yr hie dds BC bios as a Recher Oh itt enn aay py bits aap cr a prac | iY nH HA rs CO ovares “he Prine Ane hd da Wy eye aie Diet aston h aa a ei Ban ene} upotatn sun ith bet aa pa ia i ii i rane run pe fe i 2 ee sr iy a MA Buel ent uy i Perr enith ta Seek at nlite ii rs igi ey Hanis Pisa arts a Parte ae eae oh) vanes bed Ee naa Bian ‘ Sia aseaiespoea a a yw WN aed th bitin an vw eine Sanat ia Legh a on i bear tenes Ser eud aoe a poe PTE Ta tard eatav penne teat ienst “ay u) To ree oT 4 BY Tirbh (ety o ey et ane Sea * Rae era es aM pte et i aa Pra bby Fy ae ee oy oe ni Perit » eh pit. oy i e $ ce ohio hat ra i an io Beri ah 8 ae sun st Cyr teat Hts oe : fi rrethians Bintee ‘i be ct ta lat smith ii) ba pbeds oe Cie oe oe fe ae Hants as 2 8 Sei AH pe as) i rota i mits hy i soe bs by He NS sie ie batt os Ea NG: eH a eat Hi eae ines Rot ue yt une bred we Harta: ee ae 4 G a ts is 2 i =i : vie oe ar F at sp tUey et ee rn pk, ene ‘ salseeslatlty ee ae te ae oe 4 iS Het ce nase ti aan aa Rent SMITHSONIAN Miscellaneous Collections VOR Vi wel (QuaRTERLY IssuE, VoLumE I) 1457 “EVERY MAN IS A VALUABLE MEMBER OF SOCIETY WHO, BY HIS OBSERVATIONS, RESEARCHES, AND EXPERIMENTS, PROCURES KNOWLEDGE FOR MEN ””__ SMITHSON. Giiy” OF WASHINGTON PUBLISHED BY THE SMITHSONIAN INSTITUTION 1903 PRESS OF THe New ERA PRINTING COMPANY LANCASTER, PA VAS Dw ADVERTISEMENT The present series, entitled SMITHSONIAN MISCELLANEOUS COL- LECTIONS, 1s intended to include all the publications issued directly by the Smithsonian Institution in octavo form, excepting the AN- NUAL Report; those in quarto constituting the SMITHSONIAN Con- TRIBUTIONS TO KNOWLEDGE. The quarto series includes memoirs embracing the records of extended original investigations and re- searches, resulting in what are believed to be new truths and con- stituting positive additions to the sum of human knowledge. The octavo series is designed to contain reports on the present state of our knowledge of particular branches of science; instructions for collecting and digesting facts and materials for research; lists and synopses of species of the organic and inorganic world; museum catalogues ; reports of explorations; aids to bibliographical investi- gations, etc., generally prepared at the express request of the Insti- tution, and at its expense. In the CONTRIBUTIONS TO KNOWLEDGE, as well as in the MIsceEL- LANEOUS COLLECTIONS, the actual date of the publication of each article is that given in the Table of Contents, and not necessarily that of the volume in which it appears. The Quarterly Issue of the SMITHSONIAN MISCELLANEOUS COL- LECTIONS is designed chiefly to afford a medium for the early pub- lication of the results of researches conducted by the Smithsonian Institution and its bureaus, and especially for the publication of reports of a preliminary nature. Soe eAIN GIR, Secretary, Smithsonian Institution. ill ARTICLES Seventy New Malayan Mammals. Gerrit S. MILLER, JR. (Plates I-xIx, figure 1.) No. 1420. Published November Recent Studies of the Solar Constant of Radiation. C. G. AB- BOT. (Plates xx—xxi.) No. 1421. Published December GIMENO as 5oka she MRA ete ea) eh cae ee ay Set echea 8 Nae ple Dies fo ay The New Celostat and Horizontal Telescope of the Astrophys- ical Observatory of the Smithsonian Institution. C. G. AB- Hor. (Plate xxi.) No. 1422: Published December 9; SS Te TO ARI AN as IES nt CR eS et oe NAR On Some Photographs of Living Finback Whales from New- foundland. FREDERICK W. True. (Plates XXIV—XXVI.) homeo Published Mecember 9, 1903.....% 22-0 sce og A Skeleton of Hesperornis. FRepDER1Ic A. Lucas. (Plate Pomme) No. 1424. Published December 9, 1903....5...-.. A New Plesiosaur. FRepertc A. Lucas. (Plate xxvii1.) emia. etublishedeMecember GO; 1903;:. . estas as sees ak Shell Ornaments from Kentucky and Mexico. W.H. Homes. (Plates xx1x—-xxx, figure 2.) No. 1426. Published De- MIME Bee LOO Ree ok weds ove ns = 24m Pai die. Suvien w 2apntes sl Sree Be cea 8 On the Glacial Pothole in the National Museum. GEORGE P. Merritt. (Plate xxx.) No. 1427. Published December SE ART SPE oe ST MN PE aay ho ah hes oh eras Wie tse SINE ar ace mal OL epee Notes on the Herons of the District of Columbia. PAu BartscH. (Plates xxx1I—xxxvitt.) No. 1428. Published etre Be Cam) HON Siete 35h 20 01s “Snead yey Sec ous seem ueIeNG mi eho woe Es Preliminary Report on an Archeological Trip to the West In- dies. J. WALTER Fewxkes. (Plates xxxrx—xivi.) No. moe plished Miecemper GO; TOO3. . cic a- joe = + eer jo oes Form-Regulation in Celentera and Turbellaria. C. M. CHrrp. Rieretieos | bubisited Mecember,O, 1903s. . «<.s2< 2624-055 New Genera of South American Fresh-Water Fishes, and New Names for Some Old Genera. Cart H. E1GENMANN. No. iM aeeenplisned Meceimber O, TGOF. 2.52522... 5neeme >: Korean Headdresses in the National Museum. Foster H. Jenincs. (Figures 3-28.) No. 1432. Published January PAGE 74 84 QI 95 96 97 100 104 112 134 144 Vi SMITHSONIAN MISCELLANEOUS COLLECTIONS The Hodgkins Fund of the Smithsonian Institution. HELEN Watpo Burnsipe. (Plate L.) No. 1433. Published Jan- WAT Y 75 HOO 64 eros) arid oo 2s! soe «0 nee Sh re eee A Notable Success in the Breeding of Black Bears. ARTHUR B. Baker. (Plates Li-t11.) No. 1434. Published January Chinese Medicine. James M. Fiint. No. 1435. Published Jlamuaty 7; TOO... 2 vac.e 5 » «|» sala ops eee eee eae eee one Notes on the Rocks of Nugsuaks Peninsula and Its Environs, Greenland. W. C. PHaLeN. (Plates Limi—-tv.) No. 1436. Published. janwary 7; 1904... 7.00. caeadeee ence ee eee A Method of Avoiding Personal Equation in Transit Observa- hous, S. BP. LANGLEY. (Plate: ivi.) No 1446. eup- lished ‘April! 11; TOO: «402g temo sees = ese ater ee On a Collection of Fishes Made by Mr. Alan Owston in the Deep Waters of Japan. Davin STARR JORDAN and JOHN OTTERBEIN SNYDER. (Plates LvilI-Lxi1l1, figure 29.) No. 1447.. Published Aprilins 1004. ere eee eet eae Description of a New Cyprinoid Fish, Henibarbus Joitem, from the Pet Ho, Tientsin, China. Davin STARR JORDAN and EDWIN CHAPIN STARKS. (Plate Lxiv.) No. 1448. Pub- lished April 11, TQO4 .j..a0c 7a «orl etek eee pee = eon The Removal of the Remains of James Smithson. S. P. LANG- LEY. (Figure 30.) No. 1449. Published April 11, 1904.. Notes on the Breeding Habits of the Yellow-bellied Terrapin. Hucu M. Smitu. No. 1450. Published April 11, 1904.... A New Pelican Fish from the Pacific. Barton A. BEAN. (Ficure-31.) No. 1451, “Pitblished “Aprile, vO04) seve A Revision of the Paleozoic Bryozoa. E. O. ULricu and R. S. Basster. (Plates Lxv—Lxvinl, figures 32-33.) No. 1462.) Published April 11, foO4s4 heen eon. seine A Remarkable Genus of Fishes—The Umbras. THEODORE GILL. (Figures 34-38.) No. 1453. Published April 11, TIGA os ease SC eee Oh Se ki iis 2 Ree nee cae A New Occurrence of Unakite—A Preliminary Paper. W. C. PHALEN. ‘(Plates LxIxX-Lxx1, figure 39.) )No.,1454. “Pub- irshedh April il je CGO 4s) ..ccs. mete Savas eo ea eek eee Mer eect eval The Dinosaur Trachodon Annectens. F. A. Lucas. (Plates LXXI-LXx1u, figures 40-43.) No. 1455. Published April TT te OOH, «hs Sear each ted cane cine Gaeta iotev aS tS RS pe Oe AES Ae ce Classification of the Hares and Their Alles. Marcus Warp Lyon, Jr. (Plates Lxxiv—c, figures 44-45.) No. 1456. Pablished june sis, TOOA c.g acc eye tatccopsiacs Se ean eee teres 1608 230 PLATE. i LIT; EY: NE VEL. WE Vill. IX. ILLUSTRATIONS (1) Sciuropterus merens, type. (2) Funambu- iS POOSCUTUS Hyper ween io eto e's ates ss (1) Petaurista mitidula, male, Bunguran island, North Natunas. (2) Petaurista nitida, male, eastern Java. (3) Petaurista batuana, male, type. (4) Atherura sygomatica, female, type. (5) Atherura macroura, female, Trong, Lower SOLAR geese che eee The Rommel esis oie ae ses (1) Hemigale hardwicku, female, Tapanuli bay, Sumatra. (2) Hemigale minor, female, type. . (1) Paradoxurus lignicolor, male, type. (2) Paradoxurus hermaphroditus, male, Trong, WPovetye SUA ase Sa aera thie ee nec eoereeht oak (1) Paradoxurus lignicolor, male, type. (2) ~Paradoxurus hermaphroditus, male, Trong, NBO Sen MATIN ieteraute Levy tp ie ct aed ee ah eee (1) Galeopithecus ,gracilis, female, type. (2) Galeopithecus gracilis, male, Sirhassen island, South Natunas. (3) Galeopithecus pumilus, GING ap OCB ere tS cele Foca ee a pyle de Sc (1) Galeopithecus volans, female, Trong, Lower Siam. (2) Galeopithecus volans, male, Rum- pin river, Pahang. (3) Galeopithecus satura- tus, female, type. (4) Galeopithecus saturatus, ime wnat, Sala Batu ishamds a, oi 45% oye 3 (1) Galeopithecus volans, female, Trong, Lower Siam. (2) Galeopithecus volans, male, Rum- pin river, Pahang. (3) Galeopithecus satura- tus, female, type. (4) Galeopithecus saturatus, iInalembatia Bala, Batu islands 502i. sass ee « (1) Galeopithecus volans, female, Trong, Lower Siam. (2) Galeopithecus volans, male, Rum- pin river, Pahang. (3) Galeopithecus satura- tus, female, type. (4) Galeopithecus saturatus, malewdidana ala, “Batusislands. oi... 6). 2. a Vil PAGE 27, 43 47 49 50 BD Vill x XM. XV, VE VL XVIL XVI: XIX. See O45 XXIT. XXII. DEX. XXV. XT XXVII. . (1) Tupaia chrysogaster, female, type. SMITHSONIAN MISCELLANEOUS COLLECTIONS (2) Tu- paia ferruginea, female, Tanjong Dungun, Tringanu jo. (a jal @, 10) ©) 01s) © \e (6; js) w © 0 lec.e) of pigeu mtiag oe oleic) elle? a) elel a. . (1) Macacus nemestrinus, female, Tapanuli bay, Sumatra. (2) Macacus pagensi;, female, type. (1) Macacus nemestrinus, female, Tapanuli bay, Sumatra. (2) Macacus pagensis, female, type. (1) Macacus pagensis, female, type. (2) Maca- cus nemestrinus, female, Tapanuli bay, Suma- tra Sumas concolor, -adilt.%sictdes fe Sumiasconcolor, male, typer....+ eee eee Sumas. concolor,amale type ..cijae2 eae (1) Symphalangus syndactylus, male, Tapanuli bay, Sumatra. (2) Symphalangus klossiu, male. fVPE 26h .ncige wa tie Eee ee ee (1) Symphalangus syndactylus, male, Tapanuli bay, Sumatra. (2) Symphalangus klossu, male, EYES crests pated oi abpdataker ootous tae sPeme ph Maretors fe’ tetera (1) Symphalangus klossti, male, type. (2) Sym- phalangus syndactylus, male, Tapanuli bay, SUMEILEUE A /o.s iec Sein tad so alts NSTI ee a bane een Transparency of the atmosphere from bolographic Observations 4022-2 k ete athe oe eee: ae re Bolographic energy curves of the solar spectrum of a 60° glass prism. Observations of April PF [POOR vars th tare Bede neh Geese ead eye hohe o deen Distribution of radiation in the normal solar spec- trum: outside the earth's atmospheres. : a1a.-e The large ccelostat with second mirror, Smith- sonian Astrophysical Observatory............ Living finback whales. (Balenoptera physalus LE.) (@) -Preparing to” descend? (2) just feaching the fsurfacess 2 7.-eeeee eeere Living finback whales. (Balenoptera physalus Lb) a@) After spouting: — (2) Similar atti- tude; nearer views "(4)) lZestchiomaievers ery... Living finback whales. (Balenoptera physalus Le) (1), (Readyetoydescend: \i(2)iErocess: of descending—head under water, back arched... Skeleton of Hesperornis in the U. S. National Museum Che) le tej fe) <0, 6 (e) ‘Sle Je)e%b Ne lof javing a! te) (el (e bel (ee) is] eile) wile) ale) \s qaiiel shel 9) ie we, (e' (e, a) fw) aes “ol lp! 10, ©) 0 Ve, el 6! (ele) 6 (6! 0’ 6) 101 \@) © ©) @) 10) je) 6; 0) 10) 0 59 61 63 64 66 69 69 gal 71 72 ih 79 SI 88 Q2 Q2 XXVIII. XXIX. XXX. XXXI. XXXII. XXXII. XXXIV. XXXV. XXXVI. XXXVILI. SOOCV AL. XXXIX. MOL: XI. XEIT. CUE. XLIV; ILLUSTRATIONS Incomplete skeleton of plesiosaur in the U. S. National Museum. Type of Brachanchenius LUCOSL, NNGMDSTOIN te EN eavadsten, «tS isnce 4d. elm ones eia Shell gorget with engraved figure of a discus thrower, from an ancient grave near Eddyville, Kentucky Shell gorget from Mexico with engraved human figure The glacial pothole in the National Museum..... (1) Nesting site of Colony I of the Black-crowned Night Heron. (2) Adult flying (same col- ony ) (1) Nest and eggs of Black-crowned Night Heron in situ. (2) Detailed view of same. (3) Young and eggs of same, just hatched and hatching. (4) The young three days MLE SMAL CIOs ee ehaeiale vane a Mam ntt Tos 6 Ghd se (1) Young Black-crowned Night Herons seven days old. (2) Same, ten days old. (3) Same, three weeks old. (4) A favorite position in the ihee-topeanter deaving: the MeSpawess es. .ledo ss (1) Young Black-crowned Night Heron in full quvemileudnress, wie An adult: foe. 204.5: (1) Two young Green Herons 24 hours old and two addled eggs. (2) Twelve days old—on iE CORES.) Ctes cyst Sead g aepae Me seit cele d ¢ (1 and 3) The white phase of the Little Blue Heron, (2) (Great Blue Heron in a tree: (4) Little Blue Herons feeding on Anacostia river. (5) Roosting place of American Egret and cidessine EHeron., (6) American’ Egret on the marsh SHOW EMC R OM ery actate elie ar me hS ad ale alee Celts with stone handles from Santo Domingo... Stone implements from Santo Domingo and Porto Rico Stone objects from Porto Rico and Santo Do- mingo Ornamental stone pestles from Santo Domingo. . Idols with conical projections from Santo Do- MT SOM OnLE RICO nt). /S.0 ocidew sabe de» << Stone idols from Santo Domingo and Porto Rico. 9] =e) ewe) (s) ial (ey er keise) >) |e) (a, @! fe! .0) (eyiel. 382 @ land: téet or hares, rabbits, -and-pikas: . f.2s.¢¢ 384 Cl. (1, 2) Pentacrinus subangularis from Germany. (3) 4siocrinus from Kansas City, Missouri... 448 CI. Pentacrinus fossilis from Lyme Regis, England... 459 Crile Cinicrimus. sociaits trom, Kaneas. 226). 6.552: 450 FIGURE PAGE ee MOR GHIFOPOLOMYS MAAS. 2. saw age ene sss wet ses 40 2. Spider engraved on a bark tablet from an ancient grave HetOrenceneN ADAMI ete. M MeN Rigen. oS eed Mes 99 pa eonecan, hcadband andetopkmoty.ce ee one ee a else 150 PPOs. Or ALOPIMOPn erga cly eh hie lbayt eidinln diac ete 6-05 ale gal 15 5. Buttons for headband denoting rank of noblity........... 151 Semmrar tne edd cick os ae ee ee Me ENC IN Riya eras! « algie aeons 152 Pree eS@ulda ts. 5 aga.c unk eee eee HEAT oy ata tras 06 wk we ee tha 152 Pemeivil Service examinatemyMdty . salem ti ne oa) cet 6 eee 153 Sm Le TA NAG ea eat Oe hee tke das ade bce oS 154 POCECUIOMIAl “Lidice mre emer wt ts moi achicl elas freebie ah 8 sya, eile ges 155 See CCL TO Shiite eerie a Ane a ate, Ss srotans spare eis saab wv 155 Xii SMITHSONIAN MISCELLANEOUS COLLECTIONS T2. Moutner’s hat ‘and farmers hat... 2225 eee 150 13, Mourners hat-and headband’. 2.5... so ooee fis Seen 156 14.) Royal chair-bearer:s hab. si.6 ... 205.0. su wt te eee ee 157 T5.* Phys: head? trate 2.2% vache od oe noes an eee ee 158 Tos Royalmusiciai’ s Mat... sss. Se ener eae eee ee 159 Fz Royal servant's Tales .5 4... ¥ occ ase 5, secre ee eer eee 159 POSIT OUSE CAPS’ 4354-2 HE ee bane ee eed ee 160 TOsibanded Ouse Capses . x Lien: lett ale eo eerie eee eee 160 207 General s* helmet, oo2.2 3. . a2 ea somes eae ae eee 161 Bie Soldier's shelmet . (24's 25, via hake een tees cee eae eee eee 101 22. Swordsman’s hat and soldier’s and constable’s hat........ 162 2e0 Musketeers “hat .2.5, 5. SLs sone eee Cen oe ee ee 163 2ae (Gentleman's winter hoods $2.2. 40.5 sae ce ee ee 164 goreBabys hat and idétatl of samen ose eens ee 164 20: Umbrella or hat. cover andPhathoxs.. 22> aa.5 eee 105 27. Lady’s satin cap, lady’s mourning cap and lady’s winter HOGG. 215 oe ath ek ei pores tila chee ern cen ce ee 165 28. Womens hats and mode of tiair-dressingis se eee 166 29. Trismegistus owstont JORDAN AND SNYDER (section of epidermis) on4.5 Ae pee met cake ee eee eee eae 238 30: , Jaties “Siiithson eevee ticle ae oe ee eee eae aes ee 243 2u, Gastrostomus paciiens type, b. iN.) Sealine ne ane ane 255 22, Rhopalonaria capilaric i Wolitis)\i. ceca asc cee 275 22: Khopalonaria? veipstaCOemert)\ ons.) a. sae, oe eee 271 34." Buropean umbran (Umi ral uilorg cnn. a. se eter see see 295 35. Scale Of: wmiDids ate cue oe Os ee en eens ee 207 360. Elead of “unibya., dront lOve eer ree rere ee 297 a7. Navrious attitudes of tibiae asic pase eae es eee 299 as. Wester umbra. CU moraine ee eee 303 go; sElornblende in: lighter iphasevomalnenite ) erat ee 309 40. Skull of trachodon, showing the predentary bone......... 318 Als enies, of five teeth of trachodonemeacg- acter ae ee 319 42,-Group, of trachodon- teeth: «2 ae «eeycatentee ceeaeie eae eee 319 “on POre-tOot Of. iguanodon...\. «+ a-deeae aoe eee 3220 44. Enamel pattern of first upper incisor of hares, rabbits, and PUAS «oy edie ng is Oe orn salt yee se ce ree 3 ae ee 351 45a Carpus of the Veporidx and Ochotonmide. 29 saceeee eens 379 VOL: 45 PARTS I AND II SMITHSONIAN MISCELLANEOUS COLLECTIONS QUARTERLY ISSUE JULY-SEPTEMBER, 1903 SEVENTY NEW MALAYAN MAMMALS By GERRIG. S: -MIEVER; IK: Dr. W. L. Abbott has. presented to the United States National Museum four large collections of Malayan mammals of which it has been found impracticable to publish detailed accounts. The first was made on the islands of the Mergui Archipelago, off the west coast of Tenasserim, during the winter of 1900-01; it con- tains about four hundred specimens, and is noteworthy for its rich- ness in slightly differentiated insular forms of rats and squirrels. The second collection, of about two hundred and seventy speci- mens, forms part of the results of explorations among the islands of the South China Sea not previously visited,’ and on the neighbor- ing east coast of Johore. Peculiar insular species of porcupine and flying lemur, both from Pulo Aor, may be regarded as the most interesting of the new mammals found during this expedition. The third was obtained in the Rhio Archipelago, off the southern ex- tremity of the Malay Peninsula, in August and-September, 1902, and numbers about one hundred and seventy-five specimens. Two of the new mammals which it includes, a monkey and a treeshrew, show an unexpected likeness to species occurring in the Anamba and Natuna Islands. The last and most valuable of the four collec- tions was made with the assistance of Mr. C. B. Kloss during the winter of 1902-03 on the Pagi Islands, the Batu Islands, and Pulo Nias, islands of the chain lying parallel to the west coast of Sumatra.? It contains about three hundred specimens, among *For an account of the mammals obtained during Dr. Abbott’s first and second cruise in the South China Sea, see Miller, Proc. Washington Acad. Sci., I, pp. 203-246, August 20, 1900, and m1, pp. 111-138, March 26, 19or. *For an account of the mammals obtained by Dr. Abbott on the more northerly islands of this archipelago, see Miller, Proc. U. S. Nat. Mus., XXVI, pp. 437-484, February 3, 1902. I 2 SMITHSONIAN MISCELLANEOUS COLLECTIONS [VOL. 45 which are represented more than thirty hitherto undescribed species. Five of these are monkeys, two of which, a dwarf siamang and a member of a new genus Simias, differ from their known rela- tives by very remarkable characters. Study of these collections has led to the reexamination of much of Dr. Abbott’s material that has been previously reported on. The seventy new mammals here de- scribed have been found in the course of this work. TRAGULUS BATUANUS spp. nov. Type.—Adult female (skin and skull), No. 121,697, United States National Museum. Collected on Tana Bala, Batu Islands, Sumatra, February 5, 1903, by Dr. W. L. Abbott. Original number, 2226. Characters.—A large member of the Tragulus napu group with wholly black neck and black throat stripes. The black of neck ex- tends forward over face and cheeks more than in any other species except Tragulus anne, T. jugularis, and T. bunguranensis. From all of these the Batu animal is distinguished by the normal pattern of the throat markings. Color.—Type: back ochraceous heavily shaded by the black hair- tips, neither color distinctly in excess of the other. On sides the ochraceous fades to buff and the black shading becomes slightly less heavy than on back. Muzzle, loral stripe, face, except dull ochraceous line 4 mm. wide bordering loral stripe, crown, ears, entire neck to outer white throat stripes, and cheeks to level of posterior canthus of eye, glossy black. Cheek from muzzle to pos- terior canthus of eye dull grizzled ochraceous. On close inspection dull ochraceous annulations may be detected on the black portion of the cheeks and on the neck, particularly at sides. These are nowhere sufficiently numerous to break the clear black effect, ex- cept at front of shoulders, where they rapidly increase in number, producing a rather abrupt transition to the color of the back. Underparts and light throat stripes white. A brownish stripe 60 mm. long and 4 mm. wide on middle of chest. Dark throat stripes clear black; collar black, noticeably speckled with dull ochraceous. The throat pattern is in every way normal, though the black stripes are perhaps narrower than usual. At anterior termi- nation these stripes are about 10 mm. apart, and there is no brown wash on the white in front of them. Legs like back externally, the inner surface white. On hind legs the white disappears at about middle of tarsus; below this point the legs and feet are sprinkled with dull ochraceous hairs. Tail like back above, but ochraceous not as bright, white below and at tip. MILLER] SEVENTY NEW MALAYAN MAMMALS 3 Skull and teeth—The skull is very large and the nasals are unusually long, but otherwise I do not detect any tangible cranial characters. Teeth large, particularly the upper premolars. Measurements——External measurements of type: total length, 680; head and body, 595; tail vertebree, 85; hind foot, 145 (130) ; ear from meatus, 36.6; ear from crown, 30; width of ear, 24. Cranial measurements of type: greatest length, 120 (116) ;* basal length, 112 (108) ; basilar length, 105 (102) ; occipito-nasal length, IIO (105) ; length of nasals, 40 (36) ; greatest breadth of both nasals together, 13 (13) ; diastema, 15 (12.4) ; zygomatic breadth, 54 (48) ; least interorbital breadth, 32 (30.4) ; mandible, 95 (91); maxillary toothrow (alveoli), 41 (39); maxillary premolars (crowns), 22 (20.4) ; mandibular toothrow (alveoli), 47 (45); mandibular pre- molars (crowns), 23.4 (20). Specimens examined.—Four, all from the Batu Islands, one from Tana Masa, the rest from Tana Bala. Remarks.—Unfortunately the type is the only adult in the collec- tion. The immature individuals differ from it in the presence of a grayish buff wash on middle of belly, a character which is probably individual rather than due to the difference in age. The species is easily recognizable by the forward extension of the black neck area, combined with the normal throat pattern. TRAGULUS RUSSULUS sp. nov. Type.—Adult male (skin and skull), No. 121,701, United States National Museum. Collected on Tana Bala, Batu Islands, Sumatra, February 8, 1903, by Dr. W. L. Abbott. Original number, 2249. Characters—A member of the kanchil group similar to Tragulus russeus of the Banjak Islands, but upper parts not as dark, belly with less extensive yellowish suffusion, and throat pattern always normal. Skull and teeth not so large as in Tragulus russeus. Color—Type: general color above orange-ochraceous darker than that of Ridgway, the neck and outer surface of limbs brighter than body. The hairs of the back are tipped with black, forming a slight, uniform, dark clouding, much less pronounced than in Tragulus russeus. On shoulders this shading deepens abruptly into the black, slightly grizzled neck stripe. Crown a grizzle of dusky brown and buff. Cheeks dull buff. A faintly indicated buffy stripe borders upper margin of very ill-defined dark loral stripe. Ears blackish. ‘Throat pattern normal, the dark stripes and collar orange- 1 Measurements in parenthesis are those of an adult female paratype (No. 113,120) of Tragulus nigricollis. 4 SMITHSONIAN MISCELLANEOUS COLLECTIONS [VoL. 45 ochraceous, the latter clear, the former slightly grizzled by dark brown hair-tips. The dark stripes meet anteriorly. From collar a dull orange-ochraceous stripe about 5 mm. in width extends back over middle of chest to join the pale orange-buff area 80 mm. long by 40 mm. wide which occupies middle of belly. With these ex- ceptions the entire under surface of body is white. Tail white below and at tip, like back above but not as bright, and much less grizzled. Skull and teeth.—The skull and teeth closely resemble those of Tragulus russeus but are not as large. Measurements.—External measurements of type: total length, 535; head and body, 470; tail vertebrz, 65; hind foot, 122 (109) ; ear from meatus, 32; ear from head, 20; width of ear, 18. External measurements of adult female from the type locality (No. 121,697) : total length 550; head and body, 485; tail vertebrz, 65; hind foot, 109 (98). , Cranial measurements of type: greatest length, 98; basal length, 92; basilar length, 86; occipito-nasal length, 87; length of nasals, 31.4; greatest breadth of both nasals together, 15; diastema, 8.8; zygomatic breadth, 44; least interorbital breadth, 28; mandible, 79; maxillary toothrow (alveoli), 33.6; maxillary premolars (crowns), 16.8; mandibular toothrow (alveoli), 37.6; mandibular premolars (crowns), 16. Specimens examined.—Nine, all from the Batu Islands, 6 from Tana Bala, 1 from Tana Masa, and 2 from Pulo Pinie. Remarks.—The orange-buff wash on the belly is occasionally more extensive than in the type, but never conspicuously so. The throat pattern is invariably normal, and the dark stripes always come together anteriorly. In the two skins from Pulo Pinie the general color of the back, sides, and neck is lighter and more buff than in those from the other islands. RATUFA INSIGNIS sp. nov. Type.—Adult male (skin and skull), No. 115,531, United States National Museum. Collected on Pulo Sugi, Rhio Archipelago, August 26, 1902, by Dr. W. L. Abbott. Original number, 1960. Characters——Externally similar to Ratufa notabilis Miller’ of Linga Island, but size not as great. Skull noticeably smaller than that of R. notabilis, and front root of zygoma less abruptly flaring. Color.—The color so closely resembles that of Ratufa notabilis as to need no detailed description. In the type the upperparts are MPa ead, Nat. Sci. Philadelphia, 1902, p. 150, June i, 1902. MILLER] SEVENTY NEW MALAYAN MAMMALS 5 ‘-burnt-umber slightly darker than in the type of R. notabilis. Along middle of back there is a faint drab gloss, and the light annulations of the hairs are smaller and more sharply defined than in the related species. Back without the sprinkling of whitish hairs present in the type of notabilis. Underparts and inner surface of limbs cream- buff, tinged with brownish yellow, particularly on front legs. The cream-buff area on the legs is narrower than in the type of Ratufa notabilis. ‘The pale area of underparts and legs is everywhere sepa- rated from the dark adjoining regions by a distinct edging of light tawny-ochraceous. This is present in the type of R. notabilis, but much less distinct. In both species the pale flank spot is indicated by an interruption of this edging. Cheeks, feet, and muzzle whitish. Tail concolor with back above, at sides, and tip, dull whitish gray along middle beneath. The four specimens are very uniform in color characters except for the usual effects of bleaching, which are noticeable in two of the skins. Skull and teeth—The skull and teeth are readily distinguishable from those of Ratufa notabilis by their smaller size. The general form of the skull is throughout less broad and robust than in the related species, but this difference is most noticeable in the region of the anterior zygomatic roots, which flare much less abruptly and widely than in the larger animal. Measurements.—External measurements of type: total length, 780 ; head and body, 360; tail vertebrz, 420; hind foot, 76 (70) ; ear from meatus, 26.6; ear from crown, I6. Cranial measurements of type: greatest length, 64.4 (68) ;' basal length, 55 (58); basilar length, 51.4 (55); length of nasals, 21.4 (22) ; diastema, 15 (14.6); least interorbital breadth, 25.4 (27.4) ; breadth of braincase above roots of zygomata, 28 (29.4) ; zygomatic breadth, 39 (44) ; mandible, 41 (45) ; maxillary toothrow (alveoli), I2 (13); mandibular toothrow (alveoli), 12.6 (15). Specimens examined.—Four, all from the type locality. RATUFA CONSPICUA sp. nov. Type.—Adult male (skin and skull), No. 115,528, United States National Museum. Collected on Pulo Bintang, Rhio Archipelago, August 19, 1902, by Dr. W. L. Abbott. Original number, 1900. Characters—In general similar to Ratufa notabilis and R. insignis, but skull not so large as in either of these, and pale color of under- parts abrrptly contrasted with dark brown of sides. 1 Measurements in parenthesis are those of the type of Ratufa notabilis, a young adult male. 6 SMITHSONIAN MISCELLANEOUS COLLECTIONS [VoL. 45 Color.—Upperparts of type exactly as in Ratufa insignis except that the neck and shoulders are more conspicuously speckled by the light annulations of the hairs. Underparts and inner surface of legs cream-buff not tinged with yellow. This color is everywhere sharply defined against the brown. There is a faint trace of the tawny edging between the two colors on hind legs, but to a much less ex- tent than in either of the other species. The skins show practically no variation in color beyond that due to bleaching. Skull and teeth—In form the skull is similar to that of Ratufa msigmis, but in size it is not as large. Teeth as in the related species. Measurements.—External measurements of type: total length, 705; head and body, 330; tail vertebre, 375; hind foot, 77 (72); ear from meatus, 25.6; ear from crown, 13. Cranial measurements of type: greatest length, 63; basal length, 54; basilar length, 50; length of nasals, 20.4; diastema, 14; least in- terorbital breadth, 24; breadth of braincase above roots of zygomata, 27.4; zygomatic breadth, 39; mandible, 38.8; maxillary toothrow (alveoli), 12.8; mandibular toothrow (alveoli), 12.8. Specimens examined.—Seven, all from Pulo Bintang. RATUFA BALZ: sp. nov. Type.—Adult male (skin and skull), No. 121,715, United States National Museum. Collected on Tana Bala, Batu Islands, Sumatra, February 5, 1903, by Dr. W. L. Abbott. Original number, 2224. Characters—Similar to Ratufa femoralis but much darker ; back darker than belly; tail blackish. Crown and forehead not notice- ably lighter than back. Femoral spot large and conspicuous. Color—Type: upperparts wood-brown faintly washed with vandyke-brown along middle of back, and showing strong drab re- flections in certain lights. On sides the brown brightens to a brownish buff and the shorter hairs show distinct light annulations, . The speckling thus produced is inconspicuous. Crown and fore- head like sides but not quite as dark. Cheeks slightly tinged with gray. Ears and indistinct eye ring blackish. Whiskers black, their bases surrounded by a whitish patch. Outer surface of legs like sides but strongly tinged with ochraceous-rufous. This is particu- larly noticeable on front legs between shoulder and wrist, where a clear band is formed between colors of outer and inner surfaces. Feet blackish. Underparts and inner surface of legs clear orange- buff, darker along median line, paler in axillary region. Chin dark grayish. Femoral spot whitish, large and well defined. Entire tail MILLER | SEVENTY NEW MALAYAN MAMMALS 7 very dark, blackish vandyke-brown, the hairs fading to a light drabby buff below middle. This color appears faintly at surface along under side, but scarcely more than enough to throw the dark median line of short hairs into contrast. Skull and teeth.—The skull is slightly larger than that of Ratufa femoralis, the interpterygoid space is wider and the nasal bones and audital bulla are longer. Teeth as in the related species. Measurements.—External measurements of type: total length, 725; head and body, 340; tail vertebrz, 385; hind foot, 74 (70). Cranial measurements of type: greatest length, 65; basal length, 54; basilar length, 50; length of nasals, 21; greatest breadth of bath nasals together, 12; least interorbital breadth, 26; zygomatic breadth, 38; mandible, 39.6; maxillary toothrow (alveoli), 12.8; mandibular toothrow (alveoli), 13.6. Specimens examined.—Fight, all from Tana Bala. Remarks.—This squirrel is readily distinguishable from all other known members of the genus except those occurring on Tana Masa and Pulo Pinie by its rich coloration, blackish tail, and conspicuous femoral spot. The tail is nearly as dark as that of Ratufa nigrescens, but in the Mansalar animal the back also is blackish, and the femoral spot is obsolete. In Ratufa femoralis the back is distinctly lighter than the belly, while the opposite is true of the Batu species. RATUFA MASZ sp. nov. Type.—Adult male (skin and skull), No. 121,818, United States National Museum. Collected on Tana Masa, Batu Islands, Febru- ary 21, 1903, by Dr. W. L. Abbott. Original number, 2330. Characters.—Like Ratufa bale, but entire head grayish, in distinct contrast with back. Femoral spot large and well defined. Color.—The color so closely resembles that of Ratufa bale that no detailed account is required. The whole crown and forehead, how- ever, are a grizzled yellowish gray, noticeably lighter than rest of upperparts. In the grizzled area the hairs are slaty at base, then yellowish brown, followed by a conspicuous whitish cream-buff an- nulation and a dark tip. Cheeks clear grizzled gray, the general effect between the hair-brown and smoke-gray of Ridgway, the same color extending as a broad band across throat and chin. Femoral spot large and conspicuous, whitish cream-buff. Measurements.—Measurements of type: total length, 730; head and body, 340; tail vertebrz, 390; hind foot, 72 (69) ; skull, greatest length, 62; basal length, 52; length of nasals, 20; interorbital con- striction, 25; zygomatic breadth, 30. Specimens examined.—Four, all from Tana Masa. 8 SMITHSONIAN MISCELLANEOUS COLLECTIONS [VoL. 45 RATUFA PINIENSIS sp. nov. Type.—Adult male (skin and skull), No. 121,840, United States National Museum. Collected on Pulo Pinie, Batu Islands, March I, 1903, by Dr. W. L. Abbott. Original number, 2343. Characters.—Like Ratufa mase@ but darker throughout, particu- larly on the ventral surface. Head grayish in strong contrast with back. Femoral spot less well developed than in the two related forms. Color.—In general the color is like that of the two other Batu members of the group, but the vandyke-brown wash on the back is heavier, particularly in lumbar region, and the underparts are darkened nearly to ochraceous-rufous. The inguinal and axillary regions are not as dark as the rest of the ventral surface. Crown even lighter and clearer gray than in Ratufa mase@, scarcely yellower than cheeks. Femoral spot not as large as in the related forms, and so heavily grizzled by dark yellowish brown annulations as to be less conspicuous than in any of the Sumatran members of the group except Ratufa nigrescens. Skull and teeth.—The skull and teeth do not differ tangibly from those of Ratufa bale. Measurements.—Measurements of type: head and body, 325; tail imperfect; hind foot, 73 (69); skull, greatest length, 62; basal length, 52; basilar length, 48.4; length of nasals, 11; greatest breadth of both nasals together, 12; least interorbital breadth, 26; zygomatic breadth, 38; mandible, 39; maxillary toothrow (alveoli), 12; mandibular toothrow (alveoli), 13. Specimens examined.—Six, all from Pulo Pinie. SCIURUS BILIMITATUS sp. nov. Type.—Adult female (skin and skull), No. 105,072, United States National Museum. Collected at Tanjong Laboha, Tringanu, Malay Peninsula, September 29, 1900, by C. B. Kloss. Original number (Dr We Ie Abbett), 671. Characters.—Similar to the Javan Sciurus nigrovittatus Horsfield, but tail longer, back more distinctly grizzled, gray of underparts paler, and outer lateral stripe conspicuous and sharply defined. Color.—Upperparts a fine grizzle of black and cream-buff, the latter rather in excess, and brightening to buff on shoulders, neck, and flanks. Cheeks, muzzle, chin, and throat buff, brighter and more rusty than that of Ridgway. Chest, belly, and inner surface of legs pale smoke-gray, the darker bases of the hairs appearing MILLER] SEVENTY NEW MALAYAN MAMMALS 9 irregularly at surface. Lateral stripes clear and sharply defined, the inner black, the outer buff. The width of each at middle is about 10 mm. Measurements.—Measurements of type: total length, 485; head and body, 205; tail vertebrze, 180; hind foot, 46 (43). Specimens examined.—Malay Peninsula; Tanjong Laboha, Trin- ganu, I; no exact locality, 1; Tioman Island, 12. Remarks.—As compared with a series of five skins of Sciurus migrovittatus from western Java the characters of Sciurus bilimitatus are both pronounced and constant. There is no variation worthy of note, and the specimens from Tioman Island appear to agree in all respects with those from the mainland. In Sciurus nigrovittatus the pale lateral stripe is so ill defined that in some skins it would easily pass unnoticed. Horsfield, in fact, does not mention it. SCIURUS PEMANGILENSIS sp. nov. Type.—Adult female (skin and skull), No. 112,460, United States National Museum. Collected on Pulo Pemangil, off coast of Johore, June 12, 1901, by Dr. W. L. Abbott. Original number, 1062. Characters.—A pallid member of the Sciurus vittatus group some- what resembling Sciurus lautensis of the North Natuna Islands, but smaller and with upperparts a clearer, less yellowish gray. Color.—Upperparts a fine inconspicuous grizzle of black and pale very dull buff (or light wood-brown), the two colors about evenly mixed on back, the buff slightly in excess on sides, the general effect throughout very near broccoli-brown. Cheeks and outer surface of legs light wood-brown. Entire upper surface of tail like back, but more coarsely grizzled. Under surface of tail nearly clear, pale, wood-brown (or cream-buff with a slight brownish tinge) except at edges and at tip, where it is grizzled. Underparts and inner sur- face of legs bright ochraceous-buff (considerably more yellow than that of Ridgway), fading to pale wood-brown on chin. Outer lateral stripe buffy white, well defined and about 7 mm. in width. Inner lateral stripe diffuse, about 10 mm. in breadth, its color like that of sides of body but slightly darker and overlaid by a thin wash of ochraceous-buff. Feet dull buffy gray, not noticeably paler than legs. Skull and teeth.—The skull and teeth resemble those of Sciwrus lautensis except that the rostrum appears to be somewhat broader and more robust. Measurements.—Measurements of type: total length, 335; head and body, 185 ; tail vertebrz, 150; hind foot, 44 (41) ; skull, greatest length, 49; zygomatic breadth, 28. 1O SMITHSONIAN MISCELLANEOUS COLLECTIONS [VoL. 45 Specimens examined.—Twelve, all from Pulo Pemangil. Remarks——Among the squirrels of the Sciurus vittatus group with which I am familiar this species is readily distinguishable by its uniform gray dorsal surface, in which the grizzle has become obsolete. SCIURUS AORIS sp. nov. Type.—Adult female (skin and skull), No. 112,418, United States National Museum. Collected on Pulo Aor, off coast of Johore, June 5, 1901, by Dr. W. L. Abbott. Original number, 1002. Characters.—A pallid member of the Sciurus vittatus group, much like S. pemangilensis in general color, but back and sides sharply and distinctly grizzled, and red of underparts slightly less bright. Color.—Type: The type so closely resembles that of Sciurus pemangilensis that no detailed description is necessary. On back and sides the elements of the grizzle are essentially the same in the two species, but in that from Pulo Aor the colors are so sharply contrasted as to produce the finely speckled appearance charac- teristic of most of the members of the group. Underparts essentially as in S. pemangilensis but the red not as bright. Skull and teeth.—I can detect nothing to distinguish the skull and teeth from those of Sciurus pemangilensis. Measurements.—Measurements of type: total length, 335; head and body, 185; tail vertebrz, 150; hind foot, 44 (41) ; skull, greatest length, 45; zygomatic breadth, 25. Specimens examined.—Fifteen, all from Pulo Aor. Remarks.—Sciurus aoris is readily distinguishable from S. pemangilensis, by its sharply speckled back. In general color it closely resembles Sciurus lautensis, but the flanks and thighs are uniform with the back and not tinged with fulvous as in the Natuna animal. The series shows no marked variations in color. SCIURUS PENINSULARIS sp. nov. Type.—Adult male (skin and skull), No. 112,511, United States National Museum. Collected on north bank of Endau River, Pahang, June 21, roo1, by Dr..W. L. Abbott. Original number, 1078. oh Characters.—Like Sciurus vittatus, but red of underparts strongly tinged with ochraceous, and cheeks scarcely more yellow than sides of neck. Color—Type: upperparts and tail a fine grizzle of black and ochraceous-buff, the latter a little in excess on body, the former on tail. Underparts and inner surface of legs tawny, washed, particu- MILLER | SEVENTY NEW MALAYAN MAMMALS ar larly on chest, with ochraceous. Lateral stripes as in Sciurus vittatus, the outer whitish cream-buff, about 6 mm. wide at middle, the inner black and about Io mm. in width. Cheeks like back, but with a light wash of ochraceous-buff, this wash not distinct enough to produce a marked contrast with color of neck. Measurements.—Measurements of type: total length, 390; head and body, 210; tail vertebrze, 180; hind foot, 50 (47) ; skull, greatest length, 49.4; zygomatic breadth, 28. Specimens examined.—Nine, five from the banks of the Endau River, and four from Singapore. Remarks.—tThe squirrel of the vittatus group inhabiting the south- ern end of the Malay Peninsula is distinguishable from the typical Sumatran animal by its more yellowish underparts and less yellow- ish cheeks. In true Sciurus vittatus, as represented by specimens from Tapanuli Bay, the red area of the body is very nearly the same as the orange rufous of Ridgway, and the cheeks are so heavily washed with buff as to contrast strongly with sides of neck. SCIURUS PANNOVIANUS sp. nov. Type.—Adult male (skin and skull), No. 112,351, United States National Museum. Collected on Pulo Pannow, Atas Islands, South China Sea, May 28, 1901, by Dr. W. L. Abbott. Original number, 952. Characters.—Like Sciurus peninsularis, but size less, dark lateral stripes much broader and more diffuse, and tail with fairly well de- fined black pencil. Color.—Type: The general color so closely resembles that of the type of Sciurus peninsularis as to need no detailed description. The dark lateral stripes are less sharply outlined than in the mainland animal, however, and their width at middle is at least 15 mm. Tail with blackish pencil distinctly visible both above and below. Skull and teeth—The skull and teeth are not so large as in Sciurus peninsularis, and the audital builz are smaller and less in- flated, particularly in their posterior segment. Measurements.——Measurements of type: total length, 380; head and body, 193; tail vertebra, 185; hind foot, 48 (45) ; skull, greatest length, 47; zygomatic breadth, 26.4. Specimens examined.—Ten, all from Pulo Pannow. Remarks.—From Sciurus abbott and S. anambensis the Pannow squirrel is distinguished by its darker upperparts, brighter under surface, and black-tipped tail. The series of ten specimens shows no color variations worthy of note. Tz SMITHSONIAN MISCELLANEOUS COLLECTIONS [VOL. 45 SCIURUS ICTERICUS sp. nov. Type.—Adult female (skin and skull), No. 121,727, United States National Museum. Collected on Tana Bala, Batu Islands, Febru- ary 4, 1903, by Dr. W. L. Abbott. Original number, 2223. Characters.—In general appearance like Sciwrus vittatus, but size greater, color above darker, lateral stripes less distinct, and cheeks bright, clear, yellowish buff. Color—Type: upperparts and outer surface of legs a uniform fine grizzle of black and buff, the former everywhere in excess. Underparts and inner surface of legs rufous. Cheeks clear buff- yellow in striking contrast with surrounding parts. This color extends from base of whiskers to level of anterior margin of ear, and is continuous with the clear buff-yellow eye ring. This ring is 3 mm. wide over middle of eye. Posteriorly the buff area extends upward onto basal half of ear, but in this region it becomes dull and grizzled. Behind mouth and chin it becomes mixed with the rufous of underparts. The lips and chin are, however, dark grizzled gray. Outer surface of ears blackish. Feet like legs, but tinged with gray. Pale lateral stripe dull, light smoke-gray, only 5 mm. wide at middle and reduced in extent at each end. Dark stripe blackish, about 7 mm. wide, slightly longer than pale stripe. Tail like back, but more coarsely grizzled, and light bands on hairs less yellow; pencil indistinctly blackish. Skull and teeth.—The skull is similar to that of Sciurus vittatus but is much larger and the audital bullz are broader and less ele- vated. Teeth similar to those of Sciurus vittatus but larger throughout. Measurements.—Measurements of type: total length, 420; head and body, 225; tail vertebrze, 195 ; hind foot, 49 (46) ; skull, greatest length, 54 (51); basal length, 46.6 (43.6); zygomatic breadth, 32.6 (30) ; least interorbital breadth, 20 (18); mandible, 34 (32); maxillary toothrow (alveoli), 10.2 (9.6). Specimens examined.—Nine, from the following of the Batu Islandsc@RulosPinie, 1; Vana Bala: 7; Vana. Masarar Remarks.—This strikingly characterized squirrel needs no special comparison with any of the members of the group to which it be- longs. The specimens show no variation worthy of note except that the skin from Pulo Pinie has the light element in the tail less yellow than the others. 1 Cranial measurements in parenthesis are those of an adult female Sciurus vittatus from Tapanuli Bay, Sumatra (No. 114,518). MILLER] SEVENTY NEW MALAYAN MAMMALS 13 SCIURUS ATRATUS sp. nov. Type.—Adult female (skin and skull), No. 121,524, United States National Museum. Collected on North Pagi Island, Sumatra, No- vember 22, 1902, by Dr. W. L. Abbott. Original number, 2087. Characters.—A large member of the Sciurus vittatus group with clear black tail, blackish brown upperparts, and blackish gray belly. Lateral stripes nearly lost in the general darkening of the fur. Color—Type: upperparts and outer surface of hind legs an ill- defined grizzle of black and dull russet, the former everywhere much in excess, but particularly so on middle of posterior half of back, where the lighter color is scarcely noticeable except on close in- spection. Outer surface of front legs black. Cheeks dull black. Feet and ears blackish hair-brown. Tail clear black throughout, the hairs with no indication of lighter bands except at extreme base of tail, and even here very inconspicuously. Underparts and inner surface of legs a mixture of black and smoke-gray, each hair wholly of one color or the other. The gray is slightly in excess on throat and front legs, the black on chin, chest, belly, and hind legs. Black lateral stripe about 10 mm. broad. While it extends from front leg to middle of thigh it is not noticeable at first sight, on account of its slight contrast with the surrounding parts. Pale lateral stripe dull russet, more reddish than that of Ridgway, about 10 mm. wide, but so ill defined and inconspicuous that it might readily pass unobserved. Throughout the body the hairs are un- usually glossy, and along median line from crown to middle of back there is a faint sprinkling of entirely white hairs. Skull and teeth.—The skull and teeth closely resemble those of Sciurus ictericus, but the braincase is less highly arched and the nasals are broader in proportion to their length. Measurements.—Measurements of type: total length, 395; head and body, 230; tail vertebrze, 165; hind foot, 49 (46) ; skull, greatest length, 54; basal length, 46; zygomatic breadth, 32; least interorbital breadth, 18; mandible, 35; maxillary toothrow (alveoli), 10. Specimens examined.—Twenty-one, from the following localities : North Pagi Island, 8; South Pagi Island, 13. Remarks.—While the characters as a whole are very constant there are certain rather considerable variations. In most of the skins the gray hairs are in excess of the black on entire ventral sur- face, but in one (No. 121,613) the opposite is conspicuously the case. The whitish hairs on the dorsal surface are sometimes concentrated on head and neck so as to give this region a noticeably grayish cast. 14 SMITHSONIAN MISCELLANEOUS COLLECTIONS [VOL. 45 The tail is invariably clear black, except at extreme base, a char- acter which will serve to distinguish the animal from the Siporan Scirus melanogaster at a glance, though the two species are of about the same color and size. SCIURUS PINIENSIS sp. nov. Type.—Adult female (skin and skull), No. 121,800, United States National Museum. Collected on Pulo Pinie, Batu Islands, Sumatra, March 1, 1903, by Dr. W. L. Abbott. Original number, 2344. Characters—In general appearance like Sciurus mnatunensis (Thomas), but size less, and skull with short, broad rostrum much as in Sciurus tenuts. Color.—Upperparts a fine grizzle of black and dull light ochra- ceous, the latter everywhere a little in excess, but more particularly so on sides. Crown, feet, and outer surface of legs like back, but darker and more finely grizzled. Cheeks, eye ring, and ill-defined area about roots of whiskers nearly clear ochraceous. A wash of the same color on anterior inner surface of thigh. Underparts and inner surface of legs cream-buff, lighter than that of Ridgway an- teriorly and on front legs, darkening posteriorly and on hind legs almost to buff. The light and dark areas of front leg are separated on outer side by an almost clear blackish line. Tail like back, but darker above and lighter below, everywhere more coarsely grizzled; pencil black. Skull and teeth—Skull smaller than that of Sciurus natunensis and without elongation of muzzle. In size and general form it closely resembles the skull of Sciurus tenuis, but is not as broad interorbitally, the audital bulle are smaller, and the nasal bones are shorter. Teeth as in Sciurus tenuis and S. natunensis but much smaller. Measurements.—External measurements of type: total. length, 210; head and body, 134; tail vertebrae, 75; hind foot, 32 (29) ; ear from meatus, 12.6; ear from crown, 7; width of ear, 10; skull, greatest length, 35; basal length, 29.4; length of nasals, 9.4; zygomatic breadth, 20.4; interorbital breadth, 11.2; maxillary toothrow (alveoli), 5.6. Specimens examined.—One, the type. SCIURUS BALZ sp. nov. Type.—Adult male (skin and skull), No. 121,799, United States National Museum. Collected on Tana Bala, Batu Islands, Febru- ary 12, 1903, by Dr. W. L. Abbott. Original number, 2282. MILLER | SEVENTY NEW MALAYAN MAMMALS i5 Characters.—Similar to Sciurus pimiensis but slightly larger, and underparts almost white. Color.—The color throughout is as in Sciurus piniensis except that the light element of the grizzle on upperparts is paler (very nearly the buff of Ridgway), the dark line on outer side of front leg is less developed, and the underparts are nearly white anteriorly, very pale cream-buff posteriorly. Skull and teeth.—The skull is slightly larger than that of Sciwrus piniensis and the teeth are relatively smaller, but otherwise I can detect no differences. Measurements.—External measurements of type: head and body, 137; tail, — (broken) ; hind foot, 34 (31); ear from meatus, 13; ear from crown, 8.6; width of ear, 11; skull, greatest length, 36.6; basal length, 31; length of nasals, 10; zygomatic breadth, 22; inter- orbital breadth, 12.2; maxillary toothrow (alveoli), 5.6. Specimens examined.—One, the type. Remarks.—While this squirrel is very closely related to Sciurus piniensis the characters shown by the single specimen are such that it is impossible to regard the animals from the two islands as identical. SCIURUS PUMILUS sp. nov. Type.—Adult female (skin and skull), No. 121,627, United States National Museum. Collected on South Pagi Island, Sumatra, November 27, 1902, by Dr. W. L. Abbott. Original number, 2008. Characters.—Similar to Sciurus tenuis but much smaller and somewhat darker. Not as small as Sciurus fraterculus. Color.—Type: back, sides, neck, head, and outer surface of legs a fine, uniform grizzle of black and buff, the latter everywhere a little in excess, but slightly more so on sides than elsewhere. Underparts and inner surface of legs cream-buff, dulled by the appearance at surface of the slaty underfur, this especially noticeable toward sides. Tail like back but more coarsely grizzled and slightly yellower. Feet dull buffy gray, not strongly contrasted with legs. Skull and teeth.—The skull resembles that of the members of the Sciurus tenuis group, but is smaller than that of any hitherto known, S. fraterculus excepted. As compared with that of Sciurus pro- cerus, the smallest that I have at hand, it is decidedly narrower and more elongate in general outline, and the zygomata are less abruptly flaring in front. Teeth relatively as well as actually smaller than those of Sciurus procerus. Measurements.—External measurements of type: total length, 205; head and body, 125; tail vertebrze, 80; hind foot, 30 (28) ; ear from meatus, 11; ear from crown, 7; width of ear, Io. 16 SMITHSONIAN MISCELLANEOUS COLLECTIONS [voL. 45 Cranial measurements of type: greatest length, 34.6; basal length, 28.4 (28) ;' basilar length, 26.8; diastema, 8.4 (8) ; length of nasals, 10 (9.8) ; greatest breadth of both nasals together, 5 (5.3); inter- orbital breadth, 11.8 (12); zygomatic breadth, 19.6 (20.3) ; man- dible, 21; maxillary toothrow (alveoli), 5.8; mandibular toothrow (alveoli), 6. Specimens examined.—Six, all from the Pagi Islands. Remarks.—Although closely related to the Sciurus fraterculus of Sipora Island, this squirrel is distinguishable by its less dwarf size and by the absence of rufous tints on sides and upperparts. SCIURUS LANCAVENSIS sp. nov. 1900. Sciurus concolor MiLver, Proc. Biol. Soc. Washington, x1, p. 191, December 21, 1900. Part, specimens from Pulo Lankawi. Not S. concolor BLyTH. Type.—Adult male (skin and skull), No. 104,390, United States National Museum. Collected on Pulo Lankawi, off west coast of Malay Peninsula (about 75 miles north of Penang), December 1, 1899, by Dr. W. L. Abbott. Original number, rot. Characters.—Like Sciurus davisoni (Bonhote) but smaller and with scarcely a trace of the reddish areas on neck and sides. Color.—Type: back and sides a uniform grizzle of black and a brown intermediate between the buff and cinnamon of Ridgway, the lighter color somewhat in excess. Face, legs, and feet a finely grizzled gray approaching in general effect the hair-brown of Ridg- way and forming a distinct though not very striking contrast with color of back. Neck with a barely indicated reddish area behind and below ear. Ears slightly more gray than back. Underparts dull gray much like that of legs, but distinctly washed with broccoli- brown on middle of belly. Tail concolor with back above, slightly paler below, the grizzle everywhere more coarse and forming obscure cross-bars beyond middle; pencil clear black, equal to about one- sixth entire length of tail. Skull and teeth—As compared with that of Sciurus davisoni the skull of S. lancavensis is noticeably smaller and the rostrum is relatively shorter and broader. Teeth essentially as in S. davisont. Measurements.—Measurements of type: total length, 410; head and body, 210; tail vertebree; (200; hind: foot, 51 (47.5) ; skull 1 Measurements in parenthesis are those of one of the cotypes (adult male) of Sciurus fraterculus, taken from the original description. MILLER | SEVENTY NEW MALAYAN MAMMALS 7 greatest length, 53; zygomatic breadth, 29.4; median length of nasals, 15.6; greatest breadth of both nasals together, 7.6." Specimens examined.—Two, both from Pulo Lankawi. SCIURUS ADANGENSIS sp. nov. 1900. Sciurus concolor MriiEer, Proc. Biol. Soc. Washington, xt, p. 191, December 21, 1900. Part, specimens from Pulo Adang. Not Sciurus concolor BiyTH. Type.—Adult male (skin and skull), No. 104,389, United States National Museum. Collected on Pulo Adang, Butang Islands, De- cember 14, 1899, by Dr. W. L. Abbott. Original number, 153. Characters.—Like Sciurus lancavensis but smaller. Color.—The color so closely resembles that of Sciurus lancavensis as to require no special description. The sides of the neck are almost exactly concolor with back, and I can detect no trace of a reddish area along sides of body. Skull and teeth.—The skull is shorter and relatively broader than that of Sciurus lancavensis, but otherwise it does not differ appreci- ably in form. Teeth as in the related species. Measurements.—Measurements of type: total length, 395; head and body, 210; tail vertebrz, 185; hind foot, 48 (45) ; skull, greatest length, 51; zygomatic breadth, 30; median length of nasals, 14.8; greatest breadth of both nasals together, 7. Specimens examined.—Three, all from Pulo Adang. SCiURUS SULLIVANUS sp. nov. Type.—Adult female (skin and skull), No. 104,377, United States National Museum. Collected on Sullivan Island, Mergui Archi- pelago, February 1, 1900, by Dr. W. L. Abbott. Original number, 294. Characters.—Similar to Sciurus davisoni (Bonhote) from south- ern Tenasserim, but smaller and darker. Color.—Type: back a fine grizzle of black, buff, and raw-sienna, the last slightly in excess of either of the others. On sides of neck and along sides of body from axilla to inner surface of thigh the color deepens nearly to ochraceous-rufous which forms a very evi- dent contrast with the surrounding parts. Ears, head, legs, and feet grizzled gray, somewhat resembling the mouse-gray of Ridgway but with a silvery gloss. Underparts dull smoke-gray, 1 In an adult male Sciurus davisoni from Bok Pyin, Tenasserim (No. 104,392), the skull measures: greatest length, 57; zygomatic breadth, 32; median length of nasals, 17.4; greatest breadth of both nasals together, 7.6. 18 SMITHSONIAN MISCELLANEOUS COLLECTIONS [VoL. 45 with an indistinct darker wash along median line. Tail like back, but more coarsely grizzled, its terminal black area occupying about one-fourth total length of tail. Skull and teeth—The skull closely resembles that of Sciurus adangensis in form, but in size it is distinctly larger. It is smaller and less elongate than the skull of Sciurus davisoni. Thé teeth show no peculiarities. Measurements.—Measurements of type: total length, 490; head and body, 285 ; tail vertebrze, 205 ; hind foot, 51 (47) ; skull, greatest length, 55; zygomatic breadth, 33; median length of nasals, 16.8; greatest breadth of both nasals together, 7.8. Specimens examined.—Five, all from Sullivan Island. Remarks.—The series shows no variation worthy of note. The color pattern in this species is like that of the mainland animal, and the neck and side markings show no tendency to become obsolete. SCIURUS DOMELICUS sp. nov. Type.—Adult female (skin and skull), No. 104,381, United States National Museum. Collected on Domel Island, Mergui Archi- pelago, February 24, 1900, by Dr. W. L. Abbott. Original num- ber. 322. Characters.—Like Sciurus sullivanus but colors throughout slightly darker, reddish lateral areas spreading over whole ventral surface, black tip of tail longer, and skull with narrower, more elongate rostrum. , Color.—Type: The colors are essentially as in Sciwrus sullivanus but are all a shade darker, the red of the sides and neck approach- ing cinnamon-rufous. Entire underparts heavily washed with dull cinnamon-rufous, a trace of grizzled gray, however, remaining on chest. Legs, feet, and face darker gray than in the related form, but rather lighter and producing a more decided contrast. Tail as in Sciurus sullivanus except that the black terminal area is noticeably longer. Skull and teeth.—The skull differs from that of Sciurus sullivanus in its narrower general form, and in the greater elongation of the rostrum. Its resemblance to the skull of Scirus lancavensis is very close. Measurements.—Measurements of type: total length, 420; head and body, 210; tail vertebrze, 210; hind foot, 51 (48) ; skull, greatest length, 54; zygomatic breadth, 31; median length of nasals, 16.4; greatest breadth of both nasals together, 7. Specimens examined.—Two, both from Domel Island. MILLER | SEVENTY NEW MALAYAN MAMMALS 19 SCIURUS BENTINCANUS sp. nov. Type.—Adult female (skin and skull), No. 104,383, United States National Museum. Collected on Bentinck Island, Mergui Archi- pelago, March 11, 1900, by Dr. W. L. Abbott. Original number, so. Characters.—External appearance essentially as in Sciurus davi- sont but size somewhat greater, and skull with larger, more inflated audital bullze. Color.—Type: The color so closely resembles that of Sciurus davisoni from southern Tenasserim that no detailed description is necessary. The reddish area on sides and neck is, however, slightly more intense, approaching the ferruginous of Ridgway. In the type specimen the posterior half of the back is in a dull worn pelage slightly contrasting with the fresh coat on other parts of the body. Chest and belly with a distinct reddish wash, but this is not suffi- ciently heavy to obscure the usual grizzled gray. Skull and teeth.—In size and general outline the skull does not differ appreciably from that of Sciurus davisoni, but the audital bullz are distinctly larger and more inflated, a character that is particularly noticeable when the skull is viewed from below. The width of the anterior half of the bull in this aspect is fully 1 mm. greater in the island species. The teeth show no peculiarities. Measurements.—Measurements of type: total length, 465; head and body, 235; tail vertebrze, 230; hind foot, 57 (53) ; skull, greatest length, 58; zygomatic breadth, 33. Specimens examined.—Three, all from Bentinck Island. SCIURUS MATTHAUS sp. nov. Type.—Adult female (skin and skull), No. 111,920, United States National Museum. Collected on St. Matthew Island, Mergui Archipelago, December 11, 1900, by Dr. W. L. Abbott. Original number, 774. Characters.—Similar to Sciurus bentincanus but not as large and with reddish areas on neck and sides less strongly marked. Color.—Type: The color throughout is exactly similar to that of Sciurus bentincanus except that the reddish areas on neck and sides are slightly duller and less strongly contrasted with surrounding parts. Skull—The skull is similar to that of Sciurus bentincanus in form, though the size is much less. Notwithstanding the size of the skull the bullz are fully as large as those of Sciurus davisoni. The teeth show no peculiarities. 20 SMITHSONIAN MISCELLANEOUS COLLECTIONS [VoL. 45 Measurements.—Measurements of type: total length, 445; head and body, 220; tail vertebrz, 225; hind foot, 53 (50) ; skull, greatest length, 54; zygomatic breadth, 31. Specimens examined.—Four, all from St. Matthew Island. SCIURUS LUCAS sp. nov. Type.—Adult female (skin and skull), No. 104,385, United States National Museum. Collected on St. Luke Island, Mergui Archi- pelago, January 20, 1900, by Dr. W. L. Abbott. Original number, 250. Characters.—Similar to Sciurus mattheus, but ground color of upperparts and tail strongly suffused with fulvous. Color.—Type: general color above dull, light tawny-ochraceous, grizzled by the black hair-tips and buff subterminal annulations. Sides of neck and of body ferruginous, lighter than that of Ridgway and forming no very marked contrast with color of back. The ferruginous of neck is a little brighter than that of body. Tail like back, but more coarsely grizzled, the black terminal pencil well developed. Skull and teeth.—The skull and teeth exactly resemble those of Sciurus mattheus. Measurements.—Measurements of type: total length, 430; head and body, 220; tail vertebr, 210; hind foot, 54 (50) ; skull, greatest length, 54; zygomatic breadth, 31. Specimens examined.—Two, both from St. Luke Island. Remarks.—Little though it would be anticipated, the squirrel of St. Luke Island differs very noticeably from that of St. Matthew. The characters are perfectly constant so far as the rather small num- ber of specimens is concerned, and from the known stability of the colors in other members of the group there is no reason to suppose that the differences are merely individual. SCIURUS CASENSIS sp. nov. Type.—Adult female (skin and skull), No. 104,370, United States National Museum. Collected on Chance Island, Mergui Archi- pelago, December 28, 1899, by Dr. W. L. Abbott. Original number, 185. Characters.—Like Sciurus davisoni (Bonhote), but size larger, color, particularly that of tail, paler, and reddish areas of neck and sides brighter and more strongly contrasted with surrounding parts. Color—Type: back and crown a fine grizzle of black and light buff, the latter slightly in excess. On sides this becomes suffused MILLER ] SEVENTY NEW MALAYAN MAMMALS Dil with dull ferruginous which increases rapidly in intensity to form an almost clear ferruginous lateral stripe. Neck patch bright ferruginous, strongly contrasted with surrounding parts. Hind legs like back except that inner posterior surface is heavily washed with clear ferruginous. Front legs a clear silvery grizzle of ecru- drab and whitish smoke-gray. Underparts dull buffy gray washed with pale ferruginous on chest. A large, clear, orange-rufous area at groin. Tail distinctly paler than back, the light annulations of the hairs whitish cream-buff on dorsal surface, pale buff below. Black pencil well developed. Skull and teeth—The skull closely resembles that of Sciurus davisomi but is slightly larger, the rostrum is broader and the audital bullz are more inflated anteriorly. Teeth as in the related species. Measurements.—Measurements of type: total length, 465; head and body, 250; tail vertebrz, 215; hind foot, 52 (48) ; skull, greatest length, 57; zygomatic breadth, 33. Specimens examined.—Five, all from Chance Island. Remarks.—lts large size, pale color, the contrast between the upper surface of the tail and the back, and the brightness and dis- tinctness of the ferruginous markings make this species one of the most strongly characterized in the group. The series of skins shows no variations worthy of note. SCIURUS ALTINSULARIS sp. nov. Type.—Adult female (skin and skull), No. 111,975, United States National Museum. Collected on High Island, Mergui Archipelago, December 31, 1900, by Dr. W. L. Abbott. Original number, 810. Characters.—General appearance as in Sciurus casensis, but size smaller, general color paler, upper surface of tail not lighter than back, and ferruginous markings replaced by ochraceous-buff. Color.—Type: back, sides, head, and both surfaces of tail a fine grizzle of black and cream-buff, the latter everywhere in excess and palest on head, darkest on tail. Neck patch dull ochraceous-buff, slightly contrasted with color of surrounding parts. Sides strongly tinged with ochraceous-buff in region between axilla and groin. Underparts smoke-gray, grizzled with broccoli-brown along median line of chest. A small, clear ochraceous-buff area at axilla and another at groin. Inner surface of legs like belly, outer surface and feet somewhat darker. Tail with black pencil well developed. Skull and teeth.—The skull resembles that of Sciwrus casensis but is distinctly smaller. Teeth not peculiar. Measurements.—Measurements of type: total length, 437; head 22 SMITHSONIAN MISCELLANEOUS COLLECTIONS [VoL. 45 and body, 230; tail vertebre, 207 ; hind foot, 50 (47) ; skull, greatest length, 54.6; zygomatic breadth, 31. ; Specimens examined.—Five, all from High Island. SCIURUS RUBECULUS sp. nov. Type.—Adult male (skin and skull), No. 86,777, United States National Museum. Collected at Khow Sai Dow, Trong, Lower Siam, altitude 1000 feet, February 21, 1899, by Dr. W. L. Abbott. Characters.—Similar to Sciurus atrodorsalis Gray, but much larger. No black-backed phase at present known. Color.—Type: upperparts a clear fine grizzle of black and ochra- ceous, the latter somewhat in excess. On sides, legs, cheeks, throat, and along median line from chest to base of tail the ochraceous element of the grizzle is lighter and duller than on the back, ap- proaching the ochraceous-buff of Ridgway. Underparts (except the grizzled area) and inner side of legs nearly to feet, bright orange- rufous, darker and more red than that of Ridgway. The red area is completely divided by the grizzled median line. This is about Io mm. in width. Ears like back. Front feet concolor with legs, hind feet slightly darker. Whiskers black. Tail like back but more coarsely grizzled. On terminal half and pencil the tail is fringed with clear orange-rufous due to a subterminal annulation of this color on each hair. The extreme tips of the hairs forming the fringe are for the most part black, but this color is easily overlooked. On terminal third of tail there is a slight tendency toward the forma- tion of black cross-bars, and on under side through same region two faint black bands may be traced parallel with edge. Skull and teethAs compared with a skull of Sciurus atrodorsalis from Kokareet, Tenasserim, that of S. rubeculus is immediately distinguishable by its greater size, and by the relatively shorter broader rostrum. The audital bullz on the other hand scarcely ex- ceed those of the smaller animal in size, while in form they are less inflated. Frontals nearly flat anteriorly, not distinctly concave as in the allied form. Teeth as in Sciurus atrodorsalis. Measurements.—Measurements of type: total length, 440; head and body, 230; tail, 210; hind foot, 51 (48) ; skull, greatest length, 55° (50) ;* basal length, 49 (42); zygomatic breadth, 33 (20.6) ; least interorbital breadth, 20.4 (17); median length of nasals, 14.8 (14.6) ; greatest breadth of both nasals together, 8 (7). Specimens examined.—Five, from the following localities: Trong, one, the type; Tenasserim, Bok Pyin, 3, Sungei Balik, 1. 1 Cranial measurements in parenthesis are those of an adult female Sciurus 3 3 5488). SMITHSONIAN MISCELLANEOUS COLLECTIONS t. Sciurepterus merens,type. 2. Funan:bulus obscurus, type Vou. 45, PL I MILLER | SEVENTY NEW MALAYAN MAMMALS rape Remarks.—The United States National Museum contains speci- mens of two readily distinguishable squirrels identified as Sciurus atrodorsalis by Mr. Oldfield Thomas. The first is from Kokareet, Tenasserim, and is one of the series recorded in the “Annali del Museo Civico di Storia Naturale di Genova” for 1892 (vol. xxx, p. 929). The second is the type of Sciurus rubeculus, identified at my request. While there can be no question as to the close rela- tionship and general superficial likeness of the two specimens, due allowance being made for the fact that the Tenasserim skin is in the black-backed pelage, the cranial differences are such as to make their separation necessary. It seems highly improbable that the Kokareet animal is the same as true Sciwrus atrodorsalis, the type of which came from Bhotan, but the latter, as well as Sciwrus gordont Ander- son and S. hyperythrus Blyth, appears to agree with it and to differ from Sciurus rubeculus in markedly smaller size. FUNAMBULUS OBSCURUS sp. nov. (Piate I, FIGURE 2) Type.—Adult female (skin and skull), No. 121,640, United States National Museum. Collected on South Pagi Island, Sumatra, November 22, 1902, by Dr. W. L. Abbott. Original number, 2086. Characters—About the size of Funambulus insignis, but tail much shorter. Color darker than in the Sumatran animal, dark lateral stripes obsolete, and underparts iron gray. Skull with un- usually elongate rostrum. Color—Type: back and sides a uniform, very fine grizzle of black and light ochraceous, the latter slightly in excess. Outer sur- face of hind legs similar to back, but that of front legs slightly darker. Crown and cheeks a little darker than back, the cheeks perceptibly tinged with gray. Feet and ears blackish. Median black dorsal stripe about 3 mm. wide. It is well defined, and ex- tends from about middle of neck to lumbar region. Lateral stripes - reduced to mere dark shades too indistinct to be measured, or to have any definite color. They are about 10 mm. distant from the median stripe, and of approximately the same length as the latter. Underparts and inner surface of legs a grizzled iron gray. This is darker and less grizzled on legs and along sides of belly, much lighter and with a frosted appearance on throat and chest. The darker gray about matches the slate-gray of Ridgway, the lighter cannot be accurately described. Tail like back but very coarsely grizzled. Above the ochraceous predominates in the middle and the black forms a border 10 mm. wide. The extreme tips of the 24. SMITHSONIAN MISCELLANEOUS COLLECTIONS [VOL. 45 hairs at the edge are black, but there is a noticeable ochraceous subterminal band. Pencil black. Beneath the black predominates uniformly. Skull and teeth.—While the skull in a general way resembles that of Funambulus insignis the peculiarities of the latter as compared with true Sciurus are in it much exaggerated. The rostrum is greatly elongated, so much so that the skull would hardly at first glance be recognized as that of a squirrel. This elongation, how- ever, does not approach that of the rostrum in Rhinosciurus. In Funambulus msignis the distance from anterior rim of orbit to tip of nasals is about equal to breadth of braincase above roots of zygomata, while in F. obscurus it is decidedly greater than this breadth, though not, as in Rhinosciurus, approximately the same as the zygomatic breadth. The nasals are strongly rounded from tip nearly to base, so much so that near middle they almost form a ridge. Only the extreme proximal end is flat. Zygomata de- cidedly heavier than in Funambulus insignis, but not peculiar in form. Teeth similar to those of the related species but larger throughout. Measurements——External measurements of type: total length, 303; head ‘and body, 220; tail vertebra, 83; hind foot, 44 (42) ; ear from meatus, 14; ear from crown, 7; width of ear, 11.6. Cranial measurements of type: greatest length, 53.6; basal length, 45; basilar length, 43; palatal length, 24; diastema, 14.4; distance from anterior rim of orbit to tip of nasals, 25; length of nasals, 18.4; greatest breadth of both nasals together, 6.2; least breadth of both nasals together, 4; least interorbital breadth, 14; zygomatic breadth, 30; breadth of braincase above roots of zygomata, 20; mandible, 34; maxillary toothrow (alveoli), 10; mandibular toothrow (alveoli), 10. Specimens examined.—Seven, all from the Pagi Islands. Remarks.—The skins are very uniform in color. Practically the only variation shown by the series is in the distinctness of the lateral dark stripes. In four of the specimens these are nearly as well developed as the median stripe, but in none of them is there any close resemblance to Funambulus insignis. FUNAMBULUS ROSTRATUS spp. nov. Type.—Adult female (skin and skull), No. 121,801, United States National Museum. Collected on Tana Bala, Batu Islands, Febru- ary 12, 1903, by Dri Ws Abbott. Original number s22er- Characters.—Similar to Funambulus insignis, but color slightly darker, black dorsal stripes apparently broader, and skull with more elongate rostrum. MILLER] SEVENTY NEW MALAYAN MAMMALS 25 Color.—Dorsal surface a uniform, fine grizzle of black and buff, the latter in excess everywhere except on rump, shoulders, neck, and head. Black stripes well developed, about 7 mm. wide at middle of back, extending from middle of neck to lumbar region, the median faintly prolonged to crown. Region between stripes exactly similar to that outside, but appearing lighter by contrast with the black. Cheeks and outer surface of legs duller and more finely grizzled than back. Feet dark hair-brown. Hairs of tail annulated with black and orange-buff, the general effect a coarse grizzle of the two colors. Underparts and inner surface of legs cream-buff, slightly more yellow than that of Ridgway. Skull and teeth—The skull is larger than that of Funambulus imsignis, the interorbital region is relatively broader, and the rostrum is more produced. In the last character there is a close approach to Funambulus obscurus, but the nasals are flat posteriorly, as in F, insignis. Teeth as in Funambulus insignis, but somewhat larger. Measurements.—External measurements of type: total length, 265 ;' head and body, 197; tail vertebrz, 68;1 hind foot, 45 (41) ; ear from meatus, 15.6; ear from crown, 9; width of ear, 13. Cranial measurements of type: greatest length, 51.6 (49) ;? basal length, 43 (41); basilar length, 40 (38.8); palatal length, 22.8 (21.4) ; diastema, 14 (12.2) ; length of nasals, 17 (16); greatest breadth of both nasals together, 6.6 (6.4) ; interorbital breadth, 14 (12.4) ; distance between tips of postorbital processes, 22.4 (20) ; zygomatic breadth, 30 (26) ;? breadth of braincase above roots of zygomata, 22 (20.6) ;* greatest depth of braincase, 17 (16) ; man- dible, 33 (30) ; maxillary toothrow (alveoli), 9.4 (9) ; mandibular toothrow (alveoli), 10 (9). Specimens examined.—One, the type. Remarks.—The Batu Funambulus is readily distinguishable from the Sumatran F. insignis by its darker color and larger skull. For the opportunity to examine two skins and a skull of the Sumatran animal I am indebted to the Academy of Natural Sciences of Phila- delphia, through the kindness of Mr. Witmer Stone. FUNAMBULUS PENINSULZ pp. nov. Type.——Adult male (skin and skull), No. 86,776, United States National Museum. Collected at Khow Sai Dow, Trong, Lower Siam, February 18, 1899, by Dr. W. L. Abbott. 1Tail apparently imperfect. 2Measurements in parenthesis are those of an adult male Funambulus insignis from Gunong Sugi, Lampong District, Sumatra (No. 6655, Acad. emy of Natural Sciences of Philadelphia). 3 Approximate. 26 SMITHSONIAN MISCELLANEOUS COLLECTIONS [voL. 45. Characters —General color lighter and brighter than in Funam- bulus insignis. Skull larger than in the Sumatran animal, but similar to it in form. Color.—The light element in the grizzle of the upperparts is brighter and more yellow than that in Funambulus insignis and F. rostratus. The exact tint is very close to the raw-sienna of Ridgway, except in the neighborhood of the black stripes, where it is diluted with cream-buff. Front feet hair-brown. Hind feet hair- brown grizzled with yellowish. The light bands on the hairs of the tail are dull raw-sienna, but the tips are grayish. Underparts cream-buff, fading almost to white on chest, and darkening to buff on inner surface of hind legs. Skull and teeth—The skull is distinctly larger than that of Funambulus imsignis and the interorbital region is broader, but otherwise it shows no peculiarities. Teeth as in Funambulus in- SIENIS. Measurements.—Measurements of type: total length, 295; head. and body, 195; tail vertebrze, 100; hind foot, 44 (41) ; skull, greatest: length, 50; length of nasals, 16; diastema, 12; interorbital breadth, 14.4; zygomatic breadth, 28; mandible, 30; maxillary toothrow (alveoli), 8.8; mandibular toothrow (alveoli), 9. Specimens examined.—One, the type. SCIUROPTERUS MARENS spp. nov. (Pate I, FIGURE 1) Type.—Adult female (skin and skull), No. 121,531, United States. National Museum. Collected on North Pagi Island, Sumatra, January 14, 1902, by Dr. W. L. Abbott. Original number, 2206. Characters.—Like Sciuropterus lugens Thomas, but ear much smaller and color not as dark. Color.—Entire animal drab, the back and tail darker than Ridg- way’s plate 11, fig. 18, the underparts lighter and more bluish, closely approaching ecru-drab. In certain lights the shoulders, neck,. and head show a distinct wash of wood-brown or isabella-color. Throughout dorsal surface the underfur is bluish gray. Tail uni- color throughout, except at extreme base below, where it is like belly. Feet and lower portion of legs so thinly haired as not to con- ceal the skin, the hairs dull light drab. Ears and whiskers blackish. Claws light horn-color. Ears.—The ears agree in form with those of Sciuropterus lugens as described by Thomas, except that the posterior border is very slightly concave. In size they are much smaller, measuring only Jamoyy ‘Suory, ‘E€b*bg ‘on ‘areuta} ‘wrno1ovu vinsayzp “S *(adX}) ayeuray ‘wo1qvmoshz vANAaYIp “¥ *(adA}4) ayeur ‘puvnjvg visienvjag “€ “eAvf usaysva ‘66h'1zr ‘ON ‘ayeul ‘vp7272u wEstANYZagd “cs ‘“seUNTeN YON ‘purjsy[ uvindung “2zo*vor ‘on “ayeul ‘v772p7272u VISIANVIZAT “1 ANVITHAOS NVINOSHLING MILLER] SEVENTY NEW MALAYAN MAMMALS 27 15.4 mm. from meatus, and 8 mm. from crown. The measurements given in the original description of S. Jugens are 21.5 and 20 for an adult male and female respectively. Skull and teeth—So far as can be judged from the description the skull and teeth (pl. 1, fig. 1) agree with those of Sciuropterus lugens. Audital bulle very broad and flat. Measurements—External measurements of type: total length, 485; head and body, 265 ; tail vertebrze, 220; hind foot, 49 (44) ; ear from meatus, 15.4; ear from crown, 8; width of ear, 7.6. . Cranial measurements of type: greatest length, 48.8; basal length, 41.4; basilar length, 38; median palatal length, 22; diastema, 10; length of nasals, 14.8; greatest breadth of both nasals together, 7.8; interorbital breadth, 11; distance between tips of postorbital pro- cesses, 23; zygomatic breadth, 29; breadth of braincase above roots of zygomata, 20.6; greatest depth of braincase, 16.4; occipital depth, 12; mandible, 29; maxillary .toothrow (alveoli), 10; mandibular toothrow (alveoli), ro. Specimens examined.—One, the type. Remarks.—While this squirrel is evidently a near ally of the Siporan Sciuropterus lugens, it is readily distinguishable by its smaller ear and less dark color. It needs no comparison with any other species. PETAURISTA BATUANA sp. nov. (Piate II, FIGURE 3) Type.—Adult male (skin and skull), No. 121,742, United States -National Museum. Collected on Tana Bala, Batu Islands, Sumatra, February 5, 1903, by Dr. W. L. Abbott. Original number, 2233. Characters.—Like Petaurista mitidula of the Natuna Islands, but intermediate in size between this species and the large Javan P. nitida. Color.—Type: upperparts deep rufous or ferruginous, a little darkened, particularly along median line, by blackish hair-tips. Upper side of membranes and outer surface of legs burnt-sienna, darkening to black on feet, wrists, ankles, and outer edge of fore- arm. Edge of interfemoral membrane blackish brown. Under- parts and inner surface of thighs and upperarms pale ochraceous- rufous. Chin blackish. On cheeks and muzzle the rufous and ochraceous-buff are mixed together and slightly tinged with gray. Ears rufous internally, black externally. A small blackish area behind ear. Eye ring and area surrounding muzzle and base of whiskers black. Edge of flying membrane dull ochraceous-buff. 28 SMITHSONIAN MISCELLANEOUS COLLECTIONS [VoL. 45 Tail intermediate between the ochraceous-buff of belly and the rufous of back, the tip black. Skull and teeth.—The skull and teeth (pl. 11, fig. 3) exactly re- semble.those of Petaurista nitida (pl. 11, fig. 2) and P. nitidula (pl. u, fig. 1), but are intermediate in size, though perhaps most nearly approaching the Javan animal. Measurements.—External measurements of type: total length, 825 ; head and body, 405; tail vertebrze, 420; hind foot, 75 (68) ; ear from meatus, 35; ear from crown, 24.6; width of ear, 16. Cranial measurements of type: greatest length, 66.4 (72) ;1 basal length, 60 (65) ; basilar length, 56 (60) ; length of nasals, 20 (22.4) ; greatest breadth of both nasals together, 11.8 (12.8) ; median palatal length, 32.6 (33.6); diastema, 15 (16); interorbital breadth, 15.4 (16); distance between tips of postorbital processes, 36 (36); zygomatic breadth, 46.4 (49); breadth of braincase above roots of zygomata, 28.6 (28) ; greatest depth of braincase, 22 (23.6) ; occip- ital depth, 17 (17.4); mandible, 44 (49.6); maxillary toothrow (alveoli), 16 (16) ; mandibular toothrow (alveoli), 16.2 (17). Specimens examined.—Eight, all from the Batu Islands. Seven are from Tana Bala, the other from Tana Masa. Remarks.—While Petaurista batuana is more like P. nitida in size, its color is much lighter than that of the large Javan animal. It is not quite as light, however, as the small P. nitidula of the Natunas, though in all respects except size it very closely resembles this form. MUS STRIDENS sp. nov. 1900. Mus vociferans MILLER, Proc. Wash. Acad. Sci., 1, p. 208, August 20, 1900. Tioman Island. Not M. vociferans M1tiEr, April 21, 1900. Type.—Adult male (skin and skull), No. 104,992, United States National Museum. Collected on Tioman Island, off southeast coast of Malay Peninsula, October 10, 1900, by Dr. W. L. Abbott. Orig- inal number, 702. Characters.—Similar to Mus strepitans of the Anamba Islands, but ears larger and color not as dark. Color.—The color so closely resembles that of Mus strepitans? that no detailed description is necessary. The ochraceous-buff of back and sides is distinctly paler than in the Anamba animal, and the sprinkling of blackish hairs is less noticeable. Underparts and inner 1 Measurements in parenthesis are those of an adult male Petaurista nitida from western Java (No. 121,499). 2See Proc. Wash. Acad. Sci., u, pp. 207-208, August 20, 1900. - MILLER | SEVENTY NEW MALAYAN MAMMALS 29 surface of legs pale straw-yellow, not whitish as in Mus vociferans, the mainland form. Measurements.—External measurements of type: total length, 570 (550) ;+ head and body, 254 (229) ; tail vertebrae, 316 (318) ; hind foot, 46 (45.6); hind foot without claws, 44 (44); ear from meatus, 28 (23); ear from crown, 22 (18). Specimens examined.—Nine, all from Tioman Island. Remarks—This rat is a well marked form of the sabanus- vociferans group. In most characters it is intermediate between the bright Mus vociferans of Lower Siam and the dull dark M. strepitans of the Anamba Islands, but it appears to differ from both in the large size of the ears. ; MUS MATTHZEUS sp. nov. Type.—Adult male (skin and skull), No. 104,159, United States National Museum. Collected on St. Matthew Island, Mergui Archipelago, January 18, 1900, by Dr. W. L. Abbott. Original number, 243. Characters—A large member of the sabanus-vociferans group, differing from Mus vociferans in the duller, more rusty color of the back and sides, and strongly yellowish underparts. Blackish hairs of upperparts not as dark as in the mainland form. Color.—The color is essentially as in Mus vociferans* except that the ground color of back and sides is tawny instead of ochraceous or tawny-ochraceous, and the dark hairs with which the back is sprinkled are distinctly brownish, while in the related species they are nearly black. Underparts and inner surface of legs cream-buff. Measurements.—Measurements of type: total length, 600; head and body, 248; tail vertebre, 350; hind foot, 50 (48); ear from meatus, 27; ear from crown, 21; skull, greatest length, 56; zygo- matic breadth, 25. Specimens examined.—Twelve (one skull without skin), all from St. Matthew Island. Remarks.—In its dark, rich color this rat is strikingly different from the other members of the sabanus-vociferans group occurring in the Mergui Archipelago, all of which, so far as known, are paler than the mainland form. MUS STRIDULUS sp. nov. Type.—Adult female (skin and skull), No. 104,196, United States National Museum. Collected on Bentinck Island, Mergui Archi- 1 Measurements in parenthesis are those of an adult male Mus strepitans from Pulo Jimaja, Anamba Islands (No. 101,736). 2See Proc. Biol. Soc. Washington, xt, p. 138, April 21, 1900. 30 SMITHSONIAN MISCELLANEOUS COLLECTIONS [VOL. 45 pelago, March 12, 1900, by Dr. W. L. Abbott. Original number, 350- Characters.—A light-colored member of the sabanus-vociferans group, with disproportionately small feet. Color.