zium and trapezoid. In H'wmetopias insertion was upon the first metacarpal only. M. pronator quadratus (fig.9). Murie found this muscle in Odobenus but not in H'umetopias, while Miller did not find it at all in Phoca or Arctocephalus. In the Zalophus it was represented by a few fibers upon the flexor side of the interosseous membrane, while in the Phoca it was weakly though indubitably present. M. flexer carpi ulnavis (figs. 10, 21, 22, 23) in the Zalophils was fused with the palmaris longus. It was the ulnaris portion of the latter muscle, however, that was attached to the pisiform. In the Phoca it was located between and deep to the palmaris and long abductor of digit 5, and was a robust muscle. Origin was from the medial olecranol border, with a rather complex insertion, for although there was attachment to the pisiform as usual, the main thread of the tendon continued to metacarpus 5. In addition a tendinous fascia was given off mediad, this forming a second and deeper palmar fascia, and this curved deep and laterad to form part of the thick sheath of the flexor digitorum communis. In Arctocephalus and Humetopias it was very similar to that of Zalophus. In his Phoca Miller reported the same condition found by me in the same genus, save that he made ART. 15 ANATOMY OF THE EARED AND EARLESS SEALS-—-HOWELL 79 no mention of complexity in connection with the sheath of the flexor communis. M. abductor digiti quinti lengus (figs. 10, 21, 23) was really a flexor and was served by the ulnar nerve. Its homology is not certain but it seems probable that it may be a division of the flexor carpi ulnaris. It was present in Phoca only and arose by aponeurosis from the ulnar termination of the olecranon and its tendon inserted upon the first phalanx of digit 5. In the Zalophus the antibrachial extension of the pectoralis must be included with the flexors of the lower arm. In the juvenal specimen this detail proved puzzling, for after the part of the pec- toral aponeurosis inserting upon the deltoid ridge had been removed and the subdermal tissue of the manus dissected free, there remained an apparently distinct structure wth muscle fibers attached arising from the slight ridge extending from the deltoid crest to the medial rim of the trochlea. This emerged from between the biceps and the brachialis, and then fused with an extensive and touch sheet of fatty fibrous tissue covering the radial border of the forearm and metacarpus 1. With this were associated dark fibers apparently muscular in character. In Figure 19 this detail is shown as en- countered in this specimen after removal of the part of the pectoral inserting upon the deltoid crest, except that the fibrous tissue ex- tending farther upon the cranio-lateral part of the forearm is repre- sented as having been cut away. In a fresh, adult female, however, it was at once seen that this was a part of the deep pectoral, which, in a tendinous sheet, inserted upon the humerus for the entire length of its shaft, extending quite to the pollex, and with it was associated a thick, fatty, fibrous layer, entirely nonmuscular in this adult, that was most extensive over the anterior or radial border of the arm and acted not only as a buffer or shock absorber, but materially broadened the forearm. The fleshy part of the pectoralis that cov- ered the proximal part of the medial forearm was erroneously desig- nated as a superficial layer of the palmaris longus by Murie for Eumetopias, and by Miller for Arctocephalus. The extensors of the antibrachium were as follows: M. extenser digitorum communis (figs. 9, 10, 20, 21) was the most superficial of the forearm extensors, and arose in a thin sheet from the lateral epicondyle of the humerus. Its tendon passed beneath the dorsal carpal ligament close to that of the extensor pollicis longus and in both animals a branch was sent to each of the four lateral digits. This is Miller’s primus division of this muscle in his Phoca. In Arctocephalus he stated that it split into two slips, one being the same as I found the extensor communis in Zalophus and the second constituting an extensor minimi digiti. In Yumetopias and 80 PROCEEDINGS OF THE NATIONAL MUSEUM vou, 73 Odobenus the slip representing this muscle split into three tendons passing to the three middle digits. M. extensor digitorum lateralis (figs. 9, 10, 20,21) is, im most mammals, a more fitting name for this muscle than its homologue, extensor digiti quinti proprius. In the Zalophus it was located between and partly deep to those of the extensores communis and carpi ulnaris. It arose from the distal part of the lateral epicondyle and the lateral radial ligament in such a position that tension could be applied only during flexion of the forearm. It was a weak and slender muscle whose fine tendon split first into two, the lateral branch going to the lateral border of the fifth metacarpal. The medial branch again divided into two, sending one tendon to the medial side of metacarpus 5 and another to the adjoining border of the fourth. In the Phoca this muscle was quite complex and occurred in three parts which were not completely divisible proximad. Origin was considered to be from the lateral epicondyle only, although there was quite firm attachment to the capsule of the joint over the head of the radius. Three tendons developed from as many muscular slips which passed ectad of the long extensor of the pollex. The more ulnar of these split into two branches which extended to metacarpals 4 and 5. The more radial also split into two, extending to metacarpals 2 and 3, while the deeper (not shown in fig. 21) inserted upon the ulnare. For this muscle in H'umetopias Murie recognized two divisions, equal together to my one. One was a minimi digiti, with tendons to digits 4 and 5, and the other a medii digiti, with tendons to digits 3 and 5. In Obodenus these tendons went only to digits 4 and 5. Miller con- sidered that for Arctocephalus there was a single head of origin for the communis digitorum and minimi digiti. The latter divided into three main branches going to metacarpals 4 and 5. For Phoca, Miller termed this division the extensor communis secundus, men- tioning no divisions of the muscle itself but that it split into four tendons, passing to the four lateral metacarpals. The structure referred to above as the lateral radial ligament de- serves mention. It occurred in the Zalophus only, as a broad, tough band extending from the lateral epicondyle over the lateral aspect of the radius to somewhat distad of its middle. It covered the supinator brevis and was continuous along its border with the inter- osseous membrane where this adjoined. Its function was as its name implies—to add great strength to the joint. It was not an out- growth of the normal capsular ligament of the elbow, for it not only had some vestige of muscle fibers associated with it but seemed to be served by the dorsal interosseous nerve, which indicated that it may have been a relic of some primitive division of the common extensor group of the digits. No mention was made of such a structure in other eared seals. art.15 ANATOMY OF THE EARED AND EARLESS SEALS—HOWELL 8] M. extensor metacarpi pollicis (figs. 10, 20, 21) arose in the Zalophus from the whole of the fossa of the lateral ulna that was situated upon the radial side of its prominent lateral ridge, and extended in origin two thirds the length of the shaft. It ran obliquely across the radius and its tendon passed through the large groove upon the cranio-lateral termination of the radius. It then passed to a double insertion, the first to the medial base of metacarpus 1, and the other to the first phalanx of the same digit. In the Phoca it arose from the whole of the lateral face of the olecranol plate save the extreme ulnar tip. Its tendon inserted as normal upon the base of metacar- pus 1. This was termed extensor ossi metacarpi pollicis by Murie and Miller. From the former’s description, insertion in H’wmetopias and Odobenus seems to have resembled that in my Phoca, but the tendon was much weaker in the latter animal. M. extensor pollicis longus (figs. 10, 20, 21) in the Zalophus arose from that part of the fossa on the lateral ulna that was situated upon the ulnar side of its prominent ridge. Its very broad tendon ex- tended obliquely across the carpus and was inserted upon the dor- sum of the first phalanx of the pollex. In the Phoca it arose from the ulnar tip of the lateral olecranon and for some little distance distad, while insertion was as in Zalophus. For Humetopias Murie termed this muscle extensor pollicis et indicis but noted no differ- ences, and it is Miller’s extensor primi internodii pollicis. In the latter’s Phoca vitulina origin was from the posterior third of the ulna. M. extensor carpi ulnaris (figs. 9, 10, 20,21) in the Zalophus appeared very similar in its proximal portion to the extensor communis. It arose by aponeurosis from the radial two-thirds of the olecranol border of the ulna beneath the lateral anconeus. Its rather slender tendon passed to the lateral manus and was inserted upon the lateral border of metatarsus 5. In the Phoca origin was from the lateral epicondyle of the humerus, while insertion was normal as in Zalophus. M. extensor carpi radialis (figs. 9, 10, 20,21) was single in its muscular portion and arose from just dorsad of the lateral epicondyle proper. It extended next laterad to the supinator longus and its tendon was seen to be double. In the Zalophus these are doubly inserted into the lateral border of metacarpus 1 and the medial border of metacar- pus 2. In the Phoca one tendon went to metacarpus 2 and the other sent two branches to metacarpals 2 and 3 respectively. As this inser- tion was different from anything reported for the genus, I was careful to verify it. Miller wrote that in Phoca vitulina insertion was upon the first and second metacarpals, while in P. barbata it was upon the second only. 86377—28——_6 82 PROCEEDINGS OF THE NATIONAL MUSEUM VOL, 73 M. supinator longus (figs. 9, 10, 20, 21, 23) in the Zalophus was spe- cialized. Origin was from the more caudal portion of the lateral shaft of the humerus extending quite to the head and partially be- tween the brachialis and triceps. This excepts the deltoid crest, from the distal termination of which arose but a few fasciculi of this muscle. It then passed between the brachialis and extensor carpi radialis and was joined by the more distal part of the deltoid. Its iendon was inserted upon the most prominent part of the radial _ aspect of the distal termination of the radius, just deep to the tendon of the extensor metacarpi pollicis. In the Phoca origin was much more restricted and more caudad, and there was no connection with the deltoid. Murie stated that in H’wmetopias this muscle had two heads, the second from the deltoid crest, but I judge that there was little difference from conditions in the Zalophus. From the description of Avctocephalus the origin in this animal was more extensive. M. supinator brevis (figs. 9, 10, 20, 21) had its usual complex origin, chiefly from the capsule of the elbow joint, but also upon the lateral condyle of the humerus as indicated. (Fig. 9.) Insertion in both animals was upon the cranio-lateral part of the radius from its head to the pronator teres angle. The short muscles of the manus of these two families of pinnipeds were adequately investigated by Murie and Miller, and as they are not particularly pertinent to the present report, they are here omitted. MUSCLES OF TH POSTERIOR LIMB MUSCLES OF THE HIP The hypaxial muscles or psoal complex presented no especial difh- culties in Zalophus, but in the Phoca they were extremely closely associated, peculiar, and withal so tender that nice dissection was entirely out of the question. Miller found these muscles different in each phocid which he dissected, his descriptions are too involved, and at least one error is indicated, for his magnus did not go to the limb. His contribution is therefore of slight aid in this respect, nor am I able to homologize things entirely to my own satisfaction. M. pseas minor (figs. 11, 24) in the Zalophus was rather small and arose apparently from all the lumbar vertebrae, but craniad the asso- ciation with the quadratus lumborum was so close that one can not distinguish between the two muscles with certainty. A very slender separate head also arose from the centrum of the last lumbar, joining the remainder by a small tendon. The whole became stoutly tendi- nous and inserted upon the pectineo-psoal process of the innominate. In the Phoca the part that indubitably represented this muscle was yey ll ART. 15 ANATOMY OF THE EARED AND EARLESS SEALS—-HOWELL 83 the most medial of the divisions and also small. It arose apparently from the last three lumbars and inserted slenderly upon the pectineal process. Conditions have been essentially similar in the other pin- nipeds dissected, including a small Phoca vitulina by Miller; but in the case of a large specimen of this species the same author stated that this muscle was enormous and much larger than the psoas major, which seems to indicate that an error was made in either one or the other. M. psoas magnus (figs. 11, 12, 13, 24) in the Zalophus was less robust at the original but larger at the insertional end than the minor -LUMB. psoas 3-——-——-——- FN IM ——-— PSOAS MINOR—-——- eee "“——~ SARTORIUS——__ “——SRECTUS FEMORIS ——PECTINEUS™ => WN \\ \ D5 a . ji NX NN yyy a Y ADDUCTOR ANTICUS LFA “——ADDUCTORES iff fi YW ~~ Zz ADDUCTOR PosTicus~ CH iff WZ ——— —% GEMEL. INFER- YW __- SEMITENDINOSUS-—— { 2—~ LZ thf x, < TL A a “S[VSEMIMEMBRANOS US ~ « a SGRACILIS———_ Fig. 24.—VENTRAL ASPECT OF THE MUSCLES EXTENDING FROM THE INNOMINATE BONE (STIPPLED) TO THE POSTERIOR LIMB OF ZaA- LOPHUS (Z) AND PHOCA HISPIDA (P) division. Origin was from the last two lumbars and the extreme anterior sacrum. It passed beneath (dorsad of) the minor and be- tween the rectus femoris and pectineus, and inserted tendinously upon the lesser trochanter. It is difficult to understand the condi- tions in Miller’s Avctocephalus, or in his other pinnipeds for that matter. In my Phoca there was no difficulty about the insertion which was with the iliacus upon the medial tuberosity of the tibia. In this tender specimen, however, origin was entirely inseparable from a great mass of muscle extending from the thorax to insert chiefly upon the ventral part of the medio-cranial face of the ilium. About any such insertion in his specimens Miller mentions nothing. 84 PROCEEDINGS OF THE NATIONAL MUSEUM VOL, T3 A small slip also separated from this mass to insert upon the psoas process of the ventral ilium. The homology of this large mass of muscle is very uncertain, but it may provisionally be termed psoas tertius. (Figs. 11, 24.) Murie considered that in Humetopias either the two psoas muscles had fused or the true psoas minor was absent, the two muscles of this group therefore consisting of a psoas and an iliacus. For Odo- benus he said that the two psoas were closely united, and inserted by a common tendon upon the ilio-pectineal eminence—an interpreta- tion that is open to question. M. iliacus (figs. 11, 12, 13, 24) was not entirely distinct. In the Zalophus it was very small, arising from the ventral border of the ilium in intimate relation with the final shp of the quadratus lum- borum. It inserted with the psoas magnus upon the lesser trochanter. In the Phoca its fibers were inseparable from the part of the psoas tertius which joined the ventral ilium, while insertion was with the psoas magnus upon the medial tuberosity of the tibia. This muscle seems invariably to be present in pinnipeds. M. tensor fasciae femoris (figs. 12, 17) arose from the lumbodorsal fascia, in the Zalophus reaching the middorsal line over the last two lumbar vertebrae. In the Phoca the exact bounds of the muscle could not be so well defined because of the condition of the specimen. Insertion in both was upon the lateral part of the patella. M. gluteus maximus (figs. 12, 16. 17) in the Zalophus had origin con- tinuous craniad with the biceps femoris over the spines of the three sacral vertebrae. It passed over the proximal part of the greater trochanter, to be inserted upon the disto-lateral part of the same, but not appreciably onto the shaft of the femur. In the Phoca this muscle was very heavy, indeed, and arose by aponeurotic fascia over the middorsal space included between all four sacral and first two caudal vertebrae. It quickly converged to a tendinous insertion along the entire lateral border of the shaft of the femur from the lateral condyle to the distal part of the greater trochanter, although over the middle of this space there seemed to be no actual attachment to the bone. In both Kumetopias and Arctocephalus two heads were reported, but the separation seems to have been incomplete. At any vate, insertion was upon the shaft of the femur as well as the greater trochanter, thus resembling the condition in my Phoca rather than the Zalophus. Miller wrote that in Phoca vitulina there were also two heads and that one of them arose from the ilium. M. gluteus medius (figs. 11, 12, 16, 17) in the Zalophus arose from the anterior border of the ilium, and this part was not covered by the gluteus maximus. Near the insertion it was separable, except at its ventral border, into two thin sheets, these forming a V-shaped tendinous attachment to the greater trochanter. In the Phoca this anT.15 ANATOMY OF THE EARED AND EARLESS SEALS—HOWELL 8&5 muscle and the pyriformis of the present specimen were inseparable, the two parts being distinguished only by a faint line presumably between them ectomediad. Origin of the two together was from all but the ventral part of the “ lateral ” surface of the ilium practically to the acetabular border and thence upon the sacrum over the second sacral vertebrae, and upon the ridge extending to the transverse plate of the third sacral. Insertion was almost entirely fleshy upon the flattened part of the greater trochanter. In H’wmetopias Murie reported origin as also from the sacrum, presumably the spines. M. gluteus minimus (figs. 11, 12, 16) in the Zalophus seemed to be partially divisible, along its dorsal border only, into three slips. It arose from the lateral surface of the cranial half of the ilium and from two-thirds of the vertebral border, with a few fibers from the ventral surface of the ilio-sacral ligament. Insertion was strongly upon the greater trochanter. In the Phoca this muscle was entirely covered by the medius. It arose from the ventral fossa and ridge adjoining the “ lateral” surface of the ilium. Insertion was some- what tendinous upon the dorso-medial part of the greater trochanter. M. gluteus quartus (figs. 11, 12). No slip representing this muscle was mentioned by either Murie or Miller. In my Zalophus, how- ever, it could not be ignored. It was very slender and arose from the ilium craniad of the femoral process and with attachment also to the tendon of the rectus. It passed just superficial to the latter muscle to insert upon the greater trochanter in the angle between the insertions of the glutei medius and minimus. M. pyriformis (figs. 12, 16) in the Zalophus was easily separable from the gluteus minimus, caudad of which it lay. It arose from the last two sacral vertebrae beneath the caudal portion of the ilio-sacral ligament, and converged to a tendinous insertion upon the greater trochanter. In the Phoca, as mentioned, this muscle was inseparable from the gluteus medius. In Phoca vitulina Miller separated it with care. After disposing of the gluteal-pyriformis mass, the complex con- sisting of the gemelli, obturator internus and adductor posticus ap- peared in the seal as consisting of a single, strong, well-rounded muscle and much care was necessary in its dissection, partly, of course, because of the extreme tenderness of the fibers. M. gemellus superior (figs. 11, 12) lay next caudad to the pyriformis. In the Zalophus it was a much weaker muscle, arising from the superior border of the ischium just caudad of the acetabulum. In- sertion was upon the caudal aspect of the greater trochanter dorsad of the obturator fossa. In the PAoca origin was relatively farther caudad, and its tendon joined the other which, in turn, joined that of the obturator internus inserting into the trochanteric fossa of the 86 PROCEEDINGS OF THE NATIONAL MUSEUM vou. 73 femur. Murie stated that in H'wmetopias origin was from the sacral vertebrae, which seems unlikely. M. ebturator internus (figs. 12, 16) arose from the inner border of the obturator foramen and its membrane, passed over the dorsal border of the ischium between and in close association with the two gemelli, and inserted by the usual tendon into the trochanteric fossa of the femur. M. gemellus inferior (figs. 11, 12, 16, 24) in both animals arose from a similar area upon the dorsal border of the ischium—roughly the second quarter of the distance from the acetabulum to the caudal border. In the Phoca, however, there was in addition a strong tendon arising from the bicipital process and disappearing in the substance of the muscle. In the Zalophus insertion was upon the side of the greater trochanter, between the obturator internus and quad- ratus femoris. ‘At first glance it seemed to be a part of the latter muscle. In the Phoca its tendon joined the insertion of the obturator internus. For Humetopias insertion was given as between the greater and lesser trochanters. Milier referred to this muscle in parts of his text but did not describe it. M. quadratus femoris (figs. 11, 12, 16) in the Zalophus arose from the dorsal half of the caudal border of the ischium, bounded dorsad by the gemellus inferior, ventrad chiefly by adductor 1, and mediad by the obturator externus. Insertion was along the entire latero-caudal border of the femur. In Arctocephalus insertion was said to be upon the “lower half of the posterior border of the great trochanter.” For Humetopias Murie gave insertion as the outer side of the lesser trochanter. The muscle is absent in the Phocidae. M. obturator exiernus (figs. 11, 12) in the Zalophus arose not just from the border of its foramen, but from a considerable area of bone caudad and ventrad, and from all but the dorsal part of the obturator membrane. Insertion was tendinous near the disto-caudal part of the greater trochanter. In the Phoca origin was much more restricted. It arose from the obturator membrane only over its dorsal and cranial parts, from the bone anterior to the obturator foramen, and from the dorso-cranial part of the pubis, the latter part of the muscle having almost the appearance of a separate slip. Insertion was par- tially tendinous into the obturator fossa of the femur. In Hwmeto- pias insertion was said by Murie to be onto the lesser trochanter, while in Mérouwnga insertion was upon the greater trochanter; but in this Miller was probably mistaken, as he was, at least for P. hispida, when he stated that the adductors are absent in the Phocinae, for the adductors constitute the ectal part of his obturator externus, and he evidently missed the deeper and very small latter muscle. ART. 15 ANATOMY OF THE EARED AND EARLESS SEALS—-HOWELL 87 MUSCLES OF THE THIGH There was evident a decided tendency toward fusion of the gracilis, semimembranosus, and semitendinosus of Phoca, and the three to- gether formed a great muscle mass with a single prime function. No such tendency was found in Zalophus. M. semimembranosus (figs. 11, 13, 24) of Zalophus occurred in two parts. The posterior (really the inferior) arose from the caudal border of the ischium adjoining the symphysis and the origin of the gracilis. Insertion was narrow and fascial beneath the caudal border of the gracilis. The anterior division, of about the same size, arose from the caudal border of the ischium dorsad of the origin of the posterior part. Fascial insertion upon the cranial border of the tibia was entad of the cranio-dorsal border of the gracilis and was entirely hidden by the latter. In the Phoca this muscle was single and arose along the caudal border of the ischium, the more caudal part being incompletely separable from the gracilis, under cover of which muscle it extended to a fleshy insertion not directly upon the tibia but upon the’heavy aponeurosis investing the ventral belly of the semitendinosus. For both Humetopias and -Odobenus Murie called this muscle semitendinosus. In Avrctocephalus it was single, while Miller reported it double for Phoca. M. semitendinosus (figs. 11, 13, 16,17, 24) in the Zalophus arose from over the spines of the third to sixth or seventh caudal vertebrae, with fascial insertion upon the distal quarter of the shaft of the tibia (not including the malleolus). In the PAoca it occurred in two parts which were entirely distinct at origin but apparently fused more distad. The posterior division was the more superficial and much the larger. It arose robustly from the transverse processes of the first three caudal vertebrae. The anterior division arose by tendi- nous fascia from the caudal border of the ischium along an area adjacent to the semimembranosus but not reaching the dorsal spine. The insertional end of both parts together developed a stout aponeu- rosis attached along the cranial border of the tibia and the ham- string tendon. This is the muscle which Murie termed semimem- branosus. It was single also in Odobenus, Fumetopias, and Arctoce- phalus. In both animals the aponeurosis of insertion of these hamstring muscles ended in the tough fascia and connective tissue over the heel. In Phoca especially it may be said to end in a sort of ligament which extended from the head of the fibula to the most prominent part of the external malleolus, and there was also substantial anchorage in the fibrous tissue beneath the calcaneal tendon. M. biceps femoris (figs. 11, 12, 18, 16, 17) in the Zalophus occurred in three main parts. The more superficial portion of the long head was 88 PROCEEDINGS OF THE NATIONAL MUSEUM vou. 73 indeed a very remarkable muscle, origin extending from the spine of the second to that of the seventh caudal vertebrae. The fibers extended caudo-laterad to a tough sheet of aponeurosis covering the side of the lower leg and stretching from the outer condyle of the femur, the adjoining part of the tibia, and to the distal extremity of the shank. In Humetopias and Odobenus origin was a trifle farther caudad. A posterior head was very narrow and arose under the caudal border of the main muscle, but its insertion was merely a distal continuation of that of the main mass. This may be the muscle termed by Murie a levator ani. The socalled short head of the biceps was but a couple of centimeters in width and arose partly from the transverse process of the second sacral vertebra and partly from the sacro-rliac ligament. It inserted upon the aponeurotic sheet investing the dorsal border of the distal fibula. In the Phoca the biceps occurred in two divisions. The super- ficial arose by a strong, tendinous origin from the dorsal spine of the ischium. It spread fanwise to a fascial insertion over the proxi- mal seven-eighths of the lateral part of the tibia. It is thus seen that in the Zalophus this muscle was in the form of a parallelogram, closely binding the shank to the vertebral column, while in the Phoca it was in the form of a triangle with pivotal apex upon the ischium, theoretically permitting much more freedom of movement. The deep division of the biceps of the Phoca was narrow and strap- shaped, with fleshy origin from the transverse plate of the third sacral vertebra, and aponeurotic insertion over the distal fibula. This corresponds with the short head of Zalophus. Murie termed the short head in Z'wmetopias the sacro-peroneus, and Lucae designated the heads as ischio-tibialis and sacro-fibularis. In Arctocephalus there were three divisions, much as-in Zalophus. M, sartorius (figs. 11, 12, 13, 16, 17, 24) in the Zalophus occurred in two distinct sips. The proper one was very slender, arising from the ventral angle of the iliac border with insertion upon the medial patella partly deep to the tensor fasciae. The second arose from the - cranial part of the ventral border of the ilium and had fascial inser- tion upon the medial tuberosity of the tibia mediad to and continuous with that of the first division. In the Phoca it was single, arising from the ventral angle of the ilium with insertion upon the patella continuous with and mediad to the insertion of the tensor fasciae. Miller reported it as single in Phoca and double in Arctocephalus, and Murie found it single in Humetopias and Odobenus. M. rectus femoris (figs. 11, 12, 16, 17, 24, 25, 26) was a robust muscle arising from the femoral process of the ilium cranio-ventrad of the acetabulum, which is prominent in Zalophus but indicated only by a roughened area in the Phoca. Insertion was upon the patella deep to the sartorius and tensor fasciae femoris. ne art. 15 ANATOMY OF THE EARED AND EARLESS SEALS—-HOWELL 89 The vasti were separable into two divisions only. ; M. vastus lateralis (figs. 12, 25, 26) arose from the dorso-cranial part of the greater trochanter and for a very short distance distad along the lateral border. Insertion was upon the lateral patella and slightly upon the capsule of the joint. Miller considered that the vastus externus of all the pinnipeds which he dissected arose from the entire lateral border of the femur, and Murie from the “ anterior surface of the femur,” but I deem the true origin to be more restricted. M. vastus profundus (figs. 12, 25, 26) evidently comprised a fused vasti femoris and medius. In the Zalophus origin was from prac- tically the entire cranial surface of the femur, and in the Phoca, from only the proximal half. Insertion in both was upon the medial patella and adjoining part of the capsule of the knee partly deep to the insertion of the rectus femoris. Miller designated the deeper vastus as the crureus, and Murie, the vastus internus. The adductors consisted of the following muscles: M. pectineus (figs. 11, 12, 24) in the Zalophus arose from the dorso- caudal slope of the pectineo-psoas process of the innominate and the tendon of the psoas minor adjoining. Insertion was upon the distal border of the lesser trochanter adjoining that of adductor 6. In the Phoca origin was longer, from the border of the publis for a short distance directly caudad of the “ pectineo-psoas process.” Insertion was also fleshy upon the proximo-medial portion of the caudal aspect of the femoral shaft, or close to the corresponding position of. the lesser trochanter of Zalophus, absent in the Phoca. In Arctocepha- lus origin was much as I found it in Phoca. Miller considered that this muscle was Murie’s adductor brevis primus, and that there was really a second division, termed pectineus for Humetopias and Odo- benus, and called by Miller for Aretocephalus pectineo-superficialis vel femoralis. Both Murie’s and Miller’s treatment of the pectineo- adductor complex is irrational. There is no good reason, morpho- logically, for considering that the pectineus has taken upon itself com- plexity. The pectineus of Murie and the pectineo-superficialis vel femoralis of Miller were, in fact, clearly a subdivision of the true adductor mass. M. gracilis (figs. 11, 13, 24) in both animals was incompletely double at origin, the two heads being separated, craniad only, by the rectus abdominis. The more ectal arose midventrad, its fibers decussating with those of its antimere, and the more entad from the symphysis and the ischium laterad thereto. Insertion was fascial, in the Zalophus along the middle third, and in the Phoca the distal half, of the tibia. In the latter animal there seemed to be a slender tendon or ligament developing from this fascia and passing over the medial malleolus and joining the plantar fascia. In Odobenus insertion was upon almost the entire shaft of the tibia. 90 PROCEEDINGS OF THE NATIONAL MUSEUM vou, 73 RECT. FEM.—- yy WZ VASTUS PROF. Y; TVBIALIS ANT. \ --PERON. 5 +BREN. -PERON, LONG. Fic, 25.—CRANIO-LATERAL ASPECT OF THE SUPERFICIAL MUSCLES OF THE LEFT POSTERIOR LIMB OF ZALOPHUS ART. 15 ANATOMY OF THE EARED AND EARLESS SEALS—HOWEITJ: RECT. FEM AR VASTUS PROF. He VASTUS LAT. EXTERN. LAT. Lis. qi HN! P GASTROCNEM. MED.—# TIBIALIS ANT: | PERON. LONG. MTT = PERON. DG. 5 EX TEN. DIG. LONG. Lit HA FLEX. HAL. LONG. UNG hi H i RR HY) ec MAMSTRING MUSCLES EXTEN. HAL | > of j ae | =PERON. BREV. ai if SERS, Ae. ff Pom PERON. Dic. 5 EXTEN. HAL—Al “ mit \ ay. \ / s i! \ WN fae \ fi oh ANS Wi YA yl hy j/ WV ) wh ii A Pe y \ We oy { "| ! \ \\\ ww fe | ‘ \ A Wad. Fie. 26.—CRANI0-LATERAL ASPECT OF THE SUPERFICIAL MUS- CLES OF THE LEFT POSTERIOR LIMB OF PHOCA HISPIDA 91 92 PROCEEDINGS OF THE NATIONAL MUSEUM VOL, 73 In no respect, perhaps, does the tendency for subdivision of the muscles of the posterior limb of Zalophus and fusion of those of Phoca appear more pronounced than in the adductor muscles and ~ certain others near them. The adductors, however, when these show complexity, is one of the groups which it is as yet not only unwise but actually misleading to attempt to homologize too precisely with the human divisions, and we must know far more regarding the lower Mammalia than we now do before such a course can be taken with confidence. The reason for this is that save for a part of the adductor magnus, little can be proved by the innervation, and it is both simpler and more satisfactory to refer to a number of divisions by number. In Zalophus I began at the most caudaul division and worked craniad. (Figs. 11, 12, 13, 24.) Adductor I in the Zalophus was thin and arose from the caudal border of the ischium between the origins of the semimembranosus anticus and quadratus femoris. It passed laterad to all the other adductors to a fascial insertion over the medial tuberosity of the tibia and onto a slight ridge extending along the distal border of the medial condyle of the femur. Adductores 2 et 3 were apparently indivisible at origin, which was along the caudal half of the pubis laterad of the symphysis and that part of the ischial border that lay ventrad of the obturator foramen. These two divisions were undoubtedly homologous with at least a part of the true adductor magnus, for the femoral artery passed between their insertional parts. At the middle of the muscle the more ventral fibers separated into a ribbonlike slip—adductor 2—extending to a ridge upon the medial condyle of the femur. The deeper part, consti- tuting adductor 3, inserted upon a long, narrow area stretching over the caudal surface of the femur from the distomedial greater tro- chanter to a point upon the medial border of the shaft midway be- tween the lesser trochanter and the condyle. On the bone this is indi- cated by a barely perceptible ridge, corresponding to the usual linea aspera. Adductor 4 was small and arose from the pubic border just craniad of the origin of divisions 2 and 3. It passed deep (mediad) to both of the last, broadened somewhat and inserted upon a slight ridge extending over the caudal shaft of the femur from distad of the middle of the greater to the lesser trochanter. Adductor 5 was also very narrow, arising along the pubic border from the pectineo-psoas process to adductor 4. It passed superficial (mediad) to all the other adductors in this area to a fascial inser- tion over the medial epicondyle of the femur. This is the pectineus of Murie. Adduactor 6 arose from the pubis immediately deep to division 5, and inserted narrowly upon the disto-lateral border of the lesser trochanter. > | agT. 15 ANATOMY OF THE EARED AND EARLESS SEALS—-HOWELL 93 The much simpler adductors of Phoca may be termed as follows: Adductor anticus (figs. 11, 12,24) was a small muscle next caudad to the pectineus and of about the same size. Origin was from the border of the pubis and insertion was mediad to that of the pectineus upon the shaft of the femur, but there was no appreciable osteological indication of the fact. Adducter posticus (figs. 11, 12, 17, 24) was a very broad sheet of muscle arising from the caudal half of the pubic border and over the lateral surface of all of the ischium save the more caudal por- tion. It converged to an insertion that was largely tendinous upon the prominence upon the femur just laterad of the trochanteric fossa. Murie indicated that there were six adductors in E'wmetopias. As already mentioned, his adductor brevis primus is the true pectineus, and what he termed pectineus is my adductor 5. I am unable to determine the exact number of adductor divisions in Arctocephalus but judge that conditions were very similar to those in my Zalophus. Miller wrote that the adductors are absent in Phoca, but it is evi- dent that he mistook the two adductors for the obturator externus and did not dissect deeply enough to encounter the latter muscle, restricted as it is in this genus. MUSCLES OF THE LEG M. gastrocnemius (figs. 12, 26, 27,28) was single in the Zalophus and arose from the well-defined ridge upon the medial epicondyle of the femur and the capsule of the joint. It crossed to the outer side of the shank and developed a tendon (first upon its medial border) which was inserted upon the caleaneum. In the Phoca this muscle was double. The medialis was very heavy, with fleshy origin from the caudal surface of the well developed medial epicondyle. An internal tendon developed upon which the fibers from both bellies of the muscle inserted. The tendon narrowed and was attached to the caleaneum. The /atervalis was not one-tenth the size of the medial division. It arose by a slender tendon from the lateral epi- condyle of the femur, and joined the tendon of the medial division distad of the muscular part. M. plantaris (figs. 12, 25, 27,28) in the Zalophus was about one-third the size of the gastrocnemius. It arose from the lateral epicondyle of the femur in very intimate relation with, and between, the popli- teus and peroneus longus. It passed deep to the tendon of the gas- trocnemius, over the groove upon the medial calcaneum, and thence to the plantar fascia. With care two layers of this were dissected free. The more superficial divided into four tendons between the five digits, and each of these again divided, the branches running to the borders of the adjoining digits. The deeper layer also sepa- rated into four branches, these constituting sheaths for the flexor 94 PROCEEDINGS OF THE NATIONAL MUSEUM VoL, 73 longus tendons extending to the four lateral digits. These last entered the sheaths from their deep sides near the metatarsal-phalan- geal joints. In the Phoca the plantaris was much larger than the lateral, but smaller than the medial, gastrocnemius. It had an ex- tensive