ut sent sey Ley ee ad tet 44 i ters } ' ‘ vial Oye 4 ta] aly ys LH At Be, HALAL ICN wWwonne \ 1 vis 4 eye yee Vi ae Oh Ue rh 1, " ri Mi : hy, ian Bag) ra 44 AWA Nate) We iEe y 4 iM 4 ANY ahd fi Nava) Ay ‘a Ry enn’ "4 Pe wLl away PO oa yew SEP Gate RY A Taya We? Ae IM PAN telat 4: sate ca hye \ tind Vrs WY Meet aka ee moe) oath hy } ri ah a ays wiyata yp) aoa s \ RAY ata ’ yng \ Ny q ih WN N Vai gil ae ry ba Seth)» Te ie} i 1 AMIN TaN 4 yy thy f EAL | NAGA ER NI Ae 4 ay) Sy lh) UA he ie a aN yy via! BAA te that) yyy u Waker AMOR HELE ECE a ie eS, SN BAMA SE Oh eteuauenoc 4 yee ae Ne HR ent ; x) + Nahe art j wonete x Hie yy VAN RO) Pe yl Paca a ono y } ay) HW i Shite AH ANE ii M34 hae PAM Sa ¥\ ‘ teh Pansy sity era VT PY TAD by fT ith MAA a TN a ar) Hi TALE A A vs ia Ree Ae J HAA Sh Dept tua ins Acad tea eH AS DONO De é ee Pati vane NG ott BY nan ; rb ar CREM i ; iy ws 4 wate a AO ye je NOs Ny ‘ PLATA CYL NT vet ass Ae Wee ala Ag ASS yal MN eG bpaF pei ¥ re ee ; ALN iy VN RLY Lae a yay SAN, 34 Vat * Air's Va 7 NR WIR Ge f 4 PCW MII Ge ihe oats Dy nN Pie nieces yy LG MON q Pee NET IT} PIG he AD em 6 te ee > 4 sha f Ap Aw a lt isi i 4 Ab ath > i Hey f Sag Mat t i i ee ey , Lies na} fie ee Tete Xa} G erat Oe ee 1) par pe iS Wa ke 0 J ae ey in aga r, rl (# Vow Vaya ’ 1 ‘ .) AAT OTD. vob te MWY aN , 4 i aS aL SOUT TAN at AYA ACH Mts et A rt te ar Ps i At viaR! maid hate + POUR MOCHUN oO SEA FY H 44 RET RA eA r E i aks Dts Aeacdiacte ‘ ‘ cas ; ayy oa aa q ' ‘ AY) ? a, ‘ 4 fi pire NN Mi SIE a Oe ‘ : Carona! i AN: © 5 Usa eo) ri ont Bit Oy ie bade COR Ae Naka f Paae ” ial . hee > Mire AM a PUR ANN An VN Gh ' POO e be Bry t Uy | * nie AY | i nil arne wD © von i fel y) 5 ‘ ick SMITHSONIAN INSTITUTION UNITED STATES NATIONAL MUSEUM PROCEEDINGS UNITED STATES NATIONAL MUSEUM VOLUME XXXVI WASHINGTON. GOVERNMENT PRINTING OFFICE 1909 ~? ADVERTISEMENT. The scientific publications of the National Museum consist of two series—Proceedings and Bulletins. The Proceedings, the first volume of which was issued in 1878, are intended primarily as a medium for the publication of original papers based on the collections of the National Museum, setting forth newly acquired facts in biology, anthropology, and geology derived there- from, or containing descriptions of new forms and revisions of limited groups. A volume is issued annually or oftener for distribution to libraries and scientific establishments, and, in view of the importance of the more prompt dissemination of new facts, a limited edition of each paper is printed in pamphlet form in advance. The dates at which these separate papers are published are recorded in the table of contents of the volume. The present volume is the thirty-sixth of this series. The Bulletin, publication of which was begun in 1875, is a series of more elaborate papers, issued separately, and, like the Proceedings, based chiefly on the collections of the National Museum. A quarto form of the Bulletin, known as the ‘‘Special Bulletin,” has been adopted in a few instances in which a larger page was deemed indispensable. Since 1902 the volumes of the series known as ‘‘Contributions from the National Herbarium,” and containing papers relating to the botanical collections of the Museum, have been published as Bulletins. RicHarp RATHBUN, Assistant Secretary, Smithsonian Institution, In charge of the United States National Museum. JUNE 20, 1909. IIL TABLE OF CONTENTS. Page. ANNANDALE, Netson. Fresh-water Sponges in the Collec- tion of the United States National Museum. Part I. Specimens from the Philippines and Australia.—No. Serene, to, O00. 2) aa ee el 62682 _ New species: Spongilla philippinensis, 8. clementis. Bean, Barton A., and Atrrep C. WEED. Description of a New Skate (Dactylobatus armatus) from deep Water off the Southern Atlantic Coast of the United States.—No. Seeray 27 ted hee en oo Ll. SOE 4G New genus: Dactylobatus. New species: Dactylobatus armatus. Descriptions of Two New Species of Electric Rays of the Family Narcobatide, from deep Water off the Southern Atlantic Coast of the United States.—No. 1694. MOO tos ee Bre =680 New species: Benthobatis marcida, B. cervina. Casanowicz, ImmMaANUEL M. The Collection of Rosaries in the United States National Museum.—No.. 1667. April een eet a re ln we a2 DOOD Ciark, Austin’ Hopart. Comatilia, a Remarkable New Genus of Unstalked Crinoids.—No. 1668. April 27,1909%_ 361-367 New genus: Comatilia. New species: Comatilia iridometriformis. ——. Descriptions of Seventeen New Species of Recent Crinoids.—No. 1691. June 19,1909%_____-.__-.-_-_-. 633-651 New species: Eudiocrinus ornatus, Amphimetra mortenseni, Hetero- metra compta, H. singularis, Stephanometra coronata, Colobometra discolor, Cyllometra taprobanes, Crotalometra annandalei, Crinometra pulchra, C. margaritacea, C. concinna, C. insculpta, C. gemmata, Psathyrometra mira, Mastigometra micropoda, Hypalocrinus springeri, H. ornatus. —. Four New Species of the Crinoid Genus Rhizo- ermmis-—_No. 1693. June 19, 1909¢___..._......-.-... 673-676 - New species: Rhizocrinus conifer, R. brevis, R. sabe, R. robustus. a Date of publication. belt tee Ss ; VI TABLE OF CONTENTS. Page. Ciark, AusTIN Hopartr. On aCollection of Recent Crinoids from the Philippine Islands.—No. 1673. May 13, 1909%_ 391-410 New genus: Crotalometra. New species: Phanogenia minima, P. delicata, Comanthus polycnemis, Pontiometra insperatus, Cenometra Jelicata, Ptilometra pulcherrima, Stenometra arachnoides, Crotalometra eupedata, Pachylometra levigata, Iridometra exquisita. —. Revision of the Crinoid Family Comasteride, with Descriptions of New Genera and Species.—No. 1685. aune 7; 19094 8 es ee Se eke 2 ee NS New genera: Cominia, Comactinia, Leptonemaster, Comissia, Nemaster. New species: Leptonemaster venustus, Comissia lutkeni, Nemaster grandis. CocKERELL, T. D. A. Descriptions of some Bees in the U.S. National Museum.—No. 1674. May 13, 19094__._. 411-420 New species: Emphoropsis vierecki, Mesotrichia abbotti. New subspecies: Emphoropsis murthirta murina. ———., See under Robbins, W. W___.__--------------- 381-384 CouTieRE, Henrt. The American Species of Snapping Shrimps of the Genus Synalpheus.—No. 1659. January BOs NOOO Se 8 Sep5 of eh Sees oe aan eae ya ae a a 1-93 New name: Synalpheus lockingtoni. New species: Synalpheus paulsonoides, S. latastei, S. apioceros, S. townsendi, S. fritzmiilleri, S. hemphilli, S. nobilii, S. sanlucasi, 8. digueti, S. goodei, S. sanctithome, S. grampusi, S. pandionis, 8. brooksi, S. herricki, S. pectiniger, S. androsi, S. rathbune, S. paraneptunus, S. albatrossi, S. merospiniger, S. trionychis, S. bakeri, S. physocheles, 8. otiosus, S. mushaensis, S. maccullochi, S. lopho- dactylus, S. sladeni. New subspecies: Synalpheus apioceros sanjosei, S. a. mayaguensis, = S. a. leiopes, S. a. desterroensis, S. paulsoni liminaris, S. p. sene- gambiensis. = CusHMAN, JosEpH A. Ammodiscoides, a New Genus of Arenaceous Foraminifera.—No. 1676. May 13, 1909 %___ 423-424 New genus: Ammodiscoides. New species: Ammodiscoides turbinatus. EVERMANN, Barton WARREN. See under Jordan, David Starr 24 22 Se le oe bes a a Gay, M. EK. . See under Holmes, 8S. J__2.-. -2-. -- 22 32-2 6 (9 ee GIDLEY, JAMES WitiiAMs. Notes on the Fossil Mammalian Genus Ptilodus, with Descriptions of New Species.—No. 1689. June.19,.1009°% 223.0 0 2 eee eiee See New species: Ptilodus gracilis. « Date of publication. TABLE OF CONTENTS. GILMORE, CHARLES W. Osteology of the Jurassic Reptile Camptosaurus, with a Revision of the Species of the - Genus, and Descriptions of Two New Species.—No. 1666. oe LST SE CLEC NSS Seger ee New species: Camptosaurus depressus, C. browni. Hay, Oxtver P. Description of Two Species of Fossil Tur- tles, Toxochelys stenopora and Chisternon? interposi- tum, the latter hitherto Unknown.—No. 1665. April 8, (Eat Saw a Depa ai aa I ele de ee ee pe epegn aes New species: Chisternon? interpositum. On the Skull and the Brain of Triceratops, with Notes on the Brain-cases of Iguanodon and Megalo- saurus.—No. 1660. February 6, 1909%_______---_-..- Hormes, 8. J.. and M. EK. Gay. Four New Species of Iso- pods from the Coast of California. —No. 1670. April 27, eee ee A eee ee 2 et New species: Ancinus granulatus, Tylos punctatus, Actoniscus tu- berculatus, Philoscia richardsone. JORDAN, Davip Starr, and BARTON WARREN EVERMANN. Descriptions of Three New Species of Cisco, or Lake Her- ring (Argyrosomus), from the Great Lakes of America; with a Note on the Species of Whitefish—No. 1662. OTE SP ogg Se SS PBR ae 7S a a a ae New species: Argyrosomus eriensis, A. huronius, A. zenithicus. —w and Joun OTTERBEIN SNYDER. Description of a New Whitefish (Coregonus oregonius) from McKenzie River, Peete NO4 Por fr Mary 15,1909 ¢ o-oo ee New species: Coregonus oregonius. Lyon, Marcus Warp, Jr. A New Squirrel from Direction Island, South China Sea.—No. 1686. June 7, 1909 @__-. New species: Sciurus director. Additional Notes on Mammals of the Rhio-Linga Archipelago, with Descriptions of New Species’ and a Revised List.—No. 1684. June 1, 1909%._...-_..---- New species: Ratufa bulana, Mus chombolis, Galeopterus chombolis. Remarks on the Insectivores of the Genus Gym- mie NO. 1680: May 27; 1909%. 2.2. 2222. 2 eis -L- New subspecies: Gymnura gymnura minor. Mason, Orts T. Anyam gila (Mad Weave): ’ Maxilliped...< 225.0. 2.0 22522, 52. he eee Anciius depressus.. “Mandible. :.is s.2)...4.0G.25 «2-3. See ee Ancinus depresesus. First pair of legs of female. ..-..:-2 -. 0.2.2 -f20)cee see Ancinus depressus. First and second pleopod..............---2++2+22-2-e25+ Ancinus depressus. © Third pleopod ..2..-2s.-2.4- 5. =2 222 aeee ana Ancinus depressus, - Kourth pleopod:: : 02-22 +: .-2--2- 5 5a eee Ancinus depressus. Fifth pleopod. Outer branch...............----.-------- Aneinus depressus. Fifth pleopod. Inner branch.......-.-.--..1.¢----42s2ee Toxochelys stenopora. Plastron. Entoplastron; epiplastron; hyoplastron; hypoplastron; xiphiplastrom).. - 6.2... ...2..<8 12 a= see eee oe eee Chisternon? interpositum. Part of carapace. First costal plate; third costal plate; neural plate; first neural plate; fourth neural plate; first peripheral; preneural DONG... J:...255so-en ace oes eee eee eee ete = eee eee Chisternon? interpositum. Part of plastron. Entoplastron; epiplastron; hyo- plastron; hypoplastron; mesoplastron; fifth peripheral................----- Map of Quarry 13... 20.22 os0 2 See elec nies wheeeeer oe ee ce er Skull of Camptosourus. Seen from the left side._-.....-... 22022 saceeeeee ee Skull of Camptosaurus. Seen from the top..-..-......-..- 42 565-=—eeeee ee Posterior view of occipital region of skull of Camptosaurus dispar Marsh. .... - Lateral view of posterior portion of skull of Camptosaurus dispar Marsh... -. - - Outline of left premaxillary of Camptosaurus medius Marsh ........---------- Lateral view of left maxillary of Camptosouria ...- 22. = 322 sce ee ee ae External view of left dentary, Camptosaurug .-. ./324,7--= tees see a eee Tenth upper tooth; fifth lower tooth of Camptosaurus medius Marsh .. ..-.--- Internal view of right dentary, Camptosaurus dispar? Marsh.......------------ Axis and portion of atlas of Canptosaurus dispar Marsh.....-...--------------- = oa — LIST OF ILLUSTRATIONS. Ventral view of atlas and axis of Camptosaurus dispar Marsh.............--.-- Left lateral, ventral and anterior aspects of the axis of an Orthopodus? dinosaur Bemtne Wealden ofthe Isle of Wight. ...... 2 2.202. ecie ee eae eee Highth cervical vertebra of Camptosaurus brownt.......-----.--.----2-2-00-- Third dorsal vertebra of Camptosaurus browni .............-.--------------- Thirteenth dorsal vertebra of Camptosaurus browni . See PSOE ET LOSUUPUG OFOWNT.» 2 Issac ees Des Soo ws ae oueel os ak ele see ol ok Second caudal vertebra of Camptosaurus browni- te See eet, Sees Posterior caudal vertebree with chevrons, ( ipouniens browns. picts Se cheat ge Anterior chevron of Camptosaurus dispar Marsh.....................-2.0.2---- Highth cervical rib of left side Camptosaurus brownt...........-..2-.--2-+--- Posterior view of articulated scapula and coracoid of Camptosaurus dispar eT eY OL CONDLAGHUEUE. UTOWN. vac.cnceUencc es os sani ee- se acetasgeeecee Bere omaCOIe OF CUI PLOBGUTUS OTOWM 2. .e sos base ee ee ee eee kee ee PeeEEMIIMeH IS OL COMPLOSAUTUS DFOWINL. 22.00.0502. 2s. Oe ee eee earner OF Camptosaurus brown. .22252220 2056-2... 2 eee oe ee ceed ee de Right radius, ulna, and manus of Camptosaurus browni..........-........-.-- Peenbaere toot, Campiosaurus dispar Marsh.....-21..--2..-.-.-.-.0-.---cseien BemiiieL OlvCamplosaurus dispar Marsh. 2. 5222.220.-..-2---222.. 20.0. 8s. MEeaNNGl COM MCOSAUTUS UTOUNIs 0s... 2 ot Svc Uee- eis. ele cee ve ee be Ischia of Camptosaurus medius Marsh... -.-.-.--- 2G os DEN ee Pant et ioN aot mere unis of Campiosaurus dispar Marsh) s. 2.2 00.is02 2. 32 oo. ee ole oz feet temuro: Camptosaurus dispor Marsh. . 2.2252. 05 5.2... 2.2 02202.022 058 en Right tibia and astragulus of Camptosaurus dispar March............-.....-.- mipan bind foot, Camptosaurus dispar Marsh. .....-....-.-.2....--- 220 en dees Pelvic arch of Camptosaurus (Camptonotus) dispar Marsh........-..--..-.---- Lateral and front views of first sacral centrum Camptosaurus dispar Marsh... . Ungual of first digit, Camptosaurus amplus Marsh...........--.....--..--0.-- Rammer Cnn piosmurius Hanus Moarsh....c. 02.22... 02-2 Js bi bee ete Right scapula and coracoid (reversed) of Camptosaurus nanus Marsh........-- ‘ight humerus, radius and ulna Camptosaurus nanus Marsh.............-..-- ight femur Camptosaurus nanus Marsh....-..-.--------- epee tsi The last dorsal or sacro-dorsal and sacrum of Comiptawnistes y pr a ie: Re te Left femur of Camptosaurus leedsi Lydekker............----22----0022--00--- Anterior portion of right ilium of Camptosaurus depressus.......-..-..--------- Memon CaMmplOsmurun CepressuUs.. . eos oes. eee oe ee Anterior portion of right ilium of Camptosaurus depressus......-...-.---------- CST ides 60 a ee ee =a! hs A iad Snes Paints. Mirai onathopod)..<22. 2.0.2.2... ee eee ee dee Ancinus granulatus. Second leg, second pleopod of the male ................ i PIR ie edEl og) ho) iain ob So o's 2 2 la ede a Se ER Tylos punctatus. Antenna; first maxilla; maxilliped; second pleopod of the a Perm Mmcopod UROPOd .. 2... Jie. see dg ote be ele! en a IIRL Tiana Coe G25". Z's l\o,2 wie Oe hoe ON OO ee) aE ei tear cE ne 72 Bop. 2 20) fs) hie wield «e.g ie Pace wm tae Saw Sale Peper Orperinnineniman WEAVC.......-2--- = <--5-22+ ss ee heen e sa cues eee bed Bron ara rm accumonel SLIPS. =... 5-0-2 -- se ees eeenaees eee ces nase Result of adding new strips in three directions...............--....-: iseehmey es Figure 3 dissected (inside)....................- A a ERE nbs NE SR EO Position of sinistral strip; position of sinistral’strip; position of dextral strip; position of dextral strip; position of vertical strip; position of vertical strip. - Proc. N.M.vol.xxxvi—09——-11 ITNNNNNNNNWNW WD WD LO “IO~NI OD GO SD Ct or or or J Bre Pr OO ON DD bo Oo ~I Be N XVIII LIST OF ILLUSTRATIONS. Method of giving hexagonal form to base at upset.............-...--.-------- Method of finishing at the border, with two hoops of rattan...........-....---- Method of turning strips at the border, polished side out..............-..--.--- Method of turning vertical strips at the border, polished side out...........-- Method. of finishing at the border. 2222. hatle ccs ote sce eee eer Method! ot ornamentation: 2-2. este See eee ee ee Radials, basals, and infrabasals of Hypalocrinus naresianus.......-.------------ Jaro psis PAlagOniensis 22~--- 5. na4- 22s ee ee ee ee 6 eee Coregonus OP CYONTUS minis towns Lae Ss ses oe =k pe ee a ee ee Acanthoniscus spiniger -.%..0--22% - 2-02 oe a oe ae ee Acanthoniscus spiniger. Second antenna......-.5----2.--.-2s¢----+2-2+--0558 Acanthoniseus spiniger. Uropod).o222:.- 50s 5 12 Ie obese oe Aconthoniscus spiniger. “Maxilliped. 02... 2-. ye2 coe nkeeee o- Acanthoniscus spiniger. Second maxilla........--2:...-2-22--s+se0+ssseeeeee Acanthoniscus spiniger. First maxilla, inner lobe, outer lobe .......-...-...-- Acanthoniscus spiniger. Anterior view of head showing epistome with labrum. Diagram to show to relative sizes of the three forms in the genus Gymnura as determined by length of hind foot, including claws, and of basal length of Skolllee ws posse apes ae eis cole na eae ieee aici ake ere ee Outline of ventral surface of Dactylobatus armatus.....-.---.--.-.------------ Diagrammatic tergum of any complete wing-bearing segment, and the base of Be wie dorsal VIGW os 5. -<.ckc2n,ctn Ae acieieae meee SR can a ae Diagrammatic tergum of any complete wing-bearing segment; ventral view ..- Diagrammatic lateral view of any complete wing-bearing segment, external. -. - Diagrammatic view of inner surface of the pleurum of any complete wing-bear- PMS BPO MACIG io. Loe cca as slo ys le oso da pe ee ee re eee Diagram of a generalized wing and its articular sclerities or axillaries.......-.- Diagrammatic cross section of a wing-bearing segment.....-.-.--------------- Let humerus.or Pilodus gracilis. 22 saree ee ae oe =e ee - oe Proximal portions of tibia and fibula of Ptilodus gracilis...-.-...--.-----+----- Right lower jaw of Trichosurus vulpecula.......----------+-+-++++-++--+2+-+-2- Right lower jaw of Ptilodus gracilis........- 2-22-2004 st + «Scie eee Right lower jaw of Plagiaulat brecklesii.......+..-----+----+-++-+----3++--55-> Spongilla sceptrioides. Skeleton spicules; gemmule spicules. ....------------- Spongilla philippinensis. Fragment of skeleton.......-.-------------------- Spongilla philippinensis. Skeleton spicules; gemmule spicule..........------ Spongilla clementis. Skeleton spicules; gemmule spicule; fragment of skeleton. BevOGAUIS MATOHAG. «icon aR ARK Sala ean eee eae Se er co lon) THE AMERICAN SPECIES OF SNAPPING SHRIMPS OF THE GENUS SYNALPHEUS.* By Hewri Couriers, Of the Ecole Supérieure de Pharmacie, Paris. HISTORY OF THE AMERICAN SPECIES OF THE GENUS. The nominal species of Synalpheus from the coasts of America at the present time are eight in number, of which one, Alpheus precow of Herrick, is a nomen nudum. All of them appeared at first under the generic name of Alpheus: A. minus Say (1818), A. spinifrons Milne Edwards (1837), A. tridentulatus Dana (1852), A. sauleyi Guérin (1856), A. leviusculus Lockington (1878), A. saulcyi longi- carpus and A. saulcyi brevicarpus Herrick (1891). These nominal species are so imperfectly diagnosed that I have been able to retain the names of only three of them, Synalpheus minus (Say), S. brevicarpus (Herrick), and S. longicarpus (Her- rick). This list could have been augmented by Alpheus leviusculus Lockington, had it not been necessary to change the name (it having been preoccupied by Dana) to S. lochingtoni. The A. spinifrons of Milne Edwards is from Chile. The type is lost, and I have seen no form from that region which exactly corre- sponds. Although Nicolet’s drawing may be very imperfect with regard to the cephalic appendages, yet the scaphocerite seems to be much reduced, and this is confirmed by the text: “lamina basilar de las antenas esternas muy peguefia, sin llegar con mucho 4 la estremi- dad del pedtinculo de estos érganos.” I do not believe that the species, in view of this circumstance, can be placed elsewhere than in the Lavimanus group. The small claw, it is true, is described sim- ply “con algunas pelos,” but the plume of long hairs, so charac- teristic of this claw in the group, could, in spite of its constancy, very easily pass unnoticed. This plume has never before been described or figured; but the unusual prevalence of the Lavrimanus group on the American coasts compels me to recognize the importance of this curious, though apparently insignificant, character. Formerly, I identified Alpheus tridentulatus Dana with A. minus Say, by reason of the short and broad form of the frontal teeth and “Translated from the French by Miss Mary J. Rathbun. PROCEEDINGS U. S. NATIONAL Museum, VOL, XXXVI—No. 1659. Proc. N. M. vol. xxxvi—09 alt y) PROCEEDINGS OF THE NATIONAL MUSEUM. VOL. XXXVI. the slight build of the small claw. Dana’s species is from Rio Janeiro, and the species of the Brevicarrus group (to which 4. minus belongs) are not as yet known to extend beyond Bahia. But that is a purely negative assumption, of very little value, and I still, as the most plausible explanation, consider that the antennal scale was accidentally omitted from Dana’s drawing. The absence of this scale would suggest the Layvimanus group; but none of the species of this group have the external spine of the basicerite so short or the stylocerite so long. The exact identification of the form is, however, impossible because of the large number of closely related species and the slightness of the distinctive characters. Alpheus minus is no longer represented by authentic specimens, except the two dried examples, fortunately alike, preserved in the British Museum and sent to Doctor Leach by Thomas Say himself. They can be identified with a rather common species from Florida and the Bahamas; such a determination would probably not be pos- sible for all the specimens collected by Say, did they still exist. Another species from the same region, appearing to be even more common, is that which Herrick has described under the name of A. saulcyi brevicarpus, in opposition to his A. saulcyi longicarpus. Far from being closely allied varieties of a single species, these two forms are in reality widely separated and easily distinguished; furthermore, S. brevicarpus, which I had formerly considered synonymous with S. minus (Say), is also distinct, and each of these two species pos- sesses several subspecies, forming a small, well-defined group which may be designated as the Brevicareus group, which, so far as is known at present, is characteristic of the American region. I have not succeeded in identifying Alpheus saulcyi Guérin with any of the forms which I have studied; the species belongs obviously to the Brevicarrus group, being perhaps synonymous with S. brevi- carpus (Herrick). In Guérin’s drawing the chief character which recalls this species is the narrow form of the antennules; but the frontal teeth, the scaphocerite, the superior prominence of the basi- cerite are very imperfectly figured. As Guérin’s species is from Cuba, I should be more inclined to believe it synonymous with the new form which I have named S. brevicarpus querini, which is also from the West Indies, and which in other respects appears most comparable to Guérin’s figure. This resemblance is, however, much too vague for me to consider myself justified in retaining the name of saulcyi, in spite of my desire to do so. As to Alpheus longicarpus, I have been able to examine two speci- mens received from Professor Herrick; because of the dissimilarity of these two specimens, I had thought it wisest to distinguish all those which corresponded to them respectively under the name of longicarpus a and #3, but this distinction is very far from being satisfactory; for r no. 1659. AMERICAN SPECIES OF SYNALPHEUS—OOUTIERE. 3 ‘ _8. longiearpus a belongs in reality to the group which I call the Lxvimanus group, from the name of the Mediterranean species of Heller, a group which is characterized essentially by a brush of long stiff hairs on the movable finger of the small claw. I have been obliged to recognize eighteen species and subspecies belonging to it upon the American coast, so that the old specific limits have become greatly narrowed; I have retained the name longicarpus for that species which appears to be among the most widely distributed in the region of the Gulf of Mexico and the Bahamas, and which conforms to one of Herrick’s types; the species has small eggs and its larve are zoée; in regard to S. longicarpus f, the specimens which I had at first grouped under that name are found to be referable to three very distinct species, each provided with numerous subspecies; that species which corresponds to the type specimen of Herrick is S. pectiniger, new species; the other two have received the names S. brooksi and S. herricki. All three have eggs of large size, from which spring mysis larvee, and one of the species must certainly be Herrick’s nominal species Alpheus precox, without its being possible to definitely determine which. Save for the preceding exceptions, all the forms, perhaps thirty species and varieties, have had to receive new names. It is a con- siderable number, and surprised me at first. Although several pre- sent very strong resemblances to other forms of the eastern Atlantic, the Indian Ocean, and the Pacific islands, all are peculiar to the American coasts. This is true also of S. lockingtoni, which is repre- sented in the Indian Ocean by some closely allied species, which, in turn, are difficult to separate from specimens from the Red Sea, the Mascarene Islands, and from the west coast of Africa. It would seem that chese specimens represent local races of a cosmopolitan species. There is, however, a remarkable exception in S. dataste?, Chilian speci- mens of which can not be distinguished from Australian. CLASSIFICATION OF THE SPECIES. In view of the growing number of species of the genus Synalpheus, one is led to distinguish among them several groups composed of the more closely allied forms, which may be differentiated in the following manner : KEY TO THE SPECIFIC GROUPS OF THE GENUS SYNALPHEUS. ; a*. Supraorbital spines insignificant compared to the rostrum; antennules shorter than the antennie; spines of the basicerite almost equal, the exter- nal always smaller than the stylocerite; external maxillipeds oval, feebly spinous distally; first segment of the carpus of the second pair of feet very long; following feet cylindrical; ventral hook of the dactyl obsolete; telson with an oval median lobe____________________ COMATULARUM group. a”. Supraorbital spines at least equal to the rostrum in importance; antennules at least equal to the antennie; spines of the basicerite unequal, the exter- aitiabelll 4 PROCEEDINGS OF THE NATIONAL MUSEUM. VOL. XXXVI. nal often larger than the stylocerite, the internal often wanting; external maxillipeds cylindrical in form, very spinous distally ; first segment of the carpus of the second pair approximately equal to the sum of the remaining segments; following feet flat in the sagittal plane; ventral hook of the dactyl as large as the dorsal; posterior border of the telson almost straight. b*. Dactyls of the third, fourth, and fifth feet with two unequal hooks, the yentral always stronger (up to three times greater), often accompanied by a third prominence obtuse or spinous; meropodites often spinous; frontal spines always longer than wide at the base_-_-—-- NEOMERIS group. b*. Dactyls with two hooks approximately equal in width at the base; me- ropodites smooth. c'. Dactyls long and slender; hooks directed with the axis of the dactyl, little curved, the dorsal longer; scale of the scaphocerite always pres- ent: lateral spine of the basicerite slender; stylocerite longer than the basal article of the antennule. d‘. Frontal teeth always longer than wide and spinous; rostrum armed with a vertical prolongation which embraces the ocellary beak. PAULSONI group. d?. Frontal teeth squarish, at most with concave margins; rostrum with- Out, Inferior vertical prolonzsanion======—==—a == BREVICARPUS group. c*. Dactyls short; hooks strongly curved, the ventral directed normal to the lower border of the dactyl; scale of the scaphocerite ordinarily much reduced, often wanting; lateral spine of the basicerite always longer than the basal article of the antennule, thick; stylocerite short. d*. Small claw with a brush of thick and crowded long hairs normal to the dactyl; stylocerite at most equal to the basal article of the antennule; carpus of the small cheliped longer than wide. LAZVIMANUS group. ad”. Small claw without a brush of hairs; stylocerite not reaching the mid- dle of the median article of the antennule; antennal scale narrow, not reaching beyond the extremity of the same article; carpus of thessmaillicheliped|short=== = BIUNGUICULATUS group. The ComatuLarumM group is differentiated from the other groups by some very marked characters, which are almost all characters found in the Hippolytide and therefore suggest a less strong resem- blance to the “ Reptantia;” as frequently happens, there are added to these primitive characters others which show, on the contrary, an adaptation carried very far; for instance, the strongly curved hooks and the movable finger of the small chela surpassing the fixed finger; these characters are especially marked in S. comatularum, and are explained by its commensalism with the Comatulida, being implements of attachment for the Synalpheus. This group of very beautiful species appears not to occur on the American coast; the steamer 4/batross, of the U. S. Bureau of Fish- eries, has collected a new species of it at the Gilbert Islands, the more remarkable because it possesses only a few of the unusual characters of the group: The spines of the basicerite are equal and short, the first segment of the carpus of the second pair very long, and the fol- lowing feet cylindrical. On the other hand, the antennules are equal to the antenn, the rostrum is scarcely more prominent than the 110.1659. AMERICAN SPECIRS OF SYNALPHRUS—COUTIERE. 5 lateral teeth, the ventral hook of the dactyls is almost as strong as the dorsal, and the telson is straight along its posterior margin— all characters not unusual in Synalpheus. The other groups of species vary more or less from the preceding; the nearest are the Pavtsoni and Brevicarrus groups, in which the stylocerite still remains more prominent than the external spine of the basicerite, and the antennal scale is never wanting and sometimes is very wide. The external spine of the basicerite begins to predominate in the Neomeris group, but this group presents, besides, two characters of the Hippolytide but little modified, viz, the movable spines often present on the third and fourth meropodites and the frequent triun- guiculation of the dactyls of the same feet, which is a vestige of the series of spines present on the dactyl in many of the Eucyphota. There are no more than two hooks on the short and stocky dactyl in the BruneuicuLatus group, of which the ventral, the more feeble, has a tendency to become normal to the lower border of the dactyl. On the other hand, the shortening of the stylocerite and of the anten- nal scale becomes very noticeable and the finger of the small chela of the second pair carries a brush of hairs arranged in series. The species grouped under this head are few, but they show the very grad- ual connection between those which precede and the Lavimanus group, the most highly differentiated of the Synalpheids in the di- rection of the ‘“ Reptantia.” Here the antennal scale has very often disappeared without leaving any trace, the lateral spine of the basi- cerite possesses a bulk which contrasts with the slight importance of the stylocerite, and the sexual differences often become very strong, one might say exaggerated, in regard to the size of the abdomen of ‘the female and of the large cheliped of the male; finally, the finger of the small cheliped, in which the carpus has, however, remained more elongated than in any other group, bears a curious structure com- posed of from fifteen to twenty transverse rows of long stiff hairs, which are normal to the dactyl, and which diminishes in length from behind forward; this brush may be a cleansing organ in connection with the very sedentary life of these species, or it may conceal from the prey the extremity of the real prehensile chela. The only compar- able organ in the Alpheide is the tuft of long plumose hairs which is borne by the chela of the second pair in Cheirothria parvimanus Bate, a genus, moreover, very like Synalpheus. 6 PROCEEDINGS OF THE NATIONAL MUSEUM. — vot. xXxxvi. KEY TO THE SPECIES AND SUBSPECIES OF THE GENUS SYNALPHEUS. NEOMERIS group. (Represented on the American coast only by forms with the meropodites smooth and unarmed.) a*. Ventral supernumerary prominence of the third, fourth, and fifth dactyls obtuse and scarcely marked. b*. All the appendages slender. 1 Meropodites of the third pair four times as long as ase d*. Antennal spine equaling the carpocerite_____________ . fritemiilleri. d*. Antennal spine surpassing the carpocerite__S. ees clongatus. c*. Meropodites of the third pair three times as long as wide____S. nobilii. b*. All the appendages stocky ; meropodite of the third pair only 2.5 times as hover en) Wyle ee th 2 eee ee 2 PE Se S. sanlucasi. a*. Ventral supernumerary prominence of the third, fourth, and fifth dactyls spinous and very marked. »b*. Antennary spine equaling the carpocerite___________________ S. hemphilli. b?, Antennary spine surpassing the carpocerite; traces of spines on the MeELOPOGILESTOL whe hinds pai ee S. hemphilli longicornis. PAULSONTI group. * Carpocerite long, arising opposite the separation of the stylocerite from the basal article of the antennule; meropodite of the third pair from 5.5 to 5 times as long as wide. b*. Basicerite unarmed above, carpocerite 3.5 times as long as wide. ec’. Palmar prominence spinous; angles of the telson sharp, the inner spines of its posterior border 3 times longer than the outer spines. d*. Rostrum at most equal to the basal article of the antennulie; anten- nary spine not surpassing the carpocerite_________-_-_ S. townsendi. d*. Rostrum, frontal, and antennal spines more elongate. S. townsendi productus. e*, Palmar prominence obtuse; angles of the telson right, the inner spines only twice the length of the outer___________ S. townsendi brevispinis. b*. Basicerite with a right and not obtuse superior angle; posterior angles of thestelsonayveryy sharp = ee eee S. townsendi mexicanus. b*. Basicerite strongly spinous above; carpocerite ovoid, three times as long as wide. ; c*. Meropodite of the third pair less ian four times as long as wide. d*, Dactyl of the third pair about 3.2 times as long as wide at the base. . Carpus of the small cheliped spinous above; spine of the scapho- cerite equaliio the carpocerites22 22 S. apioceros. e*. Carpus of the small cheliped not spinous above; spine of the scaphocerite shorter than the carpocerite___S. apioceros sanjosei. d*, Dactyl of the third pair 3.8 times as long as wide; carpocerite very swollen, surpassing the antennule by 1.5 times its distal article; spine of the large claw continuing in a straight line its superior borders: 2232 Sees = to ney ee ore cee S. apioceros mayaguensis. . Meropodite of the third pair more than four times as long as wide. Mies . Antennal spine longer than the antennule; rostro-orbital interval acute at base; spine of the large claw preceded by a tubercle. S. apioceros leiopes. d*, Antennal spine at most equal to the antennule, rostro-orbital inter- val broad and sinuous at base; spine of the large claw continuing the superior border in a straight line____S. apioceros desterroensis. no. 1659. AMERICAN SPECIES OF SYNALPHEUS—COUTIDRE. 7 a*, Carpocerite short, arising nearly opposite the median antennulary article; basal article of the antennule equal to the following, or 1.5 times as long, or more; palmar border of the large cheliped unarmed. b+. Carpocerite three times as long as wide. e*. Meropodite of the third pair more than 3.5 times as long as wide; large chela 2.9 times as long as high; antennal spine equal to the carpo- EGIIGD spe ey A a a aed ae ere ew ee La S. lockingtoni. e*. Meropodite of the third pair less than 3.5 times as long as wide; large chela 2.5 times as long as high; antennal spine longer than the Rano cenilomatenders..- =. oa S. latastei tenwispina. b*, Carpocerite less than three times as long as wide; antennule thick, scarcely four times as long as wide; meropodite of the third pair three OMe eae OMe meses 2 eo StS ee ee S. latastei. b*. Carpocerite more than three times as long as wide; meropodite slender, five times as long as wide; spine of the scaphocerite surpassing the Pie Cee ee eee ee re ee ek S. paulsonoides, BREVICARPUS group. a*, Carpocerite cylindrical, slender, at least four times as long as wide; antennal seale from 5.5 to 6.4 times as long as wide; basicerite almost unarmed above; meropodite of the third pair from 4.25 to 4.5 times as long as wide; carpus of the second pair from ten to fifteen times as long as wide; telson 2 to 2.2 times as long as wide distally. b*. Lateral spine of the scaphocerite scarcely surpassing the scale; frontal teeth equilateral, short; eggs of large size, producing mysis. S. brevicarpus. b*. Lateral spine of the scaphocerite long and slender; frontal teeth as long as wide at the base; rostrum with concave margins; eggs small, PORES RC aE OTe S. brevicarpus guerini. a*, Carpocerite swollen, from 3.5 to 3.7 times as long as wide; antennal scale from 7 to 8.5 times as long as wide; carpus of the second pair less than ten times as long as wide; meropodite of the third pair 3.4 to 4 times as long as wide; telson 1.8 times as long as wide distally; larvie zoéze. b*. Antennules at most five times as long as wide; rostrum as long as the lateral teeth of the front. c*. Frontal teeth short, equilateral. d*., Superior spine of the basicerite feeble, as wide as long, that of the scaphocerite shorter than the carpocerite; small claw with elon- gated fingers, 2.8 times longer than wide____-________~_ S. minus. d*. Superior spine of the basicerite twice as long as wide, that of the scaphocerite equal to the carpocerite; small claw with short fin- gers, 2.6. times as long as wide____________-_ S. minus bahiensis. c*. Frontal teeth long, a little concave; carpocerite very swollen, from 3.2 to 3.5 times as long as wide; scaphocerite short; hooks of the Gaeuylsnoment Cite ee 28 ee S. minus antillensis. b*. Antennules more than five times as long as wide; rostrum shorter and narrower than the lateral teeth. c*. Carpocerite 3.5 times, meropodite of the third pair 3.4 times, as long AS) (MANGAS. ook ey ee ee een eee S. digueti. ce’. Carpocerite 3.2 times, meropodite 3.25 times, as long as wide. S. digueti ecuadorensis. 8 PROCEEDINGS OF THE NATIONAL MUSEUM. vot. xxxvi. LEVIMANUS group. a*. Carpus of the small cheliped measuring always more than one-half of the chela in the adult (proportion included between 0.54 and 0.8, but may be reduced to 0.5 in the young). b*. Lateral spine of the basicerite smaller than that of the scaphocerite. c*’. Fingers of the small chela each armed with three strong flat teeth, crossed in a vertical plane; no trace of antennal scale; spine of the scaphocerite shorter than the antennule; movable finger of the chela out of the perpendicular; eggs of large size____________ S. pectiniger. c*, Fingers of the small chela with only two teeth; spine of the scaphocerite equaling the antennule. d*. A trace of an antennal scale in the male; carpus of the small cheliped reaches 0.74 of the chela, diminishing to 0.5 and even a little less inthe; young} => 2) 2s = ees S. longicarpus. ad”, An antennal scale in both sexes; carpus of the small cheliped not exceeding 0.54 in the adult males________ S. longicarpus approxima, b?. Lateral spine of the basicerite equal to that of the scaphocerite, but both shorter than the antennule; antennal scale totally absent. c*. Carpus measuring from 0.67 to 0.8 of the small chela. d*. Proportion of T. L.: H.4=2.6:1 to 2.5:1; meropodite of the small cheliped 3 to 3.3 times as long as wide__---—_-__-=-—— S. herricki. d*, Proportion of T. L.: H.=2.7:1, the chela being more slender; meropodite of the small cheliped four times as long as wide. S. herricki dimidiatus. c*, Carpus measuring 0.65 of the small chela; proportion of T. L.: ds es Sen aa ee renee SR Toes ee a S. herricki angustipes. c*, Carpus measuring 0.56 of the small chela, proportion of T. L.: H.= 3.2:1; carpocerite 5.4 times longer than wide (instead of 4.8 to 5); feet of second and third pairs one-fourth stronger than in the pre- ceding formss:= 222k Se ee ee ee S. tanneri. a”, Carpus measuring about one-half of the small chela in the adult (0.45 to 0.52). b*. A well developed antennal scale in both sexes; ecarpocerite 6.5 times longer than wide; meropodite of the third pair 3.8 times longer than wide; small chela slender, proportion of T. L.: H.=3.33:1. c*. Spine of the basicerite and of the scaphocerite shorter than the anten- MUG 2. es oie See Se Se eee SE eee S. pandionis. c”. Spines equal to the antennule___________________ S. pandionis extentus. b*. No antennal spine. c*. Basicerite spinous above; carpus of the second pair with four seg- ments. 3. tees ae ee Be eee S. rathbune, c*. Basicerite unarmed above. d*, Large chela 2.5 times longer than’ wide, its anterior palmar spine directed obliquely downward; supraorbital spines wide, leaving between them and the rostrum U-shaped intervals; carpocerite 5.5 times longer than wide; meropodites of the third pair thick (proportion 3.3:1); eggs small; larvee zoé#__________ S. grampusi. d*. Large chela 2.7 to 3.25 times longer than wide; anterior palmar spine conical, directed obliquely upward; supraorbital spines ob- tuse, divergent; meropodite of the third pair slender (proportion 4.3 :1 to 4.5:1) ; eggs of large size; larvee mysis. “Total length to height (of large chela). xo. 1659. AMERICAN SPRCIES OF SYNALPHEUS—COUTIRRE. 9 e*, Carpus of the small cheliped measuring 0.5 of the chela. f*. Carpocerite 4.5 times as long as wide________________ S. brooksi. f*. Carpocerite 5.5 times as long as wide____ S. brooksi strepsiccros. e*, Carpus of the small cheliped measuring 0.53 to 0.57 of the chela; carpocerite 4.4 times as long as wide______ S. brooksi eleuthere. a*, Carpus measuring less than 0.5 of the small chela (0.43 to 0.4). b*, Meropodite and carpus of the third pair excavate, with a transparent UTES TNS REN es ee ee ee Se ee os al 2) ee S. androsi. b*. Meropodite of the third pair not excavate; carpus shorter than the pro- podite. c*. Brush of hairs of the small chela very reduced (about 30 hairs in 6 rows) ; antennal scale present in the male only______ S. paraneptunus. c’*. Brush of hairs of large size, composed of 15 to 20 rows. d’*. Carpocerite 3.5 to 4 times longer than wide; meropodite of the third pair 4 to 4.5 times longer than wide_______________ S. sanctithome. ad”. Carpocerite 5.2 to 6 times, meropodite 3 to 3.5 times, longer than wide. e*. Antennal scale present in both sexes___________________ S. goodei. e*, Antennal scale absent (or extremely narrow when it is exception- MMU Reo OH ET SYe) OY] §) PRE wae eek: eS, BAS eee SR She ea S. goodei occidentalis. DISTRIBUTION AND DISCUSSION OF SPECIFIC CHARACTERS. Table of distribution of the species of the genus Synalpheus. COMATULARUM GROUP. Mediterranean and West Indo-Pacific forms. American forms. African forms, comatularum Haswell. stimpsoni de Man. stimpsoni maldivensis Coutiére. carinatus de Man. . amboine Zehntner. . dbatrossi Coutiére. ratatatatntn NEOMERIS GROUP. S. neomeris de Man. S. neomeris streptodactylus Coutiére. S. nilandensis Coutiére.:.....------ S. hemphilli Coutiére. S. nilandensis oxyceros Coutiére ....| S. hemphilli longicornis Coutiére. S. graviert Coutiere. S. pococki Coutiére. ° S. merospiniger Coutiére. a S. fossor Paulson. S. trionychis Coutiére. S. triunguiculatus Paulson. RMOMKET? COULOCTE socio ccccsecceces S. nobilii Coutiére. S. physocheles Coutiére. S. demani Borradaile. S. charon Heller. S. otiosus Coutiére. S. paraneomeris Coutiére........-.- S. fritzmiillert Coutiére. S. paraneomeris prolatus, new name | S. fritzmiillert elongatus Coutiére. (=S. parancomeris oxyceros Cou- tiére). mameron? COUTIETe ....-. 2 ..-.-2.--20- S. sanlucast Coutiére. a¥or descriptions of this and other new extra-American species, see pages 89 to 93. 10 PROCEEDINGS OF THE NATIONAL MUSEUM. VOL. XXXVI. Table of distribution of the species of the genus Synalpheus—Continued. PAULSONI GROUP. Indo-Pacifie forms. American forms. Mediterranean and West African forms. CLISOIENO DIMras soe sees et paulsoni liminaris Coutiére, — paulsoni rameswarensis Coutiere. paulsoni kurracheensis Coutiéere. . S. paulsonoides Coutiére. ™m culus Lockington). ‘. lockingtoni, new name (A. levius- S. hululensis Coutiere. tn omanis) PAmISOM a1 Vettes |r emesis eisai ete aes eteteta erate S. paulsoni senegambiensis ecilimanus Paulson? Coutiére. S. mushaensis Coutiére. Seana = be 5 S. latastet Coutiére. S. maccullocht Coutiére.....-.--.-.- {fF latastei tenuispina Coutiére. S. latastei Coutiére. S. acanthitelsonis Coutiére. S. apioceros Coutiere. S. apioceros sanjosei Coutiere. S. apioceros mayaguensis Coutiére. S. apioceros leiopes Coutiére. S. hastilicrassus Coutiére.......-.-- S. apioceros desterrocnsis Coutiéere. S. townsendi Coutiére, S. townsendi productus Coutiére. S. townsendi mexicanus Coutiére. S. townsendi brevispinis Coutiere. S. tricuspidatus (Heller). BREVICARPUS GROUP. S. brevicarpus (Herrick). S. brevicarpus guerini Coutiere. S. minus (Say). S. minus bahiensis Coutiére. S. minus antillensis Coutiére. S. diqueti Coutiére. S. digueti ecuadorensis Coutiére. BIUNGUICULATUS GROUP. S. biunguiculatus (Stimpson) . (2?) S. spiniger (Stimpson). S. biunguiculatus exilipes Coutiére. S. biunguiculatus pachymeris Cou- tiére. S. neptunus (Dana). S. laticeps Coutiére. S. pescadorensis Coutiére. S. lophodactylus Coutiére. (?) S. haddoni Coutiére. . LAVIMANUS GROUP. ry S. sladeniCoutiére. S. spinifrons (M. Edwards). (?) S. longicarpus (Herrick) .......-..-- S. longicarpus approxima Coutiere. S. goodei Coutiére. S. goodei occidentalis Coutiére. S. pandionis Coutiére ..........---- S. pandionis extentus.Coutiére. S. grampusi Coutiére. S. sanctithome Coutiére. S. brookst Coutiére. S. brookst strepsiceros Coutiére. S. brooksi eleutherz Coutiére. S. herricki Coutiére. S. herricki dimidiatus Coutiére. S. herricki angustipes Coutiére. S. tanneri Coutiére. S. pectiniger Coutiére. S. androst Coutiére. S. rathbunzx Coutiére. S. paraneptunus Coutiére. S. levimanus (Heller). S. parfaiti Coutiere, xo. 1659. AMERICAN SPECIES OF SYNALPHBUS—COUTIDRE. 11 The preceding table shows the relative importance of the several groups inhabiting the American coasts, and enumerates, without description, all the species of the genus Synalpheus which are known to me. There are also included a certain number of species, yet unpublished, from other localities. This table brings out well an important fact; that is, that the species of the Pautsonr and Nromeris groups, especially the former, are the most widely distributed. This distribution accords with cer- tain unspecialized characters which are found among them, namely, the short carpocerite, the rostrum always possessing an inferior vertical prolongation, continuing the ocellary beak, and the dactyls slender and elongated. The disappearance of the rostral partition, the elongation of the carpocerite, thick or not, and the shortening of the dactylopodites indicate forms less and less allied to the Hippo- lytidz and more and more Synalphean. Before examining more closely the relation of the groups of species to one another, I ought to mention that many of the forms described have received trinomial appellations and correspond consequently to what zoological nomenclature designates as varieties, races, or sub- species, and to that which the botanists know under the name of “ petites espéeces.” In employing the trinomial name to designate cer- tain forms allied to one another, I simply wish to say that the forms represented by these names appear to me to be less distant from the species to which I attach them than the species is from another species. Most often these subspecies come from different localities and appear also to be distinct geographical races. This is, however, probably a result of the fact that the localities cited were the only ones explored. One should not forget that the stations noticed, rather numerous in Florida and at the Bahamas, for example, are restricted to a few points on the thousands of miles of shore line along the entire Pacific coast and even on the Atlantic coast south of Florida. At other times these secondary forms, related to species easy to determine, come from the same locality as the species. This expression “ same locality,” in spite of its apparent precision, is most often very vague. Two closely allied forms can find upon the same reef very different conditions of life, which isolate them as completely as if a continent separated them. One species inhabiting a sponge, another living attached by its hooks upon some species of madrepore, appear to me to represent a case of this sort. It is possible, then, that the trinomial appellation which I uniformly employ does not correspond in nature to facts exactly comparable; that certain of the “races,” “ subspecies,” or “small species” which it serves to designate merit a distinct specific name; that others, on the contrary, may be variations incompletely fixed of a species in a state of actual instability, the limits of which it has not been possible for me to fix more completely. 12 PROCEEDINGS OF THE NATIONAL MUSEUM. VOL. XXXVI. To return to the groups of species of the genus Synalpheus, the one which I designate by the name of the Paunsonr group, presents a most remarkable geographical distribution. S. pauwlsoni Nobili (per- haps identical with S. tricuspidatus (Heller) ) is a species with short carpocerite from the Red Sea and the Persian Gulf, the affinities of which are to my mind very clearly indicated; it is separated, first from the forms with carpocerite equally short, but distinguished by the spinous palm of the large chela, or by the basicerite almost un- armed above, and again from the forms with carpocerite more elon- gate; variation in this last direction leads to some forms with the carpocerite elongate and slender (S. hululensis Coutiére, S. tumido- manus Paulson, the latter very distinct on account of its large eggs, producing mysis, and the spinous angles of the telson). In another direction there are found some forms in which the carpocerite is elongate as in those preceding, but, in addition, is swollen, and of an ovoid form (S. acanthitelsonis Coutiere, 8. hastilicrassus Coutiere). In a third direction, finally, there are found some species differing from S. paulsoni by the more massive aspect of the appendages; though indicated by a form kwrracheensis, this evolutional tendency is more accentuated in the species S. datastei from Australia, which occurs without change in Chile, and which is also represented in Brazil by the form tenuwispina. In Australia again, the species S. maccullochi Coutiére differs most markedly from S. paulsona kur- racheensis by the presence of large eggs producing mysis. There are found forms derived from S. paulsoni at the Mascarene Islands and on the west coast of Africa, of which I have been able to study very unusual specimens from Cape Lopez and from Cape Verde; these are not strictly typical specimens, but can be separated only by careful study, and it is impossible for me to make them dis- tinct species, in spite of the great actual geographical isolation, which can probably be considered as absolute. On the American coasts are found exactly the same evolutional tendencies in this group; S. paulsoni and the other Indo-Pacific forms are not represented there by identical forms, but the differences are at times so slight that, without indication of locality, the identifica- tion would be very difficult. S. lockingtoni differs from S. paulsoni almost. solely by the spine of the scaphocerite being longer in the latter and surpassing the carpocerite; with the exception of the place of origin, the second species would correspond to the “ owyceros” form so often met with that it appears to be almost a constant variation among the subspecies of a given species.? “Tt is more convenient and expressive to designate by the name “ oxvyceros ” every subspecies showing this variation, but in deference to the accepted rule of nomenclature which forbids duplication of names within a single genus, I have in this paper used different names having a similar meaning, as longi- cornis, elongatus, productus, prolatus, extentus. no.1659. AMERICAN SPECIES OF SYNALPHEUS—COUTIERE. 13 5 This “oxyceros” form of S. lockingtoni exists, moreover, on the coast of Lower California. It is at present represented only by a sin- gle mutilated specimen collected by M. Diguet (Paris Museum), the characters of which I believe to have specific value, and which I des- ignate by the name S. paulsonoides. Compared to S. paulsoni, in- stead of S. lockingtoni, it differs by its appendages and especially the third pair of feet, which are more slender. This comprises the known variations in this direction, the carpocerite remaining short, with the exception, of course, of S. datastei of Chile and its tenuispina form from Brazil just now cited. The second evolutional direction (the carpocerite remaining slen- der while becoming much elongate) does not appear to be represented on the American coast; on the other hand, the forms with carpocerite long and swollen are predominant. S. apioceros Coutiére is found here, accompanied by at least four varietal forms coming from California, Florida, the West Indies, Venezuela, and Brazil; in re- gard to the last four forms, one can not say whether the geographical isolation is real, or whether it only appears so because the interme- diate connectives are not known; but the isolation of the Californian form is absolute, and yet the differences which separate it from the specimens from Florida are quite as slight and difficult to detect as are those which separate these last from the other three forms which the species assumes. While it is possible that there may be as many species quite distinct and unrelated to one another, yet it is absolutely undeniable that the characters of these five forms are less distant than are those which separate S. apioceros from S. paulsoni. Nomencla- ture does not permit the expression of this, discarding (with good reason) every hypothesis conveying the idea of a possible affiliation between species—afhiliation to which the idea of these “ lesser species ” directly leads. Another species with carpocerite less swollen is S. townsendi, very widely distributed on both shores of America. It differs most of all trom the preceding in the superior spine of the basicerite being almost or quite absent, so that these two species show a striking parallelism with S. acanthitelsonis and S. hastilicrassus Coutiére of the Mal- dives. S. apioceros and S. acanthitelsonis especially differ only in the angles of the telson, which are very sharp in the second species. The divergence is but little greater between the two others. Those species with basicerite unarmed above are important from another point of view, in that they permit of a passage to the Nromerts group by certain forms, such as S. paraneomeris Coutieére. This species, very widely distributed from the Red Sea to the Hawaiian Islands, with several “races” (probably among them the “ oxyceros” form), differs particularly from the American S. town- sendi or from 8. hastilicrassus of the Maldives by the dactyls of the 14 PROCHEDINGS OF THE NATIONAL MUSEUM. VOL. XXXVI. third, fourth, and fifth feet bearing a third and very obtuse ventral prominence. This prominence, very frequent in the NEoMErIs group, sometimes becomes a strong triangular hook, but a character still more often met with (with or without the last) is the reduction of the dorsal hook of the dactyl, the ventral hook taking the form and thickness of a wooden shoe, as in the curious species S. charon Heller. The Indo-Pacific region harbors a large number of species of this group, very easy to distinguish by reason of the characters in the striking form of the dactyls; among them those are especially curious in which exist—or reappear—the rows of spines of the meropodites, so constant in the primitive Eucyphota, and persisting in the Alpheide only on the propodite. E These forms of the Nromeris group with spinous meropodites appear to be totally absent from the American coasts. There also the parallelism with the Indo-Pacific forms is carried very far. Be- tween S. hemphilli Coutiére of Florida and S. nilandensis Coutiére of the Maldives (both of which show an “ owyceros” form) there is no difference except the presence in the latter, and the absence in the former, of the meral spines; and yet, oddly enough, one quite full-grown specimen of S. hemphilli, from the Bermudas, shows on one side only a single meral spine. It is difficult to record such ob- servations without thinking of a common origin for the two species, so completely separated at the present time, and yet so strictly parallel. Another species, S. fritzmiillert Coutiere of Florida and the West Indies, exists also in Venezuela and Brazil, where it is represented by the “owyceros” form; it is found again in California with the meropodites more slender. In Ecuador S. nodilii Coutiére, a form in which the meropodites are, on the other hand, shorter and more swollen, replaces S. fritzmiilleri. In the Indo-Pacific region, S. bakeri Coutiére, in which the supernumerary ventral prominence is very feeble, is the species most closely allied to the preceding ones. Another very striking instance of parallelism is the existence in Lower California and in the Red Sea of the species S. sanlucasi Coutiére and S. heroni Coutiére respectively, both characterized by the very massive form of all the appendages. The modes of differentiation of the species are again repeated. A form being found, for example, with a short carpocerite, one may expect to meet those with a long carpocerite, then those with more slender members, with more massive members, with basicerite spinous above, or not, with large chela unarmed or spinous on the palmar border, those in which the angles of the telson are prolonged in a spine or not, ete. One can not avoid drawing the conclusion that such a constancy in the modes of variation strongly resembles an hereditary tendency, due to the small number of species from which the genus Synalpheus must have sprung. ne no.1659. AMERICAN SPECIES OF SYNALPHEUS—COUTIPERE. 15 The Brevicarrus group is extremely like the Paunsonr group, and it is very possible that it may have been derived from the latter ; the absence of an inferior vertical prolongation of the rostrum, the fact that all the species known have the carpocerite elongated, often swollen, indicate derivative and not primitive forms. On the other hand, it is there that are found the species having the largest antennal seale (S. brevicarpus Herrick), which is a character directly opposed to the preceding; but S. brevicarpus is a species with very large eggs producing mysis, a character which appears especially to show itself in. those forms which are most highly developed. Finally, the BrevicarPus group is, up to the present time, exclusively Ameri- can. Should its presence be established in some part of the Indo- Pacific—in Australia, for instance—the American region would surely remain its true country. Compared to the preceding groups it has an unusual distribution, inasmuch as the species which com- pose it extend from the Bermudas to Brazil and from California to Ecuador, exactly lke the American forms of the PatLsonr or Neomeris groups. These last appear to have had rather an Indo- Pacific origin, if one may judge by the number and variety of the forms which represent them in that region. One is led then to wonder if the origin of the Brevicarrus group ought not to be sought for also in some of the species of the Pauxsoni groups; this might have found only in American waters the conditions which have brought about its variations in the direction of the Brevicarpus group. Being given the form brevicarpus, the most typical species with small eggs is S. minus (Say), from which one is able to derive an “ oxyceros” form from Brazil, an antillensis form with antennules short and carpocerite more swollen, and a form with antennules very long and slender, S. digueti Coutiére from Lower California. A form of this species, S. ecwadorensis, exists upon the Pacific coast of South America. The species S. brevicarpus (Herrick) with very large eggs differs in its more slender carpocerite and larger antennal scale. It is the largest species of Synalpheus known. All these species are separated by slight differences. It is probable that S. minus (Say) comprises several “races” other than those indicated here and be- haves as does 8. paulsoni in the Indo-Pacific region. The BruneuicuLtatus group includes among its species S. nep- tunus (Dana), the types of which I have been able to examine, and specimens of which have also been sent me from Australia by Mr. McCulloch. S. laticeps Coutiére, of the Maldives, is very like it. These two species have the finger of the small chela widened and spatuliform, ornamented besides with some long hairs, which in S. neptunus are arranged in rows. These hairs are directed obliquely downward, and isolated, which is not the usual disposition in the 16 PROCEEDINGS OF THE NATIONAL MUSEUM. VOL. XXXVI. group. InS. biwnguiculatus Stimpson, in the forms pachymeris Cou- titre, lophodactylus Coutiére, and pescadorensis Coutiére, these hairs are disposed in tufts formed in line, directed obliquely downward or even perpendicular to the surface of the mobile finger (S. lopho- dactylus Coutiére) ; at the same time, the chela terminating the sec- ond pair is provided with a large number of tufts of long hairs, some carried by the palm, especially on its lower face, others by the mov- able finger, these last arranged regularly in a brush. Judging by the sketch of an Australian specimen at the British Museum, taken some time previously, S. spiniger (Stimpson), with re- curved hook, would also fall in this group, but this is not significant at the present time. No American species, unless it be S. spinifrons (H. Milne Edwards), belongs to the BruneurcuLatus group; this species can be equally claimed by the Lavimanus group, this group, as I have pointed out, being the continuation, pure and simple, of the preceding one, the characters of which it extends to the extreme limit, especially those which concern the reduction of the antennal scale. The character which essentially distinguishes the two groups, in spite of its con- stancy and its importance, is itself only the variation in the cleansing apparatus present in the species of the BruneuicuLatus group; the bristles of the chela of the second pair, especially those of the movable finger, have not persisted in the Lavrmanus group, while the brush of the little chela of the first pair has acquired the quite remarkable development that I have described above. If one tried to express these facts in plain language one would say that the experiment of the numerous cleansing appliances had been abandoned in the de- scendants of certain of these species, and that a single apparatus, much more perfect, had been substituted for them. The continuity of the two groups is so evident that it leads to this conclusion: If the species of the BruneuicuLatus group are not rep- resented on the American coasts, it is because they have all undergone the variation toward the Lavimanus group, with the exception, per- haps, of S. paraneptunus Coutiere, a species particularly instructive because of the much more feeble development of the brush of bristles of the small claw. In the Indo-Pacific region, on the other hand, this variation ap- pears to be very rarely realized, since the only species which presents it up to the present time is S. sladeni Coutiére, which has not, how- ever, altogether the aspect of the American species of the Lavimanus group. A rather similar condition is observed in the relations between the Pautsonr and Brevicarrus groups, with two differences: First, the species of these two groups are still found side by side on both American coasts; second, the known distribution of the Lavimanus no. 1659. AMERICAN SPECIES OF SYNALPHEUS—COUTIERE. re group is at present much less restricted than that of the Brrvicarrus group, since it is known in the Indian Ocean, on the west coast of Africa, and in the Mediterranean. If the idea of time could be intro- duced into these data, one would say that the differentiation of the BrevicarPus group is more recent; thus would also be explained its presence exclusively in America as well as its coexistence with the Pautsoni group, from which it has come, and its relatively small number of species. All these characters are opposed to the much wider distribution of the La:vrmanus group, to the localization of the derived species of the Bruneuicunarus group and to the exces- sive development of the forms which characterize the Lavimanus group. The armature of bristles of the small claw gives to these forms a common aspect so characteristic that it can hardly be believed, in a superficial examination, that there is room for so great a number of species within a compass of differences apparently so slight. It is, however, probably true even further than I have indicated, that several of the races or subspecies have a specific value. On the whole, one may say that the species of this group tend toward the elongation of the wrist of the small claw and the suppression of the antennal scale. These two tendencies are met with occasionally within the same form, such as S. longicarpus, in which the length of the wrist varies from once to twice the width with the age of the specimens. But there are grafted onto this general plan some characters quite unexpected, such as the curious form of the fingers of the small claw in S. pectiniger, the excavated meropodites of S. androsi, and the basicerite with a longitudinally spinous superior surface, of S. rath- bune. No other group gives the impression as does this one of hav- ing sprung from a single species, by “explosion” of its characters (to employ the expression of Standfuss), characters which might be regrouped by chance like a combination of letters. Remarkable from a morphological point of view, the Lavimanus group is no less remarkable as to conditions of existence. It is the only one in which a single haul of the dredge of the U. S. Fish- eries steamer Albatross (Station 2413) has been able to bring up from 5,000 to 5,000 specimens (belonging to the two species, S. long?- carpus (Herrick) and S. pectiniger Coutiére). It is this group in which anomalies in the laying of eggs are met with most frequently ; among 227 females of S. pectiniger (Station 2413), of which I have determined the sex by examining them singly with the greatest care, I have been able to find only two or three in which the pleura were normal and the eggs present, and have been able to find none with the very large eggs carried by the normal females. The males which accompany them are 320 in number, with some closely united, all inferior in size to the normal. I propose to conduct investigations Proc, N, M. vol. xxxvi—09 2 18 PROCEEDINGS OF THE NATIONAL MUSEUM. VOL. XXXVI. to determine whether the castration so general among these females is due to a parasite, Microsporidian or Bacterian, or simply to hun- ger. Neither do the S. longicarpus which accompany them possess their maximum size; the males are largely in excess, but the eggs of the female are altogether normal. One frequently finds some anomalies of the same sort in S. brooksi Coutiére, in which the eggs give rise also to mysis as in S. pectiniger. The eggs may be reduced to two or three; they are thén very small and of a chalky aspect, and at the same time the size of the female is very much reduced. S. rathbune is known to me up to the present time only by some sterile females, very small, in which the pleura are extremely spinous as in the male. These species appear, then, to be sometimes found in very precari- ous conditions, from the point of view of their perpetuation, and the study of these conditions would probably be most interesting ; their abundance in the collections accords well with what Herrick says when he speaks of the constant fusillade which one hears on the veefs of the Bahamas from the movements of the Alpheids, and it seems to me that they would lend themselves to constant observation in an aquarium in such a manner as to make possible some “ pure cultures ” of a determined species. After an examination of the facts, it is difficult to avoid the temp- tation to draw from them some hypothetical conclusions. When one investigates the distribution of the Synalpheids known at the present time, the most striking fact is the existence of forms almost identical in regions so remote as the Red Sea and California or Florida. Now, these are very sedentary animals, which are almost never seen to swim, but live in couples in sponges or madrepores; their larvee, to be sure, could be disseminated by the currents, but the possible extent of that dissemination should not be overestimated, and when two spe- cles are separated by both the Pacific and the Indian oceans, they are certainly isolated in the most rigorous fashion. I can not repeat too often that the Indian S. paulsoni and the Californian S. paulson- oides, S. mushaenis and S. lockingtoni, S. acanthitelsonis and S. aptioceros, S. nilandensis and S. hemphilli, ete., might very well, with- out indication of locality, be considered as simply “ races,” and there is the inevitable inference that former conditions under which the antecedent species lived permitted a very vast distribution. Nothing shows that these species still exist, but at all events they have changed, since the forms which represent them in different localities are no longer exactly comparable. There is, as an exception, only the single species S. latastei, the specimens of which from Chile I can not differentiate from another—a single one, it is true—from Austra- ha. This exception, when critically examined, only goes to strengthen ss wo.1659. AMERICAN SPECIES OF SYNALPHEUS—COUTIERE. 19 very much the idea of species primitively widely dispersed; perhaps it will even render less uncertain the position of the continuous line of coast along which the dispersion might have been made. It seems to me that one might sketch this original distribution under the form of a few waves of very great amplitude, on which, at variable points, might have originated new systems of waves of the second order, of less amplitude. S. paulsoni at one extremity, S. paulsonoides at the other, would represent such systems, turning aside more or less from the wave of the first order, to continue the comparison, being able even to substitute themselves for it and to efface it. On these would be produced, by the same hypothetical mechanism, waves of the third order, of still shorter amplitude, as, for examples, the races or sub- species with trinomial appellations which are attached so obviously and so closely to S. apioceros, to S. townsendi, to S. minus, and to S. herrickt. The comparison permits us even to imagine that the characters served at first to distinguish species, and that as the waves spread and multipled they changed their original valuation and became characters of groups, and even generic characters. In order to complete the hypothesis, one might speak of the “ peb- ble” which, falling on the summit of a wave of great amplitude, might have given rise to a new system of vibrations, otherwise called anew form. If one seeks to represent the one or the other of the two possible mechanisms, insensible “ fluctuations” or sudden “ muta- tions,” one encounters the same impossibility of knowing. It is easy to see at a glance that the two modes do not exclude each other, that they are even very near to overlapping, provided we admit an ampli- tude small enough for effectual variations. There are perhaps some zoological groups which behave in a different manner from this point of view. Each molt of an arthropod is a “ mutation,” while a verte- brate “ fluctuates” in order to attain its adult characters. According to Professor Bouvier, who has in such a masterly manner demonstra- ted the reality of mutations among the Atyide and their great ampli- tude, it would not be surprising if one found among other Crustacea analogous examples. I think I can say that the Synalpheids at the present time show nothing similar. I have examined, drawn, and measured all the specimens of which I speak, excepting in the cases where the species comprised several hundreds or more than a thousand specimens, -which for want of time I have only examined. The details which ean be referred to the facts of mutation by their unusual presence in a series of specimens appear to be very few in number and without special importance. For example, first, one of the specimens of S. apioceros sanjosei has no spine on the anterior border of the wrist, which fact permits of no hesitation in its determination; second, one 20 PROCEEDINGS OF THE NATIONAL MUSEUM. VOL. XXXVI. of the specimens of S. hemphilli longicornis bears a movable spine on one of the meropodites of the third pair; it is the reappearance, very interesting, of a character present in many of the Nromerts species (Indo-Pacific), which seems to have disappeared in the American species, even those most like the preceding. It is not a “mutation” permitting one to understand the process by which a new form is originated; third, a specimen of S. mznus possesses on the anterior border of the palm of the small chela a spinous tubercle as on the opposing chela; this is a “ mutation ” which is not absolutely rare in the Alpheidx, and which I have seen even carried so far as to result in the complete symmetry of the two claws of the first pair in a very curious specimen of Alpheus dentipes Guérin. But the Alpheide have originated from forms with symmetrical claws; there, again, it is the question of the recurrence of a remote character, and not the indication of a new evolutional line; fourth, some specimens of very small size of S. Zongicarpus, and of S. brooks also, have only four segments in the carpus of the second pair; this detail character- izes the species S. rathbunw, which is far from being the nearest to S. brooksi, but it characterizes also the genus Arete, the relations of which with the genus Synalpheus are truly very remote. Concurrent with these facts, obvious, but without importance, may be cited other facts of greater weight but without proofs. The char- acters of the subspecies S. brooksi strepsiceros, S. herrichi dimidiatus, S. herricki angustipes, and of S. tanneri, occurring either among a series of typical specimens or in some localities where the typical specimens also occur, make one think of some “ mutations; ” it is, In fact, as I have said before, the case in the entire LasvimMaNnus group; close relationship, a general resemblance of the forms which make part of it, but a great variety of combinations of a small number of characters, of which one at least is absolutely constant, certain combi- nations rarely realized, while others are frequent—even taking into account certain errors in the appreciation of the scarcity or the fre- quency of the type—a distinct aspect and clear-cut, though slight, differences. These are mere impressions without proofs, but which, I believe, would occur to every naturalist who has been able to study in its entirety S0 homogeneous a group; and it would certainly be interesting to attempt for a few of the species of the Lavimanus group some “ pure cultures,” continued during several generations, supposing that one might surmount certain considerable difficulties involved in such an attempt. This work, in which the greater part of the forms described are new, necessarily allows of very few bibliographical references; those pertaining to the species of Say, Herrick, and Lockington, the names of which I have been able to retain, are given with the descriptions xo.1659. AMERICAN SPECIES OF SYNALPHEUS—COUTIDPRE. 91 of these species; for all the others, I refer to my work on the Alpheide in general,‘ and to the paper on the Alpheidz of the Mal- dives.? DESCRIPTIONS OF SPECIES. PAULSONI Group. SYNALPHEUS LOCKINGTONI, new name. Alpheus leviusculus LockineTon, Ann. and Mag. Nat. Hist., 1878, p. 478; not Alpheus edwardsi var. leviusculus Dana, 1852. I believe that I have rediscovered the species described by Locking- ton, although the specimens which represent it differ in shght details from his description. Fic. 1.—SYNALPHEUS LOCKINGTONI. ad, FRONTAL AND ANTENNAL REGION; C, CARPOCERITE ; K, LARGE CHELA; Kk’, SMALL CHELIPED OF FIRST PAIR; 1, FOOT OF SECOND PAIR; m, FOOT OF THIRD PAIR; ™’, DACTYL OF THIRD PAIR; ¢t, TELSON. The rostrum is a little longer than the lateral spines; it reaches the extremity of the basal antennular article, from which it is separated by notches which are narrow, but not sharp, at the base. The last two antennular articles are practically equal, the anten- nule being 4.4 times as longas wide; the basal article is only 1.5 @ Annales des Sciences Naturelles (8), IX, 1899, pp. 1-56. >The Fauna and Geography of the Maldive and Laccadive Archipelagoes, II, Pt. 4, 1905, pp. 852-920, 99 PROCEEDINGS OF THE NATIONAL MUSEUM. VoL. XXXVI. longer than the median; the stylocerite usually reaches the distal third of the median antennular article, but always to at least the middle of the article. The lateral spine of the basicerite is as long as the rostrum; the lateral spine of the scaphocerite is very slightly shorter than the carpocerite, which is short, beginning at the distal third of the basal antennular article, three times as long as wide, the margins almost parallel, excepting at the base, where it is shghtly swollen; it ex- ceeds the antennule by one-half the length of the distal article; the outer maxillipeds exceed the antennule by about one-half of its length. The anterior margin of the palm of the large chela terminates in a conical tubercle, short and always destitute of a spine, as Locking- ton distinctly says. I found the proportions of the chela to be: Finger 1; total length 3; height 1.1; proportion T. L.:H.=2.9:1. The anterior margin of the meropodite terminates in a triangular point. The proportions of the small chela are: Fingers 1; total length 2.36; height 0.84; proportion T. L.: H.=2.8:1. The carpus is not spinous on its superior margin; the meropodite terminates on this margin in a triangular point, its thickness being a little less than that of the palm, contained 1.9 times in its length. In the second pair the first segment of the carpus is slightly shorter than the sum of the four following ones; the meropodite is shorter than the carpus. The meropodite of the third pair is approximately equal to the carpus of the second pair, and 3.75 times longer than -wide. The proportions of the members are: Carpus 1; propodite 2; meropodite 2.15; the dactyl is very slender, the dorsal hook twice as long as the ventral. The posterior angles of the telson are right angles, not prolonged to a triangular prominence. Named for Mr. W. N. Lockington, the original describer of the species. The description by Lockington—very explicit as to the length of the antennal spines, the form of the chela of the first pair, the carpus of the second pair, the dactyl of the third pair, and the telson—appears to permit identification of the specimens of S. leviusculus with those which I have studied. The differences bear upon two points: Lockington says that the spine of the scaphocerite does not reach the extremity of the peduncle and that the movable finger of the large chela projects beyond the pollex. The first character hardly exists on the specimens that I have seen, the spine being approximately equal to the carpocerite and the fingers of the large chela equal. no. 1659. AMERICAN SPECIES OF SYNALPHEUS—COUTIPRE. 93 Lockington’s specimens came from Port Escondido, Port Mulege, and other points in the Gulf of California. Those which I have studied were collected by the steamer Albatross, of the U. S. Bureau of Fisheries, on the California coast at Station 4421, eastern point San Nicolas Island N. 26° W. 3.8 miles, 229-298 fathoms. This species is particularly im- portant because of its resemblance to S. paulsoni Nobili from the Red Sea and the Persian Gulf. There is but one difference between the two species: The spine of the scapho- cerite in S. paulsoni always exceeds the carpocerite. If, however, one were studying the two species with- out a knowledge of their source, one Besta ciidet ace 5 Fe lee would be led to make of the S. TRUAND ANTENA REGION ; a paulsoni Hie Gs Oxy CerOs ” form of NS. CARPOCERITE ; ™, PORTION OF THIRD lockingtoni. se S. paulsoni, in the group which bears its name, is a form with short carpocerite, that article arising a little below the extremity of the basal article of the an- 5 | tennule. It is a charac- ter which seems to me to be very essential for the diagnosis of the species of the group, the carpocerite being elongated and_ slender m cr elongated and thick, 7 and also important be- (4A / cause the forms with Ve short carpocerite may be considered as less de- veloped, as that article is always short in the primitive Eucyphota. Fic. 3.—SYNALPHEUS PAULSONI KURRACHEENSIS. 4, In the Indo-Pacific FRONTAL AND ANTENNAL REGION; C, CARPOCERITE ; K, i LARGE CHELA; 1, FOOT OF SECOND PAIR; m,MEROPODITE TeQ10N, S. paulsoni is ee ne surrounded by a certain number of derived forms: S. paulsoni liminaris Coutiére,in which the carpocerite is a little longer and the basicerite almost unarmed above; S. paulsoni rameswarensis Coutiére, in which the interior palmar border is spinous; and S. paulsoni kurracheensis of a more mas- sive general form, very interesting, in that it indicates the direction 24 PROCEEDINGS OF THE NATIONAL MUSEUM. VOL. XX:XVI- in which the species S. lataste7, of Australia and Chile, is differ- entiated. I have believed it possible to separate as a distinct species but S. tumidomanus is very distinct from it, as shown by the angles of the telson being clearly spinous (as Paulson has figured it), and also by the very large eggs which give rise to mysis. In these last two forms the carpocerite has be- come distinctly elongate and slender (at ee eee rk received from M. Gravier, in which the antennal scale is notably shorter than the antennule and still shorter than the carpocerite. This species is again extremely like S. lock- ingtoni from California and Lower California. Although the forms with a short carpocerite are at present less numerous on the Californian coast than those with a long one, the parallelism be- p S. hululensis Coutiere from the Maldives, which I have described in my work on the Alpheide of that archipelago under the name of S. tumidomanus Paulson; c ape least four times as long as wide). = I have also separated from S. paulsoni, Fic. 4.—SYNALPHEUS HULULEN- B pe ; p Y , _ 0) 29.9 ON ieee ee a oe under the name OnTgS: mushaensts TENNAL REGION; c, carpo- Coutiére, a specimen from the Red Sea, tween them and those of the Indo-Pacific is again accentuated by the following species which I believe should be separated from S. lockingtoni. SYNALPHEUS PAULSONOIDES, new species. The species differs from S. lochkingtoni by the following points: The antennal scale equals the antennule, and the lateral spine of the scaphocerite considerably exceeds the carpocerite, which is four times as long as wide. The carpus of the small cheliped has its anterior bor- der prolonged in a spinous prominence. The feet of the third pair are very slender. Fic. 5.—SyNnaupneus TuMt- Their proportions ‘are: Carpus 1; propo- |) yes naccon: 6m dite 2; meropodite 2.35; this last being POCERITE ; t, TELSON. 5.3 times as long as wide. The specimen, a male, is unique and its large cheliped is wanting; but it is very easily distinguished from S. lochingtoni, especially by the sleuderness of the meropodites of the third pair. It corresponds, as the Indo-Arabic S. paulsoni, to an “ ovyceros ” form of the species a ~ no.1659. AMERICAN SPECIES OF SYNALPHEUS—COUTIERE. 95 previously cited, but in a different direction. Up to the present time, in fact, I know of no form which is exactly the same in both regions. S. paulsonoides is from the island of San José, Lower California (M. Diguet, Paris Museum). On the South American coasts the species with short carpocerite of the Pavutsont group are similarly represented. The species S. latastei Coutiére, described below, is of great interest. It possesses in Brazil a “ tenuispina” form which is with difficulty separable from some specimens from Cape Lopez, in West Africa. On the other hand, it exists in Australia, for I can not differentiate from the typical Chilean specimens the unique Australian individual which I have examined. In Australia, moreover, a species with very large eggs, S. maccullochi, is shown to be closely allied, and S. paulsoni hur- racheensis, previously cited, clearly re- sembles it also, though its appendages are a little more massive. SYNALPHEUS LATASTEI, new species. The rostrum is longer than the fron- tal spines, and also wider; the antennu- lar articles are short and approximately equal; the proportion of the antennule is only 1: 3.85, the diminution in length affecting especially the basal article; ™ ee Hed den Ger gree eS te the stylocerite is shorter than in S. lock- NAL REGION ; C, CARPOCERITE; m, ingtoni; the lateral spine of the basi- — “™*0POPtTE OF TuInD moor" cerite is as long as the stylocerite; its superior spine is short and strong. The antennal scale, rather reduced, is 5.7 times as long as wide, and its long and strong lateral spine exceeds the antennule by the length of the distal article, and usually slightly exceeds the carpocerite, which arises from the same level as the median antennular article; the pro- portion of its dimensions is 1: 2.71, sometimes even 2.66; its form is more cylindrical than in S. lochingtoni. The large chela, which recalls the preceding species by the absence of any spinous prominence on the anterior margin of the palm, differs from it by its more stocky form: Fingers 1; total length 3.4; height 1.5. The small chela has the same proportions as in S. lochingtoni; the meropodites of the two chelipeds terminate on the superior mar- gin in a spinous prominence. In the second pair, the first article of the carpus, the four following and the chela are to one another as 1.2, 1.25, 1. The proportions of the third pair are: Meropodite 2; carpus 1; propodite 1.6; the me- ropodite is 3.12 times longer than wide, and consequently very stout. 772 26 The telson has the same form as in S. lockingtoni. This species seems to be frequent in Chile, from which locality I have been able to examine some ten specimens, thanks to M. Lataste, of the Paris Museum, after whom the species is named; the species is also met with in Australia (?) (one male of great length without indication of locality other than New Holland; Paris Museum). The size is greater than in S. lochkingtoni. It reaches 30.5 mm. in ‘length from the rostrum to the telson. I have been obliged to separate from the typical species, under the name S. latastet tenuispina, a large female from Desterro which dif- PROCEEDINGS OF THE NATIONAL MUSEUM. vox. xxxvt. a) aé Fic. 7.—SYNALPHEUS LATASTEI. @, FRONTAL AND ANTENNAL REGION, MALH, AUSTRALIA ; a’, FRONTAL AND ANTENNAL REGICN, FEMALE, CHILE; C, CARPOCERITE; K, LARGE CHELA} k’, SMALL CHELIPED OF FIRST PAIR; 1, FOOT OF SECOND PAIR; Mm, FOOT OF THIRD PAIR. fers from it in the following points: The antennule is less thick (four times as long as wide instead of 3.85); the scale of the scaphocerite equals the antennule, and its very sharp lateral spine exceeds the carpocerite very considerably, the latter being 3 times, or even 3.04 times, longer than wide. The meropodite of the third pair is 3.3 times as long as wide, instead of 3.1 times, as in S. Jatastez. This form approaches closely to S. lockingtoni from California and Lower California, differing from it, however, by the large and more massive chela (proportions T. L.: H. = 2.5:1 instead of 2.9:1), the no. 1659. AMERICAN SPECIES OF SYNALPHEUS—COUTIERE. "Qe fingers of which are shorter (fingers 1, height 1.4, instead of 1.1). The feet of the third pair are also more massive (meropodite 3.3 times as long as wide instead of 3.74), and the feet of the second pair, as in S. latastet, have the distal chela feeble and the first segment of the carpus equal in length to the four following ones. Re See ab ee m a i) Fig. 8.—SYNALPHEUS LATASTEI TENUISPINA. da, FRONTAL AND ANTENNAL REGION 3 C, CARPO- CERITE; A, LARGE CHELA; k’, SMALL CHELIPED OF FIRST PAIR; 1, FOOT OF SECOND PAIR; m, MEROPODITE OF THIRD PAIR. J Desterro, Brazil; Fritz Miller; one female 30 mm. long (Paris Museum). SYNALPHEUS APIOCEROS, new species. On the Atlantic coast of America the PauLsonr group is repre- sented by additional species; one of them, S. townsend, described farther on, is, up to the present time, the most aberrant form of the group known; it possesses no spine in the superior angle of the ba- sicerite, and thus closely resembles S. paraneomeris of the NromeEris group; this resemblance is further accentuated by the fact that the dactyls of the third, fourth, and fifth pairs in the preceding species possess only a trace of triunguiculation. Thus these two species mark the varietal limits of the two groups of forms. The other species, S. apioceros, also new, is, on the other hand, very typical. It is of special interest because of the great number of allied forms, American or Indo-Arabic, which may be approximated to it. The rostrum is equal to the lateral teeth, from which it is separated by intervals acute at base; the antennule is about 4.6 times as long as wide, but its basal article is 2.2 times as long as the median, and con- siderably exceeds the frontal teeth. This is a character which dis- tinguishes this species at once from S. lockingtoni. a8 - PROCEEDINGS OF THE NATIONAL MUSEUM. VOL. XXXVI. The stylocerite scarcely reaches the middle of the median article; the superior spine of the basicerite nearly equals the frontal teeth; its lateral spine does not reach the extremity of the basal antennular - article; the scale of the scaphocerite, five times as long as wide, is shorter than the antennule, its lateral spine slightly shorter than the ‘arpocerite, which last, three times as long as wide, is swollen at the base and pyriform; it is long in the sense that it takes its origin below the point where the stylocerite is detached from the basal article of the antennule and that it exceeds the antennule by about two-thirds of its distal article. The outer maxillipeds do not exceed the carpocerite. The large chela has the following proportions: Fingers 1; total length 3.15; Fic. 9.—SYNALPHEUS APIOCEROS. @, FRONTAL AND ANTENNAL REGION; C, CARPOCERITE; K, LARGE CHELA; K’’, CARPUS OF LARGE CHELIPED; k’, SMALL CHELIPED OF FIRST PAIR; l, , FOOT OF SECOND PAIR} ™, FOOT OF THIRD PAIR; m’, DACTYL OF THIRD PAIR; t, THLSON. height about 1.28; the proportion T. L.: H.=2.8:1; the anterior bor- der of the palm is swollen in a tubercle, which terminates in a spine directed slightly obliquely downward. The small chela has, as proportions, fingers 1; total length 2.8; height 0.95; T. L.: H.=2.95:1; it is thus relatively slender with short fingers; the wrist is spinous on its supero-external border; the meropodite, a little less thick than the palm, is 2.35 times longer than wide. The proportion of the two chele is about 1:3. In the second pair the first segment of the carpus equals apparently the sum of the others, and the meropodite is more slender than in 8. lockingtoni,; the chela is also notably shorter. The proportions of the third pair are: Carpus 1; propodite 2; mero- podite 2.28, this last being a little more than four times as long as ae no. 1659. AMERICAN SPECIES OF SYNALPHEUS—COUTIERE. 29 wide. The dactyl is slender, about 3.2 times as long as wide, with the ventral hook relatively shorter than in S. lochingtoni. The telson has its posterior angles accented, but not spinous. The eggs give rise to zoéa larve. : Localities : Jamaica, Albatross. Marco, Florida, 1 to 3 fathoms, H. Hemphill (Cat. No. 7000). Marco, Florida, in sponges, H. Hemphill (Cat. No. 6970), type. The species differs, then, from S. lochingtoni in many ways; the rostrum and stylocerite are shorter, the basal article of the antennule is longer, the carpocerite of different form and_pro- portions, the maxillipeds shorter, the large chela is spinous on the palmar border, the small cheliped is more slender, its carpus spinous and the second and third pairs are more slender. One is induced to separate from S. apioceros a whole series of forms of different geographic ori- gin, probably constituting Cie as many distinct species. They differ, nevertheless, very little from typical 7 specimens and correspond well to what are called “ petites espéces.” “/ 72 SYNALPHEUS APIOCEROS a SANJOSETI, new subspecies. = on This subspecies is repre- sented only in the ecollee- Fié. 10.—SYNALPHEUS APIOCEROS SANJOSEI. 4, . a = FRONTAL AND ANTENNAL REGION; IG, SPINE OF tion of the Museum at Paris LARGE CHELA; K’’, CARPUS OF LARGE CHBLIPED; by some specimens (male k’, SMALL CHULIPED OF FIRST PAIR; mm, MEROPO- DITE OF THIRD PAIR. and female) collected by M. Diguet. It is from San José Island, Lower California, and is as distinct from S. lockingtoni of California as is the preceding, from which it differs in the following details: The rostrum is usually a little longer than the lateral spines; the 5 : ) -stylocerite scarcely surpasses the basal article of the antennule; the pyre? lateral spine of the scaphocerite is notably shorter than the carpo- cerite; the carpus of the small cheliped is unarmed above. The small chela has these proportions: Fingers 1; total length 2.56; height 1.1; T. L.: H.=2.34:1; it is thus more massive than in S. apioceros, where the last proportion is 2.95 :1. The meropodite of the third pair is 3.56 times as long as wide. 80 PROCEEDINGS OF THE NATIONAL MUSEUM, Vow. xxxvi. SYNALPHEUS APIOCEROS MAYAGUENSIS, new subspecies. I have examined four specimens from Porto Rico belonging to this form. They are not, however, strictly alike. The most typical among them differ from S. apioceros in the following particulars: Rostrum a little longer than the lateral spines; sty- z locerite attaining at least the middle of the median arti- ie cle of the antennule; scale of the scaphocerite only four times as long as wide; car- Fic. 11.—SyNALPHEUS APIOCEROS MAYAGUENSIS. pocerite surpassing the an- a, FRONTAL AND ANTENNAL REGION; K, spine tennule by 14 times the See m’, DACTYL OF THIRD PAIR}; length of the distal article, longer than the spine of the scaphocerite, and very swollen at base (only 2.8 to 2.9 times as long as wide) ; the spine of the large chela continues in a straight line the anterior margin of the palm, which presents no swollen tubercle; the dactyl of the third pair is 3.8 times as long as wide and its ventral hook is more feeble than in S. apioceros; the posterior angles of the telson are right angles. The other specimens are distinguished from the preceding by slight differences in the width of the antennal scale, in the more slen- der feet of the second and third pairs, and by the more marked pos- terior angles of the telson. It is probable that more abundant mate- rial would permit of separating them also from S. apioceros. Type.—Cat. No. 24785, U.S.N.M. m Mayaguez, on coral reef. SYNALPHEUS APIOCEROS LEIOPES, new subspecies. Some females collected by M. Chaper (Paris Museum) differ FT BORE ee ee ae eee a from S. apioceros in the following GION; K, SPINE OF LARGE CHELA; m, points: PORTION OF THIRD FOOT. The lateral spine of the scaphocerite always slightly exceeds the carpocerite; the feet of the third pair are more slender, the meropo- dites being 4.4 times longer than wide; the telson has its posterior angles right angles. Venezuela, precise locality unknown. Type in Paris Museum. no. 1659. AMERICAN SPECIES OF SYNALPHEUS—COUTIERE. 81 SYNALPHEUS APIOCEROS DESTERROENSIS, new subspecies. This form is more distinctly separated from S. apioceros than the preceding; it is also more abundantly represented. The specimens (4 males, 3 females) come from Desterro (Fritz Miiller; Paris Museum). The rostrum is separated from the lateral spines by wide intervals with sinuous base; the basal article of the antennule is only twice as long as the median article, and the antennule scarcely four times as long as wide; the scale of the scaphocerite is from 4.1 to 4.3 times as long as wide, at least equal to the antennule, or longer; the lateral spine of the scaphocerite surpasses it very little, being shorter itself than the carpocerite; the maxilliped exceeds the antennule, and slightly the carpocerite; the spine of the large | chela continues the an- terior palmar border; the feet of the third oF pair are very slender, the meropodite being about 4.7 times as long as wide; the posterior angles of the telson are right angles. K All these forms, like S. apioceros, have zoéa larvee. None possess Fig. 13.—SYNALPHEUS APIOCEROS DESTERROENSIS. 4, FRONTAL AND ANTENNAL REGION; K, SPINE OF LARGE large eges. They them- CHELA; K’’, CARPUS OF LARGE CHELIPED; k’, SMALL selves present slight in- CHELIPED OF FIRST PAIR; m, MEROPODITE OF THIRD PAIR. dividual variations when the specimens representing them are numerous, and it is prob- able that they will be isolated hereafter as distinct species. I have noticed in one of the specimens of sanjoset the absence of the spine on the anterior border of the carpus. This is an example of a type of “ mutation ” with which one frequently meets in the Synalpheids, and which is of very slight importance. In the Indo-Pacific region the forms with long and slender carpo- cerite, analogous to the preceding, have as the type S. acanthitelsonis Coutiére, which differs from them almost solely by the very spinous angles of the telson, and S. hastilicrassus Coutiére, in which the su- perior angle of the basicerite is unarmed, and which consequently is closely allied to the species described hereafter. 32 PROCEEDINGS OF THE NATIONAL MUSEUM. VOL. XXXVI. SYNALPHEUS TOWNSENDI, new species. The rostrum is 1} times as long as the lateral teeth, reaching usually to the end of the proximal third of the median article of the mm \ Fic. 14.—SYNALPHEUS TOWNSENDI. dd, FRONTAL AND ANTENNAL REGION; C, CARPOCERITD ; K, LARGE CHELA; K’’, CARPUS AND MEROPODITE OF LARGE CHELIPED; k’’, SMALL CHELIPED OF FIRST PAIR; 1, FOOT OF SECOND PAIR; Mm, FOOT OF THIRD PAIR; mm’, DACTYL OF THIRD PAIR; t, TELSON. antennule. On some specimens the frontal teeth are longer and more slender. The relative proportions of the articles of the anten- nule are: 2, 1.38, 1. The stylocerite reaches almost the middle of the median article. The basicerite bears no spine above, where it terminates in an obtuse angle; its lateral spine reaches to the distal third, sometimes even to the extremity of the basal antennular article; the an- tennal scale is 5.6 times longer than wide; its lateral spine is long and reaches beyond the extremity of the Fic. 15.—SynaLpHevs Townsenn1 cCarpocerite, which is scarcely longer PRODUCTUS. da, FRONTAL AND than the antennule, and is 3.5 times as ANTENNAL REGION. long as wide. The large chela has the following relative dimensions: Fingers 1; total length 3.65 to 3.7; height 1.25; the anterior border of the palm bears a sharp spine; the carpus is very small, in the form of a coin;: the supero-external margin of the meropodite (which is twice as long as wide) is very convex, terminated by a hooked spine. no. 1659. AMERICAN SPECIES OF The small chela is one-third as long as the large one; dimensions are: Fingers 1; total length 2 The meropodite is 3.1 times longer than wide; its upper margin ending in a sharp angle. In the second pair, which is slender and elongate, the first article of the carpus measures 1, the sum of the four following is 0.83, and the chela 0.75. The relative dimensions of the third pair are: Meropodite 2.41 (five times longer than wide); carpus 1; pro- podite about 2.14, the foot as a whole being long and slender, especially the propodite; the dactyl is also elongate, its ventral hook less thick and espe- cially much shorter (about one-third) SYNALPHEUS—COUTIERE. 33 its relative 5; height 0.8 or a little less. e mr’ a tS Fic. 16.—SYNALPHEUS TOWNSENDI than the dorsal, parallel in direction. ‘The telson has sharp posterior an- with which it is BREVISPINIS. @, FRONTAL AND AN- TENNAL REGION; C, CARPOCERITE; K, LARGE CHELA; m’, DACTYL OF THIRD PAIR; t, TELSON. gles, with the inner pair of spines very slender, three times longer than the outer pair; the convex posterior border has twelve plumose hairs. Ni Fic. 17.—SYNALPHEUS TOWNSENDI MPXICANUS. @d, FRONTAL AND AN- TENNAL REGION; m’, DACTYL OF THIRD PAIR; t, TELSON. The eggs should give, in acordance with their size, zoéa larve. This species represents on the Amer- ican coasts S. hastilicrassus of the Lac- cadives and Maldives. The latter is distinguished by the large cheliped, in which neither the palm nor the mero- podite is spinous, by the shorter dactyls of the thoracic feet, in which the ven- tral hook is the more important, by the telson having more pronounced angles, and the inner spines shorter than in S. townsendi. The carpocerite is also more swollen than in the last-named species. S. townsendi shows some interesting variations. In a male specimen from Albatross Station No. 2406 (form pro- ductus) the rostrum and the frontal spines are very elongate, as is also the lateral spine of the scaphocerite. This “ oxyceros” form (very frequent in Synalpheus), is seen likewise in the Maldivian species, S. hastilicras- sus, at least as to the elongation of the rostrum. Proc. N. M. vol. xxxvi—09——3 34 PROCEEDINGS OF THE NATIONAL MUSEUM. VOL. XXXVI. This species is chiefly Atlantic, being known from the Bermudas, from Florida, the Bahamas, Cuba, and Porto Rico. It extends as far as Brazil, and, strange as it may seem, the Albatross col- lected it at the Hawaiian Islands at Station No. 3969. The species is met with, however, on the Pacific coast of America, represented by some specimens collected by M. Diguet (Paris Museum), which differ from those just described by the large chela, which has the palm obtuse, and the telson, in which the posterior angles are alto- gether right angles and the inner spines are shorter (form brevispinis). Other specimens from the same region, collected by the Albatross (form mewicanus) have the basicerite shghtly acute on the upper border; the frontal projections are short, the rostrum shorter than the basal article of the antennule; the ventral hook of the dactyls is a little larger, and the posterior angles of the telson are sharper than in typical specimens. These are very interesting differences, because they are the same which serve to separate the two species of the Maldives, S. hastilicrassus and S. acanthitelsonis, but here the characters of the basicerite and of the telson are much more marked. Named for Mr. Charles Haskins Townsend, formerly naturalist of the Albatross. Localities: North Carolina, 15 to 16 fathoms, Albatross Station Nos. 2280 and 2619. Florida: Key West (Union University collection) ; Anclote; Straits of Florida, 56 fathoms, Albatross Station No. 2640; west coast, 12.5 to 28 fathoms, Grampus Stations Nos. 5094 and 5100 and Fish Hawk Stations Nos. 7106, 7123 and 7124. Gulf of Mexico, 24 to 32 fathoms, Albatross Stations Nos. 2369, 2372, 2373 (type), 2387, 2389, 2390, 2405, 2406 (type of townsendi productus), 2407, 2409, 2410, 2411, 2412, 2414. Yucatan, off Cape Catoche, 24 to 27 fathoms, Albatross Stations Nos. 2862, 2365, 2366. St. Thomas, Albatross, and Fish Hawk Stations Nos. 6079, 6080, in 20 to 23 fathoms. Porto Rico, Mayaguez Harbor, 4 to 6 fathoms, Fish Hawk Sta- tion No. 6065. Culebra, 15 to 15.25 fathoms, Fish Hawk Stations Nos. 6087 and 6093. Vieques, 15 to 16 fathoms, Fish Hawk Stations Nos. 6091 and 6092. Bermuda, G. Brown Goode. no.1659. AMERICAN SPECIES OF SYNALPHEUS—COUTIERE. 35 Localities—Continued : Brazil, Bahia, Hartt Expedition, Station No. 173. Hawaian Islands, French Frigate Shoal, 15 to 16 fathoms, Albatross Station No. 3969. Southern part of Gulf of California, 9} fathoms (type of form mexicanus), Albatross Station No. 2826. Lower California (form brevispinis), M. Diguet (Paris Mu- seum ). Type of S. townsendi.—Cat. No. 38392, U.S.N.M. Type of S. townsendi productus.—Cat. No. 9798, U.S.N.M. Type of S. townsendi mexicanus.—Cat. No. 38393, U.S.N.M. S. townsendi is particularly close to S. paraneomeris Coutiére, a form with basicerite unarmed above, which also presents variations in the armature of that article, as does an “ owyceros” form. The difference consists principally in the supernumerary ventral promi- nence of the dactyls, absent in S. townsend?, but very characteristic of the Nromeris group, where a great number of species possess it. S. paraneomeris is one of the most widely distributed species of the Indo-Pacific region. NEOMERIS Group. SYNALPHEUS FRITZMULLERI, new species. Rostrum slender, quite distinct from the lateral spines, the margins nearly parallel for half their length; lateral spines with sharp points, generally a little shorter than the rostrum, reaching to the middle of the basal antennular article. The articles of the antennule are in the proportion: 1.5, 1.1, 1, beginning at the base; the external flagellum is bifurcate beginning at the eighth article; the stylocerite equals one-half of the median antennular article. The basicerite of the antennz bears on the upper side a strong spine, laterally a longer spine, a little shorter than the stylocerite; the an- tennular scale is narrow (6.6 times longer than wide), its sharp lateral spine reaching the extremity of the carpocerite, which surpasses the antennule by about half the distal article and is a little swollen and only three times longer than wide. The external maxillipeds reach forward to the bifurcation of the external antennular flagellum. The relative proportions of the large chela are: Fingers 1; total length 3.15 to 3.3; height 1.25; the anterior margin of the palm bears an obtuse prominence; the meropodite is 2.3 times longer than wide, its inferior margin terminating in a strong triangular point. 36 PROCEEDINGS OF THE NATIONAL MUSEUM. VOL. XXXVI. The proportion of the small chela to the large one is as 1 to 2.5. Its relative dimensions are: Fingers 1; total length 2.2; height 0.72; its meropodite also ends in a strong triangular lobe. In the second pair the first article of the carpus is approximately equal to the four others taken together; the chela is a little shorter; the meropodite measures about 0.8 of the length of the carpus. The relative proportions of the third pair are: Meropodite 2.33; carpus 1; propodite 2, or a little less; the meropodite is 3.5 times longer than wide, this proportion diminishing in adult females or in males of small size to 3.7; it reaches even 4 in a perfectly typical (/ Fic. 18.—SYNALPHEUS FRITZMULLERI. @, FRONTAL AND ANTENNAL REGION $ @, CARPOCER- ITE; K, LARGE CHELA; K’’, CARPUS AND MEROPODITE OF LARGE CHELIPED; k’’, SMALL CHELIPED OF FIRST PAIR; 1, FOOT OF SECOND PAIR}; mm, FOOT OF THIRD PAIR; m’, DACTYL OF THIRD PAIR; ™’’, REVERSE OF SAME; t, TELSON. female from Cape Florida; the two hooks of the dactyl are divergent, the ventral nearly twice as thick as the dorsal, with the anterior margin more convex; it does not directly continue the inferior margin of the dactyl, rejoining it by a concave curve of short radius in such a manner as to form a third obtuse prominence. The telson has its posterior angles obtuse, its posterior margin bears twenty plumose hairs between the two pairs of habitual spines. The eggs are of small size, and the larve are zoée. The typical specimens are from Florida, some of them living in sponges; the species is also met with in Porto Rico and in Jamaica. In these three regions the typical examples predominate, but the no. 1659. AMERICAN SPECIES OF SYNALPHEUS—COUTIERE. 37 species also occurs in the “ oxyceros ” form (subspecies elongatus), the lateral spine of the scaphocerite greatly exceeding the carpocerite of the antenna, and the rostrum being frequently longer than the lateral spines. The subspecies elongatus seems to be almost the only one in Venezuela and Brazil. Some specimens received from M. Chaper (Venezuela), others received from Fritz Miiller from Desterro, and a small specimen from Bahia (R. Rathbun, Hartt Explorations) are without exception elongatus. Aside from their longer antennal spine, these specimens have the carpocerite less thick (3.3 to 3.4 times longer than wide), and the feet of the third pair more slender, the meropodite being 3.8 to 4 times longer than wide; but this last character varies with the size, and also with the sex, in the typical specimens within a rather wide range. With the material at my disposal I can not form a conclusion as to the advisability of the specific separation of this elongatus form. The species is also found in Lower California, from which locality I have been able to study py. 19-synarpuevs a single male specimen collected by M. Diguet, = Frrzmirierr — nLon- which does not differ from the Florida speci- Np rsrmws'ut anaron, mens in regard to the antenne; the feet of the . third pair are slender, the meropodite being four times as long as wide; but I find the same figures among the small typical males of Florida, of corresponding size, so that I can not separate this specimen from Lower California, even as a distinct “ race.” Named for the naturalist, Dr. Fritz Miiller. Localities: Typical specimens— Cape Florida, Edward Palmer, 1 specimen. Key West, Union University collection, 2 specimens. Key West, H. Hemphill, 2 specimens. Marco, Florida, H. Hemphill, 10 specimens, male and fe- male, types. Florida, west coast, 28 fathoms, ish Hawk Station No. 7123, 1 specimen. St. ‘Thomas, Albatross, 1 specimen. Mayaguez, Porto Rico, Fish Hawk, 4 specimens. Lower California, M. Diguet (Paris Museum), 1 specimen. Subspecies elongatus— South Carolina, Mouth of Bull Creek, Fish Hawk, 1 speci- men, type. Florida, Eastern Dry Rock, Edward Palmer, 1 specimen. 88 PROCEEDINGS OF THE NATIONAL MUSEUM. VOL. XXXVI. Localities—Continued : Subspecies elongatus—Continued— Florida, St. Martins Reef, Lieut. J. F. Moser, U. S. N., 1 specimen. Florida, H. Hemphill, 1 specimen. Florida, Key West, Union University collection, 1 specimen. Jamaica, Albatross, 3 specimens. Venezuela, M. Chaper, Paris Museum. Bahia, Hartt Explorations, R. Rathbun, 1 specimen. Desterro, Fritz Miiller, Paris Museum. Type of S. fritzmiilleriCat. No. 6970, U.S.N.M. Type of S. fritemiilleri elongatus.—Cat. No. 38394, U.S.N.M. SYNALPHEUS HEMPHILLI, new species. The species is very like the preceding, the differences being as follows: The rostrum is always much longer (about twice) than the lateral spines; the feet of the third pair are a little shorter and thicker, their relative proportions being, carpus 1; meropo- dite 2.5; propodite about 2; the meropodite is 3.5 times longer than wide; on the dactyl the ventral hook is perpendicular to the lower border and its margins form a double curve, convex, then a ht- Fic. 20.—SYNALPHEUS HEMPHILLI. m, FOOT OF THIRD tle concave to the point ; PAIR, BERMUDAS; ™”, EXTREMITY OF FOOT OF THIRD behind, the very marked PAIR, ALBATROSS STATION No. 2409. : 3 S third prominence forms a right angle at the summit, projecting a little in a spine. It is therefore almost solely the form of the hook which dis- tinguishes the two species, for the chela of the first pair, the telson, and the carpocerite are quite alike; this character of the dactyls is not only very marked, but perfectly constant in presence and in degree. There is in this species, as in the preceding, an “owyceros” form, in which the antennal scale equals the antennule, its lateral spine much exceeding the carpocerite. The resemblance of these two forms to S. nilandensis and S. nilandensis owyceros Coutiere, of the Mal- dives, is extremely close. The differentiation from S. nélandensis, in which the supraorbital spines are equal to the rostrum, the ventral hook of the dactyl is very strong, the supernumerary hook very dis- mn! no. 1659. AMERICAN SPECIES OF SYNALPHEUS—COUTIERE. 39 tinct and spinous, is relatively easy; but in S. nilandensis oxyceros these last differences have entirely disappeared, the only ones per- sisting being the greater length of the supraorbital spines, that of the lateral spine of the basicerite, and lastly the presence of a row of five spines on the meropodite. A large female specimen from the Bermudas of 8. hemphilli longi- cornis is particularly interesting in this regard; the meropodite of the left foot of the third pair bears a well developed spine; the rest of the series is wanting; the opposite member is quite unarmed, but the suggestion arising from this circumstance is none the less in- structive, as it shows to what degree the parallelism between the Indo-Pacific and the American forms may be carried. The “oay- ceros”” forms of the species S. nilandensis and S. hemphilli appar- ently represent the first evidences of divergence of certain species originally common to two regions; judging by the much greater num- ber of species in the Indo- Pacific region, and also from the fact that there only are found the forms with spinulous meropo- dites, one might infer that that region was the center of dispersion of the group. S. paraneomeris Cou- tiére possesses the same form of hook as S. fritz- Fic. 21.—SYNALPHEUS HEMPHILLI LONGICORNIS. 4, milleri, from which it is FRONTAL AND ANTENNAL REGION; ™’, EXTREMITY OF most readily separated "~ by having the basicerite not spinous above; on the other hand, S. paraneomeris is no less closely related to S. townsendi Coutiére, the basicerite of which is unarmed above, but the dactyls of which have no ventral supernumerary prominence; so that the two groups, the Neomeris group and the Pautson1 group, have in these three species a very evident point of approximation. Named for Mr. Henry Hemphill, who has added largely to the Alpheide in the U. S. National Museum. Localities: West coast of Florida, 21 to 28 fathoms: Albatross Station No. 2409, 2 specimens, type; Fish Hawk Station No. 7123, 1 speci- men; Fish Hawk Station No. 7124, 1 specimen (type of longi- cornis. ) Bermudas, G. B. Goode, 2 specimens (longicornis.) Type of S. hemphilli—Cat. No. 9817, U.S.N.M. Type of S. hemphilli longicornis.—Cat. No. 38395, U.S.N.M. 40 PROCEEDINGS OF THE NATIONAL MUSEUM. vou. xxxvt. SYNALPHEUS NOBILII, new species. This species is represented only by a single male specimen; I do not hesitate, however, to consider it distinct from S. fritzmilleri. The differences which separate the two species (comparing two speci- mens of the same sex and of the same size) are the following: The carpus of the small chela is more massive, its width exceeding that of the palm (proportion 1.12 instead of 0.9, as in S. fritz- miillert) . The second pair is shorter and thicker, the carpus being six, instead of eight, times as long as wide. Furthermore, the first segment of the carpus is shorter than the sum of the four following. M wm ei) i Z Fic. 22.._SyNALPHEUS NOBILII. @, FRONTAL AND ANTENNAL REGION; C, CARPOCERITE; K, LARGD CHELA; k’, SMALL CHELIPED OF FIRST PAIR; 1, FOOT OF SECOND PAIR; m, FOOT OF THIRD PAIR; m’, DACTYL OF THIRD PAIR. The third pair is much more massive; the relative proportions being; Carpus 1; propodite 1.8; meropodite 2 (instead of 1, 2, 2.33) ; the meropodite is only 2.8 times as long as wide, instead of 4 times, as in S. fritemiilleri. I find no other difference, either in the carpocerite, in the dactyl of the third pair, or in the telson. Named for Dr. Joseph Nobili, the carcinologist. Locality.—St. Helena, Ecuador, one male specimen, 25 mm. long (M. Festa; Paris Museum). I would remind the reader that S. fritzmiilleri is represented in the Paris Museum by a specimen from Lower California collected by ~ no. 1659. AMERICAN SPECIES OF SYNALPHEUS—COUTIPRE. 41 M. Diguet, which is absolutely typical and which it is impossible to separate as a distinct “race,” a rare circumstance among the Synalpheids. On the other hand, S. nodilii is very easily distinguished from S. sanlucasi, in which all the appendages, namely, the antennules, the spines of the basicerite and of the anterior margin, the large chela, the feet of the second pair, and the feet and even the dactyls of the third pair are much shorter and more massive. SYNALPHEUS SANLUCASI, new species. Species of each of the specific groups composing the genera Alpheus and Synalpheus are often parallel to species of closely allied K Fic. 23.—SYNALPHEUS SANLUCASI. 4d, FRONTAL AND ANTENNAL REGION ; K, LARGE CHELA; l, FOOT OF SECOND PAIR; m, FOOT OF THIRD PAIR: m’, DACTYL OF THIRD PAIR. groups in the characters upon which the separation of specific forms is based; as, for example, in the presence or absence of the superior spine of the basicerite or of the antennal scale, the slender or swollen form of the carpocerite, the spinose or unarmed palm of the large chela, and slenderness or stoutness of the thoracic feet. When one of these characters has been recognized in the species of a given group one can almost prophesy the existence of another species pro- vided with the opposite character. This is, moreover, a well-known fact in all genera which are rather numerous in species. 49 PROCEEDINGS OF THE NATIONAL MUSEUM. VOL. XXXVI. Thus, S. sanlucasi, a form very close to S. fritemiilleri, differs from it in the shortness and more massive form of all its appendages. The frontal teeth are stronger, the rostrum, especially, being wider at the base; the articles of the antennule are approximately equal, and the proportion of total length to the width is only 4.25 instead of 5, as in S. fritezmiilleri. The basicerite has its superior spine placed higher than in the pre- ceding species, so that it reaches the extremity of the basal antennular article, and makes the lateral spine short and stout. Although the scale of the scaphocerite is as long as in S. fritemilleri, the propor- tion of its length to its width is only 6 instead of 6.6, on account of its stoutness. The carpocerite is of the same form as:in the preced- ing species, and also exceeds the antennule. The large chela has, for its relative dimensions, fingers 1, total length 2.88, height 1.33; the small chela is lacking in the type. In the second pair the carpus is only 5.6 times longer than wide, instead of 8, as in S. frétzmiilleri; the relative proportions are: First article of the carpus 1; sum of the four following ones 1.6; terminal chela 1.72, very different, therefore, from the proportions found in S. fritzmiilleri; the meropodite measures 0.7 of the length of the carpus. The proportions of the third pair are: Meropodite 1.65; carpus 1; propodite 1.56; the meropodite is only 2.53 times longer than wide. The dactyl has practically the same form as in S. fritzmillera, the differences being that the ventral hook is wider at the base, and the entire appendage is shorter than in the preceding species. The telson has not suffered the same diminution as the appendages, the proportions of its length to its proximal and distal ends being respectively 1.4 and 2.33, instead of 1.15 and 2, as in S. fritzmiillerd, the telson of which species is, therefore, wider and shorter. The eggs are of the same size as are those of the preceding species. S. sanlucasi, readily distinguishable from the two American forms S. fritemiilleri and S. hemphilli, is much more closely allied to a species which I collected at Djibouti, and to which I give the name S. heroni, the species occurring on the reefs of Héron. 8S. heroné is distinguished by the following points: The lateral spine of the basicerite slightly exceeds the extremity of the median antennular article, and the antennal scale is more reduced, the proportion of its dimensions being about 7.3; on the other hand, its lateral spine, as in the “ owyceros” forms of many species, considerably exceeds the carpocerite. The proportions of the large chela are: Fingers 1; length 3.2; height 1.32, it being, therefore, less massive than that of S. san- lucasi. Ee no.1659. AMERICAN SPECIES OF SYNALPHEUS—COUTIERE. 43 The small chela has these proportions: Fingers 1; total length 2.6; height 1.08; it is 2.45 times smaller than the large chela; the meropodite is a little more than twice as long as wide. Compared to that of S. fritzmiilleri, the small chela appears much stouter, and it would probably be the same with S. sanlucasi. In the second pair, the first article of the carpus, the sum of the four following ones, and the terminal chela are practically of the same length; the carpus is 6.5 times longer than wide. mM x Fic. 24.—SYNALPHEUS HERONI. @, FRONTAL AND ANTENNAL REGION; K, LARGE CHELA}3 k’, SMALL CHELIPED OF FIRST PAIR; 1, FOOT OF SECOND PAIR; ™, FOOT OF THIRD PAIR. The proportions of the third pair are: Meropodite 2; carpus 1; propodite 1.6; the meropodite is 2.5 to 2.6 times longer than wide, and is therefore longer and thicker than in the species from Lower California. Cape St. Lucas, Lower California; John Xantus; type, Cat. No. 6355, U.S.N.M. BREVICARPUS Group. SYNALPHEUS MINUS (Say). Alpheus minus Say, Journ. Acad. Nat. Sci. Phila., I, 1818, p. 245. Teeth of the frontal border in the form of an equilateral triangle, the rostral tooth usually a little wider at the base, and sometimes very slightly longer, than the lateral teeth. The proportions of the antennular articles are: 2, 1.5, 1; the rela- tion of the total length of the stalk of the antennule to its width is 4.8 to 5; the stylocerite reaches the distal third of the median article; the external flagellum bifurcates only at the tenth article. 44 PROCEEDINGS OF THE NATIONAL MUSEUM. you. xxxvt. Basicerite of the antenne distinctly spinous above, the spine always longer than wide at the base; the lateral spine reaches to the ex- tremity of the basal article of the antennule. te mm : a é mimi | mm FIG. 25.—SYNALPHEUS MINUS. @, FRONTAL AND ANTENNAL REGION, TYPICAL; @’, FRONTAL AND ANTENNAL REGION, SPECIMEN FROM BERMUDAS WITH BASICERITE SPINOUS ABOVE; aa, FRONTAL AND ANTENNAL REGION, SPECIMEN FROM STATION NO. 7123 WITH CARPOCERITE MORE SLENDER; C, CARPOCERITE, TYPICAL; C’, CARPOCERITE, MALE, STATION NO. 7123; ce, CARPOCERITE, FEMALE, STATION NO. 7123; €, EGG; i, OUTER MAXILLIPED; K, LARGE CHELA, TYPICAL; Kk’, SMALL CHELIPED OF FIRST PAIR, S. BREVICARPUS; kk’, SMALL CHELIPED OF FIRST PAIR, TYPICAL; kkk’, SMALL CHELIPED OF FIRST PAIR, TYPICAL (ANOTHER SPECI- MEN) ; l, FOOT OF SECOND PAIR; m, FOOT OF THIRD PAIR, TYPICAL; mm, FOOT OF THIRD PAIR, S. BREVICARPUS; mmm, FOOT OF THIRD PAIR, STATION NO. 7123; m’, DACTYL OF THIRD PAIR, NOT TYPICAL; mm’, DACTYL OF THIRD PAIR, TYPICAL; t, TELSON, S. BREVI- CARPUS; t’, TELSON, TYPICAL. The antennal scale is narrow (proportion of length to width 7, and up to 8.5), its inner border making a very obtuse angle and not a regular curve; the lateral spine is a little longer than the peduncle of no.1659, AMERICAN SPECIES OF SYNALPHEUS—COUTIBPRE. 45 the antennule, shorter than the carpocerite, which last is a little de- pressed, the proportion of its length to its width being about 3.7; it exceeds the antennule by half or even two-thirds of the distal article. The large chela is regularly ovoid; its measurements, taken along the infero-external side, are: Fingers 1, total length 3.5; height 1.35; there is on the supero-internal side, on the anterior margin of the palm, near the articulation of the finger, a strong, sharp, and rather slender spine. The small chela is in the proportion of about 2.7 to the preceding; its relative dimensions are: Fingers 1; total length 2.25; height 0.8; the fingers terminate in a simple point; the carpus is Mer: scarcely a fourth tot the entire cheliped ; the meropodite is 2.5 times longer than wide, its superior margin ter- minated by a trihedral promi- nence, not spinous. In the second pair the pro- portion between the length and width of the carpus is about 9.5 the meropodite is only 0.75 of the length of the carpus. The proportions of the third pair of feet are: Meropodite a carpus 1; propodite 1.6 to ; the proportion between the ee and the width of the meropodite is approximately 4 often a little less; the dactyl is a little curved, long, its hooks Prsote Fic. 26.—SYNALPHEUS MINUS BAHIENSIS. 4, are almost parallel, the dorsal FRONTAL AND ANTENNAL REGION; k’, SMALL nearly twice as long as the CHELIPED OF FIRST PAIR; m’, DACTYL OF ventral. THIRD PAIR. The length of the telson equals 1.06 times the width at the base, and 1.84 times its distal margin, which is regularly convex and bears about twenty plumose hairs and two pairs of feeble spines. The eggs are of small size (0.6 mm. in the nauplius stage, subse-- quently up to 1 mm.), and give rise to zoée. The length of the species does not exceed 25 mm. The typical specimens come from the region of the Bahamas and Florida, but the species extends to the Bermudas and southward to Brazil; some specimens from this last locality differ from the types and may be separated as form bahiensis; the basicerite of the antennze has its lateral spine very slender, its superior spine long and strong, the lateral spine of the scaphocerite also slender, being as long as the carpocerite; the small chela is more swollen than in the typical specimens, the proportions being, fingers 1; total length 2.8; height 46° PROCEEDINGS OF THE NATIONAL MUSEUM. vou. xxxvi. 1 to 1.09; the proportion between the length and the width of the meropodite is 2.1 instead of 2.5; the dactyls, in the third pair espe- cially, have their two hooks almost equally strong and long. The strong superior spine of the basicerite, and also the greater* thickness of the small chela, are found again, less marked, in some specimens from the Bermudas and from Florida, which it would be hardly advisable to separate as a distinct form. A specimen from Sarasota Bay, Florida, has an abnormal small claw, approximating the large one in its proportions; the fingers measure only one-third of the total length, and the anterior border is spinous. This tendency to the reestablishment of the symmetry of the two claws is not very rare in the Alpheidz; and, although leading to the same results, it is diametrically opposed to the cases of hypotypic regeneration, of which also examples are known. Other very interesting specimens differ mark- edly from the types by the width of the anten- nal seale, only 7.3 times longer than wide, recall- a ing by its form that of S. brevicarpus. Other characters of the species are in these specimens weakened in the same way; for example, the carpocerite is a little less Fic. 27.—SYNALPHEUS MINUS ANTILLENSIS. d, FRONTAL swollen, the proportion K CHELIPED OF FIRST PAIR; m’, DACTYL OF THIRD PAIR. of its length to its width (3.7) declining to 4 in the male; of the members of the second pair the carpus is 10 times longer than wide. On the other hand, the dorsally strongly spinous basicerite, the thick meropodite of the third pair, the telson widened at its distal end, and the form of the chele of the first pair, permit the determination of these examples as S. minus. They indicate in what way the variation giving rise to the species brevicarpus is accomplished. Among the other varieties of S. minus, it seems to me possible to separate a form antillensis. The specimens which are referred to this form come mainly from Porto Rico and St. Thomas. They differ from the types in the frontal teeth, which are long and narrow, es- pecially the rostrum, and in the antennule, which is only 4.2 to 4.3 no. 1659. AMERICAN SPECIES OF SYNALPHEUS—COUTIERE. 47 times as long as wide; the basicerite is not more spinous above than in the typical specimens; the carpocerite is longer, surpassing the antennules by the length of the distal article, and, especially, more ovoid (proportion 3.2 or even 3.15) ; the antennal scale is also a little wider than in the types; the small chela, as in the form bahéensis, is more swollen than in the types; it has, as its proportions, fingers 1; total length 2.38; height 0.9; there are no differences either in the form of the large chela, of the members of the second and third pairs, or in the telson. In some specimens, especially among those from St. Thomas, the dactyls of the third pair are very slender, with the superior margin only slightly convex, and the superior hook strong. I have not a sufficient series to enable me to judge of the importance of this char- acter. The specimens of the form antillensis are all of small size, 15 mm. in length at the most. The eggs are as in the typical specimens. Localities : South Carolina, 15 miles southeast of Charleston, in fragment of madrepore, R. E. Earll. Florida: Cape Florida, Edward Palmer. Elliotts Key, lat. 26° 33’ N., long. 83° 10’ W., 28 fathoms, Fish Hawk Station No. 7123 (specimen approaching brevicarpus). Harbor Key. Salt Pond Key, Stock Island. Eastern Dry Rock. Key West, Union University collection. Dry Tortugas. Florida Bay, Edward Palmer. Two miles west of Cape Romano, 15 to 18 feet, Lieut. J. F. Moser, U. 8. N. . Marco, H. Hemphill. Sarasota Bay (specimen with small chela anomalous), Union University collection. Anclote (specimen approaching brevicarpus), Thomas Low. Florida Banks, lat. 28° 56’ N., long. 28° 15’ W., 12 feet, Lieut. J. F. Moser, U.S. N. . St. Martins Reef, Lieut. J. F. Moser, U.S. N. Bahamas: Andros Island, in sponges, F. Stearns collection. Green Cay, Geographic Society of Baltimore. St. Thomas, 20 to 23 fathoms, Fish Hawk Station No. 6079 (type of form antillensis), 48 PROCEEDINGS OF THE NATIONAL MUSEUM. vow. xxxvi. Localities—Continued : Porto Rico (form antillensis) : Playa de Ponce, Fish Hawk. Humacao, 94 fathoms, Fish Hawk Station No. 6099. Bermudas (specimen with basicerite very spinous), G. B. Goode. Brazil, Plataforma, Bahia (type of form bahiensis), R. Rathbun, Hartt Explorations. Type of S. minus bahiensis——Cat. No. 38396, U.S.N.M. Type of S. minus antillensis—Cat. No. 38397, U.S.N.M. SYNALPHEUS DIGUETI, new species. This species represents the Brevicarrus group on the coast of Lower California, where it has not previously been found. It is, consequently, a very important extension of the geographic distribu- tion of this group, which thereby ceases to be an exception from the general rule. Just as the Lavrimanus group possesses at least one Indo-Pacific species, it will also be found that the Brevicarrus group has met in that region of the globe conditions inducing specific dif- ferentiation. S. digueti is very near S. minus (Say); the differentiation is diffi- cult except between adult specimens, and the more mature, the easier is the determination. The characters of the males are more decided than are those of the females. In the males the differences between S. digueti and S. minus are the following: (1) The antennule is 6 times longer than wide instead of 5 times, as in the males of corre- sponding size of S. minus; (2) the carpocerite is 3.5 times longer than wide instead of 3.7 times, and the lateral spine of the scapho- cerite is a little shorter than the antennule; (3) the meropodite of the third pair of feet is 3.5 times longer than wide instead of 3.75 times. In the females the antennule is not more than 5.8 times as long as wide, and the spine of the scaphocerite slightly exceeds the antennule, so that the tangible differences from the females of S. minus become almost none. However, the carpocerite is somewhat thicker, 3.54 to 3.58 times longer than wide, while this proportion reaches 3.7 to 3.75 in the females of S. minus; there is also a very slight difference in the thickness of the meropodite of the third pair, where the proportions are nearly 3.5 in S. minus and 3.3 in S. digueti. There is also in both sexes a slight difference in the meropodite of the small cheliped; this is at the most as wide as the palm, and gen- erally a little narrower in S. minus (proportion 0.92 to 0.96); in S. digueti, on the other hand, it is wider (proportion 1.1 to 1.18). The specimens which have just been considered do not exceed 25 mm. in length; in one large female measuring 30 mm. (also col- lected by M. Diguet) the characters are much more clearly indicated, Sno.1659. AMERICAN SPECIES OF SYNALPHEUS—COUTIBPRE. 49 the antennules especially, markedly slender, being 6 times longer than wide, as in the male; in the largest females of S. minus, which are of equal size, this proportion never exceeds 5; the carpocerite is only 3.35 times longer than wide instead of 3.7 times, as in the largest female of S. minus. This specimen presents, moreover, a very peculiar form of rostrum, the point being shorter, and, particularly, much narrower than the lateral teeth. I have noticed in a male a tendency toward this shape. The lateral spines are very slightly longer than the rostrum, forming Fic. 28.—SYNALPHEUS DIGUETI AND S. DIGUETI ECUADORENSIS. @, FRONTAL AND ANTENNAL REGION OF S. DIGUETI, MALE OF MEDIUM SIZE; @’, FRONTAL AND ANTENNAL REGION OF S. DIGUETI, FEMALE OF MEDIUM SIZE; @’’, FRONTAL AND ANTENNAL REGION OF S. DIGUETI, FE- MALE OF LARGE SIZE; dd, FRONTAL AND ANTENNAL REGION OF S. DIGUETI ECUADORENSIS, MALE; C, CARPOCDRITE OF S. DIGUETI, MALE; ¢C’, CARPOCERITE OF S. DIGUETI, FEMALE; Cc, CARPOCERITE OF S. DIGUBTI HCUADORENSIS; k’, SMALL CHELIPED OF FIRST PAIR OF S. DIGUETI ; ™, MEROPODITE OF THIRD PAIR OF S. DIGUETI. a prominence exceeding it in height, so that the rostrum seems to be situated on a lower plane. The eggs are of the same size as those of S. minus and also give rise to zoce. Twelve specimens, male and female, from Lower California (M. Diguet, Paris Museum). One very interesting form of this species is represented by two speci- mens, male and female, from Ecuador. The differences between the Eroc, N. M. vol. sxxvi—09——4 50 PROCEEDINGS OF THE NATIONAL MUSEUM. VOL. XXXVI. male and female are very slight; the frontal teeth are very short, and the rostrum a little shorter and weaker than the lateral teeth, espe- cially in the male. In both cases the antennule is 5.5 times as long as wide; the lateral spine of the scaphocerite is as long as the antennule; the carpocerite exceeds it very little in the female, a little more in the male, and is very much swollen, only 3.5 times as long as wide in the male, or 3.3 times in the female; the meropodite of the third pair is 3.25 times as long as wide in both specimens. This form of S. digueti (which might be distinguished under the name of ecuadorensis) makes the distribution of the Brevicarrus group much like that of the Pautsonr group, which has representa-— tives in Lower California and Chile, in Brazil and Florida, as well as in the West Indies. I have shown in the introduction to this paper what interest attaches to the presence or absence of the species of the Brevicarrus group elsewhere than on the American coasts, because of their close relations. of kinship with the species of the PauLson1 group. St. Helena, Ecuador; M. Festa; 2 specimens, male and female (Paris Museum). Named for M. Diguet. SYNALPHEUS BREVICARPUS (Herrick). Alpheus saulcyi var. brevicarpus Herrick, Mem. Nat. Acad. Sci., V, 1891, p. 383. : The species is also very like S. minus, from which it is distinguished by the following characters: The proportions of the antennular articles are 1.8, 1.7, 1; the pro- portion of the length to the width of the antennule is at least 5.5. The basicerite is not spinous above; it bears an angular prominence, at most as long as wide at base. The scaphocerite has a very wide scale, with the border regularly curved within; it is from 5.5 to 6.4 times longer than wide; the hairs which border it are at least twice as long as those in S. minus, but, on the other hand, the lateral spine, shorter and more obtuse, does not reach the end of the antennular stalk. The carpocerite is sensibly 4 times as long as wide and more cylin- drical than in S. minis. The large chela has the following relative dimensions: Fingers 1; total length 3; height about 1.15, varying to 1.2. In the large speci- mens the movable finger presents a second obtuse prominence between the point and the molar processes of the lower margin; the palm is less regularly ovoid than in S. ménus, and more tapering on the proximal side. The small chela measures a third of the preceding; the relative dimensions are: Fingers 1; total length 2.35 to 2.4; height 0.65 to no. 1659. AWEHRICAN SPECIES OF SYNALPHEUS—COUTIERE. 51 0.7 (a little narrower therefore than in S. minus). The meropodite is similar in the two species (proportion 2.5). _ In the second pair the proportion of the length to the width of the carpus is about 12:1; furthermore, the meropodite is 0.85 of the length of the carpus. The proportions of the feet of the third pair are the same as in 8. minus save for the meropodites, in which the proportion between the length and the width reaches 4.25. The length of the telson reaches from 1.06 to 1.15 times its large base, always more than twice (2.05 to 2.23 times) its small base, the article being visibly narrower than in S. minus. Fic. 29.—SyYNALPHEUS BREVICARPUS. ad, FRONTAL AND ANTENNAL REGION ; C, CARPOCERITE; €, BGG; K, LARGE CHELA; k, SMALL CHELIPED OF FIRST PAIR, MALE; k’, SMALL CHELIPED OF FIRST PAIR, MALE, S. MINUS (FOR COMPARISON) ; 1, FOOT OF SECOND PAIR; m, FOOT OF THIRD PAIR; m’, DACTYL OF THIRD PAIR; t, THLSON; t’, TELSON, S. MINUS (FOR COM- * PARISON). The eggs are of large size and give rise to mysis larve provided with all their appendages, comprising the chelipeds of the first pair, which are already very unequal, and those of the second pair, in which the carpus is already segmented. The size may reach 36 or even 38 mm. (female). There occurs a remarkable variety of this species represented by some specimens from Key West and also from Porto Rico (gquerint). The frontal teeth are long, especially the rostrum, which last shehtly surpasses the lateral teeth and is also wider at the base; the margins are strongly concave outside instead of being straight, as,in typical Specimens; the basicerite of the antenna is a little more spinous q 59 PROCEEDINGS OF THE NATIONAL MUSEUM. VOL. XXXVIE above; the scaphocerite has its lateral spine more elongate, so that it attains the length of the carpocerite. In these characters the specimens recall the form antillensis of S. minus, but they are perfectly distinct from it; it is as if the two species, while themselves closely allied, had effected in the same way parallel variations. While in the variety antdlensis of S. minus the carpocerite is ovoid, the feet of the second and third pairs are strong, and the telson is wide; in the variety guertni of S. brevicar- pus the carpocerite is slender and cylindrical (proportion 1:4 and even 1: 4.4), the carpus of the second pair is 12 times as long as wide, the meropodite of the third pair is 4.5 to 4.7 times as long as wide, and the telson is narrow, as in the typical specimens of the species. c The chelex of the first pair are those of S. minus. The proportions for the small chela are: Fingers 1; total length 2.26; height 0.75. T have been able to see the eggs on only one female of very small size, I give to the variety the name of BIR i ia Ge A en thd ae guerini because it perhaps corresponds GUERINI. @d, FRONTAL AND ANTEN- to Alpheus sauleyi of Guérin. In the pees ng) a’, PRONE; ¢, caRPO- fioure by that author? the rostrum is longer than the lateral spines and the basicerite appears to be spinous above. It is proper to note that in the nomenclature the trinomial appellation does not imply that this form is derived from S. brevicarpus, the contrary would be as plaus- and infested with a Bopyrid; they are ible; the forms with large eggs, always rare, may be considered as of the same volume as those of S. minus. According to the appearance derived from the species in which the eggs have the usual small size. of the mature ovary of another fe- male, I think that it is their normal size, and that this is besides another character which distinguishes these specimens from S. brevicarpus, at the same time approaching S. minus. Localities : Florida: Elliotts Key, J. E. Benedict. Harbor Key, Union University collection. Key West, H. Hemphill, Bean and King, Eliot, Union ‘University collection. @ Hist. Cuba de Ramon de la Sagra, Pt. 2, VII, 1857, p. 18, pl. 01, fig. 8, ¥ no. 1659. AMERICAN SPECIES OF SYNALPHEUS—COUTIERE. 58 Localities—Continued : Florida—Continued : Key West (form guerini), Union University collection. Dry Tortugas, Eastern Dry Rock, Salt Pond Key, and Florida Bay, Edward Palmer. Bahamas: Andros Island (some coming from sponges), F. Stearns collection. Green Cay, B. A. Bean. Porto Rico: Off Humacao, 9 fathoms, /ish Hawk Station No. 6099 (type of form guerinz). Type of S. brevicarpus guerint.—Cat. No. 24797, U.S.N.M. LA VIMANUS Group. SYNALPHEUS LONGICARPUS (Herrick). Alpheus saulcyi var. longicarpus Herrick, Mem. Nat. Acad. Sci., V, 1891, p. 3888 (part). I have previously shown that this species is closely allied, not to A. brevicarpus Herrick, but to the European species S. devimanus | (Heller). However, it is not synonymous with the latter, any more than any of the forms which follow. I have indicated in the intro- duction to this work why it was necessary to break up into several distinct specific forms A. saulcyi var. longicarpus Herrick. I have retained the original name for the above species, as it appears to me to be the most abundant of the Lavimanus group on the American coasts. The frontal border has three unequal teeth, the median narrow, a little longer, the lateral having from 2 to 2.5 times the width of the median part of the rostrum; their interspaces are in form of a V, with borders little divergent. The basal antennular article is the longer; its anterior margin is less emarginate on the inside than in the greater part of the species of the group. The relative lengths of the articles are 2, 1.5, 1. The antennule is 5 times as long as wide; the flagella are slender, the external bifurcates after the sixth article. The stylocerite reaches the distal third of the basal article. The basicerite has its superior angle obtuse; its lateral spine reaches the extremity of the median article of the antennule. The scaphocerite is almost always devoid of a scale in the male; it bears one of variable length in the female, but it hardly surpasses the extremity of the basal antennular article, and it is never more than half the width of the lateral spine, which is very strong, sharp, and exceeds the anten- nule by about half its distal article. | 54 PROCEEDINGS OF THE NATIONAL MUSEUM. _ vou. xxxvt. The carpocerite is cylindrical, rather slender, curved outward, and surpasses the antennule by the length of the distal article; it is 5 times as long as wide, and sometimes up to 5.5 or 5.6 times. Fig. 31.