—The ground color of back and sides is buff, rather darker and brighter than that of Ridgway, but scarcely approaching ochra- ceous. Back rather thickly sprinkled with dark brown hairs, sides much less so. Underparts cream-buff. Measurements.—Measurements of type: total length, 514; head and body, 222; tail vertebrz, 292; hind foot, 41 (38.6) ; ear from meatus, 25.4; ear from crown, 20; skull, greatest length, 52; zygo- matic breadth, 24. Hind foot in two other adult females, 41.4 (39.6) and 40 (38). Hind foot in two adult males, 42 (40) and 41.6 (39.4). Specimens examined.—Six, all from Bentinck Island. MUS LUCAS sp. nov. Type.—Adult female (skin and skull), No. 104,190, United States National Museum. Collected on St. Luke Island, Mergui Archi- pelago, January 20, 1900, by Dr. W. L. Abbott. Original number, Zee Characters.—Similar to Mus stridulus but with feet of normal size. Color—The color is almost exactly the same as that of Mus stridulus from Bentinck Island, though perhaps a trifle more yellow- - ish. It is therefore strikingly different from that of the dark Mus mattheus of St. Matthew Island. Measurements.—Measurements of type: total length, 525; head and body, 215; tail vertebre, 309; hind foot, 46 (44); ear from meatus, 26; ear from crown, 22; skull, greatest length, 52.4; zygo- matic breadth, 23. Specimens examined.—Three, all from St. Luke Island. Remarks.—While this species differs widely from its near geo- graphical ally Mus mattheus it rather closely resembles the Mus lancavensis of Pulo Lancawi. It is distinguishable from the latter by the more reddish cast of the upperparts, and by the less con- spicuous grizzle produced by the dark hairs of the back. MUS SOCCATUS sp. nov. Type.—Adult male (skin and skull), No..121,549, United States National Museum. Collected on North Pagi Island, Sumatra, De- cember 29, 1902, by Dr. W. L. Abbott. Original number, 2183. Characters.—A large, very dark member of the sabanus-vociferans “MILLER | SEVENTY NEW MALAYAN MAMMALS 31 group, Closely similar to Mus siporanus Thomas, but with rufous of upperparts replaced by dull buff. Color—Type: back and sides a mixture of blackish brown and dull buff, the former predominating on neck, shoulders, and fore part of back, and completely excluding the buff on lumbar region, rump, and about base of tail, the latter predominating on sides of neck and body, particularly in region of shoulders. Throughout the dorsal surface the hairs are ecru-drab at base. Crown, muzzle, and cheeks to level of lower eyelid glossy brownish black, lower half of cheeks like sides of body but more finely grizzled and somewhat darker. Ears and whiskers black. Outer surface of legs like sides proximally, but soon darkening to blackish brown, a color which covers entire dorsal surface of feet. Underparts whitish cream-buff to base of hairs. A narrow hair-brown median streak 30 mm. in length on lower portion of throat. Chest slightly discolored by a brownish stain. The color of belly extends down inner side of front leg to foot, but on hind leg it is abruptly cut off a little above middle of tibia by the blackish brown of lower half of leg. Tail blackish brown above to middle, whitish throughout terminal half. The under side of the basal half is irregularly marbled with whitish and brown. Mamme.—Mamme 8, four pectoral and four inguinal. Fur.—The quality of the fur is as in Mus vociferans, M. sabanus, and their relatives. The flattened, grooved hairs are so numerous as to give the fur a coarse, harsh texture, but they do not resemble bristles. Rump and lumbar region without noticeably elongated hairs. 5 Skull and teeth—The skull resembles that of the members of the sabanus-vociferans group, but is distinguishable by its large size, relatively short, wide, incisive foramina, and by the unusually broad, rounded, angular process of the mandible. Teeth as in Mus vociferans, but larger. Measurements——External measurements of type: total length, 592; head and body, 276; tail vertebrae, 316; hind foot, 53 (50) ; ear from meatus, 24; ear from crown, 18; width of ear, 16. External measurements of an adult female (No. 121,540) from the type locality: total length, 562; head and body, 267; tail vertebra, 295 ; hind foot, 52 (49.0). Cranial measurements of type: greatest length, 58; basal length, 50; basilar length, 47; diastema, 15; length of incisive foramen, 7.8; combined breadth of incisive foramina, 3.8; length of nasals, 23; least interorbital breadth, 10; zygomatic breadth, 27; mandible, 32.6; maxillary toothrow (alveoli), 10.2; mandibular toothrow (alveoli), EO;2. 32 SMITHSONIAN MISCELLANEOUS COLLECTIONS [VoL. 45. Specimens examined.—Twenty-four (6 skulls without skins), all from the Pagi Islands. Remarks.—This is the largest and most conspicuous member of the sabanus-vociferans group that has yet come to my notice. It is evidently a near relative of Mus siporanus of Sipora Island, but the color is without trace of rufous. ‘The series presents little variation in the color of the upperparts, and such as there is consists chiefly in the greater or less extent of the black lumbar area. On the under- parts the brown median chest line is better developed in some speci- mens than in others, though it is apparently never absent. In one skin (female, No. 121,539) it extends back to middle of belly. In most of the specimens there is a trace of the brown stain on the chest, and in four a bright yellow wash covers parts of the belly and even spreads to the posterior half of back. It is undoubtedly due to stain. MUS MASA: sp. nov. Type.—Adult female (skin and skull), No. 121,822, United States. National Museum. Collected on Tana Masa, Batu Islands, Febru- ary 21, 1903, by Dr. W. L. Abbott. Original number, 2327. Characters —A small member of the sabanus-vociferans group, with dull ochraceous back and sides, whitish underparts, and wholly brown tail. Skull like that of Mus soccatus, but not as large. Mamme, 8. Fur.—The quality of the fur is as in the members of the sabanus- vociferans group, that is, the flattened, grooved hairs, while not stiff enough to form spines, are so numerous as to make the fur coarse and harsh. The hairs of the rump and lumbar region are elongated, but not conspicuously so. Mamme.—Mamme 8, four pectoral and four inguinal. Color.—Back and sides mixed blackish brown and pale dull ochraceous or yellowish buff, the exact shade of the lighter color not easily defined, but approaching the ochraceous of Ridgway on shoulders, and fading about to buff on sides. Along middle of back the blackish brown predominates, while on sides the buff becomes almost clear. Crown like back, but more finely grizzled, and slightly tinged with gray; cheeks like sides, but duller. Outer surface of legs like sides, but tinged distally with hair-brown. Feet dull in- definite light brown. Underparts and inner surface of legs white tinged with cream-buff. Ears and tail dark brown, the latter not as dark below as above, but in no way bicolor. Skull and teeth.—Skull almost exactly like that of Mus soccatus except that it is not as large; the incisive foramina are relatively MILLER] SEVENTY NEW MALAYAN MAMMALS 33 longer and narrower, and the interpterygoid space is narrower throughout and less flaring anteriorly. Teeth as in Mus soccatus and the members of the sabanus-vociferans group. Measurements—External measurements of type: total length, 510; head and body, 235; tail vertebrz, 275 ; hind foot, 43 (41) ; ear from meatus, 22.4; ear from crown, 17; width of ear, 16. Cranial measurements of type: greatest length, 54; basal length, 45.4; basilar length, 42.4; diastema, 14.8; length of incisive foramen, 7.8; combined breadth of incisive foramina, 3; length of nasals, 20; least interorbital breadth, 9; zygomatic breadth, 14.6; mandible, 35; maxillary toothrow (alveoli), 10; mandibular toothrow (alveoli), 11. Specimens examined.—Six, all from Tana Masa, Batu Islands. MUS BALZ pp. nov. Type.—Adult female (skin and skull), No. 121,781, United States National Museum. Collected on Tana Bala, Batu Islands, Sumatra, February 12, 1903, by Dr. W. L. Abbott. Original number, 2274. Characters.—Like Mus mase but not as large; skull with relatively broader rostrum. Color.—The color so closely resembles that of Mus mas@ as to need no special description. Skull and teeth—The skull is like that of Mus mase, but the general size is rather less, and the rostrum is actually as well as relatively more robust. Interpterygoid space wider and more ex- panded anteriorly than in the related species. Measurements.—External measurements of type: total length, 465 ; head and body, 227; tail vertebrae, 238; hind foot, 41 (38.6) ; ear from meatus, 23; ear from crown, 18; width of ear, 17. Cranial measurements of type: greatest length, 52; basal length, 44; basilar length, 42; diastema, 14; length of incisive foramen, 8.2; combined breadth of incisive foramina, 3.4; length of nasals, 20; - least interorbital breadth, 9; zygomatic breadth, 25.4; mandible, 32; maxillary toothrow (alveoli), 9.4; mandibular toothrow (alveoli), 9.8. Specimens examined.—Two skins and five skulls, all from Tana Bala. MUS LUGENS sp. nov. Type.—Adult female (skin and skull), No. 121,533, United States National Museum. Collected on North Pagi Island, Sumatra, November 15, 1902, by Dr. W. L. Abbott. Original number, 2046. Characters.—A very large member of the Mus rattus group, some- what closely resembling the Mus simalurensis of Simalur Island, but 34 SMITHSONIAN MISCELLANEOUS COLLECTIONS [voL. 45 larger, darker, and less buffy. Skull with audital bullz actually as well as relatively smaller than in the Simalur rat. Color.—Type: upperparts from muzzle to base of tail bluish black, faintly grizzled with dull buff. On sides the black fades to hair-brown, and the dull buff slightly predominates. Outer surface of legs light hair-brown. Underparts and inner surface of legs smoke-gray, somewhat paler than that of Ridgway. Median line of chest with broccoli-brown streak. Feet dull broccoli-brown. Ears and tail blackish. Skull and teeth—The skull resembles that of Mus simalurensis, but is larger and relatively narrower. The audital bull are actually smaller than in Mus simalurensis and the groove below and in front of meatus is more strongly developed. Teeth as in Mus simalurensis. Measurements.—External measurements of type: total length, 436; head and body, 226; tail vertebrzee, 210; hind foot, 41 (38); skull, greatest length, 51.4 (47.4) ;1 basal length, 45.4 (41.8) ; zygomatic breadth, 23.4 (23) ;* interorbital constriction, 7 (7.4) ;1 length of nasals, 20 (18).* Specimens examined.—Eight, all from the Pagi Islands. Remarks.—The only species with which this rat needs compari- son is the Mus simalurensis of Simalur Island. It is slightly larger than the Simalur rat, and the underparts lack all trace of buff. The back is also decidedly darker and less distinctly grizzled. MUS JULIANUS sp. nov. Type.—Adult female (skin and skull), No. 112,393, United States National Museum. Collected on St. Julian Island, South China Sea, June 2, 1901, by Dr. WoL. Abbott Original number, 0372 Characters.—A moderate sized, light colored member of the Mus rattus group, resembling Mus siantanicus of the Anamba Islands in general appearance, but readily distinguishable by the large, globular audital bulla, short, deep rostrum, and the massive structure of the anterior zygomatic roots. Color.—Upperparts dull yellowish wood-brown, the back darkened by a plentiful sprinkling of blackish and dark hair-brown hairs, the sides lightened by the appearance at surface of ecru-drab under fur. Muzzle and outer surface of legs ecru-drab, the latter faintly washed with wood-brown. Whiskers mixed blackish and whitish. Under- parts and inner surface of legs cream-buff, clouded with ecru-drab 1Type of Mus sinalurensis. - MILLER] SEVENTY NEW MALAYAN MAMMALS 35 laterally and along median line of chest. Feet hair-brown. Ears and tail uniform blackish brown. Skull and teeth.—While of about the same length as that of Mus siantanicus the skull differs from that of the Anamba rat, as well as all other members of the group, in its very heavily built rostrum and anterior zygomatic roots. The plate forming outer wall of infraorbital foramen is much wider than in any other species of approximately equal size, and its upper margin is so flaring that the foramen is a very conspicuous feature of the skull when viewed from above. Audital bullz unusually large, subglobular in outline when skull is viewed from below. Teeth in no way peculiar. Measurements.—Measurements of type: total length, 379; head and body, 190; tail vertebrz, 189; hind foot, 36 (34); ear from meatus, 19.6; ear from crown, 13; width of ear, 14; skull, greatest length, 41; basal length, 37.4; zygomatic breadth, 20.6; interorbital constriction, 7. Specimens examined.—Six (three in alcohol), all from St. Julian Tsland. Remarks.—This easily recognizable member of the Mus rattus group is very different from the large Mus tambelanicus, its nearest geographical ally. In size it closely agrees with Mus siantanicus, but the cranial characters of the two species are very different. MUS GILBIVENTER sp. nov. Type.—Adult male (skin and skull), No. 104,153, United States National Museum. Collected on Sullivan Island, Mergui Archi- pelago, February 2, 1900, by Dr. W. L. Abbott. Original number, 295. Characters.—Similar to Mus cremoriventer but with coarse, spiny fur resembling in quality that of the members of the surifer group. Color.—The elements of the color are exactly as in Mus cremori- venter,’ except that the spines on the back and sides are strongly tinged with cream-buff. These spines everywhere appear conspicu- ously on the surface instead of being completely hidden by the softer elements of the fur as is the case in Mus cremoriventer. The effect on the general color is to produce a peculiar variegated grizzle of echraceous, light yellowish horn-color, and blackish brown, notice- ably different from the more uniform ochraceous of the related spe- cies, and suggesting some of the members of the surifer group. Cheeks and narrow line along sides bright, clear ochraceous. Belly and inner surface of legs cream-buff rather yellower than that of 1 See Proc. Biol. Soc. Washington, x11, p. 144, April 21, 1900. 36 SMITHSONIAN MISCELLANEOUS COLLECTIONS [VoL. 45 Mus cremoriventer. Tail uniform brown throughout, the tip with distinct brown pencil. Neither the skin of the tail nor its hairs are as dark as in the related species. Feet dirty whitish, clouded with brown. Skull and teeth.—The skull resembles that of Mus cremoriventer,* but is slightly larger; the rostrum is deeper, the incisive foramina are longer, and the audital bullz are less globular. Mandible more robust than in the related species, and concavity between condyle and angular process not as deep. Teeth as in Mus cremoriventer, but slightly larger throughout. Measurements.—Total length, 310 (317) ;? head and body, 125 (146) ; tail vertebre, 185 (171); pencil, 5 (8); hind foot, 28.8 (30°); hind foot without claws, 27 (28.5?) ; ear from meatus, 17 (17), ;}ear from icrown, 13.6 (13) ; skull; greatest leneth, 35.(44)); zygomatic breadth, 16 (15.4) ; length of nasals, 13 (11.8) ; depth of rostrum behind incisors, 7 (6) ; length of incisive foramen, 6.4 (5.6). Specimens examined.—One, the type. Remarks.—While the skull, teeth, and unicolor tail show that this rat is closely related to Mus cremoriventer, the conspicuous appear- ance at the surface of the spines in the fur of back and sides gives the animal a strong superficial resemblance to the members of the surifer group. That this peculiarity is not due to moult is shown by the fresh, unabraded condition of the fur. Not only are the soft hairs less abundant than in Mus cremoriventer, but the spines are distinctly larger. This difference is most noticeable on the back and sides, but is also apparent on the underparts, particularly in region between and just behind front legs. MUS LUTEOLUS pp. nov. Type.—Adult female (skin and skull), No. 104,276, United States National Museum. Collected on St. Matthew Island, Mergui Archipelago, January 15, 1900, by Dr. W. L. Abbott. Original num- ber, 226. Characters—A small member of the rajah-surifer group, re- sembling Mus flavidulus of Pulo Lankawi, but with tawny element in color distinctly less yellow. Color.—Ground color of back and sides buff, almost exactly simi- lar to that of Ridgway, very uniform throughout, though a little more yellow on shoulders and neck. Back thickly sprinkled with black-tipped spines, but the black is nowhere in excess of the buff, 1 Proc. Biol. Soc. Washington, xu, pl. v, fig. 2, April 21, 1900. ® Measurements in parenthesis are those of the type of Mus cremoriventer. MILLER] SEVENTY NEW MALAYAN MAMMALS 37 nor is there any tendency toward the formation of a dark dorsal line. Laterally the dark tips become less and less abundant, their color at the same time fading through bluish gray nearly to white, so that near line of demarkation between color of sides and that of under- parts the buff is essentially pure. Underparts and inner surface of legs white, distinctly tinged with cream-buff. A well defined whitish patch covers area on each side of muzzle from which whiskers spring. Eye surrounded by a well-defined dark ring. Measurements.—Measurements of type: total length, 360; head and body, 190; tail, 170; hind foot, 37 (35); skull, greatest length, 45; zygomatic breadth, 10. Specimens examined.—Twenty-three (four skulls without skins), all from St. Matthew Island. Remarks.—The series of nineteen skins presents no variation in color worthy of special comment. One specimen has the tail longer than the head and body, but in all the rest the proportions are essen- tially as in the type. MUS UMBRIDORSUM sp. nov. Type.—Adult male (skin and skull), No. 104,227, United States National Museum. Collected on Loughborough Island, Mergui Archipelago, January 24, 1900, by Dr. W. L. Abbott. Original number, 269. Characters.—Similar to Mus luteolus, but ground color above slightly more yellow, and black of back tending to concentrate into a noticeable dorsal streak. Color—tThe ground color of back and sides is intermediate be- tween the buff and buff-yellow of Ridgway, though nearer the latter. Black-tipped spines slightly more abundant than in Mus luteolus, those of middle of back forming a distinct though not sharply out- lined dorsal streak about 10 mm. wide, extending from nape to base of tail. Otherwise as in Mus luteolus. Measurements.—Measurements of type: total length, 350; head and body, 180; tail vertebre, 170; hind foot, 40 (38) ; skull, greatest length, 42; zygomatic breadth, 18. Specimens examined.—Thirteen (five skulls without skins), all from Loughborough Island. Remarks.—Specimens vary somewhat in the yellowness of the up- perparts and in the distinctness of the dark dorsal area, but the series considered as a whole is easily distinguishable from those pro- cured on other islands or on the mainland. In every individual the length of head and body exceeds that of tail. 38 SMITHSONIAN MISCELLANEOUS COLLECTIONS [voL. 45 MUS BENTINCANUS spp. nov. Type.—Adult female (skin and skull), No. 104,269, United States National Museum. Collected on Bentinck Island, Mergui Archi- pelago, March 11, 1900, by Dr. W. L. Abbott. Original number, 348. — Characters.—A rather large member of the rajah-surifer group, much resembling Mus butangensis, but color of back less strongly suffused with yellow. Color.—The color is so similar to that of Mus umbridorsum as to need no special description. The black of the back, however, shows very little tendency toward the formation of a dorsal stripe. Measurements——Measurements of type: total length, 385; head and body, 205; tail, 180; hind foot, 42 (40) ; skull, greatest length, 47; zygomatic breadth, 22. Specimens examined.—Eight (two skulls without skins), all from Bentinck Island. Remarks.—This race is distinguishable from Mus butangensis by its lighter color, and from its relatives of the Mergui Archipelago by its large size combined with the absence of a throat band. MUS CASENSIS sp. nov. Type.—Adult male (skin and skull), No. 104,249, United States National Museum. Collected on Chance Island, Mergui Archipel- ago, December 28, 1899, by Dr. W. L. Abbott. Original number, 188. Characters.—Similar to Mus butangensis, but with well developed throat band. Colr.—The color so closely resembles that of Mus butangensis as to require no detailed description.t As in this form the ground tint above is a yellowish ochraceous-buff, noticeably brighter than that of Mus bentincanus. Underparts whitish cream-buff, the throat crossed by a yellowish ochraceous-buff stripe about 30 mm. in width. Measurements.—Measurements of type: head and body, 200; tail imperfect ; hind foot, 42 (40) ; skull, greatest length, 47; zygomatic breadth, 21. Specimens examined.—Seven (two skulls without skins), all from Chance Island. Remarks.—As in the other insular forms of the Mus surifer group the tail is shorter than the head and body.? In all five specimens ' See Proc. Biol. Soc. Washington, x11, p. 190, December 21, 1900. 2 In an adult male (No. 104,284), with head and body the same length as in the type, the tail measures 178 mm. MILLER] SEVENTY NEW MALAYAN MAMMALS 39 the throat band is well developed. One female (No. 104,250), has the belly strongly darkened with gray. MUS DOMELICUS sp. nov. Type.—Adult female (skin and skull), No. 104,257, United States National Museum. Collected on Domel Island, Mergui Archipelago, February 24, 1900, by Dr. W. L. Abbott. Original number, 320. Characters.—Similar to Mus casensis, but ground color above not as bright, and throat band less distinct. Color.—As in Mus casensis the ground color above is a yellowish ochraceous-buff, but it is so much lighter than in the related form that the difference is readily noticeable when series are compared. Underparts as in Mus casensis, except that the throat band is a dull, light ochraceous-buff, which produces no strong contrast with the surrounding parts. Measurements.—Measurements of type: total length, 350; head and body, 200; tail vertebrae, 150; hind foot, 40 (38) ; skull, greatest length, 49. Specimens examined.—Seven (one extra skull), all from Domel Island. MUS PAGENSIS sp. nov. Type.—Adult male (skin and skull), No. 121,629, United States National Museum. Collected on South Pagi Island, Sumatra, De- cember 23, 1902, by Dr. W. L. Abbott. Original number, 2153. Characters.—A large, dark member of the rajah-surifer group, somewhat resembling Mus catellifer of Pulo Mansalar, but size greater, tail relatively longer, and collar obsolete or absent. Color.—Type: back and sides ochraceous, the former heavily over- laid with blackish horn-color, the latter nearly clear. The ochra- ceous area widens considerably at shoulder and extends over outer surface of legs to feet, and also across rump. On hind legs it is darkened by a brownish wash. Top of head like back. Cheeks dirty brownish buff. Feet whitish, shaded with dark brown. Tail dis- tinctly but not conspicuously bicolor to tip, blackish above, whitish below. Underparts and inner surface of legs pale cream-buff. Skull and teeth—The skull and teeth are larger than those of Mus catellifer, but do not differ appreciably in form. Measurements.—Measurements of type: total length, 406; head and body, 216; tail vertebrze, 190; hind foot, 46 (43.6) ; ear from meatus, 24; ear from crown, 17; width of ear, 15; skull, greatest length, 50.4; basal length, 40; length of nasals, 19; zygomatic breadth, 21; interorbital constriction, 8. AO SMITHSONIAN MISCELLANEOUS COLLECTIONS [VoL. 45 Specimens examined.—Eleyen, all from the Pagi Islands. Remarks.—While this species is as dark in general coloration as Mus catellifer, it is readily distinguishable by its greater size, and by the less development of the brown collar. In six of the speci- mens the throat is entirely pure white. The five others show traces of a collar, and in two the band is complete though much less no- ticeable than in the Mansalar rat. CHIROPODOMYS NIADIS sp. nov. Type.—Adult female (skin and skull), No. 121,867, United States National Museum. Collected at Lafau, Nias Island, Sumatra, March 30, 1903, by Dr. W. L. Abbott. Original number, 2413. Fic. 1.—Skull of Chiropodomys niadis (X 2). Characters.—Size as in Chiropodomys glirioides, but color not as dark, and skull with dorsal profile strongly convex. Color.—Upperparts wood-brown, everywhere paler than that of Ridgway, and tinged, particularly on sides and flanks, with drab. On middle of back the blackish hair-tips produce a slight dusky cast. MILLER] SEVENTY NEW MALAYAN MAMMALS 41 Underparts white, faintly tinged with cream-buff. Line of demar- kation sharp. Tail and ears uniform dark brown. Feet whitish,, with dusky cloudings on metapodials. Whiskers black. Skull and teeth.—The skull differs from that of Chiropodomys glirioides as described and figured by Sclatert in several important details. Most conspicuous among these is the convexity of the dorsal outline. From tip of nasals to middle of frontals the outline is essen- tially straight, distinctly less convex than in Sclater’s figure. From middle of frontals back, however, the convexity is so strong that when the nasals are held in line with those of the figure the inion falls opposite the middle of occipital condyle. Viewed from below the skull differs from that of C. glirioides in the less broadened ante- rior palatine foramina, and narrower, parallel-margined, interptery- goid space. The width of the latter barely equals one-half the length of the anterior palatine foramina, while in C. glirioides, as both de- scribed and figured, it is considerably more than half. Mandible shorter and deeper than that of C. glirioides, the angular process wider and less distinctly marked off from main portion of jaw. Teeth, so far as can be judged from the description and figures, essentially as in C. glirioides. Measurements.—Total length, 183; head and body, 81; tail ver- tebre, 102; pencil, 4; hind foot, 19 (18) ; ear from meatus, 14.6; ear from crown, 12; width of ear, 10; skull, greatest length, 24; basal length, 20.4; basilar length, 18.6; zygomatic breadth, 14; least inter- orbital breadth, 4.8; greatest breadth of braincase above roots of zygomata, 12; diastema, 6.8; length of anterior palatine foramen, 3.6; breadth of both foramina together, 1.8; greatest width of inter- pterygoid space anteriorly, 1.6; mandible, 14; maxillary toothrow (alveoli), 3.6; mandibular toothrow (alveoli), 3.2. Specimens examined.—One, the type. Remarks.—This species was first recorded by Modigliani in 1889,” and has since been mentioned by Thomas.* Neither writer, how- ever, appears to have examined the skull. Unless the figures pub- lished by Sclater are grossly inaccurate, and there is no reason to suppose that such is the case, the Nias animal is readily distinguish- able from Chiropodomys glirioides. It needs no special comparison with the Bornean forms. 1Proc. Zool. Soc. London; 1890, p. 533, pl. xiv, figs. 6a and 6b. 2 Ann. Mus. Civ. di Storia Nat. di Genova, 2d ser., VII, p. 244, 1889. 3Tbid., X, p. 942, 1802. 42 SMITHSONIAN MISCELLANEOUS COLLECTIONS [VoL. 45 ATHERURA ZYGOMATICA sp. nov. ‘(PLATE II, FIGURE 4) Type.—Adult female (skin and skull), No. 112,429, United States National Museum. Collected on Pulo Aor, off coast of Johore, June 6, 1901, by Dr. W. L. Abbott. Original number, 1009. Characters.—Like Atherura macroura, from Trong, Lower Siam, but color darker and more bluish, and zygoma shorter and deeper. Color.—Type: The elements of the color are essentially as in the mainland animal, that is, the quills are deep bluish horn-color, varied with whitish. But the horn-color is more blue and less brown than in the related form ; and the whitish areas are less developed, so that the dark predominates everywhere, instead of only on back, head and legs. Underparts, except dull whitish area on chin and another in each axilla, dark drab horn-color, darker and clearer on inner side of hind legs, lighter and speckled with whitish along sides of body. Feet, and bristles on naked portion of tail, horny blackish. Ears and whiskers blackish. Tuft of bristles at end of tail dull whitish horn-color. Skull and teeth—In general appearance the skull (pl. um, fig. 4) resembles that of Atherura macroura (pl. m, fig. 5). Its size, however, is slightly less, and in the form of the lachry- mal bone and of the zygoma it appears to differ constantly from the skull of the mainland animal. In Atherura macroura the lachrymal is fully 8 mm. in length below rim of orbit, while above it extends forward as a triangle of bone at least 5 mm. long, and is a noticeable feature of the dorsal aspect of the skull. In A. zygomatica its length below rim of orbit is usually about 5 mm. (in only one skull out of seven does it exceed 6 mm.), while the forward extension is often obsolete, and never large enough to be more than barely visible when the skull is viewed from above. Zy- goma shorter than in Atherura macroura, the jugal deeper in propor- tion to its length, more abruptly concave above, and its lower con- tour invariably—even in a specimen so young that the posterior molar is not yet in place—broken by a strongly developed concavity beneath posterior jugal suture, this concavity terminating anteriorly on the posterior upper surface of a well-marked tooth-like projec- tion. This character is present in all of the seven skulls of Atherura sygomatica, and is barely indicated in the three skulls of A. macroura that I have examined. Paroccipital processes broad and stout, not slender and elongate as in the related species. Posterior section of mandible shorter and deeper than in A. macroura. *(adAq) afeulay ‘vom apvsrima e "eiyeuns ‘Avg ynuedey ‘rot tir ‘ON ‘O[BUuld] “774 97AAPAVY a7pVS11Na ore [ J d OLE 5 der eo N 21 LGR Zi iy f SNOTLOATTIOZ) SNOANVITA IST NVINOSHLING a MILLER] SEVENTY NEW MALAYAN MAMMALS 43 Teeth as in Atherura macroura. I can detect no peculiarities in the pattern of enamel folding. Measurements—External measurements of type: total length, 720 ; head and body, 520; tail vertebree, 200; hind foot, 64 (61) ; ear from meatus, 34; ear from crown, 21; width of ear, 17. Cranial measurements of type: greatest length, 93 (99) ;1 basal length, 82 (87) ; basilar length, 77 (80) ; length of nasals, 26 (25) ; greatest breadth of both nasals together, 14 (15); diastema, 28 (28.4) ; zygomatic breadth, 45.4 (47.6) ; least interorbital breadth, 26 (28.6) ; length of zygoma from posterior rim of antorbital fora- men, 29 (31); depth of zygoma at anterior jugal suture, 8.6 (8); mandible, 58 (61) ; maxillary toothrow (alveoli), 17 (17.6) ; man- dibular toothrow (alveoli), 17.8 (19). Specimens examined.—Seven, all from Pulo Aor. HEMIGALE MINOR sp. nov. (Prate III, Ficure 2) Type.—Adult female (skin and skull), No. 121,651, United States National Museum. Collected on South Pagi Island, Sumatra, De- cember 27, 1902, by Dr. W. L. Abbott. Original number, 2173. Characters.—Like Hemigale hardwicktt Gray, but smaller and darker. Color.—Type: upperparts and outer surface of legs a buffy gray, much less yellow than that of Hemigale hardwicku. The exact shade is not easy to describe, but it is a mixture of broccoli-brown or pale isabella-color and grayish white, the former beneath the surface, the latter at the tips of the hairs. Dark markings clear black, exactly as in Hemigale hardwickw except that they are all wider and those of head and neck are ill-defined, owing to their tendency to spread over the intervening gray. Underparts light brownish buff, or pale, yellowish isabella-color, the tips of most of the hairs grayish white, these most noticeable on throat and forepart of chest. Inner sur- face of legs an indefinite yellowish gray. Feet broccoli-brown washed with buffy gray. Tail blackish brown except for two gray- ish rings at base and a grayish ventral area extending about to middle. Ears dusky brownish outside, dull buffy gray inside. Skull and teeth.—Except for its smaller size the skull (pl. 11, fig. 2) closely resembles that of Hemigale hardwicku (pl. 11, fig. 1) in general appearance. The audital bullze are, however, less ele- vated above surface of basioccipital (when skull is held upside 1 Measurements in parenthesis are those of an adult female Atherura macroura from Trong, Lower Siam (No. 84,433). 44 SMITHSONIAN MISCELLANEOUS COLLECTIONS [VoL. 45 down) and the interpterygoid space is much narrower in proportion to its length. The constriction behind postorbital processes appears to be deeper and better defined than in the larger animal, but this character may not be constant. Teeth as in Hemigale hardwicku but smaller. Measurements.—External measurements of type: total length, 760; head and body, 480; tail vertebrze, 280; hind foot, 67 (64) ; ear from meatus, 30; ear from crown, 26; width of ear, 19. External measurements of adult male (No. 121,650) from the type locality: total length, 715; head and body, 465; tail vertebrz, 250; hind foot, 65 (63). Cranial measurements of type: greatest length, 95 (99) ;? basal length, 91.4 (96) ; basilar length, 87 (91); median palatal length, 50 (52); breadth of palate between anterior molars, 16.8 (16) ; length of interpterygoid space, 15.8 (15.4) ; width of interpterygoid space at middle, 6 (7.8) ; constriction in front of postorbital processes, 19 (18.4); constriction behind postorbital processes, 12.4 (17); zygomatic breadth, 45 (47) ; mastoid breadth, 31 (32); breadth of braincase above roots of zygomata, 29.8 (31) ; mandible, 66.4 (70) ; maxillary toothrow exclusive of incisors (alveoli), 37 (39); man- dibular toothrow exclusive of incisors (alveoli), 40 (42). Specimens examined.—Six, all from South Pagi Island. Remarks.—This species is well characterized by the dark color of its head and the indistinctness of the black markings in this region. The pattern is exactly the same as that of Hemigale hardwicku, but the gray is so dull and the black so diffused that the longitudinal stripes are no longer a conspicuous feature of the animal’s markings. The dark cross-bands on the back average wider than in the related species but their outlines are nearly as well defined. The series presents no variation worthy of note. PARADOXURUS LIGNICOLOR sp. nov. (PLateE IV, FIGURE I, AND PLATE V, FIGURE I) Type.—Adult male (skin and skull), No. 121,645, United States National Museum. Collected on North Pagi Island, Sumatra, No- vember 19, 1902, by Dr. W. L. Abbott. Original number, 2068. Characters.—Not as large as Paradoxurus hermaphroditus, but general form the same. Color yellowish brown with no dark mark- ings; tail lighter than body. Teeth relatively larger than those of 1 Measurements in parenthesis are those of an adult female Hemigale hardwickit (No. 114,461), not as old as the type of H. minor, from Tapanuli Bay, Sumatra. MILLER] SEVENTY NEW MALAYAN MAMMALS 45 P. hermaphroditus, but not essentially different in form. Nasals flat posteriorly. Color.—Entire animal wood-brown, the belly and tail tinged with buff, the back somewhat lightened by the indistinct, grayish, sub- terminal annulations of the hairs. Dorsal line from crown to base of tail clouded by a wash of mummy-brown, most distinct on head and neck. Feet and outer surface of front legs tinged with broccoli- brown. The fur of the upperparts, chin, and throat is drab at base, that of belly and tail buffy throughout. Everywhere the hairs have a distinct silvery gloss, but this is especially noticeable on distal half of tail. Whiskers light wood-brown. Skull and teeth—The skull (pl. 1v, fig. 1, and pl. v, fig. 1) closely resembles that of Paradoxurus hermaphroditus (pl. tv, fig. 2, and pl. v, fig. 2), but the interpterygoid space is narrower and there is scarcely a trace of the longitudinal groove involving the region between upper rims of maxillaries. Teeth essentially as in P. hermaphroditus, but inner lobe of upper sectorial longer, and set at a somewhat different angle, so that the width of the palate is con- siderably reduced. Measurements.—External measurements of type: total length, 845; head and body, 485; tail vertebrae, 360; hind foot, 77 (75); ear from meatus, 33; ear from head, 19.6; width of ear, 25. Cranial measurements of type: greatest length, 104 (112) ;* basal length, 99 (107) ; basilar length, 94.6 (102) ; median palatal length, 47 (50); breadth of palate between sectorials, 14 (18) ; breadth of palate between front molars, 19 (21) ; length of interpterygoid space, 20 (24) ; breadth of interpterygoid space at middle, 8.6 (10.2) ; con- striction in front of postorbital processes, 17 (19) }; constriction behind postorbital processes, 10.4 (11); zygomatic breadth, 62 (58) ; mastoid breadth, 38 (38) ; breadth of braincase above roots of zygomata, 32 (33); mandible, 78 (81); maxillary toothrow, ex- clusive of incisors (alveoli), 39 (41); greatest diameter of upper sectorial, 10.2 (10.2); mandibular toothrow exclusive of incisors (alveoli), 44 (41). Specimens examined.—Two, the type, and the skull of an adult female, the latter from South Pagi Island. Remarks.—This well characterized species is probably most closely allied to Paradoxurus hermaphroditus, though in color it undoubtedly more nearly resembles P. aureus, an animal which I have not seen. It is apparently of about the same size as the Ceylonese musang, 1 Measurements in parenthesis are those of an adult male Paradoxurus kermaphroditus from Trong, Lower Siam (No. 86,793). 46 SMITHSONIAN MISCELLANEOUS COLLECTIONS [VoL. 45 and the teeth of the two animals, to judge by Blanford’s description of P. aureus, have certain peculiarities in common. Dr. Abbott writes that the female exactly resembled the male in color. GALEOPITHECUS PUMILUS sp. nov. (PLateE VI, FIGURE 3) 1900. Galeopithecus volans MiLiEr, Proc. Biol. Soc. Washington, x11, p. 193, December 21, 1900. Not Lemur volans Linneus. Type.—Adult male (skin and skull), No. 104,448, United States National Museum. Collected on Pulo Adang, Butang Islands, December 17, 1899, by Dr. W. L. Abbott. Original number, 165. Characters.—Like Galeopithecus volans from the Malay Peninsula, but much smaller. Skull with relatively shorter rostrum and broader palate than in the mainland animal. Color—Type: entire dorsal surface an intimate blending of smoke-gray, blackish brown, and isabella-color, the gray clear and pale on back of neck and sides of head, tinged with isabella-color on posterior half of back, the brown predominating on legs, feet, and greater part of membranes, the isabella-color most noticeable on crown, along edge of principal membrane, and on distal half of uropatagium. Except for the nearly clear gray area on neck and cheeks the whole upper surface of the body is so mottled and clouded that the exact color is very difficult to describe. A few small flecks of nearly pure white may be detected on posterior edge of thigh. Underparts isabella-color, rather darker than that of Ridg- way on throat, chest, and belly, fading toward ochraceous-buff on under side of membranes and limbs. | Skull and teeth—As compared with the skull of an adult male Galeopithecus volans (pl. vu, fig. 2; pl. vit, fig. 2, and pl. rx, fig. 2) from the Rumpin River, Pahang, that of Galeopithecus pumilus (pl. v1, fig. 3) is conspicuously smaller, the rostrum is relatively shorter and the palate broader. In the larger animal the least distance from orbit to anterior nares equals that between outer edges of anterior molars, while in Galeopithecus pumilus it is equal only to that from outer edge of one molar to inner edge of the opposite tooth. The supraorbital processes are less developed in the smaller animal, in which the greatest width of the process is scarcely more than one-half that of the narrowest flat portion of the interorbital region, instead of nearly equal to the entire width of this area as in the larger species. Braincase shorter than that of Galeopithecus volans, and occipital portion less distinctly outlined when skull is viewed from the side. ‘ ‘ ‘ “URIS I9MO'T “suo ah “ON “9/[8Ul 107021US, SHANXOPVANY “1 SNJ1pOLYgv Uday SNANXOpvAvYT “% (adj) apew SNOILOETIO’) §S ANVITAOSIPY NVI MILLER] SEVENTY NEW MALAYAN MAMMALS 47 The teeth are essentially like those of Galeopithecus volans, due allowance being made for their smaller size. Measurements——External measurements of type: total length, 490 (585) ;* head and body, 305 (365); tail vertebra, 185 (220) ; hind foot, 55 (69) ; hind foot without claws, 49 (62) ; front foot, 62 (77) ; front foot without claws, 55 (71) ; ear from meatus, 16 (16) ; ear from crown, I3 (13); width of ear, 13 (13.4). Cranial measurements of type: greatest length, 61 (70) ; basal length, 58 (65); basilar length, 54 (61); lateral palatal length,’ 27 (32); palatal width at front of first incisor, 11.6 (13.4) ; palatal width at space between canine and first premolar, 19 (18) ; distance between inner edge of posterior molars (alveoli), 14, (15.6) ; least distance from orbit to anterior nares, 19 (23.4) ; zygomatic breadth, 38.6 (39.6); greatest orbital breadth, 40 (45); least interorbital breadth, 16 (20) ; breadth of braincase above roots of zygomata, 24 (25); mastoid breadth, 29 (30); greatest depth of braincase, 17 (20); occipital depth, 13.8 (15); mandible, 43 (50); depth of mandible between canine and first premolar, 5.8 (5.6); depth of mandible through coronoid process, 19.6 (21) ; maxillary toothrow (alveoli), 29 (34) ; mandibular toothrow (alveoli), 30.2 (35). Specimens examined.—Two, both from Pulo Adang. Remarks.—This small Galeopithecus is so readily distinguishable from the peninsular animal, which for the present may .be assumed to represent true volans, that no further comparison is necessary. From Galeopithecus temminckii as figured by Waterhouse it differs in its smaller size, less wide palate, and much smaller teeth. GALEOPITHECUS AORIS sp. nov. Type.—Adult female (skin and skull), No. 112,428, United States National Museum. Collected on Pulo Aor, off coast of Johore, June 8, 1901, by Dr. W. L. Abbott. Original number, 1028. Characters.—Similar to Galeopithecus pumilus but not as small, palate relatively not as broad, and upper canine, lower canine, and first premolar much larger. Color.—The type is similar to the type of Galeopithecus pumilus except that the wash of isabella-color is less distinct on feet and outer edge of membranes, and the entire back is tinged with whitish cream-buff. A pale buffy spot on each shoulder and hip. A 1 Measurements in parenthesis are those of an adult male Galeopithecus volans from the Rumpin River, Pahang (No. 115,493). 2 From tip of premaxillary to front of interpterygoid space. 48 SMITHSONIAN MISCELLANEOUS COLLECTIONS [VoL. 45 young male (No. 112,427, June 5, 1901) almost exactly resembles the type of Galeopithecus pumilus. Skull and teeth_—The skull of the female is about as large as that of the male of Galeopithecus volans from the mainland. In form it more closely resembles the skull of the peninsular animal than it does the short broad skull of Galeopithecus pumilus. Palate and rostrum rather wider than in Galeopithecus volans, but not as wide as in G. pumilus. Teeth essentially as in the related species, but canines, both above and below, and first lower premolar relatively larger, particularly in the male. Measurements.—External measurements of type: total length, 625 (690) ;+ head and body, 385 (405) ; tail vertebrae, 240 (285) ; hind foot, 60 (73); hind foot without claws, 54 (65); front foot, 70 (83); front foot without claws, 63 (75); ear from meatus, 16 (22) ; ear from crown, 13 (18) ; width of ear, 12 (17). Cranial measurements of type: greatest length, 67 (77.6) ; basal length, 63 (72.4) ; basilar length, 58 (68) ; lateral palatal length, 30.4 (37) ; palatal width at front of first incisor, 12.4 (15) ; palatal width at space between canine and first premolar, 20.4 (23); distance between inner edge of posterior molars (alveoli), 15 (17); least distance from orbit to anterior nares, 23 (28) ; zygomatic breadth, Ao (47.6) ; depth of zygoma, 4 (6) ; greatest orbital length, 42 (50) ; least interorbital breadth, 17.4 (22.6); breadth of braincase above roots of zygomata, 25 (26); mastoid breadth, 30 (36.4); greatest depth of braincase, 17.8 (19) ; occipital depth, 15 (18) ; mandible, 50 (57); depth of mandible between canine and first premolar, 6.6 (7); depth of mandible through coronoid process, 20.4 (25.6) ; maxillary toothrow (alveoli), 31.6 (37); mandibular toothrow (alveoli), 33.4 (39). Specimens examined.—Two, both from Pulo Aor. Remarks.—The type is the lightest colored Galeopithecus that I have seen, but whether this character is anything more than an in- dividual peculiarity cannot now be determined. The male, so young that the permanent dentition is not fully in place, very closely re- sembles the type of Galeopithecus pumilus, but is not as small. Its measurements are: total length, 530; head and body, 340; tail verte- bree, 190; hind foot, 56 (50.4) ; front foot, 61 (56) ; skull, greatest length, 60; zygomatic breadth, 37. 1 Measurements in parenthesis are those of an adult female Galeopithecus volans from Trong, Lower Siam (No. 84,420). S+(adAj)ayeu ‘sxzr20ng snzayzigoazvy “€ “seunieN yINOG ‘puvjsy usasseyNs ‘coo*For ‘on ‘ayeut “s7779M4.5 SnIayzigoazvy *% *(adA}) ajewMey ‘s7729v4.5 SHIaYI190a]0H) “1 I MILLER | SEVENTY NEW MALAYAN MAMMALS 49 GALEOPITHECUS GRACILIS sp. nov. (PLATE VI, FIGURE 2) 1901. Galeopithecus volans Miter, Proc. Washington Acad. Sci., March 26, 1901. Part, specimens from Sirhassen Island. Not Lemur volans Linneeus. Type.—Adult female (skin and skull), No. 104,601, United States National Museum. Collected on Sirhassen Island, South Natunas, June 7, 1900, by Dr. W. L. Abbott. Original number, 461. Characters.—Similar to Galeopithecus pumilus and G. aoris but not as small. Skull more elongate than in the related species, and with much narrower palate. Color.—The color so closely resembles that of the related species as to require no detailed description. Both specimens are in the gray pelage. Skull and teeth.—The skull (pl. v1, fig. 2) is slightly longer than that of Galeopithecus aoris, but the width is distinctly less. This is particularly noticeable in the rostral and lachrymal regions. On the other hand, the zygoma is shorter and broader (deeper) than in the related species. Palate narrower than in any other known member of the genus. The teeth are rather larger than those of Galeopithecus aoris, but I can detect no tangible peculiarities in form. | Measurements.—External measurements of type: total length, 605; head and body, 405; tail vertebrae, 200; hind foot, 62 (59) ; tront foot, 73 (67) ; ear from meatus, 18; ear from crown, 14; width of ear, 14. External measurements of adult male from the type locality (No. 104,600) : total length, 410; head and body, 300; tail vertebra, 110; hind foot, 62.6 (56); front foot, 67 (60) ; ear from meatus, 14; ear from crown, 12.6; width of ear, II. Cranial measurements of type: greatest length, 68; basal length, 65; basilar length, 61; lateral palatal length, 32.4; palatal width at front of first incisor, 13; palatal width at space between canine and first premolar, 20; distance between inner edge of posterior molars (alveoli), 14.6; least distance from orbit to anterior nares, 24; zygo- _ matic breadth, 38.4; depth of zygoma, 5; greatest orbital breadth, 40.2; least interorbital breadth, 17; breadth of braincase above roots of zygomata, 21.6; mastoid breadth, 31; greatest depth of braincase, 17; occipital depth, 13.4; mandible, 49.6; depth of mandible between canine and first premolar, 6; depth of mandible through coronoid process, 20.4; maxillary toothrow (alveoli), 33; mandibular tooth- row (alveoli), 34. Cranial measurements of adult male from the type locality (No. 104,600) : greatest length, 64; basal length, 61; basilar length, 57; Se) SMITHSONIAN MISCELLANEOUS COLLECTIONS [VoL. 45 lateral palatal length, 29.6; palatal width at space between canine and first premolar, 19.6; zygomatic breadth, 37; depth of zygoma, 5; mandible, 44.8; maxillary toothrow (alveoli), 30.6; mandibular toothrow (alveoli), 31. Specimens examined.—Two, both from Sirhassen Island. GALEOPITHECUS NATUNZ: sp. nov. 190l. Galeopithecus volans Miter, Proc. Washington Acad. Sci., mt, p. 134, March 26, 1901. Part, specimen from Bunguran Island. Not Lemur volans Linneus. Type.—Adult female (skin and skull), No. 104,602, United States National Museum. Collected on Bunguran Island, North Natunas, July 16, 1900, by Dr. W. L. Abbott. Original number, 573. Characters——Intermediate in size between the large and small members of the genus. Skull and teeth essentially as in Galeo- pithecus volans but distinctly smaller. Color as in the related species. Color.—The color shows no tangible differences from that of the species just described. The specimen is in the gray phase. Skull and teeth—The skull and teeth are very similar to those of Galeopithecus volans from the Malay Peninsula, but are readily dis- tinguishable by their smaller size. In the type the lachrymal region is considerably swollen so that the rostrum is less sharply marked off from the orbit than in Galeopithecus volans, a character which may prove to be individual. Palate slightly narrower than that of Galeopithecus volans, but not closely approaching the form charac- teristic of the Sirhassen animal. The teeth show no tangible pecu- liarities. Measurements.—External measurements of type: total length, 660 ; head and body, 405 ; tail vertebra, 255 ; hind foot, 70 (64) ; front foot, 83 (77); ear from meatus, 17; ear from crown, 16; width of eat, 12.6, Cranial measurements of type: greatest length, 71; basal length, 67.6; basilar length, 63; lateral palatal length, 33; palatal width at front of first incisor, 14; palatal width at space between canine and first premolar, 22; distance between inner edge of posterior molars (alveoli), 14.8; least distance from orbit to anterior nares, 26.4; zygomatic breadth, 41.8; depth of zygoma, 5.4; greatest orbital breadth, 44.4; least interorbital breadth, 17.6; breadth of braincase above roots of zygomata, 25.4; mastoid breadth, 33.6; greatest depth of braincase, 19; occipital depth, 15.4; mandible, 51; depth of mandible between canine and first premolar, 6.8; depth of mandible through coronoid process, 22.6 ; maxillary toothrow (alveoli), 34.6; mandibular toothrow (alveoli), 36. SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 45, PL. VII 1. Galeopithecus volans, female, No. 84,420, Trong, Lower Siam. 2. Galeopithecus ve No. 115,403, Rumpin River, Pahang. 3. Galeopithecus saturatus, female (type). 4. Galeopithecus s yatus, male, No. 121,747, Tana Bala, Batu Islands. ae wr - a ~ ; ‘ ; . “ 3s . Ry - i iF af : : ee . i; ‘ i MILLER | SEVENTY NEW MALAYAN MAMMALS 51 Specimens examined.—One, the type. Remarks.—Although represented by one specimen only, this species is undoubtedly distinct from the Galeopithecus volans of the Malay Peninsula and from the form inhabiting Sirhassen Island. It appears to be more closely allied to the former, a fact in har- mony with the relationship of other Bunguran mammals, notably the giant squirrels. GALEOPITHECUS SATURATUS sp. nov. (Pirate VII, riguRES 3 AND 4; Pate VIII, Ficures 3 AND 4; Puate IX, FIGURES 3 AND 4) Type.—Adult female (skin and skull), No. 121,750, United States National Museum. Collected on Tana Bala, Batu Islands, Sumatra, February 12, 1903, by Dr. W. L. Abbott. Original number, 2278. Characters.—Slightly larger than Galeopithecus volans, and sexual difference in size less noticeable than in the other known species. First upper incisor usually with four cusps. Color very dark, the light phase not unlike the dark phase of Galeopithecus volans. Color.—Light phase (type specimen): dorsal surface mostly ochraceous-buff and seal-brown, the former predominating on body, hind legs, and interfemoral membrane, the latter nearly clear on forearms. Neck washed with gray, shoulders and back with cream- buff. Face suffused with dark brown. The colors are everywhere indescribably blended. On each hip there is a conspicuous buffy white spot, and two similar but smaller spots occur behind each shoulder. The feet and edges of the principal membranes are speckled with grayish white dots. Dark area of forearm intensified by about six conspicuous buffy white spots 3-8 mm. in length and about 2 mm. wide. Under surface of body drab washed with wood- brown. Under surface of legs and membranes tawny clay-color con- siderably lighter than that of Ridgway. Dark phase (adult male, No. 121,749) ; general color above seal-brown, faintly grizzled with russet on interfemoral membrane, lumbar region, and sides of back. On front legs and feet and over most of the membranes there is a noticeable sprinkling of whitish hairs, these most conspicuous on forearm. Neck dull, grizzled wood-brown. Head blackish, sprinkled with silvery white hairs. Hip spots indicated by a few tufts of creamy white. White spots on shoulders and forearms obsolete. Underparts as in the light phase, but wood-brown and drab rather darker, and interfemoral membrane heavily washed with an indefi- nite dark brown. Red phase (adult male, No. 121,853): entire upperparts bright cinnamon-rufous, the hairs fading to orange-buff beneath the surface of the fur. Neck slightly grizzled with wood- 52 SMITHSONIAN MISCELLANEOUS COLLECTIONS [voL. 45 brown. A slight burnt-sienna wash on shoulders. Head, face, and chin burnt-sienna. Underparts essentially as in the other phases, but everywhere tinged with rufous. Skull and teeth.—In general form the skull (pl. vu, figs. 3 and 4; pl. vit, figs. 3 and 4; pl. rx, figs. 3 and 4) closely resembles that of Galeopithecus volans (pl. vu, figs. I and 2; pl. vim, figs. 1 and 2; pl. 1x, figs. 1 and 2). In fact, when skulls of females are com- pared the only tangible difference appears to be in the form of the basioccipital and foramen magnum. In Galeopithecus saturatus this bone is narrower than in the mainland animal, and the occipital condyles are set at a less distinct angle. Consequently the foramen magnum is more nearly of the same width above and below, and the whole bony structure surrounding it stands out as more noticeably tubular. This character is equally noticeable in the skull of the male, which is further distinguished by its considerably greater size (see measurements of Galeopithecus volans under G. pumilus, p. 47). The teeth resemble those of Galeopithecus volans, but are more robust, and the first upper incisor is conspicuously larger. In four of the five skulls this tooth has four well developed cusps, while three is the usual number in the other species. Measurements.—External measurements of type: total length, 725; head and body, 459; tail vertebrze, 266; hind foot, 76 (68) ; front foot, 85 (77); ear irom’ meatus, 20; ear from, crown, 16; width of ear, 14. External measurements of adult male from the type locality (No. 121,747) : total length, 653; head and body, 403; tail vertebrze, 250; hind foot, 70 (63) ; front foot, 83 (74). Cranial measurements of type: greatest length, 78; basal length, 74; basilar length, 69; lateral palatal length, 36.8; palatal width at front of first incisor, 14.4; palatal width at space between canine and first premolar, 23.6; distance between inner edge of posterior molars (alveoli), 16.8; least distance from orbit to anterior nares, 28; zygomatic breadth, 45; greatest orbital breadth, 51; least inter- orbital breadth, 22; breadth of braincase above roots of zygomata, 27; mastoid breadth, 34.6; greatest depth of braincase, 20; occipital depth, 16.6; mandible, 54; depth of mandible between canine and first premolar, 6.6; depth of mandible through coronoid process, 23; maxillary toothrow (alveoli), 38; mandibular toothrow (alveoli), 37. Cranial measurements of adult male (No. 121,747: greatest length, 73; basal length, 70; basilar length, 65.6; lateral palatal length, 34; palatal width at space between canine and first premolar, 23; zygomatic breadth, 46; mandible, 55; maxillary toothrow (alveoli), 37; mandibular toothrow (alveoli), 37. {ITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 45, PL, VI 1. Galeopithecus volans, female, 3 2. Galeopithecus volans, male, Galeopithecus satu- No. 115,493.,Rumpin River, Pahang. 3. Gade pithecus saturatus, female (type). 4. ratus, male, No. 121,747, Vana Bala, Batu Islands. MILLER] SEVENTY NEW MALAYAN MAMMALS 53 Specimens examined.—Six, all but one from Tana Bala, the latter from Pulo Pinie. Remarks.—Of the six skins two, both males, are in the dark pelage, one male is in the red phase, and the one female, the type, is in the gray coat. The two others, both males, are in a stage intermediate between the red and dark phases. Due allowance being made for the three phases, the species in not unusually variable in color. Specimens in the dark or red _ pelages are very different from Galeopithecus volans, but those in the gray phase might readily be mistaken for the mainland animal in dark coat. But whatever the characters of the skin, the species is always recognizable by the relatively slight difference in the size of the two sexes, and by the large anterior upper tooth. GALEOPITHECUS TUANCUS sp. nov. 1901. Galeopithecus volans Miter, Proc. U. S. Nat. Mus., xxv, p. 471, February 3, 1903. Not Lemur volans Linneus. Type.—Adult female (skin and skull), No. 114,375, Unites States National Museum. Collected on Pulo Tuangku, Banjak Islands, Sumatra, January 22, 1902, by Dr. W. L. Abbott. Original number, 1454. Characters.—Color essentially as in Galeopithecus saturatus, but gray phase apparently lighter. Size much less than that of the Batu animal. Color.—Light phase (type specimen): The animal was evidently moulting when killed, as two pelages are represented; a thin, abraded, grayer coat on limbs, head, neck, and sides, and a fresh, more brown coat on back. The former closely resembles the ordinary gray phase of Galeopithecus volans, and calls for no special comment. The latter is essentially like the back in the type speci- men of Galeopithecus saturatus, but the colors are not as bright and well contrasted. Underparts as in the type of G. saturatus. Dark phase (immature’ male, No. 114,376): like corresponding pelage of Galeopithecus saturatus, except that the back is more suffused with russet, and the neck is distinctly tinged with gray. Skull and teeth—The skull is much smaller than that of Galeo- pithecus saturatus, that of the female closely corresponding with that of the female G. aoris in size. The skull of the male, how- ever, is decidedly larger than in the male G. aoris of exactly com- parable age. Teeth as in Galeopithecus aoris, except that the first 1 Apparently full grown, but permanent dentition not fully in place. 54 SMITHSONIAN MISCELLANEOUS COLLECTIONS [VoL. 45 upper incisor is larger and in the permanent dentition shows a dis- tinct trace of a*fourth cusp. Measurements——External measurements of type: total length, 640; head and body, 385; tail vertebre, 235; hind foot, 65 (59) ; front foot, 78 (71) ; ear from meatus, 18; ear from crown, 16; width of ear, 13. External measurements of immature male (No. 114,376) : total length, 550; head and body, 335; tail vertebra, 215; hind foot, 63 (57) ; front foot, 68 (61). Cranial measurements of type: greatest length, 69; basal length, 65.4; basilar length, 61; lateral palatal length, 32; palatal width at front of first incisor, 13; palatal width at space between canine and first premolar, 20.4; distance between inner edge of posterior molars (alveoli), 16; least distance from orbit to anterior nares, 24.2 ; zygo- matic breadth, 41; greatest orbital breadth, 45; least interorbital breadth, 19.4; breadth of braincase above roots of zygomata, 24; mastoid breadth, 28.6; greatest depth of braincase, 19; occipital depth, 15; mandible, 49; depth of mandible between canine and first premolar, 6.6; depth of mandible through coronoid process, 21; maxillary toothrow (alveoli), 33; mandibular toothrow (alveoli), 221Gy Cranial measurements of immature male (No. 114,376) from the type locality: greatest length, 66; basal length, 63; basilar length, 58; lateral palatal length, 30.6; palatal width at space between canine and first premolar, 19.6; zygomatic breadth, 39; mandible, 48.4; maxillary toothrow (alveoli), 30.6; mandibular toothrow (alveoli), 32.4. Specimens examined.—Two, both from Pulo Tuangku. TUPAIA CASTANEA sp. nov. Type.—Adult female (skin and skull), No. 115,608, United States National Museum. Collected on Pulo Bintang, Rhio Archipelago, August 11, 1902, by Dr. W. L. Abbott. Original number, 1872. Characters.—Similar to Tupaia splendidula but larger and darker ; underparts clear tawny-ochraceous. Color—Upperparts a fine, close mixture of black and ferruginous, the general effect slightly more red than the chestnut of Ridgway, though in certain lights appearing almost black. The hairs are everywhere very glossy. On occiput and forehead the chestnut fades rather abruptly into the light grizzled brown of face and cheeks. The pale element of this grizzle is ochraceous-buff, the dark element an indefinite blackish brown. The former is a little in excess. Outer surface of hind legs like back, outer surface of a i 1 r.* ' ~ ’ = t , i —~ i « 4 ‘ a 4 i x? A . ’ 2 ! ; 1 ‘ ke > fy ni £ ' e o = t I oi P ; a] i v. 2 ib a eee : ‘ Asa ; ; ® f ; P etal oe: . a i ‘ ri p ot a . = - / i re) S | 4 1 - of i 4 1 spurs] neg ‘ereq euvy, ‘Lel‘rzr ‘on ‘ayeur ‘s77.c2iJvs snzayzigoayvy “v uvyed ‘ary uduny “Er6‘Srr ‘on ‘apeur ‘su v702 snIayj190a7VH *% “WEIS JaMo'T ‘suo! sb ‘hg 02 snzay4igoayv+) MILLER] SEVENTY NEW MALAYAN MAMMALS 55 front legs like face. Underparts, shoulder stripes, and inner surface of legs, clear tawny-ochraceous, the hairs on the chest and throat slaty at base. Feet and ears dusky, the former distinctly darker than legs. Tail light ferruginous, heavily clouded above with blackish brown at base and near tip. The second specimen shows no marked variation from the type. Skull and teeth—rThe skull and teeth closely resemble those of Tupaia splendidula, but are distinctly larger throughout. Measurements——External measurements of type: total length, 345; head and body, 200; tail vertebre, 145; hind foot, 44 (42). External measurements of adult male from type locality: total length, 360; head and body, 210; tail vertebre, 150; hind foot, 46 (44). Cranial measurements of type: greatest length, 53 (50) ;' basal length, 49 (—); basilar length, 46 (—) ; palatal length, 28 (26) ; diastema, 3.4 (3.4); breadth of palate at middle of diastema, 5.6 (5.6) ; breadth of palate between middle molars, 10 (9) ; least inter- orbital breadth, 15.6 (14.8) ; zygomatic breadth, 25.4 (25) ; breadth of braincase above roots of zygomata, 20 (19) ; maxillary toothrow exclusive of incisors (alveoli), 19 (19); mandible, 35 (34) ; man- dibular toothrow exclusive of incisors (alveoli), 19.4 (19). Specimens examined.—Two, both from Bintang Island. Remarks.—The discovery of this treeshrew is of unusual inter- est, as it greatly extends the range of the group to which the species belongs. Hitherto three members have been known, Twupaia splendidula Gray from Bunguran Island, T. lucida (Thomas) from Pulo Laut, North Natunas, and T. chrysomalla Miller from the Anambas. The relationship of Tupaia castanea to these may be understood from the following key: Upperparts light (approximately the ferruginous of Ridgway). Head, neck, and shoulders paler than back; feet not darker than legs. Tupaia lucida. Head, neck, and shoulders not paler than back; feet darker than legs. Tupaia chrysomalla. Upperparts dark (approximately the chestnut of Ridgway). Underparts buff; total length about 320; hind foot about 41 (39). Tupaia splendidula. Underparts tawny-ochraceous; total length about 350; hind foot about Fe (13) ery ee EP VSIEES gc s s,s a, 0 0) af. cis endl oeetaamratbjoa, Sarelevere & Tupaia castanea. The distribution of these treeshrews is difficult to understand. They are unknown on the mainland of the Malay Peninsula or on any of the larger islands; but Pulo Bintang is distant from the 1 Measurements in parenthesis are those of an adult female Tupaia splendidula from Bunguran Island, North Natunas (No. 104,714). 56 SMITHSONIAN MISCELLANEOUS COLLECTIONS [VoL. 45 coast of Johore by only 11 miles, while nearly 300 miles of open water lie between it and the Anambas, and a similar expanse sepa- rates the latter group from the Natunas. This apparent anomaly is doubtless in large part due to our very fragmentary knowledge of the fauna of the larger Malayan land masses. TUPAIA PULONIS sp. nov. Type.—Adult female (skin and skull), No. 112,449, United States National Museum. Collected on Pulo Aor, off coast of Johore, June 7, 1901, by Dr. W. L. Abbott. Original number, 1023. Characters.—Similar to Tupaia ferruginea Raffles, but size larger, color of underparts much paler, and skull with rostral portion less elongate. Color—Back and_ sides ochraceous-rufous, inconspicuously speckled by the black hair tips. On sides the ground color becomes somewhat lighter, and on neck, shoulders, and outer surface of front legs it fades to ochraceous-buff, the speckling at the same time be- coming more evident. Head and feet an indefinite grizzled brown, the exact shade not far from the hair-brown of Ridgway. Shoulder stripes distinct, cream-buff. Underparts and inner surface of legs cream-buff in strong contrast with color of sides. Tail somewhat lighter than back, its upper surface a grizzle of black and dull cream-buff, its under surface clear buff at middle, grizzled buff and black along edges and at tip. Skull and teeth—The skull is larger than that of Tupaia ferruginea throughout, and in form it is relatively broader and more robust, particularly in the rostral portion. Teeth larger than thqse of Tupaia ferruginea, but of similar form. ‘The larger size of the teeth is especially noticeable in the anterior upper molar, the posterior lower premolar, and the first and second lower molars. Measurements.—External measurements of type: total length, 372; head and body, 197; tail vertebra, 175; hind foot, 42 (40). External measurements of immature? male from the type locality: total length, 385; head and body, 195; tail vertebra, 190; hind foot, 44 (42). Cranial measurements of type: greatest length, 51.4 (49) ;? basal length, 45.8 (43); basilar length, 44.6 (41.6); palatal length, 27 (25.4) ; diastema, 3 (3.6) ; breadth of palate at middle of diastema, 6.8 (5.4); breadth of palate between middle molars, 9.8 (8) ; least 1 Permanent dentition not wholly in place. 2 Measurements in parenthesis are those of an adult female Tupaia ferru- ginea from Tanjong Dungun, Tringanu (No. 105,029). MILLER] SEVENTY NEW MALAYAN MAMMALS 57 interorbital breadth, 15.6 (13.8); zygomatic breadth, 26.4 (24) ; breadth of braincase above roots of zygomata, 19.6 (19) ; maxillary toothrow exclusive of incisors (alveoli), 19 (18); mandible, 35.4 (33.6) ; mandibular toothrow exclusive of incisors (alveoli), 19.4 (18) ; crowns of first and second lower molars together, 7 (6.2). Specimens examined.—Two, both from Pulo Aor. Remarks.—lts large size and the pale color of the underparts readily distinguish this species from Tupaia ferruginea. ‘The color of the back, however, is essentially as in the animal inhabiting the neighboring portions of the Malay Peninsula, though the upper sur- face of the tail is not as dark nor as strongly suffused with yellow. TUPAIA TEPHRURA sp. nov. Type.—Adult female (skin and skull), No. 121,752, United States National Museum. Collected on Tana Bala, Batu Islands, Sumatra, February 12, 1903, by Dr. W. L. Abbott. Original number, 2276. Characters.—In size and general appearance not unlike Tupaia pheura of Sinkep Island, but upperparts darker, shoulder stripes better developed, and tail conspicuously lighter than body, its distal half silvery buff-gray. Color.—Type: upperparts mixed ferruginous and black, the for- mer in excess on neck, shoulders, and anterior half of back, the latter in excess posteriorly. On crown the ferruginous is replaced by buff. Shoulder stripes buff, well developed, about 4 mm. wide at middle. Feet and ears blackish. Underparts buff somewhat darker than that of Ridgway, particularly on middle of chest; the anterior half of belly darkened by the brownish under color. Tail light buff, the proximal half heavily overlaid with blackish brown above, the distal half grizzled with blackish and silvery, the under side clear buff throughout, except brownish line along vertebrae and brownish edging. Skull and teeth.—The skull and teeth do not differ appreciably from those of Tupaia pheura. Measurements.—External measurements of type: total length, 323; head and body, 193; tail vertebree, 130; hind foot, 41 (38.6). External measurements of adult male (No. 121,751) from the type locality: total length, 330; head and body, 180; tail vertebre, 150; hind foot, 45 (42.6). Cranial measurements of type: greatest length, 52; basal length, 46.4; basilar length, 45; palatal length, 28; diastema, 4; breadth of palate between middle molars, 9.2; least interorbital breadth, 15; zygomatic breadth, 25 ; breadth of braincase above roots of zygomata, 58 SMITHSONIAN MISCELLANEOUS COLLECTIONS [VOL. 45: 20; maxillary toothrow exclusive of incisors (alveoli), 19; mandible, 35; mandibular toothrow exclusive of incisors (alveoli), 19.4. Specimens examined.—Two, both from Tana Bala. Remarks.—The color of the underparts is exactly as in Tupaia pheura, that of the upperparts is slightly darker, and the tail is very different from that of the related species. When the hairs are in natural position the basal half of the tail does not differ appreciably in color from the back, except that the silvery tips of the hairs are more noticeable. The least disarrangement, however, brings to view the buff under color. The distal half is very different from the back, and the whole under side is very much lighter than in Tupaia pheura. Both specimens are in fresh, unworn pelage. In the male the shoulders and back are not quite as dark as in the female, and the buff of the tail is much paler. Otherwise the two specimens are essentially alike. TUPAIA CHRYSOGASTER sp. nov. (PLATE X, FIGURE I) Type.—Adult female (skin and skull), No. 121,572, United States. National Museum. Collected on North Pagi Island, Sumatra, No- vember 21, 1902, by Dr. W. L. Abbott. Original number, 2078. Characters.—Like the Javan Tupaia hypochrysa (Thomas), but. upperparts darker than in Tupaia ferruginea, tail strongly tinged with yellowish brown below, and entire ventral surface of body and. inner side of legs brownish yellow. Color.—Type: upperparts and outer side of legs a fine, incon- spicuous grizzle of black and dull ferruginous, the former in excess. everywhere except on neck and along sides of body. Face slightly tinged with gray. Feet and ears blackish, but not strongly con- trasted with surrounding parts. Ventral surface of body and inner side of legs clear brownish yellow, the exact shade ochraceous on throat, chest, legs, and along median line, somewhat tinged with clay- color elsewhere. The two colors form a rather sharp line of de- markation on legs, but on sides the fur is thin and the shades less defi- nite. Wrists and ankles tinged with blackish. Tail like back above, but more coarsely grizzled, its under side a peculiar indefinite grizzled yellowish brown, very different from the gray of Tupaia ferruginea. Skull and teeth—The skull and teeth (pl. x, fig. 1) are larger than in Tupaia ferruginea (pl. x, fig. 2) and with age the rostrum becomes decidedly more elongate than is ever the case with the mainland animal. In the latter character the skull is intermediate between that of Tupaia ferruginea and that of T. tana. Its other ‘nuesuiry ‘unsung Suofuey ‘E€6'Sor ‘on ‘vau2sut1af vivgny, *% ‘(ad Aq) apeuay ‘cazsv.soshiys ving: VT MILLER] SEVENTY NEW MALAYAN MAMMALS 59 details of form are, however, in essential agreement with the skull of Tupaia ferruginea. Measurements——External measurements of type: total length, 345; head and body, 205; tail vertebre, 140; hind foot, 44.4 (42). External measurements of an adult male (No. 197,138) from the type locality: total length, 335; head and body, 197; tail vertebre, 138; hind foot, 44 (42). Cranial measurements of type: greatest length, 56 (49.4),* basal length, 49 (44) ; basilar length, 48 (42) ; palatal length, 30.6 (26.4) ; diastema, 5 (4.2) ; least distance from rim of orbit to tip of premaxil- lary, 26.6 (22.6) ; breadth of palate between middle molars, 9 (8.6) ; least interorbital breadth, 15 (14.8) ; zygomatic breadth, 26.6 (25) ; breadth of braincase above roots of zygomata, 20 (19) ; mandible, 38 (33); maxillary toothrow exclusive of incisors, 20 (18) ; man- dibular toothrow exclusive of incisors, 20 (18.6). Specimens examined.—Ten, all from the Pagi Islands. Remarks.—This series shows no important variation in color. In a few specimens there is a faintly indicated shoulder stripe, but in the majority, as in the type, there is no trace of this. The under- parts are occasionally less bright than in the type, and the indefinite brown wash on sides of belly varies slightly in amount. Occasionally there is a slight grayish tinge at axilla. This species needs comparison with only one other, the Tupaia hypochrysa of Thomas. In the Javan animal, however, the yellow of the underparts is “a rich golden colour, almost ‘ orpiment orange’ of Ridgway,” and is confined to the throat and chest. The rest of the coloration is said to be similar to that of Tupaia ferruginea. This would make the upperparts lighter than in T. chrysogaster, and the under side of the tail gray instead of yellowish brown. TUPAIA CERVICALIS sp. nov. Type.—Adult male (skin and skull), No. 121,754, United States National Museum. Collected on Tana Bala, Batu Islands, Sumatra, February 14, 1903, by Dr. W. L. Abbott. Original number, 2294. Characters.—Like Tupaia tana, but gray neck markings paler, more extensive, and more conspicuous; black of back more sharply defined from red of sides, and teeth larger. Color—Type: the general color of underparts, tail, and back is essentially as in Tupaia tana, and requires no detailed description. 1 Measurements in parenthesis are those of an adult female Tupaia fer- ruginea from Singapore (No. 105,079). 60 SMITHSONIAN MISCELLANEOUS COLLECTIONS [VoL. 45 The sides, however, are a lighter, duller ochraceous-rufous, and the posterior half of back a deeper, more glossy black. The black area is thus better outlined laterally, while the very sharp contrast with the shoulders and neck throw it still more strongly in relief. Light neck stripes grizzled, whitish cream-buff, slightly tinged with tawny on posterior third, very sharply defined against the tawny sides of neck and blackish median stripe. Crown and face a dull grizzle of whitish and dark brown, nearly as pale as light neck stripes. Skull and teeth.—The skull is in all respects similar to that of Tupaa tana. Teeth as in the related species, but molars larger. Measurements.—External measurements of type: total length, 375; head and body, 210; tail vertebre, 165; hind foot, 46 (44). External measurements of adult female from the type locality: total length, 365; head and body, 215; tail vertebrae, 150; hind foot, 45.4 (42). Cranial measurements of type: greatest length, 59; basal length, 52; basilar length, 51; palatal length, 33; diastema, 5; breadth of palate between middle molars, 8.6; least interorbital breadth, 14.8; zygomatic breadth, 25 ; breadth of braincase above roots of zygomata, 20; maxillary toothrow exclusive of incisors (alveoli), 22; mandible, 39; mandibular toothrow exclusive of incisors (alveoli), 22. Specimens examined.—Two, both from Tana Bala. PTEROPUS GEMINORUM pp. nov. Type.—Adult female (skin and skull), No. 104,464, United States - National Museum. Collected on South Twin Island, Mergui Archi- pelago, January 28, 1900, by Dr. W. L. Abbott. Original number, 283. Characters.—Similar to Pteropus hypomelanus and P. lepidus, but color darker and teeth smaller. Color.—Type: back dark hair-brown, sprinkled with silvery whitish hairs, and with an indistinct brownish yellow wash along edges of membrane. Neck russet, blackening laterally. Head grayish ochraceous-buff. Underparts blackish, irregularly suffused with russet and sprinkled with silvery whitish hair-tips. Ears, feet, and membranes black. Skull and teeth.—The skull is in all respects essentially like that of Pteropus hypomelanus from the Natuna Islands, though the palate is usually less broad. Teeth as in Pteropus hypomelanus but smaller, the difference particularly noticeable in the second upper molar and first lower molar. *(adAq) apeuay ‘szswasvd snovovpy “2 “vayeumns ‘Avg ynuedey ‘coShr1 “oN ‘a[BUlsy ‘9702-47 8a0002 SNOILOATIO*?) SNOAUNVTTI SIJQY NVINOSHLINS = MILLER] SEVENTY NEW MALAYAN MAMMALS 61 Measurements.—External measurements of type: head and body, 215; tibia, 55; foot, 45 (36) ; forearm, 115; thumb, 62 (52) ; second digit, 105 (103) ; third digit, 257; fourth digit, 202; fifth digit, 177; ear from meatus, 23; ear from crown, 21; width of ear, 15. Cranial measurements of type: greatest length, 64; basal length, 60; basilar length, 56; median palatal length, 35; palatal breadth between anterior molars, 10; zygomatic breadth, 33; least inter- orbital breadth in front of postorbital processes, 8.4; least inter- orbital breadth behind postorbital processes, 8; distance between tips of postorbital processes, 24; greatest breadth of braincase above roots of zygomata, 23; greatest depth of braincase, 18; occipital depth, 12; depth of rostrum at middle of diastema, 6.8; maxillary toothrow (exclusive of incisors), 23.4; crown of first upper molar, 4.8X3; mandible, 49.4; mandibular toothrow (exclusive of incisors), 27; crown of first lower molar, 4.42.6. Specimens examined.—Fifteen (seven skins), all from South Twin Island. Remarks.—The series shows little variation in color. Such as there is chiefly involves the greater or less amount of blackish suffusion on the neck, and the strength of the yellowish tinge in the gray of the head. In one female (No. 104,466) the sides of the abdomen are strongly washed with silvery gray. MACACUS PAGENSIS sp. nov. (Piate XI, ricurE 2; Prate XII, ricure 2; Prate XIII, Fricure 1) Type.—Adult female (skin and skull), No. 121,653, United States National Museum. Collected on South Pagi Island, Sumatra, No- vember 17, 1902, by Dr. W. L. Abbott. Original number, 2053. Characters—Like Macacus nemestrinus but much smaller. Gen- eral color darker than in the related species, but sides of neck light yellowish brown, in strong contrast with dorsal surface. Fur.—The fur does not differ essentially from that of an adult female Sumatran specimen, except that the tail is much more thinly haired, and there are some peculiarities in the growth of the hair on head. The tail is so sparsely haired that the skin is nowhere con- cealed except at extreme base, while in M. nemestrinus no skin can be seen. This may be a seasonal character, but the fact that the scattered hairs on the tail of Macacus pagensis are all of the same light brown color, those of the upper surface showing no suggestion of the blackish dorsal stripe of the related animal, makes it seem probable that the tail is always essentially as in the type. On fore- head the hairs from a line joining anterior bases of ears all grow 62 SMITHSONIAN MISCELLANEOUS COLLECTIONS [VoL. 45 forward, while in 7. nemestrinus those in a band about 20 mm. wide above the eyebrows grow backward, meeting the others in an indis- tinct ridge. The hairs of the cheeks are all directed backward, to and surrounding base of ears. There is thus no indication of the conspicuous semicircle of antrorse hairs surrounding front of ear in the larger species. Color.—Dorsal surface from forehead to base of tail clear bister, darker than that of Ridgway, the drab underfur appearing irregu- larly at surface when hair is disarranged. Sides of body and inner surface of arms and legs isabella-color. Belly isabella-color, fading to light fawn-color on chest and throat. Outer surface of arms light russet, that of legs dark isabella-color except on thighs, which are mostly covered by an extension of the brown area of back. A similar but less extensive wash covers proximal half of upper arm. Sides of neck grayish cream-buff in striking contrast with upper surface. Cheeks and chin brown like that of back, but not quite as dark. Hands and feet dusky brownish. Tail sprinkled with isabella-colored hairs. “* Callosities fleshy brown. Palms and soles light fleshy brown.’* Skull and teeth.—The skull (pl. x1, fig. 2; pl. x11, fig. 2; pl. x11, fig. I) is very much smaller than that of a slightly younger female of Macacus nemestrinus (pl. x1, fig. 1; pl. x11, fig. 1; pl. x11, fig. 2) trom Tapanuli Bay, Sumatra. In general form, however, the two do not appreciably differ. The bony palate is concave laterally (when viewed from below, but to a less degree than in the larger animal. Its median line is nearly straight, and shows only a trace of the deep longitudinal concavity so conspicuous in M. nemestrinus in region between premolars. Audital bullz a little more swollen antero- laterally than in M. nemestrinus, but this character may be purely individual. Teeth as in Macacus nemestrinus but smaller through- out. In the larger animal the last upper molar is provided with a distinct fifth cusp on the outer posterior margin of the crown. Scarcely a trace of this can be detected in Macacus pagensis. Simi- larly the back lower molar has one less cusp than that of Macacus nemestrinus. The missing cusp in this tooth appears to be the pos- terior on inner side. Whether these differences are any more than individual peculiarities it is impossible to tell. Measurements.—External measurements of type: total length, 580 (690) ;? head and body, 435 (480) ; tail vertebra, 145 (210) ; ‘Collector’s note on label. 2 Measurements in parenthesis are those of a younger female of Macacus nemestrinus from Tapanuli Bay, Sumatra (No. 114,502). *(adAy) ayeuray ‘szswasvd snovovpy “% “eAyewNG *A ISSHLING MILLER] SEVENTY NEW MALAYAN MAMMALS 63 foot, 125 (152); foot without claws, 120 (148); ear from meatus, 31 (32); ear from crown, 28 (26) ; width of ear, 30 (32). Cranial measurements of type: greatest length, 111 (134) ; basal length, 78 (96); basilar length, 76 (92); least palatal length, 38 (52.4) ; palatal breadth between front molars, 20.6 (26.4) ; zygo- matic breadth, 71 (85) ; mastoid breadth, 62.6 (72) ; greatest breadth of braincase above roots of zygomata, 61 (68); least breadth of braincase immediately behind orbits, 42 (50); greatest orbital breadth, 60 (66); least distance from orbit to alveolus of inner incisor, 33 (47.6) greatest depth of braincase, 47.4 (48) ; mandible, 79.6 (94); maxillary toothrow exclusive of incisors (alveoli), 35.4 (41.6) ; three upper molars together (crowns), 22.6 (26); man- dibular toothrow exclusive of incisors (alveoli), 40.4 (47); three lower molars together (crowns), 23.4 (28). Specimens examined.—Two, the type, and a very young individual, both from South Pagi Island. Remarks.—The young, of which the type was the parent, closely resembles the adult in all respects, except that the tail is somewhat less thinly haired. MACACUS PHALURA sp. nov. Type.—Adult male (skin and skull), No. 121,870, United States National Museum. Collected at Siaba Bay, Nias Island, Sumatra, March 20, 1903, by Dr. W. L. Abbott. Original number, 2399. Characters——Similar to Macacus fascicularis, but general color slightly darker, and upper surface of tail conspicuously blackish. Color—Type: back and sides buff, slightly browner and more yellow than that of Ridgway, and everywhere inconspicuously grizzled with black. The sides are a trifle less yellowish than the back. Outer surface of arms like sides of body, but distinctly tinged with gray; outer surface of legs essentially like sides of body. Fingers, toes, and sides of feet indefinite dark brown. Crown like back. Forehead sprinkled with blackish hairs. Cheeks grizzled buffy gray. Underparts, sides of neck, and inner surface of legs pale ecru-drab. Tail ecru-drab below, blackish above. Near base the hairs of the upper surface of the tail are inconspicuously grizzled with buffy brown annulations. Beyond middle the blackish be- comes diluted with ecru-drab. Skull and teeth.—The skull and teeth show no tangible charac- ters to distinguish them from those of Macacus fascicularis. 1For use of the name Macacus fascicularis (Raffles) in place of V. “ cynomolgus” see Bonhote, Fasciculi Malayenses, Zoology, I, p. 3. July, 1903. 64 SMITHSONIAN MISCELLANEOUS COLLECTIONS [ VOL. 45, Measurements.—External measurements of type: total length, 940; head and body, 460; tail vertebre, 480; foot, 130 (127) ; skull, greatest length, 116; basal length, 85; basilar length, 82; zygomatic breadth, 77 ; maxillary toothrow exclusive of incisors (alveoli), 34. Specimens examined.—Four, all from Siaba Bay, Nias Island. Remarks.—Although the characters on which this species is based are slight they are very constant in the four skins as compared with an extensive series of Macacus fascicularis from various localities. in Sumatra, the Malay Peninsula, and elsewhere. PRESBYTES RHIONIS sp. nov. Type.—Adult female (skin and skull), No. 115,665, United States National Museum. Collected at Telok Pemudong, Pulo Bintang, Rhio Archipelago, August 15, 1902, by Dr. W. L. Abbott. Original number, 1888. Characters.—Like Presbytes natune (Thomas and Hartert), but color not as dark. Color.—Type: general color above, from region between ears to and including basal fourth of tail, broccoli-brown, washed with wood-brown on shoulders and tail and with chocolate on sides and lumbar region. The hairs are ecru-drab at base, and the irregular appearance of this color at surface further complicates any attempt at an exact description. From axilla to thigh may be traced a faintly indicated dark line about 25 mm. in width bordering the whitish of ventral surface. Front part of crown darker than back, the color becoming almost black on forehead and temples. Feet and hands blackish. On arms this color fades into the general tint of the body at region of elbows, but on legs it is continued, with only a slight admixture of brown, to thigh patch, and along outer side of this to join the faint dark lateral stripe. Thigh patches large and con- spicuous, as in Presbytes natune, white, slightly tinged with cream- color. Above they are sharply defined from the brown lumbar region, but below and at the sides they fade somewhat gradually into the color of the neighboring parts. Entire ventral surface, inner side of arms to elbows and of legs to knee white, tinged with gray, particularly on chest and thighs. The light area on arms and legs is continued to wrists and ankles, but beyond knees and elbows it is much encroached upon by the dark brown. The tail darkens from proximal fourth until at tip it is seal-brown. The under surface is nowhere different from the upper surface. The series shows some trifling variations in the exact shade of the dark areas, but this is chiefly due to bleaching of the hairs, and none SMITHSONIAN MISCELLANEOUS COLI VOL. 45, PL. XIII Macacus pagensis, female (type). 2. Macacus nemestrinus, female, No. 114,502, Tapanuli B Sumatra, MILLER] SEVENTY NEW MALAYAN MAMMALS 65 of the skins shows any very close approach to the blackish hues of Presbytes natune. Skull and teeth—The skull and teeth appear to be identical with those of Presbytes natune. Measurements.