fleshy origin from the caudal surface of the lateral epicondyle, the head being really separable into two parts, one arising from the ridge at this point and the other from the slight furrow adjoining. Its slender tendon passed mediad from beneath the gastrocnemius and over the depression between the calcaneum and astragalus. It then extended not to the superficial layer of the plantar fascia but deeper, to an attachment upon the plantar surface of the flexor hallucis longus. Its tendinous fibers continued, however, appar- ently to the fourth digit only. M. soleus (figs. 18, 25, 27) is present in the otariids only. In Zalophus it was very thin at origin and robust at insertion. It arose by aponeurosis from the head of the fibula and by muscle fibers from the caudal border of the shaft as well as from the aponeurosis cover- ing the peroneus brevis. Insertion was entirely fleshy upon the dor- sal surface of the calcaneal extension deep to the tendon of the gastrocnemius. This muscle was very closely involved with the peronei brevis and digiti quinti. It is lacking in the Phocidae. M. popliteus (figs. 12, 13, 27, 28) in the Zalophus was extensive but thin. Origin was by a tough tendon from the depression between the lateral condyle proper and the condyloid ridge, and by muscle fibers from the capsule of the joint, and was in intimate relationship with the plantaris. The belly expanded as usual and near the in- sertion, especially proximad, it divided into two thin sheets to allow for the passage of the internal lateral ligament. Insertion was upon the medial border of the shaft of the tibia from a point slightly distad of the head practically to the center of the shaft, but there was no osseous indication of its position. In the Phoca this muscle was rather thick but relatively-narrow. Its tendon arose from the pit cradiad of the lateral condyle, origin being a bit smaller than in Zalophus, and some of the muscle fibers also arose from the capsule of the joint. Insertion was less than 25 mm. in length and at quite some distance from the head. The flexor and extensor tendons to the digits have a habit, in diverse sorts of mammals, of wandering about, and the homologue of a hallucis or a digiti quinti muscle may be found to extend to some other digit besides the hallux or the fifth digit, respectively.. Hence one can not always judge by insertion, nor by origin either, regarding the name of a muscle. When the innervation is not diag- nostic the matter may become extremely difficult. Such is the state of affairs concerning the long flexors of Zalophus and Phoca. The one that extends to the hallux in the former does not do so in the ART. 15 ANATOMY OF THE EARED AND EARLESS SEALS—-HOWELL 95 latter and vice versa. In the former the more medial arose directly superficial to the lateral. My nomenclature is based on the fact that in the Phoca the muscle which I term the digitorum longus was located in its normal position mediad to the hallucis, and in beth my animals, the digitorum tendon is the one located nearer the calca- neum than the hallucis where they both pass over the heel. It is fully realized that the positions of the tendons at this point could be transposed, but it is necessary to have some criterion and as both muscles are served by the tibial nerve, homologizing by the innerva- tion in such specialized mammals would not be dependable. M. flexor digitorum longus (figs. 13, 27,28) in the Zalophus arose deep to the popliteus and superficial to the flexor hallucis longus. Its origin was from the caudo-medial part of the head of the fibula and from the strong tibio-fibular ligament which stretched from the head of the latter bone to a point distad three-quarters the length of the tibial shaft, which ligament marked the medial border fe the deep fascia of the shank. The tendon from this muscle passed over the more lateral of the two grooves upon the medial border of the caudal tibia, expanded and joined the tendon of the hallucis longus upon its deep surface. The two layers could be dissected apart, however, when it was seen that the digitorum longus sheet split into three branches, these going, respectively, to digits 1, 8, and 4. In the Phoca this muscle was considerably smaller than the flexor hallucis but was still a robust muscle. Origin was fleshy from the tibial side of the head of the fibula and from the adjoining border of the posterior tibial fossa as far as the internal lateral ligament. The tendon passed caudad of the internal malleolus and broadened as it extended deep to, and became fused with, the hallucis longus. After careful dissection it appeared that branches of this tendon extended to digits 2, 3, and 4. This is Murie’s flexor longus hallucis—a fact not mentioned by Miller—and its origin was simlar to Zalophus, but the precise inser- tion is not clear. Miller considered it in the same light as I do for Phoca at least, but I can judge little regarding his description of conditions in Arctocephalus. M. flexor hallucis longus (figs. 13, 26, 27, 28) is a somewhat ambiguous name as far as concerns Zalophus, but for this genus flexor fibularis would be no better. In the Zalophus it arose deep to the flexor digi- torum longus from the caudal part of the tibial border of the fibula and from the interosseous membrane, but at no point did fibers quite reach the tibia. The tendon then passed over the heel between those of the plantaris and digitorum longus, expanded, and formed the more superficial layer of the sheet common to this muscle and the digitorum longus. Further dissection showed that it split into but two branches, which extended respectively to digits 2 and 5. The 96 PROCEEDINGS OF THE NATIONAL MUSEUM vor, 73 tendons of the two long flexors which ran to the four lateral digits entered from the deep side into four sheaths which stretched from a part of the plantar fascia, as mentioned elsewhere. In the Phoca this was an exceedingly robust muscle. It arose from the caudal head of the fibula and from two-thirds of this aspect of the shaft, as well as to a slight extent from the fibular part of the interosseous membrane. .Muscle fibers ceased some 30 mm. proximad of the heel and the very broad tendon—the heaviest of the foot—passed over the groove upon the greatly specialized posterior process of the astraga- lus, in this animal actually longer than that of the caleaneum. The tone of this muscle and the form of astragalus—giving the action of a regular tendo calcaneus—is all that prevents the foot from assum- ing a platigrade position. The tendon of the flexor hallucis, after appearing upon the plantar surface, broadened and passed between the tendons of the plantaris and fiexor digitorum longus, and partly | fused with the latter tendon ectad. It split apparently into four branches, these going to digits 1, 3, 4, and 5. The one to the hallux again split, one branch extending to the dorsum and the other to the lateral side of this digit. This muscle is the flexor longus digitorum of Murie, although Miller evidently failed to notice the fact. The latter’s descriptions are very involved, and as he failed to dissect apart the two layers of tendons, no differences of significance can be noted. M. tibialis posticus (figs. 13, 27,28) in the Zalophus had fleshy origin from the extreme medial part of the head of the fibula, from the fibulo-tibial ligament mentioned under the flexor digitorum longus, and from the entire posterior tibial fossa as far as the distal quarter of the shaft. Its very large tendon passed over the most medial of the two grooves upon the posterior aspect of the tibia, down the medial border of the tarsus and metatarsus, and inserted broadly upon the terminal phalanx of digit 1, this acting as an abductor of the digit. Embedded in the tendon just mediad of the proximal part of tarsale 1 was the tarsal sesamoid bone. In the Phoca the flexor digitorum longus covered all but the medio-distal border of this muscle. It was broad but thin and arose chiefly from the poste- rior tibial fossa for almost three-fifths of the length of the bone, from the interosseous membrane mediad to the flexor hallucis longus, and from the medial head of the fibula. Its rather small tendon passed over the groove caudad to the medial malleolus and inserted upon the lateral centrale. Miller evidently followed the tendon with greater perseverance than I employed and ascribed to it considerable complexity in its attachments over a limited area. M. tibialis anticus (figs. 13,25, 26) was the most medial of the muscles of the front of the shank. In Zalophus it was rather small and arose from the head of the tibia, from the proximal] half of its shaft, = AKL. 15 ANATOMY OF THE EARED AND EARLESS SEALS—HOWELL 97 and from the deep fascia covering the extensor hallucis. Its broad tendon passed mediad to an insertion upon the medial margin of metatarsus 1. Origin in the Phoca was very similar, being from the better defined anterior tibial fossa and slightly from the adjoin- ing interosseous membrane. Its tendon passed through the deep tibial groove upon the front of the instep, went deep to the extensor hallucis tendon and inserted upon the medial border of the base of metatarsus 1, as in Zalophus. There were said to be two tendons in Eumetopias, the second inserting upon the first tarsale. M. extensor hallucis (figs. 13, 25, 26) was a weak muscle in the Zalophus arising from the interosseous membrane and slightly from the adjoining borders of both tibia and fibula. Its tendon passed over the instep between those of the tibialis anticus and extensor digitorium longus to insert upon the base of the first phalanx of the hallux. In the Phoca its origin was from the proximal part of the cranial ridge of the fibula. It lay mostly deep to the extensor digi- torum longus and its slender tendon passed over the instep just laterad to that of the tibialis anticus. It then crossed superficial to the last and ventrad of the base of the first metatarsal, after which it extended dorsad once more to insert upon the lateral side of the dorsum of the first hallucial phalanx. This was said to be of unusual volume in Humetopias. M. extensor digitorum longus (figs. 12, 13, 25, 26) in the Zalophus arose very slightly from the lateral epicondyle of the femur, from the capsule of the joint adjoining, from a small area over the part of the tibial head adjacent to the fibula, and from two-thirds of the cranial border of the fibular shaft. In addition the deeper fibers were in intimate relationship with the extensor hallucis and with the peroneal aponeurosis. In the Phoca origin was from the cranio- lateral head of the tibia. In both the tendon passed over the middle of the instep and split into four branches, which extended to the four lateral digits. Murie reported no tibial origin for Humetopias and Odobenus. Stretching from the head of the fibula to the most prominent part of the external malleolus of Zalophus there was a sort of ligament, here termed the peroneal ligament. It is a development of the usual involved aponeurosis of the peronei, and in addition had connection with the aponeurosis of insertion of the hamstring muscles. Such an aponeurosis was present in the Phoca as well, but no definite ligament was encountered nor was direct involvement with the ham- string aponeurosis noted, possibly because of the more tender state of the tissue in the latter. M. peroneus longus (figs. 12, 25, 26) arose from the lateral epicondyle of the femur. In the Zalophus it passed beneath the peroneal liga- ment and over a groove immediately caudad of the most prominent 86377—28, = ‘ 98 PROCEEDINGS OF THE NATIONAL MUSEUM \ rif // HHI WIA I, / Wy a Hh he ALIS POST. * anptalg LONG 1-FLEX. HAL. LONG. PLANTARIS Fig. 27.—CAUDAL ASPECT OF THE MUSCU- LATURE OF THE LEFT POSTERIOR LIMB OF ZALOPHUS VoL. 73 ART. 15 ANATOMY OF THE EARED AND EARLESS SEALS—-HOWELL POST. TIB.ART ERY ik GASTROCNEM.LAI- HAMSTRING MUSCLESA AW AI ~ i \\! PLANTARIS FLEX. HAL. LONG. Fig. 99 1 zz yy } L GY GASTROCNEM. MED. | MY HE: POPLITEUS Ll i Ws FLEX. DIG. LONG. TIBIALIS POST. ; I 28.—CAUDAL ASPECT OF THE MUSCULATURE OF THE LEFT POSTERIOR LIMB OF PHOCA HISPIDA 100 PROCEEDINGS OF THE NATIONAL MUSEUM vou, 73 part of the external malleolus, over another groove upon the lateral calcaneum, to the plantar surface over the peroneal groove of the cuboid, and to an insertion upon the ventro-latero-proximal part of the first metatarsal. In the Pheca the tendon passed over a groove just craniad of the external malleolus, through the deep groove upon the lateral caleaneum, and then to the plantar surface through the peroneal groove of the cuboid, here so deep that it formed a bony tunnel arched over by a process of the cuboid in contact with meta- tarsal 5. There was also a groove for this tendon upon the first tarsale and insertion was normal upon the base of metatarsus 1. Miller considered that some of the fibers of this muscle took origin also from the tibia and fibula. M. peroneus brevis (figs. 13, 25, 26). For a description of this muscle in the Zalophus, see the next. In the Phoca it was practically hidden by the peroneus digiti quinti. It arose from the proximal half of the cranio-lateral shaft of the fibula. Its tendon was in con- tact with that of the digiti quinti to distad of the calcaneum, at which point it diverged to insert upon the process directed ventrad upon the base of metatarsus 5. This has been reported as a separate muscle in the remainder of the eared seals that have been dissected, in Arctocephalus and E'umetopias arising deep to the digiti quinti and inserting upon the fifth metatarsal. M. peroneus digiti quinti (figs. 18, 25, 26) in the Zalophus certainly occurred fused with the brevis division. This arose from the lateral head of the fibula and from half the lateral border of the shaft, from the peroneal ligament and from the deep aponeurosis of the soleus. The tendon passed over the lateral malleolus just caudad of the peroneus longus tendon, then beneath the latter, and over the pero- neal groove upon the caleaneum, splitting into two branches, one going to the proximal termination of the first phalanx of digit 5 and the other to the metatarsal of the same digit. In the Phoca it arose by aponeurosis from the cranio-lateral head of the fibula. Its very slender tendon passed over the deep fibular groove directly caudad of the lateral malleolus, as in Zalophus. It then stretched distad along the lateral side of digit 5 as a well-defined tendon only as far as the basal phalanx. Miller said that in Arctocephalus it arose below the soleus, but this does not conform to his description of the latter muscle. In Odobenus Murie found a peroneous quartus, and what he con- sidered as the homologue of a peroneus tertius, the interpretation of the latter especially being doubtful, as it arose from the calcaneum. Its tendon joined that of the quartus division, and both therefore extended to the fourth metatarsal. As with the manus, the short muscles of the pes are here omitted, the full account by Murie and Muller being deemed entirely adequate. % en ant.15 ANATOMY OF THE EARED AND EARLESS SEALS—HOWELL 10] MUSCLE DIFFERENCES Before the discussion of the functional differences between the Otariidae and Phocidae is undertaken it will be necessary to digest ithe more significant of the muscle differences and to tabulate them in condensed form, so that the variations may the more easily be grasped. Certain of the myological differences between the animals dissected are of such a character that it is difficult to compare them in words, especially concisely. Many of the muscles showing slight differences or of a character which are deemed relatively unimpor- tant are omitted from the table and from the discussion. It is de- sirable to match muscular conditions of the Pinnipedia with one of the fissipeds, and solely because the domestic cat is far better known than any other this is selected for the purpose, and the nomenclature of its muscles reduced to terms comparable to those used for the pinnipeds. In the following table, then, those muscles are listed which have been found to differ in their origins or inser- tions, and the conditions in the cat (C) are given, followed by those in Zalophus (Z) and the Phoca (P). Insertion Muscle Origin Comanuibriums 2222 Saas ee lat. 44 occip. crest. Z presternum and craniad__} lat. 12 occip. crest. P lat. presternum = _—___ 22 mastoid proc. Sternomast_______-_ C ist cost. cartilage________ Bralents o-222 25. tect. cap. ant. maj_-~ Panniculus carn___- Pectoralis anti- brach. @2anerim. slipsys= 28% sisen iP 2imorm=slipss oie sey ees Z 2 norm. slips, 1 complex atlantic. C@icerv vert. 2-605 = ee L CGhvi Vertal—ow ite ee 2s Pieter. vertso-Oo es Los wenl 4 C carnivore type_________- Z modified carnivore type___ P specialized type_.-____-_-_ Chitanubrium= 2222. 225. _. Sterno - hyoid - thy- norm. ROL ae er er Z manubrium deep to pres- | norm. ternum. P ist cost. cartilage_.______ norm, Scalenus ant_______ OC rises... Mat ui seen} Scalenus med_--_-_-_ C ribs 3, PEED SRR ES enter = transv. proc. all cerv. vert. Scalenus post_---_- C ribs 6, TS; O22 SE ie mealenus 1___.*____ 7 Lis | OSs hs Se ps per eg a 4th cerv. vert. Sealenus 2__._____- 2) Tih. 2 eee Ee Es 5th cerv. vert. Ee Rimbsra; 4, S22 25 - see 5 het 3d cerv. vert. ponleniws, Qo) los 2 Pre bees by been aS uk. 4, 5, 6th cerv. vert. ) eles cl] oY 5a es Aa 2B 3, 4, 5, 6th cerv. vert. norm. norm. norm. norm. norm. elbow. 102 PROCEEDINGS OF THE NATIONAL MUSEUM VOL. 73 Muscle Pectoralis maj. su- perfic. Pectoralis maj. prof _ Pectoralis minor-____ Pectoralis abdom__- Pectoralis superfic. ant. Pectoralis superfic. post. Pectoralis prof Pectoralis Serratus mag_------ Depressor scapulae- Rectus abdominis-__- Quadratus lum- borum. * Cephalohumeral.-_-_ Humerotrapezius- -- Spinotrapezius_ -_ --_- Rhomboideus dorsi- Rhomboideus ant__- Rhomboideus dorsi-_ Atlantoscap. super-.- Origin C. cranial 44 manubrium____ C manub. and sternebrae 1, C6 sternebrae____________ Crxiphoi discon. tiie Gaus Z presterna! tip to rib 5____ A. tib, 630, x1 phoide's 23 222- Z prestern. tip to xiphoid___ P dorsad prestern. tip to xiphoid and latero- caudad to near knee. C cerv. vert. 3-7 Z cerv. vert. 3-7, ribs 1—4_- P cerv. vert. 3-7, ribs 1-2_-_ C last 2 thorac. and all lum- bar vert. Z last 3 thorac. and all lum- bar vert. P last 2 thorac. and all lum- bar vert. C from brachialis muscle _ _- Z prestern. tip; deep pect; delt. crest. P gt. and lesser tubers. -- _-- C middorsad axis to 4th thorac. Z about middorsad axis to 4th thorac. P about middorsad axis to 4th thorac. C all thoracic vert. spines_-_- Z caudal 24 thoracic vert-_-_-_ P caudal 24 thoracic vert_-_- C middle cery. to 4th tho- racic. Z med. 4 occip. crest__-_--- P nuchal ligament------_--- Ziumiddorsum = Pimiddorsum=s 332552 e=2 eee Crabsent2.. 2 22. eee hake Dvetlgge ete see al Re Insertion mid. 4g humerus. bs great. tuber. to 34 humerus. prox. 14 humerus. to latis. dorsi and pect. minor. to wrist. great tuber. to manus. 1 delt. crest and to forearm. 2 delt. crest. norm. norm. norm: norm. norm. norm. cost cart. 1, 2 and sternum. cost. cart. 8, 7, 6, 5, not ster- num. to sternum craniad to rib 1. ilium. ilium. sacrum. med. 14 occip. crest. med. 24 occip. crest. sagittal crest. dist. 14 scap. spine. delt. crest and supinator longus. scap. spine and delt. crest. 2nd 14 of seap. spine. prox. % of scap. spine. prox. 4 scap. spine. vert. border scap. prox. 3% scap. spine. glenovert. angle scap. vertebral border scap. glenovert. angle scap. vert. bord. scap. craniad of spine. vert, bord. scap. dorsad of spine. ART. 15 ANATOMY OF THE EARED AND EARLESS SEALS—HOWELL 103 Muscle Atlantoscap. infer__- mmlenius. 2-226 Trachelomastoid____ Biventer cervicis___ Episubscapularis - - _ Subscapulo- capsularis. Brachialis_ ___-___- Origin C atlas and basioccip- - ---- a NaS Se oe Reh a es ratlas bee bbs. 2 sae C nuchal ligament________- Z nuchal ligament - --_-_----_- P nuchal ligament___-_-____- © cerv: vertiAaTee soe es Z cerv. vert. 4-7 and tho- racics 1-2. Picenve Vento oa. eee C 7th cerv. to 5th thoracic vert. Z 5th cerv. to 4th thoracic vert. P 2nd cerv. to 2nd thoracic vert. C 8rd cerv. to 3rd thoracic vert. Z 2nd cerv. to 6th thoracic vert. P included with biventer cervicis. C middle 14 scapular spine- Z entire spine and caud. vert. border. Rrentire spines) 244-225 25 C prox. glenoid border of scapula. Z lat. scapula, broadly - - -_- P lat. scapula, narrowly - -_- C prox. 44 glenoid border of scapula. Z prox. 4 glenoid border of subscapularis. P axillary 14% infraspinous space. Z present in Zalophus only-_- P present in Phoca only --_- C distal 34 humerus_--____-_ Z prox. 24 humerus_-__-_-_-_~_-_ P prox. 24 humerus. ____-_- C dist. 14 axillary border scap. Z infraspin. ridge and onto vert. border. P dorsal 4 infaspin. ridge and onto vert. border. Insertion metacromion. dist. 24 scap. spine. gt. tuber. and delt. humerus. crest entire occipital crest. entire occipital crest. mastoid process. mastoid process. mastoid process. mastoid process. med. occipital crest. vertex. med. 14 occipital crest. med. 14 occipital crest. lat. occipital crest. occipital. occipital. occipital. deltoid crest. deltoid crest and supinator longus. deltoid crest. gt. tuber. of humerus. fused with infraspinatus. fused with deltoid. 3rd tenth of humerus. middle of humerus. proximad middle of humer- us. norm. norm. norm. olecranon. forearm fascia. forearm fascia. 104 PROCEEDINGS OF THE NATIONAL MUSEUM vou. 73 Muscle Origin Mricepsiates= sees Z humeral neck. 222252 P delt. crest and distal scapula. Triceps med___-_-_-- Z entire humeral shaft__ —_- P entire humeral shaft____-_ Anconeus internus__| found in the pinnipeds only_- Palmaris longus_ Eel C entepicondyle___._-_.+-_- Z entire medial face ulna, huge. iP olecranons2et fia eee Flex. digit. com- | C from palmaris long. and munis 1. flex. profundus. Flex. digit. com-|C entire ulna; entepic; munis 2. middle 14 radius. Flex: digit. com- | Z 4th tenth of ulna_—_____- munis 1. Flex. digit. com- | Z entepicondyle__..--____-_ munis 2. Flex. digit. com- | Z middle 3% radius_______- munis 3. Flex. digit. com- | Z entepicondyle_____-_____ munis 4. Flex. digit. com- | P entepicondyle____.______ munis 1. Plex. digit: com- |“Piprox: 24 ulnaceweet ie. munis 2. Flex. digit. com- | P 8rd eighth of radius______ munis 3. Flex. carpi ulnaris___ 1 Frolecranon: asa Abduct. longus. dig. 5 | P in Phoca only, from ole- cranon. Supinator longus_ _ - Z prox. 24 humerus, broadly- P prox. 44 humerus, nar- rowly. Psoasaminor= =. —— 2 | C last 2 thorae. and 1-3 or 4 lumbar vert. ZANast 3 lumbarsae kek 3 P last 3 lumbark= 30k 4 as Psoas magnus-_--___-_ Z last 2 lumbars and sacrum_ P last thorac., all lumbars, and sacrum. Piagcugae ses are AVeTuve lenin ee Gluteus quartus____| Z present in Zalophus only weak. C deltoid ridge___--_2----- C prox. ¥ humerus__-___-_-_- C entepic. and olecranon___- Pioletranon: 32s | hess C middle 4% humerus-- _-_ -__- C last 2 thorac.—7th lumbar @iventralahiumie_-2eduzes ge P! ventral iliumivo. $2222.28 Insertion olecranon. forearm fascia. olecranon. olecranon. olecranon. olecranon. norm. norm. norm. 4 lat. digits. 5 digits. digits 3, 4, 5. digits 4. digits 1, 2, 3. digit 1. 5 digits. pisiform. pisiform. pisiform and metacarp. 5. digit 5. distal radius. distal radius. distal radius. ilio-pectineal line on innomi- nate. ilio-pectineal process. ilio-pectineal process. lesser trochanter. lesser trochanter. med. tuberos. of tibia. lesser trochanter. lesser trochanter. medial tuber. tibia. ART. 15 Muscle Quadratus femoris_-_ Semimembranosus--_ Semitendinosus-_-_--- Biceps femoris-_-_---- DARtOMUS2= 222 2550-2 Vastus lateralis__-_-_-_ Vastus profundus-__-_ Rectineus.—.- 2.6 Chena i mAdductor 2 =. = AGGUCTON 222-2 -- Adductor ls _ 2. F< Adductor 2 and 3___ adductor A= => AGaUCtOY F222) - = Adductor/6u_.2_e52 AMAGUeLOL.. MaguUctor 2. 2 = Gastrocnemius lat __ ANATOMY OF THE EARED AND EARLESS SEALS—-HOWELL Origin Caschimas 23 aues tere sb TASCA ee ne Pabsentis2 Sei Ooi Saree Z caudo-ventral ischium_ ___ P caudo-ventral ischium_ -_-_ Caschiaittuber== 22 Z spines 3d—6th caudals__-_-_ P ae proc. 1-3 caudals_ caudal. ischrums)2)3 22) 4 @Aschial tuber= 222 = 222. Z, eae 2-7 caudals_-_-__-- transyv. plate 2nd sacral__ eee ischial spine_ ~~ -_--_ transv. plate 3d sacral_-__ C crest and vent. border ilium. eta angle ilium_----- ventral angle ilium___-_-__ P ventral angle ilium___--_-_ C gt. trochanter and prox. 24 femur shaft. Z gt. trochanter and prox. 24 femur shaft. P gt. trochanter. single in pinnipeds; double in cat. C cranial pubis. Z pectineo-psoal process- - - - Prcramalypubises222 2225 2—5 C symphysis pubis__------- Z position comparable to symphysis. P position comparable to symphysis. C caudal ramus ischium____ Crcrantalapubises 22s aee Z caudal ischium_-________- Z caudal 14 pubis_—_ 2 -~_-- Zmedial pubis: =e. =.= — == Zeeraniita ley Upiste were cae. _Z cranial pubis deep to 5_-_- iP;eraniglp ibis = = Preaudal Lo"pubis=22-=2=— 105 Insertion distal greater to lesser troch. whole femoral shaft. entepicond. femur and prox. tibia. 4th fifth of tibia. 2nd to 4th fifth of tibia. 2nd fifth of tibia. distal 14 tibial shaft. \sa Periaiie Siihaan tists prox. 4 tibia, and patella. ee condyle to distal shank. \prox. % of shank. prox. 14 tibia, and patella. patella. medial tuberos. tibia. patella. patella. patella. patella. distal base lesser troch. distal base lesser troch. distal base lesser troch. 3d sixth of tibia. medial third of tibia. distal 14 of tibia. whole shaft femur. 2d and 3d fifth of femur. med. tuber. tibia and epi- cond. femur. dist. gt. troch. to entepi- cond. 2d fifth of femur. entepicond. of femur. lesser troch. middle ¥% of femur. base gt. troch. 106 PROCEEDINGS OF THE NATIONAL MUSEUM VOL, 73 Muscle Origin Insertion Plantaris--25-) «.¢. C tendon over caudal cal- caneum. Z tendon over medial cal- caneum. P tendon over medial cal- caneum. Soleusi234 b. 4782 834 present in the cat and Phoca only. Flex. digit. longus__| C middle % tibia__________ to tendo flex. hallucis longus. Z fibularidheads= 8.8 eee 2 digits 1, 3, 4. P fibular and tibial heads___| digits 2, 3, 4. Flex. hallucis long__| C middle 5/7 of tibia______- tendons to all digits. Z prox. 2% interos. mem- | digits 2, 5. brane. P prox. 24 fibulavewteuey? : digits 1, 3, 4, 5. Tibialis posticus___.| C prox. 4 tibia__._....... norm. Z fibular head and prox. 34 | norm. tibia. Pipnox. 24 tibide nti: shee te norm. Extens. digit. longus_| C ectepicond. femur_-__--_~- 4 lat. digits. Z ectepicond.; tibial head; | 4 lat. digits. prox. 24 fibula. P tibialeheadias 2s eee ak 4 lat. digits. Peroneus longus__- | C prox: 4% fibula: . ste%442 all 5 metatarsals. Z ectepicond. femur____-___- metatar. 1. P ectepicond. femur__-_-_-~_- | metatar. 1. Peroneus brevis____| C distal 4% fibula_________- metatarsal 5. Z fused) with next-—- === | metatarsal 5. Piprox..34 fibula s 2 > Sheek metatarsal 5. Peroneus digiti:5_«=) (Cabsent__ =... 2.2 eee t | Z with last, prox. % fibula__} digit 5. Pifilular heads = ies eee digit 5. When one casually examines either a dissection of the animal itself or of a series of drawings of a pinniped he will likely be struck by the apparent vastness of the difference from what may be termed a normal mammal, but this difference is not as great as it seems, and a detailed study, muscle by muscle, will lead the investigator to the opposite extreme and mildly astonish him that a mammal which departs in many details so widely from the generalized type can adhere so faithfully to the fundamental carnivore plan of myological arrangement. The pinniped osteology is, of course, very specialized, and the muscle attachments must synchronize accordingly, which is the main reason why the musculature appears so complicated at first sight. In the above table the origins of the Zalophus resemble those of the Phoca in a few more instances than they differ, while the differ- ant.15 ANATOMY OF THE EARED AND EARLESS SEALS—HOWELL 107 ences of insertion outnumber the resemblances. The muscles above listed, both in origin and insertion, differ from those of the cat in about 14 per cent more instances in Zalopius than in the Phoca, but the differences in origin are more numerous than those of insertion. On the other hand, when these muscles of Zalophus are compared with those of the Phoca it is found that the origins are more con- servative, while the insertions are more prone to differ. Little or no account can be taken of muscles which may be rela- tively more or less robust because of the difficulty of comparing the cat, the lean otariid, and the fat phocid. ‘The following comparisons may, however, be made, with the cat as standard: Muscle conspicuously broader at origin: Pectoralis in both, and biceps femoris in Zalophus. Body muscles that are shorter: In both genera the sternomastoid and scalenus, the latter in Phoca especially. Occipital muscles with broader attachment: Cephalohumeral in Zalophus. Occipital muscles with narrower attachment: Cephalohumeral and splenius in Phoca,; biventer cervicis in Zalophus; and sternomastoid in both, especially the Phoca. Limb muscles whose origins have shifted distad: Triceps medialis, palmaris longus, flexores carpi ulnaris and hallucis longus, and ex- tensor digitorum longus pedis in both animals. Limb muscles whose origins have shifted prowimad: Brachialis, triceps longus, flexor digitorum longus pedis, and peronei longus and brevis, in both animals. Limb muscles whose insertions have shifted distad: Pectoralis (part), humerotrapezius (especially in Zalophus), teres major, tri- ceps longus, semimembranosus, semitendinosus, and biceps femoris, in both; deltoid, triceps lateralis, quadratus femoris, and adductores 1, 2, 3, and 5 in Zalophus; and atlantsocapularis inferior, psoas magnus, iliacus, and gracilis in the Phoca. The inclusion or ommission of some muscles in the above groupings are at times largely arbitrary, for it may be difficult to be sure whether a muscle is attached only to the tibial head, for instance, or whether it also encroaches upon the femoral condyle. Also it must be taken into account that most of the limb muscles are relatively shorter than those of the cat usually merely for the reason that the bones themselves are shorter. It is to be seen that only two of the body muscles of the Pinni- pedia listed in the tables (pp. 101 to 106) have become shorter, and these to a very slight extent. One muscle of the occipital crest has become broader and four others of this region narrower in their attachment. The origins of five limb muscles have shifted distad and five others proximad, to a very slight extent in all cases save 108 PROCEEDINGS OF THE NATIONAL MUSEUM VoL. 73 that of the triceps, and 14 of the insertions of Zalophus and 11 of Phoca have shifted distad. Further scrutiny shows that almost all of these insertions that have shifted distad are muscles of the upper arm and thigh, while of the antibrachial and shank muscles whose origins have shifted half have moved proximad and half distad, an exception being the triceps longus. On the average, therefore, the tendency has been toward a lengthening of the limb muscles, not actually but in respect to the positions of their bony attachments. DISCUSSION In the following pages Zalophus and Phoca will be discussed as two mammals that are chiefly aquatic, differing from each other in certain respects and from the normal terrestrial carnivore in others. There will be no discussion in the present paper of the probable derivation and but little regarding the relationship of the Otariidae and Phocidae, nor of certain broad principles and laws intimately correlated with the development of a mammal for an aquatic life. It may be mentioned, however, that the writer considers the serious comparison of the otariids with the bears and of the phocids with the otter (as Mivart, 1885), in an attempt to prove that the ancestry of these two pinniped families can be traced to members of groups now living, to be a rather unprofitable pastime. There are many resemblances, it is true, but it is very probable that the eared seal phylum is older than the bears. Not only is the Pinnipedia a very ancient order but the carnivore stem has had very numerous branches, and it is extremely unlikely that the protopinniped,was at all lke any living fissiped. In all the pinnipeds the relaxed position of the anterior nares is almost closed and naturally remains so between respirations even when the animal is on land, although I have seen a sea lion maintain its nostrils in a dilated position for several minutes at a time, and also a phocid when panting after considerable exertion. Tight closure can be effected both by contraction of the naso-labialis, pulling the mystacial pad against the nostril, and by contraction of the fibers of the mystacial pad itself, which radiate toward the surface and prob- ably have some voluntary muscle action. Expansion of the nasal opening is effected by flexion of the maxillo-naso-labialis, which pulls the mysticial pad laterad. This pad is much broader and thicker in the phocid, but there was apparently no difference in the operating mechanism of the anterior nares to account for it. Possibly the reason for the difference in size of the pads may be found in some variation of the tactile function of the vibrissae. In breathing on land the otariid keeps the nostrils virtually closed between breaths, opens them moderately at exhalation and widely during inhalation ART. 15 ANATOMY OF THE EARED AND EARLESS SEALS—-HOWELL 109 In the water this animal often swims partly on its side and when coming to the surface for breath, will inhale through the corner of the mouth as well as through the nose, as a human swimmer often does. Perhaps more frequently, however, it will break water with the tip of the nose and breathe through this member only, as the phocid usually, if not always, does. It may be mentioned in this connection that the epiglottis of Zalophus appears to be unusually small for the size of the animal, which one would not expect to be the case in an aquatic form which habitually comes to the surface for quick breaths. | I have encountered no statements regarding the possible duration of submergence of the Pinnipedia which I regard as both significant and trustworthy. In captivity an animal seldom experiences any incentive for lengthy submergence and I have never seen one do so for more than about two minutes. If frightened in the wild, the animal which reappears can not always with confidence be regarded as the same one which has disappeared. It seems certain that all pin- nipeds must be able to stay beneath the surface for some considerable time, while it would seem absolutely essential that the boreal mem- bers of the Phocidae, at least, which must often have to swim be- neath extensive ice floes, should have this faculty especially devel- oped. Of undoubted importance in the present connection is the _ development in this order of a large hepatic sinus, consisting of a remarkable dilation of the vena cava dorsad of the liver. For the Phocidae this was said by Murie (1874, p. 545) to have been reported and illustrated by Barkow (presumably H. C. L. Barkow in the early nineteenth century), but I have been unable to find the article to which he refers. Murie (1874) states that in his Humetopias the sinus occupied “a volume, one might almost say, greater than the glandular hepatic organ itself.” In examining the viscera of the Phoca hispida which I dissected, Paul B. Johnson encountered such a sinus, dilatable to contain perhaps 2 quarts, but in the younger . Zalophus it was much less developed. It is, therefore, possible that this sinus is developed with age and that it is largely lacking in juveniles. It is, of course, apparent that it serves as a reservoir to hold an extra amount of blood and hence to prolong submergence by just so much. Throughout sealing literature one often encounters statements to the effect that pinnipeds appear to be veritable sacks of blood. While on the subject of the viscera it may be well to mention, in passing, the habit of the Pinnipedia of swallowing stones, sometimes as large as an egg and aggregating as much as 3 pounds in weight. The reason for this action has not been determined. In the Zalophus the ear is slender and had a length of 28 mm. The external opening of the auditory tube is small and the pinna of 110 PROCEEDINGS OF THE NATIONAL MUSEUM You, T2 the ear laps at the base, so that when the ear is pulled backward by the cervico-auricular musculature, flexion of the mandibulo-auricu- lar complex furls the base of the pinna and effects complete closure of the tube. There is no conspicuous valve within the tube. In Phoca hispida conditions are considerably different, for there is no external ear or pinna. Near the orifice, however, there is a small hbrous plug which acts as a valve to close the tube upon contraction of the mandibulo-auricular. No action of the cervico-auricular mus- cles could be detected in living animals. Dissections of Ernst Huber, however, indicate that there is considerable specific variation in the mechanism for closing the ear of the Phocidae. The auditory tube is not longer in the phocid, and so the external auditory orifice is really no farther dorsad in this animal than in the otariid, but the sagittal line is higher in the latter, and this increases with age, so that in reality the top of the head is higher above the ear and the head must be thrust higher out of water for the animal to hear. In the phocid the eyes also are directed more dorsad (15° to the vertical as against 50° in the otariid), and the external nares as well (45° to the cranial axis as against about 20° in Zalophus), so that eyes, ears, and nostrils are so placed in the earless seal that these organs of sense may be utilized while the animal exposes the minimum amount of its head above the surface of the water—a definite aquatic modification developed to a greater degree than is the case in Zalophus. Quite diverse stimuli seem to have been instrumental in molding the characteristics of the neck in the otariid and the phocid. In adult bulls of the former the neck acts partially as a repository for surplus fat accumulated to sustain the animal during the breeding season. Unfortunately proof is at present impossible, but I am strongly of the opinion that at the approach of the breeding season ‘when the otariid bulls must do battle for the females, the increase in the swelling of the neck is also partially due to an enlargement and coarsening of certain of the cervical muscles, this action being caused by a hormone or similar secretion of the awakening sex glands. This had puzzled me