\ ) wy ni a ; in Na NY Voi feet Ni maha Fis ; aN 1" Nae AY hy Ae RRC tal pl fi y ; i A Vit : ah i A Gy ay" mii Ii i ; = ~ — y oe . jie r a A 2 . ‘ i U : ve i 5 yi r i a w : H 7 : F i r f i! pense - ne Ta te, u f aa 4 , ‘ Show earn rn e - , x i if t @ COGS O INS OOF EHF ofof oof Gr ote feteletelpfele! clateate 2 m: at oud 2 * ae CIP a ee eee aes eae oe haere eee et a af lemon te moe ae eh hel eka eae nr. " OOO Oe PURER we em ee where ate ete tel owe le ete mew whe wea ore ‘ob ryt were teene et a ee ee ee a etaperate wenisieeret vim e 29° ee ee oe ee a ee ——~ we eee San ee eh eed ra ah ia otra PERIOD wat ow RE eo ears il ae wee alice ay ne ns ad " Nee aed eres fapietahe wel ehorg ater TSM EP giao FPP yterm ey Aa A OSS ee ee ee ee ee . pete eo wrette a tw BPI Fr - irene avila werd aero ee 6 ’ yivieteleie viery wiutwialsieepeletereeivvint winter ginicia 2 8 9» : pes oy Dg ou earner trains bn tierce ort enhekoToaain Sanobocenaronanbeeet ene hehehbahchAsbthsne Lady abd paaanaad ah aaabeasconaneesati oe piece OMe ORL PRET e RINE Te TE TT wl phere lew wenn rman ere pehatreeite Fateypd nesting nid east ne nee e eee e ant nabinindadonnded Nee Una p Vbagin ata nainaita wind aati aH HNaengertg tte OPO Bae e lg one ob eeeh rote DiorwrecereHetewle/ePe poarvapsinsvatrincdtt iene lier fet rnd sarcoser SNH eT OE SO Soph sade inca chet lotama lh deak, Lledo Tek devised tan eee a i Naaadivnd duis Iribshchead ’ Sa nel a Seton than a et EE Ao SSE SINE TI nig ee re een nthe Minchdatnaedeteds hata inde te te ethene ie Cee t . Ne ek Mel Bat el tele kOe cel ail id pip dbihdidethdhthdetetntatetetticemes bate Ayn a ee NTN Sica tad nck ethic ddadih teak ek he et caged a mates m4 hoes Leith dbihintnth dh edie oe event erveer cated ntndhetiathdinth, fate beidhdailideccdbdbaeinibidnie in ndeshttteahdstadhdnieee te ee a a ek ee een es Saethdh Aeededind nk ee tee ee) Ce ee ee ee ee a nme ne ae ne AS SE OT ee ; ieee he eo ee bi faleiedindeelio th ak eaiadde dated indaale tntetedintnacted on ee ee St eee al eS on at etebeutero aha te eee E HEIN CT Bt fewer Sela, abaetieteinthetine’ mbar rn reel or wet wi are SUN RRS PS Se eel he PoigialCerrer«w~ 4 4 . 7 ha ink seis desiadvaraier Gn Foerster wwintehe: Ee ee ee ea ve Tice dich se dnaiincibet i ahah Oh Sk dh Hh disse skeathcin death a milcdiiediaatndindinaies iytetete ee AC oar Or arent ee A ee aah ROR ey Sow Reete OR! aera ean HE a) 19 Olena baat eres: tile err Oar eer mise POye . raeblen bach thd sh ede ty bh /thaleld coed thgik 0. vivheansl orem ender . - - PES ED ECR CE OPE Tare Riel erD pe rete Nee ob Tishaa bh Polite esdbtendein thai adslnation ne nee Were bers Mbit int Middl ited Mathie bddbid ddd ka ak ee Oe canines bd bit aa ih P ien< notin limo porta inietitind Gh letintnih ss thc deldeidat edhe nee Le tee ce, kee dh tedh-wineinihtaddb tant mend ia ny Pn POTTER YORE Wie Were ¥it oreo es Walele agarevers)scirendjelawiaie Nt aht x EI dient MMOS nd punting teh C1 OER BE SR IE al eEmenediehwiepee: one vias lop ie elmer Pet nsheantendd- sides AO tertile antes aan> nite feof anh Po ew Toe py wee tore ee = “™~ ere see ei hasnt i a te S694: 8H ew wie ie) erro ereie- aie Pete He AOTC OEE RETO rs SD etna tiny or eee . - 7 a ee a wwe iialatalnled os Oe Teer eee ter eee } hr eaiar srt vir umn indblirtt rd hoe nd De aethdiotdh, tach cadena ethdedheaes hb tee ee paras be tone bOhrindbs PMY rir eset pared —HomPned et hth hak de Ree LIA Apia edea eee hee Wpacwaenmnee tinct 2 va neh Ril flan A Th P-thtn detent th tintadindh tate dedeah-th-dh dnie-toeac dL oodak a os VECO TY EO eee re ere ewe dial ditaintle tedtee cee she? nd maaan a sinag ne ’ " nd OT Oe Oe ep lee ee ee ee ee ‘ WOU OOM EWS Ly etal ieMes ih ht SARA ee ae) le oe ot okt hae aeabdie CIV Pwd ewe . d dee en vet. CCUUSOOEVE ENON COTE: Selhdedheh A deen ie ek de ee ok cad fe anaes ee i “ vata aiisiadn bole : ’ Teer vey 4 7 . Me “ EV EV OME ET HS TEL TY TS EP Tee WE re wow ree CUES y aL ETT ONE ERS Oe ree E ECE TTT bliin ddictielh di tetedndieds te tedacielintnteteldietate a tthe b eo i ak tad \ Adind dh ft tedtah Bad x Phe IM Nelew oa eter pane ee sete . . 9 eS Me eel pe eee — i he tei IR ee eee Se an hak adalat vw nen ntti ¢ trencietiy VE ESPEN Ye ew eA - bth diedididindint tied tisha pai Fabio me bens retold noon tne ire Se Pee SRT SPT ed rine ea aoe ~“ ene A Dledadh ded i Soo ed ore ie rewere a ar ee hshthtedbibtadh teieabababtadat batdaiamh haan ee “ ee a ee So OMNI inte Penta BUEN Wen wte ew eye Ene are a Sel erie Na Mee New ete prahdnththtnghdinthdoknibardbihteanstrtedadeahsotnteahdad hanaae ended cit, aah th Pade Sad Da Wk inte ah ht Pah bdbthen cbadiahaieditaand tne te ade ete ee te ee tt deeiettairtrs COT Pepe ee 88 er Pelee 8 © ewe eee Owe Ww ee or ete eee aoe Se ee art ete ONE ET Ee Teele Wey Pe inci FR le eee ge wee on Pe La OY OM PS TS ON Ono a A Sp shy’ ws h Gettin gh eda AS Paahthdees Sadietened PVC EY rey sien oe ~ elo ntale tw sowie ofefe'= - $ Se eee Agenda ne om af eer at Pe oe st wletedyty = Se i ah a ee PO EE EEE ye He re ww ep rery es “~ POR ewe ere es weiner te Sw ew vee died ustedes Ea eT eae ne Nett ert tel tela ae aetna I We! SNE a eee ee eee ee ee Deol eee ee oe ae eee wee aig het Shahi late wow ee" Cee e ee ee ee oe eee eee eee ree I WY Cee ee See eae WEF a Be Ee TE IWR EEE ge wlele ere ew inth>thdbtintithdtidhth hen he oe ee ee . shh diandinad nds ed 2 OW PETE ew Oe re Pew eres’ series eee os Ce ee ee ee ie ee ee ee ee ee Dededhdthandhthtete een de ek ee dh ok a eae PE we Ee bi egle PON eer anes moar PORE OE YE ee Na Pay ee a we i ae Vig ye errres CO Od ew en ee ee ele SE Te Habe es yee VEN W EW OW tire er wele one SSW ew eee wel le ee ee ee ec en eee | Ov Tey rowley ls Se ee eee ee Se ee ye PI TV PEPE ee ENT TPS Pegs pref Pee Te nee i! Fe We i Lots rtntindh-2 Wjetwheioipie Ne FM weir “ i P : Stille ttadid th Roe a at ae Pie Pathe ew Se » fm “hth sh dh de Dedpattsd tarictindieT eink dtetatintaeteeeditinth thte-tatirtirtettetdit <> andttetetiindiatiae eee eee wheelie | tee awn BIS Mee « ee Ee Ry FEROS EE Ne Pee ee ey wee ey : : entree ‘ : ‘, > >, 7 tes . NG te eagle NON he Dies i NO a A ed ey ag Bd ere pees “on x a4 7 POT WY Ne Ie te ntl we) eel nl ew ten WIE ete aR ENS vipers’ Path iedielinded Tey Hew = defn Veen gates bath Badk dh Bed hdirthtlnd Halindhtietndndit Ntiethdvd etd peace, Goda eemaioh tedee ta cadetadeek inte to aetet ey 8 EES Wow Ney eet ry Ataf bh 2 dpdindinds pth chde thin cas Pe NL EET nh ON OST eS ee ie dstiet det sts TEP ONY wipe Pree Se Meee weir Wigiblw one louie “~ " vw aaleé i ee ee Pee ONT eee Pe eee PP ee ee adh detuaht tieidiththtaded et eo on ne bandied : wie eee pe aN eer Ne ¢ WK dads ic ide th Rh catalase oe ee en aE PEO ST Nee ew Ee ty ew y pth setedinds Sede I bs ak ie ake etn ey WE ae gle Sicha at gylw ews Art rtvh So 6D Mi tied £ ln OARS EI ee Td th Oh eS he Bee ee ee ea . vw Ra ee rete Tet book ah ck aii dn dn de theinecdietetetiekadk deta thse dnhodll , oe ee ee ee ww SY we aelgig¢g petty wile "1, Der wear ew CA IN welt yi ee Ww wee & VERT UE Ser ee CET POF a SNe pest Ni eager he rg ee ai Na wwe er we erwin nn - : hginiten™ 1 ONES Rw eee ibtednbs ek dtd bthdded bn hpi irish tiictttrhaiih he eee . . . Pesyty ter Neel ewig ehatatdin'e v's tied Sui dh thle Miran ON * » Crete Ayer ortho . ¥ ee ee eo aa kk bas ads tres BirAtndte Rt). eM 9 e YY wes Fe Wiig We arta eee ip COC Lye LO TY Ee le wiht wie Nearing Ne ntateebtleds thdieddh-ttos eA RAL he hah el were VCP Owe gees reece AD SAS RADE S Ss we fale ew "erway a we Wwigie-w Ce Op ey ay eet ees es PRE UNO EY WW we reas wh i LP OP ARACEAE PO f vy SCOT VOU PaO SRAUE RR EAA DEER ADR E ROD Oa ae ee re ht ha nek - ee - : Nile ylg ol ONE rere “ endl re bie Z 2 i vw t® sé dakagatinting ONE ewe lew ewer YE PIE ET ENTE VENTE T LT VUE ED EN e Wy va we rte moth tl yt TV pds rthstndieh rt redth 2-44 1st etn) Aedchntdiec nl weer ape ther Pee ge ye > PETER ys read dee ald de Y va wey ox ltt d oh ret th rth ole tet i Ree . ‘ Path dregs sis add that oer fos se vce + verre acta ore wept Coa : owe ew wee ERE REP WTI > tno oer wd seaniednaaied r n . é - t re , ore Sli dette hei tht dead at Oe rig: pnt rep vente Meld ihe wneiadinddrns nds Led c04) Sa ek owery “ Meditate Ae a Se ee ee a ee ee ie > oe eee ee OLE ETeVe we TERS OT UR. we ap ele ee een wey raw POHONTUI Wwe ire whee Y Stet dhdh Peet Lae aie th th dthidhte tad eictinttaaith Cote Y eee Ne wy ded teas - i bileetiatidt by Dh dhl tednde badd te tein dhtrdbethe) sted ated tts revs rn vl nt naan? wv rey . nw rw plat Stari ie here Phat igtw rb: 0h We wien owe eee Vee wT ww wo” Oe deata . Henrts ° eS AOS Se NEL et Ah ean | Vt ww a vl ? ¥ Sl Me aad 4 . ode } Oe ewe Se Oe Oe > PIV EP TP he we FS pare eS Slate Se ew ale ly Sete lee Sg WINE INS DE Ty va by rete ; “ ae ee ee eo hl hhc eh a hee et Bev pt Ptndeaiati Sind pla mma fe =) seirdnth ly Doers nk eh lh ec i ee en te er a ee View ore evr wees PPE PTY OE Or er ere . ~erre . ML b << , ~ ‘v* ade v ev cr UPr FTE 68 view ey TE i ERE yw rey ws aed + * ie Mt dee te ee ee ee id a hte ttetdttn op che ee ee) wees este we: —— Re SS Riek tied fists a SOOT Te wis Daal Ses . Fer Peery terre ns rtrd ate Has Cee OE Sonlaett vd Lod) Sees Seah die tide Srl ds Alesis tlode Retind hs Fe OAT SE mares roar VN rowr " (rie het ae vinte rice ewe E RE CET Ta Ley ETE ET ELT TE NEE Ye ee ere ur Mindat db LS ead ni Roney ditivt'y Liotta 8D oreo rt darth 2 sia Atle Se it SEN ia rvewrr ees Trev Ve a ae ee ea Te eo eI Waa oe ere Whe eareenne 9S ede eerie: We eee Oi Batat oh ohn da dkictntt Dri tiehd eB PDA OIE SEN val LAAN os Cpe PE yt nl en > Tf did” Sd Y CUVEE TR Oe eer et ete meee yale he ote Wit aye wlgiyiehwwints wwii Spey tetemae wevrerutotemahehe ne COV EG pee Te EET TT er ry gee COUN Yee OPE ELT Tee TE Myer ET YS er eT a ee ee ee a > + thedny \ ee See SW EG WY ETE SE e oe FEE Heres WTOC OCD wy TORTS HOG welele On eit ate ~ tel CP TET OP eT eww ee tht hee ee Sieh Minty 5 SIRS AREAS SRS SENSOR v wr Sid da ae OD A ded ee om veer er aa ee hee . » ‘oe sree TOV eww eee bd dota tidied vires ~ wy Dee ee ici hah dh, teks Ite had kt ee cere ereyr id ah th4h Goth ie Ee PETE he er ee we eo vl wie re He Fete OAT 1 th eel BID ESS Gi Cte WS) DOLL IO Sot Sab Ps ond verre. CVE eR we OO ET EE VPN ees ee owe YVeryy re ey yp eer owes t ote wy try ws Wet sO Moree oy 2 Pei ytd Aad > Ctl hae sy hands) -thdbd be teteiednd tb thae ide i tae, th aad Ah hte oe be ak a te ace a sk 6 be PAT thd ody ewrrperrs i ov hil bok nn ne he oe Peer SPR Tee PPV 8 eT Or eg wr ow i oreo oe Ve ewe ewe Uee r i t Ort dA oe 4 ‘Sy see we er wy ee UUW PE WET hy Wee ay eee ai eet CORON we i. : - vrvvy ew eg Oy eee eety ee bw he he (EUCSS Cer rey ry Whee ae eee mie eo mie pie ries : , wie "we 4 x . orreey ECT FEET CP We he eye rele e tte pe . . . . . ". ewe ow : » ie Ogres ; a ’ : Soieste vit oy fa Digs Swen er verre t+ wha awe ee? ? elke we . . re were a woo ets o wwe ew eve oP Perey ery yrs WY eee ee ee ee ee ee ei ae ow how vie ee ee Met an ae ae rye yee St We hh oe le ee ee ee se ee ee Sree aie te task A fs tds Shh vd dette aadadhte thsi dete tte taad * tL lk ee eh 2h ee a TY awe w wiel a iel gee eth Ele eae ee THE Ey re wit : Pete een Phe oe ee ee ie oer a Kar es ¢ * $3 ee.) opel senwchibcelseonsgpenhgen qupttuehadns ete A | as Ae LS RT ere eee, va _. i. * ee ey eae ee © e ~'™% oC sean toes ; ome ro ae vw ‘ 7 de Vip ‘ F é ae > .J %, <* ppl | wv ‘Miu ie 4 © ee a POH, we hy a e Vv > 1 ive wom owe see seek al ae Pa | 1 -rariety <4 » bs 98 TRASK. yf pr Kew? shy My ay teed Qh te ow. Tt AS Rae” erp bd abiebiv zy fd 3 4 ' Z ~ *etUgs A} Tr MARR WA ean nr in rane {I Ww || Sed | alata f ppg apt vnitt Na | a A A Aad ete a ONS AV ma Nagy bd, i SPECTRO AA. TEM - "Nhe "th LETTER Tat wee Nhs RA ay rey > itennan an " }eX mi aq 9 wiv i hoo ye ie ere ‘ \ ri Ve ! FuAs. a iY q vuy¥ TLL ERR hy Nb mR ee 1 a 1 du weeverest* ~ = ee Saat eal oie a Vibiay A LM | 3 1 es irine...* ctvne yh el “Woo. TOA AC hy secs A WUyEg eye atet , sek eeReeeeune rete Wy AN een : anh "\) y4 a Se fey < i g SOY wt ll nit tote TOY UL to UTE ee tebinn avetrectttcemueee rein weceeese We wenn ve 3.50 a) ei < oh Ee Swe. er. oh’ -* { ‘an UCB eye Ng Rae a iby Syouu Wy pit ee wit" SUS ISI A ” = Pure a Ae ’ é . 4 sy { 7 wWevye" Ppt aes Pity | aphid |) b aga dy t*..: vee i at 8 RAS TTT Tie ve Ay Se hiaieiad i vA vn a) ltl Se << A a yard 5 Vee ls 7 . ane push iM! vee veve Welty, Wy yor he | uly ap Pi btetee (Jt Sw vinll Niwy vy vi wwe ro Ve swe Teg tye a My X rat weer > Wesuuserce -* by ed wy Vea, ‘ ey CA ree SS new ied | yess hte levees ay WY yay ypu DE epee | ae fd. oe ae M4 Perna et OUT Re fant Sheik ee amecot nnn TF cay tg gUHTOTTE EEL aes - ¥ ne Cw & pve CU we Wey on e & in Pee Mass Le a . AALS ee weveULS. , et 4 “Naga iat ify a LA® AM ‘Sat dilttees ba STEAM ai awn girmoeneet CE NCO! PLL ye ~ gee res no | nswrse a dt Mane SSSA Aantal unnaacdtlnt reeeeet | ACS AaAga. Ae i we VUuci eg Ne a ‘ | ~ w ‘ . whe, L&@ AAS Sy ene a *~ \ ivy Bee WG” eae ih ) Svonteny NW Bee iat rah ot te UV. bd Oy yl Buy, Ppa taetatd ¢ “\.* Way i} 4 ~< ~~ Nha a coer Se ie ‘ Wy Selah and — Sy, eek eS wwe vv J oa)! |} | AAS adc ieee Wy RN a 14 4 RARDOALSE . | ts ~? hee eb hee TO Hab, Fergusson Island, Papua (fruhstorfer). 80 Colonel C. Swinhoe on new Chalciope satna, nov. r 3. Antenne and palpi grey-brown, the latter black at the sides; face and pectus greyish ochreous, the latter with a black stripe on each side ; legs ochreous grey marked with black ; head, body, and wings ochreous brown: fore wing with the inner portion filled in by a very large black elongated triangular patch, edged with whitish, its ‘upper and lower sides almost straight, its outer edge slightly evenly inwardly curved, leaving the margins narrow, its upper point very acute and nearer the apex than it is even in C. cephise of Cramer ; an oblique narrow white band through its middle, open at both ends, the ends slightly irrorated with brown, some black points on the outer white lining of the black patch, the outer margin brown ; the cilia brown with a pale inner line; some whitish suffusion on the hinder margin: hind wing dark brown ; a faint, narrow, grey band not nearly reaching either the costa or the hinder angle, where there is a little pale suffusion ; cilia grey. Expanse of wings, ¢, 2,3, inches. Hab. Nias. Allied to nothing I know of; the largest species of the genus I have yet seen. Hypetra minima, NOV. @. Palpi dark brown at the sides, whitish beneath, the tips of last joints white ; head, body, and fore wing uniform chocolate-brown: fore wing with a deep black subbasal quadrgte patch with pale edgings close to the hinder margin, excavated on its upper and lower sides, a brownish patch between it and the costa ; a curious hook-shaped deep black mark at the end of the cell, its upper part thickened and uadrate, a brownish patch between it and the costa, another brownish patch on the costa before the apex, and a small one at the apex; a series of minute black dots on the outer margin: hind wing chocolate-grey, a little paler basally, no markings; cilia of fore wing pale chocolate-brown, of the hind wing slightly paler than the wing-colour, two whitish subapical spots on it and another at the hinder angle. Un- derside: fore wing greyish brown, hinder margin white, cilia grey: hind wing pale greyish brown, the outer margin broadly darker, cilia white, greyish at the tips. Expanse of wings, 2, 1;% inch. Hab. Luzon, Philippines. Species of Indo-Malayan Heterocera. 81 Family Erebiide. We cannot believe that the genus Argiva and its allies can belong to the family Catocalide. It seems to us that Argiva has no relation whatever to the genus Catocala; their struc- ture is completely different; they have enormous’ black densely hairy extensile organs (“‘ coremata”’) upon the dorsal surface of the tegumen, almost hiding the armature of the delicate valves. These alone confirm, by the absence of the large scaptilum of Catoca/a, that the relationship is mistaken. Pl. X. fig. 20 shows the genitalia of Argiva hieroglyphica (the type of Argiva), fig. 21 that of fraaini, Linn. (the type of Catocala), copied from Pierce’s ‘Genitalia of British Noctuidee,’ fig, 23 that of the common European species, nupta, Linn. ; note the entire absence of the coremata and the asymmetry of nupta, which is usual in Catocala and its relatives. The habits of life of such of the species as are known to me are entirely different to those of the Catocalide ; they are crepuscular and mostly cave-dwellers, very quick and sharp in their short flights, the males darting forwards and. attack- ing those that pass, much after the manner of some of the butterflies of the family Nymphalide. This is particularly the case with Patula macrops; I have watched them in the caves of the Island of Elephanta in the Bombay Harbour. Genus Araiva, Hiibner. The genitalia of the different forms of Argiva are very similar ; the differences are so slight, they may arise from mounting. J give the figure of the costa of A. hiero- glyphica, the type of the genus (text-fig. 15). There appears to be no difference in any of the mounts except the “ costa,” which varies slightly under different names. Argiva lunaris. — Bocana lunaris 9, Walker, xxxi. 57 (1864). Nyctipas hieroglyphica, Hampson (part.), Phal. xii, p. 275 (1918). Hab. Celebes, Gilolo. A perfectly distinct species, sexes alike. I have both sexes from N, Celebes and Gilolo (text-fig. 18). Ann, & Mag. N. Hist, Ser. 9. Vol, ii. 6 82 Colonel C. Swinhoe on new Argiva sumbana, nov. 3. Much like the male of lunaris, but the subapical yel- lowish-white bar is not continuous, but is shorter even than in hieroglyphica and is broken into two pieces. Underside paler; two large subapical spots, well separated, another in the middle of the disk. 9. Fore wing with the basal two-thirds ochreous brown, the outer third black-brown, the comma-shaped discoidal mark as in the male; the subapical bar white and short, Fig. 15. Fig. 16. Fig. 17. Fie. 15.—hieroglyphica. ‘ Costa” more parallel sides. 0 ‘ ” : Fig. 16.—swmbana. “Costa” perhaps running narrower. Fig. 17.—luzonica, “ Costa” perhaps more curved. Fig. 18.—lunaris. “ Costa” appears narrower, but hardly in the same plane. : Fig. 19.—ceramica, ‘ Costa” somewhat differently shaped. consisting of three attached spots, the two lower ones large, the upper ones small; no other markings ; hind wing simi- larly coloured, and with a large white upper discal spot. Underside paler and more ochreous, the spots ochreous white; the subapical spots as above, but not connected ; a discal row of large spots across both wings ; the middle spot in the fore wing and the second upper spot in the hind wing pushed outwards. Expanse of wings, g 3, 8zio inches. Hab. Sumba Island (Doherty) (text-fig. 16). Species of Indo- Malayan Heterocera. 83 Argiva luzonica, nov. 3. Brown-black, the inverted comma-shaped discoidal mark very indistinct, without any distinguishing blue or white scales; the subapical streak very narrow, little more than a thick sinuous line, the colour dark chrome-yellow: hind wing unmarked. Underside paler, the subapical streak thicker and ochreous white, separated into two pieces; a faint small ochreous-white spot in the middle of the disk; hinder marginal spot suffused greyish. Expanse of wings, ¢, 3 inches. Hab. Luzon, Philippines (text-fig. 17). Argiva ceramica, nov. 3. Black-brown, the inverted comma-shaped discoidal mark very obscure, hardly traceable ; the subapical streak more curved than in the other forms, slightly thickened on the costa, but otherwise of fairly even width, ending in a point quite close to the outer margin; colour ochreous white ; no other markings. Underside paler, the subapical streak similar ; the hinder marginal space of the fore wing slightly suffused with grey. 9. Fore wing with the basal two-thirds ochreous brown, the ring of the discoidal mark the only distinct part of this mark, its tail well separated from its beginning and connected with a highly curved black line to the hinder margin; the outer third of the wing dark brown ; the subapical streak broad, narrowing hindwards, its end blunt and not reaching the outer margin; a large white spot in the middle of the disk: hind wing with the basal half dark brown, the outer half ochreous brown, divided by a series of indistinct whitish marks, Underside paler; both wings uniformly coloured, except that the hinder margin of the fore wing is slightly greyish ; the subapical streak and discal spot as above ; a minute whitish spot in the upper disk of the hind wing. Expanse of wings, ¢ 3, 2 3% inches (text-fig. 19). Hab, Ceram Island. Genus CARIONA, nov. Differs from the genus Patula in having the hind wing normal, the neuration normal. In Patula the costal half is aborted, and forms a fold turned over on the upper surface containing a large glandular patch, making the veins aborted. Section III. 8. of Hampson’s genus Nyelipao, Phal. xii. p. 286 (1913). Type, albictneta, Kallar, 6 84 Colonel C. Swinhoe on new Genus EREBUS. Erebus variegata. Nyctipao variegata, 3, Butler, Ann. & Mag. Nat. Hist. (5) xiv. p. 482 (1887) ; Hampson (part.), Phal. xii. p. 296, pl. 206. fig. 6, g (1913). Hab. Solomons. NV. caliginosa, Butler, 7. c. p. 433, which Hampson makes the female of variegata, is a distinct species; it is not the female of variegata. I have the true female of variegata, also from the Solomons (from Shortland Island); it is very similar to the male, has more white suffusion in the wings, and is much larger. Erebus ephesphoris. Phalena noctua crepuscularis, Cram, Pap. Exot. ii. p. 99, pl. 160. fig. A (1779) (nec Linn.). Nyctipao ephesphoris, Hiibner, Verz. Schmett. 272, 2675 (1827). Nyctipao ephesphoris, Walker, xiv. 1805 (1858). Nyctipao leucotenia, Guen. Noct. iii. p. 184 (1852); Hampson, Phal. xii, p. 298, pl. 207. figs. 7 3, 8 Q (1918). Hab. Amboina. Ihave one male and three females from Amboina which are undoubtedly identical with lewcotenia and with Hamp- son’s excellent figures. The type came from Amboina, Lérebus saparea, nov. @?. Chocolate-brown, tinged with ochreous: fore wing with indications of a subbasal band ; a rather broad sinuous antemedial brown band from costa to hinder margin, followed by a similar band a little before the middlé, outwardly edged with greyish ochreous from the hinder margin to the whorl- shaped discoidal mark, which is very large; its black ring strongly outwardly edged with white, which thickens on the costa and has a billhook-shaped large centre filled in with brownish ochreous, ringed with deep black, and edged in- wardly and outwardly with white; a brown thin even discal band with a slight outward curve from the costa to the outer | margin, followed by a pale and more ochreous space ; the other third of the wing as dark as its basal portion ; a large subcostal white spot before the apex, oval and excavated on its outer side, a small white lunule immediately below if, followed by five white lunular marks inwardly edged with black down the disk—the first minute, the fourth well out- wards, the row ending in an outwardly-curved white line close to the hinder margin ; cilia brown with white spots at Species of Indo-Malayan Heterocera. 85 the interspaces in the lower two-thirds of the wing: hind wing with two bands in continuation of the third and fourth bands of the fore wing, the pale ochreous-tinged space extend- ing almost to the outer margin; a large oval subapical white spot and a row of six white lunules, three and three in echelon. Underside paler and more ochreous; a black and white discoidal lunule on each wing; the subapical and discal spots as above. Expanse of wings, 2 , 4,45 inches. Hab. Sapareea, Celebes. Erebus niasana, nov. 6. Chocolate-brown; head and thorax dark brown ; abdo- men brownish grey, the first two segments filled in with black-brown, nearly pure black, the next pale grey, the rest of the abdomen darker grey: fore wing with a thick white line, a thin band from the hinder margin one-fourth from the base obliquely towards the apex curling round the discoidal whorl-shaped mark, its outer side before it begins the curl broadly pale grey, extending in a subdued form to the apex of the wing, with some pure white patches outside the band; the ground-colour of the wing above this band very dark chocolate-brown ; the black ring round the discoidal mark sinuous, the inner portion is black and confused, outwardly ringed with dull brownish ochreous ; a large triangular white subapical spot ; some indistinct blackish discal lunules, one or two of them pricked with white: hind wing with the pale grey band of the fore wing continued subbasally, followed by a thin dark brown band; a medial band, an ochreous-grey discal shade with black spear-shaped marks on its outer side; - a subapical white lunule and an indistinct submarginal lunular line. Underside with tle basal two-thirds pale and ochreous- tinged; fore wing with a subapical white spot and three in the disk; hind wing with a subapical small spot. ?. Paler than the male; the medial pale grey band ob- scure ; a whitish slightly sinuous line across the disk of the fore wing edged with brown, and continued across the middle of the hind wing ; a large subapical spot on the fore wing, with five discal white lunules, outwardly edged with black, the third and fifth with the white only indicated on the hind wing ; there is an antemedial band, a white subapical lunule, and a discal row of black lunules inwardly edged with white. Underside as in the male. Expanse of wings, ¢ 3755, 2 4,5 inches. Hab. Sitoli, Nias. 86 Colonel C. Swinhoe on new Erebus malanga, nov. 3. Head and thorax dark brown; abdomen grey with whitish and dark grey segmental bands, the first two segments black-brown: fore wing with the central band broad through- out, slightly curved, and ochreous white until it is sharply angled round the discoidal whorl-shaped mark, the upper part from the angle to the costa quite white; the black ring of the whorl is correspondingly sharply angled, its inner side inwardly edged with white, the centre portion very obscure ; the bill-hook is greyish pink ringed with black, and this colour runs right round the centre portion; all the upper portion of the wing is very dark, the subapical spot is fairly large, triangular, its lower point blunt, a small white dot outwards below it, followed by an irregular row of five white lunules outwardly edged with black, the first a double lunule, the lower lunules in a black suffusion, and a black angular patch outwardly edged with white on the hinder margin against the middle of the central band: hind wing with antemedial and medial blackish bands outwardly edged with ochreous grey ; a subapical white lunule; a much curved and recurved black lunular discal line inwardly edged with whitish ochreous, greatly protruded outwards in its middle, with a blunt square and ochreous suffusion on each side and a blackish suffused patch below the subapical lunule. Underside pale brownish ochreous, the outer marginal space suffused with brown :. fore wing with a whorl of whitish spots round the outside of the cell ; a subapical spot, seven discal spots, the fifth well outside: hind wing with a black spot in the cell, two indis- tinct outwardly curved brownish lines in the middle; a sub- apical white lunule, a small white dot below it; a discal black lunular line, disposed as on the upperside. @?. Very similar to the male. Expanse of wings, ¢ 4y45, 9 4425 inches. Hab. Malang, Java. Hrebus philippensis, nov. 3+ Chocolate-brown, tinged with ochreous: fore wing with a thin obscure whitish line from the basal fourth of the hind wing running towards the apex, but not continued beyond the whorl-shaped discoidal mark, which it curves round and thickens somewhat towards the costa; the space above this line dark brown to the apex, but the portion beyond the whorl is without the white line; the whorl line is black as Species of Indo-Malayan Ai terocera. 87 usual; on the inner side inside the black line is a narrow pinkish-ochreous stripe, its lower end curved and broadened, and joining a large black patch; a thin greyish-ochreous middle line edged with black across the wing, with a small outward angle at its middle; a blackish suffusion on the lower disk ; a subapical white rather large spot and four discal white spots in an irregular row, outwardly edged with black : hind wing with the base dark brown; an antemedial brown line with a pale outer edging ; a medial somewhat crenulate greyish-ochreous line in continuation of the middle line of the fore wing; a subapical white spot; a discal indistinct greyish-oclireous line, more or less lunular, the hollows of the lunules filled in with black, the row deeply curved above its middle and then deeply and bluntly outwardly angled below its middle; body concolorous with the wings; the first two segments of the abdomen black, the third pale grey. Under- side ochreous brown, the outer half dark, limited by a brownish thin band across both wings; the discal markings disposed as on the upperside, the white spots larger. 