—External measurements of type: total length, 1173; head and body, 468; tail vertebre, 705; foot, 150 (145). Measurements of adult male from type locality: total length, 1213; head and body, 550; tail vertebrze, 663 ; foot, 150 (147). Cranial measurements of type: greatest length, 88 (86) ;? basal length, 63 (61); basilar length, 57 (56); least palatal length, 24 (23) ; palatal breadth (between front molars), 18.8 (18) ; zygomatic breadth, 66 (65) ; mastoid breadth, 58.4 (54); greatest breadth of braincase above roots of zygomata, 52.4 (52); least breadth of braincase immediately behind orbits, 46 (43); greatest orbital breadth, 62 (57.6) ; least distance from orbit to alveolus of inner incisor, 15.6 (15.6) ; greatest depth of braincase, 43 (43) ; mandible, 61 (62) ; maxillary toothrow (exclusive of incisors), 27 (26) ; man- dibular toothrow (exclusive of incisors), 32.2 (31.8). Specimens examined.—Seven (one skull without skin), all from Pulo Bintang. Remarks.—Although closely related to Prasbysts natune, this monkey is readily distinguishable by the absence of the blackish cast on back, legs, and tail. It is common on Pulo Bintang, and Dr. Abbott notes that its cry is exactly like the shrill ka—ka—ka of P. natune and P. sumatranus. From this cry is derived the Malay name, “kaka.” In the living and freshly killed animals the palms and soles are black, the face is slaty, and the eyelids and lips are fleshy white. PRESBYTES BATUANUS sp. nov. Type.—Adult male (skin and skull), No. 121,810, United States National Museum. Collected on Pulo Pinie, Batu Islands, Su- matra, March 4, 1903, by Dr. W. L. Abbott. Original number, 23609. Characters —Similar to Presbytes swmatranus, but not as large and tail not as long. Skull somewhat broader than in the Sumatran animal. Color.—Type: belly, under side of tail to about middle, inner sur- face of thighs, and median line of chest grayish white. Remainder of fur black, the back and crown lightened by a brownish wash. Beyond middle of tail the grayish white stripe on ventral surface becomes much reduced in width, and its distinctness is further Measurements in parenthesis are those of an adult female Presbytes natune from Bunguran Island (No. 104,849). MILLER] SEVENTY NEW MALAYAN MAMMALS 65 of the skins shows any very close approach to the blackish hues of Presbytes natune. Skull and teeth—The skull and teeth appear to be identical with those of Presbytes natune. Measurements.—External measurements of type: total length, 1173; head and body, 468; tail vertebrze, 705; foot, 150 (145). Measurements of adult male from type locality: total length, 1213; head and body, 550; tail vertebra, 663 ; foot, 150 (147). Cranial measurements of type: greatest length, 88 (86) ;1 basal length, 63 (61); basilar length, 57 (56); least palatal length, 24 (23) ; palatal breadth (between front molars), 18.8 (18) ; zygomatic breadth, 66 (65) ; mastoid breadth, 58.4 (54); greatest breadth of braincase above roots of zygomata, 52.4 (52); least breadth of braincase immediately behind orbits, 46 (43); greatest orbital breadth, 62 (57.6); least distance from orbit to alveolus of inner incisor, 15.6 (15.6) ; greatest depth of braincase, 43 (43) ; mandible, 61 (62) ; maxillary toothrow (exclusive of incisors), 27 (26) ; man- dibular toothrow (exclusive of incisors), 32.2 (31.8). Specimens examined.—Seven (one skull without skin), all from Pulo Bintang. Remarks.—Although closely related to Prasbytes natune, this monkey is readily distinguishable by the absence of the blackish cast on back, legs, and tail. It is common on Pulo Bintang, and Dr. Abbott notes that its cry is exactly like the shrill ka—ka—ka of P. natune and P. sumatranus. From this cry is derived the Malay name, “kaka.” In the living and freshly killed animals the palms and soles are black, the face is slaty, and the eyelids and lips are fleshy white. PRESBYTES BATUANUS sp. nov. Type.—Adult male (skin and skull), No. 121,810, United States National Museum. Collected on Pulo Pinie, Batu Islands, Su- matra, March 4, 1903, by Dr. W. L. Abbott. Original number, 2369. Characters.—Similar to Presbytes sumatranus, but not as large and tail not as long. Skull somewhat broader than in the Sumatran animal. Color.—Type: belly, under side of tail to about middle, inner sur- face of thighs, and median line of chest grayish white. Remainder of fur black, the back and crown lightened by a brownish wash. Beyond middle of tail the grayish white stripe on ventral surface becomes much reduced in width, and its distinctness is further *Measurements in parenthesis are those of an adult female Presbytes natune from Bunguran Island (No. 104,849). 66 SMITHSONIAN MISCELLANEOUS COLLECTIONS [voL. 45 lessened by admixture of black hairs. It may be traced however to extreme tip, which is also grayish. The gray of the tail is somewhat washed with pale buff. Skull and teeth.—The skull of Presbytes batuanus is smaller than that of P. swmatranus and its width is greater in proportion to the length. The orbits are better defined from outline of braincase above, and the frontal region is less swollen. Teeth as in Presbytes sumatranus, but uniformly smaller. Measurements.—External measurements of type: total length, 1150; head and body, 485; tail vertebrze, 665; foot, 163 (161). Cranial measurements of type: greatest length, 90 (96.4) ;? basal length, 64.6 (69); basilar length, 59 (65); median palatal length, 29 (30); palatal breadth between front molars, 17.4 (18) ; zygomatic breadth, 72 (73); mastoid breadth, 58 (62); greatest breadth of braincase above roots of zygomata, 52.6 (56); least breadth of braincase immediately behind orbits, 41 (43); greatest orbital breadth, 61.4 (62); least distance from orbit to alveolus of inner incisor, 17 (21) ; greatest depth of braincase, 42.6 (44.6) ; man- dible, 64 (66.6) ; maxillary toothrow exclusive of incisors (alveoli), 28 (29); three upper molars together (crowns), 16.8 (16.8) ; man- dibular toothrow exclusive of incisors (alveoli), 33 (32) ; three lower molars together (crowns), 18.4 (17). Specimens examined.—Ten, all from the Batu Islands. Remarks.—The Batu Presbytes is readily distinguishable from its Sumatran representative by the smaller general size, and par- ticularly by the shorter tail. Of four adults of the latter from Tapanuli Bay none has the tail less than 730 mm. in length, while among the nine adults of Presbytes batuanus the longest tail is only 710 mm. ‘The series shows no variations worthy of special note. SIMIAS gen. nov. (Cercopithecide). Type.—Simias concolor sp. nov. Characters.—Skull essentially as in Nasalis, but rostrum less pro- duced and nasals not as wide. Nose like that of Rhinopithecus. Tail about one-third as long as head and body, naked except for an inconspicuous tuft of hair at tip. Ischial callosities large and con- spicuous. Teeth as in Nasalis, Rhinopithecus, and Presbytes. No cheek pouches. Remarks.—This genus combines the more important structural characters of Nasalis and Rhinopithecus with an external form un- ‘Measurements in parenthesis are those of an adult male Presbytes sumatranus from Tapanuli Bay, Sumatra (No. 114,507). SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 45, PL. XIV Simias concolor (from photograph of freshly killed individual). MILLER] SEVENTY NEW MALAYAN MAMMALS 67 like that of its relatives and strongly suggesting Macacus nemes- trinus. The nasal bones are fully as long as in Nasalis larvatus, though their width is less and their anterior extremities probably never form a distinct median projection. Among the nine skulls that I have examined none shows any tendency toward the rudi- mentary form of nasals characteristic of Presbytes and to an even greater degree of Rhinopithecus. he interorbital region is rela- tively longer and narrower than in any of the related genera, and the orbits are better defined above, two characters that give the skull a strong resemblance to that of Macacus. The facial profile is nearly straight from forehead to middle of nares. The development of the rostrum is intermediate between that of Nasalis and Presbytes, and appears to be closely similar to that in the two species of Rhinopi- thecus whose skulls are known. The teeth do not differ appreciably from those of the members of the related genera. Although the skull agrees so closely with that of Nasalis the nose is very different, show- ing no tubular elongation. As in Rhinopithecus the apertures of the nostrils are directly on the surface of the very wide upper lip; and it is only the superior nasal margin that is lengthened to give the face its snub-nosed aspect. While the structural characters of the head are essentially a combination of those of the previously known genera, the form of the body departs widely from that of other mem- bers of the subfamily Presbytine, particularly in the shortness of the arms and tail, and resembles the pig-tailed macaques. The limbs are not as robust as in Macacus nemestrinus, but the tail is of about the same relative length, and the ischial callosities are fully as con- spicuous. Although distinct from each other and from Presbytes, the genera Nasalis, Rhinopithecus, and Simias form a compact and easily recognizable group. SIMIAS CONCOLOR sp. nov. (PLates XIV, XV, XVI) Type.—Adult male (skin and skull), No. 121,659, United States National Museum. Collected on South Pagi Island, Sumatra, De- cember 3, 1902, by Dr. W. L. Abbott. Original number, 2103. Characters.—Nose and teeth essentially like those of Rhinopithecus roxellana and R. bieti as described and figured by Milne-Edwards. Skull essentially as in Nasalis larvatus but smaller, the rostrum less produced, and nasal bones narrower. Size and general form about as in Macacus nemestrinus; tail only one-third as long as head and body, naked except for an inconspicuous tuft of hair at tip. General color throughout dusky brown, the underparts darker than back. 68 SMITHSONIAN MISCELLANEOUS COLLECTIONS [voL. 45 External form.—In general external form this species very closely resembles Macacus nemestrinus, differing only in its slightly longer, less robust arms and legs, and relatively smaller, more rounded head. The hands and feet appear to be broader. ‘The tail is proportioned almost exactly as in Macacus nemestrinus. Face.—The physiognomy of Simias concolor, as shown by photo- graphs of a freshly killed individual (pl. x1v) is essentially like that of Rhinopithecus roxellana. ‘The elevated ridge above the nostrils is less abrupt than in the Tibetan animal as figured by Milne-Edwards, and the concavity extending from this ridge to the eyebrows is longer and less deep. Otherwise there is a very close agreement between the two species. Callosities.—The callosities are very large and conspicuous, even more so than in Macacus nemestrinus. In the male they are joined solidly together along the median line, but in the female they are separated by a narrow strip of softer skin. Fur.—The fur is of the same length as that of Macacus nemes- trinus, except that the underparts are thickly haired from chin to hypogastric region. ‘The latter, together with the axille and inner side of upper arm are nearly bare. On head the hair spreads uni- formly from forehead, that of middle growing directly backwards, that of sides standing out as distinct tufts over ears. On cheeks the hairs grow upwards, meeting the ear tufts and continuing them backward beneath ears. Slightly below middle of cheek the hair abruptly changes direction and grows downward, forming a thin tuft on each side of the chin. Like rest of under surface of body the chin is closely furred except the space immediately about lips. On shoulders and back of neck the hairs are slightly elongated, but not enough to form a distinct cape. ; Color.—Type: general color throughout clove-brown somewhat lightened by drabby reflections. On underparts, head, legs, lumbar region, and terminal tuft of tail, the brown is unmixed with lighter color, and on hands and feet it darkens almost to black. On neck, shoulders, outer side of upperarm, and entire back to lumbar region, most of the hairs have a buff annulation about 4 mm. in width near tip. These annulations are most abundant on neck and shoulders, where they are also paler than elsewhere, approaching a grayish cream-buff. In this region they produce a distinct grizzled effect, much like that of the same parts in Macacus umbrosus and M. fuscus. On sides of body they become less distinct than on back, and the ground color at the same time changes to a rather dark broccoli-brown. The naked portion of face and chin is bordered by SMITHSONIAN M o ou > 2 ~~ o « x Ss a = 8 % 8 ‘8 By S Q *(adAq) apeur *40702W09 SV22L1G ING SNOILOATIOZ) SNOUNY LOST NVIN¢ TAS id | DORA MILLER] SEVENTY NEW MALAYAN MAMMALS 69 whitish hairs, but these are not numerous enough to be very con- spicuous. “ Palms, soles, face, and callosities black in both sexes. Anus pinkish white in male. Iris brown.’ Skull and teeth.—The skull (pls. xv, xvr) rather closely resembles that of an adult male Nasalis larvatus, but is considerably smaller ; the rostral portion is less developed, and the upper margins of the orbits are more distinctly outlined from the braincase. As compared with the skulls of Rhinopithecus figured by Milne-Edwards, the width is less in proportion to the length, the anterior nares are narrower, the orbits are nearer together, the ridge above them is less curved, and the whole facial aspect is more suggestive of Macacus. Ros- trum decidedly more produced than in the Tibetan species. Teeth essentially as in Presbytes, but inner cusps of upper molars and outer cusps of lower molars even better developed and in height almost equal to those of the opposite side of the teeth. Measurements.—External measurements of type: total length, 740; head and body, 550; tail vertebrze, 190; pencil, 20; foot, 170; ear from meatus, 24; ear from crown, 12; width of ear, 24. Ex- ternal measurements of an adult female (No. 121,658) from South Pagi Island: total length, 650; head and body, 550; tail vertebre, 100; pencil, 10; foot, 150. Cranial measurements of type: greatest length, 106; basal length, 82; basilar length, 77; median palatal length, 36; palatal breadth be- tween front molars, 21.6; zygomatic breadth, 75.4; mastoid breadth, 62; greatest breadth of braincase above roots of zygomata, 56; least breadth of braincase immediately behind orbits, 39.6; greatest orbital breadth, 64.6; least interorbital breadth, 8.8; least distance from orbit to alveolus of inner incisor, 32; greatest depth of braincase, 43; mandible, 74; depth of mandible at posterior end of last molar, 27.4; maxillary toothrow exclusive of incisors (alveoli), 34.6; three upper molars together (crowns), 21.6; mandibular toothrow ex- clusive of incisors (alveoli), 40.6; three lower molars together (crowns), 23. Specimens examined.—Ten, all from South Pagi Island. Remarks.—This monkey is so different from all other known species that further comparison is unnecessary. The specimens show no appreciable variation, except that the females are uniformly much smaller than the males. Two young individuals (male, No. 121,655, and female, No. 121,656), 455 mm. and 465 mm. respectively in length, and with the milk dentition still in place, are light cream-buff throughout. Another (female, No. 1Collector’s note; does not refer specifically to the type specimen. 7O SMITHSONIAN MISCELLANEOUS COLLECTIONS [VOL. 45 121,662) of the same size is normal. In the type the tuft of hairs at end of tail is better developed than in any of the others. This tuft is occasionally represented by a mere lengthening of the short hairs that are sprinkled over the rest of the tail. SYMPHALANGUS KLOSSII sp. nov. (PLATE XVII, FicGuRE 2; PrLate XVIII, FicurE 2; PLATE XIX, FIGURE 1) Type.—Adult male (skin and skull), No. 121,678, United States National Museum. Collected on South Pagi Island, Sumatra, No- vember 13, 1902, by Dr. W. L. Abbott. Original number, 2032. Characters—A dwarf siamang of about half the weight of Symphalangus syndactylus. Fur shorter and less woolly than in the Sumatran animal. Throat densely furred in both sexes. Fur.—The fur is moderately long, and of a silky texture, not coarse and woolly as in Symphalangus syndactylus. It shows only a slight tendency to lengthen on neck and shoulders, and scarcely any on thighs and outer surface of upper arms. On underparts it is rather shorter than elsewhere, becoming rather thin in hypogastric region, but on throat it continues uninterruptedly to chin. Eye- brows scarcely lengthened, giving the forehead a peculiar rounded appearance. Color.—The color is black throughout, and there are no whitish hairs on face and chin. Skull and teeth—The skull and teeth (pl. xvit, fig. 2; pl. xvitr, fig. 2; pl. x1x, fig. 1) so closely resemble those of Symphalangus syndaciyins, (pl. xvi, fies Teypl. xviii, tier | pl exrx, tie. 2) atiete can detect no constant differences other than size. Measurements.—External measurements of type: head and body, to symphysis pubis, 440 (525) ;1 foot, 130 (154). External meas- urements of an adult female (No. 121,687) from the type locality: head and body, 445; foot, 123. Cranial measurements of type: greatest length, 97 (129) ;1 basal length, 75 (109) ; basilar length, 70 (104) ; median palatal length, 30.8 (59) ; palatal breadth between front molars, 16.4 (27) ; zygo- matic breadth, 65 (90) ; mastoid breadth, 64 (78) ; greatest breadth of braincase above roots of zygomata, 56.6 (66) ; least breadth of braincase immediately behind orbits, 45.4 (49); greatest orbital breadth, 58 (75); least interorbital breadth, 9 (15); least distance from orbit to alveolus of middle incisor, 23.6 (38) ; greatest depth of braincase, 46 (55) ; mandible, 68 (95) ; depth of mandible at pos- 1 Measurements in parenthesis are those of an adult male Symphalangus syndactylus from Tapanuli Bay, Sumatra (No. 114,406). 0 ‘o[eul ‘snpAqovpurs susuvpoy Putas ‘(ad Aq) ayeur *77ss074y snSuvppyguAG “% ‘eayeurns ‘Avg | nuvdey ‘gbhF11 l i 1 LOY S Pil J 107) SNOAUNWTITAOSII NVINOSHLING *(ad4q) apeur ‘2288074 snSuvjpoy duds 2 ‘exyeuns ‘Avg ynuedey, ‘o6h‘b11 ‘on ‘ayeur ‘snpdqovopuds snsunjoyguars “1 ITIAX “Id ‘SP “IOA SNOILOATIO) SQNOANVTTHOSIP NVINOSHLING MILLER] SEVENTY NEW MALAYAN MAMMALS aL terior end of last molar, 10 (15.4) ; depth of mandible through coron- oid process, 20.4 (36); maxillary toothrow exclusive of incisors (alveoli), 30.6 (45.4) ; three upper molars together (crowns), 14.8 (25) ; crown of first upper molar, 5X6 (87.6) ; mandibular molar series exclusive of incisors (alveoli), 34 (50); three lower molars together (crowns), 17 (25.4); crown of first lower molar, 5.85 (8.6X7). Weight.—Weight of type, 6.12; weight of a second adult male, 5.21; weight of four adult females respectively, 5.21, 5.78, 6.12, and 6.46 kg.? Specimens examined.—Eighteen (four young in alcohol), all from South Pagi Island. Remarks.—At Dr. Abbott’s request I have named this siamang after Mr. C. B. Kloss, to whose aid was due much of the success of the second expedition to the coast and islands of western Sumatra. The characters of the species are so striking that no special com- parison is required with Symphalangus syndactylus, the only other member of the genus at present known. Motels tnbe SPECIES DESCRIBED IN THIS ‘PAPER. Tragulus batuanus, p. 2. Tragulus russulus, p. 3. Ratufa insignis, p. 4. Ratufa conspicua, p. 5. Ratufa bale, p. 6. Ratufa mase, p. 7. Ratufa piniensis, p. 8. Sciurus bilimitatus, p. 8. Sciurus pemangilensis, p. 9. Sciurus aoris, p. 10. Sciurus peninsularis, p. Io. Sciurus pannovianus, p. II. Sciurus ictericus, p. 12. Scirus atratus, p. 13. Sciurus piniensis, p. 14. Sciurus bale, p. 14. Sciurus pumilus, p. 15. Sciurus lancavensis, p. 16. Sciurus adangensis, p. 17. Sciurus sullivanus, p. 17. Sciurus domelicus, p. 18. Sciurus bentincanus, p. 19. Sciurus mattheus, p. 19. Sciurus lucas, p. 20. Sciurus casensis, p. 20. Sciurus altinsularis, p. 21. Sciurus rubeculus, p. 22. Funambulus obscurus, p. 23. Funambulus rostratus, p. 24. Funambulus peninsula, p. 25. Sciuropterus merens, p. 26. Petaurista batuana, p. 27. Mus stridens, p. 28. Mus mattheus, p. 20. Mus stridulus, p. 20. Mus lucas, p. 30. Mus soccatus, p. 30. Mus mase, p. 32. Mus bale, p. 33. Mus lugens, 33. Mus julianus, p. 34. Mus gilbiventer, p. 35. Mus luteolus, p. 36. Mus umbridorsum, p. 37. The weight of four adult Symphalangus syndactylus from Tapanuli Bay, Sumatra, is as follows: two males, 11.79 and 12.70; two females, 9.71 and 11.56 kg. 72 SMITHSONIAN MISCELLANEOUS COLLECTIONS Mus bentincanus, p. 38. Mus casensis, p. 38. Mus domelicus, p. 309. Mus pagensis, p. 39. Chiropodomys niadis, p. 40. Atherura sygomatica, p. 42. Hemigale minor, p. 43. Paradoxurus lignicolor, p. 44. Galeopithecus pumilus, p. 46. Galeopithecus aoris, p. 47. Galeopithecus gracilis, p. 49. Galeopithecus natune, p. 50. Galeopithecus saturatus, p. 51. [VOL. 45 Galeopithecus tuancus, p. 53. Tupaia castanea, p. 54. Tupaia pulonis, p. 56. Tupaia tephrura, p. 57. Tupaia chrysogaster, p. 58. Tupaia cervicalis, p. 59. Pteropus geminorum, p. 60. Macacus pagensis, p. 61. Macacus pheura, p. 63. Presbytes rhionis, p. 64. Presbytes batuanus, p. 65. Simias concolor, p. 67. Symphalangus klossii, p. 70. LIST OF ILLUSTRATIONS (All figures, unless otherwise stated, are four-fifths natural size.) Petaurista nitidula, male, No. 104,622, Bunguran Island, North Petaurista nitida, male, No. 121,499, eastern Java. Atherura macroura, female, No. 84,433, Trong, Lower Siam. . Hemigale hardwickii, female, No. 114,461, Tapanuli Bay, Sumatra. . Paradoxurus hermaphroditus, male, No. 86,793, Trong, Lower . Paradoxurus hermaphroditus, male, No. 86,793, Trong, Lower Galeopithecus gracilis, male, No. 104,602, Sirhassen Island, South Galeopithecus volans, female, No. 84,420, Trong, Lower Siam. Galeopithecus volans, male, No. 115,493, Rumpin River, Pahang. (PLATES ) PiateE I. Figure 1. Sciuropterus me@rens, type. Figure 2. Funambulus obscurus, type. Pate II Figure I. Natunas. Figure 2. Figure 3. Petaurista batuana, type. Figure 4. Atherura zygomatica, type. Figure 5. Pirate III Figure I Figure 2. Hemigale minor, type. PLate IV Figure 1. Paradoxurus lignicolor, type. Figure 2 Siam. PLATE. WV, Figure 1. Paradoxurus lignicolor, type. Figure 2 Siam. PLATE VI Figure 1. Galeopithecus gracilis, type. Figure 2. Natunas. Figure 3. Galeopithecus pumilus, type. Pirate VII Figure I. Figure 2. Figure 3. Galeopithecus saturatus, type. Figure 4. Islands. Galeopithecus saturatus, male, No. 121,747, Tana Bala, Batu SmITHSONIAN MiIsCELLANEOUS COLLECTIONS yuphalangus klossit,male (type). 2. Symphalangus syndactylus, No."114,496, Tapanuli Bay, Sumatra. MILLER] SEVENTY NEW MALAYAN MAMMALS 73 BrAmE nivel Figure 1. Galeopithecus volans, female, No. 84,420, Trong, Lower Siam. Figure 2. Galeopithecus volans, male, No. 115,493, Rumpin River, Pahang. Figure 3. Galeopithecus saturatus, type. Figure 4. Galeopithecus saturatus, male, No. 121,747, Tana Bala, Batu Islands. PLaTE IX Figure 1. Galeopithecus volans, female, No. 84,420, Trong, Lower Siam. Figure 2. Galeopithecus volans, male, No. 115,493, Rumpin River, Pahang. Figure 3. Galeopithecus saturatus, type. Figure 4. Galeopithecus saturatus, male, No. 121,747, Tana ‘Bala, Batu Islands. PLATE X Figure 1. Tupaia chrysogaster, type. Figure 2. Tupaia ferruginea, female, No. 105,033, Tanjong Dungun, Tringanu. PLATE XI Figure 1. Macacus nemestrinus, female, No. 114,502, Tapanuli Bay, Sumatra. Figure 2. Macacus pagensis, type. PLaTE XII Figure 1. Macacus mnemestrinus, female, No. 114,502, Tapanuli Bay, Sumatra. Figure 2. Macacus pagensis, type. PLATE XIII Figure 1. Macacus pagensis, type. Figure 2. Macacus nemestrinus, female, No. 114,502, Tapanuli Bay, Sumatra. Lo | PLATE XIV Figure 1. Simias concolor, adult, greatly reduced. From photograph of freshly killed individual. PLATE XV Figure 1. Simias concolor, type. PLATE XVI Figure 1. Simias concolor, type. PLATE XVII Figure 1. Symphalangus syndactylus, male, No. 114,496, Tapanuli Bay, Sumatra. Figure 2. Symphalangus klossti, type. PLaTeE XVIII Figure 1. Symphalangus syndactylus, male, No. 114,496, Tapanuli Bay, Sumatra. Figure 2. Symiphalangus klossii, type. PLATE XIX Figure 1. Symphalangus klossu, type. Figure 1. Symphalangus syndactylus, male, No. 114,496, Tapanuli Bay, Sumatra. (Text Ficure) Figure 1. Skull of Chiropodomys niadis X 2 (page 40). RECENT STUDIES OF THE SOLAR CONSTAND (OR RADIATION By GyoG ABBOT INTRODUCTION Within the last two years the observations of the Smithsonian Astrophysical Observatory under the direction of the Secretary, Mr. Langley, have been largely for the purpose of measuring the total solar radiation, its distribution in the spectrum, and the losses which it suffers by absorption in the solar and terrestrial gaseous envelopes. In the experimental work and reduction of observations Mr. Langley has been aided by the writer, but chiefly by Mr. F. E. Fowle, Jr., whose able handling of the work I wish particularly to acknowledge and commend. Preliminary notices of this investigation have appeared in the Smithsonian Report for 1902, and in an article by the Secretary in The Astro- physical Journal for March, 1903, to which sources the reader is referred for additional information in relation to the methods of study. In the present paper will be found a summary of the results thus far reached. ATMOSPHERIC ABSORPTION It is well known that the effectiveness of the solar and terrestrial gaseous envelopes to intercept by reflection or absorption and thus diminish the intensity of the solar radiations at the earth’s surface, varies greatly for rays of different wave-length. It is customary, speaking of the matter in ready though not strictly accurate terms, to combine these two effects of reflection and absorption under the single head of absorption, but to distinguish two kinds of absorption, namely, general and selective, of which the latter includes such sudden alterations of transmission as are seep in the Fraunhofer lines, while the former denotes merely a general weakening of the radiation extending over wide ranges of wave-length. Using this nomenclature, it appears to be the general absorption of the solar and terrestrial envelopes which chiefly affects the amount of solar radia- tion at the earth’s surface, although the selective absorption of 74 ABBOT | STUDIES OF SOLAR CONSTANT OF RADIATION 75 water vapor in the atmosphere is also both very effective and very variable.* The procedure employed here to determine the general absorption of the air consists chiefly in making bolographs—that is, automatic energy spectra—of the solar radiation as often as possible throughout days of uniform and excellent sky without alteration of the sensitive- ness of the apparatus. Such energy spectra are altered in appear- ance from one to another by the varying absorption of the different thicknesses of air, so that at a little after noon the height of the curve sl Angstrom has, however, attributed much importance to the absorption of carbonic acid gas, implying by his computation that not less than 61 percent of the solar radiation which reaches the outer layers of the earth’s atmos- phere is cut off by the absorption of this gas in a vertical transmission through the air. (See Annalen der Chemie und Physik, vol. 39, pp. 309-311, 1890.) He locates the absorption of this gas principally in the bands at 2.6u and 4.34; so that, as he says, its effect is not allowed for in the procedure for obtaning the value of the solar constant of radiation adopted by Mr. Langley in his research on Mount Whitney, and which is essentially that employed here. Angstrom, while using the same method in part, adds a second term amounting to more than half the whole in his computation, solely referring to the absorption of carbonic acid gas, and thus he attains his oft-quoted result for the solar constant of radiation of 4.0 calories per square centimeter per minute. For several reasons I am inclined to think Angstrom has greatly overestimated the importance of this carbonic acid absorption term: First, as he shows, the selective absorption of carbonic acid gas is, so far as I am aware, almost wholly for wave-lengths greater than 2.5and principally in two bands between wave-lengths 2.5 and 2.85 1 and between 4.20/¢and 4.50“ respectively, where the total amount of the solar radiation is apparently less than one percent of the whole, as determined not only from the appearance of the observed bolographic solar spectrum energy curve itself, but from a consideration of the probable temperature of the sun and the distribution of energy in the spectra of bodies at high tem- perature. As a very evidently too great estimate of the energy in these wave- length regions, it may be seen that if the radiation outside the atmosphere (see plate xx1I) was of the same intensity throughout these bands as at 2.1/ the area they would include would be only about one-fiftieth the total area under the curves of plate xxm. It is of course very improbable that the height of the curve at 4.34 is nearly as great as at 2.1. Thus it would appear that the selective absorption of this gas for direct solar radiation is almost negligible. Second, if carbonic acid exercised a general absorption through the more intense parts of the solar spectrum, it is not apparent why such a general absorption is not included and allowed for in the coefficients of absorption here determined. Third, values of the solar constant computed here for the same day, but from observations made through very different thicknesses of air, are found to agree excellently, which appears to confirm the accuracy of the method of determining the atmospheric absorption which is here employed. 76 SMITHSONIAN MISCELLANEOUS COLLECTIONS [VOL. 45 is found to be a maximum for all wave-lengths, and the height falls off as the sun declines in altitude, slowly in the infra-red region of the spectrum, but more and more rapidly as we examine further and further toward the violet, or still more rapidly if we note the great atmospheric absorption bands due to water vapor. It is assumed that the atmospheric transmission for a very narrow portion’ of spectrum may be expressed by the relation— B C=, a Bo (1) where e and é, are the intensities of light of this wave-length at the earth’s surface and outside the atmospheric respectively, a the frac- tion transmitted by the atmosphere for zenith sun, m the air mass, or ratio of the length of the transmitting column of air to that for zenith sun, and # and /%, the observed and standard barometer readings respectively. Upon the bolograph the height d corre- sponding to any given wave-length is directly proportional to the amount of energy of that wave-length. Accordingly we may intro- duce a factor k constant for the single wave-length in question me 2 ad=ke= kea Bo ) and hence B log d= ms log a + log (£e,) (3) 0 As the last term of equation (3) is to be supposed constant during the day’s observations, the expression is in the form of the equation of a straight line, and if the logarithms of the deflections at the given wave-length on the successive bolographs be plotted as ordinates with the quantities (72/3/8,) as abscissz, the several points so deter- mined should fall on a straight line of which the tangent of the inclination is the logarithm of the transmission coefficient (a) for the given wave-length. Mr. Langley has stated that the attempted measures of the solar constant from a station near sea-level like Washington are sub- ject to great uncertainty from the necessity of the very large and doubtful extrapolation for atmospheric absorption. Without in the least questioning this, and while calling special attention to the great interest which would attach to a repetition of the experiments at high altitudes, I incline to the belief that the closeness with which the plotted points determined as above described lie upon a straight line for wide ranges of air mass is a reasonably sure criterion of *In our practice less than the width between the D lines. ery = TS Se aime pee ne ree ome en Tinea Or fanaa ate lea okt ici aplien ae ED tesa sete p j : | i ne a } : i waa = - ect) | Ps anit = ” i bs } ! i i ' a t i ait ; f - . i alt, i We tg 7 nn — - o _ r = = r ee = iar eaten 4 “ t - wr 4 a e arent ae i 2 = | Ts ay F 5 P fy — A " * a 4 sk 7 ~ . fhe - } Lf J { , ‘I ~ _ ¢* 4 dee . r aa 4 a z = ib p—smegs = eg a freee ted =. ‘<2 > Sr peat eet tine Conteris “SNOILVAUYASAO OIHdVUNOTOT WOU AWAHAMSOWLY AHL JO AONAAVIUSNVAL TA --e ‘SISSY 4IY 0 oce 00€ oge ore ec 002 os 091 Ov! oz! XxX “Ta ‘Sh “104A SNOILOATIO) SNOANVTISOSIP NVINOSHLING SNOILITTISIG? YILIWONVATVD 40 SWHLIYVIOT7 ABBOT] STUDIES OF SOLAR CONSTANT OF RADIATION WG, the accuracy of the extrapolation. In order to give an impression of the weight which should be assigned to the solar constant values shortly to be given, I call attention to plate xx, which contains the plots for deducing atmospheric transmission at several wave-lengths for two days, March 25, 1903, and March 26, 1903, observations for the two days being represented by circles and crosses respectively. The tangent of the angle of inclination of the plotted lines is the logarithm of the coefficient of transparency of the atmosphere for vertical transmission of a ray of the given wave-length. Plots 1 and II represent a wave-length of 1.027 »; 111 and Iv, 0.656 y»; v and vi, 0.468 # ; and vit and vim1, 0.395 ». In connection with this branch of the subject it is well to remark what the experience of meteorologists generally no doubt confirms, that the afternoon hours are found far more uniform in transparency of the air than the morn- ing hours, so that the observations of atmospheric transmission for use in computing values of the solar constant are obtained chiefly in the afternoon. forenoon observations are distinguished in plate xx by being connected by dotted lines. In order to fix our ideas both of the magnitude and the variability of the absorption of the earth’s atmosphere, the following table, show- ing the percentage of transmission at numerous different wave- lengths for the days indicated, is given. The computations upon which the table is based were made at wave-lengths specially selected to avoid large terrestrial absorption bands, and thus the table gives values of the general absorption only. A few reductions have been made to determine the selective absorption within the numerous atmospheric bands of water vapor and oxygen, but while their dis- cussion has gone far enough to show that equation (1) apparently holds good in these bands, these results are not yet far enough ad- vanced to be included in the tables. While, as another criterion of the accuracy of the method of extrapolation, it is found, in accord with what has just been said, that the employment of these observed values of transmission within the water-vapor bands would prac- tically fill up these bands in computations of the form of the solar energy curve outside the atmosphere, yet in determining the solar constant they are smoothed over and the general transmission con- stants corresponding with the smoothed curves are employed in the computation. The days included in Table I were all nearly cloudless, and thus the results represent the transmission of the atmosphere in better than average conditions. In order to bring out clearly what seems to be a marked decrease in the transparency of the air for the present [VoL. 45. SMITHSONIAN MISCELLANEOUS COLLECTIONS DeLatz % bg to oI h 1 zI z16°0 Ltg'o ‘Sz9" ‘0 ‘L6L: oO 0$6'0 1z6'0 | o16'0 | L6g'0 616°0 $gg°0 | 9Sg'0 | L¥g'o 980 98°0 | Sg’0| zgro 06'0 ogo | LL:0| gL:o 6g'0 | Gian Volvo) nL "0 L6:0 06:0 | Lg:0} zg‘o 060 | €g:0 1g'0 ogo 68°0 1g°0| 6L°0| glL‘o 96'0 6g°0| Lg:0| tg'o z6'0 | 06°0| gg'0| f£g:0 z6'0 | Sg°0| £g°0| og‘o 96'0 z6°0, 16°0| 6g°0 v6'0| £60} 16°0 98°0 ego +6:0 | 16°0| 06°0! 6g'0 06:0} 6g°0| Lg:o z6'0| z6:0| 6¢°'0 S60 | +6'0 | +60 | z6'0 $6°0 | +6'0 Oo €1 €SL-0 LS3°'0 80g'0 aii ‘0 46) 0g'o LL:o zL‘o 6L:0 blo glo | £g'0 880 6L:0 98°0 bg'0 Lg°0 | 06°'0 to Lg°0 | 6g°0. | z-1061 jo yey) pure £061 jo uvout uaearyoq SouSsouIP aBejuaied 1 00 I es “SSeyl- ai WU) OF syuat94j207) woyssrustre, [weve | vere | se nees veceee eae ee arene me eee tener atone jo uvayy pecees pe ceseees tee eeeeeceeceesceser sascereccsseseseeesseses 7 _ TOO] jo ueayy ee eeees ee Sie Sa ee he EON Teron om | secre Bor wee e weet teen ece ees cesses ersens eee e eee ceseeeses eects: oe aan cy Aqn{ [reese essen A eee ere w cence ene eene renee eeneee se eeee ter eeeees eC OKT °9z judy lees o tecccecccces cece ec ee cece ses cee sesseeses an eecetesscvese Co6I ‘Lt judy cece eee ecccees secccenecereseres st ceeeecceeeeecccerseeeeeees COG T ‘oz Yor] | see eeneeenccccscccccccccsscrccsesers poate eens ceeeseees eoesere Co61 “Cz YOIv]N seeees Stee eeeee cence rccsen eens sceeneeesssscereessceesscesesseesl Gy 6C yore | weneeeeee see ee eee enee Boece cee n een nee eeees weseeeees **COGT “Cz Areniqa,q teen eee e ee te eee eennes Cece etree se eeerece eee tresses CO6y ‘61 Areniqo.y seeee Stent e eect cece sc neeeetaeeceaersseesecersceseeeeer eer HT “Cr IaquiaaoN tee eeeeeoees Pete eeeerseceeeee cesses ceresseessesessesessstsr hy bez 13aqo~Q, sneer cece ee rece ee nee eeeccsseceeses cee B eee ecceeeseneeereree7 O67 ‘OI 19q0}9Q, veererceceee Date eenee ce ceerreseseeessensecees ceeeesserssses POH T “Cy 1aqoyIQ ste eccepeceesereces eater eeeeenes ote eces veces seecceee seeve zob1 ‘6 1aqo}Q | seecerceevon fee e cece ee eeeeereccssceereccesseesereceissrOGy “TT raquiaydag cccees vecece tole 0 kee eee cen ve csenciecccevecececessesererercesces sr Oh T *g Avy pec ee cece cveeeeteeceesscnes eer cerceeseece reese de eerecenscses ZOOI ‘Iz YOIe Poe e rere ee escenees ce eecese tecseenesssscceees eee eenees cease 1061 ‘z JaquIaaoNny eee e eee wee eee nee se stew eens ee eesrecees cree ecees cee: ceceree IOOI Cz 19q0}9Q ae “ung, yjiuay aout uounpoy al uorssruisund eS ® qunrfooy a An ae ae: oben y f aie THEA 2. Wa iJ4598 Rarer ~ 4 : ‘) : ; * £ , q 7 ’ te L ¥ . fi 14 L ‘ - é ry ‘ o-% / var r 3 ¥ WwW, I “ ¢ = : ins 3 ‘ Pp a - ‘eh — J ‘ - A : I t TT pales - rie net it 7 | : Py eee Bil WB at. : ; =e i a ie prises e 6 geome oo (suseryderp [[us Suraourss Aq paseaioul wSy, “swuseaydeip [js Rutsodsajut Aq payslulwip WS 4% “SUl] O1az jo uoitsod aaL3 0} 193jnYs Aq jo 3nd wieeg |) ‘E061 ‘Lt TIadV JO SNOILVAUMASAO “WSINd SSVID °° V JO WONLOAdS UVIOS AHL JO SHAYNO AOWANA OIHdVANO1Od %m| tod q Hy x oe h } if Wee tela) _ qd et a Ey ee a a ‘ | ———— ————— —— ae 1xx "Ta ‘SP "TOA SNOILLOATIOD SQOANVTTAOSIPA NVINOSHLING ABBOT | STUDIES OF SOLAR CONSTANT OF RADIATION 79 calendar year, the means of the general absorption coefficients have been taken for the observations of 1901-02, and for those of 1903 separately. There is an average difference of ten percent in favor of the earlier years, and this cannot, so far as I know, be accounted for in any other way than by recognizing an actual decrease in the transparency of the air, beginning somewhere between November 15, 1902, and February 19, 1903. It might be urged that the change is perhaps an annual one, as most of the results of Ig01—02 are in the autumn and those of Igo03 in the spring. But in contradiction to this view we find the observations of March and May, 1902, generally above the mean of that year, so that I incline to think the change rather extraordinary than annual in character. Such a change would imply a correspond- ing reduction in the amount of direct solar radiation at the earth’s surface, and if general over a wide area would seem to be likely to occasion some alteration of climate. Recent actino- metric observations reported by several observers in this coun- try and in Europe’ seem to strengthen the probability that the change in transparency of the air is widespread, for their measures of solar radiation at the earth’s surface have been appreciably lower of late than for the same months of former years. Several writers have suggested the possibility of the wide dissemination of fine dust clouds from the volcanic eruptions of 1902, in explanation of the lower values. It will be noted from Table I that the differences between the means of 1901-02 and 1903 are largest for short wave- lengths and diminish nearly uniformly toward the infra-red as far as a wave-length of 1.2 », which would probably be in harmony with this hypothesis; for such small dust particles might be expected to scatter and absorb the shorter wave-lengths most, not being large enough to act like an opaque screen diminishing all wave-lengths proportionally. COMPUTATIONS OF THE SOLAR CONSTANT OF RADIATION The coefficients of general atmospheric transmission resting upon measures at twenty-four different wave-lengths from 0.37 4 to 2.3 » on series of bolographic curves have been employed at the Astro- physical Observatory in connection with bolographs and actinometric data to compute the solar constant of radiation outside the atmos- phere. Referring to plate xxi, the area included underneath a spectral energy curve is directly proportional to the total radiation absorbed by the bolometer over the range of wave-lengths included *See note by H. H. Kimball, Monthly Weather Review, May, 1903. 80 SMITHSONIAN MISCELLANEOUS COLLECTIONS [VoL. 45 in the curve. But this area is not strictly proportional to the total solar radiation at the earth’s surface, as determined by actinometer observations, for the reason that the radiation has been unequally reduced at different wave-lengths by losses at the siderostat mirror, within the spectroscope, and by selective absorption at the bolometer itself. It is necessary to correct the curve so that it shall as accu- rately as possible represent the distribution of energy in the solar beam prior to these losses. Inasmuch as the coefficient of total absorption of the lampblacked bolometer strip is upward of 95 percent, it is believed that no considerable error is admitted by neglecting its differences of absorption for different wave-lengths, and no correction is applied for this. The relatrve absorption of the spectroscope for different wave-lengths is frequently determined, and that of the siderostat mirror still more frequently, for in both these optical parts of the apparatus there is rapid deterioration of the reflecting power of the silvered glass surfaces. At present this indeed forms one of the main difficulties and sources of error of the investigation, for a whole day of observing and several days of computing are required for each determination of the absorption of the apparatus, which would be determined once and for all if con- stant reflecting surfaces could be employed. With the coefficients of absorption of the apparatus thus deter- mined, each small area included under the bolographic curve for a very narrow range of wave-lengths is increased so that the total cor- rected area is then proportional to the solar radiation at the earth’s surface as measured with the actinometer or pyrheliometer. Then by the aid of formula 1, given above, and employing the trans- mission coefficient a determined from the series of bolographs of the day, each small area is again corrected till it becomes proportional to the total radiation of that wave-length outside the atmosphere. The ratio of the sum of these finally corrected areas to the total corrected area at the earth’s surface is the factor by which the re- duced pyrheliometer reading is to be multiplied to give the “ solar constant ” so-called. It is evident that these values depend directly upon the pyrhelio- meter or actinometer readings for their accuracy, so that these in- struments become here of major importance. In the work thus far a mercury pyrheliometer has been used as the primary standard, and the daily observations have been taken sometimes with it, sometimes with a Crova alcohol actinometer (specially constructed for the Institution under M. Crova’s valued supervision), and sometimes with both instruments simultaneously. It has been shown by re- : : { a : yyy ory SS ee ea ee eee eee re 7 es : ~ a a Ss S52 paced teen oe : a | ve 7 aan x #t + iH 2 j . r ; # pape SB bi ; | | | ; ; Pf } . Ve ‘ tat ae engine fe ee 0 en eect | apo erg . ee Me Se eee 7 eee Rae ee iited tre aes Leer ee hi | % ¥ i - | ah Po : P= a= ‘AAA HdMSOWLVY SHLYVA AHL ACISLNO WNALOAdS UVIOS TVWUON AHL NI NOILVIGVA FO NOILNAIALSIa ‘HLONIT FAV : es fe WO? 61 8 ZI Pik SI bi) &/ ll 11 7701 60 80 Z0 90 s0 v0 €0 y242| LO06/ Gh Avensgay 4o suc LUKKS el av AO) AY SNOILOATIOD SNOANVTISOSIJY NVINOSHLIWS NOILVIOVSY 4O ALISNFLN] ABBOT | STUDIES OF SOLAR CONSTANT OF RADIATION 81 peated comparisons of the two instruments and by comparisons of the pyrheliometer with another type that they give proportional results under widely differing conditions of wind and temperature, so that I have no question of the relative accuracy of the actinometric data employed in computing values of the solar constant within two percent. There is, on the other hand, room for question as to the absolute magnitudes of the values given, for these depend on the constants and the theory of the mercury pyrheliometer. Steps are being taken to get.further checks on this matter, and in a later pub- lication it is expected to recompute the data in accord with later information. For the present then, the values in the following table are to be held as relatively accurate and consistent among them- selves, but subject later to correction by a common multiplying factor. Taste II. Values of the Solar Constant of Radiation. From Bolographic Studies | | Calories per Sere Centimeter | Solar Constant | | | Jinute. ages Date. ecaeenet Air Mass. | = se SS | Mon ib tance | . | At the Earth’s | Outside the | of the Sun Surface. | Atmosphere. | : | hm | Cal. Cal. | 1902. Oct. 9g | Q 1.425 | 1.42 | 2.20 | 2.19 cS Go Ayes MW fy Bir, 1.624 L.44 2.21 2.19 sf 22a QvOn 2.415 1.20 | 2.18 | 2.16 ROogn web: -L9) |) ior 1.642 1.35 | 2. 34 2.28 “ coe ge he! a) oD 2.003 1.20 | 2.31 | 2:25 fo Mat. 3.4.0 59 1.429 1.34 ono | 2.26 “s <2 Oh | 1.454 1.19 2.20 2.27 Be co) 26 £7 | 1.438 1.16 2e1t | 2.10 ee commas DGG. 1.754 | 1.05 2.09 | 2.07 Apia ety) 25 45 1.463 1.19 1.97 |. 1-99 “¢ See On eel OT 1.145 ¥.29 E23 2.27 ce £0 29 2 A). | 1.308 1.05 1.93 1.97 | | General Mean. | 2.167 | | Mean of results prior to | | | March 26, 1903. | 2.229 | Mean of results after | March 26, 1903. | 2.080 The bolographs used in the computations extend from wave- length 0.374 to wave-length 2.54 with the exception of those of October, 1902, which reached only to a wave-length of 0.48y in the violet. For these latter bolographs a correction of about twelve per- cent was applied, founded on the later work, and thus the results for October, 1902, are entitled to slightly less weight on this account. All the areas have been extrapolated for the radiations lying outside at both ends of the region 0.37 to 2.54, but the corrections so applied amount to less than one percent altogether. Their magnitude was 82 SMITHSONIAN MISCELLANEOUS. COLLECTIONS [VoL. 45 determined by an inspection of the rate of decrease of successive cor- rected areas approaching the limits of the curves, and the correc- tions were checked both by computing according to Wien’s formula the probable form of the solar energy curve corresponding to the assumed solar temperature of 6000°, and by examination of the normal energy curves outside the atmosphere as computed from bolographs and given in plate XXII. I have thought it worth while to give in addition to the general mean, the means also of observations before and after March 26, when, for some unexplained reason, a fall of about 10 percent was noted in the computed solar constant. The observations of February 19,1 March 25, March 26, and April 29, 1903, appear to be entitled to the greatest weight among those given, on account of the regularity of the actinometric curves of those days and the closeness with which the plotted points for determining the atmospheric transmission coefficients lie upon straight lines, as shown for two of the days in question on plate xx. Since May 1 it has been almost impossible to get sufficient observations for computing a solar constant owing to cloudiness, but interest attaches to further determinations and these are to be made when practicable. FORM OF THE NORMAL SOLAR ENERGY SPECTRUM OUTSIDE THE EARTH’S ATMOSPHERE AND THE PROBABLE TEM- PERATURE OF THE SUN The reader has no doubt noted that, by applying corrections for atmospheric and instrumental absorption, the bolographic spectrum energy curves may be reduced in form as well as in area to represent the distribution of energy in the spectruni of the solar beam outside the atmosphere. This has been done in several instances, and in doing so the curves have been transformed from the prismatic to the normal wave-length scale by taking account of the prismatic dispersion, and several of these curves are platted in plate xxu. No account is taken in the curves, shown in plate xx1t, of selective absorption bands whether solar or terrestrial, smoothed curves only being given. It will be noted that there is a fair agreement in general form between these independently derived curves, and that they unite in ‘February 19, 1903, was the most extraordinary day as regards absence of water vapor in the atmosphere which has ever been noted here. The great water-vapor bands 2 in the infra-red spectrum were nearly filled up, and the long wave-length side of the band 2 presented an almost unrecognizable appearance. ABBOT | STUDIES OF SOLAR CONSTANT OF RADIATION 83 fixing the wave-length of maximum energy at about 0.49." Their agreement would be more exact, there can be little doubt, if it were not for the large and variable absorption of the silvered surfaces in the optical apparatus for wave-lengths at and beyond the region of maximum energy. The transmission of the spectroscope at a wave- length of 0.45 has varied on this account at different times from 33 percent to 15 percent, whereas at wave-lengths of 1 and there- abouts the transmission always approaches 90 percent. The spectro- scope mirrors are resilvered about once in two months and the siderostat mirrors still oftener. Paschen has empirically derived a law connecting temperature with wave-length of maximum radiation, which is expressed as fol- lows, where T is the absolute temperature and 4, the wave-length of maximum intensity of radiation expressed in microns: AZ, f= constant. The value of this constant for the radiation of a “ black body” or perfect radiator as determined by Paschen,? Lummer and Prings- heim,® and others is about 2900, while for bright platinum Lummer and Pringsheim give 2630 with values for other substances inter- mediate between these. Taking the higher value in connection with the observed position of maximum in the solar energy curve outside the atmosphere, we find that the sun’s radiation may be assumed comparable as regards the wave-length of maximum radiation to the emission of a “ black body ” at 5920° absolute. Readers will draw their own conclusions as to the probability that the solar temperature actually lies near this value, but it may be remarked that a further correction of the energy spectrum curve for the selective absorption of the solar envelope would undoubtedly reduce the wave-length of maximum radiation still further, and would thus incline us to the view that the interior of the sun is at a higher temperature than the above con- siderations alone would indicate. _14The wave-length of maximum energy determined by Mr. Langley on Mount Whitney was about 0.52. *Verhandlungen d. Deutschen Phys. Ges., U1, 37, 1901. ’ Paschen, Astrophysical Journal, 1x, 306, 1899. THE NEW CCELOSTAT AND HORIZONTAL, TELE- SCOPE OF THE ASTROPHYSICAL @CESERV ATO OF THE SMITHSONIAN INSTI Uation By GG SAP bon INTRODUCTION ‘About two years ago certain observations on the absorption of the solar envelope were begun at the Astrophysical Observatory of the Institution under Mr. Langley’s direction. For these experi- ments a solar image 40 centimeters in diameter was formed at the slit of the spectro-bolometer, and energy spectra were obtained of the radiation from the center of the disk and at points near the limb. As it is clear that the radiations coming from points more and more remote from the center must traverse greater and greater thicknesses of the solar envelope, this method of investigation offers a means of finding out something about the amount and quality of its absorption. It was hoped to determine also the nature of the energy spectrum of sun spots, but the arrangements then used proved inadequate to give sufficient steadiness and distinctness of the solar image even for satisfactory measures of the absorption of the solar envelope, to say nothing of measures on sun-spot spectra. The apparatus used at that time for forming the solar image con- sisted of a large Grubb siderostat of the Foucault type giving a hori- zontal southerly directed beam, a nine-inch concave mirror of 2.30 meters focus, and a small convex mirror of about a meter radius of curvature placed inside the focus of the concave mirror. These latter two mirrors both had to be out of axis in order to form the image at one side of the concave mirror, and both were within the observatory, so that the beam from the siderostat passed through an aperture in the wall to reach them. With these crude arrange- ments it is not surprising that the solar image was in very bad and changeable focus, and subject to continual and excessive disturbance by “boiling,” tremor, and bad following, besides being subject to the rotation of the field necessarily accompanying the use of the Foucault form of siderostat. A trial of a combination of two concave mirrors of 2.30 meters and 84 i ABBOT | NEW CCELOSTAT AND. HORIZONTAL TELESCOPE 85 4.70 meters focus respectively, in connection with a ccelostat and second reflection from a plane mirror east of the polar axis of the ceelostat, while doing away with rotation of the field, and with the chief defects in following, still furnished a very poor image, subject to changes of focus of ten feet between full and cloudy sunlight, and attended with serious “ boiling,’ which was not at all reduced by providing a canvas tube for the beam, but which did diminish when the sky was very thick with cirrus clouds or haze. It was thought best to arrange to cast the image to its full size of forty centimeters by means of a single concave mirror, but before ordering this, Mr. Langley, following his well-approved policy of trying a new thing on a small scale, first procured a five-inch mirror of forty feet focus, and directed that this should be set up for a pre- liminary trial. Recognizing that the “boiling” (or fluttering con- fusion of all parts of the image due to variability of the strata of air traversed by the beam) would probably prove the main obstacle to forming a well-defined solar image, he devised and directed a novel experiment of “ churning ” the column of air traversed by the beam. This experiment was made with a very satisfactory result as de- scribed by Mr. Langley in his article on Good Seeing,! and with more detail as regards apparatus in the appendix to the administra- tive report of the Secretary of the Smithsonian Institution for the year ending June 30, 1902. In brief it appeared that “ boiling” caused within the telescope tube itself could be entirely removed by churning the air, and that for high sun a churned tube pointing toward the sun and reaching forty feet above the surface of the ground sufficed to overcome the main portion of the prejudicial disturbances of the air, so that the remaining “ boiling ” of the solar image did comparatively little harm to the definition. It was observed with the forty-foot focus telescope that the vigorous churning of the air seemed to decrease those changes of focus with varying cloudiness which had been noted in the earlier work, and which were observed also with the forty-foot focus instrument when the stirring apparatus was stopped. Among the incidental advantages of the stirring may also be included the more rapid convection of the heat of absorption of the solar beam at the mirror surfaces, and consequent diminution of the alteration of figure which is always caused by unequal heating of glass mirrors. In the preliminary tests it was found that the stirring apparatus communicated vibrations to the ground sufficiently serious to pro- duce a prejudicial tremor of the image, but it was noted that the pass- 9 * American Journal of Science, February, 1903. 86 SMITHSONIAN MISCELLANEOUS COLLECTIONS [voL. 45 ing of heavy wagons produced almost as great, though intermittent, tremor, so that it appeared that very good piers would be necessary even if no stirring was contemplated. LONG FOCUS HORIZONTAL TELESCOPE A mirror of 50 centimeters aperture and 40 meters focus was ordered from the J. A. Brashear Company, and it was decided to instal this as a horizontal telescope lying in a north and south direc- tion, to be fed by a coelostat or siderostat as seemed most desirable. It was deemed necessary to provide heavy and deep-founded piers for the ccelostat and image-forming mirror, and although it was not indispensable for the work immediately in view that the pier where the image was formed should be as free from tremor as the others, it was deemed best to make the three piers alike, to provide for possi- ble contingencies. For each pier a pit was dug twelve feet square and ten feet deep, and the four sides of the pit were supported by walls of pebble and cement one foot thick. At the bottom of the cubical cavity remaining was filled in twenty-four inches of sand, and on this a twenty four-inch base of pebble and cement with a clear space of six inches all round between it and the walls of the pit. On this base was erected to the surface of the ground a hollow brick pier, with eighteen-inch walls on four sides, and a thirteen-inch wall across the center. A cap-stone eight feet long, seven feet wide, and seven inches thick completed each pier, except that for the coelostat (finally decided upon) a superstructure of brick was provided. THE C@LOSTAT The ccelostat consists essentially, as is well known, of a plane mir- ror fixed parallel to a polar axis which turns uniformly at the rate of one rotation in forty-eight hours. An instrument of this kind was described and illustrated in an article by Von Littrow* in 1863, but he ascribes the principle to August, who appears to have discovy- ered it about thirty years earlier. It is surprising that so simple and excellent a device did not come into general use sooner, but though rediscovered and employed for a time about 1880 by Mr. Langley at Allegheny, and very likely also by others, it has been only within the last five years that the coelostat has become generally known and used. Its chief merit, besides simplicity and consequent accuracy of driving, is the fact that the whole field of view remains fixed, whereas in other forms of siderostats and heliostats one point only of the re- +Wein. ber., XLVI, Il, pp. 337-348, 1863. ABBOT] NEW C@LOSTAT AND HORIZONTAL TELESCOPE 87 flected beam is fixed, and about this point all other parts of the field revolve, either with uniform or (as in the Foucault form) with a velocity variable according to the position of the heavenly body. To counterbalance these advantages the ccelostat does not in its simplest form send the beam in a fixed direction independent of the declina- tion, but when used, as is customary, to give a horizontal beam, this beam deviates toward the north of an east and west line for objects south of the celestial equator and vice versa. Furthermore, when sending a horizontal beam nearly eastward horizontally, it is clear that the mirror is employed at a very unfavorable angle for objects near the western horizon. It had been proposed here to get over these two disadvantages by the use of a second plane mirror, itself mounted so as to permit of moving it upon a U-shaped track with north and south branches close to and on the east and west respectively of the polar axis carrying the first or rotating mirror, which latter was intended to cast its beam nearly horizontally to the east in the morning and toward the west in the afternoon. This device was tried, but there were serious objections to it, the chief one being that for objects far south and near the meridian (like the sun at noon in December) the cross-sec- tion of the reflected beam was very small compared with the aper- ture of the first mirror. IMPROVED FORM OF TWO-MIRROR CCELOSTAT Fortunately a better device for solar work at this latitude was then thought of. This consists in reflecting the beam due south from the rotating mirror and thence due north from the second mirror over the top of the first. The beam from the first mirror shoots upward at an angle with the vertical equal to the sum of the angles of latitude and declination ; and for the sun at Washington this angle is about 62° at summer solstice and 16° at winter solstice. There- fore to give a horizontal northerly directed beam the second mirror is to be inclined forward 14° at the former period and 37° at the latter.t In the following table is given the diameter of mirrors for this form of ccelostat and for other instruments to furnish a fixed horizontal north and south solar beam at the latitude of Washing- ‘If it was necessary to incline the second mirror forward still more, a special support would be required; but it seems unquestionable that the Ritchey support system (Astrophysical Journal, v, 143, 1897), modified only so far as that the mirror is stuck by cement or by exhaustion of the air to the numerous balanced supporting pieces, would be quite as efficient here as for the usual case of a mirror reclining face upward. 88 SMITHSONIAN MISCELLANEOUS COLLECTIONS [voL. 45 ton and for different times of the year and day. It will be seen that in economy of mirror surface the ccelostat thus arranged has the ad- vantage. In this form of ccelostat the moving mirror is never used in very different positions, so that owing to the consequent probable constancy of figure in the mirror it seems to be well suited to long exposures in stellar photography. Comparative Sizes of Mirrors Required to Furnish a 20-inch Horizontal North and South Beam from the Sun at Washington. (Latitude 38° 53/.) | Two Mirror Ceelostat. | eo ae south. Second Now, | Miwor Reflects | Foucault Single : - lrror S1idero- Day. Hoge ._| North. Second South. sahGeah Diameter Diameter Diameter | Diameter Beam South. First Second First Second Mirror. Mirror. Mirror. | Mirror. June 21 7 A. M. 27.4 in. | 2o.6in.| 23.8dn.| -21.2 in. 30 in, g A. M. 23-6 | 20.6 23.8 21,2 27 12M. 21.8 20.6 2318) 0 ee li2e eal 25.2 Sept. 21 7 A. M. 25.2 | 22.2 uzde2 epee 2 26.2 9 A. M. 21.6 2er2 are See 21.2 23.4 12 M. 20.0 22.2 | ¥23"2) 21.2 22:2 Dec. 21 7A.M. | 27.4 | 25.6 30:4 wl eee 2 23.4 QAM Na eaeos el 25-0 |, Sommenele 2l.8 21.6 12) Boe Vie | ea2ie 25.6 36.4 21.2 20.6 It will be noted that at noon of the equinoxes the second mirror if exactly south of the first would cut off the beam, and that at sum- mer solstice it must be further south than in winter to reflect the beam clear over the first mirror.. Accordingly the second mirror is provided with a carriage and two pairs of tracks at right angles like the slide rest of a lathe, so that the mirror may be displaced to the west a little before noon when the sun is in its southern declina- tions, and can be shifted back to the east a little after noon.1 The north and south track is for the purpose of shifting the second mir- ror for different declinations of the sun, the mirror being at the south end of the track at the summer solstice. As it has been thought that this solution ef the difficulties attend- ing the use of the ccelostat will prove of interest and value to astron- omers, a large instrument of this type was ordered from the J. A. Brashear Company for exhibition at the Louisiana Purchase Expo- sition to be held next year at St. Louis. Plate xxzir is from a photo- graph of this ccelostat as now being tested at the Astrophysical Ob- "With a small instrument of this type it might be more desirable to move the first mirror and driving mechanism on an east and west track for this purpose, but this is not provided for in the large ccelostat about to be described. SMITHSONIAN MISCELLANEOUS COLLECTIONS MOL Ass else THE LARGE CCLOSTAT WITH SECOND MIRROR, SMITHSONIAN ASTROPHYSICAL OBSERVATORY. H 8 a ee « » < pede eS 60 pn, ie ae rr * - 7 ) y : _, i Se a i ee % - oot a ; a —- ar) ae is] P i 6.0 2 ig ne Z 7 ef) Zz ; . { Be) pe ere my © i eA if a *, cay o® eg ae 2, ae ee i —_ ! : —~ = 3 : 5 the 7 at > P oar p - Ao - & 5 ma i Z - is -" °® a ! * 9 5 ove 5 a [= a 7 i i as (o> 2. > 7 ¥ ’ be - ~ “ ; _ ra & ‘ 7 a - . a ¥ a 7 a> ’ - ae ? - ; , seu — > 2 , © m4 ‘ > a bo , ee . bis OP wh om a e . > % - = - f a - ‘ ss 7 , x 7 ’ F a - “_ S - - bd : cfr Ps : ) ae a a - a , } ‘ f . oy . : - f ™ 7 i, 4 : = a. - 4 ’ bs : ‘ * x ” c L 4 * I as i oe i - i - i = ' a " ; % — 1 —; é is + i 1! ee e > = ap ABBOT | NEW CCLOSTAT AND HORIZONTAL TELESCOPE 89 servatory, in connection with the long-focus mirror above men- tioned. There is also shown in the illustration a portion of the “churned ” tube of the horizontal telescope, of which more will be said later. The ccelostat carries a thirty-inch and a twenty-five-inch mirror, the former turned by a polar axis driven at the rate of one complete rotation in forty-eight hours, the latter mounted on a car- riage with traverse motions at right angles like the slide rest of a lathe. The cell of the second mirror is carried by trunnions in a fork itself capable of turning about a horizontal north and south axis, and by these two motions of rotation, with their fine adjustments, the beam may be sent in any direction whatever, though most favorably in a nearly northerly one. In actual use the reflected beam is de- pressed about 6° from the horizontal to feed the long-focus mirror,* which is 55 feet north and about 3% feet below the center of the first mirror of the ccelostat, directly under which the beam passes toward a focus on the third pier, some 85 feet further south. To provide for this depression of the beam from the horizontal, the north and south, or declination, track of the ccelostat is inclined upward at a corre- sponding angle, so that the reflected beam may always clear the first mirror. The length of travel of the lower base of the second mirror on this north and south track is five feet and the lower base itself has an east and west track six feet long on which the upper casting is moved to and fro to allow for avoiding the shading of the main coelostat mirror by the cell of the twenty-five-inch mirror between 11 o'clock and 1 o’clock near the times of the equinoxes. THE TUBE AND STIRRING DEVICE Early experiments on an artificial star with the long-focus mirror, before the completion of the ccelostat or the installation of a tube, showed conclusively that the “ boiling ” caused by irregularities of the atmosphere over the grass-grown soil between the mirror and its focus was far too great to permit anything like satisfactory defini- tion on the solar image, and therefore the novel device of a tube with provision for stirring the air by means of a blast was ordered. It consists of a main horizontal tube 24 inches in in- ternal diameter with diaphragms at five-foot intervals, and with *If the telescope was intended for stellar or lunar photography it would not be allowable to use the concave mirror quite so far out of axis as the 3° thus required; but the deformation of images from this cause, as computed by the formula of Poor (Astrophysical Journal, vu, 120, 1898) for the long-focus mirror is only about 0”.25 of arc, which is inappreciable compared with other defects of solar definition. go SMITHSONIAN MISCELLANEOUS COLLECTIONS [voL. 45 an inclined flared tube uniting with the main tube at the north end close in front of the concave mirror. At intervals of about five feet, five-inch ducts lead to air-mains 14 inches in diameter, which in turn at length unite in two twenty-inch mains leading to the intake and blast respectively of a twenty-nine-inch fan blower with direct connected 2% horsepower electric motor. It is so arranged that the openings in the telescope tube communicate with the blast and suc- tion of the blower alternately, so that the air within the tube is re- peatedly carried through the system and churned over and over. Thus the path of the beam from the ccelostat to the focus of the mir- rors is thoroughly stirred, but nothing has been done as yet to intro- duce stirring between the ccelostat and the sun. It is possible that an attempt will be made later to stir the path of the beam in the eighty feet immediately above the ccelostat, if it is found impossible to get good enough definition with the present arrangements. It should be recalled that the conditions required for bolometric work are quite different from those suited to direct eye vision or to photography. Bolometric studies require unchanging transparency of the air, else difference in the galvanometer deflection may be due to alterations in transparency of the intervening medium and not to the properties of the source of light. Thus those times when thin cirrus clouds, fog, or smoke cover the sun, which are well known by solar observers to be the times when “ boiling” is apt to be diminished, and which are the most favorable opportunities for visual and pho- tographic observations, are quite unsuitable for bolometric work. Indeed the best time for this is somewhat after noon on those clear October days when “ boiling ” is apt to be at a maximum, but cloud- iness at a minimum, and it is probable that the definition obtained in such conditions will never be the best. Trials made thus far have demonstrated the great value of the stirring apparatus, not only to diminish “ boiling,” but to preserve a constant focal length and tolerable definition. “ Boiling” is still of course noticeable, because the long reach of air above the coelostat is not stirred, but the image is far better than could be ob- tained with the earlier appliances, and owing to the massive piers and to the simplicity of driving mechanism it is less subject to jars and wandering. In later communications it is expected to describe bolometric work upon the solar image formed by the great horizontal telescope. ON SOME PHOTOGRAPHS OF LIVING FINBACK WHALES FROM NEWFOUNDLAND By FREDERICK W. TRUE It is only within recent years that works upon cetaceans have been illustrated with reproductions of photographs from nature. Earlier writers had to content themselves with drawings, and as these were quite commonly the work of unskilled hands, they were often extremely inaccurate, or even positively worthless as illustrations, From lack of knowledge the artist was usually unable to interpret the form of the various parts of the animal before him and conse- quently introduced features and combinations which had no counter- part in nature. As photography improved, opportunities were taken to obtain photographs of skeletons, of dead whales lying on the beach, etc., upon the study of which conclusions could be based without great risk of error. All the earlier photographic pictures of whales, how- ever, represented dead animals, and not unfrequently such as were in a more or less advanced stage of decomposition, whereby a life- like appearance was entirely lost. Photographic representations of living whales were still, therefore, a desideratum. Within the last four or five years some such photographs have been obtained, and the purpose at this time is to describe those which the writer made in Newfoundland in 1899. The only others with which I am ac- quainted are those taken by Dr. Racovitza and Dr. Cook in the Ant- artic Ocean in 1898 and published a few months ago among the results of the voyage of the Belgica. My own photographs are, I believe, the only ones representing living whales in American waters thus far published. They were taken from the bow of the whaling steamer Cabot, be- longing to the Cabot Steam Whaling Company, while engaged in chasing whales in the unquiet waters of Notre Dame Bay, New- foundland. They all represent the common finback, Balenoptera physalus (L.), and all the individuals were in motion. Mr. Aldrich, writing of the Arctic right whale, very justly re- marks: “ It is disappointing to see a whale, for most pictures repre- sent him as standing up like a buoy, or posing on his tail on the top gi 92 SMITHSONIAN MISCELLANEOUS COLLECTIONS [VoL. 45 of the water. The real fact is that only the top of the head and a small piece of the back are seen and perhaps the ‘ flukes,’ or, in common English, the tail, may take an occasional sweep in the air.” It is much the same with the Newfoundland finbacks. Seen alive in their native element they presented to the eye only a slight hint of that graceful form which was revealed when they were drawn out on the slip at the whaling station. Still, the part seen was that of a living animal, and therefore was possessed of a separate interest from the dead specimens on the slip. While the photographs show little of the whale’s form, they do represent with accuracy their ap- pearance in their natural environment, and also give some idea of their actions and attitudes while swimming. Among whales, as among other animals, some actions and move- ments are habitual and characteristic, while others are unusual and are repeated only at long intervals or under peculiar circumstances. Nearly all observers who have had frequent opportunities of ob- serving whales agree that many species, such as the humpbacks, indulge at times in strange antics, such as leaping entirely out of the water, rolling from side to side, taking a vertical position, with the head up or the tail up, as the case may be. While the New- foundland finbacks may, and probably do, engage in such perform- ances, nothing of the kind occurred during my observation of them. They appeared singly or by twos or threes, spouting at irregular intervals, but pursuing a quite regular course in a definite direction for a considerable distance. They came up to the surface obliquely to spout, and the top of the head became visible, but was not ele- vated perceptibly above the waves. Then the head sank down, the back came into view gradually from the shoulders backward, exhibit- ing a strong curvature, and finally the dorsal fin appeared, but little or nothing beyond. The flukes were not to be seen in any instance, nor the pectorals, nor the eye, nor any of the underparts. The photographs show a number of details not noted at the time they were taken. Indeed, the difficulty of getting the picture itself is so great that one’s faculties are entirely absorbed in the proceeding and there is little opportunity for observing particulars. The pitch- ing and rolling of the steamer in the restless waters is very discon- certing, and not less so the fact that the point at which the whale will appear is uncertain and the length of time it will remain in view very brief. I will now describe the several photographs which are here reproduced. Others were taken, but are less satisfactory or less characteristic. Plate xxiv, I, shows a very characteristic appearance of a finback SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL 45, PL. XXIV 1. Preparing to descend. 2. Just reaching the surface. LIVING FINBACK WHALES. (Balenoptera phisalus L.) SmitHsoniaNn MIscELLANEOUS COLLECTIONS VOL 45, PL. XXV 1. After spouting. 2. Similar attitude, nearer view. 3 Posterior view. LIVING FINBACK WHALES. (Balenoptera physalus L.) yUS COLLECTIONS VOL. 45, PL XXVI SMITHSONIAN 1 Ready to descend. 2 Process of descending—head under water, back arched. LIVING FINBACK WHALES. (Balenoptera physalus L.) TRUE] - PHOTOGRAPHS OF LIVING FINBACK WHALES 93 as it courses along at the surface. The whale has spouted and is pre- paring to descend. It will be noted that the entire head is below water ; the back is slightly arched and the dorsal fin is distinctly in view above the surface, though the water is breaking over it in front. The flukes are invisible and make no disturbance of the surface of the water. Nothing is seen of the pectoral fins. Plate xxIV, 2, represents a finback that has just come to the surface to spout, or has just completed that act. The top of the head is out of water and slightly inclined upward. The blowholes are seen as a dark eminence, and all about the head is a white rim of foam. There is an appearance on the right side as of the mandible projecting laterally beyond the upper jaw, but this is probably an illusion due to the waves. In plate xxv, 1, is shown a finback in a similar position, but there is no doubt in this case that the whale has spouted. Its vapory breath forms part of the haze at the left of the head. The whole upper surface of the body from about the middle of the head to the dorsal fin is above water and the blowholes are distinctly marked by a dark eminence near the left end. The dorsal fin is not visible, but would have soon appeared. A very remarkable feature of this view is the great height of the eminences at the sides of the blowholes. The apertures themselves are situated between elevations, but in the dead whales.on the slip these eminences present no such apparent height ‘as here shown. This photograph and the next lend some support to the view advanced by Buchet and Racovitza that the whales pro- ject the blowholes outward when spouting. Racovitza’s photographs, however, while admirable in other respects, and extremely interest- ing, do not show the region of the blowholes distinctly enough to throw much light on the point in question, and his sketches are rather unintelligible, and in some cases (e. g., plate 3, figs. 14, I5, etc.) cer- tainly incorrect. It would appear, at all events, that the eminences at the sides of the blowholes are raised when the whale is spouting, rather than the blowholes themselves. Plate xxv, 2, shows a finback in an attitude similar to the last, but the animal is nearer. In this the ridges on the sides of the blow- holes are extremely prominent and clearly defined. The anterior end of the head is hidden, but the line of the back is visible to the dorsal fin, over which the waves are breaking. Even beyond the fin the dorsal line is to be seen, but is not well defined. Plate xxv, 3, is a rather indistinct view of one of these finbacks from the posterior end. It shows the great breadth of the back. The blowholes appear as black spots at the interior end of the figure, and the dorsal fin, with foam about it, at the posterior end. 94 SMITHSONIAN MISCELLANEOUS COLLECTIONS [VOL. -45. In plate xxv1, I, is shown a finback ready to descend. The head has already disappeared, the back is quite strongly arched, and the dorsal fin is very distinct and entirely above water. As in the other views nothing is seen of the flukes. The finback shown in plate xxv, 2, has the head much farther down in the water and the back very strongly arched. The dorsal fin is visible, but is partially submerged. Here again the flukes are invisible. These photographs and the notes, which I made while on the whal- ing steamer, are in agreement with the observations of Packard, Scammon, Pechuel, Cocks, Balfour, and Rawitz on the same species and its close ally (which may, indeed, be identical with it) in the Pacific. All agree that under ordinary circumstances the finback rises and sounds obliquely, that the flukes are not thrown out, that the spout is vertical, and that the actions of the animal as regards. the length of time it remains below the surface, the distance it travels while submerged, and the number of times it spouts in succession, are irregular. Pechuel held that the spout was double, but my observations agree with those of Packard, Rawitz, and Racovitza, that the spout is single in the finbacks. The vapor-laden breath in all whalebone whales escapes, of course, from two separate apertures, but in the common finback, at least, the two columns unite so close to the head that they appear as one. WOdSON TVNOILVN ‘S “1 AHL NI SINNOYAdSAH AO NOLYIAMS IIAXX ‘Td “S¥ “IO A SNOILIATIOND SNOANVIIAOSIP, NVINOSHLING ene tON: OF HESPERORNIS By -PREDERIC Ay LUCAS The announcement by Professor Marsh of the discovery of birds with teeth in the chalk-beds of western Kansas was not only im- portant from a scientific standpoint, but aroused much popular in- terest. Unfortunately, toothed birds are not merely rare but usually are in a more or less imperfect condition ; the teeth, too, are so small and so few remain in the jaws that a specimen of a toothed bird is apt to prove a disappointment to the few who see them. In this last particular the example of Hesperornis regalis (plate xxvit), which formed a portion of the exhibit of the U. S. National Museum at the Buffalo Exposition, is no exception to the general rule; it is, however, one of the most complete specimens yet discovered, being the first sufficiently well preserved to admit of its being mounted. The mounting of this specimen revealed the fact that in the position of its legs Hesperornis was not only different from any modern bird, but different from all other birds. In ordinary waterfowl, such as ducks and geese, the legs, when swimming, are beneath the body, and this is also the case in such highly specialized divers as loons and grebes, in which the legs are placed far back. But in Hesperor- mis the articulations of the leg-bones were such as to show that in swimming the legs must have stood out, almost at right angles to the body, suggestive of a pair of oars. This also suggests that the legs, like oars, may have been moved together and not alternately, since an alternate motion of the legs would have had a tendency to throw the body from side to side. It is a little difficult to see just what advantage could be derived from such a method of swimming unless it was for rapid movement at the surface. This peculiar position of the legs was not found in the Cretaceous diver Baptornis, also from western Kansas, in which the legs were situated as is customarily the case. The present specimen of Hesperornis also showed that in the structure of its shoulder girdle the Cretaceous bird was more like modern birds than heretofore supposed. These details will be found described in the Proceedings of the U. S. National Museum for 1903. PP- 545-552. 95 A NEW) PEESIOSAUR By FREDERIC Ay LUCAS Among the specimens included in the Marsh collection was a. fine example of a plesiosaur which has recently been described by Dr. S. W. Williston* under the name of Brachanchenias lucasi. The specimen, which lies on its back, comprises the skull and jaws, with thirty-five consecutive vertebrz (plate xxvii). The upper por- tions of the skull and vertebrzee were unfortunately weathered away Lefore the discovery of the animal, which was found near Delphos, Ottawa county, Kansas. While the popular idea of a plesiosaur. derived from the graphic descriptions of English writers, is that of a reptile with a long, snake-like neck, yet many short-necked animals are included under that term. The present individual enjoys the distinction of being the shortest necked species yet discovered, and this, coupled with the massive head, causes the specimen to suggest a crocodile, the more that the large swimming paddles were unfor- tunately not preserved, having been washed away before the speci- men became entombed in the deposits forming the Fort Benton limestone. Dr. Williston calls attention to the fact that while plesiosaurs are not at all uncommon in the Cretaceous deposits of North America, they are for the most part represented by detached bones, or at the best isolated, if well-preserved paddles. So while thirty-two species and fifteen genera have been described from the United States, in not a single one has any considerable portion of the ‘skeleton been preserved, aside from those that have been described by Dr. Williston himself, and the skull is known in but three in- stances. As the present example shows the bones of the under side of the skull very clearly it is of special importance. *“North American Plesiosaurs,” part 1; Field Columbian Museum, Publication 73, Geological Series, vol. u, No. 1; Chicago, April, 1903, p. 57. SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 45, PL. XXVIII TYPE OF BRACHANCHENIUS LUCASI WILLISTON. INCOMPLETE SKELETON OF PLESIOSAUR IN THE U. S. NATIONAL MUSEUM. SMITHSONIAN MISCELLANEOUS COLLECTIONS VoL 45, PL XXIX SHELL GORGET WITH ENGRAVED FIGURE OF A DISCUS THROWER, FROM AN ANCIENT GRAVE NEAR EDDYVILLE, KENTUCKY. (Diameter, 5 inches.) SHELL ORNAMENTS FROM KENTUCKY AND MEXICO By W. HH. HOLMES Among the many interesting relics obtained from mounds and burial places in the Mississippi Valley are the engraved shell gorgets, a number of which are now preserved in our museums. The most recent addition to this class of objects was obtained by the National Museum from Mr. C. A. Nelson of Eddyville, Lyon County, Ken- tucky, and comes from a burial place encountered in opening a stone- quarry near Eddyville. It is a symmetric saucer-shaped gorget (plate xxrx) five inches in diameter and made apparently from the expanded lip of a conch shell (Busycon perversum). It is unusually well preserved, both faces retaining something of the original high polish of the ornament. Two perforations placed near the margin served as a means of suspension. The back or convex side is quite plain, while the face is occupied by the engraving of a human figure which extends entirely across the disk. It will be seen by reference to the illustration that this figure is practically identical in many respects with others already published.t It is executed in firmly incised lines and is partially inclosed by a border of nine concentric lines. The position of the figure is that of a discus thrower. The right hand holds a discoidal object, the arm being thrown back as if in the act of casting the disk. The left hand extends outward to the margin of the shell and firmly grasps a wand-like object having plumes attached at the upper end, the lower end being peculiarly marked, and bent inward across the border lines. The face is turned to the left; the right knee is bent and rests on the ground, while the left foot is set forward as it would be in the act of casting the disk. The features are boldly outlined; the eye is diamond shaped, as is usual in the delineations of this character in the mound region. A crest or crown representing the hair surmounts the head; the lower lobe of the ear contains a disk from which falls a long pendent orna- ment, and three lines representing paint or tattoo marks extend across the cheek from the ear to the mouth. A bead necklace hangs down over the chest and the legs and arms have encircling ornaments. The lower part of the body is covered with an apron-like garment attached to the waistband, and over this hangs what appears to be a *Holmes in Second Annual Report Bureau of Ethnology, pl. 1xxiii. 97 98 SMITHSONIAN MISCELLANEOUS COLLECTIONS [VoL. 45 pouch with pendent ornaments. The moccasins are of the usual Indian type and are well delineated. A study of this figure strongly suggests the idea that it must represent a disk thrower engaged, pos- sibly, in playing the well-known game of chungkee. Reference has occasionally been made to more or less well-defined analogies existing between the shell gorget engravings of the Missis- sippi Valley and similar designs from Mexican engraved gorgets and others occurring in various ancient manuscript books. The re- semblances are indeed striking and deserve the attention of arch- eologists. In plate xxx is presented an engraved gorget obtained in Mexico, probably in the state of Guerrero, and now owned by the Field Columbian Museum, Chicago, which will serve to illustrate the resemblances and the differences in the delineations of the two regions. The discoidal gorget is the most common form in both Mexico and the United States; but this specimen is oblong, being wide above and narrow below, conforming in a measure to the tapering form of the lip of the shell from which it was carved. The gorget is rather roughly worked out and the upper margin has been perforated for suspension, but two of the perforations have been broken away. These perforations are in the plain border which surrounds the design, but there are seven additional holes within the engraved surface; these may also have served for attaching the ornament to a garment. The human figure, engraved in rather crudely executed lines, faces to the left; the right knee is bent as in the Kentucky specimen, and the left leg extends forward. The position of the arms is not readily made out, owing to the cramped position imposed by the contracted space. What appears to be the left hand, supplied with an enormous thumb, rests against the right border of the design and grasps some kind of an implement pointed downward. The right hand extends in front of the figure against the left border and is partly broken away ; it appears to have grasped a staff terminating in what may be a rattle or possibly a symbolic device such as is often seen in ancient Mexican drawings. The mask-like features of the personage are drawn with the usual bold- ness of the Mexican work, and the eye is a conical depression sur- rounded by a curved line the ends of which open backward. The lower part of the face is covered with several groups of straight lines and a row of large teeth is shown. The ear and the ear disk are almost identical with corresponding features of the Kentucky speci- men already shown. The somewhat elaborate headdress is well en- graved, and the body and legs are covered with markings repre- senting costume. At the waist there is a belt, and the legs show encircling ornaments and indentations suggesting buttons. The de- SMITHSONIAN MiscELLANEOUS COLLECTIONS MOI 45 .u ble ex SHELL GORGET FROM MEXICO WITH ENGRAVED HUMAN FIGURE (Length, 6% inches.) HOLMES | SHELL ORNAMENTS 99 vices occupying the space beneath the human figure are carefully drawn but are so crowded together as to make interpretation difficult. These objects are presented here not that any discussion is to be based upon them but rather for the convenience of students engaged in comparative studies of the native art of the various regions. | 5 4 | Young Beas |e S |, Young Birds. | |S |. Young Birds.. > = |i] aS |g aS | |__| 81% |g\ la Beles ts (4 Wes 12103 4 fel z |Flale 3 |4jals yz} 2| 3) 4a ‘ioe aera ISIE EI mo} n 20 | | |? 391 | é | | 58 | | lenis 2 | 21 | |? ]}40] | | I 59 | Vesieas ene 3 | b 22 | |? } 42 | Bee eaGoiile |. Ne 4\ | m| | 23 | 12}42/ | iced WD | me Bt | 6 Zasey |? 43 | Peenreayi. i 6 | |n| 125 b | 44 | | |? 7631 beri 7 26 | 26 | 2145 |) |64l) | | | 8 | n | ley n 46h Samaleer ee ttOs awa 9) ieee eee | ‘2b 47 | ? | 66] | to| | |) | |29| 21 48 | | |? | 67 Ir | | alece 20 | b 49 |? 168 | i2) sal ic leet | 2 | 50 2? | 69)|? 13 | | A\|62| | n 51 | 1? |70 | 14 | Hrealaeyesaale fe ile Mire We eS 2 New eaias 15 | ie Se Sa SSSileal seals alae | | 16 pestle ets | 2154 | leita ete (leet SA cd mee Tae. | >155| | | |?) zal ee 18 es AES So eG) lenlee 75\|_| Laeiiean | ? | | | [tas | ro: ie | ae fi pe o 37 | rel [ae 4| 2 |12| 7 |39 ‘rotal@nestsvexaminedien asa ene: Mee OR “ce eggs «¢ young birds (40 in nests, 28 on branches), 68. Note.—Nests numbered 62 to 76 were not examined, but if the same average number of young to the fourteen nests be allowed, this colony should have produced 88 young in 1902. *In the above tables young in nests, b—=young in branches, c—nest well chalked but empty, empty without positive signs of having been occupied this season. 108 SMITHSONIAN MISCELLANEOUS COLLECTIONS [VoL. 45: On June 19 an examination of another colony was made, the re- sults of which are tabulated as follows: Cotony II (June 1oth, 1902) EM Nisin es eee yeste Pl als aliens : Zz 4! Young 2 Zz g, Young Birds 2 Zz F veune 2 Z om Young Birds. 2 Bia) ee a ea ee é A fare 31 4 A Tash esi legen) A I 2/3|4 4 cal sel) 2 ed asf Le 2 Sr | pes} | ae rel nb| | | 36| Ba itee Sra BD 6) 106 | b\6 2 hol | Mea b | 72 | 107 lanl 3 | 4 begets n | 73| | || | 108 [ae] Sale Ah lhl 13 | | | 74 | | 4 109 ? 5 n 40] | | 2 75) lel I10 | ¢ ? 6 c| 41] 2 | 76) | n| III 3 74 | | 42 | )o | 77 | P| 112 n | 8 | | Zale besa c| 78 | ea 113 b | 9) n | 44 | | |6 | 79| 16\ | | 114 | ? 10 Ths ara (6 80 Peale iene | ? II n | 46| ove | 81 6| | | 116 | 5 | 12 n 47 b 82 m| | 117 72 | 135 lal 2748 nb 83 6} | |118 | 4 14| | b 49 asa ale heal 119 b rele | aaa ?150 n| | 85 Bl