for some time until O. J. Murie informed me that he had noted a great increase in size of certain neck muscles (chiefly the sternomastoid I believe) during the rutting season of the caribou (Rangifer), the purpose of which is evidently to add to the fighting ability of the bulls. This is entirely comparable to the relatively great increase in the length and size of some of the perineal muscles of female mammals at the imminent approach of parturition. This cervical swelling does not take place in females and young bulls of the Otariidae. Even in old bulls submersion lightens the weight of the neck, and the more powerful musculature of this sex may theoretically handle the large neck even more agilely under agr.15 ANATOMY OF THE EARED AND EARLESS SEALS—HOWELL 111 water than is the case with females and young males. At any rate, even though the neck is relatively no longer than in the cat, it always has the appearance of being jong and flexible, capable of great con- tortion and great precision of movement. This has been developed in the pursuit of agile prey, necessitating darting movements of the head here and there, and also in the sinuous movements which this family is seen to use while swimming, thrusting the head this way and that as an accessory rudder in aid of the more posterior one (the hind feet), with the middle thorax as the fulcrum of leverage. For the reason that natation is almost exclusively by means of the fore feet, this sinuous development of the neck has not been inhibited by a more powerful stimulus, as seen to best advantage in Cetacea and to a lesser extent in the Phocidae, in both of which the neck acts as part of the fulerum upon which acts the lever of the posterior end of the animal during swimming. The mobility of the head and neck of the otariid is further increased by the part which they play in terrestrial locomotion, during which the neck violently sways back and forth, not only to maintain proper balance, but as a direct aid, chiefly through the broadened cepahlohumeral, to rhythmic motions of the forelimbs. Besides the cephalohumeral, other broad muscles which aid in diverse movements of the head are the anterior rhom- boid, splenius continuous with the trachelomastoid, and the broad- ened insertion of the sternomastoid; and the exceedingly complex third or atlantic division of the longus colli must also have an important bearing in this connection. There is apparently no reason why an agile neck would not be of great advantage to a Phoca also, and it is not meant to imply that this member is actually stiff. But the fact remains that extensive observation of both sea lions and seals shows that the latter uses its neck in a different manner. It seems actually to have extensibility and may be stretched to a surprising degree, but in retrieving a dead fish there is not the speedy action and “catch on the fly” of the otariid, but a more deliberate nuzzling and capture, nor is the head and neck moved much as an aid to terrestrial progression. The articulations of the vertebrae are doubtless as free, but the muscles are apparently used in a different manner. Largely instrumental in this lack of agility is the fact that in this family, as in the Cetacea, practically all the motive impetus during aquatic progression is fur- nished by the extreme posterior end of the animal, and the neck as well as the anterior thorax acts as the fulerum upon which works the powerful musculature of the lower back. The head, together with the forefeet, doubtless plays a most important part as a rudder, but in very circumscribed movements, for a slight twist dorsad or Jaterad would be all that was necessary. In other words, no animal that 112 PROCEEDINGS OF THE NATIONAL MUSEUM You, 73 propels itself through the water at speed by means of its posterior parts can wave its head around in all directions, but an otariid, with center of motion near the middle of its mass, can move the head in any direction it pleases if it at the same time employ its rear end in antagonistic or compensating movements. Accompanying this state of affairs is found a distinctive condition of the muscles with attachment to the occipital crest and ventrad. In contrast to the situation in Zalophus, in the Phoca the cranial attachments of the cephalohumeral, humerotrapezius, and anterior rhomboid are narrowly confined to the vicinity of the vertex, while the middle and ventral parts of the occipital crest are free of all muscles which are more characteristic of the normal occipital crest, the superior oblique being the only one found here. As the above have narrowed mediad, so the others have narrowed ventrad, and the insertions of the sternomastoid, trachelomastoid, and splenius are confined to the mastoid process. Thus the muscles here discussed are better situated to work in two planes of movement in the Phoca— directly dorsad and directly laterad—which are most effective in rudder movements, rather than muscles fitted for movement in all directions, as in the otariid. The number of vertebrae in the thoracico-lumbar series, 20, is the same as in the majority of fissipeds, which is one more than the prim- itive number as mentioned by Todd (1922), while the tendency in most orders is toward an increase in the number of this series. The almost total lack of definition of the neural spines in the anterior thoracic region of Phoca conforms to the claim by Harrison Allen (1888) that these spines are practically absent in those mammals which do not use the anterior limbs for support, such as Dipus and the bats. This statement furnishes food for thought, but the con- ditions are far from being as simple as implied, for in the Cetacea, which have surely abandoned manual support for a longer time than any other mammal, the neural spines of the anterior thorax are always well developed and at times are enormous. It might be thought that in view of the greater development of the back muscles in the Phocidae the spines might be higher than in the Otariidae, rather than so much lower, but such is not the case, and it is probable that the narrow, deeper back musculature of Zalophus, with the longer spines, has developed for movements of all sorts, especially in the sagittal plane, while the immensely broad back musculature of Phoca, with wide vertebral articulations and very low spines is in response to movements that have been so largely lateral. Photo- graphs of Mirounga (see fig. 29) show that from a prone position on land this animal can elevate the anterior part of the body, includ- ing most of the thorax, to an absolutely vertical position in a manner that one would judge to be beyond the power of any mammal. There art.15 ANATOMY OF THE EARED AND EARLESS SEALS—HOWELL 113 is no skeleton of this genus at hand but I can not find any modifica- tions in other Phocidae examined that might permit such an unusual position to be assumed. No very significant differences in the thorax proper can be detected, save a tendency toward broadening in the phocid and apparently the lengthening in the otariid, possibly due in the latter to the advantage of having bony protection against hydrostatic pressure over the greatest possible area of the abdomen. The presternum that is well developed in both animals has undoubtedly been lengthened by a forward extension of the pectoralis, which is of such prime impor- tance to the swimming of the otarid,and probably secondarily so in the phocid. In the present paper there will be no attempt to calculate by for- mulae the leverage and potential strength of the limb segments and their muscles. Such treatment of the subject gives the result a pro- found and scholarly appearance, but the writer views with the great- est distrust all such treatments, for they can not take into considera- tion the differences of fascial attachment, and no one can tell exactly what any particular muscle either can or will do. The anterior limb of the Pinnipedia as it now occurs is the result of three stimili which are hard to unravel—the fact that the prox- imal part is within the body, operative in both; phylogenetic influ- ences, of an unknown degree of resemblance; and the fact that the forelimb is the primary organ of propulsion in the Otariidae and practically inoperative for this function in the Phocidae. In the otariid the foreflipper is one of the most important, and in the phocid, one of the least important, parts of the body. On land the former animal uses this member as normally as its proportions will allow. It is extended at the wrist at a right angle to the antibra- chium, the toes being directed almost directly laterad. In the latter the manus may be used to help the animal from the water or over a rough spot, but its shortness and the thick blubber layer often pres- ent over the chest raise the manus too far from the ground for it to be of great use. Hence it is usually held somewhat pendant and abducted (from the forearm). As far as I can tell the most natural, static position of the anterior limb in both animals is with the humerus at slightly less than 90° to the scapular spine. In the otariid the antibrachium is almost extended, and in the phocid flexed to almost 90°. In the former the manus is almost on a line with the antibrachial axis, and in the latter, abducted to at least 45°—usually more. The static posture of these segments with relation to each other is shown in Figure 30. In the water the otariid moves with broad sweeps of the powerful manus, recovery being made with the radial border of the arm pre- 86377-—28 8 114 PROCEEDINGS OF THE NATIONAL MUSEUM VOL, 73 sented craniad, while the rear limbs play a very minor part as far as I have observed, never being used in rhythmic motions, but in various steering movements. During brisk swimming the forefeet _ of the Phocidae are folded against the body save when the animal wishes to make a sharp turn, at which time the outside flipper will be abducted and thrust against the water, as a man would push against a wall in making a similar movement. When a seal is merely loafing about in the water, with slight turns, rolls and such idle actions without definite idea of progression, the hind limbs may be entirely immobile while the forefeet maintain an intermittent “ fid- dling” movement, such as a man would employ while treading water. I have no doubt that when the seal is suddenly alarmed it employs its forelimbs in active swimming movements in any way which might be of assistance in starting quickly. Tt is almost impossible to determine the degree of pronation and supination of which the whole arm is capable, for so much of it is within the body covering that in the entire animal one can not follow the interaction of the joints, and after sufficient of the muscles have been cut away to determine this the results are worthless because many of the inhibitions to movement normally raised by taut mus- cles and bulk of tissue have been removed. It is probable, however, that at least in Zalophus there is less of such movement possible than in man, while in both there is more than in such a fissiped as the cat. Few conclusions regarding the scapula may be reached, and its chief stimulus for specialization in the Pinnipedia is doubtless as a scaffold upon which are hune the arm muscles, rather than as an attachment for muscles of the thorax and neck. It is relatively more rebust in the Zalophus, as one would expect, but contrary to expectations, the axillary border is relatively the longer in the Phoca; for this animal! has less need for a long lever arm for the triceps. In the Otariidae at least the scapula is unusually mobile, and slides about beneath the skin when the animal is in terrestrial movement in exaggerated manner. When resting with head low the scapulae may be in contact, projecting for several centimeters above the dorsal line, or by means of the serratus magnus and depressor scapulae muscles they may be forced well ventrad, which correspondingly lifts the body. Incidentally it may be mentioned that in most of the articulated skeletons which one sees the thorax is elevated above the ground to an unnatural extent. In the Zalophus the supraspinatus is large and powerful as an aid to extension of the humerus but in the Phoca® it is considerably weaker, as one would also expect. In both animals the infraspinatus 5At least in Mirouwnga, among the Phocidae, the supraspinous fossa is relatively quite as large as in the Otariidae. i ART. 15 ANATOMY OF THE EARED AND EARLESS SEALS—-HOWELL 115 (2 rotator) is very weak while other muscles have encroached upon the infraspinous space of the scapula, which is especially the case in the phocid, for here this space is relatively considerably larger than the supraspinous fossa. There is no very clear reason to be seen for the extension caudad of the glenovertebral cartilage of the Phoca, although it is self-evident that it has taken this course in response to stimuli supplied by the muscles attached thereto. The percentage of arm length (humerus, radius and manus), based on the bones only, to body length is in a cat skeleton 82, Zilophus 66, and Phoca 48 per cent, so it is seen that in compari- son with a fissiped the phocid arm is much reduced, while that of the otariid occupies an intermediate position. In the same order as above, the length of the humerus compared to body length is, respectively, 31, 18, and 14 per cent; of the radius, 30, 20, and 14 per cent; and of the bony part of the manus, 21, 29, and 20 per cent. It is thus seen that there has been a shortening of the two upper segments of the pinniped arm and that the proportions of one seg- ment to the other have remained almost the same as in the fissiped, save possibly that the rate of reduction in the size of the humerus has been a bit more rapid in the case of Zalophus. There has been no change in the size of the manus of Phoca relative to the entire arm, but that of the Zalophus has increased in relative size (com- pared to the more proximal segments) about one-third. Presumably, however, this osteological increase in the size of the otariid manus has not been sufficiently rapid to meet the needs of the animal and a still larger area for furnishing propulsive force has been acquired by the development of cartilagenous extensions to the digits. As an alter- native one must consider the unlikely possibility that the presence of these cartilages is due to some stimulus other than that caused by the need of a longer manus in swimming. At any rate, they have made the functional length of the manus (measured to its tip) of the Zalophus about 40 per cent of the body length, or, from a relative standpoint, fully twice as long as in the Phoca. From still another aspect, the effective length of the anterior limb operating against the water is the distance from the axilla to the tip of the flipper. In the Zalophus with the axilla at a point just proximad to the middle of the ulna, this amounts to 52 per cent of the trunk length. In the Phoca, the axilla is opposite the ulnare and the corresponding percentage is about 19. As the anterior limb is used for such very different purposes in the two families, however, a comparison of the visible portions does little but call attention to their dissimilarities. In the Otariidae the extended arm may be operated to good advantage as a swimming organ by muscles coming from other parts of the body. Thus the cephalohumeral is attached im part to the 116 PROCEEDINGS OF THE NATIONAL MUSEUM VOL. 75 deltoid crest but it also has connection with the tissue about the presternum and the border of the pectoralis. Its chief action is upon the head and neck but it also extends the humerus both in swimming, and upon the land by means of lunging, forward and back movements of the head, and is thus of definite aid in terrestrial progression. It is also involved with action of the anterior part of the pectoralis, which by means of its insertion as far distad actu- ally as the palmar tissue, adducts the arm with much power. Of the greatest importance in strong backward sweeps of the flipper are the intimately connected posterior portions of the pectoralis, the latissimus dorsi, and most of the panniculus carnosus, all of which have a long power arm upon the anterior extremity. In the Phoca these muscles perform other work and will be discussed elsewhere. Relative to length of body the humerus of Zalophus is less than 65, and of the Phoca about 45 per cent of the length of this bone in the cat. It is proportionately massive, with prominent and broad processes and ridges. Both of these details denote great power of the muscles attached. The humeral head differs somewhat in the two animals, indicating that the normal position of the humerus in Phoea is slightly more flexed and abducted than in Zalophus. Ex- tension and flexion of the humerus is limited largely by the fact that it is entirely within the body covering and that the integument limits the movements of the forearm also; but abduction of the humerus is fully as great as in many fissipeds—Canis for instance. As the arm is largely within the body, flexion of the part distad of the humerus, by the single biceps and brachialis, is limited, while exten- sion of the humerus is well provided for by a complex triceps with a leverage much greater than in the normal fissiped. In the otariid the great height and massiveness of the greater tuberosity is for supplying increased leverage to the large supraspina- tus which helps extend the humerus—a motion of much importance in recovery after a backward sweep of the foreflipper. This move- ment is not of importance to the Phoca, so the supraspinatus is smali and the greater tuberosity lower than the femoral head. The enor. mous development of the deltoid crest, extending distad from the greater tuberosity for two-thirds the distance to the condyle in Zalophus and half the distance in Phoca, is of the utmost importance in not only supplying several times the leverage to the shortened bone that these muscles could furnish to a humerus of the human type, but also, through the great elevation of the deltoid crest, in presenting an efficient lever arm for strong rotation, needed in the “ feather- ing” action of the flipper of Zalophus in swimming. The cephalo- humeral, humerotrapezius, and pectoralis have probably had far more to do with the elevation of this crest than the deltoideus. In the Phoca the deltoid crest, although not quite so long is fully as high— ANY. 15 ANATOMY OF THE EARED AND EARLESS SEALS—HOWELL 117 higher distad. It can not play the same part in the economy of this animal as in the otariid, for the forelimb is practically useless as a swimming organ. Of course the high deltoid crest may be a relic from a time when the foreflipper may have been so used, but I believe another explanation is the proper one, and this is that the high deltoid crest of the Phoca was developed in response to stimuli provided by strong antagonistic action of the muscles attached thereto; in other words, that it has acted as a static fulerum while the animal isswimming. The more caudal part of the pectoralis and the latissimus dorsi are both of decided aid in the rhythmic lateral swimming movements of the posterior body. These insert upon the r r ¢ » Bf Soldier Heyo Fowey °. ; Rocks GCOULDS » RICUTLEA ’ L BLACK POINT’ JPPRINCETON «” ’ har . ay VARANJA O> JODELLO ?omm | WESTEAD RIDA CITY Ellictt’s x Ajax *Reef Old RhOdES , pacific Key S Reef : SsAngelfish Hey Sound Pumphinkey Hey Largo Carysfort YAeer ROCK HARBOR HOMPSON YERNIER hvernier (ey Molasses Reef iy * Pickle'’s Reef Conch Reef is Reef Q & 4 [F} 20 AILWWAY ST WAICHIS SANDY WEST EXTENSION AY WHICH HAS ANY PICAL zs S =< THE FLORIDA TREE SNAILS OF THE GENUS LIGUUS By Cuartes Torrey Simpson INTRODUCTION The following pages are the result of studies and collections made in the field during a residence of nearly 30 years in southern Florida and of six visits to Cuba which permitted me to work over the island from Cape San Antonio at the west end to within a hundred miles of Cape Maisi at its eastern point. I have also visited Haiti and collected Liguus virgineus in considerable quantity. This work on the islands was not at all thorough as it was done in connection with that of getting plants and collecting the land snail fauna in general, but it gave me a considerable amount of Ziguus material and ideas of its distribution. I have made a rather careful search in Florida, having tramped the East Coast Railroad several times from Key Largo where it enters the Upper Keys to Key West at its terminus and back again. I have visited practically every one of these islands which have ever had Liguus and made collections on them. On the mainland I have been in nearly every locality in which these snails were found, and many of these places I have visited and worked over a number of times. In all I have some kind of Liguus material from over 300 Floridian localities, most of which I have personally collected. For nearly 26 years I have had an opportunity to study these snails in my own hammock within a hundred feet of my door. Long before I began to collect (1882) man had wrought great destruction to the hammocks in which they live so that certain forms were on the verge of extinction and in some localities all evidence of them was obliterated, and at present the Zzguus are almost ex- terminated in Florida. Great areas of forest have been recently cut in Cuba in order that sugar cane might be grown, and it is probable that it will be a short time only when these snails will be wiped out entirely in many localities in that island. Very much of the evi- dence necessary to a complete study of them is therefore missing, and the future student will only be able to use material collected No. 2741—.PROCEEDINGS U. S. NATIONAL Museum, VOL. 73, ART. 20. 97556—29—__1 1 a = i ES eile OR Be 7 : * ¥ + = Ser » : i U. S. NATIONAL MUSEUM 3 wo: ow: S, OL ing, BSH ° 2 * « : £ rr .?) ‘ as pos? ‘> coxa P Cape Romano ; Indiankey? |S sardine eos" chok!9S” pa ssitKerye” DADE ws Pavillion. Ges MONROE Coun * ey Morman Key } ‘ape Fierida * COUNTY ; Turkey Key£ 2 : A : r) a : "1, { Ragged oxeys Aighland Point ao Elliott's GULF OF MEXICO \ af Lanetgey Shark Poin . 48, eth é 5 Snag GY ner j } TottensQ gig pnodies srecine Pata’ maces Card*72 Shnseltiah oe, ep Re! * Seagle Xe. Y OJoekemp's East Cape 7S S Nese 3 ° ° Clivexe y? a 9 0 Man-o'- war Key Sandy) ctuetf hey : & ¢? 2 26 {Molasses Reef 0 antarion ‘ePickle’s Reef ney, Ne F Jon pes Pouanry + Conch Reef seedecneag otiihlne?” FLORIDA BAY Parra? potent & a: oggarh i ofprden Key Lignumvitae ° GZ ppeer imbeKey ptrocker's Reef © eee) ; ney Geatable Ke. od © aird a P indian Key ie nee Alligator Reef . ‘Matecumbe Key © Jewfish Key o RY TORTUGA ; if ORY TORTUGAS a Long Key sMCAgSCENT S$ gS : Onc KEY 3 ‘ ones Keys S ced * ¥ Ww MARAT! My 2, ramleeys aN ae Keys "Bahia Hondo fags le Te Cottrell Key STATUTE MILES Mullet Key fomirero Key Mule “Key Heeeteee+ FLomioa East COAST RaiLWway al : may, meneseee=+ AIOCE ON TNE EAST COAST WHICH IS SANOY Marquesas Ny “Fin washecwoman's SOU MY IN ITS UPPER PART AND ROCKY AT THE LOWEST EXTENSION xeys Soca crGrde % Ney = american Shoal WA Int Lisi OF TERRITORY WHICH HAS ANY afastercnSambo CLAIM TO BEING CALLED TROPICAL Tadie Som: oRock Key neater Bete" Western Ory Rocks ee , 97556—29. (Face p. 1.) 2 PROCEEDINGS OF THE NATIONAL MUSEUM VOL. 73 long ago. So much of the evidence as to distribution and other matters is gone that anyone attempting to work out the story of these mollusks finds himself in the position of a person who tries to read an old, torn, and faded book with many of the pages missing. In Cuba the earlier collectors seemed to think that the different forms of Liguus were merely color variations that meant absolutely nothing, so they took but little pains to get complete sets of the varieties or to record definite localities for them, and they mostly lumped all their material together and labeled it “Achatina fasciata, Cuba.” If only complete collections of Liguwus could have been made in the island and in Florida before the destruction of the. forests began and the records of localities carefully kept I am satisfied that we could not only have deciphered their past history but by their aid we could have worked out in considerable detail the geology of both regions. It has been objected that I have given too many names to color varieties of the Floridian Liguus. While I do not attach any great importance to most of these, yet I am sure that they stand for differ- ent phases of development and distribution. It seems to me that it is better to have some kind of a name for a variety, such as moséeri or castaneus, than to have to say “the small form of crenatus with narrow spiral green lines ” or “the dark chestnut variety of fasciatus having a few yellowish markings on the spire.” A considerable number of the Ziguus, which Dr. H. A. Pilsbry and I have named, are, doubtless, fertile hybrids; but they, as well as the rest, help to tell the story of the past development, migration, and distribution of these snails. I am greatly indebted to Mr. Charles Mosier, former custodian of the Royal Palm State Park, the great forest of which swarms with Liguus. Mr. Mosier made careful studies of these snails for years in this and many other hammocks in Florida and has freely given me the result of his observations. The late Mr. John B. Henderson generously made it possible for me to visit the island of Haiti and to make a half dozen trips to various parts of Cuba, also to collect in places on the Florida Keys which were difficult of access. I am also under obligatons to Dr. Carlos de la Torre, of Habana, who allowed me to examine his great collection and make notes on it, also for many valuable specimens of Cuban Liguus. This paper is respectfully dedicated to Charles Mosier, friend, master woodman, naturalist, and companion on many collecting trips, and to whose knowledge and assistance I am greatly indebted for much of what appears in these pages. —— Se AkT, 20 FLORIDA TREE SNAILS—SIMPSON 3 THE GENUS LIGUUS IN FLORIDA The genus Liguus includes about a dozen species of large land snails inhabiting the northern and northwestern parts of South America, the island of Cuba, Haiti, Cozumel, off the eastern coast of Yucatan, and the more tropical part of Florida. The six species which inhabit the islands and Florida are brightly colored and strictly arboreal in habits, and they doubtless constitute a well-defined subgenus. Cuba is an old island, geologically speaking, and has doubtless been inhabited by Lzquus for a long time, possibly back into the Miocene. It is quite likely that the genus extended into Haiti at a time when the land was much more elevated than it is to-day and that island and Cuba were united. Although Florida is separated from Cuba by a deep channel 90 miles wide it seems certain that it has derived all its Ziguus stock from that island. At first one would naturally wonder how it was possible for these arboreal snails to cross such a body of ocean and become established on the new land. Both eggs and snails sink at once in water; no bird could possibly carry either across in its beak or on its body; neither could eggs or snails be transported through the air by hurricanes. The main current of the Gulf Stream which sweeps up past the western end of the great island turns to the west as it enters the Gulf of Mexico and slowly passes around its deep central basin. It is probable that under ordinary circumstances weeks would elapse before any floating object that passed into this great vortex through the Yucatan Channel would emerge and enter the Florida Strait, and in that time any Liguus which might be car- ried over this course attached to limbs or trunks of floating trees would be sure to be washed off and perish. It would seem, then, that every means of transportation for these snails was shut out and that it would be impossible for them to migrate in any way from the great island to Florida. But when we once fully understand all the condi- tions it appears as though nature had specially arranged everything not only to make this voyage and colonization possible but abso- lutely feasible and that there have been a considerable number of successful migrations and colonizations. The Ziguus never voluntarily leave the trees on which they live except at the time when they go down to lay their eggs or make the migrations across the country for the purpose of founding new colonies, which I shall later describe. During the dry season, from November to May, in Haiti, Cuba, and Florida, these snails throw out a gummy substance which hardens and attaches the shells so firmly to the trees on which they live that they will often break with- out letting go. At this period of aestivation, as it is called, the 4 PROCEEDINGS OF THE NATIONAL MUSEUM vou, 73 animal becomes dormant, much as do those of northern countries during hibernation, only the organs do not so completely cease to function. Before the beginning of the rainy season, sometimes as early as April, the snails dissolve or let go from this epiphragm, as the gummy substance is called, and become active. Sometimes in winter if there is rain and especially if the weather is warm they do this but make an epiphragm and become dormant if it becomes dry and cool again. They live, for the most part, on smooth trees, but sometimes they are found on those with rough bark, even live oaks, and in our State, only in its extreme lower part. Although almost wholly confined to the drier hammocks in Florida I have collected them in our brackish and fresh-water swamps, in the former on the tropical buttonwoods (Conocarpus) and on cypress trees in the latter. In very rare cases they have been seen on Caribbean pine trees at the edge of ham- mocks. Their food is the confervoid growth which is found on the bark and possibly the leaves of the trees which they inhabit. Some: times, especially during rain, I have seen them clinging to the under surfaces of leaves though this was not for the purpose of keeping dry. During the warm season they freely copulate, and as they are hermaphrodites it is probable that nearly all the indi- viduals become pregnant. In the warm or rainy season from May till November tremendous showers fall in northern Cuba and trees in stream valleys bearing colonies of Liguus may be washed out and swiftly borne down on the current of the mountain torrents and into the sea. It is probable that most of our Ligwus stock is derived from the northern slope of western Cuba, and such trees thrown into the ocean are caught by a small current that leaves the main stream which passes up the Yucatan Channel, and this smaller flow hugs close to the northwest shore of the great island. Farther on it joins the Gulf Stream after the latter has passed entirely around the deeper part of the Gulf of Mexico, the junction taking place in the Florida Strait, the whole great current moving to the east and northeast and sweeping against the lower part of Florida. This current in the strait moves onward ordinarily at the rate of from 2 to 4 miles an hour, much depending on the direction and force of the wind. In time of hurricanes with wind and heavy rain from the west or southwest the speed: of this ocean river would be greatly accelerated, and trees bearing Liguwus might easily be borne from northwest Cuba to the keys in 48 hours. It is probable that some of the snails which adhere to the trees on which they live may be carried high and dry on limbs during a trip of considerable duration and distance as they are making this passage, but even if they do become immersed for quite a length of time I feel sure no damage would come to them. At one time I puta a ar Ke a eee; i. a ee Se ART, 20 FLORIDA TREE SNAILS—SIMPSON 5 lot of over 30 active Liguus into a vessel of water, weighting them down with a piece of board so that all were kept completely sub- merged for 30 consecutive hours, and when released every one of them crawled away as though nothing had happened. All were clinging to the board when they were taken out. During the rainy season the wind in the Florida Strait blows for the most part from the southeast, and this drives any drifting mate- rial diagonally across the current toward the Florida Keys and the southeast mainland. A slight, irregular flow of water takes place along this part of the mainland and the keys in a direction opposite to that taken by the Gulf Stream. This slow movement keeps close in to the shore and would tend to carry drift to the south and west, even in some cases that which had passed for a considerable distance along the main current, thus helping to dis- tribute material carrying Liguus along our southeastern shores. The greater part of the cold Labrador current goes under the Gulf Stream above Miami, and it is mixed up in this vicinity, so that it sometimes drifts to the north and again to the south. It is quite probable that the feeble current running westerly among the keys is the last vestige of this cold river. Nearly all the inner shore line of the keys and the southeastern mainland of Florida is fringed with a growth of mangroves, but on the side fronting the open sea there are sandy or rocky beaches. The trees growing in “the mangrove ” usually stand in mud or very shallow water; and wherever they grow conditions are not favorable for the landing of tree snails because, so far as I have been able to observe, they never crawl over wet mud. However, during Septem- ber, October, and November we have unusually high tides all along our lower coast, then they sometimes reach an elevation of 2 or more feet above the ordinary tide. In many cases Liguus inhabit dead trees or limbs, and such might easily be broken up during their journey down the Cuban streams or while making the sea crossing before reaching Florida. I have no doubt that in some cases these snails reach our shores on comparatively small pieces of floating timber, and these could easily drift in among the mangrove roots, where they are not too crowded, during excessively high tides and be landed on what is ordinarily high and dry ground. . If such a landing happens to be made in the edge of a hammock, the snails could crawl off and become denizens of the United States without any formality. Of course, during hurricanes the sea may be driven over the highest land on our lower coast, and it is most likely that the greater part of the Liguus have been landed and established in Florida during such storms. At one time, since I resided here, the wind during one of these storms drove the sea entirely over Elliotts and Largo Keys and far out on the mainland of Dade County. 