2. Brown with a lilac tinge ; abdomen with the first two segments black ; wings of a uniform colour, the upper dark portion of the male- only slightly indicated except towards the apex, which is dark; the whorl-shaped discoidal mark as in the male ; a broad white band across both wings, broadest on the hind wing, its outer side with points like a fringe; the discal markings as in the male. Underside pale ochreous brown; the medial white band macular on the fore wing, broad on the hind wing; the discal markings as on the upperside. Expanse of wings, ¢ 4, ? 4,45 inches. Hab. Cape Engano, Luzon, Philippines. Genus PatuLa, Guen. Patula does not possess the two curious chitinous plates in the connection between the 8th and 9th abdominal segments found in Argiva; Pl. XI. figs. 24, 25, 26, & 27 show the genitalia of the true P. macrops, drawn on the same plane as in the figure of the genitalia of Argiva. In the development of the coremata it agrees with Argiva; the structure of the valve and the shape of the penis are the chief points. The hind wing of the male has the costal half aborted, forming a fold turned over on the upper surface, containing a large glandular patch of flocculent hair; vein 4 runs to the functional apex, 5 from the middle of discocellulars, 6 to the fold, 7 and 8 very minute to near base of centre. 88 Colonel C. Swinhoe on new Patula moriola, nov. 9. More or less similar in pattern to the common Indian species P. macrops, Linn., but the antemedial line ends hindwards in two conjoined rings, the lower one touching the hinder-margin; it isa smaller insect, much paler in colour, without the purplish glow of patula, the brown colour having a distinct ochreous tinge ; it cerlainly cannot be the female of P. macfarlanei, which Hampson says is also to be found in Amboina, though the type came from Cape York in Australia, the markings being very different. Expanse of wings, 2, 575 inches. Hab. Amboina Isl. Patula oadoxia, nov. 3 9. Also very similar in pattern to P. macrops, but the outer transverse sinuous lines are farther apart on the fore wing and the submarginal line of the hind wing is not nearly so sharply doubled ; it is a very large Patula, larger even than macrops, and the colour is quite different, being paler aud more ochreous even than mortola. The genitalia, as might be expected, also differs from that of macrops; the valves of P. macrops are much broader, the penis is also different, there are larger bunches of cornuti and chitinous red, and the sacculus of the valves is much more developed (Pl. XI. figs. 26 & 27). Expanse of wings, ¢ 6,5, 2 5; inches. Hab. Alu Island, Solomons, a small island close to Short-. Jand Island. Two males, four females. Patula tpsa, nov. 3 ¢. Very similar in. pattern to macrops, but paler in colour and is a smaller insect ; the genitalia is also different ; the penis agrees somewhat with that of macrops, but the valves are much narrower; the difference is shown in the Pl; XE; fies 25. EXxpanse of wings, ¢ 5, 2 5-54 inches. Hab. Kandy, Ceylon. Family Noctuide. Brevipecten promona, nov. 3. Palpi white beneath, dark brown above; antenne Species of Indo- Malayan Heterocera. 89 grey; head, body, and fore wing dark grey, the ground- colour being white, thickly irrorated with dark grey atoms ; thorax with a brown stripe down each side: fore wing with the lines darker grey, subbasal, from the costa to vein 1 indistinct ; antemedial line slightly oblique from costa to hinder margin ; ; medial line similar, its upper part lost in a large jet-black patch from the costa, its inner side deeply excavated and edged with white, a grey line closing the cell ; postmedial line outwardly oblique from the costa, acutely angled and inwardly oblique to the hinder margin close to the termination of the medial line; marginal line crenulate, some brownish suffusion on the margin ; cilia greyish brown : hind wing pale grey, whitish towards the base and abdominal margin ; ‘terminal jine dark grey ; cilia white on the lower half, grey upwards, intersected by a grey line. Underside: both wings evenly pale grey ; a w hite subapical small patch on the fore wing, with a black spot on its inner side, which is in continuation of an indistinct grey discal transverse line. Expanse of wings 1, inch. llab. Cape York, N. Queensland (Déme/). Has some resemblance to B. captatus, Butler, from India, of which I have both sexes. Capnodes asulca, nov. ?. Head, body, and wings dark pinkish brown, very uniform in colour throughout : fore wing with a black spot in the cell and four in a cluster at the eae a curved dark mark on the costa near the apex, with a ‘disjointed white streak on its inner half; a discai transverse sinuous row of white dots from the inner end of the streak across the wing, each dot with a black dot on its inner side; a row of sub- terminal black dots: hind wing with a discal row of similar white and black dots and subterminal black dots. Underside paler ; a discal indistinct thin band and subterminal black dots on both wings ; aie brown. Expanse of wings, 2,14 inch. Hab. Khasia Hills. Diomea nasea, nov. 3. Very dark olive-brown, nearly black, very uniform in colour ; palpi white on the inner sides, a white stripe on each shoulder ; thorax and both wings with numerous round white spots : fore Wing with costal spots at equal distances apart, with minute dots immediately below them; transverse rows 90 Colonel C. Swinhoe on new of basal, antemedial, postmedial, and submarginal spots and some medial white specks, the postmedial row consisting of three rows, the others of two rows and a marginal series : hind wing with indications of a medial white line and many white spots covering the outer half of the wing: legs with white bands. Eixpanse of wings, 3, 1;p inch. Hab. Kuching, W. Borneo. Oresia camaguina, nov. &. Palpi brown ; head and collar orange; thorax and fore wing dark ochreous brown; very dark and uniform in colour on the fore wing, making the markings very obscure and difficult to trace ; a darker streak on the median vein; an oblique. straight double line from apex to hinder margin, its upper half filled in with pale dull ochreous, a narrow brown shade from its middle to the lower end of the cell, then in a straight line to the middle of the hinder margin ; two white ochreous patches on the outer margin, in its middle and at the hinder angle touching each other; cilia dark brown: hind wing white, the veins and streaks in the interspaces pale grey. ?. Much as in O. emarginata, Fabr., from the Indian region, but all the markings on the fore wing more or less obscure. Expanse of wings, ¢ 1,%, ? 1,% inch. Hab. Camaguin Island, near Manilla, Philippines (Semper). Genus SERICIA. Sericia, Guen. Noct. iii. p. 172 (1852), type speetans, Guen., from Australia. Spiredonia, Hampson, Moths India, ii. p. 457 (1894) (nec Hubner). Sericia sumbana, nov. gf 2. Fore wing narrow, much narrower than in any other species of this genus; upperside with the ground-colour pinkish grey, suffused in parts with pinkish brown; mark- ings much as in the common Indian species, S. zamis, of Cramer; the discal ocellus filling the lower curve of a figure of €, small: hind wing of the same pinkish-grey ground- colour, with the usual familiar markings. Underside much paler and brownish grey. Expanse of wings, ¢ ¢ , 2755 inches. Species of Indo-Malayan Heterocera. 91 Hab. Samba Island, south of Flores Island in the Timor Sea (Doherty). I have four males and one female of this very distinct form. Family Hypenide. Genus GLOBOSUSA, nov. 6. Antenne unipectinated, palpi long and somewhat up- turned, the first two joints thickened and with stiff paired bristles, the last joint very slender, with bristles before its end ; top of head with short thick hairs which protrude some- what in front; all the legs naked, with very long spurs ; both wings rounded in a circular form: fore wing broad, costa and hinder margin straight, cell broad, discocellulars nearly straight; vein 2 froma little beyond the middle of the cell, 3 from about halfway from it and the cell-end, 4 and 5 from the end; 6, 7, 8, and 9 deeply curved, 6 from upper end, 7, 8, and 9 stalked: hind wing with vein 2 from the middle of the cell, 3 and 4 on a short stalk, 5 from the cell- end, 6 and 7 from the upper end, 8 free, recurved, touches 7 near its base. Type, G. curiosa, mihi. A very curious-looking moth. (rlobosusa curtosa, nov. 3. Antenne grey, palpi blackish brown, legs yellow striped with black on the upperside; head, thorax, and fore wing saffron-yellow : fore wing with faint indications of subbasal, antemedial, and postmedial grey lines ; a blackish postmedial patch on the costa and black dots on the outer margin: hind wing yellowish white, indications of a recurved medial grey line, its lower part with black spots on veins 3 and 2 and two near the abdominal margin; indications of a postmedial outwardly curved grey line and black lunular spots on the outer margin. Underside uniform yellowish white; fore wing with a linear black spot in the cell, a smaller one at the end, small postmedial and subapical brownish marks ; hind -wing with a small lunular discoidal black spot. IEixpanse of wings, g, 1 inch, Hab. Saugir Island, south of the Philippines (Doherty). 92 Colonel GC. Swinhoe on new Bertula adra, nov. 3. Upperside: head, thorax, and fore wing dark olive- brown; traces of antemedial, medial, and postmedial out- _wardly curved, somewhat sinuous brown lines; a submarginal straight white Tine inwardly edged with dark brown from the costa near the apex to the hinder margin close to the angle: hind wing brownish grey, a faint brown lunule at the end of the cell ; traces of a medial outwardly curved brownish line ; a white submarginal line from close to the hinder angle, angled outwards, then crenulate upwards, and becomes obsolete before reaching ‘the costa. Underside grey: fore wing with some brownish suffusion on the upper part, whitish along the hinder marginal space; a postmedial, outwardly curved, crenulate brown line ; a straight brownish submarginal line ; the outer portion of the wing whitish: hind wing white, thickly irrorated with brown atoms ; a brown lunule at the end of the cell; two outwardly curved crenulate brown lines, outwardly edged with white, corresponding to the two lines on the fore wing. Expanse of wings, g, 1 inch. Hab, Jaintia Hills, Assam. Genus WILKARA, nov. gS. Antennee simple ; palpi upturned, very long, second joint very long, rising much above the head, densely hairy, third joint concealed by the hairs; hind legs ‘with the tibiee densely hairy, the tufts of hairs extending, reaching halfway down the naked tarsi; thorax crested; abdomen smooth : fore wing narrow, costa nearly straight, apex somewhat rounded, outer margin convex, hinder anglesomewhat rounder, hinder margin slightly convex: hind wing with the costa straight, apex and hinder angle rounded, outer margin nearly Straight : fore wing with vein 2 from the middle of the cell, 3, 4, and 5 from the lower angle, 6 and 7 from upper angle ; a long brush of stiff straight hairs from the subcostal vein crossing the upper end of the cell, with some shorter similar hairs beyond it: hind wing with vein 2 from before the middle of cell, 3, 4, and 5 from the lower end, 6 and 7 from upper end, 8 free. Type, W. nigerrima, nov. Species of Indo- Malayan FHeterocera. 93 Walkara nigerrima, nov. 3. Upperside dark uniform black, with a slight lilac tinge : fore wing with a small white dot in the middle of the cell, a white spot at the end ; a brown, nearly erect, antemedial line, a white subapical costal dot, a black apical spot ; an oblique, straight, brown, thick line from this spot right across both wings, outwardly edged with whitish, to the abdominal margin of the hind wing beyond the middle. Underside : fore wing coloured like the upperside, the costal space above the culersital vein pinkish grey, the outer veins streaked with pinkish grey; the brush of hairs grey: hind wing black, the abdominal space pale. Expanse of wings 1,4 inch. Hab. Kalim Bungo, Central Nias (Kannegieter). Bomolocha olypea, nov. 6. Head, body, and wings dark pinkish grey: fore wing with the costal line black; a large medial black aan across the wing, its inner edge upright ‘but bent inwards a little on the costa, its outer edge from. one-sixth from the apex with many outward dentations to vein 3, then with a slight inward curve obliquely to the hinder margin a little bey rond the middle ; no other markings on either wing. Underside pale uniform brownish grey, fore wing with some blackish suffu- sion on the basal half, Bxpanse of wings, ¢, 1 inch. Hab. Mahableshw ar, Bombay Presidency. Bomolocha commiztura, nov. 3. Upperside olive-brown; the ground-colour is really whitish, but the whole surface of both wings is densely irrorated with olive-brown atoms: fore wing with a black discoidal spot; traces of a whitish, outwardly curved, ante- medial line ; a postmedial white line, inwardly edged with black, outwardly oblique and incurved below the costa, then slightly sinuous, straight down with a slight incurve to vein 2, then with smooth inward curve to the hinder margin beyond the middle; traces of a white sinuous eee line; a white marginal lunular line outwardly black-edged ; cilia with indistinct white inner line: hind wing paler; an I indistinet, whitish, outwardly curved, postmedial, sinuous 94 Colonel C. Swinhoe on new line, the outer margin marked like it is on the fore wing. Underside brownish grey, with some greyish-white streaks in the interspaces. Expanse of wings, ¢, 1;%5 inch. Hab. Lombok Island, between Bali and Sumatra. Bomolocha variegata, nov. @. Palpi and head greyish ochreous, thorax greenish brown, wings greyish ochreous: fore wing with the costal line ereenish brown, a patch of that colour in a triangular form filling the cell and the basal part of the next lower interspace ; the outer part of the wing similarly coloured, an apical curved ochreous-grey streak in it which joins the ochreous-grey space between, the hinder portion of the wings ochreous grey ; marginal line brown, crenulate, and with white points; cilia ochreous grey: hind wing without markings, the margins as on the fore wing. Underside ochreous grey, as also are the body and the legs: fore wing with a white spot at the end of the cell and two subapical white spots, the latter nearly obsolete in the type-specimen. Expanse of wings, ? , 3% inch. Hab. Kina Balu, N. Borneo. Bomolocha uniformis, nov. &. Palpi, head, thorax, and fore wing dark greyish ochreous ; a blackish discoidal spot, no other markings : hind wing grey, also without markings. Underside : body, legs, and wings uniformly grey, no “markings except for an in- distinct darker grey discoidal spot on each wing. Expanse of wings 1,’ inch. Hab. Jaiutia Hills, Ag am. Family Nymphnulide. Dracenura arfakalis, nov. 3S ?. Palpi brown, white beneath; collar grey ; head, thorax, and fore wing ‘dark purplish owe : fore wing ait the veins blackish ; a black spot in the cell and another’ at the end, no other markings: hind wing pure white ; a brown marginal band with irregular inner margin, thickened some- Species of Indo-Malayan Heterocera. 95 what at the apex: abdomen with the basal half grey, with some white on the segments; anal half black, tuft white. Underside: fore wing paler, a black discoidal spot; hind wing as on the upperside ; body and legs white. Expanse of wings, ¢ 9, 1-176 inch. Hab. Arfak Mts., N. New Guinea, 4000’ (Pratt). EXPLANATION OF THE PLATES. Prats VII. Fig. 1. Maurilia instabilis, p. 71. Fig. 2. tconica, p. 71. Fig. 3. —— tunicata, p. 71. Figs. 3a, 4. —— undaira, p. 71. Puate VIII. Fig. 6. Carea subtilis, p. 73. Fig. 7. intermedia, p. 73. Fig. 8, Acontia talauta, p. 74 Fig. 9. migrator, p. 74. ‘Pruate IX. Fig. 10. Gadirtha impingens, p. 70. guineana, p. 70. Fig. 12. Amphipyra surnia (Yokohama, Japan), p. 67. Fig. 18 yama (Asama Yamay Japan), p. 67. Fig. 13.4 pyramidea (England), p. 67. Fig. 14 magna (Punjab, India), p. 67. PratreE X. Fig. 20. Argiva hieroglyphica, p. 81. Fig. 21. Catocala fraxini, p. 81. nupta, p. 81. Pratt XI. Fig. 24, Patula macrops, p. 87. Fig. 25. ipsa, p. 88. Figs. 26, 27, ordovia, p. 88. 96 Mr. A. W. Waters on IV.—Some Mediterranean Bryozoa. By Artour Wm. Waters, F.L.S., F.G.S [Plate XII. ] In my collection there are many specimens which I have intended to describe or revise, but the description of various large collections has prevented, and I am glad now to make a beginning by dealing with five interesting forms~from Naples and Oran :— Pedicellina hirsuta, Jullien. Lepralia bifurcata, sp. 0. Lepralia circumeincta, Neviani. Lepralia oranensis, sp. n. Lagenipora tgnota, Norman. Pedicellina hirsuta, Jullien. (PI. XII. figs. 1, 5.) Pedicellina hirsuta, Jullien, ‘ Bryozoaires, Mission”° du Cap Horn,’ p. 18, 1888. ‘The small specimen from Naples seems to correspond with Jullien’s description, and has large recurved spines all over the zocecium, curved and pointed at the base, and their form suggests that they were movable. ‘The peduncle is large and is also covered with spines, while the stolon is much narrower than the peduncle. In my specimen I am not able to see clearly the base of the peduncle or the adjoining stolon, but believe it is correctly drawn. The contraction near the base has no appearance of being accidental, though more complete material is desirable. This specimen was referred to in my description of the Red Sea Bryozoa*. It will be noticed that the zocecium and peduncle are very exceptionally large (calyx about 0:38 mm., peduncle about 0:11 mm.). Loc. Tle Hoste, Orange Bay, 26 met.; Naples. Lepralia bifurcata, sp.n. (PI. XII. figs. 2, 3, 4.) In specimens from Capri the zoaria have two branches bifurcating at a very wide angle (fig. 2 a). Round the zoarium there are but few zooecia, from four to eight, either surrounding an imaginary axis or slightly flattened. The zocecia are irregularly quadrate, granular, * Journ. Linn, Soc., Zool. vol. xxxi. p. 252 (1910), some Mediterranean Bryozoa. 97 having the oral aperture contracted at the side, with the part below the contraction narrower than the part above. At each side of the oral aperture there is a small, raised, rounded avi- cularium, and any of these may be replaced by a large spathulate one, in one case both avicularia being thus re- placed. Usually the spathulate avicularia are directed distally, but one is diagonal, or it may be directed proxi- mally. The bar to the avicularium has a small central denticle. The granular ovicell is globular, widely open, so that. the operculum cannot close the ovicell aperture. At the bifur- cation there is a large round opening with a raised border (fig. 3), the object of the opening is not clear. It might have been for a large avicularium, or for a radicle, but the position does not make this probable. It is much like the fossil Characodoma hallt, Maplestone*, from Mornington and Mitchell River, Victoria, Australia, which, however, has the quadrate zoarium articulated, and the ovicelligerous zocecia are surrounded by irregular nodules ; however, the shape of the zocecia is the same with the ovicell in the same position, but in C. hall: there are small triangular or spathulate avicularia replacing the semicircular or spathu- late ones of L. bifurcata. Loe. Capri, 50 fathoms. Lepralia circumeincta, Neviani. (Pl. XII. figs. 6-10.) Hippoporina circumeincta, Neviani, “ Bri. neoz. di aleune Loe. d'Italia,” pt. 8, Bull. Soc. Rom. per gli Stud. Zool. vol. v. p. 118, fig. 7 (1896); Bri. postpl. di Spilinga, p. 28, fiz. 11 (1896); “ Bri. neog. delle Calabrie,” Pal. Ital. vol. vi. p. 187 (73), pl. xvii. figs. 10, 11 (1900). Lepratia grimaldi, Jull. et Calyet, Bry. de l’Hirondelle, p. 70, pl. ix. tig. 5 (1903). Cheilopora circumeincta, Levinsen, Morph. & Syst. Stud. p. 3538 (1909). This does not appear to be uncommon at Naples, and Kirchenpauer left a manuscript description in the Zoological Station, calling it Lepralia dohrni. When the manuscript was shown to me, it was my intention to describe and figure the species, using the name given by Kirchenpauer, and I have sent away some specimens explaining that Kirchenpauer had given it this manuscript name. When my paper on the Naples Bryozoa was written it had * “Further Desc. of Tertiary Polyzoa of Victoria,” Proc. Roy. Soe. Vict. vol. xiii, n.s., p. 7, pl. ii, fig. 17 (1900). Ann. & Mag. N. Hist. Ser. 9. Vol. ii. 7 98 Mr. A. W. Waters on not come before me, nor had the specimens first met with any ovicells, but they occur from Oran and from Capri. Neviani evidently had very small pieces fossil, and did not describe any ovicell. He speaks of it as incrusting, though with frag- ments it might be difficult to be certain of this; from Naples and Capri it is unilaminate, whereas from Oran all except one piece are bilaminate, back to back. Jullien and Calvet, in describing ZL. grimaldi, do not say whether it is uni- or bilaminate. Neviani described the surface as rugose, Jullien and Calvet say with small perforations, and both are correct as regards Capri and Oran specimens, which are covered by large granules and in between there are small pores, The nature of the granules varies in different parts and in some con- ditions they are the most noticeable, while in others the pores are the most distinct, but none could be deseribed as smooth. The piece figured is very regular, but this is not always the case. There is a small triangular avicularium at one or both of the upper corners of the zocecium. There are about 27 tentacles in the Naples specimens. There are usually 4 distal multiporous rosette-plates near to the basal wall and 4-6 lateral ones. The ovicell is coarsely granular, but the granules are not so large as those figured by Jullien and Calvet. It is not raised, but shows beyond the oral aperture buried in the distal zocecium. The ovicelligerous zocecia have a much wider oral aperture than the ordinary zocecia, with the proximal edge straight, while the distal border forms the curve of a wide are (see fig. 8). The ovicell has much the same shape as that of Flustra foliacea, L., passing to the basal wall, the wall between the distal end of the zocecium and the ovicell does not appear to be calcareous. It is very difficult to know in which genus this should be placed. Neviani * made the genus Hippoporina for all species indicated by modern authors as Lepralia—that is to say, all that have a horseshoe-shaped oral aperture; he then men- tions H. pertusa, Esper, which should therefore be the type of Hippoporina. In Part II. of the same paper, also 1895, he mentions H, foliacea, Ell. & Sol., and then H. integra, sp. n., which he figures. Canu calls this the type, but it is not the first mentioned. In Part III., 1896, Neviani men- tions A. imbellis, Busk, and H. adpressa, Busk; then, further * “Bri, neoz. di aleune Localita d'Italia,” 1895, p. 109, and Waters, “ Bry. from Zanzibar,” Proc, Zool. Soc, 1918, p. 515, some Mediterranean Bryozoa. 99 on in the same paper, he describes and figures H. circum- cinetd, nov., and H. spilinge, noy. At one time, through an error in binding, I was misled into thinking H. cireumeincta was the first mentioned and therefore the type of Hippo- portna. Neviani also includes H. edaz, Busk; H. tessulata, Rss.; HH. depressa, B.; H. complanata, Norm. ; H. foliacea, KE. &§8.; H. pallasiana, Moll. Neviani also described the -genus as new in “ Bri. foss. della Farnesina,” Pal. Ital. vol. i. p. 107 (1895), where he mentions first A. foliacea, kK. & S.—that is to say, in 1895 he described it as new in two places, in one mentioning first foliacea, in the other H, per- tusa. Which of the papers was first published is not indi- cated, though in Neviani’s ‘ Publicazione Diverse’ the “Bri. neoz.’”’ comes first. Canu *, in his “ Bryozoaires des Terrains Tertiaires,” in- cludes under Hippoporina several fossil species, describing or mentioning the ovicells in all but two, but unfortunately his photographs only show them in three .cases. H. angi- stoma, Rss., is included, but with its small roundish oral aperture it does not seem closely related to many of the species mentioned by Neviani. Levinsen f places circumeincta in his genus Cheilopora, in which some of the species have the ordinary and ovicel- ligerous zocecia similar, but in circumcincta and prelucida the ovicelligerous zocecia have different and larger oral aper- tures than the ordinary zocecia. One of my specimens of prelucida with an ovicell is from Tartary f, and an ovicell has not been referred to by anyone else. It is globular, raised, perforated, and granular, about as wide as_ the zocecium, and is not directly closed by the operculum, for, as the ovicell is at a lower level than the operculum, connec- tion with the ovicell is cut off when the operculum closes the oral aperture. The operculum of L. s¢ncera has a nearly straight proximal edge with a thickened border parallel to the distal edge, and the operculum of Cyelicopora prelonga, Hincks, is very similar, so that it is unfortunate he gave the name prelonga to two species which may have to come into the same genus. At one time the dimorphism, as seen in edreumeincta, would have been thought sufficient reason for separating it gene- * Ann. de Paléontologie, vols. ii—iy. p. 77, + Morph. & Syst. Stud. p. 353, { The Tartary specimen has avicularia, as figured by Hincks, who, however, says no avicularia; so perhaps he did not recognise that they were avicularia, and in my specimens “from Singapore or the Philip- pines ” there are none. 7* 100 Mr. A. W. Waters on rically from forms in which it does not occur; but this cannot now be maintained. In Adeonellide this difference was made a generic character, but we now see that it only occurs in about half * the species. In Lepra/ia dimorphism is known in depressa, B.; bistata, Waters ; cincta, Hincks ; cleidostoma, Sm.3 circumcincta, Ney. In Hippothoa it is sometimes found, as also in many Catenicellidge; also in Caleschara and Mcnoporella waipukerensis, Waters, in Cri- brilina clithridiata, Waters, and in Schizoporella subimmersa, MacG., &e. In describing Lepralia grimaldi, Jullien says that the dimorphism of the zocecia in this species is enough to shake our confidence in the characteristic value of the oral aperture, but the reason for this is not clear, for the opercula of the ordinary zocecia will have the shape of the species both in colonies with or without ovicelligerous zocecia, and it is therefore a character of the greatest use—besides, in some cases the relationship may also be shown by the ovicelligerous zooecia. Iam not sure that Hipporina will stand as containing the present somewhat heterogeneous collection, nor do I feel at all satisfied with the family Hippopodinide, Lev., for circum- cincta has not a thin-walled zocecium, the nature and shape of the ovicell is very different in eircumeincta and prelueida, and then the slight difference in the distal wall in Chezlopora and Hippopodina is a trifling character, the difference in the rosette-plates may or may not be of importance. Under the circumstances I, provisionally, at least, adhere to Lepralia of Hincks, and to me it seems that the wisest and simplest thing would have been to have done so generally, and to have gradually removed species to other or new genera when there was sufficient reason for so doing; for as time has gone on it has become clear that many things were incorrectly grouped together under Lepralia. Loc. Naples, 45 fath.; Capri, 30 fath.; Oran, 54 fath. (specimens given by Canu); Bay of Biscay, 240 metres iw C.). Fossil. Spilinga, Calabria, post-Pliocene (NV.) ; Monteleone, Calabria, Pliocene (N.); var., Carrubare, Calabria, Upper Pliocene (NV.). * “A Structure in Adeonella,” Ann. & Mag. Nat. Hist. ser. 8, vol. ix. p. 497 (1912). some Mediterranean Bryozoa, 101 Lepralia ovanensis, sp.n. (PI. XII. figs. 11-13.) The zoarium grows as a hollow cylinder (2-3 mm. diam.), or irregularly, in places forming more than one layer. ‘The oral aperture is exceedingly long with a marked con- traction about the middle, the lower edge being nearly straight or slightly curved upw ards, and the distal half of the oper- culum is very thick, almost semiglobular. On each side of the zocecium there is a long narrow avicularium, directed distally and extending beyond the line of the aperture. Although there are several specimens, no ovicells lave been found. ‘The surface of the zocecium is irregular, but cannot be called granular, and in the Oran specimens pores are seldom visible, whereas in the Liberia specimens they are more easily followed, there being three or four on the front and some by the avicularium. There are two distal rosette-plates near the basal wall. When only some of the tubular specimens had been seen there was thought to be some similarity to Fedora excelsa, Jull., but this idea was abandoned on finding more material. There is often a groove-like mark on the dorsal surface. In a box in Jullien’s collection from Petit Tahou, Liberia, there were a considerable number of specimens, together with Cupularia canariensis and an erect Porella, and specimens therefrom of Z. oranensis were given to me in the Musée d’Histoire Naturelle, Paris. Loe. Oran, ‘zone coralligene,” 54 fathoms ; Petit Tahou, Liberia. Lagenipora tgnota, Norman. (Pl. XII. figs. 15-17.) Layenipora ignota, Norman, “ Polyzoa of Madeira,” Journ. Linn, Soc., Zool. vol. xxx. p. 309, pl. xlii. flee 10-13 (1909) ; Osburn, “ Bry. of the Tortugas Islands,” Pub. Carnegie Inst. of Washington, No. 182, p- 214 (1914). The zocecia are small, and there are very narrow vicarious avicularia placed upon a kind of mound. No zocecia have been found with two peristomial avicularia, whereas the central peristomial avicularium is well-marl ked, having a chamber much the same shape as that figured by Savigny for his Cellepora lancreti, in which the ovicells are different. The ovicells have a row of pores within the ridge, as is usual in Lagenipora, whereas L. socialis, Uincks, to which I have frequently referred *, has a pore at each corner, and as * Journ. Linn. Soc., Zool. vol. xxx. p. 174 (1907); Proc. Zool. Soe. 1913, p. 511; Proc, Zool. Soc. 1914, p. 856. 