—SYNALPHEUS LONGICARPUS. «@, FRONTAL AND ANTENNAL REGION, MALE AND FE- MALE; C, CARPOCERITE; K, LARGE CHELA; K’’, CARPUS AND MEROPODITE OF LARGE CHELI- PED; k’, SMALL CHELIPED OF FIRST PAIR, MALE AND FEMALE; kk, FINGERS OF SMALL CHELIPED OF FIRST PAIR; 1, FOOT OF SECOND PAIR, MALE AND FEMALE; ™, FOOT OF THIRD PAIR, MALH AND FEMALE3 ™’’, DACTYL OF THIRD PAIR; ™’’’, DACTYL OF THIRD PAIR OF A VERY ADULT SPECIMEN ; p’, FIRST PLEOPOD; p*, FOURTH PLEOPOD; p>, FIFTH PLEOPOD; Z, THELSON ; tt, TELSON AND UROPODS, FEMALE; wu, UROPOD, The relative dimensions of the large chela are: Fingers 1; total length 4.1; height 1.5; T. L.: H. = 2.73:1. The margin of the palm presents forward a strong tubercle, ending in a fine point. The movable finger has its point out of the perpendicular. The palm | ho. 1659. AMERICAN SPECIES OF SYNALPHEUS—COUTIPRE. 5 is prolonged behind, and the very small carpus is inserted below the principal axis of the ovoid palm. The meropodite, the superior border of which ends in a right angle, is 2.25 times longer than wide. In a comparison of the male and female, the large claws are in about the proportion of 1.3 to 1. The small claw is, to the large one, in the proportion of about 2.5 in the female, of 3 in the male, so that it is apparently of the same size in both sexes; the relative dimensions are: Fingers 1; total length 3; height about 0.8; T. L. : H.=3.75:1. The movable finger bears two teeth, the lower of which is the stronger; the fixed finger bears besides its point, two short angular prominences. The carpus shows some rather remarkable variations; in the larger ‘specimens it is constantly longer than the palm; it is usually a little shorter in medium or very small specimens of either sex. The total length of the chela being taken for a unit, the length of. the carpus may vary from 0.74 to 0.56 and even 0.5; it is always shorter in the female. In every case its distal width remains equal to that of the chela. The meropodite is about 4 times as long as wide (4.4 in the male, 3.6 in the female of large size): The ee pair is notably stronger in the male (1.08), but of similar proportions in both sexes; the first segment of the carpus is shorter than the sum of the other 4 (proportion 1.2) ; the meropodite equals twice the first carpal segment. The terminal claw is longer, in the male, than the last 4 segments, but shorter in the female; it bears, especially in the male, about ten tufts of hair. The third pair is also stronger in the male, where its relative dimen- sions are: Meropodite 2.33, carpus 1, propodite 1.7; the meropodite is 3.5 times longer than wide. In the female these dimensions. are: Meropodite 2.1, carpus 1, propodite 1.55, the meropodite being 3.8 times longer than wide. The ischiopodite is more slender and elongated also than in the male. The dactyl is short, one-sixth or more of the propodite; the two hooks are almost equal and divergent, the ventral normal to the lower border, a little thicker at the base than the dorsal, becoming proportionately longer in specimens of large size (female of 27 to 28 mm.). The sixth abdominal somite shows on either side of the telson a wide triangular point. The pleopods of the fifth somite have a very short base and a wide posterior expansion especially marked in the female, where it contributes to close the incubatory cavity. The anus is shown under the telson between two very prominent swellings. All the abdominal pleura of the male, even the second, end in a point. These last details are more marked in the Lavimanus group than in any other, and es- pecially in the two species S. longicarpus and S. pectiniger. The telson has the following relative dimensions in the male: Small base (distal) 1; large base (proximal) 1.7; height 2.3. In the rep | 56 PROCEEDINGS OF THE NATIONAL MUSEUM. _ vou. xxxvt. female the large base is double the small. The posterior margin bears 4 spines, the inner a little longer, with 4 long plumose hairs between them, and 3 pairs of simple hairs inserted above the preceding. The uropods are larger in the male, the outer especially; the latter bears on its outer margin above the transverse suture a series of 7 to 8 teeth and a movable spine between the first two. The eggs are of small size and the larve are zoée. The species may be found in sponges, but it is not probable that such is its normal habitat. I have been able to separate among the young males a rather large number of specimens different from the type and different also from the following species, S. goodei, with which, how- ‘ kK ever, they agree in hay- ing a well-developed an- tennal scale. Compared to young male longicar- pe pus of the same size, they are distinguished— . (1) By the antennal scale reaching the ex- tremity of the second ar- | a ) | f Fic. 32.—SYNALPHEUS LoNGrcarPus approxima. a, Strong; (4) by the telson FRONTAL AND ANTENNAL REGION; ¢, CARPOCERITE; g little narrower at the K, LARGE CHELA; hk’, SMALL CHELIPED OF FIRST eae : PAIR OF A YOUNG SPECIMEN; kk’, SMALL CHELIPED base ; this is contained 1.8 OF FIRST PAIR OF AN ADULT; ™, MEROPODITE OF times in the heioeht in- THIRD FOOT; t, TELSON. = ticle of the antennule; (2) by the carpocerite a little thicker (propor- tions: 1:4.6, 4.9 or 5, in- stead of 1:5.5 or 5.6); (3) by the large chela in which the anterior border of the palm ends in an obtuse point, conical and stead of 1.4 times. Excepting in the form and size of the antennal scale, which are quite similar, these specimens are also shown to be very distinct from the males of S. goodei of the same size— (1) By the carpocerite a little thicker (the same difference as with S. longicarpus) ; (2) by the large chela, invariably shorter and thicker in S. goodc?, even in the young, not exceeding 12 mm. in length (proportion of thickness 1.1); (8) by the small chela, of which the carpus and the palm are shorter in S. goodei; their sum equals only 4 times the height of the palm instead of 5 times, as in the specimens ' wo. 1659. AMERICAN SPECIES OF SYNALPHEUS—COUTIPRE. 5Y under discussion. The meropodite is also thickened in the same proportion in S. goodei; (4) by the meropodites of the third pair longer and more slender (proportion 1: 4.2 instead of 1: 3.5). I have met only one adult male which appears to be referable to this form; the carpus of the small cheliped measures 0.54 of the total length of the chela. It is accompanied by several other specimens, but they are too incomplete to permit of the appreciation of the fine distinctions which separate S. goodei and S. longicarpus. I hesitate to consider this form as specifically distinct from S. long?- carpus, although the adult specimen differs from it only by the pres- ence of an antennal scale; this is, however, much reduced. Neither is the form of the anterior palmar tubercle very constant in S. longi- carpus, as the small spine which terminates it may be absent. It seems to me sufficient to distinguish these specimens as form ap- proxima. Localities : S. longicarpus— North Carolina: Off Cape Fear, 15 fathoms, Albatross Station No. 2628, 20 to 30 specimens. Gulf of Mexico: Lat. 27° 4’ N., long. 83° 21’ 15’’ W., 26 fathoms, Alba- tross Station No. 2409, 7 specimens. Lat. 26° 33’ N., long. 83° 10’ W., 28 fathoms, Fish Hawk Station No. 7123, 1 specimen. Lat. 26° N., long. 82° 57’ 30’’ W., 24 fathoms, Albatross Station No. 2413, 4,000 to 5,000 specimens. Yucatan : Off Cape Catoche, 25 fathoms, Albatross Station No. 2362, 20 specimens. Off Cape Catoche, 21 fathoms, Albatross Station No. 2363, 15 to 20 specimens. Jamaica, in massive black sponges, 10 to 12 fathoms, J. E. Duerden. Curacao, 2 specimens. S. longicarpus approxima— Gulf of Mexico, 26 fathoms, Albatross Station No. 2409, 4 specimens, type. Gulf of Mexico, 24 fathoms, Albatross Station No. 2413, 2 specimens. Gulf of Mexico, 26 fathoms, Albatross Station No. 2414, 4 larger specimens (mutilated). Type of S. longicarpus approwima—Cat. No. 38398, U.S.N.M. 58 PROCEEDINGS OF THE NATIONAL MUSEUM. VOL. XXXVI. SYNALPHEUS GOODEI, new species. The frontal margin very forcibly suggests that of the preceding species; the rostrum is quite a little longer than the lateral spines, and the latter are more completely triangular. The articles of the antennule have as proportions, 2.3, 1.2, 1, the antennule being 5 times as long as wide; the basicerite of the antenna has its superior angle somewhat sharp; its lateral spine reaches the extremity of the median antennular article. KK Fig. 33.—SYNALPHEUS GOODEI. @, FRONTAL AND ANTENNAL REGION, MALE AND FEMALE} C, CARPOCERITE OF A YOUNG SPECIMEN %; K, LARGE CHELA; AK, LARGE CHELA OF A YOUNG SPECIMEN 7; k’, SMALL CHELIPED OF FIRST PAIR; kk’, SMALL CHELIPED OF FIRST PAIR OF A YOUNG SPECIMEN “7; k’’’, FINGERS OF SMALL CHELIPED OF FIRST PAIR; 1, FOOT OF SHC- OND PAIR; Mm, FOOT OF THIRD PAIR; MM, MEROPODITE OF THIRD FOOT OF A YOUNG SPECI- MEN “7; m’, DACTYL OF THIRD FOOT; t, TELSON; u, UROPOD, MALH AND FEMALD. The scaphocerite always possesses a scale, which is of the same dimensions in both sexes, and often reaches the middle of the distal antennular article; the lateral spine is as in the preceding species. The carpocerite is 5.2 times (in the young) to 5.7 to 6 times longer than wide. “Yor comparison with S. longicarpus approxima, fig. 32. no. 1659. AMERICAN SPECIES OF SYNALPHEUS—COUTIPRE. 59 The large chela has as its proportions, fingers 1; total length 3.5 to 3.6; height about 1.8; the relative total length is a little less in the female, but the difference in size is very slight; the form of the chela is quite different from that of S. longicarpus; the margins of the palm are nearly parallel, the anterior margin terminating in a strong tubercle which is prolonged by a point inclined downward; the carpus is inserted in the prolongation of the greater axis; the meropodite is proportionately stouter (proportion 2.05), its su- perior margin terminating in a prominent, not spinous, lobe. The small chela has the following relative dimensions: Fingers 1; total length 2.8; height 0.96 to 1; the movable finger terminates in two short teeth; the carpus is always much shorter than in 8. longi- carpus; in the largest examples its length does not surpass 0.43 of the small chela, this proportion reaching 0.74 in the preceding species ; in the young this proportion remains the same, while it is very varia- ble in S. longicarpus, the carpus is always a little less thick than the palm, the margins of which are not parallel, as in S. longicarpus; the meropodite is a little more than 3 times as long as wide. I have found no sexual differences. ‘In the second pair, the first segment of the carpus, the sum of the four following and the terminal chela are all very nearly equal; the carpus is at least 7 times as long as wide. The third pair has these proportions in the male: Meropodite 2.2; carpus 1; propodite 1.5 to 1.6; the meropodite is 3 to 3.2 times as long as wide; in the female these proportions become, respectively, 2.6; 1; 2; and the meropodite is nearly 3.5 times as long as wide, the entire appendage being more slender; the dactyl is short, very like that of S. longicarpus. The second abdominal pleuron is not spinous in the male. The telson has its wide base contained 1.24 times, and its small base about 4 times, in its height; the spines of the dorsal surface, especially in the male, are much stronger than those of the posterior border, the inner of which, a little the longer, include between them 4 plumose hairs and 2 groups of 3 simple hairs. The uropods bear upon the outer border 8 teeth in the female and from 9 to 17 teeth in the male, the first and strongest of which pro- longs the transverse suture; and there is also a movable spine placed between the two first teeth. As in 8. longicarpus, the eggs give rise to zoéx. The two species, which are very close to each other, are further connected through the forms which represent them on the Pacific coast. The Paris Museum possesses some specimens collected by M. Diguet in the Gulf of California, which are distinguished from S. goodei by the total absence of an antennal scale and by the presence of 5 to 9 teeth on the external uropod even in the male, characters | 60 PROCEEDINGS OF THE NATIONAL MUSEUM. VOL. XXXVI. which would ally them rather to S. longicarpus; but they approach S. goodei in having the small chela thicker than in S. longicarpus, as ‘ a Fic, 34.—_SYNALPHEWS GOODEI OCCIDENTALIS. da, FRONTAL AND ANTENNAL REGION: @’, FRON- TAL AND ANTENNAL REGION; K, LARGH CHELA; k’, SMALL CHELIPED OF FIRST PAIR; mM, FOOT OF THIRD PAIR; WU, UROPODS. is shown in the following table, and in having the meropodite of the third pair thicker: ioc F Ratio of eeue ee Ratio of me-| length to = ropodite to | width in me- Species. i i oneight in _| small chela. ropodite of sma . | third pair. S. longicarpus, female -....-..------- 3. 5-3. 6 4.0 3.8 S. goodei, female ...... eee Sarat 3.0 3.0 3.5 S. goodei occidentalis, female ......-. 3.2 3.5 3.3-3.4 A female is especially like S. goodei in possessing a rudiment of an antennal scale, a stylocerite longer than the basal antennular ar- ticle, the meropodites of the third pair more similar to those of the female of S. goodei (proportion 3.4), and lastly 12 spines on the outer uropod, as in the male of the last-named species. I propose to desig- nate the example from the Pacific under the name of S. goodei occi- dentalis, remarking that the female last described would probably be found to be more distinct in a more extended series. Named for the late Dr. George Brown Goode, Assistant Secretary in Charge of the United States National Museum. yo.1659. AMERICAN SPECIES OF SYNALPHEUS—COUTIERE. 61 Localities : S. goodei— Gulf of Mexico, 34 fathoms, Grampus Station No. 5088, 1 specimen. Tampa Bay, 6$ fathoms, “ish Hawk Station No. 7109, 1 specimen. Near Colon, 34 fathoms, Albatross Station No. 2147, 1 specimen. Bermudas, George Hawes, 2 specimens. Bermudas, Harrington Sound, in sponges, George Hawes, 7 or 8 specimens. Bermudas, G. Brown Goode, 20-30 specimens, type. S. goodei occidentalis— Lower California, Gulf of San José, M. Diguet, 7 speci- mens (Paris Museum). Type of S. goodei—Cat. No. 24821, U.S.N.M. SYNALPHEUS SANCTITHOMZ, new species. Although very like S. goodei, this species ought certainly to be separated from it. The basicerite has an obtuse superior angle; its lateral spine does not reach the extremity of the median article of the antennule; the antennal scale is 14 times as wide as in S. goode?, and does not ex- ceed the extremity of the median article of the antennule; its lateral spine is very slender, and shorter than the antennule. The very thick carpocerite, which is 4 times as long as wide in the female, only 3.5 times in the male, is the principal distinctive char- acter of this species. The large chela is more slender and elongated than in S. goodez, especially in the female; its relative dimensions are: Fingers 1; total length 3.66; height 1.28, in the male; and respectively 1; 4; and 1.1, in the female; the upper margin of the meropodite is strongly convex and presents no prominent anterior angle. The small chela in the male has the following proportions: Fingers 1; total length 3; height 1; the carpus measures 0.42 of the length of the chela, and the meropodite is 4 times as long as wide; in the fe- male the proportions of the chela become 1; 2.56; 0.8. The two chele are notably smaller in the female (proportions 1.15 for the large; 1.1 for the small chela). In the second pair the first segment of the carpus is smaller than the sum of the four others, and is also smaller than the terminal chela. The proportions of the third pair are: Meropodite 2.43; carpus 1; propodite 2; the meropodite is 4.4 times longer than wide; in the female the proportions become 3; 1; 2.15; and the meropodite is more than 5 times as long as wide, 62 PROCEEDINGS OF THE NATIONAL MUSEUM. vou. xxxvi. The height of the telson is 1.47 times the width of the base and 4.7- times the posterior margin; it is consequently more elongated than in S. goodei; the spines of its upper surface are weak, and the inner spines of the posterior border are more than twice as long as Fic. 35.—SYNALPHEUS SANCTITHOM®. ad, FRONTAL AND ANTENNAL REGION; K, LARGE CHELA, MALE AND FEMALE; K’’, CARPUS AND MBEROPODITE OF LARGE CHELIPED, MALE AND FEMALE; k’, SMALL CHELIPED OF FIRST PAIR, MALE AND FEMALE; 1, FOOT OF SECOND PAIR; mM, FOOT OF THIRD PAIR; t, TELSON ; wu, UROPOD. the outer; the border of the outer uropod has not more than 2 fixed teeth, without an intermediate sutural spine. The eggs give rise to zoée (0.8 mm. in greatest diameter). Localities : St. Thomas, 20 to 23 fathoms, Fish Hawk Station No. 6079, 1 -male and 1 female of very small size (9 mm.), types. St. Thomas, 20 fathoms, Fish Hawk Station No. 6080, 1 female. Type.—Cat. No. 24782, U.S.N.M. SYNALPHEUS GRAMPUSI, new species. The three frontal teeth are equal in length, but the rostrum, with parallel margins, is scarcely one-fourth of the width of the lateral teeth, which are widely rounded at their extremity; the intervals be- tween the rostrum and the lateral teeth have parallel borders, and are therefore U-shaped; the rostrum is placed at a lower level, and is continued backward by a short and narrow crest. no. 1659, AMERICAN SPECIES OF SYNALPHEUS—COUTIERE. 63 The segments of the antennule are to one another as 2.15, 1.8, 1; the stylocerite is a little shorter than the basal article; the basicerite has its upper angle a right angle, and its lateral spine very strong, reaching the last third of the distal segment of the antennule; it has the same width and length as that of the scaphocerite, which bears no trace of a scale in either sex; the carpocerite exceeds the antennule by the length of the distal article, and is 5.5 times longer than wide, a little concave exteriorly. The proportions of the large chela are: Fingers 1; total length 3.3; height 1.3; it is consequently much like that of S. gooded, and like- wise possesses a strong tubercle on the anterior margin of the palm, with a short spine directed toward the base; the meropodite has a straight, not spinous, upper margin. Fic. 36.—SYNALPHEUS GRAMPUSI. d, FRONTAL AND ANTENNAL REGION, MALE; ao, ANOMA- LOUS SPECIMEN; K, LARGE CHELIPED; k’, SMALL CHELIPED OF FIRST PAIR; J, FOOT OF SECOND PAIR; ™, FOOT OF THIRD PAIR, MALE AND FEMALE; t, TELSON; U, UROPOD. The small chela is to the large as 1 to 2.9. Its relative dimensions are: Fingers 1; total length 2.68; height 0.85 to 0.9; both fingers end in a sharp point; the carpus is 0.52 of the entire cheliped; the mero- podite is 3.9 times longer than wide. In the female these proportions become, for the large chela: Fingers 1; total length 3.3; height 1.37; and for the small chela, respectively, 1; 2.78; 0.9; the large chela is therefore a little more thick-set in the female, and the fingers of the small claw are slightly shorter. Tn the second pair the first segment of the carpus equals the termi- nal-chela, the sum of the other 4 segments being greater; the carpus is about 9 times as long as wide. 64 PROCEEDINGS OF THE NATIONAL MUSEUM. VOL. XXXVI. The third pair is very robust in the male. The relative dimensions are: Meropodite 2.33; carpus 1; propodite 1.6; the meropodite is only 3 times as long as wide; the dactyl has its two hooks almost equal, the ventral a little stronger. Tn the female the proportions remain the same, but the dimensions of the second and third pairs are noticeably less (about 0.8). The height of the telson is 1.2 times the width at the base, and 3.9 to 4 times the posterior margin, the last dimension as in the males; the border bears 4 equal and almost equidistant spines, including be- tween them, on a narrow, convex portion of the margin, 5 hairs, of which 3 are large and plumose; the spines of the upper surface are longer than those of the margin. The outer border of the uropod bears 6 to 8 spines, the first large. continuing the border of the suture, the second and following rapidly diminishing; only the last or the last two are not mobile. The eggs give rise to zo0ée. I have never met with any vestige of an antennal scale; on the other hand, a male (Fish Hawk Station No. 7124) shows an interesting variation in the length of the antennal spines, which do not reach to the extremity of the median antennular article; this specimen indi- rates the way in which S. pandionis, the next American species de- scribed, has become differentiated. One female (Grampus Station No. 5116), of which all the eggs are hatched, and the zoée are still present under the abdomen, shows the opposite variation; both the antennal spines equal the antennule, and are also thicker than usual; the stylocerite is also a little longer. Localities : Gulf of Mexico: Lat. 26° 30’ N., long. 83° 30’ W., 33 fathoms, Grampus Sta- | tion No. 5116, a female (not quite typical). Lat. 26° 33’ N., long. 88° 10’ W., 28 fathoms, ish Hawk _ Station No. 7123, 1 male, 2 females, types. Lat. 25° 50’ 15’’ N., long. 82° 41’ 45’ W., 21 fathoms, Fish Hawk Station No. 7124, 2 males, 3 females. Lat. 27° 04’ 00’ N., long. 83° 21” 15” W., 26 dathomig; Albatross Station No. 2409. Type.—Cat. No. 38399, U.S.N.M. The species is very close to the one that I described in a previous paper under the name S. lwvimanus var. parfaiti, and which should also be separated as a distinct species. It is unfortunately repre- sented by a single female, of which the small chela is missing. The frontal teeth and the rostrum are separated by wider intervals with divergent margins; the lateral teeth are wider at the base and less obtuse at the extremity; the rostrum is on a level with them, and no. 1659. AMERICAN SPECIES OF SYNALPHEUS—COUTIERE. 65 is prolonged backward by a very slight, but wide, crest. The articles of the antennule are to one another as 1.8, 1, 1. The lateral spine of the basicerite is a little longer than the anten- nule and also than the spine of the scaphocerite; the latter is a little narrower than the preceding, and carries a very well marked rudi- ment of a scale which does not, however, exceed the extremity of the basal antennular article. The large chela has as proportions, fingers 1; total length 4; height 1.5; it is more tapering distally, and the tubercle on the margin of the palm is less prominent, bearing no spine; the superior margin of the meropodite is convex, and rounded at the distal extremity. The small chela is missing. i Fic. 37.—SYNALPHEUS PARFAITI. @, FRONTAL AND ANTENNAL REGION; K, LARGER CHELA AND CARPUS; K’’, CARPUS AND MEROPODITE OF LARGE CHELIPED; m™, FOOT OF THIRD PAIR} m’, DACTYL OF THIRD PAIR; t, TELSON ; U, UROPOD. In the second pair the first segment of the carpus, the four follow- ing, and the distal chela are approximately equal. The proportions of the third pair are: Meropodite 2.28; carpus 1; propodite 1.7; the meropodite is 3 times longer than wide, as in S. grampusi. The height of the telson is 1.08 times the width of the base, 2.66 times the posterior margin, which latter has its inner spines 2.5 times wider than the outer, and also much stronger; between them are six plumose hairs and two lateral groups of three simple hairs; the spines of the superior face are shorter and stronger than in S. grampusi. Proc. N. M. vol. xxxvi—09——_5 66 PROCEEDINGS OF THE NATIONAL MUSEUM. VOL. XXXVI. The external margin of the uropod bears only two short teeth, with one long movable spine nearer the sutural tooth. The larve are zoée. The type comes from Annobon, off the west coast of Africa (Count Parfaite, Paris Museum). Named for the collector. Synalpheus levimanus (Heller) of the Mediterranean is quite dis- tinct from both of these species; the frontal margin bears three equal teeth, the rostrum being about two-thirds as wide as the lateral teeth, and in height four times its median width, or a little less. ao K mM a “a Fic. 88.—SYNALPHEUS LAVIMANUS. 4, FRONTAL AND ANTENNAL REGION, MALE AND FE- MALE: a’, FRONTAL AND ANTENNAL REGION, MALE WITH RUDIMENTARY SCALE; @”’, FRONTAL AND ANTENNAL REGION, MALE INTERMEDIATE; AK, LARGH CHELIPED; k’’, SMALL CHELIPED OF FIRST PAIR; 1, FOOT OF SECOND PAIR}; m, FOOT OF THIRD PAIR; m’, DACTYL OF THIRD PAIR; t, TELSON; wu, UROPOD. The articles of the antennule are to one another as 1.57, 1.07, 1; the stylocerite is narrow and equals the basal article. The superior angle of the basicerite is obtuse, its lateral spine reach- ing to at least the middle of the basal antennular article; the scapho- cerite is not provided with a scale in the females, where its place is only indicated by a shght prominence of the inner margin of the spine; in the males the scale is of variable length, from a very slight no. 1659. AMERICAN SPECIES OF SYNALPHEUS—COUTIERE. 67 rudiment to a scale reaching the proximal third of the median anten- nular article; its lateral spine, wider than that of the basicerite, hardly reaches beyond the middle of the distal antennular article; the carpocerite exceeds the antennule by one-half of the distal article, and is 6 times longer than wide. The proportions of the large chela, in the male, are: Fingers 1; total length 3.4; height 1.35; the anterior margin of the palm bears a tubercle which terminates in a horizontal spine; the meropodite is rounded and unarmed on its superior margin; the proportions are the same in the female, but the palm is more tapering anteriorly. The proportions of the small chela are: Fingers 1; total length 2.8; height 1; the carpus measures only 0.35 of the entire chela; the meropodite is 3.3 times as long as wide; the fingers terminate in a single point; the plume of hairs of the movable finger is less thick than in S. grampusi. In the female the small chela is only slightly narrower, the proportions being 1, 2.8, 0.92; the meropodite, with the same proportions, is also a little more slender. In the second pair, in both sexes, the first segment of the carpus, the sum of the four following ones, and the distal chela, are nearly equal, but progressively diminish slightly in length; the carpus is 10 times longer than wide. The proportions of the third pair are: Meropodite 2; carpus 1; propodite 1.6; the meropodite is 3.5 times longer than wide; the dactyl is short, terminating in two equal and slightly divergent hooks. In the female the proportions are approximately the same; the height of the telson is 1.25 times its base and 3.6 times its posterior margin, which bears two pairs of feeble spines, the inner ones shghtly the longer; between these are ten plumose hairs. The external uropod bears only two feeble teeth, with a movable spine between them. The eggs give rise to zoéx. SYNALPHEUS PANDIONIS, new species. ~ This species is distinguished from S. grampusi only by very slight differences, of which the principal one is the presence of a well- developed antennal scale. It is also very like 8. parfaiti, which it approaches especially in this last character. However, I believe that these three forms are perfectly distinct. They appear to be the result of different types of variation. The frontal teeth resemble those of S. parfatti, but this is not true of the stylocerite, which is always markedly shorter than the distal article of the antennule, as in S. grampusi: the superior angle of the basicerite is obtuse, its lateral spine reaching the extremity of the median antennular article; the scaphocerite has in both sexes a very distinct scale, which reaches the middle of the median anten- 68 PROCEEDINGS OF THE NATIONAL MUSEUM. vot. xxxvt. nular article, and is sometimes as wide as the lateral spine; the latter is equal to that of the basicerite, or is very shghtly shorter; the carpocerite exceeds the antennule by 14 times the length of the distal article; it is concave exteriorly and 6.5 times longer than wide, being, in consequence, more slender than in 8. grampusi and 8. parfaiti. The proportions of the large chela are: Fingers 1: total . length 3.3; height 1.3 in the male; that of the female is more stocky, the last dimension reaching 1.4; T. L.: H.=2.3:2.5. By its form, by the spine, which is directed obliquely downward, and is on the an- terior border of the palm, this chela is much like that of 8S. grampusi. Z K ms Fic. 39.—SYNALPHEUS PANDIONIS. 4, FRONTAL AND ANTENNAL REGION; C, CARPOCERITE, MALE AND FEMALE; K, LARGE CHELA; k’, SMALL CHELIPED OF FIRST PAIR; 1, FOOT OF SECOND PAIR}; m, FOOT OF THIRD PAIR, MALE AND FEMALE; ft, TELSON. Qe The small chela is to the large as 1 to 3, in both sexes; it is con- sequently smaller than in specimens of S. grampusi of the same size; the proportions are as follows: Fingers 1; total length 2.4; height 0.72; it is therefore more slender than in S. grampusi. The second pair has the same proportions as in S. grampusi, but it is a little more slender (proportion about 1:1.08) especially in the female. The third pair are similar; the meropodites are equal in two spec- imens, one of S. grampusi and one of S. pandionis, of the same length, but the proportion of the length to the width is 3.3 in the first case, 3.8 in the second, even more pronounced in the male; in the female of S. pandionis this proportion is reduced to 3.7, no. 1659. AMERICAN SPECIES OF SYNALPHEUS—COUTIERE. 69 The telson is like that of S. grampusi; the outer uropod bears 4 to 6 spines. The eggs give rise to z0ée. Locality: St. Thomas, 20 to 23 fathoms, Fish Hawk Station No. 6079; 2 males, 4 females. Among the specimens is a female which may be considered an “ oxyceros” form (subspecies eavtentus) of this species, as the spines of the basicerite and of the scaphocerite equal the antennule, and the antennal scale reaches the end of the median antennular article. S. grampusi and S. parfaiti are equally “ oxyceros” rela- tive to S. pandionis, which last may be considered as more prim- itive and less adapted to creeping or fixed life on account of the persistent antennal scale, the more feeble feet, the less armed uropods, ete. Type of S. pandionis.—Cat.. No. 38400, U.S.N.M. Type of S. pandionis extentus—Cat. No. 38401, U.S.N.M. SYNALPHEUS BROOKSI, new species. This species and those following (NS. tanneri, S. herricki, and S. pectiniger) constitute in the Lavimanus group a col- lection of forms closely allied, of small size and often associated. In the absence of the small cheliped it is a difficult matter to 5... 40.synanpanus PANDI- separate S. brooksi and S. pectiniger, aS — oNIS EXTENTUS. a, FRONTAL they both show curious anomalies in the Qiveccmimm number and size of the eggs; after S. longi- carpus (in company with which they are frequently found), they are among the most common forms. S. brooksi has the tridentate portion of the frontal border distinct, joined to the adjacent portions by slightly concave curves; the three frontal teeth are short and equal, the rostrum narrower; the axes of the lateral teeth are divergent. The articles of the antennule are to each other as 1.7, 1.05, 1; the stylocerite, short and wide, reaches about the middle of the basal article; the superior angle of the basicerite is very obtuse, its lateral spine reaching the middle of the median antennular article; the sca- phocerite is absolutely devoid of a scale and is reduced to its lateral spine, which is more slender and very slightly longer than the pre- ceding; the cylindrical carpocerite, a little concave externally, exceeds the antennule by three-fourths of the distal article; it is 70 PROCHEDINGS OF THH NATIONAL MUSEUM. vou. xxxvt. short, only 4.5 times longer than wide (4.4 in the male, 4.6 in the female). ; The sexes frequently differ in the proportionate size of the large chela, but this character is very inconstant. The most massive form, which I have observed in the males, correspond to the follow- Oo ai es Fic. 41.—SYNALPHEUS BROOKSI. a, FRONTAL AND ANTENNAL REGION, MALE AND FEMALE $ C, CARPOCERITE, MALE AND FEMALE; e B, EGG OF NORMAL SIZE; e€ C, EGG OF ABNORMAL SIZE FROM FEMALE, ALBATROSS STATION NO. 23623; e€ D, EGGS OF ABNORMAL SIZE FROM FE- MALE, BLAKE; AK A, LARGE CHELA; K B, LARGE CHELA, FEMALE, BLAKH; K C, LARGH CHELA, ANOMALOUS, ALBATROSS STATION NO. 23862; K D, LARGE CHELA, ANOMALOUS, FE- MALE, BLAKE; K’’, CARPUS AND MEROPODITE OF LARGE CHELIPED; k’, SMALL CHELIPED OF FIRST PAIR, MALE AND FEMALE; k’’’, FINGERS OF SMALL CHELIPED OF FIRST PAIR; l, FOOT OF SECOND PAIR; m, FOOT OF THIRD PAIR; m’, MEROPODITE OF THIRD PAIR; m’’, DACTYL OF THIRD PAIR; ¢t, TELSON, MALE AND FEMALE; t’, EXTREMITY OF THLSON 3 uU, UROPOD. ing dimensions: Fingers 1; total length 3.43; height 1.26 (Alba- tross Station No. 2362), the proportion of the length to the height being only 2.7; but it is much more frequent to find the chela be- coming more slender and this proportion equal to 2.9, 2.97, 3; the no. 1659. AMERICAN SPECIES OF SYNALPHEUS—COUTIERE. a1 extreme cases are those in which the proportion becomes equal to 3.25, the fingers being equal in length to the height of the palm. In the female there are some very similar variations. I have found in a very ovigerous female (all the eggs of which were of normal volume) these proportions: Fingers 1; total length 3.6; height 1.28; the proportion L.: H. being 2.8; it is a chela which one would not know how to differentiate from that of a male, any more by its absolute dimensions than by its size in relation to that of the animal. The extreme cases are those in which the propor- tions become: Fingers 1; total length 2.75; height 0.9; the propor- tion of L.: H. then being about 3.05, and the fingers very elongate; I have met with this last form particularly among some anomalous females, carrying few eggs, very small, and probably sterile. The most typical specimens, among those which appear to me to have been collected together, have their large chelw very dissimilar; as an example, the proportions of a male and a female of the same ‘size from Curacao are given below: | i : . Proportion of Cephalo- Large chela | E gee erporupe the large chela thorax. | (total length). | D to the cephalo- | thorax. * epee ae 3, | | Male (12 mm. long).. BOs iN I 3.00 ia ly 1,24 Female (12 mm, long) Ang. |ytoportiom.1:b { 3,29 1.05 88 It is seen here that the male and female, as is very frequently the case in the Synalpheids, differ in the length of the abdomen, which is longer and especially stouter in the female, where the eggs distend the pleura, and also in the large chela, which is smaller and more slender in the female. A constant character of the large chela is the presence of a conical tubercle, very prominent, and directed a little obliquely upward, which terminates the anterior border. The small chela in the male is to the large in the proportion of 3.3 to 3.4; its relative proportions are: Fingers 1; total length 2.7 to 2.8; height 0.9 to 0.95; the fingers each terminate in two hooks; the carpus measures 0.46 to 0.5 to 0.51 of the whole chela; the meropo- dite is 3.6 times as long as wide. In a female chosen from among the most normal specimens the small chela is to the large one in the proportion of about 2.3; its relative proportions are approximately the same as in the male, the carpus being, however, longer (0.5 to 0.6 of the total chela). In the second pair the proportions of the first segment of the carpus, of the four following ones, and of the chela are 1, 1.2, 1.2; the meropodite measures 0.9 of the carpus in the female; in the male these proportions become 1, 1.4, 1.4, and the meropodite is equal to the carpus. . 72 PROCEEDINGS OF THE NATIONAL MUSEUM. — vot. xxxvt. In the third pair the proportions are: Meropodite 2.26; carpus 1; propodite 1.62; the meropodite is no thicker in the male than in the female (proportions 4.3 to 4.5); the dactyl is short, with two teeth slightly divergent, the ventral a little stronger and shorter. Fic. 42.—SYNALPHEUS BROOKSI STREPSICEROS. @, FRONTAL AND ANTENNAL REGION; C, CARPOCERITE; k’, SMALL CHELIPED OF FIRST PAIR; 1, FOOT OF SECOND PAIR. The height of the telson in the males is a little less than 4 times its posterior margin and 1.23 times its wide base; in the females this last proportion becomes 1.1; between the posterior spines, the inner of which are the longer, are found four plumose hairs, with two pairs of simple divergent hairs; the outer uropod bears only two teeth, between which is a movable spine. The eggs are very large; I have counted at the most sixteen, and they give rise to very advanced mysis larve. When freshly laid they measure about 1.1 mm. in the long axis, and they increase to 1.6 mm. when the sixth pleosomite at- tains the height of the eyes in the larva folded in the egg; but females are also frequently found whose Fic. 48.—Synatpuevs Brooks: exxu- €888 do not exceed 0.5 to 0.6 mm. in THERE. c, CARPOCERITE; K, Larce the long AXIS} these egg@’s are, more- CHELA, FEMALE; k’, SMALL CHELIPED; ‘ m, MEROPODITE OF THIRD PAIR. eee few: (3 to 10) and have a chalky aspect in alcohol; they are probably not destined to develop. These females have in their abdominal pleura the characters of the males, as if they had been castrated by some internal parasite. Named for the late Prof. William K. Brooks, of Johns Hopkins University. The species presents some interesting variations. One male from St. Thomas differs from the types in certain points: (1) The spines of the scaphocerite and basicerite are longer; (2) the carpocerite is ane. no. i659. AMBRICAN SPECIES OF SYNALPHRUS—COUTIERE. 73 slender, 5.5 times longer than wide; (3) the fingers of the small chela are elongated, the proportions being: Fingers 1; total length 2.38; height 0.76; proportion L.: H: = 3.12:1; the carpus is 0.5 of the total length; the large chela is absent; the second and third pairs have their usual characters. This specimen may be distinguished under the name of S. brooksz slrepsiceros. _ The elongation of the carpocerite in this specimen is a variation in the direction of the species herrichi. A second variation, bearing this time on the small chela, charac- terizes some specimens from the Bahamas (3 females, 5 males) ; the carpocerite remains short (proportion 4, 4.1), thicker even than in typical S. brooksi; the spines of the scaphocerite and of the carpo- cerite, especially of the latter, are stronger and longer than in the types; the large chela is similar in the two sexes, and even slightly thicker in the females; the proportions are those of the extreme cases met in S. brooksi. | Fingers. cee Height. Sex. Le: Form louthete ig ae eco. it 3.0 Lian ly Female 2.65 Fp et Bocce Bate i 3.2 1.14 Male .....| 2.8 S. eens extreme. ...... 1 3. 43 1.26 wae O anak Dh ces oes senior nes 1 3.6 1.28 Female.. 2.8 The palm is thus shorter, while remaining as broad, in this char- acter approaching S. herricki; the proportions of the small chela are: Female, fingers 1; total length 2.45; height 0.85; carpus 0.53 of the total length; male, fingers 1; total length 2.5; height 1; carpus 0.57 of the total length; the meropodite is 3.85 times as long as wide in the females, 3.8 times in the males. The males thus seem to show particularly the tendency to the lengthening of the carpus which characterizes S. herricki, but this lengthening is almost as marked in the females, where it is hidden by the elongation of the chela and especially oe the fingers; the feet of the accent = third pairs are those of S. brooksi. These specimens may be named S. brooksz eleuthere. Localities : Bahamas: B. A. Bean, 1 specimen. Andros Island, 1 specimen. The Current, Eleuthera Island, B. A. Bean, 8 specimens, form eleuthera, type, Cat. No. 38408, U.S.N.M. Florida: Harbor Key, Union University collection, 1 specimen. Salt Pond Key, Edward Palmer, about 50 specimens. 74 PROCEEDINGS OF THE NATIONAL MUSEUM. you, xxxvt. Localities—Continued : Florida—Continued : Sugar Loaf Key, 50 males and females (several anomalous), including types. Key West, H. Hemphill, 2 specimens. Gulf of Mexico, 27 fathoms, Albatross Station No. 2372, 40 males and females. Yucatan, off Cape Catoche, 25 fathoms, Albatross Station No. 2362, 80 males and females. Vieques, 14 fathoms, /ish Hawk Station No. 6085, 1 specimen. Vieques, 124 fathoms, Fish Hawk Station No. 6095, 2 specimens. St. Thomas, 20 to 23 fathoms, Fish Hawk Station No. 6079, 2 specimens. St. Thomas, 1 specimen, form strepsiceros, type, Cat. No. 8936, U.S.N.M. Brazil, off Cape St. Roque, 20 fathoms, Albatross Station No. 2758, 1 specimen. Type of S. brookst.—Cat. No. 38402, U. S. N. M. SYNALPHEUS HERRICKI, new species. The tridentate portion is distinct from the rest of the frontal mar- gin, to which it is united by rectilinear borders; the three teeth are approximately equal in length, the rostrum a little narrower than the lateral teeth, which are at least as long as wide at the base, and usually longer. The articles of the antennule are as 2, 1.4, 1; the stylocerite reaches the distal third of the basal article. The superior angle of the basicerite is obtuse; its lateral spine reaches to at least the middle of the median antennular article; it is 1.5 times thicker than the spine of the scaphocerite, which bears no trace of a scale; it is usually, also, very slightly longer, but it may be only equal to it; the two spines are straight and parallel. The large carpocerite exceeds the antennule by the length of the distal article; it is a little concave, cylindrical, 5 times as long as wide in the males, or 4.7 to 4.8 in the females. The proportions of the large chela are very similar in the two sexes: Fingers 1; total length 3.25 to 3.4; height 1.33 to 1.35; the ratio is L.: H.=2.42 to 2.5:1; these figures apply to the males; in the females they become, respectively, 1, 3 to 3.2, 1.2 to 1.35, 2.3 to 2.5; the large chela in the female is generally proportionately broader, with the fingers a little longer; the superior margin of the meropodite is convex and unarmed; it is 2.2 times longer than wide. In the male the proportions of the small chela are: Fingers 1; total length 2.8; height 1; the carpus is always longer than the palm, measuring 0.8 of the whole chela; the meropodite is 3.3 times longer no. 1659. AMERICAN SPECIES OF SYNALPHEUS COUTIERE. q5 than wide; it is thicker than the chela (proportion 1.25), and almost as long as the carpus and the chela joined (proportion 0.79). In the female the proportions become 1, 2.45, 0.87, the fingers being longer; the carpus measures no more than 0.67 of the whole chela; the meropodite is 3 times longer than wide, and is also thicker than the chela; it measures 0.74 of the carpus and the chela together; the size being the same, the sexual differences in the length of the chela ae PE ra WN A Se a Fic. 44.—SYNALPHEUS HERRICKI. d@, FRONTAL AND ANTENNAL REGION, MALE AND FEMALE, a’, FRONTAL AND ANTENNAL REGION OF ANOTHER MALE; C, CARPOCERITE, MALE AND FEMALE; K, LARGE CHELA; K’’, CARPUS AND MEROPODITE OF LARGE CHBELA; k’, SMALL CHELIPED OF FIRST PAIR, MALE AND FEMALE; k’’’, FINGERS OF SMALL CHELA OF FIRST PAIR; 1, FOOT OF SECOND PAIR, MALE AND FEMALE; ™, FOOT OF THIRD PAIR, MALE AND FEMALE; m’, DACTYL OF THIRD PAIR OF TYPICAL MALE; m’’, DACTYL OF THIRD PAIR OF ANOTHER MALE; f, TELSON, MALE AND FEMALE; WU, UROPOD. are expressed by the proportions 1.45 for the large, 1.15 for the small chela. In the second pair the first segment of the carpus is, in the male, a little shorter than the sum of the four others, in the female a little longer; the meropodite is a little longer in the male, the whole mem- ber being more elongated (proportion 1.2). 76 PROCEEDINGS OF THE NATIONAL MUSEUM. vot. xxxvt. In the third pair the proportions are: Meropodite 2.5; carpus 1; propodite 1.55 (male) ; and 2.25, 1, 1.4 (female), by the shortening of the meropodite and the propodite. In both cases the meropodite is about 4 times as long as wide; the dactyl is small, tapering distally, with two slightly divergent hooks, the ventral the stronger and a little the shorter. The height of the telson in the male is 1.6 times its base, 3.7 times its posterior margin; in the female the height equals the base, and is 4.5 times the posterior margin; in both cases the spines of the supe- rior face are very strong, and are i larger than the inner spines of the K posterior margin; between the lat- ter are four plumose hairs and two groups of four simple divergent Fic. 45.—SYNALPHEUS HERRICKI ANGUS- hairs; the external uropod generally TIPES. K, LARGE CHELA; k’, SMALL bears four teeth on its free margin, CES UNE ADO TRAE PATE: and in addition a movable spine; in the males the teeth may be three or two in number. The eggs are of large size and give rise to mysis larve. Named for Dr. Francis H. Herrick, of Adelbert College. This species, like the preceding, shows several variations. Among the very typical specimens from Fish Hawk Station No. 7106 I find a female whose small chela is aberrant. In the females of S. herricki the meropodite and the sum of the carpus and the chela are in the pro- portion of 0.74; in the specimen cited this portion is 0.71, and the meropodite is more slender; the width is not, in fact, greater than that of the palm, which latter is less swollen at the base, its margins being parallel along its whole length, the pro- portions being T. L.: H.=3 instead of 2.6 to 2.8; the carpus is no more than K 0.65 of the entire chela, and it is as wide as the palm at its distal end. In all its other characters this female KB (form angustipes) is a true herricki; in 3 ‘ . = x lig. 46.—SYNALPHEUS HERRICKI its small chela it approaches S. brooksz. DIMIDIATUS. K, LARGE CHELA; Six females from the same station ki, SMALLS CHREIE ED One show some differences in the same di- aa rection, but still more accentuated and not exactly comparable. The palm and carpus of the small cheliped are very typical, the latter measuring nearly 0.8 of the total chela, as in the male of S. herrich?, and the palm bemg swollen at its base; the meropodite measures 0.77 of the carpus and chela together, which is also a character of the male of S. herricki (proportion 0,79), but it is 4 times as long as wide (in- no.1659. AMERICAN SPECIES OF SYNALPHBUS—COUTIERE. uy dg stead of 3.3 in the male, or 3 in the female) and its width is only 0.78 of that of the palm (instead of 1.23 in the male, 1.15 in the female, of S. herricki). This slenderness of the meropodite recalls S. brooksz. On the other hand, the large chela is equally slender, as in that last species: Fingers 1; total length 3.28; height 1.22; proportion T. L.: H. = 2.7 (1, 3.6, 1.28, 2.8 in the female of S. brooksi, in which the large chela resembles more that of the male). The anterior palmar tubercle also ends in a slender point. I shall give to these last specimens the name of S. herrichi dimidiatus. Another variation is presented by a female from A/batross Station No. 2372. The rostrum is narrower and the lateral spines wider and more obtuse than in S. herrickt. The proportions of the large chela Fic. 47.—SYNALPHEUS TANNERI. @. FRONTAL AND ANTENNAL REGION} C, CARPOCERITE; K, LARGR CHELA; k’, SMALL CHELIPED OF FIRST PAIR; 1, FOOT OF SECOND PAIR; ™, FOOT OF THIRD PAIR. are: Fingers 1; total length 3.35; height 1.39; proportion T. L.: H.= 2.4, which approaches near to S. herricki; the anterior palmar tubercle is very obtuse; the small chela differs from that of S. herricki in the short carpus, measuring only 0.56 of the whole chela; the cheliped is also more slender, the width of the meropodite being 1.2 times its ordinary width; in spite of the shortness of the carpus, the mero- podite measures 0.75 of the carpus and chela together, as in the female of herricki, the palm of the small chela being more elongate than in the types of the species (fingers 1, total length 2.6, height 0.8, ratio T. L.: H.=3.2, instead of 1, 2.47, 0.87, 2.8 in S. herricki female) ; the feet of the second pair, rather slender in S. herrichi, 78 PROCEEDINGS OF THE NATIONAL MUSEUM, vou. xxxvt. are here very stout; compared with two specimens of almost equal size (17 mm. for S. herricki female; 15.5 mm. for S. tanneri female type), these appendages are in the ratio of 0.8 in total length and thickness; the proportion of the segments of the carpus is no longer the same, the first segment being here shorter than the sum of the four others, as in the female of herricki; the feet of the third pair are also stouter than in S. herrickd, the proportions of these members being about 0.8; the relative lengths of the several segments are the same as in S. herricki male; there are also some differences in the dimensions of the carpocerite (ratio 5.43 instead of 4.8 in S. herricki female and 5 in S. herrichki male), this being more slender than in the types, and there again appreaching some proportions observed in the male. Although unique, this specimen ought, I believe, to constitute the type of a distinct species, for which I propose the name S. tanner?, in honor of the late Z. L. Tanner, formerly commander of the Adldba- tross. Localities (for S. herricki and allies) : Anclote, Florida; Capt. Thomas Low; about 150 specimens (types of S. herricki). Gulf of Mexico, lat. 25° 50’ 15”’ N., long. 82° 41’.45” W.5 21 fathoms, Fish Hawk Station No. 7124, 1 specimen (S. her- ricki). Anclote Section, Florida, 124 fathoms, Fish Hawk Station No. 7106, about 30 specimens (S. herrich?). Anclote Section, Florida, 124 fathoms, Fish Hawk Station No. 7106, 6 specimens (type of form dimidiatus). Anclote Section, Florida, 124 fathoms, Fish Hawk Station No. 7106, 1 specimen (type of form angustipes). Gulf of Mexico, lat. 29° 15’ 30’ N., long. 85° 29’ 30’” W., 27 fathoms, Albatross Station No. 2372, 1 specimen (type of S. tanne?t). Type of S. herricki.—Cat. No. 38404, U.S.N.M. Type of S. herrichi dimidiatus.—Cat. No. 38405, U.S.N.M. Type of S. herrichi augustipes.—Cat. No. 38406, U.S.N.M. Type of S. tannert.—Cat. No. 38407, U.S.N.M. SYNALPHEUS PECTINIGER, new species. While recalling the preceding species by its small size and very large eggs, this form is also closely allied to S. longicarpus Herrick, and may easily be confounded with small specimens of that species. The frontal margin has three wide teeth, the median narrower in its distal half and a little longer than the lateral, but the width of the intervals separating them is always greater than their depth. no. 1659. AMERICAN SPECIES OF SYNALPHEUS—COUTIERE. 79 The proportions of the antennular articles are 2, 1.15, 1; the flagella are stout, the external one bifurcated after the fifth article. The stylocerite is usually a little shorter than the basal article, though often equaling it, especially in the females. The superior angle of the basicerite is a right angle, its outer spine very strong, a little shorter than the first two articles of the anten- Fig. 48.—SyYNALPHEUS PECTINIGER. d, ANTERIOR HALF, FEMALE; @’, FRONTAL AND ANTEN- NAL REGION, MALE; A, LARGE CHELIPED; k, SMALL CHELIPED OF FIRST PAIR, MALE AND FEMALE; k’’’, FINGERS OF SMALL CHELIPED OF FIRST PAIR; Kk’’’, REVERSE OF SAME; l, FOOT OF SECOND PAIR, MALE AND FEMALE; mm, FOOT OF THIRD PAIR, MALE AND FEMALE; m’, DACTYL OF THIRD PAIR; ¢t, TELSON, MALE AND FEMALD. nule. The scaphocerite is absolutely without scale in both sexes; the lateral spine which alone represents it has a concave inner mar- gin, and does not even present at its base a convex prominence mark- ing the place of the absent scale; this spine, at least in its distal half, is more slender than that of the basicerite, in contrast to S. longi- 80 PROCEEDINGS OF THE NATIONAL MUSEUM. vou. xxxvt. carpus; it is also shorter than the antennule, especially in the fe- males, while in S. dongicarpus it 1s always longer. The carpocerite surpasses the antennule by two-thirds or only one- half of the distal article; it is cylindrical, a little concave on the out- side, and 6.8 times as long as wide. In the males the proportions of the large chela are: Movable finger 1; total length 3.5; height 1.2. The anterior margin of the palm bears a strong, sharp-pointed, conical prominence directed obliquely upward. The movable finger is out of the perpendicular for at least half of its length, the inferior margin of the palm rising ab- ruptly in the place of the fixed finger, which is unprovided with any point and serves only to receive in its cavity the inferior processes of the opposing finger. In the females, with the same general form, the large chela is much more slender, its proportions becoming: Movable finger 1; total length 5; height 1.3. The superior margin of the meropodite is a little convex, terminating in a right angle. The proportion of the large chele in the two sexes is about 1: 1.7. In the male the proportions of the small chela are: Fingers 1; total length 2.58; height 0.88. The carpus measures 0.62 of the total length. The meropodite is 3.8 times longer than wide. In the female the proportions become 1, 3.2, 1.1, the fingers being shorter. The carpus measures only 0.56 of the total length, and the meropodite is 3.3 times longer than wide. The proportion of the small chelze in the two sexes is hardly 1: 1.06. This appendage is quite characteristic of the species: In both sexes each of the fingers is terminated by a plate divided into three curved and obtuse teeth; on the movable finger, which appears truncated, the teeth are equal anal more and more faclinied downward; on the fixed finger the innermost tooth is reduced to an obtuse prominence. As the teeth cross each other when the chela is closed, they constitute an effective implement for dividing the prey. The proportions of the second pair are: First segment of the carpus 1; sum of the four following 1.3; chela 1.15. The proportions are the same in the two sexes, the entire member being more robust in the female. The meropodite measures 0.9 of the carpus. The proportions of the third pair are: Meropodite 2.5; carpus 1; propodite 1.64. The meropodite is 4.3 times longer than wide. In the female these proportions become 2.2, 1, 1.5, and the meropodite is a little less thick (4.1). The dactyl has two hooks directed in the same plane as the inferior border, at least in the case of the dorsal hook, which is longer and stronger than the ventral. All the abdominal pleura, in the male, terminate in a strong tri- angular point; even the second and the sixth pleosomite are pro- longed in two strong spines on both sides of the base of the telson. no. 1659. AMERICAN SPECIES OF SYNALPHEUS—COUTIERE. 81 The height of the telson, in the male, is 1.24 times its base and 5 times its posterior margin; in the female, the height hardly exceeds the base, and is 3.7 times the posterior margin. In the female, the spines of the superior face are situated on the proximal third; in the male, on the proximal half. Between the inner spines of the posterior margin, which are twice as long as the outer spines, there are three plumose hairs. The external uropod bears two teeth on its margin, and near the inner tooth a movable spine which is longer in the female. The basal spine of the uropod is strong and curved. Fig. 49.—SYNALPHEUS PECTINIGER, ABDOMEN. @ NORMAL; AQ, B?, CP, DQ, DIFFERENT DEGREES OF VARIATION IN THE FORM OF THE PLEURA; Co’, NORMAL; Eo’, ABNORMAL, The eggs are very large and give rise to mysis larve. I have men- tioned above that this species at an overstocked station (A/batross Station No. 2413, 320 males, 230 females) presents a considerable ex- cess of males with a marked sterility of the females, as if the latter were more or less completely castrated. All the females have the fourth and fifth abdominal pleura ending in a sharp point, as in the males; all, save two have the first pleuron spinous; in the very great majority of the females, even when ovigerous, the second and third 09 —-6 Proc. N, M, vol, xxxvi 89 PROCEEDINGS OF THE NATIONAL MUSEUM. vou. xxxv1. pleura also have an obtuse point. In some cases the sex is difficult to determine, as all the pleura are strongly spinous; one can arrive at it, however, to a very great degree of approximation, by noting that the third pleuron in the males is abruptly terminated in a long point, while in the female specimens, even those most doubtful as to sex, this point is wide and arched. The species appears very homogeneous and I have not been able to separate any variety from the typical specimens. Localities : Gulf of Mexico, lat. 26° N., long. 82° 57’ 30’’ W., 24 fathoms, Albatross Station No. 2418, 320 males, 230 females (with S. longicarpus). | Gulf of Mexico, lat. 25° 04’ 30’’ N., long. 82° 59’ 15’’ W.; 26 fathoms, Albatross Station No. 2414, 1 specimen. Florida, Sugar Loaf Key, 4 specimens. Bahamas, Eleuthera Island, 2 specimens, male and female (largest seen, 12 and 13 mm). St. Thomas, West Indies, 2 specimens. Curacao, Albatross, 126 males, 167 females, types. Curacao, Albatross, 2 specimens (Cat. No. 7595). Type.—Cat. No. 38408, U.S.N.M. SYNALPHEUS ANDROSI, new species. This species is represented by a single female. The frontal margin bears 3 equal, obtuse teeth, the rostrum a little less thick than the lateral teeth; the tridentate region is distinct from the rest of the frontal border. Antennular articles as 1.7, 1.15, 1. Stylocerite wide, shorter than the basal article. Superior angle of the basicerite straight, lateral spine reaching the middle of the median antennular article. The scaphocerite is reduced to its lateral spine, which is as long as the antennule, and a little wider than the spine of the basicerite. The carpocerite surpasses the antennule by more than the length of the distal article, and is 7 times as long as wide. The proportions of the large chela are: Fingers 1; total length 3.4; height 1.4; it is regularly ovoid and the anterior palmar border bears only a weak conical prominence. The meropodite is unarmed on its superior border. The small chela measures: Fingers 1; total length 2.56; height 1.2; it is consequently short and thick. The movable finger is ter- minated by only one sharp point; it is strongly curved, stout at its base, and bears an obtuse tubercle at the middle of its lower margin. The carpus measures 0.47 of the whole chela; it is less thick than the palm, both measured at the distal extremity (0.73). The small claw and the large one have nearly the ratio of 1 to 2. . no. 1659. AMERICAN SPECIES OF SYNALPHEUS—OOUTIERE. 83 In the second pair the first segment of the carpus, the sum of the four following, and the distal chela are apparently of the same length. The third pair is very characteristic of the species. Like the small chela of S. pectiniger, its form is, so far as known, unique in the genus Synalpheus. Its proportions are: Meropodite 1.75; carpus 1; propo- dite 0.92. The meropodite is 3 times as long as wide; its ventral bor- der is widened in the distal half into a flattened surface, which is a little excavated, and margined on the outer side by a transparent wing, on the inner side by a crest much less visible, but bearing some short, strong hairs. The very elongate carpus, also flattened on the ventral side, is like- wise bordered by an outer wing larger than that of the meropodite and capable of concealing it. On the inner side, the crest which bor- ders it bears 5 teeth and some hairs. The propodite itself has upon nearly all its length a crest which seems to be determined by the pressure of that article against the lower border of the meropodite Fie. 50.—SyYNALPHEUS ANDROSI. d@, FRONTAL AND ANTENNAL REGION; K, LARGE CHELA} k, SMALL CHELIPED OF FIRST PAIR; m, FOOT OF THIRD PAIR; Wie, DACTYL OF THIRD PAIR; m’’, CARPUS AND MEROPODITE OF THIRD PAIR; t, TELSON. when the leg is fully bent. In this position—which explains why the form of the carpus is more curved than is customary near its articu- lation—the distal end of this article is applied against a short non- excavate portion of the flattened meral surface, so that between it and the surface of the carpus there exists an interval closed outwardly by the two transparent superimposed plates. There exists in some species of Alpheus of the “ crinitus ” group, such as A. paralcyone, a form somewhat analogous but much less accentuated. The two hooks of the dactyl are almost equal and a little divergent. The telson bears on its posterior border 11 plumose hairs between the inner spines, which are 3 times as long as the outer spines. The external ramus of the uropod bears a small movable spine between two adjacent teeth. The type is a female from Andros Island, Bahamas; F. Stearns collection (Cat. No, 38409, U.S.N.M.). 84 ' PROCEEDINGS OF THE NATIONAL MUSEUM. vou. xxxv1. SYNALPHEUS RATHBUNA, new species. The frontal margin suggests S. goodez,; the rostrum is narrow, with parallel margins, hardly one-sixth of the width of the lateral teeth and slightly longer; the lateral teeth have almost exactly the form, inverted, of the intervals between them and the rostrum, but a little narrower. The articles of the antennule are to one another as 1.2, 1, 1; they are a little wider at the distal extremity and the flagella are stout; the stylocerite reaches the distal third of the basal article. The superior angle of the basicerite is prolonged in a strong spine reaching as far forward as the stylocerite. This is the only case that K Fic. 51.—SYNALPHEUS RATHBUN®. 4, FRONTAL AND ANTENNAL REGION; K, LARGE CHELA$ k’, SMALL CHELIPED OF FIRST PAIR; 1, FOOT OF SECOND PAIR; mM, FOOT OF THIRD PAIR; m’, DACTYL OF THIRD PAIR; t, TELSON ; U, UROPOD. I have as yet noticed in the Lzyimanvs group, so that this detail enables one to identify the species immediately. The outer spine, rather slender, reaches the middle of the distal antennular article. The scaphocerite is reduced to its lateral spine, which has a con- cave inner margin, is much wider than the spine of the basicerite, and reaches the middle of the distal antennular article. The carpocerite surpasses the antennule by hardly one-half of the same article; it is 4.6 to 4.8 times longer than wide. The proportions of the large chela are: Fingers 1; total length 3.5; height 1.25, But I have also found in one of the few specimens car- NO. —— AMERICAN saat OF SYNALPHBUS—COUTIERD. 85 rying eggs the pr oportions i; 3, 1.2. The palm bears in front a conical tubercle, not spinose, pointing opniaiely upward. The movable fin- ger slightly exceeds the fixed finger. The small chela measures: Fingers 1; total length 2.6; height 0.95, The fingers terminate in a single point. The carpus measures 0.5 of the whole chela. The meropodite is very thick, only 2.35 times as long as wide. The second pair is very remarkable in that the carpus has only four articles. I have encountered the same number in young specimens of S. longicarpus, and especially of S. brooksi, but very exceptionally. Here it is a constant character. It is not certain, to tell the truth, that the specimens examined are normal, at least the females. In about thirty of the specimens I have been able to find only five carrying eggs. Four of these females each possess but one egg, the fifth has only three. Their abdominal pleura are not only very slightly developed. but they are all terminated by a very sharp point, and the second pleuron is hardly wider than the first and the third. As the total length of the largest specimen is 7.5 mm., it is possible that I have had in my hands only dwarfed or emasculated individuals, not showing the true sexual characters of the species. Perhaps in specimens of larger size, if such exist, the second pair would have five segments in the carpus, as in the great majority of the Alpheide, the genus Arete (with four segments) being the only exception. The proportions of the third pair are: Meropodite 2 2 earpus Bull. Amer. Mus. Nat. Hist., XIX, p. 691. ¢The Ceratopsia, p. 125. Proce. N, M. vol. xxxv1I—08———7 98 PROCEEDINGS OF THE NATIONAL MUSEUM. VOL, XXXVI. into two, one for each side. Hatcher says that in Torosaurus latus there is a pair of these foramina, separated by a thick partition of bone that marks the median line of the skull. Now, to the writer it seems perfectly clear that the foramen in question represents the supratemporal fossee of the alligator and of various other reptiles. The bone rising up and dividing the fossa into two, either at the surface or deeper down, will almost certainly be found to be the true parietal. Through the enormous thickening of the postfrontals the parietal has been crowded from the upper sur- face of the skull of 7riceratops and the two supratemporal fossx have been pushed into one at the midline. As regards the fosse in the frill, they may be explained as gaps between the squamosal and the supratemporal of each side. Hatcher ? tells that these openings communicated with the temporal fosse and with the cavities at the bases of the horn cores. These passages must represent the posttemporal fosse. Possibly through them passed branches of the temporal arteries to spread themselves over the upper surface of the frill. Hatcher has, in my opinion, erred in his interpretation of the brain- case of the Ceratopsia. His errors have arisen partly from the nearly complete or complete absence of sutures in the skulls examined and partly from his not having recognized the differences between the rep- tilian skull and that of the mammals, a group with which he was more familiar. The latter cause of error is shown in his failure to mention the prootic bone, one of the most constant in the reptilian skull, and in his speaking of all of the bones of the skull in front of the exoccipitals as the alisphenoid, as if it corresponded to the sphe- noid of human anatomy. Again, on page 39 of The Ceratopsia, the organ of hearing is misunderstood because it did not present mam- malian characters. The specimen before me (Cat. No. 2416, U.S.N.M.) appears to agree quite closely with Hatcher’s fig. 8 representing the same parts of T. flabellatus, although the latter is about a third larger than No. 2416. Now, in both skulls there are undoubtedly present the basi- occipital, the basisphenoid, the presphenoid, exoccipitals, prootics, alisphenoids, and orbitosphenoids. In fact, it appears that the brain- case of the Ceratopsia was the most complete of any known reptile. Inasmuch as the sutures between the bones are mostly effaced, we are able to determine their limits only approximately, guided partly by the orifices for the nerves, partly by the courses of the sutures in living reptiles, but especially by aid from the fine skull of Campto- saurus now being studied by Mr. Charles W. Gilmore, who has gener- ously allowed the writer to examine it. @The Ceratopsia, p. 151. bIdem, p. 125, No. 1660. SKULL OF TRICERATOPS—HAY. 99 The identification of these bones and foramina may begin with the foramen ovale, that opening through which the fifth nerve escapes from the brain cavity. This has been correctly identified by Hatcher." In Cat. No. 2416 this foramen has, on the inner surface of the brain- case (Plate 2, fig. 1, 4), a diameter of about 12 mm., and it is of a somewhat squarish form. On the outer surface (Plate 1, fig. 1, 5 **) the diameter is about 16 mm. The bone here has a thickness of about 28 mm. It may be here remarked that the foramina of this skull are usually really short canals, having inner and outer ends. In the passage for the fifth nerve, not far from its inner end, there is given off a large canal which is directed forward, emerging on the outer surface of the skull about 30 mm. in front of the foramen ovale (Plate 1, fig. 1 5‘). Through this canal passed forward the ophthalmic branch of the fifth nerve. Hatcher writes that the anterior opening is the foramen rotundum, and conveyed the maxillary branch; but the structures here are identical with those in the alligator. Out of the external end of the foramen ovale (Plate 1, fig. 1, 5% *) issued the second and the third branches of the fifth nerve. Now, the foramen ovale is situated between the prootic and the ali- sphenoid bones. In the alligator the larger part of it is in the prootic; and the same is the case in Camptosaurus. Hence, in Triceratops the suture between the prootic and the alisphenoid may be provi- sionally drawn through the front of the foramen, carrying the suture up to the parietal. On the inner surface of the brain-case (Plate 2, fig. 1, 6), a little below the foramen ovale, there is seen the posterior orifice of a canal 4 mm. in diameter, for the transmission of the sixth nerve, just as may be seen in the alligator. The canal runs the length of the basisphe- noidal bone, emerging at its anterior end. In Hatcher’s fig. 8 it is indicated by the letter z and explained as being an undetermined foramen. In Hatcher’s figure just quoted there is indicated by the letters eam a small foramen just behind the foramen ovale. This is explained with a query as being the internal auditory meatus. Being on the outside of the brain-case it can not be that meatus. In fig. 24 the same foramen is said to be the external auditory meatus. The meatus properly so-called is a part of the external ear and this reptile proba- bly had no such organ. The foramen in question is that for the escape of the seventh, or facial, nerve (Plate 1, fig. 1, 7). It has the same position as in the alligator and, as in the latter animal, goes straight through the prootic bone. Outwardly it opens between two descending ridges of bone, which enclose a smooth groove, along which the nerve passed downward. A similar groove is seen in the alligator and in Camptosaurus. 4@The Ceratopsia, p. 17, fig. 8, fo. 100 PROCEEDINGS OF THE NATIONAL MUSEUM. VOL. XXXVI. On the inner surface of the brain-case, behind the foramen just described, is a large oval opening (Plate 2, fig. 1, 70) whose greater axis, 18 mm. long, is directed upward and backward. Hatcher (p. 16) has called this the foramen lacerum posterius. It is the common mouth of two short canals which emerge, the one behind the other, on the outer surface of the skull (Plate 1, fig. 1, fen and 70). Be- tween the two exterior openings there is a great ridge of bone 10 mm. thick, that ascends from the basioccipital process to the paroc- cipital process of the exoccipital. In the alligator a corresponding ridge forms the lateral boundary of the lower part of the exoccipital, and it is probable that in 7'riceratops the ridge is on the exoccipital. The anterior of the two exterior openings mentioned is the larger. It is funnel-shaped, the mouth of the funnel having a horizontal diam- eter of about 20 mm. and a vertical diameter of 15 mm. The hinder canal likewise expands as it approaches the surface, and forms a triangular foramen whose diameters are about 10 mm. and 15 mm. In the figure referred to last this opening is hidden by the ridge of bone described. The line from 70 is directed to it. Hatcher “ has identified the anterior of these foramina as the outer end of the foramen lacerum posterius, the hinder as the place of exit of the tenth and eleventh nerves. However, on page 37 he writes that he has interpreted the anterior branch from the internal foramen lacerum posterius as having conveyed the tenth nerve to the brain. He there states further that some anatomists may regard the anterior of the two outer foramina as the external auditory meatus, its internal opening as the internal auditory meatus. There is no doubt in the mind of the writer that the anterior of the outer foramina in question is the fenestra ovalis, the opening into the vestibule of the internal ear. In life it was probably partly or wholly closed by the expanded end of the stapes. So far as known to the writer, this bone has not yet been found in any member of the Ceratopsia. It was quite certainly a long slender rod, which ex- tended from the fenestra ovalis to the outer surface of theeskull, run- ning first below the paroccipital process, then behind the quadrate, reaching the skin in the notch found in the lower border of the frill. It could hardly have been less than a foot in length. A very similar stapes is found in the alligator and in most other reptiles. The hinder of the two external foramina discussed above is the proper foramen lacerum posterius, or jugular foramen (Plate 1, fig. 1, 10), and it transmitted the ninth, tenth, and eleventh nerves of its side, besides also the jugular vein. The foramen credited by Hatcher to the eleventh nerve is the anterior condyloid foramen and probably transmitted a vein. The foramen for the twelfth nerve (Plate 1, fig. 1, 72) is correctly identified. “The Ceratopsia, p. 16, fig. 8. No. 1660. SKULL OF TRICERATOPS—HAY. 101 We must now investigate further that opening on the inner surface of the brain-case (Plate 2, fig. 1, 8, 10) from which diverge the two short canals considered above, the one to the fenestra ovalis, the other to the foramen lacerum posterius. On comparison with the alligator there can be no doubt that we have here a confluence of the opening for the auditory nerve and that for the transmission of the ninth, tenth, and eleventh nerves, the lower part of the foramen hay- ing been devoted to the auditory, the upper part to the other nerves. In the alligator the latter nerves pass out through a long fissure, which is separated from the foramen for the auditory nerve by only a narrow process of bone. Had the cartilage of this process not be- come ossified the two openings would have appeared in the dried skull as a single one, as it does in 7'riceratops. Hatcher thought that the foramen for the facial nerve, his internal auditory meatus, communicated with the small cavities which he represented in his fig. 31 and indicated by the numeral III. The cavities shown there are quite certainly sections of the anterior and posterior semicircular canals, the larger section being the commissure of the two canals. The anterior section belongs in the prootic bone, the posterior in the opisthotic portion of the exoccipital, while the larger section is in the line of union of the two bones. The supraoc- cipital bone must have descended nearly to the level of these sections. Communication with these canals was had from within the skull by means of the foramen transmitting the eighth nerve, from without by means of the fenestra ovalis. In all these respects we have here the normal reptilian condition, and we have no reason for thinking that the Ceratopsia were deficient in hearing. Now, the fenestra ovalis lies between the prootic bone and that part of the exoccipital that had its origin from the opisthotic. The suture between the prootic and the exoccipital may then be drawn through the fenestra ovalis and carried upward to the supraoccipital, as it runs in the alligator and in Camptosaurus. Slightly behind and about 28 mm. below the foramen ovale, that exit for the fifth nerve, is the mouth of the canal for the internal carotid artery (Plate 1, fig. 1, car). The other end of the canal is found in the pituitary fossa (Plate 2, fig. 1, ca). The hinder open- ing appears to be in the posterior end of the basisphenoid bone. A long shallow groove (Plate 2, fig. 1, car. g) on the underside of the basioccipital process leads forward to it. From the front end of the pituitary fossa a short canal (Plate 2, fig. 1, op. 7; Plate 1, fig. 1, op. f) runs forward and opens on the outer surface of the bone. The outer opening is indicated in Hatcher’s fig. 8 by the letters pf. It seems probable that this canal conveyed to the orbit the ophthalmic branch of the internal carotid artery, a vessel that in man escapes through the optic foramen. 102 PROCEEDINGS OF THE NATIONAL MUSEUM. VOL, XXXVI. The optic foramina (Plate 1, fig. 1, 2; Plate 2, fig. 1, 2) have been correctly identified by Hatcher. Behind each of these openings is another (Plate 1, fig. 1, 3; Plate 2, fig. 1, 3) which he has correctly called the sphenoidal fissure. Through it passed to the orbit the third nerve and possibly the fourth. This fissure, or rather foramen, lies in the boundary between the orbitosphenoid and the alisphenoid bones. Indeed, in our specimen the suture between the two bones has remained open, so that the limits of the orbitosphenoidal segment may be traced. Each optic foramen is in the orbitosphenoid of its side, near its hinder edge. The bones of the two sides meet above the olfactory lobes and the suture is closed, but over the cerebral hemisphere there is a fontanel, which was closed by either the frontals or the postfrontals. As shown by Hatcher’s fig. 27, al, the orbitosphenoids met over the olfactory lobe. The cerebral hemispheres and the olfactory lobes of 7’. serratus rested on what must be regarded as the presphenoid bone (Plate 1, fig. 1, pre; Plate 2, fig. 1, pre). This is a triangular plate which, just in front of the optic foramina, descends 50 mm. below the floor of the brain-case, while anteriorly its free border rises to the place of exit of the olfactory nerves. Its lower hinder angle is thin, but the bone thickens toward the brain. Its hinder border appears to have joined the basisphenoid by a suture not closed at the death of the animal. Hatcher’s fig. 27 shows the presphenoid of 7’. horridus. The lower line leading from a is directed to it. In our specimen of 7’. serratus the sutures between the presphenoid and the two orbitosphenoids are obliterated, as they were in 7. horridus. The upper surface of each orbitosphenoid is very rough, for sutural union probably with the frontals, although Hatcher’s description (p. 18) and his figs. 9, 24, and 27 represent the postfrontals as pushing themselves below the frontals in this region. Above the optic foramen, opening into the upper part of the brain- case and near the hinder border of the orbitosphenoid, are two smaller foramina (Plate 1, fig. 1, 4, v.; Plate 2, fig. 1, 4, v.), the one behind the other. From each, on the outer surface of the bone, a groove is directed forward for a short distance. It seems probable that the hinder of these gave exit to the fourth, or trochlear, nerve. The anterior must have transmitted a blood-vessel. The olfactory nerves (Plate 2, fig. 1, 7) left the brain-case through a single orifice; at least, no bony partition separated them as was the case with T. horridus. WHowever, near the anterior end of the olfactory canal there is seen a longitudinal ridge on the upper midline, which formed a partial division of the olfactory lobe. The parietal formed most of that part of the roof of the brain-case which covered the optic lobes. Anteriorly it joined the united orbitosphenoids (Plate 2, fig 1, pa). NO. 1660. SKULL OF TRICERATOPS—HAY. 108 On the lower surface of that bone which the writer regards as the supraoccipital (Plate 2, fig. 1, soc), near its anterior end, apparently between it and the parietal, and placed right and left of the mid- line, are found two deep excavations. The mouth of each of these measures about 20 mm. fore and aft and about 15 mm. transversely. The depth amounts to 15 mm. The diameters diminish toward the upper ends of the excavations. The one on the left side (Plate 2, fig. 1, ceb. f) appears to have reached the external surface of the bone, forming a foramen. Whether the one on the other side reached the surface is uncertain, on account of some crushing. Into these excavations there penetrated probably portions of the brain. These will be considered below. Near the upper border of each orbitosphenoid there is found another excavation similar to those just described, having a some- what larger base, but not entering so deeply into the bone. The base measures 20 mm. fore and aft and 15 mm. transversely. From the front of each excavation a foramen (Plate 2, fig. 1, v) pierces the bone, as already mentioned. Professor Marsh * published a figure of a cast of the brain-case of the specimen here described. This has been reproduced by Hatcher.” This figure represents a side view of the brain. Another figure giv- ing a view of the lower surface of the brain of this specimen is pub- lished by Hatcher.* In that figure the letters VIII on the left side ought to be changed to VII; VIII should be connected with the anterior part of the mass indicated by X; and XI ought to be erased. In order to represent more accurately the brain of this specimen, a new cast has been prepared by Mr. William Palmer, of the National Museum, under the superintendence of Mr. Gilmore and the writer. The parts of this brain, as represented by the cast, are indicated in the legend affixed to each figure of Plate 3. Attention must be specially called to certain structures found on the upper surface and which filled the excavations already mentioned. The hinder pair of these is shown on Plate 3, figs. 1, 3, ceb. p. These bodies are near the boundary between the cerebellum and the optic lobes. They are probably parts of the former. Andrews” interprets a strong development of brain substance in the same region in 7ywanodon as the cerebellum; but that development formed a conspicuous band which culminated in the midline above. In 7viceratops the lateral masses are far removed from each other. Marsh’s figure of the brain does not«adequately represent these masses. On the upper surface of the anterior end of the brain there is seen another pair of processes (Plate 3, fig. 1, 3 cer. h), not rising, 7YPinosaurs of North America, pl. Lxxvit, fig. 4. >The Ceratopsia, p. 39. €Tdem, p. 37, fig. 32. @Ann. Mag. Nat. Hist., 6th ser., XIX, p. 587, pl. xv1. 104 PROCEEDINGS OF THE NATIONAL MUSEUM. vot. xxxvt. however, to such a height as the hinder pair. These probably repre- sent the cerebral hemispheres. As mentioned above, there is a fora- men placed at the front of each cerebral excavation. It probably trans- mitted blood-vessels; for certainly no nerve left that part of the brain. The extensive space between the cerebellar processes and the cere- bral hemispheres was probably mostly occupied by the optic lobes (Plate 3, fig. 1, op. 1). Doubtless, an exact model of the brain would show here and farther in front a deep longitudinal cleft; also transverse clefts in front of the cerebellum and behind the cerebral hemispheres. In Marsh’s figure of the cast of the brain the olfactory lobe seems to be too long. In the figure here shown (Plate 3, figs. 1, 3, olf. 7) it is a little too short. Its length is 55 mm. Where it escaped from the olfactory canal its divisions are seen to have been directed to the right and left, respectively. THE BRAIN-CASE AND THE BRAIN OF TRICERATOPS SULCATUS. In the U. S. National Museum there are important parts of a skull that has been identified as that of 7’. sulcatus. The specimen is Cat. No. 4286, U.S.N.M. It was collected by Mr. Hatcher in the so- ralled Laramie beds of Converse County, Wyoming, for the U. S. Geological Survey. Hatcher has presented a figure of the horn cores. The brain-case, including the occipital condyle, has been sawed from the horn cores along a horizontal plane that passed somewhat above the cerebral hemispheres. Afterwards the brain-case has been divided along the median plane, thus exposing the brain cavity. The sutures of this brain-case have been mostly, if not altogether, obliterated. The general structure is the same as that of 7. serratus, but there are some minor differences of some importance. The orbitosphenoidal segment (Plate 2, fig. 2, orbs) is not so extensively developed, since nearly the whole of the olfactory lobe lay in front of the orbitosphenoids. The supraoccipital bone (Plate 2, fig. 2, soc) is thicker than in 7. serratus. A part of the parietal is present. Whether or not it was consolidated with the supraoccipital is uncer- tain, but there seems to be an open suture. In front of its articula- tion with the supraoccipital a median sinus (Plate 2, fig. 2, sin) descends and is separated from the brain cavity by bone only 5 mm. thick. This represents probably the hinder of the two sinuses shown in fig. 33 of The Ceratopsia and indicated by the letter X. It seems to be bounded below by the parietal, and by possibly a pa¥t of the frontals. It is doubtful whether or not the orbitosphenoids of this species met above the olfactory lobe. The presphenoid lacks much of being as large as it is in 7’. serratus. The basisphenoid (Plate 1, fig. 2, bas; Plate 2, fig. 2, bas) descends @The Ceratopsia, p. 134, fig. 1138. No. 1660. SKULL OF TRICERATOPS—HAY. 105 a distance of 65 mm. below the surface of the brain, just behind the pituitary fossa. The bone in the region of this fossa has been dam- aged and replaced by white plaster. The fossa probably occupied the white area indicated by pit. f. in Plate 2, fig. 2. Its supposed hinder end is indicated there by a dotted line. The writer has not been able to see the opening for the sixth nerve in this specimen. Possibly it pierced the bone farther forward than in 7. serratus, at a point where the bone is damaged. This specimen furnishes no certain evidence regarding the opening of the temporal fossa into the so-called post-temporal foramen. The upper parts of both temporal fosse have been filled with plaster, in order to strengthen the specimen. On each side there is a passage from the temporal fossa into the cavity, or sinus, into which the so- called post-temporal foramen opens. On one side this opening seems to be partially artificial. On the other side it seems to_be natural, but is possibly the result of accident. Here the opening is about large enough to permit the passage of one’s finger. THE BRAIN-CASE OF IGUANODON, Hulke * has described a brain-case believed to belong to 7guanodon. The same specimen has been redescribed by Dr. C. W. Andrews, of the British Museum of Natural History.” A few remarks will be made on these descriptions. Hulke has designated a part of the axis of the skull as equivalent to the basisphenoid and the presphenoid.* It is evident that the presphenoid is present. The basisphenoid appears to extend forward to the notch above the letters #%. That part of the axis beyond this notch is quite certainly the orbitosphenoid. It includes the optic foramen. Judging from Hulke’s and Andrews’s accounts of this skull the ophthalmic branch of the fifth nerve left the common stem after the latter had passed wholly through the brain-case, and it then ran for- ward in a groove on the outer surface of the bone. In T7vicératops the beginning of the canal that transmits this branch is deeply buried in the bone. What Andrews regards as a foramen for transmitting a branch of the internal carotid artery into the brain cavity the present writer ‘holds to be the exit of the seventh nerve. The nerve descended along the groove described by Andrews. What in Andrews’s figure appears to be a foramen placed 18 mm. above the optic foramen may cor- respond to what in 7'riceratops is thought to be an opening for the 4Quart. Journ. Geol. Soc., XX VII, 1871, p. 199. 5 Ann. Mag. Nat. Hist., 6th ser., XIX, ‘p. 585. ¢ Quart. Journ. Geol. Soc., XX VII, 1871. pl. x1, figs. 1, 2, bps. 106 PROCEEDINGS OF THE NATIONAL MUSEUM. VOL. XXXVI. escape of the fourth nerve, as shown at the numeral 4 in the figures of Plates 1 and 2. Andrews describes three closely placed foramina as being the fenestra ovalis, the exit of the glossopharyngeal nerve, and that of the jugular vein. One of these is certainly the fenestra ovalis, another may be the fenestra rotunda. The glossopharyngeal nerve and jugu- lar vein may have passed out in company with the pneumogastric nerve. Andrews’s statement regarding the forking of the passage by which the pneumogastric nerve leaves the brain cavity agrees with what is found in 7'riceratops. As regards the cast of the brain cavity, the figure of which is fur- nished by Andrews, it seems to the writer that what is indicated as the root of the eighth nerve is really that of the seventh, while that indicated as the root of the ninth is in fact the root of the eighth. From what one sees in 7'riceratops one would expect to find the root of the ninth between the one just referred to and that of the tenth. THE BRAIN-CASE OF MEGALOSAURUS. Von Huene“ has written an interesting account of the brain-case of Megalosaurus. ‘This brain-case is short and high. That part of the lateral wall farthest in front is regarded by V. Huene as the alisphenoid. A notch at its lower end he rightly interprets as the point of exit of the optic nerve. To the present writer it appears that the region in front and above this notch is the orbitosphenoid. That portion of the wall that hes in front of the foramen ovale, rising to the parietal, must be interpreted as the alisphenoid. V. Huene records the presence of a “ meatus auditorius externus.” Now, this meatus is a part of the external ear, and as this reptile probably did not possess such an organ, the foramen in question must have some other function. To the present writer it looks as if this foramen might be the outlet of the depression marked Z in V. Huene’s fig. 1. This may be the summit of an excavation that contained such a process of the brain as has been described on page 103 as occurring in 7'riceratops. What V. Huene calls the jugular foramen is almost certainly the inner and outer passages into the internal ear. On the inside of the brain-case this foramen admitted the auditory nerve; on the outside it was closed by the base of the stapes. The jugular vein doubtless escaped in company with the ninth and tenth nerves. V. Huene fig- ures 4 foramen lying some distance above the one just mentioned and regards it as admitting to the inner ear the branches of the auditory nerve. The structures of the inner ear must le lower down in the wall of the brain-case. Excavation of the bone lower down would certainly expose the semicircular canals. It is probable that the @ Neues Jahrb. Min., etc., 1906, I, p. 1, pl. 1, text figs. 14. NO. 1660. SKULL OF TRICERATOPS—HAY. 107 opening referred to corresponds to what the present writer has inter- preted as the foramen for the fourth nerve. In V. Huene’s fig. 3 there is represented an opening, marked T?, and thought by him to be the outlet for the nerve mentioned. It appears possible that this is only the anterior end of a canal that begins at the foramen assigned by V. Huene to the eighth nerve. The Eustachian canal is mentioned by V. Huene as probably pene- trating the vestibule (Vorraum) of the inner ear. This can not be true. The Eustachian canal opens into the middle ear, that por- tion lying outside of the fenestra ovalis and containing the stapedial rod. Of the three foramina behind the foramen lacerum posterius the one marked by Y//’ probably gave exit to the hypoglossal nerve; the ones marked V//’’ and car probably transmitted veins. It is not probable that the internal carotid artery found its way into the brain cavity at a point so far in the rear. It certainly entered at the pitui- tary fossa. DESCRIPTION OF PLATES. PEATE ale External view of the rear of the skull ef two species of Triceratops. Fig. 1. Triceratops serratus. X 2. Triceratops sulcatus. X wee aie als, alisphenoid, on the crest of the alisphenoidal buttress to the _ post- occipital; bas, basisphenoidal process, broken away in fig. 1; boc. p, basi- occipital process, complete in fig. 1, mostly missing in fig. 2; car, foramen for carotid artery, not seen in fig. 2; fen, fenestra ovalis; oc, occipital condyle, partly broken away in fig. 1, in fig. 2 the suture between the basioccipital and the exoccipital is seen below oc; op. f, foramen supposed to be for ophthalmic artery ; orbs, orbitesphenoid; par. p, paroccipital process, broken away in both specimens; pre, presphenoid, small and not lettered in fig. 2; v, opening for supposed vein; 7, 2, 3, 4, foramina for cephalic nerves of corresponding num- bers; 5, external opening for ophthalmic branch of the 5th nerve; 5° °, foramen . for exit of second and third branches of 5th nerve; 6, 7, 10, 12, foramina for exit of corresponding cephalic nerves. PLATE 2. Longitudinal section of the rear of the skull of two species of Triceratops, showing the brain cavity: Fig. 1. Triceratops serratus. X 2. Triceratops sulcatus. X 2%. a. c. f. anterior condyloid foramen; a/s, alisphenoid; bas, basisphenoid, partly broken away in fig 1; boc, basioccipital; boc. p, basioecipital proc- ess; car, foramen for left carotid artery entering pituitary fossa, not seen in fig. 2; car. g, groove for right carotid artery; ceb. f. foramen? at extremity of a cerebellar process; oc, occipital condyle; op. f, foramen for exit of left ophthal- mic artery from pituitary fossa; orbs, orbitosphenoid; pa, parietal; pit. f, pituitary fossa; pre, presphenoid, not lettered in fig. 2; pro, prootic; sin, base Oe cele 108 PROCEEDINGS OF THE NATIONAL MUSEUM. VOL, XXXVI. of air sinus; soc, supraoccipital; v, opening for supposed vein; «, cavity in basioccipital; 7, 2, 3, 4, 5, 6, 7, 8, 9, 10, 12, exits for nerves of corresponding numbers. IPGATEHVos Brain-casts of two species of Triceratops. Fig. 1. Triceratops serratus. Upper surface. X 3. 2. Triceratops serratus. Wower surface. X 2. 3. Triceratops serratus. Right side. X 3. 4. Triceratops sulcatus. Right side. X 2. car, entrance of carotid artery into pituitary fossa; ceb. p process of cerebellum; cer. h, cerebral hemisphere; med, medulla oblongata; olf. 1, olfac- tory lobe; op. a, base of supposed ophthalmic artery; op. 1, optic lobes; pit, pituitary body, largely restored in fig. 2; v, veins; 2, 3, 4, optic, oculomotor, and trochlear nerves; 5, ophthalmic branch of 5th nerve; 5”*, maxillary and mandibular branches of 5th nerve; 6, 7, 8, abducent, facial, and auditory nerves; 10, pneumogastric nerve, with probably the glossopharyngeal and the spinal accessory; 12, hypoglossal nerve. U. S. NATIONAL MUSEUM PROCEEDINGS, VOL. XXXVI PL. 1 SKULLS OF TRICERATOPS. FOR EXPLANATION OF PLATE SEE PAGE 107. 1 = ae Bi, ne “ U. S. NATIONAL MUSEUM PROCEEDINGS, VOL. XXXVI PL. 2 can go SKULLS OF TRICERATOPS. FOR EXPLANATION OF PLATE SEE PAGE 107. PEs PROCEEDINGS, VOL. XXXVI U. S. NATIONAL MUSEUM "LOL 39vd 33S 3lvid 4O NOILWNW1dx4 4YO4 “SdOLVYSOIN | JO SLSVO NIVYG U31° é vedo ii Gee = "3 ' yid 109) caG 8 0 = a { Be ease ICO id as i t ' ‘ i \ | j THe" iPee ON BRAZILIAN GRASSHOPPERS OF THE SUBFAMILIES PYRGOMORPHINA AND LOCUSTINZD (ACRIDINAD OF AUTHORS). By James A. G. Rreun, Of the Academy of Natural Sciences of Philadelphia. The following study is based upon a series of one hundred and ninety-two specimens of Brazilian Pyrgomorphine and Locustine belonging chiefly to the collection of the United States National Museum. Fifty-three species are treated in the paper, of which sev- enteen are new, four new genera also being described in these pages. Of the series studied by far the greater portion is from the collec- tions made by Mr. H. H. Smith in the State of Matto Grosso and at Rio de Janeiro, while a smaller series, taken by Mr. A. Koebele in northeastern Brazil in the States of Pernambuco and Bahia, is even proportionately more interesting. My acknowledgments are due the authorities of the National Museum for their kindness in permitting me to examine the museum material. My thanks are here given also to Mr. Morgan Hebard, from whose collection was received for study a portion of the series here treated. Subfamily PYRGOMORPHIN. Genus ALGETE Bolivar. 1905. Algete Bottvar, Boletin Real Soc. Esp. Hist. Nat., V, p. 213. Type.—A. brunneri Bolivar. ALGETE BRUNNERI Bolivar. 1905. Algete brunneri Boutvar, Boletin Real Soc. Esp. Hist. Nat., V, p. 214. [Pernambuco, Brazil.] Bonito, Pernambuco, Brazil, January 27, 1883. (A. Koebele.) [U.S.N.M.] One mature male, one immature female. The above listed specimens clearly belong to this very distinct form. One individual is labeled ‘‘on cotton.” PROCEEDINGS U. S, NATIONAL MUSEUM, VOL. XXXVI—No. 1661. 109 110 PROCEEDINGS OF THE NATIONAL MUSEUM. VOL. XXXVI. Genus SPATHALIUM Bolivar. 1884. Spathalium Borivar, Anales Soc. Esp. Hist. Nat., XIII, p. 30. Included six congeneric species, of which the first, S. sommert (Bur- meister), can be considered the type. SPATHALIUM CYANOPTERUM (Blanchard). 1836. Ommexecha cyanopterum BuancHarp, Ann. Soc. Ent. France, V, p. 608, pl. xx, figs. land 2. [Chiquitos, Brazil, now in Bolivia.] Chapada, Matto Grosso, Brazil. April, September, and Novem- ber. (H. H. Smith.) rehres females. These specimens are slightly or considerably smaller than Blanch- ard’s measurement (38 mm. in length), but appear to belong to this species. The variation in size in the individuals in hand is considerable, the extremes in length of the body being 29 and 32 mm., in the length of the pronotum 7.5 and 8, and in the length of tegmen 21.2 and 25. J Genus OMURA Walker. 1870. Omura WALKER, Cat. Spec. Derm. Salt. Coll. Brit. Mus., III, p. 508. Type.—Omura congrua Walker. OMURA CONGRUA Walker. 1870. Omura congrua WALKER, Cat. Spec. Derm. Salt. Coll. Brit. Mus., ITI, p. 504. [Para, Brazil; Amazon Region; Archidona, Ecuador.] Benevides. July [one] (H. H. Smith). One male, one female. There appears to be little doubt that Walker was in error in describ- ing his specimens of this species as males, as the measurement given as well as characters in the generic description, are clearly those of the female. Stal’s Protomachus depressus is possibly distinct from Walker’s spe- cies, the shape of the rostrum and the space between the metas- ternal foveolze as described appearing different from true congrua. The species depressus was described from Peru, Bolivar also record- ing it from the Upper Amazon. In addition to the localities given above, a series of twenty-three specimens of 0. congrua from Bartica, British Guiana, have been examined by the author. Subfamily LOCUSTIN 4+ (ACRIDIN 4 Authors). Genus PROCOLPIA Stal. PROCOLPIA MINOR Giglio-Tos. Chapada, Matto Grosso. March and April. (H.H. Smith.) One male, one female. «This name should replace Protomachus Stal. See Bolivar, Boletin Real Soc. Esp. Hist. Nat., V, p. 215. NO. 1661. ON SOME BRAZILIAN GRASSHOPPERS—REHN. ble These specimens are inseparable from Sapucay, Paraguay, indi- viduals. The range of the species is here extended considerably to the northward, the type-locality, Colonia Risso, near the Rio Apa, northern Paraguay, having been the most northerly previous record. Genus CORYACRIS, @ new. This new genus is closely allied to Catreus Stal, but differs in details best expressed in the diagnosis. Fastigium somewhat produced, horizontal, sublanceolate, not deeply excavate; face regularly sloping without any marked division between line of fastigium and line of face; frontal costa sulcate; eyes prominent; antenne depressed, slightly expanded proximad. Pro- notum with the median carina straight, not appreciably elevated; lateral carinze descending ventro-cephalad on the prozona. Teg- mina elongate lanceolate, surpassing the abdomen. Prosternal spine compressed, slightly bulbous and distinctly retrorse distad; inter- space between the mesosternal lobes distinctly longitudinal, between the metasternal lobes subquadrate to slightly transverse. Supra- anal plate of the male trigonal; cerci styliform; subgenital plate compressed, rostrate. Cephalic and median limbs rather robust, the femora appreciably inflated in the male. Caudal femora with the medio-dorsal portion of the genicular margin provided with a distinct spine, genicular lobes large; caudal tibiz with an apical and nine other spines on the’ external margin, internal margin with ten spines. Type of the genus.—Coryacris dwersipes, new species. CORYACRIS DIVERSIPES, new species. Types.—Male and female; Corumba, Matto Grosso, Brazil. Female in March on highland. (H. H. Smith.) Cat. No. 12081, U.S.N.M. Size large; form rather slender in the male, slightly more robust in the female; surface of the pronotum and pleura impresso-punctate. Head with its dorsal length slightly more than half the dorsal length of the pronotum in the male, two-fifths the length of the same in the female; occiput very slightly arcuate, descending slightly in the inter- ocular region, which is equal in width to about three-fourths that of the fastigium; surface of the fastigium subhorizontal, the lateral margins subarcuate when seen from the sides, when seen from the dorsum the apex of the fastigium is seen to be acute-angulate, the margins slightly bowed in both sexes and slightly blunter and the angle broader in the female than in the male, fastigio-facial junction rounded in the female but with a slight angle in the male, the line of the face distinctly but not greatly retreating in both sexes, more pronounced, however, in the male than in the female; frontal costa rather narrow, uniform, « Kopus, signifying helmet; axpts, signifying locust. Hee PROCEEDINGS OF THE NATIONAL MUSEUM. VOL. XXXVI. subequal in width except for a slight infra-ocellar constriction in the male, sulcate throughout its length; lateral facial carinze marked, slightly diverging ventrad in the male, decidedly arcuate divergent ventrad in the female; eyes quite prominent in the male, much less pronounced in the female, subovate, in length distinctly exceed- ing the infra-ocular portion of the gene in the male, slightly exceed- ing the same in the female; antennz slightly more than twice the length of the dorsum of the pronotum, joints moderately elongate, proximal joints slightly depressed, in consequence giving a faint subensiform appearance to the appendage. Pronotum with the greatest dorsal width contained one and one-half times in the length; cephalic margin of the disk with a broad and very blunt obtuse angulation, caudal margin obtuse-angulate in both sexes, the immediate angle having a distinct, though small, rounded emargination; median carina distinct, subequal in the male, slightly arcuate on the prozona in the female; disk of the metazona flattened, Fic. 1.—CORYACRIS DIVERSIPES. LATERAL VIEW OF MALE TYPE. (X 1}) of the prozona rounding into the lateral lobes, caudal transverse sulcus strongly impressed, the two cephalic transverse sulci marked but not cutting such a pronounced gap in the median carina as the caudal one; prozona and metazona subequal in length, the prozona very slightly longer in the male; lateral angles very distinct, parallel, and horizontal on the metazona, descending on the prozona in a curve to the ventro-cephalic angle, this portion of the lateral angles being much more pronounced in the male than in the female; lateral lobes with their dorsal length very greatly exceeding their depth, ventral margin sinuate obtuse-angulate, ventral angles rounded obtuse. Tegmina about three and a half (female) to four (male) times the length of the dorsum of the pronotum, acute lanceolate, the costal lobe more pronounced in the female than in the male, the distal fourth of the costal margin gently rounding to the acute apex, sutural margin straight with a slight oblique truncation near the apex; intercalary area irregularly reticulate. Wings very slightly shorter than the tegmina when in repose. Prosternal spine distinctly NO. 1661. ON SOME BRAZILIAN GRASSHOPPERS—REHN. 113 compressed, slightly bulbous in the apical half, this being more pro- nounced in the female than in the male, retrorse, apex rather blunt; interspace between the mesosternal lobes strongly iongitudinal in both sexes, that between the mesosternal lobes as broad as long and contracted caudad in the male, slightiy transverse in the female. Supra-anal plate of the male acute-angulate, the margins slightly constricted halfway to the apex, a deeply impressed area like the inverted upper portion of an exclamation mark present proximo- mesad; cerci simple, styliform, very short; subgenital plate moder- ately full at the base, decidedly compressed distad, this portion being somewhat rostrate, strongly keeled dorsad, distinctly so ventrad, and narrowly rounded acute angulate when seen from the side, the sides of the body of the plate provided with a longitudinal cariniform fold extending two-thirds the length of the plate. Ovipositor jaws of the female robust, blunt, the margins of the dorsal pair not serrate. Cephalic and median limbs rather robust, the femora distinctly inflated in the male. Caudal femora about two-thirds the length of the tegmina, rather slender, tapering, carinse unarmed except for a slight serrulation on the dorsal carina of the male, pagina with a moderately regular but poorly impressed pattern, genicular lobes strongly rounded; caudal tibiew slightly shorter than the femora, spines of the internal margin longer than those of the external and both series ten in number. General color in the female dull walnut brown, this shade covering the head, greater part of the pronotum and pleura, while in the male the pleura alone are colored with it. Tegmina bice green in "77 COMYAURS DE both sexes, sprinkled with a number of minute — cvrtine orueap aNd olive spots near the apex. Cephalic and median "*°X°™™ “1 limbs dull bice green in the male, very dull olive-green in the female, the femora in both sexes with rows of whitish spots and a median line of purplish lead color. Caudal femora buff-yellow in both sexes, the internal face with a number of oil green blotches along the carine, very distinct in the male, much less dis- tinct and sub-obsolete in the female, carine of the lateral face with a number of very faint small green points, genicular arches and base of the lobes blackish brown; caudal tibiz with the external face ereenish yellow, marked at the base of each spine with bice green, dorsal face bice green, internal face yellowish with a green blotch at the base of each spine, the distal half of the internal margin lined with scarlet, the internal aspect of the tarsi colored with the same, spines blackish at the apex, the proximal halves of those on the external margin yellowish externally, greenish internally, those on the internal Proc. N. M. vol. xxxvi—09——8 114 PROCEEDINGS OF THE NATIONAL MUSEUM. VOL. XXXVI. margin washed with greenish, the colors on the tibize in the female being similar to those described above from the male but weaker and less sharply contrasted. Head of the male isabella color, the caudal portion of the genz maize yellow; margins of the costa drab, eyes tawny-olive; antenne becoming madder brown distad. Pronotum of the male saffron yellow in the middle of the dorsum, shading laterad to seal brown, this latter being ventrad of the lateral angles, a narrow area along the cephalic portion of the angles, some of the caudal margin of the lobes and a large patch covering the ventro-caudal portion of the lateral lobes maize yellow. This yellowish area is represented in the female by a narrow edging to the caudal margin of the lobes and a broken dull patch along the ventral portion of the same. X Measurements. | Male. | Female. mm | mm Length of bodys. ccs se6cce- sense 40 | 59 Length of pronotum.--.---5.22--.-- 8.5 | 13.5 ength' of tezmeni. See. o-. J ceee sess 37.5 49 Length of caudal femur.........---- 24. 5 | ao. .0 A female labeled Cuyaba, Matto Grosso, February, has also been examined. It is distinctly smaller than the female type with the area between the mesosternal lobes less strongly longitudinal, the caudal margin of the disk of the pronotum without the median emargination and the general color of the tegmina brighter. Genus AZOLACRIS Scudder. 1875. Holacris ScuppER, Proc. Boston Soc. Nat. Hist., XVII, p. 269. Type.—Xitphicera octomaculata Scudder. fEOLACRIS BELLA, new species. Type.—Male; Rio Purus, Brazil. March 15, 1901. (J. B. Steere.) Cat. No. 12082, U.S.N.M. Closely allied to Molacris octomaculata Scudder from Napo or Marafion,” but differing in the larger size, shorter fastigium, the dis- tinct and nonconfluent tegminal spots, the longer spines on the caudal tibiz and the coloration of the antenne. No close relationship exists with Molacris caternaulti Feisthamel from Cayenne, the remain- ing species of the genus. Size large; form very considerably compressed; surface of the pronotum, pleura, venter and abdomen impresso-punctate. Head with the visible dorsal length very slightly more than half that of the pronotum; occiput subhorizontal, the interocular region broad, « Proc. Bost. Soc. Nat. Hist., XII, p. 337. . No. 1661. ON SOME BRAZILIAN GRASSHOPPERS—REHN. 115 being equal to two-thirds the fastigial length; fastigium prominent, slightly ascending, the dorsal length equal to the exposed portion of the occiput, when seen from the dorsum evenly tapering to near the apex which is very bluntly rounded, a distinct medio-longitudinal depression present in the distal half, when seen from the side the angle of the fastigium is acute-angulate with the immediate angle narrowly rounded; face strongly retreating to the ocellus, thence very slightly so; frontal costa continuous, distinct; sulcate through- out its length, greatly compressed at its junction with the fasti- gium, thence very gradually widening to the clypeal suture; lat- eral facial carine very prominent, slightly diverging ventrad; eyes very prominent, large, ovate in outline, the length about equal to that of the infra-ocular portion of the genx; antenne about twice the length of the head and pronotum, joints elongate, strongly depressed proximad and decidedly ensiform, tapering to the slender Fig. 3.—HOLACRIS BELLA. LATERAL VIEW OF TYPE. (X 1) A and moderately acute apex. Pronotum with the greatest dorsal width contained about twice in the length; cephalic margin arcuato- emarginate, caudal margin acute-angulate; median carina repre- sented by a thread-like trace, the dorsum with a slight depression on the cephalic portion of the metazona, lateral angles inflated and covered with numerous blunt spiniform tubercles, the tubercles being present on the entire length of the angles but the distinctly inflated characcer is limited to the metazona; lateral lobes with the greatest length distinctly exceeding the depth, ventral angles obtuse, ventral margin broadly rounded obtuse-angulate, the prozonal por- tion of the lobes with a diagonal blunt ridge extending ventro-cephalad to the ventro-cephalic angle, ending in a small blunt tubercle. Teg- mina exceeding the apex of the subgenital plate by about the length of the pronotum, elongate, subequal, the greatest width contained about five and one-half times in the length; costal margin distinctly 116 PROCEEDINGS OF THE NATIONAL MUSEUM. VOL. XXXVI. arcuate proximad and distad, the median portion very slightly arcuate, sutural margin straight, apex obliquely truncate with the angles narrowly (costal) or broadly (sutural) rounded; anal area very narrow. Wings large, the greatest width contained about one and four-fifths in the length; costal margin straight except for the distal third which is considerably arcuate, apex slightly acute-angu- late with the immediate apex rounded, margin of the axillary lobe produced and rounded; posterior axillary area inflated, fenestrate, the cross veins regular, oblique proximad, becoming straighter distad; greater portion of the median area sub-fenestrate, the ‘nanes’’? subquadrate, translucent and not hyaline as in the axil- lary area. Prosternal spine erect, slender, acute, the immediate apex rather blunt. Mesosternal lobes separated by a slightly longi- tudinal interspace which is somewhat narrower than one of the lobes; interspace between the metasternal lobes shallow, transverse, slightly broader than that between the mesosternal lobes. Abdomen compressed, the dorsum somewhat carinate; supra-anal plate acute trigonal with a deep, high margined medio-longitudinal impression; cerci short, simple, styliform; subgenital plate com- pressed, greatly produced, the apex being developed into an elongate spiniform process, medio-ventral section of the plate with an irregular carina. Cephalic and median limbs rather slender. Caudal femora about three-fifths the length of the tegmina, quite slender, tapering, genicu- lar region rather large, genicular lobes rounded, promi- Fie. 4.£o- nent, pagina with a rather irregular ‘‘herring-bone”’ Lackis 3E- pattern, carina, particularly the dorsal one, weakly ser- ourue or rato-dentate; caudal tibixe almost equaling the femora in year ANY length, rather robust proximad and supplied with two or tyrz. compressed, lamellate, apically spiniform processes, ae lateral margin with seven short spines, one of which is apical, internal margin with the same number, the spines being very long, as long as the lamellate processes, proximad, slightly curved, decreasing in length distad, the terminal one being very small. General color olive green, washed on the median and costal por- tions of the tegmina with very dull and faint wine purple. Head and pronotum with a gradually expanding medio-longitudinal bar of ochraceous-buff on the dorsum; margins of the fastigium, dorsad and caudad of the eyes and the lateral angles of the disk of the pronotum blackish brown; face ochraceous, the carine lined with brown; a diagonal bar of dull buff extends ventro-cephalad over a portion of the prozonal area of the lateral lobes and the caudal portion of the gene, remainder of the lateral lobes with a broad diagonal subequal bar of olive extending ventro-cephalad, ventro- No. 1661. ON SOME BRAZILIAN GRASSHOPPERS—REHN. Lay caudal section dull buff; eyes burnt umber; antenne blackish, tipped with naples yellow. Pleura very dull buff, margined dorsad by a diagonal bar of blackish, the area immediately cephalad of the inser- tion of the median limbs of the same color. Tegmina with the anal area chromium green, the series of ocelliform spots found in this genus being in this species four in number, placed near the princi- pal veins—numbering proximo-distad the first is subcircular and entirely on the costal side, the second is of similar form and slightly on the veins, the third is roughly ovate and with the greater por- tion on the discoidal field, the fourth is elongate elliptical or sub- linear and entirely discoidal, only touching the veins, the color of the spots is deep chrome, edged with blackish brown; anal vein very dark brown. Wings with disk cadmium yellow, becoming cad- mium orange toward the periphery, fenestrate axillary area clear hya- line; marginal color blackish brown, the width of the band being about a third the length of the wing at the apex and gradually narrowing prox- imad to the merest edging. Abdo- men and venter wood brown, the former becoming cinnamon-rufous toward the apex. Cephalic limbs olivaceous, tarsi dull red brown; median limbs vinaceous' rufous. Fic.5.—owacris BELLA. TEGMEN AND WING Caudal femora very dull bottle oie green, the genicular arches and lobes dull blackish brown; cuadal tibiz very dull liver brown, the spines orange-rufous touched with black. Measurements. he Tere mm. engi OfsHody~ ete tes sass ces wale wees ss 49 Wensth OL PTONOGMM pees n arms ncines se ole 11.5 Greatest dorsal width of pronotum........-. 6 heneth Or tesment..ojacii-scace fesse nen cnt cei2 45.5 Greatest width of tegmen ..........-...-..-- 8.1 Penpthiot-caudalaemure oa cecs sence ems ani 27.6 eee a A paratypic male has also been examined. It is slightly smalle than the type, while the inflation of the lateral angles of the pro- notum is less pronounced. The color is more decidedly green and the yellows are purer with less of orange and buff in them. Genus PRIONOLOPHA Stal. PRIONOLOPHA SERRATA (Linnzus). Surmam. (C.J. Hering.) Four females. Parad, State of Para, Brazil. August. (H.H.Smith.) One male. Corumba, Matto Grosso, Brazil. March (1), April (1). Highland (1). (H.H.Smith.) Two males, two females. 118 PROCEEDINGS OF THE NATIONAL MUSEUM. VOL. XXXVI. Genus TROPINOTUS Serville. TROPINOTUS ANGULATUS Stal. | 1873. T{ropinotus] angulatus Sr&t, Oly. Kong. Vet.-Akad. Férh., 1878, no. 4, p- 53. [Bahia.] ; Bonito, Pernambuco, Brazil. February, 1880. (A. Koebele; collected on cotton.) One female. Chapada, Matto Grosso, Brazil. April (2). (H. H. Smith.) One male, two females. Corumba, Matto Grosso, Brazil. March, highland. One female. These specimens vary considerably in the length of the tegmina and wings, but all agree in the distinct angles to the lateral carine of the pronotum. The maculations of the tegmina are also variable in intensity and correspondingly in the presence or absence of those weakly indicated in the more varied form. In one ‘specimen the proximal band alone is indicated distinctly and that by two macu- lations, a larger one in the discoidal area and a smaller one in the projection of the costal area. The distal bands are very faintly indicated in this specimen and chiefly by infuscations along the sutural margin. TROPINOTUS ATTENUATUS, new species. Types.—Male and female. Chapada, Matto Grosso, Brazil. Sep- tember. (H. H. Smith.) Cat. No. 12083, U.S.N.M. Closely allied to 7. gracilis Bruner with a topotypic male of which the new species has been compared,* and from which it differs in the slightly greater size, the more distinctly sulcate frontal costa, the shorter and blunter subgenital plate and in the lesser number of spines (twelve to fourteen) on the margins of the caudal tibie. Size medium; form elongate, very slender. Head with its dorsal length contained slightly more (male) or less (female) than three times in the dorsal length of the pronotum; occiput very slightly arcuate, fastigium horizontal, a distinct median carina present on both; angles of the fastigium slightly acute, the margin slightly elevated, interspace between the eyes very slightly less than the createst fastigial width; fastigio-facial angle moderately rounded, the face considerably retreating; frontal costa narrowed dorsad, sub- equal from between the antenn to the clypeal suture in the male, very slightly expanding in the female, slightly but appreciably con- stricted at the ocellus, suleate from between the antenne to ventrad of the ocellus; lateral facial carine prominent, considerably diverging ventrad; eyes ovate, not very prominent, in length somewhat ex- ceeding that of the infra-ocular portion of the gene; antenne about equal in length to the head and pronotum, depressed, sub-ensiform. Pronotum with the greatest width contained, more than twice in the a Sao Paulo, Brazil, Sept. 14, 1900 (Hempel). NO. 1661. ON SOME BRAZILIAN GRASSHOPPERS—REHN. 119 length; cephalic margin of the disk ebtuse-angulate with the sides of the angles slightly arcuato-emarginate, caudal margin acute- angulate, very sharp in the male ; median carina moderately ele- vated, gently arcuate cephalo-caudad in the female, in the male with with the highest point at the middle of the metazona from which the FIG. 6.—TROPINOTUS ATTENUATUS. LATERAL VIEW OF FEMALE TYPE. (xX 13) carina very gradually slopes cephalad and caudad, the three trans- verse sulci hardly breaking the line of the carina, metazona half again as long as the prozona; lateral angles very distinct, regularly but not greatly diverging caudad; lateral lobes with the dorsal length slightly greater than the depth, ventro-cephalic angle obtuse, ventro-caudal angle rounded rectangulate. Tegmina exceeding the apices of the abdomen and of the caudal femora, about two and two- thirds times the length of the dorsum of the pronotum, -rather narrow; costal lobe distinct, well rounded, remain- der of the costal margin straight except toward the apex, where it is considerably arcuate, sutural margin straight, the apex narrowly rounded. Prosternal spine strongly compressed, decidedly retrorse at the apex, the tip slightly blunt; interspace between the mesosternal lobes narrow, decidedly Jongitudinal ; interspace between the metasternal lobes small, slightly longitudinal in the male, subquadrate in the female. Subgenital plate of the male acute, com- pressed, with a slight ventral carina. Cephalic and median limbs slender. Caudal femora one and three-fourths Fic. 7—Tro- (female) to one and four-fifths (male) times the length of teeew. the dorsum of the pronotum, rather slender, dorsal and Sees ventral carina serrulate, pagina with the pattern moder- nrapanp 2tely regular, genicular lobes acute; caudal tibix distinctly rronotum. but not greatly shorter than the femora, armed on the ex- Ee ternal margin with thirteen to fifteen spines, one of which is apical, on the internal margin with twelve to thirteen spines, those of the internal margin being longer than those on the external margin and slightly curved. General color ochraceous in the male, clay color in the female. Male with the dorsum of the pronotum and anal and discoidal fields 120 PROCEEDINGS OF THE NATIONAL MUSEUM. YOu. XXXVI. of the tegmina chestnut, blackish on the sides of the disk and along the principal group of tegminal veins; costal field of the tegmina pale oil green edged along the principal veins with a narrow line of buff. Female with the median portion of the disk of the pronotum cinna- mon, sides of the same broccoli brown, the dividing line an irregular one of bistre; anal and costal areas of the tegmina, aside from a pale buff humeral line, prout’s-brown, the discoidal field pale cinnamon, clouded along the costal portion with prout’s brown blotchings. A faint medio-longitudinal head stripe and equally faint postocular bars are present in both sexes; antenne vandyke brown, narrowly edged with buff proximad, the distal half entirely russet in the male; eyes prout’s brown in the male, tawny-olive in the female. Lateral angles of the pronotum edged ventrad with bistre (female) or vandyke brown (male), a faint longitudinal line of the same color crossing the lateral lobes slightly dorsad of the middle; crest edged with naples yellow. Tegmina with the distal two-thirds of the sutural margin weakly blotched with prout’s brown, the discoidal and median veins with fine buffy dashes along their proximal portions. Venter pale ochraceous, the abdomen the same in the female, wood brown in the male with the segments edged and lined with blackish. Cephalic and median limbs of the general color somewhat lined and mottled with darker. Caudal femora of the general color suffused along the dorso- median and dorso-lateral carine and along the center of the pagina with prout’s brown, genicular lobes the same; caudal tibie raw sienna in the male, dark clay color in the female, spines yellowish tipped with black. Measurements. Male. | Female. eneth of Dodywsaeacer sts ee ee eee 33.4 40. Henri Of pronouns ose...) sean eee | 10.5 12:5 enethiot tesmens so ees ee | 30.5 36.5 Length of caudal femur. ....-..-...------ | 19.5 22.°7 [ie a nh EE A paratypic male taken in June and two paratypic females taken in July and September have been examined in addition to the types. In size these specimens agree very well with the type pair, but in - color the tendency found in this genus to vary along certain lines is well exhibited. The July female has the lateral portions of the disk of the pronotum and costal area of the tegmina green, while the remainder of the dorsum is quite light. The paratypic Septem- ber female has the green much the same but darker, while the rich ochraceous discoidal field is contrasted with the very dark brown anal field and humeral region. The additional male is colored much as the type female, but has the costal area green. no. 1661. . ON SOME BRAZILIAN GRASSHOPPERS—REEN. 121 Genus COLPOLOPHA Stal. COLPOLOPHA OBSOLETA (Serville). 1831. Tropinotus obsoletus ServittE, Ann. Sci. Nat., XXII, p. 274. [‘‘Cap de Bonne-Espérance.’’] Santarem, State of Para, Brazil. (H. H. Smith.) One male. This specimen has been compared with an authentic female specimen from Cayenne received from the late Doctor Saussure. Genus HELIONOTUS,#¢ new. Allied to Draconata Pictet and Saussure’ from Colombia, but differing in the prozona being shorter than the metazona, the an- tenn being twenty-one jointed, the tegmina attingent dorsad, the subgenital plate of the male acute, the caudal femora distinctly spinose on the dorsal carina. It is immediately separated from Colpolopha Stal by the strongly spinose lateral margins of the pro- notum. Fastigium with the dorsum horizontal, not sulcate; frontal costa precurrent, constricted ventrad of the ocellus where the line of the face is slightly angulate; antenne twenty-one jointed. Pronotum with the median carina cristate, ascending to the middle of the metazona, caudal section of the same acute spiniform; transverse sulci three in number, all deeply severing the median carina; lateral angles strongly tuberculate (female) or acute spiniform (male), pro- jecting laterad. Tegmina very short, attingent (male) or overlap- ping (female), costal lobe very greatly developed; apex narrow, sinuato-truncate. Ovipositor jaws of female very blunt. Sub- genital plate compressed, the tip rostrate. Caudal limbs with the angles and dorsal surfaces serrato-dentate; caudal tibiz somewhat sinuate, external margin armed with nine or ten spines including the apical spine, internal margin with nine spines. Type.—Helionotus mirabilis, new species. HELIONOTUS MIRABILIS, new species. Types.—Male andfemale. Bonito, Pernambuco, Brazil. January, 1883. (A. Koebele.) Cat. No. 12084, U.S.N.M. Size medium (male) to large (female); form slightly compressed, moderately slender in the male, obese and fusiform in the female; surface of the thorax, except the venter, impresso-punctate, head and abdomen much smoother. Head with the dorsum a third (male) or less than a third (female) the length of the pronotum; occiput slightly arcuate, a very faint medio-longitudinal carina present; fastig- ium horizontal. blunt lanceolate in the male when seen from the dorsum, subtrigonal in the female, lateral margins slightly arcuate «“HAzos, signifying sun; voros, signifying back. b Mitth. Schweiz. Ent. Ges., VII., p. 341. 122 PROCEEDINGS OF THE NATIONAL MUSEUM. VOL. XXXVI. when seen from the side, the disk of the fastigium slightly depressed near the apex; interspace between the eyes equal to the greatest width of the fastigium; fastigio-facial angle rather narrowly rounded, the line considerably (male) or very slightly (female) retreating, a slight angle between the line of the face and the line of the fastigium; frontal costa distinct, narrow and subequal from the apical con- striction to immediately ventrad of the ocellus, where there is a decided constriction, thence the margins diverge slightly (male) or considerably (female) to the clypeal suture, the costa distinctly but shallowly suleate throughout its length; sup- plementary facial carine distinct, following in their direction, but in a more pronounced manner, the trend of the costal margins; eyes decidedly (male) or moderately (female) prominent, ovate in out- line, in length slightly longer than (male) or dis- tinctly shorter than (female) the infra-ocular por- tion of the gene; antenne slightly exceeding the length of the head and pronotum in the male, somewhat depressed proximad. Pronotum with the disk flattened, rising mesad to the crest, the greatest width of the dorsum contained about Fig. 8. HELIONOTUS MI- 5 4 J napitis. Dorsan view one and’ one-half times in the length; cephalic OF MALE TYPE. (X2) margin obtuse-angulate in the male, with an almost imperceptible angle in the female, caudal margin. pro- duced acute-angulate in both sexes, the apical spines very sharp; median carina elevated into a serrate crest, which gradually increases in elevation to the middle of the meta- zona, thence sharply descends, the crest being severed by the three transverse sulci, the incisions being marked in both sexes, but extending half- | way to the level of the disk with the interme- diate portions acute dentiforminthe male, Fic. 9. HELIGNOTUS MIRABILIS. LATERAL VIEW OF MALE TYPE. the metazonal por- we tion armed on the descending slope with long dentiform spines similar to those seen in species of Colpolopha; lateral angles pro- jecting laterad and regularly armed with numerous blunt denticles in the female and with a series of very prominent elongate spini- NO. 1661. ON SOME BRAZILIAN GRASSHOPPERS—REUHN. a5 form processes in the male; the dorsum of the pronotum is distinctly diamond-shaped, regularly narrowing cephalad and caudad, but more sharply so caudad; surface of the dorsum with a number of rather regularly placed raised points; lateral lobes with their dor- sal length distinctly greater than their depth. Tegmina about two-thirds the length of the dorsum of the pronotum, coriaceous, the principal veins alone marked; costal lobe greatly developed, the costal field at the middle of the tegmen being wider than the remainder of the same, costal margin concavely emarginate from the greatly developed but rounded lobe to the apex; costal angle of the apex acute; sutural margin slightly arcuate; apex narrow, obliquely and very decidedly sinuato-truncate. Wings reaching very nearly to the apex of the tegmina. Prosternal spine conic, erect, acute, interspace between the mesosternal lobes subquad- rate in the male, slightly transverse in the female; interspace between the metasternal lobes transverse and slightly broader than the mesosternal in- terspace in both sexes. Abdomen slightly compressed in both sexes, the dorsum distinctly keeled and _ pro- vided with spini- form points which are distinct in the male, almost obso- lete on the exposed segments in the fe- F1G. 10. HELIONOTUS MIRABILIS. LATERAL VIEW OF FEMALE TYPE. (X14) male; supra-anal plate of the male trigonal, cerci simple, acute styli- form, subgenital plate acute rostrate, distinctly carinate ventrad ; ovi- positor jaws of the female very blunt, margins hardly toothed. Cephalic and median femora somewhat inflated in the male, slenderer in the female. Caudal femora about as long as from the apex of the fastigium to the apex of the tegmina, robust proximad, slender distad, pagina with the pattern somewhat irregular and subimbricate, medio- dorsal carina acute serrato-dentate, the serrations numerous and of two grades, ventral carine similar but with the serrations less numer- ous, of equal size in the male, smaller in the female, lateral carinee with blunter and shorter denticles, genicular lobes acute; caudal tibiz slightly shorter than the femora, sinuate, spines of the internal margin slightly longer than those of the external. General colors of the male broccoli and hair brown; face cinnamon, the antennez isabelline, tips of the processes of the crest and lateral angles of the pronotum cinnamon, which color is also present as a 124 PROCEEDINGS OF THE NATIONAL MUSEUM. VOL. XXXVI. blotch along the ventral margin of the lateral lobes and obliquely across the middle of the same; eyes burnt umber; tegmina with a blackish brown spot, subovate in the male, subcircular in the female, on the ventral surface of the costal field, showing through on the dorsum, median section of the male tegmen with a longitudinal bar of cinnamon; limbs dull ochraceous with a trace of olivaceous, spotted with blackish; caudal tibie dirty olive-green, the spines narrowly tipped with black. General color of the female walnut brown, the abdomen burnt umber clouded with seal brown, the caudal limbs with russet the general color, while the tegmina are distinctly liver brown; points of the pronotal spines yellowish, the tegmina with the costal spot as apparent as in the male. Measurements. | Male. |Female. | te a -|- ee eee ee | mm. mm Length ot bod yn mecensecass aes anise | 25.5 39 enecihiohpronomiy ss eee e ee cee et i aORS 15 Greatest width of dorsum of pronotum...- - Gua|) eee Osc Length oftepmen a= --c:ccss--s522-5-2-- = 6.8 11 Leneth of. caudal femur=—- 223.2 oss-me - 2 | 16.3 23. 5 A paratypic pair have been examined in addition to the types, the data on both being to the effect that they were collected ‘‘on cotton.” They agree fully with the types in structure, differing only in the general shade of color. Genus EL4ZZOCHLORA Stal. ELAZOCHLORA TRILINEATA (Serville). 1831. Xiphicera trilineata SERVILLE, Ann. Sci. Nat., XXII, p. 272. [Brazil.] Rio de Janeiro, Brazil. December. (H.H. Smith.) One male. The general color of this specimen is very deep green, the median stripe on the head and pronotum also being more ochraceous than the buffy margins of the tegmina. ELAOCHLORA HUMILIS, new species. Type.-—Male. Chapada, Matto Grosso, Brazil. (H. H. Smith.) Cat. No. 12085, U.S.N.M. Allied to Elzochlora viridicata (Serville), but differing in the weaker and lower median carina of the pronotum, the more promi- nent eyes and the narrower tegmina. Size rather large; form moderately slender; surface of the body as usual in this genus. Head with its exposed dorsal length less than half the length of the dorsum of the pronotum; occiput very slightly arcuate; fastigium subhorizontal, a shallow medio-longitu- ° dinal depression present cephalad, form of fastigium acute-angulate, no. 1661. ON SOME BRAZILIAN GRASSHOPPERS—REHN. 125 the apex very narrowly rounded, margins very slightly elevated ; facio- - fastigial angle narrowly rounded when seen from the side, face moderately retreating; frontal costa very gradually widening ven- trad from its junction with the fastigium to immediately ventrad of Fic. 11.—ELOCHLORA HUMILIS. LATERAL VIEW OF TYPE. (X 13) the insertion of the antennz, subequal thence to the clypeal suture except for aslight constriction ventrad of the ocellus, suleate through- out its length; supplementary facial carine prominent, slightly divergent caudad; eyes quite prominent when seen from the dorsum, ovate in outline, in length very slightly exceeding the infra-ocular portion of the gene; antenne slightly longer than the caudal femora, acute, slightly depressed proximad. Pronotum with the disk flattened caudad, slightly tectate cephalad; cephalic margin very slightly angulate, caudal margin slightly acute-angulate, the apex sharp; median carina low, not elevated; lateral angles subparallel in the metazona, the diverted ridge on the lobes of the prozona with fair sized tubercles, greatest dorsal width of the disk slightly more than half the length of the same; transverse sulci three in number, metazona very slightly longer than the prozona; lateral lobes with the ventral margin arcuate, ventro-cephalic angle rounded obtuse, rie. 12 Eye. Ventro-caudal angle obtuse. Tegmina about two and one- ocutora uu- half times the dorsal length of the pronotum, sublance- miuis. Dor- olate, the greatest width of the tegmen contained about SAL OUTLINE e . or Heap ann five and one-half times in the length of the same; costal PRONOTUMOF margin nearly straight mesad, slightly arcuate proximad TYPE. (X 13) 2 i : ; and distad, sutural margin nearly straight, apex obliquely rotundato-truncate. Wings reaching to the apex of the tegmina. Prosternal spine conic, rather blunt, very slightly retrorse; inter- space between the mesosternal lobes subquadrate, the angle of the lobes very broadly rounded; interspace between the metasternal 126 PROCEEDINGS OF THE NATIONAL MUSEUM. VOL. XXXVI. lobes distinctly transverse. Abdomen slightly compressed, supra- anal plate acute trigonal, the apex quite sharp and appreciably decurved, medio-longitudinal sulcus distinct; cerci very small, sim- ple, styliform; subgenital plate compressed, the apex produced, acute rostrate, the ventral line of the produced portion carinate. Cephalic and median limbs of moderate size, the femora slightly inflated. Caudal femora about twice the length of the dorsum of the pronotum, moderately slender, regularly tapering, pattern of the pagina regular; caudal tibiz about equal to the femora in length, almost imperceptibly sinuate, spines on external margin ten in num- ber including the apical one, spines on the internal margin numbering the same. Color patiern similar to that of males of all the species of the genus. Color on the sides of the dorsum of the pronotum, margin- ing the median stripe on the head and covering the median section of the tegmina blackish, the median stripe on the head, pronotum, and anal area of tegmina tawny ochraceous. Costal area of the tegmina and that of the principal veins of the same pale oil green. Lateral lobes of the pronotum and sides of the head olive-green, the edging of the pronotal lobes, the line below the eye, and the proximo-costal streak on the tegmina buff. Eyes burnt umber; antennze walnut brown. Pleura very dark olive-green; abdomen and venter raw umber. Limbs pale oil green; the caudal femora slightly brownish; the caudal tibiz a brighter green, with the spines entirely black. Measurements. mm. henpth OP DOG see 5-see ees ae ee ee 34 ene T a OL Prono uiiMises.-- 3 ee cee eee 10 Bengthjolitesmenti as S222 ee eee 27.5 eng th. or caudal temurie ces ac os ecsneeeee ae 19.8 The type alone has been seen. ELAZOCHLORA PULCHELLA, new species. Type-—Male. Corumbaé, Matto Grosso, Brazil. (H. H. Smith.) Cat. No. 12086, U.S.N.M. Alhed to Elzochlora viridicata (Serville) and EF. humilis Rehn, but differing from both in the tegmina and wings not reaching the apex of the abdomen and in the fastigium being somewhat depressed cau- dad and projecting so that the angle of the face is not straight, but slightly concave. In the weak median carina of the pronotum it agrees with HL. humilis, but the coloration is more like FE. viridicata. Size medium; general form and surface similar to other species of the genus. Head with its exposed dorsal length slightly less than half that of the pronotum; occiput very considerably arcu- ate; the interocular area considerably depressed; fastigium mod- erately acute trigonal, the apex when seen from the dorsum very No. 1661, ON SOME BRAZILIAN GRASSHOPPERS—REIHN. 127 narrowly rounded, the surface distinctly depressed within the mar- gins; when seen from the side, the line of the fastigium rises very considerably from the interocular depression, rounding over the summit of the fastigium into the facial line; angle of the face con- Fig. 13.—ELEHOCHLORA PULCHELLA. LATERAL VIEW TYPE. (xX 2) siderably retreating, slightly concave; frontal costa very slightly expanding caudad, reaching to the clypeal suture, sulcate through- out its length; eyes moderately prominent, ovate in outline, in length very slightly exceeding the infra-ocular portion of the gene; antennz over twice the.length of the pronotum, rather slender, very slightly depressed proximad. Pronotum with its greatest dorsal width contained one and five- eighths times in the length, slightly tectate on the pro- zona, distinctly flattened on the metazona; cephalic margin subtruncate, caudal margin rectangulate; median carina moderately elevated, with an extremely sight longitudinal arcuation; lateral angles rectangu- late on the metazona, descending ventro-cephalad and armed with distinct tubercles on the prozona; transverse sulci three in number, deeply impressed, prozona and metazona subequal in length; lateral lobes very consid- erably longer than deep, the ventral margin obtuse- angulate, the caudal margin oblique, the ventro-caudal angle obtuse; all the pronotal margins except the Me. 14—FLx0- ventral onessupplied withsmall, rounded, more or less dis- cee ears tinct and regularly placed nodes. Tegmina very slightly sau view or longer than the head and pronotum together, reaching Suierodaven, .to the base of ‘the supra-anal plate, acute-lanceolate in (x 2) shape; costal margin arcuate, sutural margin straight. Wings reaching to the tips of the tegmina. Prosternal spine erect, conic, slightly blunted; interspace between the mesosternal lobes very slightly longitudinal; interspace between the metasternal lobes slightly transverse, the cephalic margin of the same obtuse-angulate. 128 PROCEEDINGS OF THE NATIONAL MUSEUM. VOL, XXXVI. Abdomen slightly compressed; supra-anal plate acute trigonal, the apex blunt, the plate tectate in transverse section, the median sec- tion slightly elevated, gradually narrowing caudad; cerci very small, simple, styliform; subgenital plate compressed, produced rostrate, the apex slightly depressed, medio-ventral line carinate and slightly lamellate before the apex. Cephalic and median femora moderately inflated. Caudal femora slightly shorter than the head and tegmina together, moderately slender, tapering, the pagina with a rather regular pattern, genicular lobes with the apex narrowed and rounded; raudal tibiee about equal to the femora in length, armed in the external margin with 11 spines including the Epica one, internal margin with 9 spines. General color pattern that found in most species of the genus. Medio-longitudinal stripe orange-rufous on the head, orange-ochra- ceous on the pronotum, becoming paler caudad and cream buff on the anal area of the tegmina; remainder of disk of pronotum, sides of the dorsum of the head and along the anal vein of the tegmina blackish, the blackish extending over the lateral lobes of the pronotum and there washed with olive-green. Face gallstone yellow; gene washed with olive-green; eyes burnt umber; antennz olive, greenish proximo-ventrad; fastigium margined laterad with blackish. Pro- notum with the deflected lateral angles greenish yellow, the usual pale blotch on the margins of the lateral lobes cream-buff. Tegmina pale oil green, margined on the costal edge with a bar of dull whitish, relieved Se ere by: an adjacent internal splotch of blackish. ‘Abdo men tawny-olive. Femora vinaceous-cinnamon, oil green distad, genicular arches of caudal femora mars brown; ead tibiae and tarsi oil green, the former with the spines on the internal margin blackish, on the external margin pale with blackish tips. Measurements. mm Length ofbody..5..Azzcccus tacee sce oo-2 = sees 28 Length of pronomm:s-eesnes= oso ue «esos ss ssce 9 Length of tegmenicis-s--ss 5... beeen aee es cea oe 16 Length’ of caudal femuriae oo... o=.enees = eae 19 A = — — 4 The type is unique. Genus CALLONOTACRIS,¢ new. A member of the Txenipode and allied to Tzeniopoda, from which it differs in the compressed form, in the strongly elevated and arcuate pronotal crest, and the slightly fossulate fastigium. Form compressed; head very much deeper than broad; fastigium strongly declivent and very slightly fossulate; frontal costa continuous with the compressed frontal costa. Pronotum cristate, strongly andes: peatgiae foe Yy; V@TOS, sieuirpide ee aKp1s, “menityine grashoppes no. 1661. ON SOME BRAZILIAN GRASSHOPPERS—REHN. 129 arched cephalo-caudad, cephalic angle of the disk considerably produced, caudal angle very greatly produced. Tegmina ample in the male, rather short in the female; costal lobe well developed. Wings well developed; axillary area slightly inflated in the male. Caudal tibiz with the spines of the two margins subequal in length; apical spine present on the external margin. The superficial resemblance of this genus to the Gdipodine genera Tropidolophus and Pyrgodera is considerable. Type.