6 PROCEEDINGS OF THE NATIONAL MUSEUM vou. 73 There are authentic records of other even greater tidal waves in lower Florida. Without a doubt this process of enforced migration began as soon as the Ligwus spread over the island of Cuba, perhaps back in Miocene or Pliocene times, but it is probable that no region existed to the northward suitable for them to live in. Certainly if any colonies of these snails existed in any part of what is now Florida before the ‘Glacial epoch they were destroyed by its cold, for they are tropical} and can not stand hard freezing. A period of subsidence occurred for the State of Florida during early or middle Pleistocene, and nearly all the eastern part of it and all the lower end to just north of the Caloosahatchee River were car- ried below sea level. A great bed of oolitic material, the Key West limestone, was deposited over the area now occupied by the Lower Keys. Another very similar set of beds was laid down along the southeast part of what is now the mainland—the Miami limestone— and a third on the southwest coast called the Lossmans River lime- stone. A set of beds was formed in what is now the Everglades, at first doubtless marine, then brackish, and finally fresh water. It is probable that a coral reef, which afterwards became the Upper Florida Keys began to develop during this subsidence, and it reached from a short distance south of Cape Florida in a curved line to what is now known as Newfound Harbor Keys, just south of Ramrod Key, thus lapping over onto what is now the Lower Keys for a distance of perhaps’5 miles. This was followed by an elevation during which the land was probably raised to about the same height as at present. The great bed of Key West limestone became a single island, reaching from east of Johnson Keys to and including Key West and from the Gulf of Mexico on the north to the Florida Strait on the south. The sea began to form a low, broad shore elevation just within what is now the southeast coast of the mainland from somewhere near Fort Lauderdale to the south, then southwest, west, and again southwest to near Whitewater Bay. This consisted of low ridges which formed one after another as the sea retreated, leaving flats and shallow lagoons between, and in these marine mollusks and other animals lived and died undisturbed. Later this became a soft oolitice limestone. A great variety of seeds of tropical trees, shrubs, and plants was conmea by the Gulf Stream, a few possibly by migrating birds, and some lighter ones by the wind and landed on the large southwest island as soon as it was high and dry above the tide, uid at once a dense hammock growth was formed over much of the higher parts of it. Seeds of the Caribbean pine (Pinus caribaea) were probably blown across from Cuba and forests established on some of it and 0 AE OLR Natl 6h a ie eS ee a este ART. 20 FLORIDA TREE SNAILS—SIMPSON i on the rocky ridge of the lower mainland. Hammocks began to be developed in protected places on this ridge, and conditions were right for the establishment of colonies of Ziguws on the soil of the United States. We can not tell whether the first successful settlement of these snails was made on the great southwest key or southeast mainland, but quite likely on the former. A gravid specimen of a form close to our present graphicus clinging to the tree on which it had made its home was thrown up on this island, either by an unusually high tide or during a hurricane. It probably came from somewhere in the neighborhood of Cabanas, about 40 miles west of Habana, where Dr. Carlos de la Torre, the late John B. Henderson, George H. Clapp, and the writer found a form of Ziqguus in considerable num- bers that has nearly all the characters of that beautiful snail. It is, however, smaller, is a richer yellow, is not quite so porcellanous, and is a little less flamed on the spire. I have little doubt that our graphicus sprang from the same stock as the small Cuban form did. The newly arrived snail multiplied and probably spread over most of the drier parts of the island, for at least recognizable fragments of it has been found on most of the keys where there is high land. Certain forms among its progeny seem to have partially lost their coloring, and one of these which Say called solidus has a broad, faint, spiral yellowish band above the periphery and another on the base. A second somewhat thin variety which has delicate coloring Pilsbry has named solidulus. This is whitish or cream colored with one or two faint yellowish narrow, spiral lines at the periphery, another at the suture, and sometimes a broad one on the base. Some of the shells which are doubtless of this variety have two or more narrow greenish or bronzy spiral lines on the last whorl, and others scarcely show a trace of banding. A shell from Ramrod Key of the lower islands is almost pure white throughout, and a very solid form which I found on Big Pine Key is ivory white, with a single narrow, bronzy peripheral line and a decidedly truncated columella to which I gave the subspecific name of crassus. Mr. John B. Henderson had specimens in his collection from Key West which I refer to this. I feel sure that if a large collection of solidus could have been made from the Lower Keys it would have shown a considerable amount of variation in coloring. Reeve bestowed the name Achatina picta on a form of solidus which has a pale, ashy yellow ground pattern with longitudinal bluish smears and a double row of squarish brown spots at the periphery and sutures, and he gave Cuba as a locality for his shell. There is a specimen in the collection of the United States National Museum with the same locality given, and many years ago I re- 8 PROCEEDINGS OF THE NATIONAL MUSEUM VoL. 73 ceived a fine shell of this form in an exchange from A. G. Wetherby labeled “Achatina fasciata, Cuba.” I only know of two authentic Floridian specimens of this subspecies, one of which is in the col- lection of the Academy of Natural Sciencies of Philadelphia and was collected by Dr. H. A. Pilsbry on Big Pine Key when on a trip to the Lower Keys with the writer in 1907. The other is a somewhat worn shell, but showing the colors perfectly, that I found on the same island in the village of Big Pine a few years later. D’Orbigny figures a shell in his Atlas which he credits to Cuba and which no doubt came from that island which bears some resemblance to pictus, but it is not that. I found a very similar shell at Luis Lazo in western Cuba that I believe is a hybrid between some form of fasciatus and solidus. It seems a little strange that this snail (pictus) should be reported from Cuba at least three times if it was never found there. I am very much inclined to think that it has actually been collected in that island as well as in Florida and that it crossed the strait and landed on Big Pine Key since the dismemberment of the large island, and so short a time ago that it has not changed any of its characters in the least. There is comparatively deep and open water close up to the big hammock in which both the specimens of pictus were found on this island. It is probable that not more than two forms of Ziquwus ever landed on the great island, although it is nearer to Cuba than any other part of Florida, and it certainly was clothed with hammock at an early date. The reason for this paucity is perhaps the fact that the Lower Keys lie almost directly across the Florida Strait from the western part of Cuba, and the strong eastward-flowing Gulf Stream would naturally carry any drifting material past them. No sooner was the great western island, the original of the present Lower Keys, formed than the forces of nature began to destroy it. Every winter several storms called “northers” sweep down from the northwest across the Gulf of Mexico, the wind often blowing to 30 or even 50 miles an hour, throwing the water of this great inland sea with tremendous force against the land of this region and cut- ting into it all along its entire northern exposure. As a consequence its northern outline is far more ragged than the southern. Not only this but the water was driven with such force that it was crowded completely through the porous foundation of the island from north- northwest to south southeast so that it came out in strong streams on the south side into the Florida Strait. This water soon formed channels through the loosely compacted rock in the eastern part of the great island, and these were enlarged by scouring and the action of the carbon dioxide in it. Later the weakened roofs fell in so that open streams reached entirely across, for the most part in exactly the _ ART. 20 FLORIDA TREE SNAILS—SIMPSON 9 direction of the wind during these storms. It is probable that the rock in the western end of this former island is of a solider type than that of its eastern part, hence the water was not driven through to any considerable extent but it entered everywhere by seepage and has eroded and dissolved it into a most complicated archipelago. I feel certain that a Liguus, the forerunner of the forms we have found on these Lower Keys, reached the big island, became estab- lished, and spread over the entire region and that it broke up into several varieties which pretty well occupied the whole area before the dismemberment took place. A thin, glassy, inflated form of crenatus was cast ashore on the elevated hammock near Fort Lauderdale, no doubt an early migrant from Cuba to our shores. Later the sea threw up a great sandy bar or bank a few miles to the east of this forest, and the intervening space has been filled with a mangrove swamp. The Liguus has been carried down and established on this bank where I found it at the New River mouth and for a couple of miles to the northward. This colony of sepientrionalis, as Pilsbry has aptly called it, is the farthest north of any in the United States. The form has spread south of New River, it has crossed Little River, and in a hammock just below the mouth of that stream I found nearly pure individuals. Near Arch Creek it has hybridized with one of the forms of fascia- tus, the shells having exactly the shape, markings, and texture of septentrionalis but a pink axial region. I have shells from the town of Jamaica in Cuba that are very close to this. It is probable that at least four forms of Ziquus that drifted from Cuba became established at the great Miami hammock. According to the Report of the State Geological Survey some of this land is elevated to a height of 30 feet above sea level, and it is most likely the highest on the southeast coast. It was probably above the sea and was dry land before any other part of this general region and has had a longer time in which it could be colonized by snails. These forms I have called miamiensis, livingstoni, mosieri, and eburneus, the first two belonging to fasciatus, the latter two to crenatus, and none of them inhabits the Upper Keys. JZ. miamiensis is doubtless a hybrid between a form near castaneozonatus and probably a typical fasciatus, the earlier whorls having a broad, more or less broken brown band and the last whorl a pattern of spiral lines, the two com- ing together abruptly. I have shells from Cuba which show similar characters. This form ranges from Ojus to a considerable distance south of Miami. The form /évingstoni is a small fasciatus usually lacking the color pattern on the upper spire, but I have a shell from Luis Lazo, in Pinar del Rio, that is exceedingly close to it in which the color pattern has almost vanished. It is found from Fort _ Lauderdale south to Long Pine Key. I have large shells of mosiert 97556—29——_2 10 PROCEEDINGS OF THE NATIONAL MUSEUM VOL. 73. from Miami that are almost identical with a specimen from Mount Guajaibon in northwestern Cuba, they being larger than the latter. It is found over nearly the entire rocky ridge of southeast Florida. Hburneus is a fine subspecies usually large and solid, somewhat porcellanous, and nearly always pure white throughout. It is almost perfectly mimicked by a shell I collected in western Cuba. It does not extend north of the Miami River but goes south to Long Pine Key in the Everglades. Thirteen subspecies of Liguus inhabit this great Miami hammock— eburneus, mosieri, luteus, cingulatus, marmoratus, livingstoni, or- natus, miamiensis, elegans, roseatus, testudineus, castaneus, and castaneozonatus—a considerably greater variety than is found in any other Floridian locality. Zuteuws occurs from Dania to Long Pine Key and in the former region as well as at Vaca is rather solid and bright colored, but in the lower Dade is thinner and lighter. It is the first form to be colonized in the incipient hammocks, and at almost every sink that has a live oak and a little scrub it will be found, while castaneozonatus is the second migrant. Marmoratus is a very variable snail with a wide distribution, and specimens found by me in the Miami hammock are very close to those from Key Vaca of Chokoloskee. A form of this has recently been found at Pinecrest, 35 miles west of Miami in the Everglades. I have no doubt but that testudineus, castaneus, and versicolor are hybrids between this and forms of fasciatus. The subspecies elegans was discovered on a small key in the southern Everglades, which was occupied exclusively by it and voseatus, and it is close to shells I collected in northwestern Cuba. It occurs occasionally along the rock ridge as far north as Arch Creek, usually in a pure form and always having the peculiar brownish flecks on the upper spire. Two peculiar forms of crenatus are confined to the southwest main- land, where they have a limited distribution—lossmanicus and capen- sis. The former is dull colored, whitish, sometimes with narrow spiral greenish lines, and the whorls are usually well rounded. The form capensis is elongated, and solid with flattened whorls occu- pying Northwest Cape exclusively and found rarely elsewhere in the cape region. It is probable that both are recent arrivals from Cuba, and they may have drifted through the Bahia Honda Channel. The fact of their limited distribution is evidence in the direction of their recent arrival. , It is probable that the land of the Cape Sable region is very recent, one of the latest additions to our territory. The area covered by the three points, Northwest, Middle, and East Capes, and the region for some distance back of them is sandy, but it was built up over an old mangrove swamp. Back of this sandy land there is a brackish swamp, and still farther back a considerable prairie, while at the ie i oa RE Da eet sini Na ee es arr, 20 FLORIDA TREE SNAILS—SIMPSON i juncture of the two latter is a line of hammocks. The entire inner country lies only just above high tide, and whenever during a hurri- cane the wind becomes westerly the water from the Gulf of Mex- ico is driven in to Florida Bay and the region to the north of Sable with great force. As the chain of Florida Keys acts as a barrier against this water passing to the eastward it is dammed up and forced over the south shore, the Sable and southwest coast areas, sometimes covering the highest land to a depth of several feet. Before there were any breaks in this upper chain of islands the dam was almost complete, and doubtless the depth of water became much deeper than at present. As there is a great deal of swamp in this territory, Ziguus can not progress from hammock to hammock as they do through the high pine woods for, as I have said, they never crawl over wet mud. There are, however, both Liguus and Oaystyla distributed abundantly throughout this whole area wherever there are suitable hammocks, but they are scattered absolutely hit and miss. At Northwest Cape only capensis is found, while in a little hammock just north of Middle Cape Dr. Edward Mercer and I got lossmanicus, castaneozonatus, with hybrids between the two, and a single, fine marmoratus. At Middle and East Capes there is quite a variety of forms, and this is true of the line of hammocks between the prairie and the brackish swamp. In one of these would be found an Oxystyla, in the next perhaps a single form of Liguus, in a third absolutely nothing, and in the next both Liguus and Oxystyla. Yet in some cases these bits of woods are only a few rods apart. ‘The explanation of this is, I believe, that trees and limbs with snails are torn off by the fury of the hurricanes in this region, probably carried along by the high water and landed on some hammock. One such piece of forest might receive a Liguus, another an Owystyla, another both, and a fourth none. The ground immediately in front of this row of hammocks is always swampy; that in the rear is wet in the rainy season and dry in the cool part of the year. The snails can not crawl over it in summer, and in winter they are fast to the trees. Some of the cape forms may have arrived during the first elevation, but most came at the second. Since the above was written a great highway, the Tamiami Trail, has been opened from Miami west through the Everglades, and at a place on this called Pinecrest, midway across the State and 40 miles from its south end, Mr. Joseph Farnham and Mr. Richard Deckert have found great numbers of Liguus. Many of these are hybrids, but they are generally rather closely related to the forms found on Long Pine Key in the south part of the Glades. It is probable that during one of the late minor subsidences this region was somewhat lower than now and that a strong tidal wave may have carried the 12 PROCEEDINGS OF THE NATIONAL MUSEUM VoL. 73 progenitors of these forms from Long Pine Key and landed them here. It is probable that at the time of the first elevation the reef of which the Upper Keys was formed was considerably developed and raised to the surface of the sea or even somewhat above it in places. but I doubt if it was clothed with hammock forest or inhabited by iand snails. The flora is composed of tropical plants and has been almost wholly derived from Cuba, but it is comparatively poor, hav- ing scarcely half as many such species as the Lower Keys and noth- ing like as many as the southeast mainland, and this fact would prove that it is much younger than either of the last-mentioned areas. Then there was a second period ef subsidence during which the land went down until the sea washed into the present bluffs at Coco- nut Grove and Miami, eroding them in places into fantastic forms. At Little River the depression amounted to some 7 feet, while at Big Pine Key it was about 3 feet. An elevation followed during which the land was raised perhaps to a few feet higher than it is at present.. The Upper Keys were elevated until they probably formed an unbroken chain or single island. ‘The sea attacked the reef, tearing it to pieces, scattering the fragments and grinding them into sand, then cementing the wreck firmly together and building it up into a solid, continuous island. This great barrier for a long time prevented floating material from landing on the mainland back of it, but it became more or less covered with hammock forest and provided a home for such Liguus as drifted in from Cuba. Several forms, either roseatus, castaneo- zonatus, vacaensis, lineolatus, and perhaps marmoratus or their immediate progenitors, drifted in from Cuba and became established long enough ago that they spread practically throughout the entire length of the long, curving island. This was the heyday of the Upper Keys, the period of their greatest development and glory. A broad, irregular land bridge reached from the region of the Mate- cumbes across to the mainland from Joe Kemps Key for 12 or 15 miles to the eastward. In places this was swamp or shallow lagoon, but there was continuous hammock-covered land, and over this migrated a considerable variety of tropical vegetation from the keys; in fact most of the flora of the south and southwest coasts of the mainland was derived from the Upper Keys in this way. This flora differs considerably from that of the rocky ridge along the southeastern mainland. It is nearly all tropical, while that of the ridge is mixed—West Indian and warm temperate—the latter part being derived from the northward. The reason. for this is that the Everglades stretches, and since the first elevation has stretched, in a broad, unbroken area from near Whitewater Bay eastward and northeastward to Cutler, forming an effectual barrier against the — arr, 20 FLORIDA TREE SNAILS—SIMPSON 13 passage of dry-land plants and to a considerable extent the migra- tation of snails. Ziguus could only cross this great marsh during a time when the sea was driven in over it by hurricanes and then on floating timber. During the time of the second elevation there was probably an almost continuous hammock along much of the south shore of the mainland; in fact a considerable amount still remains. The Liguus castaneozonatus, vacaensis, lineolatus, roseatus, and marmoratus crossed over from the keys and became established on the south and southwest parts of the mainland. By and by the sea began to seri- ously gnaw away at the great coral island. High tides occur on one side at different times from what they do on the other, and I have seen at low tide on one side of one of these coral keys many streams of water passing through, from a tiny trickle to those of the size of small rivers. Soon the carbon dioxide in the water ate the rock away until the weakened roof fell and later an open channel formed which cut the island in two. This action, aided by the fury of the sea when driven in by storms, has been continued until this once greatly elongated island has been cut into more than 30 islets and islands, each high enough to bear upland hammock, besides a vast number that are only clothed with mangroves. Of course when a key was cut in two it checked the passage of the Ligwus from one island to the other, and we can form some idea of the time of arrival of the snails by their distribution. The subspecies matecumbiensis did not arrive until Upper Matecumbe was cut off from the keys to the north and south of it, for we have never found it on any other of these islands. But it came in time to cross the land bridge, for we find it one of the most abundant of the tree snails of the Flamingo region. Oaystyla floridensis, which inhabits the keys from Largo southwestward, crossed this bridge and is very abundant in the Cape Sable area; but QO. veses, which inhabits the keys from Key West to Vaca, apparently has not passed any farther up them nor crossed to the mainland. A beautiful form, swberenatus, was common in the center of the upper chain from probably Largo to Grassy Keys. The subspecies e/liottensis has its metropolis on the island for which it is named but has been found on Old Rhodes and possibly upper Largo. It was a late arrival and probably was kept from an extensive distribution because of the cross-channels. Four subspecies of Zigwus which inhabit almost the entire range of the Upper Keys—castaneozonatus, roseatus, marmoratus, and Juteus—are also found on the mainland on the rocky southeastern ridge, although a tract of Everglades and a series of shallow sounds separate the two regions. During time of hurricanes when the sea is driven against southeast Florida with tremendous force the entire chain of islands may be overflowed and water driven in until it covers 14 PROCEEDINGS OF THE NATIONAL MUSEUM VOL. 73 the lower parts of the rocky mainland ridge. At such times many limbs and trees on these keys are torn off or uprooted and with their load of Liguwus may be easily carried across and landed on the higher mainland. There it is easy for the snails to crawl off and get into the hammocks, where they are soon established. The last general earth movement in lower Florida was a slight sub- sidence which may be continuing yet in places and of which there is abundant and widespread evidence. In the Shark River region hundreds of acres of what must have been land are to-day under a very shallow sheet of water in which littoral forest grows thickly and in the same general area and in Biscayne Bay are dead trunks of large mangroves that probably started growth on the land, but have been killed by the slight subsidence. The marly ham- mocks along the south coast are tumbling into the sea and the general chain of keys is being rapidly destroyed by the encroaching ocean. The great land bridge has been reduced to low islets and shoals. This subsistence has had a marked effect on the Liguus of the upper islands. The distribution of the Ziguus on the Upper Keys is amazing, and for a long time I was utterly unable to understand how it came about. Seven subspecies—castaneozonatus, roseatus, lineolatus, vacaensis, marmoratus, luteus, and subcrenatus—are found living on the upper and lower ends of the chain but all save lineolatus and subcrenatus are entirely absent from the central part of it. The subspecies sub- crenatus now inhabits Lower Matecumbe and has probably lived in the islands to the southwest as far as Grassy, perhaps to Vaca, while castaneozonatus and luteus occupy the northern end of Upper Mate- cumbe and the islands to the northwest. Long Key, farther down in the chain though a large island with some dry, hammock-covered land, seems to be absolutely lacking in Liguus. Yet on Lower Matecumbe, Lignumvitae, the extreme lower end of Upper Mate- cumbe and Indian Key, lying in the exact region where five of the forms I have mentioned are absent, we find no less than four tolerably well-defined subspecies of solidus, a species only found elsewhere in the United States on the Lower Keys. Why are these seven forms present in the ends of the chain and all but two absent from the center, and why should they be replaced here by a species of the lower islands?) Why is Long Key without Ziguus when it seems to be perfectly adapted for their growth? I feel sure that during the period of greatest elevation the entire set of the Upper Keys from near Cape Florida to and including the Vaca group was one great island, so high and well clothed with hammock that the seven forms I have mentioned, which were early arrivals, became distributed throughout the whole. Then at the time of the last general subsidence the middle of this chain went down just ART, 20 FLORIDA TREE SNAILS—-SIMPSON 15 enough to change the dry-land hammock into mangrove swamp, and it needed only a slight depression to do this. Five of the forms— castaneozonatus, roseatus, luteus, marmoratus, and vacaensis—were drowned out on Lower Matecumbe, but swhcrenatus and lineolatus survived, probably on a bit of hammock at the upper end of the island,. as this is a little higher than the rest of it. The snails were drowned on the southwest end of Upper Matecumbe, which is slightly lower than its northeastern part and is now nearly all swamp with only here and there dry land. Long Key was carried down just enough to drown out its Ziguus. It is probable that a very slight upward oscil- lation of this region since has made the latter island dry enough for high hammock growth as a foot of elevation would change swamp to dry land. But no Liguwus have landed on it since. During this last slight elevation of the mid chain a gravid speci- men of the variety graphicus was landed (on drift from one of the Lower Keys), probably on Lower Matecumbe, where it quickly became established and soon began to vary. Two of its forms, lignumvitae and simpsoni are now living on the near-by Lignumvitae Key, a mere dot of land just inside the regular line of the chain and separated from Lower Matecumbe by a mud flat a little more than a half mile wide. Another form which I have called delicatus is . found on Lower Matecumbe and on a smail bit of hammock in the swamp of the lower end of the upper island; also it was found by Wurdemann long ago on Indian Key, another mere point about half a mile outside the line of the regular chain and just off the lower key. I am inclined to believe that these outlying Liguus reached Lignumvitae, Upper Matecumbe, and Indian Keys by drift- ing during storms. In a part of Lower Matecumbe, where cross currents have eroded and lowered the surface until it is no longer dry land, there is a small hammock in which I found a fourth form of solidus. The shells are large, thin, of a glassy rather than porcellanous texture, the ground color being a pale yellowish gray. They have the longi- tudinal smears and a brownish peripheral band, and along this and the sutures is a double row of squarish spots, while the entire axial region is white. Because of its resemblance to pictus I have called it pseudopictus. Apparently it has rather recently developed in this isolated hammock, and I found a few partly intermediate shells. I have no doubt that it is the latest form of the genus to develop in Florida. The species solidus seems to vary very easily and apparently has broken into many forms. I found several interesting variations or perhaps hybrids of what may be this species and some other near Cabanas, Cuba, and I feel sure that before our hammocks were de- stroyed in Florida there were several variations. I found shells on 16 PROCEEDINGS OF THE NATIONAL MUSEUM VOL. 73 Lignumvitae Key which are so close to graphicus that it would puzzle an expert to tell them apart, but there were intermediates between them and lignumvitae. While the porcellanous structure is one of the chief characters of what we call solidus, it is almost absent in pseudopictus, and the axial region in the species may vary from almost dark purple to pure white. It may seem that I have recorded a large number of minor eleva- tions and depressions, mere oscillations, in lower Florida since the first appearance of the land, and I feel sure that there have been, as for most of them there is still in existence evidence in the form of raised beaches, eroded bluffs, or sunken forests. Samuel San- ford corroborates this statement and says that in places they have been rapid enough to be proved by human records.* Almost all of lower Florida is very flat, and it rises but slightly above sea level. The same authority I have just quoted remarks, on page 189 of this second report, that “A difference of 2 feet in water level means the difference between shallow lake and dry land for hundreds of square miles.” So also in the immediate vicinity of the sea a sub- sidence of less than a foot might mean mangrove swamp and a corre- sponding elevation land covered with Ligwus-bearing hammock. But the vertical movements were only slight from the first appear- ance of land in this region until now. At no time has there been a subsidence so great that the large lower island or the rocky mainland has been drowned out nor an elevation sufficient to make dry land of the bays along the southeast coast of the mainland, for in that case the warm temperate flora of the rocky ridge would have crossed over to the Upper Keys, and had there been a drown out the distribution of our plants and Liguus would have been very different now from what they are. Since the first great Pleistocene elevation I do not believe that the change of level has amounted to 15 feet. The colonization of Zzguus on the mainland of Florida came about under quite different circumstances from those of Cuba or most of our keys. In both of the latter areas there was a practically continu- ous growth of tropical forest which was exactly fitted for a home for these snails. They are hermaphrodites, and during their period of activity it is probable that most of them copulate and become gravid. At this time a considerable number of individuals come down from the trees on which they make their homes, and instead of depositing their eggs in the ground near them they obey a call to form a new colony. They travel directly away, probably going in a reasonably straight direction, and whenever they are satisfied they stop, dig out a shallow excavation among the leaves and trash, lay their eggs, cover them, and find a near-by tree for a new home. There is no *Second Report Florida Geo]ogical Survey, p. 180. ws ART, 20 FLORIDA TREE SNAILS—SIMPSON 17 special risk or hardship about this, for the snail never leaves the sheltering hammock; it can stop and deposit eggs whenever it likes; there is an abundance of food all around it. A large part of the lower Florida mainland, however, is covered. with an open growth of Caribbean pines, and such a region is hostile to the Liguus. There is practically no shelter; there is little or no food; in places where the floor of the forest is sandy it is impossible for them to progress unless they can crawl from plant to plant above it. Only here and there at intervals is there a hammock, and although the snails have a certain amount of vision yet it is probably impossible for them to distinguish one from the pineland. Yet these wanderers obey the instinct for founding colonies which was begotten during the thousands of generations that preceeded them and boldly strike out into the pine woods in order that they may reach another hammock and plant their race in a new territory. At my home I once observed one of these snails which had doubt- less left my near-by hammock and was attempting to work its way out into the open pine woods. We had rains for several days in succession, and during this time it moved away from the hammock at the rate of about 25 feet a day, and I kept close watch and set stakes to mark its onward passage. This one and others I have noticed crawled along the stems of small shrubs or grass, over fallen logs, or anything that made a firm pathway. When the weather was dry and the sun shone the Zigwus attached itself to something and did not attempt to go on until it rained again. I kept track of this specimen for several days, noting that it persistently worked away from my hammock and out into the uncharted pine woods. I have on a few occasions found them in the open forest at a long distance from any hammock and in rather numerous cases the dead shells which probably testified to the disaster in the way of exhaustion that finally overtook them. The migration is of course absolutely hit and miss, as no snail can know when it starts on such a journey that a hammock is in front of it. No doubt in cases where a number of hammocks are scattered through an area of pine forest a crawling Liguus might pass by one and, continuing on, enter another not far away. So in any general region a hammock may have a form that is absent from another that is only a few rods away; one may have several subspecies and another near by only two or three, even a single one, or rarely none at all. One of our leading botanists believes that the hammocks formerly covered most of the lower part of Florida and that the pine trees are late immigrants that are spreading and taking possession of the country, but if this were true we should find the remaining hammock portion occupied by practically the same forms of Liguus throughout. 97556—29—_3 18 PROCEEDINGS OF THE NATIONAL MUSEUM vou. 73 There is another way by which, to a limited extent, it is probable that the Ziguus may pass from one hammock to another and become established. Mr. Charles Mosier informs me that he has on several occasions seen crows flying with these snails in their beaks. It is a cw <= wee - : ~~ = ~ > ae Se Fic. 1.—DIAGRAM ILLUSTRATING THE DISTRIBUTION AND MIGRATION OF LIGUUS IN FLoripaA. THE OPEN SPACES ARE PINE WOODS; THOSH INCLOSED IN LINES, HAM- MOCKS. DOTTED LINES SHOW MIGRATION TO AND FROM HAMMOCKS. SOMB BITS OF FOREST ARH REACHED BY SEVERAL; OTHERS ARE ENTIRELY MISSED well-known fact that these and other large birds frequently prey on Liguus and other arboreal snails, and I have seen the ground in hammocks strewed with their broken shells. It is not unlikely that these birds may occasionally pick from a tree one of these snails and ART. 20 FLORIDA TREE SNAILS—SIMPSON 19 carry it away with it, just as they carry away large seeds. ‘Lhe Liguus when disturbed throw out a quantity of a mucilaginous sub- stance which is very slippery, and in such a case it might easily happen that the snail would slip and fall to the ground during the flight of the bird. If it should fall in a hammock in an uninjured condition there would be nothing to hinder it from making its home there, and if gravid laying its eggs and establishing a colony. Such distribution would be essentially the same as if the snails crawled through the pine forest. As the changes of land level and the movements of the Liguus have been somewhat complicated I will briefly recapitulate. These snails in Cuba are closely related to our forms and without a doubt are their ancestors, often scarcely removed varietally. I am sure that the close resemblance of a number of forms from the great island to those of our State are not accidental but that they indicate very close relationship. I have actually seen snails there that are very much like graphicus, livingstoni, miamiensis, septentrionalis, elegans, subcrenatus, eburneus, castaneozonatus, and mosiert, and I have no doubt that there are other contiguous forms. Trees in which these Cuban snails grew were washed out, carried with their living loads down the torrents into the sea and swept by current or the force of hurricanes along the Florida Strait; the southeast, winds of that region would bear them on to our lower shores where, because of hurricane tidal waves or very high autumn tides, they would be landed high and dry on our shores where they could form colonies. There was a great subsidence of Florida during early or mid- Pleistocene time, and the region now occupied by our Lower Keys and the southeastern part of our mainland received a deposit of oolitic limy beds. A long, curving coral reef began to grow along the south and southeast part of the State from near Cape Florida to the south side of Ramrod Key. During the elevation that fol- lowed, the sea swept this oolitic material up along our southeast coast and formed a series of low ridges as the shore retreated and when this land was high enough it became covered with Caribbean pines whose seeds were wind blown from Cuba. A single great island was raised covering the area of the present Lower Keys and seeds from west Cuba, for the most part, were borne in on the Gulf Stream and the large island partly planted with hammock growth. Two large hammocks were formed on what was then the open coast, one at Fort Lauderdale, the other at Miaini. A gravid snail much like our Liguus solidus graphicus drifted in from near Cabanas, Cyba, was landed on the lower island and be- came established in the hammock, spreading over the greater part of the area and breaking into several varieties before the land was 20 PROCEEDINGS OF THE NATIONAL MUSEUM VOL. 73 cut up into islets. One form of crenatus landed at Lauderdale and spread southward; several came to the great Miami forest and, as the land was raised and developed, hammocks spread southward and southwestward along the low, rocky ridge. It is doubtful whether the reef became high enough for hammock and snails to develop on it at this time, but the sea no doubt attacked and spread it. During a subsidence following, the coral reef probably grew up as fast as the land went down. The water of the Gulf of Mexico driven by furious northers began to erode the north shore of the great island and was driven in places entirely through it, finally cutting up a considerable part of it into islands running exactly in the direction of the wind. Sea water entered what was probably solider rock at the west end and ate it into a complicated archipelago. Liguus solidus pictus, found only on Big Pine Key, is perhaps a recent ar- rivel coming long after the great island was dismembered. When the land was elevated again it reached a higher level than now, and as it rose the sea kept tearing at the reef and consolidating it until at the time of highest land it was no doubt a single long curving island with continuous hammock and a number of Liguus which came from Cuba were thrown on it and had time to spread throughout its length. At this time there was a broad land bridge across from the Matacumbes to the south mainland. The flora of the Upper Keys is tropical but meager in species, and this would prove that the chain was much younger than the southeast mainland or the Lower Keys, which have a far richer tropical flora. Many species of these West Indian plants crossea on the con- tinuous hammock of the old bridge and now constitute almost the entire flora of the south shore and southwest coast. Through this hammock Liguus marmoratus, vacaensis, roseatus, lineolatus, cas- taneozonatus, and no doubt Oxystyla floridensis crossed to the main- land and became established there, perhaps during the time when there was the highest land elevation. MMatecwmbiensis landed on the key it was named for after the dismemberment of the Upper Keys but before the destruction of the bridge, and crossed. Dry-land connection between the south shore and the rocky ridge was pre- vented by the Everglades which stretched from Cutler to White- water Bay and had done so from the first. Several forms belonging to the Upper Keys were carried to the southeastern mainland across the bays by hurricane agency. The Cape Sable region is very recent, the capes being sand washed up on an old mangrove swamp. The sea is thrown in upon that region by hurricanes with terrific force, and as the water is driven into Florida Bay and is prevented from flowing south by the keys it overflows the Sable and southern shore regions and Liguus and Owystyles are distributed hit and miss. ART, 20 FLORIDA TREE SNAILS—SIMPSON ral Capensis and lossmanicus may be recent arrivals through the Bahia Honda Channel. Seven subspecies of Ziguwus inhabit both upper and lower ends of the upper chain of keys but all save subcrenatus and lineolatus are totally missing in the center of the chain, and they are replaced by forms of solidus, which appear to have developed from graphicus. Long Key, which has good hammock, seems to have no Liguts. During the last general subsidence the Upper Keys went down enough that all the Zigwus in the center of the chain save subcrenatus and lineolatus were drowned out, and these subspecies probably survived on a bit of dry land at the northeast and higher end of Lower Matecumbe. All the snails of Long Key were drowned and no others have reached it since. During a local but slight elevation of the center of these keys a gravid graphicus landed on the first-named island, where it multiplied and broke into variations, some of which drifted to some of the near-by islets. In Cuba and the Florida Keys the Ziguus have for ages occa- sionally come down from the trees when gravid and, obeying an instinct for founding new colonies, have wandered off into the forest, stopping and making a slight excavation where they lay their eggs and fulfill