102 On some Mediterranean Bryozoa. this has not been figured a somewhat diagrammatic figure is » given (fig. 14). In various species besides the usual row of pores there may be one or two near the centre of the area, and in a specimen from Glenelg, South Australia, the whole of the ovicell area has numerous pores. ‘This last is very closely allied to my LZ. caminata, in which a few pores may be seen between the rows. In L. costazii, Aud., besides the usual row of pores at the distal edge of the area, there is frequently another row at the proximal edge, as is also the case in L. lacinosa, Calvet, which may be costazii, Aud. The two straight sclerites of the mandible (fig. 16) are quite similar to those of L. lucida, and I only know them in these two species and Z. caminata. Something of the kind oecurs in Thalamoporella rozieri, Aud. The oblique peristomial avicularium, figured by MacGuilli- vray in his Lagenipora nitens, occurs also in the ‘ Challenger’ L. bilabiata, B.; in what has been called C. granum; in the L. lucida, H.; in LZ. diadema, MacG. LL. ignota, may be only an erect form of ZL. lucida, and there are many cases of Cellepora in which the young and the adult forms have received different names. Both have the diagonal peristomial avicularium and the long narrow avicularium. Loc. Madeira, 70 fath. (V.); Tortugas, 12 fath. (0O.) ; Oran, 54 fath. From material given by Mons, Canu. EXPLANATION OF PLATE XII. dig. 1. Pedicellina hirsuta, Jullien, x 85. a, epines, xX 250. From Naples. Fig. 2. Lepralia bifurcata, sp. n., X 25. a, natural size. From Capri. Fig. 3. Ditto. x 25, Showing the bifurcation and large round opening. Fig. 4. Ditto. x 50, Showing an ovicell and two spathulate avicularia. Fig. 5. Pedicellina hirsuta, Jullien. x 12. Fig. 6. Lepralia circumeincta, Neviani. X12. From Oran. ‘ig. 7. Ditto. xX 85. Operculum. Fig. 8. Ditto. x 85. Operculum of ovicelligerous zocecia. Fig. 9. Ditto, Lateral wall, showing rosette-plates. Fig. 10. Ditto. Distal wall, Fs 5. Fig. 11. Lepralia oranensis, sp.n. X 25. From Oran. Fig. 12. Ditto. x 85. Operculum. Fig. 13. Ditto. x 85. Mandible. Fug. 14. Lagenipora socialis, Hincks. Showing ovicell, somewhat diagrammatic. Fig. 15. Lagenipora ignota, Norman, x 50. -From Oran. Fig. 16. Ditto. xX 85. Mandible. Fig. 17. Ditto. x 85. Operculum. Mr. W. K. Fisher’s Notes on Asteroidea. 103 V.—Notes on Asteroidea.—Il. By Water K, FIsuer, Director, Hopkins Marine Station of Stanford University, California. [Plate XIII. } The Genus Freyella—In a revision of the Brisingide * recently published in this Magazine, I divided the old genus Freyella into two groups, Freyella and Freyellidea. I- made Freyella spinosa, Perrier, the type of Preyella, since no type was designated originally. ‘he old generic name was retained for those species which are distinguished by having united first adambulacral plates, a syzygial joint between the first and second adambulacral plates, conspicuous proximal marginals, the first of which is closely joined with its vis-a-vis to form a pair directly above the united first adambulacral plates, and by having, instead of two gonads to a ray, a considerable series along either side of each ray. Unfortu- tunately none of these points except the first is brought out in Perrier’s figures or mentioned in the description, since such details have generally been omitted as of no particular importance. In part they furnish a key for a natural generic analysis. Through the courtesy of Dr. H. L. Clark, of the Museum of Comparative Zoology, I recently examined an authentic example of Freyella spinosa received from the Muséum d’Histoire Naturelle. It belongs to the group which I called Freyellidea. This specimen, no. 1447, has two gonads to each ray, each gonad consisting of a good-sized clump of tubules with a single aperture to the exterior. There is no syzygy between the first and second adambulacral plates ; no syzygial joint between the upper end of the second and third ambulacral ossicles, although the interval is very narrow ; there are no supero-marginals directly above the first adambu- lacrals. ‘The first and second, and in one interbrachium also the third, adambulacral plates are joined to the corresponding adjacent plates of the next ray, although not so closely as in the other generic group, there being considerable tissue between the supposed plates. It was this feature, figured by Perrier, which led me to suppose that FP’. spinosa belonged with the group containing J’, fecunda, F. spatulifera, and others, in which the first adambulacrals are always tightly joined. For the present it is best to consider this character * Ann. & Mag. Nat. Hist. (8) xx. p. 418. 104 Mr. W. K. Fisher’s Notes on Asterordea. as of secondary importance in true Freyella, which is really not very closely related to the genus containing F. fecunda. The latter is distinguished by a syzygy, well-developed marginals for the interbrachium, and serial gonads. For the genus, which I called Freyella, I propose the name Freyellaster, with Freyellaster fecundus (Fisher) as type. In this group belong Freyellaster spatulifer (Fisher), Macassar Strait, 901 fathoms; Freyellaster scalaris (A. H. Clark), Galapagos Islands, 812 fathoms; and probably also Freyella polycnema, Perrier. The group which I termed Freyi/idea will therefore become Freyella, with Freyella spinosa as type, and Freyellidea will. drop out as a synonym. The Genus Hymenodiscus, Perrier.—In the paper.on the Brisingidz above referred to, this genus was‘not placed in the synoptical key owing to lack of data. I have since Fig. 1. Fig. 2. Fig. 1.—Hymenodiscus agassizi. An interbrachium from above first marginal plates, dotted. Fig. 2.—Hymenodiscus agassizi. An interbrachium from actinal side. a, adambulacral plates; am, ambulacral plates; ¢, interradial ; m, marginals; 0, mouth-plates. —~ examined Perrier’s type in the Museum of Comparative Zoology (no. 1448) *. The type of Hymenodiscus agassizt is almost certainly a very immature specimen, as it is small, and there are no gonads. ‘here are no skeletal arches on the rays and the greater part of the thin abactinal integument * For description see Perrier, 1884, ‘ Mémoire sur les étoiles de mer recueillies dans la mer des Antilles,’ p, 189, pls. i. & il, Mr. W. K. Fisher’s Notes on Asteroidea. 105 has been removed. The fine spinulation of the disk extends upon the base of the ray. The abactinal integument of the ray, although very delicate, contains a single layer of lattice- work holothuroid plates, some of which at the very base of the ray bear minute spinelets.. From this it would seem that the abactinal wall of the ray is destined to be similar to that of Freyel/a, unless in the fully adult animals the plates retain their embryonic character. The interbrachium resembles that of Brisinged/a, but differs in having the first marginals (those which bound the apex of thie interbrachial angle) unequal in size, as shown in the accompanying figures (figs. 1 and 2). In Brisingella these plates are equal, and the suture between the interradial ends is on a line with the interradial, or median oral, suture. There is a distinct syzygy Nonneen the first aad second adambulacral plates. The interbrachia are not so open as in Brisingella, as the inner ends of the first adambulacral plates are normally in contact, or very nearly so. In an adult specimen we would expect to find these plates still closer together. It is worth noting that in Freyellaster and in Brisinga, s. s., the first eae plates are of unequal size (see figs. 1 and 2, m; “New Genera and Species of Brisingide”). Yet in mw present juvenile form the inter- brachial angle is different from that of either Preyellaster or Brisinga, while the entire absence of costal arches, as well as of gonads, may reasonably be attributed to immaturity. It does not seem possible to identify this problematical form with any other genus, except the even less known Gymno- brisinga of Studer. Gymnobrisinga sarsti (Abhandl, Akad. Wiss. Berlin, Anhang, Abth. 2, 1884, p. 13, pl. iii., fig. 5) is based upon a ees ray only. This lacks a dorsal skeleton, and while the large pedicellaiia figured by Studer is different from those of Hymenodiscus agassizt, I am quite unprepared to offer an opinion as to the generic distinctness of the two species. The Relationships of Labidiaster.—Although Labidiaster is Rory g generally considered to be a member of the Brisingide, I would : suggest that it has few essential characters in common with that family. The genus to which it exhibits greatest structural similarity is Coronaster*, Perrier. Coronaster * See Fisher, “The Asteroid Genus Coronaster, Perrier,’ Proc, Biol. Soc. Washington, vol. xxx. pp. 23-26, Veb. 21,1917. Coronaster includes the following nominal species :—C, parfaiti, Perrier, type, C. antonit, Perrier, C. briareus (Verrill), C. volsellatus (Sladen), C. octoradiatus (Studer), C. d¢spinosus, Ives, C. halicepus, Fisher. I have examined 106 Mr. W. K. Fisher's Notes on Asteroidea. seems to be more nearly allied to Pedicellaster than to either Heliaster or to any of the recently proposed genera of Asteriidee. I would therefore place Labidiaster in the Pedi- cellasterida. I have dissected a large example of Labidi- aster radiosus, Liitken, from the Straits of Magellan. Labidiaster differs from “Brisinga, Odinia, Freyella, and similar genera in the following important particulars :— (1) Its abactinal skeleton is not duplicated in the Brisingide ; (2) forficiform, or straight, pedicellariw are present ; (3) the adambulacral plates are crowded, very short in proportion to width, and entirely unlike in form and armature the same highly peculiar plates of all Brisingide ; (4) the ambula- cralia are shorter, especially the dorsal ends, which overlap, or imbricate with, the next adoral ambulacral plate, while in the Brisingide there is no sign of imbrication, the ambula- cralia resembling the centra of chordate vertebrae, with vertical articulating adoral and aboral facets. In the Brisingide (in the narrower sense) the abactinal skeleton of the rays is variable, being in the form of trans- verse, independent, parallel ridges or coste, separated by areas of integument without plates ; or the intervals may be partially or completely filled in with more or less imperfectly developed plates immersed in the body-wall; or the arches may be absent and a tessellation of thin plates may cover the genital region of the ray ; or there may be thin plates, more or less spiniferous, together with differentiated transverse coste. In Labidiaster the skeleton of the ray is closely similar to volsellatus, briareus, and halicepus. Coronaster includes Heterasterias, Verrill, type Astertas volsellata, Sladen. In the above paper the following remarks occur :—“ The family affiliations of Coronaster are not easy to determine, its lineage being somewhat involved. The tendency to crowding in the arrangement of pedicels partakes of the Asteriidx, while its mouth-plates are quite as ‘ brisingoid ’ as those of Odinia, and perhaps more so than the oral angles of Labidiaster, two groups placed in the Brisingidz. Its skeleton is more like that of a simplified Pedicellaster than like that of Asterias or allies. Parenthetically, the mouth-plates of Pedicellaster are more prominently ‘adambulacral’ than those of any genus of the Asteriide, even of Coscinasferias, and are nearly or quite as prominent, relatively, as the oral angles of Brisinga. In Pedicellaster and Coronaster the ambulacral plates are more ‘ brisingoid,’ uncrowded, and the pedicel-pores are in two series, even if later the feet themselves lie in four ranks. In very large specimens of Coronaster the pedicel- pores form two slightly zigzag rows, much less pronounced than in small specimens of Coscinasterias (in the broader sense), and the ambulacralia are less crowded. My own feeling is that, until we arrive at a more satisfastory basis for the subdivision of the Asteriidz than is now current, it will be much better to leave Coronaster in the Pedicellasteride.” Mr. W. K. Fisher’s Notes on Asteroidea. 107 that of Coronaster. There is a longitudinal series of tri- lobate infero-marginal plates, one of quadrilobate or cruci- form supero-marginal plates, and one of cruciform median radial plates. ‘lhe marginals and radials form regular transverse series. On the basal portion of the ray there is a more or less irregular zigzag series of trilobate dorso-lateral plates. The primary plates either connect directly by their slender lobes, or these are joined by one or two overlapping, oblong, intermediate ossicles. There results an open, fairly regular, reticulate skeleton having large tetragonal meshes (except where the dorso-lateral plates frame pentagonal openings). On the outer part of the ray the longitudinal, intermediate, connecting plates and the longitudinally oriented lobes of the marginals and radials gradually disappear, so that there remains only a series of independent, transverse, slender skeletal bands, simulating those of Brisinga, but having a very different history*. The skeletal meshes contain numerous papule. The form and armature of the adambulacral plates are as in Coronaster. ‘The arrangement of the pedicellarie either in retractile wreaths surrounding the spines or in retractile transverse cushions is not unlike that found in Coronastert+. The mouth-plates of the Bri- singide, of Coronaster, Pedicellaster, and of Labidiaster are similar in general form, those of Labidiaster being relatively the smallest. The features which are chiefly relied upon to distinguish the Brisingide, and to which the family in part owes its characteristic appearance, are conspicuous by their different form in Labidiaster. Such, in the Brisingide, are the elongate and peculiarly formed adambulacral plates; the Jong needle-like subambulacral and marginal spines, with their characteristic sacculate sheaths; the variable but always non-reticulate abactinal skeleton of the rays; the presence of only crossed or forcipiform pedicellarie. The genus Rathbunaster (type Rathbunaster californicus, * Verrill, in his ‘ Monograph of the Shallow-water Starfishes of the North Pacific Coast,’ 1914, p. 352, proposes a new genus, Labidastrella, for Labidiaster annulutus, Sladen. ‘It differs considerably in structure from LZ, radiosus, especially in haying the dorsal and superomarginal plates nearly abortive distally, on the rays, beyond the genital regions.” It is evident that this tendency to lose the dorsal skeleton of the distal part of the ray manifests itself in Z. radiosus, and is carried further in L. annulatus, I agree with Kcehler that it does not form a safe basis for a generic division between two otherwise similar species (Kcehler, Ann. de l'institut océanographique, vol. vii., fasc. 8, May 1917, p. 8). + See Sladen’s figures of Asterias (= Coronaster) volsellata, ‘ Challen- ger’ Asteroidea, pl. cvii. 108 Mr. W. K. Fisher’s Notes on Asteroidea. from off California, deep water) was described by me as a neighbour of the curious polybrachiate Pycnopodia of Stimp- son. I think the genus is related, instead, to Coronaster. It is notable for the suppression of the alternate supero- marginal plates and the reduction of the abactinal skeleton to spaced circular plates without trace of connectives. The marginal and abactinal plates bear an acicular spine surrounded by a retractile sheath with an expanded distal crown covered with numerous pedicellaria. ‘Che ambulacral, adambulaeral, and oral plates are similar to those of Coronaster. In Labidiaster, Coronaster, Rathbunaster, and certain genera of the Brisingide there are two gonads to each ray ; each gonad opens upon the side of the ray at some distance from the base. All three genera, as well as the Brisingide, have a single ampulla to each tube-foot. The family Pedicellasteride, if these views are correct, would consist of the subfamily Pedicellasterine with Pedi- cellaster, Lytaster, and Gastraster, and of the Labidiasterine with Labidiaster, Coronaster, and Rathbunaster. Asterina coronata and Asterina cristatan—In the ‘ Archiv fiir Naturgeschichte,’ vol. xxxii., 1866, p. 73, von Martens describes Asterina coronata from Batjan, Molucca Islands, and from Larentuka, Flores Island, and records its occurrence at Amboina. His description states that the relation of the minor to the major radius is as 1 to 2 or 22, that the abactinal plates are so arranged that the dorsal surface has a honey- combed appearance, the plates bearing five or more spinelets, and that scattered over the dorsal surface are groups of two to four heavy spinelets with a common base, such groups being found on the sides and radial regions of the ray, but not close to the border. On the disk these special spinelets outline an irregular pentagon. In the ‘ Proceedings of the Biological Society of Washing- ton,’ vol. xxix., p. 27, Feb. 1916, I described Asterina cristata from the Caroline Islands, the special peculiarity of which is the presence of a variable number of abactinal plates (upward of fifty to a ray), elevated and tubercular in form, and sur- mounted by one to five unequal, robust, pointed spines, the largest being four or five times as long as the spinelets of the other plaies, and many times greater in diameter. These elevated plates, with their tuft of enlarged spines, 1 take to be the same as von Martens’s ‘* Biischel von 2-4 starken Stacheln mit gemeinsamer Basis,” which he says, “ stehen auf den Armen ziemlich zerstreut, sowohl auf dem Ricken als an den Seiten, aber nie ganz nahe am Rance.” Thus Mr. W. K. Fisher’s Notes on Asteroidea. 109 the chief character of the two species is the same. As Dr. H. L. Clark has suggested in a letter, the two species are probably the same, although there. exist certain discre- pancies. Von Martens does not mention subambulacral spines, but states that the furrow-spines are ‘in einer Reihe, 4 oder 5 fast gleich Grosse auf jeder Platte,” and that the actinal intermediate plates have two relatively long sharp spines. The type of Asterina cristata has two to four, mostly three, actinal intermediate spinelets, usually six furrow-spinelets webbed for about half their length, the three or four median conspicuously longer than the laterals, and usually four subambulacral spinelets, of which the two median are much longer than the laterals, I think it is possible that von Martens overlooked the small lateral furrow-spinelets, although not likely ; but certainly in no specimens seen by me are the furrow-spinelets ever subequal. The case is somewhat complicated by two specimens .of a race of coronata which I saw some years ago in the British Museum. One was contained in a box with Vepanthia macu- lata, labelled “‘ Migupou, 7 to 12 fathoms, fine sand and coral —Cuming.” The other was labelled “ Port Hssington, Australia.” In the first specimen there are twenty or twenty-five of the prominent plates to each fifth of the body. ‘he actinal intermediate plates have, 1 in the neighbourhood of the furrow, about five or six spines in a rude circle, one spine being longer than the others ; near the ambitus there are three spinelets, with often one or two standing mesad from the principal comb. The furrow-spinelets are five or six, webbed, the laterals shorter than the mesial spinelets ; the subambulacral spinelets are four or five, shorter and stouter than the furrow-spinelets, and also graduated in size, the mesial spinelets being longest *. I made no notes on the Australian specimen, but my impression is that it does not materially differ from the other, Thus the actinal intermediate spinelets are more numerous than in the types of coronata and cristata, while the adambu- lacral armature is about the same as that of cristata. The prominent abactinal plates are fewer than in cristata, and more like the condition in Japanese specimens. Dr. Seitaro Goto, in his work on Japanese Asteroidea, carefully figures and describes a species from the southern parts of Kyushu and adjacent islands which he calls Asterina nove-zelandie, Perrier, but which I believe is a form of * For the privilege of examining these and many other specimens of Asteroidea in the British Museum (Natural History) I am indebted to Professor F, Jeffrey Bell, 110 Mr. W. K. Fisher’s Notes on Asteroidea. coronata, as it possesses the prominent abactinal plates so characteristic of coronata, Thus there are records from southern Japan to northern Australia. As a beginning towards straightening the tangle of appa- rent races, I would suggest the subjoined scheme. Any further evidence for or against it, or in any way bearing upon the status of Asterina coronata, will be most welcome :— a’, Abactinal spiniform pedicellariz present ; 8 adambulacral furrow-spinelets ; 8 or 9 marginal mouth-spinelets; 12 to 14 en- langed. abactinal plates 22)... 6s iinet ws Asterina coronata eu- erces* (Fisher). (Palawan.) a’. No spiniform pedicellariz present ; furrow- spinelets 4 to6; marginal mouth-spinelets 5 or 6. b‘. Actinal intermediate spinelets usually more than 3; near the furrow 5 or 6, forming a circle or group (not a straight comb); furrow-spinelets 5 or 6; 20 to 25 prominent abactinal plates to each fist’ WEG IT. sake es tine wceles bette Asterina coronata fasci- cularis *, subsp.n, (Migupou; Port Essington?) b*, Actinal intermediate spintlets 2 or 3, but not often 4, ce’, Furrow-spinelets 4 or 5; actinal inter- mediate spinelets usually 2; promi- nent abactinal plates moderate in number (up to 25 to each fifth of body) and with as many as 25 spinelets BOMB PLAbO | ier iniew ss eerauti sate tee oes Asterina coronata coro- nata, yon Martens. (Southern Japan, Batjan, Larentuka.) ~ * Fisher, Proc. Biological Society of Washington, vol. xxx., May 23, 1917, p. 91. Ulugan Bay (near mouth of Baheli River), Palawan Island, Philippine Islands, 2 to 5 feet, mud, sand, sea-weeds. + This new race is certainly different as regards the actinal inter- mediate armature. Von Martens states that there are two spinelets in coronata. Of course, specimens may prove to be variable. M. Alvin Seale, of the Museum of Comparative Zoology, who has lived many years in the Philippine Islands, tells me he has sailed past a fairly well-known Migupou Point ; but I have not been able to locate it, with available maps, on Mindanao or on Luzon. Mr. Seale does not recall upon which of the two islands the point isfound. It is quite possible that thiy is the locality from which so many of Gray’s types were derived. t So far as true coronata is concerned, the remarks concerning the number of prominent plates and the number. of spinelets on these plates are conjectural. These observations refer to the Japanese form, described and figured by Dr. 8S. Goto (‘A Descriptive Monograph of Japanese Asteroidea,’ 1914, p. 650, pl. xix., figs, 279-281), which may, of course, be quite distinct from typical coronata of the Moluccam region, Mr. H. A. Baylis on Dicroccelium lanceatum. Ls e*, Furrow-spinelets 6; actinal interme- diate spinelets usually 5 (2 to 4); prominent abactinal plates numerous (more than 30 and as many as 50 to each fifth of body) and with not more than 15 spinelets to a plate, frequently TORORHE RE JO tia Shc tage eo Asterina coronata cris- tata (Fisher) *. (Caroline Islands.) EXPLANATION OF PLATE XIII. Type of Asterina coronata cristata (Fisher), VI.—ZJs Dicroceelium lanceatum a Parasite of the Cat? A Note on a new Variety. , By H. A. Bayuts, B.A. (Published by permission of the Trustees of the British Museum.) [Plate XIV. ] REFERENCES have occasionally been made in helminthological literature f to the occurrence of “‘ Distomum lanceolatum ” t in the cat. These cases have, however, in recent years been generally discredited, and it has been suspected that the parasites recorded belonged to one or other of the species of Opisthorchis or Clonorchis (O. felineus and C. s’nensis) known to occur in cats, these forms being more or less similar to Dicrocelium lanceatum in size and superficial appearance, though differing widely from it in their internal structure. The typical D. /anceatum is a well-known parasite of sheep and cattle, and of various other herbivorous mammals; it is algo an occasional, and probably accidental, parasite of man, having been met with some six times. Its occurrence in a carnivore, however, is a point with regard to which some scepticism is not unnatural. When, therefore, I received some time ago some ‘T'rematodes taken from the liver of a cat, I was greatly interested to find that they belonged un- doubtedly to the genus Dicrocelium, and differed from the typical D. lanceatum only in certain very small anatomical * This form is probably distributed over western Oceania. It seems to be readily separable from the Japanese form, which has been classed as true coronata, although it probably is not. + See, e.g., Leuckurt, ‘Die Parasiten des Menschen,’ I., Abth. 2, p. 860; von Linstow, ‘ Compendium der Helminthologie,’ p. 30. { Synonymy: Fasciola lanceolata Rudolphi, 1803; Distomum lan- ceolatum Mehlis, 1825; Dicrocelium lanceolatum Dujardin, 1845 ; Dicrocelium lanceatum Stiles & Hassall, 1897, 112 Mr. H. A. Baylis on Dicroccelium lanceatum. details. These specimens, of which there is a considerable number, were collected at Georgetown, British Guiana, by Mr. G. E. Bodkin, Government Biologist, during November, 1915. ‘hey were kindly handed to me for determination by the Imperial Bureau of Entomology. On a consideration of the many resemblances between these examples and the typical D. lanceatum, and of the minor points in which they differ from it, I am inclined to regard them as belonging to a well- marked variety of that species, rather than a distinct form. The one salient feature is the position of the testes, which in the specimens under consideration invariably lie symmetrically opposite to each other in the same transverse plane. All authorities are agreed in describing the testes of LD). lanceatum as being placed nearly “ tandem,” i.e, one behind the other, but somewhat diagonally, near the longitudinal axis of the body * The exact position of the testes is, as a rule, a very constant specific character in Trematodes ; but in this case the almost complete correspondence between the rest of the anatomy and that of the typical form seems to outweigh such a considera- tion. The only other differences that L have been able to find are in the somewhat smaller size of the cirrus-sac and the slightly larger average size of the eggs. Hven the coils of the uterus show complete agreement, as far as they can be traced. For the sake of comparison, however, with the type, it may be worth while to give a fairly full description of thie new variety. The length of the worms varies between 5 and 7 mm., and the maximum widths for these lengths respectively are 1°62 mm. and 2 mm. The body is flattened dorso-ventrally, narrowing considerably from side to side in front, and less so behind. The posterior end is drequently somewhat rounded; sometimes, however, it is more pointed than in the example figured. ‘l’o the naked eye the body is whitish and semi-transparent (in spirit), the masses of fully-formed eggs in the uterus being visible as blackish or brownish patches. ‘The skin is smooth. The oral sucker is subterminal, and has a diameter of * Neveu-Lemaire (‘Précis de Parasitologie humaine’) gives a figure of D. lanceatum (reproduced in Brumpt’s ‘Précis de Parasitologie,’ 2nd ed. 1918, p, 335), in which the testes are symmetrically arranged ; but there is no reference to the source of the specimen from which the original figure was drawn, and no description of the internal anatomy is given in Neveu-Lemaire’s ‘work. The figure is, in other respects, very rough and inaccurate. Mr. H. A. Baylis on Dicroecelium lanceatum. 