B> A. Very few basal sections were observed, but when ‘noted, the rather rarely developed macro- pinacoidal cleavage (100) was distinctly seen. An occasional twin was noted, and with the exception of an occasional apatite, a frag- ment of feldspar, and perhaps a zircon or two, the hornblende con- tains no included minerals. The feldspar, the light colored mineral filling the interstices between the hornblende, has undergone consid- erable change. In polarized light it appears rather clouded, though not at all commensurate with the amount of change which it has undergone. The true extent of this change may be seen by diminish- ing the illumination; then it appears that alteration is quite general, that as a rule it follows the cleavage cracks or composition planes of the twins, and that, as a result, a secondary mineral has been produced with brilliant interference tints, probably sericite. To a very small extent calcite is present in spots, as proved by faint effervescence when the specimen is touched with acid. Twinning of the plagio- clase, according to the albite law, was noted, though it is not general, and the high symmetrical extinctions indicate, together with those on the pinacoids, that the feldspar is a labradorite of a composition approximating Ab,An, to Ab,An,. It is evident that the feldspar has accommodated itself to the space remaining after the crystal- lization of the ferromagnesian constituent, and in the complete absence of quartz was the last constituent to crystallize. Additional to the feldspar and hornblende occurs magnetite, most commonly situated on the peripheral portions of, or enclosed by, the hornblende. Usually it is surrounded by a colorless or light-green rim, appearing as though formed at the expense of the amphibole. An occasional red scale of hematite was also observed, associated with the magnetite. Pyrite, already noted among the megascopic constituents, is readily diagnosed in incident light. Tabulated in the order of their abundance, there are present: Hornblende. {, pericite. As Essentials ( Calcite. Labradorite ( Magnetite. | Pyrite. Hematite. Zircon. Apatite. | Garnet. As Accessories PHALEN | THE ROCKS OF NUGSUAKS PENINSULA 189 The upper intercalated dikes (Cat. Nos. 75,482, 75,483, 75,484) are as a rule more coarsely grained than the lowermost, but in most instances the decided schistose structure is still apparent. In addi- tion to the minerals mentioned in the description of the lowermost dike, biotite is developed macroscopically, and is especially prominent in the most coarsely grained phases of the rock. There is present also a silvery, micaceous mineral, evidently a secondary product after amphibole and biotite. When viewed in thin section, the minerals appear pronouncedly coarse in grain and irregularly segregated. At times whole areas of variously oriented amphi- boles occur with no feldspar whatever; while masses of feldspar with no foreign minerals occur in the same way. The pleochro- ism of this amphibole is similar to that of the hornblende already described, namely, A, yellow with greenish tinge; B, green with a yellow cast, and C, blue or bluish-green. The absorption scheme is B=C>A. Generally the mineral appears to be quite fresh; it, however, shows bright interference tints at its edges, as compared with those of its mass. This is due to incipient alteration which has produced a light brown micaceous mineral, strongly dichroic, and with bright interference tints, as already noted. In some cases much of the unaltered hornblende still remains. The development of the feldspars reaches its largest scale, of course, in the pegmatitic facies, and here could be observed albite twinning in the macroscopic way. Here, however, the feldspar could be determined only unsatisfactorily in the ground mass of the rock, alteration having proceeded too far to admit of positive meas- urement. A fresh cleavage fragment from one of the pegmatite areas gave extinctions of oligoclase. This is too acid, however, to agree with the main mass of the feldspar, which, as in the case of the lower dike, is made up chiefly of andesite-labradorite or a plagioclase of intermediate composition. Sections which might ordi- narily have served as diagnostic material are completely transformed into brightly polarizing scales, leaving, but not always, a tiny spot, still showing the albite lamellze intact. Additional to the minerals already described, quartz and biotite occur in the pegmatitic facies of this rock. The latter mineral is best developed in the coarser portions, but it occurs also in the main ground mass, in the usual lath-shaped forms with fringed terminations, and also in scales frorh the base. In ordinary light it is brown in hue; in polarized light, strongly dichroic with rays of a brownish tint, vibrating normal to its cleavage, while those parallel to this direction are so strongly absorbed as effectually to obscure all color. 190 SMITHSONIAN MISCELLANEOUS COLLECTIONS [voL. 45 Magnetite occurs rather plentifully, especially in the region of the altered mica. It has been produced as an alteration product of the ferro-magnesian minerals and simultaneously with the secondary mica. Some original magnetite was also noted. The lower dike material, with its finer grain and homogeneous structure, represents perhaps more fairly than do any of the upper sheets the typical dioritic magma, whence the sheets have come. Its homogeneity renders it a safer criterion to judge of the constitution of the dioritic magma—hence its analysis is given, as follows: ANALYSIS OF DrorITE. (W. C. PHALEN, ANALYST.) SIO ceed oes TOE ete et eee ee 47.80 PUL Os is Rete ceegs Gee Ae SO Aas ee eae 18.24 C3) g oats tera eter ee tenet ea I eee 35 THE OT Ss Near cists ele Oe te enti) toed 9.27 INT Oe 1S ee ys SER te ete Re Ue hanes, TARE A ae oe 8.08 Ca Ole Hats eee Raya Ope enna Cd a 11.44 Ns OV iae h roch site coc eats oot eich aN CRD Ee ates, 2.24 BOM ADOVE MOO Hick cre emits catia iasiah Rhee 58 EE @©sbelOwislOOs a r.yae accom yes metres eabont eee — a @ ats yore ey euch dehcas et taped leet er teyey a one tee, oe the 1.46 © Fouche ahte TRIM eee ee ROE ae 24 INT QU Pi eres sce beed See SESE ENS eet ere 55 100.70 In addition to the rocks already described from Umanak island, there occur others of sufficient interest to warrant description. Un- fortunately the relationships of these rocks are entirely unknown, hence their value as petrographic factors in this province is largely curtailed. The one is a syenite, the other a diorite of more than passing interest, owing to the occurrence in it of an amphibole with peculiar parting. The description of these rocks with an analysis of the separated amphibole follows. Syenite-—This rock (Cat. No. 75,485) is salmon pink in color, with holo-crystalline texture, consisting forthe greater part of feldspar with an occasional vitreous quartz. The structure tends to the por- phyritic, the largest feldspars frequently having diameters of one centimeter or more and sinking from this size to microscopic dimen- sions. Occupying fissures between the feldspathic particles may be seen a green mineral, closely resembling epidote in color, while lustrous specks of pyrite are not uncommon. Cavities are scattered throughout the mass of the rock. PHALEN | THE ROCKS OF NUGSUAKS PENINSULA Ig! The texture is typically hypautomorphic granular, with porphyritic structure, and when viewed through the microscope presents a most heterogeneous appearance. This heterogeneity is in complete accord with the macroscopic appearance of the rock, for in places it is essentially syenitic, in other spots granitoid. Areas devoid of quartz are contiguous to those where silica constitutes fully one-fourth the mass. This silica does not have the appearance of typical original quartz ; it does not form distinct crystals, but appears in vein-like masses, elongated and irregularly shaped, surrounding large and distinct microclines, and frequently enclosing smaller particles of the same mineral. In this occurrence it is distinctly poecilitic. It has evi- dently, at least in part, in these spots been produced as a result of changes which the feldspar has undergone. To strengthen this hypothesis, similar irregular masses, very much smaller than those above referred to, occur in the midst of the larger feldspars. In addition to the secondary quartz, there is also present a smaller portion of undoubted original quartz. This occurs with automorphic outlines in sections from the prism zone with pyramidal terminations and in various other irregular forms. Liquid inclusions are abundant. Feldspar, of the variety microcline, constitutes the largest mass of the rock. It is associated with a small amount of orthoclase. In the former mineral a splendid development of the characteristic grating structure obtains. This constituent, as well as the orthoclase, is uni- versally filled with minute scales or lenticular particles of iron oxide, at times segregated so as comparatively to obscure the mass of the rock. These particles give, even in the thin sections, a faint reddish- brown tinge to the rock and are the basis of the salmon pink color observed in the hand specimen. These inclusions are frequently massed, especially at the boundaries of the crystals and in those portions of the rock where crystallization has taken place on a fine scale, 1. e., in the non-porphyritic portions. In the phenocrysts they are scattered in irregular and broken lines and are accompanied by a multitude of liquid inclusions. No perithitic intergrowths were noted. Of accessories there are very few; epidote and zircon were noted. The former, of undoubted secondary origin, occurs in very irregular patches, frequently elongated. It is faintly pleochroic and is much obscured by segregated iron oxide. Basal cleavage is roughly de- veloped. A scattering zircon was noted, but of apatite there is apparently none. Allusion has already been made to the abundance 192 SMITHSONIAN MISCELLANEOUS COLLECTIONS [VoL. 45 of iron oxide, hydrous and anhydrous, which constitute the pigment in the rock. Diorite (Cat. No. 75,486).—As with the diorites already described (see p. 187), so here also is the orientation of the essential constitu- ents, hornblende and feldspar, such as to produce a decided sheared or schistose structure. Besides these two constituents, an occa- sional speck of light brownish-gray silvery mica may be seen, usually included in the hornblende. Hornblende.—This mineral is present in irregular and connected grains, strongly pleochroic, with vibrations of the following color scheme: C, generally, bluish-green; B, green, and A, yellow with absorption B> C>A, with B and C lying very close together. Of inclusions, few were noted, an occasional irregularly shaped magnetite, a fragment of feldspar, and an occasional speck of brown mica, constituting the list. Upon closer examination the hornblende presents many points of rather unusual interest. Its color has been described as green; this statement, however, needs qualification, for in spots the mineral is perfectly colorless, other amphibolic characteristics remaining in full perfection. This bleaching of the mass of the crystal is appar- ently not connected with decomposition or alteration in any way. The cleavage lines of the mineral stand out full and clear. The ex- tinction angles remain as in the contiguous green portions and the mineral extinguishes as a unit. The interference tints of the bleached portion are higher than those of the green parts. A bleaching, similar to this, has been remarked by G. H. Williams? in the hornblende of cortlandite. This author says: “ The mineral (hornblende) becomes colorless and consequently non-pleochroic, while retaining the compact structure and optical behavior of the unaltered portion; later there is developed, particularly around the edges of the hornblende, a bright, emerald green substance which, on account of its lack of dichroism and feeble action on polarized light, may be regarded as chlorite.” The hornblende of this speci- men also exhibits a uniform bleaching in its peripheral portions, and the hornblende has doubtless become chlorite in these places. Very often the spaces, or boundaries between adjacent hornblendes, are filled with sericitic material which has come from the feldspars. It is possible that the waters which have produced this sericitic change may have bleached the hornblende along its path, thereby changing its composition to the less ferruginous chlorite. *Am. Jour. Sci., XXXI, 1886, p. 34. PHALEN | THE ROCKS OF NUGSUAKS PENINSULA 193 Another feature of the hornblende is a perfect parting parallel to the plane (101). This is rather indistinctly shown in plate Liv, 2, in the right portion of the figure, and in the same crystal of horn- blende there is shown the dark hornblende associated with the bleached material. This parting is far more perfect than any cleavage possessed by amphibole, not excepting the prismatic. It is present as a series of rather fine, sharp black lines, parallel to each other, and making an angle of 75.8° with the prismatic cleay- age on the brachypinacoid. Though rather rare, the phenomenon is not unique; it is first mentioned by Jermejew,’ and incidentally referred to by Williams? and Vom Rath.* Cross* and Mugge’ have also remarked it. In more recent articles, Williams* and Weidmann’ have described the phenomenon at some length, and have noted in connection with it twinning lamelle. At times in the sections under discussion a faint white band could be seen between the parting planes. The highest powers, however, failed to resolve any twinning lamelle, and it is probable that none exists in these particular sections. Though it cannot be positively stated that the parting planes are concentrated near fractures and are hence of dynamic origin,” the evidence which points to this view is very strong. As a rule they are concentrated at the boundaries of the hornblende crystals and tend to pinch out toward their centers. If now it is borne in mind that the rock under discussion has been sheared, and that its present crystalline nature is most evidently the result of dynamic meta- morphism, it will not be difficult to assign a similar explanation for the observed parting planes. An analysis of the separated amphibole is as follows: Si © Fae os fe treet rsseee niche Qneleay eRe haters 42.79 UNO tiecinins dela otaeb.dae cles Me thoe ec 15.04. ees) egy ae Re ty Reade ac at WK a hada meena Gaeta nie 5.44 FeO . 11.61 * Neues Jahrbuch f. Min., etc., 1872, p. 405. 2Am. Jour. Sci. (3), XX1X, 1885, p. 486. ®° Sitzungsber. d. Niederrh. Ges. f. Natur. u. Heilkunde, July 7, 1886 *Tschermaks Min. u. Pet. Mitth., 1v, p. 386, 1881. ° Neues Jahrbuch f. Min., etc., 1880, 1, 243. ® Am. Jour. Sci., XXXIX, 1890, p. 352. ™ Am. Jour. Sci. (4), xv, March, 1903, p. 220. 8 Williams (Am. Jour. Sci., xxx1x, May, 1890, p. 335) and Weidmann (ibid. (4), xv, March, 1903, p. 230) both assign a secondary, dynamic action as the cause of these gliding planes. 194 SMITHSONIAN MISCELLANEOUS COLLECTIONS [VoL. 45 - MeOr ike wiecee ce eee oe eer eae ¥1.00 CaO ss Myskina wo sectarian 10.24 Na.O aieuaxolctckacsvacarsueteneresattoncketenelsiehcdamenemercielsteieiena 2 89 TREO? seeitiosoece toads eee OL eI IT eee nee 47 Wt iv caret eine ratchet oe te ete verter eat pers 54 100.02 The feldspar of the rock is much kaolinized and clouded in spots. As a result of this, many tiny flakes of a highly doubly refracting mineral are present. This mineral, sericite, often penetrates the bounding planes between the hornblendes and frequently the larger cracks in the latter mineral. This process of sericitization is pecu- liar in that its effects are local, 7. e., one section of a feldspar may be completely changed, while the contiguous crystal may not have suffered any alteration whatever. Albite twinning, sometimes accompanied by pericline twinning, is frequent, though not the rule. The small extinctions, symmetrical with respect to the traces of the albite lamella, indicate a feldspar of the andesine series. Occasionally there occurs as inclusions in the hornblende and in the feldspar a fibrous mineral, evidently a member of the mica group. Its exact nature, further than this, cannot be ascertained, owing to the paucity of strong diagnostic features. It is yellow in color, strongly dichroic, with rays vibrating parallel to the cleavage planes of a yellow tint and colorless normal to this direction. The hornblende, near at hand, has been bleached, and the yellowish-brown color of the mineral in question naturally leads to the supposition that part of the iron content of the hornblende has been appropriated. It forms an intermediate link between the muscovite, which is present in slight amount, and the bleached peripheral parts of the hornblende. One other constituent deserves mention, 7. e., magnetite present as inclusions in both hornblende and feldspar. It is irregular in shape and is frequently surrounded by a pale halo. NUGSUAKS PENINSULA Kaersut.—Returning now to the mainland of Nugsuaks peninsula at Kook and traveling west to Kaersut, the gneissoid crystallines still continue overlain by. cretaceous shales and sandstones, inter- calated with coaly streaks and basalt flows. At an altitude of 880 feet above the plant beds the shales have been baked and the lignite converted to carbonite by the intrusion of a mass of horizontally bedded peridotite which forms a cliff 200 feet high. PHALEN | THE ROCKS OF NUGSUAKS PENINSULA 195 Peridotite var. picrite At first glance this peridotite (Cat. No. 75,487) might be mis- taken for a basalt, with an interlacing network of segregated olivine. It is black in spots, or rather has a blackish-brown augitic base, through which the olivine runs in zigzag courses, producing a mottled or inlaid appearance, remarkably clear and striking. A vein-like penetration of olivine into the black ground mass was also observed. On closer examination the grain of the rock proves to be rather fine, the olivine granules, however, being macroscopically distinct, while those of the ground mass are practically irresolvable. Under the microscope the following minerals were observed—oli- vine, augite, chlorite, feldspar, biotite, magnetite, limonite, hematite, and apatite. Of these constituents olivine is by far the most abundant. It occurs in perfectly automorphic forms, excepting when corroded by the surrounding magma. The crystals are in large part isolated and distinct from each other. Often very irregular masses are present. When isolated, there is, as a general rule, a development of the micropeecilitic structure, though in no case are the isolated particles broken from a parent crystal, but are per se crystalline units. In many instances the olivines present a completely shattered aspect. Fissures penetrate them in all direc- tions, often arranged radially with respect to the center and generally filled with a light-brown ferritic pigment. This phenomenon is not attributable to any pressure which the rock has sustained, for no evidences of strain are visible in any of the sections studied. Only in a single instance was undulatory extinction noted. Not only are the individual phenocrysts shattered, but the entire sections them- selves seem to be rifted in a more or less regular manner. Along these zones of fracture the rock is broken up into a series of parallel cracks, filled as with the olivines with brown and black iron oxides. We have here a phenomenon closely related to rifting, a microscopic phase of jointing, as it were. In many instances the olivines are in an advanced state of altera- tion and often the entire nucleal portions of crystals have been con- verted into light green, slightly dichroic prochlorite. This chlo- ritization furnishes the key to the explanation of the radiating cracks observed in the olivine, for the process is essentially one of hydra- tion and expansion, which produces the radial cracks observed. The process is not always nucleal, however, for simultaneous with and independent of these changes in the interior of the crystal, *See plate Liv, 3, in the automorphic olivine, represented in the middle of the illustration. 196 SMITHSONIAN MISCELLANEOUS COLLECTIONS [voL. 45 similar changes have occurred in the peripheral portions. Accom- panying the separation of prochlorite there is much light-brown iron oxide, limonitic matter, resulting from the ferruginous portions of the olivine. Often these ferritic portions are oriented in fairly parallel layers about the prochlorite nuclei. Immediately adjacent occurs a layer of limonite, bounded in turn by a rim of serpentine, while this in turn is surrounded by a border or zone of magnetite, compact and black with jagged border, the points projecting normal to the surfaces of alteration. At times the limonite is absent; in other cases limonite alone is present with no magnetite. Alteration is not at all confined to the olivine, for whole areas of pyroxene have made way for it and, curiously enough, in those spots where the changes in the pyroxene has been most profound, the olivine retains its original perfection of form and composition. Liquid inclusions are abundant and are arranged in zones or clouds, nearly always concentrically and in juxtaposition to a layer of iron ocher. Often these inclusions are so numerous as to cause these particular areas to become nearly opaque. The ground mass of the rock is formed by a violet-tinted augite, faintly green, however, in spots. It is not pleochroic. Sections from the two pinacoids are well represented. The augite, like the olivine, exhibits the micropeecilitic structure, with a particularly inter- esting development represented in the accompanying plate Liv, 4, to the right. It will be noted that there are two automorphic olivines adjacent; that to the left has its interior filled with augite, optically continuous with that which partially surrounds it. That the olivine formed first is assured; that the augite filtered in after the forma- tion of the olivine was complete forces the assumption of a primary cavity in the olivine. It seems to the writer that the phenomenon is an excellent illustration of the power of crystallization even under the most adverse circumstances, the olivine assuming its perfect form, even in spite of the intimate admixture of foreign augite mole- cules; that with the formation of the olivine phenocrysts the augite molecules were thrust apart to the interior and here segregated, optically continuous, however, with the main augitic mass without. The phenomenon is unique, only this instance having been observed in the sections studied in this series. In spite of this, I have dwelt at some length upon it with the hope that my explanation might be the means of bringing to light a more plausible solution of the phenomenon. ; The iron oxides occur in the usual forms of magnetite, hematite, and limonite, included in the olivine, augite, chlorite, etc. Most PHALEN | THE ROCKS OF NUGSUAKS PENINSULA aoe, abundantly in the prochlorite occur magnetite and hematite, without definite form, appearing in irregular scales, sometimes occupying entire centers of chlorite masses, at other times arranged peripherally in tiny elongated and jagged masses, normal to and in contact with altering olivine surfaces, as already noted under that mineral. Iron ore is also included in the chlorite, resulting from pyroxenic altera- tion. What appears to be iron ore is also relatively abundant in both the radial and parallel systems of cracks. That it is not mica is proved by its non-pleochroic character and from the fact that it is almost universally confined to the ferromagnesian constituents, end- ing abruptly at their junctions with the plagioclase. In color it is light reddish-brown, and this ends its resemblance to the small amount of biotite occurring in the rock. It is filled with flakes of black iron oxide, which seem to have resulted from it, and is with- out doubt limonite, partially changed in spots to magnetite and hematite. The remaining constituents of the rock are biotite, feldspar, apa- tite, with small amounts of chromite and pleonaste. The biotite is dichroic in dark and light shades of brown, the latter tint becoming very nearly colorless. Owing to the trifling amount of feldspar present, enough satisfactory measurements could not be made to judge of its exact composition with any degree of assurance. It exhibits albite twinning wherever it occurs, but symmetrical extinc- tions were more difficult to obtain. The highest equal extinctions were 25.5°, indicating a feldspar of composition approximating Ab, An,. Apatite occurs in relatively large amounts in the usual form of long, slender prisms in radiating aggregates. A small amount of chromite with a trifling amount of pleonaste were also detected by chemical means. An analysis of the pyroxene follows: St Oa ete eseos seeeasiensis cuba occeteha Se rreccheysie Slavs 49.49 ENTE Ose are nieve icra ars tuckewelors epee tane saaTee yas 5.45 TEs O) etre earn aos meh erecaeS etal cUare onmsenetet srs arose 1.04 TGCQUER SERA Cree here ation: Mhcova tate aver xe ieee ions he ieee 3.30 MgoO.. 15.88 Ca ix: 24.07 IN cts lee nee hs eich tren cod a os ieians ieledvnavaid seer 82 Mt Oe eyes Boned oars eainemine a mia ante 18 100.32 198 SMITHSONIAN MISCELLANEOUS COLLECTIONS [VoL. 45 Basalt (Bearing Native Iron) In addition to the occurrence of basalt in sheets, conformable with the bedding, the basalt of the region about Kaersut cuts the sedi- mentaries in dikes. Much of this basalt is decomposed, that with original vesicular structure (Cat. No. 75,495) having its cavities filled with. zeolitic material and secondary quartz. Of peculiar in- terest among the basalts is the occurrence at Kaersut of the far- famed iron-bearing rock. Its occurrence at this particular point, so far as the writer is aware, has not been reported before, though its occurrence in the immediate vicinity has been described, as at Blaafjeld (Ovifak), Mellemfjord, Asuk, Arveprindsen’s Eiland, Niakornak, Fortune’s Bay, Fiskernaes, Ekaluit, etc., where the rock has been collected and of which analyses have been made.* From a petrographic point of view this is by far the most inter- esting of all the specimens brought from Greenland, and so far as the writer is aware is the only rock of undoubted terrestrial origin known which carries native iron in any considerable amount. It resembles closely an aérosiderolite, stony iron, and for many years was supposed to be of extraterrestrial origin.? Its very close re- semblance to certain members of the meteorite family will be noted from the accompanying illustration, plate Lv. The specimen (Cat. No. 53,479) now in the petrographic collection of the U. S. National Museum was, before cutting and polishing, roughly ellipsoidal in shape, with a major axis of 17 cm. and a minor axis of 9 cm. When cut, polished, and etched with nitric acid (sp. gr. 1.42, dil. 1:10) the variation in size and structure of the metallic blebs was beautifully developed. It will be observed that the inclusions are of most irregular shapes and sizes, ranging from 1.5 cm. in greatest dimension down to mere points. They often surround patches of the basaltic ground mass, and frequently several blebs of minute size colonize and closely simulate typical graphic structure, with, how- ever, the regular outlines of the inclusions lacking. It is quite prob- able that such patches represent cross-sections of the ramifications of a single enclosed mass. While studying the thin sections of the basalt, it was observed that these metallic segregations were not at all as homogeneous as they appeared after remaining on exhibition for some years. It was plainly evident that at least two different kinds of iron are present, a black variety and a lighter variety with a silvery sheen. The irons are very irregular in outline. On etching 1 Chemical Researches on the Metallic Iron from Greenland, by J. Lorenzen, 1882, Meddelelser om Gronland, tv, pt. 11, 1893, p. 135. * Nordenskjold in Geol. Mag., 1x, 1872, pp. 88, 461. CW ON ‘Sa ‘64*ES oN) ‘ANVINHEUD ‘VIOSNINAd SHVSONN “LOSYAVA NOU! FAAILVN HLIM LIvsvd AT ‘Id ‘SP “10.4 SNOILIATIOZD SNOANVIISOSIPY NVINOSHLINS PHALEN] THE ROCKS OF NUGSUAKS PENINSULA 199 this composite nature is plainly exhibited and may be seen in detail in the accompanying illustration (plate Liv, 5). Macroscopically the two irons appear to be mixed promiscuously, and it is impossible to say with any degree of assurance which is the older, assuming that there is a difference in age. On etching, as has been remarked, the structure of these metallic inclusions is beautifully developed. The duality in composition is accentuated and a damascene luster becomes apparent. It must not be inferred, however, that the individual light and dark patches of a.given bleb are quadrilateral, for they may be and are of almost any shape whatever. They may best be described as rudely polygonal, with stippled surfaces and sharply contrasted by virtue of their light and dark shades as noted above. Though former writers, Nordensjold,t Rink,? and Steenstrup,* claim to have observed Widmanstatten figures in some irons, others have been found which do not show such markings. In the speci- men at hand, there is no regular orientation of surface markings, the parallel elongated grooves having been produced during the polishing. It has been remarked that two different irons are present. In the absence of a separation and chemical examination, such a statement needs qualification. From the physical point of view, this is true. A chemical examination may, however, reveal similarity in com- position—in this case, the marked difference in structure then might be explained as a peculiar phase of isomerism, viewed from the standpoint of the ultimate chemical molecules, and from the physical side the differences presented may be assigned to peculiarities in the orientation of the physical as contrasted with the chemical mole- cules. The solution of this problem, however, lies beyond the scope of a petrographical discussion and belongs to the sphere of the metallographer. In thin section the basaltic portion of this rock is seen to be com- posed almost entirely of pyroxene and plagioclase feldspar with abundant albite twinning. Of these two components pyroxene is by far the more abundant. It is perfectly colorless and stands out, in virtue of its strong relief, from the accompanying weakly refracting feldspar. It occurs in small irregular patches among the lath-shaped feldspar microlites. Then it occurs in larger, irregular masses, usually elongated and in forms approaching rectangles. When in * Geol. Magazine, 1x, 1872, pp. 88, 461. ? Oversigt over Kgl. Danske Videnskabernes Selskabs Forhandl., 1854, p. 1. >On the Presence of Nickel-iron with Widmanstatten Figures in the Basalt of North Greenland, K. J. V. Steenstrup, 1882, Meddelelser om Gronland, IV, pt. 11, 1893, p. 115. 200 SMITHSONIAN MISCELLANEOUS COLLECTIONS [VOL. 45 basal sections, it is observed that prism faces are poorly developed as compared with the pinacoids. There are traces of a rude and irregular parting parallel to these latter surfaces. In spots it is partially altered with the separation of yellow and black iron oxides, the former hydrous. Along the cracks chlorite is slightly developed. Twinning with the twinning plane (100) was observed in several instances. The pyroxene is closely related to the augite group. With the feldspar it is often enclosed in the metallic portions. The feldspar of the basalt is in two generations. The crystals of the first separation often attain a length of 3 mm., the microlites often sinking to ultra-microscopic dimensions. At times the larger feldspars exhibit undulatory extinction ; they were undoubtedly sub- jected to strain during the closing stages of magmatic consolidation. The feldspar is in a perfect state of preservation, colored occasionally by the presence of hydrous iron oxide, especially in the immediate region of the iron inclusions. Its high symmetrical extinction lying between 35° and 40° indicates a feldspar of the composition Ab, An,. It is, therefore, a feldspar of the labradorite-bytownite group. Magnetite (and this is the only remaining constituent noted in any considerable amount) is relatively abundant. It occurs in irregular elongated forms, often approaching a dendritic development, but the typical skeleton-like development was not seen. It is especially abundant in the region of the native iron, and here a rude fluxion structure of the entire basaltic portion of the rock was noted.* The remaining basalts, collected from the vicinity of Kaersut, are of the usual olivine and non-olivine varieties, much decomposed and presenting nothing of unusual interest. Their detailed description will, therefore, be omitted. Lying loose, just back of Kaersut, occurs a peculiar rock (Cat. No. 75,488) which deserves more than passing notice. It is brown- ish-red in color on the fresh fracture and is traversed by tiny veinlets of calcite; this latter mineral, however, is not confined to particular strips, the rock effervescing in all parts on the application of acid. Although of a coarsely crystalline character, with the exception of an occasional rather large irregular patch of tarnished olivine, the character of the ground mass cannot be made out with ease. The olivines, for the greater part, glisten i in reflected light, and present a * For a very complete bibliography on the iron- bearing basalt of Greenland, see Dr. Th. Nicolau, Meddelelser om Gronland, xxiv, 1901, p. 217, and for a brief résumé on the work done on the nickeliferous iron in the basalt of Greenland, see Gisement et Nature du fer nickelifére du Gronland, by F. Johnstrup, published in Meddelelser om Gronland, 1, pt. 1, 1893, p. 270. PHALEN ] THE ROCKS OF NUGSUAKS PENINSULA 201 brown surface, not characteristic at all. Evidently the olivine has decomposed along its cleavage planes. In spots, the clear green and rough surface of the mineral appear. The minerals, in their order of abundance, are olivine, feldspar, pyroxene, iron ore, serpentine, calcite, and apatite. These are com- bined in a hypautomorphic, holocrystalline mass with olivine, the first mineral of the consolidation. This constituent is by far the most abundant, occurring in irregular masses, with the usual hexagonal forms. It is usually much shattered and fractured, due to disintegration, which is far advanced in some of the crystals, and which is accompanied by a copious deposit of iron ore in various stages of hydration which has segregated along the cracks. In some of the olivines there is present what might be called at first sight a pseudo-cleavage—very fine straight lines or spaces, for the greater part normal to the pinacoidal cleavage, but often parallel to this cleavage. With the highest powers of the instrument this is seen to be due to lenticular masses of iron oxide, rather scattered in the body of the crystals, but becoming more abundant near the main cleavage cracks and finally disappearing in a pleochroic pig- ment present in such cracks. Liquid inclusions are very abundant. After olivine had crystallized out, the next mineral to appear was pyroxene. It occurs in irregular masses of light green color, with prismatic cleavages well developed. Extinctions range as high as 44°, indicating an augite of intermediate alumina and iron content. It is extremely abundant and shows no decomposition whatever. Magnetite, with other iron oxides, occurs in several forms in the rock, inclosed in the ferro-magnesian constituents, rarely or never in the feldspars. It is part original and part secondary, resulting from the olivine. The original crystals are in the form of cubes and octahedrons, sometimes twinned. Then there occur skeleton crystals of the most wonderful patterns, as described by many petrographers and figured by Pirsson? and Hobbs.’ Magnetite occurs as an alteration product after olivine, often clouding and rendering opaque whole phenocrysts of this mineral. Patches of olivine are also rendered partially opaque by separated red hydrous iron oxide. After these constituents, olivine, augite, and original magnetite, had separated, the feldspar accommodated itself to the residual space. It is remarkably fresh and, for the greater part, free from inclusions. It occurs in lath-shaped forms twinned according to the Carlsbad * Twentieth Ann. Rep. U. S. G. S., pt. m1, 1898-99, plate Lxxt. * Twenty-first Ann. Rep. U. S. G. S., 1900, pt. II, p. 65. 202 SMITHSONIAN MISCELLANEOUS COLLECTIONS [VoL. 45 and albite laws. The extinctions, according to the statistical method, range as high as 31.5°, which would place the feldspar in the class intermediate between those of the formule Ab,An, and Ab,An,. The composition may then be represented by the formula Ab,An, and it would be classed among the labradorites. The remaining constituents are serpentine, calcite, and apatite. Serpentine occurs as an alteration product of the olivine, appearing in cracks with a fibrous structure, with fibers normal to the altering surface. It shows aggregate polarization tints of dull gray, often becoming bright yellow. Apatite occurs in automorphic elongated prisms, enclosed in the olivine and feldspar. Calcite is also present. A partial analysis of this rock, together with a discussion of its name and place in the new quantitative classification of igneous: rocks? will be found in Pt. 11, p. 211, of this article. UJ ARARTORSUAK Westward from Kaersut, the Cretaceous rocks are seen resting unconformably upon a bluish-green, highly altered basalt. A little to the west, the beds exposed in the sea-cliffs are dislocated for several hundred feet by a fault, directly beyond which the strata are cut by three dikes, the two westerly of which are parallel with the bedding in places. These intrusive basalts are believed to be of Tertiary age. The dikes are apparently of the same rock as the intrusive sheets themselves and probably represent the vents through which the sheeted material has reached the surface. Both olivine and non-olivine varieties are represented in the Museum collection, and from the similarity in composition and general features, the detailed description of but one will be attempted, that from an inter- bedded conformable sheet at Slibstenfjeld, directly back of Ujarar- torsuak. Basalt.—This basalt (Cat. No. 75,489) is greenish-gray in color and breaks with a roughly conchoidal fracture. In the greenish- gray fine-grained ground mass occasional specks of yellowish-red iron oxide are evident to the naked eye, as also are much larger auto- morphic crystals of a light-colored and well-cleaved mineral, pre- sumably feldspar. Examined with the hand lens, though not salient, it is very abundant in lath-shaped forms. ‘The rock presents a de- composed appearance, and the acid test reveals the presence of much calcium carbonate. In thin section the rock appears holocrystalline, xenomorphic with lath-shaped feldspars, pyroxene, magnetite, hydrous iron ore, calcite, * Quantitative Classification of Igneous Rocks, by Cross, Iddings, Pirsson, and Washington. Chicago, 1903. PHALEN | THE ROCKS OF NUGSUAKS PENINSULA 203 etc. In places it shows undulatory extinction. It appears to be de- composed in spots throughout ; the ferromagnesian constituents have completely changed and the feldspar has suffered locally in the same way, while in places the latter mineral is remarkably fresh and determinations of its character may be made with ease. The posi- tions of the pyroxene and olivine, if the latter mineral were ever present, are now occupied by green and brown pigments. Mag- netite, of course, is always present. The constituents may be described as follows: Feldspar, the most abundant of the components still preserved in fresh condition, shows abundant albite and Carlsbad twinning. Symmetrical extinctions on the trace of the clino-pinacoid (o10) range as high as 35.6°. The Michel-Levy method of Carlsbad and albite twins corresponds to a plagioclase of the composition Ab,An,, which is borne out by the high symmetrical extinction given above. Thus the plagioclase may be termed a labradorite bytownite and is very basic in composi- tion. In spots this feldspar is much kaolinized; it is cracked and secondary calcite has filtered into or formed in the fissures, filling them completely. Flakes of the greenish pigment to be subsequently described are also present, as well as elongated, lenticular, and lath- shaped forms of the same coloring substance. In general these in- clusions are without regular orientation, but in the vicinity of and along the composition planes, there is a parallelism evident. At times cloud-like masses, elongated normal to the C axis, were de- tected with the number 7 objective. Magnetite, though an inclusion, is not abundant in the plagioclase. Its form is not automorphic, and part, even here, may be secondary. The residual mass of the rock is largely a conglomerate of de- composition products, of which the most abundant are iron ore, cal- cite, chlorite, and serpentine, and much material, isotropic and sec- ondary, whose properties are not at all diagnostic. These minerals are scattered promiscuously throughout the rock, in contact with and overlying each other. If we assume that the calcite areas rep- resent original pyroxene and the serpentine is taken as a rough index of former olivine, then the original rock consisted of pyroxene, feldspar, olivine, magnetite, and apatite, in the order given. From the intervening localities, at which explorations were con- ducted, namely Saviarkat, Kookangnertunek, and Niakornat on the north side of the peninsula, and Ata and Patoot on the south side, nothing of petrographic interest was obtained, owing to the very advanced state of decomposition of the rocks at these places. At 204 SMITHSONIAN MISCELLANEOUS COLLECTIONS [VoL. 45 Alianaitsunguak, however, lying to the west of Ata, olivine basalt was collected in a fairly good state of preservation. At Atanekerd- luk, on the southeastern coast, some specimens were collected, among them olivine gabbros and monzonite, the former constituting the main mass of the small peninsula. The only representation of this large areal extent is a single hand specimen (Cat. No. 75,490), somewhat decomposed, but still well enough preserved to admit of satisfactory examination. In color it is old rose, spotted here and there with what appears to be a secondary mineral, but which on examination in thin section proves to be a surprisingly fresh pyroxene. Sometimes this green. mineral forms a continuous net- work in the red feldspar. It is this latter constituent which forms the bulk of the rock and from which neariy all the secondary con- stituents have been derived. Among these latter, calcite is readily distinguished, occurring in tiny anhedrons throughout the red plagioclase ground mass. Viewed in thin section the rock presents a hypautomorphic granular appearance. Among the most abundant minerals is pyroxene. It is light brownish in hue, approaching at times a violet. Prismatic sections, both 100 and o10, were observed, the latter giving extinctions as high as 52.9°. These sections are naturally most numerous, but octagonal basal sections showing both prismatic cleavages were surprisingly well developed. Twins of the usual variety, the twinning plane being the orthopinacoid (100) were noted. With the lowest powers of the instrument the augite appears perfectly fresh. Even the highest powers prove that decomposition has not advanced very far. Even those parts of this mineral which project into thoroughly decomposed parts of the rock, consisting mainly of viridite, are plainly discernible, their crystalline outlines showing with much distinctness. Naturally in such areas the augite is much decomposed, but the evidence is plain that the greenish residual product is not chiefly derived from the augite. Along its cleavage planes foreign matter has filtered in, while a portion of such material has been produced im situ. Inclusions of hydrous iron ore and magnetite are present. Thus, to summarize, we may say that, although the augite has slightly altered, alteration has taken place only at vulnerable points and it is yet the most perfectly preserved of the constituents. In age it is older than the feldspar. Though augite presents the same appearance throughout its entire extent, so far as progress in alteration is concerned, the same is not true of the plagioclase. Though perfectly fresh in spots, it presents’ all phases of alteration to the fully kaolinized material. It exhibits a PHALEN ] THE ROCKS OF NUGSUAKS PENINSULA 205 splendid basal parting in many of its lath-shaped forms, and the process of alteration has begun along these planes of weakness, but not always, for peripheral alteration has advanced to such a degree that it must have been at least simultaneous with, if not antecedent to, the changes along the basal partings. This alteration has filled the space originally occupied by the feldspar with a greenish sub- stance, which, for want of a better term, I have referred to as viridite. In part this viridite is very dark bottle green, changing through various greenish and yellow shades to a bright saffron yellow. Un- doubtedly much of the pigment of the viridite is iron oxide in various stages of hydration. In plain polarized light and with the very high- est powers these pigments seem to be made up of shapeless patches of brownish-yellow material with occasional specks of magnetite and lath-shaped particles, all identical with that observed in the basal partings of the plagioclase. Needles of apatite are abundantly scattered throughout, while underlying the heterogeneous mass may be observed the faint grayish-blue interference tints of the feldspar. Hence it is plain that these decomposition products owe their origin to the feldspar chiefly and not to the augite. We also have in these abundant products a ready explanation of the brilliant reddish hue of the plagioclase observed in the hand specimen. In spite of its almost completely altered condition in places, the feldspar at times is even fresher than the augitic constituent. This is the exception, however, and not the rule. It is fortunate, how- ever, in that it enables the feldspathic minerals to be diagnosed with no difficulty whatever. Its extinctions range in the neighborhood of 25°, the maximum symmetrical extinctions obtained being 26.3°. The feldspar, then, is a typical labradorite of composition Ab,An,. Of the accessory constituents, olivine was noted in a few sections in an advanced state of change. It appeared with its usual comple- ment of cracks, filled with dark-brown ferruginous matter. Cavities are rather numerous; in many instances their outlines are suggestive of the former presence of olivine. Even were olivine present to this extent, it would still be classed as a minor accessory. Iron ore is present in considerable quantity. It is confined to the decomposed pofttions of the rock almost exclusively and is very irregular in outline. Although present in large masses, it appears to be at least in part secondary. Much of this dark opaque mineral is pleonaste. Chlorite is present in small quantity, bordering the augite. Calcite, in tiny anhedrons, also occurs as a decomposition product. Apatite has already been referred to. 206 SMITHSONIAN MISCELLANEOUS COLLECTIONS [VoL. 45 In summary we have then as original constituents, labradorite, augite, magnetite, pleonaste, apatite, biotite, and olivine ; as secondary constituents, hydrous iron oxide, chlorite and perhaps a small amount of serpentine. Of these original constituents the labradorite and augite only rank as essentials, the remaining constituents are accessory. The rock, therefore, may be classed as an olivine gabbro. Quartz Monzonite-——This rock (Cat. No. 75,491), of which a single specimen was secured, is light gray in color with a ground mass composed of feldspar with semi-lustrous cleavage plates. Scattered through this white ground mass lie elongated crystals of a black mineral, hornblende. In some places, locally segregated, occur bunches of mica flakes. Drusy cavities are quite frequent. An occasional pellucid quartz was also noted, but though macro- scopically evident, it is a very minor constituent. Though apparently fresh, the rock, a surface sample, has under- gone considerable alteration and weathering, and the feldspars are considerably kaolinized, presenting under the microscope grayish- brown cloud-like masses, which, however, give the dull-gray inter- ference tints. In texture the rock is panautomorphic; all the grains are crystal- line, approximately the same size, approaching automorphic forms, yet few or none possessing them. In thin section the appearance of the rock duplicates its mega- scopic features. By far the most abundant constituent is feldspar. This has decomposed to such an extent that in some spots identifica- tion is impossible, and even when freshest its identification is obscured from the large quantity of separated material, which makes the cleavage lines most indistinct. This separated material is, for the most part, kaolin. It is brownish in color and is made up of flakes and needles without any action on polarized light. It is scattered in translucent clouds throughout the mass of the feldspar. This latter mineral gives in some sections parallel extinctions ; in others, the angles range from 8° to 18°; such sections evidently approach the clinopinacoid and show an elongation of the feldspar parallel to the c axis. Though generally untwinned, in places albite lamellze were noted. These small extinctions point to the presence of albite and oligoclase. In some spots, in the neighborhood of larger quartzes, a micropegmatitic development of quartz and oligoclase was seen. Carlsbad twinning was also noted. Additional to these acid feldspars, another group of plagioclases is present of a much more basic character. These basic feldspars are twinned both according to the albite and Carlsbad laws, hence PHALEN | THE ROCKS OF NUGSUAKS PENINSULA 207 advantage could be taken of Michel-Levy’s method on sections nor- mal to o10. Extinctions in one part of the Carlsbad twins range as high as 31°, in the other part 15.5° is the observed angle. Evi- dently the feldspar is entirely different in composition from those already observed and should be classed as a labradorite with com- position very close to Ab,An,. The basic labradorite is not nearly so abundant as the more acid substance. Associated with the feldspars are the following minerals: Amphi- bole, pyroxene, zircon, quartz, apatite, and magnetite, besides some hydrous iron ore. Of these minerals amphibole is by far the most important and most interesting, owing to its variety. It occurs in at least three different types, ordinary hornblende of deep chestnut- brown color, a light green phase, while a third variety has been de- rived from the pyroxene with which it is now associated. Of these forms the ordinary variety hornblende is most common. It is strongly pleochroic, the rays vibrating parallel to C being a deep chestnut-brown, those parallel to B of a similar but lighter shade, while those parallel to A are yellowish-green. The absorption scheme is the usual C > B >A, but the difference between B and C is very slight. The hornblende is generally lath-shaped with jagged or frayed terminations. Sections parallel to 100 and o10 are present. No crystallographic terminations were noted, but basal sections (001) are frequent with prism and clinopinacoidal faces well developed, the macropinacoidal showing as a rule. The secondary hornblende is light chestnut-brown in color. It is generally associated with the unaltered portions of the original pyroxene, whence it has been derived, and crystals of light-brown pyroxene were noted with ex- tinctions as high as 45°, whose edges and corners had completely gone over to hornblende of a markedly darker hue, with extinctions ranging in the neighborhood of 20°. The third form of amphibole is usually associated with the dark and light brown varieties already described. It occupies in every instance a peripheral position and sometimes has a fibrous radiating structure—though as a rule the structure of the associated horn- blende is repeated in the green mineral. Most commonly, its form is that of the amphibole, 1. ¢., lath-shaped, or rather it forms a part of the lath-shaped crystals. Cleavage in its prism zone is well developed. Its pleochroism, bluish-green in the prism zone, with yellowish-green hues normal to this direction, together with its posi- tion and other relations, indicate a soda-amphibole of the arfved- sonite group. Evidently we have here a gradual change in com- 208 SMITHSONIAN MISCELLANEOUS COLLECTIONS [voL. 45 position from a normal hornblende to one abnormally high in soda, brought about through contact with a magma, becoming increasingly rich in soda molecules. At times the arfvedsonite mineral is slightly decomposed. The pyroxene presents nothing novel. Basal sections are rather frequent and exhibit the well-developed prismatic cleavages. It is of the augite variety. This, together with zircon, apatite, biotite, ilmenite, and magnetite, together with a scattering quartz, complete the list of original minerals. In the vicinity of the amphibole, much iron oxide has separated. It is black in shade and irregular in outline. It has been derived from the hornblende, but the residual mass of this mineral appears perfectly: fresh. Clouds of yellow ocher were noted, chiefly in the vicinity of the pyroxene, and from the analysis ilmenite also appears to be quite abundant. Fluid inclusions in the quartz are not un- common. The analysis of this rock follows: ANALYSIS OF Quartz Monzonite. (W. C. PHALEN, ANALYST.) Si@ eager ec ee odie teasl On OL ee 67.27 Ns ©) aioe tpn Te kek eae oa Me a ere 13.67 Fe.O; RiotiencatcLeneaclietetianeicmckcheuieletetatonelcnstens tel sreierencncieitc 1.83 EG Olen yee ince Aer A ne ee ee IE ee Eee) WED) rec Sa rohcre ad epdicarer nd Ste oars tone ee (CaO RP as ch Sa ae Win ee ee 1.90 Nas O Tenepretoe tearcinc, Meee arte mere OCR enna 2.79 GO Rree rae ca eed Me pean ee ae os 5.80 SOA OVemMlOO Marne toes {rer ae Aytar ears 45 EE OSbelow 100 gts trys cieaerun ek a eee 08 AA © eck cco ee Re ee ER ERE ee ce ee nO) P:Os Sietaech chomielaciieuerionais Ai Leen Cae Oe ba 16 INET) acs sala sore ale ears Seat gb ok te ca Regen eae .19 100.05 Parr il INTRODUCTION During the preparation of this article a new rock classification has appeared, based on the chemical relations which obtain in igneous rocks.t It was decided by the writer to incorporate the principles of this classification into the body of this present theme, in so far as they were applicable, but on maturer thought this was deemed inadvisable for the following reason. It was held that to introduce so much of an entirely new nomenclature into an article where an old and familiar group of names must of necessity predominate at * Quantitative Classification of Igneous Rocks, by Cross, Iddings, Pirsson, and Washington. Chicago, 1903. PHALEN | THE ROCKS OF NUGSUAKS PENINSULA 209 the present time must only tend to obscure and to detract attention from the main topic at issue. At any rate, with this idea in view, it was decided to append a second chapter, devoted chiefly to the discussion of the rock analyses and their interpretation in the light of the work already alluded to. UMANAK ISLAND Granite (Cat. No. 75,480).—Directly back of the low-lying ex- panse which has been described as stretching a mile back from the coast! there rises, to a height of 3700 feet, a typical nunatak, com- posed of granite. An analysis of this granite, comprising the boss, follows, and with it an analysis of graphic granite from Omeo, Vic- toria, Australia, by A. W. Howitt. ANALYSIS OF GRANITE—OMEOSE (2) (2) SiO Merk ines ak ae ee ee OLO8 70.91 PAE © ese en ry, sie eeacae ANG cteraieSerarstl TO? 15.32 ie: Ora re mete oat: Some econ .69 trace Rie @Ome yet Loose ace GOS -= Neer O. ok aac atte ELS .07 (Ga Oe aha ere ee ao ea ILO 58 [Nias Ore sere trs tatiana Sea Ec obecbe ee ees Os Dra REO) pers pee atin screenees eee set ETS 10.07 lhl @) eee No. Vues ete Srciew: kA eee OO) 51 MBO Mee BOE oe IS Bran Jk aie wis ane LO — 100.38 99.77 Note.—r. Greenland granite, W. C. Phalen, analyst. 2. Graphic granite, Omeo, Victoria, Australia, A. W. Howitt, analyst. Trans. Roy. Soc. Vic- toria, XXIV, pt. II, p. 120, 1888. COMPOSITION IN TERMS OF STANDARD MINERALS (1) (2) OAL ie Sherk Geee oa sentence ASO. 18.00 @pihoclase tas ste cms aol te tes 33.92 50.49 PeIbew et eee eens cece Nagase LOS 19.39 NTO Eulaltereeeeiee eae Bacieiane ao epelaastens 2.78 1.67 Niieninemeneta ct. tetitin L poe cite c ene LOl — MYM ore TAOS raaic.a Sere neiet ats ow tenate's 9) 398 — IDWODEGE.“sGo cooper dccpeosbeouens da5y/ 66 Wiollastoniteusenaemee ec rere a Loo — Hels Oia ea cacy: eee Werte raid ee surey LS20 aca 100.41 99.72 According to the quantitative scheme of classification the rock works out to an omeose, a rock very high in silica and potash, granitic in every respect. 1 See page 186. . 210 SMITHSONIAN MISCELLANEOUS COLLECTIONS [VOL. 45 ‘It will be noted that the analyses show rather wide discrepancies, only. the wide ranges in the ratios involved in the classificatory scheme allowing them to be placed in similar subdivisions; thus, for example, in the Greenland rock, the ratio Q:F is 0.6, in the graphic granite from Omeo it is 0.22. The ratio between the alkalies and lime in the former case is 15.5, and that between the potash and soda is 1.9; these respective ratios in the type omeose being 30 and 4. . Diorite (Cat. No. 75,481).—The diorite has been described as forming a series of horizontal dikes in the granite and as varying in texture from a very fine grain to a pegmatitic facies. The lower dike material has been chosen as a type of the magma, from which all the diorites have come for the reasons stated on page 190 of this article. Its analysis, therefore, is given with that of von Lasaulx’s doleritic lava from Auvergne, the diorite having worked out to an atuvergnose. ANALYSES OF DIoRITE—AUVERGNOSE (1) (2) SIO se Pea en Aine apa nce ahr cATEOO 48.57 AIS Oss ere aed etek ee ele TOraA 19.47 Re OP ea ry te pen Meee ip kant ee dees AO' Oe TQ Nic @) earn ns SRS yh Supt oe ROTO 4.25 CaOURy cnee.c Seo roche eto: LIGA 10.86 Nias Oath ne oe rena. aero 133 KS © aero ee ree a eS 82 Els © Rab oviemlOOneee eee 58 ( 8 HEOmbelaw NOU np. cosh eee = ee Sei OPM ree. Fie wes ee ask aE O — SO ae ean teed Eater ae — IN ital @) I meer resent wee, aneenecanas te ee 76 100.70 100.07 Nore.—1. Diorite (Auvergnose) Greenland, W. C. Phalen, analyst. 2. Doleritic lava from Auvergne, A. v. Lasaulx, analyst. Neues Jahrbuch f. Min., Geol. u. Pal., 1860, p. 657. COMPOSITION IN TERMS OF STANDARD MINERALS (1) (2) Orthoclalseipescen: cr etre te tee ee REE, 5.00 IN Dube tras eeret aceite vase Cae UTR OO) 11.00 ATO TtMItesaeRe econ ete eae ee Meee ae OOO) 44.76 Tl iveniter se, (ee name eae ene ECO 7 — Magnetiteshe oa peers eres oe 70 — Apatite resco cance tet ee eee .31 as Digpsiderr nae ee eee eee 7.83 Ely perstheneycaene eee eee ees Or 25.62 Olivine’. st oh eee eee ae EO eo. 5.45 FLO Shes. ce ate ae ee .48 100.77 100.14 PHALEN j THE ROCKS OF NUGSUAKS PENINSULA 211 Though the analyses show such “discrepancies, especially so as regards the content of the iron oxides and magnesia, the ratios in- volved in the placing of the rock in its classificatory position are very close. The auvergnose from the type series of von Lasaulx how- ever, would, strictly speaking, be placed in the sodi potassic subrang of the rang auvergnase, owing to a ratio value between the alkalies of 0.61, this ratio, in the Greenland rock, being 0.2. NUGSUAKS PENINSULA On page 195 there is described a rock (Cat. No. 75,488), which has been referred to as lying loose just back of Kaersut. Such ma- terial is not regarded as having very great petrographic value. Nevertheless, since the type is rather limited, so far as its exposures in the earth’s crust are known, it was deemed worthy a partial analysis to ascertain its classificatory position. From the results of this chemical examination, which follow, it will be seen that the rock is very basic, that it belongs to class 1v, the dofemane class, section four of the order hungarare, and to the domagnesic subrang of sec- tion one in the permirlic rang of this order. ,Its name therefore is custerose. PARTIAL ANALYSIS OF PERIDOTITE—CUSTEROSE (1) (2) Si@ sapere ae ene oe oss gn AOIOS 46.03 PANE Ope eee ORE, eine ne OOS 0.27 es ©) aE RAN ere Pant nc noe 3638 2.72 ee OMe eeer ee neta tae ciseietos BS LET 0.04. AN its ©) Re Pte, Wee eRe Nn: co Seve tases LONGO 25.04. (CEN) i eR see ae a a X00) 3.53 Nia Oa taser ee tran Oem mines, E20 1.48 Mit OB ean eet ere Sastre, sc5°3 .30 .40 Ni ©) Mirae epee Meuse csi isis inde. “27 = Ci Ore ee cies ok ere 1 SOS — tells Oe se aereP eaten tS cr Sfotabe, Seckva e — 04 lags O) Rae ME oc Ln oo cuclic seve oe — ny 97.45 100.09 Nore.—t!. Partial, analysis of peridotite (custerose), Greenland, W. g: Phalen, analyst. 2. Analysis of peridotite, near Querida, Silver Cliff, Custer Con Colm ean Gmbakinsyeatialyst.. Cross, 17th Ann) Reps Uso. \Ganos Disiieen: 2o4n SOO: There remains but one rock, quartz monzonite, whose microscopic analysis is given on p. 206. This rock, whose chemical analysis is given as follows, proves to be a dellenose. 212 SMITHSONIAN MISCELLANEOUS COLLECTIONS [VOL. 45 ANALYSIS OF Quartz MoNZONITE—DELLENOSE (1) (2) SiO ay Te ee ee ee MEO Teer 68.36 AV Os Sede Pere -ohevs nod ate hey tee er ALO, 13.24 BesOsa. aortacs etic ee ee Mees 1.29 He Oia ey. Streit: cee ae cos Meee en 2 AG 3.30 Nici @) Serre oe et eet nee ees ee mma 1.15 CAO den sh tice: oricce late nia ee OO 2.51 Nar Oe ee te eee a eee eee aE) 2.05 Ke Oise ein Cee eee EE TOO 5.34 lEtXO) @IDOWE WOO no sonnccacvaeoes 45 Fe OR belowalOOneee eee eee OS 203 Ahi @) Saver ae aero aa tas ei ee ee 2 © — PaO ey sees ca sg Sere Ee eee LO _ IN Tita\@) ape toe eset ane oe ee et ct emcee) ee, » 100.05 100.23 Nore.—1. Quartz monzonite—Greenland, W. C. Phalen, analyst. 2. Glassy andesite—Dellen, Sweden, H. Santesson, analyst. F. Suenonius, Andesite from Helsingland, Sweden, Geol. For. Foérh., x, 1888, p. 273. CoMPOSITION IN TERMS OF STANDARD MINERALS (x) (2) OM aii Rot een ee ee ee 2e 82 26.88 @rthoclase) 5 ay score een hie Ae 7 31.14 PN eis er Me geval Se) NM en br | OE DEO SS 720) PATIO GENIE LL ans aoe ee Ga ee EO II.12 XDA TIL OUe eat. hee ieee eta crore ee ee -- lineEMIte RA en clea sere aero ote mE oLO — Mialom tite Ai hae eles cca oralltain ge ESS 1.86 DiOPSICE” pare wees Pree reirre n eered bE OO 1.18 ISEDSRSUMEMNS S550droaueccccencece vas) 7.08 Nate ai ictare un teattoct er sres emcee a5) 2.63 100.05 100.08 From the analysis of this quartz monzonite, it will be seen that the silica is rather high. This is explained in part by the presence of free quartz. Low lime indicates a small amount of basic feldspar. The high content of titanic acid is worthy of mention. In the main, the analysis agrees fairly well with that of the type rock from Sweden. Notable differences occur, however, in the cont at of ferrous oxide, water, and titanic acid; the differences in the first and third substances, however, do not suffice to alter the classificatory position, since in rocks of the first three classes these oxides do not have the weight which they possess in similar divisions of classes four and five. Note.—In conclusion, I wish to express my appreciation of the many courtesies extended to me during the course of this work by Dr. G. P. Merrill, Dr. Wirt Tassin, and Mr. Chas. Schuchert, of the U. S. National Museum, and by Dr. David White of the United States Geological Survey. June 9, 1903. NOTES INTERNATIONAL CATALOGUE OF SCIENTIFIC LITERATURE The International Catalogue of Scientific Literature consists of a classified subject and author catalogue of all original scientific literature beginning with January 1, 1901. All of the seventeen sciences named below are included within the scope of the catalogue, one volume a year being devoted to each: Mathematics, mechanics, physics, chemistry, astronomy, meteorology (including terrestrial magnetism), mineralogy (including petrology and crystallography), geology, geography (mathematical and physical), paleontology, gen- eral biology, botany, zoology, human anatomy, physical anthropology, physiology (including experimental psychology, pharmacology and experimental pathology), and bacteriology. The organization consists of a central bureau in London (Dr. H. Forster Morley, Director) to edit and publish classified references to the world’s current literature furnished by regional bureaus estab- lished in and supported by the principal countries of the world. The system of classification adopted divides each science into spe- cific, numbered subdivisions, under one or more of which it is pos- sible to classify any paper on any subject within the domain of science. Conversely, when any subject is to be investigated, the plan is first to find the subject-heading in the classification schedule and to use the number there given instead of a page number in look- ing up the grouped references in the body of the catalogue, the pages of which bear the schedule numbers in addition to page num- bers. As, with the exception of additions, these subdivisions and numbers are the same from year to year, this method will materially aid in investigations covering a term of years. Regional bureaus are established in the following countries: Austria, Belgium, Canada, Cape Colony, Denmark, Egypt, France, Great Britain and Ireland, Germany, Greece, Holland, Hungary, Italy, India and Ceylon, Japan, Mexico, New Zealand, New South Wales, Norway, Portugal, Poland, Queensland, Russia, South Aus- tralia, Sweden, Switzerland, Victoria, Western Australia, and Fin- land. Authority over all questions of methods and administration is vested in an international convention to be held in London in 1905, I9g10, and every tenth year following. "912 A.D) 214 SMITHSONIAN MISCELLANEOUS COLLECTIONS [VOL. 45 Failing in governmental appropriation for the establishment of a regional bureau in the United States, the Smithsonian Institution felt obliged to render its fostering aid to the project, otherwise the enter- prise might have been abandoned. The Institution was at the time enabled to devote a sum of money to carry on the work here pro- visionally, which, together with gratuitous aid, rendered it possible _ to make a start. The limited means at the disposal of the Institution at first greatly hampered the work in this country, but, beginning with July 1, 1903, it was possible for the Institution to devote a sum of money to the purpose which had hitherto been otherwise employed. This will not only enable the Institution to deal more promptly with the current publications in the United States, but will render it pos- sible to make good the omissions occurring from January I, I9got, to the present time. With the exception of zoology all of the volumes of the first year’s issue of the catalogue have been published, together with astronomy of the second year’s issue. It was at first hoped that valuable aid for the work in the United States could be obtained from existing card indexes in the different scientific branches of the government, but experience has shown that, owing to the dissimilar methods used, it is practically as difficult to transpose, verify, and properly classify the references obtained in this manner as it is to procure the data at first hand. To give some idea of the extent of the work in this country it may be said that approximately 27,000 classified reference cards have been forwarded by the Smithsonian Institution to the London central bureau. The method here employed is briefly this: A numbered card rec- ord file is kept of the titles of the periodicals published in the United States which are likely to contain matter on scientific subjects; this record is systematically gone over at regular intervals and the peri- odicals called for from the Smithsonian library, which aims to re- ceive all such publications. The contents of the publications themselves are indexed separately on cards, and each card duplicated as many times as necessary in order to send to the central bureau (besides the regular reference by authors’ names) one card for each of the subjects into which the paper is classified. Duplicate author reference cards, on which are noted the assigned classification, are kept for file, and a record is kept of the entire publication on the periodical cards referred to. By this method it is possible not only to duplicate the work at any time, but to check and make good any omissions. NOTES 25 Separate publications and books are treated in like manner in regard to classification, although the methods of obtaining notice of their appearance is necessarily different. To classify properly into minute subdivisions, such as are em- ployed in this work, the vast amount of scientific matter appearing in this country is a difficult task, but every effort is used to make the references exact, and where an intricate question is involved the advice of a specialist is asked. At the central bureau a corps of referees are employed, a specialist for each science, who, to guard against error, review each reference before publication. This catalogue now furnishes aid to both librarians and students who have long needed a concise subject index to the great and ever- increasing scientific literature of the day. The United States leads in the number of subscribers to the cata- logue, the number being 97, equivalent to over 72 complete sets. The yearly subscription to the full set of seventeen volumes is $85. The individual volumes may be subscribed to for a sum pro rata to the cost of the full set. SMITHSONIAN SEAT AT NAPLES Dr. CHARLES SEDGWIcK Mutnot, Professor of Histology and Embryology in Harvard University, occupied the Smithsonian Table at the Naples Zoological Station in the autumn of 1902, devoting his time to procuring extensive series of embryos of Torpedo ocel- lata, Mustelus levis, Petromyzon, and Amphioxus, and young stages of Pristiurus and Scyllium. Owing to the exceptional opportuni- ties for procuring collections at the Station, Doctor Minot reports that he was able to obtain fine series of carefully selected stages of specimens, which, arranged in serial sections, have been added to the embryological collections at Harvard University, where they will be accessible to all competent investigators, and will serve for many years for researches in comparative embryology. Dr. CHarLes W. Haraitt, of Syracuse University, was granted the occupancy of the Smithsonian Table at the Naples Zoological Station for the months of March, April, and May, 1903, during which period it was possible for him to finish a research on the early development of Eudendrium, a genus of hydroids of whose develop- ment comparatively little has hitherto been known. The manuscript of the full report is now in the hands of the editor of the Zoolog- isches Jahrbuch for publication, and will probably appear during the current year. Doctor Hargitt found that while, in the main, the course of development in these hydroids is comparable with that Z 210 SMITHSONIAN MISCELLANEOUS COLLECTIONS [VOL. 45 of others of similar character, there are many features which are unique. The early development of the eggs and their migration into the gonophores has long been known; but in manner of growth and the physiological phenomena associated therewith, some rather unusual changes take place, which have not been generally recog- nized. In the phenomena of egg cleavage there seem to be pecu- liarities known only of special cases among other groups of inver- tebrates, which are probably due to very similar conditions in the history of the eggs. Doctor Hargitt has likewise found among this group of ccelenterates rather distinctive peculiarities, not hitherto recorded, as to the later cleavage phase and the formation of the germ layers of the embryo, which have an important bearing on problems of phylogenetic and speculative character. THe NosBet PRIZES The following data relative to the Nobel Peace Prize, to be awarded December 10, 1904, have been officially communicated to the Institution. This prize may be accorded to institutions, associations, or indi- viduals: The names of candidates for the 1904 prize must be pro- posed to the Nobel Committee of the Norwegian Parliament, before February first of that year, by one of the following named persons, who are held to be duly qualified for such action: Members of the Nobel Committee; members of parliament and government officials of different countries; members of the inter- parliamentary council; of the commission of the International Peace Bureau, and of the Institute of International Law; university pro- fessors of political science, of law, history, or philosophy, and per- sons who have received the Nobel Peace Prize. A statement of the grounds on which the proposal of a candidate is based should be submitted, with any related documents referred to. Only the printed works of a candidate are accepted as cre- dentials for a prize. Further particulars as to the award may be obtained by application to the Nobel Committee of the Norwegian Parliament, Victoria Terrasse 4, Kristiania. The Institution has also been officially informed that the Nobel prize in physics for 1901 was awarded to Prof. W. C. Rontgen of the University of Munich, for the discovery of the rays which bear his name. The prize in physics for 1902 was equally divided be- tween Prof. H. A. Lorentz of the University of Leyden and Prof. P. Zeeman of the University of Amsterdam, for their researches on the influence of magnetism on the phenomena of radiation. SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 45, PL. LVI CLUSTER OF AMETHYSTINE QUARTZ FROM RIO DO SUL, BRAZIL. (Weight 400 lbs., height 24 in., breadth 27 in.) GIANT QUARTZ CRYSTAL FROM CALIFORNIA. (Weight 309 Ibs., height 18 in., length 30 in.) NOTES 207 The Nobel prize in chemistry for 1901 was awarded to Prof. J. H. van’t Hoff, of the University of Berlin, for the discovery of the laws of dynamic chemistry and of osmic pressure in solutions. The prize in chemistry for 1g02 was awarded to Prof. E. Fischer, of the University of Berlin, for his synthetic researches in the groups of sugars and of purines. GEOLOGICAL COLLECTING EXPEDITION Dr. GreorceE P. Merritt, Head Curator of the Department of Geology of the National Museum, left Washington on August 16th and returned October 2d, going first to Ishpeming, Michigan; thence to Portland, Oregon, from which point he proceeded in turn to San Francisco, Ogden, Salt Lake, Pocatello, Laramie, and Den- ver. Field excursions for the purpose of obtaining collections were made in the Bad-lands of North Dakota, the Lower Madison val- ley of Montana, and near the town of Opal in southwestern Wyom- ing. At Ishpeming was obtained a large and exceptionally fine series of the remarkable contorted and brecciated hematite jaspers, such as have been described by Dr. C. R. Van Hise in volume xxvii of the Monographs of the United States Geological Survey. From the Quaternary bluffs on the Lower Madison river, in Montana, was obtained a large series of the beautiful wood opal for which this locality is noted, and from southwestern Wyoming an extensive series of the peculiar hot-spring deposits in the form of cylindrical trunk-like masses, lined with quartz, agate, and calcite. Through the generosity of Messrs. J. R. Wharton and A. P. Pohndorf, of Butte, Montana, there were obtained some fine ex- amples (the largest 24 inches in length) of the smoky quartzes found in the Little Pipestone district, Silver Bow county. In San Fran- cisco there was procured a magnificent milky quartz,measuring about 30 by 18 inches and weighing 309 pounds. In size and crystalline development this is believed to be one of the finest specimens of the kind in the United States. (See plate Lv1.) From dealers in San Francisco, Denver, Manitou, and Deadwood was obtained much miscellaneous material, the more important being a large mass of Wyoming moss agate, weighing 536 pounds. Through the courtesy of Messrs. Robert Forrester and W. B. Put- nam, there were obtained small but exceptionally fine series of cave aragonites from silver mines in Utah. The trip, on the whole, was highly profitable, material being ob- tained which probably could never have been brought to the Museum s ' : 218 SMITHSONIAN MISCELLANEOUS COLLECTIONS [VOL. 45 through the ordinary channels of trade or through collectors, owing to the expense of getting it out or to lack of knowledge concerning * the needs of the Museum and the possibilities of its exhibition series. A REMARKABLE AMETHYST GROUP Among the numerous accessions by the Department of Geology of the National Museum, few are of more striking interest than a recently received large and remarkably perfect group of amethystine quartz from the well-known Brazilian locality of Rio do Sul. The specimen weighs approximately 400 tbs., and, as shown in the ac- companying illustration (plate Lv1), is in the form of a botryoidal mass completely studded with short but broad pyramidal termina- tions, the prism faces showing but slightly. In size the crystals are remarkably uniform, varying from 30 to 50 millimeters in breadth and some 25 millimeters in height. The color is deep amethystine, fairly uniform, and the points, considering the great weight of the specimen and the distance of transportation, are in an exceptional state of preservation. So far as records show, this is one of the largest if not the largest group of its kind in existence. It will form a part of the exhibit of the Department of Geology at the approaching Louisiana Purchase Exposition. INTERNATIONAL EXCHANGES There was an unusual increase in exchange transmissions to and from every part of the civilized world during the year ended June 30, 1903, being an excess over the previous year of 24,421 packages, or 19 percent. The total weight aggregated 559,718 pounds for the same period, or an increase of 41 percent over that of the fiscal year 1901-02. ‘The countries which have most extensively cooper- ated with the United States in the interchange of official and scien- tific publications are Great Britain and Germany. The results of the year ended June 30, 1900, show that more parcels were ex- changed with the former than with the latter, but subsequently Germany exceeded Great Britain until the year 1903, when Great Britain again took the lead. It is of interest to note that the total difference in the number of parcels exchanged between the United States and each of these great countries during the last four years is but 344. Iron MINE aT LeEsLie, Missouri In April Mr. W. H. Holmes made a visit to Leslie, Missouri, for the purpose of studying certain traces of ancient operations re- NOTES 219 ported to occur in an iron mine near that place. Most interesting phenomena were encountered, the ancient aborigines having pene- . trated the ore body in many directions and to surprising depths, the purpose being, apparently, to obtain the red and yellow iron oxides for paint. Many hundreds of mining tools of stone were found in the ancient tunnels. The Leslie iron mine study has an interesting bearing on the technic and industrial history of the tribes. It has been a matter of much surprise, as the investigations of the ancient ° mining and quarrying have progressed, that the aborigines, seem- ingly so non-progressive and shiftless, should have conceived and carried out really great enterprises. The technical knowledge and skill displayed are of a low order indeed, but the work accomplished indicates remarkable persistence and demonstrates the existence of native capacity of high order. EXCAVATIONS AT LANSING, KANSAS In October, 1903, explorations were undertaken at Lansing, Kan- sas, with the view of determining the age of the human remains found embedded in loess-like formations near that place. The for- mations were extensively trenched by Mr. Gerard Fowke under the direction of Mr. William H. Holmes, and the conclusion was reached that the remains were of exceptional antiquity for America, but that they could not with certainty be assigned to a definite geological horizon, and that they were probably of post-glacial time. The pur- pose of the excavation was to expose the formations containing the human remains so fully that geologists of all ways of thinking might study them to advantage, thus preventing the adoption of conclu- sions based on inadequate observations. An account of Mr. Holmes’ researches in this interesting field appears in the Smith- sonian Report for 1902. The geological problems connected with this site -have been exhaustively treated by Prof. T. C. Chamberlin in the Journal of Geology, Chicago, vol. x, No. 7. FossiIL BoNE-BEDS AT KIMMSWICK, MISSOURI Mr. W. H. Holmes, Chief of the Bureau of American Ethnology, with the assistance of Mr. Gerard Fowke, has made examinations of the fossil bone-beds at Kimmswick, Missouri, with the view of determining whether there was satisfactory evidence that man was contemporaneous with the mammoth and the mastodon in that re- gion. These researches dealt with the important and ever-recurring question of the antiquity of man in America. It has been the aim of the Bureau of American Ethnology, and especially of the present 220 SMITHSONIAN MISCELLANEOUS COLLECTIONS [VoL. 45 Chief, to occupy conservative ground with respect to this subject and to so fully scrutinize the discoveries and reputed discoveries re- ported from time to time that erroneous interpretations should not prevail. No traces of man were found in direct association with the fossil remains. A STuDY OF THE SWORD Mr. Paul Beckwith, Aid in the Section of American History, United States National Museum, is at present making a thorough study of the sword and other cutting and thrusting edged weapons. A large number of these weapons belonging to civilized nations have come into the possession of the National Museum through the War Department. Many swords of different types from southeastern Asia, Malaysia, the Philippine islands, and the West Indies have been acquired through the kind interest of army officers in those areas. With the material already in hand it will be possible to cover the history of the evolution of this most interesting weapon. QUARRIES IN THE SOUTH AND WEST Early in May Mr. William H. Holmes made a trip to Georgia and Alabama for the purpose of examining quarry sites and cav- erns anciently occupied by aborigines. Examinations of aboriginal flint quarries and sites of stone implement manufacture were also made by Mr. Holmes in southern Indiana and in eastern Kentucky. RECENT PUBLICATIONS OF EHE SMITHSONIAN INSP EUTION CoNTINUED FROM List 1846-1903, No. 1376. No. Title. Series. Price. 