their mission. But in the southeastern part of our State the hammocks are scattered throughout the pineland, which is hostile to the snails. However, they obey this call that has been inherited from thousands of generations that have lived under different en- vironment and strike out into the pine woods in search of hammocks. No doubt most of them miss finding a hammock and perish; others find one and create a colony which continues the race in a different locality. Although there have been several oscillations since the great sub- sidence of early or middle Pleistocene, yet they have been but slight and in all were never great enough to drown out the snails or high- land vegetation on the greatest elevations or to make dry land of the bays on the southeast coast of our State. Notwithstanding the fact that all that I have narrated has taken place since early Pleistocene, the briefest bit of geological time, yet I feel sure that long ages were required for the development of lower Florida and to people it with Zigwus. There was much migration, much hybridizing, the development of new forms, and time was needed for them to adapt themselves to their environment. The whole must have taken many thousands of years. The death knell of these beautiful snails in Florida has been sounded, and it will be but a few years until all are gone, save it may be in the great Royal palm hammock which is a State reserva- tion. Most of the small hammocks have been destroyed, and in others still standing the snails are fading away before man. In my 29 PROCEEDINGS OF THE NATIONAL MUSEUM VOL. 73 bit of forest, into which I have carried many specimens from else- where, the birds have attacked them, and now scarcely one remains. I do not believe that it is possible to make a key that will enable anyone to identify all the material in any considerable collection of Floridian Liguus. Without a doubt most of the forms of fasciatus and crenatus freely hybridize under favorable conditions, with the result that there will be found a certain number of nondescripts that can not be placed anywhere, while some that are of purer strain vary so greatly that it is difficult to classify them. Still I believe that by far the greater part of our material may be referred to some of the names in this descriptive list. Many of the specimens in our collections have been killed in boil- ing water, which causes some of the colors of the shell to fade or otherwise change considerably. Under such treatment green is almost certain to become bronzy or dirty grayish. In some cases green lines become abraded even during the life of the animal, or they may change color. The shells of soédus and its varieties are usually decidedly porcel- lanous and brilliantly polished, while those of the other two species are less so, but pseudopictus is comparatively glassy in some cases but lacking this soft, translucent texture, and on the other hand eburneus is occasionally quite porcellanous. KEY TO THD FLORIDA LIGUUS Apex, columellar region or both, pink, purplish, or violet. Shell decidedly porcellanous. Bluish or purplish axial smears present, never reduced to a mere line. Shell rather thin. With a double sutural and peripheral row of brown spots_pictus. Without a double sutural and peripheral row of brown spots. With a brown peripheral and green spiral line or band. lignumvitae. Shell solid (green spiral lines absent) _________-_________ graphicus. Bluish or purplish axial smears absent or if present reduced to a mere line. Peripheral white zone with a central brown line______ delicatus. Peripheral white zone without a central brown line___simpsoni. Shel] not porcellanous. Shell white, with broad supraperipheral and basal spiral bands. Spiral bands) yellowyor onances 2 oabste bale ei ee roseatus. Spiral bands not yellow or orange. nS} ov cea el oe WaVG USI oVEISS tala NSN SN Ee ee castaneozonatus. Spiral bands not chestnut. Spiral bands broken into alternate light and dark axial SCrCQks 45 29) Sie Pek Sead Oey See, alternatus. Shell white, greenish, or smoky white, with few to many green spiral lines which are absent on the periphery_________________ livingstoni. OE OF GO ie is = Ss ART, 20 FLORIDA TREE SNAILS—SIMPSON 23 Shell not white. Shell flesh colored or pinkish. With a reddish peripheral line and a few brown spots on the REPO T ee SPOT eee ee eee ee ee ee ee elegans. Without a reddish peripheral line and a few brown spots on the upper spire. A few dark spiral lines may be present or absent_lineolatus. Shell not flesh colored or pinkish. Shell pale yellow to orange, or orange brown (usually with a fewsdank=spiraighnes))) wee oes Wart err Pete ied ornatus. Shell not pale yellow to orange, or orange brown. Shel] dark, irregularly marked with green. Dark spiral lines conspicuous on last whorl. testudineus. Dark spiral lines not conspicuous but absent or only faintly indicated. Shell with axial yellowish flames______ castaneus. Shell without axial yellowish flames. Shell variegated, brown, smoky yellow, or DU Spee a es versicolor. Apex, columellar region, or both not pink, purplish, or violet. Shell entirely white. Shell porcellanous. Ground color ivory white with or without a dark peripheral line. , crassus. Ground color not ivory white with or without a dark peripheral line, Ground color white, cream, or straw colored. Shelli with a broad basal and supraperipheral pale yellow band. solidus. Shell without a broad basal and supraperipheral pale yellow band. Shell with two narrow yellow peripheral bands and one at they Supure 225 eS aE AS eee solidulus. Shell not especially porcellanous. Shell of pure white, ivory white, or whitish ground color. Spiral lines greenish or bronzy. Whorls well rounded. Whorls slightly shouldered at summit_____-______ vacaensis. Whorls not slightly shouldered at summit. Whorls short and subsolid, dull colored (often with a ledge within its aperture)_—2_--2_ 2 _ = lossmanicus. Whorls not short or dull colored (last whorl often DROWAVAORL ST CENnI Shy) eee ae = ee eee mosieri, Whorls but slightly rounded. Shell with the columella curved__________ matecumbiensis. Shell with the columella not curved. Se) ES Ui el Ae 2s Se ee Be capensis. Shell not solid. Shell thin. Shoulder generally angulated (not ‘small). septentrionalis. Shoulder not generally angulated (small). elliottensis. 24. PROCEEDINGS OF THE NATIONAL MUSEUM VoL. 73 Spiral lines not greenish or bronzy. Spiral lines broad,, yellow Le force eee aie cingulatus. Spiral lines not yellow. Spiral lines reduced to mere indication of dark lines on baseiate eta de neti sel hing a i et eburneus. Shell not of pure white, ivory white, or whitish ground color. Ground color yellow to orange or greenish yellow or ashy. Double row of brown spots present at the suture__pseudopictus, Double row of brown spots not present at the suture____luteus. Ground color not yellow to orange or greenish yellow or ashy. Ground color dark with yellowish axial markins___marmoratus. DESCRIPTION OF GENERA AND SPECIES Genus LIGUUS Montfort All material referred to which was obtained by others or which has been seen in other collections is duly credited; the rest was collected by the author. I have not given synonymy partly because I do not have access to the necessary literature. Shell imperforate, oblong to ovate-conic, with simple, usually thin, unexpanded lip and obtuse, vertically wrinkled or smooth nepionic whorls which are more or less rounded; columella straight or twisted-truncate at base; color white, yellow, brown, orange, green, or bronzy. LIGUUS SOLIDUS Say Plate 1, fig. 1 1825. Achatina solida Say, Journ. Acad. Sci. Philadelphia, vol. 5, p. 122. Shell thin to solid, generally elongated, usually with rather flat- tened whorls; color white, cream, or yellow, sometimes variously painted; brilliantly polished and generally having a decidedly por- cellanous texture; axial region white, pink, or purplish. Western part of Cuba; Lower Florida Keys; central islands of the Upper Florida Keys. The only character by which this species can always be separated from fasciatus and crenatus is the brilliant porcellanous texture of the shell, and certain specimens of pseudopictus are somewhat want- ing in this, being almost glassy. Usually the color does not become darker toward the lip, but in a few solidulus which I have obtained on Stock Island near Key West it does. And some of these have narrow, spiral, bronzy lines on the last whorl exactly after the man- ner of certain forms of crenatus. Yet I believe that it should stand as a species distinct from crenatus and fasciatus, as I have no doubt that it represents a different line of development. art.20 *® FLORIDA TREE SNAILS—SIMPSON 25 LIGUUS SOLIDUS SOLIDUS Say Plate 1, fig. 7 1825. Achatina solida Say, Journ. Acad. Nat. Sci. Philadelphia, vol. 5, p. 122. Shell rather solid, usually elongated, very glossy and of a de- cidedly porcellanous texture; whorls generally somewhat flattened ; sutures impressed; color ivory white to straw colored with a broad, faint yellow band above the periphery and a wide, faint yellow basal band; axial region white. Length 44, diameter 22 mm.; length 49, diameter 24 mm.; length 63, diameter 29 mm. Key West; Sugarloaf Key; Big Pine Key; No Name Key; Litttle Pine Key. Probably once an inhabitant of all the drier islands of the lower chain of keys. Some of the specimens received from residents of these islands are not much elongated. LIGUUS SOLIDUS SOLIDULUS Pilsbry Plate 1, fig. 2 1912. Liguus solidus solidulus Pi~sspry, Journ. Acad. Nat. Sci. Philadelphia, p. 463, pl. 37, fig. 2a. Shell rather thin, slightly to moderately elongated; whorls more rounded than in solidus; color pattern generally an ivory white ground with two narrow faint yellow bands at the periphery, one at the suture and one on the base, sometimes with a few dark, spiral lines on the last whorl; whole surface glossy and porcellanous, axial region white. Length 60, diameter 28 mm.; length 60, diameter 30 mm. Entire chain of the Lower Florida Keys. I collected specimens on Stock Island which are more richly col- ored back of the aperture than on the rest of the shell, and some have from two to eight narrow spiral lines on the last whorl. I have in my collection a set of specimens received from an exchange and labeled “ Monroe County, Fla.,” one of which has no real dis- tinction of banding but is nearly uniform yellowish on the last whorl, and another has faint, broad, yellow bands that break up into blotches. LIGUUS SOLIDUS CRASSUS Simpson Plate 1, fig. 8 1920. Liguus solidus crassus Simpson, Proc. Biol. Soc. Washington, vol. 33, p. 126. Shell very solid, of medium size, with somewhat rounded whorls; columella heavy and decidedly truncated; color a uniform ivory 26 PROCEEDINGS OF THE NATIONAL MUSEUM @ vou. 73 white with sometimes a narrow, bronzy peripheral line; aperture having a strong, white ledge within; axial region white. Length of type, 43, diameter 27 mm. Big Pine Key; Ramrod Key, Frazer; Key West, Henderson col- lection. In 1885 I found the type at Watson’s hammock on Big Pine Key, a very solid shell with the tip slightly truncated. A shell from Key West was given me by the late Mr. John R. Henderson. LIGUUS SOLIDUS PICTUS Reeve Plate 1, fig. 3 1842. Achatina picta Reeve, Proc. Zool. Soc., London, p. 56; Conch. Syst., vol. 2, p. 178, fig. 10. Shell thin to subsolid, polished, of medium size, with somewhat rounded whorls; columella straight or slightly twisted and rather thin; surface straw colored or grayish yellow with a pink apex and several vertical or slightly zigzag stripes on the fourth whorl which gradually pass into a double row of squarish brown spots farther down the shell and continue along the suture and periphery of the last whorl to the aperture. These spots may be opposite or alter- nate or irregularly placed. In addition there are occasional axial bluish smears as in graphicus. Length, 41; diameter, 25 mm. Big Pine Key; Island of Cuba, probably. I have said elsewhere that I believe this form to be a native of Cuba and that it has only recently arrived on our shores. I see no difference in shells credited to this island and Big Pine. I have a dead shell that I collected near the present railway station on the latter and a fine specimen received from the late A. G. Wetherby labeled “Achatina fasciata, Cuba.” In the shells I have seen the columella is white, thin, and straight, but Pilsbry states in the manual (vol. 12, ser. 2, p. 171) that it is more or less, or not, trun- cated. It is an exceedingly rare form of which I have seen only four specimens, and it is probably extinct in Florida. LIGUUS SOLIDUS GRAPHICUS Pilsbry Plate 1, fig. 10 1912. Liguus solidus graphicus Prrspry, Journ. Acad. Nat. Sci. Philadelphia, ser. 2, Vol. 15, p. 468, pl. 37, figs. 1, 1a. Shell usually large, solid, more or less elongated, having flattened whorls and deep sutures; columella often slightly twisted but scarcely truncated; color, pale yellow with pink or purplish axial art, 20 FLORIDA TREE SNAILS—SIMPSON pari region, the third and fourth whorls painted with longitudinal or slightly zigzag blotches and strigations, and these become darker and more irregular on the next two; there is a rather broad white sutural and peripheral band with a dark central line, and it is usually bordered by a dark broken line on the spire and occasionally on the body whorl. Sometimes there is a broken brown line below the white band on the body whorl. The base may be uniform yellow or flamed with irregular brown blotches, and there are generally one or two bluish axial smears on the body and penultimate whorls. Length 69, diameter 30 mm.; length 60, diameter 30 mm. Lower Florida Keys from Little Pine Key west to Boca Chica; West Summerland Keys of the upper chain. Formerly the most abundant of any of the forms on the Lower Keys. I have never found even recognizable fragments of it west of Boca Chica Key, but it may have formerly extended to Key West. I recently re- ceived a very well-preserved specimen which had been inhabited by a land crab from Mr. Cleveland Wells of Big Pine, who collected it on the West Summerland Keys, near the lower end of the upper chain, and I found fragments of the same on one of these islets in the thick, tropical scrub. It may have reached these islets by drifting from Big Pine Key, which is only a short distance away. Its shell is one of the most magnificent of all the land snails, being large, solid, and richly porcellaneous, highly polished and finely painted. LIGUUS SOLIDUS LIGNUMYITAE Pilsbry Plate 1, fig. 11 1912. Liguus fasciatus lignumvitae Pitspry, Journ. Acad. Nat. Sci. Philadel- phia, ser. 2, vol. 15, p. 461, pl. 37, figs. 4 a—d. Shell large, somewhat elongated, thin but strong, with slightly rounded whorls; axial region pink or purplish; general color pale or greenish yellow, sometimes almost white, often becoming ashy on the spire, the second to fourth whorls having light-brown straight or wavy axial stripes, and these become broader and purplish farther down the spire; in addition there are few to numerous bluish smears and blotches on the last one or two whorls; beginning at about the fourth whorl and extending to the aperture there is often a single or double row of dark dots at the suture, and there is a whitish peripheral band with a reddish line in its center. Besides these there are generally a few to several green spiral lines or bands on the lower half of the last whorl. Length 65, diameter 30 mm.; length 50, diameter 29 mm. Lignumvitae Key, mostly on the south part; Lower Matecumbe Key; abundant on both islands and less elongated on the latter. 28 PROCEEDINGS OF THE NATIONAL MUSEUM VoL. 73 Certain specimens from the former locality are quite solid and lack spiral green lines, approaching so closely to graphicus that it is difficult to separate them. LIGUUS SOLIDUS DELICATUS Simpson Plate 1, fig. 4 1920. Liguus solidus delicatus Simpson, Proc. Biol. Soc. Washington, vol. 33, p. 133. Shell usually rather thin and elongated, with somewhat flattened whorls, straw colored to buff, sometimes having an occasional narrow bluish axial smear or dark rest line; second to fourth whorls usually with faint, longitudinal, brownish lines; there is a single, narrow spiral dark line above the sutures and on the periphery, sometimes very faintly white bordered. Axial region straw colored or pur- plish. Rarely there are a few traces of spiral green lines on the base of the shell. Length 65, diameter 28 mm.; length 55, diameter 25 mm. Lower Matecumbe Key; west end of Upper Matecumbe Key; In- dian Key (Wurdeman). I have several hundred specimens from Lignumvitae Key, but none that I can refer with certainty to this form. Rarely a specimen shows a few faint dots at the suture. Usually distinct from the other forms, but an occasional intermediate occurs. LIGUUS SOLIDUS SIMPSONI Pilsbry Plate 1, fig. 5 1920. Liguus solidus lineatus Simpson, Proc. Biol. Soc. Washington, vol. 33, p. 121. Shell large, usually elongated, thin but strong, with a pinkish axial region; pale greenish straw color, entirely lacking bluish smears or other variegation and without a dark peripheral band, usually with a few green spiral lines or bands on the base. Length 60, diameter 28 mm.; length 50, diameter 26 mm. Lignumvitae and Lower Matecumbe Keys. One shell received from Dr. H. A. Pilsbry labeled “Grassy Key.” I first bestowed the name //neatus on this form, but Doctor Pilsbry called my attention to the fact that it had been used for a Liguus by Valencennes and suggested the name stmpsoni for it. It has been very abundant on Lignumvitae Key, where it is found mostly on the northern part of the island. Certain shells of this form superficially rather closely resemble some of the specimens of subcrenatus but may always be distinguished on account of having a rosy apex, while that of the latter is milky white. ART, 20 FLORIDA TREE SNAILS—SIMPSON 29 LIGUUS SOLIDUS PSEUDOPICTUS Simpson Plate 1, fig. 9 1920. Liguus solidus pseudopictus Simpson, Proc. Biol. Soc. Washington, vol. 33, p. 122. Shell large, somewhat elongated, thin, with the axial region white, with slightly rounded whorls and well-impressed sutures, grayish white to greenish yellow, sometimes cream colored and having oc- casional bluish axial smears, the third, fourth, and sometimes the fifth and sixth whorls often having zigzag brownish lines and blotches; last whorl sometimes greenish-yellow with green, spiral lines; there is a double row of sutural squarish brown spots from columella thin and straight; texture glassy to porcellanous. There is a form with broad, brown zigzag strigations on the fourth, fifth, and sixth whorls. Length of type 50, diameter 36 mm.; length of a large specimen 64, diameter 32 mm. Lower Matecumbe Key near the middle of the island. Undoubtedly derived from the form lignumvitae and analogous to pictus from which it differs in its much larger size, its thinness of shell, duller color, and white axial region. Occasional intermediates connecting it with lignwmvitae are found. The young shells have a smoky peripheral band such as is seen in pictus. Some of the shells are porcellanous but the majority are but slightly so and a few have zreen spiral lines on the base. A form has broad, brown, zigzag stripes. LIGUUS FASCIATUS Miiller 1744. Liguus fasciatus MULLER, Verm. Terr. et Fluv., vol. 2, p. 145. 1774. Shell imperforate, oblong-conic, smooth, usually glossy and highly painted, the colors being white, yellow, brown, green, orange, and even scarlet; whorls rounded; axial region always wholly or in part pink or purplish; columella thin and straight to thick and twisted or truncate. Entire island of Cuba; Cozumel Island; lower Florida along the coast at Marco on the west to Fort Lauderdale on the east; Upper Florida Keys. In a majority of the Floridian subspecies some of the earlier whorls show brownish regular or zigzag lines or even blotches, such markings being present in castaneus, testudineus, versicolor, castaneo- zonatus, alternatus, miamiensis, elegans, and occasionally in lving- stoni; but they appear to have faded out in roseatus, lineolatus, and ornatus. Without a doubt this color pattern which we see strongly developed in Cuban shells of this species was one of the earlier char- 30 PROCEEDINGS OF THE NATIONAL MUSEUM VoL. 73 acters of fasciatus and in fact of the genus. It has become obliter- ated in all the forms of crenatus save marmoratus, in ornatus, in typical solidus, solidulus, crassus, simpson, and nearly so in delicatus. The axial region in forms of fasciatus may be pink or purplish at the tip of the spire and white at the base or the opposite, but there is always more or less color to it, while that of crenatus in all its forms is pure white. This may seem like a trivial character on which to found a species, but it is the only constant one, and with Pilsbry I believe it is one of long standing, the color having faded from the forms with white axis a long time ago. LIGUUS FASCIATUS CASTANEOZONATUS Pilsbry Plate 1, fig. 12 1912. Liguus fasciatus castaneozonatus Piuspry, Journ. Acad. Nat. Sci. Phila- delphia, ser. 2, vol. 15, p. 460, pl. 39, figs. 238, 23 a. Shell rather solid, with moderately rounded whorls; axial region pink to deep purple; surface with a white ground; third whorl with faint, irregular axial or zigzag brown striations or blotches which become darker and closer until at about the fifth or sixth whorl they form a broad, more or less solid chestnut or black spiral band. There is a similar basal band and usually a narrow reddish brown periph- eral line and all three extend to the aperture. Length 438, diameter 24 mm.; length of a large shell from Key Largo 60, diameter 30 mm. Key Vaca group; Upper Matecumbe Key northeast along the chain to Elliotts Key and on several of the small adjacent islands; Middle and East Cape Sable; Chokoloskee; the south shore of the mainland as far east as Madeira Bay; the rocky ridge of the lower east coast from Miami southwest to Long Pine Key in the Everglades; Pinecrest. A striking and beautiful Ziguwws which is very abundant and widely distributed, occupying nearly all the region in Florida in- habited by the species to which it belongs. It is apparently want- ing from Miami northward; on the west coast above Chokoloskee and on a few of the middle keys of the upper chain. At Miami it may hybridize with other forms. A Zigwus occurs in Cuba of which I have a specimen from Salto Manantiales which is extremely close to this, and I have shells from Andros de Cisneros labeled “ Isla de Cuba” very near our subspecies. It is quite possible that castaneo- zonatus or its prototype may have sprung from blaineanus, a Cuban form now confined within narrow limits. In some Floridian shells the dark band is more or less broken up, while in others it is nearly continuous. te i ee ART. 20 FLORIDA TREE SNAILS—SIMPSON jl LIGUUS FASCIATUS ALTERNATUS Simpson Plate 2, fig. 1 1920. Liguus fasciatus alternatus Simeson, Proc. Biol. Soc. Washington, vol. 33, p. 123. Shell medium sized, rather solid, with bright pink axial region; color white with a broad, spiral, supraperipheral band consisting of alternating light and dark axial chestnut lines and bars; there is sometimes a faint, narrow peripheral reddish line and there is a broad band at the base similar to the wide one above; columella straight or very slightly twisted. Length of type 45; diameter 24 mm. Timb’s hammock; Black Creek, Paradise Key, all in Lower Dade County, Fla. This form, which is probably a sport from castaneozonatus seems to be confined to a few localities in the south end of the mainland of the State. Certain shells have the broad bands replaced in places by white or yellowish. LIGUUS FASCIATUS ROSEATUS Pilsbry Plate 2, fig. 7 1912. Liguus fasciatus roseatus Pitsspry, Journ. Acad. Nat. Sci. Philadelphia, ser. 2, vol. 15, p. 448, pl. 38, figs. 11, 11a, 11b, 13, 15, 19, 19a. Shell subsolid to solid, with somewhat rounded whorls; axial region pink to purplish; surface white, with a broad supraperiph- eral band of yellow, brownish yellow, or orange; rarely this band is overlaid with a few green spiral lines; there is a similar basal band and occasionally a faint narrow reddish peripheral line; parietal wall pink or tinted purplish, darker colored along its outer edge, sometimes having deep pink streaks; columella twisted and subtrun- cate in heavy shells, thinner and straight in less solid ones. Length 45, diameter 25 mm.; length 40, diameter 21 mm. Entire area occupied by Ziguus in Florida except the extreme northeast portion, the Lower Keys, and the central part of the Upper Keys. One specimen from Pinecrest. This is the most widely distributed form of Ziguus in Florida, and it is not exceedingly variable. Some of the shells are very beautiful, the darker bands being a brilliant orange. Rarely a few dark spiral lines on the last whorl. LIGUUS FASCIATUS LINEOLATUS Simpson Plate 2, fig. 8 1920. Liguus fasciatus lineolatus Simpson, Proc Biol. Soc, Washington, vol. 33, p. 125. Shell subsolid to solid, with somewhat rounded whorls; axial area pink or purplish red; surface whitish, flesh colored, or yellowish, 32 PROCEEDINGS OF THE NATIONAL MUSEUM vou. 73 with or without a buff, greenish, or reddish spiral peripheral line; there are often one or more faint bronzy spiral lines on the base. Rarely there are one or more spiral lines on the upper part of the last whorl; columella usually twisted, almost truncate in heavy shells. Length 638, diameter 32 mm.; length 53, diameter 27 mm.; length 37, diameter 20 mm. Vaca group of the Upper Keys; all the Upper Keys from Upper Matecumbe to and including Elhotts Key; mainland from Marco south to Cape Sable; south shore of the mainland. A widely distributed and variable form. Shells from the upper end of Largo and the small keys near it are often small, solid, and have more or less flattened whorls, while in others on these same islands they are large and somewhat rounded. This latter form oc- cupies Pumpkin Key exclusively, although roseatus is found on Key Largo that is separated from it by only a narrow and very shallow strait. Pilsbry includes it with his roseatus, but it seems to me to be perfectly distinct and the two have a somewhat different distribu- tion. Its spiral lines may occur on the periphery while those of livingstoni do not. LIGUUS FASCIATUS ELEGANS Simpson Plate 2, fig. 2 1920. Liguus fasciatus elegans Stimpson, Proc. Biol. Soe. Washington, vol. 33, p. 124. Shell generally small to medium size, solid, conical, with mod- erately rounded whorls, the second to the fifth being marked with irregular brown axial stripes and blotches; axial region a rich pink, with two or more deeper colored lines on the columellar area; general surface flesh colored; there is a reddish spiral line around the periphery and at the suture and sometimes one or more greenish ones on the last whorl; columella twisted or truncated. Length of type 40, diameter 22 mm.; length of a large shell 58, diameter 30 mm. A small key east of Whitewater Bay, where this and roseatus were the only form of Ziguus; small hammock on Long Pine Key, one very large specimen; Paradise Key; Costello’s hammock; Miami; Arch Creek; Pinecrest. This form may be distinguished from lineolatus by the strigations and blotches on the earlier whorls and it inhabits an entirely different area from that subspecies, being strictly confined to the rocky ridge of the lower mainland and the Pinecrest region. I collected Liguus in the vicinity of Cabanas, Cuba, that very closely resemble this. ant, 20 FLORIDA TREE SNAILS—SIMPSON 33 LIGUUS FASCIATUS LIVINGSTONI Simpson Plate 2, figs. 3 and 9 1920. Liguus fasciatus livingstoni Stmpson, Proc. Biol. Soc. Washington, vol. 33, p. 124. Shell variable in size, solid, with rounded whorls; axial region purplish pink to deep purple, the parietal wall often showing deeper- colored streaks; surface white, usually smoky, greenish or pale yel- lowish green on the last whorl, with from a few to several spiral green lines on the last two whorls but wanting such markings at the periphery ; columella straight or twisted. Length of type 42, diameter 24 mm.; length of large shell 58, diameter 27 mm. Fort Lauderdale (Squires), south along the rocky ridge to Long Pine Key in the Everglades, rare at the lower end of its area. I have a shell of fasciatus from Luis Lazo in Western Cuba 56 mm. in length that is very close to others that I got on cypress trees at the head of the Miami River, Fla. It agrees in size, color, markings, and weight to the last detail. The species lévingstoni differs from lineolatus in having colder colors, in usually becoming darker toward the aperture and the more numerous green spiral lines which are absent on the periphery, also in lacking a reddish peripheral line; besides it has a different distribution. Recently a specimen of this has been found at Fort Lauderdale by Carl Squires. LIGUUS FASCIATUS MIAMIENSIS Simpson Plate 2, fig. 4 1920. Liguus fasciatus miamiensis Simpson, Proc. Biol. Soc. Washington, vol. 33, p. 124. Shell usually small to medium size, subsolid, with rounded whorls; axial region light to deep purple; body of the shell whitish, the fourth, fifth, and sometimes the sixth whorls having a wide, median band consisting of irregular brown blotches often on a yellowish ground, and this usually ends abruptly on the last whorl. The latter part of the last whorl ordinarily has a number of narrow, spiral green lines which extend to the aperture. Length 46, diameter 23 mm.; length 38, diameter 38 mm. Ojus south and west along the rocky ridge to Paradise Key. Doubtless a hybrid with some of the characters of castaneozonatus. I have a shell from Nuevitas, Cuba, which though larger and more elongated has the same general markings as this form. Some speci- mens of méiamiensis have a reddish peripheral line, while others scarcely show a trace of it. It is quite likely that the ancestors of this form drifted from Cuba and became established in the great Miami hammock. 34 PROCEEDINGS OF THE NATIONAL MUSEUM vou. 73 LIGUUS FASCIATUS ORNATUS Simpson Plate 2, fig. 10 1920. Liguus fasciatus ornatus Stmpson, Proc. Biol. Soc. Washington, vol. 33, p. 124. Shell subsolid, small to medium size, rather inflated; axial region pink or purplish; surface yellowish, always becoming darker toward the aperture, where it may be deep yellow, orange, pale yellowish brown, or even scarlet; there is sometimes a faint lighter-colored peripheral band and generally a few green or bronze spiral lines on the last two whorls; columella straight or twisted. Length of type 46, diameter 26 mm. Long Pine Key and hammocks along the rocky ridge to the Miami River. One specimen at Ojus. It differs from roseatus in having no distinct supraperipheral band, the entire shell being yellowish or brownish, and in always being darker on the last whorl and base. Occasionally the columella is nearly white, but the apex is colored. There are intermediates be- tween it and roseatus, and it may be a sport from or hybrid of that form. Some shells flush into scarlet at the aperture. LIGUUS FASCIATUS VERSICOLOR Simpson Plate 2, figs. 5 and 11 1920. Liguus fasciatus versicolor Stimpson, Proc. Biol. Soc. Washington, vol. 33, p. 125. Shell small to medium sized, solid, brilliantly polished, with some- what rounded whorls; axial region pink or purplish at the tip but usually only slightly colored at the columellar area. The ground color may be greenish to brownish with narrow zigzag axial yellow stripes and blotches, or it may vary to yellowish, in which case the stripes and blotches are wanting, and it may have a double row of irregular brown spots at the suture and on the periphery. There is a smoky band with a lighter center at the periphery, and it may be considerably broken up or almost entire. In some shells the general tint is bluish or bluish black. Length of type 38, diameter 22 mm.; length 40, diameter 24 mm. Long Pine Key in the lower Everglades. This island is 8 miles long and 4 wide; it is covered with a forest of Carribbean pine and has fine hammocks scattered over it. It is in one large hammock on this key that this form has its metropolis and is rarely found else- where on the island. There is an abundance of material that com- pletely connects the extremes of color in this exceedingly variable form. Very rarely there are a few faint spiral dark lines on the base of the shell. This is a wonderfully beautiful Liguus, the most ' 4 3 ; Ae ae er ee ee ‘ART, 20 FLORIDA TREE SNAILS—SIMPSON 35 variable of any subspecies I know. So far as I am aware, it is confined to this large key. LIGUUS FASCIATUS CASTANEUS Simpson Plate 2, figs. 6 and 12 1920. Liguus fasciatus castaneus Simpson, Proc. Biol. Soc. Washington, vol. 33, p. 126. Shell moderately solid, rather large; apex whitish to deep pink; columella purplish white to dark purple; surface chestnut to almost black, the upper part and sometimes all of the shell marked with pale to deep yellow, irregular axial stripes which are often zigzagged; there is a broad, sometimes double smoky peripheral band with a lighter one between, and these may be almost obliterated in very dark shells; columella slightly twisted. Length 52, diameter 28 mm. Miami to Long Pine Key. Typically this form is darker than testudineus, its near ally; it lacks the dark spiral lines of the latter and has a more distinctly defined color pattern, but there are occasional intermediates. Shells from Cox’s hammock and Paradise Key are nearly black. LIGUUS FASCIATUS TESTUDINEUS Pilsbry Plate 3, fig. 1 1913. Liguus fasciatus testudineus Pitssry, Journ, Acad. Nat. Sci. Philadelphia, ser. 2, vol. 15, p. 457, pl. 39, figs. 20 @ to 20 f. Shell subsolid, with rather rounded whorls; axial region generally pale pinkish to purplish, the columellar area sometimes almost white; color varying from yellowish to brown; in the darker shells there is a pattern of irregular, light, wavy blotches and strigations; in the lighter-colored specimens there may be only a double series of indis- tinct spots at the suture; certain examples have bluish clouds on the spire; there is usually a double smoky or dark chestnut band at the periphery, with a lighter one between, and there are dark spiral lines on the last two whorls. Columella generally thin, sometimes twisted. Length 45, diameter 23 mm.; length 48, diameter 25 mm. Miami hammocks. A beautiful, often glossy and very variable form which seems to be confined to the great Miami hammock. It differs from castaneus in being lighter colored, in its occasionally clouded surface, and in having dark spiral lines. Certain specimens closely approach some of the lighter forms of versicolor, but the former have the spiral lines which are lacking in the latter, though there are some interme- diates. There are hybrids which have numerous spiral, green lines on the last whorl. 