113 0°37 mm.*. The ventral sucker is situated -0°7 mm. behind it, and measures 0°4 mm. across. The mouth is followed immediately by a small, almost globular pharynx, measuring 0°15 mm. in length, and this is succeeded by an cesophagus _0°2 mm. long. ‘The two simple intestinal diverticula extend backwards to within a little more than 1 mm. from the posterior end. They lie, for the greater part of their length, near the lateral margins of the body. The excretory vesicle is small and inconspicuous. Its pore is terminal. The genital pore is median, situated between the two suckers and at about the level of the bifurcation of the intes- tine. ‘The testes are large compact bodies, slightly lobulated, especially on their lateral margins. They lie, as has been noted already, symmetrically opposite to each other, imme- diately behind and at the sides of the ventral sucker, and between the intestinal diverticula. Each testis measures about 0°8 mm. in length and 0°6 mm. in width. The ovary is a body of variable shape, but usually somewhat lobate; it is situated close behind the testes, but its position shows considerable variation. It appears to be rather more com- monly situated on the right side than on the left, but in three out of eight stained examples the ovary was placed behind the left testis. There is a rather large rounded receptaculum seminis, situated just dorsally to the posterior edge of the ovary. Laurer’s canal is present, and a shell- gland, not differing from that of the typical D. lanceatum. The cirrus-sac is about 0°4 mm. long and 0°15 mm. wide. It contains a coiled vesicula seminalis. The cirrus-sac partici- pates in the variability of position shown by the ovary and its associated organs. Thus, when the ovary is on the right, the cirrus-sac lies to the right of the terminal portion of the uterus; when the ovary is on the left, the positions of the genital ducts are generally reversed. The vitelline glands lie within the middle third of the body, and extend along the sides as a series of lobes of various sizes. ‘The two vitelline ducts are given off somewhat in front of the middle of the glands, and cross the body to unite into a much wider single duct just behind the ovary. The uterus fills almost the whole of the middle and posterior portions of the body, from the level of tie anterior end of the vitelline glands to the tail. Its coils, for the most * This and the following measurements are taken from an example 5 mm. long, and are therefore to be regarded as somewhat below the mean. Ann. & Mag. N. Hist. Ser. 9. Vol. ii. 8 114 Dr. J. D. F. Gilchrist on the Eggs and part, take the form of transverse folds and lateral loops. In the middle region these are confined to the space between the vitelline glands, but more posteriorly they sometimes extend laterally beyond the intestinal diverticula. The ascending limb of the uterus passes forward between, and ventrally to, the testes. The eggs are roundish-oval in shape, and when fully formed have a rather thick brown shell, usually showing an indentation on one side, so that in profile one side is convex, the other concave. ‘The eggs measure 42°5-50 wu X 30-35 p. The variety described above I propose to call Dicrocelium lanceatum St. & Hass., var. symmetricum, in allusion to the arrangement of the testes, This variety being at present known only from specimens collected from a single host, a cat, it is doubtful whether it should be regarded as a “local” variety or asa form peculiar to cats. An examination of examples of D. lanceatum from sheep or other herbivorous animals in the same locality would be of great interest from this point of view, as well as a further investigation of the parasites of cats. In any case, it would appear that the older helminthologists may have been correct in reckoning the cat among the hosts of “ Distomum lanceolatum.”’ EXPLANATION OF PLATE XIV. Dicrocelium lanceatum, var. symmetricum. Ventral view of a stained specimen. C.S., cirrus-sac; Jnt., intestinal diverticulum; Ov., ovary; #., receptaculum seminis; 7., left testis; V., vitelline glands; V.S., ventral sucker. VII. — The Eggs and Spawning-habits of the Pilot Fish (Naucrates ductor). By J. D. F. Grucurist, M.A,, D.Sc., Ph.D. In the course of .a general enquiry into the spawning- habits of Cape fishes, a mature female of the pilot fish was found. The eggs and larve of about thirty Cape fishes have been described in local publications, but, as the pilot and its peculiar habits are so well known, and have attracted attention in all parts of the world, a description of the mature eggs of this fish, hitherto unrecorded, may be worthy of a special note, and interest a wider circle of readers, more especially as the nature of the eggs seems to Spawning-habits of the Pilot Fish. 115 throw light on some peculiarities in the behaviour of the fish. The pilot fish is not uncommon in the Cape seas. The young are frequently abundant in the summer months, being found in company with .the young of Lichia amia, which they somewhat resemble in the characteristic markings of the body. The adults are well known, under the name *‘Lootsman,” to Cape fishermen, who state that they are always found accompanying a large shark, called the “ Tor- nijn Haai’’ or porpoise-shark (Charcharias melanopterus). They take up a more or less constant position near the body of the shark, and remain within a few inches of the base of the pectoral fin. The fishermen have also noted that they have the habit of(darting away from the shark towards any strange object, and then returning to their former position. This well-kuown behaviour, interpreted in other parts of the world as a guiding or piloting of the shark to its food, the Cape fishermen believe, is for the purpose of a preliminary tasting or testing of the food on behalf of the shark. On one occasion, in the month of December, a specially large pilot fish was caught on the hook by some fishermen fishing off Cape Point. It was in the company of a porpoise- shark. By placing the fish in a bucket of water, it was possible to keep it alive, and convey it to the Marine Laboratory at St. James, where it was placed in a large tank, and seemed none the worse for its capture. It proved to be a mature female with ripe eggs, which were extruded on slight pressure. These extruded eggs were readily seen, being large, though quite transparent. When placed in water, however, they became almost invisible. They did not float, and they adhered to each other and to objects with which they came in contact. The shape of the eggs was distinctly oval, though a few were more rounded. A typical example, shown in the accompanying figure (p. 116), measured 1°74 mm. in length and 1°3 mm. in greatest breadth. In another case the measurements were 1°65 x1°39 mm. There were very minute dots on the surface of the egg, and from one pole originated a single fine filament. This was of considerable length, being in one case six times the length of the egg, or about 10 mm. In most cases it was shorter, and in some it appeared to have broken off close to the egg. The filaments readily became entangled with each other, so that it was difficult to separate out any particular one with- out breaking it. At its base the filament had a broad attachment to the outer membrane, of which it is apparently 116 Dr. J. D. F. Gilchrist on the Eggs and a modification. At this point it was about -04 mm. in dia- meter, but soon diminished to about 016 mm. The filament appears to be homogeneous throughout, but, if treated with hot caustic potash, it has the appearance of a thick-walled tube. At the distal pole of the egg, opposite that from which the filament arises, there is a marked differentiation of the surtace of the egg, on a small terminal area about ‘2 mm. in diameter. This area is covered with clear polygonal markings, which vary in size, being large towards the periphery, where they fade off into the surrounding surface. Near the centre they become smaller and less distinct, and Nw at Egg and filament of Nawerates ductor. pass into a small thickened ring, in the centre of which the micropyle may be clearly seen. There is a large perivitelline space, about a fourth of the diameter of the whole egg in breadth at the middle of the egg. In the specimen figured this breadth was °32 mm. The egg proper or yolk is an ovoid mass, somewhat more oblong in shape than the outer shell of the egg. Itis clear, but granular, and no traces of vesiculations nor oil-globules were seen. In preserved material several cases were ob- served in which the yolk had shrunk away from its surrounding perivitelline substance, and, in such cases, at the distal end opposite the micropyle, a small funnel-lke Spawning-habits of the Pilot Fish. 117 projection appeared, which in the normal condition would penetrate the yolk-mass to a slight extent. It doubtless has some function in the mechanism of fertilisation, though no canal connecting it with the microphyle was detected in the perivitelline substance. The mode of origin of the filament of the egg is different from what is found in some other filamentous eggs of Teleosts. ‘Thus, in the egg of the South African species of Hemi- rhamphus and Atherina, I have noticed that, in the immature and even fairly small ovarian eggs, the filaments occur as irregular streaks on the surface of the zona radiata, but, in those of Naucrates, the filament is already free, and serves to attach the egg to the wall of the ovary. A number of such filaments are inserted at one spot, so that the ovarian eggs are often grouped in grape-like clusters, or the fila- ments become twisted on each other to form a rope-like structure, round which the eggs are grouped. The presence of filaments on fish-eggs, as a rule, has been found to be associated with the fact that they are anchored to each other or to foreign objects, floating or lying at the bottom of thesea. Thus the eggs of Hemirhamphus, Belone, ‘ and Eocetus have been found attached to each other and to sea-weed in this way, though Scombresox, another member of the same family, is said to have pelagic eggs provided with filaments. Another family, the Atherinide, all the members of which have eggs provided with filaments, so far as is known, have demersal attached eggs. There is thus a reasonable presumption that the possession of filaments indicates that the eggs are, ultimately at least, attached to some object fixed or floating in the sea, and, if we suppose that the filamentous eggs of the pilot fish are attached to the shark, with which the fish is so intimately associated, it may explain some peculiarities in its habits which have received a variety of explanations. These are not entirely convincing, partly on account of this variety, but chiefly on account of conflicting facts or of lack of confirmation. Thus the explanation that the pilot feeds on the fragments of the food of the shark is not in accordance with the fact that small fish have been found in its stomach. The same objection applies to another conjecture that it feeds on the excrements of the shark, and still another that it feeds on the parasites on the skin of the shark. An explanation of a different nature, that the pilot keeps close to the larger fish for the purpose of protection from its enemies, is a more plausible one, but is somewhat strained when its very close 118 On the Eggs and Spawning-habits of the Pilot Fish. proximity is explained as a precaution against the attack of the shark itself. It is not in accordance with the supposed amicable arrangement whereby the pilot is allowed to have a share of food or excrement, in return for its piloting services. According to actual observation, the shark is not at all disconcerted by the absence of the pilot, but the pilot is said to be greatly agitated by the loss of the shark. It has even been observed “clinging to the side of a shark,” and, on one occasion, it is stated that it was seen to leap out of the water in an endeavour to follow a shark which had been caught by hook and was being hauled on board a ship. Another peculiarity in the behaviour of the fish, which seems to be of some significance in this enquiry, is the well- known fact that it sometimes accompanies large sailing-ships, which it follows so persistently that it is drawn far away from its natural habitat. It even follows the ship into the harbour, where it is easily caught. Most of these peculiarities would be sufficiently explained if we suppose the pilot’s eggs to be attached to the rough skin of the shark, or to the bottom of the ship, which is so persistently followed. We may recall in this connection the solicitude of such fishes as the Blennies for the safety of their eggs, how they keep close guard over them, driving off any approaching intruder. The close proximity of the pilot to the shark, the darting forward towards any strange object (which seems to be an undoubted fact), the persistence in following the shark or the ship in circumstances which are unfavourable to its own welfare, would seem to indicate a very powerful motive, not dissimilar to that of the fishes which guard their eggs. The fact that the young stages of Naucrates are frequently got, but that no pelagic eggs such as those above described have, so far as I can ascertain, been procured in tow-nets, seems to have some further significance in this enquiry and to indicate that the eggs of Naucrates are not floating. The only sufficient proof of the suggestion here offered would, of course, be the finding of such eggs attached to the body of a large shark or ship, which had been accom- panied by a pilot fish, and it may be that, with the above- mentioned facts in view, the opportunity may arise for the solution of the long-standing mystery of the pilot fish. On the Sika-Deer of North China. 119 VIII.— Notes upon the Sika- Deer of North China. By ARTHUR DE CARLE Sowersy, F’.Z.8., F.R.G.S. THE opportunity has recently been afforded me of examining a fully adult Sika stag, shot by Mr. J. Holmberg, of Tien- tsin, in the Fen-chou Fu district of West Shansi, during December 1916. Previous to this, I believe, no complete specimen of this animal has ever been secured by a Kuropean ; while, as far as I know, the only reference to it in any publication is that by Pére Heude in his ‘ Mémoires concernant [Histoire Naturelle de Empire Chinois’ (tome iv. p. 210, pl. xxxvii. fig. 13), wherein he names the species Cervus grassianus, from a single pair of antlers from T’ching-lo-hsien (‘T'sing-lo Hsien), Shansi. In a paper written by me on Pére Heude’s collection of pigs, sika, serows, and gorals in the Sikawei Museum, Shanghai, and published in the ‘ Proceedings of the Zoological Society of London,’ April 1917, pp. 7-26, I suggested that the Shansi sika should be classed for the time being with Milne-Edwards’s Cervus mandarinus, though I stated then that winter skins that I had seen were lighter in colour than the figure given by Milne-Edwards. The stag which Mr. Holmberg so kindly allowed me to examine is, however, fully as dark as Milne-Edwards’s winter figure, though in this connection it is interesting to note that Mr. Holmberg states that the hinds and young that he saw with the stag were very much lighter. This agrees with my own observations. I have had no opportunity of deter mining whether or not the hinds and young of the Chihli sika are lighter than the stags ; but as a result of my, examination of Mr. Holmberg’s specimen I do not hesitate to confirm Pére Heude’s separ ration of the Shansi sika from the other Chinese forms, and, although he gave no description, the fact that he gives a figure of a pair of antlers from T’sing-lo Hsien, West Shansi, makes his name hold good, Following i isa diagnosis and description of the species :— Cervus grassianus, Heude.- Cervus grassianus Heude, ‘Mémoires concernant l’Histoire Naturelle de Empire Chinois,’ tome iv. p. 210, pl. xxxvii. fig. 13. A single fully adult male in winter pelage examined, also two winter skins of fully adult females, and a summer skin of a male, as well as two fully developed pairs of antlers, all from West Shansi. g complete, from mountains 100 miles 8. W. of Fen-chou Fu, Shansi, N. China. 120 ; Mr. A. de Carle Sowerby on the Measurements in the flesh :—Head and body 60”, height at shoulders 42", tail 8’, hind foot 164", ear 7’. Weight 165 catties = 220 lbs. (about). Colour. A general greyish brown on the head, going into brown on the forehead and a pale buff at the base of the horns and the base and backs of the ears, the inside of the ears being white. Nose dark brown ; chin dark brown, almost black, with a small white patch on either side. Area round the eye buffy-grey. The general colour gets darker on the neck, but it still retains a wash of buff or ochre. The body is dark greyish brown, with a slight indication of a darker median dorsal line. The spots are almost invisible, showing up in certain lights and quite invisible in others. The dark greyish brown of the body shades into a rich brown on the back and lower portions of the legs, getting lighter and more ochraceous on the fetlocks. There is a peculiar patch of long white hairs surrounded by black on the outer surface of the hind leg about 6 inches below the heel. The tail is black above, white beneath, the hairs being long and making the tail somewhat bushy. ‘The croup disk is white, edged with black on its upper half, the black joining up with that of the upper tail surface, so that there is no white between the tail and the back. The under surface of the belly and inner surface of thighs are white; the chest is a dark brownish grey. The hairs of the neck are considerably longer than on the rest of the body. Horns. The horns in this specimen are not very well developed, being past their prime. They measure :— Right, 193" in length. Left, 193” 54 Right, above the brow-tine 33” in circumference, below 54". Hight points, 4+ 4. Other horns examined are large, graceful, and heavy, but not so large as is usual in C. mandarinus. Skull. Condylo-basal length 322 mm.; zygomatic width 136 mm.; interorbital space 100 mm.; length of nasals 125 mm.; greatest width of palate (at post-molar) 54 mm,; greatest width of cranium 84 mm.; length of upper tooth- row 99 mm.; length of lower tooth-row 103 mm. Teeth well worn. Type. A pair of antlers in the Sikawei Museum, Shanghai, no number, from T'ching-lo-hsien (Tsing-lo Hsien), Shansi. The habitat of this species may be considered as confined to the forested and mountainous areas of that part of Shansi Sika-Deer of North China. 121 that lies west of the Fen Ho. Even here it occurs only in a few isolated districts, namely :— 1. The forest to the south of Ning-wu Fu, west of Tsing- lo Hsien and north of Ko-lan Chou, where Heude’s specimen was doubtless secured, 2. In the forested area 90 miles west of Tai-yuan Fu, known as the Chiao-ch’éng Shan. 3. In the forested area 100 miles south-west of Fen-chou Fu, known as the Ning-hsiang Hsien mountains. Formerly its range extended throughout the whole of the mountainous area of West Shansi, as well as in the moun- tains that extend in a north and south line between Shansi and Chihli; but it has been almost exterminated by native hunters for the sake of its horns, which are highly valued as medicine. Only a few isolated herds occur in the districts above mentioned, where they keep to the densest parts of the forest. ven so, they are being steadily exterminated. This sika ruts in November and December, sheds its horns about March, the new growth commencing about the end of July. Itis during August and September that this species is most sedulously hunted by the natives, for then the horns are considered to be in their prime. Following is a diagnosis of the sika occurring in the Chihli forests -— Cervus mandarinus, Milne-Edwards. Cervus mandarinus Milne-Edwards, ‘ Recherches pour servir 4 l’His- toire Naturelle de Mammiféres,’ vol. i. (text), pp. 184-186, vol. ii. pls. xxil. et xxil.a, This sika differs from C. mantchuricus, Sw., in having the white spots larger and fewer in number, in being generally lighter in colour, with less white on the croup disk, and in having the parts below the belly the same colour as the flanks, instead of white. The differences in the winter pelage are not so marked, Milne-Edwards states that the spots in C. mantchuricus in the winter pelage are so invisible as not to have been given in Sclater’s figures. (In this it resembles the Shansi stag.) In: €; mandarinus, i in spite of the general darkening of the pelage, the spots remain plainly visible. In a letter published in the P. Z. 8. 1865, No. 1, p. 142, Swinhoe retains the name mantchuricus for the Manchurian sika, having examined a living specimen at New-chwang in South Manchuria. He makes the statement that he suspects it to be the same as the deer, skins of which he secured in the Summer Palace, and which Blyth called mantchuricus Ann. & Mag. N. Hist. Ser. 9. Vol. ii. 9 123 Mr. C. Chubb on new — (P.Z. 8. 1864, p. 109), but which Swinhoe himself sub-e- quently called hortulorum (P.Z. 8. 1865, p. 1). As there is no telling where the deer confined in the Summer Palace came from originally, it being just as likely that they were brought from Manchuria as from the Imperial Hunting Grounds, owing to the fact that part of the tribute annually paid to the Imperial Manchu household from Manchuria con- sisted of game of various kinds, and since Milne-Edwards finds the Chihli species so distinct from the Manchurian form, it seems more than likely that Swinhoe’s surmise as regards the common identity of his skins from the Summer Palace and his New-chwang specimen was correct ; in which case his name hortulorum applying to the Manchurian sika is‘ later than his name mantchuricus, and so becomes a synonym, thus leaving Milne-Edwards’s name mandarinus clear for the Chihli specimen. This species occurs in a wild state only in the Imperial Hunting Grounds, north of the famous Tung Ling (Hastern Tombs), and in the Wei-ch’ang to the north of Jehol, bothin Chibli provinee, to the north and north-east of Peking. It occurs In a semi-domesticated state in the magnificent park at Jehol. Up to recent times this deer has been strictly preserved, but in 1911-12 the Manchu soldiers that were sent out of Peking and were camped in the Eastern Tombs and Imperial Hunting Grounds were allowed to kill as many as they liked, while since that date native hunters have been allowed to hunt in these districts, with the result that in the wild state the species is practically extinct. It may here be stated that unless immediate and very stringent steps are taken for their protection, both C. gras- sianus and C. mandarinus will become extinct, and the sika no longer remain on the list of North China mammals. IX.— Descriptions of new Genera and a new Subspecies of South American Birds. By aes UaUBR Eee M.B.O.U. (Published by permission of the Trustees of the British Museum.) PSEUDOCONOPOPHAGA, gen. nov. The proposed new genus, which is based on Convpopiiare melanogaster, Menetr., is distinguished from Conopophaga, founded on Turdus euritue, Gmel., by its long and narrow bill, the long tarsi and toes, the larger size, and different coloration. Type, P. melanogaster (Menetr.). South American Birds. 133 MACKENZIANA, gen. nov. Reichenbach, in 1850, proposed the generic name JVisius, and gave a figure in his Av. Syst. Nat. Voég. pl. Ixxi., which has been associated by previous authors with Thamnophalus Jeachi, Such, as the type; but, when that bird is compared with the figure, it will be easily seen, that Reichenbach could not have founded it on that species, as it is not anything like it. The species was originally, and has for many years been, placed in the genus Zhamnophilus, Vieillot, where it was equally out of place, as it is so entirely different from that genus, which was founded on Lanius doliatus, Linn. I propose, therefore, the new generic title Mackenziena, with the following characters:—Head not crested, no concealed white dorsal patch, tail much longer than the wing. Bill short and stout, the depth about two-thirds the length of the exposed culmen. The wing, which is rounded, has the fifth primary longest. The tail is also rounded and much gradu- ated, the two middle feathers longest. Coloration: the male is black, with ovate white spots and bars to the feathers, and the female is brown marked with buff. Type, MM. leachi (Such). FREDERICKENA, gen. nov. The species which I propose to separate as a new genus under the above title has also been previously placed in the genus Thamnophilus, Vieillot, with which it has no near affinity ; it may be characterized by the absence of a concealed white dorsal patch. The nuchal crest is composed of rather broad feathers with rounded tips. The bill is short and stout, the depth being equal to about one-half the length of the exposed culmen. ‘lhe wing is rounded, the fourth, fifth, and sixth primaries longest and subequal ; the seventh is longer than the third, but shorter than the fourth. The tail, which is rounded and graduated, is about two-thirds the length of the wing. , The male is almost uniform in colour, but the female has the tail and entire under surface barred. Type, Lhamnophilus viridis, Vieillot. PICROTES, nom. noy., pro Lochites, Cab. & Hein. 1859 (nec Gistel, 1848). Type, Lantus severus, Licht. SAKESPHORUS, nom. nov., pro Hypolophus, Cab. & Hein. 1859 (nec Miiller & Henle, 1837). Type, Lantus canadensis, Linn. 124 On new South Americin Birds. POLIOLZMA, gen. nov. This form is readily distinguished in having the throat aniform with the rest of the under surface. The bill, which is long compared with the other genera of this group, has the exposed culmen about equal in length to the hind toe and elaw. The wing is rounded, the third, fourth, and fifth quills longest, the second about equal to the seventh. The tail is short and nearly square, the outer feather on each side is only very slightly shorter than the rest. The feet are small and weak. ‘he male and female are entirely different in colour. I propose, therefore, that this form be separated generically under the name of Poliolema. Type, Myrmotherula cineretventris, Sclater & Salvin. DICHROPOGON, gen. nov. The species which I propose to separate generically have hitherto been associated with Hypocnemis of Cabanis, but . it differs altogether im colour as well as in its proportionate measurements. The bill is small and narrow. The wing, which is slightly pointed, has the third, fourth, fifth, and sixth primaries longest, the second about equal to the eighth. The tail, which is nearly square at the tip, is about two-thirds the length of the wing. ‘he legs and feet are proportion- ately strong, the tarsus exceeds the length of the exposed culmen by about two-fifths. Male and female quite different in colour of plumage. This genus is based on Hypocnemis pwcilonota, Cabanis- Rhopias fulviventris salmoni, subsp. 0. Adult male. Differs from the adult male of R. f. fulvi- ventris (Lawr.) in being uniform olive on the top of the head, back, and sides of face, instead of greyish brown ; upper wing-coverts pale brown, not blackish ; tail paler; the white on the throat more extensive; breast buff instead of slate- grey ; abdomen and under tail-coverts paler and inclining to buff; under surface of guills pale brown, not blackish brown. Total length 110 mm.; exposed culmen 12; wing 50; tail 37; tarsus 17. Adult female. Differs from the adult female of 2. f. fulvz- rentris in being paler both on the upper and under surface. Wing 50 mm. Hab. Colombia and Ecuador. The type, which is in the British Museum, was collected by T. K. Salmon at Remedios, Northern Colombia. THE LIFE AND LETTERS OF SIR JOSEPH DALTON HOOKER, O.M., GCS1, FRS. By LEONARD HUXLEY. Based ¢ on material collected and arranend by LADY HOOKER, | The long life of Sir Joseph Hooker, which ee exactly one } hundred years ago, covers the whole of the scientific renaissance. ‘Pre-eminent in botany, his name will always be associated with the ' making of Kew; but as Traveller and Explorer he was hardly less eminent, whether in the Antarctic, the Himalayas, or the Atlas. The history of his scientific career is also the history of many | friendships, revealed in his unceasing correspondence, most of all _ with Charles Darwin, for he was Darwin’s confidant, helper, and eritic for fifteen’ years before the ‘ ‘Origin’ appeared. _ Two Volumes. Illustrated. 