1377 Annual Report of the Smithsonian Institution, eepaper TOOT STOOD» .yeerse, cat Giclee eae ee oan LOO? fe Sarr 1378 Proceedings of Board of Regents; Report Execu- tive Committee; Acts of Congress, 1903....... R. 1902 05 1379 BADEN-PoweELLt, B. F. S. Recent aeronautical progress, and future of aerial navigation...... R. 1902 .02 1380 WricHt, Witsur. Some aeronautical experi- AMVC ES PEE a yati, eiete CREA Riess Se ER Ee Coen OO2 02 1381 Hare, G. E. Stellar evolution in light of recent PESEATEMY |S Airs sales onan el eae cites oe eee UN OO2 .02 1382) EVAL Mies). eeAu mews sOlaintheo tae seed see mL OOS 02 1383 LeBEDEW, Peter. Experimental investigation of PLRessuine Of lightest eee ae ee nem OOS .02 1384 Cox, JoHN. Comets’ tails, the corona, and the aliroOra borealiisenay seuss: Eka eee eee ek ER eLOOZ 02 1385) IGA NGDE Ye Seles i GOO ESeelnc: coarser eee oe R. 1902 02 *For sale by Superintendent of Documents, Government Printing Office. No. 1386 1387 1388 1389 1390 1391 1392 1393 1304 1395 1396 1397 13908 1399 1400 1401 1402 1403 1404 1405 1406 1407 1408 1409 1410 I4II 1412 1413 1414 1415 1416 1417 1418 1419 LIST OF PUBLICATIONS Title. BECQUEREL, HENRI. Radioactivity of matter.... Dewar, JAmMes. History of cold and the absolute ZCLOs: Bikers ne McKEnpRICK, ia G. ieeecverttnental. shoneties, ae Maver, WiLLIAM Jr. Wireless telegraphy....... Grark, @ MM: Delpherage=- ae SHARIA.) eVARRTS: myolution: of SS eurnioei eal ICLEASE Se Seen eve. a east oohsah chs: ais naval Sent tare ANDERSON, TEMPEST, and Fett, J. S. Recent eruptions of Soufriére and visit to Mont Pelee. Russe LL, I. C. Volcanic eruptions on Martinique AUTNG aes toe HI GET tee en cited en rcpsusts Chair eteas. re ee aie: Hao Hotpicu, Sir T. H. Progress of geographical knowledge ake Drorgerks WeELts, H. G. Discovery of thie fainans® DastrE, A. Life of matter. ; Macnamara, N. C. Crinislbey 5a man ee an- CHROMOLE APeSit.:u-ncvr days ”rccraedeny ie sere eo tey 9 okays Stoke Gaupry, ALBERT. Baoussé-Roussé explorations. Study of new human type, by M. Verneau.... Hotmes, W. H. Fossil human remains found near Lansing, Kansas. SKEAT, W. W. Wild iebes. oi Malay. ss cainealan: Jounston, Sir H. H. Pygmies of great Congo OES Uae Mera AIS accion saertea nahere ee oS Sirsa emia eee Ta SAEHORD Maven Crliciine an CatSepeODlenem ieee sere Jacos, Georc. Oriental elements of culture in the Oceldenitaeasere at a Means, T. H. Nile reservoir ce at GA atari Burr, W. H. Panama route for a ship canal.... Bateson, W. Problems of heredity... Howes, G. B. Morphological fictiiod: and secur progress in zoology Route, Louts. wee ae eRe ter GrosvENor, G. H. Reider in Alasten Grecory, W. K. A marine in ea te ee GrieertT, G. K. John Wesley Powell. IsrAeL, Oscar. Rudolph Virchow. : ScHUMANN, Victor. Absorption and emission of DUTT oe ties A Ute tes Sr sheet cttn es Soap ols taiesot ese True, F. W. Whalebone whales. (J press.)... Smithsonian Contributions to Knowledge, Vol. XO NONTIME HNN (LIU ME ES Ge eee eneen tL) a meee des usyare, sre ehoisitsns Annual Report of Smithsonian Institution, 1901— 1902, Pt. 2, National Museum. (Jn press.)... GraBau, A. W. Phylogeny of Fusus. (Jn press.) List of foreign correspondents of Smithsonian Institution, 1903. (Jn press.).. Smithsonian Miscellaneous Goll one ‘(Ouartenly Sea) Me OMG gt PEUGESE Tea 2h, ich acts. tonal tions aise iteyae Rare ri 7 ri ri ye Po Sertes. ER TOO? . R. 1902 R. 1902 Pe eaenOO2 R. 1902 R. 1902 fe 1902 1902 1902 1902 a 1902 1902 1902 1902 rim Pe 1902 1902 mi 1902 1902 1902 1902 1902 1902 1902 1902 1902 1902 S.C. XXIX S.C S.C. XXX M.R. 1902 M.C. XLIV MC. XLV bo No = Price. -02 -02 .02 -02 S)) bo b& oe One b&b Ome a0 b& bw bo wb our One bn v BN 9 On 222 SMITHSONIAN MISCELLANEOUS COLLECTIONS [VoL. 45 1420 Mittrr, G. S., Jr. Seventy new Malayan mam- THATS 5 nye GRC acd faiw’o.0 Sah oF ene nae ere M.C.XLWi .40 1421 Aspot, C. G. Recent studies of solar constant of LaGiation ene eee M.C. XLV 05, 1422 Axpot, C. G. New: cotaatae and fercontal tee scope of Astrophysical Observatory....... MCX. .05, 1423 Trug, F. W. On some photographs of iinines Ai back whales. he bee rok eon VISSER] .05. 1424 Lucas, F. A. A ckeleton ior THeskeromet’ VRC SGI; 05 1425 Lucas, EF, Aj” Aa mewrbiesiosatitas.s.- NEC aE Es .05. 1426 “Hotmes, W. H. Shell ornaments oan Kentucky andes ViexiCOneeerine MIC, SE AW. 05 1427 Merritt, G. P. ‘The felaciall pathiole in aie, Nae tional Museum. ae MLC. IEW .05 1428 BARTSCH, PAUL. Moiese on ihenene ue Siictetee: a @oltambtatrhs ak ae rete sore cee eee ee a nV Le GGT: .05. 14290 FEwKEs, J. W. Preliminary report on archeolog- KEM Teriyoy (oy WWVESIE Miah. bon ceopouno noo oced noes M.C. XLV sis wAlrxo) (Cisne), (C5 WI, (Coalkesoverc) ayavel A erclovelllenme, 55 o5c05 IMLIC, XULAY 05 1431 EIGENMANN, C. H. South American fresh-water ; WINES 4c 008 Jets Soe Gee xa 05 1432 JENINGS, F. H. "Korean: enddeecees, ER! 50) BMC SIL 10 1433 BurNsIpE, HELEN W. The Hodgkins fianeite MICS IEA 05 1434 BAKER, ASB: Breedingvot black beats. .......5.. MG. Sob 05 1435 Frint, J. M. Chinese medicine. . Seta eros Le CIE 05 1436 PHaten, W. C. Rocks of Greene Re ennai Cae XG ams) 1437 LANGLEY, S. P. Report of Secretary, Sinithsonian Institution for 1902-03. yu R. 1903 10 1438 SHALER, N. S. Campanicen ae earth andl moon. Cie press) eerie S:@ SOO 1439 Assot, C. G. Solar colnee eevedition: DN stroptive: ical Observatory. (Jn press.).. ee 1440 Botton, H: C. Bibliography of nem 1492— 1902. 2d. supplement. (Jn press.). Beano WIAGS SIE 1441 Travers, M. W. Researches on low Lanperitdaes: Partars set Ghn press) scdelonsr tet tautdies e..e hoki ek es ea The following numbers included in the available publications in previous lists are now out of print: Nos.: 50, 98, 440, 535, 552, 553, 630, 749, 862, 863, 865, 873, 880, 903, 908, 909, 910, 915, 929, 940, 945, 956, IOIO, 1014, 1018, 1060, 1062, 1132; TE53, 1198, 1204, 1226,)1227, 1237, 1293, 0l271,. v7. 1277, 1295, 1342, 1345, 1346, 1347, 1356. RULES OF DISTRIBUTION The publications of the Smithsonian Institution are: 1, CoNTRIBU- TIONS TO KNOWLEDGE; 2, MISCELLANEOUS COLLECTIONS; 3, ANNUAL REPORTS; 4, SPECIAL PAPERS. LIST OF PUBLICATIONS 223 No sets of these are for sale or distribution, as most of the volumes are out of print. The volumes of Contributions and of Miscellaneous Collections are distributed only to designated public libraries and to learned institu- tions in this country and abroad. A small.edition of the papers in these two series is printed separately for sale or exchange. The Smithsonian Reports are distributed by the Institution to libraries throughout the world; may be had by purchase at cost from the Superintendent of Documents, Washington, D. C., and may also generally be obtained free of charge from the applicant's Mem- ber of Congress. The papers printed in the Annual Reports are published separately for free distribution. The prices affixed to them in the lists are given as a basis for exchange. As the number of copies of Smithsonian publications is limited and entirely inadequate even to supply the demand from libraries, an applicant should state explicitly the ground of his request. Publications should be ordered by the serial number. Remittances should be made payable to ‘‘ Smithsonian Institution, Washington, D. C.” Applications for a public library should state the number of vol- umes in the library and the date of establishment, and have the en- dorsement of a Member of Congress. For the publications of the Bureau of American Ethnology, applica- tion should be made to the Bureau. Reports of the Smrrusonran Institution, NATIonAL Museum, BuREAU OF AMERICAN ETHNOLOGY, AMERICAN HISTORICAL ASSO- CIATION and DAUGHTERS OF THE AMERICAN REVOLUTION are for sale by the Superintendent of Documents, Washington, D. C. ; ai Vie ALY A 4 ‘i ‘cael: OA : i Lf! i, ‘ ae eh ‘: waren on . i lee Ne vm Cre ey i. . 7 \ Loe { “ie ide Ra ya sean Ap, eee “eh ; th i. j 7 } ‘4 wn Wha batt : v mn risen eT ge Ky ed iM Rew ah ill ” = eat vat Beso SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 45, PL. LVII THE PERSONAL EQUATION MACHINE, TRANSIT, CHRONOMETER, AND CHRONOGRAPH. WOES 45 PARTS III AND IV SMITHSONIAN MISCELLANEOUS COLLECTIONS QUARTERLY ISSUE OCTOBER-DECEMBER, 1903 A METHOD OF AVOIDING PERSONAL EQUATION IN TRANSID SOBSERVATIONS By Ss. BabA NGEEY Many years ago the writer devised and published’ preliminary accounts of a method for avoiding the so-called “ personal equa- tion” in observations of the time of transit of stars. The following quotation from the article cited will give the fundamental idea described in the former article, to which the present one adds some more recent experiments and slight modifications. The writer, after speaking of the comparative merits of and objections to a photographic method, and one where the star is seen intermittently projected on wires which are illuminated by consecutive flashes, observes : “There is one particular case, however, where the result is the same for both, that, namely, where when the flash comes, the star © is on the wire and bisected by it; in this case we know its position as accurately by the eye, as if we bisected its image on the plate by the wire of our micrometer. If we suppose, then, that by a happy accident, the flash came just as the star was crossing the first wire, this wire would be sharply defined on the disc of. the star and bisecting it, and a simultaneous record on the chronograph (made without the intervention of the observer) would evidently give us the same result as though the star had recorded its own passage by an electrical contact. Further, we may particularly notice that it is immaterial whether the star was at rest, when thus seen, or in motion. Now what we have just supposed as a single case of a favorable chance among hundreds, it will be our task to make occur, whatever the wire interval, and for any star observed.” In illustration let it be imagined that the observer is watching the passage of the star through the field of his telescope, but that * American Journal of Science and Arts, July, 1877, pages 55-60. 225 226 SMITHSONIAN MISCELLANEOUS COLLECTIONS [VoL. 45 he has no illumination of that field sufficient to discern the wires of the tally. Suddenly a flash illuminates the field, and we may suppose that by accident this occurs just at the instant when the star image is bisected by the central wire. If now the flash had recorded itself automatically on the chronograph, an observation independent of the ordinary personal equation would have been obtained, since we know where the star was at that recorded time of observation. The device I have already described consisted in making this unlikely accident occur for any star, and for numerous wires of the tally. At the time when it was first devised, such a method consisting not in correcting time observations for personal error committed, but in preventing the committal of such errors, was unique, and it had for the writer the good fortune of attracting the interest and commendation of Professor Clerk Maxwell. But the writer was prevented by distinct duties from developing it, and more recently Repsold and others have attacked the same problem in another way, with much success. Still it is possible that this older method may prove of value, and I have thought best to give this brief account of some recent trials of it. The instrument whose purpose is to illuminate the field at the required instants, was constructed from my design by A. Hilger, as follows: A conical pendulum consisting of a graduated rod with a heavy ball which may be set at any height on the rod, is suspended by two pairs of thin flat springs acting as a universal joint. At the lower end of the rod is a needle which, as the pendulum revolves, governs the rotation of a grooved arm carried by a clock-work which also drives a drum carrying fine platinum brushes which make instantaneous electrical contacts at each revolution. (Plate Lvtr.) Thus the intervals between these contacts are governed by the rate of the conical pendulum, which itself goes fast or slow according as the ball is raised or lowered on the rod, this rod being graduated so as to correspond with the declination of any star between 0° and 60°. Knowing the declination of the star, it can be arranged beforehand, therefore, that the interval between contacts is equal to the average interval between transits of the observed star across the successive wires of the tally. By appropriate electrical connec- tions this system of contacts is caused to produce a system of instantaneous discharges in a vacuum tube, which is used in place of a lamp to illuminate the cross wires of the telescope, and if the star be first seen, for instance, half way between the first and second LANGLEY] PERSONAL EQUATION IN TRANSIT OBSERVATIONS 227 wires of the tally, it will be seen at the second illumination half way between the second and third wires, and so on. We have thus arranged that flashes shall occur at recorded in- stants, without limit as to number, and at intervals such that if one, by chance, occurs when the star is bisected by a wire, it will be seen bisected by all the wires in succession. It now remains to arrange that this bisection shall actually occur. This is accom- plished by providing an independent control of the position of the contact points under the clockwork, so that with a cord in the hand of the observer he can cause the series of contacts to occur sooner or later, without altering the interval regulated by the pendulum. Thus if the flash comes a little too early for bisection of the star image by the first wire of the tally, a slight adjustment is made, delaying the succeeding flashes, and with a little practice one or two adjustments suffice to secure bisection. To add the weight of independent observations, several displacements and readjust- ments may be made during the passage of the star over the tally. As the times of all the flashes are recorded automatically upon the chronograph, an independent signal is made by the observer to mark each flash which has revealed a satisfactory bisection. I have thus far given the device substantially as described in the early paper. To test the value of the method under circum- stances admitting of distinguishing error, an artificial star was arranged to move by an accurate clockwork at about the apparent rate of an actual equatorial star. The shaft which carried the screw by means of which the artificial star was moved, had upon it an arm provided with an adjustable point which instantaneously broke an electrical circuit at each revolution. By careful adjust- ment, with the artificial star stationary, it was arranged that exact bisection took place at the middle wire of the tally, and with the point in position to break contact. Thus the artificial star was caused to record its own time of transit upon the chronograph wholly without personal equation of time. A transit instrument of 234 inches aperture and 48 inches focal length was kindly loaned me by the Superintendent of the United States Coast and Geodetic Survey, for the purpose of making an experimental test of the personal equation machine. In reconsidering the theory of the method a possible cause of failure appeared. The observer sees the star continually moving across the limited field of view, and thus his mind may uncon- sciously be influenced by the perception of motion, just as it un- doubtedly is in observations by the usual method. In order to [VoL. 45 SMITHSONIAN MISCELLANEOUS COLLECTIONS oSo0° o— ££0°0 6z0° 0 +- 0go" oO gor’ o-+ oSo'o ESor@ || eer O— CIOROM IO LONO- {So- 0) | cor-0— “Uva ZOO" Lzo er ie; tgo° £1 = gI Lo: €90° + £o° zo" — vio" zI° — SI 100° of0° + zo" Os goo: oP == 990° OP tI ozo" 6to0" = glo: £o- go° Su = Tie Come gto° 60. f1 9So° Sgo" -++ goo* OI° Lo: Co == || Ryo" GOs, | ¥90: Gor == || Olea TO es zI 6z0° gSo° + bS1- gz" £o° 90.8 —— zSo° £o° — | tor Cl Oi OMe II 160° ozi° + bri: Cae to’ 60° — gII° v1- — tgo: or’ — vzo° €1° — OI gfo° Looe — bSo- jt” fo" 200 zoo" 0) == | Ofer Go? = +60: oz 6 Szo° ior == gSo° . So" fo" Coy == zgo° 90° + tio" 220: e—— 910° 60" — 8 Ito" Zio" — glo: £0- So" Oe == || feo So. — | v£o- | Sor + blo: grr Li Lzo° Zoo” -- blo: gI° go" co goo* Oo —— tbo: 907 —— bio" a> 9 ZZO° [xo gor" 00° So" 00° == gSo- XO g10° 00" gSo° So: — S 1Zo° COOmEES = t60° oz" TI Of += zto° OMe ggo° Zo" gto: 90° — v Sgo° SO 990° to: zo" a glo: or — gzo° 10° b£o: vr" — 5 110° ovo’ + gto: 90° £o° zo" — ggo° (s{o) 990° co" -- 60° oz — Zz ££0";0 too’;o— 980°;0 ZO";0-- g0",0 £0",0-- ZOO’, ZO;sO— || VOOssO ZO;;0— | VEO".0 Vio I zo6x ‘Se Ajnf zo6r ‘Se Ajnf zo6x ‘Sz Ajnf zo6r ‘Sz A[nf zo6r ‘Sz Ang ; zo6r ‘Sz Ajnf aed IOI eae IOANG Soot, ION aeons roNy coca IOI teen IOANG seach uvoyY WOT pearesqg uvoy, WOT pearesqa, ubvofY WO | pearesqaQ, uvof WOT pearesqQ, uboy UOT | paarasqaQ, uvosyy WOT pearasqa uoIeIAsq jo 1011 uol}eIARC, JO 10114 uolIeIAaq Jo 10119 uoneiAsqd Jo 1011q, UOTJBIAR(T yo 10114 uoneiasqg jo 10114 qeig Surinssq() aUuIyoR yy aeyg Surnosqo auryoeyy TES Surmosqo re auIyoe S Aue UIIaIUyT sUIyOeTT uonenby jeuosieag Ayquay uta} Uy sUuryoR yA uonenby [euosiag Ayjuaiusayuy auryoeyy uoljenby [euosisg uoljenby [euosiag YUNA, NoWIA, Poe Tens | uonenby [euosisg yA, | MoyITA, POM [esp | uonenby [euosiag WIM | pours "WN ‘19Ar9sqQ ST maaan ‘9439800 | ‘Ww +) =Oyhe ‘dOHLAW GHAOUdWI AD GASN ANIHOVW NOLLVAOA TWNOSYAd AO LSAL mou Poy [ens~) AdATBSqQ LANGLEY] PERSONAL EQUATION IN TRANSIT OBSERVATIONS 229 overcome this objection a rearrangement of the apparatus was made so that instead of intermittently obscuring the wires of the tally, the wires were now continuously illuminated as in the usual method, but the star itself was obscured by a shutter, except at the instants when the wires had formerly been illuminated by the glow tube. As thus modified the observer by this second method sees the star only during well-separated intervals of a hundredth of a second or less, so that all appearance of motion is avoided. This device, suggested by Mr. Abbot, appears to be an improvement on the former method, and has been used in the observations presented in the accompanying table. These observations were made by three observers, €..Gs A., F: E. F., and N. E. G., of whom:F. E. F: was most experienced in observing transits without the personal equation machine, and accustomed to its use by the former method, but less practiced than the others with the machine as now employed. The general result appears to be in favor of the personal equation machine. In the case of the observer C. G. A., whose usual equation was about — 0.10 seconds when observing without the machine, the error became only —o0.016 seconds with the machine; and the average deviation was reduced slightly. Sim- ilarly with the observer N. E. G., whose direct observations were usually rather variable, but which now yielded an average personal equation of + 0.106 seconds, the employment of the machine was of marked advantage, reducing his personal equation from 0.106 to 0.029 seconds and the average deviation of it from 0.080 to 0.033 seconds. With the observer F. E. F., the usual personal equation was small and the average deviation of it was slightly reduced. So far as it has been tried, then, the advantage of observing by the aid of this device seems to be marked. ON A COLLECTION OF FISHES MADE BY einai aay OWSTON IN THE DEEP WATERS OF JAPAN By DAVID STARR JORDAN anp JOHN OTTERBEIN SNYDER The writers have recently received from Mr. Alan Owston, of Yokohama, well known as a shipmaster and as a collector in natural history, a very remarkable series of fishes obtained on long lines in the depths of Sagami bay and Suruga bay, Japan. The present paper contains notes on these species, together with field notes of Mr. Owston. The specimen of Mitsukurina owstoni and some of the others have been sent to the United States National Museum. The rest are in the collection of Leland Stanford Junior University. Family ScyYLLIORHINIDE PRISTIURUS EASTMANI Jordan and Snyder, new species “ Gobozame ” (Burdock Shark) (2rATENIOXS) One specimen, a female, 345 millimeters long, from off Izu. Type No. 7740, Ichthyological collection, Leland Stanford Junior University. The body is very slender and elongate, the head small and narrow. Head measured to first gill opening 6% in the total length; depth equal to half the distance betwen tip of snout and third gill open- ing, 11% in total length. Snout 2% in head, rather acutely pointed. Nostrils with pointed flaps on the edges, which meet and curve inward, making the nasal cavity tube-like with an outer anterior and an inner posterior opening; distance between outer anterior openings equal to cleft of mouth measured on upper jaw, or distance between anterior edge of eye and spiracle; distance between posterior openings equal to half the longitudinal diameter of eye. Width of mouth equal to length of snout; distance between anterior edge of mouth and tip of snout equal to distance between center of pupil and anterior gill opening, or to width of interorbital space. Teeth each with 7 acutely pointed cusps, the central of which is twice as high as the others; lateral cusps growing suc- cessively smaller toward outer edges of tooth; three central cusps 230 JORDAN-SNYDER] COLLECTION OF FISHES FROM JAPAN 231 distinctly visible, the others partly covered by the gums. Upper surface of tongue and roof of mouth with minute prickles. Spiracle on a level with eye, its greatest width equal to its distance behind eye, one-half the width of the first gill opening. Length of gill area equal to distance between eye and first gill opening; second, third, and fourth gill openings about equal in width, the fifth narrowest. Skin closely covered with minute, trilobed scales, each of which has a central keel. Upper edge of tail with a keel be- ginning an eye’s diameter behind base of second dorsal and extend- ing posteriorly a distance somewhat greater than length of head; keel armed by two rows of enlarged, tooth-like scales, the inner edge of each scale having a sharp cusp; about three rows of scales similar to those of the body between the rows of larger ones. First dorsal fin inserted above posterior edge of base of ventral, its distance behind tip of snout equal to 24% times the length of head; length of anterior edge of fin equal to length of snout. Second dorsal inserted anterior to end of base of anal. Free edges- of both dorsals straight. Caudal fin with a notch on the ventral edge near end of fin. Free edge of pectoral slightly convex; when depressed the fin reaches a little over half way between origin of pectoral and ventral. Ventrals sharply pointed posteriorly. Tip of anal reaching a vertical through posterior end of base of second _ dorsal. Color in spirits brownish above with indistinct clouds of a deeper shade, the more conspicuous of which are located as follows: above and a little behind base of pectoral, midway between pectoral and ventral, below bases of dorsals, on upper edge of tail and on caudal fins. Dorsals dusky, their anterior edges dark brown. Anterior edges of caudal, anal, and pectorals dark brown. Free margins of dorsals and of anal white. Pectorals dusky above, the free margins white. Tongue and inside of mouth without dusky color. The following measurements are in hundredths of the total length (snout to tip of caudal fin): Head measured to first gill opening .15; depth .085; depth of caudal peduncle .o4; width of first gill slit .017; longitudinal diameter of eye .04; length of snout .07; an- terior edge of mouth to tip of snout .065; distance between nostrils .02; width of mouth .075; length of anterior edge of first dorsal .07; of second dorsal .07; length of base of first dorsal .045; of second dorsal .o5; distance from insertion of ventral lobe of caudal to notch .20; from notch to end of caudal .06; width of base of pectoral .o5; length of anterior edge .og; length of base of ven- tral .07. 1) Oo 1) SMITHSONIAN MISCELLANEOUS COLLECTIONS [VOL. 45 Family PSEUDOTRIAKID/= PSEUDOTRIAKIS ACRALES' Jordan and Snyder, new species Oshizame (Dumb Shark) (Pirate LXII) One example, 172 centimeters long, from off Toi in Suruga bay. Type No. 12,903, Ichthyological collection, Leland Stanford Junior University. Depth of body at origin of first dorsal about 9% in total length; head measured to first gill opening 5%; snout 2% in head meas- ured to first gill opening; width of interorbital space 2# ; head broad and flat, the greatest width contained 134 times in the length, the depth 224; interorbital space slightly concave; orbit narrow and long, its diameter 8 in head, the skin below eye with a deep fold; spiracle elongate, the greatest width equal to half the diam- eter of orbit; width of space between spiracle and first gill open- ing 3% in head; height of third gill opening equal to diameter of orbit; width of space between nostrils equal to distance between anterior edge of orbit and spiracle, 4% in head; distance between an- terior edge of mouth and tip of snout equal to space between anterior edge of orbit and middle of spiracle, 354 in head; distance from angle of mouth to its anterior edge measured on upper jaw 2% in head. Teeth on jaws very small, in oblique rows, the obliquity more pronounced toward posterior part of jaws; rows on upper jaw widely spaced posteriorly. Teeth of upper jaw and of tip of the lower, each with a long, sharply pointed, median cusp; a smaller cusp on the outer posterior edge, one or two similar small cusps on the inner anterior edge, and a series of sharp ridges on the outer side; posterior to tip of lower jaw the ridges are less apparent, the central cusp grows smaller, the lateral ones larger, others being added until the teeth become broad and flat with serrated edges. Skin covered with a shagreen of minute, leaf-like prickles, each with a strong, elevated midrib—some of them with a slight denticulation on each side. Distance between origin of first dorsal fin and tip of snout equal to 4% times the length of space between eye and anterior gill opening ; fin highest behind the middle, where its height is contained 534 times in the length of base. Distance between first and second dorsals equal to 3% times the diameter of orbit; base of second dorsal equal to twice its height; laxpaanc, speechless. JORDAN-SNYDER] COLLECTION OF FISHES FROM JAPAN 233 width of free edge of fin equal to height of fin. Insertion of anal slightly posterior to that of second dorsal, the length of its base equal to that of upper jaw, measured from angle of mouth to anterior edge; height of fin contained 2% times in length of base; width of free edge a little less than height of fin. Distance between anal and origin of caudal 14% times diameter of eye; caudal with a deep notch on postero-ventral side, the dis- tance between origin of fin and notch 424 times depth of caudal peduncle at origin of fin. Anterior edge of base of pectoral below third gill opening ; width of base of fin 31% in head; length 2. Origin of ventral on a vertical a little anterior to end of base of dorsal; length of anterior edge of fin 24% in head; free edge 4. Color on both dorsal and ventral surfaces dark gray; fins includ- ing the first dorsal edged with blackish. This species seems to differ considerably from the Atlantic form P. microdon, Capello. The writers have not seen the original de- scription and figure of P. microdon, but rely on that of Gtinther, after Capello, and also on an account of an example from Long Island, published by Bean.t| Comparisons of measurements given by Bean and similar ones from the species in hand follow: _ P. microdon. P. acrales. Snout about twice as long as About 1% times. distance from its tip to mouth. Tip of snout to last gill open- 4% times. ing 5 times in total length. Height of head at angle of 141% times. mouth 11 times in total length. The distance of mouth from Distance from tip of snout to snout measured on axis of fish anterior edge of mouth 17% in equals one-third width of mouth. width of mouth; from tip of The distance from snout to angle snout to angle of mouth 17g in of mouth obliquely taken equals ‘distance between tip of snout and one-fourth the distance from last gill opening; from angle of snout to last gill opening. mouth to its anterior edge meas- ured on upper jaw, 3 times. Distance between nostrils 24% times. equals 4 times the distance be- tween eye and spiracle. *Bean, Tarleton H. The first occurrence of Pseudotriacis microdon, Ca- pello, on the coast of the United States. Proc. U. S. Nat. Mus., vol. v1, 1883, p. 147. 234 SMITHSONIAN MISCELLANEOUS COLLECTIONS [VoL. 45 Length of base of first dorsal 514 times its greatest height, 7 times its greatest height; some- 93% in total length. what more than body height at origin of fin, 83 in total length. Height of second dorsal nearly 2% times length of orbit. twice the length of orbit. Distance of caudal from end 2%. of anal base equals one-fourth the length of second dorsal base. Greatest width of pectoral 1%. equals twice height of anal. Family MirsuKURINIDE MITSUKURINA OWSTONI Jordan Tenguzame (Goblin-Shark) A huge example, 353 centimeters long, forwarded without ex- amination to the United States National Museum. This is the third example taken, the first being in the Imperial University of Tokyo, the second in the Museum of Brussels. Mr. Owston writes: ““T have several duplicates of this now, but the one I send is the longest so far obtained. This shark is taken mostly at Kozu near Odawara—on the chart 35°.16x 139°.17 E. gives the exact spot, where it will be seen there is a bank with 52 fms. on it close to depths of 300 to 400 fms. I imagine these sharks come on to this bank to breed, as mostly females are taken, and in the spring-time only. They are caught in naname (7-mesh) nets, which are set at the upper edge of the bank, so catching the fish when they come up from the deep. Oil is extracted from the liver, but the flesh is used only for fertilizing purposes. “This shark appears to be fairly well known at this particular spot only, where they call it Tengu-zame, Goblin or Elfin Shark. I showed a figure of it to half a dozen fish-mongers at Odawira, only four miles away, and not one of them had even seen or heard of such an animal. Kuma Aoki of Misaki knew nothing about this shark being found at Kozu, although I believe he has fished on the very ground. Kuma took a small one recently off Okinose, 10 miles south of Misaki, by shark lines, and they have been taken on the coast of Izu also by line.” Family LAMNIDE ISUROPSIS GLAUCA (Miiller and Henle) Nezumizame (Rat-shark) Nesumi, rat: “ Probably meaning mouse-colored.”’ Suruga bay. The species is common in Japan. JORDAN-SNYDER] COLLECTION OF FISHES FROM JAPAN bo OW UL Family SQUALIDZ LEPIDORHINUS FOLIACEUS (Giinther) Kanatsubo-zame . One specimen in good condition from off Enoura in Suruga bay. “ Kanatsubo is a vulgar term for a peculiar facial expression.” Family CHIMRIDA CHIMARA PURPURASCENS Gilbert MS. (Gilbert MS. Fishes Collected off Hawaii by the Albatross, 1902.) Kachizame One specimen 132 centimeters, No. 12,902, Leland Stanford Junior University Museum, from off Mishina, Izu in Sagami bay. Mr. Owston observes: “ The fish is called Kachizame, the exact meaning of which I have not been able to ascertain. Kachi may mean ‘a kind of gray color.’ I have a duplicate specimen. They were of a fine purplish black when fresh.” Family PLAGYODONTIDA PLAGYODUS FEROX (Lowe) (Alepisaurus esculapius Bean) Two well-preserved examples are from Misaki. Mr. Owston mentions the preservation of several duplicates and we have seen the same species in the Imperial University of Tokyo. A com- parison of the Japanese specimens with some from the west coast of North America leaves little doubt as to the identity of P. ferox and P. esculapius (Bean) with the Japanese species. A specimen collected at Unalaska, Alaska, by Dr. Jordan, has 41 dorsal, 15 anal, and 8 ventral rays; another obtained near Point Arenas, California, by Mr. Elijah Bishop, and presented to the University by Prof. Robert E. Swain, has 38 dorsal, 15 anal, and 9 ventral rays. The Japanese examples have 35 and 36 dorsal, 15 anal, 10 and 9 ventral rays respectively. A specimen from San Luis Obispo county, described and figured by Miss Flora Hartley,* has 39 dorsal (39 in description, 37 in figure), 17 anal, 9 ventral rays. All the above have the first rays of dorsal, anal, and ventral spine-like, with the anterior edge more or less roughly serrated. *Proc. Cal. Acad. Sct., 1805, p. 49, pl. 11. 236 SMITHSONIAN MISCELLANEOUS COLLECTIONS [VoL. 45 Family GONORHYNCHID& GONORHYNCHUS ABBREVIATUS Schlegel Nezumi-Gisu (Rat-Gisu or Sillago) (PiatTe LIX) A fine specimen from the Yokohama market. The Japanese species seems to differ from that of the Australian ‘seas (Gonorhynchus gonorhynchus =G. gronovii—=G. greyi=G. brevis) in the constantly larger head. In the specimen here figured the head is 424 times in length of body to base of caudal, 456 times in total length. The depth is about half the length of the head. Eye 4% in head. Mr. E. C. Starks has examined the shoulder girdle of this species ; it has the mesocoracoid arch, as usual with Isospondylous fishes. Its place is apparently with the earliest and most generalized of these forms. Family GEMPYLID PROMETHICHTHYS PROMETHEUS (Cuv. and Val.) Sumiyaki Two specimens from off Izu; the name means “ charcoal burnt.” We obtained the same species in the market of Tokyo, from off Misaki. Family LAprip@ JULIS MUSUME Jordan and Snyder, new species (een) ILY oie GONORHYNCHUS ABBREVIATUS SCHLEGEL. COLLECTIONS. OUS a ISCELLANE I IAN ° 79) rr ES = an 10g 50g 29 aah2or ae (uosulyty *S ‘MA Aq uMmeiq) “dS0ANS GNV Nvadyor INVWISVA SNYNllsidd Sag Vayar Ci Bai) "SNOILOUTIOD SAOANV'IIAOSIIA NVINOSHLING (UosUIy}Y *S “AA Ag UMeIGg) "YS0ANS ONY NvaYyOr SWASNW Sine td. “Sh to, “SNOILOGTIOND SNOANVTIAOSIJY NVINOSHLING (uosulyty °S “MM Ag umeiq) “dSGANS GNV NvddOr SAIVYOV SIMVINLOGNaSd Teale alice esi LOIN “SNOILOUTIOD SNOANVTIFOSIJN NVINOSHLING 103 80g 09 aahgoray (uosuinty *S “A Ag umesq) “YS0ANS ONY NvGYOr F/MVHIHS| SNYSLdAHOVYEL ‘U1G TITS tek, 2 “2O.\, “SNOILOATIOD SNOANVTIAOSIJY NVINOSHLING DESCRIPTION OF A NEW CYPRINOID FISH, HEMI“ BAkbUS JODIE NT FROM THE PEY He; TIENTSIN, CHINA By DAVID STARR JORDAN anp EDWIN CHAPIN STARKS A collection of fishes from the Pei Ho at Tientsin, China, made by Prof. Noah Fields Drake, of the University of Tientsin, was sent in 1898 to the Museum of Leland Stanford Junior University. These fishes were studied by Mr. James Francis Abbott, and a report published by him in the Proceedings of the U. S. National Museum, vol. xxitt, 1901. The specimens called by Abbott (p. 487) Hemuibarbus barbus, seem distinct from the Japanese species of that name. We propose for them the new specific name— HEMIBARBUS JOITENI Jordan and Starks, new species (PLaTeE LXIV) Head 4 in length to base of caudal; depth 4%. Dorsal 111, 7 (the last ray divided to its base) ; anal 8; lateral line scales 48. Body rather robust; a slight hump developed at nuchal region. Snout rather sharp as viewed laterally; projecting beyond mouth a distance slightly exceeding half the diameter of eye. Lips thick and papillose in large specimens; lower lip with a backward pro- jecting median flap. Length of maxillary barbel is from half to two-thirds of the diameter of the eye. Teeth in three rows; the outer row large; the two inner rows small; the first with five teeth, the second three, and the inner row with one. Anterior nostril ending behind in a broad flap which nearly closes posterior nostril. Large sensory canals follow the lower edge of the suborbitals and the edge of the preopercle, extending around the under side of the snout and mandible; they open to the exterior at rather wide in- tervals along their length through very small pores. Interorbital space rather wide, from 11% times diameter of eye in small speci- mens to nearly 2 in large ones; in the latter 3 slight longitudinal ridges are developed; a median ridge and a ridge at each side between outer: edges of frontals and supraorbitals. There are 7% scales in the backward downward series from first dorsal spine to lateral line and 6% in a like series from front of anal to lateral 241 242 SMITHSONIAN MISCELLANEOUS COLLECTIONS [VoL. 45 line; 16 scales in a median line between first dorsal spine and nape. Scales of lateral line slightly indented on their posterior edge; many of the scales on back with a slight angle, others with a slightly produced scallop. First dorsal spine very small, the second a fourth as long as the third, which is contained from 1 to 1% in length of head. Front of dorsal midway between snout and a point on caudal peduncle distant from base of caudal fin once and a half times diameter of orbit. Pectoral fails to reach vertical from front of dorsal by a distance equal to half the diameter of the eye. Color in spirits pinkish yellow with a longitudinal lateral series of about 8 large black spots above lateral line; smaller spots irreg- ularly placed on back and sides (on the large specimens following the rows of scales) ; dorsal and caudal with similar black spots; other fins without markings. This species differs from H. barbus, particularly in color. The black spots have been seen only in the young of the latter species. Four specimens collected by Dr. Drake in the Pei Ho at Tientsin, China. The largest, No. 8414, Stanford University, is 29 cm. in length. The species is named for the enlightened and progressive Emperor of China, officially known as Kwang-Hsu, Joi-Ten being his per- sonal name, not spoken by Chinamen but available in the democracy of science. WOR SUZ “OD J (BpewMiys o7¥49S5 Aq uMeIq ‘eHYeIG 4 YRoN) Aq pazoajjog ‘euIYyD ‘UIsjUal{ ‘aAIY OW [aq Wol4) “SHaVLS ONY NvG¥Or |NALIO£ SNEYVEINSH ADT Ta “sh "101A, “SNOILOSTION SNOANVIIFOSIJ NVINOSHLINGS THE REMOVAL OF THE REMAINS OF JAMES SMITHSON Byis. P.-LANGLEY The remains of James Smithson, founder of the Smithsonian Institution, who died June 27, 1829, at Genoa, Italy, were deposited in the little cemetery belonging to the English Church, on the Heights of San Benigno, a solitary spot planted with cypress trees, and looking down upon the Gulf of Genoa. In 1891 the Secretary of the Institution visited the grave, and, with the approval of the Regents, deposited with the Secretary of the English Church Fund a small sum to invest in Italian five-percent rentes, for its perpetual care. It was visited on two later occasions by the Secretary, who placed a bronze tablet containing a bas-relief of Smithson, in the English Church, and one also at the tomb, whence it was subsequently stolen. At this time it was understood that there was a probability that before many years the site of the cemetery might be required by the Italian authorities, and the following communication to this effect was made to the Secretary on the 24th of November, 1900, by the Committee of the British Burial Ground: 243 244 SMITHSONIAN MISCELLANEOUS COLLECTIONS [VOL. 45 7 VIA GARIBALDI, GENOA, 24 November, 1900. SAMUBL PIERPONT WANGEEY, sds slat! DiCile Smithsonian Institution, Washington. Dear Sir: The Committee of the British Burial Ground of Genoa (of which you are aware Her Majesty’s Consul is Chairman), fully realizing how keenly you are interested in all that concerns the resting place of the respected Founder of your Institution, has deputed me to write to you and lay before you the present position of our Cemetery. It will lie in your recollection that when I accompanied you some years ago up to the heights of San Benigno, you were struck by the enormous quarry which was slowly but surely eating its way towards us from the sea through the rocky side of the hill on which we stand, and excavation has lately come so close to us that the inter- vention of the Consul became necessary to arrest further advance on the plea that our property would be endangered if the quarrying were carried on. Actual blasting has in fact been put an end to for the present, and the Cemetery (although the boundary wall is now on the very edge of the excavation) remains untouched; but the local authorities who are the owners of the quarry have given us to understand that they need more stone for their harbor works, and are therefore anxious to see our graves transferred from the position they now occupy, for which purpose they would give us a suitable piece of ground in another part of the town and would also undertake the due and fitting transport of the remains. Should our answer be in the nega- tive, it is intimated to us that in five years’ time, in 1905, the term for applying the Law for Public Utility (twenty years after the date of the last burial) will have been reached, and we shall then have to give up of necessity what we are now asked to yield as a concession. Under the circumstances, the Committee have decided that it is their best policy, in the interest of all concerned, to begin to negotiate at once for the transfer on a decorous footing of the British Ceme- tery and all its tombs, and although some considerable time may elapse before this transfer is accomplished, yet it is evident that the time has now come for us to ask you to prepare your decision as to what is to be done with regard to the James Smithson remains. Are they to be laid with all possible care and reverence in new ground here, or are they to be conveyed to the United States? Awaiting the pleasure of your reply, I beg to remain, Very faithfully yours, E. A. LE MESURIER. This communication was laid before the Regents, who, at their meeting of January 23, 1901, adopted the following resolution: ResotveD: In view of the proposed abolition of the English Cemetery at Genoa which contains the remains of James Smithson, LANGLEY | REMOVAL OF REMAINS OF JAMES SMITHSON 245 that the Secretary be requested to arrange either with the English Church or with the authorities of the National Burying Ground at Genoa for the re-interment of Smithson’s remains, and the transfer of the original monument. At a meeting of the Regents held on the 22d of January, 1902, the Secretary recalled to the Board the resolution adopted at its previous meeting, and stated that the wishes of the Board with regard to the removal of the remains of Smithson had been com- municated to Mr. E. A. Le Mesurier, one of the officers of the English Church at Genoa, who, under date of December 23, 1901, had replied in part as follows: You are aware that our hope is eventually to obtain for our countrymen a separate burying-place which by an easy, and I may say obvious, arrangement might be made to give shelter not merely to British subjects but to American also. I regret to say, however, that I see no chance for the present of this most desirable consum- mation, as the authorities (apparently in consequence of the diffi- culty of finding an alternative site) have withdrawn their offer of providing us with a fresh cemetery if we allowed them to transfer at once all remains from San Benigno, where your Founder rests. The present policy of the authorities is presumably to let things remain as they are until the time comes (three years or so hence) when the Law of Public Utility will strengthen their hands as to taking possession of the San Benigno ground, of course under the obligation of transporting the remains elsewhere, which would in all probability mean a portion of the general Protestant cemetery and not a separate place of interment. When the time for the transfer approaches, it will be obviously expedient to apply to the British Ambassador at Rome (backed up, as we are confident will be the case, by the friendly offices of the Representative of the United States) to put the case before the Italian Government, so that the local authorities may be enjoined to carry out the process with all due reverence, and if possible (as it ought to be possible) to a especially reserved new Cemetery. Our Consul is most fully alive to the importance of diplomatic support and will take the initiative in due course. Doctor A. Graham Bell, a member of the Board, took occasion to reiterate the strong feeling expressed by him at the preceding meeting of the Regents, that the remains of Smithson should be brought to this country. At the regular meeting of the Regents held on December 8, 1903, the Secretary read the following letter from the United States Consul at Genoa, and that from the Committee of the British Burial Ground Fund in Genoa: 246 SMITHSONIAN MISCELLANEOUS COLLECTIONS [VoL. 45 CoNSULAR SERVICE, U. S. A. ’ Genod, Itaty, November 24th, 1903. Dr. S.-P> LANGLEY; Secretary of the Smithsonian Institution, Washington, D. C. Dear Sir: Referring to my letter of June 30th last, 1 now forward enclosed circular letter just received, which shows that the final step in the demolition of the old British Cemetery is about to be taken. If you have any desires to express as to the disposition of the remains of James Smithson, and instructions to give, these should be made known with as little delay as possible. You will see by my letter above mentioned that there is a tradi- tion here of some opposition to the removal of James Smithson’s body to America on the part of relatives in Europe. I was informed at the British Consulate that the source and character of this opposi- tion has been made known at some time to the Smithsonian Insti- tution. I can learn nothing further of it here. You are therefore in possession of such information as can be had for your guidance in adopting a course of action as to the disposition of the remains. You will notice that instructions are requested by the Cemetery Committee before January Ist, 1904. I am quite at your service in the matter of any assistance that may be needed here. I am informed by the Agent of the American Express Company here, he would transport the body to Washington for $203. This would not include taking up the body nor the coffin, the boxing, etc. A Government tax of 360 lire (approximately $72.00) has also to be paid on every body taken out of the country. The total expense at a rough estimate might come to $400.00 or to $500.00 if some- thing unusually handsome in the way of a casket were called for. All due economy would be used. Believe me Very truly yours, WititiAm HEnry BisHop, U.S. Consul. (Enclosure) GENOA, November 23, 1903. SU, We have the honor to inform you that the old British Cemetery, on the heights of San Benigno in this city, has been expropriated by the Italian authorities, and will shortly be demolished. The remains of all persons buried there will be removed to the new British Cemetery at public expense, and the tombstones will also be removed, and re-erected over the new graves, by the under- signed Committee, unless otherwise desired by the representatives of the deceased. It has been impossible to ascertain the addresses of these repre- sentatives in every case, and this letter is sent to you with reference to the grave of James Smithson. LANGLEY ] REMOVAL OF REMAINS OF JAMES SMITHSON 247 Kindly address any communication you may wish to make on this matter to Noe. Lees Esq., Care of H. B. M.’s Consul-General—Genoa before January Ist, 1904. And believe us, Yours faithfully THE COMMITTEE BritisH BuRIAL GROUND FUND Genoa. Doctor Bell renewed the proposal made by him at the previous meeting of the Regents, that the remains of Smithson be brought to this country at his expense, and after some remarks the following resolutions were adopted: Resotvep: That Doctor A. Graham Bell be appointed as a com- mittee to take charge of the matter of the removal of the remains of James Smithson from Genoa to Washington, with the request that the negotiations and removal be conducted quietly and privately. Resotvep: That upon the conclusion of this duty, all expenses involved by it be reimbursed to Doctor Bell from the funds of the Institution. Doctor Bell, accompanied by Mrs. Bell, sailed on the 15th of December for the port of Cherbourg in France, and going thence to Genoa, commenced at once the arrangements for the transfer of the remains, arrangements which would have occupied a quite indefinite time and incurred a corresponding delay, except for the aid given by the United States Consul, Mr. William Henry Bishop, which Doctor Bell gladly acknowledges. On opening the tomb in the presence of Mr. Bell, the United States Consul, Noel Lees, Esq. (official representative of the British Burial Ground Fund Association), and other witnesses, it was found that the remains of Smithson, represented by the skeleton, were in fair preservation, although the wooden coffin in which they had been enclosed had molded away. The remains were placed in a metal casket and deposited in the mortuary chapel of the cemetery, where they rested until January 2, when the casket was enclosed in a coffin of strong wood and covered with the American flag by Consul Bishop. On this occasion Doctor Bell, Mr. Bishop, and the other witnesses again assembled, and the following remarks were made: Remarks by William Henry Bishop, Esq., United States Consul. Doctor ALEXANDER GRAHAM BELL: You arrived here, my dear Dr. Graham Bell, charged by the Smithsonian Institution with the 248 SMITHSONIAN MISCELLANEOUS COLLECTIONS [VOL. 45 mission of removing to Washington the remains of the Founder of that Institution, James Smithson, who has been buried till now in the cemetery where we stand, since his death, at Genoa, in the year 1829. Having been invited by you and by the Smithsonian Institu- tion to aid you, to what extent I might be able, in this object, it has been a matter of great pride and pleasure to me that I have been allowed to do so. All the steps necessary to such removal have now been taken. We have received the authorization of the governmental heads of the Province, the City, and the British Burial Ground Fund, in which latter the title to the cemetery and the custody of the grave of James Smithson are vested, and all of these have kindly codperated with us in the work. The body of James Smithson has now been reverently raised from the earth; it has been placed in a case securely sealed, and this case stands ready to pass into the charge of the Steamship Company which will convey it to New York. I assure you that it is with a feeling of real emotion that I have just now cast the American flag over the body of this illustrious man, this noble but as yet little known benefactor, as it is on the verge of beginning its journey to the United States. The flag adopts him already, as it were, in the substance, for our country, to which he has so long belonged in the spirit. He is now about to receive there a portion of the outward veneration and homage he so supremely merits, and which, owing to the modest circumstances of his life, and his interment here in some sense almost forgotten, he has never had. Shall I admit that on taking possession of my post as Consul at Genoa, I did not even know who James Smithson was? I may say that I was surprised to learn that he was buried at Genoa; more surprised still that he was an Englishman, who had never even set foot in America. He left his great bequest to the United States, then in its infancy, through admiring confidence in our future. It is likely that many, or even most, Americans are in the same condi- tion as was I myself; for occasion has rarely arisen for taking thought as to the personality of the man. Happily this unen- lightened condition of mind is about to cease. Dr. Graham Bell, I wish you a hearty God-speed across the ocean, with your precious freight. The American people will receive it with general gratification, and, through the Smithsonian Institution, will soon delight to pay it great honor. Response by Doctor Alexander Graham Bell. Mr. Consut: It is with feelings of deep emotion that I under- take the transportation of the remains of James Smithson from the cemetery where they have so long reposed, to their last resting place in the United States. On behalf of the Smithsonian Institution allow me to thank you, Mr. Consul, for the unwearied zeal and care with which you have given me your assistance. Without your active codperation—and LANGLEY ] REMOVAL OF REMAINS OF JAMES SMITHSON 249 without your personal sympathy—it would have been difficult indeed for me to have accomplished the object of my mission here. On behalf of the Smithsonian Institution, I beg to thank you too —Mr. Noel Lees—for your courtesy and attention; and trust that you will convey to His British Majesty’s Consul General, and to the Committee of the British Church Burial Ground Fund, my thanks, and the thanks of the Institution I represent, for their ready assist- ance in furthering my mission. The United States of America will provide in Washington, D. C., a suitable and permanent resting place for the remains of her great benefactor, James Smithson, through the instrumentality of the Smithsonian Institution—the Establishment created by the Govern- ment to perpetuate his name. Remarks by Noel Lees, Esq. Doctor GRAHAM BELL: I beg to thank you heartily for the words you have said with regard to the aid you have received from the Burial Board and myself. Although we regret to lose the remains of James Smithson, we at the same time feel that in the country to which he left his money, with such charitable intent, his remains will receive the honor and glory which have so long been due to them, and we must understand that our loss is America’s gain. To us it will always remain a pleasant memory that, from the date of his burial to the present day, we have had in our custody in this pic- turesque little church-yard, the remains of a man whose foresight and kindness have enabled so many in the New World to benefit. On the conclusion of these remarks the remains were placed on board the steamer Princess Irene of the North German Lloyd line, which brought them to New York, where they arrived on the night of January 19, in the continued charge of Doctor Bell, the vessel reaching her dock at Hoboken early on the morning of the 2oth. By direction of the President of the United States, the U. S. steamer Dolphin had been detached to. meet the Princess Irene in the lower bay and to accompany her up the harbor, while a tug belonging to the Navy Yard attended at the dock to receive the remains and transport them to the Dolphin. They were received by Mr. Bell and the Secretary of the Institution, Mr. Bell accompanying the remains to the Dolphin and taking passage on her himself for Washington, where she arrived at the Navy Yard on Saturday the 23d. On Monday the 25th the remains were transported by the Naval authorities, with suitable ceremonies, to the Navy Yard gate, where they were taken ig charge by a cavalry escort furnished by the War Department, and, accompanied by Assistant Secretary of State Loomis, representing the President, by the British Ambassa- dor, the Regents and the Secretary of the Institution, and the 250 SMITHSONIAN MISCELLANEOUS COLLECTIONS [VoOL. 45 President of the Board of Commissioners of the District of Colum- bia, they were conveyed to the Smithsonian Institution, where the coffin, draped in the American and British flags, was deposited in the center of the Main Hall of the building. Dr. Alexander Graham Bell, addressing Senator Frye in behalf of the Regents, said: Mr. SENATOR: I have the honor to hand over to the Smithsonian Institution the mortal remains of its founder, James Smithson, a Fellow of the Royal Society of London, England, who died in Genoa, Italy, on the 27th of June, 1820. For nearly seventy-five years the body of Smithson has reposed in an almost forgotten grave in the picturesque little British ceme- tery on the heights of San Beningo in Genoa. City improvements have led to the expropriation of this cemetery and the removal of the remains, and at the last meeting of the Board of Regents of the Smithsonian Institution I was appointed a committee to arrange for the transfer of the remains of Smithson to this country. On my arrival in Genoa, every facility was afforded me for the accom- plishment of my mission by the provincial and municipal authorities, by His British Majesty’s Consul-General, Mr. Keene; by the Com- mittee of the British Burial Fund Association, in which is vested the ownership of the cemetery, as well as by our own Consul, Mr. William Henry Bishop, to whom I am much indebted for his valued services. On the 31st of December, 1903, the tomb of Smithson was opened in my presence, as the representative of the Smith- sonian Institution, and in the presence of the American Consul and six other witnesses. The remains of Smithson were reverently raised from the grave and placed in a metallic casket, over which the Consul of the United States cast the American flag while the witnesses stood around with uncovered heads. The casket was then left in the mortuary chapel of the cemetery, securely sealed and under guard, until the 2d of January, when it was placed in a coffin of strong wood, as demanded by Italian law, and was then transported to the North German Lloyd steamship Princess Irene, accom- panied by the American Consul and myself. . The steamer sailed from Genoa on the 7th of January, and upon arrival in the United States, the remains of Smithson were received with national honors by direction of the President, and of the Secre- tary of the Navy and the Secretary of War. The remains were brought to Washington on board the United States dispatch boat Dolphin, and have been escorted to the Smith- sonian Institution by United States cavalry. And now, Mr. Senator, my mission is ended, and I deliver into your hands, as the representative of the Board of Regents of the Smithsonian Institution, the remains of this great benefactor of the United States. LANGLEY ] REMOVAL OF REMAINS OF JAMES SMITHSON 251 Senator Frye replied: Sir, the Smithsonian Institution receives with profound gratitude the remains of its distinguished founder. Providence, every now and then, seems to place in the world a man and inspires him with a purpose to elevate his fellow men. Such a man was Mr. Smith- son, the founder of this Institution. The spirit, Sir, which prompted you to such earnest endeavor, resulting as it did in taking these remains from their resting place in a country foreign to him and foreign to us, and bringing them here where for so many years we have enjoyed the rich fruits of his splendid benefaction, your coun- trymen will appreciate. His grave here will be an incentive to earnest, faithful, wise, and discreet endeavor to carry out his lofty purposes, and, Sir, it will be to our people a sacred spot while the Republic endures. The brief but impressive ceremonies of the occasion concluded | with the following prayer, offered by the Reverend Doctor Randolph H. McKim: Almighty God, eternal source of light and truth, by whose wise providence all things in heaven and earth are governed, we give Thee thanks that Thou didst put into the heart of Thy servant whose dust we receive with reverence here to-day, to lay the foundation of this school of science, and we pray Thee that it may more and more be instrumental in the true interpretation ofthe laws of nature, and in unveiling to the mind of man the glory of God in the work of His hands, to the end that for all the generations to come, this Institution may be a beacon light of truth and of progress, to the glory of God and to the good of mankind. All this we beg through Him by whom all things were made, Jesus Christ, our Lord. Amen. The remains rest temporarily in a room which contains the few personal relics of Smithson, until their final disposal by the Regents. NOTES -ON THE BREEDING (HABITS © rs rue YELLOW-BELLIED TERRAPIN By HUGH DM SMITE In the Potomac river between Washington and salt water, the yellow-bellied terrapin (Pseudemys rugosa) in former years sup- ported a profitable fishery, but for a long time it has been uncommon and is now seldom sought, those caught being taken incidentally in winter seines hauled for fish. The decline of the fishery, while due primarily to the decrease in abundance of the terrapin, was to a considerable extent dependent on the establishment in Washington and other eastern cities of a trade in southern and western terrapins which could be caught in large numbers and sold at much lower prices than the local species. The terrapins frequented the marshy shores and heads of creeks, and were caught with haul seines and fyke nets during fall and winter. Piscataway creek was one of the best fishing grounds, and many hundreds of dozens of terrapin were taken there every winter for the Washington market. On one day in December, about the year 1883, 240 terrapin were there caught at one seine haul by Mr. L. G. Harron, now of the U. S. Bureau of Fisheries; they had been buried on a shallow bar, but had uncovered themselves under the influence of an unusually warm spell. The largest were sold for seventy-five cents apiece in Washington, which was about the average price in those days, while the smallest, six to seven inches long, were worth only fifteen or twenty cents. This species is known along the Potomac under the names “ slider,’ “ terrapin,’ and “fresh-water pullet,’ the last designation being in common use among fishermen and negroes generally. The egg-laying season is in June and July, and the place where the eggs are laid is usually a cultivated tract, often a cornfield adjoining the water. It is probable that a field would always be selected, but when there is a high steep bank the eggs are of necessity deposited on the shore. The terrapins visit the fields only during egg-laying time and only for this purpose, and some- times make their nests more than one hundred feet from the water. It has often been observed that six or eight terrapins will lay on the same shore or in the same field, their tracks being easily dis- cernible in the moist or soft sand or loam. The nest is made in 252 SMITH] BREEDING HABITS OF YELLOW-BELLIED TERRAPIN 253 sand, clay, or loam, a sandy loam or sandy clay being most fre- quently chosen. The nest, which is shaped like a carafe, is dug by the female with her fore-legs. Its size depends on the size of the animal or, what amounts to the same thing, on the number of eges to be laid; an average nest would be four inches deep and four inches wide at the bottom, the opening being somewhat smaller than a silver dollar. When on the shore, the nest is always above high-water mark. All the eggs are laid at one time, and when the laying is com- pleted, earth is scraped into and over the hole and packed tightly. The packing is accomplished by the terrapin raising herself as high as possible on all four legs and then dropping heavily, by the sudden relaxation of the extensor muscles. Immediately after covering the nest, the terrapin withdraws to the water. The size of the eggs varies somewhat with the size of the terrapin, but averages one inch by three-fourths of an inch. A six-inch terrapin lays ten or twelve eggs, while the largest terrapins, fourteen or sixteen inches long, lay as many as twenty-five to thirty- five eggs, possibly more. When a terrapin is disturbed while making a nest or laying, she will abandon the nest. On one oc- casion, when a terrapin was discovered over a nest in a cornfield, removed to see whether any eggs had been deposited, and replaced over the hole in the ground, it was found when the place was visited two hours later that she had left without laying any eggs. The eggs probably hatch during the summer, but on this point there have been no personal observations. The young, however, remain in the nest until the following spring (April 10 in one case), and when they emerge they are about the size of a twenty-five-cent piece. They go to the water at once. Considerable quantities of terrapin eggs were formerly eaten by people living on the river shores, but of late very few eggs have been thus utilized. When boiled, the eggs are regarded as a delicacy. Birds (especially crows) and other animals doubtless destroy some eggs. On one occasion a terrapin was observed over a nest on the shore, and a crow noticed on a dead tree near by. When the terrapin covered the eggs, concealed the nest, and with- drew to the water, the crow immediately dropped to the ground and began to dig into the nest. Before the observer, who was in a boat, could reach the shore, the crow had destroyed at least two of the eggs, seven others remaining. The male yellow-bellied terrapin is smaller than the female; his claws are twice as long as hers, and the under-shell is flat, while in the female it bulges centrally. A NEW PELICAN FISH FROM SEP SP Acinic By BARTON A. BEAN During a morning watch in June, 1900, half-way between Mid- way islands and Guam, on the U. S. Naval Ship Nero, there was brought to the surface what is believed to be the first pelican fish recorded from the Pacific ocean. The specimen was entangled on the sounding wire, near the sinker, the depth of the ocean at the point of capture being between 2000 and 3000 fathoms. The steamer was engaged at the time in a survey for the Transpacific telegraph cable, and it was during this voyage that the greatest depth of water ever recorded was discovered, being 5269 fathoms, in 12° 4315” N. lat.; 145° 49’ 00” W. long. GASTROSTOMUS PACIFICUS new species Length of preserved portion 14% inches (tail end of body want- ing); length of cranium ;% of an inch, equal to its width; length of upper jaw 334 inches; distance from tip of snout to origin of dorsal fin a little more than 24 length of upper jaw; the depth of body at angle of jaws is about % of the length of upper jaw; the gill slit is under the 12th dorsal ray; the first anal ray is under the 34th dorsal ray. Dorsal and anal rays high, the highest about Yé length of upper jaw. This species differs from Gastrostomus bairdu Gill, the Atlantic form, in having a more robust body and higher fin rays. The single specimen known is badly mutilated, and better material, should other captures of the fish be made, will doubtless show many points of difference from the Atlantic species. We are indebted to Dr. J. M. Flint, U.S.N. (retired), for assistance in obtaining the records of this interesting capture. The type is No. 50,724, U. S. National Museum. The accom- panying illustration (figure 31) was made by Mr. A. H. Baldwin. ‘wen pue spuryst Aempily{ usamjoq Aemypepy ‘0061 ‘ounf ‘ovany diys [eAeN “S “ (‘porojsoy) ‘urog “vy “q odd} snoy190d snmojssosspy—I1€ “SIT A REVISION ©F THE PALEOZOIG ER YOZO% By E. @: ULRICH anp. R: S: BASSEER Part I.—On GENERA AND SPECIES OF CTENOSTOMATA In 1897 Mr. Ulrich undertook the preparation of as complete a collection as possible of American fossil Bryozoa for the British Museum of Natural History. At the same time the late Professor Zittel of Munich wished a similar although somewhat smaller collec- tion for his university. The junior author was at that time assisting Mr. Ulrich, so the two decided, while preparing these collections, to revise especially the generic and specific classification of the Paleozoic Bryozoa, as thoroughly as the material at hand would permit. With the exception of two short trips into the field, Mr. Ulrich spent two years in the preparation and study of thin sections and the separation of large collections embracing representative examples of the whole class. Mr. Bassler, besides aiding Mr. Ulrich in this work, also undertook the review of all the literature of Paleozoic Bryozoa and the compilation of a card catalogue of the American forms, using as a basis a list and bibliography pre- pared some years before by Mr. Ulrich. In 1900, Mr. Bassler, collaborating with Mr. John M. Nickles, who had prepared a similar card catalogue, published! much of the information con- cerning genera and all of the synonymy learned during the prepara- tion of these collections and catalogues. During the course of our studies, particularly of the Trepostomata, some four or five thousand thin sections were prepared. Of many species we made sometimes as many as a dozen sets of sections to determine the specific variation. Naturally the immense quantity of material studied afforded many new species, and these, besides aiding the strict characterization of the previously established generic groups, also served in distinguishing certain new genera that pre- viously had been known only from species more or less imperfectly understood and therefore difficult to classify satisfactorily. At the same time a number of wholly new genera were determined, while other, perhaps equally distinct, groups require further material to prove the permanence of their peculiarities. * Bulletin 173, U. S. Geological Survey. 256 ULRICH-BASSLER] REVISION OF PALEOZOIC BRYOZOA 257 As to the new species, they are to be numbered by the hundreds. This is indicated by the junior author’s recent paper on Homotrypa,* in which the number of the Cincinnatian species of the genus previously described is more than trebled. While the specific representation of this genus of the Trepostomata is exceptional, it is still true that many of the other genera are very prolific in species and will continue to afford subjects for papers on new forms for years to come. This and the following papers of this series are based almost entirely on the collection in the U. S. National Museum, which con- sists mainly of the recently-acquired Ulrich collection. This latter collection probably contains the largest series of Paleozoic bryozoa extant and is the result of thirty years of work on the part of Mr. Ulrich, with the assistance, at different times, of Mr. Charles Schuchert and the junior author. The Museum Paleozoic series has been augmented in the last few months by the collection of fossils made by Dr. Carl Rominger of the University of Michigan. In late years, large and excellent series of bryozoa have been collected for the Museum from horizons in which the Ulrich collection is lacking. Dr. Frank Burns of the U. S. Geological Survey, during the course of his work in the Tertiary rocks, has collected many bryozoa in horizons where they were formerly supposed to be wanting. Dr. T. Wayland Vaughan has also discovered many bryozoa during the progress of his studies on the Tertiary corals of America. The study of these Tertiary collections is under way and it is hoped in the near future to publish papers on the subject. All of the five orders of bryozoa comprising the subclass Gymnolemata are represented in the Paleozoic rocks. Both species and specimens of the order Cfenostomata are usually rare. The Trepostomata, Cryptostomata, and Cyclostomata are quite abundant and commonly form a considerable part of the fauna. The Chilos- tomata, which includes so many Cenozoic and recent species, is doubtfully recognized in the Paleozoic, the single genus Paleschara probably belonging to this order. We deem it only right to introduce here an explanation concern- ing the bryozoan chapter in the American edition of Zittel’s Te-t- book of Paleontology. Several authors have alluded to the fact that the same genera occur both among the bryozoa and the corals. The explanation is very simple. The coral chapter had been translated without revision and *Proc. U. S. National Museum, XxXvi, 1903, pp. 565-501. 258 SMITHSONIAN MISCELLANEOUS COLLECTIONS [VOL. 45 printed before Mr. Ulrich had been asked to revise the chapter on the Bryozoa. When the manuscript for the latter was submitted, the author and the editor objected to the duplicate treatment of the Monticuliporoids, but withdrew their objection when Mr. Ulrich presented satisfactory evidence of the correctness of his classification. In this connection it may be well to state also that, in the opinion of the writers, no valid objection to the reference of the Monticu- liporoids and Fistuliporoids to the Bryozoa has ever been made. None of the recent critics on Lindstrom’s, Rominger’s, and Ulrich’s reference of these fossil organisms with the Bryozoa is deemed of sufficient importance to demand serious attention. As to Waagen and Wentzel,t they appear to have known little of the American literature on the subject. Had they been better informed also con- cerning the wonderful abundance and variety of the Monticuliporoids and related organisms in American Paleozoic deposits, it is scarcely probable that they would have based a new classification on a very small collection of poor specimens. Many facts bearing upon the relations of the Monticuliporoids to the later Bryozoa have come to light since the publication of the American Paleozoic Bryozoa by Ulrich in 1882. At some future time we shall marshal these facts in the hope that their publication may finally fix the position of these disputed forms. In the present series of papers the efforts of the authors are directed primarily to the consideration of the generic groups, and in only one order, the Cfenostomata, a relatively small group con- sidered in this paper, are all the known species considered. Order CTENOSTOMATA Busk. The earliest notice of the Paleozoic fossils, which seem now to be very generally accepted as imperfect remains of ancient repre- sentatives of Ctenostomatous Bryozoa, was by Nicholson and Etheridge, Jr.,in 1877. In that year these acute observers published a paper in the Annals and Magazine of Natural History entitled “On Dez 7. Portion of the creeping network of the typical form of this spe- cles, X 9, showing several nuclei. Rochester shales, Lockport, New York. Vinella? multiradiata new species. a -p: 276 8. Part of a colony attached to a crinoid Siucane euhibinay many nuclei and the intertwining of the numerous radiating stolons, X 9. Rochester shales, Lockport, New York. Ascodictyon stellatum Nicholson and Etheridge, Jr.......p. 287 9, 10. Two clusters, X 18 and X 9. Only one of the vesicles in the first figure has the surface puncte represented. Hamilton formation, Eighteen Mile creek, New York. Ascodictyon siluriense Vine......... .p. 286 11. Two isolated clusters of vesicles, X 9, Pitched to a fpaeoeene of Leptena rhomboidalis. Rochester shales, Lockport, New York. 12, A number of clusters of vesicles with connecting stolons, 9, growing on a small Orthoceras. Waldron shales, Waldron, Indiana. Ascodictyon floreale new species...............p. 288 13. Portion of a colony with numerous clusters of vesicles, X 9, grow- ing upon a Stropheodonta. Hamilton formation, 2 miles west of Alpena, Michigan. 294 SMITHSONIAN MISCELLANEOUS COLLECTIONS [voL. 45 Ascodictyon parvulum new species...............p. 288 14. Part of a colony, X 9, growing upon a crinoid column, showing vesicles arranged in clusters and others apparently without order. Chester formation, Jackson county, Kentucky. Ascodictyon sparsum new species...............P. 289 15. Several branches of the adnate colony, showing the usual scattered arrangement of the vesicles, 9. The connecting stolons are not so "clearly shown in the specimen as in the figure. Chester formation, Claxton, Caldwell county, Kentucky. SMITHSONIAN MISCELLANEOUS COLLECTIONS Vor 45 7b Ly SPECIES OF CTENOSTOMATA. SMITHSONIAN MISCELLANEOUS COLLECTIONS VoL. 45, PL LXVI SPECIES OF CTENOSTOMATA. SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 45, PL. LXvll SPECIES OF CTENOSTOMATA. VOL. 45, PL. LXVIII SMITHSONIAN MISCELLANEOUS COLLECTIONS SPECIES OF CTENOSTOMATA. i as, Le ste i , +2 7 ‘ “4 ; ea eee ‘ ee a a a A REMARKABLE GENUS OF FISHES—THE UMBRAS By THEODORE GILL I In 1726 Marsigli, in a volume on fishes in a series entitled Danubius Pannonico-Mysicus (tom. 1v), for the first time described in a recognizable manner a small fish of lower Austria which was known to watermen as the Hundsfisch (dogfish) and which he called Gobius caninus. He likened it to the Cyprinid named gudgeon (with which indeed many watermen confounded it), and hence the name he gave it, Gobius being his Latin name for the gudgeon. That name was not applied in the spirit of the modern binomial sys- tem of nomenclature and consequently has not been adopted by modern naturalists. He indicated that the little fish lives in stagnant Fic. 34.—European Umbra (Umbra umbra). After Heckel and Kner. waters and in caves, and that it is rarely seen; it is observed mostly in spring when it is carried by freshets into more frequented waters. In 1756 W. H. Kramer, in an Elenchus Vegetabilium et Animalium per Austriam inferiorem observatorum, described the same fish as found in affluents of the Danube and caves in lower Austria, and he named it Umbra; the name was given as a generic term after a consideration and comparison of the genera of Artedi. The name was given because the species harbored mostly in grottoes where the light does not penetrate and consequently lived in the shade (umbra). 295 296 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 45 In 1777 J. J. Scopoli accepted the name Umbra as a generic designation and referred it to his second order (ordo 2.), distin- guished by the subcylindrical body (corpore tereti aut teretiusculo). He interposed it between the genera Esox and Albula on the one hand and Cobitis and Anableps on the other. His diagnosis was equally applicable to almost all Cyprinodonts and doubtless, if the author had known such, he would have referred them to the same genus.t As it was, having adopted the genus from Kramer, the species described by Kramer was evidently the type. By all later authors the type was overlooked or referred to the genus Cyprinodon and not till 1842 was it properly recognized. Then the illustrious Johannes Muller found that it was a very distinct genus and related not to the Cyprinodonts but to the pikes with which he associated it in the same family. A few years afterward (1846) Valenciennes proposed to isolate it as the sole representative of a peculiar family to which he only gave the name “ Les Ombres (Umbra)” in common with the genus. He thought it was most nearly related to the Amias, from which it differed mainly in the absence of the suborbital cuirass (cuirasse sous-orbitaire) and the lingual bone; the simple non-cellular air-bladder further dis- tinguishes it. Consequently he believed it was the type of a distinct family (le type d’une famille distincte) which he proceeded to define. No representatives were recognized from America until near the close of the first half of the nineteenth century, and then several observers at nearly the same time obtained and described specimens from different localities. Jared Kirtland, in 1840, described one form under the name Hydrargyra limi from individuals obtained in northern Ohio ; Zadock Thompson, in 1842, found the same species in Lake Champlain, and, failing to recognize it as the same as Kirt- land’s, gave a new name—Hydrargyra fusca—to it; in the same year (1842) James E. Dekay gave a new name (Hydrargyra atricauda) to the fish first described by Thompson, and having obtained from Rockland county, New York, small specimens of another species, he gave to the latter the name Leuciscus pygmeus. None of these authors recognized the relations of the new forms; they all, however, appreciated the likeness to the Cyprinodonts which actually exists, although the fishes really belong to a different family, while the last (Dekay) committed in one case an incomprehensible blunder, inasmuch as his Leuciscus pygmaeus had none of the essential char- *256 Vmbra, Kramer. Maxilla inferior longior. Dentes in maxillis. Caput et Abdomen Cyprini. Membrana Branch. oss. (6). Pinnz inermes (7). Scopoli Introd. Hist. Nat., p. 450, 1777. GILL] THE UMBRAS 297 acters he assigned to the genus Leuwciscus or that truly define the family of Cyprinids. It was not till 1855 that the true relations of the American fishes were partially recognized. Two years previously (1853) Louis Agassiz had, indeed, referred to “ Charaxim [Characini] without adipose fin of I. [J.] Muller, of which a new genus occurs in the fresh waters of our northern and middle as well as western states, with a half a dozen species some of which,” he continues, “ have been unfortunately described as Leuciscus, Fundulus and Hydrargyra, with which genera they have no affinity, while other new ones have been described by Professor Baird and myself.” He concludes: “T shall call this genus Melanura, from the singular black mark which all species show on the tail.” Such is all the information communicated respecting the new genus, and no one could be certain what species were meant. But in 1855 Agassiz interposed a “ Note on Melanura Agass.” in a “ Synopsis of the Ichthyological Fauna of the Pacific Slope of North America,” affirmed that Melanura is the North American representative of the European Umbra, specified the species he would refer to it, and gave a diagnosis of it. The diagnosis contained no characteristic differentiating it from Umbra. It was regarded by Giinther in 1866 as identical with it, and such is now the opinion of all ichthyologists. In 1866 A. Gunther gave the name Umbridz to the family recog- nized by Valenciennes twenty years before and referred all the American forms, under one specific name (Umbra limi), to the same genus as the European fish (Umbra crameri). The best idea of their relationship may be conveyed by the statement that they are Pikes in the guise of Killiefishes. They have scales with a peculiar ornamentation or structure, and these extend over the head, leaving only the snout and jaws naked (figures 35, 36). Fie. 35.—Scale of Umbra. Il The results of investigations thus summarized may be summed up in a few words. 298 SMITHSONIAN MISCELLANEOUS COLLECTIONS [voL. 45 The Umbrids or mud minnows are a feeble family represented by a single genus whose species are distributed in a notable manner. A single one (Umbra wmbra or krameri) occurs in central and southeastern Europe, Hungary and the neighboring countries, and two others are inhabitants of the eastern slope of the American con- tinent and the Mississippi valley, while all the intervening countries are deprived of them. They are scarcely distinct from the pikes and, as in them, the supramaxillaries form the lateral margins of the mouth, but the jaws are not produced and the teeth are all small and “ villiform.” In appearance, indeed, they are much more like Cyprinodonts (killiefishes) than pikes. The species chiefly fre- quent still waters and are most at home in muddy and reedy ponds or in clear ponds with a muddy bottom; they are also prone to resort to. sequestered pools or caverns. They hide on or rather in the bottom, and where not one may be seen at first, numbers may be secured by drawing a net over the bottom and stirring up the mud. They are very tenacious of life and individuals have been kept by the writer for many months without change of water in moderate sized jars with water plants. Their behavior and carriage are char- acteristic and have been especially described by a number of nat- uralists, especially by Heckel and Kner (1858), C. C. Abbott (1884), and L. Kathariner (1899). IGUE From authors already cited and from personal observations a tolerable biography may be obtained and the tale is of considerable interest. In a clear pool some may be seen, if quietly approached, in mid- water, perfectly motionless, with the pectoral fins closed and stretched out and downward, at right angles to the trunk, and with the tips curved slightly forward, while all the other fins (including the ventral) are widely expanded; so they may remain for minutes at a time or with a slight movement of dorsal rays ; then one pectoral may be expanded and its surface displayed forward and backward while the other fin may be appressed to the side. Another attitude often assumed is one of balancing in midwater, when sometimes a vertical position is taken, or even the back is inclined downward; the pec- torals are then the principal fins used and are constantly moving in an undulatory manner, the fully expanded ones being for a moment outspread and horizontal to the length. Still another frequently assumed attitude is a “ statue-like ” one which has been also desig- nated by Abbott as the “ salamandrine position,’ and which is often maintained for several minutes at a time. ‘“‘ The body is frequently GILL] THE UMBRAS 299 curved when at rest, and remains so, the head being turned to the right or left, and the tail in the opposite direction. No one can fail to see the salamandrine appearance of this fish when it assumes such a position.” The use of the pectoral and ventrals in such a state of - h Vi i ! NA il i mh e Mi i) = OY, Fic. 37.