36 PROCEEDINGS OF THE NATIONAL MUSEUM vou. 73 LIGUUS CRENATUS Swainson 1821. Achatina crenata Swainson, Zool. Ill., vol. 1, pl. 58. Shell medium sized to large, thin to solid, usually not porcellanous, with more or less rounded whorls, typically white with spiral green lines which become slightly impressed near the aperture and end in small crenations on the thin edge of the outer lip; entire axial region always white. Whole island of Cuba; coast of Florida from Fort Lauderdale around to Lossmans River on the west; Upper Florida Keys. This species received the name crenatus because of the slight crena- tions at the edge of the shell where the green spiral lines end. But only the forms having these lines show crenations; and, as a matter of fact, many specimens of fasciatus that have similar lines show slight teeth. Usually lacking dark spiral lines at the periphery. It freely hybridizes with fasciatus, a fact which is proven by Liguus of the two types hatching from a single set of eggs here in Florida, and without doubt the same thing is true in Cuba. I have no evi- dence that fasciatus or crenatus hybridize with solidus within our borders, although on Lower Matecumbe Key subcrenatus is found in the same hammocks as forms of solidus, and on Upper Matecumbe one form of solidus, delicatus, and matecumbiensis, luteus, and cas- taneozonatus have been found and there is not the slightest inter- gradation. But I am inclined to believe that solidus and fasciatus hybridize in Cuba. The completely white axial region is the best distinguishing character of this species. LIGUUS CRENATUS MARMORATUS Pilsbry Plate 3, figs. 2, 3, 7, and 8 1912. Liguus fasciatus marmoratus Pirspry, Journ. Acad. Nat. Sci. Philadelphia, ser. 2, vol. 15, p. 455, pl. 37, figs. 9, 9a, 9d, 10. Shell generally elongated, thin to solid, with only moderately rounded whorls, of which the first three and the entire axial region are white; on the next whorl there are irregular alternate white and brownish axial stripes, below which the ground color becomes chest- nut to almost black, and it is marked with axial yellow or whitish stripes or flask-shaped blotches; there is a light band at the suture and on the periphery, and at the latter place it is usually bordered by two dark ones; last whorl generally showing dark spiral lines; aperture white within; columella thin and straight to heavy and truncate. Length 65, diameter 28 mm.; length 43, diameter 22 mm. Key Vaca and doubtless other islands of the Vaca group; Long Island of the Upper Keys, Key Largo; Porgy Key, doubtful; Cape Seite ee - peer ee ART, 20 FLORIDA TREE SNAILS—SIMPSON 37 Sable; Chokoloskee; Long Pine Key; Cox’s hammock; Snapper Creek; Costello’s hammock; Miami hammock; Pinecrest, south-cen- tral Everglades, not typical. A striking and very variable form, probably a hybrid, which com- monly bears the name of the “black snail”; and, considering its rarity, it has a wide distribution. Rarely a shell shows bluish or greenish cloudings like versicolor, and one specimen from Key Vaca is marked very much like the lighter-colored forms of that sub- species. In one shell from the Miami hammock and another from Snapper Creek the axial whitish strigations continue to the base, and there is a dark peripheral band without a light center. One shell from Vaca is very thin and inflated and bears some resemblance to Oxystyla resus. Without a doubt this form has been an inhabitant of the Chokoloskee region. A Mr. House, who resided there, took me in his boat to a place where there was formerly a fine hammock where he said he found the black snail, but it had been cut down and made into a field. We found dead shells there which still showed that they were marmoratus. It is a nondescript and combines characters of fasciatus and crenatus. I have no doubt but that this or an analogous form inhabits or has recently inhabited some part of Cuba, although, so far as I know, nothing like it has been found, and that it has migrated and become established on Key Vaca, from which it spread along the Upper Keys and crossed to the mainland on the old land bridge. It reached Chokoloskee on the southwest coast and has probably been swept across to the rocky ridge during time of a tidal wave. On this ridge it has again hybridized, this time with forms of fasciatus, and has produced versicolor on Long Pine Key; cas- taneus, which has spread up to the Miami hammock; and at the latter place it has developed into festwdineus. Throughout this ridge occasional shells are found which have the entire axial region pure white, and these I refer to marmoratus. LIGUUS CRENATUS VACAENSIS Simpson Plate 4, fig. 10 1920. Liguus crenatus vacaeisis Stmpson, Proc, Biol. Soc. Washington, vol. 338, De 222, Shell usually large, with convex spire, subsolid to solid, with deep sutures and the last and penultimate whorls slightly flattened, white or shaded greenish with sometimes a few spiral green or bronzy lines on the body or base; texture somewhat porcellanous; columella heavy and twisted or truncated. Length of type 54, diameter 27 mm.; length of large shell 64, diameter 33 mm. 38 PROCEEDINGS OF THE NATIONAL MUSEUM VOL. 73 Vaca Keys; Long Island; Key Largo; Angelfish Key; Sands Key; Northwest and Middle Cape Sable (?) near Flamingo (?). A fine, usually solid, somewhat porcellanous form which generally has the last whorls slightly flattened in the middle but well shoul- dered. It constantly differs from capensis in being less elongated and having a convex spire. I am not quite certain about the main- land specimens. LIGUUS CRENATUS CAPENSIS Simpson Plate 3, fig. 9 1920. Liguus crenatus capensis Stimpson, Proc. Biol. Soc. Washington, vol. 33, Dra22: Shell large, solid, much elongated, with straight sides and some- what chalky texture; whorls slopingly flattened; sutures not very deep; surface white or slightly greenish tinted on the last whorl, with a few to several spiral lines on the last or last two whorls, those above the periphery green and the basal ones ashy brown; aperture small, rather short, the outer lip not greatly oblique; columella generally thin and straight or only slightly twisted. Length of type 58, diameter 27 mm.; length 60, diameter 26 mm. Northwest, Middle, and East Cape Sable; hammock near Fla- mingo; near Cuthbert Lake 20 miles east of Northwest Cape Sable (Livingston). More elongated than any other Florida Ziguus, with very straight sides and ashy brown basal lines. A specimen found on Northwest Cape Sable in 1885 had a broad, brilliant green belt above the periphery, but it fell into the grass and was lost. The green lines are often worn away on living shells. LIGUUS CRENATUS MATECUMBENSIS Pilsbry Plate 3, fig. 4 1912. Liguus crenatus matecumbensis Prtspry, Journ. Acad. Nat. Sci. Phila- delphia, ser. 2, vol. 15, p. 446, pl. 37, figs. 5, 5a. Shell medium to large, rather thin to subsolid, varying from some- what slender to inflated, with but slightly rounded whorls and a faint angulation just below the periphery, creamy white to pale greenish yellow with a few spiral green or bronzy lines on the last whorl or two, spire rounded; aperture large, rather long; outer lip oblique; columellar area creamy to yellowish; there is sometimes a golden flush on the parietal wall; columella thin and incurved. Length 55, diameter 29 mm. Length of a slender shell 50, diameter 23 mm. Upper Matecumbe Key; Middle and East Cape Sable; Flamingo and hammocks in the vicinity; east to Madeira Bay. Very abun- dant on the mainland. ART, 20 FLORIDA TREE SNAILS—SIMPSON 39 This differs from other forms of crenatus in the large aperture, the incurved columella, and the more oblique outer lip. In most shells the colored lines become abraded during life and sometimes change from green to bronze. In some shells the yellow columellar flush is present; in others it is wanting. LIGUUS CRENATUS SUBCRENATUS Pilsbry Plate 3, fig. 10 1912. Liguus crenatus subcrenatus Pitspry, Journ. Acad. Nat. Sci. Philadelphia, ser. 2, vol. 15, p. 445, pl. 37, figs. 7, Ta. Shell large, rather elongated, thin but strong, with but slightly rounded whorls; surface glassy white except the apical region which is dull or milky white; last whorl having a few spiral green lines which are mostly on the base; columella thin and straight. Length 60, diameter 28 mm.; length 70, diameter 32 mm. Grassy, Lower Matecumbe, and Windleys Keys; Long Island; ' Key Largo. A thinner shell than vacaensis, and it has less convex sides; it is more glassy, being almost as much so as septentrionalis. Sometimes the later whorls are slightly tinted with green. The shell whose dimensions are last given and which I collected on Windleys Key is the largest Liguus I have seen from Florida. Apparently about a quarter of an inch of its apex has been broken off and sealed up, and if it were perfect it would probably measure 77 mm. It very likely had nine whorls, though most shells of this form have but eight. I collected Liguus at Cape San Antonio, Cuba, which were quite similar to this subspecies. LIGUUS CRENATUS ELLIOTTENSIS Pilsbry Plate 3, fig. 11 1912. Liguus crenatus elliottensis Pitspry, Journ. Acad. Nat. Sci. Philadelphia, ser. 2, vol. 15, p. 447, pl. 37, figs. 3, 3a, 3b. Shell small to medium sized, somewhat inflated, thin and usually fragile, with six to six and a half whorls which are generally but slightly rounded, lusterless white, sometimes having transparent gray streaks, and occasionally dark spiral lines on the lower half of the last whorl; columella thin and straight or but slightly twisted. Length 37, diameter 20 mm.; length of a large dead shell from Old Rhodes Key 44, diameter 25 mm. Elliotts Key; Old Rhodes Key; Scott’s place on Key Largo (Na- tional Museum collection). A small form with rather negative characters, being thin and simply colored. I found a number of dead specimens on Old Rhodes 40 PROCEEDINGS OF THE NATIONAL MUSEUM vou. 73 Key, some of them being considerably larger than those from El- liotts Key. Pilsbry, who has seen numerous perfect shells, says that there is a faint yellow sutural line and usually some yellowish olive green, chiefly basal lines. LIGUUS CRENATUS LUTEUS Simpson Plate 3, fig. 12 1920. Liguus crenaius luteus Simpson, Proc. Biol. Soc. Washington, vol. 33, 10h Ast Shell varying from small to large and from thin to solid; axial region white, with the columella generally twisted or subtruncate; whorls somewhat rounded, with the sutures well impressed; color varying from straw or pale yellow to deep yellow or orange, lighter to white on the earlier whorls, sometimes having from one to several narrow dark or green spiral lines on the last whorl at or above the periphery and fainter lines on the base. Length of type 63, diameter 18 mm.; length 38, diameter 18 mm. Vaca group of the Upper Keys; east end of Upper Matecumbe, one specimen; Kep Largo (C. E. Saxton), a fine young specimen; rocky ridge on the south and southeast mainland from Long Pine Key east and north to Dania; Pinecrest, very brilliant. In most shells there is a sort of ledge inside at the aperture com- posed of the dull white matter of the interior, and the outer lip is thin and darker colored. Some of the specimens from Key Vaca have the last whorl brilliantly tinted with orange, and I have shells collected back of Larkins and about Dania, the latter locality being near the northern limit of Ziguus on the east coast, which are very close to the Vaca material. It is an abundant form, and as I have remarked, it is the first to inhabit the incipient hammocks of the great rocky ridge. In some of the larger hammocks it is the only occupant. LIGUUS CRENATUS CINGULATUS Simpson Plate 3, fig. 6 1920. Liguus crenatus cingulatus SIMPSON, Proc. Biol. Soe. Washington, vol. 33, p. 123. Shell medium to large sized, thin to subsolid, somewhat shining; whorls varying from flattened to somewhat rounded, pure white, with a broad, pale yellow spiral band above the periphery and another below it; columella twisted. Length of type 33, diameter 20 mm. Long Island of the Upper Keys; Key Largo; Middle Cape Sable; East Cape Sable; Flamingo; Long Pine Key(?); Timb’s hammock; Lysiloma hammock, both in Lower Dade County; Miami; Lemon City. 5 Sa SO eee er ereemmnt a te ae ee ART. 20 FLORIDA TREE SNAILS—SIMPSON Al Although this form is widely distributed it seems to be rather rare. Exeternally it is colored much like certain specimens of Liguus fasciatus roseatus, but the axial region is pure white through- out. The yellow is paler than it is in that form. LIGUUS CRENATUS EBURNEUS Simpson Plate 4, fig. 1 1920. Liguus crenatus eburneus Simpson, Proc. Biol. Soe. Washington, vol. 33, p. 122. Shell rather solid, usually of a somewhat porcellanous texture, obese to rather elongated, with rounded whorls, pure or ivory white throughout or rarely having traces of spiral bronzy lines on the base or at the aperture; columella twisted. Length of type 52, diameter 26 mm. Timb’s hammock, type locality; hammocks along the rocky main- land ridge from Long Pine Key to Lemon City and opposite it on the peninsula; Pinechest, not typical. Has a somewhat porcellanous texture and usually is without traces of spiral lines. LIGUUS CRENATUS MOSIERI Simpson Pate 4, fig. 2 1920. Liguus crenatus mosieri Simpson, Proc. Biol. Soc. Washington, vol. 33, p. 128. Shell variable in size, subsolid, somewhat polished; whorls mod- erately to well rounded, the earlier ones white or whitish, the last ones darker, often smoky tinted or dirty greenish and having from two to several green or bronzy spiral lines which are wanting at the periphery; columella straight or slightly twisted. Length of type 45, diameter 24 mm. ; length of a large shell from Miami 50, diameter 27 mm. Hammocks from Arch Creek southward and westward along the great recky mainland ridge to Long Pine Key, being most abundant at Miami, the type locality. This subspecies is nearer the typical Cuban crenatus than anything we have, but it averages much smaJJer than that. However, I have seen shells from Cape San Antonia and other localities in that island that are no larger than the ordinary mosier7. LIGUUS CRENATUS SEPTENTRIONALIS, Pilsbry Plate 4, fig. 3 1912. Liguus crenatus septentrionalis PILSBRY, Journ. Acad. Nat. Sci. Phila- delphia, ser. 2 vol. 15, p 447, pl. 87, figs. 6, 6a. Shell thin but strong, inflated, with but slightly rounded whorls, the last usually subangulate at the periphery, with glassy surface, 42 PROCEEDINGS OF THE NATIONAL MUSEUM VOL. 732 pure white or slightly greenish, with from one to seven spiral narrow green or bronzy lines on the last whorl that may extend to the penultimate one; aperture large; outer lip thin; columella straight. Length 48, diameter 27 mm.; length 42, diameter 23 mm. Great hammock near Fort Lauderdale; hammocks along the outer shore opposite the town and to the northeastward; south side of New River near its mouth; hammock about a mile south of Fort | Lauderdale; hammock north of Arch Creek, where both typical specimens and those hybridized with fasciatus were found. A form nearly typical was taken just south of Little River stream. A well-characterized subspecies, being thin, short, and inflated, highly polished and glassy, usually with a decided peripherial angu- lation. Sometimes the spiral lines are brilliant green, or they may be bronzy; they are rarely wholly wanting. It has been found in the shore hammock more than 2 miles north of Fort Lauderdale, and this is the most northern authentic locality for Liguus in the State of Florida; I have a specimen of Liguus from the village of Jamaica, Cuba, which is very close to septentrionalis, the only difference being that the Cuban shell is a little more solid and has a slight ledge within the aperture. LIGUUS CRENATUS LOSSMANICUS Pilsbry Plate 4, figs. 4 and 9 1912. Liguus crenatus lossmanicus Pitspry, Journ. Acad. Nat. Sci. Philadelphia, ser. 2, vol. 15, p. 448, pl. 87, figs. 8 a, b. Shell usually small to medium size, inflated, and having decidedly rounded whorls, subsolid to solid; axial region pure white, the columella ordinarily strongly twisted and truncate; there is in the more solid shells a decided ledge or shoulder just inside the aperture; color white, greenish, or yellowish white, usually dull; often there are a few green or bronzy spiral lines and sometimes a dull, broad, yellowish band above and below the periphery. : Length of a shell from the type lot 40, diameter 23 mm.; length of a large shell from Middle Cape Sable 55, diameter 30 mm. Lossmans Key on the southwest coast to Middle Cape Sable; Rodgers River; John Douthett’s place near Flamingo. A peculiar form and somewhat variable for one having such a limited distribution. Most of the shells are short, solid, with ~ rounded whorls; but others which I refer to this are somewhat lighter in build, are more lengthened, and have less decidedly rounded whorls, and these usually do not show the strong inside ledge. But the two seem to intergrade. Dr. Edward Mercer and I found an extensive colony in an isolated hammock at Middle Cape Sable where some specimens were quite large and had the last whorl flattened a Me ecg ge er sue eee “08 an ART. 20 FLORIDA TREE SNAILS—-SIMPSON at the periphery and another had a broad faint yellow supraperiph- eral and a basal band. With these were a number of what were evidently hybrids between lossmanicus and castaneozonatus, having a variety of dark bandings, while others were entirely white, with a purple axial region. Fig. bee bro Fic. BES a e 1. CHOAA AH wD EXPLANATION OF PLATES PLATE 1 Liguus solidus Say, Hammock west of station, Big Pine Kay. . Liguus solidus Pilsbry, Stock Island. . Liguus solidus pictus Reeve, Cuba. . Liguus solidus delicatus Simpson, Lower Matecumbe Key. Liguus solidus simpsoni Pilsbry, figured type, Lignumvitae Key. Liguus fasciatus var. Like castaneozonatus, Salto de Marianales, Cuba. Liguus solidus Say, Big Pine Key, inflated. . Liguus solidus crassus Simpson, Watson’s hammock, Big Pine Key. . Liguus solidus pseudopicius Simpson, Lower Matecumbe Key. . Liguus solidus graphicus Pilsbry, Torch Key. . Liguus solidus lignumvitae Pilsbry. Lower Matecumbe Key. . Liguus fasciatus castaneozonatus Pilsbry, Long Pine Key. PLATE 2 1 Liguus fasciatus alternatus Simpson, Timb’s hammock, Dade County. SOMDNAARHWN ee Ee | I oo SeMNQaanKh WHE . Liguus fasciatus elegans Simpson, hammock in lower Everglades. . Liguus fasciatus livingstoni Simpson, type, Miami. . Liguus fasciatus miamiensis Simpson, type, Miami. . Liguus fasciatus versicolor Simpson, type, Long Pine Key. Liguus fasciatus castaneus Simpson, Long Pine Key. . Liguus fasciatus roseatus Pilsbry, Key Largo. . Liguus fasciatus lineolatus Simpson, type, Goodland Point, Collier County. . Liguus fasciatus livingstoni Simpson, large, head of Miami River. . Liguus fasciatus ornatus Simpson, type, Paradise Key. . Liguus fasciatus versicolor Simpson, variety, Long Pine Key. . Liguus fasciatus castaneus Simpson, black variety, Long Pine Key. PLATE 3 . Liguus fasciatus testudineus Pilsbry, ‘Miami. . Liguus crenatus marmoratus Pilsbry, variety, Key Vaca. . Liguus crenatus marmoratus Pilsbry, Chokoluskee. . Liguus crenatus matecumbiensis Pilsbry, near Flamingo. . Liguus crenatus, like small subcrenatus, Cape San Antonio, Cuba. . Liguus crenatus cingulatus Simpson, type, Miami. . Liguus crenatus marmoratus Pilsbry, Key Vaca.. . Liguus crenatus marmoratus Pilsbry, Miami. . Liguus crenatus capensis Simpson, type, northwest Cape Sable. Tiguus crenatus subcrenatus Pilsbry, Lower Matecumbe Key. . Liguus crenatus elliottensis Pilsbry, Elliotts Key, type lot. . Liguus crenatus luteus Simpson, type, Key Vaca, near Conch town. SOAS Up ow pe 10 PROCEEDINGS OF THE NATIONAL MUSEUM VoL. 73 | PLATE 4 . Liguus crenatus eburneus Simpson, Paradise Key. . Liguus crenatus mosieri Simpson, type, Miami. . Liguus septentrionalis Pilsbry, Fort Lauderdale. Liguus crenatus lossmanicus Pilsbry, Lossmans Key, type lot. Liguus hybrid of lossmanicus and fasciatus, Middle Cape Sable. Liguus crenatus from Trinidad Mountains, Cuba, close to eburneus, . Liguus crenatus from Santa Cruz del Morte, Cuba, close to mosieri. . Liguus crenatus from Jamaica, Cuba, very close to septentrionalis. Liguus crenatus lossmanicus Pilsbry, large, from Middle Cape Sable. . Liguus crenatus vacaensis Simpson, Key Vaca. O M _ NATIONAL MUSEU c U. *19999q ‘g | ol ‘sano Cb aged gas aiejd jo uondiosap 404 SQNAD AHL JO STIVNS 35uL Vdaldols ¢b aBed aes a)¥(d jo UoIdIJOsSap 404 ‘SANDIT SANSD SHL JO STIVNS 33¥L VdIdOld ee. eee eo ee U.S. NATIONAL MUSEUM OF c ae) EE SNAII ) FLORIDA TI 43 e page U.S, NATIONAL MUSEUM PROCEEDINGS, VOL. 73, ART 20 PIA E. B. Decker. FLORIDA TREE SNAILS OF THE GENUS LIGUUS. For description of plate see page 44. rs aI CONCERNING THE ORIGIN OF THE METAL IN METEORITES By Grorce P. Merri Head Curator of Geology, United States National Museum The peculiar relationship existing between the metallic and sili- cate portions of a stony meteorite has been noted by several, and particularly the earlier workers. The present writer has on more than one occasion (noted later) made reference to it and also to the work of others. In discussing the matter among his contemporaries he finds, however, a considerable difference of opinion such as has led to the preparation of this paper in which he reviews these opinions and gives the results of his own observations. In his discussion of the composition of the Lodran meteoric stone, Tschermak wrote! as long ago as 1870: Das Nicheleisen und der Magnetkies mitissen spiiter fest geworden sein als die iibrigen Mineralien, und ihre Bildung diirfte zu gleicher Zeit mit jenen Veriinder- ungen yor sich gegangen sein, welche der Olivin erlitten zu haben scheint. With reference to the metallic constituent of the Homestead meteor- ite, Gumbel wrote, five years later: Noch hiufinger erscheinen die aus Meteoreisen bestehenden kérnchen der Gesteinmasse in meist zackigen, winkelig gebogenen, oft in feine Spitzen aus- laufenden Klumpchen beigemengt, welche so innig an die nicht metalischen Theile sich ausschmiegen als ob das Hisen erst zuletzt etwa durch Reduction an der Stelle ausgeschieden worden wiire, wo es sich vorfindet.? So far as I am aware these are the first suggestions of their kind, though they seem to have attracted little attention from other workers. Before entering upon the discussion of these and other views yet to be noted, the following illustrations of actual conditions are given. Figure 1, Plate 1, is from a photomicrograph of the chondritic stone from Anthony, Kans., recently described.* The dark center (1) is troilite; the lighter border (2) metal, which forms, as it were, a Ry ag Tepe NS i i a ka Pia ce a A EE Pw 1 Sitzbericht d. k. Akad. der Wiss. II Abtheil., 1870. 2 Sitzbericht Miinchen Akad. der Wiss., vol. 5, 1875, p, 325. 3 Proce. Nat. Acad. of Sciences, vol. 10, p. 306. No. 2742.—Proceepinas U. S. NATIONAL MUSEUM, VOL. 73; ART. 21, 96270—28 l 2 PROCEEDINGS OF THE NATIONAL MUSEUM VoL. 73 binding or cementing constituent holding the fragments together. The manner in which the meta! projects into the interstices of the silicates is to be noted. Instances of this nature are common in many chondritic meteorites, both crystalline and otherwise. Figure 2 of the same plate is that of a fragment of a dark chon- dritic stone embedded with others of a quite different nature, as described in my paper entitled “The Cumberland Falls, Kentucky, Meteorite,” published in 1920.4 Attention was there particularly called to the fine threadlike forms sometimes assumed by the metal, (1) white in the figure. These veinlets or stringers vary from 1 or 2 millimeters in thickness to mere films of only microscopic dimensions, and divide and subdivide repeatedly, their ramifications reaching out and completely surround- ing or penetrating into the silicates along cleavage or fracture lines. A noticeable layer of metal, too, lies along the boundary line of the fragment, a condition which it was thought might indicate a deposition of the material since the consolidation of the stone in its present brecciated form. Figure 3 is that of a slice of a pallasite belonging to the Rékicky group, found at Admire, Kans., and described in 1902.5 The dark areas are olivine, the white (1), nickel-iron with scattered particles of schreibersite (2), and troilite (3). The feature of im- portance in the present instance, is the angular character of the olivines. It is to be noted that they are not products of crystalliza- tion, in situ. They are rather fragments, in some cases mere splin- ters, as sharply angular as so much shattered glass. These are firmly embedded in the metal with no signs of corrosion or alteration in- dicative of heat or moisture. The same is true of the Eagle station meteorite and others of its class. Figure 4 is from the Four Corners meteorite. This consists largely of a coarsely granular aggregate of metal inclosing fragments of disintegration from a fine granular pyroxenic rock, now in some cases reduced to mere sand. The metal, as shown at (1) completely incloses these silicate particles (2) and in places penetrates slightly into their interiors. The stony fragments are all unchanged, with no sign of corrosion by heat or otherwise, even when in the condi- tion of finest sand, no slag nor glass; the contact is as sharp and free from signs of alteration as though the admixture had taken place when cold. The above illustrations are sufficient, it is thought, to show that, so far as chondritic meteorites and those of the Rékicky and Four Corners type are concerned, the metal is the last constituent to con- geal, and that it is probably wholly of secondary origin. The ques- 4 Proc. U. S. Nat. Mus., vol. 57, 1920, pp. 97-105. 5Y¥dem, vol. 24, 1902, pp. 907-913. ART. 21 ORIGIN OF METAL IN METEORITES—MERRILL 3 tions for consideration are, then, (1) what is the source of the ma- terial and its manner of deposition and (2) will the same explana- tion apply to the pallasites, particularly those of the Rékicky group which seems applicable to the stony forms. Nordenskiold in his description in 1879 * of the Stalldalen meteorite expressed himself in agreement with the writers above quoted in the view that the metal is the latest formed of the constituents and sug- gested the possibility of its preterrestrial origin through the reduction of some ferruginous silicate. To quote his own words: I afseende a den nu ifriigavarande gruppen ir det af dessa meteoriters mikroskopiska struktur tydligt, att det metalliska jernet utgér dessa meteor- stenars yngsta bestandsdel och att det saledes uppkommit genom reduktion af de jernhaltiga silikaterna. This would seem to be essentially in agreement with Daubree7 who as a result of experimental work inclined to the view that the metal was due to the reduction of a highly ferriferous olivine in an atmosphere of hydrogen, with the simultaneous production of an iron magnesium silicate (enstatite). An exception was taken to this view by Fletcher,’ who pointed out the existence of pallasites like that of Krasnojarsk, rich in both iron and olivine but quite lacking in enstatite. Meunier in his article on Meteorites ® has discussed the matter in considerable detail. He wrote: Les manipulations auxquelles les fers météoriques out été sourmis par divers experimentateurs n’ont par tardé 4 montrer que ces roches cosmiques sont pro- fondement desorgansees par le fait d’une fusion pure et simple, * * *. Il etait done necessaire de rechercher une méthode propre a fournir, autrement que par fusion, des alliages de fer et de nickel semblables a ceux des météorites. And further: Le protochlorine de fer étant decomposé au rouge par Vhydrogene, on peut admettre que ce qu’on en trouve a simplement échappé 4 le décomposition et represénte la combination méme d’on la fer a été tiré pour prendre état métallique. The suggestion of Jetrofeioff and Latschinoff*® in 1888 to the effect that the meteorite consists of an isomorphous mixture of the 6 Geol, Froeningen Stockholm Forhandlinger, 1878-79, p. 60. ™ Geol. Experimentale, 1879, pp. 517, 520. 8 Introduction to the Study of Meteorites, 1908, p. 33. ® Hncyclopedie Chimique, 1884, p. 322. 1 Hs ergeibt sich auch, dass der Meteorit einem isomorphen Gemenge der Silicate Mgs SiO, und Fe, SiO, naher steht, als selbst der Olivin von Fogo. Man gelangt hiernach unwillkiihrlich zur Voraussetzung, dass der Meteorit urspriinglich ein Olivinmagma dar- stellte, welches spiter unter Einwirkung reducirender Kiérper, wie Wasserstoff der Kohlen- oxyd oder Kohlenwasserstoffen unter Abschiedung von metallischen Bisen aus dem Magma und gleichzeitiger Abscheidung von Kohlenstoff aus der reducirenden Verbindung, sich differenzirte. Die freigewordene Nieselsiiure ging auf Bildung von Augit. Diese differenz wird zugleich mit der Erhirtung der Hauptmasse des Olivin stattgefunden haben und haben sich daher kohlige Substanz und Nickeleisen hauptsiichlich an den Umrandungen der Kiérner abgesetzt. Es diirfte so auch ras Vorhandensein der Hingangs erwiihnten ebenen Aussenfliichen des Steines erkliirlich werden. (Der Meteorit von Nowo-Urei, M. Jerofeieff und P. Latschinoff in St. Petersburg, 1888.) 4 PROCEEDINGS OF THE NATIONAL MUSEUM VOL. 73 silicates Me.SiO, and Fe,SiO, which under the reducing action of hydrogen, carbon! monoxide, or hydrocarbons have been reduced and differentiated with the separation of the metal and silicate min- erals, however applicable in cases of direct crystallization from a molten magma, are scarcely so in the cases of the clastic rocks here under discussion and may be passed over. Moreover the suggestion of Fletcher already quoted, still holds good. Ideas expressed by Dr. W. Wahl with reference to brecciated struc- tures such as are shown by the Deesa and some other irons are of interest. He says: Wie aus der vorhergehenden Beschreibung ersichtlich ist, hat der Silikatan- teil unbehindert von dem Metallteile des Gesteins krystallisiert; er hat sich wie innerhalb der Hohlriume eines Schwammes, dessen Geriist der metallische Anteil war, verfestigt. Aber zur Zeit der Verfestigung der zwischenliegenden Silikatmasse muss das Metallgertist noch selbst fliissig gewesen sein, denn die Silikate sind dem Metall gegeniiber scharf idiomorph ausgebildet und _hier- durch erhielten die Hisenteile ihre zackige Begrenzung. Hs hat folglich das Magma, aus dem der Siderolith hervorging, vor dem Hrstarren aus einem inho- mogenen Gemische zweir Fliissigkeiten bestanden, die sich noch nicht entmischt hatten und von denen die eine aus den Plagioklas- und Pyroxensilikaten, die andere aus filissigem Metall und Metallverbindungen (gediegen Wisen mit Cohenit und etwas Schreibersit? Magnetkies und etwas Magnetit?) zusam- mengesetzt waren. Die verschiedenen Proportionen zwischen dem metallischen Auvteil und silikatanteil erkliren sich dann durch eine teilweise Hntmischung und durch ein Zerbrechen yon schon auskrystallisierten Silikat-partien sowie Hineingeraten derselben in den noch fltissigen metallischen Anteil. In dieser Weise entstand méglicherweise das von Daubrée beschriebene Stiick. [i. e., the Deesa Iron.] The inference here is that the metal is in a condition of fluidity such as could be imparted only by heat. If so the matter is cer- tainly open for further discussion. That it is possible the meteorites of the pallasite group may result from the direct cooling of two immiscible liquids, the metal, owing to its higher fusibility, cooling first and inclosing the gradually solidifying silicate drops, need not here be argued. ‘That, however, the brecciated structure shown in Figures 3 and 4, or the deposits of metal in the interstices of the silicates as in 1 and 2 could thus originate is doubtful. There is in this connection a view relative to these metal-silicate breccias, belonging to the Rékicky group, that may be worthy of con- sideration, and in which the question of the origin of the metal itself is not necessarily involved. Is it not possible that this breccia- tion may be due to pressure acting upon the mass of a normal palla- site after solidification rather than when the metal is in a fluid con- dition, as Doctor Wahl’s paper implies? The metal, being the more plastic, would flow, while the silicates would be crushed. In this 11Phe possible instrumentality of carbon as a reducing agent was also considered by Nordeuskiold in the paper already noted and the idea dismissed as improbable. Sas — ART, 21 ORIGIN OF METAL IN METEORITES—MBERRILL 5 way the slight amount of displacement sometimes shown by the sili- cate particles (fig. 8, plate 1, and upper plate 2) could be accounted for. It may be questionable, however, if under such conditions the original tripartite character of the metallic alloys would not be de- stroyed or disarranged. In the case of the Admire meteorite the metal gives no visible indications of any such movement. It would seem scarcely necessary to consider the possibility of the iron as having been introduced or injected in the ordinary condition of molten fluidity. The melting point of pure iron is, as given, 1,530° C.; that of nickel 1,452° C. The pyroxenes, on the other hand, fuse at approximately 1,400° C., and olivines at 1310°-1430° C. (accord- ing to Doelter). Apparently it could not then be a question of simple dry fusion as the silicates would be reduced to the condition of slag— “»rofondement désorganées,” as Meunier expressed it. Existing conditions can be explained, moreover, without assuming that the metal has at any time been in a condition of fusion. Direct reduc- tion of an ore as practiced in the early days of iron smelting, or as still practiced in the well-known Catalan process, results in the pro- duction at temperatures not above 700° or 800° C. of a spongy or pasty mass of metal. It is easy to conceive that such material, com- mingled with rock fragments and subjected for sufficient time to a moderate pressure, might give rise to the structures described, par- ticularly such as shown by the Four Corners iron.” Another feature which may have a bearing upon the subject is this. Meteoric irons almost invariably partake of the nature of the so-called “ wrought iron,” in that they are soft and malleable. Re- ports to the contrary can be accounted for only on the supposition that the material selected was a mixture. Some irons, like that of New Baltimore, Pennsylvania, can be hammered down when cold; others are more brittle but still malleable1* Fused in an ordinary gas furnace in the laboratory these soft irons yield a bead no longer malleable but hard and brittle like ordinary cast iron and with an entirely different microscopic structure. (See pl. 3.) 12] am indebted to Prof. Albert Sauveur, of the Harvard Engineering School, to whom I sent a photograph of the slice shown in fig. 4, pl. 1, for the following suggestion : “The structure to which you call my attention recalls somewhat that of wrought iron, in which we also find particles of silicates entbedded in an iron matrix. This results from the fact, as you undoubtedly know, that in the manufacture of wrought iron the reduced metal is not melted, but remains pasty, retaining some of the liquid silicates or slag very much as a sponge retains water. Also, just as further cooling of the sponge results in particles of ice within the sponge, so further cooling of the wrought iron results in particles of silicates within the iron matrix. I wonder whether such a process might have been at work in this case? It would, of course, imply reduction of an iron oxide or of an iron salt at such low temperature that the reduced iron remains below its melting oint.”’ 3 % The United States National Museum collections contain two knife blades 7 and 14 inches long hammrered out of the Coahuila and Nejed irons, respectively, by our local blacksmith in a small charcoal forge. Though easily shaped they could not be tempered. 14 These experiments have not been carried far enough nor with sufficient refinement to ‘allow the drawing of safe conclusions other than those mentioned. 6 PROCEEDINGS OF THE NATIONAL MUSEUM VOL. 73 All the evidence at present available, as I interpret it, points to the origin of the metal as introduced at a temperature lower than that of the melting point of the silicates. As above noted, a reduc- tion of a ferriferous silicate either through the aid of carbon or hydrogen is ruled out of consideration by the complete absence of any secondary or residual products. Of all other known meteoric constituents the ferrous chloride, lawrencite, would seem to best meet the apparent necessities of the case. It is found in varying though small proportions an almost universal constituent, and it is permissible to imagine its one-time presence in vastly greater quantities. It is reduced according to Meunier and, as already noted, at a temperature not exceeding 400° C. (750° F.) in an atmosphere of hydrogen. It would seem, then, not too much to assume that this mineral was, as Meunier conceived, the original source, and the frac- tional amount of chlorine found in nearly all meteorites, stony as well as metallic, but an unreduced residue. And further, it is pos- sible to conceive of a hot mass of commingled rock fragments and ferrous chloride, in which the latter is being reduced to the condi- tion of a metallic paste in which the fragments become engulfed as in Figures 3 and 4, Plate 1, or simply cemented as in Figure 1. It must not be forgotten that H. C. Sorby as long ago as 1864 suggested that the metallic constituents of meteorites were introduced into the interstices of the silicates in a state of vapor. Such a conception would seem to be particularly applicable to the metal in a stone like that of Estherville, Iowa, in which the iron is in slag or spongelike masses not always closely compacted in all its parts with the silicates.1° (Fig. 2, lower.) Tschermak’s observation on this is of interest. He wrote :7° Das Hisen verhiilt sich oft so, als ob es die letze Bildung wire eine impregna- tion welche die zum Theil krystallinische, zum Theil Tuffartige masse dur- chdrungen hat. Meteorites are unmistakably volcanic products. It is fair then to consider the original chloride itself a product of volcanic emanations as in terrestrial volcanoes. There would, in result, be this difference, however: The chloride of terrestrial vol- canoes exposed to an oxygen-rich atmosphere manifests itself almost at once as an oxide. In a heated atmosphere of hydrogen or other reducing gases such as it is possible to imagine exists at the fountain source of meteorites a contrary result would be effected and the iron appear in metallic form. This source would then be comparable to that of the metal in the basalt of Biihl bei Cassel, Germany, as described by Hitel in his re- 15 See Notes on the Meteorite of Estherville, etc., Proc, U. S. Nat. Museum, vol. 58, 1920, pp. 22-24. 16 Sitz. Kais. Akad. Wien, vol. 88, 1883, p. 253. ART 21 ORIGIN OF METAL IN METEORITES—MERRILL ; view of the Researches of F. Flade.17 The metal in this case, it will be remembered, is shown to have been reduced from magnetic pyrites. That, however, in the meteorite it was not derived from the sulphide is shown apparently by the fact that the latter is the later formed mineral of the two. Objection to such a possible source might be raised on account of the large amount of chloride demanded to produce the 10 per cent and upward of metal contained by the average stone. (Lawrencite, FeCl,=Fe 40.1 per cent, Cl 55.9 per cent.) Could it be allowed, however, it would be an aid in accounting for the enormous quanti- ties of sodium chloride in seawater and locked up in the rocks of the earth’s crust- EXPLANATION OF PLATES PLatTe 1 Fig. 1. Anthony (Kans.) stone. (1) Troilite; (2) nickel-iron. Dark, nearly black areas, silicates. 2. Dark inclosure in Cumberland Falls stone. Small white dots and stringers (1) are metal. Dark areas, silicates. 8. Admire (Kans.), pallasite. (1) Nickel-iron, (2) schreibersite, (3) troilite. Dark areas, olivine. 4. Four Corners (N. Mex.), iron. (1) Nickel-iron, (2) granular admixture of silicates and metal. PLATE 2 Upper. Polished slice of Admire pallasite, showing clastic structure and shat- tered condition of olivines. Natural size. Lower. Polished slice of Estherville mesosiderite, showing shrinkage cavities black, metal white, silicates dark gray. Enlarged about four diam- eters. 1 Silicate; 2 metal; 3 cavities. PLATE 3 Upper. Structure of Mount Joy meteroric iron—a coarse kamacite octahedrite— after fusion. Lower. Structure of Canon Diablo iron—a coarse octahedrite—after fusion. 17 Das Biihleisen hat alle Higenschaften cines extrenr niedrig gekohlten Schmiedeeisen, is infolge dessen ausserordentlich ziihe und dehnbar, aber nur schwer mit der Gestein- schneidemaschine oder mit der Siige zuzerkleinern. (Senkenbergia, vol. 2, Heft 5, Aug. 15, 1920.) O OIL ava elf bet # Lei bis hia hat “Ye Jowoods WO Neti ik sh eh a road th pte ‘or ‘boli i: mae ia ay, a oD tT Nah Pehk }y ai zt UID fo te a4 1 ¢ A mat ri Ca As RO \ ve i t ' H at. ‘ a> 4 4 hy! » : i i +5 ; fey? f ee” a as A a 4 porn Weal: sot-fostolse 12) 3 Up) rd ay fi ‘ iu ° “bas i’ Riu Atha inert) ecote Ris : Ah é rite , was YB, set, i ‘ ey ekeoediomion P22 otgaat hie 4 ballad, GROAN AY Vek RRR Me BAS: Laie a OS ie p. Neg “| pees y nal a Se du) We som 2d ath soionays (Sy. ttre joa isk) ( ihe nner ‘ : i : { ede 4 ie. ahCe my Ry Sod Rs nish Nd AS AA ssp . cvastenrt bia wrote, ot he 9) eating exis ios ne ' ' > 4 ny whe Ta eins rte *” ny ie “iy hi yh, Ki ‘ yb py ee Ke id 47 Reiss vase Some SECA EL iN aR Meendt within! { it +i i Sad | is Dowtriaht!o wahelatthh ealpaiiie. aatite Astana’ ty ‘, rey 8 Bor; ik, ei * 7 hips $5 gtanitie uo : iy yess 6 wai by WY eR te ee iwAgaies Sint e- 9 | ne vane 3 HY asi Whi ae ea eG ES LE, ORT DENT. Ber OTE OARS EL LT. ei i uso te by tae ; EA DERURIN AGT UM i OB . 