96s. net. ey te JOHN MURRAY, ALBEMARLE STREET, W. 1. SCIENTIFIC BOOKS AND SERIALS. 1 JOHN WHELDON & CO. have the largest stock-in the country of Books |. in all Departments of Science and Natural History, also Transactions and Journals of Learned Societies, etc., in sets, runs, and single volumes or numbers. Libraries or small parcela purchased for cash. SPECIAL NEW CATALOGUES: War Economy (Agriculture, Husbandry, Gardening, Economic Botany, etc.). Moerical severe) = cpeeeranhieal), 2d. each, post free. 38 «GREAT QUEER “STREET, KINGSUAT, LONDON, we. a Telephone : "Garcard Just Published, Pees 12s, 6d. THE FAUNA OF BRITISH INDIA | tren: (Including ‘Ceylon and Burma), Teacenies Published under the e authority of the Secretary of State for India j in Council. _ Medium 8vo, with Test Mustrations. _RHYNCHOTA, Vol. VII. Tawor & E Pannen, Red. Lion Court, Fleet Street, RO. 4. t i alee ple, ‘ ve SPP SE Ag abies Ske (RCA 2 CONTENTS OF NUMBER 7.—Ninth Series. I. Notes ‘i the Gatty Marine Laboratory, St. Androws.- — ee No. XLI. By Prof. M'Inrose, M.D., ‘LL.D. TBs. FRS,, he. (Plates VED cai hey aie ce ee oe ira III. New pects of Indo-Malayan rinaneerhl and Deberiptioas - of Genitalia, with reference to the Geographical Distribution of “Species rea¢in bling each other. 2 Colonel Ce sean Mz F.LS., &e. - are VII. ae eee Ta et ha Boi a8 net IV. Some Mediterranean Bryozoa, os Anoron Wa. ‘Warmns, 5 an ELS. F.GS. (Plate KIL) oe eusk ens FS at i ae OE aie ik. UP Hopkins Marine Station’ of * taiifora’ University, California. - Si (Plate KUT) 283, 32.2% as cae VI. Is Dicroccelium lanceatum a Parasite of the Cat? A Noteon ‘a new Variety. | mt Ae ein ae) . tock cog ah oat vs We Be Vil. The Eggs and d Spinwrninghabite of shis! Pilot Fish (Wanoratg ductor). By J. D. F. Gricurist, M.A., D. Sc., PhD. ote oe 4 VIII. Notes upon the Sika-Deer of North China. By ‘alan: RS pe Canty Sowensr, F.Z.8., ee aN ae a. TX. Descriptions of new Genera and a new Subnpece of ‘South American Birds. By CHARLES cea FZ. a8, M. B. 0. U.. ss aie aa : * Itis ica chat al Comtaaaioasiile. for shift Work may be sdtieae ” post-paid, to the Care of Messrs. Taylor and Francia, oe —. dion Court, Fleet Street, London. Es a DoT eg da | WITH FOURTEEN PLATES. - oa NINTH ‘SERIES. No. 8. THe ANNALS a ATURAL HISTORY, “ori INCLUDING | ZOOLOGY, BOTANY, ann GEOLOGY. CONDUCTED BY “WILLIAM CARRUTHERS, Pu. D., F.RS., F.LS., F. -G.S., ARTHUR E. SHIPLEY, M. A.B. PD FES EZ. — AND: RICHARD T. FRANCIS, F. Bs. Sa3h 4 BEING A CONTINUATION OF THE.‘ “ANNALS COMBINED WITH MESSE, LOUDON AND CHARLESWORTH’S * “MAGAZINE OF NATURAL HISTORY.’ ‘LONDON: by Simpkin, Seavohall Hamilton, Rent: & Co.; Ld. ; ; Bailliére, Paris: ‘and Hodges, Figgis, & Co., Dublin. WATKINS & DONCASTER — | : Raturalists, — Keep in stock every kind of APPARATUS wad CABINETS required by ENTOMOLOGISTS, ORNITH- OLOGISTS, BOTANISTS, &c. Also NESTING- BOXES, which should be fixed up in gardens or shrubberies at before the breeding Season. A Large Stock of Butterflies, Moths, Birds, Egys, &c. i Full Oatalogue (84 pages) mailed free to any address. 36, STRAND, LONDON, Misia ENGLAND. Hy LAND AND FRESHWATER MOLLUSCA OF INDIA. Part XII., Vol. II. Price 25s. Ge PSS he By Lieut.-Col. H. H. GODWIN-AUSTEN, F. .RS., &e. With 26 Plates and 132 pages Text, with doncripngne of ae new Species, A few Copies of other Parts of Vol. II. still on hand. Price 2ls. : TAYLOR and Franots, Red Lion Court, Fleet Stree. E.C. 4, Price 12s. 6d. Super-Royal a pp. iv, 296. With 4 Plates and 16 est figures. A CONTRIBUTION PHYTOGEOGRAPHY AND FLORA DUTCH N.W, ‘NEW GUINEA. L. Ps GIBBS, F.LS., F.R.MS. TAYLOR AND Francis, Red Lion See Fleet Street. Rates for Adverivenceia in the Rei a Magazine of Natural History, One aaa eee : Twelve Insertion. __ Insertions. _ Insertions. PAGE - ~ - - 2: (OV A eork 16 Ocach Bs Kab O each HALF-PAGE 4 02°58) 10265 YO ay ee ro: jt QUARTER-PAGE - TR Buk TAL, Bice ODE yom All applications for space to be pate to vo BS, oe aan Mr. H. A. COLLINS, 3 Birdhurst Road, | Croydon, eee THE ANNALS AND MAGAZINE OF NATURAL HISTORY. [NINTH SERIES.] No. 8. AUGUST 1918. X.—On some External Characters of Ruminant Artiodactula. —Part II. Thg. Antilopine, Rupicaprine, and Caprine, with a Note om the Penis of the Cephalophine and Neo- tragine. By R. I. Pocock, F.R.S. Tue first ger of this series of papers, supplementary to the account of the “‘ Cutaneous Glands of Ruminants” published in 1910 (Proc. Zool. Soc. pp. 840-986), was issued in the Ann. & Mag. Nat. Hist. for June of this year, pp. 426-435, It deal}with the Cephalophine, Neotraginz, Oreotragine, and Madoquine. The present communication comprises the Antilopine, Rupicaprine, and Caprine, the most inter- esting forms described being the two Rupicaprine genera Capricornis and Budorcas, of which 1 had only defective material for examination in 1910, As in the previous paper, the pagination inserted after generic and specific names refers to the original treatise published in 1910. Subfamily 4 wrrrorrvz. Genus Gazeta, Licht. In 1910 (P. Z. S. pp. 887-893) I described the preorbital, _ Inguinal, pedal, and carpal or knee-glands in the following species of this genus:—G. bennettii, subgutturosa, marica, muscatensis, dorcas, pelzelni, cuviert, rufifrons, and semme- ringtt. My descriptions were based upon fresh examples of all Ann. & Mag. N. Hist. Ser. 9. Vol. ii. 10 126 Mr. R. I. Pocock on some the species except G. semmeringzi, for which I was dependent upon a dried: skin. Since that date I have been able to confirm my observations upon additional and fresh material of G. bennetti, subgutturosa, rufifrons, dorcas, pelzelni, and semmeringii, and can now add to the hst one previously unexamined species—namely, G. dama. Some notes upon the examples of G. seemmeringii and G. dama may be of interest. Gazella semmeringit berberana.—Specimens from Somali- land (R. EH. Drake Brockman). The preorbital gland is of - moderate size or small. The pedal glands are quite normal. The inguinal glands are shallow wide-mouthed pouches external to the mamme. The carpal glands are thick pads of skin, covered with a mat of convergent hairs. In a male example the secretion from the inguinal glands smelt like sour milk. In a female the secretion from the same glands, like that from the knees, had a strong ovine scent, like that of a pen of domestic sheep, whereas the waxy secretion from the pedal glands resembled dogs’ dung in odour. The rhinarium (fig. 1, I) is a little less reduced than in typical gazelles, in which it consists of hardly more than a small irregularly pentagonal area of naked skin restricted to the septum between the nostrils (fig. 1, G, H). But in G. semmeringii its upper edge is slightly expanded and spreads a little to the right and left, partly hanging over the nostrils above. In the penis (fig. 1, B) the tubular prolongation of the urethra is short, barely projecting beyond the tip of the slightly swollen termination of the glans. It is shorter than in ordinary gazelles—e. g., G. bennettii (fig. 1, D) and G. rujfina, figured by Lonnberg in 1904. Gazella dama ruficollis.—Examples ( ¢ ? ) from the Soudan (G. Blaine). The preorbital gland is a shallow pit, quite small as compared with that of the typical gazelles. The pedal glands are quite normal. The inguinal glands consist of a pair of very shallow wide-mouthed pouches, one on each side just external to the corresponding mamma. ‘The carpal or knee-glands, on the contrary, are rather exceptionally well developed, consisting of a pad of thick skin, overgrown with a mat of mesially convergent hairs covered with scurfy secretion. The end of the penis in ane species is slightly enlarged and the urethra is prolonged as a thin tube a little beyond the tip of the glans (fig. 1, C). It has been ‘suggested that the three large white-rumped External Characters of Ruminant Artiodactyla. 127 Fig. 1. Parkes ES IVA 7 Cy ¢ WL A. Extremity of penis of Antilope cervicapra. B. The same of Gazella semmeringit. C. The same of G. dama. . The same of G. bennettiz. E. The same of Antidorcas marsupials. F. Section of the fore foot of Lithucranius walleri. G. Rhinarium of Gazella rujifrons from the front, x 3. H. The same from the side. I. The same of Gazella semmeringwi from the front, x 3. K. The same of Antilope cervicapra from the front, x 3. L. The same from the side. =) 128 Mr. R. I. Pocock on some African gazelles—G. granti, semmeringii, and dama—connect the smaller typical African and Asiatic gazelles with the springbuck Antidorcas; and Lydekker and Blaine (Cat. Ung. Mamm. iii. p. 85, 1914) adopt for them the subgeneric title Nanger, remarking that the group is replaced in South Africa by Antidorcas. Although I am only acquainted with the normal pedal glands of G. granti, 1 am unable to find in G. semmeringii and G. dama any justification for the view that they lessen the differences between the typical gazelles and Antidorcas, or that they represent the latter in north and east Africa more nearly than the other gazelles of that area represent it. In the same Catalogue another subgenus of gazelles is admitted under the name Procapra, comprismg the three central Asiatie gazelles picticaudata, przewalskii, and guttu- rosa, none of which is known to me apart from dried skins and skulls. Procapra was established by Hodgson for the reception of picticaudata, which, according to his description, differs from other gazelles in having no preorbital, inguinal, or carpal glands ; no trace of moist rhinarium, and the inter- digital fossee, described in one place as “ pores,” small. Moreover, on the positive side it possesses a large postcornual sinus, by which is meant apparently a gland behind the horns analogous to that of Rupicapra and Oreamnos. Ad- mitting the truth of these observations, and I do not see on what grounds they are to be disputed, picticaudata must. be recognized as generically distinct from Gazella, and prze- walskii, which at least resembles it in the absence of pre- orbital, inguinal, and carpal glands, must be associated with it—at all events, provisionally. Thespecies named gutturosa, on the other hand, resembles the typical gazelles in having preorbital, carpal, and inguinal glands, the first two being small and the last-mentioned large. Clearly, therefore, it must be severed from picticaudata and przewalskii, for which the name Procapra must be retained. But, according to Pallas, gutturosa possesses a preputial glandular sack, re- calling that of Moschus, Nototragus, and Sus. In this respect it differs, so far as is known, from all the species of Gazella. I propose, therefore, to dismember gutturosa from Gazella under the generic title Prodorcas. Genus AntTrpoRcas, Sund. Antidorcas marsupialis, Zimm. (p. 893). Several fresh examples of this species confirm in every External Characters of Ruminant Artiodactyla. 129 respect the constancy of the characters established in 1910, showing that, so far as the cutaneous glands are concerned, the genus Aniidorcas differs from Gazella in the absence of inguinal and carpal glands and the presence of the great dorsal gland. I may add that the rhinarium resembles that of Gazella in consisting of a small irregularly pentagonal area on the narial septum, and that the penis is also like that of Gazella, the urethral canal projecting a short way beyond the tip of the slightly swollen glans (fig. 1, E). Genus ANTILOPE, Pall. Antilope cervicapra, Linn. (p. 894). My observations upon the cutaneous glands of this antelope were based in 1910 upon two dried skins. Since that date I have seen several fresh specimens, confirming in all respects the characters previously established as distinguishing the genus Aniilope from Gazella. Two other differences are, however, supplied by the rhinarium and the penis. The rhinarium (fig. 1, K, L) is considerably better developed, and therefore less specialised than in Gazella and Antidorcas. Not only is it broader between the nostrils, but it is extended along their upper border nearly as far back as their posterior notch. In the penis, figured by Lonnberg in 1904, the urethral prolongation is longer and thicker than in Gazella and Antidorcas (fig. 1, A). Genus Litnocranivs, Kohl. Lithocranius walleri, Brooke (p. 896). I am indebted to the late Mr. F. C. Selous for the fore and hind feet and the skin of the inguinal area of this species from British East Africa. These show that the foot I examined and described in 1910 was, as suggested, distorted with respect to the glandular interdigital space. This space (fig. 1, F) differs from that of Gazella, Anti- dorcas, and Antilope in that it gradually deepens from its upper (or proximal) to its lower (or distal) end, where the thick interungual fold curves forward. In other words, the skin of the front of the pastern above the depression passes imperceptibly into the latter by a gradual inclination, without showing a.sign of the abrupt descent seen in the other genera. ‘lhe pedal gland recalls that of Rupicapra. There are two pairs of mammez, but no inguinal glands. 130 Mr. R. I. Pocock on some By their external characters, dealt with in this paper, and by their horns the genera of Antilopine here admitted may be briefly diagnosed as follows :— Genus GazeE.ia, Licht. Preorbital, inguinal, carpal, and pedal glands present, the pedal glands in the form of long and deep interdigital clefts of even depth throughout ; rhinarium a small irregularly pentagonal moist area on the narial septum, and not, or only to a very small extent, bordering the nostrils above; urethral canal usually only surpassing the glans penis to a small extent; horns in males with concavo-convex, usually sigmoid, curvature. Type, G. subgutturosa. Distribution. From Central and South-western Asia into India and North and East Africa. Far too many species of this genus appear to me to be admitted by Lydekker in the British Museum Catalogue. Genus Proporcas, nov. Distinguishable from Gazella by the presence of a preputial gland and a shorter tail, the structure of the pedal glands being unknown. Type, P. gutturosa, Pall. Distribution. Mongolia and Northern China, Genus ANnTILoPE, Pallas. Distinguishable from Gazella by the nakedness of the integumental web tying the hoofs together, by the larger rhinarium which borders the nostrils above, by the much longer and thicker elongation of the urethral canal of the glans penis, and by the spirally twisted horns. Type, A. cervicapra. Distribution. India. Genus Antiporcas, Sund. Distinguishable from Gazella by the absence of inguinal and carpal glands and by the presence of a large distensible glandular area on the back, which is peculiar to the genus. Type, A. marsupialis, Zimm. Distribution. Africa south of the Zambesi. External Characters of Ruminant Artiodactyla. 181 Genus Lirnocrantus, Kohl. Distinguishable from Gazella by the structure of the pedal glands, the floor of which gradually slopes downwards from the front of the fetlock, the cleft being deepest at its lower end, .where it is walled in by the heel-tie; also by the absence of inguinal glands and the presence of four mamme. Type, L. walleri. Distribution. British Hast Africa and Somaliland. Genus Procarra, Hodgson. Distinguishable from Gazella by the absence of the pre- orbital, inguinal, and carpal glands, the presence of a gland behind the horns, the reduced size of the pedal glands which apparently have a pore-like orifice, as in Ovis and Nemorhedus, and, it is stated, by the rhinarium being over- grown with hair. Type, P. picticaudata, Hodgs. Distribution. Mongolia, China, Tibet. Subfamily Rupreaprr az. Genus Rupicapra, Blainv. Rupicapra rupicapra, Linn. (p. 848). Several examples of the typical race of this species from the Tyrol have enabled me to verify, and in the case of some characters to extend, my observations, which in 1910 were based upon the carcases of two newly born kids and upon adult specimens living in the Zoological Gardens. Preorbital and inguinal glands are absent and the structure of the pedal glands is constant, the floor of the depression slopes gradually downwards from the front of the fetlock to the heel-tie, where the integument is folded forwards and upwards to form a ridge constituting the distal well of the depression. The walls of the depression are covered with soft, short, silky hair. Elsewhere the hair of the foot is long and coarse, and it is noticeable that the space between the hoofs and the heel-tie itself are covered with long hair. In this character the feet of Rupicapra differ from those of other genera of Rupicaprines. Even in Oreamnos, where the greater part of the interdigital cleft is hairy, the heel- tie at least is naked *. * My figure of the foot of the newly born chamois shows the point of the heel-tie to be naked. I am, unfortunately, unable to verify the accuracy of the drawing in that respect. 132 Mr. R. I. Pocock on some In 1910 I figured and described the postcornual gland of the male example then living in the Zoological Gardens when at their maximum of development, and a figure of the head of a female sketched on the same day was added to show the absence of the swelling. But in an adult female that died on Dec. 4th, 1912, I discovered the gland to be much better developed than would be expected from looking gat the living animal, in which it is covered with the hair of the parietal region. The glandular area is superficially like that of the male, consisting of a subcircular area of skin marked with grooves. In section it is seen to be composed of thickened skin thrown from front to back into four folds, making ridges separated by valleys, the ridges gradually increasing in height from the base of the horn posteriorly. It may be remembered that I described this gland in the adult female in 1910 as consisting of a crescentic groove behind the horn on each side, this description being taken from the historic preparation in the Museum of the: Royal College of Surgeons. J have no doubt that this preparation was made from a female that died during the period of inactivity of the gland, and that the difference between this specimen and the one I examined, which died in December, is purely a question of seasonal development *. The rhinarium (fig. 2, A, B) is small. It borders the nostril above as a narrow band, and it reaches inferiorly to the edge of the upper lip as a narrow vertically grooved philtrum ; but beneath the nostrils it only extends a short distance on each side of the middle line, the rest of the lower rim of the nostril being formed by hairy skin. The extremity of the penis (fig. 2, F) is slightly depressed, and the urethral canal is prolonged beyond the extremity as a pointed process which is a little longer than that of Nemo- rhedus, but shorter than that of Budorcas described below. But in the sketch published by Gerhardt in 1906 the process is at least as long as in Budorcas. Genus Capricornis, Ogilb. Capricornis sumatraensis jamrachi, Poc. (p. 855). In 1910 I gave a brief account of the superficial appearance of the pedal and preorbital glands of an example of this * Tt appears to me’to be probable that the “ postcornual sinus” described by Hodgson as present in Procapra picticaudata resembles in structure the postcornual gland of the female Rupicapra when it is in the stage of a crescentic groove. It is detectable in thenewly born young of Rupicapra in this condition. Exter nal Characters of Ruminant Artiodactyla. 133 Weed W A. Rhinarium of Repicapra rupicapra from the front, x 2. B. The same from the side. C. The same of Nemorhedus goral from the front, x 3. D. The same from the side. E. The eye and preorbital gland of Nemerhedus yoral, the gland in sec- tion showing the thickened integument overgrown with hairs, holding secretion at their bases. F, Extremity of penis of Rupicapra rupteapra. G. The same of Nemorhedus goral. H. The same of Budorcas taxicolor. I. Section of preorbital gland of Capricornis thar. K. Section of fore foot of the same, showing the large interdigital gland with its small orifice. 134 Mr. R. I. Pocock on some race, named C. thar jamrachi, which was then living in the Society’s Gardens. The death of the animal in July 1913 enabled me to make a detailed examination of these glands. The preorbital gland (fig. 2, 1) consists of a comparatively deep, thick-walled, nearly spherical sack, the cavity of which is absolutely packed with long hairs, growing nearly verti- cally from its walls and protruding as a tuft from the small, circular, non-valvular orifice. The pedal glands (fig. 2, K), alike on the front and hind legs, open by a small circular orifice on the front of the pastern at the summit of the interdigital cleft exactly as in Ovis and Nemorhedus, and, as in these genera, the orifice leads into a well-defined cylindrical tube or duct. But, whereas in Ovis and Nemorhedus this duct gradually passes into a comparatively small saccular portion of the gland bent upon the duct at an acute angle, in Capricornis the duct communicates abruptly with an immense saccular gland which occupies the entire space, bounded laterally by the bones of the feet and above and below by the anterior and posterior integument of the pastern. Inferiorly the sack reaches into the angle formed by the fold of integument constituting the heel-tie, and above it extends almost up to a point on a level with the upper edge of the false hoofs, The cavity of the sack was sparsely hairy and filled with brownish-yellow secretion. So closely are the walls of the glandular sack applied to the integument of the pastern, that I am convinced the explanation of my failure to detect the gland in the dried skin of C. argyrochetes, mentioned on p. 855 of my previous paper, lies in the occurrence of a similar condition in that species. Hence the idea I then provisionally entertained, that possibly that species has no pedal glands, may be finally dismissed. I am unable to find any justification for Lydekker’s opinion that the various forms of Capricornis should be referred to two species, C. sumatraensis, comprising nine subspecies ranging from Kashmir to Sumatra and an un- known number from China, and C. argyrochetes from Kansu and Szechuan in China. The latter does not differ so much from some of the subspecies of C. sumatraensis as some of the latter differ from each other. In the present state of our knowledge it appears to me that the only courses open to us are to regard these forms as local races of one species, the course I adopted, or as so many distinct species—a course which I prefer to leave to him who has External Characters of Ruminant Artiodactyla. 135 the time and leisure to discover and define the characters to which specific rank may be assigned. Genus Carricornutus, Heude. €> Capricornulus crispus, Temm. (p. 855). Heude separated this species of serow from Capricornis as a distinct genus Capricornulus, which Lydekker and I adopted as a subgenus. But it appears to me that the discovery of the structure of the pedal glands in Capricornis throws a different complexion on the question. In 1910 I figured and described the pedal glands of Capricornulus crispus, and pointed out that they resemble in all respects those of Nemorhedus. Moreover, the discovery of the presence of preorbital glands in Nemo- rhedus (cf. infra) lessens the differences between that genus and Capricornis, and results in the occupation by C. crispus of a position intermediate between the two so far as cutaneous glands are concerned, the pedal glands resembling those of Nemorhedus and the preorbital glands those of Capricornis. Genus Namoruepvs, H. Smith. In 1910 my examination of material of this genus was limited to dried skins of JV. goral and N. raddeanus. Since that date I have seen a fresh adult male example of the former species, which enables me to amplify and, in one particular, to correct my previous observations. Nemorhedus goral, Hard. (p. 853). A male example from Chamba, presented by Major Rodon in 1904, which died Nov. 4th, 1915. The preorbital gland was declared to be absent in this genus by Owen, Hodgson, and Ogilby. That statement, which I accepted, proves to be untrue, strictly speaking, although the gland is so small as to account for its being overlooked on dried skins or even on fresh material. Externally the gland is marked by a very small patch of nearly naked skin covered with dry scurf-like secretion. There is no invagination of the integument, but beneath the patch of bare epidermis, the dermis is thickened and glandular (fig. 2,E). The gland, although relatively smaller, may be compared in its development to that of 136 Mr. R. I. Pocock on some Adenota kob or Hippotragus niger ; but whether it repre- sents a rudimentary or vestigial condition of the pouch-like preorbital gland of Capricornis must be left an open question. The pedal glands and the struc®re of the feet resemble in every respect those of N. raddeanus, described and figured on p- 854 of my previous paper. Inguinal glands, as noticed in 1910, are absent. The rhinarium (fig. 2, C,D) is large and naked on its upper surface almost as far back as the posterior angle of the nostril, but in the middle line above, the hair grows forwards, forming an angular point. Beneath the nostril laterally there is a comparatively wide area of smooth naked skin. In front the rhinarium extends to the edge of the upper lip as a narrow grooved strip of corrugated integu- ment which expands above to right and left beneath the inner angle of the nostrils, and the expanded portion is flanked on each side by an area of smooth naked skin. The penis (fig. 2, G) is cylindrical, slightly expanded distally, then gradually narrowed to the apex, beyond which the end of the urethral canal is prolonged as a tube for a short distance. Two points of special interest may be noticed in con- nection with these observations: namely, the similarity of the penis to that of Budorcas, described below, and the presence of the preorbital gland, which serves to link Nemo- rhedus closer with Capricornis than was previously supposed to be the case. Genus Buporcas, Hodgson. Budorcas taxicolor, Hodgson (p. 856). The death of a male example of this species from N.W. Bhotan enables me to verify and extend my account of the external characters of this genus published in 1910, and based partly on this example when alive and partly upon a dried skin of B. taxicolor tibetanus lent to me by Mr. Gerrard. The rhinarium (fig. 8, A, B) is continued inferiorly to the edge of the upper lip as a narrow mesially grooved strip, which is longer than in Nemorhedus owing to the greater depth of the upper lip. Laterally an area of naked skin, narrower than in Nemorhedus, is continued with a bold curve beneath the widely expanded nostrils, and curving round their posterior extremities passes into the dorsal External Characters of Ruminant Artiodactyla. 