—Various attitudes of Umbra. After life. rest has been also likened by Kathariner to that manifested by the “ salamandroid” or “sirenoidean” lungfish of Australia named Neoceratodus. The pectorals, he declared, are completely turned downward and perpendicular, and their ends, resting on the bottom, in consequence of the weight of the body are twisted outward and 300 SMITHSONIAN MISCELLANEOUS COLLECTIONS [VOL. 45 backward; the ventrals are also extended downward and touch the bottom quite lightly. The body rests nearly parallel with the ground, and the caudal is somewhat upraised from the bottom, and the pec- torals and ventrals can be used as props for the body. Heckel and Kner have remarked of the European species, that even when the fish remains still or floats, the last three or four rays of the very erect dorsal are alone in a constantly oscillating motion. And, strange to say, they add, the fishes remain thus at rest often for hours at a time, now in a horizontal, now in a vertical position, with the head turned either upward or downward. Then suddenly they all dart from the bottom to the surface with a quick movement of the tail, gasp for air, emit it again in diving as large bubbles through the gill-opening, and for some time afterward breathe very slowly. Heckel and Kner also remarked on their peculiar movements when swimming. The pectorals and ventrals are moved alternately, like the feet of a dog running, and the dorsal makes a quick wavy motion with all its rays, the like of which is seen in seahorses and pipe- fishes (Hippocampus and Syngnathus), which is effected by a pecu- liar disposition of individual muscles for the various rays of the fin. But most of their life is spent at the bottom, and when a clear pool reveals no evidence of their existence, a little stirring up of the bottom may bring many to view. “ On disturbing them, occasionally, instead of swimming, especially if the water is very shallow, they make a forward movement, by giving these fins a leg-like motion, indicated by leaving faint traces, thus: ((((( upon the sand.” But it is also “often a voluntary movement on their part,” and, in fact, “if suddenly disturbed, they generally dart off by swimming only, and bury themselves, tail-foremost, in the mud.” They are perfectly at home there and Abbott claims that they “can pass through soft mud with as much ease [not quite!] as other fishes do through the clear waters.” They are mostly carnivorous and do not appear to manifest any special preference for anything that is of the flesh, but their capacity is limited by their own size and that of their mouth. Ina state of nature, they feed largely upon the aquatic larvee of insects as well as crustaceans, and the eggs of fishes as well as young and small fishes. They also make excursions from the water in search of insects. ‘‘ Unlike any other of our fishes,” Abbott asserted, “the mud-minnow will leap twice and thrice its length above the surface of the water to seize a fly or beetle that happens to rest upon some overhanging blade of grass or twig.” But, although mostly carnivorous, they are not entirely so, as has been asserted. Forbes (1883), indeed, found that, in the “mud GILL] THE UMBRAS 301 minnow” (Umbra limi) of Illinois, “ vegetable food amounted to forty percent, chiefly Wolffia”’ (an aquatic lemnaceous plant), and “a considerable quantity of unicellular algze was also taken by one; mollusks, eaten by two, were reckoned at five percent, all Physa; insects drop to fourteen percent, chiefly undetermined larve. No terrestrial forms were recognized. Corresponding to the greater development of the gill-rakers, we find the Entomostraca assuming greater importance in the food. These were reckoned at ten percent ; three percent additional consisting of Crangonyx gracilis,’ a small amphipod crustacean. Five specimens of the Umbra limi obtained from a pond, “covered in September with a film of Wolffia and other vegetation,” yielded to the dissector stomach contents consist- ing of sixty percent of the Wolffa; of the remainder “ about one- fourth consisting af Entomostraca” and the rest of “ unrecognized insects.” In captivity they will readily take small shreds of meat as well as their natural food. ‘ When kept in aquaria they will devour any reasonable number of flies offered them, and undertake without hesi- tation to swallow earth-worms as large as themselves. Once they take hold of a worm, they never let go, but at least secure that portion of the animal between their jaws. Not only do they allow themselves to be fed, but they will leap above the water to seize any tempting morsel held above them.” Long ago (1842) Zadock Thompson was struck by their power of accommodating themselves to different conditions and declared that they can “ live longer than most fishes without water. During droughts, as the waters subside and recede from the caves, they have the power, by a springing motion, of transporting themselves from one little puddle to another. They also have the power of partially burying themselves and living in the mud and among the moist grass-roots, after the other small fishes associated with them are all dead for the want of water. In these situations vast numbers of them are devoured by birds, musk- rats, and foxes.” This power of adaptation enables them to find a winter-home with the least waste of energy and loss of life, and to hibernate in the mud. The mud was found by Abbott in midwinter to be of “ the consistence of cheese, though, of course, it was less firm when the fish entered it, weeks before.’’ As far as he was “ able to determine the fish had burrowed tail-foremost to a depth of from four to nine inches.” In every instance “ the tail was deeper in the mud than the head, the position varying from nearly horizontal to almost or quite perpendicular.” | 302 SMITHSONIAN MISCELLANEOUS COLLECTIONS [VOL. 45 Abbott thought that when they “ had gone into winter-quarters ” in water from “ three to five feet deep, the hibernating slumber was not as profound; and when they were placed in clear water, at a tempera- ture of 40° Fahr., they almost immediately swam about, slowly at first, but with steadily increasing activity, and in from three to five minutes they were in full possession of all their locomotive powers and assumed the statue-like position common to them in summer.” Soon after the returning warmth of spring has liberated the fishes from their long confinement, or even before spring itself has arrived, they manifest renewed vigor and sexual excitement. They had previously segregated themselves, in large part at least, according to sex—males in one lot, females in another. Abbott found in the middle of one February that, “ the weather being mild and spring-like and frogs singing at mid-day,” they were in muddy ditches; a week later there were few of the fishes in the ditches, ‘‘ but a vast number of females, heavy with masses of ova,” had invaded “ the swift, clear waters of the hill-side brooks,” and they continued to ascend in spite of succeeding storms and cold—especially a heavy snow-storm. ‘Of the specimens taken from the rivulets at this time, none were males.” “Certainly the females precede the male fish to the spawn- ing-grounds.” Abbott was convinced that “while these fish at the commencement of winter seek shelter from the cold by burrowing deeply in the mud, at the approach of spring they revive synchron- ously with the maturing of the ova of the female and the milt of the male, and, having thus recovered their wonted activity (during February and March) no severity of the weather appears to deter them from seeking out exceptionally cold waters for their spawning- grounds. This was shown by the snow-storm referred to, after which the female minnows were still found passing up the brooks, forcing their way up miniature cascades with all the agility of salmon, leaping from eddy to eddy, seeking out the most distant points from their muddy summer haunts; and here, where but little water flowed, and with the long grass and twigs projecting from it thickly coated with crystal ice and glistening frost,’ Abbott “ found the plainly colored mud-minnows lying half-hidden among the pebbles and sandy ridges of the brook’s bed.” But if Umbras ascend running streams as far as they can go to spawn, it is not because they cannot do otherwise. The fact that they abound in numberless ponds shut off from all running water is sufficient evidence that they can reproduce in still water. They are said to have also spawned in confinement, as has been claimed by Seeley in case of the European species. GILL] THE UMBRAS 303 In like localities the eggs of the Umbra were found by Ryder (1886), “laid singly upon aquatic plants to which they adhere for a time by means of a thin coating of adhesive matter which in- vests them.” The eggs are about a millimeter and a half (“ 1.6 mm.’) in diameter. A peculiarity of the fertilized egg is that, at the time of the formation of the blastodisk, “ the vitellus displays a most active amoeboid activity of its substance,’ and Ryder remarked that he knew “ of no teleostean egg ” in which “ such amceboid movements of the vitellus are so pronounced and rapid.” On or about “the sixth day” after oviposition, “the young mud-minnow leaves the egg,” and “three days after hatching the air-bladder becomes apparent as a fusiform vesicle behind the pectorals and above the foregut, when the young mish is viewed as a transparent object.” Then “pigment is rapidly developed upon the upper and lateral aspects of the body, and by the sixteenth day the larve have become pretty dark in color, when observed from above.” Fic. 38.—Western Umbra (Umbra limi). After Jordan and Evermann. The mud-minnow is an excellent fish for aquaria, as it will live under conditions which other fishes cannot stand. Some kept in a small jar lived for months without change of water save for that supplied only sufficiently to compensate for what was lost by evapora- tion. The only plants for keeping the water clear were some Oscillariaceous conferve. The fishes would ascend from time to time to the surface and reject bubbles of air (carbonic acid gas) and take in with a gulping action fresh air. Some were kept for about a year by the present writer in a large glass jar; during an unusually cold period the water froze solid and the jar was broken. The lump of ice was allowed to melt gradually and all the fishes revived and swam about as lively as ever in the new receptacle furnished them. A sudden transfer of frozen fish to “temperate” or summer water was found to be injurious or even fatal. 304 SMITHSONIAN MISCELLANEOUS COLLECTIONS [VoL. 45 The European Umbra could not be induced by Heckel and Kner to spawn in captivity, and a female which was kept for a year in a small garden basin perished because, although filled to distension with eggs as large as millet seeds, it could not relieve itself of them. When one of a company dies, the rest soon follow it. The Umbras have no economic value and are even considered by many to be poisonous. For a time at least in bygone years they were frequently to be seen in the markets of Vienna generally mixed up with a loach (Misgurnus fossilis), but not sold independently. Of course they were not sought for or caught intentionally, but simply scooped up with other contents of the net. Fishermen indeed, it is said, have a superstitious belief that it brings bad luck to catch Umbras. LV: Three species of the genus and family are known, distributed and distinguished as follows: Umera UMBRA (Umbra kramert) : the European Umbra.—South- west Europe, especially Hungary. The body is light brownish and has lighter longitudinal lines on the rows of scales including and below the lateral line as well as more or less irregular darker spots scattered over the surface. There is no precaudal bar but a trace of one on the caudal fin near its base. An indistinct darker area exists on the scapular arch behind the branchial chamber. The head has a suborbital spot sometimes developed as a transverse bar. The fins are immaculate. Umepra Limi: the Western Umbra or Mud Minnow.—Basins of the Great Lakes and Ohio and [Illinois rivers, in weedy streams, ditches, and pools. The body is very dark brown relieved by about twelve to fifteen vertical narrow lighter bands, some of which often coalesce; the last is bounded by a blackish precaudal bar and there is a lighter bar on the caudal fin near its base. A silvery bar is on the scapular region behind the branchial chamber. The head has a frenal band extend- ing from snout on to operculum, but skipping eyes. The fins are immaculate. Umbra pyGM#A: the Eastern Umbra or Mud Minnow.—Atlantic slope from Connecticut southward to South Carolina in lowland streams, ponds, and pools. The body is light brown and has about as many longitudinal light lines as rows of scales, that along the lateral line being most vivid. There is a very distinct blackish precaudal bar or vertical spot and a fainter bar at the base of the fin. A silvery band is developed Grint, | THE UMBRAS 305 behind the branchial chamber. The head has a bridle-like band ex- tending from the snout on to the operculum, skipping the eye. The fins are immaculate. There are slight structural differences in proportions of body, head, and fins between the several species, but they are so unim- portant that they need not be specified here. The U. S. National Museum has numerous specimens of all the species. To Dr. Hugh M. Smith, Deputy Commissioner of Fisheries, the writer is also in- debted for the opportunity to observe living individuals of the Eastern Umbra which were obtained by digging or dipping deep into the mud at the bottom of a run near the city of Washington. It is really remarkable that, with such adaptability and such power of endurance as they possess, the range of the genus should now be so restricted. Restricted it undoubtedly is, for the European and American species must be the relics of a once widely dispersed group. Vv Here have been given the essentials of what is known respecting the habits and economy of the Umbrids. Several of the statements, however, require verification, especially such as refer to the segre- gation of the males from the females and the ascent of both up- stream. ‘The entire nature of the oviposition, also, should be espe- cially investigated; and there are the history of the development of the young, the growth from season to season, and the age at which maturity is attained to be inquired into. The present sketch is, therefore, only a preliminary from which to start. A NEW OCCURRENCE OF UNAKITE—A PRELIMI-— NAR Y GPAPERG BY WC PEALE EN INTRODUCTION In 1884, Prof. W. M. Fontaine, of the University of Virginia, sent to the U. S. National Museum a specimen of igneous rock of exceptional interest, occurring at Milams Gap (see Fishers Gap, pl. Lxrx) at the summit of the Blue Ridge, seven miles south of Luray, Va. The rock was entered and placed among the granites under the caption unakite. In the spring of 1903, in response to a letter from Dr. G. P. Merrill, head curator of the Department of Geology, U. S. National Museum, Professor Fontaine sent another consignment of specimens with the associated country rock and at the same time a detailed description of where the unakite might be found. Later in the spring the writer spent a day at the locality in the com- pany of Dr. Merrill, and in the fall he spent three days alone in the region, mapping the rocks and making collections as extensive as the limited time allowed. The following brief descriptions are the result of the two excursions. Before describing the rocks, attention should be called to plate LxiIx. It will be noticed that the boundaries of the tock masses on either side of the road, leading to and from the summit, have been left open. This was rendered necessary as it was not possible to ascertain their true extent in the limited time available; it is hoped, however, that this work may soon be accomplished. Cer- tainly from a petrographical point of view the country is one de- serving of further and detailed study. So far as the writer is aware, this Virginia locality is the second in America where unakite occurs in appreciable quantity, the first being the Unaka mountains between North Carolina and Tennessee, whence the rock derives its name. It is highly probable, however, that the rock is not nearly so limited in distribution as this statement would seem to imply. At Marblehead Neck, Mass., for example, some rather imperfect specimens were obtained a few years ago, the unakite occurring as a dike in diorite (see pl. Lxx, Db). Here the epi- dotic material has undoubtedly resulted from alteration of the colored 306 Wot 455 bles ecIEx SMITHSONIAN MISCELLANEOUS COLLECTIONS 737 @ zseash i Lite 8 Jorderc 8 78 Granite Olivine Basalt ZZ = LURAY, VIRGINIA. Unakite Hypersthene Akerite MAP SHOWING DISTRIBUTION OF UNAKITE AND ASSOCIATED ROCKS NEAR (Base map from the Luray sheet of the U. S. Geological Survey.) SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 45, PL. LXX a. Polished section of unakite. Natural size. (Milams Gap, Va.) 6. Dike ot unakite in diorite. Natural size. (Marblehead Neck, Mass.) PHALEN] NEW OCCURRENCE OF UNAKITE 307 components of the adjacent rock and is secondary. Such a peculiar combination of quartz, orthoclase, and epidote should not be uncom- mon in regions of dynamic or static metamorphism where granites . or other feldspathic rocks occur, containing the colored micas, pyroxenes, amphiboles, etc. The term unakite was applied originally in 1874 by F. H. Bradley, who, in the reference cited below, says: “This name unakite is proposed for a member of the granitic series from the Great Smoky Mts., a portion of the Unaka range of the Blue Ridge, which range forms the boundary between North Carolina and Tennessee. “The specimens thus far seen are from the slopes of the peaks, known as ‘ The Bluff,’ ‘ Walnut Mt.,’ and ‘ Max’s Patch’ in Cocke Co., Tenn., and Madison Co., N. C. The rock is said to occur also in Yancey Co., N. C., but in a comparatively inaccessible region. “The character relied on for the separation of the species is the constant replacement of the mica of common granite or the horn- blende of syenite by epidote. The amount of this ingredient present is quite variable, in some cases even exceeding one half of the whole mass. ‘The feldspar present is orthoclase of various shades of pink, forming from one fourth to one third of the whole. The quartz is mainly white, but occasionally smoky; its isolated portions form but a small part, say one fourth of the mass; it is veined in structure, but this is probably not a constant character. Small grains of magnetite are scattered through the rock, but not so thickly as in many granites. No other ingredients have as yet been detected. Mr. G. W. Hawes has determined the specific gravity at 2.79. The rock is very compact and takes a high polish and will doubtless prove a valuable material for ornamental architecture. “The deep weathering of all the rocks of the southern Appala- chians has caused the covering of these slopes with deep beds of debris, which conceal most of the outcrops and the dimensions of the bodies of unakite are therefore, as yet, unknown.” A perusal of the literature since Bradley’s brief description in 1874 reveals nothing more than mere allusions to the name and occurrences of the rock.’ It is not mentioned in the latest petro- graphical lexicon of Loewinsson-Lessing, and, so far as the writer is aware, the only reference to the Milams Gap locality is by Merrill.* Mr. F. B. Laney, who spent the last season in a survey of the building stones of North Carolina, informs me that the unakite of *Am. Jour. Science, 3d series, vu, 1874, pp. 519-520. 2 Dana, Manual of Geology, 4th ed., p. 85. Merrill, Rocks, Rock-weathering and Soils, 1897, p. 68. E. H. Williams, Manual of Lithology, N. Y., 1895, p. 128. Williams also calls attention to the occurrence of unakite in the Fichtelgebirge, Schwarzwald, and the Pyrenees. 3 Stones for Building and Decoration, 1903, p. 86. 308 SMITHSONIAN MISCELLANEOUS COLLECTIONS [VoOL. 45 that state occurs as irregularly segregated patches in a foliated or gneissoid epidotic granite, a very different rock, at least macroscopi- _cally, from the country rock at Milams Gap, and that, so far as he is aware, it is confined to Madison county, North Carolina, and Cocke and Sevier counties, Tennessee. GEOLOGY AND PETROGRAPHY The Blue Ridge at Milams Gap, Virginia, is a single range, com- posed entirely of igneous rocks. Proceeding from Stanley, on the Norfolk and Western Railroad, by the public road which runs south- eastward through the village of Marksville, the transported material of Hawksbill creek is finally left behind in large measure and the igneous rock in place is first encountered at an elevation of 1,260 fect. The rock, the hypersthene-akerite of the succeeding pages, con- tinues in almost unbroken continuity to the olivine-basalt, mapped as beginning at an elevation of 2,600 feet. The first fragments of unakite were observed at 1,640 feet, but are probably not in place. The first distinct mass of the material occurs at from 80 to 125 feet below the sharp turn in the road represented on the 2,000-foot con- tour. Much epidosite is associated with it. No sharp lines of demarcation between it and the akerite can be distinguished, and the mass cannot be more than 10 feet wide. The mass, which is prob- ably a continuation of that noted higher up on the road, trends north, 75° to 80° east. It is represented on the map as being rather regular, or as a dike. The masses seen, however, are too irregular to be classed as true dikes; they may more properly be termed irregular patches. The main mass of unakite occurs on the eastern slope of the ridge, where it is associated with akerite, a large mass of which is found in the basalt on this side of the ridge. Most of the material here is in the form of débris. It is to this spot that collectors should go for the material in quantity, “the best specimens being obtained along a foot-path that cuts off the first curve in the road in descending on the east side.”” I may add that excellent specimens may also be obtained in the formerly cultivated fields lying on the right as one continues down the road, just beyond the head of the path referred to. In order to make the discussion of the origin of the unakite more intelligible, the results of the microscopic examination of the country rock will first be presented. * All elevations were determined by means of an aneroid. 2 Fontaine’s letter. PHALEN | NEW OCCURRENCE OF UNAKITE 309 Hypersthene-akerite* Akerite, the country rock of the unakite, is a very coarsely grained, dark grayish green aggregate of feldspars and black pyroxenes with here and there an occasional speck of limpid quartz. The greasy appearance of some of the constituents is strongly suggestive of the mineral nephelite, though the chemical and microscopic analyses show that none of this mineral can be present. The albite twinning of the plagioclase feldspar is plainly seen on many of the fresh frac- tures and an occasional Carlsbad twin is also evident. In thin section the following minerals were noticed: Orthoclase, plagioclase, orthorhombic and monoclinic pyroxene, quartz, micro- cline, iron ore, apatite, and zircon, with the decomposition or altera- tion products, epidote, chlorite, and sericite. Hornblende is absent in Vie most of the slides studied. Ina light-colored segregation situated L near the upper boundary of the 7 akerite (a single instance, it may h be observed), large masses of horn- J 2 blende crystals were noticed. The 0 microscopic features of the horn- i blende will be alluded to in the sub- fo sequent descriptions. The accom- panying drawing (fig. 39) repre- y 0 sents the crystallization of the horn- 2 blende in the light-colored segrega- 0 tion of the country rock. The minerals are arranged in a hypautomorphic mosaic, which is ,,, go Elmatlende 1a iehter phase strikingly clear and beautiful when of akerite. (Scale 1”=1’,) fresh. By far the most abundant mineral is feldspar, which shows evidences of strain in the bent albite lamellae and frequency of wavy extinction. It is often asso- "1 Classified as harzose according to the quantitative scheme of Cross, Iddings, Pirsson, and Washington. For a revision of the calculation of the analyses according to the quantitative scheme, and for examining the proof, I am indebted to Dr. H. S. Washington. Dr. Washington was also kind enough to point out a discrepancy in a former paper by me on “ The Rocks of Nugsuaks Peninsula, Greenland ” (Smithsonian Miscellaneous Collections, xiv, p. 183), in which, in the calculation of the rangs, percentages instead of molecular proportions were employed. This necessitates changing omeose (p. 209) and dellenose (p. 212) to positions in the next subrangs in each instance. They are, therefore, liparose and toscanose, respectively. 310 SMITHSONIAN MISCELLANEOUS COLLECTIONS [VoL. 45 ciated with quartz, producing the micrographic structure. While much of the feldspar shows albite twinning, a larger portion shows no such phenomenon, and it is evident from the frequency of the parallel extinctions that much orthoclase must be present. It does not appear to be so susceptible to kaolinization as the more basic plagioclase. The latter is generally twinned according to the albite law, and its symmetrical extinctions, ranging as high as 18.7°, indicate a feldspar intermediate between andesine and acid labradorite, more nearly the former than the latter. Perthitic intergrowths were noticed; also a curious instance of secondary feldspar twinned according to the albite law, enclosed in a feldspar of less refrac- tive power. This is not true microperthite, and its development is undoubtedly the result of pressure, the enclosing feldspars showing the effect of strain in their wavy extinctions. The more basic feldspars, it has been noted, are the more kaolin- ized. They contain areas of brownish, cloud-like masses, non-re- active between crossed nicols. Other inclusions consist of little tufts or fibers, which polarize brightly; these are often arranged along cracks or cleavage planes and are to be referred to the mineral sericite. Fluid inclusions are present, together with minerals of earlier crystallization, iron oxide, apatite, and zircon. The colored constituents include the light-green monoclinic pyrox- ene, which, from the low extinctions and presence of characteristic partings, is to be referred to the mineral diallage. This monoclinic variety is generally associated with hypersthene, the two minerals being frequently intergrown. Many inclusions of iron oxide, both primary and secondary, are present, the hypersthene being appar- ently very susceptible to change. The secondary iron oxide is often present in masses simulating skeleton forms, which, co- alescing, form larger masses often seamed with dark green chloritic matter. Epidote often occurs along the fissures or cleavage planes. In but few sections studied was amphibole observed. It occurs in rather large patches (one inch and less) in a light-colored segregation of the country rock at an elevation of about 2,400 feet. It is the usual brown variety, strongly pleochroic, as follows: C, brown; A, brownish yellow; and B, chestnut brown with absorption scheme B=C>A. This is rather unusual, but the difference between the absorption parallel to B and C is very slight. The presence of primary quartz is of interest in this relatively basic rock. It occurs in large isolated masses and is also intergrown with the feldspar. Iron oxides (both hydrous and anhydrous), PHALEN | NEW OCCURRENCE OF UNAKITE 3 Ui apatite, and zircon, together with the alteration products, chlorite, epidote, and sericite, complete the list of minerals present. This rock has been referred to the hypersthene-akerites, a type named and described by Broggert as being essentially quartzose augite-syenites, the feldspars comprising both orthoclase and plagioclase. For convenience of comparison in the subsequent dis- cussion, a table of analyses is inserted. | | é gees Sase 6 gveg | sad ad gos BS meee & Na do o> Ae 20 eal cess ee) at oe pppoe Wi tehees fogs | ees " oo aS ob 37 . Hwa a DAS po | BES | bagees | geese | Eas ra a Bo SOR TOM oe Ne @ ef 2 & Bee ae Ley, 5 og 8 8 2 = aa ate 2 ae oe BeaSE ‘bh BE ae eee Re Eo ys SSGC# Be ao 2 mesede monee meek | o o le et Tl te oe - * . é i i & \ =: ; 5 . 7 yi * , : =n * 4% . ‘ = r *, t~ ' ; »* : a - oe: a : -,.° . ‘ sa . = ‘ . (qemais ueqy Aq paiunoyy) ‘WAASAW IVNOILVN ‘S 0 AHL NI NOGOHOVUL JO NOLATAMS wn NOILOATIOD) SNOANVTIAOSIPY NVINOSHLING IIXXT “Td ‘SP “IO A tHe DINOSAUR, ERACHODON ANNECTENS By bo AY LUCAS The skeleton of Trachodon, or Claosaurus, recently placed on exhi- bition in the U. S. National Museum, is an unusually perfect example of that group of extinct reptiles, the dinosaurs. It was included in the Marsh collection and was one of two nearly complete skeletons obtained by Mr. J. B. Hatcher some years ago on Lance creek, Wyoming. The completeness of the specimen is due to the fact that the animal was either engulfed in quicksand, and so came to his end, or that by some favorable accident, such as a cloudburst or a freshet, the body was otherwise covered with sand immediately after death, and before decomposition had set in. Whatever may have happened, the result was that the bones remained in place, the ribs being attached to their respective vertebre and the great thigh bones re- maining in their sockets, the legs even having the position they would take in walking. This is shown in pl. Lxxi1, for in mounting the skeleton the ends of the thigh bones were left as found. Some ex- amples of Trachodon have been obtained in which the impression of the skin was preserved in the surrounding rock, and from these it is known that this animal was covered with small, irregularly six-sided, horny plates, somewhat like those covering portions of the bodies of crocodiles. Unfortunately the wearing away of the rock in which the present specimen was contained had exposed some of the bones, and portions of them had been damaged and the front of the skull weathered away before its discovery in the summer of 1891. The rock containing the bones was then taken up in sections and shipped to Yale University where a large portion of the matrix was removed in order that the bones might be studied. This revealed the presence of long, slender bones, or ossified tendons, that had been embedded in the muscles overlying the backbone in the region of the back and upper part of the tail, and still situated as they were in life. These tendons are not shown in the specimen as mounted, because in order to display the vertebrz it was necessary to remove the tendons and the underlying rock; they are, however, present on the right side which is buried in the background. The object of these tendons is to afford support to the muscles of the back and tail and to strengthen 317 318 SMITHSONIAN MISCELLANEOUS COLLECTIONS [voL. 45 these regions of the body. Similar ossified tendons are found in many diving birds, such as auks and loons, the object being to brace the body against the strain it is called upon to undergo when a fish is caught and is carried in the beak. In the case of Trachodon the strain was due to the weight of the tail, which was, as we shall see, held clear of the ground in walking. The final preparation of the skeleton, its mounting, and the re- storation of the missing or damaged parts was most skilfully per- formed by Mr. Alban Stewart, of the U. S. National Museum, who devoted to it many months of patient work. Trachodon is a typical example of that group of dinosaurs called Predentata because the front of the lower jaw is formed by a bone preceding the tooth-bearing portion, and therefore termed the pre- dentary bone. This was probably encased in horn, like the beak of Fic. 40.—Skull of Trachodon, showing the predentary bone. a turtle, and served for nipping off the branches or herbage on which these animals fed. As the members of the Predentata had but three well-developed toes on the hind feet, and often no more, so that their feet resembled those of birds, the group has also received the name of Ornithopoda, or bird-footed. The animals of this division were all herbivorous and were probably preyed upon by their carnivorous relatives. The particular subdivision, or family, of which Tracho- don is a member is called the _Iguanodontide (iguana-toothed), from the name J guanodon bestowed by Mantell on the first species found in England, the name being given because the teeth were attached to the inner side of the jaw as in iguanas. In the case of Trachodon there were several series of teeth placed one. above the other, the entire series moving slowly upward, new teeth forming at the base to supply the place of those worn away at the top. This arrange- ment greatly increased the number of teeth, there being over two SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 45, PL. LXXIII SKELETON OF TRACHODON (HADROSAURUS) AS RESTORED BY B. WATERHOUSE HAWKINS. LUCAS | TRACHODON ANNECTENS 319 hundred on each side of the lower jaw, so closely packed together as to appear like a mosaic pavement. The family of /gwanodons had an extensive geographical range, for their remains have been found in Austria, Belgium, and England; and in New Jersey, Colorado, Wyoming, Montana, and Alberta, in North America. The first hint of their presence in this coun- try was the discovery of teeth in the Judith River region, Montana, in 1856; but not until a partial skeleton was found near Monmouth, New Jersey, was it fully realized that these teeth belonged to some relative of the Euro- teeth at the top. Fic. 41.—Series of five teeth of Trachodon, show- ing the new teeth at the bottom, and the old, worn Somewhat reduced. pean Iguanodon. The New Jersey specimen was described by Doctor Leidy under the name Hadrosaurus foulkii, and the bones, Fic. 42—Group of Trachodon teeth (1) seen from the outer side and (2) from above. Much reduced. which were deposited in the Academy of Natural Sci- ences of Philadelphia, fur- -nished the basis for a restor- ation of the skeleton by Mr. B. Waterhouse Hawkins. At this time the structure of the dinosaurs was not well known, and Mr. Hawkins, who was not a comparative anatomist, while recognizing the upright position of the animal, restored the missing parts from the skeleton of a modern iguana, the result being the skeleton shown in plate’ Exxtn) |” While=this restoration was far from cor- rect, especially in regard to the skull, it was a decided im- provement on that of Jguanodon shown at the Crystal Palace, 320 SMITHSONIAN MISCELLANEOUS COLLECTIONS [VOL. 45 Sydenham, which represented this dinosaur as a sort of elephantine reptile walking on all fours. That Trachodon and his kindred walked erect and carried their tails clear of the ground is indi- cated by their structure, while much light is thrown on the sub- ject by tracks made by various dinosaurs. At Hastings, England, is a series of thirty great bird-like tracks, ascribed to Jgwanodon, and showing no imprint of fore-feet nor any furrow such as would have 1 been made by the dragging of a heavy tail. Just such furrows, associated with the prints of four feet, are present in some of the famous specimens from Connecticut valley, while they are absent when the impressions are three-toed, like those at Hastings. The front foot of Trachodon had four toes, the innermost being movable, somewhat like a thumb, so that the foot could be used for grasping. The forefoot, however, was. by no means so long and slender as might be inferred from a glance at the skeleton, since the longer bones were surrounded by flesh, the effect being that of a hand with an ex- ya Aa ag ee ceedingly long palm and short, stubby fingers. Tepe odan: The forefoot of the related Jgwanodon had a thumb like a short, stout spike, standing out at right angles to the other digits (figure 43) and probably serving some useful purpose in gathering or handling food. When this spike was first found it was separated from the other bones of the hand, and was supposed to belong on the nose, after the fashion of the horn of the rhinoceros. What may be called the companion to the skeleton in the U. S. National Museum, that mentioned near the beginning of this article, is on exhibition in the museum of Yale University, and was the first complete skeleton of a dinosaur to be mounted in the United States. The Yale specimen is slightly the larger of the two, measuring a trifle over 29 feet in total length, but had suffered more from the weather and so needed more restoration. The skeleton in the U. S. National Museum is 26 ft. 4 in. long; 11 ft. 6 in. high from the base to the top of the head, and 8 ft. 2 in. to the top of the hips; the skull is 3 ft. 5 in. long, the thigh bone 3 ft. 4 in., while the track would have been about 21 inches in length and breadth. CEAS> IFICATION OF THE HARES AND THEIR ALLIES By MARCUS WARD LYON, Jr. CONTENTS PEO G ICO fie amie Sec. Me Viele SUM a ca he dwt Wenig be belie ak I]. List of names that have been applied to the existing Hares, Rab- PRRs peceed EL CUT ISASIA YO ns cree. Meme hc a1 nla PON Sa ing at chose nay Caaf apap III. List of the existing species of the Duplicidentata, arranged by Seierageandy sun cen eran toenail assisted Neen een ene tre IV. General consideration of the skeleton and teeth of the Dupli- GAGEMIbA tee Toe aerciee tains Soe We aves le eee hoe ARE. OREE ioe Mee od baie V. Table showing principal osteological differences between the fam- lbs, OYelavorrovancles eunval ILgyerneba, 45 §a6nccogounduousescoucantesue VI. Keys to Families, genera, and subgenera of See Moan Based primarily on dental characters. Based primarily on cranial chancery. Based primarily on skeletal characters othier sie fees. Of. tHe GUS BUD ie eon dire ati cethe ok ee acho rns OG ETT IO RC ERIG eG Ce OCI aC a VII. Detailed account of the genera and subgenera of the existing VALTT. IX. xX. ETA ES awe txcalllo lo ltssoms aun Gl anllent kel Gime aye fos aya oe aaa Sache stayeecten ei ee sete ie ents PATI Otel Cl ce Were Atte 5 een ee earn United States National Museum, collection Biological Survey, Department of Agriculture. ® Incomplete. LYON | 49589' Sylvilagus floridanus. . 89514" Syluilagus floridanus Salat. 16248 Sylvilagus floridanus mallurus.....-... THE HARES AND THEIR ALLIES 323 .. Florida. .Four Mile Run, Virginia. . Carlisle, Pennsylvania. 49624 Sylvilagus floridanus transitionalis..... Monroe county, New York. 11644 Sylvilagus nuttallii....................Fort Bridger, Wyoming. 920657 Sylvilagus truet...........-.+..-+-..--Metlaltoyuca, Mexico. 113432 Sylvilagus minensis.................-..Chapada, Brazil. 49665 Sylvilagus minensis, young............Chapada, Brazil. FUSER ES VIVUGEUS ISP. cade 0-6 «cs sas oe pncn ys. « ewiston,. Idaho, 94197° Sylvilagus sp.. .... Monitor valley, Nevada. 49580° Limnolagus patidicole.. sao agolelkonatae, 40645 Oryctolagus cuniculus..................-Germany. BOOKS O7VCTOlOBUS CUNICUIUS..... 0. . 910s shri car England. 49386 Oryctolagus cuniculus, lop-eared....... Domestic. 49677" Oryctolagus cuniculus, Belgian hare....United States. 57054 Romerolagus nelsom.............+---.-- Mt. Popocatepetl, Mexico. 22972° Pronolagus crassicaudatus..............South Africa. 93605° Brachylagus idahoensis................lone valley, Idaho. 935960 Brachylagus idahoensis...............-lone valley, Idaho. Q11G8 Ochotona (Pika) sawxatiis............. Cabinet mts., Idaho. 30990° Ochotona (Pika) saxatilis.............Salmon River mts., 49620' Ochotona (Pika) saxatilis.............Oregon. 49500' Ochotona (Ochotona) ladacensis......Hanlé district, Ladak. Idaho. Besides these, I have had access to a large number of skulls, mostly of American species. These are indicated at the end of the description of each genus or subgenus. The material examined has been altogether too limited for any- thing like complete work on this interesting group of rodents and much more must be compared before a satisfactory determination of various relationships can be made. Yet, in the course of study so many interesting points have been disclosed, some already pub- lished and some apparently never before recorded, that it seems advisable to set forth our present knowledge in as complete a manner as can now be done. As an aid to a clearer understanding of the text, the following outline of the classification therein adopted is here given: The existing Duplicidentata are considered as composed of two distinct families, the Leporidz, hares and rabbits, and the Ocho- tonide, pikas. The Leporide are regarded as containing the following ten dis- tinct genera, most of which have heretofore been recognized as sub- genera. Two are here described as new. *United States National Museum. * United States National Museum, collection Biological Survey, Department of Agriculture. 324 SMITHSONIAN MISCELLANEOUS. COLLECTIONS [VOL. 45 1. Lepus, with the largest number of species and the most ex- tensive geographical distribution. It contains three well-marked subgenera and a number of species which from lack of available material cannot now be satisfactorily classified. (a) Subgenus Lepus, represented by the well-known species, timidus, arcticus, europeus, campestris, and their allies. (b) Subgenus Macrotolagus, containing the jackass hares of Mexico and southwestern United States. (c) Subgenus Pecilolagus (new), containing Lepus americanus and the related species. 2. Sylvilagus, with two subgenera. (a) Sylvilagus proper, containing the wood rabbits or cotton- tails of North and South America. (b) Microlagus, containing a few small forms from western and southwestern United States. 3. Oryctolagus, represented by the rabbit of the Old World, O. cuntculus, 4. Limnolagus, containing the swamp rabbits and water hares of southern United States. 5. Brachylagus, a small short-tailed rabbit of western United States, B. tdahoensis. 6. Pronolagus (new), containing the cape hare of South Africa. 7. Romerolagus, the peculiar little rabbit of Mount Popocatepetl, Mexico. 8. Nesolagus, from Sumatra. 9g. Caprolagus, from the southern foothills of the Himalayas. 10. Pentalagus (new), from the Liu Kiu islands south of Japan. The family Ochotonidee contains but one existing genus, Ochotona, with a number of species inhabiting the northern parts or high mountain ranges of the Northern Hemisphere. It contains three well-marked subgenera. (a) Ochotona, containing Jadacensis, and allied forms from central Asia. (b) Ptka, containing alpina and the North American species. (c) Conothoa (new), represented by roylii and related species. In the preparation of this paper I have received from Mr. Gerrit S. Miller, Jr., many valuable suggestions and generous criticisms, which are here gratefully acknowledged. My thanks are due also to Dr. C. Hart Merriam for the use of all the skeletons and many separate skulls of the Duplicidentata in the collection of the Biolog- ical Survey of the United States Department of Agriculture; to Dr. Milton J. Greenman for the use of two specimens of Pentalagus LYON] THE HARES AND THEIR ALLIES 325 furnessi in the Wistar Institute of Anatomy, Philadelphia, and to Dr. J. A. Allen for the use of a skeleton of Lepus gichiganus and skull of Ochotona kolymensis in the American Museum of Natural History, New York. I have also to thank Mr. Oldfield Thomas, who presented to the U. S. National Museum a skeleton of Orycto- lagus cuniculus and two young skeletons of Sylvilagus minensts; Dr. E. A. Mearns, who collected for the National Museum a series of skeletons of Sylvilagus floridanus and the only available skeleton of Limnolagus; and Messrs. Witmer Stone and James A. G. Rehn for furnishing references and copies of Blyth’s figures of Caprolagus. i List OF NAMES APPLIED GENERICALLY OR SUBGENER= ICALLY TO THE EXISTING HARES, RABBITS, AND PIKAS ABRA Proposed by Gray (Cat. Mammals, Birds, etc., presented by B. H. Hodg- son to Brit. Mus., 2d ed., p. 11, 1863) as a subgenus of Lagomys. Type Lagomys (Abra) curzonie Hopvcson from the Himalayas of Sikkim, India. Preoccupied by Abra Leacu, 1818, a genus of Mollusca, fide Palmer, North American Fauna, No. 23, pp. 71, 860, January 23, 1904. BRACHYLAGUS Proposed by Miller (Proc. Biol. Soc. Washington, xi, p. 157, June 13, 1900) as a subgenus of Lepus for L. idahoensis MrERRIAM, the only species and the type. In the present paper it is considered a distinct genus. CAPROLAGUS Proposed by Blyth (Journ. Asiatic Soc. Bengal, xtv, 1845, p. 247) as a genus to include Lepus hispidus PEARSON. Blyth (Cat. Mam. Mus. Asiat. Soc. Calcutta, 1863, p. 133) subsequently replaced Caprolagus hispidus in the genus Lepus. Gray (Ann. Mag. Nat. Hist., xx, 3d ser., 1867, p. 225) under the misprinted 1ame Carpolagus regarded it as a distinct genus for Lepus hispidus PEARSON. Trouessart (Catalogus Mammalium, vol. 1, fase. itl, p. 664, 1897) places two species under Caprolagus as a subgenus of Lepus, hispidus PEARSON, and netschert SCHLEGEL, the latter being regarded in this paper as the type of the genus Nesolagus MAyjor. Major (Trans. Linn. Soc. London, 2d ser., vit, Zool., p. 514, November, 1899) regards Caprolagus as a distinct genus containing the three species sivalensis Major, valdarnensis WeritH, and hispidus PEARSON, the first two extinct. He also uses the name in a larger sense for a group, including (1) Caprolagus, (2) Nesolagus, (3) Oryctolagus, (4) Sylvilagus (the last con- taining (a) Limnolagus, (b) Romerolagus, (c) Tapeti, and (d) Sylvilagus), contrasted with a Lepus group containing the one genus Lepus (this the same as the genus Lepus of the present paper). 326 SMITHSONIAN MISCELLANEOUS COLLECTIONS [VOL. 45 Stone (Proc. Acad. Nat. Sci. Philadelphia, 1900, p. 462) regards Caprolagus as a distinct genus containing the following species: /ispidus PEARSON, netscheri SCHLEGEL, furnessi STONE. The last two are considered in the present paper as the types of the genera Nesolagus and Pentalagus respectively. CARPOLAGUS Used by Gray (Ann. Mag. Nat. Hist., xx, 3d ser., 1867, p. 225), typograph- ical error, for Caprolagus BLYTH. CHIONOBATES Used by Kaup (Entw. Gesch. und Natiir. Syst. Europ. Thierwelt, 1, p. 170, 1829) as a genus for two species, variabilis and borealis. It is antedated by Lepus LINNZUS. CUNICULUS First used as a name for rabbits by Meyer (Mag. f. Thiergesch., 1, pt. i, 52-53, 1790) applied to Oryctolagus cuniculus (renamed by Meyer campestris) and domesticus, angorensis, argenteus, russicus, and dauricus Europe, and to brasiliensis Brazil. The name is synonymous with Lepus LINN&Uus, containing the same three genera, cuniculus representative of Oryctolagus, dauricus = tolai (Erxleben) a member of the genus Lepus, and brasiliensis a member of the genus Sylvilagus. Cuniculus was next used by Gloger (Hand- u. Hilfsbuch Naturgesch., 1, p. 104, 1841) for Oryctolagus cuniculus renamed Cuniculus dasypus. It was proposed again by Gray (Ann. Mag. Nat. Hist., xx, 3d ser., 1867, p. 224) as a genus for “Lepus Section C” Baird (Mammals of North America, 1857, p. 575). The only species included by Gray in the genus is fodiens Kie1n which he considered the correct name for Lepus cuniculus Linnzus, which is accordingly the type. The name is preoccupied (Cuniculus Brisson, Regn. Animal, 1762, p. 13, and Cuniculus WaAGtER, Nat. Syst. d’Amphibian, 1830, p. 21) and was replaced by Oryctolagus LittjEBorG (Sveriges och Norges Ryggradsdjur, 1, 1874, p. 417). EULAGOS Used by Gray (Ann. Mag. Nat. Hist., xx, 3d ser., 1867, p. 222) as a genus for Lepus mediterraneus WaGNER, and Eulagus jud@e@e “the Holy Land Buneas, Tristram.” The word has apparently not been used by subsequent writers, who have placed these species under the subgenus Lepus of the genus Lepus. I have seen no material of either. HYDROLAGUS Used by Gray (Ann. Mag. Nat.:Hist., xx, 3d ser., 1867, p. 221) as a generic term for “ Lepus, Section F” of Baird (Mammals of North America, p. 575). Gray places in it two species: aquaticus BACHMAN, and palustris BACHMAN. The former, the first named species and the one from which the word Hydrolagus was evidently derived, may be considered the type. Trouessart (Catalogus Mammalium, 1, fasc. iii, pp. 657, 658, 1897) uses the name as a subgenus of Lepus for the following species: aquaticus BAcH- MAN, aquaticus attwateri ALLEN, palustris BACHMAN, paludicola MILLER and LYON] THE HARES AND THEIR ALLIES 327 Bancs, and truei ALLEN. In the present paper the last named species is referred to the genus Sylvilagus. As shown by Mearns (Science, n. s., v, March 5, 1897, p. 303) Hydrolagus Gray, 1867, is antedated by Hydrolagus Gitt, 1862, for a genus of fishes. (See Limnolagus.) LAGOMYS Proposed by Cuvier (Legons d’Anat. Comp., Table, 1800, characterized in Table élémentaire de l’histoire naturelle des animaux, p. 132, 1798) as a genus to include the animal previously known as Lepus alpinus, the only named species and the type. From 1800 up to 1896 Lagomnys was in general use among systematists as the generic term of the pikas. At the latter date Thomas (Proc. Zool. Soc. Lond., 1896, p. 1026) adopted the older name Ochotona Link (Beytrage Natur- geschichte, 1, pt. I, p. 74, 1795). Under Lagomys, Gray (Ann. Mag. Nat. Hist., xx, 3d ser., p. 220, 1867) included the following species: Asia: alpinus Cuvier, pusillus DESMAREST, rufescens GRAY, hodgsoni BLiytTH, nepalensis Hopcson, roylii OcILBy, hyper- boreus WAGNER; America: princips RICHARDSON, minimus LorD. Lagomys Cuvier is preoccupied by Lagomys Storr, 1780 (Prodromus Methodi Mammalium, p. 39), a substitute for the name Arctomys (fide Miller, North American Fauna, No. 12, p. 13, July 23, 1897). LAGOPSIS Used by Rafinesque (Analyse de la Nature, 1815, addendum, p. 219) as an emendation of Lagopsys used on p. 58. Nomen nudum. Used by Schlosser 1884, and Major 1809, for fossil forms. LAGOPSYS Used by Rafinesque (Analyse de la Nature, 1815, p. 58). Nomen nudum, ‘Lagopsys R. Lepus sp.; name emended to Lagopsis, ibid., p. 219. LAGOS Used in a sales catalogue by Brookes. “Cat. Anat. & Zool. Mus. of Joshua Brookes, London, p. 54, 1828.” Type: Lagos arcticus (—Lepus arcticus Ross), fide Palmer, North American Fauna, No. 23, pp. 361, 850, January 23, 1904. This questionable name is a synonym of the restricted genus Lepus. LEPUS Used by Linnzus, 1758 (Syst. Nat., 1oth ed., 1, p. 57, 1758), as a genus for four species, representing three modern genera, as follows: Timidus, Chionobates Kaur 1820. Brasiliensis, Tapeti GRAY =part of Sylvilagus Gray 1867. Capensis, Chionobates Kaup, 1829. Cuniculus, Cuniculus G LocEr, 1841 (not of Brisson, 1862) = Oryctolagus LILLyEzorc, 1874. In accordance with the code of nomenclature of the American Ornithologists’ Union, the species cuniculus would become the type, as it represents the last of the non-exotic groups to be removed. 328 SMITHSONIAN MISCELLANEOUS COLLECTIONS [VOL. 45 On the other hand, by a slight extension of the rule recommended in Science, n. S., XVI, pp. 114, 115, July 18, 1902, tumidus may be fixed as the type of the genus Lepus, and no radical changes in the generic names of the Leporide would result, as would occur by a strict application of the principle of elimi- nation. This rule is as follows: “ A generic name which is the same as that of an explicitly included species (or a cited post-Linnzan synonym of such species) takes that species as its type regardless of subsequent elimination.” Since Linnzus could cite no post-Linnzan synonym, the rule can rationally be extended to include, in the case of Linnzus, the names used by earlier writers. Under tiemidus, Linnzus cites Lepus GESNER. In the present paper, timidus is regarded as the type of the genus Lepus, and the word Lepus is retained for the species to which it has ‘been commonly applied. Pallas (Glires, 1778, pp. 1-70), Gmelin (Linnzus, Syst. Nat., 1, 1788, pp. 164-166), Schreber (Satigthiere, vi, pp. 906-918, 1792), use Lepus as the generic name of the pikas as well as of the hares and rabbits. The pikas were placed in separate genera, Ochotona LINK, 1795 (Beytrage Naturgesch., 1, pt. ii, p. 74, 1795) and later Lagomys G. Cuvier, 1798 (Lecons d’Anat. Comp. Table, 1800; characterized in Tab. Elémentaire de |’Hist. Nat. des Animaux, p. 132, 1798). From the time of Link and Cuvier until Gray in 1867, the generic name Lepus was used for both hares and rabbits and as the equivalent of the family Leporide. Gray, however, divided the genus into the following genera: Hydrolagus, Sylvilagus, Eulagos, Lepus, Tapeti, and Cuniculus. At the same time he revived Blyth’s genus Caprolagus (under the misprint Carpolagus) which had been proposed in 1845 and subsequently withdrawn by Blyth. Gray included in the genus Lepus the following species: European.—timidus LINN#&uS, hybridus DESMAREST, aguilonius BLAsIus, variabilis PALLAS, canescens NILSSON. African.—egyptius Grorrroy, hebessinicus HrmpricH and EHRENBERG, isabellinus RUPPELL, capensis LINN#US, saxatilis F. Cuvier, crassicaudatus I. GEOFFROY. Asiatic.—arabicus HrmpricH and EHRENBERG, syriacus HEMPRICH and EHRENBERG, sinaiticus HEMPRICH and EHRENBERG, nigricollis F. Cuvier, rufi- caudatus 1. Grorrroy, tolai PALLAs, tibetanus WATERHOUSE, pallipes Hopc- SON, brachyurus TEMMINCK, sinensis GRAY, altaicus BRANDT. American.—arcticus LEACH, americanus ERXLEBEN, washingtonit Batrp, campestris BACHMAN, callotis WAGNER, californicus Gray, longicaudatus GRAY, auduboni Batirp, trowbridgiit Bairp. Of these crassicaudatus is regarded in this paper as the type of the genus Pronolagus and audubonu and trowbridgit as members of the genus Sylvilagus. From 1867 to 1896 not much attention was paid to Gray’s division of Lepus, but in 1896 Mearns revived Sylvilagus and Hydrolagus as subgenera of Lepus. In the same year, 1896, Merriam described Romerolagus as a new genus of the Leporide, regarding all the other members of the family as congeneric. From that time on the idea has rapidly spread that the family Leporide no longer could be regarded as composed of but a single genus. In 1899 Major LYON] THE HARES AND THEIR ALLIES 329 revived Gray’s names and Blyth’s Caprolagus and gave one new name, Nesolagus. Trouessart (Catalogus Mammalium, 1, fase. iii, 1897, pp. 649-664) gives two genera of the Leporide, Romerlagus and Lepus. Lepus contains the fol- lowing subgenera: Lepus, Hydrolagus, Sylvilagus, Microlagus, Macrotolagus, Tapeti, Oryctolagus, and Caprolagus. All of these except Lepus are discussed elsewhere. Under Lepus as a subgenus Trouessart includes the following species: timidus LINN#&US, europeus PALLAS, mediterraneus WAGNER, tolai PALLAS, mandschuricus RAvpDE, yarkandensis GUNTHER, oiostolus Hopcason, pallipes Hopcson, dayanus BLANForD, nigricollis, F. Cuvier, ruficaudatus Is. GEOFFROY, peguensis BLYTH, sinensis Gray, hypsibius BLANFoRD, swinhoet THOMAS, hainanus SWINHOE, brachyurus TEMMINCK, omanensis THomas, arabicus HempricH and EHRENBERG, jude@ GRAY, sinaiticus HEMPRICH and EHRENBERG, egyptius AupouUIN and GEoFFROY, somalensis HruGLin, berberanus HEUGLIN, tigrensis BLANFORD, microtis HEUGLIN, capensis LINN&usS, whytei THOMAS, victorie THOMAS, crassicaudatus Is. GErorFRoy, saxatilis F. Cuvier, sale JENTINK, arcticus LEACH, grenlandicus Ruoaps, tschuktschorum Norpvouist, americanus ERXLEBEN, campestris BACHMAN. Of these, crassicaudatus is regarded as the type of a new genus Pronolagus in the present paper. Many of the others I have not examined, nor seen figures of them, but those of which specimens and figures are available un- doubtedly belong to the genus Lepus as defined further on. Major (Trans. Linn. Soc. London, 2d ser., vit, Zool., p. 541, November, 1899) uses Lepus as a genus for evidently the same species that Trouessart includes in his subgenera Lepus and Macrotolagus. Miller and Rehn (Proc. Boston Soc. Nat. Hist., xxx, pp. 177-180, December 27, 1901) have included under the subgenus Lepus, the following: americanus ERXLEBEN, americanus bairdit HaypEN, americanus columbiensis Rwoaps, americanus dalli MERRIAM, americanus macfarlani MERRIAM, americanus pheonotus ALLEN, americanus struthopus BANGS, americanus virginianus HARLAN, arcticus Ross, arcticus bangsii RHOADS, bishopi ALLEN, campestris BACHMAN, grenlandicus RuoApDs, klamathensis Merrtam, labradorius MILiEr, othus MrrriaM, poadromus MERRIAM, washingtonii Batrd. In the present paper the genus Lepus corresponds in general with Troues- sart’s and Miller and Rehn’s subgenera Lepus and Macrotolagus. I regard it as composed of a number of species whose relationships cannot be determined at the present time, and at least three distinct subgenera, (a) Lepus, includ- ing arcticus, arcticus bangsti, campestris, grenlandicus, labradorius, othus, and poadromus of Miller and Rehn’s list, together with ewropeus, timidus, and related forms of the Old World; (b) Pecilolagus, containing americanus and its subspecies, together with bishopi, klamathensis, and washingtonti; and (c) Macrotolagus, with the species included under that name by Trouessart and by Miller and Rehn. LIMNOLAGUS Proposed by Mearns (Science, n. s., v, March 5, 1897, p. 393) to replace Hydrolagus Gray (Ann. Mag. Nat. Hist., xx, 3d ser., 1867, p. 221) preoccu- pied in ichthyology by Hydrolagus Gitt (Proc. Acad. Nat. Sci. Philadelphia, 1862, p. 331). Type designated as Lepus aquaticus BACHMAN. 330 SMITHSONIAN MISCELLANEOUS COLLECTIONS [VOL. 45 Major (Trans. Linn. Soc. London, 2d ser., vit, Zool., November, 1899, p. 514) uses it as a subgenus of Sylvilagus for “S. palustris, aquaticus, &c.” Miller and Rehn (Proc. Boston Soc. Nat. Hist., xxx, pp. 183-184, Decem- ber 27, 1901) have included under Limnolagus as a subgenus of Lepus the fol- lowing species: aquaticus BACHMAN, aquaticus attwateri ALLEN, palustris BACHMAN, palustris paludicola MILLER and Banes, telmalemonus Ex.iot, and truet ALLEN. I consider Limnolagus as a genus embracing all the species just mentioned as found in Miller and Rehn, except truei ALLEN which is a member of the genus Sylvilagus. MACROTOLAGUS Proposed by Mearns (Sci., nN. s. L, p. 698, June 21, 1895; Proc. U. S. N. M., XVIII, p. 552, June 24, 1806) as a subgenus of Lepus for the jackass hares of southwestern United States and Mexico. Lepus alleni Mrarns is designated as the type. The following species are included by him in this subgenus: callotis, gaillardi, alleni (type), merriamt, melanotis, griseus, texianus, ercmicus, deserticola, and californicus. Trouessart (Catalogus Mammalium, 1, fase. iii, 1807, pp. 660-662) includes under Macrotolagus as a subgenus of Lepus, the following species: alleni Mearns, callotis WacLErR, gaillardi MEARNS, merriami MEARNS, melanotis MEARNS, texianus WATERHOUSE (including the subspecies eremicus ALLEN, griseus MEARNS, and deserticola MEARNS) californicus GRAy, martirensis STOWELL, imsularis BRYANT. The term is used by Major (Trans. Linn. Soc. London, 2d ser., vu, Zool., pp. 468, 469, November, 1899) as a subgenus of Lepus, for apparently the same group of hares that Mearns applied it to. Miller and Rehn (Proc. Boston Soc. Nat. Hist., xxx, pp. 180-183, Decem- ber 27, 1901) place in the subgenus Macrotolagus the following species: alleni Mearns, alleni palitans BANGs, asellus MILtER, californicus Gray, californicus xanti THomas, callotis WaActER, gaillardi MEARNS, insularis BRYANT, martirensis STOWELL, melanotis MEARNS, merriami MEARNS, texianus WATER- HOUSE, fexianus deserticola MEARNS, texianus eremicus ALLEN, texianus griseus MERRIAM. It is used in the present paper in the same sense as by Miller and Rehn. It is interesting to note that the hares of this subgenus were recognized by Baird (Mammals of North America, p. 574, 1857) as forming a distinct group and constituting his “ Lepus Section B.” MAMLEPUS Used by A. L. Herrera (Sinonimia vulgar y cientifica de los principales vertebrados mexicanos, p. II, 1899) as a name in a new system of nomen- clature for the broad genus Lepus of Linnean nomenclature. MICROLAGUS Proposed by Trouessart (Catalogus Mammalium, 1, fasc. iii, p. 660, 1897) as a subgenus of Lepus for Lepus cinerascens ALLEN, on characters defined by Mearns (Proc. U. S. Nat. Mus, xvi, pp. 552, 553, June 24, 1806). Cuineras- cens ALLEN, the only species named by Trouessart, is the type. It is used by Miller and Rehn (Proc. Bost. Soc. Nat. Hist., xxx, pp. 188, 189) as a subgenus of Lepus for the following species: bachmanit WaATER- LYON] THE HARES AND THEIR ALLIES 331 HOUSE, bachmani ubericolor MILLER, cerrosensis ALLEN, cinerascens ALLEN, peninsularis ALLEN. In the present paper it is regarded as a subgenus of Sylvi/agus, and includes the species just mentioned as recorded by Miller and Rehn. Baird (Mammals of North America, p. 575, 1857) recognized that the mem- bers of this subgenus formed a distinct group included under his section E. MICROTOLAGUS A curious misprint for Macrotolagus MEArns found in Elliot (Synopsis of Mammals of North America and adjacent Seas, Field Columbian Mus., Zool. Ser. II, pp. 269, 288, 1901), repeated by Allen (Bull. Amer. Mus. Nat. Hist., Pp. 607, 1903). MNUOLAGUS ‘Billberg (Syn. Faune Scandinavie, 1, Mamm., Conspectus A. (before p. 1), 1828). Nomen nudum, occurring only in a table between Lagomys and Lepus, fide Palmer, North American Fauna, No. 23, January 23, 1904, pp. 850, 95I. NESOLAGUS Proposed by Major (Trans. Linn. Soc. London, 2d ser., vu, Zool., p. 514, November, 1899) as a genus under the Caprolagus group (or a subgenus of the genus Caprolagus; Major does not definitely state) for Lepus netscheri SCHLEGEL (Notes Leyden Museum, 11, 1880, pp. 59-65). Netscheri is the only mentioned species and the type. Nesolagus is here regarded as a well-marked genus. OCHOTONA First used by Link, 1795 (Beytrage Naturgeschichte, 1, pt. ii, p. 74, 1795) as the generic name of the pikas. The following species are given: pusilla, alpina, and minor (Lepus Ochotona) Linneus, of which ochotona is the type, dauricus is an earlier name, however, for the same species. (See Palmer, N. A. Fauna, No. 23, p. 468, January 23, 1904.) It is the proper generic name of the pikas, for nearly a century called Lagomys Cuvier (Lecons d’Anat. Comp. Table, 1800) owing to ignorance of Link’s rare work. Thomas (Proc. Zool. Soc. Lond., 1896, p. 1026) seems to have been the first to have brought forward Link’s name. It is used in the present paper as a genus for the pikas and also as a sub- generic name for Ochotona ladacensis and related species. OGOTOMA First used by Gray in 1867 (Ann. Mag. Nat. Hist., xx, 3d ser., p. 220, 1867) as the generic name for Lagomys ogotoma of Cuvier and of Waterhouse, the Lepus ogotoma Patias. Gray changes the specific name to pallasi. As this is the only species placed in the genus by Gray, it is the type. The term is antedated by Ochotona Linx (Beytrage Naturgesch., I, pt. ii, p. 74, 1795), which has the same species for the type. ORYCTOLAGUS Proposed by Lilljeborg (Sveriges och Norges Rygegradsdjur, 1, p. 417, 1874) as a subgeneric name for Lepus cuniculus LINNUS which is designated as the type. It was used to replace Cuniculus Gray (Ann. Mag. Nat. Hist., 3d ser., XX, p. 225) which is preoccupied by Cuniculus WaAGLER (Nat. Syst. d’ Amphibian, p. 21, 1830), and Brisson (Regn. Animal, 1762). 332 SMITHSONIAN MISCELLANEOUS COLLECTIONS [VoL. 45 Trouessart (Catalogus Mammalium, 1, fase. iii, p. 663, 1897) uses it as a subgenus for Lepus cuniculus LINN&XUus, the only existing species. Major (Trans. Linn. Soc. London, 2d ser., vit, Zool., November, 1899, p. 514) regards it as a genus containing two species “O. cuniculus (LINN.); O. crassicaudatus (GrorFr.).”” In the present paper the latter is regarded as the type of the genus Pronolagus. Recently Thomas (Ann. Mag. Nat. Hist., ser. 7, x1, January, 1903, pp. 78, 79) following Major has used Oryctolagus in a generic sense. In the present paper it is regarded as a genus for the rabbits, heretofore commonly known as Lepus cuniculus. PICA Proposed by Fischer (Das National Museum Naturgesch. zu Paris, m1, 126, 1803) as a correction of Pika LAcEPEDE, fide Palmer (North American Fauna, No. 23, January 23, 1904, p. 537). PIKA Used by Lacépede in 1799 (Tableau des Divisions de Mammiféres, 1799, p.- 9) for the Alpine pika, called Pika alpinus, the only mentioned species, and accordingly the type. It is used in the present account as a subgeneric term to include Ochotona (Pika) alpina and its related species, embracing all the North American forms. ROMEROLAGUS Proposed by Merriam (Proc. Biol. Soc. Washington, x, December 29, 18096, p. 173) as a genus for a remarkable rabbit found on the west slope of Mount Popocatepetl, Mexico. The type and only species in the genus is Romerolagus nelsont MERRIAM. Herrera (Mem. Revista de la Soc. Cient. Antonio Alzate, xiv, 1899-1900, p. 380; La Naturaleza, 2d ser., tomo mt, 1898, p. 80) considers Romerolagus nelsoni to be a member of the genus Lagomys, a view which is entirely untenable (Nelson, Soc. Cientifica Antonio Alzate Mex., Revista Cientifica y Bibliografica, num. 3, 1901, p. 33, figs. 1-6). Major (Trans. Linn. Soc. London, 2d ser. vit, Zool., November, 1899, p. 514) regards Romerolagus as one of four subgenera of Sylvilagus, the other three being Tapeti, Limnolagus and true Sylvilagus. Stone (Proc. Acad. Nat. Sci. Philadelphia, 1900, p. 462) regards it as one of three genera forming the Leporidz, the other two being Caprolagus and Lepus. Romerolagus is here recognized as a well-marked genus of the family Leporide. SYLVILAGUS This was first proposed by Gray (Ann. Mag. Nat. Hist., 3d ser., xx, 1867, p. 221) as a generic term for “Lepus Section D” of Baird (Mammals of North America, 1857, p. 578). The species first mentioned by Gray is Syl- vilagus nanus (SCHREBER), a synonym of Sylvilagus sylvaticus (BACHMAN), whose specific name was undoubtedly the origin of the word Sylvilagus. It is evident that Gray had Baird’s account in mind and merely copied the synonymy of Sylvilagus sylvaticus as given by Baird, and in this way he happened to use Schreber’s name nanus. As the animal formerly known as Sylviagus LYON | THE HARES AND THEIR ALLIES 335 sylvaticus has had to be renamed Sylvilagus floridanus mallurus (Thomas, Ann. Mag. Nat. Hist., 7th ser., 11, October, 1898, p. 320; Allen, Bull. Amer. Mus. Nat. Hist., xu, March 4, 1899, p. 13), the latter becomes the type of the genus Sylvilagus. Two other species were placed in this genus by Gray, artemesia BACHMAN, and bachmani WATERHOUSE. The name has generally been used by subsequent writers as the subgeneric term for the cottontail rabbits (Sylvilagus floridanus and its allies) of North America. Major (Trans. Linn. Soc. London, 2d ser., vit, Zool., November, 1899, p. 514) uses it in a generic sense for Limnolagus MEARNS, Romerolagus MerrriAmM (both regarded as separate genera in the present account), Tapeti Gray (which I regard as a part of Sy/vilagus), and Sylvilagus Gray, which according to Major embraces “ S. sylvaticus, etc.” Trouessart (Catalogus Mammalium, 1, fase. 11, pp. 658-660, 1897) uses it in a subgeneric sense for the following species: sylvaticus BACHMAN (with the subspecies: transitionalis BANGS, bachmani WATERHOUSE, alacer BANGS, mearnst ALLEN, floridanus ALLEN, pinetis ALLEN, arizone@ ALLEN, holzerni MEARNS, nuttali BACHMAN, auduboni Batrp, aztecus ALLEN), grangeri ALLEN, trowbridgei Batrp, artemisia BACHMAN, arizone ALLEN (with the subspecies major MEARNS and minor MEARNS), verecrucis THOMAS, insolitus ALLEN, orizabe MERRIAM, graysont ALLEN, idahoensis MERRIAM. Of these the last I regard as the only member of the genus Brachylagus proposed by Miller as a subgenus, and trowbridgii BAirD as a member of the subgenus Microlagus of the genus Sylvilagus. Miller and Rehn (Proc. Boston Soc. Nat. Hist., xxx, December 27, 1901, 184-188) have included under Sylvilagus, as a subgenus of Lepus, the fol- lowing species: arizone ALLEN, arizone confinis ALLEN, arizone major MEARNS,. arizone minor MeraArRNS, baileyi MeErrtIAM, floridanus ALLEN, floridanus alacer BANGS, floridanus audubonii Batrp, foridanus aztecus ALLEN, Horidanus caniclums Mr1iEr, floridanus chapmani ALLEN, floridanus holzneri Mearns, floridanus mallurus THomas, floridanus mearnsi ALLEN, floridanus pinetis ALLEN, floridanus rigidus MEARNS, floridanus sanctidiegi MILLER, floridanus subcinctus MILLER, floridanus transitionalis BAnes, floridanus yucatanicus MILLER, grangeri ALLEN, graysont ALLEN, insolitus ALLEN, nuttallit BACHMAN, orizibe MERRIAM, verecrucis THOMAS. Recently Thomas (Ann. Mag. Nat. Hist., ser. 7, vitt, December, 1901, pp. 534, 539) following Major has raised Sylvilagus to generic rank. I consider Sylvilagus a distinct genus, embracing all the forms given by Miller and Rehn under their subgenera Sylvilagus, Microlagus, and Tapeti (also including Lepus truet ALLEN, which is found in their list under the subgenus Limnolagus) and all the South American Leporide. It is also used in the present paper as a subgeneric term for all the species of Sylvilagus just mentioned, except the group embracing Muicrolagus here regarded as another subgenus of Sylvilagus. TAPETI Proposed by Gray (Ann. Mag. Nat..Hist., 3d ser., xx, 1867, p. 224) as a generic term for Lepus brasiliensis LINN®US, which, as the only named species, is the type. 334 SMITHSONIAN MISCELLANEOUS COLLECTIONS [VoL. 45 Trouessart (Catalogus Mammalium, 1, fasc. ili, 1897, pp. 662-663) places the following species under it used as a subgenus, gabbi ALLEN, defilippit CorNELIA, nigronuchalis HARTERT, brasiliensis LINNUS. Miller and Rehn (Proc. Boston Soc. Nat. Hist., xxx, p. 190, December 27, 1901) have included under it (as a subgenus of Lepus) brasiliensis LINNEUS and gabbi ALLEN. Major (Trans. Linn. Soc. London, 2d ser., vir, Zool., November, 1899, p. 514) regards Tapeti as one of four subgenera (the other three being Limno- lagus, Romerolagus, and true Sylvilagus) forming the genus Sy/lvilagus. Gray describes Tapeti as follows: “Skull like Lepus, but with hinder supraorbital notch narrow, the lobes short, with a sharp inner edge; the front of the lower edge of the zygoma dilated, sharp-edged, porous above, hinder nasal opening rather narrower. Tail, none. Ears short.” This description of the skull does not agree with the skulls of Sylvilagus minensis, paraguen- sis or gabbi at hand, and I am at a loss to understand the true status of Tapeti. The available material shows that Tapeti is nothing else than a part of Sylvilagus, and it is so here regarded. lil, LIST OF THE “EXISTING SPECIES OF AE DUPEICIDEN. TATA ARRANGED BY GENERA AND SUBGENERA The species under each group are arranged alphabetically. The list includes all the names that are found in Trouessart’s Catalogus Mammalium, 1897, and in Miller and Rehn’s Systematic Results of the Study of North American Land Mammals to the Close of the Year 1900; (Proc, ,Bostom Soc. Nat. (blist., xxx, December 727 1901) together with the names that have appeared since these two works. Those species of which the writer has seen skulls or skeletons are printed in small capitals, those of which he has seen figures of the skulls or skeletons are in italic, and those of which he has seen neither specimens nor figures are in ordinary type. Lepus (Lepus) ArctTicus Ross. Lepus (LEpuS) ARCTICUS BANGSII Rhoads. Lepus (LEpUS) ARCTICUS CANUS Preble. Lepus (Lrepus) CAMPESTRIS Bachman. Lepus (Lepus) corsicanus de Winton. Lepus (Lepus) creticus Barrett-Hamilton. Lepus (Lepus) cyprinus Barrett-Hamilton. Lepus (LrEpus) EUROP#UsS Pallas. Lepus (LEpUS) GICHIGANUS Allen. Lepus (LEpUS) GR@NLANDICUS Rhoads. Lepus (Lrepus) LABRADORIUS Miller. Lepus (Lepus) lilfordi de Winton. Lepus (Lepus) oTtTHUs Merriam. Lepus (LEpus) PARNAsSIUS Miller. Lepus (LEpus) POADROMUS Merriam. Lepus (LEpus) tTiIminus Linnzus. Lepus (Lepus) timidus ainu Barrett-Hamilton. Lepus (Lepus) timidus lutescens Barrett-Hamilton. Lepus (Lepus) transylvanicus Matschie. Lepus (LEpus) vArRonts Miller. LepusS (PC@cILOLAGUS) AMERICANUS Erxleben. LYON] THE HARES AND THEIR ALLIES so Lepus (Pa@cILOLAGUS) AMERICANUS BAIRDII (Hayden). Lepus (Peecilolagus) americanus columbiensis Rhoads. Lepus (P@CcILOLAGUS) AMERICANUS DALLI Merriam. LepuS (PG@CILOLAGUS) AMERICANUS MACFARLANI Merriam. Lepus (Peecilolagus) americanus pheonotus Allen. LepuUS (PaGcCILOLAGUS) AMERICANUS STRUTHOPUS Bangs. LepUS (PC@CILOLAGUS) AMERICANUS VIRGINIANUS (Harlan). Lepus (Peecilolagus) bishopi Allen. Lepus (PC@cILOLAGUS) KLAMATHENSIS Merriam. Lepus (PC@CILOLAGUS) SALIENS Osgood. Lepus (PCCILOLAGUS) WASHINGTONII Baird. Lepus (MAcRoTOLAGUS) ALLENI Mearns. Lepus (Macrotolagus) alleni palitans Bangs. Lepus (MAcroroLacus) ASELLUS Miller. Lepus (MACROTOLAGUS) CALIFORNICUS Gray. Lepus (Macrotolagus) californicus xanti Thomas. Lepus (MacroroLacus) CALLoTis Wagler. Lepus (MAcROTOLAGUS) GAILLARDI Mearns. Lepus (Macrotolagus) gaillardi battyi Allen. Lepus (MAcROTOLAGUS) INSULARIS Bryant. Lepus (Macrotolagus) martirensis Stowell. Lepus (MAcroTOLAGUS) MELANOTIS Mearns. Lepus (MAcROTOLAGUS) MERRIAMI Mearns. Lepus (MAcROTOLAGUS) TEXIANUS Waterhouse. Lepus (MAcroTOLAGUS) TEXIANUS DESERTICOLA Mearns. Lepus (MAcROTOLAGUS) TEXIANUS EREMICUS Allen. Lepus (MAcROTOLAGUS) TEXIANUS GRISEUS Mearns. Lepus (Macrotolagus) texianus micropus Allen. Lepus egyptius Audouin and Geoffroy. Lepus arabicus Hemprich and Ehrenberg. Lepus atlanticus de Winton. Lepus berberanus Heuglin. Lepus brachyurus Temminck. LEPUS CAPENSIS Linnzus. Lepus capensis centralis Thomas. LEPUS CAPENSIS OCHROPUS Wagner. Lepus crawshayi de Winton. Lepus dayanus Blandford. Lepus etruscus Bosco. Lepus fagani Thomas. Lepus hainanus Swinhoe. Lepus harterti Thomas. Lepus hawkeri Thomas. Lepus Hypsiprus Blanford. Lepus judeze Gray. Lepus kabylicus de Winton. Lepus mandschuricus Radde. Lepus mediterraneus Wagner. Lepus microtis Heuglin. Lepus monticularis Thomas. Lepus nigricollis F. Cuvier. Lepus oiostolus Hodgson. Lepus omanensis Thomas. Lepus pallidor Barrett-Hamilton. Lepus pallipes Hodgson. Lepus peguensis Blyth. LEPUS RUFICAUDATUS Is. Geoffroy. Lepus salae Jentink. Lepus saxatilis F. Cuvier. Lepus schlumbergeri St. Loup. Lepus sechuensis de Winton. Lepus siamensis Bonhote. 336 Lepus LeEpus Lepus Lepus Lepus LEpus Lepus Lepus Lepus Lepus Lepus tolai SMITHSONIAN MISCELLANEOUS COLLECTIONS Pallas. tunetze de Winton. victorize Thomas. whitakeri Thomas. whytei Thomas. LEPUS YARKANDENSIS Gtinther. Lepus zechi Matschie. ORYCTOLAGUS CUNICULUS (Linnzus). [VOL. 45 sinaiticus Hemprich and Ehrenberg. SINENSIS Gray. somalensis Heuglin. swinhoei Thomas. syriacus Hemprich and Ehrenberg. TIGRENSIS Blanford. Sylvilagus (Sylvilagus) andinus (Thomas). SYLVILAGUS (SYLVILAGUS) ARIZONZ (Allen). Sylvilagus (Sylvilagus) arizonz confinis (Allen). SYLVILAGUS (SYLVILAGUS) ARIZON2 MAJOR (Mearns). SYLVILAGUS (SYLVILAGUS) ARIZON2 MINOR (Mearns). SYLVILAGUS (SYLVILAGUS) BAILEYI (MERRIAM). Sylvilagus (Sylvilagus) braziliensis (Linnzus). Sylvilagus (Sylvilagus) cumanicus (Thomas). Sylvilagus (Sylvilagus) defilippii (Cornalia). Sylvilagus (Sylvilagus) durange (Allen). SYLVILAGUS (SYLVILAGUS) FLORIDANUS (Allen). Sylvilagus (Sylvilagus) floridanus alacer (Bangs). SYLVILAGUS SYLVILAGUS SYLVILAGUS SYLVILAGUS SYLVILAGUS SYLVILAGUS SYLVILAGUS (SYLVILAGUS ) (SYLVILAGUS ) (SYLVILAGUs ) (SYLVILAGUS ) (SYLVILAGUS ) (SYLVILAGUS ) (SYLVILAGUS ) FLORIDANUS FLORIDANUS FLORIDANUS FLORIDANUS FLORIDANUS FLORIDANUS FLORIDANUS AUDUBONII (Baird). AZTECUS (Allen). CANICLUNIS (Miller). CHAPMANI (Allen). HOLZNERI (Mearns). MALLURUS (Thomas). MEARNSI (Allen). Sylvilagus (Sylvilagus) floridanus persultator (Elliot). Sylvilagus (Sylvilagus) floridanus pinetis (Allen). SYLVILAGUS SYLVILAGUS SYLVILAGUS SYLVILAGUS SYLVILAGUS SYLVILAGUS ( SYLVILAGUS ) (SYLVILAGUS ) (SYLVILAGUsS ) (SYLVILAGUS ) (SYLVILAGUs ) (SYLVILAGUS ) FLORIDANUS FLORIDANUS FLORIDANUS FLORIDANUS RIGIpUS (Mearns). SANCTIDIEGI (Miller). suBCINCTUS (Muller). TRANSITIONALIS (Bangs). FLORIDANUS YUCATANICUS (Miller). GABBI (Allen). Sylvilagus (Sylvilagus) grangeri (Allen). SYLVILAGUS (SYLVILAGUS) GRAYSONI (Allen). Sylvilagus (Sylvilagus) incitatus (Bangs). SyLvILaAGus (SYLVILAGUS) INSoLITUS (Allen). Sylvilagus (Sylvilagus) laticinctus (Elliot). Sylvilagus (Sylvilagus) laticinctus rufipes (Elliot). SYLVILAGUS (SYLVILAGUS) MARGARIT# (Miller). SyLvILAcus (SYLVILAGUS) MINENSIS Thomas. Sylvilagus (Sylvilagus) nigronuchalis (Hartert). SYLvILAGUS (SYLVILAGUS) NUTTALLIT (Bachman). Sylvilagus (Sylvilagus) orinoci Thomas. SYLVILAGUS (SYLVILAGUS) ORIZAB (Merriam). SYLVILAGUS (SYLVILAGUS) PARAGUENSIS Thomas. Sylvilagus (Sylvilagus) parvulus (Allen). Sylvilagus (Sylvilagus) russatus (Allen). SYLVILAGUS (SYLVILAGUS) sIMPLICANUS (Miller). Sylvilagus (Sylvilagus) superciliaris (Allen). Sylvilagus (Syilvilagus) surdaster Thomas. SYLVILAGUS (SYLVILAGUS) TRUEI (Allen). SYLVILAGUS (SYLVILAGUS) VERZCRUCIS (Thomas). SYLVILAGUS (MrIcroLAGUS) BACHMANTI (Waterhouse). Lyon] THE HARES AND THEIR ALLIES a7 SYLVILAGUS (MIcROLAGUS) BACHMANI UBERICOLOR (Miller). Sylvilagus (Microlagus) cerrosensis (Allen). SYLVILAGUS (MICROLAGUS) CINERASCENS (Allen). Sylvilagus (Microlagus) peninsularis (Allen). LIMNOLAGUS AQUATICUS (Bachman). LIMNOLAGUS AQUATICUS ATTWATERI (Allen). LIMNOLAGUS PALUSTRIS (Bachman). LIMNOLAGUS PALUSTRIS PALUDICOLA (Miller and Bangs). Limnolagus telmalemonus (Elliot). BRACHYLAGUS IDAHOENSIS (Merriam). Caprolagus hispidus (Pearson). PRONOLAGUS CRASSICAUDATUS (Is. Geoffroy). Pronolagus crassicaudatus curryi (Thomas). Pronolagus crassicaudatus nyike (Thomas). ROMEROLAGUS NELSONI Merriam. Nesolagus netscheri (Jentink). PENTALAGUS FURNESSI (Stone). Ochotona (Ochotona) curzonie (Hodgson). Ochotona (Ochotona) daurica (Pallas). Ochotona (Ochotona) koslowi Buchner. OcHONOTA (OCHOTONA) LADACENSIS Gtnther. Ochotona (Ochotona) melanostoma Biichner. Ochotona (Pika) alpina (Pallas). OcHoTona (PrKA) coLLARis (Nelson). OcHoToNA (PIKA) cUPPES Bangs. Ochotona (Pika) hyperboreus (Pallas). OcHoTONA (PIKA) KOLYMENSIS Allen. Ochotona (Pika) littoralis (Peters). OcHOTONA (PIKA) PRINCEPS (Richardson). Ochotona (Pika) pusilla (Pallas). OcHoTONA (PIKA) SAXATILIS Bangs. OcHoTONA (PIKA) SCHISTICEPS (Merriam). Ochotona (Conothoa) erythrotis Bichner. OcHoTONA (CONOTHOA) ROYLEI (Ogilby). Ochotona rufescens (Gray). Ochotona rutila (Severzow). IV. GENERAL CONSIDERATION OF THE SKELETON AND CEE OF THE DUPLICIDENTATA SKUEL The skulls of the two families Ochotonide and Leporide are widely different in nearly every respect, and were it not for the structure of the teeth and the number of the upper incisors there would be little to indicate that the two groups were closely related. The skulls of members of both of these families have so often been described and figured that there is here no need of a general description of that important part of their osteology, which will be given later in regard to less well known parts. * An earlier name for ochotona the type. ) 338 SMITHSONIAN MISCELLANEOUS COLLECTIONS [VOL. 45 LEPORIDE (Pirates LXXIV-LXXXIX) The skulls of the Leporide fall into several groups, each distinct from the other, and no specimens have been seen which show inter- mediate conditions between any two of the divisions. The varia- tions upon which these groups are founded consist principally in the shape, size, and method of the attachment of the postorbital proc- esses, the distinctness of the interparietal bone, the distance between the two vertical portions of the palate bones, or width of the choane, and the relative heaviness of the zygoma. Each of these points will be considered in detail. POSTORBITAL PROCESSES The postorbital processes are conspicuously developed in all of the Leporidz. In general and typically the process is triradiate, one arm being attached to the skull and forming the pedicle of the process, the other two arms or angles being toward the outside, one directed anteriorly, the other posteriorly. The following seven forms of postorbital processes are found: 1. Postorbital processes large and triangular, standing out from the side of the head and considerably arched from before backward. This form is best developed in the subgenus Lepus, where the process is a conspicuous triangle, one angle of which is attached to the skull, the other two angles of which are entirely free, the anterior subtending a large anterior notch, and the posterior sub- tending a larger posterior notch. Occasionally in some specimens the anterior angle of the postorbital process is directed inward, its apex meets the frontal bone and a distinct foramen is formed instead of a notch. In the subgenus Macrotolagus the posterior angle of the postorbital process is always directed inward to meet the side of the cranium and in this way forms the outer boundary of a con- spicuous foramen. Some of the Old World specimens resemble Macrotolagus in this respect; such are Lepus sp. Jumna river, India; "L. ochropus,, East Adrica;-L. ypsipius, Badak candeee: tibetanus, central Asia. In the hares belonging to the subgenus Pecilolagus the postorbital processes, while of the same general form, are much slenderer, the outer angles are not so wide, and the process is not so arched as it is in its best developed form in the subgenus Lepus. Both anterior and posterior angles are free, and help to form corresponding notches with the rest of the skull. The posterior angle and notch are larger than the anterior angle and ‘Ebb aSed aas uoneur[dxe 10g “suga7 snuasqns ‘vz ‘ec “sn josIvy snuasqns ‘vr ‘1 *(azis jermyeu 7 noqe) SHYAVH AO STTOMNS AIXX1 “1d ‘Sh "10 A SNOLLOATIO) SQOANVTTAISIP, NVYINOSHLING “Eh aded aes uojeuridxa iog ‘sugaz7 snuasqns *€ ‘sxSvzpojo4ovpy snuesqng “ “sxévjzoz1I9Mq snuesqng “rt -(azis [einqeu 2 ynoqe) SAUVH AO STTIONNS S$ 4 AXXT “1d ‘S? "10Q AT10Z) SQNOANVITAOSIPY NVINOSHLING SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 45, PL. LXXVI ‘Ebb aded vas uoneurdxa Jog “(azis jeanjeu 2 ynoqe) SLIGAVY AO STINMS “Ia ‘SP “10 A SNOILIATIOD SQOANVTIAOSIFY NVINOSHLING SMITHSONIAN MISCELLANEOUS COLLECTIONS Vov. 45, PL. LXXviil SKULLS OF RABBITS (natural size). 1. Romerolagus. 2. ronolagus. 3. Brachylagus. 4 a 4 < S > % je] ©) SMITHSONIAN MISCELLANEOUS COLLECTIONS LYON | THE HARES AND THEIR ALLIES S57 and seventh still less so, forming only a narrow arch over the vertebral canal. Not many variations are found in the cervical vertebre in the different genera of the Leporidz, but in general two forms occur: First, large rabbits, in which the individual vertebre are uniformly lengthened, and in which the costal process does not project laterally from the centrum to a marked extent, and the anterior and posterior spines of the costo-transverse process are more elongated. Cervical vertebrz of this type are found in all the skeletons belonging to the genus Lepus. Second, rabbits averaging smaller in size in which the individual cervical vertebre are uniformly shortened and in which the costal process projects further laterally from the centrum than it does in the first group, the anterior and posterior spines of which are less pronounced. The true transverse process is slightly more con- spicuous here, and often begins to project laterally from the vertebra, beginning with the fifth cervical instead of with the sixth as in the case of the more elongated type of cervical vertebre. All the genera of the family Leporidz, with the exception of the genus Lepus, belong to this section. The most extreme development of the shortened type of cervical vertebrze is seen in Pronolagus, in which the costal process stands out still farther from the body of the vertebra. The process is narrower, that is, its antero-posterior dimensions are relatively much less than they are in the other genera. The cephalad and caudad projecting spines of the costal processes are apparently not well developed, but they have a somewhat worn or damaged look in the only specimen examined. The general appearance of the cervical vertebrz in Pronolagus, when viewed from below, is much as it is in Ochotona. Thoracic V ertebre.—There are twelve thoracic vertebra, of which the more anterior are wider from side to side than they are long, while the reverse is true of the posterior vertebrae. The general size of each individual vertebra increases as one passes from before backward. The centra, often at the beginning of the thoracic series and nearly always toward the end of it, have a low ventral median ridge, which on some of the anterior lumbar vertebrz is produced into a spine, the hypophysis. In the first eight vertebrze of the thoracic series, the transverse processes are well developed and each is furnished with a large facet for articulating with the tubercle of the rib. In the ninth, 358 SMITHSONIAN MISCELLANEOUS COLLECTIONS [VoL. 45 tenth, eleventh, and twelfth thoracic vertebrz, as a rule, the trans- verse process is absent, the ribs attached to these vertebree articulat- ing with the facets on the sides of the bodies only. In each of the last four thoracic vertebree, each transverse process is replaced by a metapophysis on the anterior part of the vertebra, usually indicated by a tubercle on the eighth thoracic, and steadily increasing in size through the twelfth thoracic and still further increasing in size on the lumbar vertebrze. On the side of each of the four above vertebrze posteriorly, there is found another process, smaller than the above, the anapophysis. It is first seen in the ninth thoracic, is usually largest on the tenth, and decreases in size on the eleventh and twelfth. The large transverse processes which are found on the lumbar series are represented on the eleventh and twelfth thoracic by low lateral ridges. The spinous or neural process of the first thoracic vertebra is very short and broad, inclined slightly backward. The neural spines of the succeeding eight vertebre are all broad at the base, but spine- like at the free extremity, all are strongly directed backward. The spine of the ninth thoracic is broader than the spines of the vertebre in front of it. The spine of the tenth thoracic vertebra is still broader from before backward, and usually the spine of this vertebra is inclined neither forward nor backward, that is, as the rule, it is the anticlinal vertebra. The spines of the eleventh and twelfth are still broader, and like the spinous processes of the lumbar verte- bre are directed forward. The first vertebra presents a whole facet at the anterior edge of the centrum for the head of the first rib, a half facet at the posterior edge of the centrum for half of the head of the second rib. From the second to the eighth thoracic vertebrz, inclusive, there is a half facet at the anterior edge and another half facet at the posterior edge of the centrum. The ninth thoracic vertebra has only a half facet at the anterior edge of the centrum. The tenth, eleventh, and twelfth have a whole facet each on the centrum anteriorly. The transverse process of the first thoracic vertebra has a small concave facet for articulating with the tubercle of the rib. All the other vertebrze to and including the ninth have similar facets which increase in size up to the fifth and then decrease in size again. The tenth, eleventh, and twelfth thoracic vertebre have no transverse processes. Variations in the thoracic vertebre are fewer than they are in the cervical vertebrze of the Leporide. Some of the variations do not SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 45, PL. XCIII LUMBAR VERTEBRA OF HARES AND RABBITS (natural size). For explanation see page 446. ‘ob oSed 90s uonvuridxa 107 “SYMId ANY ‘SLIGAVU ‘SHAVH JO WUIALUAA UVAWNAT ANV SATA & AIDX “Id ‘SP “10 A SNOILOATION sac TIHOSIY NVINOSHLING LYON] THE HARES AND THEIR ALLIES 359 seem to be more than individual, and the miost that can be said is that they are tendencies rather than fixed characters. The relative length of the neural spines in. the anterior part of the thoracic series of vertebrz varies fairly constantly as follows: In Romerolagus, the length of an anterior neural spine is about twice the length of the centrum of the vertebra to which it is at- tached; in Sylvilagus, Brachylagus, Limnolagus, Lepus (Pecilo- lagus), and Pronolagus, the spine equals about two and a half times the centrum; in true Lepus it is about three times the length of the centrum, and in Macrotolagus and Oryctolagus it is a trifle over three times. The position of the anticlinal vertebra varies between the tenth and eleventh thoracic as follows: It is the tenth in Brachylagus, -Limnolagus, and Romerolagus; it is usually the eleventh in Lepus, Sylvilagus, Oryctolagus, and Pronolagus, but in some cases it is the tenth. Metapophyses are always found well developed on the last three thoracic, but in Pronolagus, Limnolagus, and most of the members of Lepus, a well-developed metapophysis is also found on the ninth thoracic; this also happens frequently in the case of the genus Sylvilagus.. But even those cases when the metapophysis is well developed on the tenth and eleventh thoracic only, it is always in- dicated by an ill-defined tubercle on the ninth. Lumbar Vertebre (pls. xcit1, xciv).—The seven lumbar vertebre are large and elongated. Each is provided with a stout, broad, spinous process, much shorter than the spines of the thoracic ver- tebrze. The metapophysis is well developed in all the lumbar verte- bree, and in the lumbar series nearly equals the spinous process in size. Like the spinous processes, the metapophyses are directed forward. On the anterior five lumbar vertebre the anapophysis is represented by a horizontal line ending posteriorly, usually, in a small projecting spine. In some cases this little spine is practically absent, while in others it is very well developed. The transverse processes of the lumbar vertebre are very large and long, projecting downward and forward. The proximal end arises from the anterior third or half of the side of the centrum. The free extremity is usually enlarged, and in the more anterior vertebrze of the lumbar series it is usually bifurcated. The total length of a transverse process in the middle of the lumbar series is usually equal to one and one-half times the length of the centrum to which it is attached. 