—Callonotacris lophophora, new species. CALLONOTACRIS LOPHOPHORA,:« new species. Types.—Male and female. Chapada, Matto Grosso, Brazil. Janu- ary (female), April (male). (H. H. Smith.) Cat. No. 12087, U.S.N.M. Size moderately large; surface of the head smooth, of the pronotum and pleura rugulose-punctate. Head with the exposed dorsal length Fig. 15.—CALLONOTACRIS LOPHOPHORA. LATERAL VIEW OF MALE TYPE. (13) equal to half (female) or two-thirds (male) the width across the eyes; occiput very slightly rounded, dipping sharply into the fossulate disk of the fastigium in the male, less sharply so in the female; disk of the fastigium moderate fossulate in the male, slightly so in the female, a very slightly raised margin surrounding the lateral and caudal borders of this area, in addition to which there is also a trace of a medio-longitudinal line of a similar structure, this latter being dis- tinctly marked in the male and hardly appreciable in the female; cephalic portion of the fastigium with a slight medio-longitudinal sulcus, cephalic margins of the fastigium sharply compressed and passing ventrad as the lateral margins of the frontal costa; lateral foveole prominent, set in a subtrigonal area; fastigio-facial angle aAogos, signifying crest; Popa, signifying carrying. 9 Proc. N. M. vol. xxxvi—09 130 PROCEEDINGS OF THE NATIONAL MUSEUM. VOL. XXXVI. rounded obtuse-angulate, face hardly retreating, subvertical; frontal costa very narrow, subequal (female) or very slightly expanding (male) dorsad of the ocellus, slightly constricted ventrad of the ocellus, the margins diverging thence and becoming fainter until hardly appreciable at the clypeal suture, sulcate except in the ventral fourth; eyes moderately (male) or hardly (female) prominent, elongate subreniform-ovate in outline, slightly longer than (male) or equal (female) to the infra-ocular portion of the gene; antennze about equal to the caudal femora in length, simple, coarse, sub- moniliform distad. Pronotum with its dorsal length about four- fifths that of the caudal femora, the median carina elevated, cristate, arcuate, produced cephalad and caudad, the height of the crest from the dorsum being slightly more (male) or slightly less (female) than a fourth the length of the same, the margin of the crest entire; cephalic Fig. 16.—CALLONOTACRIS LOPHOPHORA. J.ATERAL VIEW OF FEMALE TYPE. (xX 14) margin produced rectangulate, caudal margin produced into a lanceolate process in length but little less than equal to the remainder of the pronotum; disk not at all flattened, sloping ventrad into the vertical lateral lobes, lateral angles but slightly developed on the metazona and these rounded; transverse sulci three in number, the principal one the deepest, none intersecting or even ascending the sides of the crest; lateral lobes with the dorsal length slightly greater than the depth, cephalic margin arcuate emarginate around the side of the head, ventral margin obtuse-angulate, caudal margin oblique, the ventro-caudal angle rounded obtuse-angulate. Tegmina about twice the length of the pronotum in the male, one and one-third times the length of the same in the female, broad in both sexes, the greatest width contained four and one-half (male) or two and one-third no. 1661. ON SOME BRAZILIAN GRASSHOPPERS—REHN. 131 (female) times in the length of the tegmen; costal lobe well developed, very large and more apparent in the female than in the male, sutural margin nearly straight, apex rotundato-truncate. Wings reaching to the tips of the tegmina, quite broad in the male, the axillary field with the margin arcuato-lobate, apex of the wing. rotundato-truncate. Prosternal spine compressed, slightly retrorse, the apex caudad; interspace between the mesosternal lobes subquadrate in both sexes; interspace between the metasternal lobes transverse. Abdomen moderately compressed; supra-anal plate of the male acute-trigonal, tectate, the median sulcus narrow and not well marked; cerci of the male small, simple, styliform; subgenital plate compressed, slightly rostrate, the apex very acute when seen from the dorsum, moder- ately acute when seen from the side, ventral face of apex with a slight ridge; dorsal ovipositor jaws of the female Fig. 17.— CAL- with the edges unarmed. Cephalic and median limbs 4° TAcRI LOPHOPHORA. rather slender, moderately elongate. Caudalfemorareach- — porsat ovr. ing to (female) or considerably exceeding (male) the apex N® OF HEAD AND PRONO- of the abdomen, moderately compressed, moderately robust rus or MALE but tapering in the distal four-fifths and with a very TY?®. (« 14) slight dorsal concavity in the female, margins well elevated, pattern Fig. 18.—CaLLo- NOTACRIS LOPHO- PHORA. DORSAL OUTLINE OF HEAD AND PRONOTUM OF FEMALE TYPE. (x 12) irregular; caudal tibiz equal to the femora in length, rather robust, very slightly sinuate, spines robust, ten (female) to eleven (male) in number on the external margin, eight in number on the internal. General color pitchy brown, lined along the crest with orange-ochraceous and marked on the interocular region and fastigium with the same color, bars of which cover the region along the supplementary facial carine, reaching to the base of the labrum. Tegmina with obliquely transverse series of blotches of cream-buff, breaking up toward the margins and toward the apex. Caudal femora with three incomplete annuli of the same color; caudal tibiz orange-ochraceous, blackish distad and proximad, spines blackish at the bases and at the tips. Cephalic and median limbs blackish, with some obscure markings of dull cream-buff. Wings geranium red, margined with blackish, that of the anterior field broader than that of the remainder, disk with numbers of quadrate and oblong blotches of brownish black. Abdomen with the sides of the dorsal segments spotted along the edge with ochraceous- rufous, a large blotch of the same color at the dorsal extremity of each ventral abdominal segment. Antenne tipped with buffy; eyes chestnut. 132 PROCEEDINGS OF THE NATIONAL MUSEUM. VOL. XXXVI. Measurements. Male. | Female. mm. mm. Lengthiof bode 2 25-25-20 ee seis 32 | 44 Lensth of pronoun. osene se see eee cess 13. 5 18.2 Greatest dorsal width of pronotum -..... 5.6 | 7.8 Length, ofitepmen=.- oe seee eee eee eae 28. 3 25.5 Length of caudal femur_.--...-.....-..... 18 | 23.5 2 il nets oE A paratypic pair have been examined in addition to the types. The only difference noticed is that the light markings on the tegmina and abdomen are more numerous and much more buffy in color. Genus ZONIOPODA Stal. ZONIOPODA MIMICULA, new species. Type.—Male. Chapada, Matto Grosso, Brazil. November. (H. H. Smith.) Cat. No. 12088. U.S.N.M. Closely allied to Zoniopoda theringi; Pictet and Saussure, but differ- ing in the smaller size, the narrower, subequal and less deflected Fig. 19.—ZONIOPODA MIMICULA. LATERAL VIEW OF TYPE. (X 2) fastigium, the less angulate caudal margin of the pronotum and weaker median carina of the same. The subgenital plate is also somewhat more produced and the coloration more uniformly green. Size rather small (for the genus); surface of head and thorax (ex- cept venter) punctate. Head with its dorsal length contained one and two-thirds times in that of the pronotum; occiput moderately elevated, arcuate, interocular region and fastigium slightly declivent, the former about two-thirds the greatest width of the fastigium; fastigium rectangulate with the apex truncate, lateral margins very slightly elevated, the surface of the disk ruguloso-punctate; fastigio- facial angle obtuse-angulate, facial line appreciably but not greatly retreating, slightly arcuate; frontal costa rather broad, subequal in width, very slightly narrowed at its junction with the fastigium, separated from the disk of the latter by a slightly elevated carina, No. 1661. ON SOME BRAZILIAN GRASSHOPPERS—REHN. 133 the margins of the costa hardly reaching the clypeal suture, punctate, with a trace of a median carina dorsad, distinctly sulcate for a short distance ventrad of the ocellus; supplementary facial carine moder- ately divergent, convex arcuate; eyes moderately prominent, ovate in outline, in length exceeding the infraocular portion of the gene; antenne about twice as long as the pronotum. Pronotum with its greatest dorsal width contained about one and one-half times in the length; cephalic margin very slightly produced, arcu- ate, caudal margin obtuse-angulate; dorsum very slightly tectate, median carina hardly apparent on the prozona, distinct but very low on the metazona, lateral angles very well rounded, more appar- ent on the metazona than on the prozona; transverse sulci four in number, the cephalic not severing the carina, prozona and meta- zona subequal, dorsum of the metazona very slightly inflated; lat- eral lobes with the dorsal length slightly more than the depth, ventral margin rounded obtuse-angulate, ventro-caudal angle rather broadly rounded. Tegmina exceeding the tips of the caudal femora by nearly the dorsal length of the head, of meduim width, costa] margin with a slight indication of the costal lobe, sutural margin nearly straight, apical margin obliquely rotundato-truncate; principal veins rather heavy- - Wings equaling the tegmina. Prosternal. spine conic’ erect, blunt; interspace between the mesosternal lobes reversed wedge-shaped, narrow cephalad, the caudal width about equal to the depth; interspace between py. 20.—zowto- the metasternal lobes small, subquadrate. Supra- PoP MiMic- ULA. DORSAL anal plate lanceolate, the apex blunt, strongly ouruine or tectate, the median sulcus broad proximad, narrower — #®A4P AND PRo- : ei ; : NOTUM. (X 2) distad; cerci simple, styliform; subgenital plate very strongly produced, compressed, the apex with a narrow and deep V-shaped emargination, the line of the plate nearly straight and hardly elevated distad when seen from the side. Caudal femora reaching about to the apex of the abdomen, rather slender, pagina with a regular, close and obtuse-angulate pattern, genicular lobes slightly acute-angulate; caudal tibiz very slightly shorter than the femora, very slightly sinuate, armed on the external margin with ten to eleven spines, including the apical one, internal margin with eleven. General color uniform pale apple green; eyes buff; antennz clove brown, the proximal segments annulate distad with yellowish; caudal tibiz and tarsi scarlet, the former greenish proximad, spines greenish, tipped with black. 134 PROCEEDINGS OF THE NATIONAL MUSEUM. VOL. XXXVI. Measurements. Length of body 6 mene tly ol pono seas ee ee ere | Ds Length of tegmen 1 Length of caudal femur 4 A paratypic male (October; in campo) has also been examined. It does not differ appreciably from the type. . ZONIOPODA TARSATA (Serville). 1831. Acridium tarsatum SERVILLE, Ann. Sci. Nat., XXII, p. 283. [Brazil.] Rio de Janeiro, Brazil. December. (H. H. Smith.) One male, one female. These specimens are perfectly typical of the species, the cephalic and median femora being blackish with a broad median and a very narrow proximal incomplete annulus of red, the tibiz blackish with a median annulus of orange. Genus TROPIDACRIS Scudder. 1869. Tropidacris ScuppDER, Proc. Boston Soc. Nat. Hist., XII, p. 346. Type, as here designated 7. dux (Drury) [Gryllus (Locusta) dux Drury]. TROPIDACRIS GRANDIS (Thunberg). 1824. Gr[yllus] grandis TaunBerG, Mém. Acad. Imp. Sci. St. Pétersbourg, IX, p. 403. [Locality not stated.] Rio de Janeiro, Brazil. (H. H. Smith.) One male, one female. The uniform green tegmina are quite characteristic of this species. TROPIDACRIS CRISTATA (Linnzus). 1758. [Gryllus (Locusta)] cristatus LINN&US, Syst. Nat., 10th ed., p. 431. [Amer- ica; Arabia; Asia. ] . Para, Brazil. June. (H. H. Smith.) One male, three females. Marajo, State of Paré, Brazil. February, 1901. (J. B. Steere.) One female. Bonito, State of Pernambuco, Brazil. January 4 and February, 1883. (A. Koebele.) Two males, two females. One Para male has the tegmina distinctly pinkish. Information with one of the Bonito individuals is to the effect that it was taken ‘fon cotton.” Genus LEPTYSMA Stal. LEPTYSMA OBSCURA (Thunberg). Chapada, Matto Grosso, Brazil. August, September. (H. H. Smith.) Two males, three females. Bonito, Pernambuco, Brazil. February 18, 1883. (A. Koebele.) One damaged individual, probably a male. no. 1661, ON SOME BRAZILIAN GRASSHOPPERS—REHN. 135 These specimens have the fastigium, as a rule, slightly smaller than 5D ) ) D Paraguayan specimens, while the variation in the form of the same is quite considerable. Genus LEPTYSMINA Giglio-Tos. 1894. Leptysmina Gier10-Tos, Boll. Mus. Zool. Anat. Comp. Torino, IX, No. 184, p. 34. Included Leptysmina pallida and rosea Gigho-Tos, of which the latter is here selected as the type. LEPTYSMINA ROSEA (Giglio-Tos). 1894. Lleptysmina] rosea Giaui0-Fos, Boll. Mus. Zool. Anat. Comp. Torino, IX, No. 184, p. 34. [Buenos Ayres. ] Corumba, Matto Grosso, Brazil. March; lowland. (H. H. Smith.) Five males. These specimens are perfectly typical of this species, which is now known to range from Buenos Ayres to Corumba. In size the individuals in hand are very uniform. Specimens from Carcarafa, Argentina, determined as this species by Bruner, are seen on comparison with the descriptions and Corumba material to be the allied Leptysmina pallida Giglio-Tos. The handiest index to the two species, aside from the greater size of pallida, is the character of the prosternal spine. Genus STENACRIS Walker. The species of Stenacris available for study may be separated by the following key: a. Subgenital plate of the male with either an erect dorsal median lobe or with the margin supplied with lateral acute processes or both. b. Subgenital plate of the male with a median dorsal lobation. c. Fastigium moderately acute-angulate; eyes separated by an interspace about equal to one-third the greatest width of the fastigium; form as usual in the MENUS! Foes tater tae, cose oslo Nee ew on cera SU De sagas Ley Cyl aTodes: Rae nial very strongly acute- cee eyes seeped by an interspace about equal to one-fourth the greatest width of the fastigium; form very BleNGeE occ2 2 eG HACeae en he ee trig Pete aed wht mexicana. b’. Subgenital plage of the ale with a pair ae eae hoe Hikes processes, but no MiCenanrOreet TOMUIG! omic. a= Salo ou su Docs oe Vere et cn can. DUCIpERMis.@ a’. Subgenital plate of the ie produced and Simic without accessory lobes or styles. b. Size small (male 23-25 mm.); fastigium distinctly slender, slightly longer than STAG See SES aie ates SP aera ere ae Me tee 6 EAS 1c GRACIAS. b’. Size aa (males 32-33, eS 40-42 mm.); fostigium broad, at He as broad as eae in the female much broader than long............S. peas « This name wail have to be ned in place of Smee Simei Walker Larner chlorizans of authors). Marschall’s Gryllus vitreipennis (Zoologische Abhandlungen aus den Annalen des Wiener Museums der Naturgeschichte, 1841, p. 214, pl. xvim, fig. 6) described from Georgia is clearly the same insect as that to which Walker’s name is now applied, agreeing in structure, color pattern, and size. In consequence the species must be known as Stenacris vitreipennis (Marschall). 136 PROCEEDINGS OF THE NATIONAL MUSEUM. VOL. XXXVI, STENACRIS CYLINDRODES (Stal). 1860. Opsomala cylindrodes Stat, Kong. Svenska Freg. Eugenies Resa, Ins., p. 325 [Rio Janeiro, Brazil]. Corumba, Matto Grosso, Brazil. March; lowland. (H. H. Smith.) Three males, two females. Bonito, State of Pernambuco, Brazil. January, 1883. (A. Koe- bele.) One male. The Corumba specimens are quite uniform in size and color, the pale lateral bar being visible in all, but quite weak in one female. The Bonito individual is distinctly smaller with the vertex blunter than is the case in the other representatives, but the genitalia are of exactly the same character. This species is now known to range from the State of Pernambuco to that of Matto Grosso and the Chaco of Argentina (Resistencia). STENACRIS GRACILIS (Giglio-Tos). 1897. A[rnilia] gracilis Giau10-Tos, Boll. Mus. Zool. Anat. Comp. Torino, XII, No. 302, p. 30. [San Lorenzo, Jujuy, Argentina. ] Corumba, Matto Grosso, Brazil. March; lowland. (H.H. Smith.) Three males. These specimens are of much the same size, and have a dark dorsal edging to the lateral pale bar. It is of interest to note that the genus Leptysma presents much the same types of male subgenital plate noticed in this genus, while at least one species of that genus has the pale lateral bar supplied with a darker edging as in this species. The form is now known from the type locality, San Bernardino, Paraguay (Bruner) and Corumba, Brazil. STENACRIS COCCINEIPES (Bruner). Corumba, Matto Grosso, Brazil. March; highland. (H. H. Smith.) One male, one female. Genus OXYBLEPTELLA Giglio-Tos. OXYBLEPTELLA PULCHELLA, new species. Types.—Male and female; Chapada, Matto Grosso, Brazil. June. (H. H. Smith.) No. 12089. U.S.N.M. Allied to Oxybleptella sagitta Giglio-Tos from Paraguay and Sao Paulo, Brazil, but more robust and less compressed, with the fastigium rectangulate (male) or obtuse angulate (female). The head is also somewhat inflated in the new form and the eyes are larger and more prominent. no. 1661. ON SOME BRAZILIAN GRASSHOPPERS—REHN. 137 Size small; form slender, subequal in width; surface glabrous, punctulate on the pronotum and pleura. Head with the exposed dorsal surface slightly longer than the dorsum of the pronotum; occiput with an extremely slight arcuation when seen from the side, very slightly elevated dorsad of the line of the pronotum, interocular region descending very slightly to the fastigium, the width of the interspace between the eyes about half that of the fastigium; fastigium slightly acute rectangulate (male) or slightly obtuse rectangulate (female), the apex broadly rounded, the breadth considerably greater than the length, disk of the fastigium slightly depresso-punctate within the margins, the interocular region (female) or interocular region and fastigium (male) with a slight but distinct medio-longitudinal sulcus; fastigio-facial angle acute-angulate, the immediate angle well rounded, facial line very strongly retreating, the angle of the face being even more decided ventrad of the insertion of the antenne; frontal costa broad, considerably narrowed dorsad to meet the fas- tigium, subequal in width from between the antenne to the clypeal Fig. 21.—OXYBLEPTELLA PULCHELLA. LATERAL VIEW OF FEMALE TYPE. (X 3) suture, margins well elevated, surface of costa and face between facial supplementary carine impresso-punctate; supplementary facial carine moderately divergent dorsad, decidedly divergent ventrad, more particularly in the female; eyes rather large, elongate acute-subpyri- form, moderately prominent when seen from the dorsum, length nearly three times that of the infra-ocular portion of the gene; antenne distinctly shorter than the head and pronotum together, very appreciably depressed, very slightly ensiform, apex blunt. Pronotum with its greatest dorsal width contained one and two- thirds (female) to one and three-fourths (male) times in the greatest length, dorsum moderately arcuate in transverse section; cephalic margin arcuate (male) or arcuato-truncate (female), caudal margin with a hardly appreciable angle in the male, very slightly but distinctly obtuse-angulate in the female, lateral angles very much rounded on the prozona, apparent but rounded on the metazona; disk of the 138 PROCEEDINGS OF THE NATIONAL MUSEUM. VOL. XXXVI. pronotum subequal in width, median carina low, rather coarse, marked on all of the pronotum, transverse sulci three in number; metazona equal to (male) or exceeding (female) two-thirds the length of the prozona; lateral lobes distinctly longer than deep, the ventral margin with a very slight obtuse-angulation. Tegmina about one and two-thirds (male) to twice (female) the length of the head and pronotum together, narrow, subequal; costal mar- gin with a hardly appreciable lobe and a moderate distal arcuation, sutural margin nearly straight, apex with a slight oblique truncation, the extreme apex narrowly rounded; distinct intercalary vein present. Wings extending to the tips of the tegmina. Prosternal process transverse, the apical margin arcuate rie. 92. ©marginate, the lateral angles bluntly protuberant, not Oxyster- Tetrorse; interspace between the mesosternal lobes strongly 7H 4 longitudinal, the length about four times the width, the La. Dor- IMterspace hardly more than a third the width of one of ne meet the lobes; metasternal lobes contiguous in both sexes. ueap ann Abdomen hardly compressed; supra-anal plate of the male Pe fehie short, stout pyriform, the greatest width slightly more than mantx. the length, apex bluntly rounded; a pair of blunt tubercles (x 3) placed a short distance proximad of the apex; cerci of the male simple, robust in the proximal third, then strongly bent dorsad and tapering to the apex which is curved inward toward the median line; subgenital plate of the male acute-angu- late when seen from the dorsum, apex well rounded when seen from the side. Cephalic and median limbs very short, the femora slightly more robust in the male. Caudal fe- mora reaching about to the apex of the abdomen in both sexes, somewhat compressed, moderately robust in the proximal half, tapering distad, the genicular extremity quite small; caudal tibize appreciably shorter than the fe- mora, armed on the external margin with seven spines, on the internal margin with ten, both margins with slight 1¢.23—ox- . . “1. . YBLEPTEL- lamellate expansions distad, the tibiz appreciably broad- ,,° pux- ened in the same portion. CHELLA. DORSAL General color gamboge yellow, broad postocular bars — ovruixeor of vandyke brown, slightly darker in the male, extending 84> AND caudad over the dorsum of the lateral lobes and the same Grrnazn. portion of the pleura, represented on the costal half of (x% the tegmina by the prout’s brown infuscation of the region, a nar- row cloud along the dorsum of the pagina of the caudal femora van- dyke brown, slightly chocolate on its ventral edge; face walnut brown, the area thus colored limited by a pair of strongly divergent dark lines which extend ventrad from the insertion of the antennz; eyes No. 1661. ON SOME BRAZILIAN GRASSHOPPERS—REHN. 139 raw umber; antennz pale russet; a very narrow median line on the fastigium, occiput and pronotum seal brown in the male, in the female olive green on the head, liver brown on the pronotum; spines on caudal tibiz with the apical halves blackish. ‘ Measurements. - Male. | Female. | mm. | mm. 20 | 23.8 Length of body ............. | 23.8 Length of pronotum ........ oy | 4 Length of tegmen............ 12.8 16.2 Length of caudal femur...... 9.5 | 12 One paratypic male (June) and two paratypic females (April and May) have been examined in addition to the types. They do not exhibit any noteworthy difference from the typical pair. Genus INUSIA Giglio-Tos. INUSIA BONITENSIS, new species. Type.—Male; Bonito, Pernambuco, Brazil. January, 1883. (A. Koebele.) Cat. No. 12090. U.S.N.M. Allied to J. pallida Bruner from Paraguay“ but differing in the more prominent eyes and the broader, darker lateral bars. Size small; form elongate, slightly compressed; surface of head, pronotum, pleura and to an extent the dorsum of the abdomen punctulate, the pronotum, pleura and face with the impressions closely placed and sharply defined. Head with the exposed dorsal length contained one and one-half times in the length of the prono- tum; occiput decidedly ascending to the sub-horizontal interocular region, which latter is hardly more than a third the greatest width of the fastigium, a slight median carina present on the interocular region and occiput; fastigium acute-lanceolate; fastigio-facial angle sub-rectangulate, the interocular rostrum with its outline curving gently from the dorsal angle to a short dis- tance ventrad of the insertion of the antenne, whence the facial line is regularly and very considerably retreating; frontal costa very narrow, subequal in width, very slightly expanding dorsad of the insertion of the antenne, distinct and sulcate from the fastigio- facial angle to the clypial suture; supplementary facial carine dis- tinct, sub-parallel, slightly convex sinuato-arcuate; eyes quite promi- nent, ovate in outline, the length slightly greater than that of the infra-ocular portion of the gene; antenne lacking. Pronotum with the greatest dorsal width contained about twice in the length of the same; cephalic and caudal margins arcuate obtuse-angulate; median @ Proc. U.S. Nat. Mus., XXX, 1906, p. 660. 140 PROCEEDINGS OF THE NATIONAL MUSEUM. VOL, XXXVI. carina slight and low but quite apparent on the whole dorsum; lateral angles moderately rounded on the prozona, very slightly rounded on the metazona, the disk very gradually widening caudad; transverse sulci three in number, distinct, narrow, metazona about three-fifths of the entire length of the pronotum; latéral lobes very considerably longer than deep, the ventral margin strongly arcuato- Fic. 24.—INUSIA BONITENSIS. LATERAL VIEW OF TYPE. (X 3) emarginate cephalad, caudal section of the ventral margin straight, the ventro-caudal angle very broadly rounded. Tegmina very long and slender, exceeding the tips of the caudal femora by nearly the length of the pronotum, subequal; apex lanceolate. Wings equal to the tegmina. Prosternal spine conic, blunt, slightly retrorse; inter- space between the mesosternal lobes slightly longitudinal, the inter- space but little more than half the width of one of the lobes, the internal margin of the lobes convex arcuate; metasternal lobes contiguous. Abdomen slightly com- L pressed; supra-anal plate lanceolate, the margin notched 4 i distad of the middle for the reception of the cerci, disk % j considerably depressed; cerci robust and tapering in the 4 OF proximal half, thence sharply bent dorsad and narrowed to about a third the proximal width, tapering to the rather blunt apex, the axis of the cercus from the median bend being slightly toward the median line of the body, then curved slightly laterad at the tip; subgenital plate - Fig. 25.—INus- moderately compressed, the dorsal margins acute-angu- 4 gonrrey- late but not reaching to the inflated, sub-bulbous apex, 5%: Dorsat the latter being moderately rounded when seen from eco ee the side. Cephalic and median limbs rather slender. ?®°NoTUMor TYPE. (<3) Caudal femora surpassing the apex of the abdomen, the length being two and one-half times that of the pronotum, somewhat compressed, proximal portion regularly and evenly, but not greatly, inflated, tapering to the genicular region, genicular lobes moderately acute-angulate, pagina regularly and in the proximal section very closely patterned; caudal tibie distinctly shorter than the femora, the distal portion considerably inflated, the margins 7. No. 1661. ON SOME BRAZILIAN GRASSHOPPERS—REHN. 141 lamellate, external one armed with six spines, internal with ten; caudal tarsi with the proximal joint very considerably depressed and expanded. General color wax yellow, quite greenish on the dorsum and tending toward gamboge yellow on the caudal limbs and abdomen; postocular bars vandyke brown, broad and reaching to the apex of the tegmina, covering all but the anal area of the same; eyes russet, clouded with vandyke brown dorso-caudad; caudal tibiz clouded with brownish, -the spines yellowish very narrowly tipped with black. Measurements. Mm MEN SEO DOU Yn se sees nwa ees arc eee 17.5 enpth of pronotuimr. ose ce socs ch eckek eee 3.8 henewh of TeSMenso. Pesce. so-so a eons eae od 16.6 (ueneth oleaudalifemine: tactic. csieccnes ats. 10.6 The type alone has been examined. Genus STENOPOLA Stal. STENOPOLA PUNCTICEPS (Stal). 1860. Opsomala puncticeps Stkt, Kong. Svenska Freg. Eugenies Resa, Ins., p.325. {Rio Janeiro, Brazil.] Corumba, Matto Grosso, Brazil. March;highland. (H.H. Smith.) One female. This specimen is appreciably smaller than a specimen from Sapucay, Paraguay. The range of this species now extends from San Lorenzo, Jujuy, Argentine, east to Rio Janeiro and from Sapucay, Paraguay, to Corumba, Matto Grosso. STENOPOLA BOHLSII Giglio-Tos. Corumba, Matto Grosso, Brazil. March and May; highland. (H. H. Smith.) One male, four females. : The range of this species now extends from Resistencia Chaco, Argentina, to Corumba. Genus PARACORNOPS Giglio-Tos. 1894. Paracornops Giauto-Tos, Boll. Mus. Zool. Anat. Comp. Torino, IX, No. 184, p. 31. Type.—Acrydium longipenne De Geer. PARACORNOPS AQUATICUM Bruner. 1906. Paracornops aquaticum Bruner, Proc. U. 8. Nat. Mus., XXX, p. 663. [San Bernardino and Sapucay, Paraguay. ] Corumba, Matto Grosso, Brazil. March; lowland. (H. H. Smith.) Two males, two females. Cuyabaé, Matto Grosso, Brazil. February. (H. H. Smith.) One male. 142 PROCEEDINGS OF THE NATIONAL MUSEUM. VOL. XXXVI. One of the female individuals is quite large, considerably exceed- ing the other female and the original measurements in size. There appears to be considerable variation in the depths of the coloring of the dark lateral bars, one specimen from Corumbé having them quite faintly indicated. ; Genus MASTUSIA Stal. 1878. Mastusia Svat, Bih. till Kong. Svenska Vet. Akad. Handl., V, No.4, pp. 39 and 84. Ty pe.— Mastusia quadricarinata Stal. MASTUSIA KOEBELEI, new species. Types.—Male and female. Bonito, Pernambuco, Brazil. January 18, 1883 (male); February, 1883 (female). (A. Koebele.) Cat. No. 12091, U.S.N.M. Differing from. Mastusia quadricarinata from Peru in the much smaller size, shorter and narrower tegmina and different coloration. Bolivar’s M. spectabilis* from the Upper Amazon region is not closely related to the new form. Size small; face and surface of the thoracic segments punctate, the pronotum strongly so. Head with its exposed dorsal surface two-thirds (female) to four-fifths (male) the length of the pronotum; occiput slightly ascending very slightly arcuate; interocular space about one-half (male) to three-fifths (female) the greatest width of the fastigium; fastigium broad, very short, horizontal, the angle obtuse-angulate in both sexes, more decidedly so in the female than in the male; surface very slightly depressed within the margins; fastigio-facial angle acute-angulate, the immediate angle well rounded, facial line very decidedly retreating; frontal costa broad, subequal, but very slightly expanded between the antennz, narrowed dorsad of the antenne to joint the fastigium, a slight constriction present at the ocellus, margins distinctly elevated, surface moderately exca- vate and distinctly punctate; supplementary facial carinz promi- nent, regularly and decidedly divergent caudad; eyes moderately prominent in both sexes, acute-ovate in outline, the length about half that of the infra-ocular portion of the gene; antenne slightly shorter than the head and pronotum together (male) or no longer than the pronotum (female), slightly depressed, very slightly nar- rowed toward the base. Pronotum with the greatest caudal width of the disk contained approximately one and one-half times in the length of the same; cephalic margin slightly arcuate (male) or arcuato-truncate (female), caudal margin distinctly emarginate mesad, the sides of the emargination convex, lateral angles very well rounded; median carina distinct, but very low, principal trans- verse sulcus the only one severing the carina, metazona about a @ Anal. Soc. Esp. Hist. Nat., XIX, p. 325. No. 1661. ON SOME BRAZILIAN GRASSHOPPERS—REEN. 143 third the length of the prozona; lateral lobes slightly longer than deep, ventral margin slightly but distinctly arcuate-emarginate cephalad, the caudal section of the margin very bluntly obtuse- angulate, the ventro-caudal angle rounded. Tegmina extending slightly caudad of the margin of the metanotum, sub-lanceolate, the greatest width contained approximately three to four times in the length, apex moderately angulate. Wings very minute, lobiform, covered by the tegmina. Abdomen distinctly compressed, tectate dorsad; supra-anal plate short, transverse, the apical margin rectangu- late, disk considerably excavate; cerci moderately long, with a considerable arcuation dorsad before the middle, the apex blunt, with a sharply bent spine directed ventro-cephalad toward the median line; subgenital plate elevated and produced distad, strongly compressed and sub-lamellate in that por- tion, the apex when seen from the side well rounded. Prosternal process transverse, the apical margin rec- tangulate, the angles slightly spiniform; interspace between the mesosternal lobes distinctly longitudinal, about twice as long as wide, with the interspace about two-thirds the width of one of the lobes (male) or wedge shaped with the narrowest (caudal) portion of the interspace slightly less than half the length of the P%S. 2}. Mastusis same (female); metasternal lobes contiguous in both = wew or remaue sexes. Cephalic and median limbs very short, rather 9?" “*® slenderin thefemale. Caudal femora reaching to (female) or exceed- ing (male) the apex of the abdomen, moderately robust, tapering, genicular lobes bluntly angulate, pagina regularly and rather closely patterned; caudal tibiz distinctly shorter than the femora, slightly dilated distad, the margins distinctly lamellate in that portion, lateral margin with seven to eight spines, internal margin with ten to eleven spines; caudal tarsi with the proximal joint mod- erately depressed and dilated. General color russet, a broad postocular bar ex- tending over the dorsal half of the lateral lobes, Fig. 27.—Mastusta xor- theventral half of the tegmina and dorsal portion BELEI. LATERAL OU T- : LINE‘OF HEAD anp pro. Of the pleura vandyke brown, continued on the NoTuM oF reMaLe Tyre. sides of the abdomen as a series of blotches of the wipe same color in the female, as a bar in the male, becoming weaker distad in both sexes; ventral half of the lateral lobes, gene and pleura wood brown; face with a pair of more (male) or less (female) divergent dark lines ventrad of the insertion of the antenne; occiput with a dark brown median bar in the male; caudal tibiz vandyke brown distad, spines pale yellow, tipped with blackish. 144 PROCEEDINGS OF THE NATIONAL MUSEUM. VOL. XXXVI. Measurements. | Male. | Female. | = = | l | mm. | mm. bengthiatibodiye: = 228. s- \ ventrad and a distance dorsad of the ocellus in the male, the dorsal third biseriate punctate in the same sex; lat- \ yy eral facial carine prominent, divergent; eyes prominent, ! particularly in the male, subovate (female) or short Fra. 39.-Para. OVate (male) in outline, distinctly longer than the infra- scopascuara- ocular portion of the gene in both sexes; antenne fili- ae ew, form, elongate, in length less than half (female) or four- arexor mate fifths (male) that of the body. Pronotum very lightly . \* sellate, distinctly arcuate in transverse section; cephalic margin subarcuate with a very shallow median emargina- tion, caudal margin very slightly arcuate (male) or arcuato-truncate (fe- male), lateral angles rounded, no appreciable shoulders present ;median carina very weak and low, more apparent on the metazona than elsewhere ; transverse sulci three in number, metazona about half the length of the prozona; lateral lobes distinctly longer than deep, ven- No. 1661, ON SOME BRAZILIAN GRASSHOPPERS—REUHN. 161 tral margin arcuate-emarginate cephalad, the median angle rounded obtuse, ventro-caudal angle obtuse-angulate, the caudal margin arcuate-emarginate. Tegmina two-thirds (male) to four-fifths (female) the length of the pronotum, reaching to (male) or very slightly exceeding (female) the cephalic margin of the metanotum, subovate, the greatest width contained one and one-half times in the length, the apex very broadly and bluntly rounded in both sexes, broader, however, in the female. Wings minute. Prosternal proc- ess very low, blunt, slightly more prominent in the male than in the female; interspace between the mesosternal lobes narrow and decid- edly longitudinal in the male, broader and subquadrate in the female; metasternal lobes subcontiguous in the male, in the female separated by a subtrigonal interspace which is decidedly narrowed caudad, the narrowest space separating them being slightly less than half the greatest. Abdomen slightly keeled dorsad; furcula of the male prominent, spiniform, subparallel, as long as the segment from which they arise; supra-anal plate of the male subtrigonal, the margins con- siderably arcuate-emarginate at the base of the cerci and slightly the same before the apex which is blunted acute-angulate, a broad shallow median depression present on the proximal three-fourths, margins considerably reflexed and distinctly elevated mesad, the elevated portion covered with small excrescences; cerci of the male rather slender, elongate, reaching to about the apex of the subgenital plate, sigmoid, elevated, the whole cercus compressed, more dis- tinctly so mesad and distad, proximal third tapering, rather narrow mesad, the distal third, which is parallel to the axis of the body, slightly lamellate, apex moderately acute, the whole cercus regu- Jarly and gently curved toward the middle line to the proximal third which is slightly arcuate divergent; subgenital plate of the male inflated, recurved, the apex very close to the apex of the supra- anal plate and very blunt and low, the outline of the plate when seen from the side is moderately arcuate; ovipositor jaws of the female rather straight, the dorsal subequal proximad and mesad, the distal portion rather sharply hooked. Cephalic and median limbs moder- ately long, the femora very considerably inflated in the male. Caudal femora exceeding the apex of the abdomen in both sexes, of medium build, tapering, the greatest width contained about four times in the length, pattern of the pagina regular, genicular lobes with their apices rectangulate in the female, somewhat arcuate rectangulate in the male; caudal tibiz slightly shorter than the femora, external margin armed with seven spines, internal with ten spines. General color olive-green on the dorsum of the head, pronotum and abdomen, becoming saffron yellow ventrad and on the sides of the abdomen. Head with the face oil green in the male, dark olive- green in the female, the gene much the same but slightly paler in Proc. N. M. vol. xxxvi—09——11 162 PROCEEDINGS OF THE NATIONAL MUSEUM. VOL. XXXVI. both sexes; a broad postocular bar of blackish brown extends from the caudal margin of the eye over the dorsal half of’ the lateral lobes, touching the dorsal portion of the exposed pleura and covering the costal half of the tegmina; eyes russet in the male, mars brown in the female; antenne canary yellow proximad, suffused with dull brown distad in the male, in the female ochraceous, slightly darkened distad. Ventral portion of the lateral lobes very dull olive-yellow in the male, slightly buffy in the female. Sutural half of the tegmina paris green. The two proximal abdominal segments are clouded with blackish in continuation of the dark coloration of the tegmina; dorsum of the two apical abdominal segments, furcula, a broad margin to the supra-anal plate, the margin of the subgenital plate except the apex and the distal half of the cerci blackish brown. Limbs apple green, the caudal femora slightly yellowish proximad and ventrad, the distal extremity blackish, the dorsal face and pregenicular region oil green; caudal tibiz bottle green, blackish proximad and distad, the spines almost entirely black. Measurements. mm. mm hengthiof bod yin eec=-< seem 19 22. 5 Length of pronotum ........-.-..--- 4.3 4.5 Length of tegmem:s.- 2-2. --2-2------ 3.2 3.6 Length of caudal femur..........--- 12.6 13.1 The typical pair are the only specimens of the species in the col- lection. Genus OSMILIA Stal. OSMILIA FLAVO-LINEATA (De Geer). Pernambuco,Brazil. January 4, 1883. (A. Koebele.) One female. OSMILIA VIOLACEA (Thunberg). Rio de Janeiro, Brazil. November. (H. H. Smith.) One male, three females. Chapada, Matto Grosso, Brazil. April. (H. H. Smith.) One male. Genus POLYCHITONACRIS,¢ new name. 1859. Polysarcus SAussURE, Revue et Magasin de Zoologie, 2me ser., XI, p. 392. (Not Polysarcus Fieber, Lotos, III, p. 175, 1853.) Type.—Poli ysareus atavus Saussure. a Fougirar pentane wearing many tunics; aKpts, signifying grasshopper. No. 1661. ON SOME BRAZILIAN GRASSHOPPERS—REHN. 163 POLYCHITONACRIS ATAVA (Saussure). 1859. Plolysarcus] atavus SaussurE, Revue et Magasin de Zoologie, 2me ser., XI, p. 393. [Bahia, Brazil.] Minas Geraés, Brazil. [Hebard collection.] One female. i Rio de Janeiro Brazil. November. (H. H. Smith.) Two females. The Minas Geraés specimen is somewhat smaller than the Rio de Janeiro individuals, and has the second pronotal lobe not quite as distinctly separated from the third or the spine at the apex of the caudal femora as long. The size of all the individuals is greater than the original measurement given by Saussure (‘‘Long., 0.022”), but otherwise they do not appear separable from true atava. The measurements are as follows: mace | Ered Rio de Janeiro. mm. mm. mm. Length of body (including ovipositor jaws) -.| 27 29.8] 34.7 ene th of Pronotwms.\e esses sce sae ee 8 8&1 8 Greatest width of dorsum of pronotum ..... 7.8 8.6 8.3 Length of caudal femur.....:..2......222..-- 17 19 18.2 DESCRIPTIONS OF THREE NEW SPECIES OF CISCO, OR LAKE HERRING (ARGYROSOMUS), FROM THE GREAT LAKES OF AMERICA; WITH A NOTE ON THE SPECIES OF WHITEFISH. By Davin Srarr Jorpan, President, Stanford University, California, and Barton WarrEN EverMANN, Of the U. S. Bureau of Iisheries. In the investigations of the fisheries.of the Great Lakes system by the International Fisheries Commission of Great Britain and the United States during the summer of 1908, three new species of the genus Argyrosomus, known as Cisco or Lake Herring, were obtained. These are described in the present paper. ee iii inn —— Oy : 1 cnet SE SSS i) SR as at <> Fic. 1.—ARGYROSOMUS ERIENSIS. ARGYROSOMUS ERIENSIS Jordan and Evermann, new species. JUMBO HERRING OR ERIE CISCO. Head 42 in the length, measured to base of caudal; depth 32; depth of caudal peduncle 2! in head; eye 51; snout 3}; interorbital space 34; length of maxillary from tip of snout 3; D. 10; A. 12; scales in lateral line 71; between lateral line and origin of dorsal 8; between occiput and dorsal 32. PROCEEDINGS U. S. NATIONAL Museum, VoL. XXXVI—NoO. 1662. tee 166 PROCEEDINGS OF THE NATIONAL MUSEUM. VOL, XXXVI. Body very deep, its width contained 12 times in head; dorsal out- line curved abruptly upward behind occiput; dorsal contour of head straight; snout pointed, though rather blunt at tip; jaws about equal, the lower closing just beneath the upper at tip; maxillary extending to a point beneath anterior edge of pupil, the supplemental part about 8 times as long as wide. Gullrakers on first arch 16+29, very slender, the longest equal in length to diameter of orbit. Scales firmly attached. Dorsal inserted about midway between tip of snout and base of caudal, the highest (first) ray contained 1 times in length of head; height of adipose dorsal equal to 14 times the length of its base; height of anal contained 2 times in length of head; outline of both dorsal and anal slightly concave; origin of ventral below anterior part of dorsal, length of fin contained 14 in head; pectoral 1% in head. Color in spirits silvery, dusky on upper parts, but without blue shades in life; distal portion of dorsal, outer part of caudal, and edge and tip of pectoral dusky; other fins white. Type.—Cat. No. 62515, U.S.N.M., is from Lake Erie at Port Stan- ley, Ontario, measuring 164 inches in length, and was collected by the writers. This represents the maximum of the size of the species as seen by us. Its weight when fresh was 2? pounds. A cotype, 144 inches long, No. 13083, Stanford University collection, obtained at the same time, is a little smaller and slightly darker in color, the anal having a terminal dusky cloud. It has 11 dorsal and 11 anal rays. This species is very abundant along the northern shore of Lake Erie about the first of August. It 1s also occasionally taken in the southern part of Lake Huron, but it seems to be unknown in Lake Superior, and we did not hear of it in Lake Ontario. On the date of our visit to Port Stanley, July 29, 1908, about 1,500 pounds were taken in the gill-nets. The largest of these weighed 2? pounds, and. were about 18 inches in length. The bulk of the catch was, however, about 14 inches in length. It is said of this species that there is a ‘“oreat spurt” or large run in the spring and a short one in the autumn, before the spawning time in November. The species was also seen at Port Burwell, where large numbers of them are smoked, having an excellent flavor as thus prepared. Many others from Point Rondeau, Ontario, were seen in the Detroit market. The species may be called the “ Erie Cisco,” as it is characteristic of that lake, although other species, Argyrosomus artedi, the com- mon Lake Herring, and Argyrosomus huronius, are found in the same lake. Fishermen claim that it is found in middle water, not at the surface nor at the bottom. As a food fish it is far superior to any other lake herring, being as delicate and rich in flavor as No. 1662. NEW LAKE HERRING—JORDAN AND EVERMANN. 167 the best Whitefish, Coregonus albus and Coregonus clupeiformis. It is therefore a species worthy of careful attention from the propa- gators of fish. It is claimed that it is rapidly increasing in abun- dance and that it was virtually unknown until within the past ten years. Most of the fishermen claim never to have seen examples of 2 to 3 pounds until within four or five years. It is locally known as the “Jumbo Herring,” as it reaches a larger size than any other “Lake Herring” except the Tullibee of the northwestern lakes (Argyrosomus tullibee). It is believed by many fishermen that the Jumbo Herring is the product of a cross between the Erie Whitefish (Coregonus albus) and the Lake Herring (Argyrosomus artedi). This belief is with- out foundation. It rests on the fact that at the Put-in-Bay Hatchery attempts have been made to fertilize Whitefish eggs with the milt of the Lake Herring, in default of the milt of its own species. To test this matter, Mr. Frank N. Clark, of the hatchery at Northville, Michigan, undertook the same experiment under carefully prepared conditions. In no case was the egg of a Whitefish fertilized by the milt of the Lake Herring, and the hybridization of the two species is quite imprebable. The following are our field notes on this species: Jumbo Herring, Port Stanley, July 29, 1908. Head 4; depth 53; length 164; scales 7-9-80. D. 2, 9; A. 12. Head much deeper and less pointed than in the Lake Huron herring; lower jaw less projecting, almost even. Maxillary 3 in head. Supplemental maxillary a little narrower than in the other, not twice as long as broad, reaching front of pupil. Eye 34 in head. Suborbital to mucous channel, about 4 as broad as pupil below eye, the whole width of bone not much greater than pupil. Gillrakers about 380 below arch. Color, olive above, sides silvery, a little less bluish than in the other, the stripes on scales much less distinct. Dorsal, pectoral, and caudal edged with black. Ventral without black tips. Caudal peduncle 13 in length from adipose. Greatest depth of tail equal to its length. Cooked, the Jumbo Herring is a fine, rich, well-flavored fish, as good as Whitefish. The Lake Huron herring is rather poor and tasteless, though as good as ordinary artedi. ARGYROSOMUS HURONIUS Jordan and Evermann, new species. LAKE HURON CISCO. Head 43 in length to base of caudal; depth 44; depth of caudal peduncle 2;%, in head; eye 5; snout 33; interorbital space 34; length of maxillary from tip of snout 3; D. 9; A. 11; scales in lateral line 80; between lateral line and origin of dorsal 8; between occiput and dorsal 36. Body notably elongate, elliptical, with slender, pointed head and slender tail, less compressed than in the other species of the genus; head small, the snout long and pointed; lower jaw not closing within the upper, but extending slightly beyond it; maxillary reach- ing a point below center of pupil, its width contained 3 times in the 168 PROCEEDINGS OF THE NATIONAL MUSEUM. you. xxxvt. length. Gillrakers on first arch, 10 + 32; very slender, those near angle equal in length to diameter of eye. Lateral line almost straight. Scales large and rather loosely attached. ‘Dorsal inserted midway between anterior border of eye and base of caudal; height of first ray contained about 12 times in length of head; length of base of adipose dorsal about equal to its height; origin of ventrals below middle of dorsal, the rays slightly shorter than those of dorsal; length of first anal ray, 24 in head; caudal deeply forked ; pectoral short, about 14 in head. Color in life, clear metallic blue above, silvery below; in spirits, silvery; dusky above, light below; a very indistinct, narrow, dusky stripe along each row of scales on upper half of body; dorsal with a broad dusky margin; caudal largely dusky; a mere trace of dark color on paired fins and the anal. Type.—Cat. No. 62516, U.S.N.M., measures 14% inches in length and was taken at Port Stanley, Ontario, by the writers, on July yy b> Se Fic. 2.—ARGYROSOMUS HURONIUS. 29,1908. A cotype, No. 13082, Stanford University collection, meas- uring 17 inches long, has ten rays in the dorsal and a slightly longer pectoral. The species is known locally at Port Stanley as Lake Huron Her- ring, or Blueback. About a dozen were found mixed with about a thousand of Argyrosomus eriensis. The flesh of this species is rather dry and flavorless, something like that of the Menominee Whitefish, Coregonus quadrilateralis, and it is not to be compared as a food fish with the Erie Cisco. The species seems to be quite as common in Lake Huron as is the usual lake herring, Avgyrosomus artedi. It has not been noted from Lake Superior or Lake Ontario, in both of which lakes Argyrosomus artedi is abundant. A number of specimens were obtained from off Mackinac, but none of these was preserved. In our field notes is the following account of one of these: Argyrosomus huronius, a lake herring called Blueback, taken off Mackinac Island, by Leggett, a fisherman at Mackinac, August 3, 1908. No. 1662. NEW LAKE HERRING—JORDAN AND EVERMANN. 169 ee a eee ee Length 13% inches; head 4}; depth 4?; eye 43 in head; maxillary to end of snout 33; maxillary proper 3$; supplemental maxillary with broad hook, and more than twice as long as broad; maxillary barely reaching front of pupil; snout 33; jaws equal at tip; chin projecting; scales, 8-78-8; gillrakers, 14+ 29. Pectoral 14 in head; dorsal 1}; dorsal rays 10; anal 10. Color dark Olive with bluish luster above and sparse dark points on scales; tip of jaws dark; dark spot on preopercle as in the other species; edge of all fins dusky; maxillary rather dark; caudal slender, least depth less than half, its greatest depth a little more than half length. This seems to be the same as the Lake Huron herring we got at Port Stanley. Maxillary barely to front of pupil. The following field notes were taken in Lake Erie: Lake Huron Herring, Port Stanley. Specimen 18 inches long. Head 53; depth 5; scales 8-10-83. Dorsal 10; anal 10. Body elongate, lanceolate, not much compressed. Head pointed, lower jaw distinctly projecting. Maxillary 81 inches head, reaching front of pupil. Eye 44 in head. Snout 53. Max- illary on level of pupil. Supplemental maxillary half longer than broad, with a posterior hook. Depth of head under middle of eye 3 length of head. Sub- orbital to mucous channel nearly as broad as pupil. The bone % of breadth of eye. Pectoral 12 in head; ventral 1}; dorsal 13. Color.—Olive sides, very silvery, with lighter streaks along rows of scales. Dorsal, anal, pectoral, caudal, and ventral edged with blackish (smaller fish with the ventrals white). Depth of caudal peduncle at base of fin 2 in length from adipose fin. Gillrakers about 35 below angle. Greatest depth of tail 12 in its length. This differs from the Jumbo Herring in being more slender, less compressed, with slimmer, more pointed head, narrower and much more projecting lower jaw, and much slimmer and rounder caudal peduncle. ARGYROSOMUS ZENITHICUS Jordan and Evermann, new species. LONGJAW OF LAKE SUPERIOR, Head 34 in length to base of caudal; depth 42; depth of caudal peduncle 34 in head; eye 54; snout 32; mouth larger than in re- lated species, almost as large as in Argyrosomus hoyi; maxillary to tip of upper jaw 22 in head; scales in lateral line 72; between lateral line and base of dorsal 8; between occiput and origin of dorsal 30; D. 10; A. 12. Form of body elongate, compressed, the width contained about 22 times in the length of head; snout pointed; jaws equal in length, the lower usually closing with the upper lip, as in A. hoyt. Maxillary extending to a point below center of pupil, the supplemental part wider than that below it. Giullrakers on first arch 17-+-25; very slen der; the longest contained 6 times in the length of head. Origin of dorsal midway between tip of snout and base of caudal, its height contained 13 times in length of head; base of adipose dorsal less in length than height, the height about equal to diameter 170 PROCEEDINGS OF THE NATIONAL MUSHUM. vot. xxxvi. of orbit; caudal deeply forked, the lobes. equal; length of fin about an eye’s diameter less than length of head; height of anal contained 91 times in length of head; ventrals inserted below middle of base of dorsal; length 12 in head; pectorals unusually long, 14 in head, reaching considerably more than half way to anal. Color silvery; dusky above; dorsal, caudal, pectorals, and anal more or less suffused with black toward the borders, the anal much lighter than the others. Ventrals immaculate; none of the fins wholly black, as in Argyrosomus nigripinnis. Scales loose, falling readily. Type.—This description is taken from the type, Cat. No. 62577, U.S.N.M., a specimen 330 mm. long. It was obtained in Duluth, Minnesota, by Mr. John Coventry, of Booth and Company, in Sep- tember, 1908, it having been taken in deep water off Isle Royale. A cotype, No. 13084, Stanford University, of the same size, and taken at the same time and place, is apparently like the type, in all essential Fic. 3.—ARGYROSOMUS ZENITHICUS. details. Hundreds of specimens of this species were seen in the cold- storage plant of Booth and Company, at Duluth, “ the Zenith City.” Argyrosomus zenithicus lives in much deeper water than the ordinary lake herring, and makes a part of the autumn catch in deep water. A large part of this catch, however, is of another species, apparently undescribed, called the Bluefin. The present species is locally known as Longjaw, and is more or less confounded by the fishermen with the true Bluefin, which seems to be an ally of Argy- rosomus prognathus, and with the Mooneye Cisco, Chub, Longjaw, or Kieye (Argyrosomus hoy?), which occurs in Lake Michigan and Lake Huron, but which we did not see in Lake Superior. From the Mooneye Cisco, Argyrosomus hoyi, as seen in Lake Mich- igan, the Lake Superior Longjaw differs in being less silvery in color, with the scales thinner, looser, and more dotted with black; in having a much longer pectoral, in the longer head, longer jaws, and larger adipose fin. The Blackfin Cisco, of Lake Michigan, Argyrosomus nigripinnis, has the fins all black, the head shorter, the jaws shorter, and the snout shorter. The pectoral is long in both species. NO. 1662. NEW LAKE HERRING—JORDAN AND EVERMANN. ial The Bluefin above mentioned is also a deep water Cisco, with the body robust and the scales firm; head 44 in length; depth 33; maxillary reaching front of eye, 2? in head; fins, pale; the dorsal and caudal with the upper edge of pectoral dusky. The species reaches a larger size than the common herring, the specimen above noted, but not preserved, being 134 inches in length. Having no specimens at hand, we refrain from naming this species. The Bluefin is not a good food-fish, being rather poor and dry, the flesh rather soft. The Longjaw, described as Argyrosomus zenithicus, is a good fish for smoking. The flesh is soft, but it has a delicate flavor when fresh, though poor and bony after freezing. NOTE ON THE SPECIES OF WHITEFISH. In this connection it may be noted that the common Whitefish of Lake Superior is the species called Labrador Whitefish, Coregonus labradoricus Richardson, characteristic of the Lake of the Woods and of the Canadian lakes generally, and that it is apparently distinct from the Whitefish of Lake Erie and Lake Ontario. The Lake Superior Whitefish was first named Salmo clupeaformis by Mitchill, whose specimens came from the Sault Sainte Marie. Only the Labrador Whitefish is found at Sault Sainte Marie, where it was formerly netted or speared in large numbers by the Indians and where it still readily takes the hook. Large numbers are hooked every day, in the locks of the ship canal, by local anglers.. The Erie Whitefish does not take the hook. The technical differences separa- ting the two species are slight, but apparently constant. , Mr. Harry Marcks, director of the fish hatchery at Sault Sainte Marie, tells us that the eggs of the Superior Whitefish are different from those of the Lake Erie Whitefish, being larger and darker in color. The fry are also distinguishable, those of the Superior White- fish being much livelier and having two dark lines along each side. The Lake Superior Whitefish must therefore stand as Coregonus clupeaformis (Mitchill), or clupeiformis, if we demand correct spell- ing. The Whitefish of Lake Erie is Coregonus albus Le Sueur. The Whitefish seen by us in Georgian Bay and a series received from Che- boygan in Lake Michigan belong to Coregonus clupeiformis. The same species is found in Rainy Lake, Lake of the Woods, and Lake Winnipeg. The synonymy of the two species should stand as follows: COREGONUS CLUPEAFORMIS (Mitchill). Salmo clupeaformis MircHett, Amer. Monthly Mag., II, 1818, p. 321; Sault Sainte Marie. Coregonus labradoricus RicHARDSON, Fauna Bor.Amer., III, 1836, p. 206; Musquaw River, Labrador. Coregonus sapidissimus AGAssiz, Lake Superior, 1850, p. 344; Lake Champlain “type,” after Thompson; and Lake Superior. 172 PROCEEDINGS OF THE NATIONAL MUSEUM. vot. xxxvt. Coregonus latior AGASSIZ, Lake Superior, 1850, p. 348; The Pic, Lake Superior. Coregonus neohantoniensis PrescorT, Amer. Journ. Sci. Arts, XI, 1851, p. 342; Lake Winnipiseogee, New Hampshire. COREGONUS ALBUS (Le Sueur). Coregonus albus Lr Surur, Journ. Acad. Nat. Sci. Phila., I, 1818, p. 281; Lake Erie. Coregonus otsego CLtinvron, Med. and Phil. Register, III, about 1824, p. 188; Otsego Lake, New York. THE ISOPOD CRUSTACEAN, ANCINUS DEPRESSUS (SAY). By Harrier Ricnarpson, Collaborator, Department of Marine Invertebrates, United States National Museum, In 1818, Thomas Say* described the form which he referred to the genus Vwsa, as V. depressa. A single dried specimen of this species is to be found in the Academy of Natural Sciences in Phila- delphia which I have had an opportunity to examine, and which I? redescribed and figured in 1905. Another dried specimen of this species is to be found in the British Museum, which, according to White’ and Hansen,’ was presented to that museum by Thomas Say.¢ In 1840, Milne Edwards? redescribed V. depressa, and instituted for it the new genus Ancinus. The footnote given by Milne Edwards for Ancinus depressus reads as follows: Nesa depressa Leach, Collections du Musée brittanique de Londres. Cette espéce nous parait étre la méme que celle décrite sous ce nom par Say (Journal of the Academy of Philadelphia, I, p. 483.) . It is probable that Milne Edwards did not know that this speci- men had been presented by Say, and therefore referred it to Leach. a Journ. Acad. Nat. Sci. Phila., I, 1818, pp. 488-484. > Bull. U. S. Nat. Mus., No. 54, 1905, p. 272. ¢ List of the Specimens of Crustacea in the Collection of the British Museum, 1847, p. 105. @ Quart. Journ. Micr. Sci., XLIX, 1905-6, new ser., p. 132. € Dr. Hansen says of this specimen: ‘The specimen named seems to be the only one existing in any zoological museum; at least I have asked for material of this form in Paris and in American museums, but with negative results.” When I asked for Say’s types at the Academy of Natural Sciences of Phila- delphia I was told that they were not there, but on one occasion, when I happened to be at the museum, I accidently discovered them, f Historie des Crustacés, III, 1840, pp. 225-226. PrRoceepinas U.S. NATIONAL Museum, VOL. XXXVI—No. 1663. 173 174 PROCEEDINGS OF THE NATIONAL MUSEUM. VOL. XXXVI. In 1905, Tattersall « follows Milne Edwards in referring Vesa de- pressa to Leach. Leach, however, never described this form, the earliest description having been given by Say in 1818, and the next to follow being that of Milne Edwards in 1840. Tattersall says of this species (p. 65) in connection with remarks on the distribution of Bathycopea typhlops, an allied form: It is to be regretted that the locality of Ancinus depressus Leach is unknown. It would have been interesting to have compared the habitats of the two forms. Milne Edwards also says of Ancinus depressus “ Patrie inconnue.” Say gives the locality of the specimen of Ancinus depressus placed in the Academy of Natural Sciences of Philadelphia as Egg Harbor, New Jersey. White mentions North America as the habitat of the specimen presented by Thomas Say to the British Museum. In 1905, Tattersall instituted the family Anciniid@ for the reception of this genus and his new genus Bathycopea. In the same year Hansen created the section Ancinini of the Spharomine platy- branchiate to include this genus, as well as his new genus Ancinella and Tecticeps Richardson. Ancinus differs,- however, from any of the genera mentioned in the character of the first and second pairs of pleopoda, the first of which are single branched instead of double branched.” For this reason it can not be left where it has been placed in the classifications proposed by these authors.° Last spring in the material that came to the U.S. National Museum from Prof. A. E. Verrill was a single specimen of Ancinus depressus, collected at Woods Hole, Massachusetts, in 1885, by the U. S. Bureau of Fisheries steamer Albatross. It was found at a depth of 2-3 fathoms. The specimen is a female and, although it differs slightly from the figures given by Milne Edwards for this form, I am in- clined to think that the differences are perhaps sexual. The uropods are slightly shorter and the first pair of legs have the hand more enlarged. In the shorter uropoda, however, it agrees with the dried specimen in the Academy of Natural Sciences of Philadelphia.? As no complete figure has ever been given since that of Milne Edwards, I have thought it would be of interest to figure and redescribe this specimen, which has been preserved in alcohol, and also give some de- tailed drawings of parts which it has been impossible to study in the dried specimens. @ Wisheries Ireland Sci. Invest., 1904, II, 1905, p. 11. >See description and figures which are to follow. ©I prefer to retain Ancinus as the type and only genus of the family An- ciniide, but those who desire to follow the classification of Hansen may accept the name Spwromine colobranchiate for a fourth group to include this form. 4 See figure in Bull. U. S. Nat. Mus., No. 54, 1905, p. 272. NO. 1663. THE ISOPOD—ANCINUS DEPRESSUS—RICHARDSON. 175 ANCINUS Milne Edwards, 1840. ANCINUS DEPRESSUS (Say). Nesa depressa Say, Journ. Acad. Nat. Sci. Phila., I, 1818, pp. 485-484.—Ricn- ARDSON, Amer. Nat., XXXIV, 1900, p. 224; Proc. U. S. Nat. Mus., XXIII, 1901, p. 587. Ancinus depressus Minne Epwarps, Hist. Nat. Crust., III, 1840, p. 226, pl. xxx, figs. 17-20.—HANSEN, Quart. Journ. Micr. Sci., 1905-6, p. 152.— RICHARDSON, Bull. U. S. Nat. Mus., No. 54, 1905, pp. 271-272, fig. 282.— TTATTERSALL, Fisheries, Ireland, Sci. Invest., 1904, II [1905], p. 11-18, 65. Body oblong ovate, twice as long as wide, 6 "3"* by 125 "3. Sur- face smooth, punctate, and with a few markings. Color, in alcohol, whitish. Head very wide, much wider than long, 1$ "3". by 5 "j".; it is wider anteriorly than posteriorly, the antero-lateral angles being produced in a lateral direction and forming acute angles. The post-lateral angles are rounded. The anterior mar- gin is produced in the middle in a broad, quadrangular process between the basal articles of the first pair of antenne, and extends forward to the outer margin of these articles. It meets the frontal lamina at its anterior extremity. The eyes are round, com- posite, and situated close to the pos- terior margin of the head, but at a distance from the post-lateral angle equal to the width of one eye. The head is coalesced with the first thoracic segment about the middle, but the sides are free. The first pair of antenne have the basal article large, about twice as long as wide; the second article is half as long as the first; the third F 1 article is narrower than either of the first two and is one and a half times as long. The flagellum is com- posed of nine articles and extends to the posterior margin of the second thoracic segment. The second pair of antenne have the first article of the peduncle extremely short; the two following are sub- equal and but little longer than the basal article; the fourth and fifth are also short and subeqval, being but little longer than the two pre- ceding ones; the flagellum is composed of nine articles and extends to the posterior margin of the first thoracic segment, ANCINUS DHPRESSUS. x 142. 4 All former measurements were taken from a _ half-millimeter scale, and should be changed from millimeters to half millimeters. 176 PROCEEDINGS OF THE NATIONAL MUSEUM. VOL. XXXVI. The maxillipeds have the second, third, and fourth articles of the palp produced into lobes or processes. The mandibles have no mas- ticatory process; the cutting edge is provided with four large blunt Fic. Nese DEPRESSUS. Fig. 3.—ANCINUS DEPRESSUS. MAn- MAXILLIPED. X 773. DIBLE. X Ts. teeth; below the cutting edge is a process provided at the tip with ~ three teeth; below this process is a long spine. The Git two segments of the thorax appear to be a little shone than those following, which are subequal in length. All are of nearly Fic. 4.—ANCINUS DEPRESSUS. First Fie. 5.—ANCINUS Fic. 6—ANCINUS PAIR OF LEGS OF FEMALE. X T7i. DEPRESSUS. FIRST DEPRESSUS. THIRD AND SECOND PLEO- PLEOPOD. X 41. POD: XLS equal width, the sides of the body being almost parallel. With the exception of the first, the epimera of all the segments are distinctly separated ; they are broad, quadrangular plates with the post-lateral NO. 16638. THE ISOPOD—ANCINUS DEPRESSUS—RICHARDSON. 177 angles more or less produced, and they are all bent downwards about the middle almost at right angles with the segments. On the ventral side they are also produced in the form of a plate, covering the proximal extremity of the legs. The abdomen is composed of two distinct segments. The first is short, not as long as the last thoracic segment, and has no suture lines indicating other coalesced segments. The terminal segment is trian- gular in shape with the apex appearing somewhat truncate, owing to the sides of the segment being turned downward and inward, so that on the ventral side a funnel-like opening is formed. The uropoda are composed of a small rounded peduncle and a single slender, Fig. 8.—ANCINUS Vie. 7—ANCINUS DEPRESSUS. Fic. 9.—ANCINUS DE- DEPRESSUS. FourTH ; FIrtH PLEOPOD. PRESSUS. FIFTH PLEOPOD. X 41. OUTER BRANCH. PLEOPOD. INNER oA BRANCH. X 41. movable branch, tapering and acute at the extremity. This branch extends to the extremity of the last abdominal segment. , The first pair of legs are subschelate and have the propodus very much enlarged. All the other legs are ambulatory. The first pair of pleopoda are small, longer than wide and com- posed of a single branch furnished with long hairs. This branch is widely separated from the corresponding branch of the opposite side. The second pair of pleopoda are large, double branched, the two branches being placed side by side and attached to the peduncle, so that a sort of operculum is formed, completely covering the following pleopoda. The third pleopoda have the endopod slightly longer than the exopod; both branches without marginal sete. Fourth pleopoda with endopod and exopod of nearly equal length, and with- out marginal sete. Fifth pleopoda with both branches unjointed and without marginal sete. Owing to the difference in the structure of the pleopoda, this genus remains alone the type of the family Anciniidw. Bathycopea Tatter- sall cannot be retained in the family. Proc. N. M. vol. xxxvi—09 12 A NEW AMERICAN JURASSIC CRINOID. By Frank Sprrincer, Of Hast Las Vegas, New Mexico. The first specimen of a fossil species of Pentacrinide from Ameri- can rocks was described by Meek and Hayden in 1858," under the name Pentacrinites asteriscus, from some isolated stem ae found in the Jurassic near the southwest base of the Black Hills of Dakota. They afterwards redescribed and figured the species in their work on the Paleontology of the Upper Missouri.’ Their figures on Plate 3 were based upon the original specimen; but on page 67 the authors gave a text figure, not very accurate, of some stem frag- ments with cirri attached, which they referred with doubt to their species. This specimen, anus to the label in the U. S. National Museum, came from Red Buttes, iho neclen, a locality now included in the State of Wyoming. The deser iption was stated by the authors to apply “more particularly to the largest sized specimens,” which came from a different locality, and which, as represented by the figures on Plate 3, were considered by Dr. P. H. Carpenter’ to belong to the genus “ Hatracrinus.”” (Pentacrinus, sensu str.), al- though he perhaps based his opinion rather upon the figures given by White @ of a specimen from Utah than upon those of Meek and Hayden. So far as can be judged from a few isolated joints, there is reasonable ground to believe that the doubt expressed by the authors as to the specific identity of the two specimens is well founded; those of the typical form are nearly twice as large as the others, and the petaloid sectors on the articular face are more sharply angular. The transverse view given in the text figure on page 67 of the work cited is not correct, the structure being rather poorly “Proc. Acad. Nat. Sci. Philadelphia, x, p. 49; XII, 1860, p. 419. b Smithsonian Contributions to Knowledge, No. 172, 1865, p. 67, pl. m1, figs. 2 a, 0. ¢ Challenger Report, Stalked Crinoids, pp. 148, 297 @ Wheeler, Geol. & Geog. Surv., IV, p. 162, pl. x1m, fig. 6 a PROCEEDINGS U.S. NATIONAL MuSEUM- VoL. XXXVI—No. 1664. Aes 180 PROCHEDINGS OF THH NATIONAL MUSEUM. VOL. XXXVI. defined in the specimen. The Red Buttes specimen shows a rather obtusely pentangular, smooth stem, with straight sides, having eleven or twelve joints to the internode, and cirri tapering rapidly near the proximal end. Separate stem joints, more or less similar to both of Meek and Hay- den’s figures, have since been collected by the staff of the U. S. Geolog- ical Survey in various localities throughout the Rocky Mountain and Pacific regions, but no vestige of the crown was obtained until 1899, when the late Prof. W. C. Knight, of the University of Wyoming, in the course of some investigations among the famous Dinosaur beds near Medicine Bow, Wyoming, discovered some small slabs of lime- stone containing numerous stems and fragments of arms, with one very complete crown. This he reported as Pentacrinus asteriscus.4 In the following year Mr. H. T. Martin, of the University of Kansas, visited the Medicine Bow locality and succeeded in finding a few more pieces of the rock containing the crinoid remains, which by ‘areful cleaning have yielded some additional specimens useful for description. Through the obliging courtesy of these gentlemen this material was placed in my hands, but pressure of other matters has prevented the preparation of the necessary figures for their descrip- tion until now. The locality of these fossils is in the same region and horizon as Meek and Hayden’s Red Buttes specimens, and they probably belong to the same species. Assuming, for the reasons already given, that they are not included in the typical Pentacrinites asteriscus—of which in any event we know nothing beyond the form and size of separate stem joints—it seems proper to describe this form as a new species. I therefore propose to associate with it the name of the lamented geologist to whose researches we are indebted for its discovery. ISOCRINUS KNIGHTI, new species. 1865. ? Pentacrinites asteriscus Merk and Haypren, Pal. Upper Missouri, p. 67, text fig. (not pl. m1, figs. 2 a. b.) Specimens of moderate size. Stem smooth, long, slightly increasing in diameter distally; pen- tagonal with straight sides, except at the proximal end, where for the first few immature internodes the younger joints are stellate. Inter- nodals about 14, but varying from 12 to 17 in the mature parts; dis- tinctly crenulated at the margins; nodals not enlarged, scarcely dis- tinguishable from the others except by the cirrus sockets; these are. miber shallow, not extending to the hypozygal, or infranodal joint, but usually encroaching upon the ienneeel in which case the apposed faces of these twe joints are more or less indented, producing “Jurassic Rocks of Southwestern Wyoming. Bull. Geol. Soc. America, XI, p. 377. no. 1664, A NEW JURASSIC CRINOID—SPRINGER. 181 a marked stellate outline. Conformably to this structure the cirri are directed upward. Interarticular pores extending to the fifth internode. Cirri in whorls of five; round, long, and slender, com- posed of 40 joints or more; the proximal ones relatively short and broad—about one-third as long as wide—tapering rapidly to about half their breadth, and doubling in length in the first 8 or 10 joints, beyond which they continue uniformly about as long as wide to the end; terminal claw not preserved. Angles of stem interradial; cirri radial; axial canal in stem small, obtusely pentagonal, and apparently interradial in position. Cup forming a low cone, without any downward projection of basals or radials. Infrabasals well defined, filling half the diameter of the column facet and entirely covered by the proximal columnal. Basals large, smooth, visible in pentagonal outline, and in full contact exteriorly by their lateral faces; they form a closed ring, not pro- tuberent but flush with the plane of the radials, and about equal to them in height. Radials forming also a ring continuous with basals. Primibrachs two, united by articulation apparently bifascial. Arms simple, or bifurcating once from the sixteenth to the thirtieth ITBr, thus varying from 10 to 20; they are long, slender, with strongly oblique articulating faces, and they extend to upward of 90 brachials. Syzygies at I1Br 3+-4, and beyond throughout the erm at intervals of about 5 to 10 brachials. Pinnules long, rounded, composed of elongate joints, 15 or more in the distal pinnules, but the number in the proximal ones not observable. Disk unknown. Dimensions of mature individual. mm, VHWENEA CGE 0d LCE RON Ups AER OT aE Bh a eel Sa A AD sl re he em, Se 65 AIT GG La Rigo EYE 1S-crT) We ae] 1 8) pe ee eae ESA SB ai Te BS ne a Sear eed es ee ee BOE Bees fi RA etiivo teen eure ya bes ees Be eee Om pe “bee rey) 2 i ay 3 fn ie SS es o Wren HcieOlmeimeiis10 te Usp OlN bss eee eee ae EE Sig Be ee oe ea AS 32, Mensch or lon sest-StemspLeser wed eat: A ein ah eee Ue ee se aS 140 Dimmeter or stem at second internodes «2 = fe es a a 2 PMiamerer Or Stem) ab tentiiMbernodes.. 2 = ee ee ee ee 25 Diamererito heicht of Second internodes 2 = a a 1-2.2 Piameter to heshtrot teach danbermodes = 2-5 ao = ee ee 15.2 Diameter of stem to length of longest cirrus____________________________ 1-17 Horizon and locality—In the Shirley stage of the uppermost Jurassic. Medicine Bow, and Red Buttes, Wyoming. The occurrence at Medicine Bow was in a band of argillaceous limestone about 1 inch thick, and it is undoubtedly the remnant of a considerable colony. The upper surface of the layer is filled with disintegrated stem joints and brachials closely cemented together, while toward the lower part the crinoids had been embedded as they perished without much disturbance. Unfortunately the layer was 182 PROCLEDINGS OF THE NATIONAL MUSEUM. VOL. XXXVI. not found zn situ, all the specimens being derived from small, loose pieces in the débris. Most of the crinoids are therefore broken and imperfect, and only a few preserving parts of the arms and column were recovered. The most complete is the one given me by Professor Knight (Plate 4, fig. 1). Unfortunately the structure was not understood when found, and the stem of this specimen was almost stripped of its cirri by too energetic cleaning in the field; most of those on specimen No. 2 had been removed by weathering, but by careful manipulation of the embedded proximal part of this stem and of some other stem fragments I have developed the cirri so that their length and proportions can be ascertained. The form under consideration is clearly distinct from species like Pentacrinus fossilis and P. subangularis, in which the radials project downward over the proximal columnals, and to which type only, as clearly pointed out by Doctor Bather, in his paper on “ Pentacrinus, a Name and its History,” * the name Pentacrinus properly belongs. But it falls readily under the genus /socrinus (Agassiz, 1836). A brief excursion among original sources enables me to add a little to the very elaborate and instructive history of these names given by Bather in the work cited. /socrinus, although described by von Meyer in 1837, as stated, was actually published as a generic name by Agassiz in 1836." Speaking of this form, Bather says on page 250 that “ no figure of a fossil crinoid of our type C (/socrinus) is known to me before 1800.” In an extensive work by Daniel Briickner, en- titled “ Versuch eimer Beschreibung historischer und _natiirlicher Merkwiirdigkeiten der Landschaft Basel,” published in 23 parts, or “Stuecke,”’ from 1749 to 1763, there is on pages 2425-2431 of the twentieth Stueck a good figure—No.37—of a _ well-preserved specimen of this type, from the Swiss Jura, showing arms, stem, and cirri, accompanied by a long description and a name. The original specimens have been refigured by de Loriol in his “ Crinoides de la Suisse,” Plate 14, figs. 31-88, under the name Cainocrinus andrew Desor, a genus since considered by him to be identical with Isocrinus. The twentieth Stueck of Briickner’s work was published in 1761, and to the crinoid figured and described as above stated he gave the name HLntrochites ramosus, vel Encrinus, Lilium marinum. So not only was the type figured and described before 1800; but a name was given to it in binomial form, thus raising the question whether the real name of our genus is not the venerable and classic term Entrochites, thus for the first time brought into the domain of valid nomenclature. 4Natural Science, April, 1908, p. 252. ‘The Nomenclature of the Recent Crinoids. Austin H. Clark, Proc. U. S. Nat. Mus., XXXIV, 1908, p. 526. xo. 1664. A NEW JURASSIC CRINOID—SPRINGER. 183 However, Briickner did not employ binomial names consistently, many of those relating to crinoidal remains being polynomial, as, for instance, Entrochites fungitew adhwrens, eighth Stueck, page 888; Encrini minoris pulere ramificatum, etce.; and his incidental use of Entrochites ramosus may probably be disregarded for that reason. The case of Encrinus is much more serious. Bather credits it to Schulze (1760), who wrote it “A’nerinwm,” probably as the accusa- tive of Hnerinus. Schulze’s work was mainly a compilation from former authors, as Linck, Lhuyd, Seba, and Ellis, and he uses their names in the same manner as they did, with but small pretense to binomial application. He did not propose Hncerinum to represent a genus, but only mentioned by way of recital the fact that certain petrifactions resembling a lily have been called the lily stone, Encrinum. This is what he says: “Man findet eine gewisse Ver- steinerung, die, in Ansehung ihrer Gestalt, einige Gleichheit mit einer Lilie zu haben scheinet; daher man dieselbe anfainglich fiir die Ver- steinerung dieser Blume gehalten, und sie den Lilienstein, Encrinum, genennet hat.” ¢ On Plate 4 is a figure of a complete crown of the fossil to which he refers, and in the long description which follows he mentions it four times by the name “ Lilienstein,” but never again as H’ncrinum. It seems to me there would be as much reason for recognizing as valid names the Decacnimos (=Antedon) and Triscadecacnimos (probably =Comatula) which he transliterates from Linck, because it was the first post-Linnean use of them, as “ncerinum, which he recites as an equivalent of the name he actually uses in description—Lilienstein. Yet nobody recognizes these names, the ground of their rejection be- ing, I suppose, that they are not binomial, which Hncrinum certainly isnot. I regret to find myself led to this impression by an inspection of Schulze’s work, because there are serious troubles ahead for the name “ /acrinus,”’ from which we would be saved but for its doubtful standing there. The earliest use of the name “ “xcrinus ” in a binomial sense that TI know of was by Andreae in his “ Briefe aus der Schweiz,” published in the Hannoverisches Magazin in 1763-64, and afterwards in book form in 1776. On page 4 of this work he formally proposes the name Encrinus coralloides for certain fossils which appear to him to be a species of Hnerinus or Lilienstein not before recognized, and which had been figured on Table 8 of the eighth Stueck of Briickner’s work above mentioned. He also refers to figures of similar specimens given by Rosinus? on Table 10, A, B, C, D, E. These fossils are now supposed to be the terminal stem branches or roots of MWillericrinus, and one of them—Briiciner’s fig. A—has been * Betrachtung der Versteinerten See-Sterne und ihre Theile. p. 21. ’Testamen de Lithozois, 1718. % 184 PROCEEDINGS OF THE NATIONAL MUSEUM. Vole Sxocvale referred to M. echinatus by de Loriol.". Therefore a strict observance of the rule of priority might seem to require us to transfer the name * Enerinus ” to the crinoidal remains which we have for three-quar- ters of a century ignorantly been calling M/illericrinus, and to relegate to obscurity our still older acquaintance, #’. diiformés, until some one introduces it to us afresh under a new name. But if we hold that Andreew’s name was applied to unrecognizable fragments, and for that reason is not valid, our troubles over Hncrinus are not ended. The name was used by Blumenbach in 1779 in the first edition of his “* Handbuch der Naturgeschichte,” page 485, in a strictly binomial sense, for a genus with three species, arranged as follows: ENCRINUS: (1) asteria (Linneeus, after Guettard). (2) mylii (based on Mylius’ Greenland specimen (3) boltendi (based on Boltenius—an Ascidian.) - Here the name is taken out of the domain of Paleontology and applied to a recent crinoid—the type species being Guettard’s famous Palmier marin of Boisjourdain, best known in literature as * Penta- a Pennatulid). erinus” caput-meduse, or i present nomenclature as /socrinus asteria Linneus. In the third edition, 1788, Blumenbach again gives the genus /’n- crinus with asteria as the first species; and in 1801 Lamarck, the generally accepted father of Hncrinus as now commonly known, in- the first edition of his “ Systeme des Animaux sans Vertébres,” p. 379, recorded the genus as follows: EINcRINUS: (1) caput-medusw (=Isis asteria Linneus.) (2) liltiformis. No. 2 of Blumenbach was made the type of a new genus—U mbellu- daria, and in 1816 Savigny’ made Blumenbach’s species No. 3 the type of another genus, Boltenia. Thus by the year 1816 Hnerinus was definitely restricted, by the removal of two of its original three species, to the group with asteréa as the type. If Blumenbach’s name is to stand, the subsequent references of wsteria to Pentacrinus and /socrinus are invalid, and the reference by Lamarck of liliifor- mis to Hnerinus must likewise fall to the ground. According to the rules it will have to stand, unless theretofore validly applied to some- thing else; and unless it has been so applied, /é/iiformis can not stand under it. “Crin. de la Suisse, p. 75. +’ Mem. sur les Animaux sans Vertébres, p. 140. . NO. 1664. A NEW JURASSIC CRINOID—SPRINGER. 185 The consequences to our literature of a strict application of the rule of priority to either of these nomenclatorial discoveries would be somewhat appalling. Suppose we take— . 1. Knerinus, Blumenbach, 1779; type, 2. asteria, which is good unless preoccupied by something earlier. This will require a. A new generic name for Lacrinus Liliiformis, which has been used for nearly a century for the best known of all crinoids—one which has been figured and described as such in countless works, and specimens of which are found under that label in all the cabinets and museums of the world. b. Applying the name, so long associated with the most familiar fossils, to new, different, and unfamiliar use. ¢. Supplanting the name /socrinus after it has become thoroughly well established in lterature, and is now currently employed by all writers on the recent crinoids. Or, if we take— 2. Hnerinus, Andreae, 1776; type L. coralloides (=Millericrinus echinatus) ; this, if good, upsets Blumenbach, but does not save us from results equally direful. For it likewise requires us— a. To provide a new generic name for Z’. liliiformis. b. To apply the old name, with all its familiar association, to new and different fossil forms, occurring in the same region, well known and abundantly represented in literature under another name for seventy-five years. c. To give a new name to JMJiéllericrinus. This brings us back again to— 3. Encrinum, Schulze, 1760; no type-species stated, but the name was probably intended for the fossil commonly known as F’. liliiformis, which he figured. Schulze’s use of the term was not binominal, and the case is a hard one; but he did use some other names binomially, and it may be presumed that he intended to do so with this. To recognize his name as valid would avpid all confusion, and leave the literature as to all three of the names involved undisturbed. And in a case like this, arising in the dawn of our science, before the rules of nomenclature had become formulated, or were even practically thought of, I think that expediency and the question of practical disadvantages or benefits to the scientific public are to be consid- ered where there is a possible alternative and some room for the exercise of discretion. Here, on the one hand, is invited intolerable confusion and the overthrowing of long familiar and classic names to an extent that will bring the rules of nomenclature into disrepute ; and this without serving any useful purpose and without benefit to anybody, unless it be the satisfaction of some delver among musty tomes, as I am, in making all the trouble he can. On the other hand, there is the preservation of these names in the sense to which all 186 PROCEEDINGS OF THE NATIONAL MUSEUM. VOL, XXXVI. general zoologists and paleontologists are accustomed, without injury to anyone, or the infringing of any principle except that of an extreme technical construction of the rules. The underlying principle of the rule of priority is said, and prop- erly said, to be fiwity. Yet by insisting upon its absolute and un- bending application to all cases, without regard to circumstances, we may destroy the very fixity for which we contend. There is no law more deeply rooted in the foundations of civil government, or more essential to the welfare and stability of society, than that of the fixity of the titles to real estate based on priority. But just as that law in actual administration is subject to exceptions founded upon principles of natural justice and the dictates of public policy, so I think we may find reasonable basis for an exception to the rule of priority in nomenclature which will meet such cases as this. This would be that such names, irrespective of the actual state of the record as to their dates, should be protected under an exception to the rule, simply on the ground of long use, on the doctrine of pre- scription, which is a principle well known in law, recognized in con- tinental Europe as coming down from the civil law of Rome, and now embodied in statutes in all English-speaking countries. It is that the right of property will be upheld by the courts in favor of one who can show a long, continuous, and undisputed possession of it, under a claim of right however defective, notwithstanding he has no paper title, and even though the records may show the prior title to be in someone else. This rule of law rests upon the idea that it is for the public interest that there be an end of controversy, and that there shall be some reasonable time after which titles may be held safe from attack on any ground. And this end was attained in the beginning, not by denying or abrogating the law governing the con- veyance of property by deeds, but by invoking a simple presumption, founded on the known and usual conduct of men with regard to their interests, that where such long and undisputed possession existed there must have been a good title, the evidence of which is lost. This principle of jurisprudence is now recognized throughout the civilized world, as one of the most salutary and beneficial provisions for preventing injustice, and insuring that repose of titles which the peace and order of society demand. By virtue of its operation a title by lapse of time merely, if properly proven under all the safe- guards which are prescribed in practice to prevent the abuse of it, is as good in the actual possessor as a paper title showing priority by an unbroken chain of recorded deeds. If this be true with regard to matters of such vital importance as the titles to our landed prop- erty, why may not the same principle be invoked in favor of repose and stability of names in our scientific literature? It is not a ques- tion of “ doing justice” to any particular ancient author. The propo- no. 1664. A NEW JURASSIC CRINOID—SPRINGER. 187 sition is one of far broader significance, and involves the paramount interest of the scientific public. I am much in sympathy with the protest voiced by Dr. G. A. Boulenger, at the Dublin meeting (1908) of the British Association, against the extreme application of the rule of priority, where the effect would be, as in this case, to overthrow old and well under- stood names, or to transfer them from one object to another. He renews a suggestion made by Sir E. Ray Lankester ten years earlier, that there should be created by the International Congress some kind of committee, having the powers of a court of last resort, to decide upon the application of such an exception to the rule of priority in particular cases. In the meantime. and until overruled by some such higher au- thority, I shall maintain that, irrespective of the merits of their original titles to priority, the names of ’ncrinus and Millericrinus have become valid simply by the lapse of time, by long usage in the sense in which they are now generally understood; and that by reason of universal acquiescence in such use for nearly a century, zoologists are now estopped from disputing them. In this way, by analogy to the practice which prevails in courts of justice touching the most solemn rights of property, a presumably just conclusion can be reached independent of the rule of priority, and without impairing its force in cases to which no such considerations of public policy apply. With these two names thus firmly established, that of Jsocrinus is ipso facto confirmed, and I am enabled to proceed with further comment on the species under consideration, without the necessity of searching for a new generic appellation. In view of the generally assumed absence of infrabasals in “ Pen- tacrinus” (sensu P. H. C.) and Metacrinus twenty years ago, and in the recent species until the past year, it is interesting to find their presence now fully demonstrated in no less than six species; two fossil—this and de Loriol’s 7. leuthardi—and four recent ones within the past few months. Doederlein described them in Metacrinus acutus in November, 1907, and they were independently discovered by Mr. Austin H. Clark, who communicated the facts to me under date of November 29, 1907, in two other species of Metacrinus, and also in [socrinus decorus.” The infrabasals in our species were only observable in a single specimen, a rather small individual, in which the stem was broken off at the top joint, by which they had been covered (Plate 4, fig. 5a). As thus exposed they are perfectly distinguishable, and are somewhat larger than those figured by de Loriol. “Die Gestielten Crinoiden der Siboga Expedition, p. 20. b Proc. U. S. Nat. Mus., X XXIII, p. 671-676. 188 PROCEEDINGS OF THE NATIONAL MUSEUM. VOL. XXXVI. There can be no longer any doubt that the Pentacrinide are all either actually or potentially dicyclic, though in some species the in- frabasals are resorbed at an early stage. This has been shown by Mr. Clark to be the case in /socrinus parre (olim mullert) 7 and the observations of P. H. Carpenter (Stalked Crinoids, pp. 292-93) would seem to indicate that a similar condition prevails in J. wyville- thomson, I. asteria, and I. alternicirrus. I have found the same thing to be the case also in certain species of the paleozoic genus Ichthyocrinus. As to specific relations, it is impossible, with the material avail- able, to make any very satisfactory comparison with European species, a great many of which have been described from isolated stem joints. Although the stem as a whole often affords valuable characters for distinguishing species among the recent crinoids, and even a part of it, if the same parts can be compared, little reliance can be placed in species whose identification depends wholly upon the form and articular markings of joints whose position in the stem can not possibly be known. This has been pointed out by Carpenter Sines Crinoids, pp. 226, 298), and the fact is well shown by his Plate 22, where many different forms of columnals from the stem of 7. wyville-thomsoni are figured. Mr. Clark has recently found by dissection of the stem of a young /. decorus? that in the different parts of the same stem may be found almost every type of articular face, from stellate to round, and from a bifascial articulation with transverse ridge as in Rhizocrinus, to the radiating petaloid sectors of the usual /socrinus type. Several different forms of stem joint are found in the present species, the more common being obtusely pentagonal, while the younger: joints near the calyx become stellate. The proximal face of the nodal joint also shows a sharply stellate outline, due to the indentation by the cirrus sockets (Plate 4, figs. 9, 10, 11, 12, 18). In the associated material are fhoncenee of separate joints, besides several considerable portions of stems intact, and there is a general uniformity of size and appearance among them which indicates their probable derivation from a single species. They are uniformly different from the much larger ones on which P. asteriscus was founded, and from the Utah specimen referred by Doctor White to P. asteriscus,’ but afterwards separated from it by Dr. W. B. Clark under the name Pentacrinus whitei, because of its alternating joints. Clark’s comparison was made chiefly with the Red Buttes specimens of P. asteriscus (7), but the separation is doubtless * well founded, nevertheless, as the character on which he bases it is clear in his specimen, and can not be shown in the type of P. asteris- cus. The difference between the stem of our species and that of P. @Proc. U. S. Nat. Mus., XXXV, 1908, p. 87. bTdem, XXXV, p. 88. ¢ Bull. U. S. Geol. Surv., No. 97, p. 27. No. 1664. A NEW JURASSIC CRINOID—SPRINGER. 189 white is similar to that between the recent 7. decorus and J. parra, which is fairly constant. The most nearly related European species that I know of is de Loriol’s “ Pentacrinus ” beaugrandi, from the Upper Jurassic, Port- landian stage, near Boulogne-sur-Mer, France.*| This was the only Crinoid known to the author from the Portlandian stage, and it is the species which he originally proposed to separate from the other Pentacrinide on account of having a closed ring of basals, under the name Picteticrinus. In this he found himself anticipated by the Cainocrinus of Forbes, and in the work last cited, page 281, he aban- doned the distinction, and referred the species to Pentacrinus (sensu P. H. C.). It has similar large basals, but the arms branch lower down, the stem is more sharply stellate in corresponding portions, and the cirri much more delicate. The stem is preserved to the fourth in- ternode, which has 8 internodals, whereas ours has 14 at the same stage. Pentacrinus (Cainocrinus) andrew Desor® is similar to the French species, but with shorter basals and shorter internodes. The excellent preservation of our specimens enables us to make an interesting comparison with recent species. The stem has a con- siderable resemblance to that of 7. decorus, except in the disposition of the cirri. It must have been quite long, as the longest portion, preserved to a distance of 140 mm., shows little sign of any rounding. It is rather more pentagonal for equivalent distances. The cirri are very long and slender; the taper near the base from short and wide joints to long, narrow, and equal ones, is quite marked. The most perfect one has 44 joints, and this was probably near the maximum. The interesting thing about the cirri, however, is the fact that they are directed upward instead of downward or outward. In conse- quence the sockets do not extend to the infranodal (hypozygal) joint, but slope upward toward the supranodal, the lower margin of which is often incised by them. This is more or less the case in the genus Metacrinus, but is not usual in the recent species of Jsocranus, most of which have the cirri directed downward, though in some, as J. asteria and I. wyville-thomsoni, the socket is confined to the nodal joint, and the cirri are given off about horizontally. The basals, as shown by the five specimens figured and three others, ‘are quite uniform in their form and proportions. They form with the radials a low funnel, with smooth or slightly rounded sides, and without protuberance or projection of any kind. They are connected exteriorly by their lateral faces, giving a pentagonal outline and forming a closed ring (Plate 4, fig. 3a), as in the type for which Forbes proposed the genus Catnocrinus, instead of appearing as mere triangular points separated from each other by the radials, and tend- @Mon. tage Jur. Boulogne-sur-Mer, 1875, p. 298, pl. xxvi, figs. 23-25; Paléontologie francaise, Crinoides, XI, 2° partie, p. 278, pl. CLXxxt, figs. 1-3. > De Loriol, Crin. Foss. de la Suisse, p. 112. 190 PROCEEDINGS OF THE NATIONAL MUSEUM. VOL. XXXVI, ing more or less to project downward over the proximal column joints, as in most recent species. The bifurcation of the arms so far beyond the axillary IBr is an unusual feature, occurring in the largest specimen at the twenty- seventh to the thirtieth brachial (Plate 4, fig. 1), and in other specimens from the sixteenth to the twenty-third. I know of no Pentacrinoid in which arm division takes place so high up; nor in fact any inadunate crinoid, the nearest approach to it being found in the Carboniferous genus Poteriocrinus. There is little tendency of the arms to spread out, but they are long and slender, tending rather to he in a bundle. The general aspect of calyx and arms is somewhat like that of 7. naresianus, which it also resembles in the number and regularity of the syzygies, which is unusual in the Penta- crinide. I can trace them in two arms of specimen A (Plate 4, fig. 1) part way, and in one to the end, and can distinguish them in the. distal portion of some other arms. Beginning at I1Br 3-4, they occur at intervals of mostly about 10 brachials, but sometimes 4, 5, or 6. I give a figure of the pair next to the last, being about brachials 79+80 of that arm. (Plate 4, fig. 1a.) The type-specimens figured are deposited in the U. 8S. National Museum, where they will be available for comparison with the mag- nificent. collection of recent crinoids now being accumulated there. For convenience of reference they are designated by the letters, A, B, etc., as indicated in the explanation of the plate. EXPLANATION OF PLATHE 4. Tsocrinus knighti, new species. Fig. 1. Large specimen, A; with bifurcating arms complete and part of stem; cirri mostly lost. la. Syzygy at I1IBr 79+80 of same specimen. 2. Large specimen, B; with stem 140 mm., and part of arms. Some arms of another individual attached. 2a. Detail of stem at “a” of same specimen, showing interarticular pores, X2. 2b. Detail of same at “0b; showing cirrus sockets, X2. 3. Small specimen, C; with part of arms, some not bifurcating. 3a. Calyx and lower IIlIBr of same specimen; showing form and propor- tions of basal and radial plates, x2. 4. Small specimen, D; with part of arms, one with an axillary, and some apparently simple. 5. Small specimen, H; with two arms simple and one bifurcating at 23d IlIBr; stem detached, exposing infrabasals. 5a. Basal view of same specimen, showing infrabasals, 4. 6-8. Portions of different stems, /, G, H ; showing cirri. 7a. The longest cirrus on specimen G, X2. 9-13. Weathered stem joints associated with the other specimens; 9, 10, 11 are mature internodals; 12 is the proximal face of a nodal incised by the cirrus sockets; 13 is a deeply stellate joint from the young- est part of the stem; all, X2, 4 + ‘ I- 6 Y ‘ : Ve) , F a y Wary, sey f a S ; : at ape ie F x ee ome NS pMDI 533) ene . a x iM soba aber & ee he bi wy an. | if 5 ee SF ee id z 4 SENT ATT ; © ns a 16) F (6) g 16) a “ = i 3 0 i ° Ss ones es as ae or ag soem eeiehL))) aS 2 : " &> 2 ul e 10) Zz or oa WY) - at, 10) Ze MS Y =) Zz fa O O 09) z : : 4 : Ww : a 2 zZ = 6 = z = 3 o é < 2 Be E 16) 3 ‘ G = SS ee ee eee a eee aes Pee ee a. fe Sore DESCRIPTION OF TWO SPECIES OF FOSSIL TURTLES, TOXOCHELYS STENOPORA AND CHISTERNON? IN- TERPOSITUM, THE LATTER HITHERTO UNKNOWN. By Otiver P. Hay, Of Washington, District of Columbia. The thanks of the writer are due to the officers of the United States National Museum for the opportunity to describe and illustrate the materials which represent the two species of fossil turtles which form the subject of the present paper. TOXOCHELYS STENOPORA Hay. The remains of this species which are here described were obtained from Mr. C. M. Sternberg, of Lawrence, Kansas, in the Niobrara beds along Butte Creek, Logan County, Kansas. The catalogue number in the U. S. National Museum is 6013. The specimen presents a large part of the bones of a single individual, but in a considerably disturbed condition. Plate 5 shows the position of the various bones after the removal of the matrix that overlay them. Apparently all the bones of the skull are present, but to a consider- able extent separated from one another. Very few vertebrae have been preserved. The elements of the carapace have mostly been dis- placed ; those of the plastron to a less extent. The pelvis is missing, as well as most of the bones of the hinder limbs. The specimen is identified as 7'owvochelys stenopora, but there are not wanting some discordant characters. The individual was a small one. The carapace had an estimated length of 160 mm. and a width of 167 mm. The width is indicated by some undisturbed right and left peripherals and by the elements of the plastron. The carapace is relatively broader than that of 7. bauri Wieland. The front of the skull, including the maxille, the vomer, the pala- tines, and the prefontals, lay on the slab so as to present the palatal surface (Plate 5, 7). On being removed and cleared from matrix it shows the nasal opening and the anterior half of the orbits. The nasal opening is small, being narrow, as in the type of the species.* Its width is 5 mm., its height is 7.5 mm., but this has evidently been = a ed a “Hay, Fossil Turtles of North America, p. 172, fig. 217. PROCEEDINGS U. S. NATIONAL Museum, VOL. XXXVI—No. 1665. 191 192 PROCEEDINGS OF THE NATIONAL MUSEUM. VOL. XXXVI. “reduced by some downward crushing of the prefontal bones. The orbit seems to have had a fore-and-aft diameter of about 17.5 mm. The width of the interorbital space is 11 mm. A comparison of the snout of this specimen (Plate 5, 7) with that of the type-specimen will show that that of the latter was considerably blunter. This may be due to the greater age of the type-specimen, to sex, or possibly to a difference of species. On the plate the numeral S§ is placed on what seems to be the outer surface of the left quadratojugal; numeral 9 is on the inner surface of the left jugal. The numeral 70 is on the inner surface of the right jugal, partly covered by a costal plate. The left postorbital, 4, pre- sents its inner surface and hes against the left parietal, but not in the natural relation of the two bones. The numeral 4 is on the outer surface of the right parietal. The inner surface of the right post- frontal, 77, is shown. Attached to its hinder end, apparently in its original place, is the right quadratojugal. The numeral 7 is on the inner surface of the left squamosal. The supraoccipital, 6, has its hinder end directed upward on the plate. It did not have the great height that the same bone of the type had.t The left opisthotie, joined to the corresponding prootic, presents its upper surface, 2. Through the wrenching of the supraoccipital from its place, the brain cavity has been exposed, and is seen just above and a little to the right of the numeral 2. The pterygoids are seen in position in front of the brain cavity. The lower jaw, 3, has been little removed from the position occu- pied by the skull at the death of the animal. The upper surface of the jaw is exposed to view. The lower jaw greatly resembles that of the type of the species,” including the section of the symphysis. The skull appears to have had a length, from snout to occipital condyle, of nearly 55mm. One of the ceratohyal bones is seen at the hinder end of the left ramus of the lower jaw. The structure of the carapace can not be completely determined. Its hinder portion is missing. The nuchal, 7S, has the form of that of 7. latiremis, as figured by Case.° It is quite different from that of T. bauri Wieland.4 On the left side of the carapace there is a series of eight peripherals, 19-26, in their natural relations. It is pretty certain that the anterior one, 79, is the first. The anterior five of the right side are present, 35, 36, 27, 28, 29. The first of the right side and that of the left are separated by a distance equal to the width of the nuchal. The third peripheral on each side, 27 and 27, contains a pit for the end of a rib. Other peripherals, 30, 31, 92, and 34, have been washed forward @Hay, Fossil Turtles of North America, fig. 219. bTdem, fig. 214. ¢ Univ. Geol. Surv. Kansas, IV, pl. 82, fig. 3. @Yossil Turtles of North America, fig. 229. No. 1665. TWO SPECIES OF FOSSIL TURTLES—HAY. 193 from their original positions in the skeleton. The pygal is not present. A few neural bones are present, 72-76. These show that there was a sharp keel running along the middle of the carapace. The number 17 is placed at the side of an ossicle that had a position across the suture between two of the neurals, as in other species of the genus. Several of the costal plates are present, but some are missing. They have the form usual in the genus, the distal half being very narrow, 37-39, 41-47, 53. The plastron is present, except the left epiplastron and probably the right xiphiplastron, but the various bones have been slightly dis- turbed. The front of the plastron was covered with other bones in a way to hide it, and some of these had to be lifted temporarily. » Fic. 1.—ToOXoOCHELYS STENOPORA. PLASTRON. X &. ent, ENTOPLASTRON; epi, WPIPLAS- TRON ; hyo, HYOPLASTRON ; hypo, HYPOPLASTRON ; wiph, XIPHIPLASTRON. As accurately as possible, the plastron has been restored in fig. 1. The epiplastra and the entoplastron have not, so far as known to the writer, been hitherto observed. Each epiplastron is a narrow, curved bone 33 mm. long, whose blunt anterior end joins its fellow. These bones were not prolonged forward as they are in Chelydra and the Cheloniidz. The entoplastron is a spear-shaped bone, close to 10 mm. wide in front and narrowing posteriorly to a blunt point. The other bones of the plastron resemble closely those of the type of the species. The bridges have a width of 38 mm., which is equally Proc. N. M. vol, xxxvi—09——13 194 PROCEEDINGS OF THE NATIONAL MUSEUM. VOL, XXXVI. plastron is 40 mm. long and 11 mm. wide. The hyoplastron and the hypoplastron of the one side seem not to have come into contact with those of the other. There is a large umbilical fontanel. The scapule, 56, 57 are both present. The right coracoid, 54, is 27 mm. long. The humeri, 55, 58, are each 27 mm. long. The head of that of the right side appears behind a costal, 47, that of the left humerus behind another costal, 39. The left radius and ulna are seen between the scapula, 56, and the humerus, 4S. Some fore-foot bones appear in front of the peripherals bearing the numbers 20 and 21. A phalanx, probably of the first digit, extends from the lower jaw, 3, to the right parietal, 5. CHISTERNON? INTERPOSITUM, new species. The single known specimen of this species was collected during the summer of 1908 by Mr. C. F. Kay, of the U. S. Geological Survey, in the Livingston coal field of Montana. The formation is the Fort Union, The more exact lo- cality is given as T. 5 S., R. 19 E. This is in Carbon County, about 10 or 15 miles west of north of Red Lodge, and on or near some of the sources of Red Lodge Creek. The catalogue number of this specimen in the U. S. National Museum is 6058. The individual is represented by parts of the anterior two- Dine Or Gakarach) Memoirs Carnegie Museum, II, No. 6, 1906, p. 236, fig. 8. ¢The Anatomy of Vertebrated Animals, 1872, p. 26, NO. 1666. OSTEOLOGY OF CAMPTOSAURUS—GILMORE. 911 all other sutures are distinctly seen, failed to show any indication of an interparietal suture. Hulke* has observed that Hypstlophodon fowii, Iguanodon mantelli, and Camptosaurus (Iguanodon) prest- wichii all have unpaired parietals. Seen from above the parietals are comparatively short, heavy bones. Their lateral surfaces, which form the upper walls of the brain case, are smooth, concave antero-posteriorly and thus constricted medially into a rounded crest, and without the sharp median sagittal ridge found in Jguanodon. Anteriorly the expanded end unites with the broad plate-like frontals by an angular suture. Laterally a pro- longation of the anterior portion of the parietal curves outward to meet the postfrontals, and with these bones form the upper anterior boundary of the supratemporal fossa. Similarly the postero-lateral border turns outward, joining the squamosal with which it bounds posteriorly the upper opening of the fossa. Ventrally it encloses the upper portion of the supraoccipital. In the skull of C. medius, No. 1880, Yale Museum, the parietal has a transverse width of 51 mm. at its middle. There is no parietal (pineal) foramen in Camptosaurus. Squamosal.—The following description of the squamosal of Camp- tosaurus, found among Professor Marsh’s unpublished notes, is based on the left element of C’. medius, No. 1880, Yale Museum. The squamosal fits very snugly on the head of the quadrate, and probably exeludes the quadrate entirely from touching the paraoccipital process as in Sphenodon. In position it is most nearly related to that of Jguwana. It has four distinct processes. The postfrontal process is very thin, flat, and arched out- ward above. The sutural surface is rather more than 14 inches in length, reaching to within a half inch of the tip of the quadrate. The head of the quadrate fits closely into a pit on the under surface. Annals and Magazine Natural History, 6th ser., XTX, p. 590. xo.1666. OSTBOLOGY OF CAMPTOSAURUS—GILMORE. 919 pressed transversely with a part on the lower border of the anterior end, which curves in and was united with the dentary of the opposite side by cartilage only (see ss, fig. 8). Hulke’s description of the an- terior part of the mandible of Wypsilophodon as having a “ spout- like symphysial end” aptly describes this region in Camptosaurus. Posterior to the symphysial surface the dentary swells somewhat transversely, but remains about the same depth throughout the den- tigerous part, the upper and lower borders being nearly parallel. The lower border is almost straight, as shown in fig. 8. 2 43 eee ee hee eee 5b Greatest width anterior extremity_____-____ wf} obs) Me eee 59 Greatest width posterior extremity______________ Jj=) eee he a 45 The third cervical—The third cervical may be distinguished by its plano-concave (platyccelian) centrum and by the fact that if the planes of the articular surfaces were produced ventrally they would intersect within a foot below the ventral surface of the centrum (see Plate 12). In the succeeding cervicals just the opposite condition is found, that is, the produced planes of their articular ends would meet dorsally. In the articulated cervicals the upward curve of the posterior cervicals changes with the third vertebra to a forward and downward direction, thus giving the neck a graceful sigmoid curve very bird-like in character. The lateral surface of the centrum is constricted transversely, but flares out posteriorly. The centrum is more regularly rounded, the sides being convex vertically, and it lacks the decided ventral keel found in the cervicals which follow, NO. 1666. OSTEOLOGY OF CAMPTOSAURUS—GILMORE., 929 There is a well-developed capitular process on the side of the centrum near the anterior end just below the neuro-central suture, and a weak tubercular process near the middle lateral surface of the neural process, posterior and below the level of the prezygapophyses. As in the axis the neural canal is large, being higher than wide, with thin walls. The postzygapophyses are a pair of slender, divergent processes which extend upward, backward, and outward, their articular faces looking downward and outward. ? In none of the specimens examined did it exist as a well-defined No. 1666. OSTEOLOGY OF CAMPTOSAURUS—GILMORE. 949 foramen, as found in other members of the Dinosauria and as indi- cated by Marsh in his illustrations of this element. The superior border presents a thin, sharp edge, while the anterior is more rounded and roughened, with a thickened portion at the antero-supero angle. Measurements of Specimen Cat. No, 4282, U.S.N.M. wim. Ereainon Fenstmmcn IOlt COYaQcgnmiL 2-2-2 ee ee eae ae 115 Greatest width of left coracoid_.__._~-~----~~-.-.-~-~--+~-------~------ 156 THE FORE LIMB. The humerus—The humerus is comparatively short, being some- what expanded at both extremities. more especially the proximal. Fic, 25.—LEFT HUMBERUS OF CAMPTOSAURUS Fic. 26.—LEFT HUMBRUS OF CAMPTOSAURUS BROWNI. Howotyrn. Cat. No. 4282, BROWNI. HOLOTYPE. Cat. No. 4282, U.S.N.M. ; 4 NAT. SIZE. FRONT VIEW. d, U.S.N.M. ; 2: NAD. SIZE. BAcK VIEW. h, DELTOID CREST. FROM A PHOTOGRAPH. HEAD. FROM A PHOTOGRAPTI. The head is situated in about the middle of the proximal end and is produced considerably backward beyond the posterior border of the shaft (see A, fig. 26). It has a smooth, subspherical, articular sur- face from which a ridge-like swelling passes down into the posterior surface of the shaft. There is a heavy radial crest developed on the antero-external border just above the middle of the shaft which renders the anterior aspect of the surface concave. The crest (see d, fig. 25) appears to occupy a lower position on the shaft than it does 250 PROCEEDINGS OF THE NATIONAL MUSEUM. VOL. XXXVI. on the humerus of Zguanodon. The shaft is somewhat twisted and planes passed through the longer axes of the articular ends would cut one another at a shght angle. The outer surface of the radial crest is roughened for muscular insertion. Below the crest the shaft is constricted and is subcircular in cross section. The internal border is regularly curved from end to end, while the external is angularly convex. The radial and ulnar condyles are well defined and posteri- orly are separated by a broad, shallow depression which spasses some- what up the shaft of the bone. There is also a shallower depression between them on the distal anterior surface of the bone. The articu- lar ends of the humerus are quite rugose in the individual here described. Measurements of Specimen, Cat. No. 4282, U.S.N.M. mm. Greatest length of left humerus_______________ a es eR i le 360 Greatest width proximal end_________ al 2 ee 116 Greatest width dorsal end__________ Zia BS fr 7d 2 The ulna—The ulna is heavy above and but little expanded on : the distal end. Its length slightly ex- ceeds that of the ra- dius, but is shorter than the humerus, differing in the lat- ter particular from Trachodon, in which the ulna exceeds the humerus in length. There is the begin- ning of a massive olacrenon (see fig. 41), but it extends but little above the articular surface for the humerus. Viewed proximally the end is subtriangular in outline. The middle of the shaft in cross section is elliptical. Fig. 27.—RIGHT RADIUS, ULNA, AND MANUS OF Camp- The distal end is TOSAURUS BROWNI. HoLoTypn. Cat. No. 4282, U.S.N.M.: | ABOUT 4 NAT. SIZE. FRONT vyinw. SHows position or SOMewhat expanded ELEMENTS AS FOUND IN THE QUARRY. i, INTERMEDIUM: ~ y Ff : Seas nee pee aoe : (see ul, fig. 27), the rad, RADIUS; U, ULNARE; ul, ULNA. FROM A PHOTOGRAPH. greatest diameter be- ing transverse. The lower internal surface is slightly roughened and flattened for contact with the grooved external surface of the distal No. 1666. OSTEOLOGY OF CAMPTOSAURUS—GILMORE. 951 part of the radius (see fig. 41). The articular ends of this bone are smoother than the corresponding parts of the humerus. Measurements of Specimen, Cat. No. 4282, U.S.N.M. 7. mnt. Greatest length of left ulna_______-____ by ea eas Ae ps Wi ie Be Wh pes gin Tas 192 GrentestawiGth proximal end). ob ulna -_ 4-29) 2 See Se ee eee ho Sees aul Cistal.GnguOf Milne sas ort ae ae ee eee a2 Radius—The radius is more slender and somewhat shorter than the ulna. It is slightly expanded at the extremities, more especially the proximal end (see ra, fig. 27). In cross section the shaft is sub- circular and continues so for most of its length. On the distal exter- nal surface of the bone is a longitudinal groove which receives the roughened border of the distal end of the ulna when articulated. Viewed from the distal end the extremity is heavy and roughly sub- crescentic in outline, with an oblique end, the surface of which looks downward and outward and fits closely to the reverse bevel of the radiale. This is clearly shown in fig. 27, which represents the fore- arm and carpus of the right fore limb of C. browni, Cat. No. 4282, U.S.N.M. Measurements of Specimen, Cat. No. 4282, U.S.N.M. Grontesmilenotheotlertetagiis ee oe) Se = ae ee oY ER Aran see 232 Groaosividhnok proxamaende 95-2 e ee ee Ss 58 ce megesimng Cul sOLeCUSta linen Gta =e a a tes ee an THE FORE FOOT. In Camptosaurus there are five digits in the manus, supported by eight carpalia. The carpus has a simple arrangement, that is, a prox- imal row of three bones—the radiale, intermedium, and ulnare—and a distal row of five. The number of phalanges, beginning with the first digit or pollex, as correctly determined by Marsh, is 2, 3, 3, 3, 2. The detailed description to follow is based upon a fore foot of Camptosaurus dispar, Cat. No. 4277, U.S.N.M., from quarry 18, near Como, Albany County, Wyoming. The foot is complete with the exception of phalanges one and two of Digit I, and two and three of Digit IV. The missing parts, with the exception of the ungual of Digit I, which is unknown, will be described from other feet in the National Museum. This foot was found articulated in the field and was so received in the laboratory, and on that account there can be no question regarding the relative position of its elements. The form and proportions of the various bones of the manus are well shown in fig. 28. The carpus.—In the carpus of the typical specimen of Campto- saurus dispar, No. 1877, Yale Museum, Marsh recognized nine carpal bones, but I am unable to detect more than eight. Specimens in the 952 PROCEEDINGS OF THE NATIONAL MUSEUM. VOL. XXXVI. National Museum also show but eight. It appears that the number will vary in fully adult individuals, due to the fusion of the smaller with the larger members. The proximal row forms two distinct concave surfaces for articu- lation with the distal ends of the radius and ulna. These articulating surfaces are clearly defined, as in many of the fore limbs of the mammalia, an unusual condition as is well known to all who have attempted to articulate the limb and foot elements of most members of the Dinosauria. The contour of these opposing articular ends is distinctly shown in fig. 27, which represents the right fore foot of C. browni, and is reproduced from a photograph of the specimen as it was found in the quarry. . The radiale (r)is an irregularly shaped, block-like bone, the most robust element of the carpus. Its proximal end forms the chief sup- port for the radius, the distal end articulates with Metacarpal I and carpalia one, two, and three, and externally with the internal bor- der of the intermedium, with which it is often coossified (see fig. 27). The union with Meta- carpal I is at an angle of nearly 45° to the main axis of the foot, and in all of the feet studied, Fig. 28.—Ricur rorn root, Camprosaurus Where these elements were pre- Se ee ee, van tear SE ved, they were. lunes pee ce, carpaL two; cf, carpan rour; ¢, firmly ankylosed, particularly cana nie; i retenromt; me Z along the anterior surfaces, as I ro V, DIGITS ONE To FIvr. Unevat will be seen in fig. 28. This digit Pr aoe gece a paeaeir) was thereby rendered immovable, which is suggestive of the “ spike-like ” digit of the allied 7guanodon. The intermedium (in) viewed from the front is subtriangular in outline, the apex pointing toward the tilnare (see i, fig. 28.) Its dorsal surface articulates with both the radius and ulna, but more especially with the latter; ventrally it is supported by Metacarpal III, and a small, irregularly shaped element on the posterior internal part of this end which probably represents C,. Although apparently in position in this foot, in the right fore foot of another specimen (Cat. No. 5473, U.S.N.M.), C, lies more over the proximal end of Metacarpal III, which somewhat confirms its identification. This element, however, is about equally divided between the intermedium and radiale. The uwlnare (w) is a cushion-like bone, its transverse being greater than the antero-posterior diameter (see w, figs. 27 and 28). It has a NO. 1666. OSTEOLOGY OF CAMPTOSAURUS—GILMORE. 9583 smooth, concave proximal end forming the chief support of the ulna. The anterior surface is convex transversely, while the posterior is concave. It articulates with the intermedium by two facets on either extremity of the internal border. The intervening border being con- cave is not in contact with that of the intermedium. There are five small elements in the distal row of the carpus. Carpale one is a small ossicle-like bone wedged in between the ends of Metacarpal I and the radiale on the posterior side. It is not visible from the front aspect. In the forefoot of specimen No. 5473, this element was in position but remained as a distinct bone. In No. 4277, however, it is fused with both the radiale and Metacarpal I and is hardly recognizable. The second carpale in No. 4277 is a small, flattened, rectangular element which was retained in position on the proximal end of Meta- carpal II. Its thin, anterior border is visible in fig. 29 between the radiale and Metacarpale II. The third carpale is a wedge-shaped element coossified about equally with the posterior part of the surfaces of the radiale and intermedium. It articulates slightly with Metacarpal III, and al- though in other specimens it is visible from a front view of the foot, in this one it is not. Carpale five, seen from the front, is a lozenge-shaped element interposed between the outer distal surface of the ulnare and the proximal end of Metacarpal V. Marsh has represented this element in the foot of C. dispar as contributing to the ulnar surface, but in the specimen here described, an outer extension excludes it from the ulnar surface. The metacarpus.—Metacarpal I is much the shortest element of the series and in adult individuals is always fused with the radiale. The inner border of the proximal end rises to the level of the dorsal surface of the radiale and contributes slightly to the articulating surface for the radius. The distal end is convex dorso-ventrally with a pronounced median depression or groove. The short shaft is angular in cross section, being wider than deep. Fore and aft on the external lateral margin of the proximal end are two antero- posteriorly elongated facets which articulate with corresponding facets on the internal margin of Metacarpal II. Metacarpal II is more than twice the length of the preceding and more slender. The proximal end is subtriangular in outline, its antero-posterior diameter being the greater. The articular sur- face of this end is comparatively smooth and gently convex antero- posteriorly. The shaft is somewhat constricted but expands again at the distal end, more especially in its transverse diameter. The external border of the distal end is produced distally below the internal, which deflects the phalanges of this digit mesially. The 954 PROCEEDINGS OF THE NATIONAL MUSEUM. VOL. XXXVI. posterior border is deeply concave transversely, being a continuation posteriorly of the median depression of this end. The proximal half of the external lateral surface is concave antero-posteriorly, forming a depression which receives a rugose convex surface on the internal lateral surface of Metacarpal III. The distal lateral sur- faces are quite smooth. Metacarpal III is the longest and most robust element of the five. Its proximal end is subtriangular in outline, being longer than wide. The shaft is constricted and at its smallest part would be subcircular in cross section. The distal end is expanded both antero-posteriorly and transversely. Like Metacarpal II this end is convex antero- posteriorly and concave transversely, although the median depres- sion is not so pronounced as in Metacarpals I and II. The external lateral surface of the proximal end is roughened and during the hfe of the animal it probably had direct cartilaginous articulation with Metacarpal IV. As mentioned above, the proximal end articu- lates principally with the intermedium and slightly with carpale three. Metacarpal IV is shorter and more slender than Metacarpal IT, the shaft being more rounded. Viewed from above, the proximal end is subtriangular and has a smooth concave end which articulates closely with the distal end of carpale four. The distal end is proportionately less expanded than the preceding metarcarpals, and unlike them shows no pronounced median depression except on the anterior border. Metacarpal V is the most slender of the series, although longer than Metacarpal I. Seen from above the proximal end is triangular and like Metacarpal IV has a smooth, transversely concave articular surface for union with carpale five. On the internal lateral surface near the proximal end is a shallow concave depression which articu- lates with an outward projection on the postero-external angle of the proximal end of Metacarpal IV, forming rather a weak union with that digit. The shaft is constricted, especially antero-posteriorly, which gives it a flattened aspect. The distal end is but little ex- panded, being convex antero-posteriorly, without median depression. Measurements of metacarpals of Specimen, Cat. No. 4277, U.S.N.M. Metacarpals. iT 1vig | Ill. IV. Va ae eS ERS SE ee = | mm. mm. mm. mm. | mm. Mengebh! 222. eke «osee ces Oe rh COE EES apse ei See 26 61 76 60 40 Breadth otmroximall ends [222 an. joa ecians tne ae eee « 33 31 35 40 38 Breadth of distal end iow. Soo ae cece eee ee ee ee eee 30 35 37 32 25 Phalanges.—The phalangial formula beginning with the first digit 19 2.350) a5 2s The proximal phalanx of the first digit is short, with lateral and dorsal surfaces quite evenly rounded and palmar surface flattened, NO. 1666. OSTEOLOGY OF CAMPTOSAURUS—GILMORE. 955 The internal side is longer than the external. The proximal articular end is concave supero-inferiorly, with a blunt median keel for articu- lation with the groove on the distal articular surface of Metacarpal I. The distal end is regularly convex supero-inferiorly and concave transversely, This description is based on the phalanx of the left forefoot of C. brown. The ungual of Digit T is missing on all of the feet studied, but in the specimen shown in fig. 29 this element _has been provisionally restored. The proximal phalanges of Digits II and III are short, stout, block-like bones. The proximal articular ends are concave supero- inferiorly, with a median concave depression. The second phalanges of the above digits are much shortened, with concave proximal and convex distal ends. The palmar surfaces are flattened and produced posteriorly into thin sheets which he under the distal ends of the proximal phalanges of their respective digits. This posterior exten- sion is especially pronounced in the second phalanx of Digit TV. Digits II and III are terminated by relatively large, pointed un- guals, which were undoubtedly sheathed in horny claws, as indicated by a pair of lateral grooves on each ungual. The ungual of Digit IT is considerably depressed; the proximal end is subtriangular in out- line with concave supero-inferior surface. The median dorsal sur- face is produced posteriorly and overhangs somewhat this articular end. The proximal end of this ungual of Digit III is higher than wide and in outline is irregularly rounded. The articular end is more deeply concave than ungual two. The anterior end is deflected toward the foot of the opposite side and is sharply pointed. The proximal phalanx of the fourth digit is much depressed, especially at the distal end, and expanded transversely at the proximal end. The articular surface of this end is concave supero-inferiorly with only a faint indication of a keel on the superior part of the median sur- face. The palmar surface is flattened. The transverse extent of the distal end is much less than the proximal. It is convex supero-in- feriorly with a suggestion of a concave median groove. The second phalanx of this digit is very short and supports a small, rounded, hemispherical, terminal phalanx which in life was probably embedded within the integument of the foot. The proximal end of the latter is indicated by a concave surface. The proximal phalanx of Digit V is the most slender of the series. The upper articular end is about evenly concave in both transverse and vertical directions and tapers without noticeable constriction to the distal end, which is convex supero-inferiorly. The terminal phalanx of this digit is an irregular button-like ossicle. From the above description and figures of the manus of Campto- saurus it will be seen that the weight of the body was borne prin- cipally on the three median digits, and that through disuse the fifth 956 PROCEEDINGS OF THE NATIONAL MUSEUM. VOL. XXXVI. was becoming atrophied. While this has not yet resulted in the elimination of any of the digits, yet the fifth is fast approaching a functionless condition. Measufements of phalanges of Specimen, Cat. No. 4277, U.S.N.AM, Nl % Digits. ere | II. | Ill. | IV. | Vv. | | ae lees > eee ae | mm. | mim. mm. mm. | mm. Greatest length first row of phalanges..............---------- | a21) 26 | 34 26 | 23 Greatest length second row of phalanges............--...----|---+-+--- 24 26) @16| 6 Greatest length third row of phalanges...............- Sa era b 48 B00 200) see “Measurements of elements of left forefoot of No. 4282, is a slightly smaller individual. » Estimated, elements incomplete. THE PELVIS. The pelvis of Camptosaurus, as shown in Plates 15 and 16, is quite characteristic of the genus. It is composed of the three elements, the ilium, ischium, and pubis, usually found in the pelvis of the Krieg. 29.—LErt ILIUM OF CAMPTOSAURUS DISPAR MARSH. Cat. No. 5473, U.S.N.M.; § NAT. SIZE. EXTERNAL VIEW. END OF PUBIC PEDUNCLE MISSING. FROM A PHOTOGRAPH. Dinosauria. The long, slender post-pubis, reaching to the end of the equally long shaft of the ischium, is distinctive of this group. With the exception of the pubis, the parts described below are of specimen Cat. No. 4282, U.S.N.M. The ilium.—The ilium is an elongated plate-like bone, resembling in most of its characteristics the corresponding element of 7guanodon. The preacetabular process is long, narrow, and compressed laterally. The ilia of this specimen lack the anterior termination, but in C. dispar, as shown by the ilium of Cat. No. 5473, U.S.N.M., the end is somewhat spatulate (see fig. 29). This long, slender process extends forward and outward, overhanging the posterior ribs. The superior border of the ilium is thickened and rounded transversely, and at the posterior end descends for a distance of 90 mm., at an oblique angle, gradually thickening to the point where it reaches the infero-internal plate. Viewed laterally the ilium has a width of 87 mm. from where the superior border begins to descend to the border where the inner ~ NO. 1666. OSTEOLOGY OF CAMPTOSAURUS—GILMORE. 957 shelf is given off. The acetabular arch is bounded anteriorly by a rather wide but slender pubic peduncle directed downward and for- ward. The posterior peduncle for the ischium is heavy, broadly swelled, with two roughened surfaces, more especially the posterior, which meet at an obtuse angle. The posterior and more inferior one, which is somewhat cupped, is for the articulation of the ischium. The roughened anterior surface appears to be for the attachment of the heavy pads of cartilage and ligaments which bound the pelvic bones together. Posterior to this process the inferior border is a sharp edge, which, as it extends backward gradually turns from a vertical to a nearly horizontal position, with a rounded, thickened, posterior termination. The upper half of the internal surface is smooth and gently concave from end to end. On the lower median internal surface are roughened cupped depressions, their greatest extent being vertical, for the articulation of the sacral ribs (see w, Plate 15). These surfaces are separated by narrow, smooth, con- cave, nonarticular tracts. A shallow, roughened longitudinal groove Fic. 30.—LEFT ILIUM OF CAMPTOSAURUS BROWNI. Hotorypr. Cat. No. 4282, U.S.N.M.; @ NAT. SIZE. EXTERNAL VIEW. ANTERIOR END OF PREACETABULAR PROCESS RESTOREL AFTER THAT OF THE LEFT ILIUM OF NO. 5473 (SEE Fic. 29). FROM A PHOTOGRAPH. on the thickened internal edge of the shelf-like projection mentioned above, may be for the reception of the transverse process of the fifth sacral. The greatest vertical width of the ilium is over the ischiac peduncle. The ilium is one of the most characteristic bones of the skeleton, and while the description given above shows all of the essential char- acters of the Camptosaurus ilium, it is found that in other species there are great differences of contour and proportion. Measurements of Specimen, Cat. No. 4282, U.S.N.M. mm, Greatest length of left ilium from posterior end to center of preacetabu- LOTR TESS] Ne see oe es ee 5 ane ee eee Eee OO AE Ce AREA Te SAP) OPS) aN 392 Creates width=over: isehial process... 2. oe a 75 Greatest width from middle of acetabulum __--.2. 2-5-2 115 Sere asIne Of ACCLAPUIRIN. 45 oo 2s Cy eS Ea 161 The ischium.—The ischium is larger than the pubis and distally consists of a slender rod-like shaft which curves downward with an Proc. N. M. vol. xxxvi—09——-17 958 PROCEEDINGS OF THE NATIONAL MUSEUM. VOL. XXXVI. expanded hammer-like end (see 1a, fig..31). The distal ends appear to have been in contact as well as the borders of the lower half of their shafts. In C. browni this bone is hghtly constructed, in this respect approaching the ischia of C. medius. The proximal end has a widely expanded Y-shaped extremity with two distinct articular faces (see 7s, Plate 16), the one for the posterior peduncle of the ilium being the heavier. Its articular surface is rugose and deeply cupped. The articular surface for contact with the pubis is borne on a thin, quad- rilateral plate which extends down- ward and forward from the main ), °\ shaft of the bone, and its upper antero- ZP posterior concave border completes the acetabular boundary. The straight, truncated end of this plate is flat and thickened transversely, and when in position abuts against an opposing articular face on the posterior end of the pubis at about the middle line of the acetabulum. The lower anterior border is also thickened and is in con- tact with a depressed articular sur- face on top of the postpubis, just pos- terior to the pubic foramen. Behind, the expanded head contracts rapidly, but again widens on the inferior border la into a thin, lip-lhke, downturned, ob- turator process, against which the rod- like shaft of the postpubis rests (see Mii Plate 16). Below this process the Fie. 31.—Iscuia or Camptosaurcs shaft contracts and continues as a MEDIUS MARSH. SUPERIOR VIEW. 4 Yan Musrum: 4 nav. sizm. #t, curved, rod-like shaft to the distal ex- ILIAC SURFACH; Ja, DISTAL ENDS OF tremity. The position of the obtura- SAME. AFTER MARSH. . E tor process, well up toward the articu- lar end, at once distinguishes the ischium of Camptosaurus from Hyosilophodon fowiti, which has this process about midway between the two ends. Measurements of Specimen, Cat. No. 4282, U.S.N.M. Greatest leugthiof, left ischium=22_ = ==" ss 2 eee 545 Greatest width of proximal end] = = eee ao eae 194 Greatest width.of distal end 22) 2 ee = ee ee eee 79 The pubis.—The description to follow is based on the pubis of C. dispar, No. 1878, Yale Museum, the postpubis from C. medius, No. 1880, Yale Museum (see p’, Plate 16). As shown in fig. 82, the pubis in Camptosaurus is composed of a flattened prepubic element and an elongated, curved, slender post- NO. 1666. OSTEOLOGY OF CAMPTOSAURUS—GILMORE. 959 pubis. The anterior portion is a thin, flat, vertical blade of bone, when articulated extends forward, downward, and slightly outward. The superior border is gently concave antero-posteriorly, and near the proximal end expands transversely into a triangular, roughened surface for contact with the pubic peduncle of the ium. The proximal end is expanded transversely, forming a heavy, concave end which bounds part of the anterior and inferior borders of the acetabulum. The lower part of the proximal end has a rugose surface for con- tact with the antero-infero projection of the proximal end of the ischium. On the lower internal border of the proximal end the post- pubic bar is developed, which extends backward and downward be- neath and parallel with the ischium, reaching to the end of that bone. Between the anterior end of the postpubis and the posterior acetabu- lar surface of the prepubis is a large pubic foramen, which, in some individuals at least, does not appear to have been entirely closed posteriorly. While the notch.is closed in three individuals studied, the union of the two surfaces is not to coalescence, the suture in all instances being visible. Posterior to the up- ward projection of the superior surface of the postpubis is a shallow, roughened — depression which was in contact S B Fic. 32.—LEFT PUBIS OF CAMPTOSAURUS DISPAR MARSH. with the lower anterior PARATYPE. No. 1878, YALE MuSHUM; 7s NAT. SIZB. prolongation of tehe noe VIEW. p, PUBIS; p’, POSTPUBIS. AFTER TARSH. ischium. A cross sec- tion at the broken end of the postpubis of Cat. No. 4282, U.S.N.M., shows it to be subcircular in outline. The distal end is slightly ex- panded, especially in the ventral direction. In No. 1880, Yale Mu- seum, holotype of C. medius, in which this element is complete, the postpubis is 480 mm. in length. Its distal end is 23 mm. wide in the vertical direction. The anterior ends of the pubes do not meet medially. THE HIND LIMB. The hind limb and foot in Camptosaurus is about twice the length of the fore limb and foot and much more robust. The tibia is slightly shorter than the femur. There are four ossified elements in the tarsus, the caleaneum and astragulus being distinct. The femur.