4 : A ; cae y ree 4 ae; ter pany RD Pops Sa PES ae ED p ‘ URS FLERE TS ee SE EGS arse WRITE ora! Y x Ue +o A so tC 4 ~ - Hy — i \ af ph) i ts da f TiOket. Bre i : " eo Sy AACA AG 7 ‘ RY ae 2 0 alae bk Bid%y iy TEE et obydeegie 2a tare, a rn i s rig My i .; ‘ \ : \ we . ‘fi a , , U. S. NATIONAL MUSEUM PROCEEDINGS, VOL. 73, ART. 21 PL. 1 ORIGIN OF METAL IN METEORITES FOR EXPLANATION OF PLATE SEE PAGE 7 U. S. NATIONAL MUSEUM PROCEEDINGS, VOL. 73, ART. 21 ORIGIN OF METAL IN METEORITES FOR EXPLANATION OF PLATE SEE PAGE 7 PES U. S. NATIONAL MUSEUM PROCEEDINGS, VOL. 73, ART. 21 ORIGIN OF METAL IN METEORITES FoR EXPLANATION OF PLATE SEE PAGE 7 PL. < TERTIARY FOSSIL PLANTS FROM THE ARGENTINE REPUBLIC By Epwarp W. Berry Of Johns Hopkins University, Baltimore, Md. INTRODUCTION There are in the United States National Museum several small lots of rather indifferently preserved fossil plants collected by Chester W. Washburne in the Territories of Rio Negro and Santa Cruz during the explorations of the Hydrological Survey made for the Govern- ment of Argentina under the direction of Bailey Willis in 1911-1913. All are impressions of foliage, for the most part fragmentary, and preserved in clayey or sandy tuffs. The character of the material and its small amount render it impossible to deduce any far-reaching con- clusions; nevertheless, considerable that is of interest has resulted from its study. In striking contrast with the wealth of information regarding the Tertiary terrestrial faunas of Patagonia, very little is known about the contemporaneous terrestrial floras. In 1899 Dusén described? a ' small and rather poor collection of plants of Tertiary age from what he called the Fagus and Araucaria zones from several localities on both sides of the Strait of Magellan, and in 1925 I described’ a rather well preserved collection from Chubut Territory which ap- peared to have come from the so-called Santa Cruz formation. LOCALITIES The present collections came from the following five localities— three in Rio Negro Territory and two in Santa Cruz Territory—and the only information I have regarding them is contained on the labels accompanying them. The Rio Negro localities are all in the vicinity of Lago Nahuel Huapi, and with the collectors numbers are: 176. Folded tuffs 4°km. west southwest of Bernal (4 leagues southeast of Barriloche) ; 1Dusén, P., Svenska Exped. Magellansliinderna, vol. 1, No. 4, 1899: . 4 Berry, Edward W., Johns Hopkins University Studies in Geology, No. 6, 1925. No. 2743.—PROcCEEDINGS U. S. NATIONAL Museum, VOL. 73, ART. 22 95864—28——-_1 1 2 PROCEEDINGS OF THE NATIONAL MUSEUM VOL. 73 196. Tuff (thin bedded SS) on axis of anticline, 214 leagues above emboucheur of canyon of Rio Nirihuao into basin of Lago Nahuel Huapi; and god AT ite Tits Tm SERS 7 wz rErs <1 66 wd Gr. 64 62 Fic. 1.—LocaLitigs IN Lago NAHUEL HuApPiI REGION, RIO N®GRO TERRITORY ; 2. LocaLiry oF MirHOJA, CHUBUT TERRITORY; 3. LOCALITIES IN RIO CHALIA REGION, SANTA CRUZ TPRRITORY 198. Southeast side of Rio Nirihuao, 114 leagues above foot of can- yon, 150 yards southwest of Casa Piedra (4 leagues south of Lago Nahuel Huapi). ART. 22 TERTIARY FOSSIL PLANTS FROM ARGENTINA—BERRY 3 The two Santa Cruz localities are: 112. (One league north of Estancia Chalia) (Bob. Lively’s place in lot 77), Rio Chalia; and 116. (Bluff 44 league south of Mata Amarilla, upper Rio Chalia.) The age of the last is given as Santa Cruzian ? FLORA A total of 27 different forms are more or less satisfactorily identi- fied and 19 of these appear to be new. They comprise 22 genera in 17 families and 14 orders, and represent 4 ferns, 1 cycad, 2 conifers, 1 monocotyledon, and 19 dicotyledons. None except form genera have furnished more than a single species and no families except the Polypodiaceae and Monimiaceae are represented by more than a single species. The largest number of forms identified from any single locality is but 9. There are 16 species recorded from the Rio Negro localities and 11 from the Rio Chalia localities. None are common to the two; even the genera are all different, and they appear to be different in age as well as in the environmental conditions which they indicate. The three Rio Negro localities have but one species common to two of them, so that they may not be of exactly the same age, but the data are insufficient to affirm or deny this, and I am considering them collectively as affording certain contrasts. with the two Rio Chalia localities. These three localities in Rio Negro Territory are all in the vicinity of Lago Nahuel Huapi and apparently from what Roth called the Piso de Nahuel Huapi. They have yielded the following florule: Alsophila antarctica. Nothofagus simplicidens. Pteris nirihuaocensis., Leguminosites calliandraformis. Filicites sp. 1. Leguminosites sp. Filicites sp. 2. Anacardites (?) patagonicus. Zamia australis. Myrcia nitens. Araucaria nathorsti. Phyllites nirihuaoensis. Scirpites sp. Phyllites mollinediaformis. Fagus (?) subferruginea. Phyllites sp. (ef. Schinopsis). This florule is too small for any accurate ecologic estimate; never- theless the Zamia is the only form that is far removed from its pres- ent-day range, and Zaméa occurs abundantly in the lower Miocene coal measures of the Arauco district in Chile, where, however, it is associated with much warmer types. A comparison with the flora described from Mirhoja, Chubut Territory, nearly 2° farther south, shows no common species between the two and none of the meso- phytic warm types of the latter, so that the present florule must be considered to be a distinctly cooler temperate flora. Compared with 4 PROCEEDINGS OF THE NATIONAL MUSEUM VOL. 73 the existing flora of the Lago Nahuel Huapi district, it shows no certain indications of different temperature conditions but does seem to indicate considerable more humidity and an environment more like that found at the present time west of the Andes in Chile. If this conclusion is valid, it would mean less extremes of temperature throughout the year as compared with present conditions in Rio Negro Territory. The florule from the two localities in the Rio Chalia district of western Santa Cruz Territory includes the following 11 species: Adiantum patagonicum. Laurelia amarillana. Fitzroya tertiaria. Laurophyllwum chalianum. Rollinia (?) patagonica. Apocynophyllum chalianum. Hydrangea (?) incerta. Bignonites chalianus. Sterculia washburnii. Phyllites sp. 6 (?). Peuwmas clarki. Although occurring about 8° farther south, it comprises a much more northern and warmer climate assemblage than the previous florule, Fitzroya being the only species that seems distinctly at home in this latitude at the present time and then only in the wet environment of the Chilean side of the Andes. One species, Pewmus clarki, and sev- eral genera are common to the flora described from Mirhoja, Chubut Territory, and point to the present flora as having lived in a humid warm temperate environment. INDICATIONS OF AGE From what has been said in the preceding paragraphs, both the genera represented and the environment which they indicate point to these florules being of different ages. So much seems perfectly clear. Whether either or both should be considered Oligocene or Miocene is not so clear. The whole general question of the age of the Pata- gonian sedimentaries has given rise to a remarkable diversity of opinion, the principal contributors having been Ameghino, Roth, Gaudry, Scott, Hatcher, Ortmann, von Ihering, Wilckens, Cossmann, Wiman, Windhausen, and Matthew. The statement by the last- named author * is one of the most recent and the most useful summary. In a recent paper Schiller * mentions well-preserved dicotyledonous leaves near Barriloche overlain by tuffs partly silicified, from which he enumerates 25 species of marine mollusca representing the Pata- gonian stage. From this there is some reason for supposing that 8 Matthew, W. D., in Climate and Evolution. Annals N. Y. Acad. Sci., vol. 24, pp. 171-318, 1915. * Schiller, W., El Cerro ‘‘ Ottosh6he” de Bariloche. Bol. Acad. Nac. d. Ciencias Argen- tina, vol. 30, pp. 335-339, 1927. ART. 22 TERTIARY FOSSIL PLANTS FROM ARGENTINA—BERRY 5 the three plant localities in the Lago Nahuel Huapi region are older than the Patagonian marine beds. This coincides with my former and present conclusions based upon a study of all available evidence, although, as has been frequently pointed out, this evidence is far from complete. In former contributions * I have considered Dusén’s Fagus zone to be upper Eocene or lower Oligocene, since it occurs below the marine Magellanian, and his Araucaria zone to be upper Oligocene, since it occurs above the Magellanian and below the Patagonian.® The second might be lower Miocene, but since its flora is so unlike the lower Miocene floras of Chile I have thought it to be older. Table of distribution Rio Negro Santa Cruz] 2 g Territory Territory | 3 3 Bib 6 iL gil! 4 2/8/38) ¢ elelelale|/2/#|/2|2 onl onl onl baal mo ‘S) me oe PUlpedeedee be | fo (2 hs bok Snipewtoe: | sev] ie | eel LB. Wh Sh lie ties ge iy Sie Nsseaiace il aterdall Men Alsophila antarctica__-_----------- [sete se gil eee a ood eb eae tle Jozi 248 x Adiantum patagonicum_-__--------- Eee jem SEE Pa ae ae hate aes Pieris ‘nirthuaoensis!.__ _ 2-2 5 _| SEE SOME 2 ES cs Sa path) Widiecttes| Species lie, 2 = 4 P22 ete P92 OT eases ae wy © deere ees os eeeee = eee | ees bal ITICTUCSESDC CICS => Se ee Pe |B xo4a | pence |r ees | a es heer es] Ss Oe Amiga australis. = SS SPARS & ESSE SIR SUAS SS EE) At a oe PAU UCU IGNOU NORStbs: a2 sso et Dp ee ered Prat Oe ORT eae he (f Sie (ee pall o) A Pi croumNentanGg 2 s--2---2----- Bee Ce ete Ul, Skettis cr oles cet ete Saal ae ISGIRDILES-SPEClEBe se ales hee ETE EL: Ries 2 Bd EM eS Soa EO EA EE EY oe ay wea Wh Fagus subferrunginea__...--____-- 2 < ame ae ee | Maar eee eel ee oleae Nothofagus simplicidens__________- Re pe oa a eae fe aa |e Coal tare Rollinia (?) patagonica_____------- Stina parte eees | ETI J Pest = ste Hadrangea (?) wncerta.._.-..---_=- aa (OE SN ape es ea a Sed ee eS Leguminosites calliandraformis_-__--_'---- Se ae Brae eset een oe De Ste: Leguminosites species___-_-------- GB si Nes XX patatebta j Sit | Se reest pete Anacardites (?) patagonicus__-_-_--_-- [ri ON pee SS Seays Gate Ora SS SCRCULUG WOSNOUNNUUS 2 a esa pen od Ned a Haina mae YR WA Rabe eUMUStClUnh tae 2s POS es sified Ep tp eee aX aes | eee rae ts Laurelia amarillana____----------- eae Se teers MeO ne re een A ee (ee mS Laurophyllum chalianum___------- espana (Sele Na Sl a eae eee aes iret Meprotornitense218 1s. 262 oes hi xX alee eee Feit) rites e FST |e ae Apocynophyllum chalianum___----- ‘See So RSS PEDRO fe new [rea a ae ass ibignonies chalianus_=- = 2 2_=---* ah lat Ai PAR gl rg Fe rE a Phyllites nirihuaoensis___-_-_------- Tepes tyes |e ea Kethsaeees la Se Seerel(e 1420) ies Beat Phyllites mollinediaformis___------ ---- bh Heol aie syle ® Se ae Zee Pralitites species (ci. Schinopsis)£-__|) “-_|-22-) X |----|-=--|----|----|---- spp Pryltesspecies 6. =. 2-4-2. ss (ease re, pared Perma persth, hae Bes) Erbe & ese | } } Tf the Araucaria and Fagus zones of Dusén have any stratigraphic validity, then the florule from the Lago Nahuel Huapi region is to be correlated with these zones, since as the accompanying table of 5In First Pan Pacific Congress Proc., pt. 3, pp. 845-865, 1921. 6 According to the sections given by Hatcher and Nordenskidéld. 6 PROCEEDINGS OF THE NATIONAL MUSEUM VOL. 73 distribution shows 4 of the 16 species are identical with forms de- scribed by Dusén from these zones in the Straits of Magellan region and two additional are identical with forms described by this author from the Seymour Island Tertiary, the present occurrence repre- senting their most northern known range. This may be stated in another way by saying that in the time immediately preceding the Patagonian transgression a humid and fairly cool temperature flora extended between 41° and 54° south latitude. Since I regard the Patagonian transgression as corresponding approximately to the Burdigalian stage of the European Miocene, it would mean that the Lago Nahuel Haupi fossil flora should be correlated with the lowest Miocene or the Oligocene of the Northern Hemisphere. Al- though denominated cool temperate, it is clear from its great north and south range and its possible extension to Antarctica that the climate at that time differed from that of the present in its greater uniformity and relative greater mildness in the far south. The florule found at two localities on the upper Rio Chalia is markedly distinct from the other, not only in representing entirely different genera but in lacking any species common to the Araucaria or Fagus zones. It has, moreover, a species common to the Santa Cruz (?) flora of Mirhoja in Chubut Territory. As already men- tioned, the plants have their modern relatives far to the northward of their fossil occurrence, and the leaves are individually much larger than any in the Rio Negro florule. They ttus represent an occurrence of warm temperate types in latitude 49° south. Hatcher describes lower Patagonian marine beds from the upper Rio Chalia, and, so far as chronologic terms are concerned, there is little choice between the terms Patagonian and Santa Cruz, since I believe the latter, although partly contemporaneous with the Patagonian, extends upward to a somewhat later time. Although the evidence is far from conclusive, it points to this florule being considerably younger than the other, and to its early Miocene age. The location of both this and the earlier florule are shown on the accompanying sketch map, the base of which is Wind- hausen’s map, showing the marine transgression of the Patagonian. I have also indicated the location of the Santa Cruz (?) plant local- ity at Mirhoja in Chubut Territory. It will be noted that the localities in the vicinity of Lago Nahuel Huapi, which I regard as belonging to the pre-Patagonian Araucaria and Fagus zones, lie in an area which was transgressed by the marine waters of the long gulf depicted by Windhausen, that the localities on the upper Rio Chalia are interbedded in its marginal deposits, and that the Mirhoja locality, which I referred to the Santa Cruz, lies to the eastward of the Patagonian Gulf and was presumably a low-lying ART, 22 TERTIARY FOSSIL PLANTS FROM ARGENTINA—BERRY 7 country. If the map is correct the geography would favor warm cur- rents from the Atlantic and the land barrier would temper Antarctic influences. We know that there was a corresponding submergence of the Chilean littoral at this time, and there is no evidence of high mountains on the site of the Andes, which is also negatived by the floral evidence of equability and humidity. DESCRIPTIONS OF SPECIES Order POLYPODIALES Family CYATHEACEAE Genus ALSOPHILA R. Browne ALSOPHILA ANTARCTICA Christ (7) Alsophila antarctica CurisT in DusEN, Schwed. Siidpolar-Exp., vol. 3, Lief. 3, p. 14, pl. 3, fig. 11, 1908. This species was described for Dusén’s account of the Tertiary plants from Seymour Island, Antarctica, by Professor Christ of Basel, who considered it most like the existing Alsophila féeana and A. corcovadensis of southern Brazil. A single fragment in the pres- ent collection is identical with the illustration of the Seymour Island type except that it is slightly smaller, and as it is sterile it might as well be considered to represent the genus Polypodium. Occurrence.—Two and one-half leagues above emboucheur of the canyon of Rio Nirihuao into the basin of Lago Nahuel Huapi, Terri- tory of Rio Negro. Plesiotype.—Cat. No. 37851, U.S.N.M. Family POLYPODIACEAE Genus ADIANTUM Linnaeus ADIANTUM PATAGONICUM, new species Plate 1, Figures 5-7 There are more or less complete specimens of four pinnules in the collection, the largest and most complete being the one shown in Figure 5; a second is only about half the size of the former. Pin- nules stipitate, nearly orbicular in outline, divided nearly symmetri- cally by narrow pointed sinuses into four principal (two terminal and two lateral) lobes and the lateral lobes more or less bisected. The distal margins are undulate. Terminal sinus widest and deep- est, extending three-fourths of the distance to the base of the lamina, 8 PROCEEDINGS OF THE NATIONAL MUSEUM VOL. 73: the adjacent lobes being inequilaterally cuneate. Lateral sinuses: similar but only about half as deep. The small sinuses dividing the lateral lobes into two unequal lobules narrow, acutely pointed, and. shallow. Stipe flat, with broad band of aggregated vascular bundles. down the middle; length, 1.3 centimeters. Lamina ranging in length from 1.25 to 2 centimeters in length and from 1.75 to 3 centimeters: in maximum width. Texture subcoriaceous. Venation dichotomous, diverging as a double dichotomy in the decurrent base of the lamina, forking successively as shown in Figure 6. The veins are relatively stout but have the appearance of being immersed in the substance of the lamina. None of the specimens are distinctly fertile, but in places the distal margin shows a decidedly thickened carbonized border, as shown in Figure 7, which may represent fructifications. Some probability is furnished this interpretation, since no such thickening is shown along the distal margin of the upper right-hand lateral lobe. In most of the specimens the distal margins are more or less frayed and do not permit any checking of these features, which, while not exactly as in living Adiantwms, are suggestively similar. The genus contains upwards of 100 widely distributed existing species in the warmer parts of the world and extending southward to Chile, Paraguay, and Argentina, in some cases (A. concinnum Humboldt, Bonpland, and Kunth) over nearly 40° of latitude, so that they can not be said to be especially influenced by temperature differences. In general, the existing species are less lobate and less equilateral than the fossil, but in the absence of more repre- sentative material showing pinnules from different parts of a frond the validity of these apparent differences can not be evaluated. About a score of Cretaceous and Tertiary species have been referred to Adiantum, including the quite dissimilar Adiantites borgoéniana Engelhardt? from the Miocene of Lota, Chile. Among somewhat similar existing species the following may be mentioned: Adiantum chilense Kaulfuss of Chile, A. pensile Kunze of Brazil, A. tenerwm Swartz of Mexico and the Antilles to Peru, and A. concinnum Humboldt, Bonpland, and Kunth which ranges from Central America to Chile. Perhaps as similar a recent form as any is the old world Adiantum capillus-veneris of Linnaeus. Occurrence—About 3 miles north of Estancia Chalia, Rio Chalia, Territory of Santa Cruz. Holotype and paratypes—Cat. No. 37852, USNM. 7 Engelhardt, H., Abh. Senck, Naturf. Gesell., vol. 16, Heft 4, p. 644, pl. 2, figs. 6-9, 1891. ART. 22 TERTIARY FOSSIL PLANTS FROM ARGENTINA—BERRY 9 Genus PTERIS Linnaeus PTERIS NIRIHUAOENSIS, new species Plate 1, Figures 3, 4 Based upon small fragments of pinnae, habit of frond conse- quently unknown. Pinnae linear-lanceolate, divided nearly to the rachis into relatively long linear, ultimately pointed segments. The sinuses are usually nearly symmetrically rounded and narrower than the segments, but in some instances in maximum-sized fragments the proximal lower margin of the segment is decurrent for a con- siderable distance, subtending a space wider than the width of the segments. The rachis is stout, prominent, and somewhat flexuous. Margins entire, but faintly undulate. Texture subcoriaceous. Mid- veins of the segments diverge from the rachis alternately at wide angles and continue to the tips of the segments. They give off at acute angles numerous laterals, the distal of which are simple sub- tended by once forked laterals and these in turn by twice forked laterals. This typical venation is not constant, however, for in a great many instances there are cross connections resulting in a reticulate venation, as shown in the accompanying figure. The largest fragment seen is that shown in Figure 3 and is of a sterile pinnule. All of the specimens are much broken and dis- torted, but what I have considered to represent a piece of a fertile pinna is shown in Figure 4. This is slightly smaller than the sterile, but agrees with it in form and venation. What I take to represent a marginal indusium with its contained sori is a thick crust of car- bonaceous matter along the margins of the segments. I could not develop any spores or structural features in this thickened mass, and it may simply represent revolute margins. However, its appear- ance is significantly like fertile fragments of Pteris, and I have given this feature considerable weight in the identification of the fossils. The vegetative habit, as incompletely determinable from the present fossils, is shared among a large number of fern genera, among which I might mention Phegopteris, Goniopteris, Dryopteris, Cyathea, and Gleichenia as genera likely to occur in the Patagonian Tertiary. All of these differ in having inframarginal sori and the venation is not reticulate in Dryopteris (restricted), Cyathea or Gleichenia (used in a supergeneric sense). Phegopteris sometimes shows a similar reticulate venation, but it is more regular, as it is also in those species of Goniopteris which are reticulate. Many species of Pferis have the form of the fossil, and a similarly reticulate venation occurs in widely scattered forms in Asia and New 95864—28——2 10 PROCEEDINGS OF THE NATIONAL MUSEUM VOL. 73 Zealand, as well as in Central and South America. This in combi- nation with the supposed fertile specimens agree in pointing to Pteris as the genus to which the fossil should be referred. A large and rather questionable Pteris has been recorded by Engelhardt from the Arauco coal measures (Miocene) of Chile.® Occurrence.—¥olded tuffs 4 km. west southwest of Bernal, about 12 miles southeast of Barriloche, Lago Nahuel Huapi, and southeast side of Rio Nirihuao near Casa eden about 12 miles south of Lago Nahuel Huapi, Territory of Rio Negro. Cotypes.—Cat. Nos. 37853, 37854, 7 S.N.M. Genus FILICITES Schlotheim FILICITES species 1 Plate 1, Figure 1 Represented by the single fragment figured, showing a linear pinnule with central midvein and numerous thin subparallel laterals diverging from the midvein at wide angles. The material is too incomplete and poorly preserved to admit of its identification. The margin appears simple, but it may have been finely toothed. The laterals appear simple, but they may have occasionally been dichoto- mous. Naturally, resemblances could be pointed out to numerous unrelated fern genera. The fossil is perhaps most like the pinnules of the existing and wide-ranging Gleichenia (Dicranopteris) of South America. Occurrence.—Four km. west southwest of Bernal, 12 miles south- east of Barriloche, Lago Nahuel Huapi, Territory of Rio Negro. Cat. No. 87855, U.S.N.M. FILICITES species 2 Plate 1, Figure 2 A poorly preserved fragment of a fern pinna with short pinnulate lobes, rather coriaceous texture, and faint venation. The tips of the segments are frayed and may have been more elongate than they are depicted. This fern is undeterminable. It resembles a fragment from the Tertiary of Seymour Island, Graham Land, which Dusén ® called Pecopteris species 1. It might well be a Dryopteris. Occurrence.—Four km. west-southwest of Bernal, 12 miles south- east of Barriloche, Lago Nahuel Huapi, Territory of Rio Negro. Cat. No. 37856, U.S.N.M. § Engelhardt, H., Abh. Senck. Naturf. Gesel]., vol. 16, Heft 4, p. 643, pl. 2, figs. 1-4, 1891. ® Dusén, P., Schwed. Siidpolar-Exped., vol. 3, Lief. 3, p. 19, pl. 4, fig. 5, 1908. arT.22 TERTIARY FOSSIL PLANTS FROM ARGENTINA—BERRY 11 Order CYCADALES ' Family CYCADACEAE Genus ZAMIA Linnaeus ZAMIA AUSTRALIS, new species Plate 2, Figure 1 Frond tiny, ovate lanceolate in outline, about 5.5 centimeters long and 1.4 centimeters in maximum width, consisting of about 32 pairs of subopposite to alternate pinnules. Pinnules oriented at angles of about 65° to the rachis, to the top surface of which they are united by the whole width of their bases. They are entire, strictly linear in outline, and conspicuously truncate at their tips. Their texture is coriaceous and their few veined longitudinally parallel venation is very faint, possibly because the specimens show only the upper surface of the frond. That it is the upper surface seen and that the pinnules are attached to the upper surface of the rachis is shown by the fact that the bases of the pinnules nearly meet and the outline of the broader rachis can be made out beneath their proximal edges. This characteristic little form is one of the most interesting in the whole collection. It is based upon two nearly complete specimens and represents the southernmost known extent of the genus in either the past or the present. The genus Zamia, whose species are not at all beneutealie related to numerous fossil forms that have been described as species of Zamites, contains about 35 existing species, ranging from peninsular Florida, Mexico, and the Antilles through northern South America and along the eastern and in that region wetter Andean slopes to eastern Bolivia and northwestern Argentina. Zamia is the dominant existing cycad genus of the Western Hemi- sphere, and its range in the Tertiary was greater than at present, extending to latitude 36° 30’ north in the Eocene.*? A South American Pliocene species was recorded by Krasser** from Bahia, Brazil, but was not represented in the collections which I obtained from the same locality. A splendid species occurs in the Arauco coa fields (Miocene) of Chile.1? This last is much larger and quite different from the present form, which is very similar to the small species of Florida with underground stem. 2 Berry, Edward W., Torreya, vol. 16, pp. 177-179, figs. 1-3, 1916. 11 Krasser, F., Sitz. k. Akad. Wiss. Wien, vol. 112, ab. 1, p. 853, 1908. ~ Berry, Edward W., Johns Hopkins Studies in Geology, No. 4, p. 120, pl. 1, fig. 4; pl. 2, figs. 1-3, 1922. 1 Ws PROCEEDINGS OF THE NATIONAL MUSEUM VOL. 73 The existing southern limit of Zaméa is about 25° south latitude, the Chilean Miocene occurrence is about 38° south, and the present is in about latitude 41° 30’ south, thus over 614° south of the existing range. Occurrence.—Southeast side of Rio Nirihuao, near Casa Piedra, about 12 miles south of Lago Nahuel Huapi, Territory of Rio Negro. Holotype——Cat. No. 37857, U.S.N.M. Order ARAUCARIALES | Family ARAUCACEAE Genus ARAUCARIA Jussieu ARAUCARIA NATHORSTI Dusén Plate 2, Figures 5, 6 Araucaria nathorsti DusiN, Svensk. Exped. till Magellanslanderna, vol. 1, No. 4, p. 105, pl. 12, figs. 1-13, 1899. This species was described by Dusén from the lignitic shales near Punta Arenas, where it is so abundant that this horizon was christened the Araucaria stage. Dusén distinguishes between the leaves of sterile and fertile twigs, the latter being usually larger, broader, and more ovate (triangular of Dusén), with a broader decurrent base. The smaller leaf figured is identical with Dusén’s, Figures 5 or 11, but the larger, which predominate in the present collection, are larger and more produced pointed than any he has figured and have more veins than the smaller leaf. The smaller have 12 to 13 veins and the larger about twice as many. While inclining to doubt the possibility of assigning the detached fossil leaves to sterile or fertile shoots, I see no reason to doubt that the present collection represents the same species as that described from the Straits of Magellan. Dusén compared the fossil with the existing Chilean pine, Araucaria imbricata Pavon, and the similarity is as close as might be expected. If the two occurrences are identical in representing Araucaria nathorsti, this species ranges from about 53° south to nearly 41° south, or over nearly 12° of latitude, and this is in accord with the numerous occurrences of petrified wood of the Araucarioxy- fon type which has been reported by many explorers in various parts of this region. The fossil differs from the existing Araucaria bra- siliana in its more coriaceous texture, in which respect it is much more similar to Araucaria imbricata, differing from both in the less contracted base. ART. 22 TERTIARY FOSSIL PLANTS FROM ARGENTINA—BERRY 13 Although both belong to the Colymbea section of the genus, it differs from Araucaria araucoensis Berry * in its much larger and inferentially more crowded leaves, which also differ in outline and in the width at the basal flexure. Araucaria imponens described by Dusén* from the Tertiary of Seymour Island, Antarctica, has much the general form of the larger leaves from Rio Negro Territory with a wide base, but is more dis- tinctly lanceolate and has fewer veins—not very constant or im- portant features. Its describer considers the Antarctic species to be closer to Araucaria brasiliana than to Araucaria imbricata; but as it is represented by very meager material, there is little basis for an opinion. Occurrence—Four km. west southwest of Bernal, about 12 miles southeast of Barriloche, Territory of Rio Negro. Plesiotypes.—Cat. No. 37858, U. S. N. M. Order PINALES Family CUPRESSINACEAE Genus FITZROYA Hooker f. FITZROYA TERTIARIA, new species Plate 2, Figures 2-4 Leafy twigs, slender, branching; covered with ovate, pointed, appressed, imbricated leaves, with broadly decurrent bases. The phyllotaxy can not be determined, but as the leaf points usually rise to different levels the effect is of a spiral phyllotaxis, what- ever the arrangement at their insertion may have been. (/’. pata- gonica is said to have the leaves in alternate trimerous whorls.) These leaves are flat or convex with the contour of the twigs and not keeled; some are slightly divergent, and perhaps a majority have recurved tips, but the habit is much more appressed than in the single specimen of Fitzroya patagonica that has been available for comparison. In some fragments where the plant substance is preserved it is seen to be coriaceous and shows traces of a wide but not prominent midvein as in the recent species. Although only sterile twigs have been seen and no microscopial preparations have been made there is little doubt but that these 12 Berry, Edward W., Johns Hopkins Studies in Geology, No. 4, p. 122, pl. 3, figs. 1-4, 1922. 4% Dusén, P., Wiss. Ergeb. Schwed. Siidpolar-Exped. 1901-1903, vol. 3, Lief. 3, p. 11, pl. 1, figs. 16, 17, 1908. 14 PROCEEDINGS OF THE NATIONAL MUSEUM VOL. 73 twigs represent the genus /ztzroya. Their place of occurrence and associates point to this conclusion, as do the correspondence in size of leaves, in their broad decurrent bases and general form and midvein. They are commonly somewhat more pointed than the leaves of the small amount of recent material seen and are also more appressed. The species is said to show considerable variability in the degree of crowding or spreading of the leaves. The genus is an interesting one, the modern distribution of which has suggested that it was a relict genus, but no fossil species have heretofore been described, to my knowledge. There is also some difference of opinion regarding its position among the Coniferales. It has usually been associated with the Actinostrobinae—all of whose members have a unique disconnected range—but it is by some authors 7° removed from association with Actinostrobus, Callitris, and Widdringtonia and referred to the Cupressinaceae. Fitzroya pata- gonica is a mesophytic type of the Chilean temperate rain forest, reaching its northern limit at about latitude 40° near Valdivia and extending southward nearly to the end of South America, overlapping slightly the western frontier of Argentina and southern Patagonia, where the environment is suitable. It is a large tree and reaches its largest size in palustrine environments. A second species confined to Tasmania was described by Hooker in the monotypic genus Diselma. It has usually been considered to belong to the same genus as the Chilean tree, but recently it has been proposed to revive the genus Diselma for its reception. The question is one involving a great deal of personal equation, and whichever view finally prevails, there can be no doubt of the similarity and probable relationship between the two. Occurrence-—About 3 miles north of Estancia Chalia, Rio Chalia, Territory of Santa Cruz. Holotype—Cat. No. 37859 U.S.N.M. Class MONOCOTYLEDONAE MONOCOTYLEDONAE INCERTAE SEDIS SCIRPITIS species Dusén (7?) Plate 3, Figure 15 Scirpitis species DusEN, Schwed. Stidpolar-Exped., vol. 3, Lief, 3, p. 16, pl. 2, fig. 6, 1908. ~ The coarseness of the parallel veins and the lack of a midvein stamp these remains as stem fragments. They vary considerably in size, the fragment figured being the largest seen. As far as one may 18 Seward, A. C., Fossil Plants, vol. 4, p. 124, 1919. ART. 22 TERTIARY FOSSIL PLANTS FROM ARGENTINA—BERRY 15 judge from figures, this is identical with what Dusén considered to represent a Scirpus-like sedge from the Tertiary of Seymour Island, Antarctica. Obviously, little reliance can be placed upon this similarity, since both occurrences represent a type which might have been present in this region at any time from the Upper Cretaceous to the present and which could scarcely be expected to show differential characters. Occurrence.—Southeast side of Rio Nirihuao near Casa Piedra, 12 miles south of Lago Nahuel Huapi, Territory of Rio Negro. Cat. No. 37860 U.S.N.M. Class DICOTY LEDONAE Order FAGALES Family FAGACEAE Genus FAGUS Linnaeus (?) FAGUS SUBFERRUGINEA Dusén Fagus subferruginea DuSEN, Svenske Exped. Magellansliinderna, vol. 1, No. 4, p. 94, pl. 8, figs. 1-8, 1899. This species was described by Dusén from near Punta Arenas and Barancas de Carmen Sylva in the Magellan Strait region. It is rep- resented by a number of specimens in the present collection. I can see no valid reason for referring it to Yagus instead of Nothofagus. It is larger than the majority of leaves of Nothofagus and does resemble the leaves of the European beach; but occurring as it does in a profuse occurrence of Vothofagus, it is most unlikely to repre- sent the northern genus. Occurrence.—Four kilometers west southwest of Bernal, 12 miles southeast of Barriloche, Lago Nahuel Huapi, Territory of Rio Negro. Cat. No. 37861, U.S.N.M. Genus NOTHOFAGUS Blume NOTHOFAGUS SIMPLICIDENS Dusén Plate 2, Figures 7-9 Nothofagus simplicidens Dustin, Svenske Exped. till Magellansliinderna, vol. 1, No. 4, p. 100, pl. 9, figs. 20-25, 1899. This species was named—it can hardly be said to have been de- scribed—by Dusén in 1899. His material was abundant and came from the following localities in the vicinity of the Strait of Ma- gellan: Barancas de Carmen Sylva, Rio Beta, Rio Condor, Bagnales, and near Punta Arenas. 16 PROCEEDINGS OF THE NATIONAL MUSEUM VOL. 73 Leaves, small, subcoriaceous, ovate in general outline, with an acute apex and a slightly inequilateral cuneate base. Margin with somewhat variable but invariably simple and relatively large teeth, one to each secondary. These teeth are prevailingly dentate, from which they grade into serrate, and some specimens approach what might be called crenate-serrate. Length ranging from 1.75 to 3.5 centimeters. Maximum width ranging from 0.75 to 1.5 centimeters. Petiole stout, usually missing, but preserved in one specimen to a length of 3.5 millimeters. Secondaries 10 to 12 pairs, opposite to alternate, thin, straight, subparallel or slightly divergent, craspedo- drome, usually ascending at angles of about 45°, but subtending somewhat greater angles in the smaller and relatively wider forms. Tertiaries percurrent. Aerolation obsolete. The genus Vothofagus comprises about 17 existing austral species, confined to southern Chile, Patagonia, and Tierra del Fuego in the Western Hemisphere and to southern Australia, Tasmania, and New Zealand in the Eastern Hemisphere. According to Skottsberg (1915), the recent species are distributed as follows: 6 in New Zealand, 1 in Tasmania, 1 in Tasmania and Victoria, 1 in New South Wales, and 8 in South America. The genus is divided into evergreen and deciduous sections. The deciduous species comprise 1 in Tasmania and 5 in South America. The evergreen section con- tains 3 in South* America, all 6 of the New Zealand species, 1 in Tasmania and Victoria, and 1 in New South Wales. They are obviously related to Fagus of the Northern Hemisphere and for a long time were referred to that genus. Fagus has been recorded as a fossil associated with Nethofagus in South America, Australia, and New Zealand, but it may be questioned if the two can be separated on the basis of leaf form alone. Several forms resembling Nothofagus have been found in the Tertiary of Europe, but are equally unreliable. Recently Bandulska*® has described a Noth- ofagus from the Eocene of southern England, basing her determina- tion upon the cuticular-structure which she claims to ke able to differentiate from that of Fagus. A large number of fossil species have been described from the Tertiary of the regions where the living species occur; at least two occur on Seymour Island, Antarctica.’ In considering the 13 species and varieties which Dusén has described from the Tertiary | of Patagonia and Tierra del Fuego, one is impressed with the thought that perhaps the majority of these are the slightly varying leaves of a much fewer number of botanical species. 1@ Badulska, H., Journ, Linn. Soc. Bot., vol. 46, p. 433, pl. 39, fig. 20, 1924. 17 Dusén, P., Wiss. Ergeb. Schwed. Siidpolar-Exped. 1901-1903, vol. 3, Lief. 3, p. 10, pl. 1, ‘figs. 10, 12, 19% pl. 3, figs. 7-9; 1908. SEO 4rT. 22 TERTIARY FOSSIL PLANTS FROM ARGENTINA—BERRY 17 The present species is not uncommon in the collection studied. In the existing flora Nothofagus extends northward on the wetter Chilean side of the Andes to about latitude 33°. Occurrence.—Southeast side of Rio Nirihuao near Casa Piedra, about 12 miles south of Lago Nahuel Huapi, Territory of Rio Negro. Plesiotypes.—Cat. No. 37862, U.S.N.M. Order RANALES Family ANONACEAE Genus ROLLINIA St. Hiliare (7?) ROLLINIA (?) PATAGONICA, new species Plate 2, Figure 11 Leaves of small size, ovate-elliptical in outline, widest medianly, with a narrowly rounded apex and a cuneate base. Margins entire. Texture subcoriaceous. Venation obsolete, due to carbonization of the lamina during fossilization. Length about 4.5 centimeters. Maximum width about 2.25 centimeters. Petiole short and stout, be- tween 5 and 6 millimeters in length. Midvein straight, relatively stout and prominent. Secondary and tertiary venation not visible. It is rather hazardous to attempt an identification of this leaf as representing the genus Follinia. It agrees rather well with the exist- ing Rollinia parvifolia of northeastern Argentina. The genus has about a score of existing species of shrubs and trees in the warmer parts of South America. A single fossil species not very different from the present form has been described from the Pliocene of eastern Brazil."* Occurrence.—Bluff about 114 miles south of Mata Amarilla, Upper Rio Clialia, Territory of Santa Cruz. - Holotype—Cat. No. 37863, U.S.N.M. Order ROSALES Family SAXIFRAGACEAE Genus HYDRANGEA Linnaeus (?) HYDRANGEA (?) INCERTA, new species Plate 5, Figure 2 Leaf oval in general outline, with a rounded apex and a truncate or broadly cuneate base, widest below the middle. Margins somewhat 18 Hollick and Berry, Johns Hopkins Studies in Geology, No. 5, p. 52, pl. 2, fig. 4, 1924. 