360 SMITHSONIAN MISCELLANEOUS COLLECTIONS [VOL. 45 Each of the second and third lumbar vertebrz has a prominent ventral spine, the hypophysis. A smaller hypophysis is found in the first lumbar, and on the twelfth thoracic and the fourth lumbar vertebrae the hypophysis is usually indicated by a low ventral ridge on the centrum. The zygopophyses are prominent in the lumbar vertebre, and the articulating surfaces are directed laterally instead of horizontally as is the case with the cervical and thoracic series of vertebre. The last lumbar vertebra is peculiar in being shorter than any of the others and in having shorter and more slender transverse processes. The lumbar vertebrze of the Leporidz possess several well-marked variations, constant for certain groups and making good characters by which to determine them. The most important variations occur in the length, shape, and attachment of the transverse processes, and they may be outlined as follows: 1. Large rabbits having wide and long transverse processes with the free extremity expanded. The length of the longest process equals the length of the centrum to which it is attached and half the length of the centrum in front. The attached portion of the transverse process rises abruptly from the anterior half of the side of the centrum. All the members of the genus Lepus have the transverse processes of the lumbar vertebre of this form. 2. Medium sized rabbits having the lumbar transverse processes of the same relative length and width as in the above group. In- stead, however, of arising abruptly from the anterior half of the lateral aspect of the centrum, each process has a rather long posterior root coming from nearly the whole of the posterior half of the centrum, sharply sloping into the process itself. The skeletons having such transverse processes on the lumbar vertebrze belong to the genera Sylvilagus and Oryctolagus. 3. Medium-sized rabbits with the transverse processes slightly shorter than they are in the two preceding groups, the longest process equaling the length of the centrum to which it is attached and a fourth of the centrum in front. The processes are much broader than they are in the former groups, so that they appear much shorter than they really are. The free extremities are more expanded than they are in any of the genera except Romerolagus. The attached bases are very wide, coming from the whole side of the centrum, and the angle between the main axis of the transverse process and the side of the centrum is filled in with thin bone especially marked in the anterior part of the lumbar series, approach- LYON | THE HARES AND THEIR ALLIES 361 ing the condition found in Romerolagus. These processes are char- acteristic of the genus Limnolagus. 4. Small rabbits having the lumbar transverse process slightly more concave anteriorly than in the other rabbits. Each process is marked by a prominent longitudinal ridge. But for this pro- nounced ridge the lumbar transverse process resembles that found in the genus Sylvilagus. While this ridge is found to a greater or less extent in the other groups of the Leporidz, yet it is never so narrow nor so sharply marked off from the rest of the process. The transverse processes of this type are peculiar to the genus Brachylagus. 5. Small rabbits with the lumbar transverse processes short, the longest equaling the length of the centrum to which it is attached. The process of the first lumbar vertebra is very short and almost rudimentary. All the processes are wide and have triangular out- lines in general. The base is broad, coming from the whole side of the centrum, and the angle between the main axis of the process and the side of the centrum is completely filled with thin bone. It is an exaggeration of the condition found in Limnolagus. Even the transverse process of the last lumbar vertebra, which is usually slender in other rabbits, is here very broad, but not so broad as the transverse processes on the other lumbar vertebre. Transverse processes of this type are characteristic of the genus Romerolagus. 6. Medium-sized rabbits with lumbar transverse processes of medium length, the longest equaling the length of the centrum to which it is attached. The process is not so much expanded at the free extremity as in the case of the above groups. It is wide how- ever at the base, where it comes from the whole side of the centrum, resembling in this respect Limnolagus, but the posterior border of the transverse process is not so strongly concave as in that genus, and the process itself is more slender. Transverse processes of this type are found in the genus Pronolagus. The other variations in the lumbar vertebrz are of less importance and not so well defined as are the variations of the transverse processes. The spinous processes and metapophyses are always well de- veloped. In the anterior part of the lumbar region the spinous processes are usually a little longer than the metapophyses on the same vertebra and they are always longer in the posterior region. The spinous process is of variable shape and there are marked differences in individuals of the same species. This process is, in general, bluntly triangular, sloping obliquely anteriorly, the basal 362 SMITHSONIAN MISCELLANEOUS COLLECTIONS [VOL. 45 side attached to the rest of the vertebra, with a long sloping or usually concave posterior edge, and a more nearly vertical and shorter anterior edge. In the following hares the spinous processes are perfectly tri- angular, and are relatively not so high as in the typical form: Romerolagus, Limnolagus, Pronolagus, Pecilolagus. One skeleton of the genus Sylvilagus (No. 94,197, from Monitor valley, Nevada) also has neural spines of this shape, although in the other members of this genus which I have noted they are of the typical form. The anapophyses are very slightly developed on the lumbar vertebre of all hares. Their presence is usually indicated by a mere ridge on the side of the vertebrae, which ends in a small caudad projecting tubercle. They are least developed in the skeleton of Pronolagus and are longest in the skeleton of a lop-eared domestic rabbit, Oryctolagus cuniculus. Here they are very large, and in the three middle vertebre of the lumbar series, viz., third, fourth, and fifth, the anapophysis extends as far poster- iorly as the posterior border of the metapophysis of the next suc- ceeding vertebra. In the wild Oryctolagus, however, this great development of the anapophysis is not so pronounced, but it is much larger than on the lumbar vertebrz of any of the other skeletons. Ventral spines, or hypophyses, or ridges indicating them, are found on the first three lumbar vertebrae. The spines usually occur in three lengths: In the genera Sylvilagus and Brachylagus the hypophysis on the second lumbar is the longest, that on the first lumbar is next in length, and the shortest hypophysis 1s found on the third. Oc- casionally in Sylvilagus the first hypophysis is reduced to a ridge and the last is sometimes lacking. In all the skeletons of the genus Lepus the third lumbar vertebra bears the longest hypophysis; the first bears the shortest. In Romerolagus the first hypophysis is the shortest, the second and third are subequal, the third being a trifle the larger. In the skeletons of Limmnolagus, Pronolagus, and Oryctolagus the hypophyses are more or less in- jured. It would appear, however, in these genera that the second hypophysis is the longest. Sacral V ertebre.—The number of sacral vertebrz in the Leporidze varies from three to five, according to the age of the individual, four being the usual number. The two anterior vertebre of the sacrum are the only ones entering into the formation of the sacroiliac joint. The remaining vertebre are progressively smaller and resemble in shape the anterior caudal vertebre. The first sacral has a large neural spine, vertical or inclined LYON] THE HARES AND THEIR ALLIES 363 slightly forward like the spines of the lumbar vertebre. The neural spines of the remaining sacral vertebree become progressively smaller from before backward and they are directed caudad. In some cases the spine of the fourth sacral is very small. In nearly all cases the spines of the sacral vertebre are distinct and not fused with one another. The outline of the sacrum as a whole is triangular, the base being in front, the apex behind. The greatest width, which is at the anterior part of the massive fused transverse processes of the first and second sacral vertebra, is nearly equal to the greatest antero- posterior diameter. , In all the rabbits the sacrum presents very few variations. The differences found in the different sacra seem to be due entirely to age and individual variations. The only sacrum showing any marked deviation from the type is in Lepus sp. No. 49,621, from Jumna river, India. The posterior part of this sacrum is very narrow from side to side. The expanded wing-like portions to which the ilia are attached are very narrow from before back- ward. In general, the shape of this peculiar sacrum, as seen from below, is like a T. The adjacent tips of the first and second, and of the third and fourth of its sacral neural processes are in contact. The usual number of vertebrz in the sacrum is four; in some old individuals it is raised to five, and in some younger ones it is only three. In what follows with reference to the caudal vertebrz, four vertebrze will be considered as entering into the formation of the sacrum. Caudal vertebre.—The caudal series always includes vertebre of three different though not sharply defined forms. 1. The first one or two vertebre following the sacrum are long, and in general appearance resemble the last sacral vertebra. 2. The next three to seven cadual vertebre are shortened, have wing-like anteriorly directed transverse processes and the neural arches become progressively less developed caudad. 3. The remaining vertebre of the caudal series, four to nine in number, are merely small elongated centra without processes or neural arches. The accompanying table (page 364) shows the number of the dif- ferent forms throughout the caudal series in the available skeletons. From this table the following groups may be picked out: 1. Nesolagus, with a total of eight caudal vertebre. 2. Romerolagus, with a total of nine caudals, of which one is of the first form, five of the second, and three of the third. 364 SMITHSONIAN MISCELLANEOUS COLLECTIONS [voL. 45 | Piel gs | | si) 6. | a a | |e |e | & g oa) eo} ~~ so GENERA AND SPECIES. | Locatity. Numser. | 2] 8 | o>] & | WSS S| ia | | o | wlal2| & | Deep es LCBUS) LILES omnpeentsecea ase Sweden 15048" | Dil hye a3 sé os Be Meee OSCE ERR lop ies TyO40T TO 7 Sets Ge ut COMLPCSETUS es esa eats Kansas. Ageo2atai@ any }6 | 14 ee ss coy) We cacao een Nebraska. AQSO2 215) Steer ROM mare oe CUO LED nc adnaadacot | Siberia. ui eon? Nl alos) | | tgs ‘© (Macrotolagus) textanus....... Nevada. 94,1983 | 1 | 7 | 7 | 35 “ ( Pectlolagus) americanus | | UP SULLCILU SERS Recent eee New York. | 960) eye * FO USP ieacdisenenteide comes twersanee eae: India: Jumna river.) 49,621! | 1 | 6 | 7 | 14° Sylulagus fiortdanus., acatecsseseieees Florida. AQ-5SOUs ely O) e7ee\ ra! - Siete ei h Rca nias nee ensn oe AQS5 O77 ap Onnl ase uf So EE ea eee tenes Pace 49,588! | 1 | 5*| * * oe SOO messes cb asensoccee | ce WA Ges SoU Na tOnn 7a mts ut a MANUTUS 6.04. Virginia. 89,5143 | 1 | 6 | 6*| 13% us a transitionalts... New York. | 419; 6240) 25 fOr srs ae a subspecies........ Nevada. 94,8979 | 0) 164 | 5* m3 OS | MULL QUOD. ceca ct aneoas soak Wyoming. | 1,044) |) 16 ton \s13 SS SYLHULET Ar dee cane eee tee tn eee ee Mexico. | 92,9652 | SEUNG: neh Es SSp. PULEM CIS Sy eaciaee oes pE eRe s | Brazil; Chapada. | 113,432' | 2|5 |4 | II us COMET Sectscuoeee scat nents os ae | 49,665! | 2 | 5 }4 | 31 Brachylagus tAahoensts ...cecccvececeeee Idaho. | 93,6952 | 2/3 14 | > c Soa ohana hae oeea as | -93,5962)| 2.13) 451249 Nesolagus netschert 4... ..ccccccoecense .. Sumatra. | ba aes 8 TS OPMENOU URES ILCLSOP1 dam waeeeee ne eee ees | Mexico. Hes i7 O54 oan noes 9 Limmolagus Paludicola.s.ccccrcersecees | Florida. | 49,580" | 1/5 |5 | i Oryctolagus cuniculius, lop ear ......+. domestic. \" AO;sSOF a a 7 F er pon z os ie Ba ain see Germany. | 40,6450) er | 6) | 6 \n6 a ee Ne een: | England. | 49,648! | x |8 | 8 | 17 HC “ Belgian hare.. domestic. 49,6770 27. 1.0 eye Pronolagus crasstcaudatus ....0000..., South Africa. | 22,9724 | <2) |e Pe a 3. Brachylagus, with a total of nine caudals, of which two are of the first form, three of the second, and four of the third. 4. Limnolagus, with a total of eleven, of which one is of the first form and five each of the second and third. The other genera containing a larger number of species than the above do not show such sharply defined results, but they roughly fall into these groups. 5. Sylvilagus, having a total of eleven to fifteen caudals, of which usually the first one or two are of the first form, the next five or seven, usually six, of the second form, and the terminal four or seven of the third form. *United States National Museum. * American Museum of Natural History. * United States National Museum, Collection of Biological Survey, United States Department of Agriculture. *No specimen seen, taken from original description. (Schlegel, ‘80, p. 64.) « Series incomplete. LYON | THE HARES AND THEIR ALLIES 365 6. Lepus, having a total of thirteen to fifteen caudals, of which the first only is of the first form, the next five to seven, usually seven, of the second form, and the remaining five to seven of the third form. 7. Oryctolagus, with a total of sixteen to seventeen caudals, of which the first is of the first form, the next six or eight of the second, and about the remaining eight or nine of the third form. OcCHOTONIDA The vertebral column of the Ochotonide presents nearly as many and as marked differences from that of the Leporide as do the skulls of the former from those of the latter. Cervical Vertebre (pl. xctt, 1).—The cervical vertebre of the Ochotonide have the same general characteristics as in the Leporide. They are decidedly shortened antero-posteriorly, the laminz of the posterior ones being very narrow. This shortening in- volves the axis but not the atlas. The latter has the free extremity of the transverse processes moderately expanded. The costo-trans- verse processes of the third, fourth, and fifth cervicals are placed more obliquely to the axis of the vertebral column than the same processes are in the Leporidze. In the sixth they become horizontal as they do in the hares. The transverse process of the seventh cervical differs from that in the Leporide in not being pierced by a costo-transverse or vertebral foramen. Thoracic Vertebre.—The thoracic vertebree of the Ochotonidz are entirely different from the same series of vertebrz in the Leporide, they are 17 in number instead of 12, the thoraco-lumbar vertebrze in the two groups being 22 and 19 respectively. The first 12 of the thoracic vertebre in the Ochotonide are exactly homologous with the 12 thoracic vertebrz of the Leporide. The arrangement of the facets for the heads and the tubercles of the ribs is entirely similar. The five remaining rib-bearing vertebrze of the Ochotonidze are practically indistinguishable from one another as well as from the twelfth, except by the slightly greater size of each succeeding vertebra. The spinous processes are relatively shorter in the Ochotonide, and this is especially true in the posterior thoracic region from the twelfth onward where the spines are all low and slightly inclined forward. Each neural spine of these posterior thoracic vertebre arises by a broad base from the whole length of the neural arch; the free extremity of the process is nearly as long as the base, the posterior edge being slightly concave. 366 SMITHSONIAN MISCELLANEOUS COLLECTIONS [VOL. 45 That part of the transverse process of the thoracic vertebra which articulates with the tubercle of the rib is the same in form in the two families. Associated with this transverse process in the Ochotonide are the metapophysis and the anapophysis. Both these processes are first seen on the third thoracic as mere tubercles. The anapophysis grows larger on each successive vertebra, attaining its greatest size on the last thoracic. From the eleventh thoracic onward, the anapophysis is a well-marked process directed upward, backward, and outward. The metapophysis remains little mor: than a tubercle until the tenth thoracic vertebra is reached, where it is a well-marked process. On the eleventh it is slightly larger, and on the twelfth still larger but closely associated with the prezygo- pophysis. The metapophysis scarcely increases in size through the rest of the series and continues closely asociated with the pre- zygopophysis throughout. No ventral spines or hypophyses are found on any of the vertebree. Some of the posterior thoracic have a slight ventral ridge which is also found on all the lumbar vertebre. Lumbar Vertebre (pl. xctv, 10).—Corresponding to the greater number of the thoracic vertebre in the Ochotonide, there is a diminution in the number of the lumbar vertebree, from 7 to 5, but this is not sufficient to make the number of thoraco-lumbar ver- tebrze the same in the two families. The lumbar vertebrz of Ocho- tona are very different in form as well as in number from those of Lepus and its allies. Each vertebra is compact, with the processes broader and more closely applied to the body instead of the slenderer processes extending to a distance from the centrum such as occurs in the Leporide. The neural process is low, with the free edge as long as the whole length of the vertebra and parallel with the axis of the vertebra. The metapophyses are well developed and more closely associated with the prezygopophyses than they are in the Leporide. Anapophyses are well developed on the first and second lumbar vertebra and are a direct continuation of the thoracic series of anapophyses. These processes are slightly indicated on the third lumbar vertebra, after which point they disappear. The transverse process is little more than a tubercle on the first and second lumbar vertebra, but on the third and fourth it is a wide quadrilateral plate of bone coming from the whole side of the vertebra, sloping down- ward and outward. The transverse process of the fifth and last lumbar is a trifle longer than the other transverse processes and only about half as wide, the narrowing taking place chiefly at the expense of the posterior half of the process. There are no hypophyses, but all the lumbar vertebre (as is the case of those lumbar vertebrz of SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 45, PL XCV STERNA OF HARES AND RABBITS (natural size). For explanation see page 446. LYON | THE HARES AND THEIR ALLIES 367 the Leporide which do not bear ventral spines) possess a median ventral ridge. Sacral V ertebre.—The sacrum in the pikas is entirely different in form from the corresponding structure in the hares and rabbits. It is long and narrow, its greatest breadth being contained in its length about twice. The lateral masses that are attached to the ilia, instead of being expanded into wing-like processes with the greatest width anteriorly, are much less expanded and have nearly parallel sides. The neural spines so distinct and conspicuous on the sacra of the Leporidz are reduced in the Ochotonidz to form a low dorsal ridge, the separate spines having fused with one another. The number of vertebre entering into the formation of the sacrum of the Ocho- tonidz is four, the same as in the case of the Leporide. Caudal Vertebre—The caudal vertebre in the Ochotonide are eight in number in all the skeletons at hand except one, which has nine. In three American specimens, Nos. 91,188 and 30,990 from Idaho, and No. 49,620 from Oregon, the first caudal is somewhat narrowed, the next two are slightly wider, with faint indications of lateral projections; the rest of the series consists of short flattened bodies. The single Asiatic skeleton, No. 49,500, Ochotona lada- censis, central Asia, has mainly the same character of the caudals, but the individual vertebrz are relatively wider’ throughout. STERNUM LEPORID.A Garimo Gears); «© Vil 1,. 3-5) The sterna of the Leporide are formed of the usual three portions, presternum or manubrium, mesosternum or gladiolus, and xiphister- num. The presternum consists of one piece which is usually longer than any other single segment of the sternum excepting the xiphisternum. It is usually compressed from side to side and marked by a more or less evident ventral keel. At or anterior to its middle the first pair of sternal ribs is attached. The mesosternum -consists of four usually distinct segments. A sternal rib is attached at the point of articulation of each segment with the other, as well as at the point of articulation of the prester- num with the mesosternum. At the posterior outer aspect of the last mesosternal segment near the articulation of the xiphisternum two sternal ribs are usually attached. 368 SMITHSONIAN MISCELLANEOUS COLLECTIONS [VOL. 45 The xiphisternum is usually the longest single segment. It is attached anteriorly to the last mesosternal segment. Its posterior end is free and is terminated by a thin rounded piece of cartilage. Eight different types of sternum are found among the Leporide; some of which are characteristic of certain genera, while the others are not so sharply defined and might better be considered tendencies rather than actual developments. A greater number of ,skeletons may show that some of the types amount to nothing, as there seems to be a certain amount of variation in individuals of the same species. The different types of sterna are as follows: I. Presternum long and narrow, much compressed laterally into a keel which is most prominent anteriorly; the first rib is attached at the junction of the first and second fourths or first and second thirds. Mesosternum of four distinct segments, of which the first is narrow and compressed, the others not compressed or even flat- tened dorso-ventrally. All the mesosternal segments in general are subequal and each successive one grows wider from before back- ward. The xiphisternum about equals the presternum in length. The anterior end is considerably enlarged where it articulates with the last mesosternal segment. Its posterior and free extremity is rather pointed or only slightly larger than the narrowest portion of the xiphisternum. .One skeleton (Lepus texianus, No. 94,198), be- longing in this section, has an anomalous mesosternum composed of five distinct segments, the last of which is only slightly smaller than the other segments. At the articulation of the mesosternum with the xiphisternum, but one pair of ribs is attached instead of the usual two pairs in sterna where the fifth mesosternal segment is normally suppressed. The hares possessing sterna like the above are all large and belong to the genus Lepiis. (Piexev, 3) 4.) 2. Sternum in general very similar to the above, presternum re- latively longer, its keel less prominent anteriorly. The first pair of ribs is attached just anterior to its middle. The mesosternal seg- ments have a tendency to be less flattened. Xiphisternum shorter than the presternum, its posterior end more enlarged than it is in the case of the preceding section. Such sterna are more or less charac- teristic of the genus Sy/uilaeus. “(Ely xev, 25} 3. Sterna similar to those of the genus Lepus; presternum more conspicuously keeled, the first rib attached just anterior to its middle. Mesosternal segments more compressed from side to side than they are in Lepus. The last is much shorter than any of the other meso- sternal segments. Xiphisternum large and stout, longer than the SMITHSONIAN MISCELLANEOUS COLLECTIONS VoL. 45, PL. XCVI STERNA OF RABBITS AND PIKA. For explanation see page 446. LYON | THE HARES AND THEIR ALLIES 369 presternum, its posterior extremity about as much expanded as its anterior. These sterna are found in the genus Oryctolagus. (PI. xcv, 5.) One skeleton of Oryctolagus (No. 49,648) shows the peculiarity of five segments in the mesosternum, the fifth being very small and short. As in the case of the abnormal sternum of Lepus terianus, No. 94,198, also with five mesosternal segments, only one pair of ribs is attached to the junction of the fifth mesosternal seg- ment with the xiphisternum. 4. Presternum compressed with low keel all along ventral border, dorsal portion somewhat expanded just anterior to the attachment of the first pair of ribs which takes place at the junction of the anterior and middle thirds. Mesosternum of four distinct segments, the first compressed laterally with a low keel continuing that of the presternum, the succeeding mesosternal segments becoming suc- cessively wider and more dorso-ventrally flattened. The last seg- ment is much wider than any of the others, with a well-marked postero-lateral angle for the attachment of the last two sternal ribs. The xiphisternum is decidedly longer than the presternum; it is slender in the middle, but the ends are considerably expanded, espe- cially the posterior end. This type of sternum is found in the genus Brachylagus. (Pl. xcv, 1.) 5. Posterior two thirds of the presternum much compressed, but not keeled. The anterior third is somewhat expanded, resembling that portion of the presternum of Limnolagus. The first pair of ribs is attached at the junction of the first and second fourths. The mesosternal segments are more compressed laterally than they are in the other skeletons and each succeeding one is a little shorter than the one immediately in front of it. The xiphisternum is a trifle shorter than the presternum, is comparatively short, and about equally widened at each end. This type of sternum is found in the single skeleton of Pronolagus. (Pl. xcv1, 3.) 6. The anterior portion of the presternum is considerably ex- panded laterally, having a tendency to be intermediate in form be- tween the presterna of Lepus and those of Romerolagus and Ocho- tona. The mid-ventral line of this expanded part bears a low keel which is not extended backward on the posterior portion of the presternum. The first pair of ribs is attached to the middle of the presternum. The mesosternttm as a whole is wider than are the mesosterna previously mentioned. It is also shorter, so that its length is but little greater than that of the presternum or of the xiphisternum, both of which are about equal in length. The first, sec- ond, and third mesosternal segments are about equal in length; each 379 SMITHSONIAN MISCELLANEOUS COLLECTIONS [VOL. 45 succeeding segment is wider than the one in front. The third and fourth mesosternal segments are completely ankylosed so that the whole mesosternum is composed of but three separate pieces instead of the usual four. The xiphisternum is long and slender and about equally expanded at each end. In all the preceding sections the sixth and seventh pairs of ribs are attached to the last piece of the meso- sternum in the angle between it and the xiphisternum. In case of the form of sternum just described the sixth rib is the last attached directly to the sternum, the seventh being attached to the cartilage of the sixth near the point where the latter joins the mesosternum. This type of sternum is found in the skeleton of Limnolagus paludi- Colds (Cel excyiese) 7. In this section the sternum is very characteristic and resembles almost exactly that of Ochotona. The anterior portion of the pre- sternum is very much expanded and flattened dorso-ventrally. To the outer posterior angles of this expanded portion the first pair of ribs is attached. The rest of the presternum is long and narrow, as it is throughout the Leporide, but devoid of any ventral keel. A slight ridge indicating a keel is seen on the ventral face of the expanded portion. The mesosternum in general is very much like the same structure in Limnolagus just described. The first and second segments, however, are subequal and relatively narrower, the third and fourth segments are subequal in length, but the fourth is broader, both are relatively wider than they are in Limnolagus and as in that genus are firmly ankylosed. The xiphisternum is long and rather stout; it is about equally enlarged at either end. It is but a trifle shorter than the whole mesosternum and decidedly longer than the presternum. As in Limnolagus the seventh pair of ribs does not articulate with the sternum, but with the cartilages of the sixth pair. The above is the sternum of “Romerolagus. (PI KEV) 8. In this group the presternum is considerably enlarged in its anterior third, to about the same extent that it is in Romerolagus. Its different shape is best seen by consulting figures 1 and 4, plate xcv1. The mesosternum consists of four distinct segments, the first two subequal in length, compressed laterally; the third seg- ment is slightly shorter, not laterally compressed; the fourth segment is very short and cartilaginous. The xiphisternum is very short, much shorter than the presternum; its anterior end is considerably enlarged. Apparently six pairs of ribs only articulate directly with the sternum. This type of sternum is found in Nesolagus (Major, (OQ, pl. 20) His.uEs)s LYON | THE HARES AND THEIR ALLIES a7 OCHOTONIDE CErATEe NE Vsle)) The material for making generalizations concerning the sterna of the Ochotonidz is far from being satisfactory. Out of four skeletons but one is fully adult and there is a certain amount of variation among them. It may be that more material would show that there are two or three different types of sterna among the Ochotonide. Aside from a few minor details the sternum of the only adult Ochotona at hand, No. 91,188, is almost exactly like the sternum of Romerolagus just described, so that the genus Romerolagus can be briefly described as a rabbit with the sternum of a pika. The expanded portion of the manubrium is less developed in Ochotona than in Romerolagus, and rather triangular in outline instead of pentagonal. In other respects the two sterna are similar. The Ochotona presternum is nearly as long as the mesosternum and slightly longer than the xiphisternum. The mesosternum of Ochotona is in general very similar to that of Romerolagus. The first and second segments are subequal in length, the second being broader however. The third segment is the longest and broadest of the mesosternum. The fourth segment is the shortest and nearly as broad as the third. Both the third and fourth mesosternal segments are completely ankylosed as they are in Romerolagus and Limnolagus. The xiphisternum is considerably expanded at the proximal end, but the distal extremity is not much enlarged. It is decidedly shorter than the presternum. The seventh rib is attached along with the sixth rib to the sternum at the point of union of the mesosternum with the xiphisternum. No. 30,990 Ochotona savratilis, Idaho, is very similar to the above; it is young, however, and the third and fourth pieces of the mesoster- num are not yet fused. No. 49,620, from Oregon, has the entire mesosternum narrow and its last two segments separate. No. 49,500 Ochotona ladacensis, Ladak, central Asia, has the enlarged portion of the presternum less expanded; the mesosternum is relatively longer and decidedly narrower than is the mesosternum of O. saratilis, the third and fourth mesosternal segments are not fused. The mesosternum as a whole bears considerable resemblance to some of the mesosterna of Sylvilagus. The xiphisternum is rather short. 372 SMITHSONIAN MISCELLANEOUS COLLECTIONS [VOL. 45 RIBS LEPORIDA (Prats XCIV, 1-8) Corresponding to the number of thoracic vertebree there are twelve pairs of ribs of which seven are normally attached to the sternum by means of well developed, ossified, costal cartilages. The last three or four pairs of ribs have no ventral attachments. The ribs intervening between the sternal and floating ribs are attached by their costal cartilages to the costal cartilages of the sternal ribs. The typical rib has a head attached to the centra of two adjoining vertebrae, a well-marked tubercle a portion of which articulates with the transverse process of the more anterior vertebra and another portion of which projects superiorly, forming a spine. The most posterior ribs lack the tubercle. The first rib has the articular portion of the tubercle well developed but the spine-like part is lacking. Its place, however, is taken by a dorsally projecting spine from the transverse process of the first thoracic vertebra. The rest of the rib is made up of a long, thin, flattened shaft. In the genus Lepus the non-articular portion of the tubercles are very well developed, the eighth pair of ribs is the last pair bearing them. The second, third, fourth, fifth, and sixth ribs are very flat and broad in the ventral half of the shaft; the greatest width of the rib just behind the tubercle is very much less than the width of the shaft in the lower portion. The sternal costal cartilages in Lepus are very short and wide, as compared with the sternal costal cartilages of the other genera. In Oryctolagus conditions similar to those found in Lepus exist regarding the ribs, but the shafts of the anterior ribs are relatively less widened. The genus Sylvilagus closely resembles Lepus with respect to the form of the ribs, but like Oryctolagus the shafts of the ribs are less expanded and in some of the species, notably S. minensis, the non- articular portions of the tubercles are not so well developed and are last seen on the seventh pair of ribs. Brachylagus closely resembles Sylvilagus, especially S. minensis ; the non-articular portions of the tubercles are not very well developed and are last seen on the eighth pair of ribs. In Limnolagus the tubercles are well developed but are not con- spicuous, owing to the fact that the angle between the tubercle and the posterior edge of the rib is filled up with bone, making that part distinctly the widest portion. In the single skeleton at hand LYON | THE HARES AND THEIR ALLIES So the last rib to bear a spine-like tubercle on the right side is the seventh, while on the left side it is the sixth. The shafts of the anterior ribs are not widened ventrally. | The spines of the tubercles are very small in Pronolagus, and are last found in the seventh pair of ribs. The shafts of the ribs are relatively narrow and there is no indication of the wide expansion found in Lepus. Decidedly the widest part of the rib is just behind the spine. Romerolagus very closely resembles Pronolagus in regard to the ribs, but the spines of the tubercles are last found on the sixth pair. OcHOTONID @PrArE XIE 0)) In the Ochotonidz there are seventeen pairs of ribs of which the first seven are attached to the sternum by means of costal cartilages, the last seven pairs or about that number have no ventral attach- ments, while the three intermediate pairs are attached to the costal cartilages of the ribs in front. The ribs are all slender and weak compared with the ribs of the Leporidz and none of them possesses spine-like, non-articular portions of tubercles, but just behind the tubercles the more anterior ribs are quite broad and heavy. CEAVICLE LEPORID-E CRrAmE Sy NEVE 2s skier ) In the Leporidze the clavicle is probably always present as a small curved, slender bone, about fifteen millimeters long. In most of the skeletons, however, the clavicles are wanting, undoubtedly the result of faulty preparation. In only four of the skeletons at hand have these bones been found. The uncleaned skeletons show that the clavicle does not articulate with either the sternum or the scapula as it does in those animals where it is well developed. Ligamentary tissue extends from its inner end to the presternum, while the outer end is attached to a piece of cartilage placed on the summit of the greater tuberosity of the humerus. In his account of the genus Romerolagus (Proc. Biol. Soc. Washington, x, 1896, December 29, p. 171) Dr. Merriam says: “ The clavicle is complete and articulates directly with the sternum (fig. 33)—a thing that never happens in the genus Lepus.” Whether that is the case with any of the other genera of the Leporidz can not be told, owing to the faulty methods of cleaning the skeletons. It might be thought 374 SMITHSONIAN MISCELLANEOUS COLLECTIONS [VOL. 45 that Romerolagus with the enlarged presternum would have clavicles much larger than those in the other genera of the Leporide, but measurements show that the ratios of length of humerus to length of clavicle in Romerolagus is 3.20, in Oryctolagus 3.16, in Sylvi- lagus 4.00. OcHOTONIDE CREAT HO NG Wilpe2 In the Ochotonidz the clavicle is very well developed, the ratio of its length to that of the humerus is $. The outer end, enlarged and flattened, is connected by a ligament to the greater tuberosity of the humerus. The inner end articulates directly with the extreme anterior portion of the presternum. SGAPUE LEPORID CPLATE eC Vis aO.8) The general shape of the scapula of the Leporidz is roughly that of a right triangle, with the right angle rounded off, and directed ‘upward and forward. The acromion process is long and slender and about half the length of the actual scapular spine. Coming off at right angles to the acromion process and directed caudad is a well developed pointed metacromion. The scapule of the Leporide fall into three not very strongly marked groups: 1. Large rabbits, members of the genus Lepus in which the scapula is relatively broad, the superior border rather more convex, the antero-superior angle more rounded, and the supraspinous fossa relatively wider. 2. Rabbits averaging smaller in size, members of the genera Sylvilagus, Oryctolagus, Brachylagus, and Limnolagus, in which the scapula is relatively narrower, the superior border straighter and less convex, the antero-superior angle more pronounced and not so gradually rounded off as in the genus Lepus. The supraspinous fossa is relatively narrower. 3. Romerolagus and Pronolagus have scapula much alike and differing from the scapulz of the other genera in being longer and narrower. ‘The posterior border is more nearly straight instead of being concave as in the other two groups. The superior border is straighter. The distance between the antero-superior and the postero-superior angles is contained twice in the length of the scapula measured along the inner surface at the attachment of the ‘Leb “ohh saded aas uonyeueldxa og -(azis [eanjeu) SYHMId ANY LIGIVUa ‘SAUVH AO WINdv SNOILOATIOD JANVTIAOSIPT NVINOSHLINS LYON | THE HARES AND THEIR ALLIES S75 spine, from the superior border to the edge of the glenoid cavity. In the other rabbits the distance between the two superior angles is contained but one and one-half times in the scapular length. OCHOTONID CErATErK CV I 5) The scapula of the Ochotonidz is of a quite different type from that of the Leporide. The general outline of the bone is also that of a right triangle, but the right angle is very much rounded off and the general appearance of the scapula is more obliques “he acromion process is very long and slender, about twice the length of the actual scapular spine. The metacromion is well developed, and has about the same general proportion that it has in the Leporide. The posterior border of the scapula is long and more concave than in the Leporide ; the superior border is relatively longer and very much rounded off, so that it gradually merges into the anterior border. The distance between the antero-superior and the postero- superior angles is relatively greater in the pikas than it is in the hares and rabbits, being contained but little more than once in the length of the scapula taken along the attachment of the spine. The supraspinous fossa in the Ochotonidz is relatively much narrower, when compared with the infraspinous fossa, than it is in the Leporide. The scapula of Ochotona ladacensis differs somewhat from that of the American species. The superior border is shorter and does not merge so gradually into the anterior border, so that the antero- superior angle is more pronounced. HUMERUS LEPORIDE The humeri of the Leporide are all much alike in form and pro- portions, and the humeri of the Ochotonide differ but slightly from them in these respects. The variations in this bone in the different genera are few, hard to define, and apparently of little significance. They are as follows: The groove that subtends the internal condyle is best marked in members of the genus Lepus. It is well marked also in Sylvilagis, Oryctolagus, and Pronolagus, but in Limnolagus and Romerolagus is much less developed and in Pentalagus it is very slight. Brachy- lagus occupies an intermediate position between these last two de- grees of development. In Pentalagus the double trochlear surface at the distal extremity of the humerus differs from the same structure 379 SMITHSONIAN MISCELLANEOUS COLLECTIONS [VOL. 45 in the other genera in having its two portions very unequally de- veloped, the main one being very broad and shallow, and the external portion much reduced. Brachylagus and Romerolagus have the slenderest humeri; Orycto- lagus and especially Pentalagus have rather thick, heavy ones. All the other groups may be said to have moderately developed humeri. The external condyloid ridge is very short and poorly developed in all the hares. Romerolagus has the most prominent external condyloid ridge; Brachylagus has a ridge as wide but not so long; Limnolagus has a short and comparatively wide ridge, but only a trifle more conspicuous than the ridge in Sy/vilagus or in the other genera. OCHOTONIDE The humeri of the Ochotonidz have a strong resemblance to those of the Leporide. The head of the bone is more globular in the _ former, the bicipital groove does not encroach on its anterior surface. The tuberosities of the humeri of Ochotona bear about the same general proportions to one another that they do in the hares and rabbits. The greater tuberosity does not project so far upward in Ochotona as it does in the Leporide, so that the head of the humerus is the highest point of the bone, while in the hares and rabbits the greater tuberosity is the highest point. When viewed from the side, the head of the humerus in the pikas is seen to project farther backward than in the hares, and to form a sort of hook with the shaft of the bone. The double trachlear surface at the distal end of the humerus is shallow and wider in Ochotona than in any of the Leporide. The groove subtending the internal condyle is rather shallow and inconspicuous in Ochotona; it is developed to about the same degree that it is in Romerolagus. RADIUS AND ULNA LEPORIDA G2eArE x@\ViliieeE— 10) The bones of the forearm of the Leporide show some interesting differences among the various genera, in regard to their relative development and the length of the radius compared with the length of the humerus. The radius and ulna are always perfectly distinct, but they are in contact with one another throughout the greater part of their extent. At the point of contact there is a certain amount of fusion, but at the same time each bone maintains its individual distinctness. SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 45, PL. XCVIII BONES OF THE FOREARM OF HARES, RABBITS, AND PIKAS (natura! size) For explanation see page 447. LYON ] THE HARES AND THEIR ALLIES Sie. In the upper part of the forearm, the radius lies in front of the ulna, while at the distal extremity the radius is found toward the inner side. The radius in general is flattened antero-posteriorly, presenting a rather flattened posterior surface and a convex anterior one. The upper extremity of the radius forms the inferior portion of the sig- moid notch for articulating with the double trachlear surface of the lower end of the humerus. The lower extremity of the radius is rather enlarged and its distal surface is concave, the concavity in- clining to be double for articulating with the scaphoid and semilunar bones of the wrist. The ulna in general lies behind the radius and toward the outer side. In some cases the shaft of the ulna is more heavily built than the shaft of the radius, while in other cases it is very much reduced. The outer part of the lower end of the ulna is prolonged downward into a convex articular surface which fits into a corresponding con- cavity formed jointly by the cuneiform and pisiform bones. This projecting part of the ulna is the only part of that bone which articulates with the carpus in the case of the Leporidz. In the case of the Ochotonidz, however, there is a concave facet internal to this projection which articulates with a corresponding convexity on the cuneiform. A condition approaching this is found in Pentalagus, Romerolagus, Brachylagus, and in some skeletons of Sylvilagus and Lepus. In the genus Lepus the ulna is much reduced in size along the middle of the shaft, and except at the lower extremity it is placed almost entirely behind the radius. The radius itself is rather long and slender. The humerus and radius are usually subequal in length. In the subgenus Macrotolagus, at least so far as the limited material shows, the radius is slenderer than it is in the two other subgenera, Lepus and Pecilolagus, and it is decidedly longer than the humerus. In the genera Sylvilagus, Pronolagus,-and Limmolagus the radius and ulna are subequal in size, the ulna not being reduced in the middle part of the shaft. The ulna is situated external to the radius rather than behind it as in Lepus. Both radius and ulna as a rule are moderately slender. The radius equals the humerus in length. The condition of the bones of the forearm in Brachylagus is very similar to that just described, but the humerus is distinctly longer than the radius. 378 SMITHSONIAN MISCELLANEOUS COLLECTIONS [VoL. 45 In Pentalagus, in Romerolagus, and to a less extent in Oryctolagus the radius is slenderer than the ulna, except at the articular ends of the radius, where normal conditions exist. The humerus is distinctly longer than the radius, except in Oryctolagus where their lengths are subequal. OCHOTONIDE GREATE eX © Vililiestt) In the Ochotonidz the bones of the forearm resemble the ulna and radius of the Leporidz in general form and position. The ulna is distinctly the larger of the two bones of the forearm throughout its whole extent. As noted in the description of that bone under the Leporidz, the ulna has an extra facet at its distal surface for articu- lating with the carpus. The proximal articulating extremity of the radius maintains about the same relative size that it does in the Leporide, but the distal end is conspicuously reduced and the lower end of the ulna is correspondingly increased. The radius is dis- tinctively shorter than the humerus. CARP US LEPORID (FIGURE 45, I-3) The full number of carpal bones (nine) is found in the Leporide. An understanding of their form and position is best obtained from consulting the figures. In all the skeletons the bones have the same relative shapes, sizes, and positions. The pisiform is one of the largest carpal bones and has consider- able dorso-palmar depth, except in Pentalagus where this depth is less and has about the same relative proportions that it has in Ochotona. In the cuneiform, together with a small portion of the pisiform, is a cup-shaped depression into which fits the small rounded pro- jecting head of the ulna. The internal and median portion of the proximal aspect of the cuneiform in most hares is scarcely at all developed, but in some of the skeletons (Pentalagus, Romerolagus, Pronolagus, Brachylagus, and rarely in Sylvilagus and Lepus) there is found internal to the cup-shaped cavity for the ulna, a small convex surface which articulates with the ulna internal to the projecting convexity. The os centrale is a flask-shaped bone, only the neck of which appears in an undissected carpus. The distal aspect of the neck- like process articulates with the outer portion of the proximal end of the second metacarpal. LYON | THE HARES AND THEIR ALLIES 379 In some of the skeletons (and probably in all had they been care- fully cleaned), in the angle between the unciform and the fifth meta- Fic. 45.—Carpus of the Leporide and Ochotonide. 1. Dorsal view of right carpus of Sylvilagus foridanus mallurus, No. 16,248, Carlisle, Pennsylvania, enlarged nearly twice. 2. Proximal view of proximal row of carpal bones of same. 3. Proximal view of distal row of carpal bones of same. 4. Dorsal view of right carpus of Ochotona saxatilis, No. 49,620, Oregon, enlarged nearly four times. 5. Proximal view of proximal row of carpal bones of same. 6. Proximal view of distal row of carpal bones of same. ac, accessory ossicle; c, os centrale; cu, cuneiform ; mc,, first metacarpal; mc;, fifth metacarpal; mg, os magnum; ps, pisi- form; sc, scaphoid; sm, semilunar; ¢t, trapezium; td, trapezoid; uw, unciform. carpal, is a small bone which Forsyth Major calls the carpale 5. He regards the unciform as carpale 4 only, instead of the fused carpalia 4and 5. This small bone is otherwise known as the os vesalianum. OCHOTONID.E (FIGURE 45, 4-6) The carpus of the Ochotonidz differs in several respects from the carpus of the Leporide. The pisiform has less dorso-palmar depth ; the lunare is narrower; the internal half of the cuneiform is more largely developed, making up for the narrower lunare, and presents a well-developed convex facet for articulating with the ulna, in addi- tion to the cup-shaped cavity formed by the cuneiform and pisiform jointly. The centrale is much larger than it is in the Leporidz, and is not flask-shaped. The os magnum is reduced in size. A small os vesalianum is present. 350 SMITHSONIAN MISCELLANEOUS COLLECTIONS [VoL. 45 METAGCARE US There are few points of interest about the metacarpus of either the Leporide or the Ochotonidz; both have metacarpi of a general- ized mammalian type. The first metacarpal is very short in both families. The fifth metacarpal is about twice or about two and a half times the length of the first. The middle metacarpal is the longest, but not much longer than the two subequal adjacent ones. The middle metacarpal is relatively longer in the Leporide than in the Ochotonidz, with respect to the two metacarpals on either side Onetts The metacarpus, as a whole, is longer in the Leporidz than it is in the Ochotonide. The width of the three middle metacarpals at their bases is contained about two and one-half times the length of the middle metacarpal in all the genera except Pentalagus. In the Ochotonide, however, and in Pentalagus of the Leporidz the basal width of the three middle metacarpals taken together is contained only about one and one-half times in the length of the middle meta- carpal. PRALANGES The phalanges are relatively longer in the Leporidz than in the Ochotonidz. In both families the length of the three phalanges of the middle digit about equals the length of the middle metacarpal, which bone, as noted above, is relatively longer in the hares and rabbits than in the pikas. OS INNOMINATUM LEPORIDA The os innominatum in all the genera of the Leporidz is generally uniform in shape, differences being slight and apparently of not much importance, yet they are fairly constant for some genera or groups of genera. In the genus Lepus, the ilium is broad and shovel-like; the antero- superior angle of the crest is rounded off, but not obliquely so as in the case of the other genera. The distance from the anterior edge of the acetabulum to the most anterior point of the ilium, is less than the distance from the same point to the most distant part of the ischium, while in all the other genera the distance from the anterior edge of the acetabulum to the most distant point of the ilium is equal to or a little greater than the distance measured from the same point to the most distant part of the ischium. LYON | THE HARES AND THEIR ALLIES 381 The obturator foramen is more rotund in Lepus than in most other genera. In Oryctolagus, Sylvilagus, and Brachylagus the ossa innominata have about the same general form. There is nothing to distinguish these bones in the first two genera, but the innominate bone of Brachylagus can be distinguished by its smaller size and the greater prominence of the tubercle in front of the acetabulum and by a slenderer descending ramus of the pubis and shorter distance from the tuberosity of the ischium to nearest point of the obturator foramen. In the above three genera the ilium is not so wide and shovel-like as it is in Lepus. The anterior edge of the acetabulum is about equidistant between the extreme anterior and posterior points of the os innominatum, or just a little posterior to the equidis- tant point. The antero-superior angle of the ilium is obliquely rounded off. The thyroid foramen is usually less rotund than it is in Lepus. The genus Limmnolagus has wide ilia, much like those of Lepus, but the antero-superior angle is not obliquely rounded off. The antero-ventral angle is produced into a blunt, very short spine. The horizontal rami of the pubic bones slope backward more than they do in the other genera except Romerolagus. The os innominatum of Romerolagus closely resembles that of Limnolagus. The only marked differences, aside from its slender formation, is the more pronounced development of the short, blunt spine at the antero-ventral angle, and the straightness of the ventral edge of the ilium. In all the other groups except the genus Prono- Jagus the ventral edge of the ilium has a more or less pronounced ‘concavity on its posterior half. The genus Pronolagus has an os innominatum resembling in most respects that of Sylvilagus and Oryctolagus, but the ilium is even narrower than it is in them. Its ventral edge is straight, in this respect being like the ventral edge of the ilium of Romerolagus. OcCHOTONIDE The pelvis of the Ochotonide is very different from the pelvis of the Leporide. The most notable difference is the absence of the symphysis pubis. As none of the few available skeletons is sexed, it is impossible to say whether the difference is sexual or not. The pubic bones are widely separated from one another, but are connected by a ligament. The os innominatum of the Ochotonide as a whole is longer and amore slender than that bone in the Leporidz. The ilium is less ex- 382 SMITHSONIAN MISCELLANEOUS COLLECTIONS [VoL. 45 panded, thicker, its ventral third separated from the dorsal two- thirds by a well-marked ridge, which terminates anteriorly and ventrally in a well-marked, recurved, pointed spine. The tuberosity of the ischium is not so heavy and it lacks the processes seen in the Leporidee. The thyroid foramen is egg-shaped, the small end of the egg being directed forward toward the acetabulum. The horizontal ramus of the pubis is very short, the descending ramus long, directed obliquely backward and downward. FEMUR The femora in all the Leporidz resemble one another very closely. The bone is stout and heavy in Pentalagus, it is relatively thicker in Oryctolagus than in the remaining genera, and it is slenderest in Brachylagus, but individuals vary much in this respect. Of two Lepus campestris, one from Nebraska and the other from Kansas, one is relatively thicker than the other. The same individual varia- tion is found in the series of skeletons of Sylvilagus floridanus. In .most of the skeletons the femur is a little shorter than the tibia. In Lepus (Macrotolagus) texianus, however, it is decidedly shorter, in Pentalagus the two bones are subequal in length. The femur in the Ochotonidz has the same general characteristics that it has in Leporidz, but it is a relatively thicker and heavier bone; the third trochanter is much reduced in size, the lesser trochanter is relatively larger, and the fossa behind the great trochanter is not so deep. Its length is a trifle less than that of the tibia. TIBIACAN Dili (PraTE XCIX) The tibia and fibula show about as few variations in the different groups of the Leporidze as do the femora. The fibula is a free, dis- tinct bone above, but at a point somewhat above the middle of the tibia it becomes intimately fused with the bone and its identity is lost sight of from there on, in all the genera except Romerolagus and Pentalagus, in which genera the fusion of fibula and tibia occurs at the middle of the latter bone. The point of fusion is highest in Lepus, next highest in Sylvilagus, Oryctolagus, Pronolagus, and Limnolagus, and in Brachylagus it is just above the middle point of the tibia. In Pentalagus the tibia and fibula are relatively much heavier than they are in the other genera. In Romerolagus and Limnolagus also the tibia is relatively heavier than in the remaining SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL: 45, PL. XCIX LEG BONES OF HARES, RABBITS, AND PIKAS (natural size), For explanation see page 447. LYON | THE HARES AND THEIR ALLIES 383 genera. In both Romerolagus and Limnolagus, especially the former, the tibia is more curved than it is in the other rabbits, the inner surface of the lower part of the shaft of the tibia being concave, as it is in Ochotona. In most of the skeletons the greatest length of the foot is about equal to the length of the tibia. In Brachylagus and Sylvilagus floridanus subsp. No. 94,197, from Nevada, the foot is longer than the tibia. In the subgenus Macrotolagus and in the genus Romero- lagus the foot is shorter than the tibia. The tibia and fibula in the Ochotonide are similar in form to the same bones in Romerolagus. TARSUS (PLATE C) The tarsal bones in all the skeletons show few differences among the Leporidz, except as to size. The middle cuneiform is not fused to the base of the second metatarsal. The tarsus of the Ochotonide is in general like that of the Leporide. The middle cuneiform, however, is entirely fused with the base of the second metatarsal. The foot, as a whole, is relatively shorter than it is in the Leporide. In the Ochotonide there are some small sesamoid bones, on the plantar surface of the tarsus, which are apparently lacking in the feet of the Leporidz. These bones are discussed by Major at con- siderable length. Their true significance and functions cannot be well determined without the dissection of fresh material. The largest of these sesamoid bones is situated on the plantar surface of the base of the fifth metatarsal. In the Leporidz this bone seems to have become fused with the base of the fifth metatarsal, where it forms a prominent tubercle. METATARSUS The metatarsals in both the Leporide and the Ochotonide are four in number, the first being suppressed. According to their relative lengths the metatarsi in the two families fall into four groups, as follows: The basal width of the metatarsus, measured from the tuberosity of the fifth to an opposite point on the second is contained about three times in the length of the third metatarsal in Lepus. It is contained about two and a half times in Syivilagus, Oryctolagus, Limnolagus, Brachylagus, and Pronolagus. It is contained two times in Romerolagus. In Pentalagus and Ochotona it is contained one and a half times. 384 SMITHSONIAN MISCELLANEOUS COLLECTIONS [VOL. 45 PHALANGES Little can be said regarding the toes. The two lateral digits are relatively longer in the Ochotonidz than they are in Leporide. In Ochotona, Romerolagus, Pronolagus, Limnolagus, Oryctolagus, and Brachylagus the combined lengths of the three phalanges approxi- mately equals the length of the metatarsal to which they belong. In the other genera, Lepus and Sylvilagus, this is true of the middle digits only; in the case of the lateral digits the metatarsals are de- cidedly longer than the combined length of the corresponding phalanges. V. OSTEOLOGICAL DIFFERENCES BETWEEN OCHOTONIDZ AND LEPORID® The principal osteological differences between the two families Ochotonide and Leporide, discussed in the foregoing pages, are briefly set forth in the following table: OCHOTONIDZ LEPORIDE Skull Not arched, flat, between the orbits. much constricted Rostrum short and slender. Nasals widest in front. Palate very short. Postorbital processes lacking. Posterior free edge of malar very long. Maxilla not conspicuously fenes- trated. Mental foramen of mandible far posterior to its usual position. More or less arched, not flat, only moderately constricted between the orbits. Rostrum long and stout. Nasals not wider in front than be- hind. Palate moderately short. Postorbital processes well oped. Posterior free edge of malar only moderately long. Maxilla more or less conspicuously fenestrated. Mental foramen of mandible an- terior, in its usual position. devel- Teeth Dental formula I12C% Pm3M 2 3 Enamel not crenated in reentrant angles of upper cheek teeth. Cutting edge of first upper incisor V-shaped. _ Third upper molar entirely want- ing, Second and fifth upper cheek teeth unlike third. Last lower molar simple. Dental ton 2C2Pm$M# (In one a Pentalagus, M2). Enamel usually crenated in reen- trant angles of upper cheek teeth. Cutting edge of first upper incisor straight. Third upper molar small but usually present (absent in Pentalagus only). Second and fifth upper cheek teeth entirely similar to third. Last lower molar double. SMITHSONIAN MISCELLANEOUS COLLECTIONS HIND FEET OF HARES, RABBITS, AND PIKAS (natural size). For explanation see page 447. LYON | THE HARES AND THEIR ALLIES 385 Vertebral Column Transverse process of the seventh cervical not pierced by a vertebral foramen. Thoracic number. Thoraco-lumbar two in number. Spinous processes of thoracic verte- bre relatively short. Spinous processes of lumbar ver- tebre relatively low. Transverse processes of vertebre very short and wide. No hypophyses on any of the lum- bar vertebre. Sacrum long and narrow, twice as long as wide. vertebre seventeen in vertebre twenty- lumbar Transverse process of the seventh cervical pierced by a vertebral fora- men. Thoracic vertebre twelve in num- ber. Thoraco-lumbar vertebre nineteen in number. Spinous processes of thoracic ver- tebre relatively long. Spinous processes of lumbar ver- tebre relatively high. Transverse processes of vertebree long and _ slender. Hypophyses found on the anterior three lumbar vertebre. Sacrum wide and triangular about as wide as long. lumbar Shoulder Girdle and Upper Extremity Clavicle well developed. Scapula with antero-superior angle much rounded off. Acromion very long, three times the length of actual scapular spine. Supraspinous fossa of scapula nar- row. Dorso-palmar not pronounced. Internal half of cuneiform more largely developed, presenting a well developed convex facet for articula- tion with the ulna, in addition to the cup-shaped cavity formed by the cune- iform and pisiform together. Centrale larger, not flask-shaped. Os magnum smaller. Width of three middle metacarpals at base contained about one and a half times into length of the middle metacarpal. depth of pisiform Clavicle small and rudimentary. Scapula with antero-superior angle not much rounded off. Acromion long, about half the length of the actual scapular spine. Supraspinous fossa of scapula much wider. Dorso-palmar depth of pisiform very great, except in Pentalagus. Internal half of cuneiform not largely developed nor presenting a well developed convex facet for articulation with the ulna, though sometimes such a condition is indi- cated. Centrale smaller, flask-shaped. Os magnum larger. Width of three middle metacarpals at base contained about two and a half times into length of the middle metacarpal, except in Pentalagus. Pelvis and Lower Extremity Pubic symphysis absent. Antero-ventral angle of ilium with a well-marked recurved spine. Middle cuneiform bone of the foot intimately fused with the base of the second metatarsal. Basal width of the metatarsals taken together contained less than twice into the length of the meta- tarsals. A large sesamoid bone on the plantar surface of the fifth metatarsal. Pubic symphysis well marked. Antero-ventral angle of ilium with- out a well-marked recurved spine. Middle cuneiform of foot not fused with the base of the second meta- tarsal. Basal width of the metatarsals con- tained two or more times into the length of the metatarsals, except in Pentalagus one and a half times. A prominent tubercle on the base of the plantar surface of the fifth metatarsal. 386 SMITHSONIAN MISCELLANEOUS COLLECTIONS [VoL. 45 VI. KEYS TO FAMILIES, GENERA, AND SUBGENERA OF THE DUPLICIDENTATA, BASED ON ODENTALS CRANIAL AND OTHER SKELETAL, CHARACTERS A. Key To THE FAMILIES AND GENERA OF EXISTING HARES, RABBITS, AND PrKAS, BAsED MAINLY ON DENTAL CHARACTERS a) Dental formula: I 2, C ?, Pm 3, M 2, second upper molariform tooth unlike third in form. Family OcHotonipm, genus Ochotona, p. 431. a’) Dental formula usually I ?, C §, Pm 3, M 2 (in Pentalagus the molars are 2), second upper maxillary tooth like third in form. Family, LEpo- RIDE, p. 380. b) Reentrant angle of second upper maxillary tooth short and not crenated: c) Reentrant angle extends one-third way across the tooth, genus Nesolagus, p. 425. c’) Reentrant angle extends half way across the maxillary teeth, genus Brachylagus, p. 411. b’) Reentrant angles of second, third, fourth, and fifth maxillary teeth with sides crenated. d) First lower premolar divided into an anterior and a posterior portion by two reentrant angles, one extending from the ex- ternal, the other from the internal face to the center of the tooth. e) Anterior face of first lower premolar without reentrant angles. Genus Romerolagus, p. 420. e’) Anterior face of first lower premolar with two reentrant angles. f) Posterior limb of reentrant angles of second, third, and fourth mandibular cheek teeth with antero-posterior convolutions nearly as long as the antero-posterior diameter of the posterior portions of the teeth. Genus Pentalagus, p. 428. f’) Posterior limb of reentrant angles of second, third, and fourth mandibular cheek teeth with normal convolu- tions, not nearly so long as the antero-posterior diameter of the posterior portions of the teeth. Genus Pronolagus, p. 416. d’) First lower premolar divided into an anterior and posterior por- tion by a single deep reentrant angle extending entirely across the tooth from its external surface. g) First lower premolar with but one reentrant angle on its anterior face. (Some Mexican Sylvilagus have two.) h) Sutures of the interparietal obliterated in the adult, genus Lepus, p. 380. i) Groove on incisors not filled with cement. Subgenus Pecilolagus, p. 395. Also some Lepus and Macroto- lagus. 1) Groove on incisors filled with.cement. J) Groove deep and simple. Mostly subgenus Lepus, Pp. 304. J’) Groove deep and bi- or tri-furcated. Mostly sub- genus Macrotolagus, p. 395. LYON | THE HARES AND THEIR ALLIES 387 h’) Sutures of the interparietal distinct throughout life. k) Choane narrow. Genus Oryctolagus, p. 402. k’) Choane not narrowed. Genus Sylvilagus, p. 396. g’) First lower premolar with two crenated reentrant angles on its anterior face. 1) Internal face of anterior portion of first lower premolar without reentrant angle. Genus Limnolagus, p. 406. 1’) Internal face of anterior portion of first lower premolar with reentrant angle. Genus Caprolagus, p. 426. B. Key To THE FAMILIES, GENERA, AND SUBGENERA OF EXISTING HArEs, RABBITS, AND PikAs, BASED ON CRANIAL CHARACTERS a) Postorbital process absent, skull flat, much constricted between orbits, rostrum short and slender, nasals widest in front, bony palate very short, posterior free edge of malar very long, maxilla not conspicuously fenes- trated, mental foramen of mandible situated under last lower molar. Family OcHoTonip#, genus Ochotona, p. 431. b) Interorbital region very narrow and pinched up and strongly arched from before backward. Subgenus Ochotona, p. 438. b’) Interorbital region moderately broad, not compressed laterally nor arched from before backward. c) Brain case and whole skull very flat, rostrum slender, its origin abrupt, no foramen in the anterior part of the frontal bone. Opening in maxilla single, roundly triangular. Incisive foramina constricted into two unequal portions. Subgenus Pika, p. 438. c’) Brain case rather rounded, whole skull moderately arched from before backward, rostrum relatively long and heavy, its origin not abrupt, a small oval foramen in frontal bone. Opening in maxilla elongated triangular with a small amount of fenestra- tion just beneath it. Incisive foramina triangular in outline, not constricted into two unequal portions. Subgenus Conothoa, p. 438. a’) Postorbital process well developed, skull arched from before backward, moderately constricted between the orbits, bony palate only moderately short, rostrum long and stout, nasals widest behind, posterior free edge of malar only moderately long, sides of maxilla more or less con- spicuously fenestrated, mental foramen of mandible situated under first lower cheek tooth. Family Lrportm®. d) Postorbital process attached to the side of the skull for its whole length. Genus Limnolagus, p. 406. d’) Postorbital process not attached to the side of the skull for its whole length, bounding well marked foramina or notches. e) Postorbital process small, without anterior limb, the single pos- terior limb pointed, directed outward and backward, bounding a distinct notch. Bony palate long. f) Nasals relatively short, as broad in front as behind. -g) Incisive foramina elongated triangle in outline, postorbital process smaller and pointed. Genus Caprolagus, p. 426. 388 SMITHSONIAN MISCELLANEOUS COLLECTIONS [VoL. 45 g’) Incisive foramina, not triangular in outline, small, their sides approximately parallel, postorbital process larger and blunter. Genus Pentalagus, p. 428. f’) Nasals relatively larger, wider behind than in front. h) Audital bulla much reduced in size, smaller than foramen magnum. Genus Pronolagus, p. 416. h’) Audital bulla normal, as large as the foramen magnum. Genus Romerolagus, p. 420. e’) Postorbital process large with distinct anterior and_ posterior limbs, bony palate relatively shorter, usually less than its poste- rior third formed by the horizontal plates of the palate bones. 7) Interparietal not present as a distinct bone in adult life. Genus Lepus, p. 389. J) Skull shortened and arched, posterior limb of postorbital not touching the sides of the cranium. k) Size large, postorbital process wide and heavy, nasals and rostrum wider and heavier. Spines of lumbar vertebree well marked. Subgenus Lepus, p. 304. k’) Size medium, postorbital process rather slenderer, nasals and rostrum narrower. Spines of lumbar vertebre low and triangular. Subgenus Pecilo- lagus, p. 395. 7’) Skull elongated, less arched, posterior limb of postorbital meeting side of cranium and enclosing a large foramen. Subgenus Macrotolagus, p. 395. 7’) Interparietal present as a distinct bone in adult life. 1) Choane very narrow, pharyngeal vault high, distance be- between vertical plates of palate bones less than least length of bony palate. Posterior limb of postorbital not meeting side of skull. Genus Oryctolagus, p. 402. l’) Choanz not remarkably narrow, the distance between the vertical plate of palate bones equal to or greater than the least length of bony plate. m) Skull and rostrum shortened, arched, audital bullz very large. Postorbitals slender, neither its an- terior nor its posterior limb touches the side of the skull. Genus Brachylagus, p. 411. m’) Skull and rostrum not relatively shortened, audital bullae not greatly enlarged, posterior limb of post- orbital usually in contact with side of the cranium, rostrum heavier. Genus Sylvilagus, p. 306. n) Posterior limb of postorbital always meeting side of cranium, and inclosing a clavate slit nar- rower than the process itself. Subgenus Sylvilagus, p. 401. n’) Posterior limb of postorbital not meeting the side of the cranium, or if it does, enclosing a clavate slit as wide or wider than the process itself. Rostrum slenderer. Subgenus Micro- lagus, p. 402. LYON] THE HARES AND THEIR ALLIES 389 C. Key To THE FAMILIES AND GENERA OF HaArEs, RABBITS, AND PIKAS, BASED ON SKELETAL CHARACTERS OTHER THAN CRANIAL (Because of insufficient material, the genera Caprolagus, Pentalagus, and Nesolagus can not be incorporated in this key.) a’) Transverse process of 7th cervical vertebra, with no vertebral foramen, thoracic vertebra 17, lumbar vertebra 5, sacrum much longer than wide. Acromion about 3 times the length of actual scapular spine, foot short, basal width of metatarsals contained 1% times into length of longest metatarsal. Family Ocnoronip®. Genus Ochotona, p. 431. a) Transverse process of 7th cervical vertebra, with no vertebral foramen, thoracic vertebre 12, lumbar 5, sacrum as wide as long. Acromion process about half the length of the actual scapular spine, foot long, basal width of metatarsals contained at least 2 times and usually 2% times into the length of longest metatarsal. Family Lrporrpa. b) Cervical vetrebre lengthened individually, ulna much reduced in size, smaller than radius and behind that bone, ventral half of anterior ribs much broadened. Genus Lepus, p. 304. b’) Cervical vertebrae individually shortened, ulna and radius subequal in size, the ulna toward the outer side of radius. Ventral half of anterior ribs not broadened. c) Transverse processes of the lumbar vertebrae not conspicuously ridged, and the angle between the process and the side of the centrum not conspicuously filled in with thin bone. d) Vubercles of anterior ribs not well marked. Genus Prono- lagus, p. 4106. d’) Yubercles of anterior ribs usually well marked. c) Presternum with well marked keel, xiphisternum large and stout, equally enlarged at both ends, longer than presternum. Genus Oryctolagus, p. 402. e’) Presternum with keel not well marked, xiphisternum shorter than presternum. Genus Sylvilagus, p. 390. c’) Transverse processes of lumbar vertebre conspicuously ridged or else angle between the process and the side of centrum con- spicuously filled in with thin bone. f’) Anterior portion of presternum conspicuously enlarged, like presternum of Ochotona. Genus Romerolagus, p. 420. f) Anterior portion of presternum not conspicuously enlarged. g) Size small, each transverse process of the lumbar verte- bre with a well marked longitudinal ridge. Genus Brachylagus, p. 411. g’) Size larger, angle between transverse process and side of centrum of lumbar vertebre filled in with thin bone. Genus Limnolagus, p. 400. il, DETAILED ACCOUNT OF THE GENERA AND SUB- GENERA OF THE EXISTING HARES, RABBITS, AND PIKAS FAMILY LEPORID/E Genus LEPUS Linnzus 1758. Lepus Linnzus, Syst. Nat., 1oth ed., 1, p. 57. 1829. Chionobates Kaur, Entw. Gesch. und Natur. Syst. Europ. Thier- welt, I, p. 170. 390 SMITHSONIAN MISCELLANEOUS COLLECTIONS [VoL. 45 1867. Lepus Gray, Ann. Mag. Nat. Hist., xx, 3d ser., p. 222. 1900. Lepus Major, Trans. Linn. Soc. London, 3d ser., vit, Zool., p. 514, November, 1899. Type.—Lepus timidus Linneus. Geographical Distribution —The genus Lepus is found throughout nearly every portion of the world except Australia, Madagascar, and South America. Diagnosis.—Pelage soft, a patch on throat different in color and texture from surrounding fur, ears as long as or longer than head, tail short but plainly evident. Hind feet long, heavily clothed with hair, claws not conspicuous. Sutures of interparietal obliterated in the adult, postorbital processes large and triangular, with distinct anterior and posterior limbs. Palate short, its least length two and one-half or less times the length of m*, choane wide, about four times length of mt. Teeth normal as described on page 301. Skull (pls. LXXIv, LXXV, LXXX-LXXxv).—The postorbital proc- esses are large and spreading, usually distinctly triangular, standing out from the side of the head and considerably arched from before backward. The sutures of the interparietal are obliterated at an early age, so that in adult life the interparietal is not distinguishable as a separate bone. Apparently the only other genera of the Leporidz in which this condition occurs are Pronolagus and Pentalagus. The bony palate is very short, but not relatively so short as it is in Brachylagus. The horizontal plates of the palate bones form be- tween a fourth and a third of the bony palate. The posterior por- tion of the palate bone, bordering the edge of the maxilla, caudad of the posterior edge of the bony palate, is very slightly developed and scarcely enters into the formation of the roof of the mouth. The posterior palatine foramina are of moderate size and located between the palatine plate of the maxilla and the horizontal plate of the palate bone, at the anterior outer angle of the horizontal plate. As in Brachylagus, the choane are wide in the genus Lepus. The length of the bony palate, taken at a point midway between the median line and the dental alveoli, is decidedly less than the least distance between the vertical plates of the palate bones. The incisive foramina are wide; their greatest width much exceeds the length of the bony palate, measured midway between the median line and the dental alveoli. The zygoma is deep, but not thickened. The foot-like extremity of the zygomatic process of the squamosal is shorter in Lepus than LYON] THE HARES AND THEIR ALLIES 391 in most of the other genera. The external lateral length of the squamoso-malar suture is contained about twice in the superior border of the zygoma, measured from the anterior end of the squamoso-malar suture to the antero-inferior angle of the orbit. The antero-inferior angle of the zygoma is only slightly enlarged in Lepus. The posterior free projecting extremity of the malar is short. The audital bullz, the fenestration of the maxilla, and the shape of the mandible throughout the genus show nothing distinctive and may be said to represent normal degrees of development. Teeth (pl. xci, 6).—The following are the dental characters of the genus Lepus: Front upper incisor with longitudinal groove in anterior face, more or less deep, simple or branched inter- nally, filled or unfilled with cement. First upper maxillary tooth has typical folding of enamel on the anterior surface, a deep median reentrant angle, on either side of which is a smaller reentrant angle. The first lower molariform tooth, divided into two portions by a single reentrant angle from the external face, has a small reentrant angle on its anterior surface, and a shallow broader reentrant angle on the external surface of the anterior half of the tooth. The second, third, fourth, and fifth upper molariform teeth show each a deep re- entrant angle extending from the internal face about three-quarters the distance across the tooth; the adjacent edges of this angle are al- most in contact with one another throughout their whole extent. The enamel of the reentrant angles is beautifully crenated. The lateral diameter of the posterior half of each of the second, third, and fourth lower molariform teeth is about four-fifths the lateral diameter of the anterior half of the tooth. The last upper molar is a small elliptic cylinder; the last lower molar is a small double cylinder in form, the anterior half of which is larger and elliptical, the posterior ielivtenete: Vertebral Column.—tThe cervical vertebre (pl. xci1, 7-9) of Lepus have a form characteristic for the genus and are uniformly lengthened as compared with the same series of vertebrz in the other genera. The costo-transverse process does not project laterally from the centrum to a marked extent, and the anterior and posterior spines of the same are more elongated. The thoracic vertebree show practically nothing characteristic of Lepus. The neural spines in the anterior part of the series equal about two and a half to three times the length of their centra. The anticlinal vertebra is usually the eleventh, but it may be the tenth. A well-developed metapophysis is found on the ninth thoracic. 392 SMITHSONIAN MISCELLANEOUS COLLECTIONS [VoL. 45 The lumbar vertebrz (pl. xctII, 1, 2) have transverse processes peculiar to this genus. They are wide and long with the free ex- tremity considerably expanded. The length of the longest process equals the length of the centrum to which it is attached and half the length of the centrum in front. The attached portion of the transverse process arises abruptly from the anterior half of the side of the centrum. The anapophyses are indicated by a mere ridge or tubercle. Hypophyses are developed on the first three lumbars, the first being the shortest and the third the longest. Except in absolute size the sacra in all the skeletons but one differ in no respect from the sacra of the other genera. The single excep- tion is No. 49,621, from the Jumna river, India. The narrow posterior part of this sacrum is very narrow from side to side. The expanded wing-like portion to which the ilia are attached is very narrow. In general its shape, as seen from below, is like a T. The adjacent edges of the first and third, and of the third and fourth sacral neural processes are in contact. The total number of caudal vertebrz varies from thirteen to fifteen ; of these the first is long and with a complete neural arch, the next five or seven, usually seven, are of the form having more or less evi- dent broad transverse processes, the remaining five to seven are small rudimentary centra, without processes. Sternum and Ribs (pls. xctv, 1, 2; XCv, 3, 4).—The sternum of the genus Lepus is not very characteristic. The presternum is long and narrow, compressed laterally into a keel, which is most prominent anteriorly. The first rib articulates with it at the junction of the first and second fourths, or first and second thirds. The mesoster- num consists of four distinct segments, of which the first is narrow and compressed, the remaining three are not compressed, and are often flattened dorso-ventrally. All the mesosternal segments in general are subequal in length, and each successive one grows wider. The xiphisternum is about equal to the presternum in length; its anterior end is considerably enlarged where it articulates with the last segment of the mesosternum. Its posterior free extremity is rather pointed and only slightly larger than its narrowest portion. The form of the ribs is quite characteristic for the genus Lepus. The spine-like portions of the tubercles of the ribs are well developed and prominent. The eighth pair of ribs is the last pair bearing these tubercles. The second, third, fourth, and fifth ribs are very flat and broad on the ventral half of their shafts. The greatest width of one of these ribs, just behind the spine and tubercles, is very much less than the width of the shaft in its lower portion. Seven pairs of ribs LYON] THE HARES AND THEIR ALLIES 393 articulate with the sternum. The sternal costal cartilages are very short and wide as compared with the same structures in the other genera. Shoulder Girdle and Upper Extremity.—The scapula (pl. xcvu, I, 2) in Lepus, especially in the larger members, has a tendency to be relatively broad, with the superior border rather more convex than it is in the other genera, with the antero-superior angle more rounded, and with the supra-spinous fossa relatively wide. The radius and ulna (pl. xcvit1, 1-3) show characteristics peculiar to the genus. The ulna is much reduced in size along the middle of the shaft, and except at the lower extremity it is placed almost entirely behind the radius. The radius itself is rather long and _ slender. The humerus and radius are usually subequal in length. The carpus, the metacarpus, and the phalanges of Lepus are entirely similar in form and position to these same bones in any of the Leporidz as detailed in the general account of the wrist and the hand. Pelvis and Lower Extremity.—The ilium is broad and shovel-like in the genus Lepus,; its antero-superior angle is rounded off, but not obliquely so, as in the case of the other genera. The distance from the anterior edge of the acetabulum to the extreme anterior point of the ilium is less than the distance from the former point to the most distant point of the ischium. In all the other genera the former distance is usually equal to or a little greater than the latter. The obturator foramen is usually more rotund in Lepus than in the rest of the family. The femur and the tibia and fibula (pl. xcrx, 1) of Lepus are typical for the family, as detailed in the general account and show nothing that is peculiar to the genus. The basal width of the metatarsus is contained two and a half or more often three times into the length of the third metatarsal, as is commonly the case among the Leporidz (pl. c, 6). The combined length of the phalanges of the two lateral digits is less than the length of the metatarsal to which they belong. The combined length of the phalanges of the middle two digits approxi- mately equals the lengths of the metatarsals to which they belong. The genus Lepus possesses a number of skeletal characters quite peculiar to itself and serving to separate it at once from all the other genera of the Leporide. Among these are form of postorbital proc- esses, shortness of palate, reduced size of ulna, shape of anterior ribs, shape of transverse processes of lumbar vertebra, and the elongated form of the cervical vertebrae. Not one of these charac- 394 SMITHSONIAN MISCELLANEOUS COLLECTIONS [VoL. 45 ters is found in any of the other genera. To these might also be added the obliteration of the sutures of the interparietal in the adult, which is nearly as characteristic. Thus there is considerable foundation for Major’s view that the existing hares and rabbits fall into two supergeneric groups, one con- taining Lepus, and one containing the other genera, which he desig- nates as Caprolagus. It is the writer’s opinion, however, that Major’s group, Caprolagus, contains many genera as well defined as Lepus. The genus Lepus has the most extensive geographic distribution and the greatest number of species of any of the Leporide. As would be expected, there are several groups of these species, consti- tuting at least three well-marked subgenera for forms from North America and northern Eurasia described below. There are a num- ber of skulls at hand, however, from central and eastern Asia and from Africa, which it is impossible to classify satisfactorily, owing to lack of material, and no provision is made for these in the present discussion. For a list of these specimens see page 335. Subgenus LEPUS Linnzus 1758. Lepus Linnzus, Syst. Nat., toth ed., 1, p. 57. 1900. Lepus Major, Trans. Linn. Soc. London, 2d ser., vim, Zool., p. 514, November, 1899. Type—Lepus timidus LINNZUs. Geographical Distribution—Mainly the northern portions of the New and Old Worlds. Diagnosis.—Size large, ears moderate, greatest length of skull nearly 100 mm.; skull broad, postorbital broader than notch it sub- tends, its posterior limb not touching side of cranium. The skull is broad and strongly arched. The rostrum is shorter and broader and the brain case is broader and the nasals shorter and broader than they are in the other members of the genus Lepus. The postorbital processes are broad and triangular, neither the anterior nor the posterior limbs touch the sides of the cranium, but help to bound well-marked anterior and posterior notches. Teeth (figure 44, 19-21 ).—The groove on the front of the incisors is fairly deep, is simple and usually filled with cement. This subgenus apparently possesses no other skeletal character. Species included in this subgenus, see p. 334. LYON] THE HARES AND THEIR ALLIES 395 Subgenus P@CILOLAGUS, new Type.—Lepus americanus ERXLEBEN. Geographical Distribution—The Canadian, UHudsonian, and Transition zones of North America. Diagnosis —External, cranial, and dental characters essentially as in subgenus Lepus, but size smaller, greatest length of skull about 75 mm., portorbital process slenderer than in Lepus, its posterior fimb free behind, rostrum more pointed, groove on incisors simple, shallow, and without cement. Spinous processes of lumbar vertebre low and triangular in form. Sin pls. LXKV, 15 LXXVI,-1, 45 LXXXIV; LXXxV, 6-12).