—Unfortunately the femora are lacking in Cat. No. 4282, U.S.N.M.,.but in a second, larger individual both are present, and two more, representing as many individuals, were found in the collection. All are remarkably free from distortion, and in a fine state of preservation. The description to follow is based upon the femora of No. 5818 (see fig. 33). 260 PROCEEDINGS OF THE NATIONAL MUSEUM. VOL. XXXVI. The femur is the most robust bone of the skeleton. It has a curved shaft, and is compactly and strongly built throughout. The forward arctuation of the shaft distinguishes this bone from all other known American predentate dinosaurs excepting Laosaurus and Dryosau- rus. A compressed inner or fourth trochanter (0), fig. 33) of the pendant type is developed on the postero-internal margin of the shaft. Dollo has called attention to this type of trochanter in /guanodon as indicating a powerful caudo-femoral musculature, as in some birds. In Camp- tosaurus, however, it appears more prom- inently developed than is found in any of the European representatives of this group. The fourth trochanter begins to develop somewhat above the middle of the shaft, the apex being directed downward and in- ward toward the distal end of the bone. Just anterior_to the trochanter is a shallow vertical depression with a markedly rugose surface which extends out on the internal surface of the trochanter. The head (A, fig. 33) is well developed and subglobular in form, and is attached by a short, thick neck at nearly a right angle to the main axis of the shaft. The articular surface of the head is somewhat rugose, and this rugosity is continued along the superior surface of the greater trochanter to its external border. A prominent lesser tro- chanter (a, fig. 33) rises on the antero- external surface of the upper part of the shaft, nearly to the height of the greater gts 33.—RIGHT FEMUR OF trochanter, as a transversely compressed AMPTOSAURUS DISPAR 3 Marsu. Car. No. 5818, blade. Posteriorly the lesser trochanter VSNMG NAT Sue js separated from the shaft by a deep, NNE HW. da, LESSER : : TROCHANTER; b, INNER TRo- Narrow cleft. Behind the head is a pro- oe ae cuis. how «) DOMBced: groove separated from a second PHOTOGRAPH. concave depression on the posterior median surface of the shaft by a heavy, rounded, longitudinal swelling. The superior surface! of the greater trochanter (2, fig. 42) is wider and more gently rounded antero-posteriorly than the head. The distal end has the usual condyle shape, though the inner condyle (ce, fig. 33) is much more robust than the outer. The two condyles are separated by a deep, intercondylar groove, wider at the bottom than the top. Both condyles project decidedly backward. The anterior inter- condylar groove (2, fig. 42) is wide and of moderate depth, con- Ss me Se No. 1666. OSTEOLOGY OF CAMPTOSAURUS—GILMORE. °61 trasting strongly with the deep, narrow, almost “ tunnel-shaped,” anterior-intercondylar groove in 7’rachodon. The shaft of the femur appears to be somewhat twisted, due to the alteration of the aspect of its surfaces, that which at the proximal end is external becoming at the distal end anterior. There is a small roughened area on the extero-posterior angle of the distal portion of the shaft for the attachment of muscles, and just above the beginning of the inter- condylar depression or groove on the anterior surface is a second roughened patch. The outer surface of the outer condyle is decidedly concave, forming a wide, longitudinal depression at the distal end, in which a large tendon must have passed. Measurements of Specimen, Cat. No. 5818, U.S.N.M. mm. Peston cil kOLarloni torent se ees eee See oe) be 2 te eee 592 Greatest diameter of proximal end_-___-_—~--__-_—- gees te IC ee 190 eeeniuer Wiameter of Gisinl end. 2-226 3 ee 178 The tibia—The tibia in Camptosaurus is shorter than the femur, in which respect it may be distinguished from Laosaurus, Dryosaurus, and Hypsilophodon fou. It is constricted medially, but greatly ex- panded at either extremity, the longer axes of the expanded ends being at nearly right angles to one another. The proximal end of the tibia shows a division of the articular surface into two condyles, which project posteriorly, separated by an inter- condylar groove, the internal one (c¢, fig. 34) being the heavier. A large enemial crest (6, fig. 34) projects outward from the upper end of the shaft in front of the external condyle. The distal end is divided into two mal- leoli, of which the inner is the shorter and heavier, its articular surface looking down- ward and forward, while the external is longer and thinner and looks directly downward. These are separated on the front surface of the bone by a shallow Fic. 34—Ricur rrr ayp : a & Sih ASTRAGULUS OF CAMPTO- groove. As in 7'riceratops, the outer mal Se ne Me se leolus falls below the superior border of Cat. No. 5818, U.S.N.M. ; . é NAT. SIzb. INNER VIEW. the astragulus. Its flattened anterior sur- G, ASTRAGULUS ; b, BNEMIAL face was in contact with the distal ex- ee ee Lom TE. 4 : FROM A PHOTOGRAPH. tremity of the fibula. The posterior sur- face of the distal end of the tibia is angularly convex transversely. 962 PROCEEDINGS OF THE NATIONAL MUSEUM. VOL. XXXVI. Measurements of Specimen, Cat. No. 5818, U.S.N.M. mm, Greatest length: of right tibiae *= see Se ee ee ee 5381 Greatest diameter of proximalvend==2." 5. | 2s ee eee 222 Greatest diameter, of idistal ends = 2 a a ee ee 90 The fibula—The fibula of No. 1877, Yale Museum, is a slender bone, being somewhat shorter than the tibia. It has a flattened shaft above, which is subcylindrical below, with flattened expanded ex- tremities. The face of the proximal end which was applied to the tibia is concave antero-posteriorly, while the outer surface is convex. The articular surface of this end is nearly straight antero-posteriorly, but roughened and gently rounded transversely. The expansion of the upper end takes place more especially toward the anterior border which overhangs considerably the constricted shaft. As with the tibia, planes passed through the longer axes of the expanded articular ends would cut one another at nearly right angles. Internally the distal extremity is plain and was closely applied to the opposing flattened surface of the outer anterior face of the tibia. The articular end is roughened and concave and viewed from below is semicircular in outline. The shaft has a small medullary cavity. The distal end articulates closely with the upper surface of the calcaneum. THE HIND FOOT, The astragulus—The astragulus, although closely applied, was not ankylosed to the tibia, even in adult individuals. Its upper surface is the counterpart of the inner articular surface of the tibia. The upper surface (see a, fig. 34) is deeply concave from front to back and divided by a ridge, which marks out two portions cor- responding to the inner and part of the outer tibial malleolus. These two surfaces meet one another at an obtuse angle. The distal sur- face is convex antero-posteriorly, with a broad, shallow depression medially, making this surface slightly concave transversely. The- anterior margin is a thickened lip. The posterior margin is stout, and viewed from above, terminates in a heavy, angular, posterior point. The inner end is comparatively thin and probably non- articular, although a prominently developed projecting knob on the antero-internal angle appears to have been closely apposed to the distal inner surface of the fibula. The anterior half of the external depression of the dorsal surface is nonarticular and did not come in contact with the tibia, but formed the walls of an open- ing leading down between this bone and the os calcis. The outward knob-like projection overlaps the outer surface on the distal part of the tibia. On the anterior internal surface, just below the supe- rior border, is an elongated pit. There is no ascending process on the astragulus of Camptosaurus. No. 1666. OSTEOLOGY OF CAMPTOSAURUS—GILMORE. 263 Measurements of Specimen, Cat. No. 5818, U.S.N.M. STOMLCShait ARS VELSG (GLA ING LOI ane et ok ae et eS) eee Ee ee eS ee? 139 I TEnMLeStManLeLO-DOSLCLIOL. .CismMetenr a2 sss bs ai tae ere eae ati 95 The calcaneum.—Seen from the outside, the caleaneum is subcres- centic in outline. The upper surface is divided by a diagonal ridge into two articular faces. The anterior one, which is above the heavier end of the bone, receives the distal end of the fibula. The posterior articular portion has its surface below the level of the fibular surface and is deeply concave antero-posteriorly, and receives the outer part of the tibial malleolus. This portion has the greatest transverse diameter. The posterior margin is rounded transversely and turns up as a thickened rounded hp. ‘The ventral surface is convex from front to back, the inner border being in contact with the outer edge of the astragulus. The outer surface vertically is gently concave. The caleaneum is represented in two individuals in the National Museum, Cat. Nos. 5473 and 5961, also in the holotype of C. dispar, No. 1877, Yale Museum, which, being the most complete element, furnished the following measurements: Measurements of Specimen, No. 1877, Yale Museum. Greatest extent antero-posterlorly of caleaneum_______~___ === ST Greinlesthextent transversely: of -caleamenmess. 2" 2 is See 5D Distal tarsals—The distal row of tarsals (see ¢, fig. 35) in Camp- tosaurus consists of two flattened, cushion-like bones, which remain - distinct, never fused with the metatarsals, as Hulke has suggested in Zguanodon, and whose separateness would be restricted to the embryo. The external one, viewed from above; is subtriangular in outline. Its proximal surface is shallowly cupped, the narrowed portion being directed backward. The distal articulating surface is concave antero-posteriorly and fits closely to the proximal end of Metatarsal IV, as shown in fig. 35. There are deep longitudinal pit- like depressions on the anterior and internal surfaces. Both proxi- mal and distal articulating surfaces are smooth. The tarsal borne by Metatarsal III is an irregular elliptical shaped bone, somewhat thinner than the outer. Its distal articulating surface and all of the surrounding edges are roughened and pitted. This surface is shghtly convex, while the proximal surface is gently concave and quite smooth except near the edges. In specimen, Cat. No. 4277, U.S.N.M., these elements have been retained by the matrix in their mutual relationship with the metatarsals, so there can be no question concerning their exact position. Even without this positive evidence, the conformation of their surfaces with the proximal ends of the metatarsals would have enabled one to place them accurately. 264 PROCEEDINGS OF THE NATIONAL MUSEUM. — you. xxxvt. Measurements of Specimen, Cat. No. 4277, U.S.N.M. mm. Greatest transverse diameter. of outer tarsal-—=_—2 =) 2 eee 52 Greatest antero-posterior diameter of outer tarsal___________ 2230 Se 72 Greatest transverse diameter: of Inner tarsal-= 29202 eee 64 Jreatest antero-posterior diameter of inner tarsal_______________________ 92 The metatarsals —There were three functional digits in the hind foot, the first being rudimentary and the fifth wanting. The meta- tarsals are much longer and heavier than the metacarpals, the third being the longest. The proximal ends of the second, third, and fourth metatarsals are in the closest mutual apposition, their shafts being closely applied for about one- half of their length. The third and fourth support the two cushion-like tarsalia. The metatarsal of the first digit is a short, splint-like, ir- regularly curved bone, with a smooth, rounded proximal end compressed _ laterally. The thickened rugose distal articular end looks out obliquely from the main axis of the shaft, and sup- Fic. 35.—RIGHT HIND FOOT, CAMPTOSAURUS' DISPAR MarsH. Cat. No. 4277, U.S.N.M.; 4 NAT. SIzE. SHEN ports two phalanges. FROM THE FRONT: t, TWO TARSAL BONES OF THE DISTAL f row ; J, II, III, IV, FIRST TO FOURTH DIGITS. UNGUALS It appears quite 32 OF DicIts J, III, AND IV DRAWN FROM THE FEET OF Marsh has described OTHER INDIVIDUALS. it as being rudi- mentary and probably did not reach the ground. It articulates loosely with Metatarsal II, lying in a broad, shallow, longitudinal depres- sion on the inner proximal half of this metatarsal. The second metatarsal is slightly longer than the fourth, the third being longest. The proximal end is compressed transversely, but is much lengthened antero-posteriorly. The face which is applied to the third is plane, while the outer surface is irregularly concave. The articular end is gently convex antero-posteriorly, and has a roughly pitted surface. Just below the middle of the shaft on the antero- no. 1666. OSTEOLOGY OF CAMPTOSAURUS—GILMORE. 965 external border a thin lip-like process is developed which laps over and is closely applied to a roughened surface on the antero-internal border of the shaft of Metatarsal III. At this point the shaft is bent outward, which throws the distal part away from the median bone. The distal end is stout, and subquadrilateral in outline, ex- cept that it is bisected by a deep, rounded notch on the posterior border. This end is rounded antero-posteriorly. The pits for the attachment of lateral ligaments are large and fairly deep. The proximal end of the third metatarsal is roughly triangular in outline, the thickened portion being in front. The articular surface is roughened and slightly concave for the reception of the flattened tarsale of the distal row. The inner surface of the proximal half is plane, but below, this border of the shaft is rounded. Viewed anteriorly the bone remains about the same width, being slightly ex- panded at the distal end. Viewed laterally, however, the shaft con- tracts rapidly from the proximal toward the distal pulley-like articu- lar end. The external surface of the proximal end has two oblique faces, which are opposed by the excavated internal surface of the fourth metatarsal. The pits for the attachment of lateral ligaments are large but shallow. Seen from the proximal end the fourth metatarsal is subtriangular, the apex being directed outward. The slightly roughened articular surface in front is concave and receives the distal tarsale closely. The internal surface of the upper half is roughened and fits closely to the upper surface of the third metatarsal. The shaft is com- pressed antero-posteriorly and is wider than thick. As in the second metatarsal the lower half diverges outwardly and is free from the median or third metatarsal. The distal end is expanded, more espe- cially antero-posteriorly than laterally. The distal articular end is rounded and roughened on the posterior but smooth on the anterior half. The pits for the lateral ligaments are large, the internal being shallow, the external very deep. The numbers of the phalanges, beginning with the first, are 2, 3, 4, 5, and correspond, as Hulke @ has pointed out, to the first, second, third, and fourth toes in the foot of existing lizards and birds. Hypsilophodon fowit has the same for- mula. The proximal phalanges are rather long, with pulley-shaped distal and concave proximal ends. The phalanges have their articu- lating ends closely applied to one another. A median rounded verti- cal ridge on the proximal end fits into a corresponding depression on the proximal end of another, thus forming a strong union which would allow but little lateral motion. There are deep, well-defined lateral pits for the attachment of ligaments on most of the phalanges. The second, third, and fourth phalanges of Digit IV are considerably more shortened than the other phalanges. The ungual phalanges ¢Phil. Trans. Roy. Soc. London, CLXXIII, 1882, p. 1053. 266 PROCEEDINGS OF THE NATIONAL MUSEUM. vou. xxxvt. are long, pointed, and slightly curved longitudinally and laterally, being deflected inward. They are somewhat depressed, except the ungual of Digit I, which is higher than wide, and more sharply pointed. They were undoubtedly incased in compressed pointed claws, as indicated by a pair of lateral grooves on each ungual (see fig. 35). The ungual of the second digit is the more robust of the series. . Measurements of the right hind foot of Specimen, Cat. No. 4277, U.S.N.M. Digits. at Il. II. IV. mm. mm. mm. mm. Greatestlenrtinoiimetatamsal sic. cms nas alco cteteele's ele ie eater iene 133 212 234 202 Greatest antero-posterior diameter proximal end of metatarsals ...--- 15 118 106 0a Greatest transverse diameter of proximal end of metatarsals.......-.- 6 55 66 81 Greatest transverse diameter of distal end of metatarsals...........-- 26 64 76 45 Greatestilensth"istirow phalanges.\...----s-euesesc esse cece pee 58 95 85 63 Greatest-length2d' row phalanges <2 5.- 4. cee es cme een ecce ce =e a5g 69 53 39 Greasestilensthi3dirow phalansesescececeseseoes soecn- sense ences s|osmenents 94 43 30 Greatestilength 4th row phalanges: 2-8-2 cs ge sts eae ncewigeo cle tae ani Gee ae seem ieee a9g0 29 Greatest length 5th row: phalanges) 2) se eee ae a see tan ee ne ce ey See ra sass eran) sree ere ee asl “Measurements from other individuals of about same proportions as No. 4277. THE GENUS CAMPTOSAURUS. In a paper published in December, 1879,¢ Prof. O. C. Marsh pro- posed the genus Camptonotus, and at the same time described two species, C'. dispar and C. amplus, both from the Morrison beds (Atlan- tosaurus beds of Marsh) of the Upper Jurassic. In the same article he proposed the family Laosauride to include the two genera Lao- sourus and Camptonotus, In 1881,’ without comment, he proposed the family Camptonotide, under which the following genera were included: Camptonotus, Laosaurus, Nanosaurus, and Diracodon. In 1882 °¢ he briefly defined the family, including in it the European genus Hypsilophodon, Diracodon being removed to the Stegosauride. The name Camptosaurus was proposed by Marsh in 1885 @ to re- place Camptonotus, preoccupied. Dollo,’ in 1888, gave an emended definition of the family, using the older term Camptonotide, under which he placed the two genera Camptonotus and Hypsilophodon. To distinguish the former from the latter, “ Camptonotus” is defined as follows: “Two phalanges in manus-digit-V. Preacetabular process of the ilium slight. No rudiment of pes-digit-V.” In 1889, Lydekker’ referred three species to the genus Camp- tosaurus, C. leedsi, C. valdensis, and C. prestwichii, Seely’s genus @ Amer. Journ. Sci., XVIII, 1879, pp. 501-503. bIdem, XXI, 1881, p. 423. ¢Idem, XXIII, 1882, p. 84. @Tdem, XXIX, 1885, p. 169. € Compt. rend. Acad. Paris, CVI, 1888, pp. 775-777. f Quart. Journ. Geol. Soc. London, XLV, 1889, pp. 47, 48. NO. 1666. OSTHOLOGY OF CAMPTOSAURUS—GILMORE. 967 Cumnoria being considered a synonym. In 1890 he characterized the genus as follows: Teeth simpler than in the typical group of Jguanodon. Cervical vertebrze opisthocoelous; sacrals flattened inferiorly and not anchylosed ; manus with five normal digits. Lium typically deep with short and pointed pre- and postace- tabular processes, the latter having a distinct ventral plate; pubis relatively stout and long as ischium. Femur slightly longer than tibia, with curved shaft and pendant inner trochanter; typically four functional digits in pes. In 1894," Marsh published a restoration of the skeleton of Camp- - tosaurus dispar, and additional characters relating to the osteological structure were noted. Four months later,’ he briefly described the two species, C. medius and C. nanus, and at the same time character- ized the genus as follows: Premaxillaries edentulous with horny beak. Teeth large, irregular, and few in number. ’ Amer. Journ. Sci., XLIV, 1892, pl. v. ¢Tdem, XLVIII, 1894, pl. v. NO. 1666. OSTEHOLOGY OF CAMPTOSAURUS—GILMORE. 273 Detailed description—The skull and jaws of the holotype are un- known, and the complete neck with the first dorsal is all that is preserved of the vertebral column of this individual. The cervicals, posterior to the third, are all opisthocoelous, the second and third, as in the other species of the genus, being plano-concave. The median ventral keel is wider and heavier, and the centra are not so deeply excavated laterally as in C. browni. The cervical region of the holotype of C. medius being still inclosed by the matrix, direct comparisons with that species could not be made. . The first dorsal, as in C. browni, is opisthocoelous and shows the same transition of the parapophysis from the anterior end of the centrum in the ninth cervical to a position well up on the side of the neural arch on this vertebra. It is the first vertebra of the series to bear a true spinous process—a short spine 11 mm. high, slightly thickened and roughened at its upper extremity. The neuro-central sutures are plainly shown on all of the vertebra preserved. The sacrum, No. 1877a, Yale University Museum (see Plate 13), although considered by Marsh as pertaining to the holotype, in all probability belongs to another individual. This is shown by a com- parison of the measurements of the cervical and sacral vertebrae with the homologous parts of Cat. No. 4282, U.S.N.M., the proper associa- tion of the different elements of this skeleton being unquestionable. Referring to the table of measurements on page 248, it will be seen that the cervicals are smaller and the sacrals larger than those of No. 4282. On account of the difference in proportions noted, I am inclined to believe the sacrum represents a distinct individual, and should, therefore, as it was included in the material first de- scribed, be considered a paratype, to which the catalogue number 1877a has been given to distinguish it from the holotype, No. 1877. In the sacral region of No. 1877a there are five vertebra united by suture, of which the posterior four represent true sacrals, or those which give support to sacral ribs. This interpretation excludes the anterior vertebra of the series which may be regarded as sacro-dorsal (a, Plate 13), as shown by the weak diapophyses which carried a single-headed rib. Marsh has described the peculiar peg-and-notch articulation (see 3, fig. 837) of the centra of this region and makes ° it an important feature of the genus, but after a study of several sacra I am inclined to believe too much stress has been laid on this character, as it appears to be a variable one. For example, in the sacrum of C. browni this peculiarity is only weakly developed be- tween S, and S, (see fig. 17), and in (C’. nanus there is only the suggestion of such an articulation. In the sacrals under discussion this articulation is faintly shown between S, and §,, more pro- nounced between S, and §,, and strongly developed between S, and Proc. N. M. vol. xxxvi—09—18 974 PROCEEDINGS OF THE NATIONAL MUSEUM. VOL. XXXVI. S, (see fig. 87). The ventral surfaces of all the centra are flattened, with a slight, median, longitudinal depression, which at once dis- tinguishes them from the keeled or hzmally rounded sacral centra of C. browni. The ventral surface of the sacro-dorsal is regularly and evenly rounded and is without the decided keel present in that of C. browni. The two spinous processes preserved (see Plate 13) rise as thick- ened plate-like spines, the upper termination being thickened trans- versely, especially on the anterior part of this end. The spine of the sacro-dorsal is heavier anteriorly than the spine of sacral one, and probably indicates the culmination in the development of the spinous processes. The border below the thickened termination is compressed and presents a sharp anterior edge. The posterior bor- der is somewhat thickened and has a shallow, vertical groove which may have received the sharp anterior edge of the spine posterior to \ jim fF / 1s uh aw a — is Mi) _ iid) Fig. 37.—(1) LATERAL AND (2) FRONT VINWS OF FIRST SACRAL CENTRUM CAMPTOSAURUS DISPAR MARSH. No. 1877a, YALE MUSEUM. PARATYPE. % NAT. SIZE.. @, ANTERIOR FACE FOR ARTICULATION OF FIRST SACRAL RIB; 0, POSTERIOR ARTICULATING SURFACE FOR SECOND SACRAL RIB. (3) POSTERIOR SACRAL VERTEBRH OF SAME. 3 NAT. SIZE. SHOW- ING PHG-AND-NOTCH ARTICULATION; TOP VIEW: d@, ANTERIOR END; p, POSTERIOR END. AFTER MARSH. “it, which would give the plate-like appearance of the three median sacral spines, as shown by Marsh in his first restoraton of C. dispar. In No. 1878, Yale Museum, the ilium, pubis, and ischium are pre- served, the two former elements being incomplete, that is, the ilium lacks most of the preacetabular and postacetabular processes, and the pubis most of the post-pubis, which, as shown in Plate 15, repro- duced from a photograph of the specimens, has been restored in plaster. In the pelvic bones are found the chief differences which serve to separate this species from the other members of the genus. The ilium, in proportion to its length, is considerably deeper and more robust than in any of the other species (compare figs. 29 and 30). The preacetabular notch is wide and the acetabular notch deep. The oblique border of the supero-posterior end is much longer—in most individuals a third longer than in C. browni—and terminates pos- teriorly as a thickened angular end. NO. 1666. OSTEOLOGY OF CAMPTOSAURUS—GILMORE. 975 The preacetabular process of this particular individual is un- known, and in the first figure of this element it was represented as being a very short, blunt process (see fig. 36). As previously stated, the first figures given are now known to be erroneous. The shaft of the ischium of (. dispar is more robust, and at its dis- tal termination has a larger hammer-like development of this end than is found in any of the known species (compare 7s, Plates 15 and 16), the greatest diameter being 106 mm. The pubis does not show any especial differences, while the post-pubis is almost wholly restored. The limb and foot bones do not exhibit any especial characters except a more robust development than is found in the smaller spe- cies. It is at once distinguished from (. amplus by the greater size and the compressed, sauropod-like ungual of Digit I of the latter, as compared with the rounded, claw-like element of (. dispar (com- pare fig. 35 with fig. 38 and Plate 17). Of the pes of No. 1877 I need mention only the extreme shortness of Metacarpal I as figured by Marsh. When compared with com- plete elements it appears that the upper third is missing, which would account for the extreme brevity of this digit as originally figured. In the same cut Digit III is represented as bearing the heaviest ungual, when, as shown by the pes of Cat. No. 4277,U.S.N.M. (see fig. 35), Digit II carries the most robust terminal phalanx of the hind foot. A cast of the foot in question now before me shows an ungual attached to the second digit whose small articular surface in- dicates at once that it has been wrongly placed, and in all probability pertains to Digit IV. The manus, if correctly associated with the pes just discussed as pertaining to the same individual, shows variation from the same elements in Cat. No. 5473, U.S.N.M., being much lighter in construc- tion, the hind feet having about the same dimensions. The associa- tion of the fore and hind feet of this latter individual is undoubtedly correct. The fifth metacarpal is unusually short. The ungual of Digit TIT is very small, weak, and sharply pointed. Carpale five, as shown in the earlier figures of the manus of (. dispar, contributes to the articu- lating surface for the ulna and is so shown in the cast of the right fore foot. The unusual position of this element would hardly point to the true interpretation of the proper articulation of these bones, and it has been depicted otherwise, as shown in fig. 28. In the right foot I can only recognize seven carpal elements, although the eighth is probably fused with the radiale and Metacarpal I, and has thus lost its identity. Carpale three is almost wholly articulated with the proximal end of Metacarpal III. It is wedge-shaped, the thick- ened part being posterior. Carpale two occupies a posterior position 276 PROCEEDINGS OF THE NATIONAL MUSEUM. VOL. XXXVI. on Metacarpal II and is not visible from the front view as in the foot of Cat. No. 4277, U.S.N.M. (see fig. 28). In other respects the foot does not differ materially from the description already given in that part of the present paper devoted to the osteology of Campto- SQUIUs. The restoration (see Plate 18) of Camptosaurus dispar published by Marsh ¢ was based on the typical specimens, and while it depicts well the general appearance of the animal, it is now known to be erroneous in the following particulars: The thoracic region is too long by at least three, and possibly five, vertebrae; all of the presa- cral vertebre are rib-bearing, and thus there are no true lumbars; the anterior caudal vertebrae, as shown in the restoration, have the arches and transverse processes too high above the centra, the latter, as shown in fig. 18, being below the level of the pre- and postzygapo- physes. The spines should also be more inclined backward and de- crease more rapidly in height posteriorly; the transverse processes are continued too far posteriorly, as shown by two specimens in the National Museum, where they end on either the twelfth or thirteenth from the sacrum. Other minor corrections have been touched upon in the previous pages, so need no mention here. As shown by the skeleton, C. dispar is an animal of quite robust proportions, only exceeded in size by C. amplus. Marsh estimated its length as being 20 feet. After a review of the typical specimens as compared with the other species of the genus, C. dispar may now be distinguished by the fol- lowing characters: Specific characters.—Typically of large size. Cervical centra with heavy keel. Four sacrals with peg-and-notch articulation. Sacral centra flattened inferiorly. Sacro-dorsal without ventral keel. Ilium deep, and heavy in its proportions. Ischium stout, with greatly en- larged distal extremity. Ungual of Digit I of pes rounded and pointed. CAMPTOSAURUS AMPLUS Marsh. Camptonotus amplus Marsu, Amer. Journ. Sci. (3), XVIII, 1879, p. 503. Camptosaurus amplus Marsu, Amer. Journ. Sci. (8), XXIX, 1885, p. 169; 16th Ann. Rep. U. S. Geol. Surv., 1894-95, Pt. 1, 1896, p. 196. Camptosaurus amplus Norcsa, Foéldtani Kozloény, Budapest, XX XI, 1901, Ds 210! Camptosaurus amplus Hay, Bull. No. 179, U. S. Geol. Surv., 1902, p. 501. Holotype——No. 1879, Yale University Museum, was collected by Mr. Arthur Lakes from “ Big Canyon Quarry,” Jurassic (Morrison Beds), in the vicinity of Como, Albany County, Wyoming. It con- sists of a right pes nearly entire. The first description is as follows: A second species of this genus, about three times as large as the one [C. dis- par] just described, is represented by various remains, among which is a left @ Amer. Journ. Sci., XLVII, 1894, pl. vi. No. 1666. OSTEOLOGY OF CAMPTOSAURUS—GILMORE. 977 hind foot nearly entire. There were three functional digits in this foot, the first being rudimentary and the fifth entirely wanting. The metatarsal of the first digit is a splint, much curved, and with the apex above. The terminal phalanx of this digit is much compressed, not round as in the smaller species. The second metatarsal is of much greater length. The terminal phalanx of this digit is longer in proportion than that of the preceding species. The third and fourth digits were large and powerful. The main dimensions of this foot are as follows: energie secon): Meta ran sels e ees eeeeeewn 2 tye Ee 295 realest aciaimMeler OL PTORIMAlmentese ese Oe ee Are eve ee 113 Henetheor third. metatarsal. 22 eee to Eg eS lke Oe Lee eRe Ai. ceil gee 345 TeALesteciameLler OL proximal! ends sess owe So ee oe rs ae ee 150 ParaAnSVersesdiameber- OL, GIStHL CNG a smes essa) a See ee a ed 102 Eee ta Peimil Meta tnrsn le... meee Ae ey a 305 MenenneOrtirst.: pialanx: OL Chit dCi ene ts aie ee es ee ee 140 Mero wMOmnrst pilalainx Of Second: digit. 2242.20 28 ek UUs ake 120 The remains of the present species are from a lower horizon in the Jurassic than those described above, but within the limits of the Atlantosaurus beds. An examination of the type specimen shows it to be a right instead of a left hind foot, as originally described, and it has been so re- mounted, as shown in Plate 17, reproduced here from a photograph. As Marsh has already pointed out, C. amplus may be distinguished from the other species of the genus, (1) by its great size, (2) by the compressed terminal phalanx of the first digit. i, ss-uwevar om This phalanx, in nearly all of its details, re- = rmsr pierr, = Camp- sembles the ungual of the third digit of the Mangu. No. 1879, pes in the opisthoccelian dinosaurs, as shown in = Yau Musnum. Hoto- TYPE. % NAT. SIZE. fig. 38. ; la, ARTICULAR END. A second specimen, No. 1887, Yale Museum, Roven ovriine consisting of portions of the skull and lower jrticoprims swarm jaws (see Plates 7, 8, and 9), may, on account of its large size, be provisionally referred to this species. It was collected by Prof. O. C. Marsh from deposits in the Garden of the Gods, Colorado Springs, Colorado, in 1886. With this specimen was found the following note in Professor Marsh’s handwriting: “Part of this animal and various Sauropoda bones were taken out by Professor Kerr in 1878.” Diligent inquiry has failed to locate the repository of the parts-of this specimen collected by Professor Kerr, but their association with Sauropoda remains would indicate the Jurassic age of the deposits in which they were originally found. On page 1159 of volume 2 of Nicholson and Lydekker’s Manual of Paleontology, they remark: “It is not improbable that the large Iguanodont from the Upper Jurassic of the United States, de- scribed as ‘Camptosaurus amplus,’ should be referred to this group of Zguanodon, since it has but three functional digits in the pes.” 978 PROCEEDINGS OF THE NATIONAL MUSEUM. vou. xxxvt. While the first digit in C. amplus is undoubtedly rudimentary, still it is at once distinguished from the remnant of the first metatarsal in Iguanodon by the heavy articular distal end of Metatarsal I, and the presence of two distal phalanges. The distal row of tarsals, as in C. dispar, consists of two elements (see Plate 17), although, com- paratively speaking, somewhat reduced in size. Principal measurements of Specimen, No. 1879, Yale Museum. Digits. Is Il. Ii. IV. Greatestdeneth:.of metatarsals seecenrs oe eaie iets eee eee (a) 305 345 295 Greatest length of first row phalanges.........---.....------ Shik: oes 66 117 124 84 Greatest length of second row phalanges. .-...........----------------- 43 92 87 69 Greatest length of third row phalanges...........-...-.-----202-------|--------| 103 60 44 Greatest leneth of fourth row phalanges) s-s eee. soce- senate coer eae meee mans Ws Seeieer (0) 32 « Incomplete. » Missing. CAMPTOSAURUS MEDIUS Marsh. Camptosaurus medius Marsu, Amer. Journ. Sci. (3), XULVIIT, 1894, p. 85, pl. iv; 16th Ann. Rep. U. S. Geol. Surv., 1894-95, Pt. 1, 1896, p. 196, pl. nur; Mon. Geol. Surv., XXVII, 1897, p. 502, text figs. 58, 59. Camptosaurus medius Norcsa, Féldtani K6zlény, Budapest, XX XI, 1901, p. 210. Camptosaurus medius Hay, Bull. No. 179, U. 8. Geol. Surv., 1902, p. 501. Holotype—No. 1880, Yale University Museum. This specimen was collected by Mr. W. H. Reed from “ Quarry 13” in the Jurassic (Mor- rison beds) of Albany County, Wyoming." "It consists of a well-pre- served disarticulated skull and jaws; 59 or more vertebra, represent- ing all parts of the vertebral column, of which at least half have the arches and spinous processes complete; two humeri, two ulne, one radius, one femur, tibia and fibula, astragulus, caleaneum, left pes, two ilia, ischium, pubis, numerous ribs, and ossified tendons. The species was never properly defined by Marsh, the original de- scription consisting of the few lines which follow: ‘ There are at ieast two small species of the genus (C. medius and C. nanus, noticed below). * * * OC. medius was about fifteen feet long. * * * The skull, brain, and teeth of C. medius are shown on PI. IV. It is now known that the skull as figured ® was reconstructed, not from the disarticulated elements of the holotype alone, but the nasal region and the lower posterior section of the skull and mandible were drawn from No. 1887, Yale Museum (see Plates 7 and 8), a very much larger individual which undoubtedly pertains to a distinct “Tn his Bibliography and Catalogue of Fossil Vertebrata of North America Hay cites the occurrence of this species in Colorado. So far as known, it has not been found outside of Wyoming. 516th Ann. Rep. U. S. Geol. Surv., 1894-95, Pt. 1, 1896, pl. Lm. 4 NO. 1666. OSTEOLOGY OF CAMPTOSAURUS—GILMORE. 2979 species, as stated on page 277. Taking into consideration these facts in connection with many inconsistencies discovered in the figures, I do not believe that the skull, as illustrated, should be considered typical of the species. The teeth, as shown in fig. 9, are of the holo- type, but I am unable to find that they show any distinctive charac- ters. While the skull may show specific differences when the crania of the other forms are known, such comparisons can not be made at present, owing to the lack of material. The parts of the type-specimen, as enumerated above, represent a considerable portion of the entire skeleton, and it is unfortunate that they have not yet been made ready for study. At present only the disarticulated elements of the skull and jaws, limb, foot, and pelvic bones (see Plate 16) have been freed from the matrix, the other parts being still largely enveloped in the hard, concretionary mass in which they were originally entombed. Even though Marsh failed to define this form, I believe it to repre- sent a valid species intermediate in structure between C. dispar and C. brown. From C. dispar (compare Plates 15 and 16), its nearest ally, it may be distinguished by the lighter structure of the pelvic bones, that is, the ilium is considerably narrower in proportion to its length. The oblique border of the supero-posterior end is somewhat shorter. The ischium is much more slender, and the hammer-like expansion of the free end less robust, its greatest diameter measuring 73mm. Throughout the skeleton of C. medius appears to be lighter and more delicately constructed. The femur pertaining to the type- specimen is proportionately very short, but I am inclined to the opin- ion that this is either due to severe crushing, or the preparator, in joining the two ends, has omitted a section of the shaft of the bone, which would account for the extreme brevity of this element. Except in its smaller size, the hind foot shows no essential differ- ences when compared with that of C. dispar. The following are the measurements of the left pes of No. 1880: mm. (Tee SDUeUe TH Ol LIMeLabarsals Ieee en woke tts So) ee eee he ey ee ee 2 74 CrebesimencnnvOnemeLgtarsnin lites ts 2k 28 as St eee 148 ReaenielenctinOofemetacarsal eles 2) 28. te eet ets 2 181 (rCnCes te Len ll ¢Ol amet tarsal, Verse. 2 2 le Ae ee a 151 _ It is anticipated that upon the complete preparation of the type material, characters will be disclosed by which this species may be more completely defined, but for the present it must rest on those shown by the pelvis, although the dissimilarity shown by these bones certainly suggests other and even more important differences in the other portions of the skeleton. 280 PROCEEDINGS OF THE NATIONAL MUSEUM. — you. xxxvt. CAMPTOSAURUS NANUS Marsh. Camptosaurus nanus Marsu, Amer. Journ. Sci. (8), XLVIII, 1894, p. 85, pl. v., fig. 3; 16th Ann. Rep. U. S. Geol. Surv., 1894-95, Pt. 1, 1896, p. 196, pl. Lv, fig. 2. Camptosaurus nanus WaLcort, Science (2), XI, 1900, p. 28. Camptosaurus nanus Nopcsa, Foldtani Koézlény, Budapest, XXXI, 1901, p. 210. Camptosaurus nanus Hay, Bull. No. 179, U. S. Geol. Surv., 1902, p. 501. Holotype.—Cat. No. 2210 U.S.N.M.*¢ was collected by Mr. W. H. Reed from the upper Jurassic (Morrison beds), “ quarry 13,” near Como, Albany County, Wyoming,’ in 1882. The elements preserved are as follows: Portion of atlas, axis, 7 cervical, 16 dorsal, 4 sacral, and 34 caudal vertebrae, numerous cervical and thoracic ribs, right fore limb (scapula, coracoid, humerus, radius, and ulna), 2 femora, 2 tibiz, left fibula, 2 ilia, 2 ischia, portion of left pubis, 2 metacarpals, 1 carpal, and parts of ossified tendons. The original description is as follows: The smallest species of the genus ©. nanus was not more than 6 feet in length, and perhaps 4 feet in height when standing at rest. One of the striking features of this diminutive species is its long sigmoid scapula, shown in fig. 3, Plate V. This is in strong contrast with the short, straight scapula of C. dispar, seen on the same plate, fig. 2. Detailed description —The cervical vertebrae of the present species, as compared with the corresponding parts of the larger forms, show no particular distinguishing characters, although they differ in minor details. The entire neck and the first two dorsals remain articulated, and thus the true relationship of these parts is accurately displayed. The neck shows the same graceful upward curve of the cervicals as is found in the related species. The transition of the parapophyses from the anterior lateral border of the centrum of the ninth cervical to the elevated position on the side of the neural arch on the tenth (considered the first dorsal), agrees with the conditions found in @. browni and in the holotype of C. dispar. The odontoid is all that remains of the atlas. The axis, which lacks a portion of the spinous process, is otherwise very similar to the same bone in (. dispar. The second, or axis intercentrum, is not co- ossified as in the other species. In the narrowness of the ventral keel and the deep lateral depressions on the sides of the centra the re- maining cervicals appear to approach the vertebree of @. browni “Marsh's original accession number is (1561). No. 1881 is the number under which this specimen was catalogued while in the Yale University Museum. + Bull. No. 179, U. S. Geol. Surv., p. 501. Hay gives the locality as Wyo- ming (?). On the original field labels found with the specimens the locality was given in full, and may now be considered as absolutely determined. No. 1666. OSTEOLOGY OF CAMPTOSAURUS—GILMORE., 981 rather than those of C’. dispar. The opisthocoelian nature of the an- terior vertebre is not so pronounced as in the larger species. The dorsal series, though all are not articulated, appears to be com- plete; consisting of 16 vertebrae, many of which lack parts of their processes. The only marked characters in the dorsals of C. nanus are to be seen in the spinous processes of the posterior portion of the series, which are comparatively thin and lack the enlarged upper ex- tremities of the same vertebra in C. dispar and C. browni. The arches also appear to be relatively higher, with a corresponding enlarge- ment of the neural canal, in this respect approaching C. prestwichii. The sides of the centra (considered as a whole) are less concave in the longitudinal direction and appear to be more evenly rounded ven- trally. The centra gradually increase in length from the first to the thirteenth, the four- teenth being the shortest of the posterior dorsals (26mm.). The fifteenth and sixteenth dorsals may be distinguished from the others by the enlargement of their posterior ends in adap- tation to the correspond- ing surfaces of the Fic. 39.—SacruM OF CAMPTOSAURUS NANUS MARSH. s s ; No. 22 J-SIN.M.- 3° NAT. SIZE! rPH. ecentra with which they Car. No 10, U.S.N.M. 4 NAT. SIZE ee ‘ F VIEWED FROM THE RIGHT SIDE. @. 2y9., PREZYGA- articulate. The SIX- POPHYSES; art. §8., ARTICULAR SURFACH FOR FIRST SACRAL RIB; il., ILIUM; oO. t., OSSIFIED TENDONS; p., teenth or sacro- ¢ Bese 2 : acr dorsal PARAPOPHYSIS OF THE FIFTRENTH DORSAL; p. il., (see S. dor., fig. 39) 1s PREACETABULAR PROCESS OF ILIUM; p. 2yg., POSTZYGA- é c ° POPHYSES; S1, S82, S88, S4, SACRALS ONH TO FOUR, suturally united with a sath era fe K fe RESPECTIVELY; tr., TRANSVERSE PROCESS OR DIA- the centrum of the first POPHYSIS; 1, SPINE OF FIFTEENTH DORSAL; 2, SPINE OF SIXTEENTH OR SACRO-DORSAL ; 3, 4, AND 5, SPINES OF sacral. Its ventral sur- OR AES 0 , 4p AND 5 SACRALS ONE, TWO, AND THREE. face lacks the decided keel found in the sacro-dorsal of C. browni. The more detailed description of this vertebrae and the succeeding sacrals will be found in that part of the present paper relating to the osteology of Camp- tosaurus, pages 242 to 244. The sacral vertebrae show the most distinc- tive character of the species in the absence of the peculiar peg-and- notch articulation of the centra, which forms an important feature of the larger species. By some this difference might be considered of more than specific importance, but on account of the close similarity (except in size) of nearly all of the other elements to the homolo- gous parts of the larger species, and the lack of other diagnostic characters, I can see no reason for separating this form generically. It is of interest to note the similarity in this respect of C. nanus to C. prestwichii as described and figured by Hulke (see fig. 43). 989 PROCEEDINGS OF THE NATIONAL MUSEUM. VOL. XXXVI. Most of the caudal vertebra are articulated, the first alone of the anterior ones being detached, but its size and other characteristics at once show its proper place in the series. So closely do they resemble the caudals of C. browni, described elsewhere, that I shall mention only a few of their important features. The second caudal bears the first chevron. Science, XVI, Sept. 12, 1902, p. 435. 994 PROCEEDINGS OF THE NATIONAL MUSEUM. vot. xxxvt. preacetabular process is terminated by a somewhat more angularly pointed end than the rounded, more spatulate type found in C. dispar (see fig. 29). With the exception that the ischiac process is not so robust, in nearly all other respects the ilia appear very similar to those of the better known species. The contour of the posterior end, due to the damaged condition of both elements, can not be de- termined at this time. Turning now to the sa- cral and caudal vertebre associated with the ilia, it may be observed that the sacrals are characterized by the ankylosis of all their centra and their quite evenly rounded hemal sur- faces, there being just the faintest indication of the presence of a median keel upon Sacrals I, II, and III. None shows the flat- tened ventral surfaces ob- served in the paratype of C. dispar, No. 1877a, Yale Museum. In this respect, except in size, they more nearly resemble the sacrals of C. nanus. From the fragmentary evidence, it appears there were at least seven verte- bre ankylosed by their centra in this region, the most anterior of which is a sacro-dorsal, the poste- rior one likely representing a sacro-caudal, as in C. browni. Attached to the rock, which also holds the left ilium, are a number of spinous processes, which, if they pertain to the sacrum, show the spines as being much narrower antero-posteriorly, and without the heavy expanded tops of C. dispar. The caudal centra show no distinctive characters, but, as in the other species, the anterior caudals show the same obliquity of the centrum, the small neural canal, the wide transverse processes, and the slightly biconcave cupping of the articular ends, The diameter 4 NAT. FROM A PHOTOGRAPH. SIZE. Fic. 47.—ANTERIOR PORTION OF RIGHT ILIUM OF CAMPTOSAURUS DEPRESSUS. CAT. No. 5820, U.S.N.M. No. 1666. OSTEOLOGY OF CAMPTOSAURUS—GILMORE. 295 of the anterior caudal centra is greater vertically than transversely. One of the sacro-caudals shows it to have had a transverse process whose greatest width, at the point of origin at least, is in the vertical direction. The more distal caudal vertebrae, of which there are three present, show the same cylindrical shape, with long pre- and _post- zy gapophyses, as found in C. brown. The principal dimensions of the holotype are as follows: num. Vertical depth of ilium from middle of acetabular border (estimated) ____ 105 Distance between pre- and post-acetabular notches___________-_-_________- 141 WaAdEhsOL PUDLS Gor mMMm from tneanterion ends. 4 = 105 ve LesSte el SUN Ole sderO-C OLS iilemieemenat. somes sere 8 ee 58 iGredtestMenechrOL frst Sacral sue waste Leeson tT See el ee 53 realest LeuStoLOL SECON: Sa Gall = semen oe Ney ee i ee 56 Greatest vertical depth of sacro-dorsal (anterior end)__________________ 70 Greatest length of caudal centrum bearing chevron (second?) ~~.» 51 Greatest width anterior end of caudal centrum bearing chevron (second?) —_ 66 Greatest height anterior end caudal centrum bearing chevron (second?) —— 76 A second specimen, Cat. No. 5820, U.S.N.M., consisting of the well preserved anterior portion of a right ilum, collected by Mr. J. L. Kenney from the Morrison beds of the Jurassic, near Como, Albany County, Wyoming, undoubtedly pertains to this species (see fig. 47). CAMPTOSAURUS BROWNI, new species. Holotype.—Cat. No. 4282, U.S.N.M. From the Jurassic (Mor- rison beds), Quarry No. 13, 8 miles east of Como, Albany County, Wyoming. Collected by Mr. Fred Brown during the years 1885 and 1886. Named for the collector, whose discoveries of important fossil specimens have done much to further the science of paleontology. The typical specimen consists of a considerable portion of the skeleton, and since the elements have been listed on page 203, it ap- pears unnecessary to again enumerate them. As this skeleton is the basis for that part of the present paper devoted to the osteology of Camptosaurus, where a detailed description of the bones compos- ing it will be found, it is necessary here to discuss only those char- acteristics by which it differs from the other known species. Specific characters—Illium of moderate depth, with long pre- and post-acetabular processes; the hinder part especially narrow; su- perior border slightly convex with oblique posterior portion short.. Seven vertebra, of which five are considered sacrals, united by suture in sacral region. Peg-and-notch articulation confined to the posterior members and extending into the anterior caudals. Anterior sacral vertebrae compressed transversely. Last dorsal with ventral keel. Ischia slender with light expanded distal ends. Typically the skeleton represents an animal about 16 feet in length, intermediate in size between C. dispar and C. medius. While it ap- 296 PROCEEDINGS OF THE NATIONAL MUSEUM. vou. xxxvt. proaches the former more nearly in size, it resembles the latter in the comparative lightness of its skeletal structure. | The cervical vertebra of C. browni, when compared with those of C. dispar, show a much narrower ventral keel and deeper lateral depressions. The dorsal region, on account of the lack of material, I am unable to contrast with the other species, excepting C. nanus, the dorsals of which may be readily distinguished on account of their small size and the thinness of the spinous processes, which are with- out decided thickening of their superior terminations. The last, or sacro-dorsal, has a short but decided ventral keel which at once distinguishes this element from the smoothly rounded hemal surface of this vertebre in C. dispar, or the faint keeled ones as found in C. depressus, C. nanus, and C. prestwichit. The sacral region of C. browni resembles C. depressus most nearly in having seven vertebre united by suture, of which five are con- sidered true sacrals in the former species. The sacrum differs from C'. dispar in the increased number of sacral vertebra, the absence, in the anterior elements, of the peg-and-notch articulation, and its con- tinuance into the caudal region, and by the more compressed and keeled anterior centra. The absence of the peg-and-notch articula- tion in the sacrals of C. nanus and C. prestwichii at once separates them from C. browni. Outside of the anterior caudals, as noted above, the other vertebra of the tail show no distinctive features. As known at the present time, the ium is one of the most char- acteristic bones of the entire skeleton of Camptosaurus, and since this element is present in all of the holotypes of the American species, excepting C. amplus, it offers a basis of comparison equal in impor- tance to the differences displayed in the several species. The length of the ilium (see fig. 30), which wants the extremity of the pre- acetabular process, is quite equal to that of the average adult indi- vidual of C’. dispar (see figs. 1 and 3, Plate 14), although its greatest depth is considerably less. The preacetabular process when complete was long, the superior border slightly convex, and the post-acetabular portion long and especially narrow, the angular oblique border of the supero-posterior end being short. Compared with the ilia of the other species, the difference in the form of the hinder half, coupled with its other proportions, would, apart from other evidence, indicate the specific distinctness of its owner. In contour the ilium is intermediate between C’. medius on the one hand and @. depressus on the other, as may be seen by comparing figures. Compared with C’. depressus, it shows a much wider preacetabular notch, a deeper acetabulum, and a greater depth of the bone as a whole. The pathological condition of the right ilium of C. brownd is of interest in showing to what extent the shape of a bone may be modi- No. 1666. OSTEOLOGY OF CAMPTOSAURUS—GILMORE. 997 qs SEES see fied by external injury. On the posterior half, the comparatively thin, plate-like part of the ilium is divided vertically, the two halves swelling out to form the walls of a cavity which extends downward, emerging on the ventral border. The cavity is longer than wide, measuring on the upper border of the opening 86 mm. in a longi- tudinal direction and 46 mm. in the transverse, the ventral exit being considerably smaller. As indicated by a deep depression on the dorsal border, the injury was probably received from above. The exostosis of the bone was greatest on the front side of the cavity where it measures 72 mm. in width. The normal diameter of this part of the ilium, as shown by one of the opposite side, is only 21mm. A second injury was found on one of the caudal vertebree near the root of the tail, as indicated by the pathologic condition of the spinous process, which is considerably enlarged and has near its base an elongated opening which perforates the bone. While the wound in the ilium must have been an exceedingly painful one at the time of infliction, it in no: way utterly disabled the animal, at least to the extent of leading to its death, for, as shown by the speci- mens, all of the broken margins of the bone had healed. Although these injuries may have been inflicted by some of its large carnivorous contemporaries, the position of the wounds suggests the idea that this individual was a female who might have received the injuries during copulation. The ischia of C. browni are comparatively slender, and while there is a considerable expansion of their distal extremities, they lack the massiveness of those of (. dispar. Tn the lightness of the struc- ture of these elements, they approach C’. medius most nearly. The fore limbs and feet show no distinguishing characteristics. The principal measurements of the vertebre and other parts will be found in that part of the paper devoted to the osteology of Camp- LOSQUT'US. GEOGRAPHICAL AND GEOLOGICAL DISTRIBUTION. In North America, camptosaurian remains have been found in southeastern Wyoming, in Albany and Carbon counties;* Colorado, near Canon City, and in the “ Garden of the Gods”, near Colorado Springs; South Dakota, in Custer County, in the vicinity of Buffalo Gap. Beyond the limits of the United States, specimens which have been referred to Camptosaurus have been found in England, Isle of Wight, and Hungary. All of the American species, with the possible exception of C. de- pressus, are from the Morrison beds (Atlantosaurus beds of Marsh), 4JIn the Journal of Geology, XIII, No. 4, 1905, p. 348, Dr. S. W. Williston reports the occurrence of Laosaurus remains in Fremont County, near Lander, Wyoming, in deposits considered lower Cretaceous in age. 998 PROCEEDINGS OF THE NATIONAL MUSHUM. — vou. xxxvt. of the Jurassic, but Marsh’s statement that “they occur in successive deposits of the same general horizon, the smallest species below, the ’ largest above,’ can not be verified, and is not borne out by the structural characteristics of the typical specimens. Moreover, as shown by the original field labels still remaining with the type material, (. dispar, C. medius, and C. nanus came from the same quarry (No. 13) and in all probability were found at the same level. Although the holotype of C. amplus was found in the same general area, it is from another quarry in a region where it is difficult to trace stratigraphic horizons. It also appears from Marsh’s own writings that he was not quite clear as to their stratigraphic posi- tions, for in 1879” at the close of his description of C’. amplus, he says: “ The remains of the present species are from a lower horizon in the Jurassic than those described above |C. dispar], but within the limits of the Atlantosaurus beds.” He thus places the larger species at a lower level, which is contradictory to his later statements. Fia. 48.—SEcTION oF QuaRRY 13. Mapr BY Mr. Frep Brown IN 1884. Since Quarry No. 13 has furnished four of the holotypes per- taining to the genus Camptosaurus, besides a vast quantity of other material, the exact stratigraphic position of the bone-bearing layer is of considerable interest. A clue to the position of this layer was found in a rough section of the strata exposed in working this quarry, made by Mr. Fred Brown in 1884 (see fig. 48). The fossils occur here in a layer of sandy clay,’ as I have deter- mined from the matrix, still adhering to bones, and, as seen in Brown’s section, the bone-bearing layer (“pay streak”) is inter- calated between layers of marl or clay, green below and brownish above, all three layers lying between bands of sandstone. 4 Amer. Journ. Sci., XLVIII, 1894, p. 85. bTdem, XVIII, 1879, p. 503. ¢On page 199 of the present paper, Mr. W. H. Reed is quoted as also noting the sandy nature of the matrix in which the fossils occur as being unusual. «a _ No. 1666. OSTROLOGY OF CAMPTOSAURUS—GILMORBE. 999 In comparing the conditions here with the sections so carefully worked out by Dr. F. B. Loomis at Como Bluff and Little Medi- cine, in an area a few miles to the west of that under discussion, it appears that this sandy layer may be tentatively correlated with No. 28 of his section (see Plate 20), which he describes as follows: No. 28 is a gray sandstone in which the rich Bone Cabin Quarry is situated, and also the Stegosaurus Quarry. The sandstone varies extremely in hard- ness, being, in the south part of Bone Cabin Quarry, soft and mixed with con- siderable clay so that it is workable with an awl. Im the northern part of the quarry, however, there are bands of the firmest sort of sandstone. In Como Bluff the layer is clay with merely an admixture of sand. Bone Cabin Quarry — has yielded a great variety of genera: Diplodocus, Morosaurus, Brontosaurus, Allosaurus, Ceratosaurus, Oamptosaurus, Stegosaurus, as well as_ several genera of carnivorous Dinosaurs; also Compsemys and Goniopholis. The correctness of the above correlation appears to be indicated (1) by the similar nature, lithologically, of the materials composing the bone horizon, (2) a similarity in the over and underlying strata, (3) the likeness of the faunas from the two localities. If later investigations show this provisional correlation to be correct, it is of the utmost importance as definitely locating the horizon from which the holotypes of the following species have come: Campto- saurus dispar, C. medius, C. nanus, C. browni, Dryosaurus altus, and Diracodon laticeps. Among the other dinosaurian genera recognized from quarry No. 13 are Stegosaurus, Allosaurus, Coelurus, and Morosaurus, as well as the turtle, Glyptops and the crocodile Goni- opholis, and fish remains, which, however, are too fragmentary to admit of identification. By comparing the faunal lists of the two quarries, it will be observed that they are quite alike, although Bone Cabin Quarry predominates in representatives of the Opisthocoelia (Sauropoda), Quarry 13 in members of the Orthopoda (Predentata). This observation would also apply to the relative numbers of indi- viduals of each group found in the two quarries. Quarry No. 13, as shown by the maps, was especially rich in stegosaurian and campto- saurian remains. With the permission of Dr. F. B. Loomis, I reproduce (Plate 20, figs. t and 2) sections of the Little Medicine and Como Bluff exposures, which, according to his measurements, show the 5-foot sandy layer No. 28 to be within 60 feet of the overlying Cretaceous (Dakota). This is the highest known horizon of the Jurassic in which camptosaurian bones have been found, and the discovery at this level in the famous “ Bone Cabin Quarry” of a skeleton of CO. nanus (see Plate 19), strengthens somewhat the assumption of the contemporaneity of this layer with the bone horizon of Quarry 13. ¢ Bull. Amer. Mus. Nat. Hist., XIV, 1901, pp. 189-197, pl. xxvu, figs. 2, 3. 300 PROCEEDINGS OF THE NATIONAL MUSEUM. you. xxxvt. The finding of Camptosaurus remains in the Lakota, near Buffalo Gap, South Dakota, appears to extend the geological horizon of this genus. The occurrence of these fossils is described by Mr. N. H. Darton? as follows: The formation has yielded a large number of cycads, notably those described by Mr. Lester F. Ward. These and associated plants are regarded by Mr. Ward as Cretaceous in age. In 1898 I discovered saurian bones in or near the cyead horizon at Buffalo Gap, but as they are of new species it is difficult to obtain from them any evidence bearing on the age of the formation. If it were not for the evidence of the flora, these bones would be regarded as late Jurassic in age. * * * The bone bearing beds are in the middle of the ~ Lakota formation, or about 90 feet above the unconformity on the Unkpapa sandstone which is approximately the horizon that has yielded the cycads between Edgemont and Minnekata, near Blackhawk and elsewhere about the hills. The vertebrate fauna of the above horizon, as now known, con- sists of Toplitosaurus (Stegosaurus) marshi (uucas) and Camp- tosaurus depressus, new species, while the presence of a sauropodous dinosaur is indicated by some fragmentary bones found associated with the type material. That the bone-bearing layer at Buffalo Gap is later than the fossil horizon in Quarry 13, Como, Wyoming, appears quite probable, al- though the evidence as yet is insufficient to conclusively establish the fact. MHoplitosaurus has its nearest ally in Polocanthus of the Weal- den. While the former genus is known by a single fragmentary speci- men only, it may, from its geological position, represent a form in- termediate between Stegosaurus of the Jurassic, and Nodosaurus, Stegopelta, and Anchylosaurus of the American Upper Cretaceous. Although the typical specimen of C. depressus is fragmentary, such parts as are preserved appear to show that of all the known forms of Camptosaurus this species approaches the Wealden /gua- nodon most closely, as indicated by the narrowness of the ilium and the coosification of the sacral vertebree. As has been pointed out in the preceding pages, the Camptosaurus remains from Quarry 13, when compared with 7guanodon, show a more generalized structure, which suggests a somewhat greater an- tiquity for the beds in which they are found. In this connection it is a significant fact that of the several European species referred to Camptosaurus, the only valid one is (. prestwichit from the Kim- meridge Clay, and its affinities appear to be nearest to C’. nanus, the holotype of which was found in Quarry 13. Corroborative evidence is furnished by the abundant remains of Stegosaurus found in the above quarry (see Plate 6), which genus is so closely allied to Dacen- 421st Ann. Rept. U. S. Geol. Surv., Pt. 4, 1899-1900, p. 527. NO. 1666. OSTEOLOGY OF CAMPTOSAURUS—GILMORE. 801 trurus® (Omosaurus) from the Kimmeridge that Marsh believed them to be identical. It is unfortunate that the early paleontologists rarely gave any precise location, much less the exact geological horizon from which typical specimens were obtained, so that the faunas of the upper and lower parts of the American Jurassic, except in a few instances, have never been differentiated. The whole fauna has been included under the term Upper Jurassic, and only in the last few years have a few authorities separated some of the upper part as lower Creta- ceous. The vertebrates found in the Lakota at Buffalo Gap point to its being the equivalent of the Wealden of England. Assuming, as many authorities do, that the Wealden is really Jurassic, these beds would then represent the uppermost part of that formation. The above evidence, then, is in favor of the contemporaneity of the Buffalo Gap horizon with the Wealden, and indicates that the age of the Quarry 13 bone layer is greater than the Wealden. Such evi- dence as is shown by the Camptosauride not only supports Hatcher’s contention ? that the lower members of the Morrison (Atlantosaurus Beds) are below the Wealden, but that they are of greater age than the Purbeck and possibly equivalent to the Kimmeridgian. RESTORATION OF CAMPTOSAURUS. Marsh gave us the first skeletal restoration of Camptosaurus, here reproduced as Plate 18. While this reconstruction gives a good idea of the animal as a whole, it is now known, as has been pointed out earlier in the present article, to be in error in several particulars. The most striking change brought about by this more recent. study is the shortening of the presacral region, which was made too long by Marsh, owing to an overestimate of the number of presacral vertebra. In his figure (see Plate 18) there are 30 presacral vertebra, 9 of which are considered as belonging to the cervical region, thus leaving 91 thoracic vertebra. Two specimens in the U. 8S. National Museum agree in having 16 dorsalseach. If, then, this latter number is correct, the presacral series would be shortened by 5 vertebrie, making the proportions of the animal markedly different from the first conception of its appearance (compare with Plate 19). Even though it ulti- mately be found that Camptosaurus has 18 dorsal vertebrz (a possi- bility indicated by the occurrence of that number in the holotype of C. prestwichii and in the allied Zguanodon), it would still mean the shortening of the column by 3 vertebra, which would have lessened the distance between the fore and hind limbs, producing a more compactly built animal than appeared in the first reconstruction. @Science (N. S.), XVI, 1902, p. 435. 5 Memoirs, Carnegie Musuem, III, 1903, p. 68, 802 PROCEEDINGS OF THE NATIONAL MUSEUM. VOL. XXXVI. Although there is a considerable disparity in length between the fore and hind limbs, there appears to be some evidence to show that the bipedal mode of progression was not habitual. While I do not wish to be understood as believing that the upright position was not frequently assumed, still it appears to me that the quadrupedal posture was used more frequently than has been generally supposed. This is shown by the compact, ossified carpus, with smooth, well-defined articulating surfaces, which is supported by comparatively short and stout metacarpals, whose function was that of support rather than prehension. When compared with those of animals whose mode of progression is normally bipedal, the suggestion advanced here be- comes more apparent. Z'rachodon has slender, elongated metacarpals and imperfectly ossified carpus, and 7yuwanodon also has a tendency toward the lengthening of the metacarpals, though not so marked as in the former genus. The curved femur is also indicative of a flexed limb, which would have equalized somewhat the difference in length between the fore and hind legs. This character of the femur is in striking contrast to the straight femur of both /guanodon and Trachodon, a provision, as in the Proboscide, for the support of ereat weight. The obliquity of the anterior caudal centra indicates a rapid dropping of the tail as it leaves the sacrum, which is also suggestive of a normal quadrupedal position. In the two genera mentioned above the caudals extend straight out from the sacrum without appreciable ventral deflection. Through the courtesy of Dr. W. D. Matthew, of the American Museum of Natural History, I am enabled to present in Plate 19 the first skeleton of Camptosaurus to be mounted, which gives a truer conception of the animal than is obtained from the earlier recon- structions. As seen in the figure, the head is comparatively small, being carried on a gracefully curved neck of moderate length. The thoracic region, which has been given 16 dorsals, is of moderate length, borne on stout, clawed limbs, of which the hinder are longer and stouter than the fore. In life this animal was evidently strong and agile in movement. The tail was long, nearly equaling half the total length of the skeleton, and in life it probably served as a bal- ancing organ when the upright bipedal posture was assumed. Unlike many of the other predentate dinosaurs, there have been no dermal scutes nor ossicles found, so we have no knowledge as to the dermal covering, i Gore eee ~ Me tf Epes dea, FES oe SS rt * Sm ob ae ae ee ane f : jae ae yet te ** CS ae eet he a | aie BT araeY AS,” meh, | POE eda oe £ oe hie Yad Att ae ar eran u Bee eae ne. 52" 4 St tO aby ree AD FE 3. 7 os f ‘ Ns eee a ae Pe ee oe mY, ao a EXPLANATION OF PLATE 6. Diagrams 5, 7, and part of 4, of Quarry 13, near Como, Albany County, Wyoming, worked by Mr. Fred Brown for Prof. O. C. Marsh, during the years 1884, 1885, and 1886. The numbered bones show the positions in which the various elements of the holotype of Camplosaurus browni, Cat. No. 4282, U.S.N.M., were found in the quarry. The unnumbered bones scattered about pertain chiefly to members of the Stegosauria. A, Plesiotype of Diracodon laticeps, Cat. No. 4288, U.S.N.M. B, Series of caudals and dermal plates of Stegosaurus, Cat. No. 4714, U.S.N.M. The scale is about 4 feet to the inch. 304 s. Ss MUSE U. S. NATIONAL MUSEUM PROCEEDINGS, VOL. XXXVI PL. 6 ‘ ' ’ ‘ 1 ' ' ' 4 ‘ ' 1 ' ’ ‘ ‘ ' ' ‘ ' ‘ ‘ ' ‘ ' !' ' — NATIONAL MUSEUM are ae . = 3 PROCEEDINGS, VOL. XXXVI PL. 6 Diagram5. Map oF Quarry 13, NEAR Como, WYOMING. Proc. N. M. vol. XX Xvi—()9——_929 EXPLANATION OF PLATE 7. Frc.1. Anterior part of skull, Camptosaurus amplus? Marsh. No. 1887, Yale Museum. About 2 nat. size. 2. Explanatory figure of same: 7, lachrymal; m, maxillary; nd, nasal ; no, mnarial orifice; pmz, premaxillary ; prf, prefrontal; 80, supraorbital. 206 U. S. NATIONAL MUSEUM PROCEEDINGS, VOL. XXxvI Hoy we Bmx ANTERIOR PORTION OF SKULL OF CAMPTOSAURUS AMPLUS?. EXPLANATION OF PLATE 8. Frc. 1. Posterior part of skull, Camptosaurus amplus? Marsh. No. 1887, Yale Museum. #2 nat. size. External view. 2, Explanatory figure of same: ang, angular; d, dentary; ju, jugal; q, quadrate; J, quadratojugal; sur, surangular. 308 U. S. NATIONAL MUSEUM PROCEEDINGS, VOL. XXXVI PL. 8 POSTERIOR PART OF SKULL OF CAMPTOSAURUS AMPLUS?. EXPLANATION OF PLATE 9. Fic. 1. Posterior part of skull, Camptosaurus amplus? Marsh. No. 1887, Yale Museum. 2 nat. size. Internal view. 2. Explanatory figure of same: ang, angular; art, articular; co, coro- noid; f, internal mandibular foramen?; h, hyoid; p, art, prearticu- lar; pt, pterygoid; q, quadrate; 8, splenial. 310 U. S. NATIONAL MUSEUM PROCEEDINGS, VOL. XXXVI PL, ) POSTERIOR PART OF SKULL OF CAMPTOSAURUS AMPLUS 2. ties s Grae EXPLANATION OF PLATE 10. Fic. 1. Posterior portion of skull, Camptosaurus dispar Marsh. Cat. No. 5473, U.S.N.M. 4 nat. size. Seen from above. From a photograph. 2. Explanatory figure of same. al. sp., alisphenoid; c, occipital condyle; f, frontals; p, parietal; pf, postfrontal; poc, paraoccipital process or opisthotic; so, supraoccipital. : 312 U. S. NATIONAL MUSEUM PROCEEDINGS, VOL. XXXVI POSTERIOR PART OF SKULL OF CAMPTOSAURUS DISPAR. bes 10 EXPLANATION OF PLATE 11. Fic. 1. Occipital region of skull, Camptosaurus dispar Marsh. Cat. No. 5478, U.S.N.M. 4 nat. size. Posterior view. From a photograph. 2. Explanatory figure of same. 06s, basisphenoid; exo, exoccipital; f, fora- men magnum; oc, occipital condyle; p, parietal; pf, postfrontal; poc, paraoccipital process or opisthotic; p. pt., process which meets ptery- goid; so, supraoccipital. 314 U. S. NATIONAL MUSEUM PROCEEDINGS, VOL. XXXVI PL. 11 CLO ZZ OCCIPITAL REGION OF SKULL OF CAMPTOSAURUS DISPAR. Pf bs EXPLANATION OF PLATE 12. Complete neck and posterior portion of skull Camptosaurus dispar Marsh. Cat. No. 5473, U.S.N.M. 2 nat. size. Viewed from the left side. From a photo- graph. 316 HES ah PROCEEDINGS, VOL. XxXxvI U. S. NATIONAL MUSEUM “dvdsid SNYNVSOLdUNVD 40 TINMS JO LUV d GNV MOAN % A 2! a" : Th a © EXPLANATION OF PLATE 13. Sacrum of Camptosaurus dispar Marsh. No. 1877a, Yale Museum. About 24 nat. size. Paratype: viewed from the left side. a, sacro-dorsal; 1, 2, 3, and 4, sacrals one to four, respectively. From a photograph. 318 U. S. NATIONAL MUSEUM PROCEEDINGS, VOL. XXXVI Peas SACRUM OF CAMPTOSAURUS DISPAR. | ny : EXPLANATION OF PLATE 14. Fig. 1. Left ilium Camptosaurus dispar Marsh. Cat. No. 5473, U.S.N.M. 4 nat. size. Pubic peduncle restored after Cat. No. 5818, U.S.N.M. . Left ilium Camptosaurus nanus Marsh. Cat. No. 2210, U.S.N.M. Holo- type. 4 nat size. 3. Left ilium Camptosaurus browni. Cat. No. 4282, U.S.N.M. Holotype. ~ hat. size. 320 2 U. S. NATIONAL MUSEUM PROCEEDINGS, VOL. XXXVI SS Se — ~ ae SSW 7 Ji Gee! Fe Se a eS > Pe |! ['4A¢ oe ILIA OF THREE SPECIES OF CAMPTOSAURUS. PL. 14 EXPLANATION OF PLATE 15. Pelvic arch Camptosaurus dispar Marsh. No. 1878, Yale Museum. Para- type. One-fifth nat. size. Internal view of right side: 4a, acetabulum ; il, ilium: is, ischium; p, pubis; p’, postpubis; 2, depressions for articulation of sacral ribs. From a photograph. 322 U. S. NATIONAL MUSEUM PROCEEDINGS, VOL. XXxXvI PL. 15 PELvic ARCH OF CAMPTOSAURUS DISPAR. EXPLANATION OF PLATE 16. Pelvic arch Camptosaurus medius Marsh. No. 1880, Yale Museum. Holotype. About one-fourth nat. size. Viewed from the right side: a, acetabulum; i, ilium; is, ischium; p, pubis; p’, postpubis. From a photograph, 324 U. S. NATIONAL MUSEUM PROCEEDINGS, VOL. XXXVI PL. 16 PeELvic ARCH OF CAMPTOSAURUS MEDIUS. EXPLANATION OF PLATE 17. Right hind foot Camptosaurus amplus Marsh. No. 1879, Yale Museum. Holo- type. Viewed diagonally from the front. Less than one-third nat. size. From a photograph. 326 U. S. NATIONAL MUSEUM PROCEEDINGS, VOL. XXXVI PL. 17 RIGHT HIND FooT OF CAMPTOSAURUS AMPLUS. EXPLANATION OF PLATE 18. Restoration of Camptosaurus dispar Marsh. Based upon specimens in the Yale Museum, One-thirtieth nat. size. After Marsh. 328 PROCEEDINGS, VOL. XXXVI PL. 18 U. S. NATIONAL MUSEUM iN VQA HY sees crete aa a Se Le WN OIAY YG ¢ SSI \ = I) | ‘uvdSIG SNYNVSOLMNVD JO NOILVYOLSSY * any 7 val! IXPLANATION OF PLATE 19. Mounted skeleton of Camplosaurus nanus Marsh. No. 6120 American Museum of Natural History. About 7s nat. size. Through the courtesy of Dr. W. D. Matthew. 390 19 Fes PROCEEDINGS, VOL. XXXVI U. S. NATIONAL MUSEUM *SANVN SNYNVSOLMWVD JO NOLA1SxS GALNNOW EXPLANATION OF PLATE 20. Fig. 1. Section across Como Bluff, south side of Como anticline. bho Como Morrison (Atlantosaurus Beds of Marsh). Shirley=Baptanodon Beds of Marsh. “28. Sandy clay” is the equivalent of 28 of fig. 2 . Section across ‘Bone Cabin Draw,” south side of Little Medicine anticline. “28. Gray sandstone ” contains bone layer of the famous ‘*t Bone Cabin Quarry,” and probably equivalent to the fossil layer of Quarry 13. Both figures after Loomis. 332 U. S. NAT 7150 6770 01234 Triass oa pert sraren ets Cor ae ) i a —— Ss. U. §. NATIONAL MUSEUM PROCEEDINGS, VOL. XXXVI PL. 20 TRIASSIC SHIRLEY COMO DAKOTA ALTITUDE 01234 6 8 10 Red Sandstone. SS SS ee Roos Fic. 1 Triassic. Shirley. Como. | Oskote. Fic, 2 SECTION OF JURASSIC ExPOsURES, NEAR MEDICINE Bow, WYOMING. anton bi _eehde & : 3 ore ; i wdian ARS ypaRNVECaXT OmeaAUL YO MOITOSS a ; . THE COLLECTION OF ROSARIES IN THE UNITED STATES NATIONAL MUSEUM.2 By Immanuet M. Casanowicz, Of the U. S. National Museum. INTRODUCTION. The rosary is a string of beads, generally formed into a circlet or loop, used' for keeping count of prayers or formulas repeated in re- ligious devotions. The materials of which it is made range from nat- ural berries or common wood to costly metals and precious stones. It is best known from its use by Roman Catholics in devotion to the Vir- gin Mary, to which is also due the name “rosary,” as will be seen further on. But long before they came into vogue in Europe and among Christians, mechanical devices for counting the repetition of prayers or mystical sentences were in use among various oriental peoples, and at present some form or other of rosary is used by about three-fourths of the world’s inhabitants. Man’s natural tendency to «Viterature consulted: William Tayler, The Rosary in India, Journ. Soc. of Arts, XXI, No. 1068, London, May 9, 1873, pp. 461-470. Monier M. Williams, Indian Rosaries, The Athenzum, No. 2624, London, Feb. 9, 1878, p. 188, and, Buddhism in its Connection with Brahmanism and Hinduism, New York, 1889, p. 383. L. Austin Waddell, Lamaic Rosaries, Their Kinds and Uses, Journ. Asiatic Soc. Bengal, LXI, Pt. 1, 1892, pp. 24-88, and, The Buddhism of Tibet, or Lamaism, London, 1895, pp. 202-211. J. M. James, Descriptive Notes on the Rosaries (jiu dew) as used by the Different Sects of Buddhists in Japan, Trans. Asiatie Soc. Japan, IX, Yokohama, 1881, pp. 173-182. Ignaz Goldzieher, Le Rosarie dans l’Islam, Revue de |’Histoire des Religions, Jan.—Juin, 1890, XXI, pp. 295-300. Rev. Herbert Thurston, S. J., The Archeology of the Rosary Beads, The Month, London, April, 1901, pp. 383-404, and History of the Rosary in all Countries, Scientific American Supplement, No. 1870, New York, April 5, 1902, pp. 21960-21968. John R. Volz, Beads in the Catholic Encyclopedia, II, New York, pp. 361-862. Much interesting information on Chinese and Japanese rosaries was also derived from manuscript notes of Miss Eliza R. Scidmore, which she deposited with the larger numebr of rosaries described in this paper in the National Museum. Mr. William E. Safford, of the Department of Agri- eulture, and Mr. Wirt Tassin, of the National Museum, kindly aided in identify- ing the material of the rosaries. PROCEEDINGS U, S, NATIONAL Museum, VOL, XXXVI—No, 1657. 333 834 PROCEEDINGS OF THE NATIONAL MUSEUM. VOL. XXXVI. iteration, especially of prayers which have assumed a_ traditional form, to the recital of which a particular merit or potency is attrib- uted, must have early suggested some means of assuring accuracy of the count, such as the fingers, pepo, knotted cords, gradually evolv- ing into che string of beads. Following the succession in time of the appearance of’ the rosary in the several religious systems which are known to use it, the paper will treat first of the form and manner of its use among the Hindus or Brahmans; second, among Buddhists; third, among Mohammedans; and, fourth, among Christians." I. THE HINDU OR BRAHMAN ROSARY. The Hindus are generally believed to have first evolved the rosary. “Tt is not unreasonable to conjecture,” says the noted Indianist, Monier M. Williams, “that the original invention of the rosary is due to India. * * * No other country in the world stands in such need of aids to religious exercises. * * * The pious Hindu not only computes his daily prayers as if they were so many rupees added to his capital stock in the bank of heaven, but he sets himself to re- peat the mere names of his favorite gods, and will continue doing so for hours together.” It is first mentioned in the Atharva Veda.¢ The Sanskrit name for Me rosary is japamala, “ muttering chaplet, and sometimes smarana, “ remembrancer.” Corresponding to the two great religious sects into which the Hindus are mainly divided there are two rosaries, different in material and number of beads used by them. The rosary of the votaries of Siva is a string of 32 or 64 rough berries of the rudraksha tree (Elacocarpus ganitrus) each generally marked with five lines, the roughness perhaps symbolizing the auster1- ties connected with the worship of Siva, and the five lines standing for the five faces, or the five distinct aspects of the god. That of the followers of Vishnu is usually made of the wood of the tulasi, or holy basil (Ocimum sanctum), a shrub sacred to Vishnu,* and generally - consists of 108 smooth beads. Hindu ascetics (yogis) are said to some- times wear beads made of the teeth of dead bodies. The rosary is used by: the Hindus to count the repetition of the names and epithets @A. V. Williams Jackson, Persia, Past and Present, New York, 1906, p. 395, mentions that in connection with the funerary rites of the Parsees, or Zoroas- trians in Persia, ‘‘ The priest, with a rosary of beads, asks each of the mourn- ers how many prayers he will offer in memory of the deceased.” But nothing further could be learned on the nature and use of the Parsee rosary. b’The Atheneum, February 9, 1878, p. 188. ¢ Compare E. Washburn Hopkins, The Religions of India, Boston and Lon- don, 1895, p. 557. d“ The tulasi shrub is pervaded by the essence of the great god Vishnu and his wife Lakshmi, and is itself worshipped daily as a deity.” J. G, Frazer, Lec- tures on the Early History of Kingship, London, 1905, p. 156, no. 1667. COLLECTION OF ROSARIES—CASANOWICZ. 335 of the deity. High-caste Brahmans merely employ it to assist them in counting up hate daily prayers, while the ascetics consider the operation of counting a means of promoting contemplation and mental abstraction, which is so highly prized by the Hindus. Devo- tees attach great importance to the size of the beads, which may vary from small seeds to heavy balls as big as a human skull. Rosaries are also worn by the Hindus as necklaces, and the Vishnu chaplet of 108 tulasi beads plays an important part in the ceremony of con- firmation, or initiation, which children undergo at the age of 6 or 7, when such a rosary is passed around their neck, and they are at the same time taught some sacred formula or sentence to be recited by them. There is no example of a Brahman rosary in the National Mu- seum’s collection. But in form and use it resembles the Buddhist rosary, of which it apparently was the parent. II. THE BUDDHIST ROSARY. In the Buddhist rosary of 108 smooth beads may be .recognized its Brahman origin. In fact, the rosary and even prayer itself, must be considered an accretion upon the simple original system of Buddha, in which “ personal divinity has almost faded into a mere metaphysical idea.” The rosary in Buddhism is accordingly es- pecially peculiar to the northern school (the Mahyana, or great vehicle), with its belief in the merit and efficacy of meditation, and in the potency of repeating mystic spells and formulas. But, though thus borrowed from the outside, the rosary has attained in Buddhism its widest diffusion and most general application. It forms an es- sential part of a Buddhist monk’s equipment. The Buddhists give the number 108 of the beads a symbolic signifi- cation of their own: The number 108 is said to correspond to a like number of mental conditions, or sinful inclinations, which are to be overcome by the recitation of the beads. The number 108 seems to have a special signification in the tradition and philosophy of Buddhism; 108 Brahmans were summoned at Buddha’s birth to fore- tell his destiny. The Burmese foot prints of Buddha have some- times 108 subdivisions; the Aahgyur, the Tibetan sacred writings of Buddhism, are composed gf 108 volumes, and the white pagoda at Peking is surrounded by 108 columns. So also in Japan, on the festival of the dead (the bommatsuri or bonku), which is observed from the 13th to the 15th of July, 108 welcome fires (mukaebi) are “Tn the Buddhist Forty-two Points of Doctrine, article 10, is written: “ The ‘man, who in the practice of virtue applies himself to the extirpation of all his vices, is like to one who is rolling between his fingers the beads of a chap- let. If he continues taking hold of them one by one he arrives speedily at the end. By extirpating his bad inclinations one by one a man arrives thus at perfection.” Compare Dr, Zerfii in Journ, Soc, Arts, May 9, 1873, p. 469. 836 PROCEEDINGS OF THE NATIONAL MUSEUM. VOL. XXXVI. lighted along the shores of the sea or lake or river by which a city or village is situated. One hundred and eight rupees are commonly given in alms, while in China 108 blows are an ordinary punishment for malefactors. Alongside of the full rosary of 108 beads, employed by the monks, there are in vogue rosaries of 18 and 16 beads, representing, respec- tively, the 18 lohans, or chief disciples of Buddha counted by the Chinese, and the 16 rohans of the Japanese. The common people, moreover, use indifferently rosaries of 30 or 40 beads. The material of the Buddhist rosaries varies according to the taste, wealth, and rank of the owner. The commonest are made of seeds, wood, pebbles, shells, glass, or bone; the more costly of jade, turquoise, coral, amber, silver and gold, and even of pearls and other gems. Marco Palo relates that the king of Maabar (that is, Malabar), whom he visited about 1290 A. D., had a necklace of 104 (doubtless an error for 108) large pearls and rubies to count his prayers upon. Much in favor for rosary beads is the wood of the sacred Bo-tree (Indian Pilpal, a species of fig, Ficus religiosa), under which Sakya Muni attained to the state of Buddha. The countries in which the Buddhist rosary is most widely used are Tibet, China, and Japan.? A. TIBETAN ROSARIES, oe The rosary in Tibet—called trengwa, “ string of beads ”—is not only an essential part of the outfit of the lamas, as the Buddhist monks are called there, but 1s everywhere in appearance. The patron god of Tibet, Cheresi or Padmapani, is represented with a rosary in his hand, and nearly every man and woman carries a rosary, holding it in the hand, or attached to the girdle, or wearing it around the neck as a necklace, or twisted around the wrist as a bracelet. Lay- men also use it to assist in ordinary calculations, like the sliding balls of the Chinese, in their business transactions. The material is not only varied according to the taste and wealth of the owner, but is also determined by the particular sect to which the devotee belongs, and the deity to whom worship is to be rendered. The head lama of a large and wealthy monastery may have rosaries “Compare Lafcadio Hearn, Glimpses of Unfamiliar Japan, London, I, p. 107. %As regards Burma, Mr. Waddell (Journ. Asiat. Soc. Bengal, LXI, p. 25) relates that he met several Burmese monks “ possessed of a rosary, called ‘Bodhi,’ consisting of 72 black subcylindrical beads, which I understand were composed of slips of a leaf inscribed with charmed words and rolled into pellets with the aid of lacquer or varnish.” He adds (p. 33) that the Burmese “seem to use their rosary for repeating the names of the Buddha Trinity, namely, Phra or Buddha, Tara or Dharma [law], and Sangha | the congrega- tion], and the number of their beads in their rosary is a multiple of 5 by 35, as with the Jamas,” NO. 1667. COLLECTION OF ROSARIES—CASANOWICZ. 337 of pearl and other precious stones, or of silver and gold. The yellow rosary made from the ochery yellow wood, supposed to be from the Bodhi tree, usually in form of spherical beads about the size of a pea, is used for all kinds of devotions. But prized above all are beads made from the bones of a holy lama. Lay people, however, use rosaries composed of any sort of bead, and the same chaplet may contain beads of a variety of sizes, materials, and colors. The full Tibetan rosary of 108 beads is usually divided by three beads of a different size or material into four groups of 27 beads each. The two ends of the string before being knotted are passed through three extra beads, called do dzin (spelled rdog hdzin), “ retaining beads ” or “ union holders,” as they keep the proper rosary beads in position and indicate the completion of a cycle. They symbolize the Buddhist triad—the Buddha, the doctrine (dharma) and the community (sangha). Attached to the main string are two small pendant strings, having each 10 smaller beads, or metal rings, one terminating in a miniature dorje or vajra (the conventionalized thunderbolt of Indra), the other in a tiny bell (drilbu). These pendants are used as counters (drang dzin) to keep count of the number of times the rosary is said. A bead of the dorje string is sid down to mark a single recital of the rosary, while those of the bell string note each ten repetitions. They thus serve to register the utterance of 108 multiplhed by 10 multiplied by 10 equaling 10,800 prayers or formulas. In the beads of lay people both counter strings record only units of cycles, which suffice for the smaller amount of bead telling done by the laity. Sometimes there are two additional pendants terminating respectively in a magic peg (purbu) and a wheel (Aor fo). There are also attached to the rosary string small odds and ends, such as keys, tweezers, toothpicks, ete. The formula most frequently repeated by means of the rosary, and which is uttered at the conclusion of any other prayer that may be recited, is Om mani padme hum! which is commonly rendered “ Salutation to the jewel in the lotus flower!” in allusion to Pad- mapani (Sanskrit Avalokiteshvara), the mystical reflex or repre- sentative of Buddha, who is believed to have: appeared on earth from a lotus flower. He is held in special veneration in Tibet as the protector and patron of the country, and is believed to be -reincar- nated in the Dalai Lama, the head of Tibetan Buddhism, by the - emission of a beam of light. _ 1. Tibetan rosary.—Consisting of 108 disk-shaped shell beads, divided into four groups of 27 beads each by three red coral beads. The three retaining beads are a large spherical amber bead, a smaller ; disk-shaped one, and a conical one of coral. The four counter _ strings, with 10 silver beads on each, terminate in various ornaments. This form of rosary is in common use among the lamas. Length, Proc. N. M. vol. xxxvi—09——22 » 338 PROCEEDINGS OF THE NATIONAL MUSEUM. vou. xxxvt. 25 inches. Kumbum, Tibet. (Plate 21, Cat. No. 167271, U.S.N.M.) Lent by Hon. W. W. Rockhill. 2. Tibetan rosary.—Consisting of 108 disks cut from human skulls, divided into four sections of 27 each by three larger disks of conch shell, with two retaining beads of amber and wood, respectively, but without counters. Such rosaries are especially used in the worship of Dorje jig-ch’e (Sanskrit, Yama), the king of the dead. Length, 25 inches. Tibet. (Cat. No. 130387, U.S.N.M.) Lent by Hon. W. W. Rockhill. 3. Tibetan rosary.—Made of small disks‘of rosewood, with four red coral beads as dividers. It has no counters, and the dividing beads, as also the three retaining ones, have to be counted to complete the number of 108. Beads of reddish color, usually of red sandalwood, are used in the worship of the fierce Tamdrin, the special protector of Lamaism. Length, 15 inches. Ta-chien-lu, China. (Plate 22, fig. 1, Cat. No. 167267,,.U.S.N.M.). Lent by Hon. W. W. Rockhill. 4. Tibetan rosary—The same as No. 38. Ta-chien-lu, China. (Cat. No. 167267, U.S.N.M.). Lent by Hon. W. W. Rockhill. 5. Tibetan rosary—Consisting of 108 disks of yellow wood, with the dividing beads of the same material, only shghtly larger and thicker. It has only two retaining beads and no counters. It is the special rosary of the Gelupa, or reformed school of lamaism. Length, 25 inches. Batang, China. (Plate 22, fig. 2, Cat. No. 131058, U.S.N.M.). Lent by Hon. W. W. Rockhill. 6. Tibetan rosary —tConsisting of 108 spherical beads of yellow wood, without counters and with only one retaining bead. Said to have been brought from Lhasa, the holy capital city of Tibet. Length, 39 inches. Ladakh, Tibet. (Cat. No. 178120, U.S.N.M.) Gift of Dr. W. L. Abbott. 7. Tibetan rosary.—The same as No. 6, only the beads are smaller in size. Length, 26 inches. Ladakh, Tibet. (Cat. No. 178119, U.S.N.M.) Gift of Dr. W. L. Abbott. B. CHINESE ROSARIES. The Chinese name for rosary is su-chu. The full or long rosary consists, like the Tibetan, of 108 beads, and is also usually divided by three beads of a different size or color into four groups. The shorter rosary has 18 beads, corresponding to the 18 chief disciples of Buddha, or /ohans. The ends of the string are passed through two retaining beads, a large globular one and a smaller oblong or oval one. The large bead sometimes contains a sacred relic or charm. “The rosaries lent by Mr. Rockhill have also been described by their owner in Notes on the Ethnology of Tibet, by William Woodville Rockhill, in the Report of the U. S. National Museum for 18938, pp. 736-738 and pls. 35-387. No. 1667, COLLECTION OF ROSARIES—CASANOWICZ. 339 The Chinese official necklace, worn by dignitaries on state occa- sions, is the Buddhist rosary which was made a part of the court costume. These official sw-chus are often made of costly materials and adorned with fine carvings. They are here represented by the following two numbers. 8. Chinese official “ su-chu.”—The 108 beads of the main string are palm wood balls five-eighths of an inch in diameter. The dividing, as also retaining, beads are of silver, richly enameled, measuring 14 inches in diameter. The three counter strings have each ten beads, likewise of enameled silver, but of smaller size, being only one-half inch in diameter. From the retaining beads is suspended:a silk ribbon embroidered with small glass beads of diverse colors to repre- sent the Swastika and other symbols, with a silver enameled medal- lion, measuring 2+ by 1? inches, in the center, and terminating in an oblong or oval bead 2 inches long. Such an oval bead is also at the end of each of the three counter strings, each 14 inches long. They are called the “ four dewdrops,” which they resemble, or the “ dis- ciple beads,” or the “ regents of the four heavens.” They typify the emperor, father, mother, and the teacher to whom a Chinese sub- ject owes reverence and obedience. Length, 8 feet. China. (Plate 23, Cat. No. 202869, U.S.N.M.) Gift of Mr. Yang Yu, Chinese minister to the United States, 1897. 9. Chinese official “ su-chu.”—Made of glass beads. The 108 beads of the main string, five-eighths of an inch in diameter, are amber colored; the dividing and retaining beads, 1 inch in diameter, are green, while those on the counter strings and the medallion or disk on the pendant ribbon are of rose color. Length, 3 feet 8 inches. China. (Plate 24, Cat. No. 5559, U.S.N.M.) Lent by Miss Eliza R. Scidmore. 19. Chinese rosary.—Consisting of 108 globular beads made of plum stones. Finely carved, so that on each bead, measuring one- half of an inch in diameter, are seen five human figures in the midst of flowers and trees. Length, 4 feet 7 inches. China. (Cat. No. 5526, U.S.N.M.) Lent by Miss Eliza R. Scidmore. 11. Chinese rosary.—Consisting of 108 globular beads made of rhinoceros horn, terminating in two retaining beads and a tassel of white silk. Length, 4 feet 8 inches. China. (Cat. No. 5541, U.S.N.M.) Lent by Miss Eliza R. Scidmore. 12. Chinese rosary.—Consisting of 108 globular beads made of ebony. Used by: pilgrims. Length, 41 inches. China. (Cat. No. 5540, U.S.N.M.) Lent by Miss Eliza R. Scidmore. 13. Chinese rosary.—Consisting of 108 ovoid beads made of ebony, with the dividing and retaining beads of reddish peat) Length, 40 “The information on the Chinese rosary is Thaelee derived Arar Miss Scid- more’s notes. 840 PROCEEDINGS OF THE NATIONAL MUSEUM. VOL. XXXVI. inches. China. (Cat. No. 5521, U.S.N.M.) Lent by Miss Eliza R. Scidmore. 14. Chinese rosary.—Consisting of 108 globular beads made of polished palm wood, with the dividing and retaining beads of white glass. Length, 44 inches. China. (Cat. No. 5544, U.S.N.M.) Lent by Miss Eliza R. Scidmore. 15. Chinese rosary.—Consisting of 108 pearl-colored glass beads; the dividing and retaining beads are green. Length, 35 inches. China. (Cat. No. 5522, U.S.N.M.) Lent by Miss Eliza R. Scidmore. 16-18. Three Chinese rosaries.—Consisting each of 108 globular beads made of black wood. Length of each, 36 inches. Hoihau, China. (Cat. No. 154242, U.S.N.M.) Collected by Dr. Julius Neumann. 19. Chinese rosary.—Consisting of 18 olive-shaped beads, probably made of some wax or resin composition, each being carved into an image of one of the 18 Johans, or saints, with their special attributes. (See illustration to No. 50.) The term lohan (Japanese, rohan; San- skirt, arhant) is applied to those disciples and followers of Buddha who have attained the highest degree of perfection. Length, 23 inches. China. (Cat. No. 130388, U.S.N.M.) Lent by Hon. W. W. Rockhill. 20. Chinese rosary—tConsisting of 18 peach-stone shaped beads, probably made of some wax composition. Each bead represents in low relief on one side the image of a lohan with his attribute, on the other the grotesque head of a demon. With two retaining beads of lapis lazuli and agate, respectively. Length, 17 inches. China. (Cat. No. 55138, U.S.N.M.) Lent by Miss Eliza R. Scidmore. 21. Chinese rosary.—Consisting of 18 beads made in shape of wal- nut shells, but probably of some wax composition. On each bead is carved in low relief, on one side, the image of a Johan, on the other a Chinese inscription, perhaps the formula Omito Fat (“O, infinite Buddha!”), which is usually repeated by Chinese Buddhists on the rosary. With one retaining bead of agate. Length, 23 inches. China. (Cat. No. 5507, U.S.N.M.) Lent by Miss Eliza R. Scidmore. 22. Chinese rosary.—Consisting of 18 beads made of plum stones, each carved into the head of a lohan. With one retaining bead of malachite. Length, 144 inches. China.. (Cat. No. 5508, U.S.N.M.) Lent by Miss Eliza R. Scidmore. 23. Chinese rosary.—Consisting of 18 beads made of plum stones, cut into the form of vases with flowers Length, 17 inches. China. (Cat. No. 5510, U.S.N.M.) Lent by Miss Eliza R. Scidmore. 24. Chinese rosary.—Consisting of 18 beads made of peach stones, each finely carved in intaglio with the figure of a Johan with his spe- cial attribute, surrounded by flowers and trees. Length, 20 inches. China. (Cat. No. 5515, U.S.N.M.) Lent by Miss Eliza R. Scidmore. No. 1667. COLLECTION OF ROSARIES—CASANOWICZ, 841 25. Chinese rosary.—Consisting of 18 beads made of peach stones, each cut into the form of the fish-shaped templed drum, called in Chinese mo-yii, in Japanese mokugio. Length, 19 inches. China. (Cat. No. 5509, U.S.N.M.) Lent by Miss Eliza R. Scidmore. : 26. Chinese rosary.—Consisting of 18 beads made of the dried and polished fruit of Hlaeocarpus. Length, 18 inches. China. (Cat. No. 130403, U.S.N.M.) Lent by Hon. W. W. Rockhill. 27. Chinese rosary.—Consisting of 18 beads made of the fruit of the Trapa bicornis of China, which resembles a buffalo’s head with two blunt horns. Length, 21 inches. China. (Cat. No. 5512, U.S.N.M.) Lent by Miss Eliza R. Scidmore. 28. Chinese rosary.—Consisting of 18 oblong amber beads, with two retaining beads. Length, 17 inches. China. (Cat. No. 5503, U.S.N.M.) Lent by Miss Eliza R. Scidmore. 29. Chinese rosary.—Consisting of 50 beads of alternating bone cylinders and oval blue glass pearls. Attached to the retaiming bead is a lizard or marmot of jade. Length, 26 inches. China. (Cat. No. 5518, U.S.N.M.) Lent by Miss Eliza R. Scidmore. C. JAPANESE ROSARIES. In the Japanese jiu-dzu the Buddhist rosary attained its highest development. The sho-zoki jiu-dzu, or the rosary used by all sects in common, consists of 112 beads of a uniform size, divided by two large beads, called oya-dama, or parent beads, into two equal parts. They are distinguished into upper parent bead, ten-no oya-dama, also called, father, sun, Buddha, ete., and lower parent bead, chz-no oya-dama, mother, moon, Bo, divine spirit, which inspired and per- fected the enlightenment of Buddha. The ends of the string before being knotted are drawn through the two parent beads which have for this purpose a third opening. From the upper parent bead ex- tend two pendent strings on which are strung 21 beads, rather smaller than those on the main string, in the following manner: Immediately above the large parent bead, on the left side pendent string, is a solitary bead. Beyond this the strings are knotted. Then come five beads on each string when they are again knotted. Still again there are another five beads on each pendant, which then terminates in an elongated bead, called dewdrop, tswyu-dama. The use of the solitary bead is that in holding the rosary, with the upper parent bead uppermost, it should be in the left hand; this will insure a right signification to each bead during prayer. The collective _ name of these pendent beads is kami-deshi, superior disciples. Ex- tending from the lower parent bead are three strings on two of which are five small beads, called shima-deshi, or inferior disciples, each terminating in a dewdrop bead, while the third has ten small beads, without a dewdrop. They are used as counters and are called 842, PROCEEDINGS OF THE NATIONAL MUSEUM. VOL. XXXVI. khadzu-tori. The four dewdrop beads are also termed shi-ten-no, the four regents who are supposed to preside over the four quarters of the universe. The rosary represents metaphorically the Buddhist pan- theon; and the position of the four dewdrops at the ends of the strings on which all the other beads are strung, thus keeping in har- mony and order the entire rosary as it is intended to be used, is sup- posed to be symbolic of their actual positions of power and authority in the universe. On the main string, at an interval of seven beads on either side from the upper parent bead, are two beads, rather smaller than the others and generally of a different material, and again, at an inter- ral of fourteen beads from these, on either side, are other two of the same kind. They are sometimes erroneously called shi-ten-no, the four regents, or shi-bosatsu, the four saints. They indicate where a special invocation is to be uttered while the rosary is lifted to the forehead with a reverence. A smaller rosary of 16 beads, corresponding to the 16 Japanese rohans, or chief disciples of Buddha (analogous to the 18 dohans of the Chinese), is chiefly used by lay people on ceremonial and social oceasions. It has only one parent bead, or oya-dama, and one elon- gated, tapering bead in form of a vase or pagoda (similar to the retaining beads in the Tibetan rosary), called fwsa-dome, “tassel stopper,” and terminates in a silk tassel. Frequently it is spaced by two saints’ or busatsu beads of a different substance. Moreover, the Japanese rosary varies in the number as well as the arrangement of the beads with the different sects. The rosary, according to Miss Scidmore, who traveled extensively in Japan, plays an important part not only in the religious life but also in the social etiquette of Japan. It is carried by monks and lay people on all occasions of religious celebrations, on visits of cere- mony or condolence, at funerals, etc. There is always a hook on the wall or on posts of the ceremonial or tea room, on which to hang the jéu-dzu, and a amique or historic rosary is a much appreciated ornament for a tea room. Among the treasures of the Imperial Museum in Tokyo is the jiéu-dzu of the regent Shotoku Taishi, the Constantine of Buddhism in Japan, who died in 621 A. D. All the soldiers in the late Russo-Japanese war carried rosaries with them. The dead are always buried or cremated with a rosary slipped on the wrist, and the mourners in a funeral procession likewise carry each a rosary. Jiu-dzu shops, marked by a gigantic rosary on the outside, flourish at every place of popular pilgrimage and at some of the larger tem- ples, and a rosary that has been consecrated over the sacred flame and incense smoke of a venerated temple is greatly valued by the devout. No. 1667. COLLECTION OF ROSAPIES—CASANOWICZ,. 343 30. Japanese rosary.—Consisting of 112 globular beads made of cherry wood. It is the sho-zoki jiu-dzu described above, which is used by all sects. The parent, disciple, regent, and saints’ beads are of the same material, differing only in size. Length, 6 feet 6 inches. Japan. (Cat. No. 130, 683, U.S.N.M.) Collected by Mr. Romyn Hitchcock. 31-32. Two Japanese rosaries.—Consisting each of 112 globular beads made of plum-tree wood. The same as the preceding No. 30. Length, 6 feet. Japan. (Cat. No. 130688, U.S.N.M.) Collected by Mr. Romyn Hitchcock. 33. Japanese rosary.—Consisting of 112 small globular beads made of cherry wood. Used by the Nichiren sect, which was founded in the middle of the thirteenth century A. D. Its rosary is similar to the sho-zoki jiu dzu, differing only in the size of the beads, which, as a rule, are very small for convenience of carrying and for being more easily manipulated. Length, 32 inches. Japan. (Cat. No. 5525, U.S.N.M.) Lent by Miss Eliza R. Scidmore. 34. Japanese rosary.—Consisting of 112 beads made of mother of pearl. The two parent beads are of amber, the four spacing or saints’ beads are of red coral. This rosary is used by the Shin-Gon sect, which was founded 805 A. D. Length, 4 feet. Japan. (Plate 25, fig. 1, Cat. No. 5555, U.S.N.M.) Lent by Miss Eliza R. Scidmore. 35. Japanese rosary—Consisting of 112 beads made of smooth peach stones. The beads on the pendant strings, as also the parent and spacing beads, are of glass. Length, 8 feet. Japan. (Cat. No. 5545, U.S.N.M.) Lent by Miss Eliza R. Scidmore. 36. Japanese rosary.—Consisting of 112 beads of some dark-brown seeds, with the beads on the pendant strings and parent and spacing beads of glass. Length, 26 inches. Japan: (Cat. No. 5550, U.S.N.M.) Lent by Miss Eliza R. Scidmore. 37. Japanese rosary.—Consisting of 112 flattened beads made of ebony. Used by the Zen sect, which was founded at the beginning of the thirteenth century A. D. This rosary has the two parent and four spacing or saints’ beads, but no pendant strings with their dis- ciple beads. The ends of the strings run out from the upper parent bead, extending about 4 inches in length and terminating in a knot without tassel. The four spacing or saints’ beads are here placed at intervals of 18 beads, so that by means of the two parent and four saints’ beads the string is divided into six sections of 18 beads each. The parent and saints’ beads are of glass. Length, 28 inches. Japan. (Cat. No. 5528, U.S.N.M.) Lent by Miss Eliza R. Scid- more. 38. Japanese rosary.—Consisting of 174 flattened beads made of black wood. The parent, pendant, and spacing beads are of glass. Perhaps used by the lay people of the Zen sect. Length, 6 feet. 344 PROCEEDINGS OF THE NATIONAL MUSEUM. vou. xxxvt. Japan. (Cat. No. 5547, U.S.N.M.) Lent by Miss Eliza R. Scid- more. 39. Japanese rosary.—Consisting of 100 flat beads made of plum- tree wood, without spacing beads. Length, 4 feet 10 inches. Japan. (Cat. No. 5519, U.S.N.M.) Lent by Miss Eliza R. Scidmore. 40. Japanese rosary.—Consisting of 82 globular glass beads. The pendants of the lower parent bead are without beads. There are also wanting the spacing beads. Perhaps used by the Monto or Ikkoshin sect, which was founded at the beginning of the thirteenth century A.D. Length, 22 inches. Japan. (Cat. No. 5548, U.S.N.M.) Lent by Miss Eliza R. Scidmore. 41. Japanese rosary.—Consisting of 80 beads made of the fruit of Elaeocarpus, dried and polished. At an interval of ten beads on either side of the parent beads are three smaller glass beads. The pendant strings have likewise glass beads. Length, 5 feet 3 inches. Japan. (Cat. No. 5543, U.S.N.M.) Lent by Miss Eliza R. Scidmore. 42. Japanese rosary.—Consisting of 58 beads made of carved cherry stones. With one parent bead of glass followed by an elon- gated stopper bead (fusa-dome) of mother-of-pearl and terminating in a tassel formed of eight silk cords. The four spacing beads are of amber. Length, 24 inches. Japan. (Cat. No. 5517, U.S.N.M.) Lent by Miss Eliza R. Scidmore. 43. Japanese rosary.—Consisting of 69 glass beads, without any attachments excepting two tassels. It was hung on the arm of a temple image in Kioto. Length, 25 inches. Japan. (Cat. No. 5535, U.S.N.M.) Lent by Miss Eliza R. Scidmore. 44. Japanese rosary.—Consisting of 54 beads made of the nuts of the Pride of India (also known as tree of paradise, bead tree, or holy tree—Melia azedarach). The parent beads are of black wood, while those on the pendant strings are of glass. There are no dewdrops nor spacing beads. Length, 41 inches. Japan. (Cat. No. 5542, U.S.N.M.) Lent by Miss Eliza R. Scidmore. 45. Japanese rosary.—Consisting of 100 beads of “ Job’s tears” (Cotx lachryma-jobi), with only one pendant string from either par- ent bead, the other evidently having been worn off. Length, 4 feet. Japan. (Cat. No. 5534, U.S.N.M.) Lent by Miss Eliza R. Scid- more. 46. Japanese rosary.—Consisting of 192 beads made of black wood. It is probably made up of two different strings, as the beads are of unequal size. At irregular intervals are two or three glass beads. From either of the parent beads extend two strings with tassels, but without beads. Length, 5 feet-4 inches. Japan. (Cat. No. 5539, U.S.N.M.) Lent by Miss Eliza R. Scidmore. 47. Japanese primitive necklace, so-called “ Shinto rosary.”—It consists of a string of 30 glass pieces in regular alternation of one No. 1667. COLLECTION OF ROSARIES—CASANOWICZ. 845 in form of the toe of a bear (the sacred animal of the Ainus), one of a globe, and the third of a tube or cylinder, with one of the latter serving as a tassel stopper (fusa-dome). Length, 25 inches. Japan. (Plate 25, fig. 2. Cat. No. 5520, U.S.N.M.) Lent by Miss Eliza R. Scidmore. 48. Japanese rosary.—Made of flattened mahogany beads peculiar to the Jodo sect, which was founded by Honen Shonin at the end of the twelfth century A. D. Its rosary consists of two strings of beads reeved one within the other. One usually has 40 flat beads with one parent bead; the other 27 of the same size as the 40, alternating with 28 smaller ones, and likewise one parent bead, thus making a total of 95 beads, exclusive of-the two large parent beads. On the second larger string is a metal ring, sufficiently large to allow the string to pass freely through it. Attached to this ring are two pendant strings, on one of which are ten small round beads, on the other six, both terminating with dewdrop beads. On the smaller string of 40 beads the single prayers or formulas are recited, while the larger string of 55 and the two pendant strings with their 16 beads are used as two sets of counters in the following way: The string with 40 beads is placed, with the parent bead uppermost, over the first joint of the forefinger, while the other string with 55 beads is held between the second and third fingers of the same hand and used as a first set of counters. The upper string is then turned by the thumb, one bead at a time for each prayer or formula uttered, beginning with the bead next to the parent bead, until it comes round to its starting point, when one bead of the lower string, starting likewise from the parent bead, is slipped through between the fingers, one bead for every revolution of the upper string, until the whole has been ex- hausted, when recourse is had to one of the small pendant beads to register the fact. ‘The whole process has then to be gone over again, so that by the time the whole of the 16 beads has been used 35,200 prayers will have been recited. The invention of this double rosary is ascribed to Awanosuke, one of the personal attendants of the founder of the Jodo sect, its ob- ject being that it should be manipulated only with the left hand, thereby leaving the right hand free to minister to the needs of his master, thus combining praying and working at one and the same time. In the present example the upper string has 36 and the lower 30 beads, all of the same size. Length, 28 inches. Japan. (Plate 26, fig. 1. Cat. No. 5527, U.S.N.M.) Lent by Miss Eliza R. Scidmore. 49. Japanese rosary.—Consisting of 18 beads made of walnut shells cut in the shape of skulls. Upon the parent bead are carved two groups of nine figures each, representing the 18 disciples (rohans). The tassel stopper is of mother-of-pearl. The two cords which extend from the parent bead are tied with three peculiar knots. 8346 PROCEEDINGS OF THE NATIONAL MUSEUM. VOL. XXXVI. . Length, 27 inches. Japan. (Plate 26, fig. 2. Cat. No. 5516, U.S.N.M.) Lent by Miss Eliza R. Scidmore. . 50. Japanese rosary.—Consisting of 12 olive-shaped beads, prob- ably made of some wax or resin composition, each being carved into an image of a saint, with his special attribute (compare. above No. 19). The parent bead and tassel stopper are of jade. Length, 27 inches. Japan. (Plate 26, fig. 3. Cat. No. 5505, U.S.N.M.) Lent by Miss Eliza R. Scidmore. 51. Japanese rosary.—Consisting of 16 beads of the same. material and workmanship as those of No. 50. The parent bead, the tassel stopper, and two spacing beads are of agate. Length, 21 inches. Japan. (Cat. No. 5504, U.S.N.M.) Lent by Miss Eliza R. Scidmore. 52. Japanese rosary.—Consisting of 16 beads in shape of walnut shells, but probably made of some wax composition. On each bead is carved in low relief, on one side, the image of a saint, on the other, some animal or bird. The tassel stopper is of agate. Length, 28 inches. Japan. (Cat. No. 5506, U.S.N.M.) Lent by Miss Eliza R. Scidmore. 53. Japanese rosary.—Consisting of 18 beads of plum stones. On each are finely carved in intaglio four human figures, surmounted by an open lotus flower and surrounded by plants and animals. The parent bead is of amber, while the tassel stopper and two spacing beads are of glass. Length, 22 inches. Japan. Cat. No. 5511, U.S.N.M.) Lent by Miss Eliza R. Scidmore. 54. Japanese rosary.—Consisting of 27 beads of smooth plum stones. The parent bead and two spacing beads are of white glass, while the tassel stopper is of green glass. Length, 18 inches. Japan. (Cat. No. 5551, U.S.N.M.) Lent by Miss Eliza R. Scidmore. 55, Japanese rosary.—Consisting of 26 beads of smooth plum. stones. The two spacing beads and the tassel stopper are of green glass. Length,18 inches. Japan. (Cat. No. 5552,U.S.N.M.) Lent by Miss Eliza R. Scidmore. 56. Japanese rosary.—Consisting of 21 beads of smooth plum stones, with four spacing beads of pink-colored glass. Length, 21 inches. Japan. (Cat. No. 5529, U.S.N.M.) Lent by Miss Eliza R. Scidmore. : 57. Japanese rosary.—Consisting of 21 beads of smooth plum stones. The parent bead is of white glass, the tassel stopper of bone, and the two dividing beads are of green glass. Length, 15 inches. Japan. (Cat. No. 5531, U.S.N.M.) Lent by Miss Eliza R. Scid- more. 58. Japanese rosary.—Consisting of 20 beads of the fruit of 7aeo- carpus, dried and polished. The parent bead and tassel stopper are of porcelain, while the two spacing beads are of amber. Length, 12 inches. Japan. (Cat. No. 5549, U.S.N.M.) Lent by Miss Eliza R. Scidmore. —ee a See | ee NO, 1667. COLLECTION OF ROSARIES—CASANOWICZ. 347 59. Japanese rosary.—Consisting of 18 beads made of palm nuts. The parent bead is of glass, the tassel stopper of mother-of-pearl. Length, 26 inches. Japan. (Cat. No. 5530, U.S.N.M.) Lent by Miss Eliza R. Scidmore. 60. Japanese rosary.—Consisting of 18 beads made of palm nuts. The parent bead and the spacing beads are of glass, while the tassel stopper is of porcelain. Length, 20 inches. Japan. (Cat. No. 5538, U.S.N.M.) Lent by Miss Eliza R. Scidmore. 61. Japanese rosary.—Consisting of 16 beads made of palm nuts. The parent bead and tassel stopper are of porcelain, the two spacing beads of yellow glass. Length, 15 inches. Japan. (Cat. No. 5536, U.S.N.M.) Lent by Miss Eliza R. Scidmore. 62. Japanese rosary.—Consisting of 27 beads made of cumuna pods. The two spacing beads are of reddish agate, the parent bead is of glass, and the tassel stopper of bone. Length, 134 inches. Japan. (Cat. No. 5537, U.S.N.M.) Lent by Miss Eliza R. Scid- more. 63. Japanese rosary.—Consisting of 27 beads made of dwarfed peach stones, with the parent bead of black wood. Length, 13 inches. Japan. (Cat. No. 5533, U.S.N.M.) Lent by Miss Eliza R. Scid- more. 64. Japanese rosary.—Consisting of 20 beads of the berries of Pride of India (Melia azedarach). The parent bead and tassel stopper-are of porcelain, the two dividing beads are of glass. Length, 12 inches. Japan. (Cat. No. 5554, U.S.N.M.) Lent by Miss Eliza R. Scid- more. ? 65. Japanese rosary.—Consisting of 20 beads of the berries of Pride of India (JJelia azedarach), with one parent bead of glass. Length, 13 inches. Japan. (Cat. No. 5532, U.S.N.M.) Lent by Miss Eliza R. Scidmore. 66. Japanese rosary.—Consisting of 19 beads of pine nuts. The parent bead is of glass, the tassel stopper of bone. Length, 17 inches. Japan. (Cat. No. 5502, U.S.N.M.) Lent by Miss Eliza R. Scid- more. 67. Japanese rosary.—Consisting of 18 beads made up of various fruits, nuts, and berries, with two parent beads of fruit and tassel stopper of bone. Length, 23 inches. Japan. (Cat. No. 5553, U.S.N.M.) Lent by Miss Eliza R. Scidmore. 68. Japanese.rosary.—Consisting of 34 globular beads made of red wood. The two spacing beads and the parent bead are of glass, while the tassel stopper is of horn. Length, 19 inches. Japan. (Cat. No. 5546, U.S.N.M.) Lent by Miss Eliza R. Scidmore. 69. Japanese rosary.—Consisting of 23 beads of alternating wooden models of a pagoda and beads of mother-of-pearl, quartz, and glass, with parent bead and tassel stopper of agate. Length, 12 inches. 848 PROCEEDINGS OF THE NATIONAL MUSEUM. VOL. XXXVI. Japan. (Cat. No. 5514, U.S.N.M.) Lent by Miss Eliza R. Sceid- more. 70. Japanese funeral rosary from the Senkoji temple—Consisting of 11 cylindrical wooden beads roughly cut. It is worn around the wrist by mourners at a funeral. Length, 9 inches. Japan. (Cat. No. 5501, U.S.N.M.) Lent by Miss Eliza R. Scidmore. 71. Japanese rosary.—Used for the Hiaku mam-ben devotion, when the formula, Vamu Amida Butsu/, “ Hail, infinite Buddha!” (short- ened into Vem-butsu/), which the Japanese usually repeat by means of the rosary, is recited a million times. This special devotion was instituted in Kioto in 1831 A. D., on the occasion of a devastating plague, and its celebration is reserved for times of calamity, such as pestilence, war, and famine. At certain popular temples, however, it is almost continually observed by the pilgrims. For this service a rosary of 1,008 large wooden beads is used. The present specimen consists of 897 flat wooden beads, with 2 parent beads, from one of which extend 2 pendent strings with 5 smaller beads on each. Length, 20 feet. Japan. (Cat. No. 5556, U.S.N.M.) Lent by Miss Eliza R. Scidmore. Ill. THE MOHAMMEDAN ROSARY. The Mohammedan rosary, called swbha, in Persia, tasbih (from the Arabic verb sabbaha, “to praise,” “to exalt”), consists of 99 beads, divided into three equal portions by a stone or bead of dif- ferent shape or, in the more costly varieties, by tassels, called shamsa (“servant ”), made of gold thread or variegated silk. The Moham- medans use the rosary for the recital of the 99 attributes of God, as, “the mighty ” (al-aziz), “the holy” (al-kuddus), “the merciful” (av-rahman), “ the loving ” (al-wadud), “ the forgiver ” (al-ghafar), etc. A hundredth bead of larger size, called the imam (“leader”), or a tassel in its place, is frequently added for the essential name of God, Allah. Other devotional formulas recited by means of the rosary, are the ejaculations known as the takbir: “God is very great ” (Allahu akbar); the tasbih: “I extol God” (subhana illah); the tahmid: “God be praised” (al-hamdu Villahi), and the tahlil: “There is no deity but God” (la ilaha illa illah). Great merit, ac- cording to tradition, is attributed by the prophet to the recital of the hundred names of God, or to the repetition of these formulas. “ Verily,” he is reported to have said, “ there are ninety-nine names of God, and whoever recites them shall enter into Paradise,” and “ Who- ever recites this sentence (the tasbih and tahmid) a hundred times, morning and evening, will have all his sins forgiven.” 4 According to Mr. R. A. Stewart Macalister, in the Palestine Exploration Fund Quarterly Statement for July, 1908, p. 172, “There is another variety of rosary less commonly used, with 101 pellets corresponding to the 101 names of the Prophet.” NO. 1667. COLLECTION OF ROSARIES—CASANOWICZ. 849 Mohammedan rosaries are frequently made of date stones. Special value is attached to beads, the material of which originated in the sacred cities of Mecca and Medina. It is generally assumed that the Mohammedans borrowed the rosary full-grown from the Buddhists. The Mohammedan tradition (hadith) pushes back the use of some mechanical contrivance for counting prayers to the time of Mohammed. It is related that the prophet reproached some women for using pebbles in repeating the tasbih, takbir, ete., and recommended that they should count them on their fingers. In a tradition, collected in the third century A. H. (ninth century A. D.), is related that Abu Abd al-Rahman, son of Abu Bekr, the first calif, who died about 53 A. H. (673 A. D.), see- ing in the mosque groups of worshipers, reciting under a leader 100 takbirs, 100 takhlils, and 100 tasbihs by means of small pebbles, re- proached them with the words, “ Rather count your sins, and I shall guarantee you that nothing of your good works will be lost.” Ab- dallah, son of the calif Omar, who died 73 A. H. (692 A. D.), seeing one picking up pebbles while praying, said to him, “ Do not do that, for this comes from Satan.” All this may point to the adoption of some counting device at the time when the recitation of the above- mentioned formulas became a practice, the date of which, however, can not be fixed with certainty. The use of pebbles in the repetition of these litanies would seem to mark a primitive form of the subha, the point of departure in the evolution which resulted in the rosary, that is, in threading beads on a string, which may have been copied from the Buddhists. It also shows that the rosary at the time of its appear- ance met with some opposition from the conservatives and the rigorists of the religious discipline. In fact, as late as the third century A. H. (ninth century A. D.) the use of the swbha, as an instrument of prayer, was in vogue only among the lower classes and looked down upon by the theologians and higher classes. When the pious ascetic Abu-I-Kassim al-Gunejd (died 279 A. H.—909 A. D.) was found with a rosary and expostulated with, since he “ belonged to the better world,” he apologized with the words, “I could not renounce an object which was the means of bringing me nearer to God.” Even in the seventh century A. H. (thirteenth A. D.) Abu Abdallah Mam- med al-Abdari, called Ibn al-Hajj (died 737 A. H.—1336 A. D.), complains over the exaggerated use and esteem of the swbha as being contrary to the primitive simplicity of Islam. The Wahabis, followers of the reformer Abd al-Wahhab (1691- 1787 A. D.), who opposed all practices not sanctioned by the Koran and tradition, regard the rosary as an abomination and count the ’ names of God on their fingers. 72. Mohammedan rosary.—Consisting of 100 globular beads made of olive wood, divided into three sections by two vase or bottle- 850 PROCEEDINGS OF THE NATIONAL MUSEUM. VOL. XXXVI. shaped beads. The two ends of the string pass first through the hun- dredth bead, then through a fusiform or spindle-shaped tube, and lastly through two smaller beads, terminating in a green tassel. Length, 40 inches. Cairo, Egypt. (Plate 27, fig. 1. Cat. No. 155166, U.S.N.M.) 73. Mohammedan rosary.—Consisting of 97 beads made of blood- stones, with two dividing beads of chalcedony. Between the ninth and tenth beads from one end of the string an oblong rectangular piece of chalcedony is inserted, probably to complete, with the two dividers, the number 100. The string terminates in a tassel of silk and gold thread tied with an artistic knot. Length, 34 inches. (Plate 97, fig. 2. Cat. No. 179075, U.S.N.M.) Collected by Dr. G. Brown Goode. 74. Mohammedan rosary—Consisting of 91 beads made of horn. On either side of the two dividing beads, which are vase shaped and inlaid with silver dots, are three coral beads, and at either end of the string two coral and one amber beads. The ends of the string pass through two small beads of horn and a fusiform tube, terminating in a green tassel. Length, 31 inches. (Plate 27, fig. 3.° Cat. No. 179075, U.S.N.M.) Collected by Dr. G. Brown Goode. 75. Mohammedan rosary.—Consisting of 102 beads made of com- position, alternating three brown-colored and one, somewhat larger, black. Length, 5 feet. Monastery of Mount Sinai, Syria. (Plate 27, fig. 4. Cat. No. 154561, U.S.N.M.) Gift of Mrs. Layyah Barakkah. 76. Mohammedan rosary.—Consisting of 100 beads made of black wood, divided into three sections by two beads of bone. The ends of the string are passed through an oblong piece of slate. Length, 45 inches. Paris, France. (Cat. No. 76709, U.S.N.M.) Collected by Mr. John Durand. 77. Mohammedan rosary.—Consisting of 99 beads made of bone, divided into three sections by two date stones. The ends of the string pass through a large bead made from a piece of conch shell. This style of rosary is used by the Mohammedans in China. Length, 30 inches. China. (Plate 22, fig. 3. Cat. No. 167300, U.S.N.M.) Lent by Hon. W. W. Rockhill. IV. THE ROMAN CATHOLIC ROSARY. The ordinary Catholic rosary consists of 150 small beads divided into decades by 15 larger beads. To these beads, forming a chaplet, is usually attached a pendant, consisting of a cross, one large and three small beads. The devofion begins with the invocation, “ In the name of the Father, the Son, and the Holy Ghost.” Then the Apostles’ Creed is recited on the cross, a pater noster (the Lord’s prayer) on the larger bead and three Ave Maria (Hail Mary) on the three smaller beads, closing with the gloria (Glory be to the NO. 1667. COLLECTION OF ROSARIES—CASANOWICZ. 851 Father, etc.). This forms the introduction to the rosary proper. Then follow decades of aves, counted by the smaller beads, each decade preceded by a pater noster, for which a larger bead is used, and followed by a gloria. The 150 aves correspond to the number of Psalms, hence from an early period the devotion was called “ Our Lady’s psalter.” For each decade a subject, or ‘“‘ mystery,” in the life ef Christ and Mary is set for meditation, the 15 mysteries being divided into 5 joyful, 5 sorrowful, and 5 glorious. The 5 joyful mys- teries are: the annunciation (Luke i, 26), the visitation (Luke i, 39), the nativity (Luke 11), the presentation (Luke ii, 21), and the finding in the temple (Luke 11, 41) ; the 5 sorrowful mysteries are: the agony in the garden (Matthew xxvi, 36), the scourging (Matthew xxvii, 26), the crowning with thorns (Matthew xxvii, 29), the carrying of the cross (John xix, 17), and the crucifixion (Matthew xxvii, 35); the 5 glorious mysteries are: the resurrection (Matthew xxviii), the ascen- sion (Luke xxiv, 50), the descent of the Holy Ghost (Acts ii), the as- sumption of Mary into heaven, and the coronation of Mary in heaven (the two last mysteries are accepted on the authority of tradition). This arrangement of definite mysteries does not occur prior to the fif- teenth century. The earlier and more widely accepted practice was to assign an incident of Christ’s life to each ave and to insert some short clause, commemorating the incident, into the ave itself. The rosary most in use, however, consists of five decades of small beads for the aves and five larger beads for the pater nosters, called the “lesser rosary.” Otherwise it is arranged in the same way and recited in the same manner and order as the “ greater” or “ full” rosary. The en- tire devotion of 15 decades may be said on it by counting it three limes.? Rosaries are usually blessed with prayers and holy water by some duly authorized ecclesiastical person and become thereby sacramen- tals, that is, instruments of grace.” The name “rosary” (Middle Latin, rosarium), which came in vogue for the devotion, and the string of beads by which it is per- *In a rosary book entitled: Jesus, Maria, Joseph (dated 1663), the 15 linysteries are comprehended in the following three verses: She’s told, she visits, He’s born, offered, found, F He prays, is whipped, is crowned, carries, is killed, Rises, ascends, sends down: she dies, is crowned. >Outside of the Roman Catholic Church, rosaries are in use among the Sopts in Egypt. They generally consist of 42 beads, or sometimes of 81, and are employed to count the repetitions of the Kyrie eleison (Lord, have merey upon us!). Compare Alfred J. Butler, The Ancient Coptic Church of Egypt, Oxford, 1884, II, p. 238. In the Orthodox Church when a novice is consecrated into the “second grade of monastic life,” he is given, among other things, a chaplet (called in Russian, chotki, in Greek, kombologion, or proseukhe) to count prayers and protestations by. Compare D. Sokolof, A manual of the Orthodox Churches, New York and Albany, 1899, p. 151. 352 PROCEEDINGS OF THE NATIONAL MUSEUM. VOL. XXXVI. formed since the fifteenth century, is commonly explained as a meta- phorical designation, meaning a wreath or chaplet of spiritual roses. The corresponding words, corona, chaplet, Rosenkranz, capellina, all convey the idea of a garland. Garlands of flowers were much worn at that period, and it was also the custom to place such garlands as a mark of respect or admiration upon the heads of persons or statues.2 Rosaréwm was also not uncommonly used (like florilegium) in the sense of an anthology, or a collection of choice extracts. Others trace the name to the title “Mystical Rose,” by which Mary is addressed in the litany of Loretto, or to the beads being originally made, commonly, of rosewood. In the middle ages many other names were applied to prayer beads, as pater noster beads, patriloquium, devotiones, precaria, precula (little prayers), serta (chaplets), nu- meralia, calcula, computum (counters), stgnacula (marks), ete. The word “bead” (beade or bede) originally meant a prayer; to “bid the beads” and to “ pray” were synonymous. The expression “ bedes byddyng” is found in the Vision of Piers the Plowman. So, also, Spenser in his Faerie Queene: All night she spent in bidding of her bedes And all the day in doing good and Godly deeds. In a bull of 1571 Pope Pius V (1566-1572) ascribes the inven- tion of the “ rosary, or Psalter of the Blessed Virgin,” to St. Dominic (1170-1231), the founder of the Dominican order. This has been commonly understood of the string of beads, and the natural infer- ence would be that the suggestion came to western Europe through the crusaders, who observed the Mohammedans using their subha. Legend has it that the Virgin Mary handed St. Dominic a rosary from heaven as a weapon against the Albigense heresy and the in- “A pretty story of a garland which is met with since the beginning of the thirteenth century, and with which the Rev. Herbert Thurston, in the Scientific American, already quoted, would connect the name “rosary,” may find here a place. The legend, as given by Father Thurston, is this: “A youth was ac- customed to make a wreath of roses or other flowers every day and to place it upon the head of Our Lady’s statue. He became a monk, and in the cloister his occupations no longer permitted him to observe this pious practice. Being much distressed, he asked counsel of an aged priest, who advised him to say 50 aves every evening (in some versions it is 150, in others 25), which would be accepted by Our Lady in lieu of the garland. This the young man faithfully observed until one day, being upon a journey, he had to pass through a lonely wood where robbers were lying in wait. They were employed in watching him, feeling sure of their prey, when he, unsuspicious of their presence, remembered that his aves were not yet said and forthwith stopped to say them. Then to their surprise the robbers saw a most glorious lady stand before him and take one after another from the lips of the kneeling monk 50 beautiful roses, which she wove into a garland and placed upon her head. The robbers, so the legend tells, conscience stricken at the vision, were all converted to a better life, and themselves soon after entered the monastery.” “cc \ eee i Le Ta No. 1667. COLLECTION OF ROSARIES—CASANOWICZ. 853 fidels.*. But both the practice of often repeating prayers and the employment of some device for recording the number of repetitions can be traced to a much earlier date, so that St. Dominic can only be considered as the originator and propagator of the present form of the rosary and the method of devotion (150 aves and 15 pater nosters). Thus Sozomenus (about 400-450) relates in his ecclesias- tical history (book v1, chapter 29) that the Egyptian abbot Paul, who died in 341, recited daily 300 prayers which he counted by peb- bles gathered in his cloak, dropping one as he finished each of the prayers. The same means for reckoning prayers is related to have been used by St. Godoric, an English saint who died in 1172. The first undoubted mention of the use of a string of beads for counting prayers is that of Lady Godiva, wife of Leofric, in the eleventh cen- tury, who, when dying, bequeathed to the monastery of Coventry, which was founded by her, “a circlet of gems, which she had threaded on a string, in order that by fingering them one by one, as she succes- sively recited her prayers, she might not fall short of the exact num- ber.” ’ The practice of repeating the same prayer a number of times, often amounting to more than a hundred, must have spontaneously led to the adoption of some contrivance for keeping an accurate rec- ord. It would seem, therefore, that though the Buddhist and Mo- hammedan varieties of bead chaplets preceded the Christian in order of time, there is not necessarily a causal connection between them. As regards the arrangement of the chaplet into 50 or 150 beads, divided into decades, the total number of 150 corresponds, as men- tioned above, to the number of Psalms. For the recital of a certain number of pater nosters, which was originally the prayer repeated on the chaplet, as its designation, pater noster beads, in nearly all Euro- pean languages proves, was a substitute for the Psalms for those monks who had not sufficient education to learn them in Latin. Just as the Psalms were divided into fifties, so that the recitation of 50 or two fifties or three fifties was a common form of devotion, it was natural that 50 paters, or twice or thrice 50, should be enjoined on those who could not read. And as many still used the fingers to count with it was natural to subdivide the beads into tenths. *Jn his encyclical of September 2, 1885, Leo XIII attributes to the power of the devotion of the rosary the suppression of the Albigense heresy in the twelfth and thirteenth centuries, the victory of the Christians over the Turks in the naval battle at Lepanto, near the Echinades Islands, on October 7, 1571, as also in the battle at Temesvar in Panonia and at Corfu in 1716. After the victory of John of Austria over the Turkish fleet at Lepanto Pius V established the festival of “Our Lady of Victory,” which Gregory XIII (1572-1582) two years later changed to the feast of the rosary, which since then has been ob- served on the first Sunday of October as the anniversary of the battle at Lepanto. > William of Malmesbury, Gesta Pontificum Anglicorum, book tv, chapter 2, edition of 1596. Proc. N. M. vol. xxxvi—09——23 354 PROCEEDINGS OF THE NATIONAL MUSEUM. you. xxxvt. The number and arrangement of the beads were, however, not always uniform. Representations on tombs from the fourteenth to the sixteenth centuries exhibit rosaries divided into nines, sevens, sixes, and fives. On some the chaplets count 80, 75, 40, or 33 beads, often without divisions of any kind. . Besides the “ Dominican” rosary, or the “ Marian Psalter,” de- scribed above, which is used in common by all Catholics, there are other varieties of chaplets used by particular religious bodies, or for special devotions. So the chaplet of St. Bridget of Sweden, which consists of 63 beads for the aves, to commemorate the 63 years which Mary is supposed to have lived, divided by seven beads for the paters, the crown of Our Lady, in use among the Franciscans, has 72 aves, based on another tradition of Mary’s age, and others more. During the middle ages the patenotriers, paternosterers, i. e., makers of rosaries, represented an important branch of industry. In Lon- don a street, Paternoster lane, was called after them. In Rome there is still a street, near St. Peters, called Via Dei Coronari—corona being a variety of pater noster, or rosary. The existence of the name in various countries shows that the production of the rosary was a matter of commercial importance. Considerable artistic skill and costly material went into the manufacture of these instruments of piety, which were also worn as personal ornaments. In the inventory of the plate and jewels of Charles V, King of France, in 1380, there are enumerated 19 rosaries made of rose-tinted amber, jet, coral with pearls for markers (sezgnault), gold beads, rings of gold, blue and white enamel, jet beads with eleven gold crosslets (croizettes), black amber and pearls, coral alternating with beads of silver, and two instances of gold beads of Damascus work which were filled with musk. So, again, in the inventory of the Princess of Orleans, Valois, in 1408, there are entered a rosary of amethysts and jasper with a stud (bouton) of pearls, another of jet with nine little bells (dandins) of gold and a jewel with nine pearls as a pendant, and another again of jet with nine gold markers and a gold figurine of St. Christopher attached. Analogous to the attachment of keys, tweezers, etc., to the Tibetan rosary, various objects, such as-signet rings, cameos, brooches were often suspended from the Christian rosary in the middle ages. As a consequence a certain worldliness and extravagance entered into the use of these objects of devotion, which the authorities tried to check. Thus the municipal council of Regensburg, in 1485, decreed that none should possess more than three or four rosaries, and that these should not exceed the value of 10 gulden.* And various monastic @Compare Johannes Janssen, Geschichte des deutschen Volkes seit dem Aus- gang des Mittelalters, Freiburg i. B., I, 8th edition, 1883, p. 877. Janssen adds: “As three fat oxen could then be purchased for 12 gulden, this seems a pretty generous allowance.” NO. 1667. COLLECTION OF ROSARIES—CASANOWICZ. 855 ordinances are extant prohibiting monks from having beads of coral, crystal, amber, etc., and nuns from wearing beads around their necks. On the other hand, beads were openly carried as a sign of penance, especially by bands of pilgrims visiting in procession the shrines, churches, and other holy places of Rome, and wearing of the beads at one’s girdle was a distinctive sign of membership in a religious confraternity. The religious military orders, notably the Knights of St. John (founded in the twelfth century), adopted the rosary as part of the equipment of the lay members, who were required by their constitution to say 150 paters each day. By the devout beads were especially valued if they had been worn by a person of known sanctity, or if they had touched the relics of some saint, in which case they were believed to be the instruments of miraculous power and healing virtue. ‘The oriental Christians affect rosaries made in Jerusalem and other holy places of Palestine. Another contrivance for counting prayers in the middle ages was the so-called “ decade rings,” or “ rosary rings.” They were finger rings having ten knobs, or bosses, at intervals all around a hoop; some had an eleventh knob of larger size, indicating ten aves and one pater. An additional twelfth knob marked the repetition of the Creed. Sometimes the knobs were separated from one another by three small beaded dots, perhaps symbolic of the Trinity. They were worn by some classes of the religious during the hours of repose, so that on awakening during the night they might repeat a certain number of prayers, marking them by the beads or knobs on the ring.“ 78. Catholic rosary.—The full or greater Dominican rosary of 15 decades of beads for the aves and 15 larger ones for the paters are made of ebony. The cross, of the same material, is framed in silver-plated nickel, with the figure of Christ on one side and a crown of thorns with a burning heart inside, of the same metal, on the other. Length, 6 feet 94 inches. (Plate 28, fig. 1. Cat. No. 179075, U.S.N.M.) Collected by Dr. G. Brown Goode. 79. Catholic rosary.—The full or greater Dominican rosary. The beads for the aves are of glass, while those of the paters, as also the three introductory beads are of composition. Length, 4 feet 4 inches. (Cat. No. 179075, U.S.N.M.) Collected by Dr. G. Brown Goode. 80. Catholic rosary.—The lesser Dominican rosary of five decades of beads for the aves and five larger beads for the paters, made of mahogany, with four sets of double circles, or “ eyes,” carved on each. The cross is likewise formed of beads. Length, 5 feet 3 inches. (Plate 28, fig. 2. Cat. No. 179075, U.S.N.M.) Collected by Dr. G. Brown Goode. *Compare William Jones, Finger-ring Lore. Historical, Legendary, Anec- dotal, London, 1890, pp. 248-258. : 856 PROCEEDINGS OF THE NATIONAL MUSEUM. VOL. XXXVI. Si. Catholic rosary.—The lesser Dominican rosary of five decades of beads made of ebony, with the cross of the same material. Length, 4 feet 9 inches. (Cat. No. 179075, U.S.N.M.) Collected by Dr. G. Brown Goode. 