137 portion of the rhinarium, which is much shorter from before backwards than in Nemorhedus, being considerably more overgrown with hair. Fig. 3. A. Muzzle of Budorcas taxicolor from the front, x 3. B. The same from the side. C. Genital area of Budorcas taxicolor. p., pendulous extremity of penis ; t., long tuft of hair protruding from the prepuce; m., mammze arising from glandular elevation ; s., scrotum. The feet resemble in essential particulars those of the dried example figured in 1910 (p. 852) and described (p. 856), except that on the fore foot there is no trace of the 138 Mr. R. I. Pocock on some transverse ridge of integument just where the hair of the pastern ceases in the interungual space. There is no trace of definite pedal gland, although the hair at the bottom of the interdigital depression in front is stuck together with secretion, indicating activity of the skin at that spot. The hind foot is like the front foot. There is no trace of preorbital gland or of inguinal glands in the ordinary sense of that term; but the two mamme (fig. 3,C, m.) on each side, set as far out from the middle line as the outer edge of the scrotum, are close together, one in front of the other, in the centre of a distinct swelling like a small udder. When the skin is cut away, this swelling is seen to be caused by a blackish glandular mass like a small bunch of grapes, and blackish secretion could be squeezed through a single pore on the posterior teat with the use of considerable pressure. This unusual condition of the mammary gland in the male is worth putting on record, although, pending the examination of other specimens of Budorcas, it must be regarded, I think, as pathological in one individual. The penis (fig. 3, C, p.) is provided with a pendulous prepuce, three inches long, rising from the abdomen six inches in front of the scrotum. Just within the orifice of the pre- puce the skin is highly glandular and overgrown with long hairs, which protrude from the aperture to form a tuft three or four inches long. The glans penis (fig. 2, H) is apically attenuated and provided with a straight, moderately stout, urethral prolongation projecting some little way beyond - the tip of the glans. Except for the greater elongation of the free portion of the urethral canal, the glans penis is very like that of Nemorhedus. One of the chief interests connected with Budorcas is involved in the claim that the genus is related to Ovibos, whose uncertain position in the Bovidze was expressed by Lounberg’s ascription of it to a special subfamily Ovibovinee (Proc. Zool. Soe. 1900, pp. 142-167). Judging from the characters dealt with in this paper it does not appear to me that the claim of close relationship between the two forms can be maintained, and I am disposed to regard the resem- blances between them in horn-growth, robustness of build, etc., as independently acquired. The differences between them may be tabulated as follows. For most of the characters relating to Ovibos I am indebted to Lénnberg’s paper :— External Characters of Ruminant Artiodactyta. Budorcas, ad. 3. Rhinariumwell developed, about 14 mm. deep above the nostrils, 26 mm. wide between them, and extended beneath them as a naked strip of skin and passing inferiorly to the edge of the upper lip asa mesially grooved band (philtrum) about 7 mm. wide. Preorbital gland absent. Hoofs narrower, more pointed in front, integument between them naked. Mamme 4, the anterior and posterior on each side almost in contact, but very widely separated from those of the opposite side, the four together arranged in a transverse oblong about five times as wide as long. Prepuce distally pendulous, distal portion of its cavity not provided with longitudinal ridges, but thickly beset with coarse long hairs protruding at all seasons some 4 inches from the orifice asa long tuft. Glans penis markedly attenuated at the apex, the urethral canal pro- longed for a considerable distance beyond the tip. 139 Ovibos, ad. 3. Rhinariumgreatly reduced, about 8 mm. deep above the nostrils and only a little more between them, not extending beneath them and not continued inferiorly to the edge of the upper lip. Preobital gland present, inyagi- nated. Hoofs broad, wide in front, in- tegument between them thickly hairy except for the naked heel-tie. Mamme 4, arranged so as to form the normal four-sided figure, which is only a little wider than long, the anterior being separated from the posterior on each side by a considerable space. Prepuce distally pendulous, distal portion of its cavity provided with longitudinal folds and clothed with fine hairs only in the winter, but these do not form a long protruding tuft. Glans penis blunt at the end, the urethral canal not extending be- yond its tip. But although the differeaces above tabulated exclude the idea of relatiouship between Budorcas and Ovibos, sufficiently intimate to warrant the removal of Budorcas from the Rupicaprine, as now understood, and its association with Ovibos in a special subfamily, they by no means justify the conviction that Ovibos is not a specialised Rupicaprine. The description, for example, of the preorbital gland applies to that of Capricornis or Capricornulus, and the termination of the urethral canal in Nemorhedus is nearly intermediate in development between those of Budorcas and Ovibos ; the arrangement of the mamme is normal for the Ruminantia, as a whole, including the typical Rupicaprines ;_ the structure of the feet may be easily derived in imagination from that of Oreamnos or even of Nemorhedus, in which the gland has reached the retort-like stage, which in the Caprinz precedes its total suppression, as attested by Ovis and Capra, and the reduction of the rhinarium in Ovidos is foreshadowed 140 Mr. R.I. Pocock on some in Rupicapra, except for the total suppression of the phil- trum. In this respect Ovibos is highly specialised and unique, so far as its possible allies are concerned. On the evidence before me, I consider that if the Ovi- bovine be maintained as a special subfamily of Bovide, the Rupicaprinz, as at present understood, should be split up into three subfamilies, the Rupicaprine for Rupicapra and Oreamnos, the Neemorhedine for Nemorhedus, Capricornulus, and Capricornis, and the Budorcine for Budorcas. But if the conservative course of maintaining the Rupicaprine in its recognised comprehensive sense be followed, then Ovibos should, I think, be one of the genera of this somewhat heterogeneous assemblage. Subfamily Caprrz. Genus Ovis, Linn. Ovis musimon, Schr., and O. vignet, Blyth (pp. 859-861). Since 1910 I have examined representatives of the two species previously recorded, namely Ovis vignei and OU. musi- mon, without finding anything to add or alterations to make to my previous description of the cutaneous glands, except to remark that in the case of O. musimon the naked condition of the interungual integument noticed in one specimen is quite exceptional, and that as a very general rule that species and O. vignei are alike with respect to the hairiness of the areain question. Possibly the variation noticed is seasonal, as appears to be the case in Ammotragus lervia. The rhinarium of O. vignei is quite characteristic of the genus. It extends as a narrow bar above the nostrils almost back to their posterior termination, the internarial septum is narrow, the area beneath the septum is a little expanded, and a narrow philtrum cleaves the upper lip, but there is no naked area of skin bordering the nostrils below. The penis of O. vignei (fig. 4, D), as in O. aries, ends in a blunt gland-like enlargement, bent downwards distally. From its underside the very long filiform termination of the urethral canal arises, and passes forward on the left side of the glandular thickening. Genus Pseupotrs, Hodgson. Pseudois nayaur, Hodgs. (p. 863). Specimens examined since 1910 confirin in every respect External Characters of Ruminant Artiodactyla. 141 the constancy of the characters upon which I separated this species from Ovis—namely, the suppression of the preorbital, inguinal, and pedal glands. The rhinarium (fig. 4, F, G) resembles in a general way that of Ovis vignei, but the nostr ls are more dilatable and the “philtrun” less well defined, hardly a trace of it remaining. In one specimen the hairs of the upper lip are only separated by a very narrow parting, which is com- pletely overlapped and concealed by the hairs to the right and left of it. The naked underside of the tail (fig. 4, H) is marked on _ each side above the anus with a wide and moderately deep glandular depression, corresponding with the subcaudal gland of Capra, but smaller. The glandular portion of the end of the penis (fig. 4, B) is longer and straighter than in Ovis vignei, but the filiform termination of the urethra is approximately as long as in that species, and much longer than in the following genera. The length of this tube and the absence of strong ‘‘ Caprine”’ smell in the male are two points in which Pseudois comes nearer Ovis than Capra. In the suppression of the specialised cutaneous glands Psewdois is Caprine and not Ovine. Genus Ammotraeus, Blyth. Ammotragus lervia, Pall. (p. 862). My notes upon this species, published in 1910, were taken from the examination of a living specimen. Several dead examples that have passed through my hands since that date confirm in every respect the statement then made as to the absence of the preorbital, inguinal, and pedal glands. A peculiarity I drew attention to in 1910—namely, the smoothness of the interdigital depression in the example examined— proves to be inconstant, although the hairs of this area when developed are not so long as in Ovis and Pseudois. Possibly the variation is seasonal. or instance, in a specimen (¢) that died on Nov. 11th, the interdigital cleft was clothed with short hairs down to the heel-tie, as is normal in the Caprine series. In a second that died on March 5th, the interdigital cleft was naked. A third, which died on Feb. 10th, exhibited a condition intermediate between those of the other two. In the newly born young the space is covered with hair. The rhinarium (fig. 4, M) presents no features of special Ann. & Mag. N. Hist. Ser. 9. Vol, 11. Et 142 Mr. R. I. Pocock on some Fig. 4. iis ms Be / y A. Extremity of penis of Hemitragus jemlatcus. B, The same of Pseudots nayaur. C. The same of Ammotragus lervia. D. The same of Ovis vigner. E. The same of Capra egagrus. F. Rhinarium of Pseuwdois nayaur, showing absence of philtrum, x 3. G. The same from the side. H. Lower side of base of tail of Pseudois nayaur, showing the pair of glandular depressions above the anus. I, Rhinarium of Hemitragus jemlaicus from the side, x 3. K, The same from the front. L. The same of Capra egagrus, X 3. M. The same of Ammotragus lervia, X 3. External Characters of Ruminant Artiodactyla. 143 interest, being typically Ovine or Caprine in structure, with the narrow ‘“ philtrum ” well developed. There is a well-marked subcaudal gland above the anus as in Pseudois. The gland-like termination of the penis (fig. 4, C) is very like that of Ovis vignet in shape and curvature, but the filiform termination of the urethra is a little shorter than in that species. According to Lydekker, the males of this animal are not malodorous (Cat. Ungulates, i. p. 123). That is quite untrue. The males have a very decidedly goaty odour in the breeding season. It is also untrue that the typical race of this species is distinguished by ‘‘ an indistinct median face stripe.” A pair imported from Morocco and exhibited in the Gardens a few years ago showed no trace of such a ‘stripe. Genus Capra, Linn. (p. 864). I have nothing to add to what I said in 1910 regarding the suppression of the preorbital, pedal, and inguinal glands in various species of this genus. The rhinarium conforms in type to that of Ovis and Ammotragus, the “ philtrum” being better defined than in Pseudois. In an example of C. egagrus from Crete, I found the supranarial extension of the rhinarium (fig. 4, L) larger than in most examples of domesticated goats; but this varies to a certain degree in the latter, as also does the width of the naked area of skin beneath the nostrils laterally The subcaudal gland was a deeper pocket than those observed in Ammotragus-and Pseudvis. The penis (fig. 4, E) also is constructed very much as in those genera, and has a well-defined, but rather short, glan- dular termination, which, on the right side, as in the other genera, curls beneath the tubular filiform termination of the urethra, which is shorter than in Ovis, Ammotragus, and Pseudois. Genus Hemirracus, Hodgson. Hemitragus jemlaicus, Hodgs. (p. 866). Additional specimens confirm my previous statements with regard to the suppression of the preorbital, inguinal, and pedal glands. Hodgson’s assertion that the rhinarium (fig. 4, I, K) is larger in Hemitragus than in Capra is perfectly true. The supranarial extension is considerably deeper, and, similarly, ibs 144 External Characters of Ruminant Artiodactyla. the extension beneath the inner angles of the nostrils in front is wider. In the penis (fig. 4, A) the glandular termination is more elongate and less bulbous than in Capra and the filiform termination of the urethra is shorter. It is the shortest, indeed, that is found within the limits of the Caprine. The subcaudal gland is represented externally by a shallow depression above and at the sides of the anus. Note on the Penis of the Cephalophine and Neotragine. In my paper published in the issue of this Journal for June 1918, I regret that I overlooked at the time Lonnberg’s descriptions and figures of the penis of Cephalophus natal- ensis and of Sylvicapra gri immia (Ark. Zool. Stockholm, (5) v. no. 10, pp. 2-3, figs. 1-2, 1909). He shows that in C. natal: ensis the urethral canal has a very long filiform prolongation resembling that of Guevet mazxwelli figured by Garrod (P. Z. 8. 1877, p. 10, fig. 20), whereas in S. grimmia the tubular pro ngation is quite short, only overlapping the glans to a small extent. Now, C. natalensis is so closely related to C. dorsalis as hardly ‘to admit of a doubt as to identity in the structure of the penis in the two species. In that case the penis of C. dorsalis I described as being without the tubular urethral prolongation must have been defective, owing to mutilation. Lonnberg’s observations show that Cephalophus differs from Sylvicapra not by the suppression of the urethral prolongation, as I stated, but by its develop- ment and length, which affiliate the former genus with Guevei. In the case of the Neotraginz, it may be recalled that Garrod (op. cit. p. 11, fig. 21) described the penis of Ourebia nigricaudata as possessing a long slender urethral prolongation considerably overlapping the slender tip of the glans penis, whereas, according to Lonnberg’s observations (op. cit. p. 4, figs. 83-4), the urethra does not surpass the tip of the glans in Raphicerus campestris and Neotragus living- stonianus. The penis of the example of Nototragus mela- notis in which I found the preputial gland agrees with that of Raphicerus campestris. On Four new Species of the Genus Demodex. 145 XI.—On Four new Species of the Genus Demodex, Owen. By Sranwey Hirst. (Published by permission of the Trustees of the British Museum.) Demodex soricinus, sp. n. @. Asmall species, the cephalothorax being fairly wide. Body a little more than three times the width of the cephalo- thorax. Abdomen pointed posteriorly and somewhat longer than cephalothorax+capitulum. Capitulum much wider than long. (The spines on the capitulum cannot be seen in the unique specimen, which lies ventral side uppermost.) Total length 119 pw. Host: Sorex vulgaris. Demodex apodemi, sp. n. @. A very minute but fairly elongated species. Body about 44 times as long as the greatest width of the cephalo- thorax. Abdomen a little less than twice the combined length of cephalothorax and capitulum. Capitulum (at base) wider than the length. Spines on dorsal surface of capitulum well developed, being pointed at the end as in D. musculi etc. Total length 139 p. 3. Body from a little more than 4 up to about 5 times as long as width of cephalothorax.. Capitulum when fully extended about as long as wide. Male sexual aperture situated above interval between second and third pairs of legs: Penis fairly long and slender. Host: Apodemus sylvaticus. Demodex longior, sp. n. 9. An elongated species of comparatively large size, resembling J. canis in many respects. Body sometimes nearly nine times as long as the width of the cephalothorax. Abdomen about 22 times the combined length of cephalo- thorax and capitulum. Capitulum wider than long; the spines on its dorsal surface are short and somewhat curved, Total length 280 p. 6. Abdomen about twice as long as the cephalothorax+ capitulum. Body more than 6 times as long as the cephalo- 146 Mr. O. Thomas on thoracic width. Male sexual orifice situated above the interval between the legs of the first and second pairs. Note.—In one male specimen the tracheal tubes leading from the capitulum are quite distinct; each is at first double, but afterwards fuses to form asingle wide lateral main trunk. Host: Apodemus sylvatieus. Demodex nanus, sp. n. ?. A minute species very like that present in Sorex vul- garis casteaneus. Length varying from less than 3 up to slightly more than 3} times the width of the cephalothorax. Abdomen considerably shorter than combined length of cephalothoraxand capitulum. Capitulum usually much wider than long ; the spines on its surface apparently obsolete or absent. Total length 87-102 yw. Host: the black rat (Rattus rattus), a number of specimens collected by the author from a freshly killed rat. Note.—Hahn has already described a species of Demodex (D. ratt’) from a house-rat said to be Mus rattus. I have not been able to consult his original description, which is referred to by Gmeiner. ‘The latter says the species is like that of the dog. From this one would infer that it was an elongated form of comparatively considerable size, similar to that found in Rattus norvegicus. It is probable, indeed, that the rat from which Hahn’s specimens were taken was really Rattus norvegicus, the brown or Norwegian rat (syn. dfus decumanus). It is, of course, possible that two species occur in /tattus rattus, as is certainly the case in Apodemus sylvaticus. XIT.—WNew Species of Gerbiilus and Taterillus. By OLDFIELD THOMAS. (Published by permission of the Trustees of the British Museum.) Gerbillus allenbyi, sp. n. A small species, with short feet and tails probably allied to G. agag. General colour much more mouse-grey than the usual tone of gerbils, markedly greyer than G@. gerbillus; head, shoulders, and most of the upper surface near “ cinnamon-buff,” but new Species of Gerbillus and Taterillus. 147 the middle dorsal area greyer, though this difference may be less marked in older specimens. Under surface less absolutely pure white than usual, the hairs, especially in the inguinal region, with a slight tinge of buffy. Postorbital light patches present, but not very sharply defined ; below them on each side, between eye and ear, there is a distinct patch of grey hairs. Ears with proectote buffy, the rest whitish; post- auricular white patch sharply defined. Hands and feet white, but a slight tendency to buffy appears on the wrists ; soles all hairy except for a small round patch on the heel. Tail not proportionally long; dull buffy, little lighter below; its terminal dark crest inconspicuous. Skull of the general build of that of G. gerbidlus, but the bulles smaller. Supraorbital beads little developed. Dimensions of the type (measured in flesh) :— Head and body 70 mm. ; tail 95; hind foot 24; ear 9. Skull: greatest length 26-2 ; condylo-incisive length 23 ; zygomatic breadth 14°5 ; nasals 9°6 ; interorbital breadth 5:2 ; breadth of brain-case 13°3 ; zygomatic plate 3°9; palatal foramina, anterior 4°4, posterior 2°2; greatest horizontal diagonal diameter of bulla 9°2; breadth of bulla at right angles to last, exclusive of meatus, 5:7 ; upper molar series 4. Hab. Coast region of Palestine. Type from Rehobot, near Jaffa. Type. Young adult male. B.M. no. 14. 5.29.5. Original number 8. Collected 3rd February, 1914, by TT’. Aharoni. Presented by the Hon. N. Charles Rothschild. This is evidently the species which Nehring * assigned to G. longicaudus, Wagn. But Wagner’s animal, which I have seen in Munich, was from Egypt, and was clearly referable to G. gerbillus, as has been shown by Anderson and de Winton. The Palestine gerbil seems to be related to G. agag, Thes., but is readily distinguishable by its less bright colour, greyer back, and the greyish patches between eye and ear. I have named it in honour of the general to whose forces the country where it occurs owes release from the barbarian domination under which it has suffered for so many centuries. Gerbillus acticola, sp. n. Near G. pygargus, but the bullee larger. Size and colour as in G. pygargus, of the same light desert-colour—quite unlike that of G. dunni of Central * SB. Ges. Fr. Berl. 1901, p, 173, 148 Mr. O. Thomas on Somaliland. Compared with a series from Shendy, the ground-colour is warmer, being near ‘‘ warm buff” in py- gargus, while it is “pinkish cinnamon” in aetécola; but the variation in the colour of these desert-animals is so great that not much stress can be laid upon it. Sides lighter, line of demarcation high up. Postorbital and postauicular white patches well marked. Fore limbs wholly, hind limbs mostly white. Hind soles with a nearly naked stripe running along the inner side almost to the base of the hallux. Tail buffy: above, white below; the terminal crest inconspicuous, brown. Skull of the same stoutly built elongated form as in pygargus, the supraorbital beads similarly strongly developed. Bulle of similar shape, but decidedly larger than in any of the considerable series available of pygargus and pyramidum. Dimensions of the type (measured in the flesh) :— - Head and body 118 mm.;. tail 144; hind foot 29; ear 15. — Skull: greatest median length 32°5; greatest diagonal length 32 ; condylo-incisive length 28'5 ; zygomatic breadth 17:4; nasals 12:7; interorbital breadth 6°6; breadth of brain-case 14°5; breadth between meatal edges 16°3 ; zygo- matic plate 4°7; palatal foramina, anterior 5°4, posterior 3 ; bull, horizontal diagonal length 12 ; breadth at right angles to last, excluding meatus, 7; greatest diameter in any direction 12°7; upper molar series 41. Hab. Coast region of N. Somali. Type from Berbera, other specimens from Bulhar. Type. Adult female. B.M. no. 7. 11.5.4. Original number 32. Collected 30th July, 1905, and presented by br. R. E. Drake Brockman. Nine specimens. This Somali representative of G. pygargus, distinguished by its larger bulle, is the specics mentioned on p. 119 of Dr. Drake Brockmai’s ‘ Mammals of Somaliland’ (1910) as the Coast Gerbil, a title I have Latinized as above. Gerbillus vallinus, sp. n. A Gerbillus with an unusual amount of the soles naked and with very large bulle. Size about as in G. peba. Fur long and loose... General colour strong sandy buffy, near “ cinnamon-buff,” not so inclined to russet as in G. pela. Line of demarcation on sides not very sharply defined. Lighter postorbital and post- auricular markings scarcely perceptible. Hars short, their new Species of Gerbillus and Taterillus. 149 proectote buffy like the general colour. Fore limbs wholly in the white area, without any darker colour on their front surface. Soles less haired than in other members of Gerbillus, the naked area extending from the heel along the middle of the sole to the level of the base of the hallux, but the region of the pads is closely and profusely hairy, as usual in the genus. ‘J'ail at base pale buffy above, whitish below— its terminal portion lost in the type. ~ Skull remarkable for the great size of the bullae, which tend to recall those of Desmodillus and far exceed those of any other member of this genus. ‘he posterior breadth of the skull is therefore unusually great. Muzzle slender. Supra- orbital beads present. Zygomatic plate more projected forward than in most species of Gerbéllus, and almost approaching the projection characteristic of Yaterdllus ; the same is the case In G. peba. Palatal foramina, both anterior and posterior, large and well open. Bulle greatly swollen, the anterior edge oF the meatus also inflated; a well- marked vacuity just beneath the opening of the meatus. Dimensions of the type (measured in the flesh) :— Head and body y2 mm.; tail (60+); hind foot 30; ear 15. . Skull: greatest median length 29; greatest diagonal length 30; coudylo-incisive length 27 ; zygomatic breadth 16; nasals 11:2; interorbital breadth 6; breadth of brain- ease 14°3; breadth between outer edges of meatal inflations 16°8; zyg comatic plate 4°8; palatal foramina, anterior 5:2, posterior 2°55; greatest horizontal diagonal diameter of bulla 10°7 ; greatest diameter in any direction 12°2 ; upper molar series 4°2. Hab. Bushman-land. Type from Tuin, near Kenhart, Hartebeest River, near 29° S., 21° E. Type. Adult male. B.M. no. 12. 1. 11. 2. Presented alive by Maj. H. A. P. Littledale to the Zoological Society, by whom it was transferred on death to the National Collection, This well-marked species is readily distinguishable by its greatly enlarged bulle, which tend to approach in size those ei Deemodillus auricularis, obtained in the same region by Major Littledale. The hind feet of this animal are also more naked than in other members of Gerbillus, but have, however, the characteristic distal cushion which distinguishes the genus from Déepodillus. 150 On new Species of Gerbillus and Taterillus. Taterillus gyas, sp. n. A Taterillus with decidediy larger skull than any other, Size rather, but not conspicuously, larger than in 7’, emind. General colour above strong and dark, near “ cinnamon,” or even approaching “tawny”; sides cinnamon-buff. ars rather large. Hands and feet white; soles quite without any trace of the usual transverse band of fur. Tail long, its basal half brownish above, dull buffy below; terminal tuft well developed. Skull conspicuously larger and more heavily built than in any known Taterdllus. Interorbital region rather more parallel-sided than usual, the supraorbital ridges strongly developed. Posterior palatal foramina extending from the level of the front root of m! to the middle of m?. Bulle of average proportional size. Dimensions of type (measured in flesh) :— Head and body 127 mm. tail (damaged in type, 175 mm. in another specimen of about the same size) ; hind foot 34; ear 21. Skull: greatest length 39; condylo-incisive length 35; zygomatic breadth 19°5; nasals 15°6; interorbital breadth 7:3; breadth of brain-case 15°83; zygomatic plate 7:3; palatal foramina, anterior 7°2, posterior 4°6; horizontal diagonal diameter of bulla 10°2 ; upper molar series 5:5. Hab. Kamisa, Dinder R., Sudan. Type. Adult female. B.M. no. 14. 3.8. 24. Original number 55. Collected 26th December, 1913, by Willoughby P. Lowe, and presented by Abel Chapman. Two adult and six young specimens examined, . This Yaterillus is remarkable for its large size and the complete absence of the hairy band across the soles. It thus considerably resembles the members of the genus Taterona. But its elongate posterior palatine foramina show that its place really is in this genus, all the more that 7. gracilis proves to be variable in the development of the same hairy band. In that species the band is commonly absent, fairly often slightly or partially developed, and occasionally fully developed, all extremes occurring in any one locality. This species ranges eastwards from the Gambia to Upper Nigeria, where it occurs side by side with 7. nigeri@ on the Bauchi Plateau. ‘The latter was first described from a single specimen, but about a score of gerbils have been more recently sent by Mr. Fox, and were all supposed to be of the same species as the first. I now find, however, that they On a new Duiker from Zanzibar. TF are mostly referable to 7. gracilis, only four belonging to T. nigerie, which may be distinguished by its larger size, longer anterior palatine foramina, and uniformly longer feet, and these in all four examples have well-developed sole-bands. XUIL—A new Duiker from Zanzibar. By OLDFIELD ‘l'HOMAS. (Published by permission of the Trustees of the British Museum.) Tue British Museum has received from Dr. W. M. Aders, Government Biologist at Zanzibar, native skins of three local Ungulates, two antelopes and a Potamochwrus. One of the former is that of a Nesotragus moschatus, but the other repre- sents a duiker quite distinct from any species hitherto described. In honour of its donor, to whom the Museum is indebted for many Zanzibar mammals, it may be called Cephalophus adersi, sp. n. Allied to C. weynsi*, but with whitish bands across thighs and a white tufted tail. Size and general characters about as‘in C. weynsi of the Congo. Line along nape with reversed fur, as in that species. General colour of withers and nape dark brown (near mtmmy-brown), which gradually becomes more rufous (darker than “‘avellaneous”’) on the shoulders and flanks, and posteriorly on the rump passes into deep rich chestnut-rufous (“‘ mahogany-red”” where richest). Under surface whitish, not sharply defined laterally, the hairs pale drabby at base, whiter terminally ; a mesial rufous patch on the chest. Fore limbs with the avellaneous rufous of the shoulders passing down without interruption, but on the hind-quarters there is a broad whitish band running across the outer side of the hips and separating the chestnut-red of the rump from the rather paler red of the legs ; this band is more or less rufous white where it commences on the sides above the inguinal glands, but becomes nearly pure white posteriorly, where it * Figured and described, Ann. Mus. Congo, ii. p. 15, pl. vi. (1901), 152 Dr. G. A. K. Marshall on Alcides, Schénh. contrasts prominently with the mahogany-red rump. Fore and hind feet deep rufous speckled with white, but how far these white specklings may be an individual abnormality I have no means of judging. Tail, without tuft, about 2 inches in length, the tuft well marked, its hairs rather more than an inch long, wholly white, though there is a narrow rufous line running along the top of the tail basally. Middle of neck to rump about 24 inches. Hab. Zanzibar. Type. Native skin. B.M. no. 18.5. 