— Phe members of this subgenus are hares of medium size, the skull being about 75 mm. long. The skull is less strongly arched than it is in the true Lepus, but not so flat and long as in Macrotolagus. The nasals and rostrum are narrower than they are in Lepus. The post- orbital processes are much slenderer, less triangular than they are in Lepus or Macrotolagus, free both in front and behind. They bear considerable resemblance to the postorbital seen in Oryctolagus. Teeth (figure 44, 18).—The first upper incisors bear each a simple shallow groove not filled with cement. The remaining teeth are as in true Lepus. Vertebral Column.—The spines of the anterior dorsal vertebre are relatively shorter than they are in true Lepus, being equal to about two and a half times the length of the centra, instead of three times. The spinous processes of the lumbar vertebrae (pl. xc, 2) are also relatively lower in Pwcilolagus than in Lepus; they are tri- angular in outline, with the posterior edge nearly straight, while in the other subgenera the posterior edge is much cut out. Species included in this subgenus, see pp. 334 and 335. Subgenus MACROTOLAGUS Mearns 1896. Macrotolagus Mearns, Proc. U. S. Nat. Mus., xvit, p. 552, June 24, 1806. 1899. Macrotolagus Major, Trans. Linn. Soc. London, 2d ser., vit, Zool., pp. 468, 469, November, 1899. Type.—Lepus allen’ MEARNS. Geographical Distribution—Mexico and the southwestern United States. Diagnosis.—Size large, as in subgenus Lepus, ears longer than head or hind foot, skull narrow, posterior limb of postorbital touch- ing side of cranium, enclosing an oval foramen. 396 SMITHSONIAN MISCELLANEOUS COLLECTIONS [VOL. 45 Skull (pls: LXXIV, 15. LXXV,,25 Lexx; Dex xin)—— Pherskulleor Macrotolagus is large and slender. It is not arched as in the sub- genus Lepus. The nasals are larger and narrower, the brain case is narrower, and the choanze somewhat narrower. The postorbital processes are larger, rather longer than they are in the true Lepus, and their posterior angles are always attached to the side of the cranium enclosing large foramina. Teeth (figure 44, 11-17).—The groove on the front of the in- cisor teeth in this subgenus may be perfectly simple and not filled with cement as it is in Lepus (Macrotolagus) californius, but more often it is deep and filled with cement as in L. (M.) texianus, where the groove is simple, and in gaillardi, alleni, merriami, where the groove is bifurcated internally, and in callotis and asellus where the grove is trifurcated internally. This is the only group of the Leporide in North America that shows this complicated folding of the groove internally, but at the same time some members of it do not show it. When found in a North American rabbit this folding of the groove on the incisor is diagnostic of the subgenus, but when not found it is without significance. Vertebral Column.—The length of the neural spines is a little greater than it is in either the subgenus Lepus or Pecilolagus. In the latter two it is about three times the length of the centrum in the anterior part of the thoracic series of vertebrze; in Macrotolagus it is a little over three times the length of the centrum. Upper Extremity (pl. xcviit, 2).—The radius is slenderer than in the other members of the genus Lepus and longer than the humerus, while in the other subgenera, the humerus and radius are about equal in length. Species belonging to this subgenus, see p. 335. Genus SYLVILAGUS Gray 1867. Sylvilagus Gray, Ann. Mag. Nat. Hist., xx, 3d ser., p. 221. 1896. Sylvilagus MEARNS, Proc. U. S. Nat. Mus., xvitt, p. 551, June 24, 1806. 1899. Sylvilagus Major, Trans. Linn. Soc. London, 2d ser., v1, Zool., p. 514, November, 1899. Type.—Syivilagus floridanus mallurus (THOMAS). Geographical Distribution—In general, North America south of the northern border of the United States and in South America. Diagnosis—External and dental characters as in Lepus. Skull essentially like that of Lepus except that sutures of interparietal are always distinct; the postorbital process is long and narrow, its LYON] THE HARES AND THEIR ALLIES 397 posterior limb touching the side of the cranium and helping to bound a clavate foramen. Rarely the process fails to meet the cranium. Skull (pls. LXv1I, 7-10; LXVII, I, 2; LXXXVI; LXXxvi1).—The post- orbital processes are long and narrow, attached to the skull by a very broad pedicle. The whole anterior part of the process, with the exception of a millimeter or two, is attached to the skull and a very small anterior notch is thus formed. The posterior part of the postorbital process is long and narrow, not triangular. Its inner posterior edge usually touches the cranium and the process thus forms a narrow clavate slit. The amount of fusion of the postorbital process is subject to consider- able variation. In one individual, Sy/vilagus foridanus mearnsi, from Illinois, complete fusion of the postorbital with the side of the skull takes place and the usual clavate slit is obliterated. In Sylvilagus arizone and its forms the anterior portion of the postorbital process is not so extensively fused with the frontal bone and a larger anterior notch is found than in the more typical forms. In the most highly differentiated forms of the subgenus Microlagus, viz., bach- mam, the postorbital process is attached by a comparatively narrow pedicle, a large notch is found in front, and as the posterior end of the postorbital does not meet with the cranium a large notch is found posteriorly instead of the usual clavate slit. In No. 63,957, Sylvilagus bachmani, from Posts, California, the posterior extremity of the postorbital almost touches the cranium. Skulls of S. (Microlagus) cmerascens show conditions ranging from those seen in No. 63,957 to conditions very similar to those found in typical Sylvilagus, but the posterior clavate slit is always relatively wider. Intermediate conditions are thus found from the extreme freedom of the post- orbital seen in S. bachmani, No. 35,131, Nicasio, California, passing through S. cinerascens and then S. arizone and normal typical Sy/v- lagus, to the atypical skull, from Illinois, No. 22,409, where the postorbitals are entirely fused to the sides of the cranium. The bony palate in true Sylvilagus is rather short, slightly larger relatively than it is in Lepus or Brachylagus, but shorter than in any of the other genera. Its length half-way between the median line and the dental alveoli about equals the greatest width of the incisive foramina and also the distance between the vertical plates of the palate bones. In the Sylvilagus arizone group the palate is shorter than in typical Sylvilagus; its length half-way between the median line and the dental alveoli is less than the width of the incisive foramina and about equals the distance between the vertical plates of the palate bones, which are here closer together than in 398 SMITHSONIAN MISCELLANEOUS COLLECTIONS [VOL. 45. any of the rabbits except those of the genus Oryctolagus. The por- tion of the palate bone that borders the maxilla caudad of the posterior edge of the bony palate is poorly developed in Sylvilagus. The posterior palatine foramina are of moderate size and located between the palatine plate of the maxilla and the horizontal plate of the palate bone, at the anterior outer angles of the horizontal plate. The zygoma of Sylvilagus is thin and shallow. The foot-like extremity of zygomatic process of the squamosal is short, the external lateral length of the squamoso-malar suture is contained about two times into the superior border of the malar measured from the anterior end of the squamoso-malar suture to the antero-inferior angle of the orbit. The antero-inferior end of the zygoma is only slightly enlarged. The posterior free extremity of the malar is short. The audital bulle, the external auditory meatus and the fenestra- tion of the maxillz in Sy/vilagus show nothing distinctive and may be said to represent normal degrees of development. The angular process of the mandible is rather larger and more rounded, as compared with Lepus. The ascending ramus is of moderate development, resembling that of Lepus. Teeth (pl. xct, 9; figure 44, 2 and 3).—The following are the char- acters of the teeth of the genus Sy/vilagus, which, with the exception of the front upper incisors and first lower premolar, and smaller size throughout, are not different from those of the teeth of Lepus: First upper incisors with longitudinal groove in anterior face always simple, usually shallow and filled with cement only in a few speci- mens from Mexico. The first upper maxillary tooth has typical fold- ing of enamel on the anterior surface, a deep median reentrant angle on either side of which is a smaller reentrant angle. The first lower molariform tooth has a small reentrant angle on its anterior surface and a shallow, broader reentrant angle on the external surface of the anterior half of the tooth. In some of the specimens of Sylvi- lagus, from Mexico, the anterior face of this tooth is marked by two reentrant angles, instead of the usual one, the tooth in this case resembling the first lower premolar of Limnolagus. The second, third, fourth, and fifth upper molariform teeth show each a deep reentrant angle extending from the internal face about three-quarters the distance across the tooth; the adjacent edges of this angle are almost in contact with one another throughout their whole extent. The enamel of the reentrant angle is crenated. In the second, third, and fourth lower molariform teeth the lateral diameter of the pos- terior half of each tooth is about four-fifths of the lateral diameter LYON] THE HARES AND THEIR ALLIES 399 of the anterior half of the tooth. The last upper molar is small with an elliptic cross section. The last lower molar is a small double cylinder in form, the anterior half of which is farger and elliptic in section, the posterior half terete. Vertebral Column.—the cervical vertebre (pl. xci1, 5) of Sylvi- lagus have nothing to distinguish them from the same vertebrz of many other genera, being noticeably different from those of Lepus and Pronolagus only. The cervicals are uniformly shortened, the costal processes project laterally farther from the sides of the centrum, the anterior and the posterior spines of these processes are less prominent, than in Lepus. The true transverse process is more conspicuous and is often found to project laterally from the side of the fifth vertebra. The length of a neural spine in the anterior thoracic region is about twice the length of the centrum, to which it is attached. The anti- clinal vertebra is usually the eleventh, but often the tenth. Meta- pophyses are usually seen only on the tenth, eleventh, and twelfth thoracic vertebrze, sometimes on the ninth. The lumbar vertebre (pl. xciv, 14) are scarcely distinguishable from the same in Oryctolagus. ‘The transverse processes instead of arising abruptly from the anterior half of the lateral aspect of the centra as they do in Lepus, arise from the whole side, so that the angle between the process and the centrum is partly filled in with bone, but this is not comparable to the filling in that takes place in Limnolagus and in Romerolagus. True anapophyses are lacking, the neural spines and metapophyses show nothing of interest. The hypophysis on the second lumbar is the longest, that on the first is next in length, and the shortest is found on the third. Sometimes the first hypo- physis is reduced to a ridge and the third is lacking. The sacrum differs in no way from the description given in the general account. The caudal vertebre vary from eleven to fifteen in number. The larger number is the more usual. Of the different forms (see p. 363) of vertebre in this region, one, or often two, is of the first form; five to seven of the second, usually six; four to seven of the last form, usually six or seven. The skeleton with the lowest number of caudals, eleven, is Sy/vilagus minensis THomas. The decrease takes place in the last form, which is here reduced to four in number. Sternum and Ribs——The sternum (pl. xcv, 2) of Sylvilagus is not very characteristic. In general it resembles the sternum of Lepus. The presternum is relatively larger, the keel less prominent anteriorly. The first pair of ribs is attached just in front of its 400 SMITHSONIAN MISCELLANEOUS COLLECTIONS [VoOL. 45 middle. The mesosternal segments have a tendency to be less flat- tened than they are in Lepus. The xiphisternum is shorter than the presternum, its ends moderately enlarged, the posterior end being more enlarged, relatively, than in Lepus. The ribs (pl. xctv, 5) are somewhat peculiar, but there is nothing to distinguish them from the ribs of Oryctolagus. The shafts of the anterior ribs are only moderately enlarged in their ventral portions. The spine-like portion of tubercles is well de- veloped on all the skeletons except that of Sylvilagus minensis where the spines are poorly developed and are last seen on the seventh pair of ribs instead of on the eighth as in the others. Shoulder Girdle and Upper Extremity—The scapula (pl. xcvir, 7) of Sylvilagus is different in form from the same bone of Lepus, Romerolagus, and Pronolagus, but shows nothing tangible by which it can be distinguished from the scapulz of the other genera. It is relatively narrow, the superior border relatively less convex, the antero-superior angle moderately pronounced, and the supraspinous fossa relatively narrow, when compared with the scapula of Lepus. Regarding the humerus of Sylvilagus there is nothing peculiar by which it can be distinguished from the humeri of many of the other genera. The forms, relations, sizes, and positions of the radius and ulna (pl. xcvir, 4) are alike in Sylvilagus and in Limnolagus. These bones are subequal in size; the ulna is not reduced in the middle of its shaft, and is situated external to radius rather than behind it. Both bones are moderately slender. The radius is equal to the humerus in length. The carpus, the metacarpus, and the phalanges of Sy/vilagus are entirely similar in form and position to these bones in any of the Leporida, as detailed in the general account of the wrist and hand (p. 378). Pelvis and Lower Extremity.—The os innominatum has about the same general form in Syvilagus and in Oryctolagus. The ilium is not wide and shovel-like. The anterior edge of the acetabulum is about midway between the extreme anterior and posterior points of the os innominatum or just a little posterior to the middle point. The antero-superior angle of the illum is rather obliquely rounded off. The obturator foramen is apparently less rotund than is the case with Lepus. The femur, the tibia, and the fibula of Sy/vilagus are typical for the family, as detailed in the general account (p. 382), and show nothing that is peculiar to the genus. LYON | THE HARES AND THEIR ALLIES 401 The greatest length of the foot about equals the length of the tibia, but in one skeleton, No. 94,197, from Nevada, the foot is longer than the tibia, in this respect resembling the foot and tibia of Brachylagus. The basal width of the metatarsus is contained two and a half times into the length of the third metatarsal, as is commonly the case among other Leporide. The combined length of the phalanges of the two lateral digits, as well as of the two middle ones, is usually approximately equal to the length of the metatarsals to which they belong. The genus Sylvilagus stands second to Lepus in point of number of species and extent of geographic distribution. It is a fairly homo- geneous group. A few forms have been classed as a separate sub- genus, an arrangement which is here retained. The teeth on the whole are formed as they are in typical Lepus. Certain of the Mexican species, however, have the first lower premolar as it is in Limnolagus. In the form of the cervical vertebrze and the shape of the transverse processes of the lumbar vertebrz, the relative size of the radius and ulna, and in the shape of the pelvis and the ribs, sternum, and scapula, Sylvilagus seems to be generally similar to Oryctolagus. The teeth in these two genera are essentially alike. There are two subgenera of Syivilagus—Sylvilagus proper and Microlagus. No skeletons of the latter are available for examination. Subgenus SYLVILAGUS Gray 1867. Sylvilagus Gray, Ann. Mag. Nat. Hist., xx, 3d ser., p. 221. 1896. Sylvilagus Mearns, Proc. U. S. Nat. Mus., xvui, p. 551. June 24, 1806. 1899. Sylvilagus Major, Trans. Linn. Soc. London, 2d ser., vil, p. 514, November, 1899. Type—Sylvilagus floridanus mallurus (THOMAS). Geographical Distribution.—Same as for the genus. Diagnosis and Description.—Size larger, skull (pls. LXxv1, 7, 8; LXXVII, 2; LXXXVI} LXXXVII, I-10) heavier, rostrum often heavier and not so pointed. Postorbital processes long and narrow, attached to skull by a very broad pedicle. The anterior part of the process short, attached to the side of the skull for nearly its whole length. Posterior part of the process long and narrow, the posterior edge touching the cranium, the process thus enclosing a very narrow clavate slit, much narrower than the width of the postorbital, which helps to form it. Species in this subgenus, see p. 330. 402 SMITHSONIAN MISCELLANEOUS COLLECTIONS [VOL. 45 Subgenus MICROLAGUS Trouessart 1897. Microlagus TRouESSART, Catalogus Mammalium, 1, fase. i11, p. 660. Type.—Sylvilagus (Microlagus) cinerascens (ALLEN). Geographical Distribution—Western and southwestern United States and northern Mexico. Diagnosis and Description.—Externally like Sylvilagus, size small, skull. (pls: LXXVI,,0,. 10; EXXVIl, Ij oGeVES EX KV 7 I oe rostrum rather straight, narrow, pointed ; postorbital process attached to skull by a narrow pedicle, the anterior portion of the postorbital more or less free from the skull, and helping to form a well-marked notch. Posterior portion of postorbital rather slender. It may or may not touch the side of the cranium; when it does, it incloses a clavate foramen, fully as wide as the posterior limb of the process. For the species belonging to this subgenus, see pp. 336, 337. Genus ORYCTOLAGUS Lilljeborg 1867. Cuniculus Gray, Ann. Mag. Nat. Hist., 3d ser., xx, p. 225. 1874. Oryctolagus LiLLJEBOoRG, Sveriges och Norges Ryggradsdjur, 1, Dp. 417. 1899. Oryctolagus Major, Trans. Linn. Soc. London, 2d ser., vu, Zool., November, 1899, p. 514. Type—Lepus cuniculus LINN-EUS. Geographical Distribution—Southern and western Europe and northern Africa. Diagnosis —Externally similar to Lepus, but ears and hind feet relatively shorter. Postorbital process large but not wide and triangular; its posterior limb does not touch side of cranium. Sutures of the interparietal distinct. Choanze very narrow, incisive foramina and least length of palate subequal, width of choanze much less than either. Teeth essentially as in Lepus. Skull (pls. LXXVI, 2, 5; LKXXVII, 5; LXXXVIII; LXXXIX, I-5).—The postorbital processes are large, but not wide and triangular; they do not stand out from the side of the skull to a marked extent. The process is arched. Neither the anterior nor the long and rather pointed posterior portions meet the side of the cranium in the wild rabbits. Anterior and posterior notches are thus formed. In two of the skulls of domestic rabbits at hand, a lop-eared and a Belgian hare, the anterior angle of the postorbital meets the frontal bone, and in the lop-ear forms the outer boundary of an irregular foramen, while in the Belgian hare the whole anterior angle is fused to the cranium so that even the foramen is obliterated. In both these speci- mens posterior notches are present. LYON ] THE HARES AND THEIR ALLIES 403 The interparietal is present as a distinct bone in Oryctolagus. The bony palate is relatively long. The horizontal plate of the palate bone well developed. It enters into the formation of the bony palate to a greater extent than it does in Lepus, where it forms a fourth to a third of the bony palate, and to a less extent than it does in Romerolagus where it is about half of the bony palate. The portion of the palate bone that borders the maxilla caudad of the posterior edge of the bony palate is moderately developed and thus helps to form part of the roof of the mouth along the posterior dental alveoli. The distance between the vertical plates of the palate bones, that is the width of the choanz, is very slight, less than it is in any other genus. The pharyngeal vault is high. The length of the bony palate, measured midway between the median line and the inner edge of the dental alveoli, is about equal to the greatest width of the incisive foramina, which are narrow in this genus, and it is much greater than the least distance between the two vertical plates of the palate bones. The posterior palatine foramina are of moderate size and situated in the usual position at the anterior outer angles of the horizontal plates of the palate bones. The anterior portion of the zygoma is deep, as it is in Lepus, but the under edge of the posterior portion is cut away, rendering it shallower. The antero-inferior angle of the zygoma is enlarged and flares outward to a rather marked extent. The foot-like extremity of the zygomatic process of the squamosal is enlarged. The lateral length of the squamoso-malar suture is contained between one and one and a half times in the superior border of the malar, measured from the anterior edge of that suture to the antero-inferior angle of the orbit. The posterior free extremity of the malar is long. The audital bulla, the fenestration of the maxille, and the shape of the mandible in Oryctolagus show nothing distinctive of the genus, and may be said to represent normal degrees of development. Teeth (pl. xc, 5; figure 44, 22).—First upper incisors, with longitudinal groove on anterior face, always simple, moderately shallow and unfilled with cement. First upper maxillary tooth has typical folding of enamel on the anterior surface, a deep median reentrant angle, on either side of which is a smaller reentrant angle. The first lower molariform tooth has a small reentrant angle on its anterior face, and a shallow broader reentrant angle on the external surface of the anterior half of the tooth which is separated from the posterior half by a single reentrant angle. The second, third, fourth, and fifth upper molariform teeth show each a deep reentrant angle 404 SMITHSONIAN MISCELLANEOUS COLLECTIONS [VOL. 45 extending from the internal face about three-quarters the distance across the tooth. The adjacent edges of this angle are almost in contact with one another throughout their whole extent and crenated. In the second, third, and fourth lower molariform teeth the lateral diameter of the posterior half of each tooth is about four-fifths the lateral diameter of the anterior half of the tooth. The last upper molar is very small, with an elliptical outline. The last lower molar is also small, consisting of a larger anterior elliptic portion and a smaller posterior portion, circular in outline. Vertebral Column.—The cervical vertebre (pl. xctI, 6) of Oryctolagus have nothing to distinguish them from the same series of vertebrz of several other genera, being noticeably different only from the two extremes, Lepus, with elongated cervicals on the one hand, and Pronolagus, with more shortened ones, on the other. The cervicals are uniformly shortened, the costal processes project relatively far from the sides of the centrum; the anterior and the posterior spines of these processes are less prominent than they are in Lepus. The true transverse process is more conspicuous and is often found to project laterally from the side of the fifth vertebra. The length of the neural spines in the anterior thoracic region is about three times the length of a centrum. The anticlinal vertebra is the eleventh, on the lop-eared domestic it is the tenth, however. Metapophyses are well developed on the tenth, eleventh, and twelfth thoracic and are indicated on the ninth by small tubercles. The lumbar vertebre (pl. xciir, 4, 5) are scarcely distinguishable from the same series of vertebre in Sylvilagus. The transverse processes do not arise abruptly from the anterior half of the lateral aspect of the centra, but from the whole side, so that the angle be- tween the process and the centrum is partly filled in with bone, but in no way comparable to the filling in that takes place in Limnolagus and in Romerolagus. Anapophyses are best developed in this genus of all the Leporide. In the skeleton of the lop-eared domestic, these processes are very large; in the three middle vertebrz of the lumbar series, the third, fourth, and fifth, the anapophyses extend as far caudad as the posterior border of the metapophyses of the next suc- ceeding vertebra. In the two skeletons of wild Oryctolagus, how- ever, this great development of the anapophyses is not so pro- nounced, but they are much larger than they are on the lumbar vertebree of any of the other skeletons. The hypophyses are injured in the skeletons of the wild animals, but from what remains of them it would appear that the second hypophysis is the longest, as it is in the lop-eared domestic. LYON] THE HARES AND THEIR ALLIES 405 The sacrum of Oryctolagus differs in no way from the description given in the general account, p. 362. The caudal vertebree are about seventeen in,number, one being of the first form (see p. 363), six to eight of the second, and the last eight or nine are the rudimentary elongated centra, without neural canal or processes. Sternum and Ribs——The sternum (pl. xcv, 5) in Oryctolagus is in general as in Lepus. The presternum is more conspicuously keeled ; the first pair of ribs attached just anterior to its middle. The mesosternal segments are more compressed from side to side than they are in Lepus. The last is much shorter than any of the other mesosternal segments. The xiphisternum is large and stout, longer than the presternum, its posterior end about as much enlarged as its anterior. There is nothing by which the ribs of Oryctolagus can be distin- guished from the ribs of Sy/vilagus. The shafts of the anterior ribs are scarcely enlarged in their ventral portion. Spine-like portions of the tubercles of the ribs are well developed, and are last seen in the eighth pair. Shoulder Girdle and Upper Extremity.—The scapula (pl. xcvit, 3) of Oryetolagus is different in form from the same bone in Lepus, Romerolagus, and Pronolagus, but shows nothing tangible by which it can be distinguished from the scapulz of the other genera. It is relatively narrow, the superior border relatively convex, the antero-superior angle moderately prominent, and the supraspinous fossa relatively narrow, as compared with the scapula of Lepus. Regarding the humerus of Oryctolagus there is nothing by which it can be distinguished from the humeri of many of the other genera. The form, relative sizes, and positions of the radius and ulna are nearly alike in Oryctolagus, Sylvilagus, Pronolagus, and Limnolagus; that is, the ulna is not reduced in the middle of its shaft and is situated external to the radius rather than behind it. In Oryctolagus the ulna is somewhat heavier than the radius and both bones are rather heavier than they are in the other genera mentioned above. The radius is equal to the humerus in length. The carpus, the metacarpus, and the phalanges of Oryctolagus are entirely similar in form and position to these same bones in any of the Leporidz, as detailed in the general account of the wrist and hand (p. 378). Pelvis and Lower Extremity.—The os innominatum has about the same general form in Oryctolagus and in Sylvilagus. The ilium is not wide and shovel-like except in the lop-eared domestic. The 400 SMITHSONIAN MISCELLANEOUS COLLECTIONS [VOL. 45 antero-superior angle is rather obliquely rounded off. The anterior edge of the acetabulum is about midway between the extreme anterior and the extreme posterior points of the bone. The obturator foramen is less rotund than it is in Lepus. The femur, the tibia, and the fibula of Oryctolagus are typical for the family Leporidz, as detailed in the general account (p. 382), and show nothing that is peculiar to the genus. The greatest length of the foot about equals the length of the tibia. The basal width of the metatarsus is contained two and a half times in the length of the third metatarsal, as is commonly the case among the Leporide. The combined length of the phalanges of the two lateral digits, as well as of the two middle ones, is approximately equal to the length of the metatarsals to which they belong. Species in the genus. One, O. cuniculus (LINNZUS). Oryctolagus 1s a well-marked genus. In spite of a rather extensive geographic distribution, only one species has been recognized. It seems to have few relationships with the other Old World genera. With Sylvilagus in America, it has certain points of resemblance, in the shape of the transverse processes of the lumbar vertebra, of the scapula, in the relative size of the radius and ulna, in the shape of the pelvis and of the ribs and sterna. The skulls of the two genera taken as a whole are not markedly different, and the teeth are essentially alike. Oryctolagus has been extensively domesticated, and many stable varieties produced, among them the Belgian hare, but the true generic characters are never lost, and some of them, as the keeled presternum and the large anapophyses of the lumbar vertebre are even better marked in the domestic rabbits than they are in the wild ones. Genus LIMNOLAGUS Mearns 1867. Hydrolagus Gray, Ann. Mag. Nat. Hist., xx, 3d ser., 1867, p. 221. 1897. Limnolagus MEarNs, Science, n. s., v, March 5, 1897, p. 393. 1899. Limnolagus Mayor, Trans. Linn. Soc. London, 2d ser., vit, Zool., November, 1899, p. 514. Type—Limnolagus aquaticus (BACHMAN). Geographical Distribution—Austro-riparian faunal area of North America, in general the South Atlantic and Gulf states. Diagnosis —Externally similar to Sylvilagus, but pelage harsher ; ear, tail, and hind foot relatively shorter, the latter scantily haired. Skull much as in Sylvilagus, but postorbital process fused to side of cranium for its whole length, forming neither notch nor foramen. LYON] THE HARES AND THEIR ALLIES 407 Teeth essentially as in Lepus, except first lower premolar which has two or more reentrant angles on anterior face. Seu (pls. UXXvi, 3,0; LXXKVIM, Aj LXXXVIIL: LXXXIX,(6-10)— The genus Limmnolagus possesses postorbitals quite different in form and attachment from the postorbitals of any of the other genera of the Leporide. The whole process is fused to the side of the frontal bone, so that only a very minute notch is found anteriorly, and no notch, foramen, or slit is found pos- teriorly, except in rare and anomalous cases, in which a small fora- men may be formed by incomplete fusion of the postorbital with the side of the skull. The fused postorbital process has about the same general shape as has the unfused process in the genus Sylvilagus. An atypical specimen (No. 64,029, Kissimmee, Florida) shows the manner in which the process is attached. The hind-end of the process, instead of meeting the skull directly as it does in those genera where the posterior end of the process is in contact with the side of the skull, is met by an outgrowing process from the cranium. In this specimen a small foramen is enclosed between the posterior part of the postorbital process and the above outgrowing process from the cranium. Pirate LXXV (About five-sixths natural size) . Lepus (Pecilolagus) americanus dalli, No. 36,209, Alaska. 2. Lepus (Macrotolagus) alleni, No. 59,223, Sonora, Mexico. 3. Lepus (Lepus) europeus, No. 105,831, Switzerland. + PLaTE LXXVI (About five-sixths natural size) 1, 4. Lepus (Pecilolagus) americanus dalli, No. 36,209, Alaska. 2, 5. Oryctolagus cuniculus, No. 49,645, Germany. 3, 6. Limnolagus palustris, No. 2,089, Society Hill, South Carolina. 7, 8. Sylvilagus (Sylvilagus) floridanus transitionalis, No. 49,624, Monroe county, New York. 9, 10. Sylvilagus (Microlagus) bachmani, No. 44,416, San Luis Obispo, Cali- fornia. Pirate LXXVII (About five-sixths natural size) 1. Sylvilagus (Microlagus) bachmani, No. 44,416, San Luis Obispo, Cali- fornia. 2. Sylvilagus (Sylvilagus) floridanus transitionalis, No. 49,634, Monroe county, New York. 3. Pronolagus crassicaudatus, No. 22,972, South Africa. 4. Limnolagus palustris, No. 2,089, Society Hill, South Carolina. 5. Oryctolagus cuniculus, No. 49,645, Germany. Pirate LXXVIII (Natural size) ta, 1b, tc. Romerolagus nelsoni, No. 57,954, Mt. Popocatepetl, Mexico. 2a, 2b. Pronolagus crassicaudatus, No. 22,972, South Africa. 3. Brachylagus idahoensis, No. 93,696, Ione valley, Idaho. Pirate LXXIX (Natural size) 1. Brachylagus idahoensis, No. 93,696, Ione valley, Idaho. 2a, 2b, 2c. Pentalagus furnessi, No. 5,583 (Wistar Institute), Liu Kiu islands. 444 SMITHSONIAN MISCELLANEOUS COLLECTIONS [VOL. 45 Pirates LXXX, LXXXI (About three-fifths natural size) . Lepus (Lepus) othus, No. 15,878, St. Michaels, Alaska. . Lepus (Lepus) europeus, No. 1,857, Bavaria. . Lepus (Lepus) tschuktschorum, No. 86,459, Petropaulski, Kamschatka. Lepus (Lepus) variabilis, No. 37,137, Sweden. . Lepus (Lepus) campestris, No. 49,623, Kansas. . Lepus (Lepus) labradorius, No. 23,132, type, Labrador. . Lepus (Lepus) grenlandicus, No. 114,849, Bache peninsula, Greenland. . Lepus (Lepus) varronis, No. 105,834, Switzerland. ON ANAK wWN Puiates LXXXII, LXXXIII (About two-thirds natural size) 1. Lepus (Macrotolagus)alleni, No. 50,292, Sonoyta, Mexico. 2. Lepus (Macrotolagus) merriamt, No. 84,646, Fort Clark, Texas. 3. Lepus (Macrotolagus) asellus, No. 36,009, type, San Luis Potosi, Mexico. 4. Lepus (Macrotolagus) callotis, No. 35,678, Guadalajara, Mexico. 5. Lepus (Macrotolagus) californicus, No. 60,867, U. S. and Mexican bound- ary line at Pacific ocean. : 6. Lepus (Macrotolagus) texianus eremicus, No. 63,119, U. S. and Mexican boundary line, monument No. 77. 7. Lepus (Macrotolagus) texianus, No. 63,118, U. S. and Mexican boundary line, monument No. 64. 8. Lepus (Macrotolagus) gaillardi, No. 35,700, White Water, Chihuahua, Mexico. PiLates LXXXIV, LXXXV (About five-eighths natural size) . Lepus ruficaudatus, No. 38,039, Indie. . Lepus ochropus, No. 34,735, Kilima-njaro, Africa. Lepus hypsibius, No. 84,076, Rupshu, Ladak. Lepus tibetanus, No. 62,126, Ladak. Lepus yarkandensis, No. 62,132, Yarkand river, eastern Turkestan. Lepus (Pecilolagus) washingtonu, No. 6,265, Washington. Lepus (Pecilolagus) americanus dali, No. 36,209, Alaska. . Lepus (Pecilolagus) americanus struthopus, No. 23,123, Labrador. . Lepus (Pecilolagus) americanus virginianus, No. 101,844, Massachusetts. 10. Lepus (Pecilolagus) americanus macfarlani, No. 14,467, type, Fort An- derson, British America. NOH CON AA Ew ‘Oo Pirates LXXXVI, LXXXVII (Nine-fourteenths natural size) 1. Syivilagus (Sylvilagus) floridanus mallurus, No. 1,227, District of Co- lumbia. 2. Sylvilagus (Sylvilagus) floridanus yucatanicus, No. 37,772, type, Merida, Yucatan. 3. Sylvilagus (Sylvilagus) floridanus mearnsi, No. 22,400, Illinois. 4. Sylvilagus (Sylvilagus) truei, No. 112,787, Mexico. 5. Sylvilagus (Sylvilagus) margarite, No. 63,217, type, Margarita island. 6. Sy'vilagus (Sylvilagus) subcinctus, No. 35,663, Hacienda El Molino, Mexico. 7. Sylvilagus (Sylvilagus) brasiliensis gabbi, No. 37,794, type, Talamanca, Costa Rica. LYON | THE HARES AND THEIR ALLIES 445 8. Sylvilagus (Sylvilagus) paraguensis, No. 121,423, Sapucay, Paraguay. 9g. Sylvilagus (Sylvilagus) arizone, No. 58,863, Fort Lowell, Arizona. 10. Sylvilagus (Sylvilagus) auduboni, No. €0,921, Coast Range mountains, U. S. and Mexican boundary line. 11. Sylvilagus (Microlagus) cinerascens, No. 60,871, Tecate river, U. S. and Mexican boundary line. 12. Sylvilagus (Microlagus) bachmani, No. 35,131, Nicasio, California. 13. Sylvilagus (Microlagus) bachmani ubericolor, No. 35,371, type, Beaver- ton, Oregon. Pirates LXXXVIII, LXXXIX (About five-eighths natural size) . Oryctolagus cuniculus, domestic, Belgian hare, No. 105,402. . Oryctolagus cuniculus, common domestic, No. 62,704. . Oryctolagus cuniculus, common domestic, No. 36,831. . Oryctolagus cuniculus, wild, No. 3,124, England. . Oryctolagus cuniculus, wild, No. 49,645, Germany. . Limnolagus paludicola, No. 64,029, Kissimmee, Florida. . Limnolagus palustris, No. 2,089, Society Hill, South Carolina. . Limnolagus aquaticus, No. 3,779, Warren county, Mississippi. . Limnolagus aquaticus, No. 84,109, Mississippi. o. Limnolagus attwatert, No. 97,031, Richmond, Texas. HO ON AM BW ND H PLATE XC (Natural size) 1. Ochotona (Ochotona) ladacensis, No. 84,063, Ladak, central Asia. 2. Ochotona (Pika) princeps, No. 66,678, Nelson, British Columbia. 3. Ochotona (Conothoa) roylei, No. 36,814, Baltistan. Pirate XCI (Enlarged three times) Lower AND Upper CHEEK TEETH, RIGHT SIDE 1. Brachylagus idahoensis, No. 23,104, Goose lake, California. . Ochotona ladacensis, No. 84,062, Ladak, central Asia. . Romerolagus nelsoni, No. 57,954, Mt. Popocatepetl, Mexico. . Limnolagus paludicola, No. 64,029, Kissimmee, Florida. . Oryctolagus cuniculus, No. 49,648, England. . Lepus campestris, No. 61,367, Madison, Minnesota. . Pentalagus furnessi, No. 5,583 (Wistar Institute), Liu Kiu islands. . Pronolagus crassicaudatus, No. 22,972, South Africa. . Sylvilagus floridanus, No. 49,588, Florida. © ON AM SW N Prate XCII (Natural size except Figure 1, which is enlarged twice) VENTRAL VIEWS OF CERVICAL VERTEBRE 1. Ochotona saxatilis, No. 91,188, Cabinet mountains, Idaho. 2. Romerolagus nelsoni, No. 57,954, Mt. Popocatepetl, Mexico. 3. Brachylagus idahoensis, No. 93,595, lone valley, Idaho. 4. Limnolagus paludicola, No. 49,580, Florida. 5. Sylvilagus floridanus, No. 49,587, Florida. 6. Oryctolagus cuniculus, No. 49,645, Germany. 7. Lepus (Lepus) campestris, No. 49,623, Kansas. 8. Lepus (Peacilolagus) americanus virginianus, No. 969, New York. 9. Lepus (Macrotolagus) texianus, No. 94,198, Newark valley, Nevada. 10. Pronolagus crassicaudatus, No. 22,972, South Africa. 446 SMITHSONIAN MISCELLANEOUS COLLECTIONS [voL. 45 Pirate XCIII (Natural size) LUMBAR VERTEBRA, SEEN FROM THE SIDE. 1. Lepus (Lepus) campestris, No. 49,622, Nebraska. . Lepus (Pecilolagus) americanus virginianus, No. 969, New York. . Pronolagus crassicaudatus No. 22,972, South Africa. . Oryctolagus cuniculus, No. 40,648, England. . Oryctolagus cuniculus, lop-ear, domestic, No. 49,386. in & WwW WN PLATE XCIV (Natural size, except 9 and to enlarged twice) Riss AND LUMBAR VERTEBRZ 1. Lepus texianus, No. 04,198, Newark valley, Nevada. . Lepus campestris, No. 49,623, Kansas. . Pronolagus crassicaudatus, No. 22,972, South Africa. . Orvctolagus cuniculus, No. 49,645, Germany. . Sylvilagus floridanus mallurus, No. 89,514, Four-mile Run, Virginia. . Limnolagus palustris, No. 49,580, Florida. . Brachylagus idahoensis, No. 93,595, lone valley, Idaho. . Romerolagus nelsoni, No. 57,954, Mt. Popocatepetl, Mexico. 9g and 10. Ochotona saxatilis, No. 91,188, Cabinet mountains, Idaho. 11. Romerolagus nelson, No. 57,954, Mt. Popocatepetl, Mexico. 12. Brachylagus idahoensis, No. 93,595, lone valley, Idaho. 13. Limnolagus paludicola, No. 49,580, Florida. 14. Sylvilagus floridanus mallurus, No. 89,514, Four-mile Run, Virginia. COON AM WwW N Pirate XCV (Natural size) STERNA AND COSTAL CARTILAGES, VENTRAL VIEW. 1. Brachylagus idahoensis, No. 93,595, lone valley, Idaho. . Sylvilagus floridanus, No. 49,587, Florida—clavicles shown. . Lepus (Macrotolagus) texianus, No. 94,198, Newark valley, Nevada. . Lepus (Lepus) campestris, No. 49,622, Nebraska. . Oryctolagus cuniculus, No. 49,645, Germany—clavicles shown. wk WwW bd PLATE XCVI STERNA AND COSTAL CARTILAGES, VENTRAL VIEW. 1. Romerolagus nelsoni, No. 57,954, Mt. Popocatepetl, Mexico—clavicles shown. Natural size. 2. Ochotona saxatilis, No. 91,188, Cabinet mountains, Idaho—clavicles shown. Twice natural size. 3. Pronolagus crassicaudatus, No. 22,972, South Africa, natural size. 4. Nesolagus netscheri, copied from PI. 39, fig. 18, Major ‘99. 5. Limnolagus paludicola, No. 49,580, Florida, natural se. PLATE XCVII (Natural size) SCAPULZ. . Lepus (Lepus) campestris, No. 49,623, Kansas. . Lepus (Macrotolagus) texianus, No. 94,198, Newark valley, Idaho. Oryctolagus cuniculus, No. 49,648, England. Pronolagus crassicaudatus, No. 22,972, South Africa. Ochotona saxatilis, No. 91,188, Cabinet mountains, Idaho. . Limnolagus paludicola, No. 49,580, Florida. NOH Ayub ow LYON] THE HARES AND THEIR ALLIES 447 7. Sylvilagus floridanus, No. 49,588, Florida. 8. Brachylagus idahoensis, No. 93,696, Ione valley, Idaho. 9. Romerolagus nelsoni, No. 57,954, Mt. Popocatepetl, Mexico. Pirate XCVIII (Natural size) BoNEs OF THE ForE ArM 1. Lepus (Lepus) campestris, No. 49,623, Kansas. . Lepus (Macrotolagus) texianus, No. 94,198, Newark valley, Nevada. . Lepus (Pecilolagus) americanus virginianus, No. 969, New York. . Sylvilagus floridanus, No. 49,586, Florida. . Limnolagus paludicola, No. 49,580, Florida. . Brachylagus idahoensis, No. 93,696, Ione valley, Idaho. . Pronolagus crassicaudatus, No. 22,972, South Africa. . Oryctolagus cuniculus, No. 49,648, England. 9. Romerolagus nelsoni, No. 57,954, Mt. Popocatepetl, Mexico. 10. Pentalagus furnessi, No. 5,583 (Wistar Institute), Liu Kiu islands. 11. Ochotona saxatilis, No. 91,188, Cabinet mountains, Idaho. ON An BW DN PLATE XCIX (Natural size) BONES OF THE LEG 1. Lepus campestris, No. 49,622, Nebraska. 2. Oryctolagus cuniculus, No. 49,645, Germany. 3. Pronolagus crassicaudatus, No. 22,972, South Africa. 4. Pentalagus furnessi, No. 5,583 (Wistar Institute), Liu Kiu islands. 5. Sylvilagus floridanus, No. 49,586, Florida. 6. Ochotona saxatilis, No. 91,188, Cabinet mountains, Idaho. 7. Ochotona ladacensis, No. 49,500, Ladak, central Asia. 8. Brachylagus idahoensis, No. 93,696, Ione valley, Idaho. 9. Romerolagus nelsoni, No. 57,954, Mt. Popocatepetl, Mexico. 10. Limnolagus paludicola, No. 49,580, Florida. PLATE C (Natural size) Hinp Feet 1. Ochotona ladacensis, No. 49,500, Ladak, central Asia. . Romerolagus nelsoni, No. 57,954, Mt. Popocatepetl, Mexico. . Brachylagus idahoensis, No. 93,696, Ione valley, Idaho. . Pentalagus furnessi, No. 5,583 (Wistar Institute), Liu Kiu islands. . Limnolagus paludicola, No. 49,580, Florida. . Lepus campestris, No. 49,623, Nebraska. . Pronolagus crassicaudatus, No. 22,972, South Africa. . Oryctolagus cuniculus, No. 49,645, Germany. . Sylvilagus floridanus, No. 49,586, Florida. 0 ON AMF WwW DN NOTES PALEONTOLOGICAL COLLECTIONS FROM EUROPE Mr. Charles Schuchert, Assistant Curator of the Division of Stratigraphic Paleontology in the National Museum, left Washing- ton, June 24, 1903, for an extended trip through Europe, return- ing November 2. His object was to study in the field and in the museums the Ordovicic, Siluric, and Devonic faunas and their stratigraphic sequence, to obtain large and instructive materials for the Louisiana Purchase Exposition, and to represent the National Museum at the Ninth International Geological Congress held at Vienna during the latter part of August. He spent six days in collecting Siluric fossils on Gotland, an island province of Sweden replete with well-preserved fossils. Nearly all the localities are along the sea-shore and the cliffs known as the Klint. The late Professor Lindstrém’s collector, Mr. Anton Florin, assisted in gathering the fossils. Mr. Schuchert next devoted sixteen days to a study of the mu- seums in St. Petersburg and to making Ordovicic collections in the Baltic region west of Reval. The preservation of these fossils is excellent and collecting here is as good as about Cincinnati. Nearly all the localities in Estland are on baronial estates, and every facil- ity was extended by the owners for making the collecting both easy and profitable. During this time Mr. Schuchert was directed and instructed by Professor Dr. Frederick von Schmidt of the Imperial Academy, and by Dr. August von Mickwitz of Reval. Both of these gentlemen also presented much fine material, so that a repre- sentative collection of the varied trilobites, cephalopods, brachiopods, and cystids are now in the National Museum. From Reval Mr. Schuchert traveled to Berlin, Dresden, and Prag, where the various natural-history museums were visited and the paleontologic collections examined. With Dr. Jaroslav Perner a day was spent in the Moldau valley, above Prag, in examining a number of Barrande’s localities for fossils of the Siluric and De- vonic strata. Another day was spent in collecting Konieprusian fossils at Barrande’s famous locality, Zlaty Kun, near Beraun, and some good material was purchased of Mr. Fritsch at Prag. At Munich two days were spent in an examination of the great col- lections made under the direction of the late Professor Zittel. 448 SMITHSONIAN MISCELLANEOUS COLLECTIONS = a = © 4 &c fx nH iS) t gq ~ fe BG Co ~~ NN OH tf aS n v a = ~ 4 “ 10 = a o N WN ~~ = 3» RIS FROM GERMANY; : TACRINUS SUBANGUL 1,2, PEN KANSAS CITY, MISSOURI. NOTES 449 Another day was spent at Eichstatt, Bavaria, in examining the “ Solenhofen ” quarries, from which region so many interesting Jurassic fossils have been procured. More than one hundred speci- mens were purchased of Mr. Fritz Ehrensberger, among which are some large cuttle-fish bones or pens of Leptoteuthis gigas, one of which is 38 inches and another 24 inches in length. Arrangements also were made for the purchase of several large fishes and a fine Rhamphorhynchus, showing both wing and tail membranes. Twelve days were spent in collecting Devonic fossils in the Eifel, around Coblenz on the Rhine, and in Hessen-Nassau. In the first and last-named places Mr. Schuchert had as his instructor Professor Dr. E. Kayser of Marburg, and about Coblenz, Dr. O. Follmann, Although the season was very wet and collecting con- sequently much interrupted, a representative collection of Middle Devonic fossils of the Calceola shales was obtained. At Bonn the large natural-history establishment of Dr. F. Krantz was visited and a number of interesting fossils purchased. Two of these are large Lias Holzmaden crinoids, Pentacrinus subangu- laris, illustrated in plate ci. A Cerithiim gigantewm, more than 24 inches in length, was also procured. From another dealer and paleontologist, Dr. B. Sttirtz, an excellent series of Lower Devonic starfishes and crinoids was obtained, making the National Museum collection of these rarities second only to that of the British Mu- seum. Two days were spent in collecting Cretaceous bryozoa about Maastricht, in southern Holland, a locality prolific in these fossils. At Brussels the splendid paleontologic collections of the Royal Mu- seum were examined, chief among which is the dinosaur /guanodon bernissartensis, to which an entire room is devoted. In this museum is also the De Koninck collection of Carboniferous invertebrates. A day each was given to examining the Carboniferous stratigraphy about Visé and Tournai, and small collections were procured from local dealers. At London considerable time was devoted to examining the great and well-exhibited fossil collections of the British Museum of Nat- ural History, and also the collection of the Geological Survey. From Robert Damon, of Weymouth, England, was purchased a lot of fine fossils, one of which is an excellent Jurassic slab of Trigonia, while another slab is replete with crinoid heads of Penta- crinus fossilis and is illustrated in plate c1t. 450 SMITHSONIAN MISCELLANEOUS COLLECTIONS [VOL. 45 As a result of this trip sixteen boxes of European fossils have arrived at the National Museum—a good nucleus for comparative studies with the American faunas. A NOTEWORTHY CRINOID Frank Springer, Esq., of East Las Vegas, New Mexico, recently presented to the United States National Museum a slab, nearly five feet square, covered with the Uinta crinoid (Uintacrinus socialis GRINNELL). This slab is a piece of one mass originally fifty feet long by twenty feet wide, and was collected by Mr. Springer in the upper part of the Niobrara foraminiferal chalk (Upper Creta- ceous), near Elkader, Logan county, Kansas. The slab shown in plate cri has about 140 bodies or crowns of this crinoid, but orig- inally there were several times as many, because the slab, three- eighths of an inch in thickness, is composed of them. This ten- rayed crinoid was.a free floating species, and at times probably held itself to foreign objects by its long, slender arms. One specimen on this slab preserves several complete arms having a length of 36 inches, but the largest individuals are estimated to have had an arm-spread of 8 feet 4 inches, “a size far exceeding that of any other known crinoid, recent or fossil.” CHARLES SCHUCHERT. SHIELDS FROM WESTERN SUMATRA Dr. W. L. Abbott has recently sent to the National Museum a number of wooden shields from Nias and Batu islands, lying off the western coast of Sumatra. Each of these shields is carved from a single piece of the strong light wood of the oil tree (Diptero- carpus), and consists of an elliptic body, concavo-convex in trans- verse section, with projecting club-like ends. The shield is strength- ened by a midrib on the exterior, which is expanded into a boss in which the hand-grip is excavated. The thin wings on each side of the midrib are strengthened at intervals with a horizontal sewing, of bark cord or rattan strips, through perforations. This sewing is done with two pairs of cords parted to each side at the start; each pair is then twisted between the perforations, one free end from each pair is passed through the hole in opposite directions, and twisted with the remaining cord as before. This results in what seems to be a continuous two-strand cord extending evenly on both sides of the shield, the holes being hidden. By this sewing the thin wings of the shield are made very strong with but little increase (3922 xb 9z1IS) ‘ANVIONG ‘SIDAUM ANAT WOU S77/SSOY SANTIYIVINGT PLT neTt ac siveT OVA SNOILOA1107) SNOANVTTHOSIP, NVINOSHLINS SMITHSONIAN MiscELLANgous CoLLecrions Vou. 45, PL. CII - 45, PL. UINTACRINUS SOCIALIS FROM KANSAS. (Size nearly 5x 5 feet.) NOTES 451 of weight. This treatment has not been previously described. Some- times bands of rattan are passed across the faces of the shield and held in place by a cross-sewing through holes. On the upper pro- jection of the shield, the midrib, and the edges, thin wedge-shaped pieces of iron have been inserted as a protection against cutting weapons. The Nias shield shows plainly the development of this defensive weapon from the club, and it is evidently still mainly used for parrying, though it has widened out under the influence of the spear and of cut-and-thrust weapons. The solid projecting ends and the midrib preserve the vestiges of the club, from which it has been derived. The average Nias shield measures 41 inches in length and 9% inches in width. WALTER HouGH. EARTHQUAKE IN PERU The Department of State has courteously sent to the Smithsonian Institution the following despatch from United States Consul A. L. M. Gottschalk, dated Callao, Peru, March 5, 1904: “T have the honor to report that an earthquake-shock of much severity was felt in Lima and Callao on Friday, March 4, at about 5:20 A.M. Oscillation was from southwest to northeast; duration variously estimated at from 30 to 50 seconds. Considerable dam- age of the minor sort was done to public edifices, churches chiefly, in Lima and Callao, in the way of broken statues, fallen cornices, etc. One of the towers of the cathedral of Lima is said to be badly cracked and may have to be torn down. During the vibrations church bells were tolled and for a few moments general consterna- tion prevailed. I believe that only one life was lost, though a num- ber of persons were more or less injured by falling plaster, etc. -No house is without broken crockery and bric-a-brac, and at the Con- sulate at Callao all the furniture, including an iron safe, was shifted about an inch. The shock was preceded, at 12 M., by a seismic movement so slight as to be barely perceptible. “ At Callao greater damage was done owing to the ramshackle nature of buildings, a number of which (small dwellings) collapsed. There was no appreciable damage to wharves or shipping. Pas- sengers aboard the S. S. Mexico entering port state that the shock was felt by them ‘as if the ship had suddenly struck a rock.’ “ At Chorrillos, a fashionable coast resort near by, some damage was done to houses and to the sea wall. President Candamo’s villa there was in part destroyed during the family’s occupancy. 452 SMITHSONIAN MISCELLANEOUS COLLECTIONS [VOL. 45 “ Canete and Chincha to the south, and all the ports north of us as far as Trujillo, report feeling the shock severely. News is not at present obtainable from Cerro de Pasco, the railway having suffered from land-slides. ‘No tidal wave accompanied the shock here. “ The earthquake of March 4, 1904, is generally admitted to have been the severest felt in Peru since 1868. The last phenomenon of the kind experienced here was on September 20, 1897, and was of a much lighter nature.” A RECENT VISIT TO THE COFFEE MOUNTAINS OF LIBERIA The following extract from a despatch, dated March 2, 1904, from Consul General Ernest Lyon at Monrovia, Liberia, has been officially communicated to the Smithsonian Institution by the De- partment of State: “We left Carysburg Sunday afternoon at 3:30 for the Coffee mountains, a place of interest to all tourists who come to Liberia. The distance to the principal mountain in the range is about twenty- five miles. The range received its name because the aborigines found coffee growing on its summits. The path to the mountains lies through the dense forest which abounds in rich timbers and interesting flora. The timber consists of a variety of valuable woods, some of which are the following: red and gray mahogany, four varieties of oak, cedar, rosewood, mangrove, whitegum, burr- wood, mulberry, sassawood, brimstone, redpeach, pepperwood, white-mangrove, persimmon, ironwood, poplar, and greasypeach. If it will be of interest to the department, this office will furnish samples of some of the valuable woods. “The flora is not only interesting but possesses value for me- dicinal purposes. The forest for one hundred miles whither I have been is lined with a variety of these valuable herbs. There is scarcely a leaf or berry or root, according to the intelligent native medicine-man who was paid to accompany us, which cannot be utilized as a medicine for the diseases common to Africa; for in- stance, the corsa is a species of leaves which may be used as a purgative or for the cure of ulcers; the corsa is used also for weak- ness and imbecility as well as for a general stimulant; and there are many others, as the Xmas-leaf, cherrybark, alum, senna, etc., each of which has its medicinal use. “The next day at If A.M. we reached the foot of the principal mountain. It was impossible to reach the top, but after a climb NOTES 453 of two hours’ duration we reached a portion of the summit. From this lofty peak, several hundred feet above the level of the sea, we could see the capital city, Monrovia, and many of the towns on the St. Paul river, lying apparently at its base. The gigantic size of the trees so. impressed us that we took actual measurement of one of them, which we found to be from 25 to 35 feet in circumference, and we judge the trunk extended without a limb from 100 to 125 feet high. This abundance of timber suggests the possible wealth of the lumber industry in this country. Strange this mountain is still unexplored, and its summit is still the home of wild and dan- gerous animals. Here also is the center of the hopes of the gold concessionist. After many experiments they have been led to be- lieve that the vein which will ultimately reward their endeavors lies somewhere around these mountains. A number of mining ex- perts are now there attempting to locate this vein.” THE Scotia ANTARCTIC EXPEDITION Through the courtesy of the Honorable, the Secretary of State, the Smithsonian Institution is enabled to present the following brief but interesting report on the progress of the Scotia Antarctic Expe- dition, which has been submitted by the Honorable John Barrett, United States Minister to Argentina: “In view of its probable interest in scientific circles, I have the honor to submit the following brief report of the trip of the Scotia Scientific Expedition to the Antarctic regions, the data for which have just been given me by the leader of the expedition, Mr. W. S. Bruce, F.R.S.E. “The expedition was organized by public subscription in Scot- land and received very material assistance from all the leading Scot- tish scientific societies, especially the Royal Scottish Geographical Society, the Royal Society of Edinburgh, and the Perthshire Natural Science Society, as well as from several departments of the British Government which supplied some valuable instruments and charts. “The object of the expedition was a study of meteorology and oceanography, both physical and biological, in the Antarctic regions. The expedition was also to lend assistance, in case of necessity, to Dr. Nordenskj6ld of the Swedish expedition, but no assistance was necessary in view of the fact that relief had already been sent to the Antarctic by the Argentine and Swedish Governments. “ The personnel of the expedition was as follows: Mr. W.S. Bruce, F.R.S.G.S., Leader; Mr. R. C. Mossman, F.R.S.E., Meteorologist ; 454 SMITHSONIAN MISCELLANEOUS COLLECTIONS [VOL. 45 Dr. J. H. H. Pirie, Surgeon and Geologist; Dr. R. N. R. Brown, Botanist; Mr. D. W. Wilton, Zoologist; Mr. Allister Ross, Tax- idermist; Mr. W. A. Cuthbertson, Artist. “The Scotia left Scotland on November 2, 1902, and, after stops at Madeira and the Falklands for coal, met the ice on February 2, 1903, in 59° S. lat., 40° long. W. (approximate). She then visited the South Orkney islands and thence made a cruise of about 5,000 miles to the south and east between longitude 16° W. and 45° W. and as far south as 70° 25’. In this region a deep sea was found of an almost uniform depth of 2,500 fathoms, the deepest sounding being 2,739 fathoms. Trawls were lowered several times in these depths and brought up fishes, star-fishes, sea-lilies, sea-cucumbers, sponges, crustaceans, shell-fish, and smaller organisms. Tempera- tures, salinities, currents, and the color of the sea were observed at all depths from the surface to the bottom. “This cruise ended on March 25, 1903, the Scotia having covered about 4,000 miles of previously unexplored sea. The ship anchored in a bay in Laurie island, South Orkneys, since named Scotia bay, and before the end of the month was beset and frozen in for the winter, not being released until November 24, 1903. During this time a house was built and a meteorological observatory (first class) was established, as well as a magnetic observatory. Throughout the winter baited sea-traps were set through a hole in the ice which secured a large number of zoological specimens and no less than 2,000 pounds of fish for food. Every day, except when severe weather prevented, a dredge was dragged by means of a continuous rope and two holes cut through the ice and cleared from day to day. Four expeditions were made during the spring, three with sledges and one with boats for purposes of hydrographic and land survey. As a result of these expeditions and of survey work in the vicinity of the ship a 25-inches-to-the-mile map of Scotia bay and a 2- inches-to-the-mile map of the rest of Laurie island and the adjacent islands were prepared. During this local survey about 500 sound- ings were taken—almost every one of them involving cutting through ice from 3 to 5 feet thick. With the exception of the chief engineer, who died of heart- disease, all the members of the expedition and crew enjoyed excel- lent health. Mr. Bruce attributes this largely to the fact that, in preference to preserved meats, penguins and fish and, in the spring, penguin eggs in liberal quantities were eaten. “The Scotia left the South Orkneys on November 27, 1903, and arrived at Port Stanley December 2 and at Buenos Aires December NOTES 455 24. While she was in dry-dock here [ visited her, together with Mr. Ames, the Secretary of Legation, by invitation of Mr. Bruce and of her commander, Captain Robertson, and found her stoutly built, splendidly equipped, and full to overflowing of various scien- tific specimens. “ Meteorological and magnetic observations have been continued at Scotia bay under the superintendence of Mr. Mossman. Mr. Bruce has offered to the Argentine government the use of this station and his supplies there and passage thither by the Scotia for four scientists if the government will maintain the station dur- ing the coming year and undertake the relief of those stationed there at the end of that time. “The Scotia will leave for the South Orkneys on or about the 15th instant and, after a few days at Scotia bay, will proceed to the south and east for further deep-sea explorations, returning, in all probability, by way of Cape Town and reaching Scotland toward the end of June.” RECENT PUBLICATIONS OF THE SMITHSONIAN INSTITUTION CoNTINUED FROM LIST OF JANUARY, 1904, IN PUBLICATION No. 1419 No. Title. Series. Price. 1442 Lanctey, S. P. James Smithson.. es as .10 1443 PETTIGREW, J. B. Various modes of. flight. . = Re 1867 05 1444 ALDRICH, a. VI. Catalogue of N. A. Diptera Un ROSS) Meee eres 65.5 eee bases Neale ne eee en ees M.C. XLVI 1445 Smithsonian Miscellaneous Collections (Quar- terly Issue), vol. xLv, Parts 3 and 4 (containing TAA Oe IAGO) Men Son tory ees sec charters = ote SP M.C. XLV 1.00 1446 LANGLEY, S. P. Personal ene in transit observations (Quarterly Issue) . NEG ex V: 05 1447 JorDAN, D. S., and SNyover, J. O. ‘Owston’s col: lection of Japanese Fishes (Quarterly Issue)... M.C. XLV my 1448 JorDAN, D. S., and Starks, E. C. New cyprinoid fish from China (Quarterly Issue). MC y Ie: 05 1449 Lanctey, S. P. Removal of remains OF. James Smithson (Quarterly Issue). MiG. 2a .05 1450 SmitH, H. M. Breeding habits of yellow bellied terrapin (Quarterly Issue) . MN@y XUV) .05 1451 BEAN, B. A. New pelican fish from the Pacific (Quarterly I. ase oetactend hickiae cron ltoment Gace M.C. XLV 05 1452 Utricu, E. O., and Bassier, R. S. Revision of Paleozoic bryozoa, Part 1 (Quarterly Issue)... M.C. XLV .20 1453 Girt, THroporeE. Remarkable genus of fishes. ite WUmbras (Ouarterly Issue). ..302....02-2.. MC, XLV .05 1454 PHaten, W. C. Unakite in Virginia (Quarterly HRCA EAMES pes toch whe ANG ciche aus cco gs aust Shuey pet wales ae aS 1455 Lucas, F. A. Dinosaur Trachodon Annectens 7 (Quarterly Issue). Eee EG aE .05 1456 Lyon, M. W., Jr. Classification. of hares and their allies (Quarterly Issue) . Pee cate ae ey LEN 85 1457 Smithsonion Miscellaneous Collections, “vol. XLV vee (containing 1419 and 1445). Le Pe ae ee a ee ne SOLEAV, 2.00 456 SMITHSONIAN MISCELLANEOUS COLLECTIONS [VOL. 45 No. Title. Series. Price. 1458 Powett, J. W. Twentieth Report, Bureau of { 2.00 paper’ AMenicam thiol OowaElCOSnOO Mae aneneniene ie (2.15 cloth 1459 Draper, HENRY, and RitcHEy, G. W. Silvered glass telescopes (/mepress) nen sno ee iene ees COO lay 1460 International Exchanges. Brussels convention of 1886, proclaimed by United States, 1889........ 05 1461 Classified list of Smithsonian publications avail- able > April i OOA se eg: eye eae A iran ep erence es 05 ‘For sale by Superintendent of Documents, Government Printing Office, Washington, D. C INDEX Abbot, C. G. Recent studies of the solar constant of radiation 74 The new ceelostat and horizontal telescope of the Astrophysical Observatory 84 Abbott, W. L., Malayan mammals col- lected by 1 Sumatra shields presented by 450 | abbreviatus, Gonorhynchus 236 Abra 325 Acestrorhamphus 146 Acestrorhynchus 146 Acnodon 147 acrales, Pseudotriakis 232 adangensis, Sciurus 17 agellus, Ptychocladia 290 Aggregatum 282 Akerite 309 Allonema 266, 279 botelloides 281 botellus 281 fusiforme 284 minimum 284 moniliforme 282 subfusiforme 283 waldronense 283 altinsularis, Sciurus 21 Amethyst group in National Museum 218 Amulets, stone, from West Indies 126 Angstrom, K., cited 75 Anisitsia 144 annectens, Trachodon 317 Antarctic expedition of the Scotia 453 antiqua, Rhopalonaria 269 aoris, Galeopithecus 47 Sciurus 10 Archeological trip to West Indies 112 Ardea herodias 111 Ardetta exilis 110 Ascodictyon 266, 285 filiforme 288 floreale 288 parvulum 288 siluriense 286 sparsum 289 stellatum 287 youngi 289 Ascodictyonide 266, 285 Astrophysical Observatory, new ccelo- stat and horizontal telescope of 84 observations of, on solar constant of radiation 74 Ateleopide 239 457 Ateleopus japonicus 239 Atherura zygomatica 42 atratus, Sciurus 13 attenuata, Rhopalonaria 268 Audital bull of leporide 346 Baker, Arthur B. A notable success in the breeding of black bears 175 bale, Mus 33 Ratufa 6 Sciurus 14 Balenoptera physalus got Bartsch, Paul. Notes on the herons of the District of Columbia 104 Basalt from Greenland 198, 202 Bassler, R. S. See Ulrich, E. O., and Bassler. batuana, Petaurista 27 batuanus, Presbytes 65 Tragulus 2 Bean, Barton A. A new pelican fish from the Pacific 254 Bears, black, success in breeding of 175 Beckwith, Paul, on the sword 220 Bell, A. Graham, remarks on removal of Smithson’s remains 248, 250 bentincanus, Mus 38 Sciurus 19 Bibliography of hares and their allies 440 bilimitatus, Sciurus 8 Bishop, William H., letter from, on removal of Smithson’s remains 246 remarks on removal of Smithson’s remains 247 Bone-beds at Kimmswick, Missouri 219 Bone-carvings from West Indies 131 Botaurus lentiginosus r1o botelloides, Allonema 281 botellus, Allonema 281 Boulengerella 147 Brachanchenias lucasi 96 Brachylagus 325, 411 Breeding of black bears, success in 175 Bryconodon 146 Bryozoa, Paleozoic, revision of 256 Burnside, Helen Waldo. The Hodg- kins Fund of the Smithsonian Insti- tution 168 Butorides virescens 110 candidissima, Egretta 111 caninus, Gobius 295 capillare, Heteronema 278 capillaris, Rhopalonaria 271 458 Caprolagus 325, 426 carbonarium, Heteronema 279 Carpolagus 326 Carpus of Leporide 378 of Ochotonide 379 Carvings from West Indies 128, 131 casensis, Mus 38 Sciurus 20 castanea, Tupaia 54 Caudal vertebre of Leporide 363 of Ochotonide 367 Cave exploration in West Indies 114 Cerianthus 139 cervicalis, Tupaia 59 Cervical vertebre of Leporide 356 of Ochotonide 365 Cestoplana 136 Child, C. M. Form-regulation in Ccel- entera and Turbellaria 134 Chimera purpurascens 235 Chimeride 235 China, new cyprinoid fish from 241 Chinese medicine 180 Chionobates 326 Chiropodomys niadis 40 Cheek teeth of Ochotonide 355 chrysogaster, Tupaia 58 Claosaurus 317 Clavicle of Leporide 373 of Ochotonide 374 Ceelentera, form-regulation in 134 Cceelostat of the Astrophysical Observa- tory 84 cerulea, Florida 111 Coffee mountains, Liberia, visit to 452 Colossoma 148 concolor, Simias 67 conferta, Vinella radiciformis 275 Conothoa, detailed account of 438 conspicua, Ratufa 5 contextum, Heteronema 279 crinoid, a noteworthy 450 Ctenostomata 256, 258, 266 Cuniculus 326 Cyprinoid fish, description of new 241 Dinosaur trachodon annectens 317 Diorite from Greenland 187, 192, 210 District of Columbia, notes on herons of 104 domelicus, Mus 39 Sciurus 18 Duplicidentata, keys to families, genera, and subgenera of 386 list of existing species of 334 skeleton and teeth of 337 skull of 337 Earthquake in Peru 451 eastmani, Pristiurus 230 egretta, Herodias 111 candidissima 111 SMITHSONIAN MISCELLANEOUS COLLECTIONS Eigenmann, Carl H. New genera of South American fresh-water fishes 144 Epidosite 314 Eulagos 326 Europe, paleontological collections from 448 Evermannella 146 exilis, Ardetta 110 Farrington, O. C., on extraction of glacial pothole 100 Femur of Leporide 382 Ochotonide 382 Fenestration of maxille of leporide 346 ferox, Plagyodus 235 Fewkes, J. Walter. Preliminary report on an archeological trip to the West Indies 112 Fibula of Leporide 382 of Ochotonide 383 filiforme, Ascodictyon 288 Fish, cyprinoid, description of a new 241 Fishes, collection of, from deep waters of Japan 230 new genera of, from South Amer- ica 144 the umbras, a remarkable genus of 295 Flint, James M. Chinese medicine 180 floreale, Ascodictyon 288 Florida ceerulea 111 foliaceus, Lepidorhinus 235 Fontaine, W. M., unakite from 306 Form-regulation in Celentera and Tur- bellaria 134 Fowke, Gerard. Excavations at Lan- sing, Kansas 219 Frye, William P., remarks on removal of Smithson’s remains 251 Furambulus obscurus 23 peninsule 25 rostratus 24 fusiforme, Allonema 284 specimen Galeopithecus aoris 47 gracilis 49 natune 50 pumilus 46 saturatus 51 tuancus 53 Gastrostomus pacificus 254 geminorum, Pteropus 60 Gempylide 236 Geographical distribution of Leporide 439 of Ochotonide 440 Geological collecting expedition 217 Gilbertella 147 gilbiventer, Mus 35 INDEX Gill, Theodore. A remarkable genus of fishes—the umbras 205 Glacial pothole in National 100 glauca, Isuropsis 234 Gneiss from Greenland 184 Gobius caninus 2095 Gonorhynchide 236 Gonorhynchus abbreviatus 236 gracilis, Galeopithecus 49 Granite from Greenland 186, 2009 Greenland, rocks of Nugsuaks penin- sula 183 Museum Hares and their allies, classification of 321 Hargitt, C. W., occupant of Smith- sonian seat at Naples 215 Headdresses, Korean, in National Mu- seum 149 Hemibarbus joiteni 241 Hemigale minor 43 herodias, Ardea 111 egretta III Herons of the District of Columbia 104 Hesperornis, skeleton of 95 Heteronema 266, 278 capillare 278 carbonarium 279 contextum 279 Hodgkins fund of Smithsonian Insti- tution 168 Holmes, W. H. Aboriginal quarries examined by 220 Examination of fossil bone-beds at Kimmswick, Missouri 219 Excavations by, at Lansing, Kan- sas 219 Notes on aboriginal iron mine in Missouri 218 Shell ornaments and Mexico 97 Holoprion 145 Holopristis 145 Holoshesthes 144 Hornblende from Greenland 192 Hough, Walter. Shields from western Sumatra 450 Humerus of Leporide 375 of Ochotonide 376 Hydrolagus 326 Hypersthene akerite 309 ictericus, Sciurus 12 Idols from West Indies 120, 129 Incisors of Leporide 349, 350 of Ochotonide 355 insignis, Ratufa 4 from Kentucky International Catalogue of Scientific Literature 213 International Exchanges, increase in transmissions of 218 Interparietal bone of Leporide 340 459 Iron mine, aboriginal, in Missouri 218 ishikawe, Trachypterus 240 Isuropsis glauca 234 Japan, deep-water fishes from 230 japonicus, Ateleopus 239 Jenings, F. H. Korean headdresses in the National Museum 149 joiteni, Hemibarbus 241 Jordan) DES sand scnyder in OseOn a collection of fishes from Japan 230 Jordan, D: Ss and Starks}, E.G) De= scription of new cyprinoid fish from China 241 julianus, Mus 34 Julis musume 236 Kaersut from Greenland 194 Kansas, excavations at Lansing 219 Kentucky, shell ornaments from 97 keokukensis, Rhopalonaria 272 klossii, Symphalangus 70 Korean headdresses in National Mu- seum 149 krameri, Umbra 304 Labride 236 Lagomys 327 Lagopsis 327 Lagopsys 327 Lagos 327 Lahilliella 144 Lamnide 234 lancavensis, Sciurus 16 Langley, S. P. Method of avoiding personal equation in transit ob- servations 225 Removal of remains of James Smithson 243 Lees, Noel, letter from, concerning Smithson’s grave 247 Remarks on removal of Smithson’s remains 249 Le Mesurier, E. A., letter from, on re- moval of Smithson’s remains 244 lentiginosus, Botaurus r1o Lepidorhinus foliaceus 235 Leporide, audital bulle of 346 carpus of 378 caudal vertebre of 363 cervical vertebre of 356 clavicle of 373 femur of 382 fenestration of maxilla of 346 fibula of 382 genera and subgenera of 389 geographical distribution of 439 humerus of 375 incisors of 349, 350 interparietal bone of 340 lumbar vertebrae of 359 460 SMITHSONIAN Leporide, mandible of 347 metacarpus of 380 metatarsus of 383 os innominatum of 380 palate of 341 phalanges of 380, 384 point in sternum of 367 posterior nares of 341 postorbital processes of 338 radius and ulna of 376 ribs of 372 sacral vertebre of 362 scapula of 374 skulls of 338 sternum of 367 tarsus of 383 teeth of 349 thoracic vertebre of 357 tibia of 382 vertebral column of 356 zygomatic arch of 345 Leporide and Ochotonide, osteological differences between 384 Leptoplana 134 Lepus 327, 380, 304 Liberia, recent visit to Coffee mountains of 452 lignicolor, Paradoxurus 44 limi, Umbra 304 Limnolagus 329, 406 Liparidide 238 Lodge, R. H., black bears bred by 175 Lodge, W. R., black bears bred by 175 Lucas, F. A. A new plesiosaur 96 A skeleton of Hesperornis 95 The dinosaur Trachodon annectens 317 lucas, Mus 30 Sciurus 20 lucasi, Brachanchenias 96 lugens, Mus 33 Lumbar vertebre of Leporide 359 of Ochotonide 366 luteolus, Mus 36 Lyon, Ernest, on Coffee mountains of Liberia 452 Lyon, M. W., jr. Classification of the hares and their allies 321 McKim, Randolph H., prayer of, on arrival of Smithson’s remains 251 Macacus pagensis 61 pheura 63 Macrotolagus 330, 395 merens, Sciuropterus 26 Malayan mammals, seventy new I Mamlepus 330 Mammals, Malayan, seventy new 1 Mandible of Leporide 347 Markiana 145 mase. Mus 32 Ratufa 7 MISCELLANEOUS COLLECTIONS matthzus, Mus 29 Sciurus 19 medialis, Rhopalonaria 270 Medicine, Chinese 180 Merrill, G. P. Notes on geological col- lecting expedition 217 On the glacial pothole in the Na- tional Museum r100 Metacarpus of Leporide, 380 of Ochotonide 380 Metatarsus of Leporide 383 of Ochotonide 383 Mexico, shell ornaments from 97 Microlagus 330, 402 Microtolagus 331 Miller, Gerrit S., jr. Malayan mammals 1 minimum, Allonema 284 minor, Hemigale 43 Minot, C. S., occupant of Smithsonian seat at Naples 215 Missouri, examination of bone-beds at Kimmswick 219 study of iron mine at Leslie 218 Mitsukurina owstoni 234 Mitsukurinide 234 Mnuolagus 331 Moenkhausia 145 moniliforme, Allonema 282 Monzonite from Greenland 206 multiradiata, Vinella 276 Mus bale 33 bentincanus 38 casensis 38 domelicus 39 gilbiventer 35 julianus 34 lucas 30 lugens 33 luteolus 36 mase 32 mattheus 29 pagensis 39 soccatus 30 stridens 28 stridulus 29 umbridorsum 37 musume, Julis 236 Myleocollops 147 Mylossoma 148 Seventy new nevius, Nycticorax Nycticorax 104 Naples Zoological Station, occupants of Smithsonian seat at 215 results obtained by C. M. Child at 134 National 100 Korean Headdresses in 149 natune, Galeopithecus 50 Museum, glacial pothole in INDEX Nelson, C. A. from 97 Nesolagus 331, 425 Newfoundland, photographs of finback whales from o1 niadis, Chiropodomys 4o Nobel prizes 216 Nugsuaks peninsula, rocks of 183 Nycticorax Nycticorax nevius 104 Nyctinassa violacea 110 obscurus, Funambulus 23 Ochotona 331, 431, 438 Ochotonide, carpus of 379 caudal vertebre of 367 cervical vertebre of 365 cheek teeth of 355 clavicle of 374 genera and subgenera of 431 femur of 382 fibula of 383 geographical distribution of 440 humerus of 376 incisors of 355 lumbar vertebre of 366 metacarpus of 380 metatarsus of 383 os innominatum of 381 phalanges of 380, 384 radius and ulna of 378 ribs of 373 sacral vertebre of 367 scapula of 375 skulls of 347 sterna of 371 tarsus of 383 teeth of 354 thoracic vertebre of 365 tibia of 383 Ochotonide and Leporide, osteological differences between 384 Ogotoma 331 Olivine-basalt 315 Ornaments, shell, from Kentucky and Mexico 97 Orthomyleus 148 Oryctolagus 331, 402 Os innominatum of Leporide 380 of Ochotonide 381 Osteological differences between Ocho- tonide and Leporide 384 Othonophanes 145 Owston, Alan, fishes collected by 230 owstoni, Mitsukurina 234 Trismegistus 238 pacificus, Gastrostomus 254 pagensis, Macacus 61 Mus 39 Palate of Leporide 341 Paleontological collections from Europe 448 461 Shell gorget obtained | Paleozoic bryozoa, revision of 256 pannovianus, Sciurus 11 Paradoxurus lignicolor 44 parvulum, Ascodictyon 288 Pelican fish from the Pacific 254 pemangilensis, Sciurus 9 peninsule, Funambulus 25 peninsularis, Sciurus 10 Pentalagus, detailed account of 428 Peridotite from Greenland 195 Personal equation in transit observa- tions 225 Peru, earthquake in 451 Petaurista batuana 27 pheeura, Macacus 63 Phalanges of Leporide 380, 384 of Ochotonide 380, 384 Phalen, W. C. A new occurrence of unakite 306 On the rocks of Nugsuaks penin- sula, Greenland 183 physalus, Balenoptera g1 Piaractus 148 Bical 332 Pictographs, Porto Rican 132 Rikassse 438 piniensis, Ratufa 8 Sciurus 14 Plagyodontide 235 Plagyodus ferox 235 Plesiosaur, a new 96 Peecilolagus, account of 395 Point in sternum of Leporide 367 Porto Rico, archeological objects from RED Posterior nares of leporide 341 Postorbital processes of Leporide 338 Pothole, glacial, in National Museum 100 Pottery from West Indies 126 Presbytes batuanus 65 rhionis 64 Pristiurus eastmani 230 prometheus, Promethichthys 236 Promethichthys prometheus 236 Pronolagus, detailed account of 416 Pseudemys rugosa 252 Pseudotriakide 232 Pseudotriakis acrales 232 Pteropus geminorum 60 Ptychocladia 267, 289 agellus 290 Publications, recent, Institution 220, 455 of Smithsonian pulonis, Tupaia 56 pumilus, Galeopithecus 46 Sciurus 15 purpurascens, Chimera 235 pygmea, Umbra 304 examination of Quarries, aboriginal, 220 462 radialis, Vinella 274 radians, Vinella 276 Radiation, recent studies of solar con- stant of 74 radiciformis, Vinella 275 conferta, Vinella 275 Radius of Leporide 376 Ochotonide 378 Ratufa bale 6 conspicua 5 insignis 4 mase 7 piniensis 8 regalis, Hesperornis 95 Regurgitating sticks from West Indies 130 repens, Vinella 274 rhionis, Presbytes 64 Rhopalonaria 266, 267 antiqua 269 attenuata 268 capillaris 271 keokukensis 272 medialis 270 robusta 269 tenuis 270 venosa 268 vetusta 271 Rhopalonariide 266, 267 Ribs of Leporide 372 of Ochotonide 373 robusta, Rhopalonaria 269 Rocks of Nugsuaks peninsula, Green- land 183 Romerolagus 332, 420 rostratus, Funambulus 24 tubeculus, Sciurus 22 rugosa, Pseudemys 252 russulus, Tragulus 3 Sacral vertebre of Leporide 362 of Ochotonide 367 Santo Domingo, archeological objects from 112 saturatus, Galeopithecus 51 Scapula of Leporide 374 of Ochotonide 375 Schuchert, Charles, on paleontological collections in Europe 448 Sciuropterus merens 26 Sciurus adangensis 17 altinsularis 21 aoris 10 atratus 13 bale 14 bentincanus 19 bilimitatus 8 casensis 20 domelicus 18 ictericus 12 lancavensis 16 lucas 20 SMITHSONIAN MISCELLANEOUS COLLECTIONS Sciurus mattheus 19 pannovianus 11 pemangilensis 9 peninsularis 10 piniensis 14 pumilus 15 tubeculus 22 sullivanus 17 Scotia Antarctic expedition 453 Scylliorhinide 230 Shell-carvings from West Indies 131 Shell ornaments from Kentucky and Mexico 97 Shields from western Sumatra 450 siluriense, Ascodictyon 286 Simias concolor 67 Skull of duplicidentata 337 Smith, Hugh M. Notes on breeding habits of yellow-bellied terrapin 252 Smithson, James, removal of remains of 243 Snyder, J. ©: See Jordan, DiS; and Snyder. soccatus, Mus 30 Solar constant of radiation, studies of 74 South American fresh-water fishes, new genera of 144 sparsum, Ascodictyon 289 Springer, Frank, crinoid presented by 450 Squalide 235 Starks, E. C. See Jordan, D. S., and Starks. stellatum, Ascodictyon 287 Sterna of Ochotonide 371 Sternum of Leporide, point in 367 Stichonodon 146 Stone implements from West Indies 116 stridens, Mus 28 stridulus, Mus 29 subfusiforme, Allonema 283 sullivanus, Sciurus 17 Sumatra, shields from 450 Sword, study of, by Paul Beckwith 220 Syenite from Greenland 190 Sylvilagus 332, 396, 401 Symphalangus klossii 70 Tapeti 333 Tarsus of Leporide 383 of Ochotonide 383 Teeth of Leporide 349 of Ochotonide 354 Teeth of Ochotonide 354 Telescope, horizontal, of Astrophysical Observatory 84 tenuis, Rhopalonaria 270 tephrura, Tupaia 57 Terrapin, yellow-bellied, breeding habits of 252 notes on INDEX Thoracic vertebre of Leporide 357 of Ochotonide 365 Tibia of Leporide 382 of Ochotonide 383 Trachodon annectens 317 Trachypteride 240 Trachypterus ishikawe 240 Tragulus batuanus 2 russulus 3 Transit observations, personal equation in 225 Trismegistus owstoni 238 True, F. W. On photographs of living finback whales 91 tuancus, Galeopithecus 53 Tubularia 142 Tupaia castanea 54 cervicalis 59 chrysogaster 58 pulonis 56 tephrura 57 Turbellaria, form-regulation in 134 Ujarartorsuak, Greenland 202 Ulna of Leporide 376 of Ochotonide 378 Witch wewOs, and Bassler, R. S. vision of Paleozoic bryozoa 256 Umanak island, Greenland 184, 209 Umbra krameri 304 limi 304 pygmea 304 Re- Umbras, account of the 295 umbridorsum, Mus 37 Unakite, new occurrence of 306 venosa, Rhopalonaria 268 Vertebral column of Leporide 356 vetusta, Rhopalonaria 271 Vinella 266, 273 multiradiata 276 radialis 274 radians 276 radiciformis 275 conferta 275 repens 274 Vinellide 266, 272 violacea, Nyctinassa 110 virescens, Butorides 110 waldronense, Allonema 283 West Indies, archeological trip to 112 Carvings from 128, 131 Whales, living finback, photographs of QI Williston, S. W. Description of plesio- saur by 96 | Wood-carvings from West Indies 128 youngi, Ascodictyon 289 Zoological station, at Naples 134, 215 zygomatica, Atherura 42 Zygomatic arch of Leporide 345 wt ¢ tw by rere ev pe pee ne PSI U nn) er ves Severity iain warty entrar Sain (bn tire yere ser terran iti ta Carte ty iar hn ke vA Pry reror nant POSTE ore sh Omri tt ereran iy tery Murat titan Geletetien rar Dra ray tener es eens la Leh cree ha ae et hei cs ean earieery ath ak A nT Pete one Treen wee he riinie Pon enerty er rier te yy ae TOU eeety rrirran ete ret artery NOP OWEN Hed eI opbag addi e EY Me efit tio reel ed Poser eer Wiad Woah hee & bum dad CITT Tea ren tiny crt eal are th ier oa fer er tar eure Tn PLN Cool PM nrien nial rewired erin) iva ee CO eye ta nr Ee) ean yay peer verti PS rer teat hr ht er cna: > errr ee i SPAM create ie Orne LN re er che ni rene DERI Y prea ata Difiedtpaavartep sures ab Wotan one Oe tninern ss rayne ante. Tene te Pern rary at COM anni vier he Sanbiy rer errr ant bree at rr rN OT etary pene era HT e CHU i i ar Pern even Pareriirt ints Pear tery a neni) tai cae er niet More inte ry tate UNPenT enh eyen ne SON ea eee Ve enn Teer ie arr ah hii rere ary Cyd Puree sneer) Nes oe ee renty" erica Ree inlet sted ster er tech eet eee et merc) AMEE UBEM ST ue need el Mew eet Ol fe nbs poo hela tlhes bit pbhee ie Ce neers rt ie CTSA ROR ron Cire rie a er ey nian Thien ca herr ares TVS Tee PR ert ere he Rata I Oe erie ih rari er inn Sel Iraniv Tika evn turn Ter POO ere erty Ore Pee a ae a ee ore AP haere tre oe UT MPTP ty ear vs era) Seater. oer wry ee] TOT A er eit Per near Rene ene ren ray Teds eects rey eet LOPE er een) Oar entre nn Orr, ete vas Tne es enter ery Cera a tO OTTO UN Ce rar ein ty Seer AN fey ena very Var arin @ WLP Puna a Yay Yu CP wore tr et er PoP Tea e ee ere ree Chir erteenny Tarte ereoven tile eee ne | rem wy) O21 OS RIMMER NEM He tre Mbt rf Pieri Cronin ir aerreters tater Deer y Werke emer ear Poo ar ae trite tnt art roe wine saiTa iri ye) Tren rare NLPOUN er rn ie, veers e arrie eri" ora renin H pore t tyne tnt} a cari, ieee Thabo) PRs CIO ee AC RR a Ue aU Wel ae hey CVO Ce DOS PR Mad OAT COA Reis Mar eae ar ebreS ey A r fiat wat Oren ia he ta OSAP REHM DT Ea AP hd ol LOUD he ree ante 1 pod yn beds duty 4 aa Kid q irae eA AL MAL Lata CR LI rar ee Pr kere r el Pina Puen hy pany eT TA dee Reh ma TORU er Ment Mey hs Wb ear aed chin we heerire HAGAN CCAS ater as Dee ta) na ea Cada sah ee gk ats (ta aa MT CAL Oa Le ee mney (PER saa ee a { t ata Lister peur irmehiara® (Pele it CIN MIA DA RRO OI a a SAS Ok eR NTA he een TC PIV Ae MALI ARM SODA UC aT er Tae or Pare Nea a Write RDA ed TAL nee MOPAR YIN CT Ch ag cg ART OLoA GAGA OA ORG td ca teat tate ec te a LMUAMI TIER Cat cage ie ae cmt an eee oy eri aT a Heat had Onda I a cee A ea wided Sil DA URUE ee Me VeH nim ance velieliarwe nr Ue UFURU EY eee HERE DOORS EDR Meher Ue UEOEIS DEKKER ORTENY er LARUE rte Dae Petrey Os Pain tery yer eee Hele ie Oe das be eta tie he unten bias hel ADSL ES OUT at Retr ec Teen ere ihY eee ake Ratios APRA Raa VTA eet ae esa Pease ea tite bento é Pt Ph f TM Or Tea eae a a ere CUE (eC eLC Slt att) ae ean os eee aE ear ec Tr ON ie (bn err tia Oe) ne be fl PROMO Loek id cl te AREER ce TR Le at a Deen RAO at Dacohnen ts tl Ha bade Da okie pe oe ea at (ta aa ds fi OLEATE tans Alt Leo eee eh at WAR ULea tO re cme arto Pe ent Deaths Wee etn tat ot Pe a one a ee UP bebe leur (aga i Cierra ireesr an ech ratchet ht nies ft erat sr ait erp hey prey aprile FORTE Le Uareme ie tre pede EES WELK APIECE OE bep aire | OER. Beier EUaT HT hemees Ue h iE tele Teese OOO abbr Len DADO paid SALOU ALC OTe nt ater | Ua ret TURD eee ae Le aire ne LATE es na La TOUR GTC Sencha tL ae te ee een A AIA eR oe Chae ey er eT UCT Pre Pr ante Tet eer rena hari ne UA Us Raat es TCT CeO ae et A OTe hy Peed nn Cs nD ev en aT UA trea LO eC RP Pear rt eae te ta rari Ne APR Oe na TPN TET AMON eae Pn RT DRACO nt hairs aren ny (ha OTS Rn eT ar enn Oe Pa ocala da tema ea ati: re SATs (ina ect oaks Te ay CRT Rent rey oatrta irs orient Cem ne ron i HOM arte a et hit MUM er ate te Pea fy APAMIADAUeRO WROD RT RASA R AOR dl PLU ba ake Oe Uae DDO er en tia es enn bah) SOOO ARR MC CO HOT OM TTL routs Oo (rome Een (Paha eu ' ADAG De a RE LA ee a St Tee re TP a ae CUR Cee areata eae a ta UMA keg a Uae are i CORA he Lo Cer ea ROU eS Cees wee ee Ty arn PHEW HORT EEL eeHeUe WH We (sje we ile ee OM uaa aa if f AVeb iota bi na bole Eta eR ca Cea tc ee aaron ac Wea te Pe AAG UAIA AA ei Ne UL MRL RU eR TP evs er OP as ea Te fea TR Pa DPA nA Tiilbe UTA PACU Lat Trair a kta utah i Thy aati oe eo Baar nee Sonal DRC Ne BOP aI oe eR CP ea ea tra hs Oa TON e eatery mnie tnt ra wea Panne a hese Pex ate ta ea te beh Qe htegeh SH Osea ee lieve uelt Roipin UE VE OUUEdreiiebs HAE Arde vst raiedie (Mee l UIReb eR UIE “ aaa thy CI De ROE La ene TCT Rate Cea ech ee MAA hon nae UU tT UCR Oa i i Pa odurHeA ne i GL Ce as a erred eccrine ree er ini Heine ett OA IC on Uke an UU re t MOTT TRSaLU CT aT a ha en oe A eR er ord Te Prt cor Dae eae esa bar a tion Arba Wai eat a bel oO Pee Pe Hee TAMA LG LP eg SO MRT tl Dk er ea Delete ef ae We VEOH RS GiGe rel yrGitrveu LEA LHe cu Suche ats bees tty wend Aen iuy nee Patan ue A UC Ua Us AU ene bares a Phar Vee de tee tet C8 bk UA ORM OPEC ea Ane oer ety on ED NEY PETA ea Ua a LAT UWe even Oeste ebilsterg Hie ehpe pene liu poled Mons byiroruc he neloe mekst teks s na earn? UICC TCR Lon ore tater fat OU H Me be US ae gay ear Palen aD AUN a Ud et Set sari ober eb eh Ube ee oc a Nh SU ete Cer ea mana ar bahar ne ay her Iba TSTOT Da rt PUPA DAAC Utd ad A eS Ea ee Tek Dw Brita aire iy Par abs dees a eats Tae ” UM obra A aad RU RT RP ne rR TR) ean Vn ear Our eri LL Nee Oke rn et Nn tr ee ir Cu aera fi ORM ba ae aS OD oO nae eae Oa Wi r HPD De Deeb CuatnrreLes bel ph Gite: Usbeeme tet heeetic| a0) eM ta Cee a a ote wane aa ata iia eis rae ee ey ea tenn fer a Herre SUM reg nae He oe teen Me beD ey Porat at PAAR dt St ee aa a eT ea cen eta Marae SAAR UR ALOT Cae eee aR Uta UO Ed Coa PE eta UU CLEL DOANE SP EDA BOUL De lieie mere hetaltrticuie leben uene ie wele betietep es eieeete fel Wide migtepeder gears by bled letome iru. le bs Ged) Ci aU Dae er a eta a OL Pat ak el Pa AalISIod (hata Dat a ahs Cha ea ra tears OD ota ON er al ad oP LD aa USA MU he Ten ty a ek ee enter (ya) oe en (tara: BURT AI ot UR eat Ot Ue a a aa ee OEE Ne lie MMW Ke ROM We He Gobel DEPrsle Om Dr tmyitienibe Goan d ere, 0 TSN arene ae ka Ue CE Dae Ut ne ee ee Fe et il ae RS Ra iy Ps Pa ee AS pea hepednt eset lieheit li, + eer ree nn Per eier ey fern q fires ae is diivete ye bot Perret au H Whee ARM aera iran avr mis a mare r Corer a i i Herta hr te Bie f n earn i nena Daan ge) Pyle iran SMITHSONIAN INSTITUTION LIBRARIES “WII 3 9088 01421 4365 DE Hi f f ee as ren fi vy aor r f i rw egies A Pa ‘i tea i we i f i i r vey ( bee f al ie f ou a i ay aa c ual ee a c aL i f i f V isefet (i ' Pe, f n ay, NN