82. Catholic rosary—tThe lesser Dominican rosary of five decades of very large oval beads coarsely made of wood, probably worn by some religious orders, perhaps Franciscans, at the girdle. Length, 4 feet 5 inches. (Cat. No. 179075, U.S.N.M.) . Collected by Dr. G. Brown Goode. : 83. Catholic rosary.—The lesser Dominican rosary of five decades of beads made of olive wood, carved with intersecting circles. In place of the cross is a bronze medal, three-fourths of an inch in diameter, with the bust of Pius IX and the date 24 (the number of years of his reign) on the obverse; on the reverse is the figure of the Pope, in full pontificals, on his throne, attended by cardinals, and the Latin words, “ Ecumenical Council, 1869,” referring to the Vati- ean Council, which was opened in that year. This rosary was blessed by Pius IX in 1873. Length, 37 inches. Rome, Italy. (Plate 29, fig. 1. Cat. No. 168294, U.S.N.M.) Collected by Gen. John A. Halderman. 84. Catholic rosary—The lesser Dominican rosary of pearl-col- cred glass beads. The place of the pater beads is taken by oval metal plaques engraved with the image of Mary and an invocation to her. Tt has no introductory beads. Length, 28 inches. (Cat. No. 179075, U.S.N.M.) Collected by Dr. G. Brown Goode. 85. Catholic rosary.—tThe lesser Dominican rosary of black glass beads, rose cut. Length, 26 inches. Philippine Islands. (Cat. No. 205535, U.S.N.M.) Collected by Mr. A. J. Gies. 86. Catholic rosary—The lesser Dominican rosary of ivory beads, faceted, while the pater beads are barrel shaped. In place of the cross is a copper medal, 1} inches in diameter, having on the obverse the image of Mary crowned, with the infant Jesus in her arms; on the reverse, a much-effaced Latin inscription. Length, 40 inches. (Plate 29, fig. 2. Cat. No. 179075, U.S.N.M.) Collected by Dr. G. Brown Goode. 87. Catholic rosary.—The lesser Dominican rosary of Job’s tears. The cross of ebony is set in brass and has the same appurtenances as the one described under No. 78. Length, 33 inches. (Plate 29, fig. 3. Cat. No. 179075, U.S.N.M.) Collected by Dr. G. Brown Goode. 88. Catholic rosary.—Consisting of seven sets, each having. seven beads, made of composition. This rosary is used in honor of the seven sorrows of Mary, namely, the prophecy of Simon (Luke 11, 35) ; the flight into Egpyt (Matthew 11, 13); the losing of Jesus in the NO. 1667. COLLECTION OF ROSARIES—CASANOWICZ. 857 temple (Luke 11, 48) ; seeing Jesus carrying the cross (John xix, 17) ; standing under the cross (John xix, 25); the piercing of Jesus’s side with the lance (John xix, 34), and the lowering of Jesus’s body into the sepulcher (Matthew xxvii, 60). In place of the pater beads are seven brass plaques, representing each, on one side, Mary with seven swords piercing her heart; on the other, the incidents in Christ’s life enumerated above. The cross is formed of beads. Length, 33 inches. (Plate 29, fig. 4. Cat. No. 179075, U.S.N.M.) Collected by Dr. G. Brown Goode. 89. Catholic rosary.—The lesser Dominican rosary of five decades of small blue glass beads, while the five pater beads are of the seeds of the Abrus precatorius (called “ crabs’ eyes,” or “ jumble beads”). Inclosed in an egg-shaped box of bone. Length, 12 inches. Diam- eters of the box, 1 inch by three-fourths of an inch. Madrid, Spain. — (Plate 29, fig. 5. Cat. No. 167020, U.S.N.M.) Collected by Dr. Walter Hough. 90. Catholic rosary.—The lesser Dominican rosary of small black glass beads. Instead of the cross are two oval brass plaques bearing the image of Mary. Length, 22 inches. (Cat. No. 179075, U.S.N.M.) Collected by Dr. G. Brown Goode. 91. Catholic rosary.—The lesser Dominican rosary of small green glass beads. In place of the cross is an oval brass plaque bearing the image of the Virgin of Guadelupe of Mexico. The paters are marked by double beads of the same size and color as the aves. Length, 26 inches. Mexico. (Cat. No. 179075, U.S.N.M.) Collected by Dr. G. Brown Goode. 92. Catholic rosary.—The lesser Dominican rosary of wooden beads, painted black. Worn at the girdle by members of the Fra- ternity of the Misericordia (Arciconfraternita de Santa Maria della Misericordia) in Italy. Length, 50 inches. Pisa, Italy. . (Cat. No. 153893, U.S.N.M.) Collected by Dr. G. Brown Goode. 93. Catholic rosary.—The Franciscan rosary of seven decades of beads made. of composition. This rosary is used for the devotion in honor of the seven mysteries in the life of Mary, namely, the concep- tion (Luke i, 26) ; the visitation (Luke i, 39) ; the nativity (Luke ii) ; the adoration of the magi (Matthew ii); the presentation (Luke ii, 21) ; the finding in the temple (Luke 1i, 41), and the apparition after the resurrection to Mary. The rosary is provided with two rings for suspending from the girdle. Length, 6 feet 8 inches. (Plate 30, fig. 1. Cat. No. 179075, U.S.N.M.) Collected by Dr. G. Brown Goode. 94. Catholic rosary.—The lesser Dominican rosary of black glass beads. An oval bronze medal, 1? and 12 inches in diameter, which takes the place of the cross, has on one side the bust of St. Ignatius 858 PROCEEDINGS OF THE NATIONAL MUSEUM. VOL. XXXVI. Loyola (1491-1556), the founder of the Society of Jesus, on the other the figure of St. John of Nepomuk, the patron saint of Bo- hemia, who, according to tradition, was martyred in 1383. The rosary is provided with two rings to be suspended from the girdle. Probably worn by Jesuits. Length, 55 inches. Plate 30, fig. 2. Cat. No. 179075, U.S.N.M.) Collected by Dr. G. Brown Goode. 95. Catholic rosary—Used in the devotion of the crown of our Lord. Consists of 33 beads made of wood for the paters, to commem- orate the years of Christ’s hfe on earth, and five for the aves, in honor of the five wounds. The cross is substituted by a brass medal, 14 inches in diameter, engraved with the instruments of the passion and the latin words, ‘ The passion of Christ save us, the passion of Christ comfort me.” Between the ave beads is inserted a piece of bone, 14 inches high, carved with the faces of Christ and Mary, and that of a skull. Length, 47 inches. (Plate 30, fig. 3. Cat. No. 179075, U.S.N.M.) Collected by Dr. G. Brown Goode. 96. Catholic rosary.—Consisting of three sets of nine beads each, made of composition, separated by an oval brass plaque, having on one side a representation of the Trinity,.on the other the gloria in Latin. Length, 21 inches. (Plate 30, fig. 4. Cat. No. 179075, U.S.N.M.) Collected by Dr. G. Brown Goode. 97. Catholic rosary.—The Franciscan chaplet of seven decades of beads made of composition. (See under No. 93.) The cross of wood is inlaid with mother-of-pearl. Length, 424 inches. (Cat. No. 179075, U.S.N.M.) Collected by Dr. G. Brown Goode. 98. Catholic rosary—The Franciscan chaplet of seven decades of small purple-colored glass beads. (See under No. 93.) Length, 33 inches. Philippine Islands. (Cat. No. 205535, U.S.N.M.) Col- lected by Mr. J. A. Gies. 99. Catholic rosary.—Consisting of 51 beads made of composition, strung on a cord, with crosses, medals, and figurines at irregular inter- vals. Length, 18 inches. (Cat. No. 179075, U.S.N.M.) Collected by Dr. G. Brown Goode. 100. Catholic rosary—Consisting of 33 beads made of olive wood. Used in the devotion of the crown of our Lord. (See under No. 95.) Length, 44 inches. (Cat. No. 179075, U.S.N.M.) Collected by Dr. G. Brown Goode. 101. Catholic rosary—Consisting of 33 small blue glass beads. Used in the devotion of the crown of our Lord. (See under No. 95.) Length, 19 inches. (Cat. No. 179075, U.S.N.M.) Collected by Dr. G. Brown Goode. 102. Catholic rosary.—Consisting of 26 beads of Job’s tears and composition alternating. Length, 13} inches. (Cat. No. 179075, U.S.N.M.) Collected by Dr. G. Brown Goode. NO. 1667. COLLECTION OF ROSARIES—CASANOWICZ. 859 . 103. Catholic rosary —Consisting of eight decades of small pink- colored glass beads. Length, 36 inches. Philippine Islands. (Cat. No. 205536, U.S.N.M.) Collected by Mr. J. A. Gies. 104. Catholic rosary.—Consisting of thirteen decades of beads made of black wood, without the introductory beads. Used by the Tagalogs of Luzon, P. I. Length, 34 inches. Luzon, P. I. (Cat. No. 216990, U.S.N.M.) Collected by Mr. Frank F. Hilder. 105. Catholic rosary.—Consisting of nine decades of beads made of black wood. Used by the Tagalogs of Luzon, P. I. Length, 33 inches. Luzon, P. I. (Cat. No. 216990, U.S.N.M.) Collected by Mr. Frank F. Hilder. Sheth 4 iyi Ra | LBs pliah raf bee hates ef r fas (ft eae Sitwell ies hy fobs aera te see tant pin kbeal Te tina Uainuhos thors sean PEI ERY ER es oh tee otra die io gt go dite Poth eee ee) sed Fox 4 3 an fan! Api lo Sareea Ae AR) ape Gass hae pate ee an m syed i emit Fe he Pee « ie, nye ~ ve, des Wy 4 Netaie Shag 2084 cdi 4 i, if Owe ‘ 0 he. ey aN | e ' +? ¥ 4 - 2 . » i Pi» es ¥ ans aah us 7 bad ok +4 - aro ¢ . ai Pao . . ® . i. 4 7 oe Cl ‘ : +. % : ae fe an ee: i f <7 » . es ioe r , thy rO J ic vit . a be vr *| fp Tete > » a> Le scat ee x fl a a ‘ ¢ n abet VE _— j “3 i U. S. NATIONAL MUSEUM PROCEEDINGS, VOL. XXXVI PL. 21 TIBETAN ROSARY OF SHELL BEADS. FOR REFERENCE TO PLATE SEE PAGE 337. rh j a) ns) i vr EXPLANATION OF PLATE 22. ee Fig. 1. Rosrewoop Rosary. TA-CHTEN-LU. 2. YELLOWWoop Rosary. Bartana. 3. MoHAMMEDAN Rosary, OF BoNE AND Date SEEps. HSI-NING-FUv. Pi22 PROCEEDINGS, VOL. XXXVI U. S. NATIONAL MUSEUM id C€PCerectet °% UX Aetonananaatayd ‘ 3 7 Cy Tmreesuccssstesees : ceccceseree® | _eore &, TIBETAN AND MOHAMMEDAN Rosaries. FOR REFERENCE TO PLATE SEE PAGES 338, 350. PL. 23 PROCEEDINGS, VOL. XXXVI U. S. NATIONAL MUSEUM ll yee rt EP EEREFE -CHU. CHINESE OFFICIAL Su FOR REFERENCE TO PLATE SEE PAGE 389, U. S. NATIONAL MUSEUM PROCEEDINGS, VOL. XXXVI PL. 24 CHINESE OFFICIAL SU-CHU. FOR REFERENCE TO PLATE SEE PAGE 389. EXPLANATION OF PLATE 25. fs Fig. 1. Rosary or MorHer oF PEARL. JAPAN. 2. Surnto Rosary, oF GuAss Pieces. JAPAN. PL. 25 > x x x a fe) > 122) © re a WwW WW 2) fe) a a © "e JAPANESE ROSARIES. FoR REFERENCES TO PLATE SEE PAGES 343, 344. tee, mk: 60 Ceeececcqceeooeee®® iad . "te U. S. NATIONAL MUSEUM EXPLANATION OF PLATE 26. a ae Fig. 1. DousLe Rosary oF THE Jopo Sect, oF MAHOGANY. JAPAN. 2. Rosary oF SKULL-SHAPED BEADs, OF WALNUT SHELLS. JAPAN. 3. Rosary witH BEaps CARVED TO REPRESENT ROHANS. JAPAN. U. S. NATIONAL MUSEUM PROCEEDINGS, VOL. XXXVI PL. 26 JAPANESE ROSARIES. FOR REFERENCES TO PLATE SEE PAGES 345, 346. ac el = = — yi . a ee 4 Ss EXPLANATION OF PEATE 27. Fig. 1. MoHamMMEDAN Rosary, oF OLivE Woop. Cairo, Eayrr. 2. MoHAMMEDAN Rosary, OF BLOODSTONES. 3. MoHAMMEDAN Rosary, oF Horn. 4. MoHAMMEDAN Rosary, oF Composition. Mount Sinat, Syria. LS 27 PROCEEDINGS, VOL. XXXVI U. S. NATIONAL MUSEUM ¢ 93300.220002\ see eee eee 8 tf aD DDDD32D3DD DDB sae 8 94 3.8 6 8 8 MOHAMMEDAN ROSARIES. FOR REFERENCES TO PLATE SEE PAGES 349, 350. EXPLANATION. OF PLATE 28. Fig. 1. Carnoric Rosary, oF Exony. 2. CatrHouic Rosary, or MAHOGANY. U. S. NATIONAL MUSEUM PROCEEDINGS, VOL. XXXVI PL. 28 ROMAN CATHOLIC ROSARIES. FOR REFERENCE TO PLATE SEE PAGE 355, EXPLANATION OF PLATE 29. Fig. 1. Carnoric Rosary, oF OLIvE Woop. Roms, ITAzy. 2. CarHouic Rosary, oF Ivory. 3. CarHoLic Rosary, OF JoB’s TEARS. 4. CatrHouic Rosary, oF CoMPOSITION. 5. CarnHoric Rosary, OF GLASS AND SEEDS OF ABRUS PRECATORIUS, WITH Ea@a- SHAPED Box. MAaprip. PROCEEDINGS, VOL. XXXVI PL. 29 U. S. NATIONAL MUSEUM cn = GA)...» ade a i ~, kh i ils Saar, > . eS 3 P-)-B-d- P D-O™ ne ee a a ae al aeeag eee ee) Pes eS 92939 3-3-3-3-2-..1y... BD 2° +d? 2990200005. D309. "220299 j >. D 32-9. 229:9-9-90939-9-2939999099"" D329-9 ROMAN CATHOLIC ROSARIES. PAGES 356, 357. EXPLANATION OF PLATE 30. Fig. 1. Carnoxic Rosary, or Composition, witH Two Rivas. 2. Carnouic Rosary, or BLack Guass Braps, with Two Rivas. 3. Carnotic Rosary, or Woop, with CARVED PIECE OF BONE. 4. CarHouic Rosary, or Composition. U. S. NATIONAL MUSEUM PROCEEDINGS, VOL. XXXVI PL. 30 ROMAN CATHOLIC ROSARIES. FOR REFERENCES TO PLATE SEE PAGES 357, 358. COMATILIA, A REMARKABLE NEW GENUS OF UN- STALKED CRINOIDS. By Austin Hosart Criarx, Collaborator, Division of Marine Invertebrates. The elaborate definition of the genus Actinometra (the family Comasteride as now understood) given by Dr. P. H. Carpenter in the Challenger Reports, and the various additional characters noted by him in different places in the same work, would seem to have established its status definitely, and to have demonstrated conclusively that it formed a well-circumscribed unit, very sharply contrasted with the aggregation of species called by Doctor Carpen- ter Antedon,; in other words, that the recent free crinoids without basals and with ten or more arms fall naturally into two well-defined structural types, separated from each other by more numerous and more important characters than exist between the various specific groups within the two types. At the beginning of my studies I had become convinced from the available material that the division of these forms into two sharply contrasted groups was very artificial, and could not stand the test of modern systematic methods, in the light of our greatly increased knowledge. I therefore proposed to recognize, instead of the two genera “Antedon” and “Actinometra” used by Carpenter, five great divisions (the families Comasteride, Zygometridx, Himerometride, Tropiometridz, and Thalassometride) which covered exactly the same ground, except that the number of included species was nearly, if not quite, doubled. Each of these families appeared to me to be separated from the others by characters of just as great importance as the Comasteride (the old Actinometra) was from any one of them. My suspicions in regard to many of the characters relied upon to differentiate “Actinometra” from “Antedon” have recently been con- firmed in a most conclusive manner. In examining some comatulids taken by the U. S. Bureau of Fisheries steamer Albatross in deep water between the Bahama Islands and Cape Fear, North Carolina, I found a most peculiar form which, according to the structure of its oral pinnules and brachials, belongs to the Comasteride, but PROCEEDINGS U. S. NATIONAL Museum, VOL. XXXVI—No. 1668. 361 862 PROCEEDINGS OF THE NATIONAL MUSEUM. VOL. XXXVI. which, judged upon its other characters, might be placed in the Antedonide, Himerometridx, Thalassometride, or Tropiometride, while certain features find a parallel only in Rhizocrinus and Bathy- crinus. Briefly stated, the essential features of the Comasteride are, an eccentric mouth, large central or subcentral anal tube, ambulacra quite without calcareous plating, absence of sacculi, stout cirri with short joints, and the full complement of pinnules, the lowest of which bear terminal combs. The first and the last are the characters upon which most reliance is usually placed. In this new form the mouth is always central, the anal tube small and marginal, sacculi are pres- ent (though rare), the cirri are exceedingly slender, with greatly elongated joints, the six pinnules following the first two (P, and P,) are lacking, the pinnule ambulacra are provided with large side plates, and the two lowest pinnules are combed. In its central mouth this new species agrees with all the other recent species except those of the Comasteride, and the same is true in regard to the sacculi; the delicate cirri find counterparts only in the Antedonide, in the genera Hathrometra and /ridometra, the deficient pinnulation recalls Perometra and [Typalometra among the Antedonide, Cyllometra and Colobometra among the Himerometridx (with this exception, that the first inner pinnule, on the fourth brachial, P,, is always the first to disappear in these genera, while it is invariably present in the new form), and also Atelecrinus of the Atelecrinide; the develop- ment of an ambulacral plating is found perfected only in the Thalas- sometride and in the Tropiometride. Taken as a whole, the great- est resemblance in size and general build is to the little littoral Antedonids of the East Indies belonging to the genus /ridometra, Z. nana in particular, and this in spite of the fact that it has the deepest habitat of any of the Comasteride. The development of plating along the ambulacra in the family Comasteride was first demonstrated by Mr. Frank Springer, who showed its presence in a new species of Comaster, C. iowensis, from the Tortugas. I have since found it to be a constant feature of the West Indian species of this genus, even including C. lineata, in which I detected it in specimens previously examined by Carpenter. No less interesting than the adults are the young, as shown by a single specimen with an arm length of probably about 7 mm. Had it not been found with the fully grown, it could very well have passed as a new species of Thaumatocrinus, or the representative of a genus intermediate between 7haumatocrinus and Antedon. Thaumatocrinus renovatus was based upon a very peculiar coma- tulid which had been dredged by the Challenger in 50° 01’ S. lat., 128° 04’ E. long., at a depth of 1,800 fathoms. Doctor Carpenter’s diagnosis of the genus, as given in the Challenger Reports, is: Calyx composed of a centro-dorsal, basals, radials, and primary interradials, the latter resting on the basals and so separating them no. 1668. A NEW GENUS OF UNSTALKED CRINOIDS—CLARK. 863 laterally; that on the anal side bears a short jointed appendage; mouth central and protected by five large oral plates which occupy the greater part. of the disk, and are separated from the calyx inter- radials by two or three rows of small irregular plates; five arms only. Doctor Carpenter discusses at considerable length the similarity between Z’haumatocrinus and a number of palewozoic forms, and, al- though he includes it in the “Comatulidee” as understood by him, he is inclined to regard it as a very anomalous type, exhibiting certain atavistic characters. The type-specimen of Thawmatocrinus renova- tus is exceedingly small; the total width of the calyx across the disk is barely 2 mm., and the height of the centro-dorsal and radials to- gether is about the same. The arm structure of Thaumatocrinus is identical with that of Pentametrocrinus and Decametrocrinus, which together form the family Pentametrocrinde, and in its pinnule and cirrus structure, in so far as it can be made out, it also agrees with the conditions found in those genera. The arm and pinnule structure of the Pentametro- crinid is unique in its simplicity among the comatulids, which in itself suggests that it is a family of a remarkably primitive type. The adult Pentametrocrinus differs from the adult Antedon in its more generalized and presumably more primitive structure; if we take a young Antedon and generalize it by supplementing its single interradial (anal) with four others like it, we would find a 7hauma- tocrinus as a result. Zhaumatocrinus in every detail except such as, from analogy with Antedon, may be safely ascribed to immatu- rity, agrees with Pentametrocrinus; hence it seems probable, as the atavistic Thaumatocrinus bears a very similar relation to the young Antedon to that which the primitive Pentametrocrinus does to the adult Antedon, and as Thaumatocrinus and Pentametrocrinus agree in all essentials fully as well as the young and the adult of Antedon agree, that Zhaumatocrinus is in reality nothing more nor less than the young of Pentametrocrinus. But there is still further evidence. In Decametrocrinus, which is a meristic variation from Pentametro- crinus, differing only in having twice as many radials and arms, the ends of the five basal rays appear externally in the angles of the calyx, dividing the ten radials into groups of two; but in Proma- chocrinus, which is a similar meristic variation from Heliometra or a closely allied genus, the ends of the basal rays appear exter- nally under the center of alternate radials. Promachocrinus prob- ably has young much like those of the closely allied Heliometra, in which there is but a single interradial plate, the anal; now the divi- sion of each radial in the young of Promachocrinus, and the growth of each resultant half to the same size as the single undivided radial in Heliometra, would, as the anal plate is lifted out from between the posterior radials, produce a certain amount of tortion of the 864 PROCEEDINGS OF THE NATIONAL MUSEUM. VOL. XXXVI. calyx, and might very well result in causing the basals in the adult to occupy a position under alternate radials. The ap- pearance of the basal rays in Decametrocrinus between the radials instead of under alternate ones is a difference of considerable mor- phological importance, suggesting that the young are of quite differ- ent structure. Decametrocrinus is a meristic variation from Penta- metrocrinus; Thaumatocrinus, probably the young of Pentametro- crinus, has interradial plates of equal size in all the interradial areas; now if the young of Decametrocrinus were of the Thaumatocrinus type with five equal interradials which, during growth, were shoved out from between the radials at an equal rate, the basal rays in the adult would maintain exactly the same relation to the radials as the basals did to the radials in the young, instead of being twisted about into a semiradial position as in Promachocrinus. Thus a compari- son of adults of the ten-rayed genera of the Pentametrocrinidz and Antedonide leads to the same conclusions as a comparison of 7’hau- matocrinus with the Antedon larva, namely, that 7hawmatocrinus is the young of Pentametrocrinus. By this reasoning I had sometime ago reached the conclusion that Thaumatocrinus was very close to Pentametrocrinus, and probably the young of it, and I therefore placed it next to Pentametrocrinus in the family Pentametrocrinide. Here the matter rested, for noth- ing further could be done without additional facts to prove or dis- prove the results attained by purely speculative processes. The young example of this new comatulid is so like 7hawmato- crinus in certain ways as to convince me that I was right in my tentative treatment of that genus. It represents, however, a more advanced stage; the five large, strong orals are present as in 7’. reno- vatus, surrounded by small irregular plates; the basals are not evi- dent externally; the radials are in lateral contact, and just above their apposed lateral edges in the angles of the calyx are five large interradials which appear to have been recently thrust forward from between them, but which are not yet undergoing the process of re- sorption. In the large and persistent orals and interradials (though displaced) this young specimen resembles 7’haumatocrinus, though it is true that it differs from it in the approximation of the radials and in the absence of external basals; but there can, I think, be no reason- able doubt that these differences are merely the result of its greater development. The arm structure, so far as it is elaborated, resembles that of the adult. This new form is undoubtedly referable to the Comasteride, as evidenced by the characteristic pinnules with short joints, coarsely spinous on their distal ends, finely spinous dorsally, the comb on the terminal portion of the first pair, the coarsely spinous overlap of the brachials, and their spinous dorsal surface. This appears to out- xo. 1668. A NEW GENUS OF UNSTALKED CRINOIDS—OLARK. 365 weigh all the other features. The ambulacral plating is unlike that developed in the Antedonide, Tropiometride, or Thalassometride, in that there is only a single series of plates instead of both side and covering plates; these appear to represent the side plates of other forms, and not the covering plates, as does the single series in Rhizo- crinus. The development of covering plates is an uncertain quan- tity, and one upon which too much stress has previously been laid. Hartlaub and Minckert both divided the old genus “Antedon” into two sections, one with and one without them; but both included in the “ plated ” section Carpenter’s “ Basicurva group,” which contains species in which they are not found. Moreover, in the Tropio- metride they are extraordinarily developed in Calometra, more or less imperfectly developed in Asterometra and Ptilometra, and quite undeveloped in 7ropiometra; but more curious still, while they make their appearance in the pentacrinoid stage of the species of Thalasso- metridz, they are not found in the young of Ptilometra. It is evi- dent, therefore, that, though a valuable index to the systematic posi- tion of the comatulid species, they must be treated with great cau- tion, as they appear to be very lable to sudden development, as is the ease in /eliometra, in very unexpected places, and to equally dis- — concerting suppression. The central position of the mouth, while interesting, is of no real importance; it is usually nearly, and often quite, central, in Phanogenia, Comatella, and Comaster, and often more or less eccentric in Heliometra and in certain of the Himero- metridz. Sacculi are somewhat uncertain organs at best, while cirri are so very variable that the occurrence of a new type need cause no trouble; the resemblance to the cirri of 7ridometra is not borne out by the finer structure; for instance, in /ridometra, as in all comat- ulids heretofore known, the opposing spine is single, whereas in this new form it is forked. I propose to recognize this new comatulid as follows: Genus COMATILIA, new. Centro-dorsal discoidal, moderate in size; cirrus sockets marginal, usually in a single row. Cirri about XX, 9-10, about one-fifth of the arm length, very slender, the second and following joints much longer than broad, the third and fourth the longest, about four times as long as their prox- imal diameter; cirrus joints all with expanded ends; no dorsal spines; _ opposing spine forked in its distal half, or ending in a bunch of fine spines. Arms 10; first four brachials oblong, broader than long, then ob- liquely wedge-shaped, at first as long as broad, soon becoming longer than broad and very long terminally; large interprimibrachs present, rounded, not contiguous. 366 PROCEEDINGS OF THE NATIONAL MUSEUM. VOL. XXXVI. Disk naked; mouth central, very large; anal tube small, marginal. PP ,, 25 45 db) co a absent; PP , and , long and slender, with a large terminal comb; pinnules from PP ,,, gradually increasing in length, distally reaching approximately the length of the oral pinnules; the first three to six after PP ,,. bear globular genital glands at the base; side plates developed on the outer three-fourths of the pinnule am- bulacra, best developed in the distal half, Genoty pe-—Comatilia tridometriformis. COMATILIA IRIDOMETRIFORMIS, new species. Centro-dorsal moderate, discoidal, the bare polar area flat, usually with a more or less developed low rounded median tubercle, 1 mm. to 1.5 mm. in diameter; cirrus sockets usually in a single, but often in a more or less partially double or triple marginal row. Cirri XVI-X XVIII, 9-10, 5 mm. to 6 mm. long, exceedingly slen- der and delicate; first joint short, about twice as broad as long; second about twice as long as broad; third and fourth the longest, about four times as long as broad; fifth slightly shorter; following joints gradually decreasing in length, the antepenultimate being about half again as long as its distal diameter; penultimate joint about as long as, or slightly longer than, broad; second and following joints strongly “ dice-box shaped ” with expanded ends; after the fourth the proximal ends becoming less, the distal more expanded, so that the cirri as a whole bear a close resemblance to those of certain of the Antedonide, as Hathrometra and [ridometra; distal ends of the cirrus joints over- lapping all around, but slightly more dorsally than ventrally, the dorsal overlap, however, exhibiting no tendency to project anteriorly: opposing spine terminal in position, directed obliquely forward, arising from the entire distal half (or rather more) of the dorsal side of the penultimate joint, about equal to one-half of the distal diameter of the joint in length, usually forking transversely in its distal half, more rarely breaking up into a number of small spines; terminal claw approximately equal in length to the penultimate joint, mod- erately stout, and evenly curved. Ends of the basal rays visible as rather prominent tubercles in the interradial angles of the calyx; radials very short in the median line, but extending up in the angles of the calyx, separating the bases of the IBr, for a distance about equal to one-half of the basal diam- eter of those joints, the condition in general resembling that found in Calometra multicolor and in Bathymetra,; two to four large oval or round interprimibrachs are found in each interradial area, which, however, are usually not quite contiguous; IBr, comparatively small, oblong, very short, between three and four times as broad as long; IBr, (axillary) broadly pentagonal, about twice as broad as long, the lateral edges about as long as those of the [Br,, making with them an obtuse angle. no. 1668. A NEW GENUS OF UNSTALKED CRINOIDS—CLARK. 367 Arms 10, about 30 mm. long; first brachial very short, oblong, about four times as broad as long, united basally with its fellow, diverging at approximately a right angle distally; second brachial usually about twice as large, wedge-shaped; third and fourth brachials (syzygial pair) somewhat longer than broad; following brachials very obliquely wedge-shaped, at first about as long as broad, but almost immediately becoming longer than broad and gradually in- creasing in length, being terminally two or three times as long as broad, or even longer, with expanded ends; after about the sixth the brachials develop strongly produced and overlapping distal ends, which are armed with a row of comparatively coarse spines. Syzygies occur between the third and fourth brachials, again between the thirteenth and fourteenth, and distally at intervals of two oblique muscular articulations (“in alternate joints ”).. Mouth central and very large; anal tube small and marginal; disk naked, except for the previously mentioned interprimibrachial plates. No pinnules on the fifth to the tenth brachials, PP., 5, 4, », ¢ a being absent; P, 4 mm. to 4.5 mm. long, with twenty joints, slender, and tapering evenly distally; first joint not so long as broad, second and third about as long as broad, the remainder about half again as long as broad; the joints are somewhat constricted centrally and have expanded and overlapping distal ends which are armed with fine spines, and a finely spinous dorsal surface; terminal comb very prominent, composed of six to eight large teeth, trapezoidal or blunt triangular, their bases in contact, somewhat higher than the trans- verse diameter of the joint which bears them, and recurved; P, sim- ilar to P,; P,; (the next pinnule) 3 mm. long with twelve joints, the first two not so long as broad, the third about as long as its proximal diameter, the remainder somewhat longer than broad; all but the first have greatly expanded distal ends, armed with comparatively coarse spines; a very round and prominent genital gland is found on the second-fourth or second-fifth joints; following pinnules slowly increasing in length, the joints, except the two first, slowly becoming more and more elongated; distal pinnules about 4.5 mm. long, or about the length of the oral pinnules; the genital glands are found on three to six pinnules on either side of the arm and are always small and situated basally, like that on the first genital pinnule; large side plates are developed along the sides of the pinnule ambulacra of the third and following joints, becoming more perfect distally. Color (in spirits)—Brownish white, the perisome yellow brown; probably yellow in life. Type—Cat. No. 25460, U.S.N.M., from Albatross Station No. 2670; between the Bahama Islands and Cape Fear, North Carolina; 280 fathoms. 4 Dy & “ r DESCRIPTION OF A NEW SPECIES OF LEATHERBACK TURTLE FROM THE MIOCENE OF MARYLAND. By Wixi1am Pater, Of the U. S. National Museum. During July, 1908, while Mr. D. B. Mackie and myself were search- ing for fossils along the Calvert Cliffs, in Calvert County, Maryland, we discovered some hard, stony plates of a peculiar nature, which puzzled us for some time. The first piece found was in a lump of wet, sandy clay at the line of high water. It stood on edge, and could only be examined, without disintegrating, on its under surface. After considerable examination and discussion, we decided that it was either a piece of crude Indian pottery or a piece of baked clay, possibly the bottom of an oven or burning place. Attention was then attracted to a larger mass of drier clay near by, which, after a few minutes of picking and cleaning, developed a similar back and edge exposure, and also an upper surface divided into many small, interlocking, flat plates, the whole being 562 mm. long and 130 mm. wide, and composed of about 100 plates. Greater interest in the find being thus excited, much of the débris was examined, with the result that many separate plates were found, and also a few pieces in which several or many were still united. An examination of the face of the cliff was then made, disclosing not only the place from whence the material had recently fallen, but also, some 30 feet above the beach, a clearly-defined section of what was evidently an inverted shell or carapace about 4 feet (1,220 mm.) across. The ends were turned upward for about 8 inches (204 mm.), and a very distinct and heavy ridge projected downward from its center. A number of steps were cut into the cliff and efforts made to get near the remains, but the treacherous nature of the wet clay and the fact that the shell was located in an overhanging portion of the cliff compelled us to abandon the attempt for a time. On another visit several days later, with tools and ropes. we were unfor- PROCEEDINGS U. S. NATIONAL MUSEUM, VOL. XXXVI—No. 1669. Proc. N. M. vo). xxxvi—09——24 369 370 PROCEEDINGS OF THE NATIONAL MUSEUM. VOL. XXXVI. tunate enough to find that the remaining part of the shell and much of the cliff had fallen, and were mixed in fragments with the earher débris. It became necessary, therefore, to carefully dig over all the talus material, and to wash much of it also, in order to extract the plates, which were generally separated. In this way many hundreds were recovered, together with a few broken and badly disintegrated vertebrae, ribs, and other bones, one piece only being in fair condition, A superficial examination of this material, which required several days’ labor to bring away, led to the conclusion that the remains probably belonged to a mammal resembling a Glyptodon, but later, on comparing the plates and the best of the bones with the skeletons of several turtles, it became clear that they represented an animal closely related to the living Leatherback Turtle, though differing in several respects. Dermal scutes of this same species had been previously picked up from the beach bordering the cliffs by Dr. F. W. True, and later by myself, but their true nature, though often commented on, was not satisfactorily determined until the remains herein described were found. PSEPHOPHORUS CALVERTENSIS, new species. Dermatochelys, J. Mister, Ueber die fossilen Reste der Zeuglodonten von Nord-Amerika, 1849, p. 34, pl. 27, fig. 7. Upper Eocene Zeuglodon beds of Alabama. Tdem, p. 407, 872 PROCEEDINGS OF THE NATIONAL MUSEUM. VOL. XXXVI. edly in size and shape and other particulars, but the differences are those of degree only, and as far as our present knowledge extends, the species may all be considered as belonging to the same genus. During our many trips along the Calvert Cliffs, we found between tides several pieces of bones evidently belonging to turtles. They were all heavy and black. One piece appeared to be the central portion (about half) of a scapula, and on comparison with a scapula of D. coriacea would seem to belong to P. calvertensis. It differed from the same part in 2. coriacea in being slenderer and flatter. The piece, 64 inches (165 mm.) long, appears to indicate that the scapula is shorter than that of D. coriacea. The following is a list of species which have been referred to Psephophorus: PSEPHOPHORUS POLYGONUS Von Meyer. Psephophorus polygonus Von Meyer, N. Jahrbuch, 1846, p. 472, and 1847, p. 579: Ber. Mit. Freund. Naturw., 1851, p. 83—Von Hauer, Verh. k.—k. geol. Reichsanst., 1868, p. 887, and 1870, p. 342; Fucus, Verh. k.—k. geol. Reichsanst., 1874, p. 220—SreLry, Quart. Journ. Geol. Soc. London, 36, 1880, p. 406, pl. 15—Woopwarp, Proc. Geol. Ass. London, 11, 1889, p. 13. P[{sephophorus] polygonus, DoLLo, Ann. Soc. Sci. Bruxelles, 11, 1887, p. 139; Bull. Mus. Roy. Hist. Nat. Belg., 5, No. 1, 1888, p. 83. The type-species, known only from a few bones and many scutes, from near Neudorfl, Austria, in Pliocene sandstone. Specimen, in- cluding the type material, now in the museum of the Imperial Geo- logical Survey, Vienna. The slab of hard sandstone in which the remains are preserved is 460 mm. by 410 mm. wide. PSEPHOPHORUS PESUDOSTRACION (Gervais). Sphargis pseudostracion Gervais, Dict. Univ. Hist. Nat. (Ch. d’Orbigny), 11, 1848, p. 56; Zool. et Pal. Franc., 2d ed., 1859, p. 438, pl. 9, fig. 1. P[sephophorus] (Sphargis) pseudostracion, Dotto, Bull. Mus. Roy. Hist. Nat. Belg., 5, No. 1, 1888, p. 83. Type-locality, Vendargues, near Montpellier (Hérault), France. Miocene. First considered to be a fish, Ostracion sp. PSEPHOPHORUS RUPELIENSIS (Van Beneden). Sphargis rupeliensis VAN BENEDEN, Bull. Acad. Roy. Belg., 5d ser., 6, (1888), p. 665.—Woopwarp, Proc. Geol. Ass. London, 11, 1889, p. 18. P[sephophorus] (Sphargis) rupeliensis Dotio, Bull. Roy. Hist. Nat. Belg., 5, No. 1, 1888, p. 838. Type-locality, Boom, Belgium. Middle Oligocene. PSEPHOPHORUS SCALDII (Van Beneden). Macrochelys scaldii VAN BENEDEN, Bull. Acad. Roy. Belg., 2d ser., 31, 1871, p. 13.—Do11o, Bull. Mus. Roy. Hist. Nat. Belg., 5, No. 1, 1888, p. 75. P[sephophorus| (Macrochelys) scaldii, Dotto, Bull. Mus. Roy. Hist. Nat. Belg., 5, No. 1, 1888, p. 88. Type-locality, Antwerp, Belgium. Pliocene and Miocene. no. 1669. NEW SPECIES OF LEATHERBACK TURTLE—PALMER. 3783 PSEPHOPHORUS sp., Lydekker. Psephophorus sp., LYDEKKER, Cat. Foss. Rept. Brit. Mus., 1889, Pt. 5, p. 224. Sussex, England. Middle Eocene. PSEPHOPHORUS EOCANUS Andrews. Psephophorus eocenus ANDREWS, Geol. Mag., 4th ser., 8, 1901, p. 440, fig. 3; Desc. Cat. Tert. Vert. Faytim, Egypt, 1906, p. 275. Type-locality, Qasr-el-Sagha beds, Egypt. Middle Eocene. EXPLANATION OF PLATE 31. Psephophorus calvertensis. A. Portion of a rib, probably the tip, showing striations. B. Under side of a scute. C,C. Under side of scutes, showing the central pit. D. A minor lateral ridge, 9 mm. thick. E. Portion of the medium ridge, 200 mm. long; center, 20 mm. thick; edge, 12 mm. thick. F. Edge piece of the plastron, 135 mm. long; 10-14 mm. thick. G. Piece of a median ridge, 155 mm. long; center, 21 mm. thick; edges, 8-11 mm. thick. Lateral view. PL. 31 PROCEEDINGS, VOL. XXXVI U. S. NATIONAL MUSEUM A NEw SPECIES OF LEATHER-BACK TURTLE. FOR EXPLANATION OF PLATE SEE PAGE 877, —s FOUR NEW SPECIES OF ISOPODS FROM THE COAST OF CALIFORNIA. — By 8S. J. Houmes and M. E. Gay, Of the University of Wisconsin, Madison. The specimens of the new species here described were collected on the coast of California by Dr. S. J. Holmes and sent to the U. 8S. National Museum. — ANCINUS GRANULATUS, new species. Body very broad and much depressed, contractile, evenly and densely granulated. Thorax with parallel sides. Head twice as broad as long; front produced into a rectangular lobe between the bases of the antennules; a small lobe on the anterior margin on either side of the median one. Eyes small and. round. Antenne nearly equal in length; the flagellum of the first somewhat longer than the peduncle and composed of about 10 joints. Second antenne with the flagellum about 10-jointed and longer than the pe- duncle; both furnished with setee having a brush of radiating hairs at the tip. Mandibles with the palp situ- ated behind the middle, the last two joints furnished on the dis- | tal part of the outer margin Fic. 1.—ANCINUS GRANULATUS ; 9, FIRST with setose spines, those of the TRE: last joint increasing in length toward the tip. Palp of the maxilli- peds with the first joint very short, the second and third joints as wide as long and produced into a rounded setose lobe on the inner margin; fourth joint produced at the distal end of the inner margin into a rounded setose lobe; last joint oblong, distally rounded and setose, and about three-fourths the length of the preceding one. PROCEEDINGS U. S. NATIONAL Museum, VOL. XXXVI—No. 1670. BY (3) 846 PROCEEDINGS OF THE NATIONAL MUSEUM. VOL. S¥XVI. First pair of legs with a broad hand having a very convex anterior margin; palm comprising nearly all the posterior margin of the hand, evenly curved, and furnished with a long spine at the upper end. The following legs increasing in length posteriorly and fur- nished on the margins with long spines and fine cilia. First abdominal seg- ment very short. Termi- nal segment triangular with slightly sinuous mar- gins, the tip narrowly rounded when seen from above, but having a deep notch on the lower side. Uropods with a_ single movable, styliform = ra- mus, which nearly or quite reaches the tip of the abdomen. Length, 8 mm. = b Locality — Near Coro- Fig 2.—ANcINUS GRANULATUS; a, sEconD Luc; b, nado Island, California, SECOND PLEOPOD OF THE MALE. THESE FIGURES, from Py depth of 3 fathoms c ra . WHICH WERE KINDILY SUPPLIED BY Miss H. RICH- ARDSON, ARE DRAWN TO A LARGER SCALE THAN THE Ty pe.—Cat. No. 39046, PRECEDING ONES. U S N.M . on . . This is the only species of the genus known to occur on the Pacific coast. It is broader and flatter than Ancinus depressus Say of the coast of New Jersey and has a more broadly tri- angular terminal segment of the abdomen. TYLOS PUNCTATUS, new species. Oblong, covered with scattered short spines or acute granulations. Eyes nearly round. Firs’ antenne single jointed, scale like. Second anten- ne less than one-fifth the length of the body, not reaching the middle of the first thoracic segment : a hook-like process on the second joint of the peduncle; third joint nearly as long as the two preceding; flagellum slightly longer than the last joint of the peduncle, the third joint nearly as long as the two preceding; fourth joint short, rie. 3.—rTyxos pune- conical, and furnished with numerous sete at its THEM distal end. Lateral lobes of the head with two triangular projec- tions in front of the eyes. No. 1670. FOUR NEW SPRCIBS OF ISOPODS—HOLMES AND GAY. 377 Thoracic segments subequal, the epimera in all produced back- ward and rounded at the posterior angle. Legs very spiny, the terminal part of the claw marked off by an apparent suture from the longer basal portion; first pair of legs with an acute lobe near the distal end of the anterior margin of the second joint; fourth joint produced and rounded in front. Third abdominal segment and to a less extent the fourth produced backward at the outer posterior angle; lateral process of fifth segment small. Last segment truncated and four or five times as broad as long. Uropods nearly semicircular in outline, armed with a few scattered spines, the small ter- minal joint furnished with a few spines and several sete. Length, 10 mm. Locality——San Diego, Cali- fornia, in sand near the beach. Type. — Cat. No. 39047, Fic. 4.—TYLOS PUNCTATUS ; a, ANTENNA; Md, FIRST MAXILLA; map, MAXILLIPED; plpo, U.S.N.M. SECOND PLEOPOD OF THE MALE; plps, FOURTH No other representative of © FLS0FOD; ar, uEoron. the family Tylide is known from the west coast of North America. bp, ACTONISCUS TUBERCULATUS, new species. Body elliptical in outline and furnished with small tubercles. Head deeply inserted, with an acute median lobe and prominent rounded lateral ones. Eyes oval. Antenne not one-third the length of the body, the second joint of the peduncle a little longer than the third and about twice the length of the first; fourth joint longer than the third but not quite so long as the fifth; flagellum with four evident joints and a minute terminal fifth joint. The peduncle is bent between the second and third, and the fourth and fifth joints. Maxillipeds with a rounded setose inner lobe; palp short and broad, the first joint much wider than long, the second triangular with slightly lobulated inner margin, the tip with a brush of long sete. Legs similar, spiny, a long ciliated spine on the lower margin of the fifth joint. 878 PROCEEDINGS OF THE NATIONAL MUSEUM. VOL. XXXYI. Basal joint of uropods large, similar to the coxal plates of the pre- ceding segments, and‘ setose on the distal margin; rami extending about to the tip of the peduncle, the outer one inserted at the middle of the inner mar- gin of the basal plate, the inner one near the base; both tipped with sete. Length, 3.25 mm. Locality.—San Diego, California,on moist ground near the seashore. Type.—Cat. No. 39048, U.S.N.M. This species seems to be closely allied to Actoniscus ellipticus Harger from the At- lantic coast. The body is somewhat broader and the lateral processes of the segments are more nearly rectangular in outline, espe- cially in the abdomen, and more prominent. PHILOSCIA RICHARDSONZ, new species. AE a ORR ae Pe Body oblong-oval, covered with short CULATUS. minute spinules. Head twice as wide as long; frontal margin arched; lateral angles subacute. First thoracic segment longer than the following ones, the last three segments pro- duced backwards at the lateral angles. Antenne about one-half as long as the body, the last joint of the peduncle about as long as the third and fourth; flagellum ‘tt ri articulate, nearly as long as tle fifth joint of the peduncle, the first and third joints sub- equal and a little longer than the second; last joint ending in a spine. Legs similar, increasing gradually in length from before backwards, and very spiny. Abdomen abruptly much narrower than the thorax, the lateral angles of the third, fourth, and fifth seg- ments produced backwards; last segment over twice as broad as long, with the posterior margin concave on either side of the narrowly rounded tip. Basal joint of Fic. 6.—PHILOSCIA RICHARDSON. No. 1670. FOUR NEW SPRCIES OF ISOPODS—HOLMES AND GAY. 879 the uropods about as broad as long; outer ramus slender, acuminate, subconical, with the outer margin nearly straight and the inner one somewhat convex; inner ramus about one-third the length of the outer, subconical, with narrow blunt tip which is armed with one or more sharp spines; scattered short spines occur on both rami. Length, 5 mm. Locality.—San Diego, California, on moist swampy ground. Type.—Cat. No. 39049, U.S.N.M. Named after Miss Harriet Richardson, who has contributed so much to our knowledge of the North American Isopoda. i NOTES ON TWO SLUGS OF THE GENUS VERONICELLA. By W. W. Ropsins and T. D. A. CockErReELL, Of Boulder, Colorado, Slugs of the genus Veronicella are numerous in the tropical regions vf both hemispheres, 154 species having been described up to the present time. Dr. D. F. Heynemann * has given a map of the world, showing the distribution of the known species. In this map Austra- lia is shown to be without Veronicella, with the exception of a couple of species found at Brisbane; and these latter, according to Henry Tryon,” have almost certainly been introduced. Nothing is known of Veronicella in New Guinea, where the related but very distinct genus Prisma takes its place. The group to which the names A¢opos, Prisma, and Padangia have been applied extends from north Queens- land through New Guinea to Amboina, Celebes, the southern Philip- pines, Sumatra, the Malay Peninsula, and Cochin China. Westward it occupies the same general regions as Veronicella, but eastward it seems to occur to the exclusion of it. Leaving the Austro-Malay region and passing on to the islands of the Pacific, we again meet with species of Veronicella, in New Caledonia, the Loyalty Islands, the New Heb- rides, and even as far as the Fijis and Tahiti. The species of the Pacific Islands, noticeable for their rather small size, have been de- scribed from rather inadequate material, with the exception of V. willeyi Collinge ° from Lifu, Loyalty Islands. Further details have been especially desired in the hope of throwing some light on the origin of these animals. They are more or less arboreal, and the pre- sumption would be that they were originally carried to the islands on floating trees; but the available ocean currents appear to set wholly from the American side, implying South American origin and a voyage of extraordinary length. It does not appear possible at the present time to demonstrate any close affinity with either the South American or Asiatic groups of Veronicella; but when we know more about the “Die Geographische Verbreitung der Nachtschnecken, 1905, pl. 2. > Queensland Agricultural Journal; July, 1899, p. 5. ¢ Willey’s Zoological Results, Pt. 4, 1899. PROCEEDINGS U.S. NATIONAL MUseEuM, VOL. XXXVI—No. 1671. 381 389 PROCEEDINGS OF THE NATIONAL. MUSEUM. you, xxxvt. anatomy of the various neotropical and Malayan species this may be done. On the whole, in spite of the currents, it seems likely that the Pacific species are of Asiatic origin; and perhaps it may be assumed that in past times the currents of the Pacific Ocean were very different from what they are to-day. This is especially suggested by the fact that whereas the North Equatorial current, passing the Hawaiian Islands, comes from the American coast, the Hawaiian fauna shows every indication of having arrived from the opposite direction.* Pilsbry’ draws attention to the antiquity and comparative homo- geneity of the mid-Pacific snail-fauna, and considers that the former (late paleeozoic or early mesozoic) existence of a mid-Pacific conti-- nent is strongly indicated. This may be true, but it hardly seems possible to account for the Veronicellé as remnants of the ancient fauna; and we are thus left to assume the agency of ocean currents, unless it can be believed that man carried these slugs about during . his early migrations. VERONICELLA AGASSIZI Cockerell. This species was described® from a specimen obtained by Dr. Alex- ander Agassiz in Tipaerui Valley, Tahiti. It was not possible to determine the anatomical characters, and nothing was known of the variation. Mr. R. W. Doane, of Stanford University, when recently in Tahiti, kindly collected a considerable series of specimens, which have made it possible to put the species on a much soimder basis. The original description of the external features remains valid; the specimens received are very uniform in appearance, the younger ones being paler than the adults. The dorsal surface is granular with small warts; its color coffee-brown, marbled with black, with no dis- tinct dorsal band. Some of the younger specimens show slight indi- cations of a dorsal band, consisting merely of a stripe on which the mottling is absent, and along the margins of which it tends to accu- mulate.? The general form of the animal is shown in figs. 2 and 5. V. gilsont Collinge,’ from the Fiji Islands, appears to be the nearest ally. So far as the external measurements go, there is no substantial difference. The color of V. gilsoni appears to be distinctive, and at present can not be matched in any of the Tahitian material. The 4An ancient elemert of the Hawaiian flora, including several endemic genera of Composite, appears to be of American origin; while an apparently more recent group is Polynesian (Wallace, Island Life, 2d ed., pp. 325-6). The Ta- hitian Composite show Malayan affinities. > Proce. Acad. Nat. Sci. Phila., 1900, pp. 568-581. PTO. 1, (S. Nat. UUs, kos Len Desoo: 4@The same sort of thing, more exaggerated, is seen in the South American Veronicella fusca Heynemann. ¢ Journ, of Malacology, VII, p. 179, no. 1671. NHW SLUGS OF GENUS VERONICELLA—COCKERELL. 383 anatomy of V. gilsoni is unknown. The following table of Pacific Veronicellw will assist in the separation of the species: Dorsum with a very distinct light yellow band; female orifice close to sole; URGE SURO TLCS eet UO) ee a en ee a Pe A te ae. V. willeyi Collinge. Dorsum without a band, or with at most traces of one, not lighter than the EPA REPEL (GUO U GSR I BS a eS ee ole 1. Ground color dirty yellow, with small blackish bloteches___V. gilsoni Collinge. RAEN MCL OTC ur: SCO COVDRO Wise otek ee one ee ee ek Dy 2. Form very broad, the breadth about half the length (according to the figure) V. brunnea Collinge, (Hvidently based on a juvenile.) Form narrower, the breadth less than half the length; accessory glands 8 V. agassizi Cockerell. Our largest V. agassizi is 36 mm. long and 14 mm. broad, the sole 3.5 mm. broad and the female orifice 3 mm. from sole and 23 mm. from anterior end. The average measurements were: Length, 25.6 mm.; breadth, 11.6; breadth of sole, 2.7; female orifice from sole, 2.1; female orifice from margin, 2.8, and from anterior end, 14.8. .The Temale orifice is distant from the head 56.4 per cent of the total length. The jaw is normal, strongly ribbed, as is sufficiently shown in fig. 1. The teeth are normal for the genus, with blunt dark points. The median tooth is small and narrow, the cusp not projecting be- yond the basal plate. The first laterals are large and simple in form (fig. 4). The marginal teeth, as usual, are much smaller than the laterals. The male generative organs are normal for the genus. The acces- sory glands are strikingly different from those of V. willeyi, being much shorter (little more than half the length of the dart-sac) and fewer in number (eight in the specimen examined). In V. willeyi they are not only numerous, but conspicuously longer than the dart- sac. The intestine is formed essentially as in V.* wélleyi and V. . brunnea, but the stomach is covered by the liver. (See fig. 3.) VERONICELLA SCHIVELYZ BAHAMENSIS Dall. This form was described by Doctor Dall“ from Nassau and Little Abaco, Bahamas. The typical V. schivelyw Pilsbry comes from Ber- muda and is exceedingly similar to the Mexican V’. moreleti Crosse and Fischer. Upon comparing the descriptions of V. moreleti and schivelye (external characters), nothing distinctive was found, except that the female orifice was slightly more posterior in schivelyw, and the dorsal bands were rather nearer to each other than either to the edge of the body. In the bahamensis form the bands are more or less evanescent in the adult though well-marked in the young. In all a strongly distinctive feature is the projection of the foot beyond the body posteriorly. 4Smithsonian Misc. Coll., XLVII, p. 446. 884 PROCEEDINGS OF THE NATIONAL MUSEUM. VOL. XXXVI. As the anatomy of the schivelyw forms was unknown, we were anxious to dissect specimens. Doctor Dall has very kindly sent us a specimen of bahamensis collected in the Bahamas by Mr. Riley. It is only half the size of the types, and unusually pale in color. Un- fortunately, the generative organs were not developed, but fig. 8 represents the under surface, showing the projecting foot, and fig. 7 one of the alimentary canal and liver. On the whole, it seems not unlikely that the Bahama and Bermuda slugs have been introduced from Mexico, and are in fact V. moreleti. This can not be demonstrated, however, until much more material is obtained, both of the Mexican and insular forms. The species of Florida, Cuba, and Jamaica, so far as known, are not of the moreleti type. The measurements of the specimen of V. schivelyw bahamensis examined by us are as follows: Length, 38 mm.; breadth, 16; breadth of sole, 5; female orifice from the sole, 2.5; from margin, 3.5; from anterior end, 21. The female orifice is distant from the head 55 per cent of the total length. EXPLANATION OF PLATE 32. Veronicella agassizi Cockerell. Fic. 1. Jaw. 2. Diagrammatic transverse section through the body. 0. ¢, body cavity. 3. Male generative organs. dc. gl., accessory glands. d. s., dart-sac. p., penis. 1”. m., retractor muscle, v. d., vas deferens. 4. Teeth. 0b. p., basal plate. c¢., central tooth. /., lateral. 5. View from ventral side. X22. Female generative orifice. 6. Alimentary canal and liver. cr, crop. e., esophagus. int., intestine. l., liver. st., stomach. Veronicella schivelye bahamensis Dall. . Alimentary canal and liver. Lettering as in fig. 6. 8. View from ventral side. Natural size. 9 Female generative orifice. PROCEEDINGS. VOL. XXXVI PL. 32 U. S. NATIONAL MUSEUM 0 in i nl Two SLUGS OF THE GENUS VERONICELLA. FoR EXPLANATION OF PLATE SEE PAGE 388. ] ANYAM GILA (MAD. WEAVE): A MALAYSIAN TYPE OF BASKET WORK. By Oris T. Mason, . Late Head Curator, Department of Anthropology, U. S. National Museum, In the W. L. Abbott collections of basketry from southwestern Malaysia, in the U. S. National Museum, are a number of specimens made in a variety of technic not known in America. It was first made public by Mrs. L. E. Bland, of the Penang Resi- dency, Straits Settlements, in the Journal of the Straits Branch, Royal Asiatic Society, No. 46. Mrs. Bland studied the art among the Malay women of Tanjong Kling, Malacca, and also mentioned the same ware from the province of Wellesley, from Siamese terri- tory, from elsewhere on the peninsula, as well as from Sumatra and other islands. . The baskets are made from narrow strips of pandanus, or screw pine, leaves, of which there are many species. In the specimens de- seribed by Mrs. Bland the “ mengkuang” (Pandanus fascicularis) was used. The material is prepared by the old women, who cut the long, prickly leaves with a woman’s parang, or native knife,* and carry them home in large bundles on their heads. They next dry, or “Jayor,” the leaves over a fire of sticks and cut off the thorns that grow down the spine. This divides the leaf into two wide strips, for which purpose a smaller knife, called “ pisau,” is used.? The women then divide the half leaves into uniform strips by means of a rude gauge, or “ jangka.” ¢ This is a flat piece of wood with brass spikes fixed into one end at regular intervals, governed by the width of the required strips, varying from { inch to 1 inch. At the same time the thorny edges are removed. The strips are next made supple and smooth, or “ lurut,” with a “ pulurut,” a piece of hollow bamboo pulled over a leaf many times “See Journ. Roy. Asiatic Soc., Straits Branch, pl. 4, fig. a; ane pl. 15, Proce. U. S. Nat. Mus., XXXV. bTdem, pl. 4, fig. DB ¢Tdem, pl. 4, fig. c. PROCEEDINGS U. S. NATIONAL MUSEUM, VOL. XXXVI—No. 1672. + Proc. N. M. vol. xxxvi—09——25 385 326 PROCEEDINGS OF THE NATIONAL MUSEUM. you. xxxvt. by the left hand with a curling movement.* The smooth strips are folded into compact bundles and soaked in cold water for three nights, changing the water twice a day. After this they are bleached in the sun for a day and are then ready for use. Mrs. Bland’s account of the preparation of the materials is most helpful, and, before giving a detailed description of the drawings and processes shown, an abstract of her description will throw much light on the native practice. The basket starts from six strands crossed in the middle of the bottom and the fabric is built up by working in three directions, braiding, not weaving. The strands go from left to right in the process. After the work reaches the rim of the basket, the strips are turned back over a rattan split and inwoven, as will be fully ex- plained. The strips of pandanus are glossy on one side only. In this they resemble the split roots and cane of the Pacific coast Indians and of the tribes of the Gulf States. In order to have the basket smooth on both sides, the native women work their splits and strips in pairs. But the “mad weave” maker proceeds on an entirely different plan. After the basket has been wrought from bottom to rim in single ply, the strips are inwoven backward to the center of the bottom by tucking under, the glossy side being outward and the ends of strips being cut off invisibly. On the way back pretty designs are frequently made by curling and folding the strips between thumb and forefinger. Mrs. Bland speaks of these as “rice grains” and they are worked into stars or hexagons, which are further bunched into single or combined geo- metric patterns.? The “mad weave” is worked up into various shapes, but the hexagon is the prevailing form. All of them—square, oblong rec- tangular, oval, and diamond shape—start with the six-sided motive and are brought by skill into other designs. In 1906 Henry Balfour, esq., of Pitt-Rivers Museum, Oxford, England, took up the “mad weave” and reproduced it in tapes of three colors. He described the system as an under-two-and-over-one (4) and an under-one-and-over-two (7) system of interweaving. The difficulty comes from having three sets of parallel strands. In 1907 I was so fortunate as to secure the cooperation of some arts and crafts friends, with the result that Miss Edwina H. Fallis, of Denver, Colorado, through Mrs. Wright Jones produced a one-ply specimen, and Mrs. Mary Wright Gill, of Washington City, worked out the mad weave detail now to be described and illustrated. “Straits Branch Journal, No. 46, pl. 4, fig. d. bTdem, pl. 6. NO, 1672. A MALAYSIAN TYPE OF BASKET WORK—MASON. 3887 The technic is from pandanus strips of various widths, in close twill, in three directions, to be spoken of as vertical, dextral, and sinistral; the terms right oblique and left oblique may replace the last two. When the work is finished the surface is made up of rhombs or diamond-shape checks, giving the appearance of cubes and six-pointed stars. As before mentioned, the fabric is twice wrought, or two ply, so as to have both the inside and the outside of the basket expose the glossy side of the leaf. The technic here detailed is based on Cat. Nos. 219963 and 236282 in the U. S. National Museum. The former is in the W. L. Abbott collection from Rumpin River, Pahang, and the latter is from Malacca, sent to the Museum by Mrs. L. E. Bland. < : x S Fig. 1.—MrruHop OF BEGINNING MAD Vic. 2.—MbprHoD OF ADDING ADDITIONAL WHAVE, STRIPS. Fig. 1 shows how the work is started from the center of the bottom, with six prepared strips, two sinistral, two vertical, and two dextral, crossing at the middle, one of each pair passing over and under one strip in each of the other pairs.- This is the foun- dation. , Fig. 2, a-c, illustrates how strips are next added. A vertical strip (a) passes down over two sinistrals, under one dextral and over one dextral. The dextral strip (4) passes upward over one sinistral, under one sinistral, and over two verticals. The sinistral strip (c) passes upward over two dextrals, under two verticals and over one vertical. ; 388 PROCEEDINGS OF THE NATIONAL MUSEUM. VOL. XXXVI. Fig. 3, a, shows the result of adding more strips in the three directions and fig. 4, looking from the inside of the basket is. fig. 3 Fic. 3.—RESULT OF ADDING NEW STRIPS IN Fie. 4.—FIGURE 38 DISSECTED THREE DIRECTIONS. (INSIDE). dissected. The sinistral strips are uniformly under two and over one vertical and over two and under one dextral (figs. 5@ and 5d). Hic. 5 a, POSITION OF SINISTRAL STRIP}; Db, POSITION OF SINISTRAL STRIP; C, POSITION OF DEXTRAL STRIP; d, POSITION OF DEXTRAL STRIP; €, POSITION OF VERTICAL STRIP; 7, POSITION OF VERTICAL STRIP. No. 1672. A MALAYSIAN TYPE OF BASKET WORK—MASON. 889 The dextral strips are uniformly over one and under two sinistral and over two and under one vertical (figs. 5¢ and 5d). The vertical strips are uniformly over one and under two dextral and over two and under one sinistral (figs. 5e and 5/). MAM Fic. 6.—METHOD OF GIVING Fig 7.—METHOD OF FINISH- HEXAGONAL FORM TO BASE, ING AT THE BORDER, WITH AT UPSET. TWO HOOPS OF RATTAN. In fig. 6 is shown a continuation of the bottom in single technic, with the rough side of the strips outward, indicating the method of giving the hexagonal form to the base at the upset. Fig. 7 illus- Fic, 8.—METHOD OF TURNING STRIPS AT THE BORDER, POLISHED SIDE OUT. trates the interweaving of strips to form sides and the turning at the border over two hoops of rattan. At the lower end of these hoops the dextral and sinistral strips cross (fig. 8) and are inwoven Fic. 9.—METHOD OF TURNING VERTICAL STRIPS AT THE BORDER, POLISHED SIDE OUT. back with the polished side out, over the whole surface of the basket and are cut off where they meet, usually at one side of the bottom, or at the upper edge of the design, if there be one. 390 PROCEEDINGS OF THE NATIONAL MUSEUM. vou. xxsvt. The vertical strips are carried over the upper hoop—the alter- nate ones moving from outside and from inside—and are inwoven yh | iy Aree, ZY Zi Fic. 10.—MertTHoD OF FINISHING AT THE BORDER. back over the rough surface, leaving the polished side out (figs. 9 and 10). Fic. 11.—METHOD OF ORNAMENTATION. The ornamental designs, which are not common in the Abbott collections, are formed, as mentioned, by curling or twisting the outer layer of strips as they are inwoven (fig. 11). le ole i ee ON A COLLECTION OF RECENT CRINOIDS FROM THE PHILIPPINE ISLANDS. By Austin Hosarr Cruark, Collaborator, Division of Marine Invertebrates, U. S. National Museum. The second consignment of crinoids received from the United States Fisheries steamer A/batross, now working among the Philip- pine Islands, proves to be a collection of very considerable interest, and well worth a separate report. The first consignment included fifty-two species, four of which, however, were laid aside until addi- tional material could be obtained for comparison; these are described herein. The present lot contains twenty-nine species, fifteen of which (or about half) were not met with before, while of the species represented in the first lot thirty-one (more than half) are not in- cluded. A few of the Challenger species yet remain to be rediscoy- ered, while several genera, known from both north and south of the islands, have not as yet been discovered there. Endoxocrinus alternicirrus has been rediscovered for the first time, ° and Hypalocrinus naresianus, dredged by the Challenger and the Siboga, has been again found. Metacrinus wyvillii, known from the Kermadec and Ki islands, proves to occur also in the Philippines, while a peculiar comatulid type, first found off the Meangis Islands and later at South Africa, is represented in the collection. Family COMASTERIDA. Genus COMASTER L. Agassiz. COMASTER SENTOSA (P. H. Carpenter). Obtained at station 5249, Rakiputan Strait, between the northern end of Samal Island and the west coast of Davao Bay; 23 fathoms. COMASTER MULTIRADIATA (Linnezus). A single specimen from station 5249, Rakiputan Strait, between the northern end of Samal Island and the west coast of Davao Bay; 23 fathoms; has 20 arms (all the II Br series present), with PROCEEDINGS U. S. NATIONAL MUSEUM, VOL. XXXVI—No. 1673. 391 8399 PROCEEDINGS OF THE NATIONAL MUSEUM. VOL, XXXVI. oblong and strongly overlapping. brachials; but the cirrus joints agree with those given for C’. multiradiata by Carpenter. A critical study of a very large series shows that Comatula fimbri- ata, Actinometra borneensis, and Actinometra coppingeri should all be referred to the Linnean Astertas multiradiata. Other specimens were obtained at stations 5161, between Simaluc Island and Tawi Tawi; 16 fathoms; and 5165, south of San Gasangz Island (Tataan group) ; 9 fathoms. Genus PHANOGENIA Lovén. PHANOGENIA MULTIBRACHIATA (P. H. Carpenter). This species was secured a station 5248, Rakiputan Strait, between the northern end of Samal Island and the west coast of Davao Bay; 18 fathoms; station 5249, same locality; 23 fathoms; and station 5254, same locality; 21 fathoms. PHANOGENIA CARPENTERI (A. H. Clark). One specimen was dredged at station 5153, east of Port Dos Amigos, Tawi Tawi; 49 fathoms. This is the species which was described by Prof. Johannes Miiller under the name of Alecto multifida, which name, however, he ex- plicitly states to be a substitute for multiradiata Lamarck, not multi- radiata Goldfuss. The elements of the III Br series are united by synarthry, and this led Doctor Carpenter to place the species in his “ Parvicirra group,” corresponding, in part, to the genus Comanthus; but all except the II Br series, which are 4 (3+4), are 2, a condi- tion quite unknown in Comanthus, the proximal pinnules are slender instead of stout, and terminal combs occur at intervals along the mid- dle and distal pinnules; these combs also are abrupt, with a few long, curved teeth. In spite of the synarthrial instead of syzygial union of the III Br and IV Br series, the affinities of this form are distinctly with the members of the genus Phanogenia as restricted, and to that genus it must be assigned. The specimen at hand has about forty arms. PHANOGENIA MINIMA, new species. Centro-dorsal stellate, without cirri; radials entirely visible, trape- zoidal, proximally twice, distally three times (or rather more), as broad as long; I Br, entirely free laterally, four times as broad as long, decreasing slightly in diameter anteriorly; I Br, (axillary) broadly pentagonal, twice as broad as long; II Br 4 (8+4), rarely 2; III Br 2 (142); I1V Br 2 (142), but rarely present. About forty arms 70 mm. long; brachials and pinnules as in P. gracilis, but proportionately more slender and delicate. no. 1673. RECENT CRINOIDS FROM THE PHILIPPINES—CLARK. 8398 Color—Chrome yellow, with numerous longitudinal narrow lnes on the radial and division series, and transverse lines on the arms, of dark brown. Type—Cat. No. 25469, U.S.N.M., from AJ/datross station 5108; west of Nasugbu Point, central Luzon (east of Simo Bank); 16 fathoms. This delicate little species appears to be quite distinct from any of those heretofore described; the absence of regular division series be- yond the III Br, these last being 2 (1+2), are sufficient to distin- guish it at once. Though very small, the type-specimen has the appearance of being fully developed. PHANOGENIA DELICATA, new species. Centro-dorsal large, discoidal, with a broad flat polar area 5 mm. in diameter; cirrus sockets in a single, slightly irregular, marginal row. Cirri XTII-XIX, 24-25, moderately stout, 20 mm. to 25 mm. long; first joint about twice as broad as long, the following gradually in- creasing in length to the fourth or fifth, which is as long as broad; next three or four joints one-third to one-half again as long as broad; following joints decreasing in length, the terminal fourteen or fifteen being about twice as broad as long; the second and following joints have their distal dorsal edges slightly prominent; after the seventh or eighth, which is a well-marked transition joint, the distal dorsal edge becomes strongly produced and finely serrate, this production very gradually narrowing anteriorly, at the same time gradually becoming crescentic, and gaining in height; the antepenultimate joint bears a subterminal transverse ridge, bending forward on each side as it decreases in height; opposing spine small, median, erect, trans- versely elongate, not reaching one-fourth the dorso-ventral diameter of the penultimate joint in height; terminal claw longer than the penultimate joint, moderately stout, evenly tapering, and moderately and evenly curved. Ends of the basal rays visible as small tubercles in the angles of the calyx, bridging over the narrow clefts between the centro-dorsal and the primary division series; radials only very slightly visible over the ends of the basal rays; I Br, almost entirely, or quite, con- cealed, very short, almost entirely united laterally; I Br, (axillary) broadly pentagonal, twice as broad as long, the anterior angle rather produced, the lateral edges free; II Br 4 (3+4), well separated laterally; II Br, united for the proximal two-thirds, separated by a broad U-shaped gap distally; III Br 2, very rarely 4 (3+4). Arms thirty or thirty-one, slender, 70 mm. to 80 mm. long; first two brachials approximately equal, slightly wedge-shaped, about twice as broad as the median length, the first almost entirely united 894 PROCEEDINGS OF THE NATIONAL MUSEUM. VOL. XXXVI. interiorly; third and fourth brachials (syzygial pair) half again to twice as broad as long, oblong; next three brachials oblong, about twice as broad as long, then becoming wedge-shaped, almost triangu- lar, half again as broad as long, after the proximal third of the arm becoming gradually less obliquely wedge-shaped, though remaining of about the same proportionate length, and in the terminal portion of the arm becoming wedge-shaped, and-as long as or longer than broad. The elements of the division series have everted and promi- nent, finely spinous, distal ends; the brachials have strongly produced and overlapping, finely spinous, distal ends. Syzygies occur between the third and fourth brachials, again between the eleventh and- twelfth or twelfth and thirteenth, and distally at intervals of four oblique muscular articulations. Disk 13 mm. in diameter, with a few scattered calcareous granules about the central anal tube; mouth marginal and radial. Py 15 mm. long, slender, with forty joints, all approximately as long as broad; terminal comb arising suddenly, with seven teeth, triangular, longer than broad, basally in apposition, rather longer than the diameter of the joints which bear them, rather strongly incurved; P, similar, shghtly less stout basally, equal in length or slightly longer; P, small and weak, 6 mm. long; P, similar, but 5 mm. long (twenty joints); P, 6 mm. or 7 mm. long, consider- ably stouter than the two preceding pinnules, with the distal ends of the joints strongly produced, and bearing a genital gland; following pinnules slightly stouter, with larger genital glands, and slowly increasing in length; distal pinnules very slender, about 9 mm. long, with about twenty joints, the first two short, then increasing in length to about twice as long as broad and decreasing again in the terminal portion. Color (in spirits).—White, the arms beyond the basal portion with broad broken lateral lines, and dorsal bands, of violet; cirri white, with occasional bands of light violet. Type—Cat. No. 25463, from Albatross station 5153; east of Port Dos Amigos, Tawi Tawi; 49 fathoms. Genus COMATULA Lamarck. COMATULA PECTINATA (Linnzus). One specimen from station 5152; off Tawi Tawi; 34 fathoms. no. 1673. RHCENT CRINOIDS FROM THE PHILIPPINES—CLARK. 8395 Genus COMATELLA A. H. Clark. COMATELLA NIGRA (P. H. Carpenter). This species was obtained at station 5253; Rakiputan Strait, be- tween the northern end of Samal Island and the west coast of Davao Bay; 28 fathoms.* Genus COMANTHUS A. H. Clark. COMANTHUS NOBILIS (P. H. Carpenter). Specimens were obtained at station 5249; Rakiputan Strait, be- tween the northern end of Samal Island and the west coast of Davao Bay; 23 fathoms. . Station 5250; same locality; 23 fathoms. Station 5253; same locality; 28 fathoms; and Station 5254; same locality; 21 fathoms. The examples from stations 5249 and 5250, and one of those from station 5254, have each two well-developed cirri. COMANTHUS BRIAREUS (Bell). I have recently been able to study a large series of specimens of this -and allied species, and have succeeded in dividing them up into cer- tain specific groups which appear to be circumscribed by definite and well-defined characters, and which will, I believe, stand the test of future investigations. Briareus was first described by Bell from Port Denison, Australia, but he entirely overlooked its very obvious affinities, placing it in the genus “Antedon,” between two species of Zygometra. The division series subsequent to the III Br are described as “two joints, no syzygy,” but in the figure most of them are 4 (3+4). Carpenter ealled attention to both of these errors in 1888, and at the same time described a supposedly new species, divaricata, from Banda; divari- cata differs from briareus only in having the centro-dorsal small and stellate, with no trace of cirri, this being in briareus a thin disk with from fifteen to twenty partially developed cirri. The present series shows all variations between the two extremes, and it therefore be- comes necessary to consider the two identical. The type of dévarcata js slightly smaller than that of briareus, though more developed ; but @ Under the name of Antedon bassett-smithi Professor Bell has described (and, fortunately, figured) a specimen of Comatella stelligera from the Macclesfield Bank, just west of the Philippine Islands. He refers this supposed new species to Carpenter's “ Spinifera group,” and discourses at great length upon the “ ex- traordinary divergences exhibited by the syzygies of this species.” Had the “species” been referred to the “Stelligera group” of ‘“Actinometra,” instead of to the “Spinifera group” of “Antedon,” this peculiarity, as well as all the other supposed peculiarities, would have been found to be quite normal. Being now correctly identified, there is no longer any danger that ‘‘Antedon” bassett- smithi “‘ will severely shake our faith in the value of the site of the syzygy as an aid in specific diagnosis.” 396 PROCEEDINGS OF THE NATIONAL MUSEUM. VOL, XXXVI. the difference is not great, and the size at which any given comasterid may become mature is always variable. Carpenter’s magnifica, from the Philippine Islands, is very close to briareus, representing merely a somewhat more advanced stage along the same phylogenetic line, and many specimens of the latter approach it more or less; but mag- nifica 1s sufficiently well marked off to retain its present status as a distinct species. The Philippine specimens referred to briareus were compared with ~ a typical Australian example and no differences were found. The Albatross dredged Comanthus briareus at station 5138; be- tween Jolo and Pangasinan islands; 19 fathoms. Station 5142; north of Jolo (town) ; 21 fathoms. Station 5147; off Balinpongpong Island (south of Jolo); 21 fathoms. Station 5148; Rakiputan Strait; between the north end of Samal Island and the west coast of Davao Bay; 18 fathoms. Station 5249; same locality; 23 fathoms; and Station 5254; same locality; 21 fathoms. COMANTHUS POLYCNEMIS, new species. In my previous list I referred, rather doubtfully, to Comanthus ulternans, a specimen resembling C. briareus in all respects, except that the division series distal to the II Br are 2 until the last division series is reached, which is usually 4 (3+4). The present collection contains numerous examples of the same form which appears to be quite constant, and I therefore propose to recognize it under the name of Comanthus polycnemis. It may be diagnosed as follows: Centro-dorsal pentagonal or stellate, with no trace of cirri; general build and proportions as in C. briareus; rays dividing four to six (usually four or five) times; II Br 4 (8+4) ; subsequent divisions 2, except that the outermost divisions, especially on the outer side of each III Br series, are usually 4 (38+4). The scarcity of 4 (8+4) division series, which, though always present, are confined to the outer parts of the rays, distinguish this species at once from all others. Bell’s variabilis, and the multifida of Miiller and Carpenter, placed by the latter in the “ Parvicirra group,” and therefore presumably considered by him as near briareus, belong in reality to the genius Phanogenia and are near P. typica. Type.—Cat. No. 25467, U.S.N.M., from Albatross station 5249; Rakiputan Strait, between the northern end of Samal Island and the west coast of Davao Bay; 23 fathoms. The Albatross dredged this species at Station 5139; between Jolo and Pangasinan Islands; 20 fathoms. Station 5147; off Balinpongpong Island (south of Jolo); 21 fathoms. no. 16738. RECENT ORINOIDS FROM THE PHILIPPINES—CLARK. 897 Station 5248; Rakiputan Strait, between the northern end of Samal Island and the west coast of Davao Bay; 18 fathoms. Station 5249; same locality; 23 fathoms. Station 5250; same locality; 23 fathoms. Station 5251; same locality; 20 fathoms. Station 5959; same locality; 28 fathoms. Station 5253; same locality; 28 fathoms. Station 5254; same locality; 21 fathoms. There are slap specimens with no definite locality given. One of the specimens from station 5249 is four rayed. COMANTHUS DUPLEX (P. H. Carpenter). A single specimen was secured at station 5252; Rakiputan Strait, between the northern end of Samal Island and ‘ie western coast of Davao Bay; 28 fathoms. COMANTHUS ROTALARIA (Lamarck). This species was found at station 5218, east of the northern end of Burias Island (south of Luzon) ; 20 fathers. Station 5248; Rakiputan Strait, between the northern end af Samal Island and the western coast of Davao Bay; 18 fathoms. Station 5253; same locality; 28 fathoms; and Station 5254; same locality; 21 fathoms. COMANTHUS ALTERNANS (P. H. Carpenter). One specimen was dredged at station 5252; Rakiputan Strait, between the northern end of Samal Island and the western shore of Davao Bay; 28 fathoms; and another at station 5254; same locality; 21 fathoms. Family HIMEROMETRIDA. Genus PONTIOMETRA A. H. Clark. PONTIOMETRA INSPERATUS, new species. Centro-dorsal hemispherical, the dorsal pole small, shghtly convex; cirrus sockets arranged roughly in three alternating rows, more or less crowded. Cirri XX, 41-52, 30 mm. long, stout, but tapering in the distal half, and comparatively slender at the tip; first joint three times as broad as long, second and third twice as broad as long, the following gradually increasing in length to about the seventh, which is about one-third broader than long; next three or four similar, the joints then gradually decreasing in length, those in the distal third of the cirrus being uniformly twice as broad as long; sixth to eighth and following joints with the distal dorsal edge everted in the shape of an open V-shaped ridge, composed of an apical round tubercle, and two lateral more or less elongate tubercles; distally this ridge gradually becomes less and less V-shaped, composed of four or five tubercles, 898 PROCEEDINGS OF THE NATIONAL MUSEUM. VOL. XXXVI. the apical tubercle, however, remaining in the same position, and therefore occupying a place below the center of the now almost straight tubercular ridge; as the joints decrease in length distally, the apical tubercle gradually disappears, and the transverse ridge moves gradually to a median position; on becoming median, it at first has usually four tubercles, this being later reduced to three, while beginning on about the fifteenth from the end there are only two, the last three or four joints before the penultimate bearing only a single median tubercle; opposing spine comparatively large, aris- ing from the entire dorsal surface of the joint,*the apex subterminal, equal to about half the diameter of the penultimate joint in height; terminal claw somewhat longer than the penultimate joint, rather stout, and strongly curved. Ends of the basal rays not visible; radials short in the median line, but extending far up into the angles of the calyx, reaching the disk, and separating the bases of the I Br,; I Br, slightly trapezoidal, about twice as broad basally as long, very widely separated laterally, well rounded dorsally; I Br, (axillary) pentagonal, about as long as broad; II Br 2, one series only being present. Arms eleven in number (in the type), very widely separated. All the arms are broken off at the syzygy between the third and fourth brachials. First brachial shghtly wedge-shaped, half again as broad as long, interiorly united for the anterior half or two-thirds, diverg- ing distally as approximately a right angle; second brachial nearly square; third (hypozygal) oblong, three times as broad as long. P, slender, 15 mm. long, evenly tapering, with twenty to twenty- five joints, the first twice as broad as long, the second and third squarish, the following about twice as long as broad, shorter termi- nally. Color.—Purplish brown. Type—Cat. No. 25468, U.S.N.M., from AJdbatross station 5145; off Jolo town; 23 fathoms. Genus CENOMETRA A. H. Clark. CENOMETRA DELICATA, new species. Centro-dorsal discoidal, the polar area slightly concave; cirrus sockets marginal, arranged in two closely crowded alternating rows. Cirri XIX, 31-33, 20 mm. long, comparatively slender; first joint short, the remainder subequal, about twice as broad as long; joints somewhat flattened dorsally, after the tenth bearing very small blunt paired median dorsal tubercles; opposing spine triangular, arising from the entire surface of the penultimate joint, blunt, the apex median in position; terminal claw about as long as the penultimate joint, rather stout and strongly curved. Ends of the basal rays visible as flattened tubercles in the angles of the calyx, but difficult to differentiate from the centro-dorsal ; no. 1673. RECENT OCRINOIDS FROM THE PHILIPPINES—CLARK. 8399 radials visible, but short; slightly divergent distally; I Br, trape- zoidal, decreasing slightly in diameter anteriorly, proximally about two and one-half times as broad as long; I Br, (axillary) broadly pentagonal, about twice as broad as long, the lateral edges about as long as those of the I Br,; II Br 2; III Br 2; IV Br 2, only found on the outer side of the I Br series; all the joints to and including the second brachial with prominent lateral processes, the outer edges of which form a line parallel to the axis of the division series. Thirty-five arms (in the type) 85 mm. long, more slender than in the other species of the genus; brachials as in C. wnicornis, but pro- portionately slightly longer. The pinnules are in general like those of C. wnicornis, but P., is much more slender, 9 mm. long, with sixteen or seventeen joints, the first nearly twice as long as broad, becoming squarish on the third, and then gradually becoming about one-third longer than broad; from the fourth onward the distal dorsal edge of the joints is strongly produced, standing out in the form of a coarsely spinous crescentic ridge. © Color.—Deep violet, the cirri, P,, and the dorsal side of the other pinnules, bright yellow. Type.—Cat. No. 25465, U.S.N.M., from Albatross station 5248; Rakiputan Strait, between the northern end of Samal Island and the west coast of Davao Bay; 18 fathoms. Another specimen, entirely deep violet, with thirty-four arms, was dredged at station 5249; same locality; 23 fathoms. CENOMETRA UNICORNIS (A. H. Clark). Two fragmentary specimens from station 5108; west of Nasugbu Point (off Simo Bank) central Luzon; 16 fathoms. Genus STEPHANOMETRA A. H. Clark. STEPHANOMETRA TENUIPINNA (Hartlaub). One mutilated specimen was dredged at station 5174; off Jolo town; 20 fathoms. Genus CYLLOMETRA A. H. Clark. CYLLOMETRA MANCA (P. H. Carpenter). One specimen from station 5154; off Tawi Tawi (between Simaluc and Tawi Tawi); 12 fathoms; and others from station 5212; off Masbate; 80 fathoms. Genus OLIGOMETRA A. H. Clark. OLIGOMETRA PULCHELLA A. H. Clark. One small mutilated specimen from station 5248; Rakiputan Strait, between the northern end of Samal Island and the west coast of Davao Bay; 18 fathoms. 400 PROCEEDINGS OF THE NATIONAL MUSEUM. VOL. XXXVI. 4 Family TROPIOMETRIDA. Genus CALOMETRA A. H. Clark. CALOMETRA CARDUUM A. H. Clark. One fine specimen from station 5166; off Simonor Island; 97 fathoms. Genus PTILOMETRA “A. H. G@lark. PTILOMETRA PULCHERRIMA, new species. Centro-dorsal large, columnar, the sides parallel, terminated by a group of five large tubercles each radial in position, arising from an otherwise flat polar area; small circular space bounded by the ends of these tubercles light in color, this hght color extending in interradial lines between the summits of the tubercles and thence to the periphery of the polar area; cirrus sockets arranged in ten columns of three or four each, the radial areas being separated from each other by a low, rounded ridge, and the pairs of columns in each radial area being separated by a broad, shallow groove about twice as broad as the rounded interradial ridges. The centro-dorsal is 5 mm. long and 6 mm. in diameter. Cirri long and slender, XX XV, 80-85 (the less developed as few as 68), 75 mm. to 80 mm. long; first joint short, second about twice as broad as long, the following gradually increasing in length to the sixth or seventh, which is as long as broad; following fifteen to twenty joints between one-third and one-half again as long as broad, then very slowly decreasing in length, the terminal thirty to thirty-five joints being about twice as broad as long; the cirri are somewhat com- pressed distally; after the seventeenth to the twentieth joint the median part of the distal dorsal edge begins to project as a sharp and slender spine, directed diagonally forward; this spine gradually increases in length, at the same time arising from more and more of the dorsal surface of the joints, on the short distal joints arising from their entire dorsal surface, with a slightly convex proximal and more strongly concave distal edge, equaling in height about one-half the vertical diameter of the joints; terminal eight or ten joints tapering rather rapidly, at the same time increasing slightly in proportionate length, so that the antepenultimate and penultimate joints are very small and about as long as broad ; opposing spine equal in length to the diameter of the penultimate joint, blunt, the distal edge forming a straight line with the distal edge of the penultimate joint, arising from nearly or quite the whole of the dorsal surface of that joint; terminal claw slightly longer than the penultimate joint, compara- tively stout and strongly curved. Disk naked; all but one of the ambulacra (which divides immedi- ately) are given off in well-separated pairs, so that nine ambulacral — “NO. 1673. RECENT CRINOIDS FROM THE PHILIPPINES—CLARK. 40] grooves reach the mouth; brachial ambulacra naked; pinnule ambu- lacra with small but well developed side and covering plates. Ends of the basal rays visible as small tubercles in the angles of the calyx; radials short, of approximately uniform width all around the calyx, slightly convex proximally and correspondingly concave distally; I Br, oblong, three times as broad as long, in close apposi- tion laterally, with an indicated broad rounded median keel; I Br, broadly pentagonal, two and one-half times as broad as long, the lat- eral edges about half as long as those of the I Br,, in close lateral apposition and sharply flattened; II Br 2, resembling the I Br, but proportionately slightly longer, with the same indicated broadly rounded median keel, sharply flattened laterally and in close apposi- tion. Twenty arms 100mm. long; first two brachials slightly wedge- shaped, about twice as broad as long, sharply flattened and in close apposition exteriorly, the first closely united interiorly, the second in close apposition and flattened interiorly; third and fourth brachials (syzygial pair) oblong, about one-third broader than long; next four or five brachials oblong, about three times as broad as long, then be- coming wedge-shaped, and after the twelfth or fifteenth triangular, two and one-half times as broad as long, in the terminal part of the arm wedge-shaped again and slightly longer, the arm ending abruptly with six or seven very small and short brachials, and curving inward between the terminal pinnules, which exceed the arm in length by 4 mm.; arms broadly convex dorsally in their basal portion (the first seven or eight brachials sharply “ wall-sided”), very gradually be- coming narrower, and distally strongly carinate; at about the twelfth or fifteenth brachial a broadly rounded median keel begins to be indicated} this gradually narrows distally, and on the twentieth to the twenty-fifth the median portion of the distal edge begins to be slightly prominent; this increases slowly in extent at the same time narrowing, so that brachials in the outer half of the arm are bluntly carinate, with the median portion of the distal edge produced, and in the terminal portion sharply carinate, with prominent overlapping spines. P, 8 mm. long, strongly prismatic, slightly less stout than the succeeding pinnules, with sixteen joints, the first short, the second and third squarish, the remainder very slightly longer than broad, be- coming about one-third again as long as broad distally. P, 12.5 mm. long, with nineteen joints, the first twice as broad as long, the second squarish, the remainder very slightly longer than broad; the more distal joints exhibit a tendency toward a slight production of their distal edges at the prismatic angles; the terminal three or four joints taper rather more rapidly than usual. P, 14 mm. long, with seventeen joints, the first twice as broad as long, the second squar- Proc. N. M. vol. xxxvi—09——26 402 PROCEHDINGS OF THE NATIONAL MUSEUM. VOL, XXXVI. ish, then very gradually becoming longer than broad, and twice as long as broad in the terminal portion; the last six joints taper rather rap- idly, and the last two joints are minute; from the fourth joint onward small spines are ee on the distal border, on the distal angle of the pinnule, and the last four joints have, in addition, a somewhat longer spine developed on the corresponding interior angle. P, 15 mm. long, with sixteen joints, all but the basal proportionately longer than those of P,, the distal with long spines on their prismatic angles; P, similar, slightly longer; P, similar, 17 mm. long; P,, is 19 mm. long, with twenty joints, becoming elongate distally, with long spines at the prismatic angles; in the terminal part of the arms the pinnules slowly decrease in length. The abrupt distal taper gives the pinnules the appearance of having been broken off at the tip and sub- sequently repaired. ° Color.—Y ellow-brown. Type.—Cat. No. 25466, U.S.N.M., from Albatross station 5252; Rakiputan Strait, between the northern end of Samal Island and the western shore of Davao Bay; 28 fathoms. family THALASSOMETRID®. Subfamily THALASSOMBE TRIN 4. Genus STENOMETRA A. H. Clark. STENOMETRA ARACHNOIDES, new species. Centro-dorsal moderate, columnar, broader basally than long, de- creasing slightly in diameter distally, the bare polar area 2 mm. in diameter; cirrus sockets arranged in ten columns of two each, closely crowded. Cirri XX (XII in the type), 61-65, 30 mm. long; first four joints subequal, averaging twice as broad as long, rather prominently over- lapping all around; fifth joint nearly half again as long as broad, a more or less marked transition joint; sixth joint about the same length or slightly shorter; next five joints approximately squarish, then gradually decreasing in length, those in the distal half of the cirrus being twice as broad as long, or even slightly shorter; the fourteenth or fifteenth and following joints bear prominent dorsal spines. Disk and ambulacra well plated. Ends of the basal rays visible as dorso-ventrally elongate tubercles in the angles of the calyx; radials concealed, or with the distal, coarsely spinous, margin just visible over the ends of the basal rays; I Br, very narrow, chevron shaped, with abruptly everted, coarsely spinous edges, in close apposition laterally; I Br, (axillary) rhombic, twice as broad as long, the edges concave, abruptly everted and coarsely spinous all around, a high sharp median keel in the proximal two-thirds, NO. 1673. RHCENT CRINOIDS FROM THE PHILIPPINES—CLARK. 403 Arms ten, but all broken off near the base; first two brachials externally, and second and third internally, sharply flattened later- ally; first brachials interiorly united; second brachial large, shield- shaped, deeply incising the very narrow first brachial; first two brachials with more or less everted and coarsely spinous edges; arms with a very narrow, sharp, and moderately high median carination. The pinnules are essentially as in S. hana. Color.—Chrome yellow. Type.—Cat. No. 25470, U.S.N.M., from Albatross station 5154; off Tawi-Tawi (between Simaluc and Tawi-Tawi) ; 12 fathoms. A much mutilated specimen found in a jar with a specimen of Amphimetra discoidea (and therefore probably taken in shallow water), from Port Denison, Australia, certainly belongs to this genus, and possibly to this species. It is slightly smaller than the type, with the keels less produced, and with the spinous edges of the lower joints less pronounced, differences which are in all probability due to im- maturity. The occurrence in the East Indian region of a littoral species of Thalassometride, a family there and elsewhere especially character- istic of the deep-water “ Oceanic ” faunal division, is a fact of very considerable interest. CROTALOMETRA, new genus. Centro-dorsal large, conical, as long as or longer than broad, the cirrus sockets large, arranged in two columns of usually two each in each radial area. Cirri X-XX, 60-80, very long and strongly flattened; first five joints short, then longer than broad, becoming short again distally ; distal joints with the distal dorsal edge produced; all the joints with the edge all around somewhat prominent. Ends of basal rays visible as small tubercles in the angles of the calyx; radials short and bandlike, of uniform height, or concealed ; I Br of moderate length, rounded dorsally, in close lateral apposition and strongly wall sided, the lateral edges everted. Ten to twenty arms; II Br 4(8+4) ; first two brachials in close lateral contact, and sharply flattened, the lateral edges everted ; first four brachials in close apposition and sharply flattened interiorly ; arms stout and.rugged; first nine or ten brachials oblong, about twice as broad as long, tubercular; following brachials triangular, about as long as broad, becoming wedge-shaped and somewhat longer than broad terminally. Syzygies occur between the third and fourth brachials, again between the fourteenth-fifteenth to seventeenth- eighteenth, and distally at intervals of four to ten (usually six or seven) oblique muscular articulations. 404 PROCEEDINGS OF THE NATIONAL MUSEUM. VOL. XXXVI. P, large and stout, but becoming slender distally, strongly flattened exteriorly; following pinnules decreasing rapidly in stoutness, and somewhat in length; distal pinnules stout, strongly prismatic, longer than P,. Color.—Y ellow. enoty pe.—Crotalometra eupedata. Carpenter’s Antedon valida, the systematic position of which has puzzled me greatly, belongs to this genus, and is closely related to C. eupedata, though apparently perfectly distinct; Antedon incerta also should be referred to this genus. The Antedon magnicirra described from South Africa by Professor Bell is likewise a member of this genus, and I have examined two or three additional species from the East Indian region. CROTALOMETRA EUPEDATA, new species. Centro-dorsal large, conical, 5 mm. long and 5 mm. broad at the base, the bare polar area forming the apex of the cone; cirrus sockets arranged in ten columns, one at the outer side of each radial area, so that the cirrus columns of adjacent radial areas are in close apposi- tion, the two columns in each area being separated by a space about equal to their own width; one or two cirrus sockets to each column, usually succeeded by one or two obsolete and more or less obliterated ones. Cirri X-XX, 66, 90 mm. long; first three joints two and one-half times as broad as long; fourth twice as broad as long; fifth half again as broad as long; sixth slightly longer than broad; seventh a transi- tion joint, dull in the anterior three-fourths, highly polished and flat- tened in the distal fourth, not quite twice as long as broad; eighth to eleventh or twelfth joints about twice as long as broad, then grad- ually decreasing in length, becoming squarish about the twentieth, and.twice as broad as long distally; after the fifteenth joint, the distal dorsal edge begins to project slightly, though this is scarcely notice- able until the twentieth is reached, after which it increases in height, becoming more sharply rounded in end view, and arises gradually from the whole dorsal surface of the joint, so that the dorsal profile of the terminal third of the cirrus is strongly serrate; opposing spine a blunt tubercle, the apex subterminal, arising from the whole dorsal surface of the penultimate joint; terminal claw very long and slen- der, twice as long as the penultimate joint, only slightly curved; all the cirrus joints have the distal ends all around slightly projecting and very finely spinous, making the cirri rough to the touch; this projection is slightly more marked on the ventral side than laterally. Disk covered with small plates, very thickly set near the ambulacra, but becoming more scattered toward the periphery in the inter- ambulacral areas; disk ambulacra lined with large regular plates; no. 1673. RECENT CRINOIDS FROM THE PHILIPPINES—CLARK. 405 plating on brachial and pinnule ambulacra very highly developed; perisome of the arms completely covered with rather large inter- brachial plates, so that the arms and pinnules, when the covering plates are closed, are completely encased in a calcareous covering. Ends of the basal rays visible as small, though prominent, tuber- cles in the angles of the calyx; radials of uniform width all around the calyx, short, somewhat over four times as broad as long, the anterior edge set with small scattered spines; I Br, short, of uniform height, the posterior border convex, the anterior concave, about three times as broad as long, the posterior edge slightly prominent, the lat- eral edges in very close apposition, and rather prominently everted, the crest of the resultant ridge finely spinous; I Br, broadly pentagonal, the lateral edges about as long as those of the I Br,, about twice as broad as long, the lateral edges everted and finely spinous like those of the I Br,; like the I Br, and the first two brachials it bears a single small rather prominent rounded tubercle near each lateral margin. Arms ten, stout and rugged, gradually becoming slender distally, 150 mm. long; first brachial longer exteriorly than interiorly, concave anteriorly, the interior edges closely united, the exterior in close apposition, everted and spinous like those of the preceding joints; second brachial about twice as large, irregular in shape, strongly convex posteriorly, in close apposition and strongly flattened with everted and spinous edges, both exteriorly and interiorly; third and fourth brachials (syzygial pair) half again as broad as long, flat- tened exteriorly and interiorly, the edges less everted than those of the preceding joints; following five brachials approximately oblong, rather strongly tubercular, about three times as broad as long, after the twelfth becoming triangular, about as long as broad, this propor- tion remaining unchanged until near the arm tips, where the brach- ials become wedge-shaped, and somewhat longer; the distal edges of the brachials in the outer two-thirds of the arm are overlapping and finely spinous. Syzygies occur between the third and fourth brachials, again between the fourteenth-fifteenth to seventeenth- eighteenth, and distally at intervals of four to ten (usually six or seven) oblique muscular articulations. P, large and very stout, strongly flattened exteriorly, with seven- teen or eighteen joints all broader than long; the pinnule tapers rather rapidly after the proximal third, so that the terminal portion is delicate, with very small joints; P, 7 mm. long, stout basally, though not nearly so stout.as P ,, tapering rapidly, so that the distal half is slender; it is composed of fourteen joints, the first three broad, the fourth about as long as broad, the remainder somewhat longer than broad; the first six joints of P, have the distal side very strongly concave, forming two sharp keels, one external along the 406 PROCEEDINGS OF THE NATIONAL MUSEUM. VOL. XXXVI. flattened outer side, the other internal; the external keel is armed with fine spines; the distal joints are prismatic, with the angles somewhat produced; P, has a similar double carination, but, while the exterior keel is much lower, it persists in a raised and very spi- nous line to the tip of the pinnule; the ends of the distal joints are much more spinous than in P,; P, 6 mm. long, more slender than P,, being in about the same proportion to that pinnule as it is to P,, with twelve joints, at first broad, becoming about as long as broad at the fifth, and longer than broad distally; the pinnule is strongly prismatic, the ridges and the distal ends of the joints being spinous; but the two basal keels are only slightly marked; P, and the follow- ing pinnules from P, onward similar to P,, but slightly more slender, with the joints proportionately shghtly longer ; distal pinnules 12 mm. long, rather stout, strongly prismatic, with eighteen joints, the first crescentic, the second strongly trapezoidal, about as broad as its greater length, the remainder about half again as long as broad; the external ridge is somewhat produced, and finely spinous. Color.—Bright yellow, the cirri lighter. Type.—Cat. No. 25462, U.S.N.M., from Albatross station 5236; off the east coast of Mindanao, north of Lianza Bay; 494 fathoms. A young specimen with arms 60 mm. long was dredged at Station 5116; north of Maricaban Island (between Luzon and Mindoro) ; 200 fathoms; it possesses one II Br series of 4(3+4). Genus PARAMETRA A. H. Clark. PARAMETRA COMPRESSA (P. H. Carpenter). One specimen from station No. 5255; off Davao (town) ; 100 fath- oms; and another very fine example from station 5166; off Simonor Island; 97 fathoms. Subfamily CHARITOME: TRIN 4. Genus PACHYLOMETRA A. H. Clark. PACHYLOMETRA LEVIGATA, new species. Centro-dorsal thick-discoidal or more or less columnar, the polar area flat, 2.5 mm. or 3 mm. in diameter; cirrus sockets arranged in three columns in each radial area, the two outer columns converging distally, usually meeting beyond the middle column; columns usually separated by more or less developed ridges. Cirri XXX-XXXV, 14-15, 15 mm. long; first two joints about twice as broad as long; third joint about as long as broad; fourth to seventh or eighth half again as long as broad, then gradually de- creasing distally, the third and fourth from the end being only no. 1673. RECENT CRINOIDS FROM THE PHILIPPINES—CLARK. 404% slightly longer than broad; the terminal joints are again about half again as long as broad; distal ventral ends of the joints slightly prominent; distal dorsal ends of the outer joints sometimes slightly thickened; opposing spine minute, terminally situated, directed obliquely forward, often barely indicated or altogether absent; ter- minal claw nearly as long as the penultimate joint, moderately slen- der and moderately curved. Disk completely covered with a pavement of small plates; brachial and pinnule ambulacra well plated; large irregular plates over the genital glands. Ends of the basal rays usually visible as dorso-ventrally elongate tubercles in the angles of the calyx, but sometimes quite concealed ; radials concealed ; I Br, usually entirely concealed in the median line, slightly visible over the ends of the basal rays, sometimes visible as a narrow line along the proximal border of the I Br,; I Br, rhombic, twice as brdad as long, all the sides somewhat incurved, rising in the proximal half to a large rounded tubercle; IT Br 4 (3-+4), rarely 2, in close apposition and strongly flattened laterally. Arms twelve to fourteen, 100 mm. long; first eight or nine brachials wedge-shaped or almost oblong, about twice as broad as long, more or less tubercular, then becoming triangular, about as long as broad, in the terminal portion of the arm wedge-shaped and-longer than broad. Syzygies occur, in arms springing direct from a I Br axillary, between the third and fourth brachials, again between the fourteenth and fifteenth or fifteenth and sixteenth, and distally at intervals of six to sixteen oblique muscular articulations; in arms springing from a II Br axillary the first syzygy is between the first and second or second and third brachials. The pinnules resemble those of P. angusticalyx, but the expansion of the genital pinnules is somewhat more marked. Color.—Yellow. } Type—Cat. No. 25464, U.S.N.M., from Albatross station 5236; off the eastern coast of Mindanao, north of Lianza Bay; 494 fathoms. Genus GLYPTOMETRA A. H. Clark. GLYPTOMETRA TUBEROSA (P. H. Carpenter). I tentatively refer to this species a very young specimen from station 5236; off the eastern coast of Mindanao, north of Lianza Bay; 494 fathoms. The lateral eversion of the I Br and lower brachials is very pronounced, and the median carination characteristic of tuberosa is indicated on the I Br and lower brachials; tubercles, however, have not as yet appeared. This specimen is remarkable for the extraordinary size of the external ends of the basal rays. 408 PROCEEDINGS OF THE NATIONAL MUSEUM. VOL. XXXVI. Family ANTEDONIDZ. Genus IRIDOMETRA A. H. Clark. IRIDOMETRA EXQUISITA, new species. Centro-dorsal hemispherical, nearly covered with cirrus sockets, a very small convex polar area bare. Cirrli XL-L, 14-15, exceedingly slender and thread-like, 10 mm. long; first joint short, second half again to twice as long as broad, the following very greatly elongated, becoming shorter again on the ante- penultimate, which is three times as long as broad, and the penulti- mate, which is twice as long as broad; the latter decreases slightly in diameter distally, and may bear a minute terminal opposing spine, though this is usually absent; terminal claw about three-qtarters the length of the penultimate joint, slender, and slightly curved; the articulations of the joints are greatly expanded, except in the terminal five or sIx. Radials even with the edge of the centro-dorsal; I Br, short, deeply - incised in the median line, in contact basally; I Br, (axillary) rhom- bic, twice as broad as long, the anterior angle produced. Ten arms, 40 mm. long; first brachial much longer outwardly than inwardly, very deeply incised by the second brachial, barely in con- tact basally over the anterior angle of the I Br,; third and fourth brachials (syzygial pair) slightly longer interiorly than exteriorly, about as broad as long exteriorly; next four brachials oblong, about twice as broad as long, then becoming triangular, about as long as broad, after the end of the proximal third becoming obliquely wedge- shaped, rather longer than broad, and very gradually increasing in length distally. Syzygies occur between the third and fourth brach- ials, again between the ninth and tenth and fourteenth and fifteenth, and distally at intervals of three oblique muscular articulations. P, 6 mm. long, moderately stout basally, becoming slender distally, somewhat stiffened, with thirteen joints; first joint twice as broad as long, second half again as long as broad, third and following three or three and one-half times as long as broad, becoming slightly shorter terminally; P, similar, but more slender, 3 mm. long, with nine joints, the first twice as broad as long, the second shghtly longer than broad, the third twice as long as broad, the remainder greatly elon- gated; P, 2 mm. long, with eight joints, not tapering so rapidly as P, (therefore appearing somewhat stouter), and bearing a small genital gland in the distal portion; the joints have slightly overlapping distal ends; following pinnules similar; distal pinnules 6 mm. long, similar to those in other species of the genus. Color (in spirits.)—White, with blotches of brown on the arms and pinnules; perisome brown. Type—Cat. No. 25471, U.S.N.M., from Albatross station 5178; north of Tablas Island; 78 fathoms. No. 1673. RECENT CRINOIDS FROM THE PHILIPPINES—CLARK. 409 Family PENTACRINITIDZ. Genus ENDOXOCRINUS A. H. Clark. ENDOXOCRINUS ALTERNICIRRUS (P. H. Carpenter). Two fine specimens of this interesting species were dredged at station 5236; off the eastern coast of Mindanao, north of Lianza Bay; 494 fathoms; one has thirty, the other thirty-one arms, 115 mm. long from the basals. They agree perfectly with Carpenter’s description and figure, and with the Challenger specimen in the U.S. National Museum. The color is a shghtly brownish white. Carpenter included as uncertain Challenger station 210, off Pan- glao and Siquijor (875 fathoms), in his lst of localities; the rediscovery of the species in the Philippine Islands suggests that he was right in referring his unlabeled specimens to this station. GenussLey PAL OCG RINUS Ac El. Clark. HYPALOCRINUS NARESIANUS (P. H. Carpenter). Three specimens, each with arms about 150 mm. long, were secured at station 5236; off the eastern coast of Mindanao, north of Lianza RADIALS, BASALS, AND INFRABASALS OF HYPALOCRINUS NARESIANUS. Bay; 494 fathoms. They agree perfectly with a specimen at hand from the Challenger collection which possibly was taken off Panglao and Siquijor in 375 fathoms. The peculiar dorso-ventral flattening of the short proximal cirrus joints, and the double dorsal tubercles on the more distal joints, characteristic of the genus, are well marked. The terminal 30 mm. to 35 mm. of the arms have only rudimentary pinnules, so that the arms present the same curious “ rat-tailed ” appearance considered by Carpenter as especially characteristic of Metacrinus. This condition is exhibited by the two specimens figured by Prof. Doderlein which were obtained by the Stboga off Celebes; 410 PROCEEDINGS OF THE NATIONAL MUSEUM. VOL, XXXVI. but in the two Challenger specimens I have examined the arm tips are broken off, as delineated in all Carpenter’s figures. One of the specimens was dissected to determine the presence or absence of infrabasals. They were found to be present, resembling closely those of Jsocrinus decorus. Genus METACRINUS P. H. Carpenter. METACRINUS WYVILLII P. H. Carpenter. One broken specimen from station 5236; off the eastern coast of Mindanao, north of Lianza Bay; 494 fathoms. It agrees well with a specimen taken near the Kermadec Islands in 630 fathoms, in the U. S. National Museum. DESCRIPTIONS OF SOME BEES IN THE U.S. NATIONAL MUSEUM. By T. D. A. Cockere tt, Of the University of Colorado, Boulder. The late Doctor Ashmead described many genera of bees, some of which were based on species hitherto unknown. The descriptions were mostly in the form of tables, and in several cases the new species were merely mentioned by name, the detailed descriptions being reserved for a later occasion. Owing to the pressure of other work and Doctor Ashmead’s illness, the opportunity for preparing the projected descriptions never came, and in consequence the species concerned remained very imperfectly known. Through the kindness of the National Museum authorities, I have been allowed to borrow the principal species referred to, and accordingly offer detailed de- scriptions of them. In the cases of the species of I/icrandrena, Croci- _ saspidia, Perditomorpha, and Cenonomada, although Ashmead gave no separate specific descriptions, he published enough information in the course of the generic diagnoses to satisfy the technical require- ments, and the specific names must be credited to him and dated from 1899. I have added some notes on the Philippine Island bees described by Doctor Ashmead, which I examined a few years ago when in Washington. Genus CASNONOMADA Ashmead. CHNONOMADA BRUNERI Ashmead. Cenonomada brunert ASHMEAD, Trans. Amer. Ent. Soc., XXVI, 1899, p. 68 (no locality given). . Tetrapedia gaullei VacHAL, Revue d’Entomologie, January, 1904, p. 22. (Tucuman, Argentine. ) Male.—Length about 10 mm.; black and lemon yellow; hair of head and thorax above very pale fulvous, of cheeks, pleura, etc., white; wings dusky, the nervures and stigma ferruginous. Head broad; eyes large and prominent, pale green, converging below; facial quad- PROCEEDINGS U. S. NATIONAL MUSEUM, VOL. XXXVI—No. 1674. 411 412 PROCEEDINGS OF THE NATIONAL MUSEUM. — you. xxxvi. rangle higher than its breadth in middle; labrum, basal half of mandibles, clypeus except upper part (where the lower edge of the black has the outline of a very broad W), a triangular supraclypeal mark, lateral face marks narrowing to a point on orbit nearly level with anterior ocellus, and line along posterior orbits, all yellow ; scape much dilated, yellow with a large black triangle on one side above; flagellum ferruginous, strongly blackened above, reaching about to anterior part of scutellum, the apical joint attenuated and directed to one side at end, forming a sort of hook; second antennal joint sunken in apex of scape; third much longer than fourth, though not long; ninth and tenth more or less tuberculate beneath; mandibles with a strong inner tooth; tongue long and slender, its length about 4, 165; labial palpi normal for the genus, four-jointed, second joint about 765 » long, first about 1,920 »; blade of maxilla elongated, broad basally, but the apical two-thirds slender, with small erect bristles along the inner edge; maxillary palpi six-jointed, the first three joints at least twice as thick as the other three, the first less than half length of second; lengths of joints in p: (1) 102, (2) 255, (38) 955, (4) 204, (5) 170, (6) 136; maxillary comb very strongly de- veloped, paraglosse about 1,020 » long, thus much shorter than first joint of labial palpi; mesothorax and scutellum dull, yellowish-brown, the hind part of scutellum dull orange; pleura and metathorax black, the pleura shining with sparse punctures; postscutellum yellow, as also upper edge of prothorax, interrupted in the middle, two short stripes on anterior middle of mesothorax, a dull spot above each tegula, and small spots on axille; tubercles orange, strongly produced and pointed; tegule fulvous, rather large; stigma small; marginal cell broadly rounded at end; basal nervure meeting transversomedial ; third submarginal cell longest, receiving second recurrent nervure at beginning of its last third; second submarginal broad, almost if not quite as long below as first, and receiving first recurrent nervure beyond the middle; third transversocubital nervure strongly and abruptly bent; tibiee all yellow (hind tibie black at extreme base) ; anterior and middle femora yellow, partly black above; hind femora greatly swollen and thickened, bulging beneath near base, black, with a broad yellow stripe on outer side; hind tibiz thick but not conspic- uously abnormal, but with an extremely large and broad dark ferru- _2 «The mouth-parts are described from a slide-mount made by Mr. Craw- ford. Compared with Exomalopsis (H. solani Cockerell), the paraglossz of OC. bruneri are much longer, the tongue is longer, and the maxillary palpi are very much shorter in comparison with the blade or galea, being hardly half its length, whereas in the Hxomalopsis the palpus is little shorter than the galea. The maxillary blade of C. bruneri shows some approach to the condition found in Hntechnia. No. 1674. DESCRIPTIONS OF SOME BEES—COCKERELL. 418 ginous spur; anterior basitarsus yellow and the small joints ferru- ginous, both with dense white hair behind; middle basitarsus yellow with the apex black, the small joints dark, the last ferruginous; hind basitarsus’ black, broadened and flattened, with short dark fuscous hair on inner side; small joints of hind tarsi dark, the first three with successively decreasing pencils of fuscous hair; abdomen yellow, with the hind margins of the segments very broadly black, so that in the middle line there is more black than yellow; extreme base of fifth segment black (probably also the others, were they uncovered) ; apical plate broadly truncate, yellow with the apex broadly and the sides very narrowly black; venter yellow at sides and black in the middle. Female——More robust, but very similar in general appearance. Scutellum and lateral margins of mesothorax dull orange; post- scutellum yellow, but the yellow marks on prothorax and mesothorax wanting; face much broader; scape slender and with more black; flagellum normal, but the third‘antennal joint is at least as long as the next three together; clypeus with more black; lateral face marks reduced to irregular triangles, not going above level of antennx; legs without yellow, except on the anterior and middle knees; middle tibie and tarsi brownish with curious short glittering hair; hair on inner side of middle and hind basitarsi dark fuscous; hind tibize and basitarsi broad, with a large glittering scopa; hind tibial spurs slender and normal; abdomen marked as in the male, but first seg- ment black with a transverse yellow band not quite reaching the margins; fifth segment with a heavy fringe of fuscous hair; venter. black, with long white hairs fringing the segments. Habitat—Carcarana, Argentine Republic (Z. Bruner). One of each sex. In Friese’s table of Zetrapedia® this runs to 31, and runs out because of the coloration of the venter of abdomen. It thus falls into the 7. picta group of Friese. Since writing the above, I have corresponded with Doctor Friese, who would place the insect (along with Holmberg’s Chacoana) in Epicharis. To this I can not assent, as H'picharis is derived from Tetrapedia, mainly by the reduction in the joints of the maxillary palpi; the subgenus Z'picharoides, which most resembles Cawnonomada, has these palpi three-jointed. Cwno- nomada has six-jointed maxillary palpi, and thus goes with Jetra- pedia. On-the other hand, I must agree with Doctor Friese that Caenonomada.is the same as Chacoana. Holmberg’s description of Chacoana melanoxantha appeared in 1903, and so Cwnonomada has priority. Mr. Schrottky wrote me in 1906 that he had seen a speci- men of Chacoana melanoxantha from Asuncion and ascertained that it was not an L'picharis. @Ann,.k. k. Naturhist. Hofmuseums, Wien, 1899, p. 278. & 414 PROCEEDINGS OF THE NATIONAL MUSEUM. VOL. XXXVI. Genus EMPHOROPSIS Ashmead. EMPHOROPSIS MURIHIRTA MURINA (Ashmead MS.), new subspecies. Meliturgopsis ASHMEAD, Trans. Amer. Ent. Soc., XX VI, 1899, p. 62.—CocK- ERELL, Ann. Mag. Nat. Hist., Jan., 1901, p. 49. (No species cited in either place.) Male.—In all respects like 2. murihirta Cockerell, except that the face markings are ivory white instead of yellowish; the hair of the thorax is mouse-grey mixed with black, without the yellow tint; and the abdomen beyond the first segment is rather densely beset with long pale hair, not mixed with black. Habitat—San Francisco County, California, October (collector unknown). Typical #. murihirta is from Los Angeles. It has in the male much black hair on the second and following abdominal seg- ments, and a silvery-white fringe just before the apex. Type.—Cat. No. 12237, U.S.N.M. A male specimen without locality, labeled as representing another species of Meliturgopsis, is Anthophora pacifica Cresson. 'This insect is much like the male of A. porterw Cockerell, but among other char- acters tne labrum is longer and conspicuously turned up at the end. It occurs in California. EMPHOROPSIS VIERECKI, new species. Emphoropsis, new species, COCKERELL, Canadian Entomologist, July, 1905, p. 265. Allied to 2’. pascoensis Cockerell, but hair of face and vertex with- out black intermixed. Colorado and New Mexico. I supposed in 1905 that Mr. Viereck was about to describe it, but as he did not do so, I provide a name. The type is in the collection of the American Entomological Society. ° ALLODAPE PHILIPPINENSIS (Ashmead). Prosopis philippinensis ASHMEAD, Journ. N. Y. Ent. Soc., XII, 1904, p. 5. Allodape philippinensis ASHMEAD, Proc. U. S. Nat. Mus., XXVIII, 1904, p. 149. Although Doctor Ashmead corrected the generic reference in his list, he did not indicate that the species was previously described under Prosopis. The original description included a note stating that the reference to Prosopis was provisional, and not really correct. I have examined the type, an interesting little species, best distin- guished by the fact that the hind margins of the abdominal segments are very narrowly testaceous. The description almost exactly agrees with that given by Bingham for Allodape marginata Smith, an insect only known by the unique type in the British Museum, reputed with doubt to be from the East Indies. I suspect that A. marginata really came from the Philippines, and is the same as A. philippinensis, NO. 1674. DESCRIPTIONS OF SOME BEES—COCKERETCL. 415 COELIOXYS MANIL (Ashmead). _—> Coeliorys manile ASHMEAD, Canad. Entomologist, XA XVI, p. 281. The last dorsal segment (female) is broadly rounded, much like that of C. lanceolata Nylander. MEGACHILE ROBBII (Ashmead). Megachile robbii ASHMEAD, Proc. U. S. Nat. Mus., XXVIII, p. 128. (Female. ) The ventral scopa is very pale fulvous, black at tip; the tegule are red with a black basal spot. The unique type is from Manila. Genus MESOTRICHIA Westwood. This genus, usually considered a synonym of Xylocopa, appears to be valid, as Ashmead _ states.¢ Cyaneoderes from it. I can not satisfactorily separate MESOTRICHIA C/ERULEA (Fabricius). Bombus veruleus FAapricius, Syst. Piez., 1804, p. 345. Xylocopa semiarmenia LATREILLE; WIEDEMAN, Mag. f. Ent., IV (** New Caledonia.” ) (atsPaby. p. 421.—LEPELETIER, Hist. Nat. Ins. Hym., II, 1841, p. 200, as synonym. (Java.) Xylocopa cerulea LEPELETIER, Hist. Nat. Ins. Hym., II, 1841, p. 200.—IFr1Esg, Abt. Nat. Ver. Bremen, 1904, p. 184. (Buitenzorg, Java.) Koptorthosoma ceruleum CAMERON, Proc. Zool. Soc., London, May, 1901, p. 34. (Malay Peninsula.) Cyaneoderes fdirchildi ASHMEAD, Trans. Amer. Ent. Soc., XXVI, 1899, p. 70. (Java; the specimens collected by D. G. Fairchild at Buiten- zorg, 1896.) MESOTRICHIA ABBOTTI, new species. Differing from J/. cwrulea as follows: M. cerulea (female). Larger; anterior wing 19-20 mm. long. Only two submarginal cells, the first transverso-cubital nervure absent, or represented by a faint streak. First abdominal segment quite densely clothed with blue hair; sides of second fringed with blue. Wings fuscous, with pinkish-purple iridescence. Supraclypeal shining. ridge prominent and M. abbotti (female). Smaller; anterior wing not over 16 mm. long. Three complete submarginal cells, the first transverso-cubital nervure strong. First abdominal sparsely and incon- spicuously clothed with blue hair; sides of second without blue. Wings darker, the purple stronger. Supraclypeal ridge less prominent. Habitat—Trong, Lower Siam (Dr. W. L. Abbott). Three females. Also from Trong, collected by Doctor Abbott, is a female of genuine M. cerulea, with the face narrower than the average of the Javan specimens, but evidently conspecific with them. TI accept as the genu- @Trans. Am. Ent. Soc., X XVI, p. 71. 416 PROCEEDINGS OF THE NATIONAL MUSEUM. you. xxxv1. ine If. cwrulea the insect described by Lepeletier, who particularly calls attention to the peculiarity in the venation. The Xylocopa cerulea of Bingham," is I/. abbotti, as his figure very clearly shows. Fabricius described his Bombus ceruleus from New Caledonia, and Bingham says that it ranges to New Caledonia. Vachal has recently reported on a collection of bees from New Caledonia, and includes neither J/. cwrulea nor any relative of it. I think there can be no doubt that “ New Caledonia” was an error, the real range of the insect being from Java to Siam. Another blue-haired species, Xylocopa grubaueri Friese, has been described Upper Perak, Malacca. It is very distinct from those now under discussion. The male of I. cwrulea was briefly indicated by Ashmead from the Javan specimens under his generic description of Cyaneoderes. It is large, black, with the hair of the head and thorax (so far as can be seen from the specimens, which have been in spirit) greenish or olivaceous brown, not at all blue. The eyes are very large, and approach above, leaving only a narrow space between them and the large ocelli. The face is without light markings; the copper-red hairs on the labrum are very brilliant. The wings are a little lighter than in the female. The abdominal segments are red at the extreme base, as becomes conspicuous when they are unusually extended. The hind tibiz have at the apex within a large obtuse shining tubercle, the end of which is directed posteriorly. The flagellum beyond the base is ferruginous beneath. Type.—Cat. No. 12238, U.S.N.M. Genus PERDITOMORPHA Ashmead. Perditomorpha ASHMEAD, Trans. Amer. Ent. Soc., XXVI, 1899, p. 86. This is a genus of Panurgines close to Camptopwum, from which it is distinguished by the absence of light markings on the face, the simple spur of middle tibia, the very narrowly subtruncate marginal cell, and the transversomedial nervure meeting the basal. These remarks all apply to the female, the male of Perditomorpha being unknown. In the type of Camptopwum (C. frontale Fabricius from Europe) the basal nervure falls far short of the transversomedial, and the second submarginal cell is longer than the first. In the South American C. ochraceum Friese, OC. submetallicum Spinola, and C. flaviventre Friese, which are before me, the basal nervure also falls far short of the transversomedial, and the apex of the marginal cell is much more remote from the costa than it is in Perditomorpha. The spur of the middle tibia is finely denticulate in all these species, as also in Parafriesea® prinit (P. brasiliensis Schrottky, Camptopeum print Holmberg). @Fauna of British India, Hymenoptera, I, p. 544, pl. 1v, fig. 8. » Friese refers this genus to Perdita. _It is very distinct from Perdita, but scarcely separable from Calliopsis. No. 1674. DESCRIPTIO! COCKERELL. ALT In Acamptopawum * the body 1s somewhat hairy, the basal nervure meets the transversomedial, and the second submarginal cell is searcely longer than the first. In all this there is close approximation to Perditomorpha, but in other respects the bees are not very similar. In Spinoliella (S. zebrata Cresson, S. obscurella Cresson) the vena- tion is essentially as in Camptopewum, and the middle tibial spur (female) is very finely denticulate. In Psenythia (P. philanthoides Gerstaecker, P. annulata Ger- staecker) the middle tibial spur is strongly obliquely dentate; the basal nervure almost meets the transversomedial in P. philanthoides, but falls some distance short of it in P. annulata. In Hypomacrotera the end of the marginal cell is as in Perdito- morpha, but the basal nervure falls short of the transversomedial. In Greeleyella the basal nervure meets the transversomedial, but the end of the marginal cell is not at all as in Perditomorpha, and the first recurrent nervure meets the first tranversocubital. In Hesperapis the shape of the third discoidal cell and the end of the marginal are very different from those in Perditomorpha. Parandrena is easily known from Perditomorpha by the broad fovex, which are as in Andrena. All things considered, Perditomorpha is nearest to Acamptopeum, but apparently sufficiently distinct. Should they be merged, Ash- mead’s genus has priority. In regard to the mouth parts Perditomorpha runs in the table in Annals and Magazine of Natural History, July, 1902, p. 42, to Hes- perapis, to See it is not closely allied. PERDITOMORPHA BRUNERII Ashmead. Female—Length 9-10 mm.; Andrena-like in appearance; black, with a shining ferruginous-red abdomen; pubescence rather short, white, fuscous at apex of abdomen, a few infuscated hairs on scutel- lum, more or less fuscous on middle tibiz in front, and coarse and strongly fuscous on upper outer side of hind tibie. The white hair of the face, cheeks, pleura, and sides of metathorax is abundant and conspicuous. Head broad, facial quadrangle about square, the inner orbits practically parallel; clypeus shining, densely punctured, with a median raised line; no Andreniform facial fovez ; front closely punc- tured; sides of vertex shining, with a considerable impunctate area; ocelli pothex large, in an cemely low triangle; antennz very short, scape black, flagellum black basally and suffused with black above, but otherwise eS chestnut-red ; second antennal joint rather large; third much longer than fourth, the latter being broader than long; labrum @Cockerell, Trans. Am. Ent. Soc, XX XI, 1905, p. 320. Proc.N.M.vol.xxxvi—08——27 418 PROCEUDINGS OF THE NATIONAL MUSEUM. VOL. XXXVI. ordinary, not notched; mandibles with strong inner tooth; labial palpi short, four-jointed, the first joint conspicuously shorter than the other three combined (its length about 145 »), and the whole palpus hardly 340 » long; mentum long, about 1,530 p», its breadth near apex about 290 »; maxillary palpi short and rather thick, six- jointed, much shorter than blade of maxilla, though considerably more than half its length, the joints measuring in p: (1) 135, (2) 69, (3) 50, (4) 42, (5) 35, (6) 60; the ends of the fourth and fifth joints are very oblique; maxillary blade narrow, obtusely pointed, about 600 » long, with subapical bristles; maxillary comb well developed, but the teeth rather short and blunt; maxillary stipes about 1,275 p long, with plumose hairs at the base.¢ By some accident the tongue was lost in the single preparation I was able to make. Thorax robust, shining; the scutellum moderately convex, very sparsely punctured, but the hind margin delicately fluted; meso- thorax rather sparsely but strongly punctured, with distinct median and parapsidal grooves; postscutellum very hairy, as also the meta- thorax, except the large triangular nude basal area, which is smooth and shining, without sculpture; legs ordinary, the hind legs, and also the ventral hairs of the abdomen, carrying much yellow pollen; claws cleft; pulvillus large; spurs simple; tegule dark brown; wings cusky hyaline, distinctly reddish, nervures and stigma ferruginous; stigma large; marginal cell narrowly subtruncate, appendiculate; two submarginal cells, the second at least as long as the first below, narrowed a little more than half to marginal above, receiving both recurrent nervures, the first fully twice as far from its base as the second from its end; transversomedial nervure very oblique; abdomen dull, the broad hind margins of the segments more shiny, and pale golden-hyaline; segments two to four very sparsely fringed with white hairs, not enough to form bands; venter very hairy. MHabitat——Carcarana, Argentine Republic (Z. Bruner). Two fe- males, one being Ashmead’s type. | NOMIA (CROCISASPIDIA) CHANDLERI (Ashmead). Crocisaspidia chandleri ASHMEAD, Trans. Amer. Ent. Soc., XX VI, 1899, p. 6S. Female.—Length about 14 mm., robust, dull black, the first four abdominal segments with broad marginal bands, of which the middle third is lacking, of a most brilliant turquoise blue; face, prothorax, pleura, and sides of metathorax with much white hair; sides of mesothorax posteriorly, and basin of postscutellum covered with “In Hesperapis (H. rhodocerata Cockerell) the maxillary palpi are really very different, in that the first joint is much shorter than the second, and the joints are more narrowed basally. The stipes is also very much shorter in pro- portion, The maxillary palpi of Perditomorpha are nearly the same as those of Hypomacrotcra (Cockerell and Porter, Ann. Mag. Nat.. Hist., Dec., 1899, p. 419), but the labial palpi are very different. No. 1674. DESCRIPTIONS OF SOME BEES—OCOCKERELL, 419 dense white tomentum; scutellum with lateral lobes, and postscu- tellum produced, exactly as in NV. scutellaris Saussure, from Madagas- car; antenne black, the flagellum stout, greyish-pruinose ; mesothorax with dense but rather shallow punctures; scutellum very densely punc- tured; tegule large; anterior wings fuscous-black, with violaceous tints; posterior wings hyaline; legs black, the coarse scopa of hind legs black; anterior and middle tibiz each with a large patch of white hair on outer side, occupying all but apical part of anterior, but little more than basal half of middle ones; some white hair also behind the hind knees. Habitat.—Jombene Range, East Africa (Chanler-Hohnel Expedi- tion). The specific name should apparently have been chan/eri, not chandleri. The above description disagrees in some important par- ticulars with Ashmead’s brief account, and the type specimen bears a specific name dedicating the insect to Doctor Abbott. It is, however, the true (and unique) type, as Doctor Ashmead showed it to me when I was in Washington some years ago, remarking that he had labeled it under the impression that it was caught by Doctor Abbott, and would have to change the name. The name Crocisaspidia may very well be used in a subgeneric sense, for the species of the group of Nomia scutellaris, namely : (1) WV. scutellaris Saussure. Madagascar. With entire white bands; wings not very dark. (2) N. maculata (Friese). Grotfontein, Southwest Africa; Langenburg, Lake Nyassa. Abdominal segments one to four with bluish-white spots on each side; scopa white. (3) N. nigripes (Friese). Ondonga, Southwest Africa; Old Calabar, West Africa; Chinchoxa, Africa; Togo, Africa. Much like maculata, but scopa black, ete. (4) NV. amabilis Cockerell. Benguella. Similar in most respects to chandleri, but postscutellum without white tomentum, and its lobes much more pointed; while the abdominal markings are of quite a different tint, a clear blue with purplish shading, whereas in chand- leri, they are of a deeper and greenish-blue by comparison. (5) WV. chandleri | chanlert| (Ashmead). NOMIA (HOPLONOMIA) QUADRIFASCIATA (/ amead). Hoplonomia quadrifasciata ASHMEAD, Journ. N. Y. unt. Soc., XII, p. 4. I have examined the type; the abdomine’ oands are green, tinged with orange-vermilion. HALICTUS PHILIPPT . WSIS Ashmead. Halictus philippinensis ASHMEAD. oc. U. S. Nat. Mus., XXVIII, p. 128. A species with the general - spect of H. pectoralis,; the hind spur of hind tibie long pectinate; third submarginal cell of the short type; hair of abdomen at lateral bases of segments one and two, and across on three and four. 4920 PROCEEDINGS OF THE NATIONAL MUSEUM. VOL. XXXVI. HALICTUS MANIL Ashmead. Halictus manile ASHMEAD, Canadian Entomologist, XXXVI, p. 281. A very ordinary looking species, with basal hair-bands on abdomen ; hind spur of hind tibia pectinate with four teeth, two long; third submarginal cell short. Genus MICRANDRENA Ashmead. The group which Robertson? calls Opandrena is readily divisible into two very distinct series, which seem to deserve subgeneric rank under Andrena. These are as follows: Abdomen evidently punctured; basal nervure practically meeting transverso- medial; apex of marginal cell rounded, not on costa; end of first transverso- cubital nervure not close to stigma. Opandrena Robertson (type, eressonii Robertson) Abdomen impunctate or practically so; basal nervure falling some distance short of transversomedial; apex of marginal cell more pointed, and on costa; end of first transversocubital nervure very near to the large stigma. Micrandrena Ashmead (type, pacifica Ashmead) Micrandrena also includes Andrena ziziw Robertson, A. personata Robertson, and A. fragariana Graenicher. Unfortunately there is an- other type of Andrena, that of A. flavoclypeata Smith, which though placed by Robertson in Opandrena, does not fit into‘either of the groups defined above. It is near to I/icrandrena, but the basal nery- ure almost meets the transversomedial, falling only a little short of it, and the end of the first transversocubital is not close to the stigma. A. flavoclypeata is larger than Micrandrena, but A. ziziwformis Cockerell, from Virginia, falls with it according to the characters cited, and yet has the stature and appearance of a M/icrandrena. Those who accept the whole series as one, following Robertson, must use Ashmead’s name J/tcrandrena, which was published in 1899, while Opandrena was not published until 1902. In Ashmead’s description of Micrandrena it appears that the facial fovee are wanting; this is not really the case, they are quite distinct and practically as in A. zizie, of a seal-brown color, appearing white in certain lights. ANDRENA PACIFICA (Ashmead). Micrandrena pacifica ASHMEAD, Trans. Amer. Ent. Soc., XX VI, 1899, p. 89 (no locality cited ).—CocKERELL, Psyche, X, 1903, p. 75 (California). Female.—Stature and appearance as in A. z/ziw Robertson; differ- ing from ziziw as follows: Anterior middle of clypeus very shiny, with sparse punctures; flagellum dark, not ferruginous beneath; area of metathorax rougher; wings yellowish, nervures clear ferruginous, second submarginal cell broader; apical depression of second ab- dominal segment stronger and a little larger. Habitat—Alameda County, California, June (collector not stated on label.) @Trans. Amer. Ent. Soc, XXVIII, p. 193. DESCRIPTION OF A NEW ISOPOD OF THE GENUS JHROPSIS FROM PATAGONIA. By Harrier Ricnarpson, Collaborator, Division of Marine Invertebrates, U. S. National Museum. A species of Jwropsis, heretofore undescribed, was collected by the U. S. Bureau of Fisheries steamer Albatross during its cruise off the east coast of Patagonia in 1888. The list of species in this genus now includes Jwropsis brevicornis Kehler, J. mari- onis Beddard,? J. curvicornis® (Nicolet), //. lobata Richardson, J. dollfusi Norman,’ //. rathbune? Richardson, and the present species. JZZROPSIS PATAGONIENSIS, new species. Body oblong-ovate, about two and two-thirds times as long as wide. The lateral parts of the thoracic segments are yellow. Most of the head and most of the abdomen as well as the middle of the dorsal region of the thorax is colored reddish brown. ; The head is wider than long and has the post- lateral angles rounded, the antero-lateral angles acute. The anterior margin is produced on either side of the median line in a small triangular proc- ess. In the concavity formed between the two is placed a small lobe, the anterior margin of j.a:nopsts Li aera Rae which is produced in the middle ina small point. ar The eyes are placed about halfway between the anterior and the Ann. Sei. Nat., (6), XDX, 1885, p. 7. > Challenger Report, XVII, 1886, p. 20. © Stebbing has recently shown that Je@ropsis neo-zealandica Chilton is a syno- nym of Jeropsis curvicornis (Nicolet). Ceylon Pearl Oyster Fisheries Re- port, Pt. 4, 1905, p. 51. ; 4 Historia de Chile, III, 1849, p. 263, pl. 3, fig. 10. € Proc. U.S. Nat. Mus., XXI, 1899, p. 857. f Ann. Mag. Nat. Hist., (7), LV, 1899, p. 291,, pl. 5, figs. 2-8. 9 Trans. Conn. Acad. Sci., XI, 1902, p. 298, pl. 40, figs. 53-55. PROCEEDINGS U. S. NATIONAL Museum, VOL. XXXVI—No. 1675. 