25.1. Presented and collected by Dr. W. M. Aders. By its reversed nape-hairs and general type of coloration, with brown fore back and rufous rump, this striking duiker shows relationship to C. weynsi, but it is at once distinguished by the whitish bands which run across the thighs and show up the brilliant rufous of the rump, and by the wholly white tail-tuft, that of C. weynsi being prominently blackish above. These characters are so marked that, although the specimen is a native skin, without head or hoofs, I feel justified in describing it, but hope Dr. Aders may soon be able to obtain a complete example of so striking an animal, on whose discovery he is to be congratulated. XIV.—Notes on Alcides, Schinh. (Curculionide, Coleoptera). By Guy A. K. Marsuatt, D.Sc. ConsIDERABLE Confusion exists in collections with reference to the strikingly marked species of Alcides related to A. delta, Pasc. Pascoe’s original description (Journ. Linn. Soc. Lond., Zool. x. 1870, p. 460) was based on three speci- mens, from Ceylon, Ceram, and Amboyna respectively ; of these he selected the Ceylon specimen as his type, and the other two examples now prove to belong to a quite distinct species. Subsequently he gave a figure of A. delta (ibid. xi. 1871, pl. ix. fig. 10), but instead of illustrating his type he unfortunately selected a so-called ‘“ variety,” which turns out to be yet a third species, and was later described by Kirsch under the name of triangulifer (Mitt. Mus. Dresd. i. 1875, p. 40). Probably misled by Pascoe’s figure, Aurivil- lius in 1891 (Nouv. Arch. Mus. Paris, (8) il. p. 218) sunk triangulifer as a synonym of delta, and thus it stands in Bovie’s ‘ Catalogue of Alcidine ? (Wytsman, fasc. 71). Dr. G. A. K. Marshall on Alecides, Schdnh. 153 As a matter of fact, A. delta differs from both the other forms mentioned above in a very striking character ; nor- mally in Alcides each tarsal claw is deeply cleft, but in A. delta the claws are simple and soldered together at the base. This character is also found in yet another new and allied species, likewise from Ceylon. I have not so far observed it elsewhere in the genus, though Lacordaire men- tions its occurrence without citing any species. A. triangu- lifer presents a somewhat intermediate condition, the inner division of the claw being very much reduced. ‘The following table will serve to discriminate the mem- bers of this group, all of which possess a similarly-shaped large patch of silvery-white scales on the side of the meso- and metasternum, and this also occurs in the very differently marked A. kirschi, Pase., from Labuan :— 1 ( 4). Tarsal claws simple, connate at the base ; sides of prothorax not constricted in front; genze of ¢ not produced down- wards at the apex. 2 ( 3). Transverse impression at base of elytra shallow, base of prothorax not lower than the apex; apical edge of rostrum produced into a short point in the Md dle i520 0 30 Oy eth Cae ok eee delta, Pasc. 3 ( 2). Transverse impression on elytra very deep, base of prothorax distinctly lower than the apex; apical edge of rostrum shallowly emarginate in the middle... ephippiatus, sp. n. 4 (1). Tarsal claws cleft; sides of prothorax markedly constricted in front; genz of ¢ with a tusk-like downward pro- cess at the apex. § (14). Elytra with a large common triangle formed of broad pale stripes enclosing a black triangle; front tibiz with an internal tooth placed nearly in the middle. 6( 9). Peduncle of submentum narrowly ob- long (2X1) and shallowly constricted at the extreme base only; anterior pale stripe on prothorax running trans- versely upwards along the edge of the granulate area, not covering the post- ocular lobe. 7( 8). Apical margin of rostrum rounded; scutellum pointed at apex; post-hume- ral stripe on elytra not uniting behind with posterior angle of the pale tri- amigle, .cs.is. STS cls > ood 3/4 sta ceramodelta, sp. n. 8( 7). Apical margin of rostrum with a short sharp point in the middle; scutellum 154 Dr. G. A. K. Marshall on Alcides, Schénh. rounded at apex ; post-humeral stripe on elytra uniting broadly with the posterior angle of the triangle ...... muir, sp. D. 9 ( 6). Peduncle of submentum subtriangular, broad at apex and very strongly nar- rowed behind; prothorax with an ill- defined pale stripe covering the whole postocular lobe and running obliquely backwards on to the disk. 10 (11). Shoulders of elytra produced outwardly into a sharp angle; sete at apex of dibizo Wlaekislt 1.54% Pasiciaicn oisisie ofc Ks stamodelta, sp. 0. 11 (10). Shoulders of elytra obtuse; setze at apex of tibiz reddish. 12 (13). Elytra broader, broadest at the shoulders and narrowing gradually behind; edeagus of g with the median lobe narrowed to a point at the apex ; pro- thorax with an oblique blackish stripe running from the eye almost to the DESO). oss kininase unis ae eaters triangulifer, Kirsch. 13 (12). Elytra narrower, almost parallel-sided from the shoulders to beyond the middle; edeagus of J with the me- dian lobe dilated at the apex, its apical margin very broad and sinuate ; prothorax with the black mark behind the eye contined to the non-granulate HDC! ATER A wa aaielea s Se sae Syma Javanodelta, sp. n. 14 ( 5), Elytra without any distinct triangular markings, the oblique discal pale stripes diverging from the middle to the shoulders instead of converging towards the scutellum; the internal tooth on the front tibise much nearer to the base than to the apex ......,. magicus, Pase. Alcides delia, Pasc. So far as is known at present the true A. delta is confined to the lowlands of Ceylon. Alcides ephippiatus, sp. un. 3d ¢. This species has the same general facies and pat- tern as A. delta, as well as the simple and connate claws, but differs as follows :—The pale markings are usually covered with a dark pink or pinkish-brown powdering, and the stripes on the elytra are generally narrower, so that the enclosed black triangle is larger; the infra-humeral stripe is reduced to one-half the length or less; in the V-shaped apical patch the outer arm (on interval 7) is only half as long as the Dr. G. A. K. Marshall on Alcides, Schénh. 155 inner (on interval 3), whereas in delta they are equal or nearly so. The rostrum is proportionately much shorter, and the apical edge is shallowly emarginate in the middle. The dorsal outline of the prothorax is much more convex, so that the basal margin is well below the plane of the apical. The elytra are proportionately shorter, the basal transverse impression being much deeper, so that the dorsal outline is strongly convex; intervals 3 and 4 are not so markedly costate at the base, and the scales that form the pale markings are much smaller, most of them being very deeply fringed at the apex. Length 10-134 mm., breadth 43-54 mm. Cryton: Dikoya, 4000 ft. (type), and Bogawantalawa, 5000 ft. (G@. Lewis) ; Kandy (E. EL. Green). The deeply sinuous dorsal outline of this species renders it easily recognizable. It appears to be the mountain representative of A. delta in Ceylon. Alcides siamodelta, sp. n. 2. Closely resembling Pascoe’s figure of A. triangulifer (1. c.), except that the shoulders of the elytra are produced outwardly into a sharp angle. Other distinctions are :—In triangulifer the 7th joint of the funicle is elongate and equal to or longer than the club (4: 3-4), in the transverse pale band forming the base of the triangle on the elytra the intervals are distinctly granulate, the apical setz on the tibize are reddish, and the tarsal claws have the inner divi- sion unusually short and slender; in siamodelta the 7th joint of the funicle is transverse and distinctly shorter than the club (23:4), the intervals are not granulate in the transverse band of the elytra, the apical sete on the tibic are blackish, and the tarsal claws are normal, the inner division being about three-fourths the length of the outer. Length 94-103, breadth (at shoulders) 54-6 mm. Frencu Inpo-Cuina: Laos (type) ; Sram. Alcides triangulifer, Kirsch. So far as I know at present this insect is confined to the Malay Peninsula, Burma, and the Nicobars. Insects recorded from Borneo under the name of A. delta will probably be found to belong to a distinct species. Alcides javanodelta, sp. n. 3 2. Apart from its narrower form and shorter rostrum, 156 Dr. G. A. K. Marshall on Alcides, Schinh. extremely similar to A. triangulifer. In addition to the characters given in the key, the following distinctions have been noted:—The mentum is quite flat (an ¢riangulifer it bears a shallow longitudinal impression); the proportions of the 7th funicular joint to the club are 2-22: 34-4 (in triangulifer 4: 3-4), and the intermediate tibiz are simply angulate in the middle internally (in ¢rzangulifer there is a sharp tooth). But its most striking character is the broad dilatation at the apex of the median lobe of the edeagus, for in all other species of the group this organ is pointed at its tip, as is usual in the genus. Length 83-12} mm., breadth 33-53 mm. Java. All the specimens of this group that I have seen from Java belong to this species. There isin the British Museum a single specimen labelled Singapore (Coll. Atkinson), but it seems possible that the locality may be erroneous. Alcides ceramodelta, sp. n. 3 ?. While this species agrees with triangulifer, as com- pared with deléa, in the structural characters mentioned in the key, it differs from it in the pattern of the prothorax, which quite resembles that of delta and ephippiatus, the general colour being blackish brown, with the usual oblique lateral pale stripe above the cox, a transverse subapical pale band running along the anterior edge of the granulate area, and a pale central stripe. The general form is broader in proportion to its length than in any of the other species. ‘The rostrum is propor- tionately short and stout, and its apical margin is rounded, with traces of very feeble undulations ; the peduncle of the submentum differs from that of all other members of the group (except A. muiri) in its more narrowly oblong form. In the autenne the 7th joint of the funicle is shorter than the club (8:4)*. The prothorax is very similar in shape to that of triangulifer, but the granules are slightly smaller and there is no trace of the shallow median stria. The scutellum is bluntly pointed at its apex, whereas in all the other species it is broadly rounded. ‘The intervals on the elytra are more distinctly granulate than in delta and rather less carinate than in ¢riangulifer, thus giving the * By actual measurement; owing to the club being pointed, it appears relatively shorter than it really is. Dr. G. A. K. Marshall on Alcides, Schénh. 157 elytra a somewhat smoother appearance. The legs are markedly shorter than in ¢riangulifer, but the tarsal claws are similar, the inner division being much reduced; the median tooth on the middle tibiz is almost as long as that on the front pair. Length 124-13, breadth 64-62 mm. Ceram (type); AmBorna (A. R. Wallace). Alcides muiri, sp. 0. ¢. Pattern similar to that of A. delta and A. ceramodelta, except that the post-humeral stripe on the elytra unites broadly with the posterior angle of the pale triangle on each side; the edges of the pale markings rather ill-defined. Very similar in structure to hs +") jie \ A. Muzzle and rhinarium of Connochetes grow from the front. x 4. (The vibrissee shorter than in nature.) B. The same from the side. \ C. The same with the upper lid of the nostril raised to show the orifice of the sack penetrating the septum. D. Extremity of penis of the same from below. E. The same of Gorgon taurinus from the left side. like the entrance to the chamber itself, is revealed when the upper lid of the nostril is raised and concealed when it is in its normal depressed position (fig. 2, B, C). Owing to the scanty clothing of hair on the dorsal side of the.muzzle, the rhinarium is not so well defined above and behind as in Damaliscus; it extends less than halfway 218 Mr. R. I. Pocock on some round the upper lid of the nostril. Viewed from the front it is exceedingly wide and laterally attenuated, with a concavo-convex, sinuous upper edge. The philtrum, which is broad, angular, and ungrooved, is inferiorly abbreviated, ending in a point a little above the middle of the upper lip, the lower portion of which is continuously hairy across the middle line. The surface of the rhinarium is transversely striated, not roughened and tessellated. In his paper on the anatomy of the Gnu, Lonnberg (K. Vet.-Akad. Handl. xxxv. no. 3, p. 48, 1901) paid no special attention to the rhinarium, contenting himself with a reference to the descriptions published by others, notably by Sclater and Thomas in the ‘ Book of Antelopes,’ vol. i. This brief description, however, contains no mention of the pouches in the internarial septum, because they are com- pletely concealed in dried skins, No doubt this fact accounts for their having hitherto apparently escaped detection. At all events I have not come across any record of their occurrence. I am unable to suggest any explanation of the function” of these pouches, unless they act as traps for the maggots of parasitic dipterous insects (Cstrus) whose usual habit it is to pass up the true nostril into the narial passages, where they frequently set up serious disorders in Ruminants. At all events, these parasites would be innocuous in the pouches. The penis (fig. 2, D) differs from that of Damaliscus py- gargus in being apically attenuated, without trace of the coidate thickening at the end, and in the termination of the urethral canal in a short process on the left side of the apex, beyond which it projects for a very short distance. Genus Gorgon, Gray. Gorgon taurinus, Burch. (p. 906). An example of G. taurinus ‘albojubatus, four and a half months old, had the muzzle constructed as in Connochetes gnou, except that the peculiarities were less exaggerated ; it was less depressed and narrower and the rhinarium seen from the front was deeper from above downwards and the shortened philtrum showed a narrow groove. The preorbital gland was scantily clothed with long hair and its surface was mesially depressed and saucer-like. The penis (fig. 2, E) of an adult male of the typical race was less attenuated apically than in that of Connochetes gnou and the urethral canal was not prolonged beyond, the end of the glans. From evidence supplied mainly by the digestive tract, External Characters of Ruminant Artiodactyla, 219 Lénnberg (K. Vet.-Akad. Handl. xxxv. no. 3 (1901) ; Arkiv. Zool. v. no. 10, p. 21 (1909)) was of opinion that the Guus are phylogenetically related to the Bovine, the latter being the descendants of antelopes closely akin to Conno- chetes and Gorgon. It appears to me, however, to be certain that the Gnus must be regarded as highly specialised forms of Bubalis; but I cannot admit that the latter are in any way nearly affiliated to any form of Bubaline. ‘he evidence, on the other hand, that the Bovine are specialised Tragela- phinee is, in my opinion, complete. The usually recorded differences between the Gnus and Hartebeests in cranial and cornual characters are well known. Using the muzzle as a basis the two groups may be distin- guished as follows :— a. Muzzle comparatively narrow; rhinarium cleaving the upper lip approximately to its inferior edge, its depth about half its width, its surface roughened and reticulated ; no pouches in the internarial septum within the MING Sse etyae Scte cles ties aXe ea tes a ess Bubalis, Damaliscus. a’, Muzzle comparatively very broad; rhina- rium not extending to inferior edge of upper lip, its depth less than half its width, its surface transversely striated; a pair of pouches penetrating the internarial septum WANE NAT GNC TIOSETI Sy wersigixtseg ye el siete sPela’ ave, « Connochetes, Gorgon The Bubaline constitute a compact group of Bovidee showing comparatively slight range of variation so far as the external features dealt with in this paper are concerned. The muzzle is expanded, the rhinarium is reduced, the nostrils are valvular and lined within the orifice with longish hair for the exclusion of foreign bodies. The preorbital gland is large and is either provided with a narrow duct-like in- vagination (Damaliscus, Bubalis) or has a flat, slightly convex or slightly concave surface (Connochetes, Gorgon). Inguinal glands are absent, and there is normally, at all events, a single pair of mamme. Pedal glands are well developed only on the fore feet, where they consist of a long deep interdigital pouch with a long orifice, but not so long as in the Antilopine, on the front of the pastern. In the hind feet this gland is aborted and represented merely by a shallow depression. The penis at most has a short tubular urethral prolongation. Subfamily Orverz. Genus Oryx, Blainville. My account of the cutaneous glands of this genus pub- lished in 1910 was based upon an examination of dried skins 220 Mr. R. I. Pocock on some and living animals only. Since that date I have had the opportunity of seeing fresh carcases of two very distinct species, namely O. gazella, the type of the genus, and O. leucoryx, which should rank, I think, as a distinct genus Oryx gazella, Linn. The muzzle (fig. 3, A, B, and 4, F) is broad and depressed, y) y , a) Le ch LANNY LEIS ( LAUGH Z - | i » vy II i A. Muzzle of rhinarium of Oryx gazella from the front. x B. The same from above. ae with the nostrils narrow, elongated, valvular, and hairy right up to their lower rim. The smooth rhinarium is reduced in size, moderately broad between the nostrils, and extending laterally as a comparatively narrow strip all along the upper rim of the nostrils. From the dorsal aspect it is crescentic, the hairs of the dorsal side of the face extending External Characters of Ruminant Artiodactyla. 221 far forwards between the nostrils, more than halfway along their length, forming a field with an evenly convex antero- lateral border. From the front the upper edge has a sinuous curvature, and the depth of the rhinarium down the middle line is about equal to the width of the internarial septum ; the inferior edge is slightly angled, but is not continued as a philtrum down the upper lip, which is continuously hairy across the middle line *. Preorbital and inguinal glands are absent, as Owen and Ogilby correctly recorded. The pedal glands ou all four feet consist of dilated hair- lined pouches, opening by a narrow passage and a small orifice on the front of the pastern just above the summit of the folded interungual web. They resemble those of O. beisa described in 1910, except that the orifice is small and sub- circular (cf. infra). Oryx beisa, Rupp. (p. 907). I am indebted to the late Mr. F. C. Selous for the fore and hind foot of an adult example of this species from British East Africa. In these the glands were moderately large and saccular, with a narrow cylindrical exit passage and circular orifice. In 1910 I described the orifice of the gland observed on the dried feet of an immature specimen as consisting of an elongated slit. The shape assumed by the orifice in this case was probably due to shrinkage of the skin when drying. At all events, the glands of the specimen brought for me by Mr. Selous resembled those of the fresh specimen of O. gazellia described above. Genus A’coryx, nov. Differs from Oryx in possessing a preorbital gland, a more reduced rhinarium, and curved horns. Type, 4égoryx algazel, Oken. Aigoryx algazel, Oken (p. 909). In 1910 my notes on this species were restricted to the statement that an example living in the Gardens showed the presence of preorbital glands by patches of secretion on the face about one inch in front of the eye, thus disproving the assertions of Owen and Ogilby that the preorbital gland is absent. * In the figures illustrating the muzzle of the antelopes described in this paper, no attempt has been made to indicate by shading the trans- ser and vertical convexity of the rhinaria, which thus appear to be too flat. 229 Mr. R. I. Pocock on some In an example of the typical race of this species from Northern Nigeria, the gland (fig. 4, B) consists of a » thickened area of skin concealed and overgrown by hair basally adherent with secretion. The glandis about 30 mm. long and 6 mm. thick and slightly elevated, resembling the Fig. 4. 7 wh?” rN TH Sy s ene Ben Mi ca Pa) wey vy eile ste win 1 j co NYA aa wl Ut t Wy \ Wy ng my a vs he ue HAS mS Of I aN Fr RE PTs oe ACN ae i ‘ a NV} uy ' ; ae Wer 4 Pugupra Monee : Mone aT} Cty Awtient A My AY AES. u #7 Bir — LM lg h 1; 4 1 if! Wa | yf \e fowl ee 3 ie (oe ech \ A. Muzzle and rhinarium of 4%goryx algazel from the front. x 3, B. Preorbital gland of the same in longitudinal section. C. Extremity of the penis of the same from the left side. D. The same from the right side with urethral process pulled down. E. Extremity of penis of Hippotragus niger with urethral process straightened. F. Muzzle and rhinarium of Oryx gazella from the side. corresponding gland of Hippotragus, although shorter as compared with its thickness. The muzzle (fig. 4, A) in its general features is like that of Oryx gazella, but the rhinarium is considerably more reduced. When viewed from the front it is much wider as External Characters of Ruminant Artiodactyla. 223 compared with its height, the height in the middle line being slightly less than that of the upper lip and much less than the width of the internarial septum. As in other members of this subfamily the inguinal glands are absent, there are two pairs of mamme, and the pedal glands are present and constructed as in Oryz. The penis (fig. 4,C, D) is remarkable for the thickness and length of the tubular prolongation of the urethral canal, which projects some distance beyond the ovate termination of the glans and is nowhere adherent to it. It rises from the underside of the cylindrical portion of the penis, and, although normally closely applied to the left side of the well- defined terminal portion, is in reality separable from it. It is thick at the base and gradually attenuated apically. The penis closely resembles that of Addazx as described and figured by Garrod (P.Z.S.1877, p. 10, fig. 18) ; but Gerhardt’s figure of the penis in Addag has a much shorter urethral prolongation, not overlapping the tip of the glans (Verh. Deutsch. Zool. Ges. xvi. p. 153, 1906). Genus Hrppotracus, Sund. Hippotragus niger, Ham. (p. 909). Specimens examined since 1910 confirm in every parti- cular the facts stated in that year as to the structure and incidence of the cutaneous glands. The new facts here recorded relate to the rhinarium and the penis. The muzzle (fig. 5, A, B) is less depressed than that of Oryz, and the sculptured rhinarium is relatively larger and more normal, with a better defined naked tract below the nostrils, a well-developed philtrum mesially grooved, broad at the base, and narrowed below where it reaches the inferior edge of the upper lip; the nostrils are more expanded with the upper lid more swollen, so that from the anterior aspect the upper edge of the rhinarium appears to be biconvex with an angular median depression and from the dorsal aspect the anterior edge is seen to be transversely truncated. The hairs of the upper side of the nose extend some distance between the nostrils, although not so far as in Oryx, so that the posterior border of the rhinarium is strongly concave. The penis (fig. 4, E) resembles that of Agoryx, described above, in the thickness, length, and freedom of the tubular prolongation of the urethral canal, but the termination of the glans is less markedly bulbous. The Orygine, apart from Addax which requires examina- tion, are a remarkably uniform group with respect to the 224 Lewternal Characters of Ruminant Artiodactyla. structure of the penis, the presence of four mamme, the absence of inguinal glands, and the presence on all four feet of flask-shaped pedal glands with narrow exit passage and small circular orifice just behind the summit of the interungual web. pe es =e id = Sze AS =~ Nt i) . a} bd ’ yy ULE REY Y, ia = oe Gt PT Aus “fy Feng Sal A. Muzzle and rhinarium of Hippotragus niger from the front, X 3. B. The same from above. Setting Addaz aside, the genera discussed in this paper may be distinguished as follows :— a. Rhinarium sculptured, with distinct philtrum reach- ing lower edge of upper lip; upper rim of dilated nostrils swollen; horns rising erect from head .. Hippotragus. a’, Rhinarium smooth, without philtrum, upper lip entire ; upper rim of narrow nostrils not swollen ; horns inclined backwards in the plane of the forehead. b. Preorbital gland present as a thickened pad of skin like that of Hippotragus; rhinarium shallow; hors eurved» .).. ./. +. 60h Se emer os ugoryx. '. Preorbital gland absent; rhinarium deeper; horns Straight 5s... occ s ace eee eee” OFTE, On new Species of Syntomidee, Nymphalide, &e. 225 Addax differs from the three above enumerated genera in having broad rounded hoofs, the interdigital web exceedingly thick above, the pedal glands represented by a short narrow cylindrical tube, corresponding to the duct of the gland in Hippotragus and Oryx, and the horas spirally twisted (Proc. Zool. Soc. 1910, pp. 910-911). XXV.— Descriptions from the Joicey Collection of new Species of Syntomide, Nymphalide, and Hesperidee, and Two Genera of Syntomide. By W. J. Karyn, F.E.S. ALL the new species herein described will be figured after the war. The striking new Chlorippe from Haiti is so far unique. It is a ¢@ and could scarcely be a 2? ab. of cherubina, the species it doubtless comes closest to. In Cuba Chlorippe laure occurs, but the present insect is certainly not a @ of that species, although it is highly possible that laure occurs in Haiti. The new race of Anea xenocrates from French Guiana, although different in the ¢ from the typical species, has a 9 (only a single specimen) that is exceedingly like the ? at Tring of the type-form from Bolivia. The female of this species appears to be exceedingly rare, and it was rather surprising to get a single pair from quite a new locality. Syntomide. TIGRIDANIA, gen. nov. Proboscis well developed. Palpi long, upturned, reaching well above head, and separated widely at base, but meeting above the head. Antenne bipectinate in both sexes, longer in g. Legs fairly long. Hind tibize with two pairs of spurs of nearly equal length and strong spines on the tarsal joints. Fore wing with vein 3 a long way before end of cell and distance between veins 2, 3 less than that between 3 and 4. Veins 4, 5 from angle of cell ; a fold between 5,6, extending across cell; 11 from cell; 7, 8, 9, 10 stalked. Costa greatly bulged at base. Hind wing with the lower diséocellular very short and oblique; veins 2 and 4 on a long stalk, - 3 absent, 5 present, 6 and 7 from upper angle. = ; Type, quadricincta, Kaye. The genus comes nearest to Sarosa, from which it differs markedly in the position of veins 2 and 3 of the fore wing. 226 Mr. W. J. Kaye on new Tigridania quadricincta, sp. n. Fore wing smoky transparent, with the costa broadly black from before discocellular to apex, which is very broadly black. Discoidal spot black ; outer margin with an extension inwards along vein 2, and inner margin with an extension along vein 106 heavy black. Hind wing bluish transparent with hardly any smoky appearance; outer and inner margins heavily black and costa clothed with pale yellowish hair. Abdomen with four yellowish segmental rings. The last four segments black. Fore coxe whitish beneath. Frons’ white; gule pale yellowish; tegule with two pale yellowish spots. Mesothorax with a long central pale spot and patagia with a pale area at base and a pale stripe beyond middle. Metathorax with two pale spots. Expanse 66 mm. Hab. Upper Amazons, Rio Ucayali. Type in Coll. Joicey. Autochloris crinopoda, sp. n. Head black with blue scaling at vertex ; tegule black with blue patches ; shoulders with white spots. ‘Thorax and patagia black. Abdomen black with indistinct sublateral blue patches; last four segments crimson. Hind tibiee with dense orange tufts of hair. Fore wing hyaline with heavy black margins; a heavy black discoidal blotch and a similar blotch between cell and inner margin. Hind wing hyaline, with the outer margin broadly black and a small black discoidal mark. Abdomen below with only the last two segments crimson. » Fore cox with exterior patches of white scales. Expanse 41 mm. Hab. Cayenne. Lge Ab. lutea, nov. Abdomen with the last three segments yellow, the fourth only yellow laterally. Hab, Ecuador, Sarayacu (C. Buckley). Saurita pebasa, sp. n. Head black, tegule black ; patagia with large red patclies ; shoulders with red patches. Abdomen black. Fore wing smoky black, darker about the discocellulars and with a pale Species of Syntomide, Nymphalide, de. 227 transverse area across the disc. Hind wing smoky black, darker at apical area and inuer margin. Expanse 22 mm. fab. Peru, Pebas Loreto, 1913. Chrostosoma guianensis, sp. n. Head black with some blue scaling behind the eyes, Thorax black; patagia with red spots and a red spot on the shoulder. Metathorax with a blue spot. Abdomen black, legs and palpi black, Fore wing hyaline, smoky, with dark scaling at base, along inner margin, and at apex. Hind wing smoky hyaline, with apex and inner margin narrowly darker. Hxpanse 28 mm. Hab. British Guiana. Chrostosoma halli, sp. n. Head and thorax black. Shoulders with red patches. First abdominal segment with a pair of subdorsal red spots. Abdomen black with some metallic-green scaling, especially on last three segments. Abdomen beneath white on first three segments and with orange sublateral patches on fifth and sixth segments. Fore wing yellowish hyaline, with the costa narrowly black beyond the cell and with the apex black. Outer margin very narrowly black. Hind wing yellow hyaline; outer margins narrowly black, becoming broader at anal angle. Expanse 33 mm. Hab. Guatemala, Barrios, 22. xii. 12 (A. Hail). Pheta serpensis, sp. n. Head black ; frons metallic green. Tegule orange with a few blue-green scales. Patagia orange. Coxe vermilion- red. Abdomen above orange with a broad expanding median stripe of blackish brown. Abdomen beneath with a large white valve covering the basal segments. Last five seg- ments black-brown. Fore wing with costa as far as discoidal cell orange, and inner margin for half the distance orange. Wings hyaline. Discoidal spot black, rather rectangular. Outer margin broadly black; apex broad, black. Hind wing transparent, the margins black. Antenne with the tips white. Exxpanse 26 mm. Hab. Lower Amazon, Serpa, Jan.