18 PROCEEDINGS OF THE NATIONAL MUSEUM VoL. 73 irregularly sublobate. Texture coriaceous. Length about 5 centi- meters. Maximum width about 3.5 centimeters. Midvein stout. Secondaries 4 or 5 irregularly spaced pairs, diverging from the mid- vein at varying angles, camptodrome. ‘ The identity of this incomplete leaf is exceedingly problematical, and it may possibly represent some member of the Bignoniaceae. Both alternatives are represented in the existing flora of northern Argentina. Hydrangea is an ancient genus, cosmopolitan in existing floras and with several well-defined Tertiary species in the Northern Hemisphere, represented in Patagonia by Hydrangeiphyllum affine Dusén.*® Occurrence.—One and one-half miles south of Mata Amarilla, upper Rio Chalia, Territory of Santa Cruz. Holotype.—Cat. No. 37864, U.S.N.M. LEGUMINOSAE INCERTAE SEDIS Genus LEGUMINOSITES Bowerbank LEGUMINOSITES CALLIANDRAFORMIS, new species Plate 3, Figures 13, 14 This species is based upon the single specimen figured, obviously representing a leaflet of some member of the alliance Leguminosae. Leaflet small, very inequilaterally obovate. Apex broadly rounded, but slightly unsymmetrical. Base sessile, markedly inequilaterally cuneate. Margin entire. Texture subcoriaceous. Length about 11 millimeters. Maximum width about 5 millimeters. Midvein stout, much curved. Secondaries thin, ascending, curved, camptodrome; three from the extreme base, a slender one inside the midvein, and two coarser ones outside the midvein. This leaflet is much like those of the genus Calliandra Bentham, which contains over 100 existing species, the majority in the warmer parts of South America. Two fossil species have been described from the Pliocene of Potosi, Bolivia. Since similarly veined leaflets occur in Cassia, Caesalpinia, and other genera of this alliance, and since the present material is so limited, it is referred to the form genus Leguminosites. Occurrence.—Canyon of Rio Nirihuao, 214 leagues above its em- boucheur in the basin of Lago Nahuel Huapi, Territory of Rio Negro. Holotype—Cat. No. 37865, U.S.N.M. 2 Dusén, P., Svenska Exped. till Magellansliind, vol. 1, No. 4, p. 102, pl. 10, fig. 5, 1899. aRT.22 TERTIARY FOSSIL PLANTS FROM ARGENTINA—BERRY 19 LEGUMINOSITES species Plate 3, Figure 17 Leaflet small, inequilaterally subelliptical, with about equally rounded apex and base. Margins entire. Length about 9.5 milli- meters. Maximum width, above the middle, about 3.5 millimeters. Petiolule stout, curved, about 1 millimeter in length. Midvein not prominent. Balance of venation obscure. This obviously represents some member of the Leguminous alliance, but it affords no reliable generic features, and is therefore referred to the form genus Leguminosites. Occurrence.—Southeast side of Rio Nirihuao, 150 yards southwest of Casa Piedra, about 12 miles south of Lago Nahuel Huapi, Terri- tory of Rio Negro. Cat. No. 37866, U.S.N.M. Order SAPINDALES Family ANACARDIACEAE Genus ANACARDITES Saporta (7?) ANACARDITES (?) PATAGONICUS, new species Plate 5, Figure 1 This is one of those leaves that is not especially characteristic. Its sessile base and slightly falcate outline suggest that it may repre- sent a leaflet of a pinnate leaf. I identify it with considerable hesi- tation as a leaflet of some member of the Anacardiaceae which contains several South American genera similar to the fossil. Medium-sized, ovate-falcate in outline, widest medianly, with an acute tip and a slightly inequilateral more narrowed, sessile base. Margins entire. Texture subcoriaceous. Length about 5.5 centi- meters: Maximum width about 1.75 centimeters. Midvein stout, prominent and curved. Secondaries thin, numerous, subparallel, camptodrome. The generic name used was proposed by Saporta for leaflets of this family whose generic assignation was uncertain. A score of species have been described from the Tertiary of North America and Europe. Fossil species in South America have been found in Trini- dad and Brazil. Occurrence-—Southeast side of Rio Nirihuao, 150 yards southwest of Casa Piedra, about 12 miles south of Lago Nahuel Huapi, Terri- tory of Rio Negro. Holotype.—Cat. No. 37867, U.S.N.M. 20 PROCEEDINGS OF THE NATIONAL MUSEUM VOL. 73 Order MALVALES Family STERCULIACEAE Genus STERCULIA Linnaeus STERCULIA WASHBURNII, new species Plate 4, Figures 1-7 Since these leaves are variable and abundant at a single outcrop, I have considered that they represent a single species of Sterculia, a genus whose leaves are notoriously variable as to form and loba- tion. It may be described as follows: Leaves variable in size and form, palmately three to five lobed. Lobes ovate with rounded tips or conical with acute tips. Sinuses openly rounded, shallow or extending about halfway to the base. Base ranging from decurrent to cuneate to truncate, depending on the number and attitude of the lobes, which may be directed obliquely upward or laterally. Petiole long and exceedingly stout, in one specimen preserved for a length of 3 centimeters. Margins entire. Texture coriaceous. Length ranging from 4 to 8 centimeters. Maxi- mum width ranging from 2.5 to 8 or more centimeters. Primaries three, stout, diverging from the base, or subbasal in some of the forms with a decurrent base, diverging at acute angles. In the five- lobed forms the lateral primaries give off a short distance above their base a stout lateral which runs to the tip of the lower lateral lobe. Secondaries thin, camptodrome. Tertiaries thin, simple and percur- rent, or flexed medianly, or sometimes forked medianly. Named for the collector, Chester W. Washburne. This species has the general features of leaves of this genus, which first appear in considerable abundance in mid-Cretaceous floras of various parts of the world. The genus is common in the warmer parts of South America, at the present time ranging southward to the Argentine Mesopotamia (about latitude 30°). A second fossil Argentine species, not unlike but perfectly distinct from Stereulia washburnii, has been described from the supposed Santa Cruz beds of Chubut Territory.” Occurrence.—Bluff about 114 miles south of Mata Amarilla, upper Rio Chalia, Territory of Santa Cruz. Cotypes—Cat. No. 37868, U.S.N.M. 20 Berry, Edward W., Johns Hopkins Studies in Geology, No. 5, p. 220, pl. 9, figs. 5, 6, 1925. arT.22 TERTIARY FOSSIL PLANTS FROM ARGENTINA—BERRY oF Order LAURALES Family MONIMIACEAE Genus PEUMUS Persoon PEUMUS CLARKI Berry Plate 2, Figure 10 Peunwus clarki Berry, Johns Hopkins Studies in Geology, No. 6, p. 204, pl. 5, fig. 2, 1925. This species was described from the supposed Santa Cruz beds of Mirhoja, Chubut Territory. The specimen from Santa Cruz terri- tory is similar to the type in every respect except that it is slightly ~ narrower, in consequence of which the sessile base is more acute. The genus is monotypic in the existing flora of Chile, ranging, as an evergreen tree, from about latitude 30° to latitude 42°. The present fossil occurrence carries its range much farther south and extending it over nearly 7° of latitude in Argentina. In this con- nection it is possible that the leaf from the Tertiary of Seymour Island, Antarctica, described by Dusén*! as Phyllites species (2), may represent a second fossil species of Peumus. Occurrence.—Bluff about 114 miles south of Mata Amarilla, upper Rio Chalia, Territory of Santa Cruz. Holotype.—Cat. No. 37869, U.S.N.M. Genus LAURELIA Jussieu LAURELIA AMARILLANA, new species Plate 5, Figure 3 Leaf broadly lanceolate in outline, widest medianly and about equally pointed at the apex and the base. Base narrowly cuneate and decurrent. Margin entire for its basal third, above which it has somewhat irregularly and widely spaced undulate-crenate teeth. Texture coriaceous. Length about 6 centimeters. Maximum width about 2.5 centimeters. Petiole stout, 7 to 8 millimeters in length. Midvein stout, prominent on the under side of the leaf. Secondaries about 5 alternate pairs, diverging from the midvein at acute angles, thin, long ascending, inclined to be somewhat flexuous, camptodrome, but sending branches into the marginal teeth. Areolation obsolete, a few tertiaries seen, as shown in the illustration. This greatly resembles the existing Chilean Laurelia aromatica Sprengel in all of its features. The genus contains three species in 21 Dusén, P., Wiss. Ergeb. Schwed. Siidpolar-Exped., vol. 3, Lief. 3, p. 16, pl. 1, fig. 15, 1908. 22 PROCEEDINGS OF THE NATIONAL MUSEUM VOL. 73 the existing flora, one in New Zealand and the other two in Chile in the region between 36° and 42° south latitude. If botanical sys- tematists are correct in their opinion that these belong to the same genus, then we are bound to presuppose that it had a geological history unless we are prepared to subscribe to the once fashionable but now absurd notion that a genus can originate more than once and in different areas. Excluding the highly problematical fossil forms from the Northern Hemisphere which have been referred to Laurelia, a fossil species has been described by Dusén 2? from the Tertiary of Seymour Island, Antarctica. This is a somewhat fragmentary specimen of a larger size, with less ascending secondaries and more pointed teeth than Lauwrelia amaril- lana, and also less similar to the existing Lawrelia aromatica. Odcurrence.—Bluff one-half league south of Mata Amarilla, upper Rio Chalia, Territory of Santa Cruz. Holotype.—Cat. No. 37870, U.S.N.M. Family LAURACEAE Genus LAUROPHYLLUM. Goeppert LAUROPHYLLUM CHALIANUM, new species Plate 5, Figure 4 Leaves elongate-lanceolate, widest below the middle, with an ex- tended gradually narrowed tip and a more abruptly acute base. Margins entire. Texture subcoriaceous. Length about 9 to 10 centi- meters. Maximum width about 1.5 centimeters. Midvein stout, somewhat flexuous, prominent on the under side of the leaf. Second- aries thin, numerous, diverging from the midvein at acute angles, long ascending, eventually camptodrome. Tertiaries obsolete. This species is represented by several incomplete specimens and is evidently Lauraceous. As it is impossible to determine its generic position with certainty, it is referred to the form genus Laurophyl- lum. It may represent the genus Vectandra, although similar leaves occur in several existing Lauraceous genera of the warmer parts of South America. A typical species of Vectandra is present in the supposed Santa Cruz beds of Chubut Territory,”* so that this genus is known to have ranged farther south during the Tertiary than it does at the present time. Occurrence —Three miles north of Estancia Chalia, Rio Chalia, Territory of Santa Cruz. Holotype.—Cat. No. 37871, U.S.N.M. 2 Dusén, P., Schwed. Siidpolar-Exped., vol. 8, Lief. 3, p. 4, pl. 1, fig. 5, 1908. 23 Berry, Edward W., Johns Hopkins University Studies in Geology, No. 5, p. 224, pl. 8, fig. 1, 1925. ART. 22 TERTIARY FOSSIL PLANTS FROM ARGENTINA—BERRY 23 Order MYRTALES Family MYRTACEAE Genus MYRCIA De Candolle MYRCIA NITENS Engelhardt Plate 3, Figures 1-9 Myrcia (Cryptomyrcia) nitens ENGELHARDT, Abh. Senck. Naturf. Gesell., vol. 16, Heft 4, p. 679, pl. 10, fig. 7, 1891. Myrtiphyllum, bagualense Dustn, Svenska Exped. till Magellansliind., vol. 1, No. 4, p. 108, pl. 11, figs. 7-9, 1899. This species was described by Engelhardt from the lower Miocene of Coronel, Chile. It is exceedingly abundant in all sizes in the present collection, and the larger leaves (such as those shown in figs. 7-9) are identical in every respect with Engelhardt’s Chilean type. With these, and grading to much smaller but otherwise simi- lar leaves, are a series of forms identical with those from southern Patagonia which Dusén described as Myrtiphyllum baqualense. Every gradation of size is represented in the present collection, and there can be no doubt but that a single botanical species is represented. It might possibly be argued that Dusén’s, which come from over 11° farther south, were normally smaller, because of the possibly more severe climatic conditions. Among the leaves re- corded by Dusén** from the Tertiary of Seymour Island, Antarctica, there is an apical fragment (PAyllites species 16) which very prob- ably represents this same species. Occurrence.—Four km. west southwest of Bernal, 12 miles south- east of Barriloche, Lago Nahuel Huapi, Territory of Rio Negro. Plesiotypes.—Cat. No. 37872, U.S.N.M. Order GENTIANALES Family APOCYNACEAE Genus APOCYNOPHYLLUM Unger APOCYNOPHYLLUM CHALIANUM, new species Plate 5, Figure 5 Leaf oblong, acutely pointed at the apex and base. Margins entire. Texture coriaceous. Length about 11 centimeters. Maximum width about 3.25 centimeters. Petiole stout, about 2.25 centimeters in length. Midvein stout, prominent. Secondaries thin, numerous, diverging from the midvein at wide angles, subparallel, ending in an acrodrome marginal vein parallel with and close to the leaf margins. 24 Dusén, P., Schwed. Siidpolar-Exped., vol. 3, Lief. 3, p. 18, pl. 2, fig. 10, 1908. 24 PROCEEDINGS OF THE NATIONAL MUSEUM ‘VOL. 73 This form somewhat resembles Myrcia costatoides Engelhardt ?° of the lower Miocene of Chile, but is much larger and stouter with a much longer petiole. It does not conform to the features of any Myrcias with which I am familiar. There are, of course, leaves of this general type in various unrelated existing genera. Some figs have similar leaves, e. g., the existing Ficus pulehella Schott, but the venation is not quite the same and the inframarginal vein is usually farther within the margins and arched from secondary to secondary. Somewhat similar leaves occur in various Myrtaceae and Gutti- ferae, but the type is especially characteristic of a considerable num- ber of genera of the Apocynaceae and can be exactly matched among existing species of Plwmzera and Allamanda. For these reasons I feel justified in referring it to the form genus Apocynophyllum. Nothing of this kind was found in the supposed Miocene flora from Mirhoja, Territory of Chubut.2° All of the existing genera with leaves like the fossil find their home in the warmer parts of South America, and none extend farther southward than northern Argentina. Occurrence.—About 3 miles north of Estancia Chalia, Rio Chalia, Territory of Santa Cruz. Cotypes.—Cat. No. 37873 U.S.N.M. Order PERSONALES Family BIGNONIACEAE Genus BIGNONITES Saporta BIGNONITES CHALIANUS, new species Plate 5, Figure 6 Leaflets of medium size, ovate, widest medianly and about equally pointed at the apex and base. Margins entire. Texture subcoriace- ous. Length about 7 centimeters. Maximum width about 3.5 centi- meters. Midvein stout, prominent. Secondaries about 5, mediumly stout camptodrome pairs; the basal pair are stoutest and opposite and run close to and parallel with the lower lateral margins to the middle of the leaf, simulating lateral primaries. Tertiaries thin, usually forming a double series of meshes between adjacent secondaries. This form presents features allying it with various existing genera of Bignoniaceae of the existing flora in the warmer parts of South America. Several genera range southward as far as northern Argentina. *> Engelhardt, H., Abh. Senck. Naturf. Gesell., vol. 16, Heft 4, p. 680, pl. 9, fig. 6, 1894. °° Berry, Hdward W., Jobns Hopkins University Studies in Geology, No. 5, pp. 185-252, pls. 1-9, 1922. ART. 22 TERTIARY FOSSIL PLANTS FROM ARGENTINA—BERRY 25 Occurrence—Three miles north of Estancia Chalia and 114 miles south of Mata Amarilla, upper Rio Chalia, Territory of Santa Cruz. Holotype.—Cat. No. 37874, U.S.N.M. DICOTYLEDONAE INCERTAE SEDIS PHYLLITES NIRIHUAOENSIS, new species Plate 3, Figures 18, 19 Leaf or leafiet tiny, orbicular in outline, with large irregular crenate teeth. Length 6 millimeters. Maximum width 4.5 milli- meters. Apparently sessile. Venation relatively enormously stout and prominent. Midvein flexuous, terminating in an apical tooth. Secondaries 3 or 4 on each side, alternate, stout, often forking, recurv- ing as marginal veins in the teeth. Areolation irregular and mostly rectangular, gradually diminishing in strength. This typical form is based on the single specimen figured. It is somewhat similar to a slightly larger fragment from Barancas de Carmen Sylva, recorded by Dusén as E'scaloniiphyllum species.?* The venation suggests a relationship with the Cunoniaceous genus Weinmannia, but this resemblance does not warrant a decision. Occurrence.—Southeast side of Rio Nirihuao, 150 yards southwest of Casa Piedra and about 12 miles south of Lago Nahuel Huapi, Territory of Rio Negro. Holotype.—Cat. No. 37876, U.S.N.M. PHYLLITES MOLLINEDIAFORMIS, new species Plate 3, Figure 16 Leaf small lanceolate, coriaceous, with toothed margin and cras- pedodrome secondaries. Based on the single specimen figured and possibly representing a narrow Nothofagus. Shows considerable re- semblance to Mollinedia seymourensis Dusén** of Tertiary of Sey- mour Island, Antarctica. Occurrence.—Southeast side of Rio Nirihuao, 150 yards southwest of Casa Piedra, about 12 miles south of Lago Nahuel Huapi, Terri- tory of Rio Negro. Holotype—Cat. No. 37877, U.S.N.M. PHYLLITES species (cf. SCHINOPSIS) Plate 3, Figures 10 to 12 More or less complete specimens of small linear-lanceolate leaves or leaflets are not uncommon at this locality. They do not show suf- 27 Dusén, P., Svenska Exped. Magellansliinderna, vol. 1, No. 4, p. 102, pl. 11, fig. 5, 1899 23 Dusén, P., Schwed. Siidpolar-Exped., vol. 3, Lief. 3, p. 4, pl. 1, fig. 18, 1908. 26 PROCEEDINGS OF THE NATIONAL MUSEUM VOL. 73 ficiently characteristic features to warrant even a surmise as to their botanical affinity. They may be briefly characterized as follows: Outline linear lanceolate, widest medianly, acutely and about equally pointed at the apex and base. Margins entire. Texture relatively coriaceous. Length 4 to 5 centimeters. Maximum width 2.75 to 7 millimeters. Petiole stout, about 3 millimeters in length. Midvein mediumly stout. Balance of the venation obscure; a few thin oblique camptodrome secondaries can be made out, thus ruling out compari- sons with members of the family Myrtaceae, which are represented in most South American Tertiary floras, including that described from the supposed Santa Cruz beds of the Territory of Chubut.?® They might represent the leaflets of some Sapindaceous genus, and they are also not unlike those of the Anacardiaceous genus Schin- opsis, which is represented by good material from the Miocene of Chubut.°*° Occurrence.—Southeast side of Rio Nirihuao, 150 yards southwest of Casa Piedra, about 12 miles south of Lago Nahuel Huapi, Terri- tory of Rio Negro. Cat. No. 37878, U.S.N.M. PHYLLITES species 6 Dusén (7?) Phyllites, sp. 6. Dustn, Schwed. Siidpolar Exped., vol. 3, lief. 3, p. 18, pl. 1, fig. 3, 1908. A specimen indistinguishable from the one recorded by Dusén from the Tertiary of Seymour Island, Antarctica, but too frag- mentary to be reliable, is present in the collection from near Mata Amarilla, upper Rio Chalia, Territory of Santa Cruz. Cat. No. 37879, U.S.N.M.) 2 Berry, Edward W., Johns Hopkins University Studies in Geology, No. 6, p. 225, pl. 2, fig. 6, 1925. 30Tdem, p. 208, pl. 1, fig. 2. Se ART, 22 TERTIARY FOSSIL PLANTS FROM ARGENTINA—BERRY 20 EXPLANATION OF PLATES PLATE 1 1. Filicites species 1. 2. Filicites species 2. 3. Sterile, and Fig. 4. Fertile pinna of Pteris nirthuaoensis, new species X 3. 7. Adiantum patagonicum, new species. 5. The most nearly complete specimen found. 6. Same, slightly restored, < 4. 7. Enlarged segment of distal margin showing supposed sori. PLATE 2 Fig. 1. Zamia australis, new species. 2-4. Fitzroya tertiaria, new species. 2. Natural size. 3 and 4. Enlarged xX 5. 5, 6. Araucaria nathorsti Dusén. 7-9. Nothofagus simplicidens Dusén. 10. Peumus clarki Berry. 11. Rollinia (?) patagonica, new species. PLATE 3 Fies. 1-9. Myrcia nitens Engelhardt. 10-12. Phyllites species (cf. Schinopsis). 13, 14. Leguminosites calliandraformis, new species 14 enlarged X 4. 15. Scirpites species Dusén (?) 16. Phyllites mollinediaformis, new species. 17. Leguminosites species. 18,19. Phyllites nirihuaoensis, new species 19 enlarged X 10. PLATE 4 Fias. 1-7. Sterculia washburnii, new species. PuatTE 5 Fic. _ . Anacardites (?) patagonicus, new species. . Hydrangea (?) incerta, new species. Laurelia amarillana, new species. . Laurophyllum chalianum, new species. . Apocynophyllum chalianum, new species. . Bignonites chalianus, new species. O Oop wd wig Mm bine 4 ieseTe : Seo Te WE Pa | e ROUSSE HTL BTM op SATE oeS asioods RaW Aaplliiecn sale , : 4 Siie dd oes ~ ssinega wert we childs Aorhuddontmes : s ‘ . ‘ ac WANT hay HOR) co wed Ag Oe Oe aS : aida Wort ky RH OBAS halen Qi, 1H a b ] U. S. NATIONAL MUSEUM WALD ee ANA NOSE PROCEEDINGS, VOL. 73, ART. 22. ARGENTINE TERTIARY PLANTS FOR DESCRIPTION OF PLATE SEE PAGE 27 PL. U. S. NATIONAL MUSEUM PROCEEDINGS, VOL. 73. ART. 22. PL. 2 NY 4 10 11 ARGENTINE TERTIARY PLANTS FOR DESCRIPTION OF PLATE SEE PAGE 27 PROCEEDINGS, VOL. 73, ART. 22. PL. 3 U. S. NATIONAL MUSEUM ARGENTINE TERTIARY PLANTS U. S. NATIONAL MUSEUM PROCEEDINGS, VOL. 73, ART. 22. ARGENTINE TERTIARY PLANTS FOR DESCRIPTION OF PLATE SEE PAGE 27 PL. U. S. NATIONAL MUSEUM | 4 ; 6 5 ARGENTINE TERTIARY PLANTS PROCEEDINGS, VOL. 73, ART. 22. PL. 5 ma FOR DESCRIPTION OF PLATE SEE PAGE 27 Ailes ae mn se a CLA ah de M4 : ve 7 ths ter ; iW fy Ps a ; Wie . i “sf $ . vy r a ‘ » . t + - * ] ‘ ‘ ! i a f t ~ ' 1 i ‘ “4 NOTES ON AMERICAN TWO-WINGED FLIES OF THE FAMILY SAPROMYZIDAE By J. R. Mariocu Of the Biological Survey, United States Department of Agriculture Through a coincidence Dr. F. Hendel and the present writer recently published papers’ dealing with South American Sapro- myzidae. Doctor Hendel included all the genera of the family known to him, while I deait only with South American species. As some of the new genera erected contain species which are not rare, it is not remarkable that several synonyms resulted, and those which I am certain of are noted below. Genus NEOMINETTIA Hendel This generic name was proposed by both authors (Malloch, p. 9), and contigua Fabricius was cited as genotype by both. As Hendel’s paper antedates mine, the genus should be credited to him. Genus DRYOMYZOTHEA Hendel My genus Dryomyzoides is the same as this, but my genotype, advena Malloch, is apparently distinct from his, se¢inervis Hendel. Genus DEUTOMINETTIA Hendel This genus was erected for a species, pulchrifrons Hendel, which has most of the characters of Minettia Robineau-Desvoidy, differing in having the scutellum haired on the disk. I have before me a species which is evidently referable to the genus; and though the face is a little more convex than in normal J/inettia, it is not so much so that I would place it in the group with markedly convex face. I have not seen the genotype, so can not say whether it has the mid tibia with well-developed bristles on the posterior surface; but if it has, and its similarity to the other included species leads me to believe so, it might be better to base the generic distinction on 1Hendel, Encyclopédie Entomol., Diptera, vol. 2, pp. 103-142, 1925, and Malloch, Proc. U. S. Nat. Mus., vol. 68, art. 21, pp. 1-35, 1926. No. 2744.—PRocEEDINGS U. S. NATIONAL MUSEUM, VOL. 73, ART. 23 96271—28——_1 5 2 PROCEEDINGS OF THE NATIONAL MUSEUM VOL. 73 the presence of these rather than on the form of the face and the setulose scutellum. In this case the genus would contain the geno- type and four species known to me—frontalis Macquart, assimilis Malloch, geniseta Malloch, and the species described below. Only the genotype and the last of these have the scutellum haired on the disk, and these two and frontalis have the frons with a pair of, velvety black marks in front. However pulchrifrons may not have mid tibial setulae. I present below a key for the separation of the above-mentioned species and the new one described herein. KEY TO THE SPECIES OF DEUTOMINETTIA 1. Frons with two black marks in front; scutellum in some species with distinct setulose hairs on disk in addition to the four marginal bristles___________ 2 Frons yellow, unspotted; scutellum with the disk bare, only the four mar- ginal ‘bristles, resents. ee a ea eal A ee ee 4 2. Wings without, dark Markings=-—=2> 22 eee bimaculata, new species. Wings with’ distinct dark°markingti0) _U9I99 19 S190 50 WS Silt te see 3 3. Only the third wing vein dark at apex; disk of scutellum setulose. pulehrifrons Hendel. Costal margin dark brown, more broadly so from opposite inner cross vein to beyond apex of fourth vein, the dark color on apices of third and fourth veins narrowly divided by a longitudinal hyaline streak inwardly, both cross veins broadly dark brown, fifth vein faintly clouded; disk of scutel- Wun Dare abel ne A eye ee A ha frontalis Macquart. 4. Cross veinsS.of wings. very, distinetly clouded2 2-2. ee 58 Cross veins of wings almost imperceptibly clouded____--___ geniseta Malloch. 5. Costa without an apical suffusion; mid tibia with about three rather long posterior bristles; hind femur with at least one evident preapical antero- ventral bristleccl 22 2f4) 2) pe gE ey assimilis Malloch. Costa with a trace of an apical suffusion; mid tibia with about seven short posterior setulae; hind femur without an evident preapical anteroventral bristle ocps<-6-n ee a ee ee ee approximata, new species. Should the presence of scutellar setulae be considered as the dis- tinctive generic character, the four species lacking these would then require to be placed in a separate genus, but the absence or presence of similar hairs has not been considered as sufficient grounds for the erection of genera in related families such as Helomyzidae. A careful examination of Hendel’s description of his genus Allominettia and its genotype, maculatifrons Hendel, leads me to conclude that this is the same species which I have identified as frontalis Macquart.? Hendel’s specimens came from Peru, while those I had came from Costa Rica. He makes no mention of the mid tibial bristles, but there is nothing remarkable in that, as this character has been ignored by all writers who have dealt with the family until the appearance of my recent papers on the Oriental species of Homoneura sens. lat. As indicated in my previous paper, Sn eee ee eee SS 2Hist. Nat. Dipt., vol. 2, pt. 8, p. 346, from Brazil. | ART. 23 FLIES OF THE FAMILY SAPROMYZIDAE—MALLOCH 3 definite decision as to the identity of Macquart’s species may depend upon the discovery of his type. e DEUTOMINETTIA APPROXIMATA, new species Male—Glossy fulvous yellow, the center of frons shining; an- tennae and palpi yellow. Both cross veins of wings distinctly but not broadly, clouded, costa apically slightly suffused with brown. Anterior orbital bristles over half as long as posterior pair; ocel- lars minute; postverticals long; arista plumose; face almost flat, glossy, orbits whitish dusted. Thoracic bristles long and strong, anterior sternopleural very short and fine. Mid tibia with about seven short posterior bristles; mid femur without anteroyentral bristles. Inner cross vein a little beyond middle of discal cell. Length, 6 mm. One male, Trinidad River, Panama, May 2, 1911 (A. Busck). Type.—Cat. No. 40965, U.S.N.M. DEUTOMINETTIA BIMACULATA, new species Male and female—tTestaceous yellow, distinctly shining. Orbital stripes glossy, ceasing at anterior orbitals, a velvety deep black mark on each side from anterior orbital to anterior margin of frons which extends posteriorly, forming a wedge-shaped mark between orbit and eye almost as far as posterior bristle; parafacials silvery. Abdomen with a faint dark apical line on each tergite in male. Legs yellow. Wings hyaline. Halteres yellow. Frons slightly widened anteriorly, with some microscopic surface hairs, all bristles long and strong; arista plumose; face slightly bulging out over mouth, a little convex; eye narrowed below, slightly emarginate on lower half behind; two strong bristles on lower part of occiput. Thorax with three postsutural dorsocentrals, about eight series of introdorsocentral hairs, the intra-alar strong, disk of scutellum black setulose, and the anterior sternopleural bristle very weak. Fore femur without an anteroventral comb; all tibiae with preapical dorsal bristle, mid pair each with a series of eight or more distinct posterior setulae. Inner cross vein of wing close to middle of discal cell; apical section of fourth vein but little longer than preceding section; first posterior cell slightly narrowed apically. Length, 6-7 mm. One male, Trinidad Rio, Panama, May 1, 1911; allotype and one male paratype, same locality, March 16 to 23, 1912 (A. Busck). Type—Cat. No. 40708, U.S.N.M. Genus ASILOSTOMA Hendel } This genus was erected for the reception of a single species, ender- lent Hendel, from Bolivia. Before me there is a species undoubt- 4 PROCEEDINGS OF THE NATIONAL MUSEUM VOL. 73 edly belonging to this genus, and another which agrees with the gen- eral description of the genotype, but differs in the bristling of the vertex and in some other respects. I deprecate the erection of monobasic genera and, as I find included in some related genera species which differ in a similar manner, I do not propose to erect a new genus for the species now before me. All of the species have the basal and second antennal segments of about the same length, the first haired below, the third not less than six times as long as wide at middle, the arista plumose, face promi- nently protruded and convex, thorax with two pairs of postsutural dorsocentrals, no presutural (posthumeral), and one sternopleural; scutellum bare, flattened above, and with four bristles. The vena- tion is different in palpalis from that of the other species. In three the wings are marked with fuscous, and the second, third, and fourth segments of fore tarsi of all five are compressed. I present below a key for the identification of the species, the geno- type being unknown to me except from the description. KEY TO THE SPECIES OF ASILOSTOMA ~ 1. Distance from bases of antennae to lower margin of face not half as great as that from bases of antennae to vertex; frons testaceous yellow, except the dark ocellar spot, face concolorous, the lower lateral angles and most of the cheeks glossy black; labrum large, its area almost as great as that of face, glossy black; palpi deep black, and much dilated; anterior pair of orbital bristles about as long as posterior pair, and close to middle of frons. palpalis, new species. Distance from bases of antennae to lower margin of face not much less than that from bases of antennae to vertex; frons and face black; labrum nar- row; palpr slender -SM Sieh ces eee Sa PERE a 2 2. Anterior pair of orbital bristles not half as long as posterior pair; legs entirely “stramimeousss ele lusetes bee ace ee eee eee 3 Anterior pair of orbital bristles fully as long as posterior pair; legs yellow, fore tibia and fore metatarsus, except apex of latter, black. enderleini Hendel. S.crrons and face. clossy Dlatkio 2. Lee ee ee ee pallipes, new species. Frons with a large velvety black mark in front, the remainder glossy black__4 4) aceventirely jelossy black_cisy_ sieves soup seh eh eee atriceps, new species, Face and cheeks entirely glossy yellow_---------_----_ flavifacies, new species. ASILOSTOMA PALPALIS, new species Male.—Head shining testaceous yellow, ocellar spot, a large mark on each side of upper half of frons, and the lower lateral angle of face and contiguous portion of each cheek, black; upper side of basal segment of antennae, and the third segment except base fuscous; palpi black, arista yellow, the hairs brownish. Thorax and abdomen brownish yellow, a dark streak over each humerus, and a velvety black vitta along upper half of pleura and sides of abdomen. ART. 23 FLIES OF THE FAMILY SAPROM YZIDAE—MALLOCH 5 Legs testaceous yellow, a mark on apices of fore femora, another near bases of fore tibiae, and the fore metatarsi except their apices black, the dilated portion of fore tarsi whitish. Wings marked with fuscous as in Figure 2. i Frons fully twice as long as wide, the surface uniform in texture: ocellars miscroscopic; cheek almost linear; face concave below an- tennae, the lower half prominently convex (fig. 1); third antennal segment about seven times as long as its width at middle. Thorax appearing finely granulose on dorsum, with two or three closely placed series of miscroscopic intradorsocentral hairs, and a series of similar hairs in line with each series of dorsocentrals; prescutellar acrostichals lacking. Fore femur without bristles or anteroventral Fies. 1-2.—1. HEAD OF ASILOSTOMA PALPALIS FROM SIDB. 2, WING oF ASILOSTOMA PALPALIS comb; tibiae with the preapical dorsal bristle very short except on mid pair. Venation as in Figure 2. Length, 5.5 mm. | One male, Barro Colorado, Panama Canal Zone, July 27, 1923 (R. C. Shannon). Type.—Cat. No. 40710, U.S.N.M. ASILOSTOMA PALLIPES, new species Female.—Head glossy black, antennae testaceous, third segment black except extreme base; aristae and the hairs, except those on basal half on upper side, white; palpi fuscous. Thorax colored as in palpalis, but without black markings. Abdomen brownish black. Legs entirely pale yellow. Wings marked with fuscous as in Figure 3. Halteres yellow. Frons a little longer than wide, posterior orbitals about middle of frons, and fully twice as long as anterior pair; antennae as in palpa- lis; face but slightly concave below antennae, and quite prominently convex below, much as in genotype, the labrum narrow, cheek about as high as width of third antennal segment. (Fig. 4.) Thorax not so long as in patpalis, the anterior pair of dorsocentral bristles much 6 PROCEEDINGS OF THE NATIONAL MUSEUM VOL. 73 nearer to suture, and the intradorsocentral hairs farther apart. Fore femora with two or three fine posteroventral bristles. Venation as in Figure 3. Length, 3.5 mm. ; One female, Trinidad Rio, Panama, March 23, 1912 (A. Busck). Type.—Cat. No. 40709, U.S.N.M. ASILOSTOMA ATRICEPS, new species Female.—Head black, upper portion of frons to below the upper orbital bristle on sides and to a point a little higher in middle dis- tinctly shining, anterior portion deep velvety black, with a white dusted mark on each margin above level of bases of antennae; face glossy, with a purple tinge; basal two antennal segments testaceous yellow, third fuscous; aristae white, the long hairs at base above dark; palpi fuscous. Thorax pale brown, more or less dusted on dorsum and not noticeably shining except on sides of mesonotum, propleura, scutellum, and metanotum testaceous yellow, a large mark in front Figs. 3—4.—3. WING OF ASILOSTOMA PALLIPHS. 4. HEAD OF ASILOSTOMA PALLIPES FROM SIDE of each wing base velvety black. Abdomen brown, shining. Legs pale stramineous, the fore tarsi palest. Wings grayish hyaline, with faint brown clouds along fifth vein, outer cross vein, apex of second vein, and on a short subapical section of third and fourth veins. Hal- teres yellow. Vertex with inner pair of bristles long and strong, outer pair Jacking, postverticals minute, anterior orbital bristles very small; third antennal segment about six times as long as wide; arista long haired basally above, the hairs decreasing rapidly in length apically, the lower side short haired; face in type damaged, but evidently not so much swollen as in some of the other species. Thorax normal, the two pairs of dorsocentral bristles long and strong. Legs long, fore tarsi thickened. Inner cross vein a little beyond middle of discal cell; second vein slightly arched with costa, third and fourth slightly convergent on apical sections to near apices. Length 2.75 mm. One female, Higuito, San Mateo, Costa Rica (P. Schild). Type.—Cat. No. 40954, U.S.N.M. ee Oa er | oe Peebles Sr ee NF 4 z *e ART. 23 FLIES OF THE FAMILY SAPROMYZIDAE—MALLOCH < ASILOSTOMA FLAVIFACIES, new species Female.—Head as in preceding species but the face and cheeks are honey yellow. The thorax is colored as in atriceps but the anterior margin is more yellowish and vittate, and the sternopleura is yellow, not dark, and the dark mark on sides of metathorax is deeper black. Abdomen shining fuscous. Wings with but faint indications of the cloud on fifth vein and outer cross veins, none on apices of other veins. Structurally similar to atriceps. Length, 3 mm. Type and one defective paratype, Higuito, San Mateo, Costa Rica (P. Schild.). Type.—Cat. No. 40955, U.S.N.M. Genus BLEPHAROLAUXANIA Hendel This genus has the base of third vein setulose to beyond the inner cross vein, the face convex below, and two pairs of normal back- wardly inclined orbitals. The most striking character of the genus is the presence of fine hairs on the upper surface of the third antennal segment which are as long as, or longer than, the width of the seg- ment itself. There is no other genus except the next following one so far described in which such hairs occur. BLEPHAROLAUXANIA TRICHOCERA Hendel This is the only known species of the genus, and occurs in Peru. It is a yellow species, with the wing veins mostly browned. The arista is very long haired and the thorax has three pairs of postsutu- ral dorsocentrals. I have not seen the species. Genus PLATYGRAPHIUM Hendel This genus lacks the presutural (posthumeral) bristle, has the third antennal segment three times as long as wide, and long haired on upper surface as in the preceding genus. It differs from that genus in having no presutural bristle, the third wing vein without bristles, and the arista with the hairs about half as long as width of third antennal segment. Platygraphium penicillatum Hendel is the only known species, and it is recorded only from Bolivia. It is yellow in color, with the abdomen browned, and has the wings grayish hyaline, with the base and costa yellowish. It is unknown to me. Genus ERIURGUS Hendel This genus lacks the presutural bristle, has the wing veins without bristles, and the third antennal segment orbicular and without hairs 8 PROCEEDINGS OF THE NATIONAL MUSEUM VOL. 73 above. The distinguishing character is the presence of long fine hairs on the posterior and ventral surfaces of the fore femora and tibiae. Hriurgus pilimanus Hendel is the only known species of the genus and occurs in Peru. It is entirely yellow in color and similar in most respects to Dryomyzothea setinervis Hendel The genotype from which the description was made is a male and it is possible the female may not have long hairs on the forelegs. Genus ALLOMINETTIA Hendel For discussion of this genus see under Deutoméinettia in this paper. Genus SCUTOLAUXANIA Hendel This genus has the scutellum with hairs above, and the stem of veins 2 and 8 setulose as in Xenochaetina Malloch. The thorax has two pairs of postsutural dorsocentrals, and the arista is long haired. Scutolauwania piloscutellaris Hendel is the only known species of the genus; it occurs in Peru.