499 PROCEEDINGS OF THE NATIONAL MUSEUM. you. xxxvi. posterior margins of the head and a distance from the lateral margin equal to the width of one eye. The first pair of antenne have the first article large; the second is about one-half as large as the first; the third is as long as the second, but narrower; the fourth is half as long as the third; the fifth is a little longer than the fourth; the sixth and seventh are subequal and only about half as long as the fifth. The second antenne have the first article very short; the second is about three times longer than the first; the third is quite long, about three times longer than the second; the fourth and fifth are about equal in length and each is a little shorter than the. third; the flagellum is composed of twelve articles. The antennz are gen- iculate at the articulation of the third and fourth articles. The first and fourth segments of the thorax are subequal in length; the second and third are subequal and are the longest; the fifth is the shortest; the sixth and seventh are subequal and are a little longer than the fourth but not quite as long as the third. The sides of the segments are almost straight and the epimera are not indicated. The abdomen consists of a single segment. The posterior margin is deeply excavate on either side of an acute median point. The post-lateral angles are also acute. Abopt one-third the distance from the post-lateral angles the sides of the abdomen are produced in a small, but pronounced tooth, just above a small excavation in the lateral margin. The uropoda are placed in the posterior excavations of the posterior margin, and consist of a large peduncle, about twice as long us wide, and a minute inner branch, tooth-like, and an outer branch, which in a dorsal view is apparent only as a bunch of hairs. In a ventral view the outer branch is placed in an excavation, is minute, and does not reach beyond the posterior margin of the peduncle. The legs are all similar and terminate in biunguiculate dactyh. Three specimens, all females, come from U. S. Bureau of Fisheries station 2770, east coast of Patagonia. They were collected by the steamer Albatross at a depth of 58 fathoms in gray sand and black specks. Type.—Cat. No. 39240, U.S.N.M. AMMODISCOIDES, A NEW GENUS OF ARENACEOUS FORAMINIFERA. By Josepn A. CusHMAN, Of the Boston Society of Natural History. While examining the original material upon which Dr. Axel Goés based his paper on the Foraminifera of the expedition to the Gala- pagos Islands, a few species included in that report from the Gulf of Mexico were studied. Among these, certain specimens were found under the name of Ammodiscus incertus V’Orbigny, which were at once seen to be peculiar. Instead of the ordinary plano-spiral test of that species these specimens were found to,be really conical, especially the young, the first few revolutions forming a hollow cone. This por- tion stands out as a large prominence in the younger specimens, espe- cially when seen in the view shown in fig. 8. An illustration of this young alone is shown in fig. 4. In later growth the revolutions form a low, flaring cone in the reverse direction. This character may be best seen in the diagrammatic sections, figs. 5 and 6. In occasional specimens, instead of reversing the direction, the cone developed in the young simply becomes more flaring. The specimens were all from the Gulf of Mexico, from Albatross Station 2383, from 1,181 fathoms. In all there were eighteen specimens of varying size. The specimens — from the Pacific in the Goés collection were all of the typical Ammo- discus tenuis or A. incertus forms, all plano-spiral throughout their development. In the early development of the conical species there is a certain resemblance to Gordiaminna and Turritellella, but the later portion is very different. Each of the eighteen specimens had a microspheric proloculum, or initial chamber. The early coils are very uniform in size, and in this respect are again like the other two genera to which reference has already been made. In texture the specimens are much PROCEEDINGS U.S. NATIONAL Museum, VOL. XXXVI—No. 1676. 423 424 PROCEEDINGS OF THE NATIONAL MUSEUM. vou. xxxvi. like the ordinary Ammodiscus, but the color is a peculiar, dark red- dish brown. Proc. Acad. Nat. Sci. Philadelphia, 1856, p. 186; Des Chiites R. Oregon. ¢ Bull. U. S. Fish Commission for 1894, p. 204, pl. xxr, fig. 3. @U. S. Pacific R. R. Survey, 1858, p. 326, pl. ixv1. no. 1677. NEW WHITELDISH FROM OREGON—JORDAN AND SNYDER. 429 species figured by Bean, which must be the original Coregonus wil- liamsoni. Mr. Bean tells me that Girard’s type-specimen is no more in existence. Though we are not sure which species exists in the Des Chiites River, and perhaps both may be found there, we must assume that Girard’s figure is correct. We have no specimens of Coregonus williamsoni from the Colum- bia Basin, for direct comparison with Coregonus oregonius. TExam- ples from Sicamous, on Shuswap Lake, British Columbia (Coll. GC. H. Kigenmann), and numerous fine examples from the Truckee River, California,’ which may be considered as, representatives of C. wil- “iamsoni, seem not to differ from each other, but agree closely with the figure and description published by Girard. They differ notably from Coregonus oregonius in the heavier body, deeper caudal pedun- cle, shorter and less pointed snout, and in having a comparatively small adipose fin, its base being only about two-thirds as long as that of the anal. One poorly preserved specimen of the form called Corcgonus cismontanus, from Beaverhead River, Montana, appears to agree very closely with the above. Both the Beaverhead and Shusuap specimens are smaller than those from the Truckee River, and neither will serve to show slight differences if any such exist. Measurements of the specimens referred to are here given. _ The types of Coregonus cismontanus, from Horsethief Springs, Madison River, Montana, are still smaller and less satisfactory. Coregonus couesi Milner, from Chief Mountain Lake, the head of the Saskatchewan, is doubtless the same as Coregonus cismontanus. Coregonus couesi and Coregonus cismontanus represent at the most a subspecies of Coregonus williamsoni, with possibly smaller adipose fin. Fin rays and measurements of Coregonus williamsoni. = i Scales, Seales, Seales Loeality. piace | quel lateral | transverse |} before aye: YS. line. series. dorsal. Penn yt See ee ee ae ee 13 11 82 949] 32° 12 ht 84 8+ 9 31 13 | 11 85 8+9 29 Beaverhead River, Montana..................--. | 13 | 11 81 8+9 30 Shuswap Lake, British Columbia.....-.......- Boe 13 11 83 8+9 33 | 2 : A Depth of Baer Head in | Snout in) Depth of Locality. Jength. | head. body. Segue REMC KBE USIVCT, =< ec 3.1octseteniotehlc os sete oe el ceminnics aeee betes 4.9 3.2 5 14 4.7 3.2 5.2 15 ail 3.3 4.8 14 enVerMedG River, MiOmMbaiarssen one. foc lee ee ol 4.4 Sue, 4.7 13 Shuswap Lake, British Co:umbia......-..-.--.--2+-<2...0-- | 4.3 3.8 4.8 14.3 According to our present view, the status of the whitefishes of the Pacific slope may be expressed thus: In the Columbia River are three @These were collected near Floriston, California, by Mr. S. J. Mandeville. A30 PROCEEDINGS OF THE NATIONAL MUSEUM. — you. xxxvi. species, one of which, Coregonus oregonius differs widely from the others, and is not known to be represented in any other basin. Core- gonus williamsoni occurs also in streams to the northward of the Columbia, and in the Great Basin of Nevada. It is represented east of the divide in Montana by a slightly differentiated form, Coregonus couesi,s of the Saskatchewan, which probably includes Coregonus cismontanus” of the Upper Missouri. The third species, Coregonus coultert (Eigenmann, from the Upper Columbia River, Kicking Horse River at Field), of which we have specimens from Diamond Lake, Washington, is a species well-distinguished by its slender body and large scales, there being but 60 in the lateral line. Its relation- ships are not close to any other known species. Farther east, this group or genus Prosopium, to which all these species belong, is represented by the Menominee whitefish, Coregonus quadrilateralis. Prosopium is distinguished from Coregonus proper by the short, few gill-rakers, the slender body, and the small, inferior mouth, above which are the large preorbitals, which Milner compared to a mask, zpoowmor. *Qept. U. S. Fish Commission for 1872, 1874, p. 88; Milner, Chief Mountain Lake. > Jordan, Bull. U. S. Fish Commission, IX, 1889, p. 49, pl. rx, fig. 89; Horse- thief Creek, Montana, THE ISOPOD CRUSTACEAN ACANTHONISCUS SPINIGER KINAHAN REDESCRIBED. By Harrinr Ricrarpson, Collaborator, Division of Marine Invertebrates, U. S. National Museum. In 1847 Adam White? gave the name Acanthoniscus spiniger to a new Isopod which Mr. Philip H. Gosse found in Jamaica. ‘This form was not described at that time. Six years afterward the latter naturalist ® referred in the following way to this isopod: “A curious little dark grey Oniscus, every segment of which is armed with two spines, was numerous; it has been described by my friend, Mr. Adam White of the British Museum, under the name of Acanthoniscus spiniger.” The first description which was published of Acantho- niscus spiniger was that of Kinahan in 1859.¢ Kinahan’s descrip- tion is based on the original specimen in the British Museum, which was the only specimen he had seen. In 1885 Budde-Lund? placed in his family Onisci the genus Acanthoniscus of which he said: “ Ad tri- bum Oniscorum sequentia genera in natura mihi ignota pertinere videntur.” He had evidently not seen the specimen in the British Museum for he gives no description, and although he does not place the genus in either section I, Armadilloidea, or section Il, Onis- coidea, yet he refers to it at the end of his section Oniscoidea. WKina- han did not place the genus in any family. About 1877 Mr. H. G. Hubbard, the entomologist, made collections in Jamaica. Some of his collections were given to the Museum of Comparative Zoology and some came to the U. S. National Museum after his death. Among the insects was a specimen of Acanthonis-' cus spiniger, which was turned over to me last winter. The label accompanying it reads: “ Oniscus spiniger. Jamaica.” As I had not seen a specimen of this species before, and as the only description of it is that given by Kinahan, I thought it would be of interest to “List Crust. Brit. Mus., 1847,-p. 99. +A Naturalist’s Sojourn in Jamaica, 1851, p. 65. € Proc. Dublin University, I, 1859, p. 197, pl. 19, fig. 4. ¢@ Crust. Isop. Terrestria, 1885, p. 242. PROCEEDINGS U. S. NATIONAL Museum, VOL. XXXVI—No. 1678. 43 432 PROCEEDINGS OF THE NATIONAL MUSEUM. VOL. XXKVI. redescribe and figure it. No general figure has ever been given, although Kinahan gave detailed drawings of the uropod and the terminal abdominal segment. It seems strange that, although this isopod is said to be numerous in Ja- maica, no specimens exist, so far as I know, in any museum except the British Museum. Family ARMADILLIDIDZ. ACANTHONISCUS Kinahan. Acanthoniscus IINAHAN, Proce. Dublin University, I, 1859, p. 197. ACANTHONISCUS SPINIGER Kinahan. — Acanthoniscus spiniger WHitr (nomen nudum), List Crust. Brit. Museum, 1847, p. 99.—Gossr, A Naturalist’s Sojourn in Jamaica, 1851, p. 65.— KINAHAN, Proc. Dublin University, I, 1859, p. 197, pl. 19, fig. 4.—BuppDE- Lunp, Crust. Isop. Terrestria, 1885, pp. 241-242.—RicHarpson, Bull. U.S. Nat. Mus., No. 54, 1905, pp. 687-638, 592 footnote. Body oblong ovate, capable of rolling up into a ball. Color, in alcohol, dark brown, with irregular spots of ight brown. ; Head much wider than long, with the front emarginate and the lateral angles acutely produced. ‘The eyes are large, bulbous, com- posite, and situated at the post-lateral angles of the head. On the posterior mar- gin of the head are three spines, one in the median line and one on either side, close to the eye, the median spine being smaller than the other two. The first pair of antennae are rudimentary and incon- spicuous. The second pair are broken and the flagellum lost. The first article of the peduncle is short; the second and third are long and subequal; the fourth is one and a half times longer than the third; the fifth is about one and a half times longer than the fourth. Fig. 2.—ACANTHONISCUS SPINI- The first segment of the thorax is longer (pr Suce’? session hee than any of those following. The lateral parts are produced in large, rounded processes, which extend down- ward and upward, surrounding the posterior portion of the head. This segment is armed with two extremely long spines, one on either Fig. 1.—ACANTHONISCUS SPINIGER. no. 1678. ACANTHONISCUS SPINIGER REDESCRIBED-RICHARDSON. 438 side, which are nearly three times the length of the segment. Be- tween these two spines are three short ones on the posterior margin, one being in the median line. Anterior to these spines are four small tubercles, two on either side of the median line in longitudinal series. Lateral to the long spine, halfway between it and the lateral margin, is one small spine on either side. The six follow- ing segments are about equal in length. Each is armed with two extremely long spines, one on either side of the body. Between these long spines are three small spines on the posterior margin, one in the median line. Anterior to these spines are two small tubercles, one on either side of the 16.8.—AcantHon- ? ISCUS SPINIGER. median line. Lateral to these — Urorop. (Unper A\ spines are two small ones, half- = °™”? as way between them and the lateral margin, one small Feaaasen) anterior one and a posterior one, which gradually increases in length, that on the seventh segment being about half as long as the longest spine. The first two segments of the abdomen have the lateral parts covered by the last thoracic segment. The lateral parts of the three following segments are greatly produced, the pos- terior angles being acute. These five segments are about equal - in length; the third and fourth are armed with two small spines on the posterior margin, one on oon elbher side. Of the: median: line. Maxittieup. The sixth, or terminal segment, Fae is widest at the base, contracted about the middle with the posterior half widely rounded and notched in the middle, a small triangular process on either side of the notch. On the anterior portion of the segment are two long spines, equal in length to twice the length of the segment, placed one on either side of the median line. The peduncle of the uropoda re- Fis. 5.— Acantuon- : ? . ISCUS SPINIGER. sembles in form the lateral parts of the third, — gnconn Maxmua. fourth and fifth thoracic segments; the inner pos- Sos terior angle is acutely produced, the outer angle being rounded. The inner branch is inconspicuous in a dorsal view, being con- cealed beneath the abdomen; it is attached at the inner antero- lateral angle of the peduncle and does not quite reach the tip of Proc. N. M. vol. xxxvi—09 28 434 PROCEEDINGS OF THE NATIONAL MUSEUM. VOL. XXXVI. the abdomen. The outer branch is produced in a long spine, ex- tending half its length beyond the inner posterior angle of the peduncle; in a dorsal view it is in- serted on the inner lateral margin at the anterior angle. An anterior view of the head shows the labrum produced on either side so as to extend much beyond the epistome. The inner lamella of the first maxille carries two plumose processes. Although this specimen does not agree in every respect with the de- scription of Kinahan, I can not but think it is the same species. Kina- han does not mention the small spines placed between the two long spines on the posterior mar- gin of the thoracic a b Fic. 6.—ACANTHONISCUS SPINIGER. First Maxiuia, a. Inner Lose, Segments. He also b OUTHR ROBE, xX Tt. = - ] does not mention the’ 4p. 7.Acaesoe presence of spines on the abdomen, but in his — rscus_ sprytenr. ; : : i ANTERIOR VIEW OF figure of the terminal abdominal segment I think 4... suowine he intended to represent them in the two long © uPrstome witu lines in the center of his figure. His representa- 9 ““""™ tion of the uropod does not agree with the specimen I have, but the shape of the terminal abdominal segment is so similar that I am inclined to think that there must be some error in the figure of the uropod.? 4 Since preparing the above description I sent a copy of my figure to Doctor Calman of the British Museum for comparison with the type. In his answer, just received, he says that he is almost certain that my specimen is Acantho- niscus spiniger. He mentions the fact that in the type-specimen there are two teeth instead of three on the posterior margin of the thoracic segments after the first. On the first segment, the middle tooth is extremely small. On the hinder edge of the head are only two teeth, placed a little in front of, not on, the margin. He also noticed a difference in the shape of the uropod, but thinks this may be due to its being in a slightly different position from my sketch. He very kindly made drawings of the type for me. Although I ‘am aware of these discrepancies, I hesitate to consider my specimen other than Acanthoniscus spiniger when the resemblance is so strong and the locality the same. Moreover the type-specimen is probably a dried specimen and some allowance must be made for change in contour owing to its condition. When Kinahan described it, twelve years after it was collected, it was probably in no better condition than it is now, ADDITIONS TO THE LIST OF PHILIPPINE BIRDS, WITH DESCRIPTIONS OF NEW AND RARE SPECIES. By Enesar ALEXANDER Mearns, Associate in Zoology, U. S. National Museum. This is the seventh of a series of papers on Philippine birds, pub- lished by the writer, adding, in all, 56 species to the list of those pre- viously known from the islands.* It is the writer’s intention soon to publish a list of Dr. Paul Bartsch’s Philippine collection of birds, in - which two additional forms will be described. The following are additions to the species recorded from the Phil- ippine Islands: STERNA LONGIPENNIS Nordmann. NORDMANN’S TERN, I collected five specimens (Nos. 14065-69, author’s collection) of this species, in Basilan Strait, off Zamboanga, Mindanao, April 19, 1906. Four were preserved as skins (Cat. Nos. 200770-73, U.S.N.M.) and one in alcohol. LOBIPES LOBATUS (Linneus). NORTHERN PHALAROPE, On several occasions I had seen flocks of phalaropes on the seas surrounding the Philippine Islands; and on April 19, 1906, three specimens were collected in Basilan Strait, which connects the Sulu and Celebes seas. These are Cat. Nos. 200774-6, U.S.N.M. TANYGNATHUS MEGALORHYNCHOS (Boddaert). GREAT-BILLED PARROT. On January 23 and October 8, 1906, this large green parrot was found in considerable numbers on Sarangani and Balut Islands of the Sarangani group, off southern Mindanao. An adult female (Cat. No. 200811, U.S.N.M.) was collected on Balut Island, January 23, 1906. The following are believed to be new to science: 7See Proc. Biol. Soc. Washington, XVIII, January 20, 1905, pp. 1-8: Feb- ruary 21, 1905, p. 73; February 21, 1905, pp. 83-90; June 29, 1905, p. 185; Philippine Journ. Sci., II, October, 1907, p. 853; pp. 355-860. PROCEEDINGS U. S. NATIONAL Museum, VoL, XXXVI—No. 1679. 496 : oo 436 PROCEEDINGS OF THE NATIONAL MUSEUM. VOL. XXXVI. PHAPITRERON SAMARENSIS, new species. SAMAR BROWN PIGEON. Type.——Cat. No. 161096, U.S.N.M. Adult female. Collected in March, 1888, on the island of Samar, Philippine Islands, by Dr. F. 8. Bournhs. Characters.—Resembling Phapitreron brevirostris, but with fore- head, chin, and throat whiter than in P. albifrons McGregor. From P. brevirostris it is readily distinguishable by the white forehead, chin, and upper throat, also by the much greater amount of coppery amethystine reflections on the crown, breast, and upper back, and by the usual absence of green reflections on the side of the hind head below the suborbital white stripe; and the under parts are much paler. From P. albifrons it may be at once distinguished by the different color of the under parts, which lack the olivaceous-gray on the breast, and by its more ochraceous abdomen; also the coppery reflections on the breast and upper back of P. samarensis are entirely absent in P. albifrons. Measurements of type (from well-made skin of female).—Total length, 230 mm.; wing, 124; tail, 95; culmen and cere (chord), 14; tarsus, 19; middle toe with claw, 26. “ Eyes pink. Feet, base of mandible, and around eyes, dark purple. Tip of bill black. Ege taken from ovary.” (KE. S. Bourns.) Measurements of adult male (Cat. No. 161095, U.S.N.M., collected in April, 1888, on the island of Samar, by Dr. F. S. Bourns).— Wing, 129 mm.; tail, 95; culmen and cere (chord), 14.5; tarsus, 20; middle toe with claw, 27. MUSCADIVORES PALMASENSIS, new species. PALMAS ISLAND FRUIT-EATING PIGEON. Type.—Cat. No. 200889, U.S.N.M. Adult male. Collected Jan- uary 21, 1906, on Palmas Island, in the Celebes Sea, Philippine Islands, by Edgar A. Mearns. (Original number, 13889.) Characters.—Closely related to Muscadivores picheringi from Mangsee Island, north of Borneo, on the west side of the Sulu Sea; also to Muscadivores langhornei Mearns, from West Bolod Island, southeast of the Sulu Sea and near the island of Basilan. From the type of pickeringi (Cat. No. 15732 U.S.N.M.) it differs in being paler, with much less vinaceous color on the chin, throat, breast, and under tail-coverts. The wing is 10 mm. shorter. From Jang- hornei, which it resembles more closely in the coloration of the under parts, it can be distinguished at a glance by the dark color of the mantle, rump, and upper tail-coverts. Adult male (type, killed January 21).—Head, neck, upper back, and under parts, llac-gray, purest on the upper side of neck and no. 1679. LIST OF PHILIPPINE BIRDS—MEARNS, 437 upper back, washed with vinaceous on crown, ear-coverts, and breast, fading to whitish around base of bill, and shading to drab-gray on legs and crissum; scapulars, back, rump, and wing-coverts mouse gray, lustrous in a certain light; wing-quills and upper tail-coverts dark mouse gray, with subdued reflections of violet, coppery, and green; rectrices lustrous golden green above, smoke gray below; flanks, axillars, and lining of wings clear gray. An adult male topotype in fresh plumage, shot by Dr. Paul C. Freer, October 7, 1906, only differs from the type in being appreciably darker. ‘The sexes are practically alike in size and color. Colors of soft parts—Two mated pairs, about to breed, had the soft parts colored exactly alike, January 21, 1906: Iris red; eyelids and feet vinaceous; claws dusky purplish gray; bare space surround- ing eye, pale plumbeous; bill pale bluish gray at tip, darker— plumbeous—at base. Testicles functionally enlarged. Measurements of two adult males (type and topotype measured fresh by the author).—Total length, 420, 480 mm.; alar expanse, 735, 750; wing, 240, 240; tail, 156, 160; culmen (chord), 20, 20; tarsus, 32, 34; middle toe with claw, 46, 49. OTUS STEEREI, new species. TUMINDAO SCOPS OWL, Type.—Cat. No. 210752, U.S.N.M. Adult male. Collected by Edgar A. Mearns, October 13, 1906, on Tumindao Island, off Sitanki Island, Philippine Islands. (Original number, 14421.) Characters—Very similar to the Celebesian Ofus menadensis, from which it may ‘be distinguished by being larger, with upper parts darker, with more of the black vermiculations; black centers to the feathers of the under parts much less conspicuous; feathers of tarsus more heavily cross-barred with blackish. Of the Philip- pine species it is most closely related to Otus cuyensis McGregor, but is darker and much smaller, having the same white, black-tipped seapulars, but with the entire plumage darker and more heavily marked, and the wing about 15 mm. shorter. It bears no close resemblance to any other Philippine species. Measurements.—Wing, 157 mm.; tail; 84; culmen from cere (chord), 15; tarsus, 33. Iris yellow; bill and feet greenish (from fresh specimen). The stomach of the type contained insects. Named for J. B. Steere, known for his studies of Philippine birds. PRIONITURUS MALINDANGENSIS, new species. MOUNT MALINDANG RACQUET-TAILED PARROT. Type.—Cat. No. 200887, U.S.N.M. Adult female. Collected at 5,000 feet altitude on Mount Lebo, a spur of Mount Malindang, Misamis Province, northwestern Mindanao, Philippine Islands, May 14, 1906, by Edgar A. Mearns. (Original number, 14151.) 438 PROCEEDINGS OF THE NATIONAL MUSEUM. VOL. XXXVI. Characters.—Closely related to Prioniturus waterstradti Roths- child, from Mount Apo, southeastern Mindanao, from which it may be distinguished by its larger size, much yellower coloring of under side, greener, less brownish back and scapulars, longer tail, but much shorter naked shafts to the central pair of feathers. Adult female (type and only specimen).—Upper parts green, with slight touches of bluish on forehead, and a light greenish brown rump, as in P. waterstradti; wings bright green above, with con- cealed inner webs blackish, and outer webs narrowly edged with yel- low; edge of wing pale yellow; rectrices green above tipped with dull black, beryl green on under side, the shafts and spatules of the central pair dull black, with webs all green to the naked shafts; entire under parts golden green; axillars and lining of wings oil green; under side of primaries dull black, broadly bordered on the inner webs with beryl green. Iris dark brown; bill pale horn color, faintly plum- beous at base and tip of mandibles; feet and claws plumbeous (from fresh specimen). Measurements of type (female).—Wing, 153 mm.; tail, 79 (to end of lengthened central pair, 126); culmen from cere (chord), 20; tarsus, 16; middle toe with claw, 25. Measurements of adult female topotype of P. waterstradti (Cat. No. 192136, U.S.N.M.).—Wing, 145 mm.; tail, 75 (to end of length- ened central pair, 151); culmen from cere (chord), 18; tarsus, 15; middle toe with claw, 25 (skin). YUNGIPICUS SIASIENSIS, new species. SIASI PIGMY WOODPECKER. © Type—Cat. No. 210765, U.S.N.M. Adult male from Siasi Island, Philippines, collected October 12, 1906, by Edgar A. Mearns. (Origi- nal number, 14401). Mr. E. Hargitt, in the original description of “ /Jyngipicis ram- say?,’* gave the type-locality as “ Northeast Borneo;” but the same author ¥ says that the type was an adult male collected by A. Everett in the “Sulu Islands.” An adult male specimen (Cat. No. 211344, U.S.N.M.) collected by Dr. Paul Bartsch, Feburary 23, 1908, on Papahag Island, off 'Tawi- Tawi, one of the southern islands of the Sulu group, agrees with Har- gitt’s Yungipicus ramsayi. Adult male (type).—Similar to the male of Y’. ramsayi, but with a smaller bill and without white markings on the upper surface of the primaries or secondaries except a small concealed white spot on inner webs of several secondaries, and with much less white on the inner @Tbis, 1881, p. 598. » British Museum Catalogue of Birds, XVIII, p. 335. No. 1679. LIST OF PHILIPPINE BIRDS—MEARNS. 439 margins of the inner webs on under side of wing; and the orange- yellow of the under parts is much more restricted, being confined to a narrow band across the chest. Adult female (Cat. No. 210746, U.S.N.M., killed at the same time ‘and place as the type, with which it was apparently mated).—Similar to the male, but lacking the elongated scarlet feathers on the edge of the posterior half of the crown and occiput, the entire upper surface of head and neck being dark brown. The white markings of the upper and under sides of the wings are restricted to the same extent as those of the type, and tend to form a very narrow brownish-white margin to the inner webs of the innermost secondaries, below, instead of forming squarish detached white spots as in ¥. ramsayi; the orange-yellow pectoral band as in the male. Measurements of YVungipicus siasiensis—Adult male (type): Wing, 83 mm.; tail, 47; culmen, 19.5. Adult female (Cat. No. 210764, U.S.N.M.; topotype) : Wing, 85 mm.; tail, 49; culmen, 20.2. Measurements of Yungipicus ramsayi—Adult male (Cat. No. 911344, U.S.N.M.): Wing, 85 mm.; tail, 47; culmen, 18. RHINOMYIAS RUFICAUDA MINDANENSIS, new subspecies. MINDANAO RUFOUS-TAILED FLYCATCHER, Type.—Cat. No. 190247, U.S.N.M. Adult male, collected by the writer at Pantar, Mindanao, Philippine Islands, August 24, 1903. (Original No. 12929.) The series of this genus in the U. S. National Museum shows that there are three geographical forms of the rufous-tailed flycatcher, from the islands of Basilan, Mindanao, and Samar, respectively. Compared with the Samar and Mindanao forms the Basilan form, Rhinomyias ruficauda ruficauda, has the under parts whiter, middle of chest grayer, sides of chest and flanks a grayer brown, with entire side of head slate-gray. R. mindanensis and samarensis both have brown cheeks, and differ from each other in size, the Mindanao form being larger. The upper surfaces are of a lighter, more olivaceous, and less rufescent color. Measurements of Rhinomyias ruficauda (Sharpe). Locality. Date. Collector. Museum No Collector’s No. | Sex and age. Depth at angle of gonys. | Tarsus. 201263 | 13939 | Adult..... Isabella, Basilan..| Feb. 2,06] 72 | 58 |..... "| 10.5 | 17.5 | Mearns. 440 PROCEEDINGS OF THE NATIONAL MUSEUM. VOL. XXXVI. Measurements of Rhinomyias ruficauda mindanensis, new subspecies. 5 7 area, : i) (e) = 5 ‘ A oD iS 8 z Locality. Date. ‘ g 3 5 2 Collector. 2 | 3 v | te Re eee S o 3 = a f= 2 a a 'S) 72) S a Ss) A a mm. 190247 | 12929 | Male ad...| Pantar, Mindanao..| Aug. 24,03 17 Mearns.@ 210837 | 14123 |...do.......| Catagan, Mindanao.| May 13,06 18 Do. 211161 74) Mail eto es sic ace Goes ss eee May 21,06 17.3 | Schroder. 200953 | 14075 | Malead...| Zamboanga, Min- | Apr. 25,06 18 Mearns. danao. 190246 | 12893 | Femalead-.| Pantar, Mindanao..| Aug. 13,03 16 Do. 210836 | 14199 |...do.......| Catagan, Mindanao.| May 23,06 17 Do. PAOSSS | LA2I 3S oeGOssenvcelaecee OC sae Are ee May 25,06 16 Do. Measurements of Rhinomyias samarensis Steere. HGLSOG See cece Male ad....| Samar Island, P. I.| Mar. —,88 | 73 | 62 OUR ines oe | 16 Bourns. 16TH ee Soe ianemaleiade|vaeenC0teess ees ne Apr. 23,88] 73 | 60 LOSE | Sas | 17 Worcester. NET SOD) See een eel ame me bane Oe ee eee Apr. 20,88 | 68 | 56 EO eee: | 16 Do. TGTS04s Vessel 00355) 2 |-2 5. GOt ese Apr. 2,88 69 | 56 OFS) eee | 15.8 Do. 161503 | ert sam: ena | Figs it (A eRe es Apr. 18,88} 71/60 | 11 |...... | 16 Do. usbr ees | eee wees « Type. CRYPTOLOPHA MALINDANGENSIS, new species. MOUNT MALINDANG FLYCATCHER-WARBLER. Type.—Cat. No: 202360, U.S.N.M. Adult male. Summit of Grand Malindang Mountain (altitude slightly above 9,000 feet), Misamis Province, northwestern Mindanao, Philippine Islands, June 6, 1906. Collected by Edgar A. Mearns. (Original number, 14275.) Characters.—Similar to Cryptolopha mindanensis Hartert, from Mount Apo, Mindanao, but smaller, less yellow above and below, and with a distinct yellowish-white post-ocular streak extending to the occiput; bill flesh color instead of yellow on base of mandible; feet grayish flesh color instead of plumbeous. Adult male.—Upper parts olive-green, darkest on the crown; wings and tail dark brown, broadly bordered with olive-green on the outer webs, but with outer rectrix white to the base, edged with pale yellow basally and with olive-brown terminally on outer web; second rectrix dark brown at base of inner web, white on terminal two-thirds, yellow on basal half of outer web, and olive-brown on terminal half; third rectrix edged with white at tip of inner web and yellow on outer web at base; loral and post-ocular streak yellowish white; entire under parts sulphur yellow, obscured by pale olive-green centers to the feathers; under tail-coverts plain sulphur yellow; sides of chest and flanks olive-green, streaked with yellow; axillars and lining of wings pale sulphur yellow; cheeks pale sulphur yellow, mottled with very pale olive-green ; iris hazel ; bill brownish black, flesh color at base of mandible; feet grayish flesh color, claws brown (from fresh specimen). No. 1679. LIST OF PHILIPPINE BIRDS—MEARNS. 441 Comparative measurements of Cryptolopha mindanensis and C. malindangensis.—Adult males (from skins) : Wing, 58, 56 mm.; tail, 48, 46; bill from nostril, 7, 7; tarsus, 21, 21. Material—F ive specimens of Cryptolopha mindanensis and 11 of C. malindangensis. Range.—F rom 5,000 to 9,000 feet on the Malindang Mountains. PSEUDOTHARRHALEUS MALINDANGENSIS, new species. MOUNT MALINDANG WOOD-ACCENTOR. Type.—Cat. No. 210853, U.S.N.M. Adult male. Summit of Mount Malindang, northwestern Mindanao, Philippine Islands, alti- tude slightly above 9,000 feet, June 6, 1906. Collected by Edgar A. Mearns. (Original number, 14277.) Characters.—The largest known species of Pseudotharrhaleus ; gray of cheeks and supraorbital stripe obscured by heavy markings of brown; feathers of chest heavily marked with black centers. Adult male (type and only specimen).—General color above burnt umber, washed with Vandyke brown on rump and upper tail-coverts; tail darker; wing-quills brownish black, with outer webs broadly margined with the same color as the upper parts and extending to the outer webs of the under side of wing; head sepia above, without an appreciable supraorbital stripe; sides of head grayish brown, maculated with bister; chin and upper throat dirty whitish; much obscured by dusky macules occupying the centers of the feathers; middle of chest gray, heavily marked with blackish centers to the feathers; sides, crissum and under tail-coverts, axillars, and lining of wings like the back, this color shading to wood brown on middle of belly ; iris brown; bill plumbeous-black; feet and claws brown (from fresh specimen). The following measurements were taken from the type specimen, freshly killed, by the writer: Total length, 196 mm.; alar expanse, 212; wing, 66; tail, 90; culmen (chord), 16.5; bill from nostril, 10.8; from occiput to tip of bill, 42; tarsus, 28; middle toe with claw, 25. This bird was usually found in hollows under mossy logs. Its note resembles the alarm call of the American Pipilo fuscus mesoleucus. BRACHYPTERYX MALINDANGENSIS, new species. MOUNT MALINDANG SHORTWING. Type.—Cat. No. 202137, U.S.N.M. Adult female. Summit of Grand Malindang Mountains, altitude 9,000 feet, Misamis Province, northwestern Mindanao, Philippine Islands, June 5, 1906. Collected by Edgar A. Mearns. (Original number 14269.) Characters.—Most closely related to Brachypterye brunneiceps Grant and B. mindanensis Mearns. Smaller than brunneiceps, about equaling mindanensis, coloration very dark; russet of front of head 449 PROCEEDINGS OF THE NATIONAL MUSEUM. — you. xxxvt. intensified to almost a burnt umber, and not extending backward beyond the eyes; edge and lining of wings, slate color, instead of rusty. Adult male.—(Cat. No. 202136, U.S.N.M. tarsus, 26, 25, 25. CHIBIA CAGAYANENSIS, new species. CAGAYAN SULU DRONGO SHRIKE. Type.—Cat. No. 191894, U.S.N.M. Adult female from Cagayan Sulu Island, in the western part of the Sulu Sea, Philippine Islands, collected February 26, 1904, by Edgar A. Mearns. (Original num- ber, 13285.) Characters.—Very similar to Chibia palawanensis, differing only in its somewhat larger size, shallower forking of the tail, the nar- rower and very much smaller spangles on the breast, and in the absence of metallic green on the upper tail-coverts. Measurements of skin (type and only specimen).—Length, 260 mm.; wing, 136; tail, 126; emargination of tail, 16; culmen (chord), 28.5; tarsus, 24.5. Measurements of two adult female topotypes of Chibia palawanen- sis (Cat. Nos. 161330 and 161331, U.S.N.M.).—Length, 157, 150 mm.; wing, 133, 131; tail, 126, 130; emargination of tail, 20, 20; culmen, 28,222 tarsus, 93; 95.5. Remark.—I saw a number of these birds on the island of Cagayan Sulu on my brief first visit. REMARKS ON THE INSECTIVORES OF THE GENUS GYMNURA. By Marcus Warp Lyon, Jr., Second Assistant Curator, Division of Mammals, U. S. National Museum. An examination of nearly two dozen specimens of the genus Gymnura in the United States National Museum shows that in addi- tion to the two distinct species usually recognized, a third form, from the northern portion of the Malay Peninsula, must be considered. It is here described for the first time as a subspecies of the older of the two species. As there is some lack of agreement among authors regarding the limits of the genus Gymnura, it will not be out of place to con- sider some of its characteristics before describing the new subspecies. Dobson, Flower and Lyddeker,? and Trouessart’ have included under the term Gymnura certain small short-tailed species, which other writers, Thomas* and Max Weber,’ for instance, regard as constituting a separate genus Hylomys. I quite agree with the opinion of these latter, and in order to show the distinctness of Gymnura from /Tylomys figures of their skins and skulls are shown on Plates 34 to 37. In order to make the relations of the genus Gymnura still clearer, the entire animal and the extracted skull of the unique specimen of the recently discovered and related genus Podogymnura,! an animal unknown to the authors cited, are shown on the plates 36 and 37. Gymnura is here considered as one of three genera forming the subfamily Gymnurine, the other two being 7ylomys and Podogym- nura. In respect to size and external characteristics ylomys and Podogymnura appear to be closely related to one another. (Plate 37.) An examination of the teeth, however (Plate 36), shows that these “Monogr. Insectivora, 1882, p. 5. > Introd. Study of Mammals living and extinct, 1891, p. 620. ¢Cat. Mamm. Suppl., 1904, p. 126. @ Ann. Mag. Nat. Hist., 6th ser., II, November, 1888, p. 407. € Die Siiugetiere, 1904, p. 379. f Mearns, Proc. U. S. Nat. Mus., XXVIII, No. 1402, p. 437, May 13, 1905. PROCEEDINGS U. S. NATIONAL MUSEUM, VOL. XXXVI—No. 1680. , 449 Proe.N.M.vol.xxxvi—09——29 450 PROCEEDINGS OF THE NATIONAL MUSEUM. VOL. XXXVI. two genera are very distinct. The premolars both above and below in Hylomys are 4, in Podogymnura %. ‘The three anterior premolars in Zylomys are small, practically functionless teeth, while in Podo- gymnura the penultimate premolar is well developed and trenchant, standing nearly as high in the tooth row as the last premolar (Plate 36). The premolars in Gymnura are $,as they are in Zylomys. The two anterior of them are small and practically functionless, like the three anterior premolars of /Zylomys; the two posterior are large and trenchant, like the two posterior premolars of Podogymnura. (Compare Plates 35 and 36.) Of the three genera Gymnura appears to be the most generalized and primitive, and from it or its ancestors the other two genera have been derived. In the shortening of the tail and simplification in the structure of the premolars //ylomys has departed further from Gymnura than has Podogymnura. ‘The last upper molar of Podogymnura and Hylomys shows a more simple structure than the corresponding tooth of Gymnura, which has a distinct posterior heel, lacking in the other two genera. The genus Gymnura is said to present the most generalized structure known among placental mammals.* An account of its anatomy is given in Dobson’s Monograph of the Insectivora, 1882. The geographic distribution of the genus coincides with part of the Malayan subregion of the Oriental region, namely, Tenasserim, the Malay Peninsula, Sumatra, and Borneo, and a few immediately adjacent islands. So far as known, it does not occur on others of the larger or of more remote smaller islands of the Malayan Archipelago. Although the existence of two species in the genus had been pointed out by Giebel in 1863,” yet they were usually considered as mere local varieties, or color phases, until Jentink reaffirmed their specific dis- tinction in 1881.° Jentink and other writers have described certain peculiarities of shape in the skull and teeth by which the two species in the genus may be distinguished, but I have been unable to detect any other peculiarities than size and color for distingushing them. The skulls and teeth, however, show many individual variations, but none of them are constant for definite geographic areas. The char- acters assigned by other writers may probably be explained as the result of examining too small a number of specimens. The different members of the genus may be distinguished by the following key: KEY TO THE SPECIES AND SUBSPECIES OF GYMNURA. a. Color uniformly white or whitish, size large, hindfoot 66-74 mm., basal > Zeitschr. Ges. Naturw., XXII, p. 277. ¢ Notes Leyden Mus., III, pp. 166-168, no. 1680. INSECTIVORES OF THE GENUS GYMNURA—LYON. 451 b. Color black with some white on head and shoulders, size medium or small, hindfoot 58-68 mm., basal length of skull 67-78__Gymnura gymnura, p. 451 a.’ Size medium; hindfoot, 61-68 mm.; basal length of skull 71-78. G. g. gymnura, p. 451 b.2 Size small; hindfoot 58-59 mm.; basal length of skull 67-71. G. g. minor, p. 453 GYMNURA GYMNURA (Raffles). Diagnostic characters—Size small or medium; hindfoot 58-68 mm.; basal length of skull 61-78; color generally black, with white markings on head. See fig. 3, Plate 34. Color.—General color black or blackish, except lips, chin, cheeks, an irregular V-shaped patch on top of head, and terminal fourth or more of tail white or whitish. About the shoulders and on upper back are numerous long hairs with long white or whitish tips. Similar hairs, but much shorter, occur on the throat. Pelage—The pelage is composed of two kinds of hairs, short (10-15 mm.) soft underfur of a dull blackish, brownish or dark drab-gray color, except about lips and cheeks, where it is white or whitish ; and long, coarse, bristly hair, 50-60 mm. in length uniformly black in color, except in the region of the light markings about head and shoulders, where the long hairs are white throughout. their extent or else have long white tips. About the head the long hairs are much shorter than over the rest of the body. On the under- parts the long hairs are relatively few, soft, and short. Tail.—The tail is covered with small scales about 10 to the centi- meter, each scale subtended by about 3 hairs. On the dorsal surface of the tail the hairs are inconspicuous and a little longer than a scale in length; on the underside they are more conspicuous and equal 3 or 4 scales in length. Distribution.—Tenasserim, Malay Peninsula, Sumatra. Remarks.—This species is separable into two distinct races, a smaller one from Tenasserim, and Trong, Lower Siam, described below as new, and the typical race found on the lower end of the Malay Peninsula and on Sumatra. GYMNURA GYMNURA GYMNURA (Raffles). 1822. Viverra gymnura Rarries, Trans. Linn. Soc. London, XIII, p. 272. (Type- é locality, probably Bencoolen, Sumatra.) 1827. Gymnura rafiiesii Lesson, Man. Mamm., p. 171. (May 1827. See Pal- mer, North Amer. Fauna, no. 23, 1904, p. 304.) 1827. Gymnura rafiesii, Horsrirtp and Vicors, Zool. Journ., III, p. 248, pl. vt. ' (October, 1827. See Palmer, North Amer. Fauna, no. 23, 1904, p. 304.) Diagnostic characters.—A large race of Gymnura gymnura, hind foot, 61-68; basal length of skull, 71-78. Color.—As described above under G. gymnura. 452 PROCEEDINGS OF THE NATIONAL MUSEUM. VOL. XXXVI. Skull and teeth—These are relatively large and heavy, otherwise they do not differ from those of the race described below. See fig. 2 4, Plate 35. Distbutione Ue U. S. National Museum contains specimens from Rumpin River, Pahang, Malay Peninsula; Singapore; Tapanuli Bay, west coast of Sumatra; the Siak region of eastern Sumatra, and Pulo Tebing Tinggi, a low alluvial island off the east coast of Sumatra. This form is probably generally distributed throughout Sumatra and the lower extremity of the Malay Peninsula. LENGTH OF HIND FOOT, INCLUDING CLAWS, IN MILLIMETERS 67 68 69: 70 “71.72 738 74. 75 76. 7% 78 79° 80 '8i) 82) 83 Geese BASAL LENGTH OF SKULL IN MILLIMETERS @ GYMNURA GYMNURA GYMNURA, % GYMNURA GYMNURA MINOR, © GYMNURA ALBA. DIAGRAM TO SHOW THE RELATIVE SIZES OF THE THREE FORMS IN THE GENUS GYMNURA AS DETERMINED BY LENGTH OF HIND FOOT, INCLUDING CLAWS, AND OF BASAL LENGTH OF SKULL. MEASUREMENTS ARE IN MILLIMETDRS. EACH DOT REPRESENTS AN ACTUAL SPECIMEN. Remarks.—What is apparently an albino specimen of this race is recorded by Schneider® as Gymnura alba. He remarks, however, that it is distinctly smaller than white examples that he has seen from Borneo. Aside from color, difference in size is apparently the only manner by which Bornean and Sumatran examples Tay be dis- tinguished. Schneider himself thought his example only a “sport.” “Zool. Jahrb. Syst., XXIII, 1905, p. 89. no. 1680. INSHCTIVORES OF THE GENUS GYMNURA—LYON. 453 GYMNURA GYMNURA MINOR, new subspecies. Type—Skin and skull of adult male, Cat. No. 86783, U.S.N.M., collected at Trong, Lower Siam (on Khow Nok Ram, 2,000 feet altitude), January 12, 1899, by Dr. W. L. Abbott. Diagnostic characters.—Similar in all respects to the typical race, but averaging distinctly smaller throughout. See fig. 1, Plate 34. Color.—The color of .Gymnura gymnura minor does not differ essentially from that of G. g. gymnura, but there is a tendency to have more white about the head, neck, shoulders, and upper portion of back, as well as a narrower and less conspicuous black super- cilliary stripe. Skull and teeth—These possess the same relative shapes and pro- portions in Gymnura gymnura minor as they do in G. g. gymnura, but average distinctly smaller throughout. See fig. 1, Plate 35. Measurements.—A series of four adults, all from Trong, give the following extremes of measurements: Head and body, 311-3835 mm.; tail, 216-241; hind foot, with claws, 58-59; basal length of skull, 67-71; upper tooth row (all teeth, alveoli), 40-42. See table, page 455, Specimens examined—Four from the type-locality, 2 from an altitude of 2,000 feet, 1 from 1,000 feet. The altitude for the remain- ing specimen is not known. Distribution —Trong, Lower Siam, extending northward into Tenasserim. The small size of individuals from the latter country — has been mentioned by both Dobson® and Blandford.’ Bonhote ° remarks that “the Malayan form appears to be smaller than the ’ Sumatran race.” GYMNURA ALBA Giebel. 1863. Gymnura alba GieseL, Zeitschr. Ges. Naturw., XXII, 1863, p. 277. Type- locality, Borneo. ) 1876. Gymnura rafiesii var. candida GUNTHER, Proc. Zool. Soc. London, 1876, p. 425. (Type-locality, “ Labuan, the mainland opposite Labuan, and Sarawak,” Borneo.) Diagnostic characters——Color entirely white or essentially so, largest member of the genus, hind foot 66-73 mm., basal length of skull, 78-85. Color—Everywhere white or dirty white or cream color, with the exception of a few long hairs on the back which have long black tips. See fig. 2, Plate 34. Pelage and tail—Except for color these have the same characters in Gymnura alba that they do in G. gymnura. @ Monogr. Insectivora, p. 4. b’FWauna British India, Mammals, p. 221. ¢ Journ. Fed. Malay Sates Mus., III, p. 38, 1908. 454 PROCEEDINGS OF THE NATIONAL MUSEUM. VOL. XXXVI. ——— a na ras — _ Skull and teeth. —These possess the same relative shapes and pro- portion as they do in Gymnura gymnura, but are distinctly larger throughout. See fig. 8, Plate 35. Measurements.—See table, page 455, and diagram, page 452. Distribution.—Borneo and the off-lying island of Labuan. Speci- mens in the U. S. National Museum are from along the Sempang River, southwestern Borneo. Remarks.—I have been unable to find any constant character, aside from size and color, to distinguish between Gymmnura gymnura and G. alba. As in point of size the two species of Gymnura overlap one another slightly, it would be impossible to identify positively a large albino specimen of G. gymnura or a small individual of G. alba. A black individual is mentioned by Dobson “ as possibly coming from Borneo. The size relationships of the three forms on the basis of hind foot and basal length of skull are aoa by the diagram, page 452. @ Monogr. oe ora, p. 4. 455 1, 4, INSECTIVORES OF THE GENUS GYMNURA—LYON. *(1TOOATR) no. 1680. ‘ad AY, a “SUINS PelIp WOT] 10311M AG Poinsvay q “SJUSTIOINSBIUL S,10}99][0D v £ Gr | CeAGaN ey iL Siall Wan UG re] AOP es eames We mete net ha TP OR 6 sss |e eae ae em OD aes peers S OD 555) (OSES Rie Toscana. ems ODP 5 roll & os eae Ee aot od Sap | Of | 2°82 | 69 | 822 | OF |. --- pasojo ynq ‘yourystq |7777 7777" wom AWUBTS 777777 Op?" GRQ0P Ts | tooo" or concen ODS TA aay a ceo od oh} ee | eos | FL | 29S | Och |” “payeszeqtiqo Aywoyoelg |°- "7" uzom Ayounstq |7* ~*~” OD? S/CReey la |oes era RRP etre poe she od oTh | €6y | 9°94 | 99 | O6T | c6B |--*" pesopo ynq ‘youystq |777 7777" Ieam Jo sav, |" "aTeULeT | T8SGpT |--77 777777777 OE Tocks cece capesonas od ce | e-eo | os | cL | 08s | Shh |- ~~“ peyereqqo Apeonoerg [----""° uiom Ajouystq |" ~-** Ope} PeGGp Las Ore gee aes SD Ar ree od | | “IOATY sue Gr | 9'TS 18 | TZ | €9¢ (Sofas) eee mgt ee ote i Oe Ss “--maoMun AT[BaTOeIg |" **~ SIV | S8SSPT | -Wleg :oouI0g “MA “S | “DQ]D “9 ce | om | 19 | 89.) eee | We [777s uedo pu yourst |77777 7 TQM YG 2," OD | PSLOR I > een does Ope pug vom at od gf (9p | TL) 89. | 91e | 988. 720777" poyesoyttqo ATTeaN, 777777 wom AMouNsiq |°~"ayeutag | $8198 |77*77 Te OD. age nace ee od G9E €F LDF IOS RGEC Mae GGre ia oe are uedo puv youlysiq |77" 77777" uioM AYYSg |77 7 >> ~ ODF a SRZ9B aula = Maca eae OD" aie =. ere ae od 9g | SEF TZ | 6¢ Teco Neer esc ce DSSOTOMNG TOU SIG «pam aaaen one cae OD pees maleate aR | 66PES |°~“SuoTy, :urerg MOT |----"* Lou ninuwhb “p G18 | 2h | Ges | 39 OP Gy MOGEE len enka tenis Se SOP Sanaa a eae IeaM JO Sedv1y, |" ~~" Oia BLGEP Ey tau aioe ai a OD sign ean muae es ies od G‘9E 67 GL | 89 COG re OLS as rae cae MOQ OND UGH OUI) ST (im eae eer CELOVANUIT) Si tien O [ema OG EP. V: [We teenies eae aaa cg rl hagas Ao ete pet: | “1SULL, G'Sé 67 | $°9L | 19 ORGHSITGSE Sites n poyeroqz[qo Ayan |--- "7 mIOM AQOUNSIq |" "7" ~ Op***| 6ZEFFI | SUIQET oOINg :ByeuINg |-*-""-""**" "ee kOUL | “IOATY OF SF Qu ASO sal OS care | a Uta an Si pera OD gsoon |i eased gan uiOM Won |" "7" * Op-"~| PLIFFT : HOIVBUNDEH |e cnt ne een IF 1¢ SLIGO" ECO Sr a GOP |ierehie mies om eee OR ea oo gage UIOM AOUNSIC, |" -“eTeuLeg | TLIPPT | 7-777 OD Saisie cbt ce Aa ee od S| Sh) 82) 29 | 09% | S9e |--"peyerezTIGo ATTeOHOBIg |"-"“UIOM ATYSYS AOA |-""""-Op""*| ELTPFL pa eve oir Ine ai peace | “IOATY ¢ 8s 0g LL | 99 SOG aR GL Ge |wimen hale a je Whevaihg (OKO): ie ieee eh ee ae OD rs “*""""Op"""| GLIPFT | ABIS 94IT tee ae ae eas rE “oq SE oF TZ | 19 CO eL Cha laame pesoyo ynq “youTysIq |" ~"" 77" mioM AQOUTSIG |-*"" ~~ SIGN | ISShIT | Wnuedey, :eyeurng |----- "77777" POG Rare Pegg ul micieis (=| srisiere!s | siniaietein| siciaige\n | nictato otal etkions oie aicaeioeicln|ei ec a OMS TOU |e tae Aa OLC at ELC CLOOCS (1s ayaa aaa OUT| *IOATY cee} S| est] 29 | OL6 | OTF |~ ~~ pozeroqITGQo ATTeoNOVIg |"-~~~>-"~ uI0M AYYSIS | ---* aeW | 68PSIT | UIdurny :suvyeg | ~-punuwhb pinwwhb “pH | | | N ig aw : ce 48 53 ra g s Fa fe 8 fon Le] o | _ n — son= & ae ee 5 ‘ainjns . z ‘ON : Belo as 5 5 ES a 3 = | prouayds-ozidyoorseg 99 aes “429 Ayy1200T JEM AN seerslge| aie} el" z 88445 Bo re 3 a eoa2c = + fe EIS oS = S 5 S ‘DQ]D “4) pup DUNUWAD DINUWA JO sJUAWaLNSDIUL [DVUDLI PUD JDUWIAILT 456 PROCEEDINGS OF THE NATIONAL MUSEUM. vou. xxxvt. EXPLANATION OF PLATES. Puate 34 (about + natural. size). Fic. 1. Skin of Gymnura gymnura minor, adult female, Cat. No. 86784, U.S.N.M., 9° “ 3. Trong, Lower Siam, 2,000 feet altitude, collected by Dr. W. L. Abbott. . Skin of Gymnura alba, adult female, Cat. No. 145585, U.S.N.M., Sempang River, western Borneo, collected by Dr. W. L. Abbott. Skin of Gymnura gymnura gymnura, adult male, Cat. No. 144178, U.S.N.M., Little Siak River, east coast of Sumatra, collected by Dr. W. L. Abbott. Puate 35 (2 natural size, dorsal, lateral, and ventral views of skulls). Fic. 1. Gymnura gymnura minor, Cat. No. 86783, U.S.N.M. Type, Trong, Lower 2. 3. Siam. Gymnura gymnura gymnura, Cat. No. 144171, U.S.N.M., Little Siak River, Sumatra. Gymnura alba, Cat. No. 145584, U.S.N.M., Sempang River, western Borneo. PLATE 36 (twice natural size). Skulls of Hylomys suillus (Miiller and Schlegel), Cat. No. 124229, U.S.N.M., Tenasserim, Dr. W. L. Abbott, collector, and Podogymnura truet Mearns. Type, Cat. No. 125286, U.S.N.M., Mount Apo, Mindinao, Philippine Islands, Dr. E. A. Mearns, U. S. A., collector. jie ales . Right mandibular ramus of Podogymnura. . Dorsal view of mandibular teeth of Hylomys. . Dorsal view of mandibular teeth of Podogymnura, . Palatal view of part of skull of Hylomys. . Palatal view of part of skull of Podogymnura, . Lateral view of facial portion of skull of Podogymnura. Fa. ol m Co bo oO “I Right mandibular ramus of //ylomys. . Lateral view of facial portion of skull of Hylomys. ile bo PLATE 37 (about 7 natural size). External appearance of Podogymnura truci Mearns. Type, Cat. No. 125286, U.S.N.M., Mount Apo, Mindanao, Philippine Islands. . External appearance of Hylomys suillus dorsalis Thomas. 'Topotype, Cat. No. 1248328, U.S.N.M., Mount Kinabalu, British North Borneo. U. S. NATIONAL MUSEUM PROCEEDINGS, VOL. XXXVI PL. 34 SKINS OF THE THREE FORMS OF GYMNURA (ABOUT ONE-FOURTH NATURAL SIZE). FOR EXPLANATION OF PLATE SEE PAGE 456. U. S. NATIONAL MUSEUM PROCEEDINGS, VOL. XXXVI PL. 35 SKULLS OF THE THREE FORMS OF GYMNURA (TWO-THIRDS NATURAL SIZE). FoR EXPLANATION OF PLATE SEE PAGE 456. PL. 36 PROCEEDINGS, VOL. XXXVI U. S. NATIONAL MUSEUM "QGp 3DVd 33S JLV1d 4O NOILVNV1dXxa HO4 ‘(AZIS TIVUNLVN 3OIML) VHANWADOGOd GNV SAWOTAH VYSN35 3SHL JO S11NxS a ad - - U. S. NATIONAL MUSEUM PROCEEDINGS, VOL. XXXVI PL. 37 EXTERNAL APPEARANCE OF PODOGYMNURA AND HYLOMYS (ABOUT FIVE-SEVENTHS NATURAL SIZE). FOR EXPLANATION OF PLATE SEE PAGE 456. DESCRIPTION OF A NEW SNAKE FROM PANAMA. By Lronnarp STesNEGER, Curator, Division of Reptiles and Batrachians, U. S. National Museum. Among a small lot of reptiles collected by Mr. Frank E. Read at Bocas del Toro there is a specimen of a little snake which appears to be undescribed. Bocas del Toro is situated on a small island at one of the entrances to the bay of Chiriqui, on the Atlantic side and near the western extremity of the Republic of Panama. MESOPELTIS LONGIFRENIS, new species. Diagnosis—A small unpaired postmental in contact with anterior lower labial only; median dorsal scale row not enlarged; loreal three times as long as high, entering eye; no preocular; ventrals 165; sub- caudals 98; supralabials 8. Habitat—Panama. Type-specimen.—Cat. No. 38750, U.S.N.M.; Bocas del Toro, Republic of Panama; Frank E. Read, collector. Description of type-specimen.—Body only moderately compresed and slender; tail at vent suddenly diminishing both in height and width; eye moderate, with vertical pupil; rostral small, triangular, ‘as high as wide, scarcely visible from above; internasals small, their suture less than one-half that between the prefrontals, descending low on the side of the face and entering the orbit broadly; frontal longer than wide, longer than distance from tip of snout, shorter than the parietals but longer than the suture between them, wider than supraoculars; nasal undivided, with a horizontally oval nostril; loreal very long, three times as long as high, entering eye below pre- frontal; no preocular; two postoculars, upper largest; temporals 1+; supralabials 8, fourth, fifth, and sixth entering eye, sixth larg- est; mental small, followed by a small postmental separating the first pair of lower labials and also the mental from the anterior chin- shields; 8 lower labials, first five in contact with anterior chin-shield ; anterior chin-shields large, followed by one on the right side and PROCEEDINGS U. S. NATIONAL MUSEUM, VOL. XXXVI—No. 1681. 457 458 PROCEEDINGS OF THE NATIONAL MUSEUM. you. xxxvt. two on the left; behind these a wide plate somewhat wider than the adjacent first ventral; 15 rows of smooth scales without pores, median row not sensibly larger than the others; 165 ventrals; anal single; subeaudals 98. Color (in alcohol) brownish gray with a median — series of about 42 pale brown ocelli, edged with black, on the body, the last six descending to the ventrals so as to form transverse bands, many of the ocelli, especially anteriorly, divided medially, the halves more or less alternating; on the two or three outer scale rows a series of similar spots below the dorsal ones, each edged in front and behind by a blackish border, between which a whitish interval; tail above with about 17 alternating pale brown and brownish gray bands with a small whitish spot in the middle of the latter on the sides; head and nape dark brownish gray with indistinct marblings of pale brown and yellowish; third, fourth, and fifth labials marked with whitish; underside pale with dark brownish gray dots and spots which tend to form two parallel series on the ventrals. Dimensions. mm Total Jenethl-5 202522 25s aa see ee ee 4858 Tip. Of Sno utto sven Gs sae Se ae ae eee Vieni (0) tips08 shal = oa ee ee ee ee 158 Remarks.—The present species differs from J/esopeltis dimidiatus (Guenther) and d/. sanniolus Cope in having a much smaller un- paired postmental which is only in contact with the first pair of lower labials, and in a longer snout with a much longer and lower loreal. In the coloration of the back and sides it resembles Cope’s Leptog- nathus argus which, however, has a short snout and, moreover, possesses preoculars. From d/. sanniolus it differs furthermore by having a larger number of subcaudals, and from J/. dimidiatus, as well as from Z. argus, by the much smaller number of ventrals. U. S. NATIONAL MUSEUM PROCEEDINGS, VOL. XXXVI PL. 38 DORSAL VIEW OF DACTYLOBATUS ARMATUS (ONE-HALF NATURAL SIZE). FOR EXPLANATION OF PLATE SEE PAGE 459. DESCRIPTION OF A NEW SKATE (DACTYLOBATUS ARMATUS) FROM DEEP WATER OFF THE SOUTHERN ATLANTIC COAST OF THE UNITED STATES. By Barton A. Bran anp Atrrep C. Ween, Of the Division of Fishes, U. S. National Museum. Among the fishes collected by the Bureau of Fisheries steamer ° Albatross in deep water off Charleston, South Carolina, during 1885 - and 1886 were a male and female specimen of skate, closely allied to aja but differing greatly from this genus in having the middle rays of the pectorals very much produced. This genus may be called Dactylobatus in allusion to the finger-like processes of the pectorals. DACTYLOBATUS Bean and Weed, new genus. Disk subcircular, not rhombic as in Raja; middle rays of the pectorals produced as a finger-like process on each side; dorsal surface armed with spines and prickles but apparently without the patch of differentiated spines near the outer edge of the pectorals that is found in the males of Raja; tail abruptly marked off from the disk as in Raja, not tapering gradually as in RAinobatus and Narcine; tail armed with spines and prickles on the dorsal surface and with a rather wide dermal flap on each side; two dorsal fins near the end of the tail; a rudimentary caudal fin present. Type of genus.—Dactylobatus armatus. DACTYLOBATUS ARMATUS Bean and Weed, new species. Disk nearly circular, with the snout slightly projecting and with about six or seven of the middle pectoral rays produced to form a finger-like or flap-like process, armed on the ventral surface with a double row of sharp, hooked spines having their points directed toward the meson. This armament is continued forward along the edges of the disk, nearly to the snout; middle line of back and tail with a single row of enlarged hooked spines with a row of smaller ones on each side of it. A single large blunt spine on each shoulder and a row of five on each orbital ridge; rest of the dorsal surface and the dorsal fins sparsely covered with fine prickles. Color in alcohol, ashy with large black spots which coalesce to form blotches of various sizes and shapes; the belly white, clouded with PROCEEDINGS U. S. NATIONAL Museum, VOL. XXXVI—No. 1682. a 460 PROCHREDINGS OF THE NATIONAL MUSEUM. VoL, XXXVI. dark. Where what appears to be a layer of hardened mucus is re- moved, the color appears as a pinkish brown with dark brown spots. Two specimens are known—a male, number 62914, from Albatross station 2624, 32° 36’ N., 77° 29’ 15’’ W. at a depth of 258 fathoms, OUTLINE OF VENTRAL SURFACH OF DACTYLOBATUS ARMATUS. and a female, number 62915, from station 2666 or 2667, about 31° N. by 80° W., at a depth of about 270 fathoms. The male was taken October 21, 1885, and the female May 5, 1886. no. 1682. DHSCRIPTION OF A NEW SKATE—BEHAN AND WHED. Table of measurements. Male Om, SIMITLE son cave corsa atehatnemtcl a titcisl vac CCE cite ee Bete eee ree NOS Medi owedee ses Be 429.4 RIIINOTAN TR CER are eis Wana cen bona Rint hi aale Biaicde aoe’ sokay wha See Re msalainte state hate ek Lotte iw ¢20.5 PePIITUNOTIVOLIULSL Ila riches ood Soc cele eMcce e ciet cq tna Jamie's Caen ee abiedick es 17.7 BuouUrOOnG.Of Pectoral Ain «cece. cee ce cece clea ss HEALS Cuvke per ee en Bate 15.3 Bunt une Tolling Colbers.Of PUPS! 2. voc cece ee al conse ode sence ede ee nce 3:7 PMMA IOI OS DALACLOS 5 4.0 nc bic an nc an cn. cece g cab dsah wok dace Mesto gence ie 4.9 PL EPRE UO TO UII COMOS IEG. sat. selosc csc acleas nuns tates cucewauescduenceaccusentes 2.6 BBM MMe TOMIUS AISt PLUSH 8 oc. cece reste m ane cmee oekewess Jie eee bee Seth 6.0 MAUOTITO TOLER MAS PPG LUESE oe ech odu sco dodo bea cle oe ateewneaanaee ae cece buco csc 8.0 cnatulte pea ersiellite) ate} ag C0) 011-0 Ea a oy i oie, a eae ey oS eke oe 4.1 PMOM EIR SAL att Ae arata the ctarclera,5 Statens! x. wera te Lid tia cis ata we DE pe ocean eee ba eelee we bemewweslen 14.9 syotopitin prog abe G nyo) Vag heY ay ao Cfo) cio 0 1237) 1 eae a gg Opn rate SVMIRCU RRM CEU IITNCO LLU Lat ey an ks >a Pe, Sata en Set PIS a ne aa ames ofa ie malw bra cis clea aeGele we Oe wep y Pare SUA BUWER COMLGIS Ol DUPUS: 2 okcccnnc-owclcacwasdcecoccdccws ves echt csweeeclnos ey 2.4 UCL IDR ECECOIT S PLLACIOS 2 cco. vc ls aaiacle'vine.ae winovdes caoedue se tue e Sein tenaas pene geen ek 2.2 VOR D CRE LIMOS TLS + ae seth. sv Satekncies We Dasha cuts cap athelel oe ahiclc bee bte Oia d o temian evil DE MIOHOCVE REIL TS LL Slittee. S.y28 fee e cet knee oeen cade yc combs delet lute ocean dues 4.5 SEA IDIBRUWE CI IAS LT SlLSmeb eis ecu cal nc vee nc woe nat toc ie ct acmie «cure crema secu ee oul Fingers extend beyond line of disk about.................2.20--2-ceeeeeeeeeeee See 2.5 Sen PRC MSIL ECL AVON VOL Gs scsctsie sini Acre amie wi dctiae nto cvs mjole'cia ke ae Heer auies we peanle eRe 14.5 411; inches. ®102 inches. ¢ 87; inches. 461} inches, — Female, _ — NNW SEKENNNOWSNANS RORSEL ST ape piAT Sats CIMT CES Cine. eo eT eeRGre ot CAOWNrONwMsearnooarnaanwac~e This species is called armatus in allusion to the peculiar armament of the ventral surface. A LIST OF BIRDS COLLECTED BY DR. PAUL BARTSCH IN THE PHILIPPINE ISLANDS, BORNEO, GUAM, AND MIDWAY ISLAND, WITH DESCRIPTIONS OF-THREE NEW FORMS. By Evear ALEXANDER MEARNS, Associate in Zoology, U. S. National Museum. On the voyage of the United States Bureau of Fisheries steamer Albatross to the Philippine Islands Dr. Paul Bartsch accompanied that vessel as a representative of the United States National Museum, and made collections of birds whenever opportunity afforded, being assisted in this work by the regular staff of the Albatross. Collections were thus made at Midway Island during November 7 and 8, 1907; at Guam Island, November 19 to 21, 1907; at Sandakan, Borneo, and vicinity, March 1 to 3, 1908; but by far the largest collection was gathered in the Philippine Islands, where many im- portant gaps were filled in the United States National Museum series of birds. The following forms are here described as new to science: Ramphalcyon capensis smithi and Pycnonotus goiavier suluensis from the Philippines, and Collocalia bartschi from Guam. The following-named Philippine species are new to the collection of the U. S. National Museum: Loriculus bonapartei, Yungipicus ramsayi, Rhinomyies ocularis, EHdoliosoma everetti, Pericrocotus marchese, Anthreptes wiglesworthi (female), and Chibia suluensis. The collection also contained several species not previously re- ceived from the Philippine Islands. BIRDS COLLECTED IN THE PHILIPPINE ISLANDS. MEGAPODIUS CUMINGI Dillwyn. Cat. No. 211299, adult. Taal Voleano, Batangas Province, Luzon, Decem- ber 27, 1907. GALLUS GALLUS (Linnzus). Cat. No. 2113800-1, male adults. Taal Voleano, Batangas Province, Luzon, December 27, 1907. Cat. No. 211302, female adult. Taal Volcano, Batangas Province, Luzon, December 27, 1907. PROCEEDINGS U. S. NATIONAL Museum, VOL. XXXVI—Nco. 1683. 463 464 PROCEEDINGS OF THE NATIONAL MUSEUM. VOL. XXXVI. OSMOTRERON VERNANS (Linnezus). Cat. No. 211308, male adult. Taal Voleano, Batangas Province, Luzon, December 27, 1907. PHAPITRERON BREVIROSTRIS Tweeddale. Cat. No. 211309, male adult. Davao, southern Mindanao, May 17, 1908. MUSCADIVORES NUCHALIS (Cabanis). Cat. No. 211307, —— adult. Port Binang, Laguna Provinee, Luzon, January 8, 1908. Cat. No. 211306, male adult. Lapae Island, Sulu group, February 17, 1908. MUSCADIVORES AENEA (Linnzus). Cat. No. 211304, male adult. Dumurug Point, Masbate, April 19, 1908. Cat. No. 211305, female adult. Dumurug Point, Masbate, April 18, 1908. MYRISTICIVORA BICOLOR (Scopoli). Cat. No. 211303, male adult. Lapac Island, Sulu group, February 17, 1908. MACROPYGIA TENUIROSTRIS Bonaparte. Cat. No. 211310, male adult. Taal Volcano, Batangas Province, Luzon, December 26, 1907. STREPTOPELIA DUSSUMIERI (Temminck). Cat. No. 211311, male immature. “Taal Volcano, Batangas Province, Luzon, December 27, 1907. PUFFINUS LEUCOMELAS (Temminck). Cat. No. 211223, female adult. Near the Philippine Islands, latitude, 13° 12’ N.; longitude, 131° 20’ E., November 24, 1907. HYDROCHELIDON HYBRIDA (Pallas). Cat. No. 211264-6, female adults. Manila Harbor, Luzon Island, May 4, 1908. STERNA BERGII BOREOTIS Bangs. Cat. No. 211263, female adult. Batangas Bay, Luzon Island, June 7, 1908. OCHTHODROMUS GEOFFROYI (Wagler). Cat. No. 211283, female adult. Jolo, Sulu Island, March 7, 1908. Cat. No. 211284, male adult. Tataan Island, Tawi Tawi group, February 19, 1908. ZEGIALITIS DUBIA (Scopoli). Cat. No. 211285, male immature. Majayjay, Laguna Province, Luzon, December 19, 1907. Cat.. No. 211286, female immature. VPansipit River, Batangas Province, Luzon, December 25, 1907. Cat. No. 211287, male adult. Davao, southern Mindanao, May 17, 1908. FEGIALITIS ALEXANDRINA (Linnzus). Cat. No. 2112S8S-91, females. Jolo, Sulu Island, March 7, 1908. NUMENIUS VARIEGATUS (Scopoli). Cat. No. 211296, male adult. Davao, Mindanao, May 16, 1908. ACTITIS HYPOLEUCOS (Linnzus). Cat. No. 211292, female adult. Taal Voleano, Luzon, December 26, 1907. Cat. No. 2112938, — adult. Jolo, Sulu Island, March 7, 1908. GLOTTIS NEBULARIUS (Gunnerus). Cat. No. 211295, female adult. Davao, Mindanao, May 17, 1908. RHYACOPHILUS GLAREOLA (Gmelin). Tagal name: ‘“ Manunagtog.”’ Cat. No, 211294, male adult. Majayjay, Luzon, December 19, 1907. no. 1683. BIRDS COLLECTED IN THE PHILIPPINES—MEARNS. 465 ORTHORHAMPHUS MAGNIROSTRIS (Vieillot). Cat. No. 211297, male adult. Tataan Island, Tawi Tawi group, February 19, 1908. Cat. No. 211298, female adult. Tataan Island, Tawi Tawi group, February 19, 1908. “ Eggs well developed.” The U. S. Bureau of Fisheries has recently sent us a third specimen collected by the Albatross Philippine Expedition : Cat. No. 211631, — adult. Palaui Island, off northeastern Luzon, November 18, 1908. PYRRHERODIAS MANILLENSIS (Meyen). Cat. No. 211270 (=15279 Mearns), female adult. Altitude 4,500 feet, near Baguio, Benguet Province, Luzon, July 6, 1907. Governor Pack asked Doctor Mearns to skin this bird for him. The governor subsequently gave the specimen to Dr. Paul Bartsch. DEMIEGRETTA SACRA (Gmelin). Cat. No. 211271, male adult. Dumurug Point, Masbate, April 18, 1908. Cat. No. 211272, male adult. Capunuypugan Point, Mindanao, May 10, 1908. Cat. No. 2112738, female adult. Capunuypugan Point, Mindanao, May 10, 1908. NYCTICORAX MANILLENSIS Vigors. Cat. No. 211274, male adult. Pansipit River, Luzon, December 25, 1907. Cat. No. 211275, female young. Pansipit River, Luzon, December 25, 1907. BUTORIDES JAVANICA (Horsfeld). Cat. No. 211277, female young. Taal Volcano, Luzon, December 27, 1907. Cat. No. 211278, male young. Taal Volcano, Luzon, December 26, 1907. Cat. No. 211280, male adult. Silago River, Samar, April 12, 1908. Cat. No. 211281, female adult. Silago River, Samar, April 12, 1908. Cat. No. 211282, male adult. Baganga River, eastern Mindanao, May 13, 1908. BUBULCUS COROMANDUS (Boddaert). Cat. No. 211276, female adult. Baganga River, eastern Mindanao, May 13, 1908. Adult female in nuptial plumage. IXOBRYCHUS CINNAMOMEA (Gmelin). Cat. No. 211279, female immature. Pansipit River, Luzon, December 25, 1907. ‘ DENDROCYGNA ARCUATA (Horsfield). Cat. No. 211268, male adult. Davao River, southern Mindanao, May 16, 1908. Cat. No. 211269, female adult. Davao River, southern Mindanao, May 16, 1908. ANHINGA MELANOGASTER (Pennant). Cat. No. 211267, female. Pansipit River, Luzon, December 27, 1907. SPILORNIS HOLOSPILUS (Vigors). Cat. No, 211318, male adult. Surigao River, northeastern Mindanao, May 8, 1908. PONTOAETUS LEUCOGASTER (Gmelin). Cat. No. 211312, male immature. Bongao Island, Tawi-tawi group, Febru- ary 23, 1908. ELANUS HYPOLEUCUS Gould. Cat. No. 211314, male adult. Tacloban, Leyte, April 11, 1908. Cat. No. 211315, male adult. Surigao River, northeastern Mindanao, May 8, 1908. Proc.N.M.vol.xxxvi—09——30 466 PROCEEDINGS OF THE NATIONAL MUSEUM. VOL. XXXVI. CACATUA HAEMATUROPYGIA (P. L. S. Miiller). Cat. No. 211318, male adult. Silago River, Samar, April 12, 1908. Cat. No. 211316, male adult. Dumurug Point, Masbate, April 19, 1908. Cat. No. 211317, female adult. Dumurug Point, Masbate, April 17, 1908. Cat. No. 211319, female adult. Siasi Island, Sulu group, February 18, 1908. TANYGNATHUS LUCIONENSIS (Linnezus). Cat. No. 211321-3, males. Dumurug Point, Masbate, April 19, 1908. Cat. No. 211324, female. Dumurug Point, Masbate, April 19, 1908. Cat. No. 211320, male. Romblon, Romblon Island, March 26, 1908. Cat. No. 211325, male. Lapac Island, Sulu group, February 17, 1908. LORICULUS BONAPARTETI Souancé. Cat. No. 211326, female adult. Jolo, Sulu Island, March 7, 1908. EURYSTOMUS ORIENTALIS (Linneus). Cat. No. 211353, male adult. Dumurug Point, Masbate, April 19, 1908. Cat. No. 211354, male adult. Silago River, Samar, April 11, 1908. Cat. No. 211356, male adult. Capunuypugan Point, Mindanao, May 10, 1908. Cat. No. 211355, female adult. Mati, eastern Mindanao, May 15, 1908. RAMPHALCYON CAPENSIS GIGANTEA (Walden). Cat. No. 211364, male adult. Baganga River, eastern Mindanao, May 18, 1908. Cat. No. 211365, male adult. Bongao Island, Tawi-tawi group, February 23, 1908. Cat. No. 211366, female adult. Bongao Island, Tawi-tawi group, February 23, 1908. The typical Mindanao specimen is decidedly bluer—less greenish blue—on_ the upper surfaces than the pair from Bongao Island. RAMPHALCYON CAPENSIS SMITHI, new subspecies. MASBATE STORK-BILLED KINGFISHER. Type.—Cat. No. 211363, U. S. N. M., adult male, collected by Dr. Paul Bartsch at Dumurug Point, Masbate Island, April 18, 1908. Characters.—Intermediate in color between 2. c. gigantea and R. ec. _gouedi. The largest Philippine form; head, buff; under parts, ochra- ceous-buff. Measurements of type (adult male).—Wing, 157 mm.; tail, 96; exposed culmen, 80; tarsus, 17. Geographical range.—Southeastern Luzon, Masbate, and others of the middle islands of the Philippines, where the ranges of 7. ©. gigantea and PR. c. gouldi approach each other. There are several specimens of this form in the Menage collection, gathered by Messrs. Worcester and Bourns, now in the Bishop Museum at Honolulu. Named in honor of Dr. Hugh M. Smith, Deputy Commissioner of Fisheries and director of the Albatross Philippine expedition. ALCEDO BENGALENSIS Gmelin. Cat. No. 211388, female adult. Pansipit River, Luzon, December 25, 1907. Cat. No. 211384, male adult. Taal Voleano, Luzon, December 27, 1907. Cat. No. 211385, female immature. Taal Voleano, Luzon, December 27, 1907. Cat. No. 211386, female adult. Montalbon, Luzon, January 1, 1908. no. 1683. BIRDS COLLECTED IN THE PHILIPPINES—MEARNS. 467 HALCYON GULARIS (Kuhl). Cat. No. 211367, female adult. Majayjay, Laguna Province, Luzon, De- cember 18, 1907. Cat. No. 211368, female adult. Santa Cruz, Laguna Province, Luzon, De- cember 20, 1907. Cat. No. 211369, male adult. Near San Fernando, Union Province, Luzon, March 16, 1908. Cat. No. 211370, male adult. Silago River, Samar, April 12, 1908. HALCYON WINCHELLI Sharpe. Cat. No. 2113871, male adult. Papahag Island, Tawi-tawi group, February, 23, 1908. HALCYON CHLORIS (Boddaert). Cat. No. 211372, female immature. Taal Voleano, Luzon, December 27, 1907. Cat. No. 211373, male immature. Pansipit River, Luzon, January 18, 1908. Cat. No. 211374, female immature. Pansipit River, Luzon, December 25, 1907. Cat. No. 211375, female immature. Port Binang, Luzon, January 8, 1908. Cat. No. 211376, male adult. Dumurug Point, Masbate, April 19, 1908. Cat. No. 211377, male adult. Panabutan Bay, Mindanao, February 5, 1908. Cat. No. 211378, female adult. Davao, Mindanao, May 17, 1908. Cat. No. 211379-SO, male adults. Lapaec Island, Sulu group, February 17, 1908. Cat. No. 211381, female adult. Bongao Island, Tawi-Tawi group, February 23, 1908. Cat. No. 211882, male adult. Bongao Island, Tawi-Tawi group, February 23, 1908. PENELOPIDES PANINI (Boddaert). Cat No. 211390-1, male adults. Dumurug Point, Masbate, April 18, 1908. Cat. No. 211392-3, male adults. Dumurug Point, Masbate, April 19, 1908. Cat. No. 211394, female adult. Dumurug Point, Masbate, April 18, 1908. Cat. No. 211395, female adult. Dumurug Point, Masbate, April 19, 1908. PENELOPIDES AFFINIS Tweeddale. Cat. No. 211396, male adult. Capunuypugan Point, Mindanao, May 10, 1908. CRANORRHINUS LEUCOCEPHALUS (Vieillot). Cat. No. 211387, male adult. Capunuypugan Point, Mindanao, May 10, 1908. Cat. No. 211388-9, male adults. Mati, eastern Mindanao, May 15, 1908. MEROPS AMERICANUS P. L. S. Miiller. Cat. No, 211357, male adult. Dumurug Point, Masbate, April 19, 1908. MEROPS PHILIPPINUS Linneus. Cat. No. 211360, male adult. Naguillian, Union Province, Mareh 15, 1908. Cat. No. 211361, male adult. Near Naguillian, Union Province, March 16, 1908. Cat. No. 211358, male adult. Pansipit River, Luzon, December. 25, 1907. Cat. No. 211359, male adult. Pansipit River, Luzon, January 8, 1908. Cat. No. 211862, male adult. Bongao Island, Tawi-Tawi group, February 23, 1908. HEMIPROCNE COMATA (Temminck). Cat. No. 211351, female adult. Silago River, Samar, April 12, 1908. Mindanao and Samar islands are included in the range of Hemi- procne comata comata. The specimens of H. c. major in the U. S. National Museum are from Luzon and Panay islands. 468 PROCEEDINGS OF THE NATIONAL MUSEUM. VOL. XXXVI. COLLOCALIA TROGLODYTES Gray. Cat. No. 211346, male adult. Near Majayjay, Luzon, December 19, 1907. Cat. No. 211347, adult. Near Majayjay, Luzon, December 19, 1907. COLLOCALIA MARGINATA Salvadori. Cat. No. 211348, male adult. Near Majayjay, Luzon, December 19, 1907. Cat. No. 211349, female adult. Montalbon, Luzon, January 1, 1908, Cat. No. 211350, female adult. Capunuypugan Point, Mindanao, May 10, 1908. PYROTROGON ARDENS (Temminck). Cat. No. 211352, male immature. 'Tacloban, Leyte, April 11, 1908. EUDYNAMIS MINDANENSIS (Linneus). Cat. No. 211328, male adult. Dumurug Point, Masbate, April 17, 1908. Cat. No. 211329, male adult. Dumurug Point, Masbate, April 19, 1908. Cat. No. 211330, female adult. Dumurug Point, Masbate, April 19, 1908. Cat. No. 211331, female immature. Dumurug Point, Masbate, April 17, 1908. CENTROPUS VIRIDIS (Scopoli). Cat. No. 211338, male immature. ‘Taal Volcano, Luzon, December 26, 1907. Cat. No. 211332, male adult. Dumurug Point, Masbate, April 18, 1908. Cat. No. 211333, female adult. Dumurug Point, Masbate, April 18, 1908. Cat. No. 211335-7, male adults. Baganga River, eastern Mindanao, May 13, 1908. Cat. No. 211334, female adult. Mati, eastern Mindanao, May 15, 1908. CENTROPUS JAVANICUS (Dumont). Cat. No. 211339, male adult. 'Tacloban, Leyte, April 11, 1908. DASYLOPHUS SUPERCILIOSUS (Cuvier). Cat. No. 211327, male adult. Magdalena, Laguna Province, Luzon, December 18, 1907. XANTHOLZAMA HAMACEPHALUM (P. L. S. Miiller). Cat. No. 211345, male adult. Taal Voleano, Luzon, January 18, 1908. YUNGIPICUS RAMSAYI Hargitt. Cat. No. 211844, male adult. Papahag Island, off Tawi Tawi, February 23, 1908. YUNGIPICUS VALIDIROSTRIS (Blyth). Cat. No. 211342, male adult. Magdalena, Laguna Province, Luzon, December 20, 1907. Cat. No. 211343, female adult. Magdalena, Laguna Province, Luzon, Decem- ber 20, 1907. Cat. No. 211341, male adult. Taal Volcano, Luzon, December 27, 1907. CHRYSOCOLAPTES LUCIDUS MONTANUS Grant. Cat. No. 211340, female adult. Mati, eastern Mindanao, May 15, 1908. An extreme example of montanus. This is a species of eastern Mindanao, intergrading with Jucidus at Pantar, western Mindanao, instead of being a mountain form. HIRUNDO GUTTURALIS Scopoli. Cat. No. 211447, female adult. Santa Cruz, Laguna Province, Luzon, De- cember —, 1907. Cat. No. 211448, female immature. Majayjay, Luzon, December 19, 1907. HIRUNDO JAVANICA Sparrman. Cat. No. 211449, female adult. Romblon, Romblon Island, March 26, 1908. no. 1683. BIRDS COLLECTED IN THE PHILIPPINES—MEARNS. 469 CYORNIS PHILIPPINENSIS Sharpe. Cat. No. 211500, male adult. Taal Voleano, Luzon, December 26, 1907. Cat. No. 211501, male adult. Taal Voleano, Luzon, December 27, 1907. Cat. No. 211502, female adult. Taal Volcano, Luzon, December 27, 1907. MUSCICAPULA WESTERMANNI Sharpe. Cat. No. 211517-8, female adults. Baguio, Benguet Province, Luzon, March 138, 1908. GERYGONE SIMPLEX Cabanis. Cat. No. 211503, male adult. Taal Voleano, Luzon, December 26, 1907. Cat. No. 211504, female adult. Taal Voleano, Luzon, December 27, 1907. Cat. No. 211505-6, female adults. Taal Volcano, Luzon, December 26, 1907. Cat. No. 211507-10, adults. Taal Voleano, Luzon, December 26, 1907. Cat. No. 211511, adult. Taal Voleano, Luzon, December 27, 1907. Cat. No. 211512, male adult. Naguillian, Union Province, Luzon, March 15, 1908. GERYGONE RHIZOPHORZ Mearns. Cat. No. 211518, female adult. Tataan Island, Tawi-tawi group, February 19, 1908. This specimen is doubtfully referred to this species on ac- count of its darker coloration and larger size than Gerygone simplex. The sides of the head are nearly as dark as in G. rhizophore from Mindanao, and, like it, the crissum is pale and the under tail-coverts white; but the upper parts are much paler, and the bill smaller than in typical rhizophorae. HYPOTHYMIS AZUREA (Boddaert). Cat. No. 211514, female adult. Port Binang, Luzon, January 9, 1908. Cat. No. 211515, male adult. Capunuypugan Point, Mindanao, May 10, 1908. Cat. No. 211516, male adult. Baganga River, Mindanao, May 13, 1908. RHIPIDURA CYANICEPS (Cassin). Cat. No. 211498, male adult. Olongapo, Luzon, January 8, 1908. Cat. No. 211494, “‘ female?” adult. Olongapo, Luzon, January 8, 1908. RHIPIDURA NIGRITOROQUIS Vigors. Cat. No. 211495-G6, male adults. Taal Volcano, Luzon, December 26, 1907. Cat. No. 211497-S, female adults. Taal Volcano, Luzon, December 26, 1907. Cat. No. 211499, male adult. Tataan Island, Tawi-tawi group, February 20, 1908. RHINOMYIAS OCULARIS Bourns and Worcester. An alcoholic specimen, taken on Pangamian Island, Sulu group, February 18, 1908, measures as follows: Length, 595 mm.; wing, 77; tail, 63; tarsus, 22.5; middle toe with claw, 18. ARTAMIDES KOCHI Kutter. Cat. No. 211408, male adult. Silago River, Samar, April 12, 1908. ARTAMIDES PANAYENSIS Steere. : Cat. No. 211406, male adult. Dumurug Point, Masbate, April 18, 1908. Cat. No. 211407, female adult. Cataingan Bay, Masbate, April 17, 1908. EDOLIOSOMA EVERETTI Sharpe. Cat. No. 211409, male adult. Lapaec Island, Sulu group, February 17, 1908. PERICROCOTUS MARCHES Guillemard. Cat. No. 211410, female adult. Jolo, Sulu Island, March 7, 1908. 470 PROCEEDINGS OF THE NATIONAL MUSEUM. VOL. XXXVI. LALAGE NIGER (Forster). Cat. No. 211411, male adult. Pansipit River, Luzon, December 25, 1907, Cat. No. 2114124, male adults. Naguillian, Union Province, Luzon, March 16, 1908. ; Cat. No. 211415, female adult. Naguillian, Union Province, Luzon, March 16, 1908. Cat. No. 211416, male adult. Dumurug Point, Masbate, April 19, 1908. IOLE EVERETTI Tweeddale. Cat. No. 211462, male adult. Capunuypugan Point, Mindanao, May 10, 1908. IOLE GULARIS (Pucheran). Cat. No. 211464, male adult. Montalbon, Luzon, January 8, 1908. Cat. No. 211465, male adult. Montalbon, Luzon, January 1, 1908. Cat. No. 211466, no label—adult. (Montalbon, Luzon, January 1, 1908?) Cat. No. 211467, male adult. Port Bineng, Luzon, January 8, 1908. Cat. No. 211469, female adult. Tacloban, Leyte, April 11, 1908. IOLE GUIMARASENSIS Steere. Cat. No. 211468, male? Dumurug Point, Masbate, April 17, 1908. IOLE CINEREICEPS Bourns and Worcester. . Cat. No. 211465, female adult. Romblon, Romblon Island, March 25, 1908. POLIOLOPHUS UROSTICTUS (Salvadori). Cat. No. 211479, male adult. Baganga River, eastern Mindanao, May 18, 1908. Cat. No. 211480, female adult. Baganga River, eastern Mindanao, May 18, 1908. PYCNONOTUS GOIAVIER GOIAVIER (Scopoli). Cat. No. 211470, male adult. Taal Volcano, Luzon, December 26, 1907. Cat. No. 211471, male adult. Taal Volcano, Luzon, December 27, 1907. Cat. No. 211472-8, female adults. Taal Voleano, Luzon, December 27, 1907. Cat. No. 211474, male adult. Port Dupon, Leyte, May 7, 1908. PYCNONOTUS GOIAVIER SULUENSIS new subspecies. SULU GUAVA BULBUL., Type.—Cat. No. 211475, U.S.N.M., adult male from Jolo, Sulu Island, Philippines, collected March 7, 1908, by Dr. Paul Bartsch. Characters.—Similar to the Luzon Guava Bulbul (Pycnonotus goiavier goiavier), but smaller, with relatively shorter tail, broader supraorbital white stripes, and a paler auricular patch. Comparative measurements. No. of ~ r Name. speci- Sex. Locality. Wing. | Tail. | neway | Ho mens. } mm. | mm. mm. mm. Pycnonotus goiavier goiavicr . . - 10 | Male...... Luzon Island ..... 84 | 88 15.6 | 20.8 Pycnonotus goiavier suluensis . - GEER ae Mindanao Island. . 80.7 | 81.1 15.3 | 19.9 DORs aes aece eee wee Iii Eeedore soc | Basilan Island....| 83 86 16 21 WO Sas sass Shoe ee Se Sie soe Se | Sulu Island....... 81.7 | 82 15.8 | 19.3 Pycnonotus goiavier goiavicr --- 6 | Female....; Luzon Island..... 80.3 | 84.5 14.7) 20.3 Pycnonotus goiavier suluensis. . Ge zdokeesee Mindanao Island..| 79 79.3 14.8 | 19.1 BO Rae ee to ee ne oe 1 seo eee ee Sulu Island ...... 77 85 15.5 | 18.5 The Philippine forms of Pycnonotus goiavier are distinguishable from P. analis (Horsfield) by their smaller size and considerable color differences. no. 1683. BIRDS COLLECTED IN THE PITLIPPINES—MEARNS. Ava Material ewamined.—Twenty-three skins of Pycnonotus goiavier gotavier from the northern Philippine islands of Luzon, Mindoro, Panay, Samar, and Leyte; 20 skins from the southern Philippine islands of Mindanao, Basilan, and Sulu. PYCNONOTUS GOIAVIER SULUENSIS. Cat. No. 211475, male adult. Jolo, Sulu Island, March 7, 1908. Type. Cat. No. 211476-7, male adults. Jolo, Sulu Island, March 7, 1908. Cat. No. 211478, female adult. Jolo, Sulu Island, March 7, 1908. PETROPHILA MANILLENSIS (Forster). Cat. No. 211618, male adult. Twin Peaks, Benguet wagon road, Luzon, March 138, 1908. Cat. No. 211619-20, females, immature. Baguio, Benguet Province, Luzon, March 13, 1908. Cat. No. 211614, female adult. Majayjay, Luzon, December 20, 1907. Cat. No. 211615, male adult. Taal Voleano, Luzon, December 27, 1907. Cat. No. 211616, female adult. Summit of Taal Volcano, Luzon, January 18, 1908. Cat. No. 211617, female, immature. Port Binang, Luzon, January 8, 1908. COPSYCHUS MINDANENSIS (Gmelin). Cat. No. 211610, male adult. Between Magdalena and Majayjay, Luzon, December 18, 1907. Cat. No. 211611, female adult. Taal Volcano, Luzon, December 27, 1907. Cat. No. 211612, male, immature. Romblon Island, March 26, 1908S. Cat. No. 211615, female adult. Lapac Island, Sulu group, February 17, 1908. PRATINCOLA CAPRATA (Linnzus). Cat. No. 211519, male adult. Pansipit River, Luzon, December 25, 1907. Cat. No. 211520, male adult. Taal Voleano, Luzon, December 26, 1907. Cat. No. 211521-2, male adults. Taal Volcano, Luzon, December 27, 1907. Cat. No. 211523-4, female adults. Taal Volcano, Luzon, December 27, 1907. Cat. No. 211525, female adult. Baguio, Benguet Province, Luzon, March 18, 1908. ORTHOTOMUS FRONTALIS Sharpe. Cat. No. 211482, male adult. Capunuypugan Point, Mindanao, May 10, 1908. ORTHOTOMUS RUFICEPS (Lesson). Cat. No. 211488, male adult. Sitanki Island, February 26, 1908. Cat. No. 211484, female adult. Sitanki Island, February 26, 1908. CISTICOLA CISTICOLA (Temminck). Cat. No. 211485, male adult. Taal Volcano, Luzon, December 27, 1907. Cat. No. 211486, male adult. Pozorrubio, Benguet Wagon Road, Luzon, March 12, 1908. Cat. No. 211487-S, male adults. Near San Fernando, Union Province, Luzon, March 16, 1908. Cat. No. 211489, female adult. Near San Fernando, Union Province, Luzon, March 16, 1908. Cat. No. 211490, 19, 1908. Cat. No. 211492, female adult. Jolo, Sulu Island, March 7, 1908. CISTICOLA EXILIS (Vigors and Horsfield). Cat. No. 211491, male adult. Tacloban, Leyte, April 11, 1908. adult. Lemery, Batangas Province, Luzon, January 472 PROCEEDINGS OF THE NATIONAL MUSEUM. VOL, XXXVI. MEGALURUS PALUSTRIS Horsfield. Cat. No. 211481, female adult. Near Naguillian, Union Prevince, Luzon, March 15, 1908. ACANTHOPNEUSTE BOREALIS (Blasius). Cat. No. 211605, female adult. Between Magdalena and Majayjay, Luzon, December 18, 1907. Cat. No. 211606, female adult. Pansipit River, Luzon, December 25, 1907. Cat. No. 211607, female adult. Pozorrubio, Luzon, March 12, 1908. Cat. No. 211608, male adult. Baguio, Benguet Province, Luzon, March 13, 1908. Cat. No. 211609, female adult. Baguio, Benguet Province, Luzon, March 13, 1908. ARTAMUS LEUCORYNCHUS (Linnzus). Cat. No. 211397-S, male adults. Taal Voleano, Luzon, December 26, 1907. Cat. No. 211399, female adult. Taal Volcano, Luzon, January 18, 1908. Cat. No. 211400, female adult. Majayjay, Luzon, December 19, 1907. Cat. No. 211401, female adult. Port Dupon, Leyte, May 7, 1908. Cat. No. 211402, female adult. Jolo, Sulu Island, March 7, 1908. Cat. No. 211403, male adult. Papahag Island, Tawi Tawi group, February 23, 1908. Cat. No. 211404, female adult. Papahag Island, Tawi Tawi group, February 23, 1908. : Cat. No. 211405, 23, 1908. adult. Papahag Island, Tawi Tawi group, February CEPHALOPHONEUS NASUTUS (Scopoli). Cat. No. 211435-6, female adults. Baguio, Benguet Province, Luzon, March 13, 1908. Cat. No. 211487, male adult. Naguillian, Union Province, Luzon, March 15, 1908. OTOMELA LUCIONENSIS (Linnzus). Cat. No. 211488, female adult. Near Santa Cruz, Luzon, December 18, 1907. Cat. No. 211439-40, males, immature. Magdalena, Luzon, December 20, 1907. Cat. No. 211441, female immature. Pansipit River, Luzon, December 25, 1907. Cat. No. 211442, male adult. Taal Voleano, Luzon, December 27, 1907. Cat. No. 211443, male adult. Olongapo, Luzon, January 8, 1908S. Cat. No. 211444, male adult. San Fernando, Union Province, Luzon, March 16, 1908. Cat. No. 211445, male adult. Tacloban, Leyte, April 11, 1908. HYLOTERPE ALBIVENTRIS Grant. Cat. No. 211446, male adult. Sablan, Benguet Province, Luzon, March 15, 1908. PARDALIPARUS ELEGANS (Lesson). Cat. No. 211597-9, male adults. Baguio, Benguet Province, Luzon, March 13, 1908. : Cat. No. 211600-1, female adults. Baguio, Benguet Province, Luzon, March 13, 1908. Cat. No. 211602, female adult. Near Sablan, Benguet Province, Luzon, March 15, 1908. no. 1683. BIRDS COLLECTED IN THE PHILIPPINES—MEARNS. 478 CALLISITTA MESOLEUCA (Grant). Cat. No. 211603, male adult. Baguio, Benguet Province, Luzon, March 13, 1908. Cat. No. 211604, female adult. Baguio, Benguet Province, Luzon, March 13, 1908. ZOSTEROPS BASILANICA Steere. Cat. No. 211530, male adult. Papahag Island, Tawi Tawi group, February 238, 1908. ZOSTEROPS MEYENI Bonaparte. Cat. No. 211526, male adult. Taal Voleano, Luzon, December 26, 1907. ZOSTEROPS WHITEHEADI Hartert. Cat. No, 211527-9, female adults. Baguio, Benguet Province, Luzon, March 13, 1908. DICHUM PAPUENSE (Gmelin). Cat. No. 211533, male adult. Basiao Island, off Samar, April 16, 1908. Cat. No. 211534, male immature. Basiao Island, off Samar, April 16, 1908. DICKUM PYGMZEUM (Kittlitz). Cat. No. 211531, male adult. Dumurug Point, Masbate, April 19, 1908. Cat. No. 211532, female adult. Dumurug Point, Masbate, April 19, 1908. LEPTOCOMA SPERATA (Linnzus). Cat. No. 211537-41, male adults. Basiao Island, off Samar, April 16, 1908. Cat. No. 2115424, female adults. Basiao Island, off Samar, April 16, 1908. Cat. No. 211545, female adult. Silago River, Samar, April 12, 1908. CYRTOSTOMUS JUGULARIS JUGULARIS (Linnzus). Cat. No. 211546, female adult. Between Magdalena and Majayjay, Luzon, December 158, 1907. Cat. No. 211547, male adult. Santa Cruz, Luzon, December 20, 1907. Cat. No. 211548, female adult. Taal Voleano, Luzon, December 26, 1907. Cat. No. 211549, female adult. Taal Volcano, Luzon, December 27, 1907. Cat. No. 211550, male adult. Pansipit River, Luzon, January 18, 1908. Cat. No. 211551, male adult. Naguillian, Union Province, Luzon, March 15, 1908. Cat. No. 211552, male adult. Romblon, Romblon Island, Luzon, March 25, 1908. Cat. No. 211553-4, males young. Dumurug Point, Masbate, April 17, 1908. Cat. No. 211555, male young. Dumurug Point, Masbate, April 19, 1908. Cat. No. 211556-60, female adults. Basiao Island, off Samar, April 16, 1908. CYRTOSTOMUS JUGULARIS WOODI Mearns. Cat. No. 211561—2, male adults. Jolo, Sulu Island, March 7, 1908. Cat. No. 211563, female adult. Jolo, Sulu Island, March 7, 1908. Cat. No. 211564, male adult. Sitanki Island, February 26, 1908. Cat. No. 211565-6, female adults. Sitanki Island, February 26, 1908. Cat. No. 211567, male immature. Papahag Island, Tawi Tawi group, Feb- ruary 23, 1908. Cat. No. 211568-9, male adults. Papahag Island, Tawi Tawi group, Feb- ruary 23, 1908. Cat. No. 211570-1, female adults. Papabag Island, Tawi Tawi group, Feb- ruary 23, 1908. There is also a female alcoholic specimen taken on Pangamian Island, Sulu group, February 13, 1908. ANTHREPTES MALACCENSIS (Scopoli). Cat. No. 211536, female adult. Sitanki Island, February 26, 1908. 474 PROCEEDINGS OF THE NATIONAL MUSEUM. VOL. XXXVI, ANTHREPTES WIGLESWORTHI Hartert. Cat. No. 211535, female adult. Jolo, Sulu Island, March 7, 1908. This is a topotype. An adult male specimen from Pangamian Island, Sulu group, taken February 18, 1908, was also received in alcohol. MOTACILLA OCULARIS Swinhoe. Cat. No. 211451, female adult. Taal Volcano, Luzon, December 26, 1907. MOTACILLA BOARULA MELANOPE (Pallas). Cat. No. 211452, female adult. Majayjay, Luzon, December 18, 1907. Cat. No. 211453, female adult. Majayjay, Luzon, December 19, 1907. ANTHUS HODGSONI Richmond. Cat. No. 211456-7, male adults. Baguio, Benguet Province, Luzon, March 13, 1908. Cat. No. 211458-9, female adults. Baguio, Benguet Province, Luzon, March 13, 1908. Cat. No. 211460-1, female adults. Near Sablan, Benguet Province, Luzon, March 15, 1908. ANTHUS RUFULUS Vieillot. Cat. No. 211454, female adult. Magdalena, Luzon, December 20, 1907. Cat. No. 211455, male adult. Pansipit River, Luzon, December 25, 1907. ALAUDA WATTERSI Swinhoe. Cat. No. 211450, — adult. Magdalena, Luzon, December 20, 1907. MUNIA ORYZIVORA (Linnzus). Cat. No. 211621, male young. Pozorrubio, Benguet Road, Luzon, March 12, 1908. MUNIA JAGORI Martens. Cat. No. 211623-4, male adults. Tacloban, Leyte, April 11, 1908. Cat. No. 211625, female immature. Tacloban, Leyte, April 11, 1908. Cat. No. 211626, male adult. Jolo, Sulu Island, February 17, 1908. MUNIA CABANISI Sharpe. } ‘at. No. 211627, male adult. Pozorrubio, Benguet Road, Luzon, March 12, 1908. ; Cat. No. 211628, female adult. Pozorrubio, Benguet Road, Luzon, March 12, 1908. Cat. No. 211629, female adult. Naguillian, Union Province, Luzon, March 15, 1908. Cat. No. 211622, female, immature. Naguillian, Union Province, Luzon, March 15, 1908. ORIOLUS CHINENSIS Linnzus. Cat. No. 211427, male adult. Taal Volcano, Luzon, December 27, 1907. Cat. No. 211428, female adult. Taal Volcano, Luzon, December 26, 1907. Cat. No. 211429, — immature. Taal Valcano, Luzon, December 27, 1907. Cat. No. 211430, male adult. Naguillian, Union Province, Luzon, Mareh 15, 1908. Cat. No. 211431, male adult. Mati, southeastern Mindanao, May 15, 1908. There is also an adult alcoholic specimen from Pangamian Island, Sulu group, taken February 13, 1906. DICRURUS BALICASSIUS (Linnzus). Cat. No. 211483, — adult. Without a label, but said by Doctor Bartsch to have been taken on the Pansipit River, Luzon. g CHIBIA SULUENSIS (Hartert). Cat. No. 211432, male adult. Lapaec Island, off Siasi Island, February 17, 1908. Measurements of skin: Length, 284 mm.; wing, 148; tail, 145; culmen, 35; tarsus, 25. no. 1683. BIRDS COLLECTED IN THE PHILIPPINES—MEARNS. 475 ZETHIOPSAR CRISTATELLUS (Gmelin). Cat. No. 211424, female adult. Taal Voleano, Luzon, December 27, 1907. Cat. No. 211425, male adult. Pansipit River, Luzon, December 27, 1907. SARCOPS CALVUS (Linnzus). Sarcops lowi Suarrr, Trans. Linn. Soc., 1877 (Sibutu Island). Cat. No. 211419, male adult. Papahag Island, Tawi Tawi group, February 23, 1908. SARCOPS CALVUS MELANONOTUS Grant. Cat. No. 211417, male adult. Dumurug Point, Masbate, April 19, 1908. Cat. No. 211418, male adult. Tacloban, Leyte, April 11, 1908, LAMPROCORAX PANAYENSIS (Scopoli). Cat. No. 211420, female adult. Taal Voleano, Luzon, January 18, 1908. Cat. No. 211426, male adult. Taal Voleano, Luzon, January 18, 1908. Cat. No. 211421, male adult. Pansipit River, Luzon, January 18, 1908. Cat..No. 211422, female adult. Dumurug Point, Masbate, April 17, 1908. Cat. No. 211428, male adult. Papahag Island, Tawi Tawi group, February 23, 1908. CORONE PHILIPPINA (Bonaparte). Cat. No. 211434, male adult. Bongao Island, Tawi Tawi group, February 23, 1908. BIRDS COLLECTED IN BORNEO. HIRUNDO JAVANICA Sparrman. Cat. No. 211581, female adult. Sandakan, Borneo, March 2, 1908. PYCNONOTUS PLUMOSUS Blyth. Cat. No. 211573-4, female adults. Sandakan, Borneo, March 3, 1908. PYCNONOTUS SIMPLEX Lesson. Cat. No. 211575, male adult. Sandakan, Borneo, March 3, 1908. This ap- pears to be one of the white-eyed forms of P. simplex mentioned by Doctor Richmond, in the Proceedings of the United States National Mu- seum, XXVI, 1903, page 506; but the color of the iris was not noted by Doctor Bartsch. COPSYCHUS NIGER Wardlaw Ramsay. Cat, No. 211572, male adult. Sandakan, Borneo, March 3, 1908. ZEGITHINA VIRIDIS (Bonaparte). Cat. No. 211576, male adult. Po Bui Island, off Sandakan, Borneo, March 1, 1908. Cat. No. 211577, male immature. Sandakan, northeastern Borneo, March 3, 1908. ANTHREPTES MALACCENSIS (Scopoli). Cat. No. 211582-5, male adults. Sandakan, Borneo, March 3, 1908. Cat. No. 211586, female adult. Sandakan, Borneo, March 38, 1908. Cat. No. 211587-92, male adults. Po Bui Island, off Sandakan, Borneo, March 1, 1908. Cat. No. 211593-6, female adults. Po Bui Island, off Sandakan, Borneo, March 1, 1908. UROLONCHA FUSCANS (Cassin). Cat. No. 211578, male adult. Sandakan, Borneo, March 3, 1908. Cat. No. 211579-80, female adults. Sandakan, Borneo, March 3, 1908. A476 PROCEEDINGS OF THE NATIONAL MUSEUM. VOL. XXXVI. BIRDS COLLECTED AT GUAM ISLAND. HALCYON CINNAMOMINUS Wn. Swainson. Cat. No. 211254, female adult. Guam, November 19, 1907. COLLOCALIA BARTSCHI, new species. GUAM ISLAND SWIFTLET. Type.—No. 211255, U.S.N.M. Adult female, collected on the island of Guam, Pacific Ocean, November 20, 1907, by Dr. Paul Bartsch. Characters——Most closely related to Collocalia unicolor amelis Oberholser, described from specimens taken in the Benguet High- lands of Luzon, Philippine Islands. Its naked tarsus and much paler coloration at once separate it from Collocalia fuciphaga; and the absence of a whitish band across the rump distinguishes it from members of the Collocalia francica group. An adult male specimen (Cat. No. 200566, U.S.N.M.) of this species which I collected on Guam Island, July 20, 1905, is labeled “ Collocalia u. amelis” in Mr. Oberholser’s handwriting. Although Oberholser mentioned this specimen under Collocalia unicolor amelis, and included the island of Guam in the area of distribution of that form in his “ Monograph of the Genus Collocalia,” published in the Proceedings of the Academy of Natural Sciences of Philadelphia, pp- 193 and 194, issued July 26, 1906, he did not include it in the table of measurements which concludes the description, which cir- cumstance suggests that its small size led him to consider its identity with amelis doubtful. The wing is shorter than in Collocalia u. amelis while the tail is, relatively, a little longer, with the tail-feathers and upper tail-coverts much broader. The color differs from amelis in being paler on the hind-neck and sides of rump, the back a slightly paler and warmer brown, and the under parts decidedly paler, becoming almost silvery whitish on the chest. This new species is named in honor of Dr. Paul Bartsch in recog- nition of this important contribution of birds to the collection of the U. S. National Museum, made in connection with the operations of the Albatross Philippine Expedition. The following table gives comparative measurements: No. of a : are maj]. | 2xposed | Tar- suet Sex. Name. Locality. Wing. | Tail. culmerteltarae mm. mm. mm mm. 1)| Malone. s22 Collocalia bartschi.......- Guan dissere-.e-- ee 108 53 3.9 8.8 1 | Female....} Collocalia bartschi......-. Guam Tdiss22 35 ee 108 53 3.94 8.6 1 (?) Collocalia unicolor wni- | Ceylon............--..- 120 55 4.5 | 10 color. | 1 (?) Collocalia unicolor uni- | Ceylon........--.----- 116 54 4 9.5 color. 63) Male--e see Collocalia wnicolor amelis.| Benguet Prov., Luzon 116 50 4.8 9.9 Tdi Peck: 6 | Female....) Collocalia unicolor amelis.| Benguet Provy., Luzon 116 50 5 10 OBS IS no. 1683. BIRDS COLLECTED IN THE PHILIPPINES—MEARNS. 47% RHIPIDURA URANIZ Oustalet. Cat. No. 211256, male adult. Guam, November 20, 1907. Cat. No. 211257, male adult. Guam, November 21, 1907. MYZOMELA RUBRATRA (Lesson). Cat. No. 211258, male adult. Guam, November 20, 1907. Cat. No. 211259, male adult. Guam, November 21, 1907. APLONIS KITTLITZI (Finsch and Hartlaub). Cat. No. 211260, male adult. Guam, November 19, 1907. Cat. No. 211261, female adult. Guam, November 19, 1907. Cat. No. 211262, female, immature. Guam, November 19, 1907. BIRDS COLLECTED AT MIDWAY ISLAND AND VICINITY. STERNA FUSCATA CRISSALIS (Lawrence). Cat. No. 211229, immature. Midway Island, Hawaiian Islands, November, 1907. OCEANODROMA LEUCORHOA (Vieillot). Cat. No. 211225, male adult. Near Midway Island, Hawaiian Islands, lati- tude 27° 20’ N., longitude 172° 45’ W., November 5, 1907. Cat. No. 211226, male adult. Near Midway Island, Hawaiian Islands, lati- tude 27° N., longitude 179° E., November 9, 1907. Cat. No. 211227, adult. Near Midway Island, latitude 26° N., longitude 174° 16’ E., November 11, 1907. Cat. No. 211228, female adult. Near Midway Island, latitude 26° N., longi- tude 174° 16’ E., November 11, 1907. ‘“‘ Obtained at 10 p. m.” OCEANODROMA TRISTRAMI Stejneger. Cat. No. 211224, female adult. Near Midway Island, Hawaiian Islands, latitude 28° 15’ N., longitude 176° 30’ W., November 6, 1907. The skin of this specimen of the very rare Tristram’s fork-tailed petrel measures as follows: Length, 245 mm.; wing, 175; tail, 118; exposed culmen, 17.5; tarsus, 27; middle toe with claw, 26. PUFFINUS CUNEATUS Salvin. Cat. No. 211222, female adult. Near Midway Island, Hawaiian Islands, latitude 28° 15’ N., longitude 176° 30’ W., November 6, 1907. PORZANULA PALMERI Frohawk. Cat. No. 211250, male adult. Eastern Island, Midway Island, Hawaiian group, November 7, 1907. Cat. No. 211251, female adult. Eastern Island, Midway Island, Hawaiian group, November 7, 1907. Doctor Bartsch informs me that this species had been introduced from Laysan Island some years previously and was abundant at the time of his visit. CHARADRIUS DOMINICUS FULVUS (Gmelin). Cat. No. 211230-1, male adults. Sand Island, Midway Island, Hawaiian Islands, November 8, 1907. Cat. No. 211232-3, male adults. Eastern Island, Midway Island, Hawaiian Islands, November 7, 1907. ARENARIA INTERPRES (Linnzus). Cat. No. 211234, male adult. Sand Island, Midway Island, Hawaiian group, November 8, 1907. : Cat. No. 211235-9, female adults. Sand Island, Midway Island, Hawaiian group, November 8, 1907. 478 PROCEEDINGS OF THE NATIONAL MUSEUM. VOL. XXXVI. NUMENIUS TAHITIENSIS (Gmelin). Cat. No. 211242-6, female adults. Sand Island, Midway Island, Hawaiian group, November 8, 1907. Cat. No. 211247-9, female adults. Eastern Island, Midway Island, Hawaiian group, November 7, 1907. PISOBIA AURITA (Latham). Cat. No. 211240-1, female adults. Sand Island, Midway Island, Hawaiian group, November 8, 1907. TELESPYZA CANTANS Wilson. Cat. No. 211252-3, male adults. THastern Island, Midway Island, Hawaiian group, November 7, 1907. Introduced from Laysan Island. ADDITIONAL NOTES ON MAMMALS OF THE RHIO- LINGA ARCHIPELAGO, WITH DESCRIPTIONS OF NEW SPECIES AND A REVISED LIST. 3y Marcus Warp Lyon, Jr., Second Assistant Curator, Division of Mammals, U. S. National Museum. Since the publication of the very complete list of the mammals of the Rhio-Linga Archipelago, by Mr. Gerrit 8. Miller, jr., and a short paper by me on the mammals of Batam Island, collected by Mr. C. Boden Kloss,? Dr. W. L. Abbott has presented to the U. S. National Museum mammals from five other islands of the Rhio-Linga Archi- pelago. These specimens are mentioned in detail below, among them being three species hitherto undescribed. The islands recently visited by Dr. W. L. Abbott are Bulan (or Bulang), Jombol (or Chombol), Galang, Setoko, and Penjait Layer. The first three are shown on the map of the Rhio-Linga Archipelago published with Mr. Miller’s paper. Setoko appears on the map lying just northeast of Rempang, but is not named. Penjait Layer is a small island not shown on the map, but lies to the south of Setoko, from which it is separated by a strait about one-third mile wide. At the conclusion of this paper is given a list of all the mammals known to occur on the islands of the Rhio-Linga Archipelago based upon the present material and the two papers mentioned. MANIS JAVANICA Desmarest. 1822. Manis javanica DESMAREST, Mammalogie, Pt. 2, p. 387. Skin and skull of an adult male, from Pulo Bulan, Cat. No. 144418, U.S.N.M. Measurements: Head and body, 600 mm.; tail from anus, 505; hind foot, 99; weight, 9 kilos (20 pounds) ; greatest length of skull, 101; zygomatic width, 31.5. The zygomatic arch is complete and bony on each side. On the right side the posteriorly directed zygomatic process of the maxilla has met the anteriorly directed process of the squamosal. On the left side a similar condition exists, but the ossification has taken place in such a manner that a distinct @Proc. U. S. Nat. Mus., XX XI, No. 1485, pp 247-286, Sept. 11, 1906. bTdem, XXXI, No. 1502, pp. 653-657, Jan. 16; 1907. PrRoceepiInas U.S. NATIONAL MUSEuM, VOL. XXXVI—No. 1684. 479 480 PROCEEDINGS OF THE NATIONAL MUSEUM. VOL. XXXVI. bony segment is intercalated between the process of the maxilla and that of the squamosal, presenting the appearance of a true jugal or malar bone. “Caught a J/anis in Bulan. It was up a small tree about 20 feet and I could not at first tell what it was. I cut the sapling down and seized the J/anis by the tail as he was running away.” W. L. A. SUS RHIONIS Miller. 1906. Sus rhionis Mitter, Proc. U. S. Nat. Mus., XXX, No. 1466, p. 749, June 13, 1906. (Type-locality, Pulo Ungar.) The skins and skulls of two pigs collected on Pulo Jombol are clearly referable to this species, agreeing in all essential respects with the original series from the islands of Ungar, Sugi Bawa, and Great Karimon. For external and cranial measurements see table below. SUS VITTATUS Miiller and Schlegel. 1889-44. Sus vittatus MiLLer and Scuiecert, Verh. Natuur. Gesch. Nederl. Bezitt. Zool., p. 172, pls. 29, 32. (Type-locality, Sumatra, restriction by Jentink, Notes Leyden Museum, XXVI, p. 175, Oct. 16, 1905.) The skin and skull of an adult male from Pulo Penjait Layer is indistinguishable from the Sumatran Sus vittatus and is quite dis- tinct from Sus rhionis Miller, found elsewhere on the Rhio-Linga Archipelago. For external and cranial measurements see table below. External and cranial measurements of Sus rhionis and Sus vittatus. ; Cat. No. Cat. No. pone 144379, | 444 : eee nearly j|adult male, Dimensions. Bae adult fe- Pulo anna male, Pulo} Penjait Sus rhionse.| Jombol, | Layer, Sus ‘Susrhionis.| vittatus. Head and body 6, ain ic wc sioctes oe ane actions ae eee Seece oe ee ceees 1,150 1,025 1,160 SB SIV Gs Fo Tee ree cee are See eo ee eee eee eee 255 195 220 Lind footie. ese... Sacks ade hae tases See cee eee re rai na ere 259 244 250 Height-at shoulder@ss oc. ooo 0.5 ee eee ee er en eee 560 490 500 Weirhtin pounds (Gnilclos)ias= 228 ee eee en are oe eee eee 95(43) 58(26) 94(43) Upperilengthiofskwll)<7i7iee..-. see see tee eae 297 270 311 Basal length > 232 sore janoc oe ee tere oe eae RE Ee On ne Oe ae eee 265 241 272 Basilar length 2 3252-2 se ss See a ee ee ee ee ees 250 225 256 Palatal leneth 2. -22 4. <3 cee oe eee ee ee ee ee eee 187 166 192 Width of palate‘at pm 2... 438 ke es ee eee ee oes 30 27 31 beast width: of palate attron hol mtsenmce. aes aeneeeee ee eee 21 21 27 Width ot palateinclidine 77) S822 ese epee eee ee ee 58 56 63 AyEomatic breadth = fa 32s cases im ee cree oe ae ne 124 107 136 Least interorbital ‘breadth= 22... — ee ae seer ae eee eee 62 53 65 Parietal constriction: : <= i258. seo eee eee ee eee 25 Ns a eben 36 Nasal breadth at posterior extremity of premaxilla................-- 29 26 31 enethionnasals-= 5. 2 . 2S ee eee eerie an 142 128 142 Occipital depth'to basions. (420 a ee ee as 82 71 91 IMBNOTD Ose chs tenn ee De ar eee : 232 206 239 Maxillary toothruw, exclusive of canine..............-- ce 97 (5) 95 Second upper molar: 05 Sb) Soe Ss eee ere 1815.5 20X17 19X16 MhindsuppersaMolar’ 2+. oS. fees oe ee ee ee ee 28X18. 5 29X18. 5 Mandibular toothrow, excluding anterior pm............-...--:.--- 90 b) 92 Seeondilower molar:---.2 0. !-2 22252. ee EPRERLA GRRE Se Bias! ee ate ae L725 19X14 | 16.512 EEbirg lOWer MOI. os 2.5 see dec ck cee ero eee eee 27.15 (>) 30X15. 5 a Collector’s measurements. b Last molars not entirely through alveoli. no. 1684, MAMMALS OF THE RHIO-LINGA ARCHIPELAGO—LYON. 48] TRAGULUS PERFLAVUS Miller. 1906. Tragulus perflavus Mitter, Proc. U. 8S. Nat. Mus., XXXI, No. 1485, p. 251, Sept. 11, 1906. (Batam Island.) 1907. Tragulus perfiavus Lyon, Proc. U. S. Nat. Mus., XXXI, No. 1502, p. 6385, Jan. 16, 1907. (Batam and Galang islands.) Twenty-seven specimens of this well-marked species of mouse-deer have now been sent to the National Museum. At the time it was de- seribed it was known by but a single specimen taken by Mr. C. Boden Kloss on Batam Island. Since then Mr. Kloss has collected 9 addi- tional specimens on Batam, and 3 from Gong Hill, Pulo Galang. Doctor Abbott has recently collected 6 on Pulo Setoko and 8 on Pulo Bulan. The amount of individual variation in this large series is not great so far as color and markings are concerned, consisting chiefly in the intensity of the yellow color on the upper parts of the body behind the shoulders. In some individuals the black tips of the hairs on the back are quite conspicuous, thus obscuring the yellow color. The neck, however, in all the specimens is always strongly yellowish, without admixture of darker colors, and in none of the specimens does it make any approach to the neck coloration found in 7raqgulus flavi- collis Miller from Pulo Sugi. The type-specimen of 7. perflavus is about the average of the series so far as color is concerned. In external and cranial measurements there is more individual variation in the series than there is in the color and markings. See table. External and cranial measurements of Tragulus perflavus. 7 esse Se? Nie aeroumeede pa age Ke =) » ,| of iw oa 2 | |.s8 Ss. ed/8 Sis, ; Zz foe |.ee A=| col pe les Locality. 1A Sex. Age. a |e |em| Ae sien (HERS Seep Stoke En #)/ PF 1318/8 | tH | of | 82 sa Op \Sl-/135| 2/2/48 | 6) ge less | o Fed Sieh ee ek: > ao Sip | Hila || Fl ejo is |S" | | | | mm.|mm.| mm.) lbs. kilos.. mm.| mm. | mm. | mm 143196 | Male....; Immature...) 465 75 |130| 4 $B) 4.9) Pass 0 los lesa 143197 | Female .| Adult......- 5385 | 80] 1387 | 7 3.2 | 96.0 | 49.5 | 35.0) 41.5 +.| 143199 |... d 85 | 137 | 5h | 2.5 | 92.3 | 45.5 | 34.6 | 43.0 -| 144422 |... 81 | 121 | 4 1S S00 el? poe tae alee 144426 |... 80 | 135 ; 53 | 2.4 | 89.0 | 45.5 | 37.0} 40.5 144427 |... 75 | 132 | 62 | 3.1 | 93.5 | 47.7 | 37.0 | 42.0 144423 80 | 129} 5 | 2.3 | 90.0 | 44.5] 37.0 | 43.7 .| 144424 |...do.... ae | 80 | 131 | 5s | 2.5 | 89.0 | 47.0 | 34.5 | 41.6 -! 144425 |...do..... Lees GO.r ese | 484! 90/130! 5 | 2.3 | 88.5 | 45.4! 36.0 | 41.0 144394 | Female .| Young...-.- CFM aT i Ua fit Uy 0 ee aes ocDy Al Oule es) a aacae gE Z ED ee eee eee Goss ee- 467 | 75 | 134 | 43) 1.9 | 83.4 | 42.0 |......).-.... 144395 2003-552 | Nearly adult} 510 | 80 | 131 52 | 2.6 | 89.8 | 45.9 | 37.4 | 41.7 144401 saO2As AGG s cee 503 | 70}125| 6 | 2.7 | 92.6 | 47.5 | 34.8 | . 39.5 PASOG tl. = Osa 22 |-5-= doo 2.2 5 530 | 84 | 136 alee lee 48.0 | 38.0 » 43.0 144397)-\2 2:00%2.-<| 22-22 ah Aer eee | 582 | 85 | 187 | 64 | 3.0 |100.0 | 49.4 | 35.5 | 41.0 144399 | Male....|....-. (ite eons 475 | 67 |130| 5 | 2.3 | 91.5 | 49.5 | 39.0 | 42.6 144398 x ae he 0 eee aOR Ae 1S em Di | 93.7 | 51.0 | 35.6 | 41.6 143200 | Female .|....-. [2 ee ees S| ee 120. |. 2 see. | 92.0 | 48.0 | 37.4 | 44.0 143205 it ee Rees GOs 4-22 | 635 | 78 | 137 | 7% | 3.5 | 96.8 | 47.2 | 38.0 | 43.2 144439) 1; 22d0re.0-15--.- 5 6 Coe eae {icv erate | 127 ia wnta| aac ae O90 2 | 45..2, |, 00-5 | 414 1421250) 00". --.|---.~ dGs5se | 620 | 85 | 137 | 5% | 2.5 | 97.6 | 48.8 | 37.0 | 41.3 143202 | Male....| Nearlyadult) 583 | 82 | 131 | 4% | 2.2 | 89.0 | 46.0 | 37.0 | 40.5 143203 |...do....-. Adult.......| 585 | 77 | 130 | 5% | 2.5 | 91.7 | 47.3 | 37.0 | 42.0 PARADE) SHO. 22 | 2-28 lt ees Pee IS eee 1 pees | Bae <-| 91-8 | 47.9 | 36.5 | 42.0 } } a Type. Proc. N. M., vol. xxxvi—09——31 482 PROCEEDINGS OF THE NATIONAL MUSEUM. VOL. XXXVI. SCIURUS PENINSULARIS Miller. 1903. Sciurus peninsularis MILLER, Smiths. Misc. Coll, XLV, p. 10, Nov. 6, 1903. (Type-locality Pahang.) Six squirrels of the vittatus group from Pulo Bulan are referable to Sciurus peninsularis Miller. This is the same species that I have else- where called Scturus vittatus vittatus.% For measurements see table, page 483. RATUFA BULANA, new species. . Type.—Skin and skull of adult female, Cat. No. 144412, U.S.N.M., collected on Pulo Bulan, Rhio-Linga Archipelago, March 23, 1907, by Dr. W. L. Abbott. Original number, 5130. Diagnostic characters—Very similar to Ratufa insignis Miller,? from Pulo Sugi, but differing in having the naso-frontal suture shorter and the foramen leading from the orbital fossa into the pos- terior nares distinctly smaller. Color.—Type: General effect of upper parts of body ‘calneile color; posterior portion of upper part of head same, but finely griz- zled by lighter annulations of similar color of the hairs; anterior por- tion of upper surface of head dark hair brown, slightly and finely grizzled with whitish; outer surfaces of thighs, forelegs, and a narrow line along sides of body russet ; tail generally similar in color to the back; the hairs both above and below with lighter bases and the under side of the tail in the middle line with short appressed cream-colored hairs; under parts of body, throat, chin, inner side of fore and hind legs and sides of head beneath ears whitish; upper surface of hands and feet whitish, but irregularly suffused with a bright russet color, especially about the toes; sides of head and nose whitish, but dark- ened by brownish tips to the hairs; ears in general blackish brown, but with many lghter hairs on the inside. Series: The series of Ratufa bulana is quite uniform in color and none of the specimens depart much from the color of the type. One specimen, an adult female, Cat. No. 144408, U.S.N.M., is rather lighter in color, being generally clay-color above. Skull and teeth—The skulls of Ratufa bulana differ from those of FR. insignis, its nearest ally, in two very constant features, the rela- tively short naso-frontal suture and the small size of the foramen leading from the orbital fossa into the posterior nares. In 2. insignis the nasals taken both together have an hourglass constriction just posterior to the middle, which is lacking in #. bulana owing to the “Smiths. Mise. Coll., XLVIII, p. 278, Feb. 4, 1907; Proc. U. S. Nat. Maus, XXXI, p. 653, Jan. 16, 1907 ; Proc. U. S. Nat. Mus., XXXIV, p. 626, Sept. 14, 1908. > Smiths. Misc. Coll., XLV, p. 4, Nov. 6, 1903. no. 1684: MAMMALS OF THE RHIO-LINGA ARCHIPELAGO—LYON. 488 shorter naso-frontal suture. I can detect no differences between the teeth of the two species. Measurements.—See table below. Specimens examined.—F ive, all from Pulo Bulan. Remarks.—Ratufa bulana is very closely related to R. insignis. The skins of the type-specimens appear sufficiently distinct, but the type of A. insignis is evidently in an unworn and unbleached pelage, while the pelage of the series of 2. bulana is much bleached. One of the paratypes of 2. insignis, Cat. No. 115532, U.S.N.M., is practically indistinguishable externally from Cat. No. 144410, one of the para- types of &. bulana. All of the specimens of the latter species, how- ever, have the forelegs and thighs more russet than does the series of R. insignis. The cranial characters serve to separate the two forms instantly. Haeternal and cranial measurements of squirrels. im] Se a = te P= = Thesd 3 8 = a Rs § oF a 2/22/89) % |) as Name. Locality. AZ, Sex. Age. & £ |SE/+e] ss | ‘ee en 3 So Ae keys ors Sp Bolg (Bo (se) 2 | se ® ‘3 = 2 b HO oe | feast tae fen Peo / tela | | 3 mm.| mm.| mm.| mm. | mm.| mm. Ratufa bulana......... Pulo Bulan.| 144411 Adult 347 | 403 80 | 65 41 27 Do | d 144408 SOs: 330 | 385 72 | 64.5 | 39.3] 26.7 144409 |... do:s2 328 | 405 79 | 64 39.5 | 27.6 144410 |... doz: 342 | 413 82 | 66.5 | 40.2 | 27 c144412 |__. dower. 327 400 84 | 65.3 | 39 26. 8 144402 dos 218 197 51 | 50.9 | 31.3] 18.4 144403 does: 210} 190 54 | 50.3 | 31.2] 17.9 144404 dos. 222 | 200 53 | 52.3] 31.3] 183 144405 |... dose: 220 | (d) SOR ee lee ceo eae 144406 do2? 226 | 196 51 | 51.4 | 31.8] 19.6 144407 donee 220 | 190 52 | 50 SL ome Lae a Collector’s measurements. c Type. b Measured by writer after relaxing feet in water. d Defective. MUS ASPER Miller. 1900. Mus asper Miturr, Proc. Biol. Soc. Washington, XIII, p. 145, Apr. 21, 1900. One specimen, skin and skull of an adult female, from Pulo Setoko. This is‘the second specimen of the I/us asper group of rats known from the Rhio-Linga Archipelago. The other is the type of J/us batamanus from Pulo Batam, which I wrongly referred in the orig- inal description * to the Mus jerdoni group. A reexamination shows it to be a member of the Mus asper group. The infrequency with which rats of this group have been taken in the Rhio-Linga Archi- pelago is surprising, in view of their common occurrence on the Malay Peninsula, Sumatra, Banka, and Billiton. For measurements of the Setoko specimen, see table, page 485. @Proc. U. S. Nat. Mus., XXXI, No. 1502, p. 654, January 16, 1907. 484 ; PROCEEDINGS OF THE NATIONAL MUSEUM. VOL. XXXVI. MUS LINGENSIS Miller. 1900. Mus lingensis M1LLeR, Proc. Washington Acad. Sci., II, p. 206, Aug. 20, 1900, Of this rat, which is so widely distributed in the Rhio-Linga Archipelago, Doctor Abbott secured 10 individuals (8 skins and skulls, and 2 in alcohol) on Pulo Jombol. For external and cranial measurements, see table, page 485. MUS FIRMUS Miller. 1902. Mus firmus Mituer, Proc. Acad. Nat. Sci., Philadelphia, 1902, p. 155, June 11, 1902. Two specimens from Pulo Setoko do not differ from d/us firmus as found elsewhere in the Rhio-Linga Archipelago. For measurements, see table, page 485. MUS CHOMBOLIS, new species. Type.—Skin and skull of adult female, Cat. No. 144393, U.S.N.M., collected on Pulo Jombol, Rhio-Linga Archipelago, March 10, 1907, by Dr. W. L. Abbott. Original number, 5100. Diagnostic characters.—Similar to Mus firmus Miller, but slightly darker in color, smaller in size, and with distinctly smaller skull. Color.—Upper parts and sides of head and body, a coarse grizzle of brownish black and pale ochraceous buff, the brownish black in excess along the dorsal line. Along the sides the slate gray of the underfur shows through on the surface. Underparts and inner sur- faces.of legs dirty cream color; feet dull brownish, tail and ears dull dark brownish. Pelage, ete—The pelage consists of three types of hair—the rela- tively short, dark, gray underfur; soft, weak hairs, with dark-gray bases, light ochraceous buff subterminal rings, and short blackish- brown apices; relatively long, soft, grooved, brownish-black bristles. Tail concolor, dark brownish, 10 scales to the centimeter in the middle portion; each scale subtended by 3 hairs, each of which is equal to about 1 scale in length. Skull and teeth—The skull of Mus chombolis in general resembles that of J/. firmus, but is distinctly smaller and lhghter throughout, especially noticeable in the rostral portion, which is much shorter and narrower and much less deep; the bulle are smaller; the incisive foramina shorter; the anterior nares smaller. The maxillary tooth row is distinctly shorter in J/. chombolis than it is in WM. firmus and the individual teeth smaller; the incisors are also smaller and weaker. @Proc. Acad. Nat. Sci. Phila., 1902, p. 155, June 11, 1902; Proc. U.S. Nat. Mus., XXXI, No. 1485, p. 266, Sept. 11, 1906. no. 1684. MAMMALS OF THE RHIO-LINGA ARCHIPELAGO—LYON. 485 Measurements.—External: Head and body, 195 mm. (245)¢; tail, 230 (255) ; hind foot with claws, 45 (47.5). Cranial: Greatest length of skull, 46.8 (52.7); length of nasals in middle line, 17.9 (20.5) ; basal length, 41 (47); zygomatic width, 23.5 (26.2); width of ros- trum at antorbital foramen, 8.5 (10.3); width of brain case above roots of zygomata, 19 (21.2); least depth of rostrum, 8.7 (10.3) ; maxillary tooth row (alveoli), 8.2 (10). Specimens examined.—One, the type. Remarks.—Although the rats of the Mus firmus group hitherto known from the various islands of the Rhio-Linga Archipelago are referable to typical firmus (type-locality, Linga Island), the single known rat of this group from Pulo Jombol appears too different to be regarded as belonging to the same species as the others. That its small size is not due to immaturity is shown by a considerable amount of wear of the teeth, the closure of skull sutures, which remain open in young animals, and the perfect development of angles and ridges on the skull. An examination of many skulls of J/us firmus from the islands of the Rhio-Linga Archipelago fails to show any that approach the skull of J/. chombolis in its general small size, short rostrum, and smaller teeth. The somewhat darker color of J/. chom- bolis can not be considered as characteristic, and externally, aside from its smaller size, it can scarcely be differentiated from I/. firmus. External and cranial measurements of rats. a 8, lis aks se (Wie oh eaaksticd res Tan | s le ee ee v 2 5 » “a I — § . Name. Loeality. » Sex. Age. se) Sale abcd |") ae fe} = H see It eed 3 Bo Z eats etree lees eS 3 mam |a 8 & | % 3 o oI = 4 Bb fas} oO ee ae | oO Ne |e mm.|mm.| mm. | mm. | mm. | mm. Mus lingensis......... Pulo Jombol | 144384 | Female .} Adult...| 204} 148 39 | 44.5 | 22 6.6 IGE ak ee cor ae eelacece QOn seen 144385 |...do-.. BEG 214 160 41 | 46.8 | 22 6.7 Oe ese aralcaee GOL eek 144386 |...do dots. 216 171 42 | 47 21.6 6.9 iLO) ieee tee ee ees GOAse cae 144390 |...do-.. BedOsanes QOD eee 41 | 46.5 | 20.3 ib LD fc}, CS ae ae oe Gee Cdnes as. 144383 | Male....|...do-..... 220 178 44) | 48.2) |nasen. 6.6 (De Ape ese ee see GOupe eas 144387 |\...doz.-.- Sedo: =e 220 171 45 | 48.6 | 21.7 1.2 LO pele a adele a a (Close eee 144388 |...do....-. ed IG Ce ree ed 214 164 44 | 47.7 | 22.8 6.8 1 Foe Surah San one | ae Gee 144389 |...do..... Pe AdGt oss. 207 151 44 | 47 20. 2 6.8 Mus chombolis........|...-- G02.ce-e 144393 | Female .|...do....}| 195 | 255 45 | 46.8 | 23.5 8.2 MUS TTMUS 22 22-2- Pulo Setoko.| 144428 |...do..... ye dGeeene 226 | 242 50 | 52 Aisa t 9.7 LD yoy sey > Ce ee ee eee Goris so: 144499 | Male....|...do..... 247 241 51 | 56.5 | 26 9.3 BMS ISDETS men ke eae san do.......| 144430 | Female .|...do..... 135 99 RAMs Roeseces 1558 5=s15. 8 1 “ Collector’s measurements. AONYX CINEREA (Illiger). 1815. Lutra cinerea Ivuicer, “Abh. Akad. Berlin, 1811, p. 99, 1815.” (Type- locality near Batavia, Java.) Skin and skull of an adult female, Cat. No. 144434, U.S.N.M., from Pulo Setoko. The size of the teeth in the specimens of clawless 4 Measurements in parentheses are those of the type of Mus firmus. 486 PROCEEDINGS OF THE NATIONAL MUSEUM. VOL. XXXVI. otters in the U. S. National Museum is quite variable. The above specimen and one from Great Karimon Island, Cat. No. 122840, U.S.N.M., have remarkably large and heavy teeth as compared with a skull from Tapanuli Bay, Sumatra, and a skull from northern Borneo. A very young,skull from Pulo Sebang, Rhio Linga Archi- pelago, has small teeth about like those of the Sumatran skull. The significance of this variation in the size of teeth is not clear. It is not sexual. A similar variation in the shape of the ascending ramus of the mandible is seen, no two of them being exactly alike. See Plate 39. Measurements.—Head and body, 495 mm.; tail, 290; hind foot, 95; weight, 6 pounds (2.7 kilos); basal length of skull, 82; zygomatic width, 57.5; maxillary tooth row (alveoli), 29. “The morning I left Setoko the natives brought me a fine female elawless otter which they had hit over the head with a paddle while swimming across the slat [strait]. One day a large otter swam across the s’lat close ahead of the schooner, but my men were too slow for him, -** * * They doubtless feed upon shell fish, among other things, and I know they.eat crabs.” W. L. Axssorr. ARCTOGALIDIA FUSCA Miller. 1906. Arctogalidia fusca MiLtErR, Proc. U. S. Nat. Mus., XXXI, No. 1485, p. 269, Sept. 11, 1906. (Type-locality, Pulo Kundur.) An immature male from Pulo Bulan, Cat. No. 144420, U.S.N.M., is indistinguishable from Arctogalidia fusca Miller. External measurements by collector: Head and body, 475 mm.; tail, 545; hind foot, 90; weight, 1.6 kilos (34$ pounds). Cranial meas- urements: Greatest length, 99.4; basal length, 93.5; zygomatic breadth, 53; postorbital constriction, 19; width of brain-case above roots of zygomata, 33; maxillary tooth row (alveoli), 34.6. GALEOPTERUS CHOMBOLIS, new species. Type.—Skin and skull of adult female, Cat. No. 144375, U.S.N.M., collected on Pulo Jombol, Rhio-Linga Archipelago, March 3, 1907, by Dr. W. L. Abbott. Original number, 5091. Diagnostic characters—A medium-sized member of the genus, closely related to Galeopterus tuancus (Miller) ,* but having wider zygomata, more inflated mastoids, and smaller first upper incisors. Color.—The color of the type and two adult female paratypes differs in no essential respects from that of flying-lemurs in the gray pelage phase from the Malay Peninsula, being, perhaps, a trifle paler in general color effect. Another paratype, an adult male, is in the “ved” phase, having the general color effect of cinnamon-rufous, 4 Smiths. Misc. Coll., XLV, p. 58, Nov. 6, 1903. _ no. 1684. MAMMALS OF THE RHIO-LINGA ARCHIPELAGO—LYON. 487 very light and clear on the under parts, darker and mixed with black- ish above. The usual white flecks are found on the feet and legs and a few on the back. Skull and teeth—The skull and teeth of Galeopterus chombolis are very similar to those of G. ¢wancus in general appearance. There are, however, several minor constant differences found in the animal from Jombol. The zygomata are wider, the lateral area of the mastoid much greater, the first upper incisor distinctly smaller, the third upper incisor somewhat larger, and the notch at the superior end of anterior border of premaxilla much larger and more angular. Measurements.—See table below. | Specimens examined.—One adult male and 3 adult females, all from Pulo Jombol. Remarks.—Aside from G. tuancus, the only other species with which G@. chombolis needs to be compared is @. tellonis (liyon).* Unfortunately, I have not been able to make a direct comparison between the two species, but the description of G@. tedlonis shows that the mastoid inflation is even less than in G. tuancus, and consequently much less in G. chombolis. Measurements of Galeopterus. a S . ee pate = 3 ey = a 8 5a Sues Ea = a 2S © oak) | oO =) Le] 3 a> | nm 5 2 ir) Pie qd po bape 2 salda| F |g8|s¢ Name. Locality. | © Sex. Age. 2 Seiad] & | S8/)24 es A q 3 | 2S see a Venpse 3 3) iu) EAE a | es o Z oO i=} S = iS on 8 o & eae deed oe $ & | © an 3 [os] heed = is — ) ope |e ps Ay Ee ies mm.|mm.| mm. | mm.| mm. | mm. | mm G. tuancus ......- Pulo Tu- | 114375 | Female..| Adult...| 385 | 235 | 62 68.4 | 44.3 | 19 32. 4 anku. | G. chombolis ..... Pulo Jom- | 144372 | Male....|...do..... 370 | 220 | 60.5 | 65.9 | 42 16.6 | 30.4 bol. Oneness ee ae eed Once. | 144o7o) |eMemAlen |" d0s.~ 22 390 | 260 | 62 69.7 | 42.9 | 18.7 31.8 PDOs 520. sos 4] 520052. 3(1443750|.. do... Beds eee 400 250 | 61 69.3 | 43.4 | 17.8 32. 3 Dar ae ace S2adOre 2-22) T4487'7%¢ | dor 25-7 bAdO25% 370 | 250 | 61 68.1 | 41 18 31 “ Collector's measurements. » Measured by writer from dried skin. ¢ Type. CYNOPTERUS MONTANOI Robin. 1881. Cynopterus montanoi Rosin, Bull. Soc. Phil. Paris, 7th ser., V, p. 90. (fype-locality, Malacca.) One specimen each from Pulo Bulan ‘ind Pulo Jombol, and two from Setoko. ‘This species appears to be widely distributed through- out the archipelago. For PICASUPEMIGN is, see table, page 488. ——— —! > ha ee “Ann. Mag. Nat. Hist., Sth ser., I, p. 189, February, 1908. 488 PROCEEDINGS OF THE NATIONAL MUSEUM. VOL. XXXVI. - EMBALLONURA PENINSULARIS Miller. 1898. Hmballonura peninsularis MILLER, Proc. Acad. Nat. Sci. Philadelphia, 1898, p. 328, July 25, 1898. (Type-locality, Trong, Lower Siam.) - Four specimens from Pulo Bulan. For measurements see table, below. This species is probably identical with Hmballonura monticola Temminck.*. Temminck’s standard of measurement was probably the pied du roi. His 1 inch and 7 lines as length of forearm then equals nearly 43 mm. Mr. Miller in describing 2’. peninsularis con- verted 1 inch and 7 lines on the basis of the English inch, which gave the length of forearm as 40 mm. and made all the other measure- ments correspondingly smaller. In the absence of specimens from Java for actual comparison, I have used the name peninsularis. MYOTIS MURICOLA (Gray). 1841. Vespertilio muricola Hopeson, Journ. Asiat. Soc. Bengal, X, p. 908 (Nomen nudum) 1846. Vespertilio muricola Gray, Cat. Spec. Draw. Mamm. Birds Nepal and 'Thibet, presented by Hodgson to Brit. Mus., p. 4. One specimen, an adult female, preserved in alcohol, from Pulo Setoko. For measurements see table below. Measurements of bats, = 2 ee Z s dja |58 Z E |e. | 38 Locality ~ | Sex and age. | 5 | 23 | Se Name. socality. _ ex and age. | 4 ib he > = 8 5 So u% | ag | 4 < a | ¢ 4 | €eres ose EA ae Gal ea ae f=) gag eee | a | v i) iS | o) oS u as Pre. ica) a & ial oe ) = mm.|mm.| mm. \mm.| mm. | mm. | mm. | mm. Cynopterus montanoi, Pulo Bulan.| 144417 | Male adult..) 82 7 | 61 23 | 13 15 28.6'|° 9.5 DOS faowece esse | PuloJombol | 144381 | Female adult} 75 7 | 59 23 | 13 15.5 | 26.6 | 8.9 DOS soos5her ae | Pulo Setoko.|. 144431 | Male adult..| 84} 11 | 64 24 | 13 15 30 9.5 DO ase kobe seal dOsssse5 144432 |... _- dors ot 83 | 10 | 62 24/14 | 14.5 | 28.5) 9.2 Emballonura penin- | Pulo Bulan.| 144413 |_.... dOssce.: 40 | 11 | 43 1607) | 98 eee sularis. | DORRL Aes cones once GO-c. <= 144414 |__... OURS seer 40} 11} 41 LS ew 10: | Scie DO Seat el eres Go:: ¥s:8 144415 | Femaleadult| 43 | 10 | 42 150) 725 | VOL sera ee DOr lees eee soe (| P44416 0). dOzs=nse 45 | 12 | 44 aby ss 12 13.9 | 4.6 Myotis muricola. . - | Pulo Setoko.| 144433 |..... dost 40 | 34 7 11 8.7 | 5.3 MACACA FASCICULARIS (Raffles). 1822. Simia fascicularis RAFFLES, Trans. Linn. Soc. London, XIII, p. 246, 1822. (Type-locality, Sumatra.) Skin and skull of an adult male, Cat. No. 144419, U.S.N.M., col- lected on Pulo Bulan. This specimen is grayer and less reddish than the majority of examples of this species. 4 Tydschr. Natuur. Gesch. Physiol. Leiden, V, p. 25, 1838, type-locality, Java. no. 1684. MAMMALS OF THE RHIO-LINGA ARCHIPELAGO—LYON. 489 Measurements by collector: Head and body, 472 mm.; tail, 530; hind foot, 126; weight, 10% pounds, 4.9 kilos. Cranial measure- ments: Basal length, 83; zygomatic width, 75.3; maxillary tooth row (alveoli), 35.6. PRESBYTIS CRISTATA (Raffles). . 1822. Simia cristata RAFFLES, Trans. Linn. Soc. London, XITI, p. 244, 1822. (Type-locality, Sumatra. ) Skin and skull of an adult male, Cat. No. 144871, U.S.N.M., col- lected on Pulo Jombol. Measurements by collector: Head and body, 510 mm.; tail, 660; hind foot, 151 ; weight, 15 pounds, 6.8 kilos. Cranial measurements: Basal length, 66.8; zygomatic width, 71.4; maxillary tooth row (alveoli), 32. LIST OF MAMMALS OF THE RHIO-LINGA ARCHIPELAGO, The name of each species is followed by the names of the islands on which it occurs. When the species is not represented by actual speci- mens in the U. S. National Museum collection, but is noted as occur- ring on certain islands by Dr. W. L. Abbott or Mr. C. Boden Kloss, the name of the island is printed in italics. Manis javanica. Bulan, Aundur, Penuba, Sanglar, Sinkep. Tragulus flavicollis. Sugi. Tragulus formosus. Bintang. Tragulus sp., kanchil group. Batam, Little Karimon ?, Penuba. Tragulus lutescens. Sugi Bawa, Jani. Tragulus sp., napu group Durian ?, Little NKarimon ?, Moro Kechil, Penuba. Tragulus nigricollis. Sinkep. Tragulus nigrocinctus. Wuandur, Great Karimon. Tragulus perflavus. Batam, Bulan, Galang, Setoko. Tragulus pretiellus. Bakong, Sebang. Tragulus pretiosus. Linga. Tragulus rubeus. Bintang. Tragulus subrufus. Linga, Sinkep. Sus oi. Batam, Kundur, Ungar. Sus rhionis. Bakong, Batam, Durian, Great Karimon, Jombol, Little Karimon, Moro Kechil, Sanglar, Sugi, Sugi Bawa, Ungar. Sus vittatus. Penjait Layer. (Some of the islands listed under Sus rhionis, from which there are no specimens, may possibly have on them the present species instead of S. rhionis.) Ratufa bulana. Bulan. Ratufa carimonensis. Great Karimon. 490 PROCEEDINGS OF THE NATIONAL MUSEUM. VOL. XXXVI. Ratufa condurensis. Wundur. Ratufa confinis. Sinkep. Ratufa conspicua. Bintang. Ratufa insignis. Sugi. Ratufa notabilis. - Linga. Sciurus carimonensis. Great Karimon. Sciurus condurensis. WKundur. Sciurus peninsularis. Batam, Bintang, Bulan, Little Karimon, Linga, Penuba, Sanglar, Sebang, Sinkep, Sugi. Sciurus tenuis. Batam, Linga. Rhinosciurus laticaudatus. Linga. Sciuropterus amenus. Wundur. Nannosciurus pulcher. Sinkep. Mus asper. Setoko. Mus batamanus. Batam. Mus chombolis. Jombol. . Mus concolor. Batam. Mus firmus. Bakong, Batam, Great Karimon, Linga, Moro Besar, Sebang, Setoko, Sugi, Sugi Bawa. Mus fremens. Winga, Sinkep. Mus lingensis. Bakong, Batam, Bintang, Great Karimon, Jombol, Linga, Moro Besar, Moro Kechil, Penuba, Sebang, Sin- kep, Sugi, Sugi Bawa. Mus “ vattus.’ Bakong, Batam, Great Karimon, Kundur, Moro Kechil, Sugi, Sugi Bawa. Felis “ tigris.” Bintang, Penjait Layer, Setoko., Paradoxurus brunneipes. Wandur. Paradoxurus “ hermaphroditus.” Batam. Arctogalidia fusca. Bintang, Bulan, Kundur. Arctogalidia simplex. Batam, Linga, Sinkep. Viverra tangalunga. Bintang, Linga, Aundur, Arctictis binturong. Bintang, Kundur. Aonyx cinerea. Great Karimon, Sebang, Setoko. Tupaia castanea. Bintang. i Tupaia ferruginea batamana. Batam. Tupaia malaccana. Linga, Sinkep. Tupaia pheura. Sinkep. Tupaia tana. Linga. Galeopterus chombolis. Jombol. Galeopterus temminchii. Bakong, Batam, Bintang, Great Kari- mon, Kundur, Penuba, Sebang, Sug. Pteropus vampyrus malaccensis. Linga. Cynopterus montanot. Bulan, Jombol, Kundur, Penuba, Sanglar. Setoko, Sugi. no. 1684. MAMMALS OF THE RHIO-LINGA ARCHIPELAGO—LYON. 49] Emballonura peninsularis. Bintang, Bulan, Karimon Anak, Sanglar. Myotis muricola. Setoko. Macaca fascicularis. Bakong, Batam, Bintang, Bulan, Durian, Great Karimon, Aundur, Linga, Moro Kechil, Sebang, Sug. Macaca nemestrina. Batam. Presbytis cana. Batam, Kundur. Presbytis cristata. Bakong, Batam, Bintang, Jombol, Linga, Sebang, Sugi. Presbytis rhionis. Bintang. EXPLANATION OF PLATE 39. View of under side of skulls and left half of mandibles of Aonyx cinerea, about ? natural size. Fig. 1. Cat. No, 144434, U.S.N.M., adult female, Pulo Setoko, Rhio-Ling: Archipelago. 2. Cat. No. 114466, U.S.N.M., adult female, Tapanuli Bay, west coast of _ Sumatra. 3. Cat. No. 34904, U.S.N.M., Kinabatagan River, British North Borneo. 4, Cat. No, 122840, U.S.N.M., Great Karimon, Rhio-Linga Archipelago. [Notr.—Since the page proofs of this paper have been made up, Messrs. Old- field Thomas and R. C. Wroughton have published in the Annals and Magazine of Natural History, eighth series, Volume III, pages 439-441, May, 1909, descrip- tions of the following six new mammals from the Rhio Archipelago: Presbytis cristata pullata, p. 489, Batam, Bintang. Sciurus vittatus nesiotes, p. 489, Batam. Sciurus seimundi, p. 440, Kundur. Rhinosciurus leo rhionis, p. 440, Karimon, Kundur, Batam, Bintang. Mus rattus rhionis, p. 441, Bintang, Batam. Sus anderson, p. 441, Batam.—M. W. L., Jr.] Msi at . NATIONAL MUSEUM PROCEEDINGS, VOL. XXXVI PL. 39 SKULLS OF CLAWLESS OTTERS. FOR EXPLANATION OF PLATE SEE PAGE 491. REVISION OF THE CRINOID FAMILY COMASTERIDA, WITH DESCRIPTIONS OF NEW GENERA AND SPECIES. By Austin Hoparr CriarK, Collaborator, Division of Marine Invertebrates, U. S. National Musewm. The work of the steamer A/batross, of the United States Bureau of Fisheries, especially within the last two years, has resulted in the accumulation of a magnificent collection of comasterid material, practically every known and numerous heretofore unknown species being represented. This has been studied in connection with the remarkably comprehensive collection belonging to the zoological museum of the University of Copenhagen (previously studied by Drs. C. F. Liitken and P. H. Carpenter), for the privilege of exam- ining which I am indebted to the generosity of my friend, Dr. Th. Mortensen; with the very fine collection of Japanese comasterids deposited in the U. S. National Museum by Mr. Frank Springer; with the collection made by the German steamer (azelle in Aus- tralia, sent to me for study through the kindness of Drs. W. Weltner and R. Hartmeyer, and with the collections of a number of Amer- ican museums. Still other collections have been examined at differ- ent times, and the notes made on them have proved of considerable value, the most important of these being the collection at the Museum of Comparative Zoology, which contains specimens identi- fied by Carpenter, and that of the Boston Society of Natural History. After the completion of the work on this “ revision,” the authori- ties of the Indian Museum at Calcutta, through the superintendent, Dr. N. Annandale, sent me the specimens collected by the steamer Investigator, which proved to be a collection of more than usual interest. It was with considerable gratification that I found, after a critical study of the 7nvestigator material, no changes of any kind were necessary, and I was thereby induced to publish the “ revision ” in its present form, believing that, if such a large collection as that of the Znvestigator did not alter in any way the general scheme, there was a reasonable hope of at least as long a life as is enjoyed by most “ revisions.” PROCEEDINGS U.S. NATIONAL Museum, VoL. XXXVI—No. 1685. 493 494 PROCEEDINGS OF THE NATIONAL MUSEUM. VOL. XXXVI. The history of the family Comasteride may be said to date from that most excellent memoir of Dr. P. H. Carpenter on the genus Actinometra. In this memoir he gives a detailed account of the systematic treatment of the various species of the genus by previous authors, and ably reduces to order a systematic chaos scarcely sur- passed in the whole subject of zoology; for even so great a zoologist as Prof. Johannes Miiller, in the only monograph then published on the unstalked crinoids, had placed a single species (under four dif- ferent specific names) under the subgenus Actinometra among the exocyclic forms (twice) ; under the subgenus A/ecto among the endo- eyclic forms, and again under the subgenerically incertw sedis, in the heterogeneous group Comatula. Since 1879 the genus Actinometra has been accepted in the sense in which it was used by Carpenter. Nine years afterwards he split it up into eight specific groups, distributing these among four “ series,” and this arrangement has been used ever since. Shortly afterwards the genus was raised to family rank, on a par with the family “An- tedonide,” covering the genus “Antedon” of Doctor Carpenter. This I have recently shown to be an unnatural division. In the course of my work I found the genus Actinometra becoming some- what unwieldy, and I accordingly split it in two sections, one small and one large; but with the accession of new material the large divi- sion proved not to be natural, and I split that into two parts. Enormous collections from the Philippine Islands having been re- ceived, a still further change was seen to be necessary, and the last of the divisions created was shattered into three fragments. I dis- carded the appropriate and euphonious name Actinometra proposed by Professor Miiller in favor of Comatula of Lamarck, of which it is a pure synonym, with the same type. While the name Comatulide was first employed to cover the family in place of Actinometride (not available because of the disuse of Actinometra), I soon found that confusion with the Comatulade of Fleming (1828) and Comatu- lidee of d’Orbigny (18: 52) and succeeding authors, with a more or less comprehensive range of meaning, but never so restricted as to cover the “Actinometridx ” alone, made a change desirable, and I there- fore substituted ‘“* Comasteride,” the name being derived from that of the next oldest genus. The most important discovery made in regard to the Comasteride since the publication of Carpenter’s memoir in 1879 is that of Mr. Frank Springer, who in 1903 described and figured a strongly de- veloped ambulacral plating on the arms and pinnules of a new species from the Tortugas. I have since found these plates to be universally present in the species of the “ Fimbriata group” from the West Indies, well developed even in “Actinometra” lineata, in which I de- tected it in some of the Challenger specimens previously examined by —_ No. 1685. REVISION OF CRINOID FAMILY COMASTERIDAI—CLARK. 495 Carpenter. These plates appear to represent the side plates of the Pentacrinitide, Tropiometridx, Thalassometride, Antedonide, etc., although performing the function of both side and covering plates. The latter are phylogenetically more advanced structures; whereas side plates are phylogenetically the thin produced ventrolateral bor- der of the pinnulars and brachials which has become separated from the parent ossicle by suture, so covering plates are the produced inner distal angles of the side plates which have become secondarily sep- arated off. KEY TO THE GENERA OF THE COMASTERIDA, a’, Six pinnules following the first pair absent; mouth always central. (1) COMATILIA. a’, All pinnules present; mouth usually more or less eccentric. bd’. IBr: and and first two joints after each axillary united by syzygy. (2) CoMATULA. b?. IBr; and: and first two joints after the first axillary united by synarthry. c’. Cirri present. ad’. Cirri without dorsal spines (ten arms). e’. First two pinnules much stouter than the succeeding; cirri long, slender ands munrerots)(CXay))) 2-2 ee ee (3) COMINIA, e’. First two pinnules more slender than the succeeding; cirri short and stout, few in number (to X XVII) _--_-_____- (4) COMACTINIA. .@. Cirri with dorsal spines or projections (ten or more arms). e’. Distal pinnules exceedingly slender with greatly elongated joints which have expanded articulations; ten arms_(5) LEPTONEMASTER. e’. Distal pinnules comparatively stout, the joints rarely over twice as long as broad, the articulations not expanded. f’. Ten arms; synarthrial tubercles prominent; pinnule joints with the distal ventro-lateral angle produced_________ (6) COMISSIA, f. More than ten arms; synarthrial tubercles not developed; no pro- duction of the distal ventro-lateral angles of the pinnule joints. g. First brachial bearing a pinnule on arms arising from a IIBr or subsequent axillary; a syzygy between the second and third brachials. h*. Brachials in distal half of arm exceedingly short, almost dis- cola sambulaeras naked! =. 023 (7) CAPILLASTER. h*. Brachials in distal half of arm triangular or very obliquely wedge-shaped, nearly or quite as long as broad; pinnule ambulacra with large side plates__________ (8) NEMASTER. g°. First brachial never bearing a pinnule. h*. All division series 2; a syzygy between the first two brach- VEILS Sy SORA Oe ly A Se ce (9) CoMATELLA. h*®. Several or all of the division series 4 (8+4). i’. Proximal pinnules more slender than the succeeding; combs occur at intervals on the distal pinnules__(10) ComMasTeEr. #. Proximal pinnules stouter than the succeeding; no combs on tine: distal, pimnnules= = "=e (11) CoMANTHUS. ce’. Cirri absent; centro-dorsal a thin pentagonal or stellate plate. ad’. First brachial bearing a pinnule; first syzygy between the first two | Urez KON EWS te ee OE ee ee ee ee (7) CAPILLASTER. 496 PROCEEDINGS OF THE NATIONAL MUSEUM. VoL. XXXVI. d’, First brachial never bearing a pinnule, e’, Proximal pinnules more slender than the succeeding; terminal comb short, with long curved teeth, appearing at intervals along the distal “pinnules 2) ?iJ 0S aoe Se ee (10) COMASTER. e*, Proximal pinnules stouter than the succeeding; terminal comb long with short rounded teeth, confined to the proximal pinnules. (11) CoMANTHUS. SUPPLEMENTARY KEY TO GENERA CONTAINING SPECIES WITH TEN ARMS ONLY. a’. Six pinnules following the first pair absent; large side plates developed along the pinnule ambulacra ; opposing spine forked or branched. (1) CoMATILIA, a*, All pinnules present; no ambulacral plating; opposing spine single. bd’. [Br and 2 and first two brachials united by syzygy____---- (2) COMATULA. b°. IBr: and » and first two brachials united by synarthry. ce’. Cirri without dorsal spines or projections. d’. First two pinnules much stouter than the succeeding; cirri long, Slender] and numerous (NW) ae ee eee (3) CoMINTA. @d, First two pinnules more slender than the succeeding; cirri short and Stout, few oi UME GEO; exe Vel)) ee eee (4) COMACTINIA. ¢. Cirri with dorsal spines or projections. d@. Distal cirrus joints (except the penultimate) considerably longer than broad; synarthrial tubercles not developed; distal pinnules exceed- ingly Slender, with greatly elongated joints which have expanded ATCICUIATIONS = 22 22 ee Se ees (5) LeEPTONEMASTER. @. Distal cirrus joints considerably broader than long; synarthrial tubercles prominent; distal pinnules comparatively stout, the joints rarely over twice as long as broad, the articulations not ex- pand edis+ Sot see Ss he Sees eee eee ee Ae (6) COoMISSIA. SUPPLEMENTARY KEY TO GENERA CONTAINING MULTIBRACHIATE SPECIES. a’. [Bri and 2 and first two joints beyond each axillary united by syzygy. (2) COMATULA. a’. IBr: and » and first two joints beyond the first axillary united by synarthry. b*. First brachial bearing a pinnule; first brachial syzygy between the second and third brachials. c. Brachials in distal half of arm exceedingly short, almost discoidal; EER OARONU UE: KGd ig: lm aye): 2) 0 Pemeeintneny eas Laie ee, Ae a Ce eee (7) CAPILLASTER. c. Brachials in distal half of arm triangular or very obliquely wedge- shaped, nearly or quite as long as: broad; pinnule ambulacra with large. Bide platen sao = ee eee ake tee ee (8) NEMASTER. b°. First brachial never bearing a pinnule. c’. Division series all 2; first brachial syzygy between the first two brachials on all. but the: otltermost.anms= 2-2 ee ee (9) COMATELLA. c’. Several or all of the division series 4 (3+4). d‘. Proximal pinnules more slender than the succeeding; combs occur at intervals along the distal pinnules___..___________ (10) COMASTER. d’. Proximal pinnules stouter than the succeeding; no combs on the distal, pinnulées = 2 ees ea a oc ee ~(11) CoMANTHUS. No. 1685. REVISION OF CRINOID FAMILY COMASTERID CLARK. 49% 1. Genus COMATILIA A. H. Clark. 1909. Comatilia A. H. CiarK, Proc. U. S. Nat. Mus., XXXVI, p. 365. Genotype.—Comatilia iridometriformis A. H. Clark (new species). Distribution Only known from between the Bahama Islands and Cape Fear, North Carolina. Depth—Two hundred and eighty fathoms. 2. Genus COMATULA Lamarck (emended). 1758. Asterias (part) Linnzeus, Syst. Nat., 10th ed., II, p. 663. 1772. Asteria (part) Brinwnicu, Zodlogia fundamenta, p. 230 (emendattion). [1812. Comatule (part) LAmMARcK, Extrait du cours de zodlogie du mus. d@hist nat. sur les animaux sans vertébres, p. 85 (no definition) ]. 1816. Comatula LAMARCK, Hist. nat. des animaux sans vertébres, IT, p. 530, emended 1908—A. H. Crark,, Proc. U. S. Nat. Mus., XX XIII, p. 685. 1841. Actinometra J. Mtiuer, Archiv fiir Naturgesch., 1841, I, p. 140. Genotype-—Comatula solaris Lamarck (new species). Distribiition—Northern Australia to the Mergui Archipelago, China, and the Philippine Islands, ? Madagascar, ? Society Islands. Depth—tittoral and sublittoral. 3. COMINIA, new genus. 1908. Comanthus (part) A. H. CrarK, Proc. Biol. Soe. Washington, XXI, p. 220. Genotype—Comanthus decameros A. H. Clark, 1908. Description —Centro-dorsal discoidal, bearing numerous marginal cirri in roughly three irregular and crowded more or less alternating rows. ; . Cirri XL, 16-17; first joint very short, second slightly longer than broad to about twice as long as broad, third-sixth two and one-half to three times as long as broad, the following decreasing in length, the last two being squarish ; opposing spine represented by a low tubercle ; no dorsal spines or projections; terminal claw about as long as the penultimate joint, moderately stout and moderately curved. Ends of the basal rays very prominent in the angles of the calyx; radials concealed; IBr, short, oblong, widely separated laterally ; IBr, (ax) broadly pentagonal, about twice as broad as long. Ten arms; first seven or eight brachials slightly wedge-shaped, then triangular about as broad as long. Arms rugged and tubercular basally, but not enlarged or swollen. P, long, stout basally but becoming slender and flagellate distally ; P, slightly smaller and slightly less stout basally; following pinnules shorter and more slender, the distal pinnules increasing to about the length of P,; comb confined to PP,, ., and ,. Distribution—Only known from the Korean Straits, Depth.—One hundred and seventy fathoms. Proc. N. M. vol, xxxvi—09 32 498 PROCEEDINGS OF THE NATIONAL MUSEUM. VoL. XXXVI. 4. COMACTINIA, new genus. 1840. Comatula (part) J. MULLER, Archiv fiir Naturgesch., 1840, I, p. 311. 1841. Alccto (part) J. Mtxiuer, Archiv fiir Naturgesch., 1841, I, p. 148. 1849. Comatula (Alecto) J. MULLER, Abhandl. d. k. preuss. Akad., 1847, p. 250. : 1878. Antedon (part) PourTaLis, Bull. Mus. Comp. Zool., V, p. 214. 1881. Actinometra (part) P. H. CarPentrr, Bull. Mus. Comp. Zool., IX, No. 4, p. 154. 1908. Comaster (part) A. H. CuarK, Proc. U. S. Nat. Mus., XX XIII, p. 685. 1908. Phanogenia (part) A. H. Ciark, Proc. U. S. Nat. Mus., XXXYV, p. 124. Genotype.—Alecto echinoptera J. Miiller, 1841. Description.—Centro-dorsal rather large, discoidal, the cirri ar- ranged in a single marginal row. Cirri short and stout, IX—X XVII, 8-12; basal joints short, then two or three half again to twice as long as broad, the following de- creasing in length, being about as long as broad distally; opposing spine, small, erect, median in position; no dorsal spines-or projec- tions; terminal claw about as long as the antepenultimate joint (which is longer than the penultimate) stout and strongly curved basally, becoming slender and nearly straight distally. Ends of the basal rays visible in the interradial angles; radials concealed; IBr, very short and band-lke, closely united laterally ; IBr, (ax) triangular, usually about twice as broad as long; IBr, and first two brachials in lateral contact, though not laterally flattened. Ten arms; proximal brachials discoidal, or oblong, then becoming triangular, at first broader than long, later about as long as broad, and wedge-shaped terminally. Oral pinnules longer than, but not quite so stout as, those succeed- ing; middle pinnules with more or less developed spinous edges and dorsal processes on the joints. Distribution Caribbean Sea, northward to South Carolina and southward to Brazil. Depth—Sublittoral, and down to 262 fathoms. 5. LEPTONEMASTER, new genus. Genotype.—Leptonemaster venustus, new species. Description.—Centro-dorsal a thin flat disk; cirrus sockets in a single marginal row. Cirri XV—-XX, 12-15, long and slender; first joint short, second half again as broad as long to nearly square, third about twice as long as its terminal diameter, fourth the longest, two and one-half to three times as long as its proximal diameter, fifth a transition joint, not quite so long as the fourth, with a dark band about its center; following joints gradually decreasing in length, the antepe- nultimate being very slightly longer than broad, or squarish, and the No. 1685. REVISION OF CRINOID FAMILY COMASTERID.E CLARK. 499 penultimate squarish or not quite so long as broad; second to sixth joints slender, moderately constricted centrally (* dice-box shaped ”), with prominent articulations, rounded in cross-section, then becom- ing rather strongly compressed laterally (the distal portion of the cirrus therefore becoming broader in lateral view) and less and less “ dice-box shaped;” transition and following joints with a small, though prominent, sharp subterminal dorsal spine; opposing spine slightly marked, median, arising from the entire dorsal surface of the penultimate joint; terminal claw somewhat longer than the penultimate joint (about as long as the antepenultimate), moderately stout, and moderately curved, the curvature being strongest in the basal portion. Ends of the basal rays visible as rather prominent tubercles in the angles of the calyx; radials entirely hidden or slightly visible over the ends of the basal rays, separated distally; [Br, short, nearly four times as broad as long, the proximal edge convex, not in contact basally, rounded and widely free laterally, the sides of adjacent IBr, making with each other an angle of about 90°; IBr, (ax) triangular, the anterior angle somewhat produced, about one and one- half times as broad as long, the very short lateral edges making an obtuse angle with those of the IBr,. Ten arms; first seven brachials approximately oblong, then becom- ing obliquely wedge-shaped, and after the tenth triangular, about as long as broad, and terminally obliquely -wedge-shaped and longer than broad, with somewhat expanded articulations; after about the sixth the brachials develop strongly produced and overlapping distal ends. Syzygies occur between the third and fourth brachials, again between the tenth and eleventh to twelfth and thirteenth, and distally at intervals of three oblique muscular articulations. Disk naked; mouth and anal tube about equally eccentric. P, the stoutest, and much the longest, evenly tapering to a flagellate tip; P, considerably shorter and much more slender than P,; P, not much more than one-third, P, one-third the length of P,; fol- lowing pinnules increasing slowly in length, the distal being nearly as long as P,, with elongated joints which have expanded articula- tions, and spinous distal ends. Distribution —Caribbean coast of Central America, Gulf of Mex- ico, and northern coast of Cuba. . Depth.—F orty-two to 163 fathoms. LEPTONEMASTER VENUSTUS, new species. Centro-dorsal a thin flat disk, the small cirrus sockets arranged in a single crowded marginal row, usually five to each radial area, Cirri XV—XX, 12-15 (most commonly 13 or 14) 10 mm. long, as described above. 500 PROCEEDINGS OF THE NATIONAL MUSEUM. VOL. XXXVI. IBr series and calyx elements as described. Ten arms 70 mm. to 90 mm. long; first brachial short, slightly wedge-shaped, about three times as broad as the exterior length, en- tirely separated interiorly by the anterior apex of the IBr,, the interior edges diverging at an angle of approximately 90° or slightly less; second brachial irregularly quadrate, slightly larger than the first; third and fourth brachials (syzygial pair) oblong, about half again as broad as long; next three brachials approximately oblong, about twice as broad as long, then becoming obliquely wedge-shaped, and after about the tenth triangular, about as long as broad, further out on the arm becoming very obliquely wedge-shaped (almost trian- gular) about as long as broad, and in the terminal portion longer than broad. After about the sixth the brachials develop strongly produced and overlapping distal ends. P, 10 mm. long, with about thirty-five joints, moderately stout basally and evenly tapering; terminal comb with 13 to 15 teeth, pre- ceded by two or three more or less rudimentary; teeth spade-shaped or triangular, longer than broad, slightly longer than the lateral diameter of the joint which bears them, well separated, and incurved ; basal joints of the pinnule broader than long, the proportionate length gradually increasing, so that the joints from the middle on- ward are approximately squarish; the joints have prominent dorsal projections with the apex at the distal end, and strongly produced distal edges, these characters dying gradually away after about the middle of the pinnule; P, much more slender than P,, 7 mm. long, the joints after the fifth squarish; first two joints with strong dorsal processes or broad carinations, that of the second the stronger; fol- lowing joints with rounded dorsal processes and prominent distal edges; terminal comb rather long with sixteen fully developed and five or six smaller and more rounded teeth; teeth proportionately sheghtly longer and better developed than the teeth of P,; P, about 4+ mm. long, slender and delicate, the first two joints disproportion- ately large, about. half again as broad as long, the second with a much produced distal dorsal angle-or even distal half of the dorsal side; third joint squarish; following joints shghtly longer than broad; third and following joints as far as the comb, as in P,, with strongly produced coarsely spinous distal ends; comb as in P,; P, 3.5 mm. long, shightly more delicate than P,, with no enlargement of the two basal joints and no comb; first two joints short, third longer than broad, the following increasing slightly in length, being about half again as long as broad distally; third and following joints with pro- duced and coarsely spinous distal edges; P, similar to P,, 4 mm. long, with sixteen joints, but slightly stouter; following pinnules similar to P., increasing very gradually in length; distal pinnules 8 mm. to 9 mm. long, slender, -with about 21 joints, the first two not so long as No. 1685. REVISION OF CRINOID FAMILY COMASTERID®—CLARK. 50] broad, the third slightly longer than broad, the remainder becoming elongated and about three or four times as long as broad distally ; third and following joints with expanded articulations and coarsely spinous distal ends. Color (in spirits).—Brownish white, the perisome dark brown. Type.—Cat. No. 25457, U.S.N.M., from Grampus station 5104, off the west coast of Florida; 51 fathoms. 6. COMISSIA, new genus. 1908. Comaster (part) A. H. CLARK, Smiths, Miscell. Coll. (Quarterly Issue), tive ps 202: Genotype.—C omissia liitheni, new species. Description.—Centro-dorsal discoidal, the bare polar area broad and flat, the cirrus sockets arranged in two closely crowded alter- nating rows. Cirri XV-XXV, 16-24, resembling those of Capillaster; the fourth is a transition joint. Ends of the basal rays visible as prominent tubercles in the angles of the calyx; radials very slightly visible over the ends of the basal rays, or quite concealed; IBr, short and broad, closely united later- ally, more or less concealed by the centro-dorsal; IBr, (ax) triangu- lar, about twice as broad as long, free laterally; synarthrial tubercles prominent. Ten arms; first brachial short, slightly wedge-shaped, between three and four times as broad as long exteriorly, interiorly united; second brachial larger and much more obliquely wedge-shape:!: third and fourth (syzygial pair) somewhat longer interiorly than exte- riorly, about twice as broad as the interior length; following one or two brachials almost oblong, about three times as broad as long, then becoming triangular, about twice as broad as long, in the terminal part of the arm becoming very obliquely wedge-shaped, about as long as broad; brachials after the second with prominent and finely spinous distal ends and a very finely tubercular or spinous dorsal surface which in the terminal portion gradually become obsolete, so that the ends of the arms are practically smooth. Syzygies occur between the third and fourth brachials, again between the eleventh and twelfth to fourteenth and fifteenth, and distally at intervals of three oblique muscular articulations. Disk naked, or with small scattered calcareous granules; mouth subcentral; anal tube small and marginal. P, the longest; following pinnules decreasing gradually in length and slightly in stoutness to P,, which is less than half as long as P,, with somewhat less than half as many joints; following pinnules remaining similar for some time, then gradually becoming more slender and increasing in length to about the length of P, distally; 502 PROCEEDINGS OF THE NATIONAL MUSEUM. VOL. XXXVI. the joints of the middle and distal pinnules are slightly “ dicebox- shaped,” with a finely spinous surface and with the distal ends pro- duced ventrally into two long sharp spines, one on each side of the perisome ; this modification of the joints in the more proximal of the pinnules affects only the distal portion, but later encroaches more and more upon the proximal part, soon involving almost all of the joints. COMISSIA LUTKENI, new species. 1908. Comaster coppingeri (part) A. H. CLarK, Smiths. Miscell. Coll. (Quar- terly Issue), LII, p. 202 (ten-armed specimens). Centro-dorsal discoidal, the bare polar area broad and flat, 4 mm. or 5 mm. in diameter; cirrus sockets arranged in two closely crowded alternating rows. Cirri XV-XXV, 16-24 (usually 18-21) 7 mm. to 17 mm. long, comparatively small and rather stout; first joint over twice as broad as long, second and third nearly or quite as broad as long, fourth half again to nearly twice as long as broad, a transition joint, usually rather darker than the preceding, but light colored and with a pol- ished surface in the distal fourth; following joints decreasing in length, after the eighth being about twice as broad as long; occasion- ally the fifth is a transition joint instead of the fourth, in which case the two are about of the same size; fourth and following joints with the dorsal and dorso-lateral distal edge everted and finely spinous; this eversion of the distal edge of the joints gradually narrows ante- riorly, on the last two or three joints becoming merely a single blunt spine or tubercle; concurrently with its shortening, it gradually attains a crescentic form, so that in lateral view the joints from the fourth onward appear to be furnished with low dorsal spines which arise gradually from the whole dorsal surface, at first terminal, gradually becoming subterminal in position, and on the antepenul- timate joint almost median; opposing spine median, arising from the entire dorsal surface of the penultimate joint, short and blunt, reach- ing not more than one-third the distal diameter of that joint in height; terminal claw about as long as the penultimate joint, stout, and moderately curved. Post-radial elements as given in the generic description; the arms are 70 mm. to 75 mm. long. P, 12 mm. to 15 mm. long, slightly stouter than the succeeding, though not especially large, with about thirty-five joints, at first about twice as broad as long, very gradually becoming longer and about as long as broad after the twelfth or fifteenth; terminal comb prominent, arising abruptly, with sixteen teeth, bluntly triangular, nearly twice as long as broad at the base, basally in apposition, about as high as the transverse diameter of the joints which bear them, rather strongly recurved; P, similar, 10 mm. to 12 mm. long; P, No. 1685. REVISION OF CRINOID FAMILY COMASTERID®—CLARK. 503 similar, 8 mm. to 10 mm. long; P,6 mm. long; P,; and following pinnules 6 mm. long without combs, composed of sixteen joints, the first three not so long as broad, the remainder about as long as broad ; distally the pinnules gradually increase in length and become more slender, being distally 8 mm. long with twenty-three to twenty-five joints, the first two short, the third and following longer than broad, becoming about twice as long as broad in the outer portion. The lower pinnules have the corners of the joints considerably cut away as in /Zeliometra,; the joints of the middle and distal pinnules are slightly “ dice-box shaped ” with a finely spinous surface, and with the distal ends produced ventrally into two long, sharp spines, one on each side of the perisome; this modification of the joints in the more proximal of the middle pinnules affects only the distal portion, but later encroaches more and more upon the proximal part of the pinnules, soon involving almost all of the joints. Color (in spirits) —Bright yellow, the skeleton lighter. Type.—Cat. No. 25513, U.S.N.M., from Albatross station 5153; east of Port Dos Amigos, Tawi Tawi; 49 fathoms. % Genus CAPILEASTER A. H. Clark. 1758. Asterias (part) Linnaus, Syst. Nat., 10th ed., II, p. 663. 1772. Asteria (part) BrRUNNIcH, Zodlogia fundamenta, p. 230 (emendation). 1816. Comatula (part) LAMarRcK, Hist. nat. des animaux sans vertébres, EG p530; 1836. Comaster (part) L. Acassiz, Mém. Soc. de sci. nat. de Neuchatel, I, p. 193. 1841. Actinometra (part) J. MtLuer, Archiv fiir Naturgesch., 1841, I, p. 140. 1849. Comatula (Actinometra) (part) J. Mtuirer, Abhandl. d. k. preuss. Akad., 1847, p. 246. 1849. Comatula (Alecto) (part) J. Mtiier, Abhandl. d. k. preuss. Akad., 1847, p. 258. 1909. Capillaster A. H. CLARK, Proc. Biol. Soc. Washington, XXII, p. 87. Genotype.—Actinometra sentosa P. H. Carpenter, 1888. Distribution Madagascar to northern Australia, the Philippines, and Japan. Depth.—tLittoral and sub-littoral; rarely down to 160 fathoms. 8. NEMASTER, new genus. 1879. Antedon (part) RarupBun, Trans. Conn. Acad. Sci., V, p. 157. 1880. Actinometra (part) P. H. Carpenter, Journ. Linn. Soe. (Zool.), XV, p. 213. 1908. Comaster (part) A. H. CLarK, Proc. U. S. Nat. Mus., XX XIII, p. 685. Genotype.—Nemaster grandis, new species. Distribution Caribbean Sea and Atlantic coast of South America south to Bahia. 504 PROCEEDINGS OF THE NATIONAL MUSEUM, — you. xxxv1. Depth.—Littoral, and down to 194 fathoms. Diagnosis.—In general same as Capillaster; I1Br 4 (8-+4) ; T1Br 8 (243), or irregular; brachials at first oblong, then triangular, about as long as broad, wedge shaped and longer terminally; ter- minal comb usually repeated en inner side of proximal pinnules; side plates developed along the ambulacra. NEMASTER GRANDIS, new species. Centro-dorsal thick-discoidal, the polar area 5 mm. in diameter, deeply concave; cirrus sockets marginal, arranged in three closely crowded alternating rows. Cirri XXV-XXX, 30-85, about 40 mm. long, large and stout; first joint short, about three times as broad as long; following joints gradually increasing in length to the sixth or eighth, which, with the three following, is squarish, then gradually decreasing, the joints from the twelfth or fifteenth onward being about twice as broad as long, but the last two are almost square again; a transition joint occurs between the seventh and the twelfth, proximal to which the joints have a dull, finely pitted surface, distally a highly polished surface, the pits widely scattered or absent, and dorsal projections; transition joint not especially marked; joints proximal to the transi- tion joint with practically straight sides and no modification of the dorsal distal edge; transition and following joints with the distal dorsal edge projecting as a transverse ridge, coarsely dentate (usually tridentate), the ridge being equal in length (transversely) to about half the diameter of the joints; distally the ridge gradually narrows, becoming bidentate, and in the terminal four to seven joints resolves itself into a single spine, which on the antepenultimate becomes sub- terminal in position; all the transverse ridges appear as rather promi- nent spines in lateral view; opposing spine prominent, though short, rather stout, arising from the whole dorsal surface of the penulti- mate joint, about equal in length to one-third the diameter of that joint, the apex subterminal or submedian, the distal edge usually making much less of an angle with the transverse diameter of the joint than the proximal, giving the spine the appearance of leaning forward; terminal claw considerably longer than the penultimate joint, stout basally, slender distally, strongly curved proximally, but becoming nearly straight in the distal portion. Ends of the basal rays visible as low tubercles in the angles of the calyx, but with difficulty differentiated from the adjacent parts; radials concealed in the median line, but visible as a rather prominent triangle in the angles of the calyx, the apex of which separates the lower corners of the IBr,; [Br, oblong, rounded dorsally and laterally, about three times as broad as long, widely separated laterally; IBr, (ax) pentagonal, one-third to one-half again as broad as long, the No. 1685. REVISION OF Spelt FAMILY COMASTERID® LARK. 5Q5 lateral edges diverging distally and ‘about equal in length to those of the IBr,; I1 Br 4 (3-44) ; ; [11 Br 3 (2+3), the division series being separated by a distance about equal to the breadth of the I1Br series; if the full IIIBr series is not present, those on the exterior sides of the IBr series are most frequently absent, so that there is an approxi- mation toa 1, 2, 2, 1 arrangement. Twenty-four to thirty-one arms, about 200 mm. long; first brachial wedge-shaped, rather large, not quite half again as broad as the exterior length, almost entirely united interiorly; second and third brachials (syzygial pair) not quite so long as broad; next four brachials oblong, about twice as broad as long, then wedge-shaped, and after three or, four triangular, about as long as broad; in the terminal portion of the arms the brachials become wedge-shaped, nearly or quite twice as long as broad. The distal edges of the brachials project slightly and are beset with fine spines. Syzygies occur between the second and third brachials, again between the fourteenth and fifteenth to twenty-second and twenty-third (usually in the vicinity of the eighteenth) and distally at intervals of three oblique muscular articulations. Mouth marginal and radial; anus central; disk naked, about 30 mm. in diameter; side plates developed along the brachial and pin- nule ambulacra. Py 30 mm. to 35 mm. long, stout, much stouter than the succeeding pinnules, but tapering evenly to a slender and flagellate tip, with forty to forty-five joints, all of which are approximately squarish ; a slight prominence is visible on the dorsal side of the distal edge of the second joint, which rapidly becomes larger and increases in width on the succeeding joints, after about the seventh taking the form of a strong coarsely spinous eversion of the dorsal edge of the joints; this disappears near the proximal part of the distal comb; comb composed of fourteen teeth, arising abruptly; first tooth low and triangular; second oblong or slightly trapezoidal, usually slightly broader basally than high, the following becoming more obliquely trapezoidal and relatively somewhat higher, the terminal teeth being truncated-triangular; the more proximal teeth are not equal in height to more than three quarters of the lateral diameter of the joints which bear them, but the later teeth, owing to the dis- tal tapering of the pinnule, become about equal to the lateral diameter ; Pp about 25 mm. long, considerably less stout than Pp, but otherwise similar to it; P, about 20 mm. long, much less stout than Pp, similar to it; P, and following pinnules slender and delicate, about 10 mm. long; P, bears a comb distally, but the following pinnules are with- out combs; PP., , ,, and «a, », - have the first two joints dispropor- tionately large and produced dorsally into large carinate processes; distal pinnules slender, about 12 mm. long, with about twenty-five 506 PROCEEDINGS OF THE NATIONAL MUSEUM. VOL. XXXVI. joints, the first over twice as broad as long, the second about as long as broad, the third longer than broad, the remainder about half again as long as broad. The distal ends of the joints are slightly everted and finely spinous; the dorsal surface is beset with fine spines, and the last four joints bear long recurved spines. Type.—Cat. No. 25459, U.S.N.M., from Albatross station 2146; off Colon; 34 fathoms. 9. Genus COMATELLA A. H. Clark. 1874. Actinometra (part) LUTKEN, Mus. Godeffr. Cat., V, p. 190. 1908. Comaster (part) A. H. Criark, Proc. U. S. Nat. Mus., XX XIII, p. 685. 1908. Phanogenia (part) A. H. CLARK, Proc. U. S. Nat. Mus., XX XV, p. 124. 1908. Comatella A. H. CiarK, Smiths. Miscell. Coll. (Quarterly Issue), LIE p: 207. Genotype.—Actinometra nigra P. H. Carpenter, 1888. Distribution.—Ceylon to Fiji, Tonga, Samoa, and Japan; West Indies, St. Paul’s Rocks, Atlantic coasts of southern Europe and northwestern Africa. Depth—Imn the Indian and Pacific oceans, littoral, and down to 140 fathoms; in the Atlantic 73-830 fathoms. 10. Genus COMASTER L. Agassiz. 1816. Comatula (part) LAMARcK, Hist. nat. des animaux sans vertébres, TT, p. 530. ; 1836. Comaster L. AGAss1z, Mém. Soc. de Sci. nat. de Neuchatel, I, p. 193. 1841. Alecto (part) J. MULier, Arehiy fiir Naturgesch., 1841, I, p. 147. 1849. Comatula (part) J. MtLrer, Abhandl. d. k. preuss. Akad., 1847, p. 262. 1866. Phanogenia Lovin, Ofversight k. Vetensk.Akad. Férhandl., 1866, No. 95 p: 251: 1879. Actinometra (part) P. H. CARPENTER, Proc. Roy. Soc., XXVIII, p. 386. Genotype—Comatula multiradiata Lamarck, 1816 (not Asterias multiradiata Linneus) =Alecto multifida J. Miiller, 1841.4 Distribution.—N orthern Australia to Luzon and the Mergui Archi- pelago. Depth.—uittoral and sublittoral. ile Genus COMANTHUS A. H. Clark. 1816. Comatula (part) LAMARCK, Hist. nat. des animaux sans vertébres, II, De Dae 1841. Actinometra (part) J. MUtwier, Archiv fiir Naturgesch., 1841, I, p. 140. 1841. Comaster (part) J. Mttrer, Archiv fiir Naturgesch., 1841, I, p. 140. 1841. Alecto (part) J. MULier, Archiv fiir Naturgesch., 1841, I, p. 144. 1849. Comatula (Actinometra) (part) J. Mtwier, Abhandl. d. k. preuss. Akad., 1847, p. 256. 4Cf. Proc. Biol. Soc. Washington, XXII, p. 87. . No. 1685. REVISION OF CRINOID FAMILY COMASTERIDA—CLARK. 507 1849. Comatula (Alecto) (part) J. Mittier, Abhandl. d. k. preuss. Akad., 1847, p. 260. 1891. Goldfussia (not of de Castelnau, 1848) Norman, Ann. and Mag. Nat. Hist., [6] VII, p. 387. 1908. Phanogenia (part) A. H. CLark, Proc. U. 8. Nat. Mus., XXXV, p. 124. 1908. Comanthus A. H. CLuark, Proc. Biol. Soc., Washington, XXI, p. 220. Genoty pe-—Comanthus intricata A. H. Clark, 1908. Distribution —South Africa westward and northwestward, along the southern coast of Asia and the entire coast of Australia, through- out the East Indies, to southern Japan, the Kingsmill (Gilbert) Islands, Fiji, and Samoa.“ Depth.—Littoral and sublittoral. “Carpenter records C. rotalaria (“Actinometra parvicirra”) from Peru, in South America, and others have since accepted this record. This Peru is, how- ever, undoubtedly Peru or Francis Island, situated approximately in lat. 1° 30’ S., long. 176° 00’ E., in the Gilbert group, north of Fiji. sake 4 Wa 1 its ve ie Ne aes 7 eae 8 ve pees A NEW SQUIRREL FROM DIRECTION ISLAND, SOUTH CHINA SEA. By Marcus Warp Lyon, Jr., Second Assistant Curator, Division of Mammals, U. S. National Museum. On his way to Borneo, in 1907, Dr. W. L. Abbott stopped for a day at Direction Island, where he secured a single specimen of the new species of plantain squirrel described below. Direction Island, also called Pulo Mankotan and Pulo Pengiki Kichil (or Paneeky Ketchil), lies in the South China Sea in latitude 0° 14’ 39°” north and longitude 108° 1’ 53’’ east. Politically and geographically it is a member of the Tambelan group, of which it is the most south- eastern. An account of the mammals of this group and of some adjacent islands was published by Mr. Gerrit S. Miller, jr.,2 in 1900. Doctor Abbott says of the island: “ It is about three-fourths mile long by about one-fourth mile wide, and 500 to 600 feet high. The surface is rocky, but covered with trees except at the southwest corner, where a clearing has been made by the Orang Laut, who occasionally visit the island. Here a few cocoanuts have been planted, also some bananas and papaya. A number of squirrels were heard, but only two were seen, of which one was shot. No other mammal was seen, although there were doubtless rats. There were many white fruit pigeons. Turtles had been laying their eggs on the sand beach, where there were also many tracks of Varanus lizards.” SCIURUS DIRECTOR, new species. Type.—Skin and skull of an immature (large permanent upper premolar just displacing the milk tooth) male, Cat. No. 145392, U.S.N.M., collected on Direction Island, South China Sea, May 1, 1907, by Dr. W. L. Abbott. Original number, 5152. Diagnostic characters —A “red ”-bellied member of the vittatus group characterized by a more ruddy cast to the entire pelage than usual. 4@Mammals collected by Dr. W. L. Abbott on islands in the South China Sea, Proc. Wash. Acad. Sci., II, pp. 208-246, August 20, 1900. PRocEEDINGS U. S. NATIONAL MUSEUM, VOL. XXXVI—No. 1686. 509 510 PROCEEDINGS OF THE NATIONAL MUSEUM. VOL. XXXVI. Color.—Upper parts of head and body, a fine grizzle of black and ochraceous or ochraceous-buff, both colors about equally mixed, the ochraceous being somewhat paler over the shoulders than elsewhere, and darkest posteriorly and in the region of the thighs; upper sur- face of the tail very similar to upper surface of head and body in color, but the grizzling is coarser and in certain lights the tail ap- pears finely annulated; outer sides of thighs and arms similar to back, but with finer grizzling and the ochraceous predominating; upper surfaces of feet dull ochraceous darkened by the blackish bases of the hairs showing through; light side-stripe, about 55 by 5 mm., be- tween buff and ochraceous-buff; dark side-stripe, about 50 by 7 mm., blackish finely sprinkled with ochraceous-buff; underparts and inner side of fore and hind legs a color something between Ridgway’s ochraceous-buff and ochraceous-rufous ; underside of tail a very coarse grizzle of blackish and ochraceous, the latter color predominating in the middle line; inner side of ears and an orbital ring, the lower half of which is most pronounced, ochraceous-buff ; outer side of ears similar to adjacent parts of head; cheeks and base of whiskers similar to rest of head, but grizzle very fine and the ochraceous-buff pre- dominating. Skull and teeth—These show no special peculiarities; the audital bulle and teeth, however, are smaller than they are in the majority of species of squirrels of the vittatus group. Measurements.—External measurements taken by collector: Head and body, 190 mm.; tail vertebrae, 182; hind foot, with claws, 49. Cranial measurements: Greatest length, 46.4; basal length, 39; zygo- matic breadth, 26.7; interorbital constriction, 16; mandible, front of symphysis to back of condyle, 29.5; maxillary toothrow (alveoli), 9.2; mandibular toothrow (alveoli), 8.7. Specimens examined.—One, the type. Remarks——Compared with its geographical neighbor, Sciurus ab-— bottii Miller, of Big Tambelan Island, S. director is conspicuously more ruddy throughout, being ochraceous or ochraceous-rufous where S. abbottii is only buffy or ochraceous-buff. Among the forms of the S. vittatus group of squirrels in the National Museum S. tedongus Lyon.’ from the island of Banka, most nearly resembles S. director, but is less ruddy, except on the belly, which has about the same color in the two species. @Proc. Wash. Acad. Sci., II, p. 224, August 20, 1900. bProc. U. S. Nat. Mus., XXXI, p. 591, December 18, 1906. THE THORAX OF INSECTS AND THE ARTICULATION OF THE WINGS. By Roperr Evans Snoperass, Of the Bureau of Entomology, U. 8S. Department of Agriculture. I. INTRODUCTION. This paper is an attempt to show the unity of thoracic structure that prevails throughout all the orders of insects. It is hoped that it will be of special service to systematists in entomology and that it will meet with approval from students of morphology. The material on which the paper is based was all drawn from the U. S. National Museum and the dissections have been deposited in the museum. The work has been done under the direction of Dr. A. D. Hopkins, of the U. S. Bureau of Entomology, and has grown from an attempt to determine thoracic homologies in the Coleoptera, especially in the family Scolytide. It is published by the approval of Dr. L. O. Howard, chief of the bureau, as a contribution from the office of Forest Insect Investigations. The author is indebted to Doctor Hop- kins not only for the opportunity of carrying on the work but also for a great deal of help in doing it and for the verification of observa- tions. Assistance has also been received from other members of the entomological staff of the bureau, among whom are Mr. Nathan Banks, Mr. A. N. Caudell, Mr. D. W. Coquillett, Mr. R. P. Currie, Dr. H. G. Dyar, Mr. Otto Heidemann, Mr. E. A. Schwarz, and Mr. H. S. Barber and also Mr. J. C. Crawford, of the U. S. National Museum. Some of the drawings on the plates were used by the writer in a former paper on the thorax, published in the Proceedings of the Washington Entomological Society (1908), and are here reproduced with the permission of the editors of that journal. No new theory is presented. The writer claims that the dia- grams forming text figures 1 to 6 represent simply the facts. All schemes of thoracic symmetry in consecutive circles are discarded on the ground that they are supported only by the imagination. PROCEEDINGS U. S. NATIONAL Museum, VoL. XXXVI—No. 1687. 511 512 PROCEEDINGS OF THE NATIONAL MUSEUM. VOL. XXXVI. The following statements sum up the principal conclusions: (1) There is no reason for believing that the parts of any thoracic segment are derived from more than one metamere, though the primi- tive thoracic region may have been composed of more than three segments, remnants of the supernumerary ones being possibly repre- sented by the intercalary plates of some of the Aptera; (2) the thoracic sclerites are subdivisions of an original undivided segmental wall; (3) the sclerites of the pleurum are homologous throughout all the orders and modifications are brought about principally through the coalescence of the pleurites; (4) the tergum consists of a primitive undivided notal plate carrying the wings and, in the adult meso- and metathorax of all the principal orders, except the Orthoptera, of a second postnotal or pseudonotal plate developed in the membrane behind the first and having no connection with the wings; (5) the divisions of the notum are secondary, though similar in most of the orders, and are not necessarily homologous, while modifications are brought about through a stronger subdivision into distinct regions and even into separate sclerites. It is unfortunate for modern entomology that there are so many species of insects. Entomologists early had to specialize as Coleopter- ists, Dipterists, Lepidopterists, and so in each order a scheme of anatomy and a nomenclature grew up which satisfied the needs of the worker in that order but had no necessary connection with those of workers in other groups. It is true that Andouin in 1824 worked out a system of comparative external anatomy and proposed a universal set of names for the sclerites. It is true also that his names have been in large part employed by nearly all subsequent entomolo- gists. But in the actual application of Andouin’s names to the scler- ites of the thorax, specialists in the various orders have differed widely on account of their ignorance concerning the correspondence of parts in different insects. Recent entomologists who have attempted to enforce a uniformity of nomenclature based on a more thorough knowledge of insect structure are confronted with the nonconform- ing masses of literature which must form the basis of work by present and future students in each order. However, even if systematists never can employ a uniform system of names, it can not be denied that it is best to know the true homology of the parts as far as this can be determined. It is impossible to follow the rule of priority in selecting anatom- ical terms, for the name must be descriptive of the part to which it is apphed. The earlier entomologists also paid little attention to the duplication of parts in successive segments, but gave a separate name to every piece. Andouin did away with this system in 1824 and firmly established a nomenclature based on the belief that each thoracic segment is a modification of one plan of structure. He NO. 1687. THE THORAX OF INSECTS—SNODGRASS. 513 should be taken as the Linnwus of thoracic nomenclature, and there is not suflicient reason on any ground for applying new terms to the parts he named. In the study of the wings the venation nomenclature established by Comstock has been adopted. No attempt has been made to prove or to disprove Comstock’s interpretations of the veins in the main part of the wing. While a study of the basal structure may show definitely that some particular vein is absent as a distinct trunk at the base, it still remains an open question whether this vein is actually gone or is fused with the one before or behind it. The general venation must furnish the evidence in most such cases. II. THE SEGMENTATION OF THE HEAD AND BODY. A few decades ago an insect was defined as a creature consisting of a head, a thorax subdivided into three segments, and an abdomen composed of 10 or 11 segments. Such a definition, however, would not satisfy the demands of most present-day entomologists, and it is interesting to contemplate the shock some antievolutionary forefather of entomology would receive could he now see in print the statement that an insect is composed of 40 segments. This said forefather might be in some measure pacified, however, were he to learn that the insects themselves have not been required to keep pace with the ideas of entomologists concerning them. 1. SEGMENTATION OF THE HEAD. If the question as to how many embryonic metameres form the head of an insect could be decided by a vote among present and past students of the subject, the six-segment theory would undoubtedly be established. The problem of head segmentation has been attacked from both an anatomical and an embryological standpoint, but, since the embryologists attempt to discover the actual facts of de- velopment, it would seem that deference should be paid to their opinions. Furthermore, the embryologists agree more closely among themselves than do the anatomists. Although the number of head metameres claimed by the former varies from four to seven, this dis- crepancy is not what it appears to be in figures, for the chief point of disagreement is whether the three preoral segments apparent in the embryo are actual metameres or are only secondary divisions. The real question is thus reduced to one between six'and seven segments. On the other hand, the anatomists describe from four to nine seg- ments without any alleviating circumstances. The number of theories seems to agree closely with the number of theorizers. Comparative anatomy as a key to morphology has been so thoroughly deposed in vertebrate craniology that we must regard it with great suspicion in Proc.N.M.vol.xxxvi—09——83 514 PROCEEDINGS OF THE NATIONAL MUSEUM. VOL. XXXVI. entomology. On the other hand, its advocates may, of course, point out that adult insects are so specialized that even the embryo in many respects does not repeat phylogeny. Yet it has never been shown that the head is a case in point. It certainly at an early em- bryonic stage consists of a series of segments and no evidence has been offered to prove that these embryonic segments are not true metameres. The principal anatomists who have mapped out the head on purely anatomical grounds are Newport (1839), Janet (1899, 1900), and Verhoeff (1905). Newport went at the subject in the simplest manner possible. He virtually drew circles around the head corre- sponding with the plates of the dorsal surface, namely, (1) the lab- rum, (2) the clypeus, (3) the front (clypeus posterior), (4) the small sclerites sometimes found about the bases of the antenne, and (5) the epicranium. He thus had five segments which were composed laterally and ventrally by whatever fell between two of the inclosing lines, Later anatomists have not been satisfied with this direct and simple arrangement by Newport. Janet (1899, 1900) makes out nine head segments which he arranges in three sets of three each and then points out the nice conformity in which the set of three thoracic segments follows. He does not even intimate, however, by what natural law the head should be a multiple of the thorax, or why his theory should be more plausible by making it such. Verhoeff (1905) discredits embryology as a guide in the study of the morphology of the insect head, and, on purely anatomical grounds, elaborates a scheme of eight segments for the head of the Dermaptera. The labrum, the clypeus, and the front, according to his plan, are the first three terga, while the sterna of these segments form the epi- pharyngeal membrane and the anterior part of the throat. These segments constitute the “ protocephalon.” Following them are an antennal segment (a preantennal segment being absent in Der- maptera), and a premandibular segment, constituting the “ deuto- cephalon.” Finally, the three jaw segments form the “ tritocephalon.” The mentum and submentum form the sterna of the labial and max- illary segments, respectively. This view assumes that the maxille originate behind the labial palpi. In the Chilopoda the so-called maxillz are much more like the ligula and labial palpi of insects than like the insect maxilla, while the chilopod labium consists principally of the leg-like palpi, thus suggesting that the hexapod first maxille are the chilopod second maxilla. If this should be true, then the theory advanced by Banks (1893) that the poison fangs have co- alesced with the second maxille in Chilopoda to form the first max- ille of Hexapoda appears more possible. Otherwise, Banks had to assume that the poison claws moved forward past the bases of the second maxille and then fused with the first maxilla, A combina- NO. 1687. THE THORAX OF INSECTS—SNODGRASS. 515. tion of the poison claws and the labial palpi of the chilopods would form an organ much more nearly resembling the insect maxille than would a union between the poison claws and the first maxille of chilopods. All reasoning of this fascinating sort, however, simply shows the limitless extent to which morphological theorizing can be carried on anatomical grounds. Verhoetf’s theory of head segmentation has been severely criticised by Heymons (1905) on the ground that the facts of embryology utterly refute it, and that it does not conform with the segmentation of the nervous system. The simplest embryological view holds that there are four segments in the head—a preoral, a mandibular, a maxillary, and a labial seg- ment. This is advocated by Lowne (1892), who regards the three embryonic divisions of the preoral region as secondary. Bengtsson (1897, 1905) adopts this view concerning the preoral region, but he finds four segments in the postoral part of the head. Almost all students of the embryology of the insect head, however, regard the three preoral divisions as true metameres. Hence, embryologists are divided in opinion mainly between six and seven head segments. The principal advocates of six segments are Zaddach (1854), Hux- ley (1878), Viallenes (1887), Wheeler (1893), Heymons (1895), Packard (1898), Riley (1904), and Holmgren (1904, 1907). The advocates of seven head segments are Folsom (1899, 1900). and Comstock and Kochi (1902). But these authors are supported also by Bengtsson (1897, 1905) and by Bérner (1904) in so far as they find four postoral segments, though they recognize only one preoral segment. The seven-segment theory is based mainly on Folsom’s (1900) ob- servation that seven pairs of ganglia appear in the head soon after involution, and that in Anurida maritima a pair of appendages or “ superlingue ” appear back of the mandibles, corresponding with the fourth pair of ganglia. These appendages fuse in most insects with the lingua of the embryo to form the hypopharynx of the adult, but in many lower forms they remain as the lateral lobes of the hypopharynx or “endolabium” and have been misleadingly called the “ paraglosse.” Borner (1904) finds that the hypopharynx of nearly all insects having incomplete metamorphosis is a compound structure formed of the median * glossa” and the lateral paired ele- ments, which he calls the “ maxillule.” (The reader must remember that the terms “ glossa ” and “ paraglossx ” have been inconsiderately applied by some recent entomologists to the parts of the hypopharynx or “endolabium,” while they properly belong to the outer or true labium.) Borner thus recognizes four postoral segments. Hansen (1893) suggested that the “ paraglossx ” (superlinguz, maxillule) of 516 PROCHEDINGS OF THE NATIONAL MUSEUM. VoL. XXXVI. Machilis ave homologous with the first maxille of Crustacea, and Folsom concurs in this view. Apparently no one has compared them with the paragnatha of Crustacea. Holmgren (1907), on the other hand, claims that these superlingual processes arise from the premandibular segment and are innervatea from the tritocerebrum. It would seem that he must refer to a different pair of appendages, namely, the second antennal rudiments or “ intercalary appendages.” His cbservations were made on a fly larva (Phalacrocera). Bengtsson (1897, 1905), however, describes an endolabium in Phalacrocera which includes “ paraglossve,”’ equivalent to the super- lingue of lower insects. Holmgren (1907) refutes this idea entirely, and claims that Bengtsson’s so-called endolabium of fly larvee is not the endolabium of lower insects but simply the terminal lobes of the ordinary outer labium, of which Bengtsson’s “ ectolabium” is the mentum and submentum. He furthermore asserts that what Beng- tsson takes for nerves going to this endolabium from the superlingual ganglion are simply muscle fibers, though Bengtsson (1905) had stoutly defended his former observations (1897). The best summarized statement of the segmentation of the head is that made by Comstock and Kochi (1902). Although some work has been done since, but little new information has been added. The preoral part of the head consists of three embryonic segments cor- responding with the three lobes of the brain, namely, the protocere- brum, the deutocerebrum and the tritocerebrum. The first segment has no appendages, but it imnervates the eyes; the second is the antennal segment; the third carries the “ intercalary appendages ”— vestigial organs observed by many embryologists in the Aptera (Wheeler 1893, Uzel 1897, Claypole 1898, Folscm 1900), possibly in the Diptera (Holmgren 1907), and in the Hymenoptera (Biitschli 1870). These rudimentary appendages correspond with the second antennee of Crustacea. The postoral region of the head and the mouth parts are certainly derived from at least three embryonic segments, or, according to many embryologists, from four. The first 1s the mandibular segment. The possible second is the one under dispute, but so many embryolo- gists have described two small appendages back of the mandibles which fuse with the median lingua to form the hypopharynx that their existence can not be doubted, 2nd it is reasonable to suppose they represent a segment. Riley (1904), however, shows that these superlingual appendages, or maxillule, are absent in the embryo of Blatta, and he doubts that they are actual appendages where ob- served. Berlese (1906) also does not recognize a superlingual seg- ment. Following this doubtful metamere is the segment of the first maxille, and finally that of the second maxille or labium. NO. 1687. THE THORAX OF INSECTS—SNODGRASS. 517 Comstock and Kochi (1902) attempt to assign the various head sclerites of the adult to individual segments of the embryonic head. Riley (1904) in studying the cockroach arrives at different results, but he discredits the reliability of all attempts to map out the adult head according to segments. In discussing Comstock’s view, he says: “ My results have convinced me that so intimate a relation between primary segmentation and the sclerites can not be shown.” Of course, the ventral part of the preoral region becomes dorsal so that the mouth, which is originally on the middle of the ventral surface of the head, comes to be situated anteriorly. Thus the labrum, clypeus, and front are developed from a primitive ventral surface. So, in a general way, the other sclerites arise from definite regions, but they are simply secondary divisions of a continuous head capsule, and the notion that they are modified terga, pleura, and sterna of the head metameres appears to be entirely unsupported by actual evidence. The last head segment is the one that chiefly concerns us in a study of the thorax. All embryologists seem to agree that its body forms the sclerites found in the neck of the adult and that only its fused append- ages, the labium, become asociated with the head, except when there is a gular plate present, which sclerite is derived from its sternum. This embryonic segment can, therefore, hardly be spoken of as a head metamere. It is the segment of the neck and this, in the adult, has received the name of “ microthorax.” Hence we would. accept six primitive head segments, providing the apparent superlingual segment is a real one, and one microthoracic er neck segment. 2. SEGMENTATION OF THE Bopy. The foregoing discussion of the segmentation of the head has been made more extensive, perhaps, than a mere introduction to the study of the thorax would require. But the writer wishes to illustrate to anyone not familiar with the subject the utter futility of attempting a study of metamerism on an anatomical basis. The embryology of the thorax has never brought out much more than that three segments compose it, except in the Hymenoptera, where the first abdominal segment is fused with the thorax. Hence there are no embryological facts concerning the thorax by which anatomists can be held in check, but, with the unfortunate example of both the vertebrate and the insect head in mind, one must certainly regard with much doubt all theories of thoracic metamerism based on a study of the plates form- ing the very apparent three segments in the adult. Riley (1904) makes the following appropriate statement : It would seem that the definitive sclerites can afford little or no evidence as to the primary segmentation of insects. This is certainly true of the head sclerites 518 PROCEEDINGS OF THE NATIONAL MUSEUM. VOL. XXXVI. and I see no reason why it should not apply to other regions of the body. Sclerites originate from mechanical causes and do not necessarily have any relation to the primary segmentation. : Lowne (1892) in discussing the prevalent notion of the dual struc- ture of the thoracic segments states that he does not admit it proved, and does not see that it helps in the understanding of the morphciogy of the insect segment. The writer, then, wishes to say emphatically that he discards every- thing but plain statements of the facts in the description of the thorax. Since, however, modifications of the same plan of thoracic structure recur throughout the insect orders, this fact can be taken as evidence that all the sclerites, especially those of the pleurum, have not been produced independently in the different orders. Many writers have supposed that each thoracic segment consists of two united segments. The arrangement of the plates on any typical segment would suggest this—the dividing line on the side passing between the episternum and the epimerum, on the back be- tween the scutum and scutellum, and on the venter between the sternum and sternellum. Some authors have adduced further evi- dence of the dual nature of the segment from the apparent division of the coxa in some orders into an anterior and a posterior part. Banks (1893), arguing from the coalescence of segments in the Chilopoda, concluded that the thorax of insects is formed of five segments, the first, third, and fifth retaining the legs, the second and fourth bearing the wings. He regards the coxve as double and cites the meso- and metacoxal appendages of J/achilis as examples of remnants of the ventral appendages of segments two and four. He points out that in Scutigera (the highest chilopod) the small terga, after the first segment, are united with the larger ones so that the first segment bears only one pair of legs while the following bear two pairs each. It is only a step from this to suppose that in M/achilis the second leg of each pair has become rudimentary, forming the coxal appendages, while the first of each pair has persisted as the functional walking appendage. Banks does not seem to regard the cervical sclerites of insects as representing a segment in the thoracic series. Patten (1890) gave other reasons for regarding each segment as double, adduced from a study of the mouth parts and the nerves. Walton (1900) still further supports this theory by a study of the cox. He concludes that in both the Chilopoda and the Hexapoda the coxa is composed of an anterior part, “coxa genuina,” and a posterior part, “coxa meron.” These two coxal segments falling in line with the episternum and.epimerum, and the arrangement of the thoracic muscles, form his basis for believing the entire segment a compound of two primitive segments. NO. 1687. THE THORAX OF INSECTS—SNODGRASS. 519 Now, it is only in the mesothorax and metathorax of Mecoptera Neuroptera, Trichoptera, and Lepidoptera that the coxa is actually a double structure. In these orders the coxa genuina of Walton carries the trochanter, while the coxa meron is attached to the coxa genuina only. In other orders in which the coxa shows a more or less evident division this division is in the coxa genuina itself, the coxa meron being absent, and is of the nature of a strengthening of the coxa by opposite ridges on its inner walls. In the Neuroptera and Trichoptera at least it can easily be demonstrated, by a study of larval and pupal forms, that the “coxa meron ” is simply a detached extension of the epimerum, which fuses upon the posterior side of the true coxa. It is, hence, not a part of the primitive coxa at all, and the apparent double coxa in these orders is a purely secondary con- dition. (See special descriptions under Neuroptera, p. 564, and Tri- choptera, p. 565, also p. 542 and figs. 144-148.) Comstock and Kochi (1902) show that the plates of each segment may be arranged into two subsegments, but defer any opinion as to whether they represent two primitive segments or not. It will be found that all these theories are purely imaginative. Embryologists have not shown that the plates of any thoracic seg- ment are derived from more than one metamere. However, it may be true that two, three, or four segments primarily existed where there is but one in insects as we now know them. Verhoeff (1902, 1903, 1903a, 1903c, 1904, 1904a) is the principal elaborator of this theory. Verhoef bases his ideas on a study of the Aptera, the Embiide, and the Dermaptera, and especially on a comparison of Japyx with the Chilopoda. He first points out the tendency in the Chilopoda toward the suppression of every alternate segment by a fusion with the preceding larger spiracle-bearing segment. In /apy« there are remnants of extra segments between the pro- and mesothorax, and between the meso- and metathorax, represented principally by well- developed tergal and sternal plates. Thus the thorax would consist of six segments in three pairs, namely, the microthorax and prothorax, the stenothorax and mesothorax, and the cryptothorax and meta- thorax. Verhoetff observes, however, that this arrangement does not correspond with that of the Chilopoda, because the small segment in Japyx is associated with the large segment following instead of with the one preceding. Then, as if to remedy this discrepancy, he further discovers traces of still other thoracic segments, one between the stenothorax and the mesothorax and another between the crypto- thorax and the metathorax. Finally, by the aid of small presternal plates (“ vorplatten ”) he is able to construct the following table of complete uniformity in segmentation between Scolopendridxe and Japygide (Verhoeff, 1904a) : 520 PROCEEDINGS OF THE NATIONAL MUSEUM. VOL. XXXVI. SCOLOPENDRID A: JAPYGIDA. Head. Head. Maxilliped segment. Microthorax. First leg-bearing segment. Prothborax. Intercalary segment. Presternal plates (vorplatten). Second leg-bearing segment. Stenothorax. ° Intercalary segment. Small intercalary ring. Third leg-bearing segment. Mesothorax. Intercalary segment. Presternal plates (vorplatten). Fourth leg-bearing segment. Cryptothorax. Intercalary segment. Small intercalary ring. Fifth leg-bearing segment. — Metathorax. Intercalary segment. Presternal plates (vorplatten). Thus, it is supposed that two pairs of Scolopendrid segments—a leg-bearing and an intercalary segment in each pair—have been re- duced to one segment in ordinary insects. This reduction has re- sulted not from a combination of segments but from a suppression first of the interealary segments of the chilopod and then of the alternate remaining leg segments. The intercalary segments of Scolopendride, in other words, are not the small segments of Japyax, but are the much more rudimentary traces of segments between these and the large segments. Verhoeff’s own statement (1903c) is as follows: The intermediate segments (zwischen-segmente) of insects are reduced pri- mary segments, inherited from Chilopodan ancestors and which have united into a double segment with the large primary segment immediately behind, while the intercalary segment of the original double segment of the Chilopods has become almost entirely extinct. According to this theory, then, the primitive thorax consisted of ten segments. However, all but three of these have been eliminated in all but the very lowest insects, and the eliminated segments have taken no part in the formation of the plates of the body wall in pres- ent-day insects. It is certainly no difficult matter to show that the sclerites are formed during postembryonic growth and are purely secondary divisions of the body wall of one segment. Hence, this theory of Verhoefl’s is entirely tenable, since it deals only with condi- tions which are presumed to be obliterated before the thoracic plates begin to form. However, it must be admitted that all this elaborate scheme is based on an excessive use of the imagination. No proof is adduced to show that the intermediate and intercalary sclerites of /apya are not secondarily developed plates or even subdivisions of the prin- cipal segments. Desguin (1908), in reviewing this notion of the multiple nature of the thorax in Aptera, concludes that neither the anatomical nor the embryological evidence is sufficient to prove whether these intermediate sclerites represent true segments-or not. Borner (1903) also gives a good criticism of some of Verhoeft’s ex- travagant theories. No. 1687. THE THORAX OF INSECTS—SNODGRASS. 521 Verhoeff extends his view of the many-segmented structure of the insect body to the abdomen (1903a, 1903c, 1904). Here he finds, in the region of the first seven ordinary segments, seven primary seg- ments and seven secondary ones. Beyond these are two genital seg- ments, then the segment carrying the cerci, and finally, in the lowest insects, traces of three more beyond the last—the pygidium, the metapygidium, and the telson. The gonapophyses and the cerci are earried by the fifteenth, sixteenth, and seventeenth primitive seg- ments, which are the eighth, ninth, and tenth persisting segments. Verhoeff thus makes out a total of twenty abdominal segments. Add to these the ten thoracic segments, one microthoracic segment, and nine head segments, and an insect assumes the dignity of a creature of forty segments! Ill. THE MICROTHORAX. Embryologists have shown that the sclerites of the neck, the second maxille of the head, the hind part of the subeesophageal ganglion, and the gular plate, when present, are all derived from one metamere. They usually reckon this metamere as the last segment -of the head, while anatomists call its cervical parts in the adult the microthorav. This term has become pretty well established and will be adopted in the present paper, but not implying that it is a part of the true thorax. On the other hand there is no reason for calling it a head seoment. In many of the lower insects its appendages, the second maxille or jabium, are not attached to the head but are suspended from the gular membrane and associated much more closely with the microthoracic sclerites than with the head (Spodromantis, 25, Sm.). The fact that the microthoracic ganglion is fused with the true head ganglia preceding it signifies nothing more than does the fusion of the first abdominal ganglion with that of the metathorax. It is only when the sternal plate becomes transferred to the ventral surface of the head, as the gula, that the microthorax takes any part in the actual formation of the head. Verhoeff (1902) regards the segment of the maxillipeds or poison’ claws in the Chilopoda as the equivalent of the microthorax in insects. This. however, is denied by Silvestri (1902), who says that the maxilliped segment of the Chilopoda is the prothorax of insects. Verhoeff (1903b) then further shows that in Scolopendra there are four pairs of nerves going to this segment, of which the second is the largest and innervates the appendages. In Polypsitlota striata, a Mantid, he discovers the same four pairs of nerves arising from the subeesophageal ganglion back of the labial nerves and going to the microthorax and salivary glands. Here, however, the second is the weakest and obviously because there are, according to Verhoeft’s view, no microthoracic appendages, the labium not being regarded as such. 599 PROCEEDINGS OF THE NATIONAL MUSEUM. VOL. XXXVI. There is evidently a lack of harmony here unless it be that the first maxille of the Chilopoda correspond with the superlingue of the Insecta, the second maxille with the maxille, and the poison claws with the labium. In this case we could regard the microthorax of insects as the maxilliped segment of the Chilopods, which, from superficial appearances, would not seem impossible. The sclerites of the microthorax are well known. They have been studied extensively by Verhoeff (1902) and occur in nearly all the orders of insects. They are specially well developed in the Odonata (5, 6, 7, 8, 9, 12, Me and 1 mz, 2 mi, 3 mi, 4 mi), in the Orthoptera (24, 25, 36, 37,45), and in the Euplexoptera (93), but occur in a more reduced condition in many of the other orders, such as the Coleoptera (95, M/Z) and the Diptera (174, mz, mi). In the Orthoptera and Euplexoptera they often form an almost complete segment present- ing tergal, pleural, and sternal plates. Verhoeff has gone so far as to identify all the pleurites of a thoracic segment in the microthorax, but undoubtedly this is establishing homologies on a too imaginative basis. Comstock and Kochi (1902) regard the gular sclerites of the head as the microthoracic sternum, and in some of the Euplexoptera (93) the microthoracic sternites are so large and so associated with the head as to suggest the gular sclerite of the Coleoptera. We may conclude that there is no reason for regarding the micro- thorax as anything more than the neck segment whose sclerites are reduced to the small neck sclerites and the gular plate when the lat- ter is present, whose ganglion has fused with the last head ganglion, and whose fused appendages become attached to the head in most cases and constitute the labium. It should not be included, in reck- onings of the number of segments forming the head, as one of the head segments. (See note on page 595.) IV. THE THORAX. In a former paper (1908) the writer gave a brief account of the structure of the insect thorax. This description can now be amplified by illustrations taken from all the principal orders. For convenience the subject will be divided under three heads, namely, (1) the ter- gum, (2) the pleurum and coxa, and (3) the sternum. 1. THr TERGUM. The word tergum is here used to designate all the chitinized parts of the dorsum of any segment. It is generally used interchangeably with the term “ notum,” but where the tergum consists of two plates the latter name, nofwm, will be restricted in this paper to the first or wing-bearing sclerite, and the term postnotum or pseudonotum (Verhoeff, 1903) applied to the posterior or post-alary plate. The notum is the plate which, by diversities of its surface topography, NO. 1687. THE THORAX OF INSECTS—SNODGRASS. 5283 becomes divided into the more or less definite regions usually called the prescutum, scutum, and scutellum, while the postnotum remains undivided and is the postscutellum. The postnotum does not occur in the Orthoptera; it does not occur in the nymphs of any insects, even though well developed in the adults; it does not occur in the pupe of Neuroptera and Coleoptera at least; and it is never present in the prothorax. Therefore it is most probably not a primitive tergal plate, and the term pseudonotum fits it very well. Verhoeft (1903) gave this name to the postnotal plate of the Euplexoptera (Dermaptera), though he may not have intended its general uso 1 the sense here applied. Text figures 1 and 2 é, diagrammatically rep- ik Tg R 7 \ M resent the relation as of the notum (JV) and the pseudonotum (PN) to each other and to the wing, the last being carried en- tirely by the notum. Fig. 3, representing a segment in side view, shows the pseudo- notum continuous lat- par SS SS : ‘ Pph Tic. 1.—DIAGRAMMATIC TERGUM OF ANY COMPLETE WING- BEARING SEGMENT, AND TIE BASS OF THE WING, ‘erally with the epi- DORSAL VIEW; 14, FIRST ANAL VEIN; ANP, ANTERIOR merum (Epm). This NOTAL WING PUSS anr, LINE OF pee VEN- 2 TRAL NOTAL RIDGE (ANE OF ric. 2); AC, AXILLARY 1s the most frequent corD; AM, AXILLARY MBEMBRAND; C, COSTA; Cu, condition. thou oh CUBITUS; Em, LATERAL i [ARGINATION OF NOTUM; , ; = M, MEDIA; Mb, MEMBRANE BETWEEN NOTUM AND often there Sie! line PSEUDONOTUM ; N, NOTUM; PN, PSEUDONOTUM OR POST- NOTUM; PNP, POSTERIOR NOTAL WING PROCESS; pnr, between the two and Se 4 LINE OF POSTERIOR VENTRAL NOTAL RIDGE (PNR OF sometimes they are FIG. 2); Pph,; POSTPHRAGMA; f&, RADIUS; Sec, SUB- te costa; 7g, TEGULA; UV, LINE OF MEDIAN OR V-SHAPED V oT ) ‘ > ’ Ys ’ ’ only articulated or VENTRAL NOTAL RIDGE (V OF FIG. 2). merely contiguous. This figure and figure 2, giving a ventral view of the tergum, both show the postphragma (Pph) depending from the posterior edge of the pseudonotum. though it is often restricted to the middle of the latter. The pseudonotum always carries the postphragma. Verhoeff re- gards it as a development of the postphragma, but it is probably a better statement of the facts to say that the phragma is a develop- ment of the pseudonotum, for in the lower insects the latter is a large flat plate, while the phragma may be simply a thin fold pro- jecting downward from its posterior edge. In the mesothorax of Lepidoptera, Hymenoptera, and Diptera this condition, however, is reversed, the phragma being developed to a great size, although the pseudonotum itself is not reduced. The postphragma is really a 524 PROCEEDINGS OF THE NATIONAL MUSEUM. you. xxxvt. chitinization of the infolded intersegmental membrane behind the pseudonotum, for it is always composed of two closely appressed or fused lamine. The first is directly continuous with the pseudo- notum, the second is connected with the notum of the segment fol- lowing, generally by membrane but sometimes directly, as when the segments are fused. The pseudonotum is conspicuous in the metathorax of Coleoptera (182-140, PN). It is the plate that Straus-Diirckheih (1828) named the “tergum” in MWelolontha vulgaris (185, PN), but most authors have followed Audouin (1824) and Newport (1839) in call- ing it the “postseutellum.” This name is appropriate when a scutum and scutellum can be distinguished. Berlese (1906) recognizes and fig- ures the plate in the Coleoptera, but he refers it to the abdomen, calling it the “ acrotergite ” of the first abdominal segment. Such a disposition of the sclerite, how- ever, 1s clearly impossible on ac- count of its intimate connection, an articulation (7) in beetles, with the epimera of the metathorax. In Fie, 2.—Diacrauiavrc Terqua of any the mesothorax of Coleopteraaihngra DE er ea ee is no pseudonotum unless the two WING =ROGHas = AVR, iamenkon NOTA ge Sula plates (127, 128, 181, q) - RIDGE; Aph, PREPHRAGMA; AwxC’, AXIL yoking’ the mesonotum to meta- LARY CORD; Hm, LATERAL EMARGINA- * x : : TION. OF ‘THD NOTUM: Md. Mwemenann motum! are ruduments. Gaines BETWEEN NOTUM AND PSEUDONOTUM; pupee of beetles do not show a N, NOTUM; PN, FSEUDONOTUM; PNP, : POSTERIOR NOTAL WiNe Process: PNR, PSeudonotum even in the meta- rors nora zens, "mm, gost. thorax. In Dendroctonus vale DION OF THE NOTUM =) Wo wsaieen | (122, 26) Shand eae ae eee RIDGH OF NOTUM, THE ENTO- yelytinwm (123) it is easy to see ae that no pupal plate intervenes be- tween the metathoracic notum or wing-bearing sclerite (V,) and the first abdominal tergum (/7). The latter can be identified by the first abdominal spiracles. In the Plecoptera the pseudonotum is a large, simple plate in both the meso- (75) and the metatergum. It is partly overlapped by the notum (V). Ina nymphal tergum, however, there is no trace of it (76), and its site is entirely membranous (J/b). It is similar in Neuropteran adults (142), but lacking in the pupa (141). In the Lepidoptera (149), the Hymenoptera (169), and the Diptera (174 and 179) the pseudonotum is present in both segments and is easily distinguishable as the tergal plate behind the wing bases. In the mesothorax of these orders it carries the large phragma (150, iti ~ NO. 1687. THE THORAX OF INSECTS—SNODGRASS. 52 t 163, 170, 179, Pph or Pph.) that projects posteriorly through. the metathorax and almost shuts off the cavity of the thorax from that. of the abdomen. In a Tipulid pupa (173) the mesopseudonotum (PV) is present as a large plate intervening between the two wing-bearing plates (V, and V,). It is interesting to note here that the halter is ¢ wing-like structure (IV,). The pseudonotum has been discussed at considerable length because the fact has apparently not been recognized by other authors that the postscutellum or pseudonotum is an independent plate in its origin and is, hence, not one of the divisions of the notum, as is the prescutum, secutum, or scutellum. The notum (1) or wing-bearing plate of a meso- or metathoracic tergum is diagrammatically illustrated in figs. 1,2, and 3. In its sim- plest form it is an undivided plate, convex dorsally. On its ventral] surface (fig. 2) the anterior and posterior margins are thickened, forming the anterior notal ridge (AV/) and the posterior notal ridge (PNP). The latter is generally folded forward a short distance on the ventral surface, forming a free posterior reduplication (/?d) which often overlaps the sclerite following. The lateral margins of the notum are produced into two processes which carry two of the articular sclerites of the wing base. These lobes are the anterior notal wing process (ANP) and the posterior notal wing process (PNP). The posterior edge of the notum, formed by the posterior reduplication, usually appears as a marginal thickening which is con- tinued outward on each side as a corrugated, cord-like thickening of the anal edge of the basal membrane of the wing. These thickenings may appropriately be called the avilary cords (AxvC). They are important characters in determining the posterior limit of the notum. The anterior notal ridge bears the anterior phragma or prephragma (Aph). This outline might be taken to represent the structure of the meso- or metatergum of primitive winged insects, for it is approximately that of nymphal and some pupal forms though these lack the notal wing processes. A pseudonotum is never present in nymphal stages and the prephragma is usually but little developed. The tergum of the nymph is illustrated in the Odonata (15), in the Mantide (31), in the Acridiidz (56, 58) and in the Perlide (76). The same sim- plicity is exhibited by the pupal tergum of Coleoptera (122, 123, 126). Ina Tipulid pupa (173), however, the pseudonotum is pres- ent in the mesothorax as a distinct plate (PN,) between the two wing-bearing plates (V, and V,). As will be shown later, the notum of the adult is commonly divided more or less distinctly into sev- eral regions or even sclerites. But a study of nymphs and pup shows conclusively that these notal divisions are secondary char- acters in the growth of the individual. The tergum consists at 526 first of one plate—the notum, from the entire lateral margins of which the wings develop. Behind the notum is added, in the adult stage, the pseudonotum (postscutellum) as a distinct plate, while the so-called prescutum, scutum, and scutellum are formed as secondary divisions of the notum. These notal regions, moreover, are not homologous in all the orders. This can be proved by a study of the ventral surface of the notum, which presents certain fundamental characters common to nearly all insects. Two of these are the ante- rior and posterior notal ridges (fig. 2, ANR, PNP) already de- scribed; a third, and the most important one, is the V-shaped ridge (V), the entodorsum of Amans (1885), located on the posterior half of the notum, having its apex forward and the bases of its arms fused with the posterior ridge. A comparison of text figs. 1 and 2 will show that the three ventral ridges (AVP, V, PVR) form three transverse lines on the surface of the notum (anv, v, pnr). These ridges and their surface lines are undoubtedly homologous structures in all insects. They mark off the area of the notum into four regions, as follows: (1) A narrow anterior marginal band in front of the line of the anterior ridge; (2) a large bilobed region situated between the line of the anterior ridge (an) and that of the entodorsum (v) and carrying the notal wing processes (AVP, PNP) ; (3) a triangular space between the line of the entodorsum and that of the posterior notal ridge; and (4) a narrow posterior marginal band terminating laterally in the axillary cords (4#@C) and form- ing the posterior free edge or reduplication of the notum. This typical simplicity of structure is illustrated in the Orthoptera by Blatella (88, 40) and Gryllus (49, 50), and in the Veuroptera by Corydalis (142, 143). It will be observed that the pseudonotum (PN) is absent in the Orthoptera, but well developed in Corydalis. These four regions of the Orthopteran notum are very suggestive of the four divisions of the tergum as ordinarily recognized, but a comparison with Corydalis (142) at once shows that the term “ postscutellum ” can not be applied to any part of the Orthopteran tergum, for the name belongs to the pseudonotum, which is absent in Orthoptera. Hence, all identifications of a “ postscutellum ” in Orthoptera, supposed to be homologous with that of the higher orders, are erroneous. A comparative study of the orders shows that the posterior line (pn) is generally absent, that the notum is very commonly divided by lines or actual sutures into three subdivi- sions, and that these lines or sutures are not determined by the ventral ridges and do not bear the same relation to them in the different orders. Thus we have the following premises: (1) The ventral ridges of the notum are constant in all the orders and are, hence, fundamental homologous structures; (2) three notal divisions are of general PROCEEDINGS OF THE NATIONAL MUSEUM. VOL. XXXVI. NO. 1687. ‘THE THORAX OF INSECTS—SNODGRASS. 527 occurrence but are not constant nor do they present the same relation to the ventral ridges in the different orders. From these facts it fol- lows that the notal divisions are not necessarily homologous wher- ever they occur, though they may be so within limited series, and that they are simply secondary adaptations to some common demand upon the notum. The above statements and conclusion can be verified by a study of the species illustrated on the plates. Since the subdivisions of the notum are best developed in the highest orders these will be described first. The ordinary names of prescutum (psc), scutum (sct), and scutellum (scl) will be used to designate the notal regions, but the reader must bear in mind that they are not used in the different orders in a homologous sense, and that the abbreviations on the fignres do not designate parts necessarily homologous. 'The ‘“ postscutel- lum ” will be called the pseudonotum (PV). The notal divisions are probably as well shown in the mesothorax of a Tipulid fly as in any other insect. In Holorusia grandis (174, 175) the prescutum (psc) is a large plate with its posterior margin produced posteriorly in a large V-shaped angle, having no relation to the anterior notal ridge. The lateral posterior angles are pro- duced into two small lobes (175, ~) lying opposite the anterior angles of the wing bases. The scutum (sct) is a wide plate carrying the anterior notal wing processes (AVP). The scutellum (sc/) consists of a median elevated shield and of a depressed area on each side. The posterior wing processes belong to the latter, though they are sep- arated from it by a tongue of membrane. The ventral V ridge is present just as in the diagram (fig. 2), but it marks only the apex (v) of the scutellum, its lateral parts not showing on the surface. Thus the three divisions of the Tipulid notum are but slightly influenced by the ventral ridges. The pseudonotum (P.V) is well developed, con- sisting of a median and two lateral plates, the latter articulated with the epimera (174, pm). In a Tabanid (179, 180) the third division of the notum (scl) is distinct but the first (psc) and the second (sct) are not separated mesially. The lateral angles (w) of the prescutum, as in Holorusia, lie opposite the wing bases. In the Hymenoptera similar divisions of the notum occur, as is well shown in the drawing (160) of Parasiobla, a Tenthredinid. An examination of the ventral surface reveals the V ridge present but situated entirely behind the suture between the scutum and scutellum. These two plates, furthermore, are easily separable along this suture and, hence, the latter can in no way be compared with the dividing line between the scutum and scutellum of the Orthop- tera. In the example given (160) the prescutum (psc) is perfectly exposed, but it is more commonly hidden in the Hymenopteran meso- thorax by the pronotum (169, V,), which is attached to and overlaps 528 PROCEEDINGS OF THE NATIONAL MUSEUM. VOL. XXXVI. the anterior part of the mesothorax. The pseudonotum (P.V) is also usually hidden on account of its projecting downward before the metathorax. It can easily be shown, however, by removing the mesotergum from the surrounding parts (163, 170). The metanotum in both the Diptera and Hymenoptera is reduced in size and the subdivisions are not well marked (174, 169). The metapseudonotum is present in both orders but is generally very narrow in the Diptera. In the Hymenoptera it is usually a large plate (160, 169, PV.) continuously fused on the sides with the meta- epimera (/’pm,), though in some cases it is narrow and scarcely distinguishable from the metanotum (164). In the Hymenoptera there occurs a fusion of the first abdominal segment with the metathorax. This fact has led to a great deal of discussion among entomologists and to the production of an immense amount of literature. Latreille (1821) first described the rear part of the apparent Hymenopteran thorax as being a part of the abdo- men and named it the “segment médiaire.” Newman later (1833) called it the “ propodeon.” Packard (1866), by a study of the devel- opment of Bombus, proved that the first abdominal segment is actually transferred to and becomes consolidated with the meta- thorax. A great many other writers have written a great many opinions about it and about the opinions of other writers, and Gosch (1881) has furnished a voluminous historical account of all the opinions of all these writers up to his time. To his Contribution to the History of Entomology (1881) the reader is referred if he is interested in this phase of the subject. If not, the examination of a few-specimens of the insects concerned will probably suflice. The abdomen of Cimbex (166) shows clearly enough that the first abdominal segment (164, 77) is much more closely attached to the thorax than to the rest of the abdomen. That the part in question is the first abdominal segment is proved by its spiracles (J Sp) and by the structure of the metathorax and of the rest of the abdomen. The metathorax of Cimbex (164) has, in addition to the attached part (/7), its full complement of sclerites. The notum (1) and pseu- donotum (/?V) are present dorsally and the episternum (Z’ps) and the epimerum (Z'pm) laterally. In the abdomen itself, if the first segment behind the one in question is counted as the first, there would be present only nine segments in all, and the absurdity would be forced upon us of referring the female gonapophyses to the seventh and eighth segments. Hence, arguing from either end, the conclusion would be that the median segment belongs to the abdomen. In Parasiobla (160) this segment could never be regarded as more than a slightly transposed part of the abdomen. However, in the higher forms, of which Pepsis (169) isa good representative, the median segment is so intimately grown into the metathorax that it certainly does not NO. 1687. THE THORAX OF INSECTS—SNODGRASS. 599 appear to belong to the abdomen. Yet in the metathorax there is present the true metathoracic notum (V,), and the pseudonotum (PN,) identifiable by its fusion with the metathoracic pleura. Hence the large corrugated plate (77) behind the pseudonotum has no place in the metathoracic anatomy, and its abdominal origin is proved by its spiracles (J Sp). Thus a study of comparative anatomy proves con- clusively that the “segment médiaire ” is at least the first abdominal tergum which has been transferred to the thorax, and Pepsis indi- eates that the entire first abdominal segment is so transferred. If this is true, then the ventral part disappears as a distinct plate in the higher families. In the Lepidoptera the mesonotum has much the same appearance as in the Diptera and Hymenoptera, but differs in details of structure. In the Cosside (149, 150) it consists of a large scutum (sct) and scutellum (sel) separated along the line of the ventral V-ridge, and of a very narrow prescutum (150, psc). The postscutellum (P.V) is present, but normally (149) is almost hidden between the mesothorax and the metathorax. In the Sphingide (155, 156) the prescutum depends vertically from the anterior edge of the scutum (155) and carries the prephragma (Api). The prescutum in Phassus is, there- fore, much more nearly the equivalent of the anterior division of the notum in the diagram (fig. 1) than in either the Hymenoptera or the Diptera and, hence, is more similar to the Orthoptera (88) and Neuroptera (142). In the Diptera, it will be recalled, the prescutum is large and extends back to the bases of the wings. In the Hy- menoptera it is remote from the wing bases. In Phassus (150) the scutum carries both the anterior and the posterior-wing processes of the notum, while in Protoparce (156) the posterior processes arise from the scutellum. This is due to the fact that here the lateral parts of the scutellum are not defined by the ventral V-ridge, but appear simply as depressed areas at the sides of the median elevated part of the scutellum as in Holorusia (175) and Tabanus (180). In this case the separation between the scutum and scutellum laterally is simply a matter of topography. . The metathorax of Phassus (149, 151) is larger and more like the mesothorax than is usual among the higher insects. The prescutum (psc) and the scutellum (sc/) almost meet on the median line, thus separating the scutum into two lateral plates (sct). The posterior wing processes (PVP) arise in the angles between the scutellum and the scutum. The scutellum in all the forms so far described carries the axillary cords of the wings (Aa() at its extremities. These cords, which are distinct in nearly all insects, are, hence, diagnostic of the location | of the scutellum in Lepidoptera, Hymenoptera, and Diptera, defining its posterior margin, and consequently, the posterior edge of the Proc.N.M.vol.xxxvi—09—34 : 530 PROCEEDINGS OF THE NATIONAL MUSEUM. VOL. XXXVI. notum. The postscutellum (PN) lying behind them is always a plate more or less distinctly separated from the notum, but connected or continuously fused with the epimera. Tn the Coleoptera there occur two special modifications of the meta- thoracic tergum which set the beetles apart in this respect from all the other orders. One of these characters is the forward extension of a median tongue of the scutellum toward the prescutum, cutting the scutum into separated lateral halves. The second character is the division of each lateral scutal plate again into two by lines formed by special transverse ventral ridges laterad of the apex of the V-ridge. In a separate paper (1909) the author has shown that Audouin’s interpretation of the coleopteran tergum is untenable, that, in order to make out his four transverse tergal sclerites, Audouin has represented certain parts as continuous which in nature are separate, and in other cases has made separations where none occur. A Carabid, Calosoma scrutator (132, 133), presents a very simple arrangement of the metatergal subdivisions characteristic of the beetles. The prescutum (psc) consists of a large quadrate median part and of two narrow lateral arms widened terminally into the triangular anterior notal wing processes (AV/). The median part is separated by a membranous area (mb), the “ toile” of Straus-Diirck- heim (1828), from the anterior extension of the scutellum on the floor of the median notal groove (G@). On its anterior ventral edge the prescutum carries the prephragma (A p/h) mesially and the cup-shaped muscle apodemes (J//)) laterally. The scutum (sct, sct) is divided into four plates by the median approximation of the prescutum and scutellum, and by the lines (132, w) formed by the special transverse ventral ridges (133, w). The posterior division on each side carries the posterior notal wing process (P VP). The scutellum (132, scl) presents a median enlargement carrying the tongue extending for- ward on the floor of the median notal groove (G) and determined by the entodorsum or ventral V-shaped ridge (133, V), while laterally it extends to the bases of the axillary cords (A #C() as a narrow marginal postscutal strip on each side, determined by the posterior notal ridge (PV 2). Behind the notum, and entirely separated from it by a flexible suture, is the pseudonotum (postscutellum) (PV), carrying the post- phragma (Pph) and articulating at its extremities (7) with the epimera. Dytiscus dauricus (136, 137), is very similar in its metatergal struc- ture to Calosoma (182, 133). The lateral arms of the scutellum, how- ever, are larger and, in addition to carrying the axillary cards, they support the combined bases (7) of the anal veins of the wing. The ventral ridges (137, V, w) are much larger than in Calosoma. NO. 1687. THE THORAX OF INSECTS—SNODGRASS. 581 The higher families of beetles, illustrated by ydrophilus triangu- laris (134), Melolontha vulgaris (135, 138), and Cyllene robinia (140), have a prescutum somewhat different in appearance from that of Calosoma and Dytiscus. In Hydrophilus (134) and Cyllene (140) its median part (psc) is narrow and arched forward, and the mem- branous area (mb) back of it is extended transversely. The scutel- lum (sc7) appears to have a long median tongue () by itself entirely separating the scutum into lateral halves. The anterior scutal subdi- visions, in front of the transverse dividing lines (w), which are in- complete in ydrophilus, are reduced to turgid antero-lateral corner lobes. The lateral extensions of the scutellum in each of these genera fuse laterally with the parts in front of them, so that the axillary cords (A#C’) appear to be attached to the margins of the scutum (154, 140). Melolontha vulgaris departs still more widely from the Calosoma-Dytiscus type. The median part of the prescutum is represented entirely by the very large prephragma (135, 138, ph), which is supported by the lateral parts of the prescutum and sepa- rated from the scutum and scutellum by an extensive membranous area (mb). The scutum (135, sct, sct) is divided, as in the other genera, into lateral halves by a median tongue of the scutellum (sc/) on the floor of the median notal groove (G), but the transverse ridges (138 w)are coincident with the anterior scutal margins and do not subdivide the scutal plates. The scutellum (sc/) is not defined later- ally, but two triangular postero-lateral divisions of the scutum, the “ scapulaire posterieure ” of Straus-Durckheim (1828) carry the pos- terior notal processes (PVP) and the axillary cords. The pseudonotum is well developed in all of these genera (134, 135, 140, PV) and carries the postphragma (Pph). The latter is spe- cially large and of complicated structure in Melolontha (159, Pph). The mesonotum of beetles is apparently constructed on the same plan as the metanotum. A pseudonotum is lacking. In Calosoma (127) and Dytiscus (128) the prominent shield-shaped area (sc/) corresponds with the median part of the scutellum of the metanotum, lateral arms extending from it which carry the axillary cords (AaC). Laterad of the median shield are the separated halves of the scutum (sct) carrying the posterior wing processes of the notum (PVP). In front of it is a large complex prescutal part (psc) carrying the anterior phragma (Ap) and laterally the anterior notal wing processes (AVP). Two little plates (7) lie between the mesonotum and the metanotum. These may be rudiments of a mesopseudonotum, but they are more closely connected with the metanotum than with the mesonotum. On the ventral surface of the mesonotum (181) a V-shaped ridge (V) is present similar to that of the metanotum. In most other beetles the parts of the mesonotum are so blended that any plan of structure closely corresponding with that of the metanotum 532 can not be made out. Yet a progressive modification from the Cal- osoma-Dytiscus type can be traced through Hydrophilus (125), Cyllene (129), and Dendroctonus (124). It is thus clear that the same fundamental structure of the notum obtains throughout the Coleoptera, the Lepidoptera, the Hymenop- tera, and the Diptera, but that the notal subdivisions are not neces- sarily determined by it. In the Neuroptera a tergum of diagrammatic simplicity is found. In Corydalis (142, 148) the scutum (sct) and scutellum (scl) are separated along the line (142 v) of the ventral V ridge (148 V). Anteriorly and posteriorly are narrow marginal areas defined by the anterior and posterior notal ridges (AVR, PNR). The first of these might be called the prescutum, but the second is simply the pos- terior notal reduplication (Ad) and does not correspond with the postscutellum of higher orders, for this is the pseudonotum, which is well developed in Corydalis (142, PN). In the Euplexoptera the metanotum of Spongiphora (96) consists of an undivided plate carrying both the anterior and the posterior wing process (AVP, PVP), while articulated near the middle of its posterior margin are two long arms (f) bearing the axillary cords (da7C). In the mesonotum (90) there is a large anterior triangular part carying the anterior notal wing processes ( These are all premolars, as afterwards recognized by Cope. 616 PROCEEDINGS OF THE NATIONAL MUSEUM. VOL. XXXVI. Type-locality.—Sweet Grass County, Montana. Horizon—Fort Union formation. The description as given by Douglass is as follows: Crown of ps semi-elliptical in a lateral view; crown not high, and upper por- tion of cutting edge not extremely convex; eleven distinct and two posterior indistinct ridges on the crown; mi nearly one-half the length of ps, with four external ® and six internal ®% tubercles; anterior portion of tooth narrower than posterior portion. The last premolar is larger than that of Ptilodus trouessartianus Cope, but not so large as that of P. mediavus Cope. Measurements of type (after Douglass). é mm. Teneth Of pees) 20 2 ee ee oe ee ae (2b [Tastee hength of mise) os Sa ee BRE pep gd: CARE wh See ae oooh 3.7 [3.4] This is the first species of Ptilodus described from the Montana Fort Union and is apparently quite distinct from the New Mexico Torrejon species described by Cope. In the National Museum collec- tion there are portions of upper and lower jaws representing four in- dividuals that are probably referable to this species. They are two left lower-jaw fragments, each containing p,; a lower-jaw fragment containing p, and m,; and an upper-jaw fragment containing the three anterior premolars. These are, respectively, Nos. 6077, 6078, 6079, and 6080, U. S. National Museum collection. They are from the same locality and horizon as the type of P. montanus. PTILODUS GRACILIS, new species. Type-specimen.—A. partial skeleton, including a nearly complete skull and lower jaws, the greater portion of the pelvis, one femur, the distal portion of the humerus, a radius and portion of an ulna, the proximal end of a tibia, a terminal phalanx, portions of vertebra, and other fragments. (Cat. No. 6076, U.S.N.M.) Ty pe-locality— Sweet Grass County, Montana. Horizon.—F¥ort Union formation.° General definition —Dental formula, it, c+, pm$, m3. In size this species is about intermediate between P. mediwvus and P. trouessar- tianus. It is but shghtly smaller than P. montanus Douglass. The lower jaw is far more slender in proportions than either P. mediwvus or P. montanus; the fourth premolar is relatively higher and has 14 ridges on either side of the crown. Lower m, has but 5 tubercles in @This is evidently a typographical error and should read “ four internal and six external tubercles.” 5 In carefully remeasuring the type I find these measurements to be: Length of ps, 7.8 mm.; length of mi, 3.4 mm. ¢ Other forms from these beds indicate that this horizon is about equivalent to, as reported by Douglass, or perhaps somewhat older than the Torrejon beds of New Mexico. No. 1689. THE FOSSIL GHNUS PTILODUS—GIDLEY. 617 the outer row of cusps, as compared with 6 in P. montanus. The upper dentition of this species can at present be compared only with that of Ptilodus plicatus Cope, since the upper molariform teeth of other species of the genus are not known. The principal differences thus shown are: (1) The smaller proportions of P. gracilis, which is about one-fourth less in size; (2) the proportionately greater antero- posterior length of the cheek teeth; and (3) the greater number of cusps in p*, there being 6 cusps instead of 5, as in the type, or 4, as in the neotype of P. plicatus. The number of these cusps may be more or less variable, but are probably of specific importance. DETAILED DESCRIPTION. The dentition—The teeth of the present specimen being fully adult and but little worn afford an excellent opportunity both for a de- tailed and general study of their characters. The single upper incisor preserved is placed in the premaxillary near but not closely appressed to the median line. it is a relatively long, rounded and gently curved tooth, with a pointed: tip. The crown is slightly compressed on the posterior face with lateral angles dividing the surface into two unequal areas, the posterior one being the smaller and less convex. Near the summit of the tooth the pos- terior area is subdivided obliquely by a sharp ridge running from the apex upward and outward to the outer main angle, so that the tip of the crown is roughly triangular in cross section. The next tooth in the upper series, which from its position is appar- ently a canine, is separated from the incisor just described by a con- siderable diastema. It much resembles the incisor in general form, but is smaller, and a posterior oblique ridge runs to the inner main angle instead of the outer one. A considerable diastema separates the canine from the premolar series. | The first upper premolar, pt, is a triangular tooth composed of three nearly equal subpyramidal cusps whose apexes point slightly backward. Premolar* resembles p*, except that it is quadrate in outline and has four subequal cusps. This tooth is somewhat higher crowned than the others and is the largest of the anterior three pre- molars. Premolar*® equals p? in length, but is narrower and _ less robust. The crown is quadrangular and bears six cusps of nearly uniform size arranged in two longitudinal rows. The anterior pair is somewhat smaller than the others, and the cusps are closely joined to those of the median pair. This tooth strikingly resembles the upper true molars of Plagiaulax (Bolodon). All the cusps of the anterior premolars have the peculiar wrinkled appearance noted by Cope, Marsh, and others. In classing the next tooth of the upper series with the premolars. I differ from the opinions of Cope and Osborn, who called the cor- 618 PROCEEDINGS OF THE NATIONAL MUSEUM. VOL. XXXVI responding tooth in Ptilodus plicatus the anterior true molar. How- ever, a critical examination shows that, though resembling in general form the molar next behind it, it differs from that tooth in several important particulars, while in position and function it properly belongs with the antemolar series. Like the first true molar, p* is much elongated anteroposteriorly and is multituberculate, but here the real resemblances between the two teeth end. P* is made up of two rows of cusps of unequal length, the inner, composed of nine tubercles of nearly uniform size, being the longer, while the tubercles of the outer row, seven in number, are very unequal, some being larger and some smaller than those of the inner row. The third tubercle from the front in the outer row is the largest and principal cusp of the tooth; in consequence the transverse diameter of the crown is greatest at this point. This and the three other larger cusps of the outer row have the characteristic wrinkling of the enamel seen in the anterior premolars and is almost a duplicate of the outer main cusps of p*; if detached, it might readily be mistaken for one of them. There is a marked difference in the character of these cusps and the smooth-surfaced tubercles of the true molars. In contrast with p*, m* is composed of two subequal rows of tuber- cles of nearly uniform size and a third less prominent and much shorter inner row, which, in the present specimen and in P. plicatus, is developed only along the posterior half of the crown; thus the widest transverse diameter of this tooth is at the extreme posterior end instead of across the anterior half, as in p*. The last tooth of the series, m*, is wider transversely, but is much shorter than m‘. It has three rows of tubercles, the inner and outer rows being fused into an almost undivided ridge. The cusps of the median row are shehtly curved and point forward. The lower incisors are comparatively long, slender, and moderately curved. They are oval in section at the base, but are somewhat flat- tened on the inner faces and are sublanceolate near the tips. . 8 Width of palate between fourth premolars______ = 11 Distance from occipital condyle to posterior border of palate __---_-_- =| as) Mranisverse-diamecerioL Occipital Cond yest ee ee ne 9.2 Distance from) center iOfsslen oid Oss e to mis = eee eee eee ae ney eee ilit Distance from) condyle,oPlower jay, tO) ila ae en 11 Kength of humerus (estimated) 25. = =) 2 ee 24 Lengthiox qdistal portion preserved = css. - Pa a eek on eee 16 Greatest diameter of head otihtimen tga. se ee eee ee 4.9 ‘Dransyerseidiameter of) sla hte. teh ess Fs he ee Se ZAG Width seross: condyles 1.2 .- = 2 fe i, | os i, (ee 5.6 Width across ar tieular ss Ort ace (emer Ory) eee se eee 3. 4 Width across articular Ssurtace {qpOSteriOm) mass. as ann eee ee 2316 AAD LCTOPOSLELLOM diameter Ol MRO Gest eete ae eee es ee 3.4 Notalslength? oi Pelvis CEs time nities) hee ee te eee era ae 37 Tenet wok Sew Mai, 2 Ee: es oe es Be a 2 eee 1285 Diameter: of ‘acetabulum = 5 see a ee eee ar fo hength)-Of emurie 22225. se 2 ee oe Si ee 32+ Diameter: of shatt: 5 == 32 5 eee Se ee ee 4 PTILODUS SERRATUS? ®@ (Marsh). A left lower fourth premolar (Cat. No. 6088, U.S.N.M.), found in the same deposits from which the types of P. montanus and P. “ Halodon serratus Marsh, Amer. Journ. Sci., XXXVIII, 1889, p. 87, ply aim. fig. 14. ~“ NO. 1689. THE FOSSIL GENUS PTILODUS—GIDLEY. 623 gracilis were obtained, represents a third and much smaller species of Ptilodus from these beds. This tooth is about the size of the type of P. trouessartianus Cope, but does not agree with it in proportions, the latter being relatively much lower crowned. In this respect it corresponds much more closely and agrees also in size with the type of “ Halodon serratus” Marsh, from the Ceratops beds of Converse County, Wyoming. For the present I provisionally refer it to Marsh’s species, although there are minor differences which indicate that it may later be placed in a new species. ° PTILODUS FORMOSUS? © (Marsh). A fourth and still smaller species is represented in this collection from the Fort Union by two lower fourth premolars (Cat. Nos. 6089 and 6090, U.S.N.M.). These teeth, which are only 3.2 mm. in length, agree in size, proportions, and the number (12) of enamel ridges with Marsh’s “ Zalodon formosus,” from which they can not at present be distinguished. The type of this species is also from the Ceratops beds. ZOOLOGICAL RELATIONS. Owing to the absence of good material, well-defined characters other than those presented in the teeth have hitherto been wanting. In consequence there has been a wide diversity of opinions regarding the relation of the Multituberculates to other great groups of the Mammalia. Earlier writers, studying Jurassic forms, classed them with the Marsupialia. Marsh, in 1880," first proposed placing the group in a distinct order, which he named the AJdlotheria, although recognizing their marsupial affinities. In defining the order he said: ¢ “These characters alone do not indeed separate the Plagiaulacide from some of the Marsupials, and future discoveries may prove them to belong in that group, where they would then represent a well- marked suborder.” Later Cope* proposed the name JI/ultituber- culata for this same group, which he considered a suborder of the Marsupialia. Still later he suggested their relationship to the Mono- tremes. At present most authorities rank the group as a full order, which is variously classed with the Marsupials or the Monotremes. As has been stated by Osborn,’ the relationship of the group to the Marsupialia, which was first proposed by Falconer, had not been questioned until the discovery, by Poulton, of the early-shed multi- tubercular teeth of Ornithorhynchus. This led. Cope to suggest the reference of the Multituberculates to the J/onotremata, a view which @ Halodon formosus Marsh, Amer. Journ. Sci. XXXVIII, 1889, p. 179, pl. vil, fig. 36. 6 Amer. Jour. Sci., (3) XX, 1880, p. 239. ¢ Amer. Naturalist, XVIII, 1884, p. 687. @The Structure and Classification of the Mesozoic Mammalia, Journ. Acad. Nat. Sci., Phila., (2), IX, 1888, p, 254. 624 PROCEEDINGS OF THE NATIONAL MUSEUM. VOL. XXXVI. has found rather wide acceptance. In his classification of the Meso- zoic Mammals“ Osborn stated his own views as follows: While the Multituberculata are widely separated from the mammals of the second group [including the Trituberculate forms] they are so closely related to each other by the unique structural and functional adaptations of the denti- tion, that the discovery in one genus of a single taxonomic character, which is distinctive, will probably determine their position either with the Monotremata or Marsupialia or in an independent order; * * *, The separation of these genera from the Diprotodonts justifies the prediction,. as a result of future discovery, that the Multituberculata will prove to be the last representative of a very ancient phylum which reached too great a degree of specialization and den- tal reduction at the close of the Cretaceous to survive or leave descendents in the recent period. Whether they are to be considered as a branch of the mono- treme or of the marsupial stock is an unsettled question. Via. 7.—RIGHT LOWER JAW OF TRICHOSURUS VULPE- alse oe ci In his latest _classification of the Mam- NER view. 3Nat. size. Malia,’ Osborn lists. the Allotheria under the subclass Protherta as a doubtful order of uncertain systematic position. Falconer and Owen referred Plagiaulax to the Diprotodontia, but differed in their opinions regarding its probable habits and taxonomic relations. Falconer compared Plagiaulax with Hypsiprymnus (Potorous) and sought to prove that the former was a salatory herbivorous marsupial, allied to the Rat-Kangaroos. Owen °¢ just as strongly contended that it was carnivorous in habits, and more prob- ably related to the extinct carnivorous 7'hy- lacoleo. Owen’s conclusions regarding the. carniv- orous habits of Plagiaulax lose much of their force since it is now apparent, through a study f tl eetitaaGe, £5P ion s that are F1G. 8.—RIGHT LOWER JAW of the present specimen ‘of Piilodus, that his: ~ 42 pasoncs ealceie! principal arguments were based on an error in Type. Iynervinw. 3 2 =! c NAT. SIZE. the interpretation of a most important factor, ‘ namely, the normal position of the jaw. Viewing the lower jaw of Ptilodus, properly articulated with the upper (see figs. 7 and 8), it is observed that passing forward it pitches downward at a considerable angle, bringing the plane of the tooth-row below the condyle, and the incisors into a semiprocumbant position as in the Diproto- donts. It will be noted also that the greater part of the thin cutting blade of p, does not come in contact with the upper teeth, but stands free in the mouth. If the lower jaw of Plagiaulax is thus placed @ Journ. Acad. Nat. Sci. Phila. (2), IX, 1888, p. 254. > Bvolution of the Mammalian Molar Teeth, 1907, p. 11. ¢ Fossil Mammalia of the Mesozoic Formations, 1871. no. 1689, THE FOSSIL GENUS PTILODUS—GIDLEY. 625 (see fig. 9), here also the condyle is above the tooth-row and not be- low it, as stated by both Owen and Falconer. The premolar teeth likewise drop away from the level of the molar series, forward, so that the anterior ones could scarcely have come in contact with any teeth of the upper jaw. It is further observed that, as in Ptilodus, the ridges on the sides of the cutting blades viewed laterally run which have always been described as being “ oblique” in the fossil forms, are after all placed in the same relative position in the mouth as those of the ridged premolars of living species. Assuming this like premolars did not oppose teeth of like structure in the upper jaw, the carnivorous characters pointed out by Owen seem to disap- pear, while the general resemblances to the less ever apparent. . f The resemblances pointed out by Falconer ¢ between Plagiaulax and Potorous (‘“ Hypsi- in the main substantiated in the present speci- OF PLAGIAULAX BECKLE- men of Ptilodus, which also possesses some opie eon BETS : 4 f VIEW. ABOUT NAT. SIZE. important characters in common with some of tion of the teeth, however, has been carried to a greater degree, both in reduction in numbers of the molars and in the development of the premolars, than in any ofthe living Diprotodonts. sions regarding the zoological position not only of Ptilodus and the Plagiaulacide, but of the Allotheria in general. A fact of first im- portance is that neither in the skull nor skeleton of the Montana tremes, while every character is marsupial, as shown in the general arrangement and function of the teeth and the development of the skull and skeleton. The unequal development of the fore and hind position of the cheek-teeth all indicate definite affinities with the Diprotodonts. At the same time the reduction in numbers of the molars and the extreme specialization of the premolars confirms namely, that the Multituberculata may be the last representatives of a very ancient phylum that became extinct in the early Tertiary. From the foregoing therefore.it appears that the Allotheria repre- nearly at right angles to the plane of the molars; thus these ridges position for the lower jaw and recognizing the fact that the blade- specialized Diprotodonts become more than prymnus”) a genus of the Macropodide, are |... 9 icy Lowen saw the Phalangeride, as 7richosurus and Phalanger. The specializa- This preliminary study leads apparently to the following conclu- specimen are there any indications of affinities suggesting the Mono- limbs, the character of the incisors, the form of the palate, and the Osborn’s conclusion regarding the philetic position of the group, sent an extinct group of multituberculate eutherian mammals closely “Quart. Journ. Geol. Soci. London, XIII, 1857, 626 PROCEEDINGS OF THE NATIONAL MUSEUM. VOL. XXXVI. related with but not ancestral to the Diprotodont division of the Marsupialia, with which division they may be now classed as an Infraorder, or Superfamily, their relationship dating back to a common ancestry somewhere in the Jurassic or even to earlier Tri- assic times, as was suggested by Cope. Regarding the probable character of the food, upon which Ptilodus and its related genera subsisted, the specimen from Montana described above seems to throw considerable light. An examination of the type of P. gracilis (Pl. 70) shows Cope’s® statement, that the grind- ing teeth are “ weak in structure,” is incorrect, and his supposition that it was necessary for the animal to swallow its food without mastication is not admissible. On the contrary it will be seen that the grinding area is comparatively very considerable, occupying nearly three-fourths of the actual contact space between the upper and lower dental series, and although the tooth-crowns are low they are relatively broad and massive.’ To add to their efficiency the molars are well supplied with an array of short stout tubercles, well adapted to crushing and grinding small hard substances but very poorly adapted to cutting or masticating meat. In specimens of old individuals the much worn condition of the tubercles of the molars as compared with that of the lower cutting-premolar suggests that the latter may have been used for the purpose of cutting only soft materials, such as the skin and pulp of fruit, while the molars were employed in grinding harder substances, such as seeds. The evidence that Ptilodus and Plagiaulax were not carnivorous in habits seems rather conclusive, but as to whether they were in- sectivorous, herbivorous, or frugivorous there may still be some differences of opinion. I am inclined to consider them as frugivorous, since the incisors were well fitted for picking small fruits or berries, while the large cutting blades of the lower premolars were admirably adapted to cutting or sliceing the rinds of tough-skinned berries, or to chopping up fleshy fruits held against the blunt-pointed premolars of the upper jaw. For masticating the seeds of such small fruits and berries the multituberculate molars were amply sufficient. EXPLANATION OF PLATE 70. Ptilodus gracilis Gidley. (Type-specimen.—Cat. No. 6076, U.S.N.M. All figures twice natural size.) Fig. a. Skull with lower jaw in position, side view. b. Right ramus of lower jaw, inside view. c. Lower jaws in normal position, viewed from above. d. Skull, top view. e. Skull, palate view. f. Distal portion of left humerus, posterior view. U. S. NATIONAL MUSEUM PROCEEDINGS, VOL. XXXVI PL. 70 PTILODUS GRACILIS GIDLEY. FOR EXPLANATION OF PLATE SEE PAGE 626, O haop A vnal Tia FRESH-WATER SPONGES IN THE COLLECTION OF THE UNITED STATES NATIONAL MUSEUM.—PART I. SPECIMENS FROM THE PHILIPPINES AND AUS- TRALIA. By Netson ANNANDALE, Superintendent of the Indian Muscum, Calcutta. The collection of Spongilline in the United States National Museum consists very largely of specimens named by Mr. Edward Potts, whose Monograph of the Fresh-water Sponges*® must ever remain a classical work on the group. Since Mr. Potts gave up active work on the sponges, however, a considerable number of specimens have been added, which the authorities of the Smithsonian Insti- tution have been kind enough to send me for examination. As these specimens are accompanied by duplicates of all the named American species in the collection, and as the Indian Museum pos- sesses an almost complete set of the species recorded from Europe, Asia, and Africa, I hope that it may ultimately be possible for me to determine all those that are determinable. In the meanwhile, stress of official work renders it difficult for me to attack the Ameri- can species, and I propose, therefore, to deal separately with those from the Philippines and Australia. Genus SPONGILLA. Subgenus EUSPONGILLA Vejdovsky. SPONGILLA SCEPTRIOIDES Haswell. Spongilla sceptrioides Haswety, Proc. Linn. Soc. N, S. Wales, 1882, p. 209.—v. LENDENFELD, Zool. Jahrbiicher, II, 1887, p. 89. Haswell’s original description is very brief, and Lendenfeld adds little of importance to it. There is a specimen in the collection under review which is labeled, “ Fresh water Sponge with winter eggs Queensland, Australia Apr. 4-”. This, I beleve, to represent Has- well’s species, although I have had some doubts as to the identity. Tt will be well, therefore, to describe the specimen in some detail. *Academy of Natural Sciences, Philadelphia, 1887. PROCEEDINGS U. S. NATIONAL Museum, VOL. XXXVI—No. 1690. 627 628 PROCEEDINGS OF THE NATIONAL MUSEUM. VOL. XXXVI. It apparently formed an irregular mass some 10.5 cm. long and 2 cm. thick, coating a piece of stick, but has unfortunately been much damaged in transit and now consists for the most part of loose powder and gemmules. The color (dry) is a pale gray. The surface so far as it remains, is smooth, with fairly large oscula (about 3 mm. in diameter), which are not raised on eminences. The external mem- brane has wholly perished. The substance of the sponge is com- pact, the primary radiating fibers, but not the secondary transverse ones, being visible in a vertical section to the naked eye as slender white threads. The gemmules, which are prac- - tically colorless, are numerous throughout the sponge. a x 240 eae @ x240 Fig. 1.—SPONGILLA SCEPTRIOIDES. ad, SKELETON SPICULES j; b, GEMMULE SPICULES. The largest skeleton spicules measure 0.35 mm. by 0.021 mm. They are straight or feebly curved and are covered with extremely minute projections in the central part of their length, the ends, which are sharply and cleanly pointed, being smooth. The projections are so minute that it is often difficult to see them. They are conical in out- line, somewhat broad at the base in comparison with their length, and are rarely sufficiently numerous to give the spicules a roughened look under a low power of the microscope. I can find no flesh spicules. The gemmule spicules measure from 0.126 mm. to 0.147 mm. in length. They are slender in proportion (transverse diameter about 0.0042 mm.) and straight or feebly curved. The spines which cover them with fair uniformity are about half as long as the spicule is thick; those in the middle are straight, those at either end curved and directed backward. As a rule the spicule terminates at either end in a single straight spine. No. 1690. NEW FRESH-WATER SPONGES—ANNANDALE. 629 In general structure the gemmules closely resemble those of Spon- gilla lacustris. 'They are spherical and measure on an average 0.52 mm. in diameter. There is a thick granular coat, in which the spicules are arranged close together and tangentially, while an outer layer of horizontal spicules can be detected on the surface of some gemmules. The aperture of the gemmule, which is single, is provided with a stout foraminal tubule, which is generally more or less curved and projects through the granular coat. Remarks.—It is clear that this sponge is a close ally of S. lacustris, from which it may be distinguished by the absence of free spicules and by the armature of the aperture of the gemmule. From my Spongilla proliferens it is distinguished by its more compact and massive structure as well as its lack of free spicules. Spongilla sceptrioides has been recorded from New South Wales and Queensland. SPONGILLA PHILIPPINENSIS, new species. The sponge has evidently formed a sheet of considerable size ad- herent to some solid body but has been broken into small pieces in the Fic. 2.—SPONGILLA PHILIPPINENSIS, FRAGMENT OF SKELETON, X 70. type-specimens, which are about 1 cm. thick. The surface is smooth. with numerous oscula level with it. There is no trace of branches. Proc.N. M.vol.xxxvi—09—T-44 630 PROCEEDINGS OF THE NATIONAL MUSEUM. VOL. XXXVI. Externally the sponge appears to have been bright green in color, but the basal parts are yellowish. The texture is ight and friable, by no means elastic. In vertical section both radiating and transverse fibers are visible to the naked eye and the sponge has a distinctly reticulate appearance, although the vertical mterspaces are much more conspicuous than the horizontal ones. Wide circular canals penetrate the sponge in a course parallel to its base. Comparatively little spongin is present. Under the microscope it is evident that the radiating fibers are much more coherent and reg- ular than the trans- verse ones. On the external surface of the . sponge a network of 4 horizontal spicules can : be distinguished. basal — structureless a There is a delicate membrane. The ecto- dermal membrane has perished. The skeleton spicules measure 0.174 mm. to 0.278 mm. in length and on an average 0.021 mm. in greatest transverse diameter. They are very sharply FG. 3.—SPONGILLA PHILIPPINENSIS. @, SKELETON pointed at both ends, SPICULES; b, GEMMULD SPICULE. straight or nearly SO, smooth or somewhat sparsely covered with extremely minute projec- tions, the ends being always smooth. There are no flesh spicules. The gemmule spicules are very variable in length, measuring from 0.0798 mm. to 0.122 mm. in length and about 0.0031 mm. in transverse diameter. They are cylindrical, straight or nearly so, armed with somewhat irregular spines, which are often slightly retroverted at the two ends. “Sometimes there is a single straight spine at either end, but often the spicule ends abruptly and is surrounded by a ring of spines in such a way as to suggest a rudimentary rotule. There are few gemmules, those that exist occurring singly in the substance of the sponge and being free. They have a blackish color, are spherical, measuring on an average 0.609 mm. in diameter. Each is provided with a single aperture, to which a short, straight, rather stout foraminal tubule is attached. The inner chitinous coat is *%240, xo.1690. NEW FRESH-WATER SPONGES—ANNANDALE. 631 rather thin, but the granular coat is well developed and contains many spicules, which are arranged horizontally or nearly so as a rule, but sometimes to a slight extent tangentially. Habitat—Camp Keithly, Lake Lanao, Mindanao, Philippines. Altitude 2,250 feet. Mary Strong Clemens, collector, January, 1907. Type-specimen.—Cat. No. 7718, U.S.N.M. Remarks.—With the exception of EH phydatia fortis from Luzon, this appears to be the first fresh-water sponge recorded from the Philippines. It appears to be quite distinct from the other form discovered with it and here described; but it is just possible that it may be a form of S. sceptrioides. Pending the acquisition of further information regarding the latter species, however, I prefer to con- sider it a new species. All the specimens I have seen are dry. Subgenus STRATOSPONGILLA Annandale. SPONGILLA CLEMENTIS, new species. In general appearance and color this sponge, judging from dry specimens, closely resembles S. philippinensis, but the surface is usually covered with a network of deep, broad furrows which sep- arate small elevated areas of a more or less circular form. The oscula occur on these elevated areas and are large and numerous. Probably in the fresh sponge the furrows are roofed in by the ectodermal mem- brane. In vertical section the transverse fibers of the skeleton are seen to be stouter and more regular than those of S. philippinensis, being hardly inferior to the radiating fibers in these respects, so that the skeleton forms a much more regular network than is the case in the other sponge. There is a stout chitinous membrane, which sends bunches of hol- low root-like processes downwards at intervals. These do not appear to be in any way connected with the primary skeleton fibers. There are numerous scattered skeleton spicules in the basal membrane. The skeleton spicules are smooth, as a rule, but occasionally bear a few irregular spines; they are somewhat bluntly pointed at the ends, as a rule regularly but feebly curved. They measure on an average 0.252 mm. in length and 0.021 mm. in greatest transverse diameter. There are no flesh spicules. The gemmule spicules are slender, cylindrical, nearly straight. In the middle they bear minute irregular projections, which only take the form of actual spines towards the two ends. Each end terminates in a stout, straight spine, surrounded by a row of smaller spines at right angles to it. None of the spines are retroverted. There are very few gemmules indeed. They occur singly in the basal membrane and are apparently closely adherent to the support 632 PROCEEDINGS OF THE NATIONAL MUSEUM. VOL. XXXVI. of the sponge. Each measures about 0.325 mm. in diameter (the shape being spherical) and is provided with a single straight for- aminal tubule on the summit. The granular coat is feebly developed, - but there is a strong outer chiti- nous coat in continuity with the basal membrane. The gemmule spicules le in this coat parallel or almost parallel to the surface ‘of the gemmule but crossing one another at all angles. Habitat—Camp Keithly, Lake b Lanao, Mindanao, Philippine Is- lands. Altitude 2,250 feet. Mary *£40. Strong Clemens, collector, Jan- uary, 1907. Ty pe-spectmen.—Cat. No. 7719, U.S. NOT. Remarks.—This sponge, which I a have much pleasure in naming x240 after its discoverer, is evidently very distinct from S. philippinensis (with which it was apparently found in close associa- tion), differing in its shorter and smoother skeleton spicules, more reg- ular skeleton, thicker basal mem- brane, and adherent gemmules with their ill-developed granular coat. It approaches those Fic, 4.—SpoNGILLA CLEMENTIS. a, SKELETON spicuLes; forms I have re- b, GEMMULE SPICULE; ¢C, FRAGMENT OF SKELETON. cently grouped to- gether in a new subgenus (Stratospongilla’), but differs from them in its slender gemmule spicules. On the whole, despite this differ- ence, I think that it should be associated with them. — “In an account of the fresh-water sponges collected by Prof. Max Weber in S. Africa published in the Zoolog. Jahrbticher, 1909. DESCRIPTIONS OF SEVENTEEN NEW SPECIES OF RECENT CRINOIDS. By Austin Hopart Crark, Collaborator, Division of Marine Invertebrates, U. S. National Museum. The authorities of the Indian Museum, Calcutta, have recently done me the honor of intrusting to me for study the very important collections of recent crinoids brought together by the steamer /nves- tigator during the course of her work in the Indian Ocean. Many of the new species are represented by a considerable number of speci- mens, and of these cotypes have been retained and deposited in the U. S. National Museum. The types themselves are in the Indian Museum. The completed report on the collection will be published as one of the series of /nvestigator monographs. I wish here to record my appreciation of the kindness shown me by the authorities-of the Indian Museum through the superintendent, Dr. N. Annandale, and by Dr. F. A. Bather, at whose suggestion the collections were sent to me. Family ZYGOMETRIDA. Genus EUDIOCRINUS P. H. Carpenter. EUDIOCRINUS ORNATUS, new species. Centro-dorsal a thin disk, the bare polar area flat, 2.5 mm. in diam- eter, the cirri arranged in a single marginal row. Cirri XVIII, 17-18, 10 mm. long; first joint twice as broad as long, second nearly or quite as long as broad, third to fifth twice as long as the proximal diameter, sixth slightly shorter, a more or less marked transition joint; following joints gradually decreasing in length, the terminal joints being only slightly longer than broad; penultimate joint about as long as broad. The third to the sixth joints are very strongly “ dice-box shaped,” with the distal edge all around produced, except on the dorsal side; from the seventh onward both these fea- tures become less marked, and the cirrus becomes somewhat com- Proceepinas U.S. NATIONAL Museum, VoL. XXXVI—No. 1691. 634 PROCEEDINGS OF THE NATIONAL MUSEUM. VOL. XXXVI. pressed laterally. There are no dorsal spines; opposing spine sharp, prominent, arising from the entire dorsal surface of the penultimate joint, equal to about half the diameter of that joint in height; ter- minal claw equal in length to the penultimate joint, stout, and strongly curved. Disk with a few rather large plates along the ambulacra, and well plated in the anal area. Ends of the basal rays visible as small tubercles-in the angles of the calyx; radials projecting shghtly beyond the centro-dorsal, gently concave distally; IBr, and IBr, united by syzygy, forming an ob- long syzygial pair from one-third to one-half again as broad as long, the lateral edges straight, barely in apposition basally, the ventro- lateral border slightly produced. Five arms, 85 mm. long; first brachial oblong, about three times as broad as long; second slightly wedge-shaped, about the same size; third and fourth (syzygial pair) slightly longer on one side than on the other, half again as broad as the median length; next three brachials approximately oblong, twice and one-half as broad as long, the following becoming triangular, as broad as long, and after the proximal fourth of the arm wedge-shaped, as long as broad, and in the terminal portion somewhat longer. The lower brachials have on each side, as far as the lowest pinnule on that side, a shghtly produced ventro-lateral edge, corresponding with that on the [Br series; the brachials have a somewhat concave dorsal surface and very prominent distal ends, everted on the proximal, strongly over- lapping on the distal, which gives the animal a curiously ornate ap- pearance. Syzygies occur between the third and fourth brachials, again between the eighth and ninth, and distally at intervals of three, more rarely four, oblique muscular articulations. P, 5.5 mm. long, moderately stout basally, tapering evenly to the tip, rather strongly prismatic, with twelve joints, the first short, the second not quite so long as broad, the third and fourth squarish, the following gradually increasing in length, being nearly or quite twice as long as broad terminally; P, similar to Pc, with the same number of joints, but somewhat stouter and not tapering so rapidly; P, 8.5 mm. long, much stouter than P., gradually tapering from the base to the tip with twelve or fifteen joints, the first three about as long as broad, the following very gradually becoming elongated and about twice as long as broad distally; the pinnule is rounded-prismatic ; P, similar to P,; P, 6 mm. long, slender, cylindrical, less stout basally than P., gradually tapering and becoming very delicate in the term1- nal portion, with fifteen or sixteen joints, the first short, the second and third about as long as broad, the following gradually increasing in length and becoming nearly or quite three times as long as broad in the terminal portion; P, similar to P,; following pinnules similar, no. 1691. SEVENTEEN NEW SPECIES OF CRINOIDS—CLARK. 635 gradually decreasing in length to 5 mm., then very slowly increasing, reaching a length of 10 mm. distally. The distal ends of the joints of the lower pinnules are more or less produced and spinous. Type-specimen.—Cat. No. TC, Indian Museum; lat. 14° 04’ 30’ N.; long. 93° 51’ 00’’ E.; 41 fathoms. Cotype.—Cat. No. 25478, U.S.N.M., from the same locality. Family HIMEROMETRID 4. Genus AMPHIMETRA A. H. Clark. AMPHIMETRA MORTENSENI, new species. Centro-dorsal thick-discoidal, the bare polar area flat, 4 mm. or 5 mm. in diameter; cirrus sockets arranged in two closely crowded alternating marginal rows. Cirri XVITI-XX, 30-42 (usually about 35), 25 mm. to 30 mm. long; first joint short, about three times as broad as long, second and third about twice as broad as long, the following gradually increas- ing in length to the ninth or tenth, which is nearly, though never quite, as long as broad; next five to seven joints similar, the following gradually decreasing in length, in almost the whole of the terminal half of the cirrus being about one-half again as broad as long; from the twelfth or fourteenth onward sharp median tubercles or small spines are developed on the dorsal side of each joint, those on the last few joints occupying a position slightly proximal to median; opposing spine much larger than the processes on the preceding joints, triangular, the apex median, arising from very nearly the whole of the dorsal surface of the penultimate joint, equal to about half the diameter of that joint in height; terminal claw longer than the penultimate joint, moderately stout basally, but gradually be- coming slender distally, moderately curved. Radials concealed, or just visible beyond the centro-dorsal; IBr, oblong, very short, in close lateral apposition; IBr, (axillary) very broadly pentagonal, almost triangular, the lateral edges not quite so long as those of the IBr,, about two and one-half times as broad as long; I1Br 4 (8+4); H1Br 4 (8+4); division series and first two brachials in close lateral apposition and laterally flattened, the dorsal carination only of Pp being visible exteriorly; synarthrial tubercles usually prominent. Twenty to twenty-five arms 150 mm. long; first brachial slightly wedge-shaped, short, about three times as broad as its exterior length, almost entirely united interiorly; second about the same size, but more pronouncedly wedge-shaped; third and fourth (syzygial pair) oblong, half again as broad as long; next five or six brachials oblong, nearly or quite four times as broad as long, then slowly becoming wedge-shaped and then almost triangular, four times as broad as long, soon becoming wedge-shaped again, and in the outer half of 636 PROCEEDINGS OF THE NATIONAL MUSEUM. VOL. XXXVI. the arm oblong and very short, though somewhat longer again terminally. The proximal discoidal brachials are somewhat swollen, and most of the brachials have slightly overlapping distal ends. Syzygies occur between the third and fourth brachials, again be- tween the thirteenth and fourteenth to thirty-first and thirty-second (usually somewhere between the sixteenth and twenty-fifth, with sometimes an extra one from two to four or five brachials beyond the first), and distally at intervals of two to thirteen (usually eight to twelve) oblique muscular articulations. Pp 7 mm. long, moderately stout basally, but tapering rapidly and becoming slender in its distal half, with about twenty-five joints, at first three times as broad as long, becoming twice as broad as long at the sixth, and squarish in the terminal portion; some of the lower joints are bluntly carinate; P, 10 mm. long, with 30 joints, shghtly less stout basally than Pp and tapering somewhat less rapidly; joints at first about twice as broad as long, becoming as long as broad at about the eighth, and somewhat longer than broad terminally; P, 15 mm. long, stouter than P,, tapering evenly to a delicate tip, with 30 joints, at first about half again as broad as long, becoming squarish at the eighth or ninth, and about twice as long as broad at the tip; P,, 22 mm. long, stouter than the preceding, with 30 joints, at first broader than long, becoming squarish about the tenth, and longer than broad terminally; the pinnule is more or less carinate in its proximal half and has a moderate supplementary ridge on the distal half of the outer side; P, resembling P.,, but very slightly longer and slightly stouter and more carinate; P, like P,; P, 10 mm. long, re- sembling P,, but somewhat more strongly carinate proximally; fol- lowing pinnules gradually decreasing to 7 mm. in length and losing the basal carination, then increasing to 12 mm. distally. On some arms P, is small as described for P,, and again P, may also be small, while occasionally P, and P, are similar and P, is greatly enlarged; sometimes PP., ,, and , are as described for PP, ,, and ,. On one or more of the inner arms of each ray P, is often much larger than on the outer, while the adjacent pinnules are reduced. Ty pe-specimen.—Cat. No. 42B., Indian Museum; Port Blair, Andaman Islands. Cotype.—Cat. No. 25479, U.S.N.M.; from the same locality.’ T have dedicated this species to Dr. Th. Mortensen, of Copenhagen, in recognition of his valuable contributions to the knowledge of the Echinoderms. Genus HETEROMETRA A. H. Clark. HETEROMETRA COMPTA, new species. Centro-dorsal discoidal, the bare polar area flat, slightly convex or slightly concave, about 5 mm. in diameter;.cirrus sockets arranged in a single more or less irregular marginal row. xo. 1691. SHVENTEEN NEW SPECIES OF CRINOIDS—CLARK. 637 Cirri XVITI-XNITII, 31-35, 23 mm. to 25 mm. long; first joint very short, the next three nearly two and one-half times as broad as long, the following gradually increasing in length to the sixth or seventh, which is about as long as broad; next five to seven joints usually slightly longer than broad (sometimes squarish), the follow- ing gradually decreasing in length, the terminal fifteen or rather more being half again to twice as broad as long; at about the fifteenth joint dorsal tubercles are developed, at first involving only the distal portion of the dorsal surface, later arising in a slightly convex line from near the proximal end, the apex being subterminal; these tuber- cles are narrow, laterally occupying only a small portion of the median part of each joint, and are slightly rounded dorsally; on the last three joints the tubercles become somewhat sharper, more erect, and move to a median position; opposing spine small (though larger than the tubercle on the preceding joint), blunt, arising from the entire dorsal surface of the joint, the apex median or submedian in position, in height equal to about one-third the diameter of the penul- timate joint; terminal claw somewhat longer than the penultimate joint, rather stout and strongly curved. Ends of the basal rays and radials concealed; IBr, very short and band-like; IBr, (axillary) short, almost triangular, two and one-half times as broad as long; IT Br 4(3+-4), in apposition laterally, though not laterally flattened; IIBr, entirely united interiorly; I11Br 2, rarely 4(3-+4); IVBr 2, but rarely present. Sixteen to twenty-five arms 110 mm. long; first two brachials wedge-shaped, three times as broad as long exteriorly, the first in- teriorly united; following four or five brachials oblong, about four times as broad as long, then gradually becoming wedge-shaped, almost triangular, about three times as broad as long, and less oblique and somewhat longer on the outer portion of the arms. The dorsal sur- face of the arms is perfectly smooth. Syzygies occur between the third and fourth brachials, again between the thirteenth and four- teenth to twentieth and twenty-first (usually in the vicinity of the fifteenth) and distally at intervals of seven to eleven (most commonly eight or nine) oblique muscular articulations. Pp 7.5 mm. long, moderately stout basally, but tapering rather rapidly in the proximal half and slender distally, with twenty-five joints, at first twice as broad as long, becoming squarish after the tenth; the first four joints are strongly carinate, this carination de- creasing from this point onward and disappearing after the middle of the pinnule; P, 13 mm. long, slightly stouter than Pp basally, tapering gradually, and becoming slender in its distal third, with twenty-six joints, at first twice as broad as long, becoming squarish after the ninth and somewhat longer than broad in the terminal por- tion; the first seven or eight joints are rather strongly carinate and in 638 PROCEEDINGS OF THE NATIONAL MUSEUM. you. xxxvt. addition have a low sharp ridge running along their exterior surface at the base of the carinate processes; P, similar to P, and of the same length, but the low ridge just described may be traced to about the twelfth joint; P, 9 mm. long with nineteen joints, similar to the two preceding pinnules, but slightly less stout; P, small, 6 mm. long, tapering rapidly in the proximal half and becoming very slender distally, with sixteen joints, at first twice as broad as long, becoming squarish about the ninth, and longer than broad distally, the first six joints carimate ‘like those of the preceding pinnules; P, similar, 5.5 mm. or 5 mm. long; P, and the following pinnules 6 mm. long with seventeen joints, at first twice as broad as long, becoming squar- ish about the eighth and twice as long as broad terminally; the pin- nules are about as stout basally as the two preceding, tapering rapidly in the proxunal half and becoming very slender distally; the carina- tion of the proximal joints is slightly marked on the first four; this carination later becomes restricted to the second and third joint only, and disappears entirely in the outer half of the arm. Ty pe-specimen.—Cat. No. 4 F.=* %!4, Indian Museum; Pedro Shoal, north of the Laccadive Islands. Cotype.—Cat. No. 25480, U.S.N.M.; from the same locality. HETEROMETRA SINGULARIS, new species. Centro-dorsal discoidal, the bare polar area flat, 1.5 mm. in diam- eter; cirrus sockets arranged in a single crowded, more or less irreg- lar marginal row. Cirrt XVII, 21-25, 12 mm. long; first joint short, second about twice as broad as long, third somewhat longer, fourth about as long as broad, next two slightly longer than broad, the following grad- ually decreasing in length, the terminal fifteen being one-third or one-half again as broad as long; at the seventh subterminal dorsal spines begin to develop which soon become long and prominent; opposing spine large and long, much larger than the spines on the preceding joints, triangular, the apex terminal, arising from the whole surface of the penultimate joint and about equal to the diam- eter of that joint in length; terminal claw nearly twice as long as the penultimate joint, slender, abruptly curved proximally, becoming nearly straight distally. Disk with a few calcareous granules in the anal area, especially on the anal tube. Radials short, oblong, the dorsal surface with numerous prominent rounded tubercles; IBr, short, oblong, shghtly over four times as broad as long, in close lateral apposition; IBr, (axillary) broadly pentagonal, almost triangular, twice as broad as long, the lateral edges shorter than those of the IBr,; I1Br 4(8+4); joints up to and including the second brachial exteriorly and the fourth interiorly, no.1691. SHVENTEEN NEW SPECIES OF CRINOIDS—CLARK. 639 as well as the first two joints of the first three pinnules, in close appo- sition and sharply flattened, the lateral edges somewhat produced. Eleven arms (in the type), 40 mm. long; first two brachials sub- equal, wedge-shaped, about twice as broad as the exterior length, the first interiorly united; third and fourth (syzygial pair) slightly longer interiorly than exteriorly, nearly three times as broad as the interior length; next four brachials oblong, nearly four times as broad as long, then becoming almost triangular, about three times as broad as long, then gradually lengthening (though remaining almost triangular) to about twice as broad as long, and at a point somewhat beyond the end of the proximal third rather quickly becoming wedge- shaped, almost oblong, about two and one-half times as broad as long. From about the ninth onward the brachials have prominent distal ends, though they do not overlap the bases of the succeeding joints. Syzygies occur between.the third and fourth, ninth and tenth, and fifteenth and sixteenth brachials (the second sometimes omitted), and distally at intervals of seven to ten oblique muscular articulations. Py 4.5 mm. long, moderately stout basally, but tapering rapidly in the proximal half, and slender distally, with 20 joints, at first about twice as broad as long, becoming squarish after the eighth; the second to the seventh joints are rather strongly carinate; P, similar, very slightly longer and stouter; P, 6 mm. long, considerably stouter and stiffer than the preceding, and rather more strongly carinate basally, with about 20 joints, the first 7 (except for the carinate processes) squarish, the remainder slightly longer than broad, becoming about half again as long as broad distally; the ridge in the distal half of the outer side is but little marked; P, 3 mm. long, much smaller than any of the preceding, with about 12 joints, at first broad, becoming squarish about the fifth, and nearly twice as long as broad distally ; the second-fifth joints are carinate; following pinnules similar and about the same length, the joints becoming gradually longer and the basal carination gradually less; distal pinnules, 5 mm. long. On the arms arising from a [Br axillary, P,, P,, and P, are usually as described for Pp, P,, and P,, and P, is much smaller, as described for P,; but occasionally P, is enlarged and similar to P,, as described, instead of being small lke P,. Type-specimen.—Cat. No. TA, Indian Museum; southern por- tion of Malacca Strait. Genus STEPHANCMETRA A. H. Clark. STEPHANOMETRA CORONATA, new species. This species is most closely related to S. tenuipinna. Cirri XXII-XXIII, 25-80, 22 mm. long, resembling those of S. tenuipinna; the longest joints are about one-third longer than broad; the ninth, tenth, or eleventh is a well-marked transition joint. 640 PROCEEDINGS OF THE NATIONAL MUSEUM. — you. xxxv1. Radials projecting slightly beyond the edge of the centro-dorsal ; IBr, oblong, short, about three and one-half or four times as broad as long, not in lateral apposition, with a rounded ventro-lateral process in the proximal half; IBr, (axillary) broadly pentagonal, twice as broad as long, the lateral edges about half as long as those of the IBr,, produced into a rounded prominent ventro-lateral process; synarthrial tubercles rather prominent; IJBr, H1IBr, and IVBr (when present) 2; elements of division series and first brachials with prominent rounded ventro-lateral processes. Thirty-three or thirty-four arms 120 mm. long, in general resem- bling those of S. tenuipinna. P, 14 mm. long, stout, stiff, and spine-like, with fourteen joints, the first two somewhat broader than long, the third to the fifth squarish, the remainder becoming gradually elongated and twice as long as broad distally; P, and P, exactly like P,; P, 10 mm. long with ten joints, resembling the preceding; P, 7 mm. long, spine-like as the pre- ceding, but somewhat more slender, with eight joints; following pin- nules decreasing gradually in length, P, being 5 mm. long with eight joints; subsequent pinnules remaining of similar length, but de- creasing in stiffness and increasing in the number of joints, P,, being 5 mm. long with twelve joints, the third squarish, the distal twice as long as broad, only shghtly stiffened proximally; distal pinnules slender, 9 mm. long. Type-specimen.—Cat. No. 18 H=,*%4,, Indian Museum; “India.” Cotype.—Cat. No. 25481, U.S.N.M.; from “ India.” Family COLOBOMETRID ZK. Genus COLOBOMEDRA Al oH. Clark COLOBOMETRA DISCOLOR, new species. Cirri XVITI-XXIT, 29-40 (usually about 35), 25 mm. to 30 mm. long, slender, resembling those of C. perspinosa, but with the distal ends of the joints not so strongly spinous. Radials projecting shghtly beyond the centro-dorsal; IBr, oblong, slightly over twice as broad as long, the ventro-lateral borders slightly produced into a thin border, by which they are in apposition; IBr, (axillary) broadly pentagonal, twice as broad as long, the lateral edges somewhat more than half the length of those of the [Br,, making with them a straight line, and with the same ventro-lateral projection ; a slight constriction is usually present just below the lateral angles. Ten arms, 80 mm. long, rather slender, resembling in general those of C. suavis. P, absent; P, 6.5 mm. long, small, tapering rapidly to a slender and delicate tip, with 15 or 16 joints; first jomt twice as broad as no.1691. SHVENTEEN NEW SPECIES OF CRINOIDS—CLARK. 641 long, second somewhat longer, third about as long as broad, the fourth Reviiar. the following very eradually increasing in leneth to about half again as long as broad, and becoming squarish again in the terminal 4 or 5; -P, 15 mm. long, moderately stout and very stiff and spine-like, with bau 20 joints, the first about twice as broad as long, the second slightly longer, the third nearly half again as long as broad, the remainder about twice as long as broad; beginning on the second joint there is a faintly indicated, broadly rounded keel running along the middle of the outer side, as on P,; on the third and. follow- ing joints the distal dorsal edge projects in the line of this keel in a narrow fringe of spines, which broadens on succeeding joints, the spines at the same time becoming longer, and is supplemented by additional spines on the ventro-lateral angles of the joints; P,, similar to P,, usually about 1 mm. shorter; P, 10 mm. long, resembling P, and P,, though not quite so stiff, with 15 joints; P, and following pinnules very slowly decreasing in length and stiffness, at the same time becoming more slender, with the spines on the distal ends of the joints less and less pronounced; P, is 8 mm. long and P,, is 7 mm. long, each with 15 joints ; from this point the pinnules very gradually increase to 10 mm. in length distally, the distal pinnules being slender, comparatively littlesstiffened, with 20 to 22 joints, which have eid ately everted ends armed saith fine spines; the distal pinnules are somewhat compressed laterally. Ty pe-spectmen.—Cat. No. 9C., Indian Museum; lat. 14° 04’ 30’ Nes tone. 93° 51°00’ E.; 41 fathoms. Deine. —Cat. No. 25482, U.S.N.M.; from the same locality. Genus GC VULOME TRA A. H. Clark. CYLLOMETRA TAPROBANES, new species. Centro-dorsal thin, discoidal, the bare polar area flat, 2 mm. to 3 mm. in diameter; cirrus sockets arranged in a single, slightly irregu- lar, crowded marginal row. Cirri XX-XXIJ, 25-29, 12 mm. or 13 mm. long; first joint short, the next about two and one-half times as broad as long, the following slowly increasing in length to the fifth or sixth, which is twice as broad as long, and the tenth or twelfth, which is half again as broad as long, and still further increasing, so that the antepenultimate and one or two of the preceding joints are about as long as broad; fifth to seventh and succeeding joints with the distal dorsal edge prominent, forming a low transverse ridge which slowly moves ante- riorly, attaining a median position on about the twelfth, and grad- ually narrows distally, becoming reduced to a small median tubercle on the last twelve; opposing spine prominent, rather slender, median, equal in height to about one-half the diameter of the penultimate 642 PROCEEDINGS OF THE NATIONAL MUSEUM. VOL. XXXVI. joint; terminal claw slightly longer than the penultimate joint, mod- erately slender, and moderately curved, rather more proximally than distally. Radials projecting very slightly beyond the centro-dorsal, slightly separated distally; [Br, oblong or slightly trapezoidal, four times as broad as long; IBr, (axillary) broadly pentagonal, twice as broad as long; synarthrial tubercles moderately developed. Arms 10, about 80 mm. long, resembling those of C. studeri; dis- tal ends of the brachials very slightly, if at all, produced. P, absent; P, 4.5 mm. long,small and slender, with about 14 joints, the first short, the second slightly longer, the third squarish, those in the distal portion being half again as long as broad; P, 8 mm. long, stouter and stiffer than P,, though not especially enlarged, with 15 to 17 joints, the first short, the second and third squarish, the re- mainder one-third to one-half again as long as broad, becoming again somewhat shorter at the extreme tip; the joints in the distal half have slightly enlarged distal ends; P, 6 mm. long, less stout than P., but similar to it, with 14 joints; P, 5 mm. long, slightly less stout. than P,, but similar, with 12 joints; P,; and following pinnules 4 mm. long, about as stout as P,, but not stiffened, with 12 joints, the third squar- ish, the remainder longer than broad, becoming half again as long as broad in the distal half; the distal ends of the component joints are shghtly everted and spinous; distal pinnules slender, 7 mm. long, the joints smooth. Type-specimen.—Cat. No. 2°42, Indian Museum; off Colombo Light House, Ceylon; 264 fathoms. Cotype.—Cat. No. 25483, U.S.N.M.; from the same locality. Family THALASSOMETRIDZE. Subfamily THALASSOMHTRIN At. Genus CROTALOMETRA A. H. Clark. CROTALOMETRA ANNANDALEI, new species. Centro-dorsal columnar, the tip truncated conical as in Astero- metra, 5 mm. long by about 5 mm. broad at the base; cirrus sockets arranged in ten columns of usually three each, the columns of adja- cent radial areas being closely crowded and more or less alternating, the two columns of each radial area being separated by a slightly con- cave median area of about half their width; polar area with five more or less marked interradial ridges which terminate in five small tubercles about the apex. Cirri comparatively slender, XXX, 62-79, 65 mm. long; first three joints approximately equal, short, about twice as broad as long, the following gradually increasing in length, becoming squarish on the fifth or sixth and half again or nearly twice as broad as long on the no.1691. SHVENTEEN NEW SPECIES OF CRINOIDS—CLARK. 648 eighth or ninth; next three or four joints similar, the length then very slowly decreasing, the joints in the middle of the cirrus being squarish and those in the distal part about twice as broad as long; eighth, ninth, or tenth a transition joint; shortly after the transition joint the median part of the distal dorsal edge begins to become prominent; this very slowly increases in height, arising from pro- gressively more and more of the dorsal surface of the joints, which become progressively more and more carinate, so that in the terminal forty-five or fifty the dorsal surface is produced into a sharp, thin keel, straight in front, convex posteriorly, the outer edge parallel with the median line of the cirrus, in height equal to about one-third the diam- eter of the joints which bear them; opposing spine small and blunt, arising from the entire surface of the penultimate joint, the apex subterminal or central, in height equal to about one-third the diam- eter of the penultimate joint; terminal claw small, about equal in length to the penultimate joint, stout, and moderately curved. The cirri are rounded in the basal third, then becoming strongly com- pressed laterally and, when viewed from the side, somewhat broader. Ends of the basal rays visible as dorso-ventrally elongated tuber- cles in the angles of the calyx; a deep and narrow cleft between the ‘radials and the centro-dorsal; radials very narrow, convex proxti- mally, concave distally, with a small, sharp tubercle in the median part of the proximal border; [Br, about three times as broad as long, the proximal border convex, the distal concave, in close lateral appo- sition, and extending rather well up into the angles of the calyx; the lateral edges are more or less denticulate, and there is a low, though sharp, serrate median keel; IBr, (axillary) slightly longer than broad, shield-shaped, the posterior border: produced into a rounded projection incising the IBr,, the anterior edges concave, the anterior angle somewhat produced, the lateral edges rather strongly denticulate; it bears a sharp serrate median keel in the proximal two-thirds; I[Br 4 (3+4), strongly convex dorsally, in close appo- sition and sharply flattened like the IBr series, the lateral edges somewhat produced and strongly denticulate; ITBr,,, centrally con- stricted with the lateral angles produced as in the other species. Twenty arms, 115 mm. long; first brachial short, slightly longer exteriorly than interiorly, interiorly united, somewhat incised by the second, which is nearly twice as large and has a rounded posterior projection; these two brachials, like the IBr, and ., have a slightly marked median carination; third and fourth brachials (syzygial pair) not quite so long as broad, somewhat constricted centrally; next five or six brachials almost oblong, about twice as broad as long, the surface rather strongly concave, then becoming wedge-shaped and soon triangular, nearly as long as broad, and after the middle of the arm wedge-shaped again and about as long as broad. The arms are 644 PROCEEDINGS OF THE NATIONAL MUSEUM. VOL. XXXVI. at first evenly rounded dorsally, but after the basal third they gradu- ally become compressed and more sharply rounded dorsally, and in the outer half very narrow and very sharply rounded dorsally, though not really carinate; after the basal third of the arm the brachials. develop shghtly projecting and finely spinous distal edges. The dorsal (but not the dorso-lateral) side of the fourth and following brachials is covered with fine short spines, which gradually become coarser after the proximal third of the arm and tend to arrange themselves into longitudinal lines; joints of the division series and arm bases with strongly denticulate borders. Syzygies occur between the third and fourth brachials, again between the twenty-fifth and twenty-sixth to thirty-fifth and thirty-sixth (usually in the vicinity of the twenty-ninth) and distally at intervals of five to seventeen (usually seven to ten) oblique muscular articulations. P, 12 mm. long, moderately stout in the proximal half, but becom- ing slender distally, with about twenty joints, all of which are approx- imately as long as broad, and the basal two-thirds of which are strongly carinate; P, 10 mm. long, similar to Pp, but less stout basally; P, 6 mm. long, much more slender than P,, tapering evenly from the base to the tip, with fifteen joints, the proximal four or five squarish, then longer than broad, and about twice as long as broad terminally. P,, similar, 6 mm. long; P, and following pinnules 5 mm. long with about thirteen joints, less slender distally than the preced- ing; the joints have shght overlapping spines developed on the distal edge along the dorsal crest; distal pinnules 10 mm. long, rather slender, with about twenty joints, the first short and crescentic, the second trapezoidal, about as broad distally as its median length, the following half again as long as broad, the terminal four or five dis- proportionately small; the dorsal crest is sharp and somewhat spinous. Ty pe-specimen.—Cat. No. 20A.= £492 , Indian Museum; Malay Archipelago; 30 fathoms. Cotype—No. 25484, U.S.N.M.; from the same locality. This species is named for Dr. N. Annandale, the superintendent of the Indian Museum, through whoses courtesy the exceptionally inter- esting collections of that institution have been sent to me for study. Subfamily CHARITOMETRIN 42. Genus CRINOMETRA A. H. Clark. CRINOMETRA PULCHRA, new species. Cirri XX-XXIV, 18-20, moderately slender, 30 mm. to 40 mm. long. Ends of the basal rays visible as rather large tubercles in the angles of the calyx; radials concealed, or at most forming a /\-shaped ridge over the ends of the basal rays; IBr, very nar- no.1691. SHVENTEEN NEW SPECIES OF CRINOIDS—CLARK. 645 row, chevron-shaped or crescentic, or entirely concealed; [Br, (axil- lary) large, rhombic, half again to twice as long as broad, the edges all around smooth and prominent, with a moderate rounded median carination; II Br 2, the first very short, the axillary rhombic, about twice as broad as long; [11 Br 2, similar to the I1Br, developed interiorly in 1,2,2,1 order. The division series are perfectly smooth dorsally, in close lateral apposition and sharply flattened; the edges of the component joints are slightly prominent, and the axillaries have a slight broadly rounded median ridge, most pronounced on the first. One specimen has one IiBr series, and one III Br series 4 (3-+4). Thirty arms, 150 mm. long, resembling, except in ornamentation, those of the other species of the genus; after the third or fourth brachial strongly overlapping distal ends are developed, the middle of which is swollen into a broad tubercle which may extend back- ward to the proximal end of the joint; after the thirtieth brachial this gradually disappears. The pinnules are essentially as in the other species of the genus. Type-specimen.—Cat. No. 25473, U.S.N.M.; from Albatross Sta- tions Nos. 2319-2350, off Havana, Cuba; depth between 33 and 279 fathoms. CRINOMETRA MARGARITACEA, new species. Cirri XX, 13-15, 20 mm. long. Ends of the basal rays visible in the angles of the calyx, bearing one or more long tubercles; radials concealed; I Br, very short, five or six times as broad as long, the edges parallel and slightly curved; IBr, (axillary) rhombic, about two and one-half times as broad as long; IIBr 2; I11Br 2, developed interiorly, but never present in the full series. The division series and first two brachials are shghtly convex dorsally and are in close lateral apposition and sharply flattened later- ally; the first eighteen or twenty brachials are also sharply flattened laterally. The axillaries.and preceding joints are separated in the outer part of their contiguous surfaces, forming rhombic water pores; the first and second brachials are similarly separated interiorly. The ornamentation consists of moderately large blunt tubercles dis- tributed evenly over the surface of the division series, becoming gradually less marked after the second brachial and disappearing altogether at about the end of the proximal fourth of the arm. The IiBr and I11Br series and the first two brachials have a low but prominent rounded narrow median carination; this is continued onto the arm bases in the shape of prominent median tubercles on each joint which disappear at about the end of the proximal fourth of. the arm. : Twenty-one to twenty-nine arms, resembling in structure those of other species of the genus. Proc.N.M.vol.xxxvi—09——45 646 PROCEEDINGS OF THE NATIONAL MUSEUM. VOL. XXXVI. The pinnules are of the type common to most of the species of the genus, but are somewhat more slender, the genital pinnules not being so much expanded. T'ype-specimen.—Cat. No. 25472, U.S.N.M.; from Albatross Sta- tion No. 2154, off Havana, Cuba; 310 fathoms. CRINOMETRA CONCINNA, new species. Cirri XX, 14-18 (usually 15 or 16) 25 mm. to 30 mm. long. Ends of the basal rays visible in the angles of the calyx, bearing one or more long tubercles; radials concealed; IBr, very short often more or less concealed by the centro-dorsal, curved and band- like or narrowly crescentic; IBr, (axillary) rhombic to approxi- mately triangular, two and one-half times as broad as long, the lateral edges as long as those of the IBr,, and often, like them, re- duced to a point; IIBr 2 (once 4 (3 + 4) and twice 4 united in two synarthrial pairs in eight specimens) ; I1Br 2, developed interiorly ; edges of the joints to the third brachial everted and raised, usually broken up into high blunt tubercles which intermingle more or less with similar high blunt, more or less confluent tubercles on the dorsal surface of the joints; division series (except the IBr) and first two brachials usually with a high, rather narrow, median ridge, higher than the tubercles on the dorsal surface of the joint; this is sometimes partially or entirely broken up into two or three dorso-ventrally elongate tubercles, larger than any of the others on the joints. The proximal edge of the axillaries and the inner proximal edge of the second brachial are curved upward, while the distal lateral angles of the joints preceding the axillaries, and the inner distal angle of the first- brachials, are cut away, leaving prominent openings, which serve as water pores. The division series are only very slightly convex dorsally, and are in very close lateral apposition; the first sixteen brachials are flattened laterally. Thirty arms 150 mm. -long, resembling those of other species of the genus; the lower brachials to about the fifteenth have strongly everted distal ends, which are usually more or less crenulate, or may be tubercular; there is usually a prominent central tubercle, dorso- ventrally elongate, and also some more or less obsolete tubercles on the dorsal surface; from the fifteenth onward the brachials are almost perfectly smooth dorsally. Ty pe-specimen.—Cat. No, 25476, U.S.N.M.; from A/lbatross station No. 2342, off Havana, Cuba; 201 fathoms. CRINOMETRA INSCULPTA, new species. Cirri XX, 15-18, 25 mm. to 30 mm. long. Ends of the basal rays visible in the interradial angles‘as a cluster of high tubercles, with difficulty separable from the similarly modified no.1691. SHVENTEEN NEW SPECIES OF CRINOIDS—CLARK. 647 surface of the surrounding skeletal elements; radials concealed; IBr, usually concealed except in the angles of the calyx; very short; IBr, (axillary) triangular, three or four times as broad as long; I1Br 4 (3+4) and 2, usually both in the same specimen, but the former always in the majority; IIIBr 2 (1 + 2), or 2 after a I[Br 2 series (rarely, when developed exteriorly, 4 (3 + 4) or 4 (1 + 2;3+ 4)), developed interiorly in 1, 2, 2, 1 order, but never present in the full series. The elements of the division series are in close apposition, no water-pores being present. The division series and lower brachials are but slightly convex dorsally, and are in close lateral apposition and sharply flattened. The elements of the IBr series are thickly and evenly covered with prominent tubercles resembling those on the dorsal pole of the centro-dorsal. These sometimes arrange themselves in a more or less linear series in the median line, or there may be a more or less distinct median keel, which, however, is never very well marked. This evenly tubercular ornamentation may encroach some- what upon the lower elements of the II Br series, and always extends a considerable distance up into the angles of the calyx and between the I1Br series, narrowing to a point anteriorly, as does the somewhat similar ornamentation in Mariametra subcarinata. The elements of the I11Br and II1Br series and the lower brachials have more or less (usually strongly) crenulate or tubercular edges, and the dorsal sur- face usually bears a few small scattered tubercles; along the median line they bear large and prominent, dorso-ventrally elongate, narrow, dorsally rounded tubercles, which form a conspicuous narrow cari-_ nation. The lower brachials have very strongly tubercular or dentate distal ends, in the center of which is a single large tubercle, these large tubercles forming a median line of prominent tubercles, which continues the carination of the division series out onto the arms, gradually dying away and disappearing at about the end of the proximal fourth. The prominent eversion of the distal edges of the brachials becomes distally less and less strongly dentate, at the same time becoming less and less erect, until at about the twentieth brachial it becomes merely a moderately marked, finely spinous overlap, and so continues to the ends of the arms. The brachials to about the twentieth are sharply flattened laterally. The pinnules are as in other species of the genus. Type-specimen.—Cat. No. 25477, U.S.N.M.; from Albatross sta- tion No. 2753, off the windward coast of St. Vincent; 281 fathoms. CRINOMETRA GEMMATA, new species. Cirri XX, 12-15, 20 mm. to 25 mm. long. Ends of the basal rays visible as elongate tubercles in the angles of the calyx, usually covered with short, fine spines; radials con- cealed, or just visible over the ends of the basal rays; IBr, very 648 PROCEEDINGS OF THE NATIONAL MUSEUM. VOL. XXXVI. short and band-like, of uniform height, strongly curved, the proximal edge everted and dentate, and with a row of small pointed tubercles, sometimes more or less confluent, midway between the anterior and posterior borders; IBr, (axillary) rhombic, twice and one-half as broad as long, the anterior and posterior angles approximately equal, the lateral edges about equal to those of the IBr,; I1Br 4 (8+4) (in one specimen twice 2); III Br 2 (142), but only present in a single instance, developed interiorly. The division series are in close lateral apposition and are sharply flattened laterally; they are strongly convex dorsally, so that the dorsal portion of Pp is exposed. The division series and arms to the fourteenth or eighteenth brachial are thickly covered with numerous uniform, small, sharp, conical tubercles, which exhibit a tendency to arrange themselves in hori- zontal rows; these are more numerous and more slender along the edges of the division series. Seen without a glass, the proximal por- tion of the animal has the appearance of being finely and evenly granulated. Nineteen to twenty-one arms, 100 mm. to 125 mm. long, resembling, except for the basal ornamentation as described, those of other species of the genus. The pinnules are essentially as in the other species. Ty pe-specimen.—Cat. No. 25474, U.S.N.M.; from A/batross station No. 2330; off Havana, Cuba; 121 fathoms. Family ANTEDONID. Genus PSATHYROMETRA A. H. Clark. PSATHYROMETRA MIRA, new species. Centro-dorsal conical, rounded at the apex, 4 mm. broad at the base and 4 mm. high, divided into five radial areas by five shallow interradial furrows, each equal in width to nearly or quite the diame- ter of the adjacent cirrus sockets; cirrus sockets closely crowded, regularly arranged in two converging columns in each radial area, with a single socket, the remnant of a third column, between the distal ends of the first sockets of the outer columns, which come to- gether just beneath it. Cirri XL, lacking in both specimens. Ends of the basal rays visible as small tubercles in the angles of the calyx, but with difficulty separable from the general surface of the centro-dorsal and radials; radials even with the edge of the cen- tro-dorsal in the median line, but extending up in the angles of the calyx and entirely separating the bases of the IBr,; IBr, oblong, shghtly over twice as broad as long, evenly rounded dorsally and laterally; IBr, (axillary) broadly pentagonal, about as long as broad, the lateral edges not quite so long as those of IBr,, convex, the lateral angles somewhat produced outward. no.1691. SHVENTEEN NEW SPECIES OF CRINOIDS—CLARK. 649 Ten arms, all broken off at the base in the two specimens at hand; first brachial slightly wedge-shaped, about twice as broad as its ex- terior length, entirely free interiorly; second brachial considerably larger, approximately oblong, not quite so long as broad; third and fourth brachials (syzygial pair) not quite so long as broad; the re- mainder of the arms and the pinnules, so far as can be judged from the fragments, are similar to those in other species of the genus. The synarthrial tubercles are very slightly marked. Type-specimen.—Cat. No. 9G=7%!2, Indian Museum; lat. 11° 31’ 40’’ N., long. 92° 46’ 40’’ E.; 188-220 fathoms. | Cotype.—Cat. No. 25485, U.S.N.M.; from the same locality. Genus MASTIGOMETRA A. H. Clark. MASTIGOMETRA MICROPODA, new species. Centro-dorsal low hemispherical, 4 mm. in diameter at the base, the polar area slightly convex or flattened; cirrus sockets closely crowded, very numerous, in four or five alternating rows. Cirri L-XC, 16, about 10 mm.*long; first two joints short, rather over twice as broad as long, third as long as broad to about one-third longer than broad, fourth and fifth slightly longer; succeeding joints subequal, about as long as broad; third to sixth joints slightly “ dice- box shaped,” the remainder with the ventral surface practically straight and the dorsal with a slight median concavity (in lateral view) ; no trace of dorsal spines or overlap; cirri becoming somewhat compressed in the distal two-thirds, and therefore appearing very shghtly broader in lateral view; opposing spine represented by a shght tubercle, terminally situated, which may be obsolete. Scattered calcareous granules are present along the disk ambulacra, and single interradial plates may be present between the IBr,. Radials even with the edge of the centro-dorsal; IBr, very short, five or six times as broad as long, of uniform height, not quite in apposition basally, the lateral edges diverging distally; IBr, (axillary) triangular, about half again as broad as long, the anterior angle somewhat produced, the proximal border as long as the proxi- mal edge of the IBr,. Ten arms, probably about 80 mm. long, their structure being the same as in J/. flagellifera. The distal intersyzygial interval is three oblique muscular articulations. P, 15 mm. long, much stouter basally than the succeeding, though tapering to an exceedingly slender and delicate flagellate tip; P, 9 mm. long; following pinnules gradually decreasing in length. The pinnules are of the same proportions and structure as are those of M. flagellifera. Ty pe-specimen.—Cat. No. 14 H., Indian Museum; “ ?India.” Cotype.—Cat. No. 25486, U.S.N.M.; “ ?India.” 650 PROCEEDINGS OF THE NATIONAL MUSEUM. VOL. XXXVI, The only specimen in the collection with a definite locality is a small and much broken one, which was dredged off Colombo Light, Ceylon, in 263 fathoms. Family PENTACRINITIDA. Genus HYPALOCRINUS A. H. Clark. HYPALOCRINUS SPRINGERI, new species. Stem slender, 4 mm. in diameter, rounded-pentagonal in cross section, the sides smooth, flat, or very slightly convex; interarticular pores extending to the eighth node; internodals, 10 (rarely 9 or 11) of equal size, each face slightly over twice as broad as high; nodals slightly longer than the internodals, the small transversely oval cirrus sockets touching the distal (lower) border and extending upward to the proximal fourth of the joint face; neither the supra- nor infra- nodals are modified in any way. i Cirri slender and delicate, twelve times the diameter of the stem (48 mm.) in length, with 50 joints; first joint very short, the following gradually increasing in length to the fourth, which is twice as broad as long, and further increasing to the sixth, which is about as long as broad; following joints slightly longer than broad, but in the ter- minal fourth becoming again about as long as broad; from the twen- tieth or twenty-third joint onward small but prominent median dor- sal tubercles are developed; terminal claw small and blunt, conical, twice as long as broad at the base, slightly longer than the preceding joint. Infrabasals present, resembling those of Jsocrinus decorus; basals prominent externally, rhombic in outline, just contiguous by their lateral angles, strongly convex exteriorly, bearing from one to three prominent tubercles; in dorsal view the basals form a figure similar to that made by the basals of /socrinus decorus ; radials large, strongly convex proximally, slightly concave distally, about half again as broad as long, ornamented with a few coarse, high, tubercles, irregu- larly placed; IBr, oblong, about twice as broad as long, without orna- mentation; the lateral edges are just in apposition, but are not flat- tened; they are cut away somewhat anteriorly and posteriorly, form- ing small rhombic pores on the lines of articulation between the IBr, and the radials, and the IBr, and ,; [Br, (axillary) short and broad, triangular, twice and one-half as broad as long, the anterior edges everted and produced into a high scalloped ridge; IBr, and , united by syzygy; IIB 2, the distal edges of the joints standing out in high prominent scalloped ridges; II1IBr 4 (3+-4), the distal edges of the I11Br,, ., and , forming high scalloped vertical ridges. About twenty-five arms 140 mm. long, the terminal 30 mm. being slender and with only very rudimentary pinnules, as in Metacrinus no. 1691. Sy Aa hsealad NEW SPECIES OF CRINOIDS—CLARK. 651 and in JZ. naresianus ; fe eeachial very y obliquely wedge-shaped, the distal edges Tonnage a straight line with those of adjacent first brachials, and standing out in a high scalloped vertical ridge or bear- ing two or three high tubercles, the interior edges entirely united; second brachial smaller, wedge-shaped, about twice as long out- _ wardly as inwardly, the distal ede everted as in the preceding; following brachials obliquely wedge-shaped, about twice as broad as long, after about the twelfth becoming oblong, at first half again as broad as long, gradually increasing in length, after about the middle of the arm being about as long as broad, and in the terminal portion half again as long as broad; the great eversion of the brachials gradually dies away as the joints become oblong, giving place to a slight prominence of the distal edge of the brachials, which in the terminal portion of the arm becomes a rather strong overlap. Syzygies occur between the second and third or third and fourth brachials (more rarely between the fourth and fifth), again between the fifteenth and sixteenth to thirty-first and thirty-second (usually in the vicinity of the twentieth), and distally at intervals of from four to nineteen oblique muscular articulations, the interval being long in the proximal, short in the distal part of the arm. The pinnules are in general nee those of H. naresianus. Ty pe-specimen.—Cat. No. — z!- , Indian Museum; lat. 13° 47’ 497” N., long. 73° 07’ 00’’ E.; 636 ean Cotype.—Cat. No. 25487, U.S.N.M.; from the same locality. This species is dedicated to Mr. Frank Springer, the eminent authority on the Crinoidea. HYPALOCRINUS ORNATUS, new species. In general like H. springeri, but a smaller and more. delicate species. Stem as in HZ. springeri, but only 3 mm. in diameter; cirri propor- tionately more slender, 30 mm. long (ten times the stem diameter) with 40 joints, the dorsal tubercles commencing at about the seven- teenth; basals as in /socrinus decorus, without ornamentation; radials without dorsal ornamentation, but with the distal edges everted and produced into a high, thin, scalloped overlapping ridge; IIBr 4 (3-+4). Eighteen to twenty arms, about 95 mm. long from the radials. Ty pe-specimen.—Cat. No. £524, Indian Museum; Andaman Sea; 200 fathoms. Cotype.—Cat. No. 25488, U.S.N.M.; from the same locality. DRAGONFLIES OF THE MISSISSIPPI VALLEY COL- LECTED DURING THE PEARL MUSSEL INVESTIGA- mons ON THE MISSISSIPPI RIVER, JULY AND AUGUST, 1907. By Cuartes Brancu WILson, Department of Biology, State Normal School, Westfield, Massachusetts. INTRODUCTION. Many opportunities were afforded for the collection of dragon- flies in connection with the pearl mussel investigations on the Missis- sippi River and its tributaries during the summer of 1907. So far as practicable these opportunities were improved and: a list is here presented of the different species obtained, with their geographic and seasonal distribution. Some of the territory visited had been previously worked over by dragonfly investigators in a more thorough and satisfactory manner, but much of it also was new and is here reported upon for the first time. Then, too, all the previous work had been disconnected, confined to fl single State or even a single locality, and hence there was not the same chance for correlation and comparison. The present is the first attempt, so far as known, to collect from any considerable extent of the Mississippi River and its tributaries; and while it is confessedly deficient in many particulars, it neverthe- less affords a general outlook that may be of some value. The itinerary of the trip, so far as dragonfly collecting was concerned, was as follows: The time between July 6 and 12 was spent at St. Paul in examining some of the numerous small lakes with which that city and Minne- apolis are surrounded. Leaving St. Paul on the 12th, a run was made down the Mississippi to Prescott, Wisconsin, where the party remained until the 15th. On that date the St. Croix River was ascended as far as Stillwater, Minnesota, where a stop of twenty-four hours was made, during which time dragonflies were collected from both banks of the river and from a small lake in the outskirts of the town. Returning to Prescott the next forenoon and continuing down the Mississippi, the par stopped at La Crosse, Wi isconsin, PROCEEDINGS U. S. NATIONAL AKUSEUM: VoL. XXXVI—No. 1692. 653 654 PROCEEDINGS OF THE NATIONAL MUSEUM. VOL. XXXVI. from the 20th to the 22d, and reached McGregor and Prairie du Chien on the 25th. From Prairie du Chien on the 26th the Wisconsin and Kickapoo rivers were ascended as far as Wauzeka, where the banks of the Kickapoo and the marshy land adjacent to them were thoroughly examined for dragonflies. A return was made to McGregor the — same afternoon. Again continuing down the Mississippi, the party stopped at Muscatine, Iowa, for twenty-four hours and reached Bur- lington in the same State on August 3. Here a stop was made until the 6th, and again both banks of the river were thoroughly examined for specimens. From Burlington the Mississippi was descended to Grafton, Ili- nois, which was reached on the 11th. On the 12th the Illinois River was ascended as far as Hardin, Illinois, and a return was made on the 13th, collecting along both banks. On the 13th also a run was made down the Mississippi to the mouth of the Missouri River, which was ascended for a few miles and back again, the party reaching St. Louis on the evening of the 13th. Leaving there the morning of the 14th, a long run was made down the Mississippi to Cairo, [lhnois, at the mouth of the Ohio River, which was reached on the 19th. During this entire run not a solitary dragonfly was seen except one or two specimens of Libellula pulchella Drury flying across the river, The Ohio was then ascended on the 20th to Paducah, Kentucky, at the mouth of the Tennessee River. The next day the party started up the Tennessee and reached Riverton, Alabama, on the 24th. Here a stop was made till the 26th and considerable collecting was done. We started back down the river on the 26th for Paducah, where the trip ended on August 30. The Mississippi River was thus covered from St. Paul to Cairo through the States of Minnesota, Wisconsin, Iowa, Missouri, and Illinois, and there were included also 40 miles of the St. Croix River, 20 miles of the Wisconsin and Kickapoo rivers, and 40 miles of the Illinois River. The Ohio River was then covered from Cairo to Paducah, between the States of Illinois and Kentucky, and the Ten- nessee River through the States of Kentucky and Tennessee into Alabama, making in all a distance of nearly 2,000 miles. | With the exception of a partial duplicate series, which the author was kindly allowed to retain and for which his sincere thanks are tendered to the honorable Commissioner of the Bureau of Fisheries, all the specimens collected are now in the U. S. National Museum and admit of ready reference. A few notes descriptive of the general physical characteristics are given under such of the localities as seem to demand them. Where No. 1692. 5 ovaiehiseimabaes OF THE MISSISSIPPI VALLEY—WILSON. 655 they are omitted it is to be understood that the conditions were those usually found along the river—high and heavily wooded banks with a cleared area at the immediate margin of the river, more or less covered with tall grass and weeds. LIST OF SPECIES. I, LAKH AMELIA, MINNEAPOLIS, .MINNESOTA, JULY 6 AND 7. This is a small lake west of Minneapolis which serves as one of the feeders of the Minnehaha River. It is surrounded by high and dry banks which are entirely cleared, leaving only here and there a bush or tree. The immediate shores are covered with a dense growth of grass and weeds, back of which are cultivated fields. 1. ANAX JUNIUS (Drury). Common around the shore and inland half a mile or so; a very strong flier and hard to capture except when mating. 2. TETRAGONEURIA SPINIGERA (Selys). Common flying along the shore and very close to it inland; frequently alights and can then be easily captured. 3. LIBELLULA PULCHELLA Drury. Common not merely around the lake but far inland over the potato and corn fields; too wary to be easily caught. 4. LIBELLULA EXUSTA Say. A single male captured near the shore, the dorsal surface of whose abdo- men was already (July 6) deeply pruinose. 5. GOMPHUS SPICATUS Hagen. A single male captured in the grass along shore; no others seen, although carefully searched for. 6. LEUCORHINIA INTACTA (Hagen). Abundant everywhere along shore and for some distance inland. 7. CALOPTERYX MACULATA (Beauvois). Found in the gorge of the Minnehaha River, the outlet of Lake Amelia, below the falls; fairly common. 8. CALOPTERYX AZQUABILIS Say. Found in the same gorge, but not as common as C. maculata. g. LESTES INEQUALIS Walsh. Two specimens seen and one captured; not common. ro. LESTES UNCATUS Kirby. A single male captured in the grass near the shore. tr. ISCHNURA VERTICALIS (Say). Fairly common in the grass alongside the lake; only orange females found, no black one. 12, ENALLAGMA EBRIUM (Hagen). ; Found in company with the preceding species everywhere, even at a dis- tance from the lake shore in the woods and cultivated fields; females as common as the males. 13. ENALLAGMA HAGENI (Walsh). Found everywhere, fairly swarming in the grass along shore and for some distance inland; eaten by Gomphus spicatus. 656 PROCEEDINGS OF THE NATIONAL MUSEUM. VOL. XXXVI. II. LAKE PHALEN, ST. PAUL,*MINNESOTA, JULY 8. This is a small lake northeast of St. Paul and partially outside the city limits. Its western and northern shores are covered with prime- val woods, while the southern and eastern banks are low and swampy in places and clothed with dense underbrush. At the northwest corner the lake is connected with another much smaller lake or pond by a short stream which winds through an intervening strip of low marshy ground. The banks of the smaller pond are densely wooded, except a narrow strip along the shore and around the outlet. On the stream connecting the two bodies of water and around the shores of the smaller pond the dragonflies were especially abundant, and most of the species recorded were collected there. The species of 7Vetragoneuria was the one most abundant on Lake Phalen itself. 1. ANAX JUNIUS (Drury). Fairly common around the shore; several seen closely enough for satisfac- tory identification, but none captured. 2. HSHNA JUNCEA VERTICALIS (Hagen). Fairly common; a female caught off the side of the trolley car just as it stopped; body very beautifully colored when alive, but fades almost immediately after death; actively feeding along the shore rather than over the water. 3. LIBELLULA PULCHELLA Drury. Very common; sexes about equally abundant; two females captured by hand, which had evidently recently emerged from their pupa cases. 4. LIBELLULA QUADRIMACULATA Linnezus. Two females secured along the shore; the only ones seen, 5. CELITHEMIS EPONINA (Drury). A single pair captured, which were the only ones seen. 6. GOMPHUS VILLOSIPES Selys. Common, squatting on the bare ground, logs, and rocks; found in company with G. spicatus, but is considerably larger; strong and pugnacious, catebes und eats the smaller dragonflies like Leucorhinia and NSym- petrum. 7- GOMPHUS SPICATUS Hagen. Three captured, all females; smaller than preceding but habits similar; feeds largely on damselflies, like Hnallagma and Ischnura. 8. TETRAGONEURIA CYNOSURA (Say). Common everywhere; all secured were males; hovers over the water but rarely alights; very pugnacious, attacking and driving away ‘(fomphus and even 4?shna. No spinigera seen at this lake, no cynosura at Lake Amelia. g. ERYTHEMIS SIMPLICICOLLIS (Say). Two pairs secured, both sexes in full color and not yet beginning to become pruinose, to. LEUCORHINIA INTACTA (Hagen). Common everywhere, the sexes about even in numbers. no. 1692. DRAGONFLIES OF THE MISSISSIPPI VALLEY—WILSON. 657 11. LEUCORHINIA PROXIMA Calvert or FRIGIDA Hagen. 12. 13. 14. 15. 16. 17. 18. 19. 20. Both sexes secured. Males with yellow spots along the back of the thorax and the first five abdominal segments; seventh, eighth, and ninth seg- ments much dilated; two large yellow spots on the ventral surface of second segment; lower appendages fused with a much shallower noteh between them than in intacfa. Females with the basal half of the wings, or at least to the inclosing of the triangle, colored ved; dorsal yellow spots like the male but not as distinct; ventral surface of second Lo Seventh segments becoming pruinose. SYMPETRUM RUBICUNDULUM (Say). A single pair secured; no others seen; easily distinguished even at a dis- tance by its brilliant red color. ENALLAGMA SIGNATUM (Hagen). Quite common flying about the floating alge. ENALLAGMA CARUNCULATUM Morse. Fairly common, but not as plentiful as #2. ebrium and EF. hageni. ENALLAGMA EBRIUM (Hagen). Both sexes very common in the grass along shore. ENALLAGMA HAGENI (Walsh). Most common of all the damselflies; both sexes found everywhere in the grass and weeds along shore. ENALLAGMA ANTENNATUM (Say). Rare, only a few seen; found on rushes over the water. ISCHNURA VERTICALIS (Say). Quite common in company with Hnallagma hageni; both sexes secured. LESTES INEQUALIS Walsh. Rare, only a single female secured. LESTES VIGILAX Hagen. A little more numerous than ZL. inequalis; mostly males. III. BEAVER LAKE, ST. PAUL, MINNESOTA, JULY 10. Nearly east from St. Paul, a small lake with a portion of the banks high and sandy and a portion low and swampy, everywhere covered with a dense growth of vegetation, underbrush, weeds, and grass. The two dragonflies which were most abundant were found on the high sandy banks, while the damselflies were captured in the low and wet places. I. LIBELLULA EXUSTA Say. Abundant everywhere, most common with quadrimaculata in the under- growth close to the shore. When it alights it squats like a Gomphus on the rocks, stumps, and even on the ground. It is gregarious, as many as fifteen or twenty alighting on the same spot; it is also inquis- itive and many were caught that actually alighted inside the net as it was being carried. The males are predominant and are all pruinose thus early, even the two antehumeral stripes showing clear white. 2. LIBELLULA QUADRIMACULATA Linnzus. Everywhere in company with the preceding; when it alights it does not squat but perches on a twig, holding its body horizontal even if the twig is vertical. It is gregarious, like the preceding species, from 658 PROCEEDINGS OF THE NATIONAL MUSEUM. VOL. XXXVI. To. TI. T2. twelve to fifteen or twenty alighting on the same stalk or twig. It is not wary, but neither is it inquisitive like the preceding: species. These two species of the genus were present in great numbers, the others but sparingly. LIBELLULA PULCHELLA Drury. Only a few seen and none of them secured. LIBELLULA LUCTUOSA Burmeister. . Two females secured in fine color. - ANAX JUNIUS (Drury). A single pair secured close to the shore. TETRAGONEURIA SPINIGERA (Selys). Common everywhere, but not in great numbers like JLibellula erusta and L. quadrimaculata. - LEUCORHINIA INTACTA (Hagen). Common, but in small numbers; at Curve Lake, which is badly polluted by drainage from some harvester works in the immediate vicinity, this was the only dragonfly to be seen. . LEUCORHINIA PROXIMA Calvert, or FRIGIDA Hagen. This is the same species as was found at Lake Amelia; only a few speci- mens were seen, and these were all secured. ENALLAGMA HAGENI (Walsh). Very common; hundreds secured by a single sweep of the net through the long grass along shore. ENALLAGMA EBRIUM (Hagen). Not as common as the preceding; about in the proportion of one to ten. NEHALENNIA IRENE (Hagen). Both sexes fairly common, but not in such numbers as /schnura posita. ISCHNURA POSITA (Hagen). Found in company with Hnallagma hageni and E. ebrium everywhere; not as plentiful as the former, but more so than the latter. IV. MISSISSIPPI RIVER, BETWEEN ST. PAUL AND HASTINGS, JULY 12. . LIBELLULA PULCHELLA Drury. Not common; only a few seen flying across the river. . LIBELLULA QUADRIMACULATA Linneus. Common on the bluffs on the east bank of the river; both sexes secured. . LIBELLULA LUCTUOSA Burmeister. Both sexes captured upon the same bluffs on which L. quadrimaculaia was taken. PLATHEMIS LYDIA (Drury). Not common, only two specimens secured, both females. - GOMPHUS FRATERNUS (Say). A couple of males were secured from the river bank just below St. Paul. LEUCORHINIA INTACTA (Hagen). Common everywhere along both banks of the river. . LEUCORHINIA PROXIMA Calvert or FRIGIDA Hagen. The same species as previously recorded, obtained from the bluffs along the east side of the river. . TETRAGONEURIA CYNOSURA (Say). Common everywhere, flying over the water and along the banks, no. 1692. DRAGONFLIES OF THE MISSISSIPPI VALLEY—WILSON. 659 9. TETRAGONEURIA SPINIGERA (Selys). AS common as the preceding; some dead ones seen floating in the water, with fish jumping for them. ro. ARGIA APICALIS (Say). Both sexes found around the heaps of clam shells on the east bank of the river near Pleasant Farms. ir. ARGIA MSTA PUTRIDA (Hagen). Both sexes found in company with A. apicalis. 12. HZSHNA JUNCEA VERTICALIS (Hagen).’ Many seen flying across the river and along the east bank. 13. ANAX JUNIUS (Drury). Many seen flying across the river and along both banks. It was noted that the species of 7etragoneuria mated most often during the hour preceding sunset. Many couples could then be seen flying over the water, and they would approach the boat and even alight on it. Gomphus, Anaw, and A’shna were each seen plunging into the water after insects. Anav went in like a kingfisher, sub- merging its whole body and evidently grasping the insect with its feet. None of them apparently wet its wings while doing this, at least not enough to hinder it at all in its flight. V. PRESCOTT, WISCONSIN, JULY 13 TO 15. 1. LIBELLULA PULCHELLA Drury. Not common, only a few seen flying across the river. 2. LIBELLULA QUADRIMACULATA Linnezus. Only four individuals of this species seen. 3. ZESHNA CONSTRICTA Say. Common flying about in the woods near the St. Croix River; four females ‘secured. It alights on the sides of tree trunks or hangs vertically downward from the underside of a twig or a leaf, and in this position quietly munches the insect it has secured. It is more active toward night, coming out of the woods and flying about over the water. 4. GOMPHUS VASTUS Walsh. Common on the rocks along the shore of the river; all that were secured proved to be males. 5. GOMPHUS FRATERNUS (Say). Not as common as G. vastus; the two that were secured were females. 6. DIDYMOPS TRANSVERSA (Say). Two males caught in the woods along shore, others seen but only in a single restricted locality. 7. ARGIA MCESTA PUTRIDA (Hagen). Both sexes common along the rocky shore close to the water. 8. ARGIA VIOLACEA (Hagen). Both sexes common in company with A. masta putrida. g9. LESTES RECTANGULARIS Say. Three males taken in the grass along the banks some distance from the water, where it was shady; four other males were taken along a slough on the opposite side of the river. 1o. GOMPHUS EXTERNUS Selys. A single male taken on the river bank in front of the town. ‘X sy) 660 PROCEEDINGS OF THE NATIONAL MUSEUM. VOL. XXXVI. 11. LIBELLULA LUCTUOSA Burmeister. Both sexes found at a small pond south of the town. 1z. PLATHEMIS LYDIA (Drury). Both sexes found at the same pond with Libellula luctuosa. 13. CALOPTERYX AQUABILIS Say. A single pair seen in the slough opposite Prescott. 14. HETHRINA AMERICANA (Fabricius). A single specimen seen on the island opposite the town. 15. ENALLAGMA HAGENI (Walsh). Sparsely scattered aiong the river’s edge. 16. ENALLAGMA EBRIUM (Hagen). A few found with FH. hageni. VI. STILLWATER, MINNESOTA, JULY 15. On the St. Croix River; the banks of the river are high and dry and well wooded, except an area just opposite the town where formerly stood a large sawmill. The refuse accumulating from this null has formed a terrace along the river’s edge elevated well above the water and without a shred of vegetation anywhere upon it. 1. LIBELLULA QUADRIMACULATA Linnzus. Found by the hundreds in the old lumber yard on the bank of the St. Croix opposite Stillwater; every stick, stub, and bush alive with them. They were very tame, alighting not merely on the net but also on the hand and arm and all over the clothing. This and the other four species here listed were the only dragonflies seen. 2, LEUCORHINIA INTACTA (Hagen). Common, but not nearly as numerous as Libellula guedrimegculata, 3. ARGIA TIBIALIS (Rambur). Both sexes flying about in the open sunshine in company with Lidellula quadrimaculata and Leucorhinia intacta, 4. ARGIA APICALIS (Say). A few males found in company with A. tibialis. 5. PLATHEMIS LYDIA (Drury). Both sexes found on the river bank a little below the lumber yard. VII. LILY LAKE, STILLWATER, MINNESOTA, JULY 16. A small sheet of water on the high ground to the west of the town; its western and northern banks are covered by dense underbrush, the eastern and southern banks cleared and occupied by dwellings. From the southeast corner proceeds a small outlet, winding about through soft, marshy land. The dragonflies were most abundant along this cutlet and on the margin of the lake in its immediate vicinity. 1. EPICORDULIA PRINCEPS (Hagen). A few seen patrolling the shore; one male captured. 2. LIBELLULA LUCTUOSA Burmeister. Both sexes quite plentiful in one restricted area at the northwest corner of the lake. 3. LIBELLULA PULCHELLA Drury. : Common, many of the females just out of their pupa cases, NO. * f 1692. DRAGONFLIES OF THE MISSISSIPPI VALLEY—WILSON. 661 10. II. I2. 13. 14. ERYTHEMIS SIMPLICICOLLIS (Say). Common everywhere. . PLATHEMIS LYDIA (Drury). Fairly common, many of the females just emerged from their pupa cases, LEUCORHINIA INTACTA (Hagen). Common everywhere; the most numerous species seen. TETRAGONEURIA SPINIGERA (Selys). A single female taken and one or two others seen. DOROCORDULIA LIBERA (Selys). Several seen flying about over the small stream which serves as the outlet to the lake; distinguished readily by its inflated abdomen; hard to eatch, but both sexes secured. LIBELLULA QUADRIMACULATA Linnezus. Found in company with Dorocordulia libera and quite common; but it is very wary here, and it was extremely difficult to secure even a single specimen. VIII. RED WING, MINNESOTA, JULY 17. LIBELLULA PULCHELLA Drury. Several seen flying across the river. ANAX JUNIUS (Drury). Many seen patrolling the river banks. PLATHEMIS LYDIA (Drury). Both sexes seen along the river bank just above town. GOMPHUS EXTERNUS Selys. A single pair captured on the river bank. - GOMPHUS VASTUS Walsh. Common everywhere; most of the specimens secured were males. - GOMPHUS CRASSUS Hagen. A single female secured in company with G. vastus. GOMPHUS FRATERNUS (Say). Both sexes fairly common. GOMPHUS AMNICOLA Walsh. A single female secured in company with G. fraternus. LESTES INEQUALIS Walsh. Both sexes common in shady places near the woods. ARGIA TIBIALIS (Rambur). Both sexes common along the river bank. ARGIA APICALIS (Say). ; Found in company with A, tibialis, but not as plentiful. ARGIA MCESTA PUTRIDA (Hagen). A few individuals found along the river bank. ENALLAGMA HAGENI (Walsh). Found in the grass along the river bank. ENALLAGMA EBRIUM (Hagen). Found in company with H. hageni. IX. WINONA, MINNESOTA, JULY 19. - LIBELLULA PULCHELLA Drury. Several seen flying across the river. Proc.N,M.vol.xxxvi—09——46 N\ \ ° 662 PROCEEDINGS OF THE NATIONAL MUSEUM. VOL, XXEVL. 2. PLATHEMIS LYDIA (Drury). Both sexes taken along the river bank. 3. ARGIA MGESTA PUTRIDA (Hagen). A few in the open spaces close to the water’s edge. 4. ISCHNURA POSITA (Hagen). A few taken in company with EL. hageni and EF. ebrium. 5. ENALLAGMA HAGENI (Walsh). Fairly common in the tall grass along the river banks. 6. ENALLAGMA EBRIUM (Hagen). Found with H. hageni, but not so numerous. X. HOMER, MINNESOTA, JULY 20. Very few dragonflies or damselflies seen; a few individuals of Libellula pulchella Drury, Anaw junius (Drury), and Plathemis lydia (Drury) observed flying across the river. Lestes vigilax Hagen, L. rectangularis Say, and Argia moesta putrida (Hagen) taken in small numbers along the river banks. XI. REEDS LANDING, MINNESOTA, JULY 18, 1. ANAX JUNIUS (Drury). Common patrolling the shore or flying over the water. 2. GOMPHUS FRATERNUS (Say). Both sexes captured along the sandy shores. 3. GOMPHUS VASTUS Walsh. More specimens, including both sexes, of this species were here secured than at any other locality on the river; the banks of the river with alternating reaches of sand and gravel seemed peculiarly attractive to these dragonflies. 4. SYMPETRUM ALBIFRONS (Charpentier). Both sexes were captured in the tall weeds along the edge of the woods; they seem to prefer shady spots. 5. SYMPETRUM RUBICUNDULUM (Say). A single male was found in company with S. albifrons. 6. CALOPTERYX MACULATA (Beauvois). Both sexes common along the steep banks where the grass reaches to the water’s edge and there is a swift current. 7. CALOPTERYX AZQUABILIS Say. Both sexes of this damselfly were found in company with C. maculata, but were not so numerous. 8. LESTES VIGILAX Hagen. Both sexes found in the tall weeds and grass back from the water; not very common. 9. LESTES RECTANGULARIS Say. Both sexes secured in company with L. vigilax. to. ARGIA MCESTA PUTRIDA (Hagen). A few specimens seen along the river banks. XII. LA CROSSH, WISCONSIN, JULY 20 TO 22. Just to the north of the city the banks of the Mississippi are low and swampy and traversed by numerous streams and bayous. The no. 1692. DRAGONFLIES OF THE MISSISSIPPI VALLEY—WILSON. 668 railroad tracks cross and recross this region in many directions and afford a convenient means of reaching localities that would otherwise be inaccessible. Much of the collecting was done along these railroad tracks. 1. LIBELLULA PULCHELLA Drury. Common everywhere around the outskirts of the town. 2. PLATHEMIS LYDIA (Drury). A single colony, including both sexes, of this species was discovered at a small pond just north of the railroad tracks; none was seen anywhere else. 3. ANAX JUNIUS (Drury). Very common over the marshes and along the La Crosse River above the city. 4. EPICORDULIA PRINCEPS (Hagen). A few seen patrolling the banks of the La-Crosse River. 5. SYMPETRUM RUBICUNDULUM (Say). A few individuals secured in the edge of the woods back of the railroad tracks. 6. GOMPHUS FRATERNUS (Say). - Both sexes captured on the gravel along the river bank. 7. GOMPHUS VASTUS Walsh. In company with G. fraternus and more numerous. 8. PERITHEMIS DOMITIA (Drury). ’ Both sexes obtained at the lake in the park. 9. ENALLAGMA HAGENI (Walsh), E. EBRIUM (Hagen), and E. SIGNATUM (Hagen). Found together in the long grass and weeds along the river bank; the first-named species the most abundant. to. LESTES RECTANGULARIS Say and L. VIGILAX Hagen. Found a little distance back from the water, near the woods. 11. ARGIA MGSTA PUTRIDA (Hagen) and A. APICALIS (Say). Found in the shrubbery along the water’s edge, the last mentioned the most abundant species. 12. ISCHNURA VERTICALIS (Say). A few found in company with the species of Enallagma. XIII. BROWNSVILLE, WISCONSIN, JULY 23. After leaving La Crosse the only Neuroptera seen were at Browns- ville, Wisconsin, Crosby’s Slough, Minnesota, and Victory, Wiscon- sin. At each of these places the high and wooded banks yielded three species of Argia, namely mwsta putrida (Hagen), tibialis (Rambur), and apicalis (Say), their relative abundance being in the order named. Where the shores became sandy and less steep two species of Gomphus, vastus Walsh and fraternus (Say) were predominant, flying over the water and patrolling the banks. Apparently these two genera did not intermingle to any extent, but each was colonized by itself. XIV. LANSING, IOWA, JULY 24. Only two species of Gomphus, vastus Walsh, and externus Selys, were seen at this station or along the river above and below it. 664 PROCEEDINGS OF THE NATIONAL MUSEUM. VOL. XXXVI. XV. PRAIRIE DU CHIEN, WISCONSIN, JULY 25. Again only two species of dragonflies seen, Gomphus eaxternus Selys, and Plathemis lydia (Drury) ; both of these were fairly com- mon. The river banks were low, flat, and sandy, and in the scattered weeds were obtained Argia tibialis (Rambur), A. apicalis (Say), and Jschnura verticalis (Say), none of the three at all common. XVI. HORSESHOE LAKE, OPPOSITE McGREGOR, IOWA, JULY 26. Another excellent example of colonization, Perithemis domitia (Drury), was found here by the hundreds, the males out on the lily pads in the open lake, the females in the weeds along the shore. This species was:seen at only a few other places, and then only sparingly. E'picordulia princeps (Hagen), and Libellula pulchella Drury were the only other dragonflies seen; the former were much the more numerous and were patrolling the pickerel weed and rushes along the shore. The lily pads and rushes were further tenanted by three species of E’nallagma, hageni (Walsh), ebriwm (Hagen), and signatum (Hagen), none of them at all numerous. In addition there were a few specimens of /schnura verticalis (Say). XVII. WAUZEKA, WISCONSIN, JULY 26. This was 15 miles up the Wisconsin River and a half mile up the Kickapoo River, on the banks of the latter. The ground was all marsh land, soggy and wet, with standing water everywhere. Here was found a colony of Plathemis lydia (Drury), both sexes of which were present in large numbers, flying about over the water. The only other dragonflies seen were Gomphus externus Selys, a pair of which were captured in one of the dry spots on the marshes, while others were seen flying over the water and along the flat, dry banks of the Wisconsin River. There were also a few males of Perithemis domitia (Drury) seen on the marshes, and Libellula pulchella Drury flying over the fields farther back from the water. Of the damselflies, Hetewrina americana (Fabricius) was found along the grassy banks of the Wisconsin River in limited numbers, while Lestes vigilax Hagen was found in the rushes on the marshes. XVIII. GUTTENBURG, IOWA, JULY 27. Here was a flat, dry shore, raised considerably above the water and covered thickly with weeds. On it were found Libellula luctuosa Burmeister and Gomphus externus Selys, while flying about over the water were /'picordulia princeps (Hagen) and a few males of Perithemis domitia (Drury). Of damselflies there were found Enallagma hageni (Walsh), Ischnura verticalis (Say), and Lestes rectangularis Say. xo. 1692. DRAGONFLIES OF THE MISSISSIPPI VALLEY—WILSON. 665 XIX. CLINTON, IOWA, JULY 30 AND 31. From Guttenburg to Clinton the shore’ was dry and sandy and yielded nothing but Argia tibialis (Rambur) and Argia apicalis (Say), and the two species of Gomphus, vastus Walsh and externus Selys. ~ At Send Prairie, Illinois, the sand was raised in high bluffs along the shore, and here the two species of Gomphus were specially abun- dant. Elsewhere even the species of Avgia were scarce and no other kinds were seen. At several of the landings just above Clinton not even a single specimen of dragonflies or damselflies could be found by careful and long-continued hunting. Taken all in all, this was the most barren section of the river encountered during the entire season, except that between St. Louis and Cairo. XX. LE CLAIRE, IOWA, JULY 30. The shores at this place were high, dry, and sandy, and there were very few dragonflies or damselflies to be seen, H'pecordulia prin- ceps (Hagen), Tetragoneuria cynosura (Say), Perithemis domitia (Drury), and Argia tibialis (Rambur) comprised all that could be found, and of the first three only a single specimen was seen. XXI. MUSCATINE, IOWA, AUGUST 1. Here the shore was low and covered with a rich growth of weeds, in which there was the greatest variety of dragonflies and damselflies, in the smallest space, of any locailty on the river. Only a few moments could be spent in collecting, but in that time 14 species were secured. 1. ZESHNA JUNCEA VERTICALIS (Hagen). Two specimens taken in the high bushes back from the shore. 2. EPICORDULIA PRINCEPS (Hagen). A single specimen seen patrolling the river bank. 3. LIBELLULA -PULCHELLA Drury. Common everywhere along the banks and over the water. 4. LIBELLULA LUCTUOSA Burmeister. A few males found in the weeds along shore. 5. GOMPHUS AMNICOLA Walsh. A single pair captured on the rocks at the water’s edge. 6. ERYTHEMIS SIMPLICICOLLIS (Say). Males common on the river bank, but only a few females seen. 7. PACHYDIPLAX LONGIPENNIS (Burmeister). A single specimen taken in the thick undergrowth. 8. PERITHEMIS DOMITIA (Drury). A single female captured at the water’s edge; no others seen. 9. SYMPETRUM VICINUM (Hagen). Fairly common along the edge of the woods back from the river bank. 10, ARGIA MCESTA PUTRIDA (Hagen). ; Common everywhere in the dry and open places on the banks. 666 PROCEEDINGS OF THE NATIONAL MUSEUM. VOL. XXXVI. tr. ARGIA TIBIALIS (Rambur). Found in company with A. masta putrida, and about as common, 12. LESTES VIGILAX Hagen. A single specimen taken in the thick weeds. — 13. ISCHNURA VERTICALIS Say. Also rare; a few specimens secured in a patch of long grass. 14. ENALLAGMA GEMINATUM Kellicott. Rare; a few found in company with J. verticalis. At a landing a couple of miles above the town on the river bank an hour’s search revealed nothing but Argia mesta putrida, and seem- ingly the locality was fully as favorable as this other one. XXII. BURLINGTON, IOWA, AUGUST 3 TO 6. H . ANAX JUNIUS (Drury). Several seen and one found mutilated on the shore. 2. HSHNA CLEPSYDRA Say. A single male captured in the woods near the river. 3. TETRAGONEURIA CYNOSURA (Say). Several seen along the water’s edge. 4. LIBELLULA PULCHELLA Drury. Common everywhere, particularly along the western bank of the river. 5. PLATHEMIS LYDIA (Drury). Rare; only a few males seen. 6. GOMPHUS DESCRIPTUS Banks. A couple of females secured on the rocks near the river. 7. ERYTHEMIS SIMPLICICOLLIS (Say). Common on both banks near the water. 8. PERITHEMIS DOMITIA (Drury). A single pair taken on the west bank. 9. PACHYDIPLAX LONGIPENNIS (Burmeister). More common on the eastern bank of the river. 1o. SYMPETRUM RUBICUNDULUM (Say). Found in company with Pachydiplax longipennis on the eastern bank of the river. 11. ISCHNURA VERTICALIS (Say). Common everywhere in the grass along the water's edge. 12. ARGIA TIBIALIS (Rambur). Found in the more open places and very ‘common. 13- ARGIA APICALIS (Say). Found with A. tibialis and nearly as common. 14. ARGIA MSTA PUTRIDA (Hagen). Common on the western bank, but none could be found on the eastern side; prefers the rocks and sand along the water’s edge. 15. ENALLAGMA ANTENNATUM (Say). A single male secured in company with Jschnura. 16. HETHRINA AMERICANUA (Fabricius). A single male was secured from O’Connell slough which had escaped from its pupa case so recently that its color was not yet defined. There was no evidence of colonization here, but a fairly even dis- tribution of all the species. Several individuals of Libellula pul- no. 1692. DRAGONFLIES OF THE MISSISSIPPI VALLEY—WILSON. 667 chella Drury were observed going to roost for the night in the tall ironweed along a dried-up overflow bottom. When roosting they flatten back against the vertical stem of the weed instead of holding their bodies horizontal as is done when they alight in the daytime, possibly as a protection against rain. This was not the right kind of a shore for Gomphus, and hence only the single pair was seen. XXIII. QUINCY, ILLINOIS, AUGUST 9. Only a half-hour could be spent here, and in that time the following species were either seen or secured: Libellula pulchella Drury, Ashna juncea verticalis (Hagen), Pachydiplax longipennis (Bur- meister), Gomphus amnicola Walsh, Argia tibialis (Rambur), and A. mesta putrida (Hagen). XXIV. HANNIBAL, MISSOURI, AUGUST 10. Two hours in the afternoon and the same period the next forenoon were spent here in collecting, but with limited results. There were hundreds of Libellula pulchella Drury flying across the river and over the inland fields, but the only other species found were Gomphus externus Selys, G. amnicola Walsh, Pachydiplax longipennis (Bur- meister). Argia mesta putrida (Hagen), A. tibialis (Rambur), and A. apicalis (Say), and of these there was only a single specimen of each of the first three. XXV. THE ILLINOIS RIVER, AUGUST 12. In passing up the river from Grafton to Hardin two distinct colonies of Hrythemis simplicicollis (Say) were found. The first was 10 miles above Grafton, where the east bank of the river was covered with hundreds of this species, including both sexes, while many were flying across the river. The other colony was 4 miles farther up the river, at the head of an island. Here the island seemed to be the headquarters from which the dragonflies flew out in every direction. Just below Coon Creek an abundance of Libellula pulchella Drury was observed, and they could be seen flying over the inland fields. Repeated observations seemed to indicate that in general the flight was from the shady to the sunny side of the river, from east to west in the forenoon and from west to east in the afternoon. XXVI. COON CREEK, ILLINOIS, AUGUST 12. This was between the colony of Zibellula and one of Erythemis, and there were found here, naturally, these two species, though in limited numbers, and beside them Pachydiplax longipennis (Bur- meister), fully as numerous as either of the preceding, Celithemis eponina (Drury), Tetragoneuria cynosura (Say), and Gomphus plagiatus Selys. 6638 PROCEEDINGS OF THE NATIONAL MUSEUM. you. xxxvt. Of damselflies there were the two species of Argia, apicalis (Say) and tibialis (Rambur). XXVII. HARDIN, ILLINOIS, AUGUST 12 AND 13. Here were found a few specimens each of Libellula pulchella Drury, Zetragoneuria cynosura (Say), Anax junius (Drury), and Gomphus amnicola Walsh, together with large numbers of /schnura verticalis (Say), and a few males of H’nallagma piscinarium Wil- lhamson. XXVIII. MISSISSIPPI RIVER FROM GRAFTON TO CAIRO, ILLINOIS, AUGUST 183 TO 20. No stops were made between Grafton and the mouth of the Mis- sourl River, but the dragonflies and damselflies were as common as they had been and could be seen along either bank flying over the water or in the bushes. A run was made up the Missouri for 8 or 10 miles and back, but not a solitary dragonfly was seen, and this con- tinued all the way down to Cairo. Repeated landings were made and the banks diligently searched for specimens, but without finding even one. This abrupt demarka- tion is no doubt due to the muddy water poured in by the Missouri River. No dragonfly larva could rightly be expected to live in such a medium, and their absolute refusal is what might naturally be looked for. XXIX. JOHNSONVILLE, TENNESSEE, AUGUST 21. On ascending the Ohio River the dragonflies began to appear again, and were as numerous as ever on reaching Paducah, at the mouth of the Tennessee River. For the entire length of this latter river to Riverton, Alabama, Gomphus was particularly abundant and could be seen at all hours of the day flying over the water. The first stop for collecting was made at Johnsonville, and here were found Libel- lula pulchella Drury, Macromia teniolata Rambur, Erythemis sim- plicicollis (Say), Pachydiplax longipennis (Burmeister), Plathemis lydia (Drury), Argia masta putrida (Hagen), A. tibialis (Ram- bur), A. violacea (Hagen). The shore at this particular place was not suitable for Gomphus, and none was secured. XXX. SAVANNAH, TENNESSEE, AUGUST 23. Here the shores were favorable for Gomphus and three species were caught—vastus Walsh, notatus Rambur, and one undetermined. No other dragon flies seen. XXXI. RIVERTON, ALABAMA, AUGUST 24 TO 26. The banks of the river were high and dry, except in one place in the outskirts of the town, where were a few small swampy ponds. no. 1692. DRAGONFLIES OF THE MISSISSIPPI VALLEY—WILSON. 669 Here were found large numbers of Libellula pulchella Drury with Plathemis lydia (Drury) and a species of J/acromia. Along the river bank pulchella was not as numerous, and there were associated with it Hrythemis simplicicollis (Say), Pachydiplax longi- pennis (Burmeister), Jacromia teniolata Rambur, Gomphus vastus Walsh, G. notatus Rambur, two species of /7etwrina, one of which was americana (Fabricius), Argia tibialis (Rambur), Argia violacea (Hagen), and a species of Anaw, of which none could be obtained. SUMMARY. Certain facts must be kept in mind while endeavoring to summarize these observations. 1. With few exceptions the examination of each locality was con- fined to a period of only a few hours duration. Hence the species obtained would represent the fauna of the locality for that day only, and would give but few suggestions in reference to its fauna at other times, or to seasonal changes. 2. The dates for each of the localities examined were different. While this would have little practical influence for neighboring locali- ties visited within a few days of each other, it would mean a great deal when the interval was increased to a month, or even two months. 3. There was a continual progress in the localities visited from Minnesota, one of the extreme Northern States, to Alabama, one of the extreme Southern. Hence the geographic changes would cause considerable differences in the fauna, irrespective of the seasonal changes, and by thus combining the two their separate influence would be much augmented. In spite of these difficulties, however, there are certain conclu- sions which may be fairly drawn from the observations which have just been recorded. 1. A small fresh-water lake or pond, surrounded by shrubbery and vegetation, furnishes the ideal breeding place for dragonflies and damselflies, with which even such a river as the Mississippi, with its numerous sloughs and bayous, is scarcely worthy of comparison. The larve of these insects evidently prefer clean to muddy water as a medium in which to live; in witness whereof may be cited the fact that not a solitary specimen of the Neuroptera was seen on the Mis- sourl River or on the Mississippi between the mouth of the Missouri and the mouth of the Ohio. 2. Only a single species was found in all the localities visited. This species, Libellula pulchella Drury, may therefore be taken as the most widely distributed in the Mississippi Valley, both geographically and seasonally. A close second was furnished by Pachydiplax longi- pennis (Burmeister), which appeared in nearly all the localities. Furthermore, neither of these species was found colonized anywhere. 670 PROCEEDINGS OF THE NATIONAL MUSEUM. VOL. XXXVI. 3. The genus Gomphus is chiefly the guardian of the river. The species may be seen at all times of day patrolling the river’s surface with tireless vigilance or squatting upon the shore and watching their surroundings intently, and woe betide the luckless insect that comes within their reach. All of the species observed are remarkably alike in their habits so that it is practically impossible to distinguish them until after they are caught. This genus also was found universally distributed and not colonized. 4. The other dragonflies and the damsels, on the contrary, were found in colonies, each made up of a few closely related species that harmonize well with one another and restricted in its area with fairly well-defined borders. A few of these colonies are worthy of special mention. A. The first was at Beaver Lake in St. Paul and was made up of Libellula exusta Say and L. quadrimaculata Linneus for the dragon- flies, and Hnallagma hageni Walsh for the damsels. The lake is small and surrounded by a scattering growth of underbrush and rank grass. There were hundreds of the dragonflies among the bushes and shrubs, while the tufts of grass were so loaded with Enallagma that a single sweep of the net secured over two hundred. While other species were found, as given in the list, it was only after long and careful search and in such small numbers as to count for nothing beside the myriads of the three species mentioned. B. Another colony was found on the bank of the St. Croix River, opposite Stillwater, Minnesota. Here had been formerly a large saw- mill, and the river bank for a long distance was packed with sawdust, bark, and edgings to the depth of several feet. Flying about over this area and alighting on the projecting sticks were swarms of Libel- lula quadrimaculata Linneeus, sometimes a dozen or more on the same stick, and with them were numerous specimens of Avgia tibialis (Rambur), particularly along some piles of old slabs back from the water. The most careful search revealed only three other species, and in such small numbers that they could only be regarded as stragglers, C. A third colony was found in Horseshoe Lake, a part of the river surrounded on three sides by islands and opposite the town of McGregor, Iowa. This was a colony of Perithemis domitia (Drury), and H'picordulia princeps (Hagen), the former flying over the lily pads by the score, the latter patrolling the pickerel weeds and rushes along the shore. The only other dragonfly seen after long and care- ful search was Libellula pulchella Drury, which had evidently come across from the mainland. D. Special mention should also be made of the two colonies of Krythemis simplicicollis (Say) observed on the Illinois River, one vo. 1692. DRAGONFLIES OF THE MISSISSIPPI VALLEY—WILSON. 671 10 miles above Grafton and the other 4 miles farther up the river. These have already been referred to on page 667. This isolation of species into colonies prevailed throughout the entire length of the various rivers visited and in some of the small lakes. It is a very different condition from what is found in other lakes where fifteen or twenty species, or even more, can be secured in a single afternoon; witness Lake Amelia and Lake Phalen. It leads naturally to the next conclusion— 5. With the exception of such genera as Gomphus and Anaa and such species as Libellula pulchella Drury, the individual range of any dragonfly or damselfly is in all probability very small. The members of one of these colonies just noted are probably natives of the locality. They were born there, they spend their lives in hunting the insects that surround the water, they lay their eggs in the same water, and then die. Continued observation of such a colonized area for many years would doubtless reveal much that would be of interest in its bearing upon colonization in general, as well as upon the distribution of species. aa, t eT eee re Oipoet= Re AS neg ed tees tet hs FOUR NEW SPECIES OF THE CRINOID GENUS RHIZOCRINUS. By Austin Hopart Crarg, Collaborator, Division of Marine Invertebrates, U. S. National Museum. In his report upon the stalked crinoids collected by the Challenger, Dr. P. H. Carpenter admitted only two recent species of the genus Rhizocrinus, R. lofotensis, and PR. rawsoni. 'The former is credited with a geographical range extending from the Lofoten Islands and Nantucket (Mass.) southward to Uruguay, and with a bathymetric range of from 80 to 1,900 fathoms; the latter is said to inhabit the Atlantic coasts of southern Europe and northern Africa, and to occur among the outlying groups of islands and throughout the West Indies at depths of from 73 to 1,277 fathoms. The first record of lofotensis is, of course, the original description of Sars, in 1864, when the genus was founded; the first record of rawsoni (under the name ldofotensis) is that published in 1870 by Fischer,* who had obtained specimens off Setuval, Portugal, this antedating by two years Sir Wyville Thomson’s record of the Porcupine specimens, cited as the first by Carpenter. In 1883 Professor Perrier had described a supposedly new crinoid which had been dredged by the 7’ravailleur off Morocco under the name of Democrinus parfaiti. This new genus was strongly criticised by Carpenter, who placed the type-species under the synonymy of Rhizocrinus rawsoni, as understood by him. Two years later (a year after the publication of the Challenger report) Perrier described, under the nanie of /lyocrinus recuperatus, a very remarkable species, in all essentials a Bathycrinus, but having separate basals. This was also subjected to severe criticism by Carpenter. In the next year Korotneff reported the discovery of a large species of “Rhyzocrinus ” in the Straits of Sunda,.where it was easily obtainable before the eruption of Krakatoa.? Aside from a few additional records by Rathbun, Koehler, Chun, Grieg, and Agassiz (the last proving to be a Bathycrinus), nothing @ Actes de la Soc. linn. de Bordeaux, XXVII, p. 351. > Bull. de ’acad. roy. de Belgique [3], XII, p. 558. PROCEEDINGS U.S. NATIONAL Museum, VOL. XXXVI—No. 1693. 675 674 PROCEEDINGS OF THE NATIONAL MUSEUM. VOL. XXXVI. additional was learned in regard to this genus until 1907, when Pro- fessor Déderlein published his report on the- stalked crinoids of the Siboga expedition. In this he includes two new species, one obtained by the German steamer Valdivia off east Africa, the other by the Siboga in the East Indies. The latter, 2. weberi, is possibly the species recorded by Korotneff in 1886. Doctor Carpenter considered both Rhizocrinus lofotensis and R. rawsoni extremely variable in all their characters, and so broad are the diagnoses he gives that there appear to be no hard and fast lines of division between them. This, taken in connection with the enor- mous geographic and bathymetric ranges of the two forms as under- stood by him, led me to believe that each of Carpenter’s species was in reality a composite of several. This opinion was confirmed by an examination of several hundred specimens from the West Indies and Gulf of Mexico in the United States National Museum and in the Museum of Comparative Zoology at Cambridge. Rhizocrinus lofotensis does not occur on the American side of the Atlantic, its place being taken, from Cape Cod to Florida, by 2. ver- villi, which differs in being somewhat larger, generally stouter, with a less expanded basal cone and much shorter columnars. ; RHIZOCRINUS CONIFER, new species. Basals separated by distinct sutures; calyx very long, conical, ex- panding evenly from the top of the stem, the diameter at the distal end of the radials about three times that at the distal (lower) end of the basals; length of the calyx six times the distal diameter of the basals and over twice the distal diameter of the radials; a more or less marked constriction at the level of the base or the middle of the radials. The length of the calyx from the top of the stem to the end of the radials is 11 mm. Stem moderately stout, probably about 150 mm. long; in the type 80 mm. long to the end of the thirtieth columnar ; first columnar very short and discoidal, second about four times as broad as long, third not quite twice as broad as long, fourth about one-third longer than broad, fifth about twice as long as broad, after the seventh or eighth becoming three times as long as broad; second to fourteenth or fifteenth shghtly barrel shaped, then becoming more cylindrical, with very slightly swollen ends. The longest columnars are about 3 mm. long by 1 mm. in diameter. The first brachials are trapezoidal, longer than their proximal width. Ty pe-specimen.—Cat. No. 22679, U.S.N.M.; from Albatross station No. 2756; off Ceara, Brazil; lat. 3° 22’ 00’’ S., long. 37° 49’ 00’’ W.; 417 fathoms; bottom temperature, 40.5° F, \ No. 1693. FOUR NEW SPECIES OF CRINOIDS—CLARK. 675 RHIZOCRINUS BREVIS, new species. 1888. Rhizocrinus rawsoni P. H. Carpenter, Challenger Reports, XXVI, Zoology, p. 267, fig. 19; p. 262 (part); p. 268 (SS. Investigator, 15 miles N. by E. of Panama; 800 fathoms). The two specimens figured by Carpenter, which were dredged by Capt. E. Cole, of the cable ship /nvestigator, off Colon (not Panama), appear to be well worthy of specific recognition. They are not merely an extreme variety of rawsondi as supposed by Carpenter, but represent a definite type, presenting valid characters. The calyx may be described as very short, conical, the width at the distal ends of the radials twice that at the distal (lower) end of the basals, one-fifth greater than the length; the radials are slightly longer than to half again as long as broad; the basals are separated by sutures. The stem is not figured; but it is probably much like that of R. robustus. : The two figured specimens (of which the one represented by figure 19A may be considered the type) are in the zoological department of the British Museum; two others are in the geological department of the same institution. RHIZOCRINUS SABZ&, new species. Calyx four times as long as the diameter of the proximal part of the column, and about three times as long as the diameter across the radials; the basal cup is somewhat swollen, so that the greatest diam- eter of the calyx is between the topmost columnar and the radials, usually nearer the latter, instead of across the radials as usual; this diameter is one-fifth greater than that across the radials, and is 1.8 times the diameter of the upper part of the stem; basals separated by distinct sutures. The calyx measures 4.5 mm. in length from the top of the stem to the distal end of the radials. Stem very stout, as in ?. rawsonii, the longest columnars being 2.6 mm. long by 1.7 mm. broad, with a rather prominent constriction. The arms resemble those of R. rawsonii, and are 19 mm. long from the radials. Type-specimen.—Cat. No. 22700, U.S.N.M., from off Saba, 200 fathoms, taken by Capt. E. Cole, of the cable steamer /nvestigator. RHIZOCRINUS ROBUSTUS, new species. Calyx conical, moderately long, expanding evenly from the distal portion of the basals to the radials; summit of stem proportionately small, causing the calyx to appear disproportionately large; calyx three-fifths to nine-tenths again as long as the diameter at the radials and about five times as long as the proximal diameter of the stem. N D) 676 PROCEEDINGS OF THE NATIONAL MUSEUM. VOL, XXXVI. There is sometimes a slight constriction about the basals or radials, but more commonly none. The basals are separated by distinct sutures. Stem in large specimens 190 mm. to 280 mm. in length, compara- tively slender, with 74 to 106 joints, the longest of which are 1.6 to twice as long as broad, about 3 mm. long. The columnars are all approximately cylindrical, those in the proximal third showing a very slight tendency toward a barrel-like shape, the remainder to a slight central constriction, but neither is as marked as in the related species; the lowest 30 mm. or 40 mm. of the stem has the articulations more or less swollen and produced, this soon giving place to numer- ous fine radicular cirri and stout irregular branching roots. Ty pe-specimen.—Cat. No. 22680, U.S.N.M., from Albatross station No. 2401; Gulf of Mexico, off Pensacola, Fla.; lat. 28° 38’ 30’’ N., long. 85° 52’ 30’ W.; 142 fathoms; green mud and broken shell. R. rawsonti may be at once distinguished from any of the above species by its almost cylindrical calyx, which is twice as long as broad at the radials, the latter dimension being usually less than half again as great as the diameter of the proximal part of the stem. The calyx of 2. parfaiti (which is a perfectly valid species) is more inclined to conical in its shape; the diameter across the radials is twice that across the distal end of the basals, while the length is one and one-half times the breadth at the radials.¢ The species of Rhizocrinus may be conveniently grouped as follows: (1) Basals anchylosed, without sutures: 2. lofotensis, R. verrill. (2) Basals always separated by distinct sutures. (a) Stem comparatively slender, the longer columnars being at least twice as long as broad; calyx distinctly conical: 2. conifer, P. brevis, R. robustus, R. chuni. (b) Stem very stout, the longer columnars but little longer than broad; calyx approaching the cylindrical: 2. rawsoni, R. parfaiti, R. webert, R. sabe. Déderlein has found that the species of Bathycrinus are, like the species of Rhizocrinus, divisible into two groups, one with the basals anchylosed into a solid basal cup or ring, the other with the basals separated by suture. This was, however, known long ago, for Per- rier’s Zlyocrinus recuperatus is a species belonging to the latter group, and was the first species of it to be described. A very good figure of it was published by Perrier in his Explorations sous-marines, page 273, figure 193 (1886). a@The data are taken from the figure published by Professor Perrier in his Explorations sous-marines (1886). DESCRIPTIONS OF TWO NEW SPECIES OF ELECTRIC RAYS, OF THE FAMILY NARCOBATIDA, FROM DEEP WATER OFF THE SOUTHERN ATLANTIC COAST OF THE UNITED STATES. By Barton A. Brean and Atrrep C. Weep, Of the Division of Fishes, U. 8S. National Museum. In a lot of fishes collected by the steamer Albatross of the U. S. Bureau of Fisheries in deep water off the southern Atlantic coast of the United States were found two electric rays, with rudimentary (functionless) eyes, differing specifically one from the other and, like- wise, from the only known species under the genus to which these are now assigned, namely Benthobatis Alcock.* BENTHOBATIS MARCIDA Bean and Weed, new species. Disk broadly ovate, body abruptly narrowed at the caudal edge of the pectoral fins so that the ventrals appear to be inserted entirely on BENTHOBATIS MARCIDA. the tail, which thus appears very long. The vent is just midway between tip of snout and end of caudal fin. The width across pectoral 7Ann. and Mag. Nat. Hist., August, 1898, p. 144. Zool. of R. I. M. S. Investi- gator, Calcutta, 1899, p. 17, pl. xxvi, fig. 1. Type of the genus Benthobatis mores byi. PROCEEDINGS U. S. NATIONAL Museum, VOL. XXXVI—No. 1694. Proc. N. M. vol. xxxvi—09——47 677 678 PROCEEDINGS OF THE NATIONAL MUSEUM. VOL. XXXVI. fins is equal to the distance from tip of snout to caudal edge of these fins. The lax skin makes it difficult to see the true shape of the ven- tral fins. They appear to be less developed than in other members of the family. They are adnate to the tail their entire length and have the rays rather feebly developed. Two dorsal fins are developed, the second much larger than the first, the first dorsal inserted slightly in advance of the caudal edge of the ventrals; caudal large, obovate, with the dorsal edge nearly straight and the ventral edge obliquely rounded; tail with a very distinct lateral fold; eyes want- ing or rudimentary, their position indicated by a pale spot situated 14 cm. in front of the spiracles; teeth flat or concave, broadly rhombic with a small backward projecting point and arranged in quincunx; spiracles large, the edges not fringed or tuberculate; nasal valves confluent into quadrangular flap; a large electric organ devel- oped between the head and pectoral fin on each side. The skin over the entire body is very lax and flabby, making it difficult to see the true shape of the creature and take its measurements. The species is known from a single specimen, a female 49 cm. long, taken at station 2660 by the steamer Albatross of the Bureau of Fisheries, May 3, 1886, at a depth of 504 fathoms, in the course of an exploration of the eastern coast of the United States. Measurements. Om. POtsToleme ay ass eo eee (1975 in.)__ 49.0 Motalelens ths without cad al == eee 40.5 Timor snout to- end of ventrals_.-— <2 Hip, of Shout to:end Of PeCtOLalS == = = = ee PALA Tip. of snout‘to origin of first-dorsal__-—_--.....-=-= == eee 28.5 Tip of snout to origin -of secend dorsal=—__- = Se 32.6 Minot SnOUb LOLS PICACIGS Sc = ee a ee ee 9.8 Tip of snout to nostrils_____--.--____--__-__-=--~-==- === === _—— (EU Tip of snout to first gill slits_______--------_--_--+-____-__________—_ === 12.0 Tip of snout to last gill slits_____________--_-------_____________-__-_— 16. 0 Tin-ob snout 40, MOURN <9 === = eee 8.7 Tip of snout to vent____-_-__-_------_-----~---~------------------------ 24.5 Vent to end of caudal... ee eee Length of first dorsal hase_____---------------------------_------------ 2.8 Length of second dorsal base_—__-_-_---=----L—_---~-___-___---___-__=__= 4.0 Wueneth of Cad alls= 282 S10) Diagonal height of first dorsal from origin to highest point==== 322) === 4.5 Diagonal height of second dorsal_____-__------------~~--_---__----__-___ 6.5 Depth of caudal from highest point to horizontal projection of lowest part_ 6.0 AVG EM ACKOSS PeCtOLAIS: [2 ee ZALES Width eaCrOsSS velleha (So se see 14.0 AVI hbetween SPllaCles= 9-22. 2s Se 4.3 Width petween NOStrilS.. 22" =" 2. ee 3.8 Widtheoft mouths 22 623 ee ae ee 2.0 Widthebetweel frst Cllushtss2 = 22 eee 5.3 Width petween last /eill Slits. == = eee 4,4 Length of opening of each gill slit about___--_---------___-___-__--___-- a0) xo.1694. TWO NEW ELECTRIC RAYS—BBAN AND WEED. 679 Color of dorsal surface, light fawn-color, with a few scattered white spots about 1-2 mm. in diameter. The color becomes lighter toward the edges of the body and fades gradually into the dirty white of the belly. Type.—Cat. No. 62916, U.S.N.M. Marcida, loose, soft, lacking substance. BENTHOBATIS CERVINA Bean and Weed, new species. Disk considerably narrower than long, its width 2.5 em. (one- sixth) less than the distance from tip of snout to end of pectoral fins; this is somewhat wider than is represented in the figure of B. moresbyi,; length of disk slightly less than half of total length; eyes situated about 0.7 cm. in advance of the spiracles and much less re- duced than in B. moresbyi and B. marcida; they may be slightly functional as the orbit seems to be somewhat developed; the external opening is about 1 mm. in length. Teeth rhombic, arranged in quin- cunx, occupying nearly the whole width of the jaw and each tooth: has the surface flat or concave with a sharp point projecting back- ward. Ventral fins about as represented in B. moresbyi; second dorsal much larger than the first and the caudal well developed evenly above and below with posterior margin rounded. In B. moresbyi the first dorsal is represented as being much larger than the second. Nasal valves confluent, forming a quadrangular curtain; an electric ‘battery on each side between head and pectoral fins. Measurements. Om. Snes rg te tY% lean Aen 2 pe SS ee See eee oe Bee eee { seal ; BOSSES SSS SRE Tap Oe oa ee a (18 in.) OLN Tay GUEKCH UG al: 2 ek ee 8 ee ee ee ee ee Pater Men pimOrend Oc Venbraile fins oss. 3148 a ee ee ee es 20. 0 Pom ne LOLen nO lM MeChOral ims asee Ss as Se 15.5 eS eLOLOri Me OLS GOLSAle: = = mes Jy LS oe ee Se ees 18.8 Wen eee LOROriolmeOleSeCOMG COTSH le 9-20. ks ans 2 eS PAL UT SOMES ETE TQ SS aT ep SIS Be eas 2 Ta mga Ree I ef ee in Sp 6.0 ee Tee amie nme TLO Gl ine wre e ey nk oon e R eers Poe ee 4.5 Leitich (eS LDP SU RS a ee ee eS 7.8 TsSSTeRCAEL TRO) TREVSE Pp wear NT SN pee SE a a ee 10.3 TLSGTCVEA AED GCG) TEQUOYDUT yo OS ENG TANS ig hg ads Res ar Se ee 5.3 TL JER RETED, TROD PRPS TOTES he Se Fa ee ah TE SI a a SPS a me 16.3 \VCEEILE (fy CLOG ISTE CURB UE tse ee ee Fs OE ea ce IG} 74 ere eNOSSmMCChORallse = ere omen eo te eet Me eee a 13.0 ONS GPS “SS GUTOISISS. “Sy EteTgr ev RS ae eS ee 9.8 Mardi OInveCMns Oli Glee sae ne See SS St 2 Na ee 2.6 VV DGHELGY Toye rmeeyetey Maks eit PS eS es ee ee 260 COC SEIGU EDS GUE TONY ol ge a eS a Se te FN a a a ee 1.5 Pant HCL Sipe IN een [yee ees ee SS Se Nn eee 4.5 Sohne bie tee Mes ey Str olll eee ae eee ee a SE Ba TIBUO REED, COME” GEAUUITD SUIS TL OY UT ye SY yee cece a 0.6 ene imihiSieC ONS Seas — se ae ee SS ee Se ee 1.6 680 PROCEEDINGS OF THE NATIONAL MUSEUM. vou. xxxvt. Cm. Length of second: dorsal base222~ = t= = a Se Se ee 259) Leneth: of caudal! Se! = 22 ea ae. eee ee ee ee 5. D Depth*:of “careless Sey ek tS eh Se em ae a ee ee ee ce ae. Dingonal ‘height of first dorsal =< =2= 2) 2 ea en 2.5 Didgonal height of second dorsal: 222222222 S eee 3.5 Color.—Upper surface hight fawn color, fading at the edges into the dirty white of the belly; a few white spots scattered over the upper surface of the body; these are -much less prominent than de- scribed in B. moresbyi. Entire body and fins, except the caudal, enveloped in a loose flabby skin. This species is well differentiated from B. moresbyi by the differ- ence in color, greater width of disk, difference in relative size of dorsal fins, and in the greater development of the eyes, which may be due to the less depth of water in which it lives. This species is known from a single specimen, a female 33 cm. (13 inches) long, from station 2664, steamer A/batross, of the U. S. Bureau of Fisheries, lat. 29° 41’ N., long. 79° 55’ W., depth 373 fathoms; bottom coral sand. Collected May 4, 1886. Type.—Cat. No. 62917, U.S.N.M. Cervina, like a deer, from the fawn color. We would acknowledge our indebtedness to Dr. Theodore N. Gill for helpful suggestions in the preparation of these descriptions. ay ia OB he Page. PN IONTE TOM e', clele ass scat se ecccm ces wie s-< 601 Pichardsonte so. sce -:-s-+55=5 +2 soe 427 Acrydium flavo-fasciatum.............-..- 157,158 | Actinometra...........- eee eS 361, 494, 497, 498 IWHEAERNSISS 252 25.22 SB Seuss 392 | CODPINECLED-. Saat sce secs ee 392 | MIIORUN ME ae eee hearse ose kes 494 PCC see Sk ee Sot caged Wee hs sk 506 AMVAGI Tae sets ste eette acess 507 BSGNGLOSH Ie ae toe ae et donee es oes 503 PA BUUDISAYPOLICOS(. 2...) cesanccetke ccs cede 464 ACLOMISCUS CllIDtICUS =< 5- so... 5s nos ows ee 378 ipIperculatus. 3: 225=9s20e eset ee 377 Additional Notes on Mammals of the Rhio- Linga Archipelago, with Descriptions of New Species and a Revised List, by M. W. TUNA ELE Siri Peters Ses oh Se ol ae ili eR cI 479 Additions to the List of Philippine Birds, with Descriptions of New and Rare Species, Meroe MEATS. conse... Shae cceee ss 435 “mmalitis alexandring... 2.2.5... 22.5. .ee 2s 464 SVD A Se eee See ee 464 OPC TEV aha ks UR ee Se i ee 475 f@olacris....... Pa cheiecdy ate ee 42k) SM ot re 114 (STARE Gee Vee eR oa See See eae 114 GCapemaulbie Reo so.t. ses esas ce. kee 114 ectemaciiaba: 2655 \ ley nn sek 114 ES STENT (2) 013) (0 0 ¢ ena 666 GOBSHIC Aes a. 2 one see oe ots hees anes 659 juncea verticalis.......... 656, 659, 665, 667 /ethiopsar cristatellus.......--2----2-Jeeebes 475 AOAIIGRKWALUCISL: 2. cmeee eee lle koce cee aces. 474 Alcedo bengalensis.............-. Une ie sheeree 466 JUICE 1 Se ee a Rae a 494, 498, 506 ICL Tuhe tayo} fs 2 ee eS I Se Se ee Se 498 FPS (Nida ERR eee ae ee 392, 506 iNTSTES SE SESE © en a ener ee ca 144 NRC OILS eee oat See hoe men moos 144 PRP ODM er eren ine ao oe aol es ee 2s SEO 109 IDELAMOTH Se -- 522 5ace ese 677 Australia.—Fresh-water Sponges in the Col- lection of the U. S. National Museum. Part I. Specimens from the Philippines and, by Nelson Annandale...............- 627 Authors ).—On Brazilian Grasshoppers of the Subfamilies Pyrgomorphine and Locusti- ne (Acridine of, by James A. G. Rehn... 109 | Bartsch in the Philippine Islands, Borneo, Guam, and Midway Island, with Descrip- tions of Three New Forms.—A List of Birds collected by Doctor Paul, by E. A. Mearns. 463 | Basilan Island thick-head.........---..----- 442 Basketwork. Anyam Gila (Mad Weave): A Malaysian Type of, by Otis T. Mason..... 385 Bathycopes typhlops:.-2=2------=pee==e——— 174 Bathycrinus 2:2. 55-2 3-5-2 eee eee 362, 673 Bathymetra.-. 222-225-244 eee ee ee 366 Bean, Barton A., and Alfred C. Weed.—De- scription of a New Skate (Dactylobatus armatus) from Deep Water off the South- ern Atlantic Coast of the United States.... 459 —— Descriptions of two New Species of Electric Rays, of the Family Narcobatide, from Deep Water off the Southern Atlantic Coastiof the United States: 2-2 2=- seeaae 677 Bees in the U. S. National Museum.—De- | scriptions of some, by T. D. A. Cockerell.. 411 | Belostomidie..s 2222 526526 282 eee eee ee 561 B6naecus.oacc 22S 8 2 2s yase ae eee ee 532 haldemanum)s 22. 5.2: 2. 2s55- eee 561 Benthobatis..... 2s. .css- seen eee ee eee 677 Gervina ese Sees eee 679 MALCIOA St ac4 i205 43 seeee eee 677, 679 MOTLESD less. ete eee ee 679, 680 Birds collected by Dr. Paul Bartsch in the Philippine Islands, Borneo, Guam, and Midway Island, with Descriptions of Three New Forms.—A List of, by E. A. Mearns.. 463 Birds, with Descriptions of New and Rare Species.—Additions to the List of Philip- pine, by E. A. Mearns......--------------- 435 Biuneuiculatus.:=2-2--- --- === eee 4,16,17 Blatellay s.2fo22 45. ee eee ee ee 526 PerMaNice >. 2-2 3ss- eae oe 555 Bolodon.5-.\..\o-<22 + s--=- hee eee eee 612 Bolteniawei= c.k Uswle. wae ae 398 UAC OQ RUNS Bere 2 oe arate mbm elerctale stoieeacers 399 Céntropusjavanicuse:...2sces.n.4-a. sees 468 VAEUUUS ele Sons Scic iste toe eee eee 468 Cephalacanthidiess.=---2-c.-2-- eee toe eeee 604 Cephalophoneus nasttus....-...-.520----.0-- 472 Gérastospinus VENOSUS:--..2--2----2---6-55- 560 Ceratosaurus nasicornis. - --- - bins ateate antes 227 Wernher. tere see ae os Se ae ote eae 537, 551 Chacoana melanoxantha: <2. 0252. soe anes 413 Charadrius dominicus fulvus.....:...---.--- 477 Gnariiometrinw ts sees see ee cso eeaae 406, 644 Cheirothrix parvimanus.....-.-..----------- 5 Chelonigees \ ss eer se a ee cse af fee eeiece 210 @HipigiegPaAyenSiS! 2 saeco 222s -ealeoweee cece 447 DalAWANGRSIS= 220 oso--2 sneer eee ee 447 SUIMENSISE =. -- cece te a ene eee ones 463,474 China Sea.—A New Squirrel from Direction Island, South, by M. W. Lyon, jr--..-.-...- 509 @HINESeTOSATIOS: 2.2 = han socio lee se coe fas 338 CHInO Re Ne ose soe eee eae rmeate Senses 612 PlicattiSee san eek owen wae se emacs 612,614 Chiselzemowthya@ker eee aceis- eerie 425 Chisternon?interpositumess. 4-2 se= -2 2 194 Chisternon? interpositum, the Latter hitherto Unknown.—Description of Two Species of Fossil Turtles, Toxochelys stenopora and, Dy#Oliverey vila vor eee sea seer 191 Chrysocolaptes lucidus montanus..........-- 468 Gimibe ee he oe es ean ROH see eee 528 AIMELICHM AEE acne a focsee eee eee 567 Cisco, or Lake Herring (Argyrosomus), from the Great Lakes of America; with a Note on the Species of Whitefish.—Descriptions of Three New Species of, by David Starr Jordan and B. W. Evermann......--..--.- 165 G@ishicolacishicel ders se sa= case eee ae eae 471 GxaIS Sah st Gane taoe See See eeae 471 @ijheromaresaliSes = 5 sss2a22 cate ce eras salem 565 Olanipenece tes caeras Jor Saks Sac eeeite ae on ener 607 Comatilia, a Remark- able New Genus of Unstalked Crinoids 361 Descriptions of Sev- enteen New Spe- cies of Recent Crinoids=aseecceee 633 Clark, Austin Hobart. 684 INDEX. Page. Page Clark, Austin Hobart. Four New Species of Comaster mul tiradiatar 2-0-2. ss seeee ee 391, 392 the Crinoid Genus SONUOS@ 522 cose eee ee ee 391 Rhizocrinus. ...... 673: |, Comasteridae sss. 25.) 282) es 391 On a Collection of with Descriptions of New Genera Recent Crinoids and Species.—Revision of the from the Philip- Crinoid Family, by A. H. pine Islands....-.. 391 Clarkew 2 sooth eee 493 Revision of the Cri- Comatellaxeo 5-2 ona 365, 395, 495, 496, 506 noid Family Co- MUGT a, oS aee 45.0 ee Ae ee 395 masteride, with Comatilial 23:23) (es eae 4 aes 365, 495, 496, 497 Descriptions of New Genera and Species: 24 soe 493 Coast of Californiaa—Four New Species of Isopods from the, by S. J. Holmes and M. E. Ce eis a de See he eee 375 Coast of the United States.—Description of a New Skate (Dactylobatus armatus) from Deep Water off the Southern Atlantic, by RAC Bean and-A./C.- Weed. 22.025 .2- =~ 459 Coast of the United States.—Descriptions of Two New Species of Electric Rays, of the Family Narcobatide, from Deep Water off the Southern Atlantic, by B. A. Bean and ACEC. IMMCAU Sa- cpraer: Sey tae eth Sees aren 677 Cockerell, T. D. A., Descriptions of Some Bees in the U. S. National Museum............. 411 — and W. W. Robbins.—Notes on Two Slugs of the Genus Veronicella ...........- 381 Goelioxys: lanceolata 522 os s-ceeceneena as 415 ATH Leese seo, SEs oA en: 415 CeeluTnus Trapilis-- cece act =e Soa ane 198 Collection of Recent Crinoids from the Philip- pine Islands.—On a, by Austin H. Clark.. 391 Collection of Rosaries in the United States National Museum, by Immanuel M. Casan- 0 Ye ee Ran Se ane oe 333 Collection of the United States National Mu- seum. PartI. Specimens from the Philip- pines and Australia.—Fresh-water Sponges in the, by Nelson Annandale.............. 627 Collocalia bartsehig.. 23 sp cctea bee sects cok 463,476 TUGL DBA Pais 23 tes oe eee oan 476 TIBROUIN A Sao aaa So see ne 468 troglodiybess>*-.-.sesssso-% ate aeeee 468 UNniColOr AmMGlis | eases ease eae 476 @olohometrat=- 7 aes: a2 cee ete 362,640 LONIST610) (0) one eRe a ete 640 PSISPINOSA ses. 2325 cseset zene Se 640 Colobometridas- 269. occ cet -Sacteecee taka 640 Colpolophit..25-5-02 0226 SYS seat bases eee 121 obsoleta 5: 54 fa. osesee epee oes 121 | Comsachinias 22 -- Wess soe oa seer Ses 495,496,498 | Comanthis cert soe teres 395,495, 496, 497, 507 altermans. 2.) ssa ee ee 396, 397 DRIAPQUS cec3 ee ate ne Eee eoee 395 Gecameros. 2-2 eee oe eee eae 497 Guplexse heat. sos sees ee 397 HGR OS MPEO Nh tae SI cape a ne ye 507 TG DU ISS Aone eee seer ene eee 395 | POlVEHEMIS >< ese set ee eee 396 | ROvAl Anan. os so os 82 oe see 397, 507 Comaster.......--- 365,391, 495, 496, 498, 501, 503, 506 GOPDINGEL Sao ki< ce eteie cs sise omiewiae ae 502 DOWELISISS ore oieis sinieicisisisleivivie visteoeieeite 362 | | Crinoids.—Comatilia, Comatilia, aremarkable New Genus of Un- stalked Crinoids, by A.H.Clark. 361 iridometriformis. J. 2.222. 555-568 366, 497 Comatala) 2:20.28. 20s ee eee 183, 394, 494, 495, 496,497, 498, 503, 506, 507 Cematula;fimibriata, .3. : ese. 40 = ee eee 392 Tre tiradigtas 1 san. eee ee 506 pectinata-<.25. 2c2se< tee eee 394 solaris... ..230 543 aS eee 497 Comatule consol eet ee a eee 497 Comatuluml..:2..2 sos! Seep eee eee 3 Cominiain:2 shane et eee ececeee eee 495, 496, 497 Comissia.s.2a2g2305-00e eee eee 495, 496, 501 luitkenties: 3:5. [65 eee 501, 502 Copsy chusanindanensis:.-.= 4292 - sees eee 471 Niger sos 32 ses ee eee 475 Coarezonus albus... 22 >. epee ee eee 167,171,172 CISMONLANUS 2) 22. 352. eee 429, 430 clupeaformis.e. 24-65-e eee 171 COURS. 5 eset sae eee 429, 430 Goullerts 222.0! 2222.38 eee 430 labrad oricus. .co<. so ace ese eee 171 WatiOry .2..s22 eee ck eee 172 neohantoniensis -.....<--'-..-sia see eee M2 OTePOniss 3-6 ~Jaeee eeeee 425, 428,429 (Coregonus oregonius) from McKenzie River, Oregon.—Description of a New Whitefish, by David S. Jordan and J. Otterbein Snyder. S625 J o.2.--e sues oe eee eee 425 Coregonus otsego.....-.-- Joa eee eee 72 quadrilateralis. -- =~ -ccmees 168, 428, 430 SApIGISSIMUS-< 2 2c one a eee eee 171 Williamson sec eyes sieeiees 425, 428, 429 Corone philippina.. 22. poems Ae eee eee 475 Coryacris: 3:2 2g eseesss:e=r eee amore eres 111 diversipes: ..22.<-.02 ee eee eee 111 Corydalis: 2.6: 2260525. -29sodeanaaee eee 526 COmuUlad 2. 52s-cneeeoseeeasee 539, 542, 564 Coutiére, Henri, The American Species of Snapping Shrimps of the Genus Synalpheus 1 Cranorrhinus leucocephalus.........--..---- 467 Crinoid.—A New American Jurassic, by Prank Springer. — 2. 2dcncc cash eoe eee 179 Crinoid Family Comasteridz, with Descrip- tions of New Genera and Species.—Revision of the; by: A... Clark. 2-2: 4... ee 493 Crinoid Genus Rhizocrinus.—Four New Spe- cies of the, by. A. El. Clarkin. -9-secee ees 673 a Remarkable New Genus of Unstalked Crinoids, by A. H. Clarks. 5.02 +28: 422 2e525.25 ee eee 361 Crinoids.—Descriptions of Seventeen New Species of Recent, by A. H. Clark..-.-....- 633 Crinoids from the Philippine Islands.—On a Collection of Recent, by Austin H. Clark.. 391 INDEX. 685 Page. Page. MOM GPHO NT Ee Go tae daeewceececcusvessa 644 | Description of a New Skate (Dactylobatus SUSIE pare ean a aes Sic einaienibts 646 armatus) from Deep Water off the Southern DIETETIC le iy els Ree ae as a 647 Atlantic Coast ofthe United States, by B. A. PENNA Sek ae mac encal se 646i |i eBeanvand' A; C. Weed... 2.2... cos sh eacccses 459 DUGOUT Ch Ec a ae a 645 | Description of a New Snake from Panama, by ovrilel ft; eas Rage ee eae 644 ||; Leonhard Stejnoger.........-.-.00-2--s0s- 457 ESTES OY CCAR Ss ae en 411 » Description of a New Species of Leatherback Retreat tetpieame eee sae hed ee 403,642 | Turtle from the Miocene of Maryland, by Benim Blel <<- 24 sks see G22. |), 2W illiam, Palmer. 2: 25° fos cesar seenectace 369 Bi pedatas. sen. a. toc nse 404 | Description of a New Whitefish (Coregonus Crustacean Acanthoniscus Spiniger Kinahan oregonius) from McKenzie River, Oregon, redescribed.—_The Isopod, by Harriet by David 8S. Jordan and J. Otterbein Sny- MICU ANOS UE Se ore a mato. icc an sweets 431 RRR eie a SaaS den ree seed. oaaeisemne ce aa 425 Crustacean, Ancinus depressus (Say).—The Description of Two Species of Fossil Turtles, Isopod, by Harriet Richardson..........-..- 173 Toxochelys stenopora and Chisternon? in- Cryptolopha malindangensis..........------ 440 terpositum, the Latter hitherto Unknown, mindanensis.........---- AA ateaaas th aby iOliversbs Have sa. 00e steno ddcecenee 191 Cumnora (Camptosaurus) dispar....-..-.---- 270 | Descriptions of New and Rare Species.—Ad- Cumnoria (Iguanodon) prestwichii.......--. 289 ditions to the List of Philippine Birds, with, MLSS iC HI sc ees Wick ce ne sce afte 285 Dy: H. NAY M@arnsti2 2. ac. Sach anees aeetasee 435 COREG Aa he 416 Descriptions of New Genera and Species of falrcbuigia Veo.) sock bees oa: 415 Fishes from Japan and the Riu Kiu Is- ON aL relic) 0) ha): re 531, 538, 562 lands; by: On Say Gers. 22.5 See seeeoes 597 CAD SYST: hits os Ses Oe Sea ec 362,399,647 Descriptions of New Genera and Species.— THENCE ee BoE se eee aos 399 Revision of the Crinoid Family Comaster- SUN DEO LG) yeep, ate nd ie me a Se a 642 IG cSsWithy Dy Acrel. Clatks .oe aes aa ee 493 TADRODANESS so ose acm oce ae Sa cease 641 Descriptions of New Species and a Revised Cynopterus montanoi............-.------- 487, 490 List.—Additional Notes on Mammals of the Grornis philippmensiss: 22525. 2.+. oe OR eae eae See 469 DECUOLANIS LS 2 2 Sota tae tee coe 419 Gidley, James Williams.—Notes on the Fos- philippinensis 52 oe 419 sil Mammalian Genus Ptilodus, with De- | Biorion Sarmamems = 623 scriptions of New Species._..._...._......- 611 | SI eo A etal of 623 Gila (Mad Weave): A Malaysian Type of ieplocanthasaurus..... 3 S22 csesces cee ee 146, 147 The Thorax of, by. R. E. Snodgrass........ 511 | furcillata: =< 9. 1 "S225 Jess. ee 147 Interpositum, the Latter hitherto Un- | Jordan, David Starr and Barton Warren known.—Description of Two Species of Fos- | Evermann.—Descriptions of Three New sil Turtles, Toxochelys stenopora and | Species of Cisco, or Lake Herring (Argyro- Chisternon?, by Oliver P. Hay..........-.- 191 somus), from the Great Lakes of America; Tees eee oe te Aa Oe ie Gas BE 328 607 | with a Note on the Species of Whitefish... . 165 BI Ee eo caeiecinitts aatye Sen en eee won 608 | ——and John Otterbein Snyder.—Description (KOM ase ceo tnce se caecke ee ee esos Sasa ser 607 | of a New Whitefish (Coregonus oregonius) jwattl:| CRESS «pe Satna Reb ann Geach conaaese ae 139 from McKenzie River, Oregon...........-- 425 ONTECTISIS 2h Wes eae eae oe eae 139 | Jumbo herring or Erie cisco. ..............-.- 165 Mallides me: oh jae (2 ee aoe oa nae 139 | Jurassic Crinoid——A New American, by Investigations on the Mississippi River, July Frank Springer: . <2. .-. 2. .-s2-eseseee eee 179 and August, 1907.—Dragonflies of the Mis- Jurassic Reptile Camptosaurus, with a Revi- sissippi Valley collected during the Pearl | sion of the Species of the Genus, and De- Mussel, by Charles Branch Wilson .....-.. 653° | scriptions of Two New Species.—Osteology TolBiGiherleepSs-.sssceeesccs sce once eee 470 |. ofthe; by. C. W. Gilmore:-.22:---- sec ceeeen 197 COETCN IR aes eee Se eee en 2 ee 470 | Kinahan redescribed.—The Isopod Crusta- SULMATASCLISIS see ce ce tree nae eae eee 470 | cean Acanthoniscus spiniger, by Harriet SUlATIS A eee ee oe cs eee eS See 470 Richardson: <2 22..c22- 20 sacs pesesee eee 431 iridometta|: |. hee 6 ee ace eeseesae 362,365,408 | Koptorthosoma ceeruleum ....-.....-.-.---- 415 OXQUISItas oso. ane caseseeeoeee 408. |) Eiesvimanuss 252 tees see 1, 4, 16, 17, 53, 55 AAT eee ee hs Sate i 5 ee 362 | Lake Herring (Argyrosomus), from the Great Ischnoptera hyalina: 2-.2-2t Wake Huron Cistos-<-e-- 2+ eee eee 167 ASHCTI: Sins nae c-ee wees aS 184,188,189 | Lakes of America; with a Note on the Species decordssess- oe 187, 188, 189, 410, 650, 651 of Whitefish.—Descriptions of Three New ani Pht ina eee ee etry. ates as 180 Species of Cisco, or Lake Herring (Argy- lenthardit=s-*s8eo.e 42. = ee ae 187 rosomus), from the Great, by David Starr NALCSANUSS oe Meee eee hae 190 Jordan and B. W. Evermann.............. 165 DAKE Me. 5 oe Soe cee re ee oe 188.189" || dualage niger: <2 e222 ae eee ae 470 wyville-thomsoni................ 188,189 | Lamprocorax panayensis............-------2 475 Isopod Crustacean Acanthoniscus spiniger U@OSAULUS SS one enee ce aas ae 216, 260, 261 Kinahanredescribed,by Harriet Richardson 431 (COTSOLS oar ale ae iain lalate erate etree 236 Isopod Crustacean, Ancinus depressus (Say), Wearalenas Mint 32. oe ce= no see eee 144 Dysblarmiet/ Richardson: 2 setecs meen oacces 173 | Leatherback Turtle from the Miocene of Isopod of the Genus Jeropsis from Pata- Maryland.—Description of a New Species gonia.—Description of a New, by Harriet of, by William! Palmer-<-22222-2*e--ecoee 369 ICH ARGSOH tate een aaa ok Sees 491 | Leelotettixs 22-2 steno eee See ee 159 Isopods from the Coast of California—Four Wiridis. =... S.n2css0 sate ee 159 New Species of, by S. J. Holmes and M. E. TLS PISINA Soe gen oe ae eee eee 552 CAG Niet lo aaah oa See eae ha els Pt ee ee 375-| Leptocoma sperata- 02 on. 4es- eee 473 Ixobrychus cinnamomea.............---....-- 465 | Leptegnathus:arguss22- <5. coe eee eee 458 TOLOPSIS WTO VICOLMIS: so. seceeeewieceenisace ccs 491 || Leptomerinthoprora.- 4.2 esses eee eee 149 GUmMVICOMNISe. «Sasceaa. face ae eee 421 sequalisie2 soos Sasesee 150 GOLTAS eee et owas acc aos eee 421 brevipennis...:...... 149, 150 INDEX. 689 Page. Page OMUONEMIASUEL. ©... ecco eee ee 495, 496, 498 | Mariametra subcarinata..................--- 647 VWeNUStus.< oo... 2 reaction de CRE Ab allachyc}5]e4 (2) ba en a eS 623 1 PUSS. 5- oe ee ee 134 | Maryland.—Description of a New Species of BESGiineessess < fs aeeicesnnsdscee 134 Leatherback Turtle from the Miocene of, by [OU SU SO ne ee eres 135 Wino RalmMer sn. sss e tastes ace eee eee 369 POAREIN er tet See tata Satie ease care 135 | Mason, Otis T. Anyam Gila (Mad Weave): PLES ETSY Sohal Sled el le an as 135 A Malaysian Type of Basketwork. ........ 385 ESTAS ING UANSS 25 ss0 Se ceece scene Gop Gols Galen! PM ASHE ONE A... Jeno. tkics a daak aecmecnceoek ne 649 MOGI PUALIS = | oot c cece ee 659, 663, 664 nil 2*ey2\ RU) os ee a tS 649 PUD Ae eae Se tones cece ic CSE sice.c 553, 655 UTC Y OO UME soe sea cere rere oer 649 FEL Slee ee ea ne 51, GOs, O00 | MMBSDISIAs 2725 iaccccs kone seb st oScceeenceaens 142 Leucorhinia intacta............ 655, 656, 658, 660, 661 ROG DElGICeE ante men veceesaee semen 142 WrOstMaa seen css tected = 657, 658 Quadricarinatass =o . to ha2 sacs see 142 inelale Mnripenwms= 2s. So scec. cesses. eee n ce ee 348 MOrGsBTITSapiIlSsseseeee secon ana 207, 227 USMS saa eS eoc cote asa acm 198 Motacilla boarula melanope.....-.-.--------- 474 OCINATIS! 30 =s0 5 Se eee ec ee eae a 474 Mount Malindang flycatcher-warbler-.-.-.-.-- 440 racquet-tailed parrot ..... 437 shorbwing 25%. 45-922 see © 441 SINWOL-OY. Gisee meer e ona aecte 443 wood-accentor .....-.-.-.- 441 NEtITI STD eLeU) abeies ser eee ear eatr 623 NAIM OP rat Sas Gregan iocoeeeebeeee ccm 474 JES 0 ik Sie Pat SC eis ne AS ee ee See o 474 TAY ZUVOLS ser eat stems eer a Safe car serie 474 VIISAS HOR te Aken ees aie ee eae aoe = 483, 490 Pai LamasNlS so-t a stants ars see eto te nsec 490 HOM DOS Saha cooee at ao aes sees 490 CORCGIOD ses Sac ae certo eae Serer ee 490 TIENTS Se sise oas-fote metal See el tee 484, 485, 490 ANEIMRUSSS See ste oe ce ere eee ges 490 jerdoni..... eos eee ae eee seem 483 [Go¥eo1 as) CEES Cee eee Ae py ae pers soso 484, 490 OEE OLS mph cia arcane ee eteere at oer ote =i 490 IMMISCAOLVOLES) ON GBes ae asl annie =i see = = = 464 NPICHAI Ses iee Gos ee eee ees 464 PAMMASBUSIS® oe pe. ceca ei maar 436 Mickerinpli: Joe eae che seam eeios 436 Muscicapula westermanni........-.---.----- 46 Museum.—The Collection of Rosaries in the United States National, by Immanuel M. (WASATOWICZ: tse se sat ornare aa ees 333 Mussel Investigations on the Mississippi River, July and August, 1907.—Dragonflies of the Mississippi Valley collected during the’ Pearl by-C. B. Wilson... 2: ---..-04--- 653 IMG OFS MAN COLA gene mieten tere sate al eee le 488, 491 Mivristichvons i1GOlOIre =e «t-te stalemate a areal 464 Miv zp me ley Til DU eta sect 2 aie lees rele ee 477 IMPreeehiolsonGekt pasar | ac eees ob Sern are tee 173,175 INSMNOSCIUINIS DULCHON 3 2 oceania ee ae 490 Narcobatidee, from Deep Water off the South- ern Atlantic Coast of the United States.— Descriptions of two New Species of Electric Rays, of the Family, by B. A. Bean and BA CMWWCR LYS Seid or. Gee ee ae ee ate ae 677 National Museum. Part I. Specimens from the Philippines and Australia.—Fresh- water Sponges in the Collection of the United States, by Nelson Annandale...........-.. 627 Nehalennia arenes = 26 c2 dec2 feat ak sccn~mninn a 658 EN BIMVASUOD Scot asreee one canine etatae la alata 495, 496, 503 STANGIS oe sce ease es ee cares 503, 504 ING OTMET Sees stu ee Ort ta ets 2 Sot ete 4,6, 35 Neuronia ocellifera................--.-- 539,542,565 New American Jurassic Crinoid, by Frank SjomietiGin: 6 Ai ohapenigHo sae en oomecsscs a: 179 of Philippine Birds, with Descriptions of, 435 New Forms.—A List of Birds collected by Dr. Paul Bartsch in the Philippine Islands, Borneo, Guam, and Midway Island, with Descriptions of Three, by E. A. Mearns.... New Genera and Species of Fishes from Japan and the Riu Kiu Islands.—Descriptions of, by:J,..O.. Snyders s.2 2st. ee as eee New Genera and Species.—Revision of the Crinoid Family Comasteridz, with Descrip- tions:of ny Ay tas Clarke. oestrone New Genus of Arenaceous Foraminifera.— Ammodiscoides, A, by Joseph A. Cushman. New Genus of Unstalked Crinoids.—Coma- tilia, A Remarkable, by A. H. Clark....... New Isopod of the Genus Jeeropsis from Pata- gonia.—Description of a, by Harriet Rich- ardsonss: cdsasSa20 aes eee eee New Skate (Dactylobatus armatus) from Deep Water off the Southern Atlantic Coast of the United States.—Description of a, by B; A. Bean and A.C. Weed=.-522 8-22 .ees New Snake from Panama.—Description of a, by Leonhard'Stejneger-’s. 22. = seen. see New Species and a Revised List.—Additional Notes on Mammals of the Rhio-Linga Archipelago, with Descriptions of, by M. W. 463 597 493 423 361 421 New Species.—Notes on the Fossil Mamma- lian Genus Ptilodus, with Descriptions of, by Js Wie Gidley. Ss... see eee New Species of Cisco, or Lake Herring (Argyo- somus), from the Great Lakes of America; with a Note on the Species of Whitefish.— Descriptions of Three, by David Starr Jor- damand B..W, Eyermannes- 2. = eee eee New Species of Electric Rays, of the Family Narcobatide, from Deep Water off the Southern Atlantic Coast of the United States.— Descriptions of Two, by B. A. Bean and -A...C.- Weed... ..2.2. 223s cesses eee New Species of Isopods from the Coast of Cali- fornia.—Four, by S. J. Holmes and M. E. 611 165 677 New Species of Leatherback Turtle from the Miocene of Maryland.—Description of a, by William: Palmenscc o2-3-5.5 = oe New Species of Recent Crinoids.—Descrip- tions of Seventeen, by A. H. Clark .....-... New Species of the Crinoid Genus Rhizocri- nus.—Four, by A. H. Clark.........-.. a as New Species.—Osteology of the Jurassic Reptile Camptosaurus, with a Revision of the Species ofthe Genus, and Descriptions of Two, by Charles W. Gilmore ..........-. New Squirrel from Direction Island, South China Sea; by Mow. Lyon; ren seessee ae New Whitefish (Coregonus oregonius) from McKenzie River, Oregon.—Description ofa, by David S.Jordan and J. Otterbein Snyder. Nomiajamablliss-2.2. .0--eese soe ee eee chandlerl.282 2555--h asset zeet (Crecisaspidia) chandleri............. Nomia (Hoplonomia) quadrifasciata......... TRAD Phill RR i SE Ci i a ee PRAISE cle Oh 2 thro cis aes jane Aw na)s> ome SOT Oe Se ee ee PN STCTMRT NITES CU BND 0c, 2.5 bu ooa'a 2 pemenicty ~ oae's «oo PGnHHen Ee RAlATONS en sono bbe ew scasenns Note on the Species of Whitefish.—Descrip- tions of Three New Species of Cisco, or Lake Herring (Argyrosomus), from the Great Lakes of America; with a, by David Starr Jordan and B. W. Evermann.............- Notes on Mammals of the Rhio-Linga Archi- pelago, with Descriptions of New Species and a Revised List.—Additional, by M. W. Notes on the Braincases of Iguanodon and Megalosaurus.—On the Skull and the Brain of Triceratops, with, by O. P. Hay......-.- Notes on the Fossil Mammalian Genus Ptilo- INDEX. 691 Page. Page. She Ox VOIBDLOllA sce e e's: ia aise s oGusiepee ee suees 136 419 pumahella:.c 5. 3a. Jc toe 136 419 SHPIBA.: Oo“. ou ud sess Sone 136 419 | Pachydiplax longipennis..................2- 534, 435 553, 665, 666, 667, 668, 669 AAACN VIONUCUNAG oc. c- soe ecco ees Se sl, eoaneee 406 anrusticglye- Ses. sss oeeane ee 407 Towiratiqess: 2: ds dca nt ee 406 PaGanoiakece oe mecciatte theo. sto ek eee 375 Palmas Island fruit-eating pigeon............ 436 165 | Palmer, William. Description of a New Spe- 479 dus, with Descriptions of New Species, by . Vis Wis Gra EE ete ee ee meters 3 Vee Notes on Two Slugs of the Genus Veronicella, by W. W. Robbins and T. D. A. Cockerell. DOGS EN ig oe a ae A es a ee ra variegatus Nycucorax maniwensis..00..-.22.2--. <--aegsniane eels aera 363 iPerdidomorpha brMimMeriis.-..2--- 2s aeae ae 417 HerciutomorpN denne sess oes eee ee eee 411, 416, 417 IPericrocOUlsS: DASTGDESE. -). cceklaes ween 463, 469 Perithemis domitia...........- 663, 664, 665, 666, 670 I BLOMIELT AS Pt. barca cc ie oe ai isla ns eee ees 362 Petropuilamanillensis: -.. 2.52 --c-nene+-2- ieee eal POZOLGUUM~ ois veticad ce ecthc Pec cee eae 158 Pialacnocenan se. shes 2 oh teee a Se eee 516 PHANOPeMIAS ose. 2 acess en- 365, 392, 498, 506, 507 GaNpenterle 5 < So: ban es acesne sees 392 Gelicatalvs. ct: oe a aaes oe canter 393 692 INDEX. Page. Page. Phanopeniaminimidereree sence cee ater eee 392 | Psephophorus polygonus............-- 370, 371, 372 miultibrachiataess.- oss. see eecee 392 pseudosiracioness-+ sce sees 372 iPhapitreron albiironss-2-- eeeen ee meee ee eee 436 Séaldit 7 ee. eases ee 372 IPLEVILOSHEIS. 2 san oes ose een 436, 464 (Macrochelys) sealdii......... 372 SAIMALFENSIS: 5.o...6 Jeet ce Sesceae 436 | (Sphargis) pseudostracion.... 372 IPHASSUSH eee we ene ee eee eee ee eee 529 rupeliensis. .....-- 372 argentilertis S2ost ec. oe secewsac eae 565 | Pseudotharrhaleus malindangensis..........- 441 MAN SW ari Shs cee cee noe eee 539,565 | Pseuphophorus rupeliensis...............-- 372 Philippine Birds, with Descriptions of New | -Pteronarcys californicas.- s2s2.--=ecsasse= 532, 560 and Rare Species.—Additions to the List | Pteropus vampyrus malaccensis...........-- - 490 OL bys. 2a. Mears. 25. 2.5. 2eoee Seana 435. | Ptilodussc42 es eee a ee 611, 612, 613 Philippine Islands, Borneo, Guam, and Mid- | STACIIS St econ see eee ~ 616, 623, 626 way Island, with Descriptions of Three MmedteyWSs-se ee ees 612, 613, 614, 615, 616 New Forms.—A List of Birds collected by MONENUSs = eee ee eee 615, 616, 622 Dr. Paul Bartsch in the, by E. A. Mearns. 463 DlicatUS oc ee eee 614, 617, 618 Philippine Islands.—On a Collection of Re- S@ITAMIS! So oes, 20 5 ce aoe eee 622 cent Crinoids from the, by Austin H. Clark. 391 TLOUESSAER ANUS! soe eee eee 614, 623 Philippines and Australia.—Fresh-water Ptilodus, with Descriptions of New Species.— Sponges in the Collection of the United Notes on the Fossil Mammalian Genus, by States National Museum: Part I. Speci- ds" Wis Gidley 225 25.2 cadtes 2 oes ee 611 mens from the, by Nelson Annandale...... 627.1. Ptilometras . et ea ee ee 365, 400 HUOSCia PICHALASONG? =. 22sec. eee 2 =e 378° pulcherimaes= 220s secon 400 PICLELICRUMUS oe tte eee Sane one -e eeee ee eee 139) (Putinus cumeatuse see. o- = e e ee 477 Pisobia. aurita 3.22.22 vaest Maen tte ote 478 JeOUCOMEIAS: 6S eee eee 464 plapiatlacidee = 222 Sas. ce eae eee eee 613: | Pycnonotusianalish=323205-252240 =e 470 PLEDGE TELE 2d pete eke ee Se) Sed = 612, 624, 625, 626 POlaVICPS 2-2. co 5. one eee 470 Plathemis lydia....... 658, 660, 661, 662, 663,664,668 | COlAVIEr se ee ee 470, 471 eebPlatyphylax'designata. =. .-2-c-2s+.4-6-.- =. 565 | Suluensisees- ee: 463, 470, 471 Subfasciata.5 2392 Ses. cee see e ee 565 | plumosus 22222 ss-2 eee eee 475 HOCOLYMUNUPA Hens alas oo eae Se 449, 450 | simplex: -.. Yo see eee 475 Pol OlOpRUSTUNOSHCHUS.. 4-92. eae A470: || (Pyrgomorphine 2,22 kee ee 109 Eto} Lolef ny 0 bi hierar ere ee me aS Seaae 300 | Pyrgomorphine and Locustine (Acridine of Polyehitonacris;----. 2 os esas ee eee ee oe 162 Authors).—On Brazilian Grasshoppers of LRT Ne ae a eric ee 163 the Subfamilies, by James A. G. Rehn..... 109 Rahs pSilo ber Striataaaa- cic ote settee) ae O21, | “Pyrotresom Srdenss. a: 9.2622 see eee 468 IPOIVSAr USE: (tea acne thee ate een ee 162 | Pyrrherodias manillensis..-.-...........:.-. 465 StAVUSZ Cesc a sewn owl once 1625163 | Pyzrrhulaleucogemys=:--..<-_-.-2--..-see 445, 446 POMACETME Ges: sao ose oe Se COS en eee 600 Steereis) a eo eee 445, 446 PONOMEtTArs- «ess sec soe ones eee OO7. | (Raja... 5-6 se eee ee 459 INSPOLABUS 2.4 ewes ee ae 397 | Ramphalcyon capensis gigantea. .-..-......- 466 Pontosetus leucopaster.2.-.-2---=.--+-2-22s2- 465 POULGIS ore ee eee 466 Pornzanulapalmeri-. 2.9. Jo-5 sete eee ee 477 | SMUG: 952 oe 463, 466 IPOLEIIOCrINUS!* .-Einn.2 25 so ances a= aoe ae e = 190 Rare Species.—Additions to the List of Phil- Eratincola Gapratacs......2..< = ses see eee 489 Prioniturus malindangensis..............--- 437 | condurensis. -..-- - Peete gee wera 490 waterstradti:s.. 2.222.228 weeeeeie 438 CONBNIS =. 22 2¢ 490 PTONOIGDHA see ate re eee cera 117 Y CODSPiCUaS 226 8 oot ee ee 490 Serratare tec. ses cee eensces 117 iisipnis == 2 ss. se ee ee 482, 483, 490 IPRISINAA Sead cd oss sees ie aes See See ee ee 375 notabiliss 25 c.c0 os aes sae eee 490 PLOCOPI AM as Rosco sod de cose ore eee 110 | Rays, of the Family Narcobatide, from Deep MUM ORI seers seen eae ee eee 110 Water off the Southern Atlantic Coast of IPTOMAGCHOCMIMUS .2 2 )ce56 asa ose eaen eee 363 the United States.—Descriptions of Two Prosopis :philippinensis. --:.----..-2--2-2s-s.- 414 New Species of Electric. by B. A. Bean and PVOSOMUMIM ss coe ise Ae Sse ess om aes see B30) 9 AS CW Cede ele ee 677 PNOURCU eae eae a 5 ee he cite ascie Bek tae eee 624 | Recent Crinoids.—Descriptions of Seventeen Protomachus depressus::2 2-24. 25-- se -neee 110 New Species of, by A. H. Clark...........- 633 PrOhOparce Gimewlata. «tos sees e eee es 565 | Recent Crinoids from the Philippine Psrenyihig Somes. oss seeac Saas seem 417 Islands.—On a Collection of, by Austin I. puilanthoidess J2c-. -oe-em-cseeese 417 GClatk oo soc. cosa 2s San eee eee eee 391 PSAP YROMOU A ewe te A 2a acl eesc See 648 | Rehn, James A. G. On Brazilian Grasshop- TUT Be caa ee Foe een ee 648 pers of the Subfamilies Pyrgomorphinz and Psephophorus........--: SOSA SSE a eee ee 373 Locustine (Acridine of Authors)........- 109 calyertensis. =. ---.----- 370, 371,372 | Remarkable New Genus of Unstalked Cri- GOCSONUS 7. S23. seeceree cers eee 373 noids.—Comatilia, A, by A. H. Clark...... 361 INDEX. Page. Remarks on the Insectivores of the Genus Gymnura; by Marcus Ward Lyon........- 449 Reptile Camptosaurus, with a Revision of the Species of the Genus, and Descriptions of Two New Species.—Ostrology of the Juras- sic, by Charles W. Gilmore...............- 197 Revised List.—Additional Notes on Mam- mals of the Rhio-Linga Archipelago, with Descriptions of New Species and a, by M. “TLRS 0 9 ie eo ea Re a Ba 479 Revision of the Crinoid Family Comasteride, with Descriptions of New Genera and Breciess pyr. rl. Clark. <2. 5..- sb -Sas0ceee 493 Revision of the Species of the Genus, and De- scriptions of Two New Species.—Osteology of the Jurassic Reptile Camptosaurus, with Be pviCharles W. Gilmore.: .. 5.52.0 cc 284 197 amomyies OCWMaris: ....-0.2--.. hen. os 05 463,469 ruficauda mindanensis......-.- 439 ruGesuda.2 32220... - 439 Rhinosciurus laticaudatus..............-- eae Rhio-Linga Archipelago, with Descriptions of New Species and a Revised List —Addi- tional Notes on Mammals of the, by M. W. Luaritiall iF Riel Sees ae eee ie es ee Spe ae 479 BU NOUTA CVANICADS:-\5- soko ens tee Oe 469 AUPTILORYUIS soa on oe scien t o eee 469 RIES TAUREN a eters a cae ete rs tee 477 ADC LUTE: We yo Sos eet eos oe wae Sona 188, 362, 365 TBIS coe De AE occ 7 So. Sore es ee 675,676 Bianiiet ey aa ot oe aoe 674 Rhizocrinus.—TIour New Species of the Cri- noid Genus, py A. TH. Clark-. 222.222.2202. 673 Rhizocrinus lofotensis........-..-. 673,674, 675,676 PAU S OTM Se es ctar sa 2 so mye wete ee ee 673,676 EQUUSUIS ee one kee eee 675 BaD ese erase: Gece eee eaten oe 675 MEHR Lee ee. oo. ee cee 674,676 WHORES Roos Jc So sann BEA 674,676 Rhyacophilus glareola.................-..--- 464 Richardson, IHarriet. Description of a New Isopod of the Genus Jeeropsis from Pata- ONT R ee atas Sane 421 The Jsopod Crusta- cean Acanthoniscus Spiniger Kinahan redescribed........- 431 The Jsopod Crusta- cean, Ancinus de- pressus (Say)....--- 173 Riu Kiu Islands.—Descriptions of New Gen- era and Species of Fishes from Japan and avee aN id [a CSch fC (2 pan ores So ear ieee 597 Robbins, W. W., and T. D. A. Cockerell. Notes on Two Slugs of the Genus Veroni- Poblaete nese ee seen oe ee Sete Se 381 Ranan Catholic vosary sc. s22e2e0 222 ten hese 350 Rosaries in the United States National Mu- seum.—The Collection of, by Immanuel M. GAVE Nake 6/ Aa arene a Se ae 333 PANO ChUPEALONIMISe).\ 2.) =. 2. fe ere sate ee 171 Sarwan Prowl pigeon -....... 2... -..2.25s0-- ees 436 PBUCCICONTUIS ested 2c Ce ale cinacte uate ele 475 melanonovus: se jo. 32.4 50-0 ee 475 Proc.N.M.vol.xxxvi—09——48 693 Page. (Say).—The Isopod Crustacean Ancinus de- pressus, by Harriet Richardson. .......... 173 BCALICHUD VIG Sa1 32 2k eon niae ch ooue sate ee ere 602 SODISLOCOTES 20 Proc ceaic: len); ote ee ee 157 GINGKIC AM Ges Zeck ce Gah saeee 158 Meni cr say eo ae Ok) Ag) i ee 157, 158 GAVOMAswatlacc2 fo kale. eee 157 DEMens = 2, es ee eee eee 158 Scitropterus-amoonus: . -2522- 22.22.82 e-eee 490 DCMUMMS ADOOUMU . <2 2552 ons dat. eee 510 GAnlIMONenSIS <5 ub 4s ae eee 490 CONGUBCNSISE £35 50c Sos. tea ees 490 CUPeCLONS FOF ia. See. cnc Soke cose s en OS OLO pPeninsulanise es ce: <2) eee 482, 490 LEMON GUIS? 24S. Se a ae 510 LONWS ese aeee ces oe to ee seen 490 IVR Lah eR DULG gta ree ahaha oss eee ee 510 WALL UEIS 22 aio. 2 co eee 482 Sevlopocnyptopse —.—-) « sasce2 2 Seen 537,541, 551 MCUiLCTan se eo eee oe tpiiretiae ia tee oe eee 518 Seventeen New Species of Recent Crinoids.— Descriptions of, by Nelson Annandale.... 633 Shinto rosanyees vest a ae aoe ae sees 344 Shrimps of the Genus Synalpheus.—The American Species of Snapping, by Henri (Gyo gi) Keli Naa Seley cnc g sn a RMR ye esas SS 1 Siasi pigmy woodpecker.../-.. /32..-.-4.-.. 438 SPN ai SunINAaMe;nsis; feces = oe ee es 562 SUMMA LenS tab aey 1) ih te = 5) Sed Ae Nock eee eee 489 fasciculatiss (ooo esse wek Ts eee 488 Siphostoma yoshi. ....... cee ae eeces eae 597 DILOsuea Vy DEMMISS sss see: os. eee eee 567 Skate (Dactylobatus armatus) from Deep Water off the Southern Atlantic Coast of the United States.—Description of a New, by B.A. Bean ands”. C. Weed... 222.2.2-- 459 Skull and the Brain of Triceratops, with Notes on the Brain-cases of Iguanodon and Mega- losaurus.—On the, by O. P. Hay......... 95 Slugs of the Genus Veronicellan—Notes on Two, by W. W. Robbins and T. D. A. Cockerell 22% see en ees a. eee 381 SMAranae dooce oe score acess oon ete 334 Snake from Panama.—Description of a New, by: Leonhard !'Stejneper= 252 222i ee 457 Snapping Shrimps of the Genus Symal- pheus.—The American Species of, by Henri Goulitte, 22-2 eee cn noe ee eee il Snodgrass, Robert Evans, The Thorax of In- sects and the Articulation of the Wings. - - Snyder, John Otterbein, Descriptions of New Genera and Spe- cies of Fishes from Japan and the Riu Kiu Islands... -... and David Starr Jor- dan. Description of a New White- fish (Coregonus oregonius) from Mckenzie River, 511 597 425 IS(2] (Ts Fe eR ee Se A re crs Sees eee a ee et 609 South China Sea—A New Squirrel from Direction Island, by M. W. Lyon, jr....... 694 INDEX. Page. Page. Southern’ Atlantic Coast of the United Sphzrominze colobranchiatz.........------- 174 States.—Description of a New Skate (Dac- platy branchiate.2e-2- 25. sees 174 tylobatus armatus) from Deep Water off Sphargis pseudostracion...........-.-------- 372 the, by B. A. Bean and A. C. Weed ...--- 459 rupeliensiS=-. “S225 Slee eee 372 Southern Atlantic Coast of the United Sphecits speciosuss s5 5-22 s2aee see eee ee 567 States.—Descriptions of Two New Species Spilomis holospilus 23-3222 ss: eseeeee eeeeeee 465 of Electric Rays, of the Family Narcoba- Spiniger Kinahan redescribed.—The Isopod tide, from Deep Water off the, by B. A. | Crustacean Acanthoniscus, by Harriet Beanyand Aw ©, Weed i:.2 0. near semen sees 677. |) Richardson. 220. 2s eee eens 431 Snathiivums cess. cate oa see eee sen 109) | ‘Spinoliella/obscurellasss-- 22--- 2 Sase=e ee eee 417 CYANOPteLum.. 4 - se se sees = 110 pebrata.c. voces sass estes taeseeee 417 Species. — Additions to the List of Philippine Spodromantis puttatas.s.-2 22-2 5e eee ee 539, 555 Birds, with Descriptions of New and Rare, | Sponges in the Collection of the U. 8. National Dy BAG Menrmsj.i2.cu'c <:se na eaeue ciee 435 | Museum. Part I. Specimens from the Species and a Revised List.— Additional Philippines and Australia.— Fresh-water, by Notes on Mammals of the Rhio-Linga Ar- Nelson Annandales-. 32.2. sea ee eee 627 chipelago, with Descriptions of New, by Spongilla o.oo. 4. 22 oa eee pa ee oe 627 INAV Vic ME TyOL, i sci nance nee ee oeieae ay: 479 | ~ cClementis2: 222-5 222 sslece Sat oeeeee 631 Species.—Notes on the Fossil Mammalian lacustris: 2 222.22 staccteees aoe oe 629 Genus Ptilodus, with Descriptions of New, philippinensis. ....-.- cre ae 629, 631, 632 yd We Gidleys sete wianeeee as teeee. = 611 proliferens:)...-2.b2 eset 629 Species of Cisco, or Lake Herring (Argyroso- sceptrioidés:..< f:.242:- 22. aee eee 627, 629 mus), from the Great Lakes of America; Spongiphora-coss-- esses) 2. - sehen 532 with a Note on the Species of Whitefish.— apicidentata...-4cs5-1 eae 539, 561 Descriptions of Three New, by David Starr brunmneipennis--- 2 -- eae _ 561 Jordan and B. W..Evermann............-. 165 | Springer, Frank. A New American Jurassic Species of Electric Rays, of the Family Nar- Grindids. 2. ce ciceeeenace saeee 179 cobatid, from Deep Water oil the South- Squirrel from Direction Island, South China ern Atlantic Ccast of the United States.— Sea.—A New, by M. W. Lyon, jr..--.-.-.-- 509 Descriptions of Two New, by B. A. Bean Steere bullfinch >: <5 5. eS 5 eee 445 and Jac ©: Weed. aoa! 2) sss soaaee se oomens 677. | Sterosaurusmarshil-eosec.. 42 =e eee 292 Species of Fishes from Japan and the Riu Kiu suleatus.22 2 vc SA eee 198 Islands.— Descriptions of New Genera and, Stejneger, Leonhard. Description of a New LabysO SSnyderiece 65 ost cece ee cer ac 597 Snake irom: Panama? 2.2 22. eeeae- eee 457 Species of Fossil Turtles, Toxochelys steno- St@HECRIS S-—- =. .saeeaeae ciao heeeee acre Seer 135 pora and Chisternon? interpositum, the Chlorizans-2.-3--/2.-5-+5==nes see 135 Latter hitherto Unknown.— Description of COCCINGIPES.- sa se.- === 2 8, 10, 64, 67, 68, 69 ils eo\0) Leh: eee eee et 9 OxbeNHNS 2 n8 sce ee 8, 10, 6% CONGO) pea ee ee een 9, 10, 82 paraneomeris...... 9,13, 27, 35, 39, 90, 91 apioceros.... 6,10, 18,19, 27, 29, 13, 80, 31 OXYCerOS 2 2c eeSe re) desterroensis......-.. 6,10,31 prolatus_.2 25. 7.252 9 leiopes............--- 6,10,30 paraneptunus............:. 9, 10, 16, 86. mayaguensis......... 6,10,30 Parlaitis <2 cece ot. moeee 10, 67, 68, 69 sanjosel.c.-. S)s.0- 6, 10, 19, 29 paulsoni....... 10, 12,15, 18, 19, 23, 24, 92 Paketltetes ete Sakae sesh 9,14, 91 kurracheensis . 10,12, 23,25, 92 biunguiculatus........---. 10, 16, 87, 93 Mand aRIS ss Ss aes oe 10, 23, 92 @xilipes-._ 5... .- 10,93 rameswarensis........-- 10, 23. pachymeris.....- 10 senegambiensis.......-- 10, 92 brevicarpus. ....- 1, 2,7, 10, 15, 46, 50, 52 aa paulsonoides......... 7,13, 18, 19, 24, 25 guerini = <2.25 325 2,7, 10, 53 pectiniger.... 3,8,10,17, 18, 55, 69, 78, 83 | brooksi.... 3,9, 10,18, 20, 69, 73, 76, 77, 85 pescadorensis. 255-5 :\sirx alae 10, 85, 87 eleuthers ..<..2..2c-5- 9, 10, 73 puysecheles.j252 25. csscnsee see see 9,91 strepsiceros -<-..-...---.- 10,73 MOCochI ws oes SEE oe he g RATS ewan eis .g ee aoe 9 PAMIDUNS sssasc =: eee 8, 10, 18, 20, 84,87 BBATOR ES sas ies: sce eee ens 9,90 Sanctithomss=---- 2ashesoeeee 9,10, 61 Gomeiwanim 2:28 5-2 cates 4,9 Sanlueasixu--so5s-5 sees 6,9, 41, 42, 4% PUT NEAT IR ey Deeg ee eee 9 Sladenin’s <5) sci oieeeee sees 10,16, 93: GISUBII S55 sci wotsee ene 7, 10, 15, 48, 50 Spiniironss... -. =. = s--seeesee see 10,16: ecuadorensis ...-------- 7, -0,65 Bpinigens. 255 5-ssu ase 10, 16,17 fOSSOTSS Fs At Ne et ES aces 9,91 SUM PSONIS-4 2 once ee eae aed g fritzmiilleri... 6,9, 14,35, 38, 39, 40, 42, 43 Maldiyensis:.-25- 5. +=. 9 elongats)::-2.2-.c4- 6,9, 38 THONOTIE oc. -\j2 sana 8,10, 20, C9, 7& [ey 00 (3S Sor tees ages ES as eae 9, | Synalpheus.—The American Specics of Snap- 10, 56, 57, £8, 59, 60, 61, 62, 63, 84 ping Shrimps of the Genus, by Henri Cou- occidentalis .......-.- 9, 10, 60, 61 GOLEM ~ 22 0s ose Soe as ee ee eee I SPAIDUSH. - 2:5 cones 8, 10, 62, 65, 67, 68,69 | Synalpheus townsendi .. 6,10, 13,17, 27,32, 33,35, 39, Rerawinie es setae sao sarees 9,90 brevispiniss.:: 2.2222 6 TRENT GTi ee os ee eee ' 10 MOXICANUMS Sc eee 5 ose 6, 10,35 hastilicrassus........ 10,12, 13,31, 33,34 productus-.. -22- 6,10, 35 hemphilli.......:.. 6,9,14, 18, 38, 39, 42 tricuspidatuss=2-.-- -22-cessoee 10,12, longicornis........ 6, 9, 20,39 ERO GC Wise es: 13 eee 9,91 erOMMe set oe eer meena Pe sese Os trip snIewlaiss...2 a2 abe 9,91 herricki....-.- 3,8, 10, 19, 69, 73, 74, 77, 78 tumidomantisss225.5-..222 10, 12, 24, 92 angustipes..--<---= SAO RO noel oy NPM aN ect e eat geo See eee eta 597 dimidiatus....-- SAO Pe 0o> || Da panuwsatravuSe.- 2% an ace e-5 «eas qeeeee 568 WUVIENSIS! 22s. Ss .cceMe Sse HO 2aeO 2k Meni OpOmas hc asc es ate noe eet mere 128 POMITMAIUS Ecc cine eee see eae OS66;90" || Meeni pode = 2. ae) meee ae ee ne me eee seer 128 War Dariaitine mesos 64 | Teeniopteryx fasciata..............------.--- 560: TAbASUEI Ss. acs 7,10, 13, 18, 24,25, 26,27 | Tanygnathus lucionensis........-...--.-.---. 466° feniwispinas .-SosS- eat 7,10, 26 megalorhynchos..........----- 435 IWC Spam toner: duccabesade HO; Lia) Uelespy za CaMtansen os. or = en enem seein aa 478 POG RAMS FONTS 2-1 Sees = creme 1,7, | Tetragoneuria cynosura.... 056,658, 665, 666, 667, 668 10, 12, 13, 18, 21, 23, 24, 25, 26, 27, 29 Spinivera-sesos ens C55, 658, 659, 661 longicarpus........-----.---- yon One Ve wrapeGias. «ae acs tee seems atime een eee 413. 17,18, 20, 53, 55, 56, 57, Pale Freese ee eee 411 59, 60, 69, 79, 80, 85-93 10) (61: eee RE ras Se 413 ‘approxima........ 8,10,57 | Tetropium velutinum.....---------------- 524, 562 lophodactylus....-...-------- WOW Gs93) ||\huithelassomebtridce 2.) i225. 2s. .secese see 402, 642 macculloghie Ts: be = 255 HOS 25.9" Mhalassometrinee.<2ac.sk och e cee eee ee sec 402, 642: PMELOS PUN SOR 2) <,3)5ae ~ Sse ctorsta ches OPOO RN ratimabOCHIBUSS...\280s.0ee see se ose asec ees 362 PYNITA Ter ees eon 1,2, 7,10,15, ENO V AUIS eee Cason sacle 362. 19, 20, 43, 46, 48, 49, 50, 51, 2,87 | The Isopod Crustacean, Ancinus depressus ‘ULM GTISIS 3 oo eee a aiainte= 7,10, 48 (Say), by Harriet Richardson........-..--- 173 DAHICTISIS: 222.7 arenes eu OAS Dh eromorpias.+ao2- a= cee ce =