—Mar. 1914 (A. Hall). 228 Mr. W. J. Kaye on new Pheia nanata, sp. n. Head black with vertex of head, tegule, frons, and shoulders with metallic-green spots. Abdomen with first segment with sublateral red spots and a series of faint dorsal green spots. Fore coxe brilliant vermilion-red. A large white valve covering basal segments beneath. Fore wing transparent with costa, discoidal spot and outer margin black, the last broad at apex and expanding inwards at vein 2. Hind wing with the costa and cell filled up with dark scaling. Apex rather broadly black. Expanse 26 mm. Hab. Peru, Rio Pacaya, Lower Ucayali, Aug.—Sept., 1912. Related to Pheta hemapera, Schs. Rhyncopyga discalba, sp. n. Frons black, vertex of head black. Collar orange, tegule orange. Thorax and abdomen black. First two joints of palpi orange. Coxze and valve covering basal segments white. Underside of last five abdominal segments orange. Fore wing with the basal half transparent, Discal half of wing dull black, containing a large white discoidal spot. Median vein heavily scaled with blackish. Hind wing transparent with a broad black apex. Expanse 19 mm. Hab. Panama, Bugaba. Related to 2. flavicollis, Druce. Cosmosoma ochreipennis, sp. 0. Palpi orange ; frons yellowish. Blue spots behind antenne. Tegulee black with metallic-blue spots. Patagia black with a central orange streak. Hind tarsus black above, orange beneath. Thorax black. Abdomen black, segmented with orange and with subdorsal metallic-blue spots, the last five segments with a subsidiary second row of blue spots, Fore wing transparent yellowish, the costa yellow, apex broadly black, and outer margin narrowly black, wider at tornus. Hind wing transparent yellowish with a narrow black outer margin. Expanse 32 mm. Hab. Peru, Contamana, Rio Ucayah, xix. 1912. Gymnelia semicincta, sp. n. Frons black, between antenne bluish black. Tegule with brilliant blue patches. Patagia black. ‘Thorax black. Species of Syntomidee, Nymphalide, cc. 229 Abdomen with first segment above orange, becoming paler at sides. A broad black dorsal fascia running down the remaining segments, edged on the sides with orange seg- mental bands and interspaces of bluish scales, especially on the sixth and seventh segments. Fore wing with a small bunch of white scales at base. Costa yellowish, becoming orange beyond the cel]. Inner margin orange on basal half. Outer margin black, the apex very broad, the remainder very narrow. Wing-membrane yellow. Hind wing slightly less yellow than fore wing. Inner margin rather broadly black, outer margin narrow, Expanse 25 mm. Hab. Colombia, Valparaiso. Mesothen demicostata, sp. n. Palpi black; vertex of head metallic blue-green; legs orange. Tegule and patagia edged orange. Metathorax and first five segments of abdomen with orange segmental bands.- Fore wing yellowish transparent. Costa on the central area bright orange; basally and on apical third black. Apex rather narrowly black and outer margin very narrowly black. Inner margin narrowly orange, except at base, which is black. Hind wing yellowish transparent, with outer margin narrowly black. Expanse 28 mm. Hab. W. Colombia (San Antonio), 5800 ft., Nov. 1907 (Ml. G. Palmer). Rhyncopyga semirufa subochrea, subsp. n. Fore wing lighter, more ochreous than in semirufa. No dark discoidal mark and with the dark marginal band greatly narrowed at tornus. Between discocellulars and marginal band a broad ochreous shade. Hind wing paler than semz- rufa and with a slightly narrower marginal band. Fore wing below with distinct ochreous postdiscal band. Expanse 26 mm. Hab. N. Peru, River Tabaconas, 6000 ft. (A. H#. & F. Pratt), 1912. PSEUDODIPTERA, gen. nov. Proboscis absent; palpi slightly downcurved ; antenne bipectinate, with long branches. Thorax and second seg- ment of abdomen clothed with hair. Fore wing long ; vein 3 long hefore end of cell; 4, 5 on a short stalk ; 6 from middle of discocellulars, curving down greatly towards vein 5; 7, 8, 9, 10, and 11 stalked. Hind wing small, greatly cut away Ann. & Mag. N. Hist. Ser. 9. Vol. ii. 17 230 Mr, W. J. Kaye on new at apex; veins 3 and 5 widely separated, 4 absent, 6 absent. A short veinlet in the cell. Type, muszforme. Pseudodiptera comes nearest to Apisa, from which it differs in having veins 3 and 5 of hind wing widely separated at origin and in having vein 6 of fore wing from middle of discocellulars. Pseudodiptera musiforme, sp. 0. Palpi black; frons with large white spot. Head black with metallic-blue spot between antenne. Tegule with white patches. Patagia black with white spot at base of wing. Below, fore coxee white and white patches at base of tibiz. A broad orange stripe on underside of abdomen. Abdomen above black with dark green metallic segmental bands. Fore wing transparent, the margins narrowly black. Discoidal spot narrowly black, connected with outer margin by a short black streak along vein 5. Inner margin with a black extension inwards midway. Hind wing transparent, with the costa and cell filled up with blackish, Expanse 24 mm. tio Hab. Congo, Oubangui-chari, Tschad. Type in Coll. Joicey. Family Hesperida. Subfamily Paweuizrr az. Pseudosarbia camptcola, sp. n. Head, thorax, and abdomen dull brownish black. Fore wing above dull brownish black, with a broad, macular, creamy-whitish, transparent, median band, commencing on costa as a small whitish dot succeeded by a rather square spot within the cell; a much larger and more transparent spot between veins 2, 3, and a creamish-white, more opaque spot lying beneath, but not reaching the inner margin by about 1-2 millimetres, Cilia same as the ground-colour, except for a large white area at tornus. Just beyond cell is a broad regular white band from costa to vein 4, with the veins showing through brownish. Between veins 3, 4 near cell is a small white comma-like mark. Fore wing below as above, except that instead of a small white dot on costa at commencement of band there is a pale yellow streak. Hind wing above dull brownish black with a broad white band from vein 8 to vein 2 divided up into sections by the dark brewn veins. Cilia at apex brown, becoming white thence to tornus, where it is considerably longer. Hind wing Species of Syntomide, Nymphalidee, &e. 231 below as above, except for a straight yellow streak within the cell which runs beyond the discocellulars along the fold in place of vein 5. Abdomen beneath with paired white spots on sternites 4, 5, 6, 7, and 8. ?. Like the male, except that all the white markings are broader. Expanse, ¢ 52 mm., ? 58 mm. Hab. 8. Brazil, Parana, Ponta Grossa, 1 ¢, 30. 3.1910 (W. J. Kaye). Uruguay (2. Trimen). 3 type in Coll. Kaye. @ type in Coll. Joicey, The habitat of this striking “skipper” is open grassy campo in S. Brazil at 3000 ft. elevation. Hardly another butterfly was to be seen where the ¢ was caught, although a close search was made at the time for further specimens of what I recognized at the time as a rarity. On the label of the ? specimen labelled Uruguay it is stated “Mr. W. C. Hewitson had this Hesperid from me [Rowland Trimen] to describe and figure together with the specimen of Papilio hellanichus (also from Uruguay); but although he attached to it the label ‘ Apheka’ I have not found that he published any description or figure of it.— R. Trimen.” The type of Papilio hellanichus, once in the Trimen col- lection, was acquired with the whole collection by Mr. Joicey. Family Nymphalidae. Chlorippe speciosissima, sp. 0. @. Fore wing ochre-yellow with two black transverse marks, the one within the cell flat V-shaped, the other lying along discocellulars. A pale transverse band across disc, straight to vein 3, then set back and broken; a conspicuous blackish spot surrounded with reddish ochreous near tornus between veins 2 and 3. A dark shade in subapical area ' containing two pale round spots. Subterminal black line regular preceded by a crenulated black band which merges in the dark subapical area. Hind wing ochre-yellow with a small round black spot within the cell, lying close to origin of vein 7. Costal area brownish black with a square whitish patch in middle, which represents the end of a transverse band which is almost obliterated. A large black spot sur- rounded with reddish ochreous between veins 2, 3. Sub- terminal line black, regular to vein 2, where it is strongly . toothed and edged externally with grey. A heavy black inner crenulated band also strongly toothed at vein 2. Outer margin crenulated. Underside of hind wing pinkish silvery 232 Mr. R. Broom on the with the upperside markings showing through, and with a well-defined central whitish band becoming more or less merged with the ground-colour at anal angle. Expanse 82 mm. Hab. Haiti, no precise locality. Type in Coll. Joicey. Anca xenocrates punctimarginale, subsp. n. &. Differs from xenocrates xenocrates from Bolivia in the fore wing by having no blue scaling at tornus and in the blue subapical spots being widely separated and showing no tendency to unite inwards. Hind wing with a series of rather small triangular blue marginal spots, not a band as in the Bolivian form. @. Shows much less difference from type-form. The margin of hind wing is yellow banded as in the 9 from Bolivia. There is an extra yellow spot between veins 3, 4, smaller than that between veins 2, 3. EXxpanse 82 mm. Hab. French Guiana, St. Jean de Maroni. Lion Type in Coll. Joicey. The occurrence in French Guiana of a species only known hitherto from Bolivia and the Upper Amazons (Pebas) is strange, and at first suggests specific difference and not sub- specific. But the species is rare, the ? exceedingly so, and its range may lie across the interior of Brazil where it could easily remain undetected. The species has been chiefly known from Eastern Bolivia, but the few specimens known from Pebas belong to the same form with a blue marginal hind-wing band in the ¢. XX VI.— Observations on the Genus Lysorophus, Cope. By Rosrert Broom. With a Note, by Prof. W. J. Souuas. So much has already been written about this little vertebrate by Broili, Case, v. Huene, Moodie, Finney, and Williston that it might seem doubtful wisdom to add another paper to the already extensive literature, and more especially as my observations are on specimens already carefully examined ‘by Case and v. Huene; but when one considers that ‘Lysorophus is the most remarkable land vertebrate that has been discovered for many years, and that opinions not only differ as to its affinities but also as to the interpretation of a number of the cranial elements, a further review of even the present evidences seems justifiable. Genus Lysorophus, Cope. 233 There is no lack of material. The Chicago Museum has 200 nodules, each containing much of the skeleton of a specimen: the American Museum, New York, also has many nodules, and in the American Museum nine skulls have been chiselled out, one or two in very good condition. In Tiibingen there are 24 skulls, and at Munich a consider- able number more. As the extensive literature has been reviewed by Williston and others, it will be unnecessary to enter into this in detail. To Broili we owe the first really good figures of the skull, but there are one or two points in his interpretation that I, in common with all later writers, do not accept, and from his conclusion as to the affinities of the genus I also differ. Case gives a brief description of the more conspicuous elements of the skull, and reproduces Broili’s and Williston’s figures. As these two figures differ in a number of points, one could have wished that Case had given an original figure of his own interpretation, and his description, while pointing out the different views, does little to clear up the matter. Williston gives us clear definite views as to the structure of the skull and skeleton, and equally clear opinions as to the affinities of the genus. Von Huene, the latest worker on the genus, has just issued a paper on Lysorophus in the ‘ Anatomischer Anzeiger,’ and. another paper is in the press describing the specimens in the American Museum. Though these two papers are appearing in the same year, I believe that the one in the ‘Anatomischer Anzeiger’ to be the later. On one or two points the opinions expressed differ in the two, and it is therefore well to know which is the latest. Von Huene has figured a number of the better skulls in the American Museum, and gives us clear opinions not only on the structure, but also on the affinites of the genus. The skulls in the American Museum, though comparatively few in number, are mostly well preserved, and there is scarcely a point in the structure that cannot be made out in one or other. The best figures published of the top of the skull are those of Broili and Williston, and they differ, apart from inter- pretations, only in the relative width of the nasal region. While neither is altogether correct, a composite of the two would give the truth. The difference arises from the peculiar state of affairs in front of the prefrontal. Broili correctly recognises a round opening here which he regards as the nostril. It is also shown in Williston’s specimen. The most natural conclusion would seem to be that this is * 234 Mr. R. Broom on the the nostril, but two of the American Museum specimens seem to indicate that the opening extends somewhat inwards and forwards, and one would like to see a specimen showing the perfect snout to feel quite sure that this opening is the nasal opening and not perhaps also an opening for some sensory organ. There is a small premaxilla—possibly toothed. It is figured by v. Huene. The maxilla is slender and carries about ten teeth. Its posterior end articulates, I believe, with the palatine. It forms the floor of the nasal opening Fig. 1. Pal. ~ Ss NY SS SS SS RES Restoration of the underside of skull of Lysorophus tricarinatus, Cope, X 5 and perhaps its posterior border. The doubt lies in the fact that in the specimens it is impossible to be quite sure whether the bridge of bone which connects the prefrontal with the maxilla is a part of the prefrontal or a part of the maxilla or a small independent bone. One specimen shows most of the palate. The bones area little crushed and fractured, and the interpretation I give is made with some hesitation (fig. 1). Von Huene figures the specimen, but his interpretation differs somewhat from mine, Genus Lysorophus, Cope. 235 which agrees pretty closely with Broili’s. I consider v. Huene in error in regarding that there are “ two large, elongate internal nares, separated by a narrow bridge.” The large supposed left choana of v. Huene I regard as the median vacuity between the prevomers, and the narrow bridge as the right prevomer. The figure I give will show how I interpret the palatal structures. The prevomers form a horseshoe-like arrangement with posterior processes passing back to the parasphenoid and apparently articulating with the pterygoids. The teeth on the prevomers are well shown in this specimen. In front there are about 6 and about 8 on each side. The paiatines are delicate bones extending from the maxillz to the pterygoids. Between the palatines and prevomers are, I believe, the internal nares. The ptery- goids extend back as rather delicate bones to meet the quadrates. The parasphenoid is a very large bone, which forms nearly the whole of the base of the posterior two- thirds of the skull. The supposed suture figured by v. Huene between the parasphenoid and the basisphenoid is, I think, a fracture merely. The ficure I give of a transverse section of the skull (fig. 3) shows the relations of the pterygoid to the para- sphenoid, and also the elements of the back of the mandible. In Broili’s figure A of the side view of the skull, there are seen in the orbital region some deep-seated elements. These are also shown in two of the American Museum specimens. In what might be regarded as the sphenethmoid region there appear to be three elements with a deep posterior notch. In one of the New York specimens an almost exactly similar appearance is shown, and further back an elongated element very like an epipterygoid in appearance. Though these elements have been seen by Broili, neither he nor anyone else appears to have expressed any opinion as to what they were. After considering many possibilities I have come to the conclusion that they are ossifications or calcifications in the cartilaginous brain-case. The anterior elements look as if separated by sutures, but, whereas all true sutures in the skull and even cracks are filled with the red clayey matrix, these divisions are formed of clear calcite which probably indicates that they were originally formed by hyaline cartilage. Further, in a second specimen the ossification appears to be entire. ‘The posterior narrow vertical element is also, in my opinion, an ossification of the cranial cartilage. It certainly has much superficial resemblance to a reptilian epipterygoid. It articulates with the parietal above and passes down to at least near to the pterygoid. It thus answers in position to the epipterygoid. 236 Mr. R. Broom on the But though in front it has a smooth edge the posterior edge is irregular, as if indicating an ossification in cartilage. The anterior ossification or ossifications probably correspond to the sphenethmoid of Siredon or the frog, and the posterior to the ossification seen in Dinosaurs, Crocodiles, and birds, and usually, but I think wrongly, called alispbeuoid. The quadrate is large and its upper half is largely hidden by the squamosal. There need not, I think, be the slightest i y Lower jaw of Lysorophus tricarinatus, Cope, X 5. A represents a section at aa. Ang., angular; D, dentary; P.Art., prearticular; S.Ang., surangular. Fig. 3, fai Fe. Pe. S.Ang PL Art. J} Ang. Section across skull and jaw of Lysorophus tricarinatus, Cope, X 5. The section of the lower jaw is near the point indicated by 06 in the figure of the jaw. The outer corners of the parasphenoid are sepa- rated by cracks or sutures. They are believed to be parts of the parasphenoid. ' Ang., angular; Pa., parietal ; Pa.Sp., parasphenoid ; P.Art., prearticular ; Pt., pterygoid ; S.dng., surangular. doubt about this bone being the squamosal—the view also held by Williston and v. Huene. The occiput has recently been figured by v. Huene from one of the American Museum specimens and also from one of the Tibingen specimens. His drawing of the American Museum specimen is not in my opinion quite accurate, the American specimen agreeing closely with his figure of the Tiibingen specimen. ‘The main difference between the two Genus Lysorophus, Cope. 237 is that in the drawing of the American specimen the ex- occipital is represented as very small. This is, I think, wrong, the exoccipital being large, as represented in the drawing of the Tiibingen specimen. The drawing v. Huene gives of the occipital condyle is thoroughly satisfactory, showing that the articulation is as much basi- as exoccipital. Von Huene’s identifications of the fenestra ovalis and fora- men for the vagus are probably correct. The large bone situated by the sides of the supraoccipital has been very variously identified. By Broili and Case they have been called squamosals, by Williston epiotics, and by v. Huene supratemporals. That they cannot be squa- mosals requires no argument, the undoubted squamosals lying in front. Nor can they, I think, be regarded as supratemporals. From their being quite behind the parie- tals, and at the sides of the supraoccipital and far behind the jaw, it is very doubtful if they im any way roof the temporal region. They may be epiotics, but we do not know any forms in which epiotics take up this position. They further appear to overlap the supraoccipital, and to be thus membrane bones. It seems to me that they, how- ever, answer all the requirements of the tabulares. They lie on the outer part of the paroccipitals, are behind the parietals, and articulate with both the parietals and squa- mosals, and to form the upper lateral parts of the occiput. The lower jaw has never been fully described. Von Huene figures one of the specimens in the American Museum, but with one or two of his interpretations I do not agree. He has also examined some jaws in the Tubingen Museum, but they have apparently not yielded any fresh light. The American Museum specimen, no. 4761, shows something of ‘the jaw, but not nearly so much as two other specimens not numbered. Between these three specimens practically all details can be made out (fig. 2). The dentary forms about two-thirds of the jaw. It comes to a sharp point in front and forms with its neighbour a short feeble symphysis. It articulates on the outer side behind with the surangular and angular. The splenial is a small bone lying on the inside of the lower part of the dentary just behind the symphysis. It forms the lower margin of the jaw in this region. The angular forms nearly the whole of the lower border of the jaw, passing in front between the dentary and the splenial. From two of the American Museum specimens I incline to differ from v. Huene, and believe that the splenial does not form part Ann. & Mag. NV. Hist. Ser. 9. Vol, ii. 18 238 Mr. R. Broom on the of the symphysis. The surangular forms the upper half of the back of the jaw as indicated in the figure. Von Hueneis, I think, in error in regarding the large opening in the side of the jaw in specimen 4716 as natural. Only a small part is, | believe, a natural opening, the rest due to faulty pre- paration. In other specimens the lateral opening is quite small, as indicated in the figure. I find no evidence of a coronoid element. Inside the jaw is a large prearticular. The articular is evidently quite small, and possibly carti- laginous. Though the structure of the skull of Lysorophus may now be said to be pretty well known, there is still some little doubt as to the affinities. Lysorophus agrees closely with no known animal, recent or extinct. With Williston I agree in holding that Lysorophus is not a reptile. All known reptiles are either Cotylosaurs or are manifestly derived from Cotylosaurian ancestors, but Lysorophus is neither a Cotylo- saur nor can it have been derived from a Cotylosaur. The supposed reptilian resemblances are entirely fallacious. Von Huene in his recent paper, though correctly figuring and describing the occipital condyle, says: “ this condyle is intermediate between the true reptilian condyle and the true amphibian condyle .... The structure of the condyle shows a great resemblance to that of the Theromorphs and of Turtles.”” In Theromorphs and Turtles the condyle is a tripartite condyle, of which the upper two-thirds are formed by the exoccipitals and the lower third by the basioccipital. In most Chelonians and Theromorphs the exoccipitals come close together, and the basioccipital is squeezed out from the foramen magnum. Inall generalised forms the condyle is a projecting rounded structure which articulates with the arches of the atlas and with the intercentrum. In Lyso- rophus the whole articulation is with the centrum of the atlas, which fits close into the broad hollowed out surface formed by the basi- and exoccipitals, The presence of a large articular surface on the basioccipital seems at first sight to be a non-Amphibian character, but, as Watson has recently pointed out, this is the primitive Amphibian condition. The early Stegocephalians of the Lower Car- boniferous, such as Pteroplax, have the basicecipital forming practically the whole of the articulation, the exoccipitals only very gradually in later forms taking the place of the basioccipital. So that, so far from the occipital condyle of Lysorophus indicating any reptilian affinities, it is really in a more primitive condition than is found in any other Permian or later Amphibian. Genus Lysorophus, Cope. 239 Doubtless Williston is right in regarding Lysorophus as a mud-borrowing animal, and many of its specialisations are due to this habit, such as the greatly elongated snake-like body with very numerous vertebre, great reduction of the limbs, relatively small size of skull, loss of the arches, and advanced position of the quadrate. And ‘the somewhat similar characters, acquired by convergence in other groups which have similar habits, have given rise to some striking superficial resemblances to Lysorophus in the Gymnophiona, the Amphisbenans, and the Typhlopide. But, apart from all modifications in Lysoruphus due to a burrowing habit, the skull is undoubtedly fundamentally an Amphibian skull, and the only known Amphibia, recent or extinct, with which it seems at all allied are the Urodela, and, more remotely, the Anura and the Gymnophiona. Note by Prof. W. J. Soutas. Some years ago Dr. Broom obtained, through the kindness of Dr. Matthew, two specimens of Lysorophus, and these he presented to me for investigation by serial sections; at the same time he made a most generous aldition to this gift by placing in my hands, to dispose of as I thought fit, a paper embodying the important conclusions to which he had been led from his study of the specimens in American museums. My own study is now completed, and I hope soon to give a full and exact account of the structure of the skull in all its details. This will confirm all the more important con- clusions of Dr. Broom, and in justice to him I can no longer withhold from publication the paper which he entrusted to me in 1914. One or two minor emendations ought, perhaps, to be made. Thus, the vacuity between the vomers, as it is repre- sented in fig. 1, does not really exist; these bones are without thickened margins and meet in the middle line; and, again, the articulare of the lower jaw is a comparatively large and important bone. On the other hand, there can be no doubt that the cranial walls include, as Dr. Broom suggests, a large “sphen- ethmoid” and “alisphenoids.” These are shown in section in the accompanying figures (figs. 4 & 5). The whole anatomy of the skull recalls in a striking manner that of Siren or Menopomus, and to my mind Lyso- rophus is without doubt an ancestral Urodele. It presents some remarkably interesting primitive characters, 240 On the Genus Lysorophus, Cope. Fig. 4. Ptr: Fa. Q Pt. Pa_S. ae ~T ou, A B Transverse sections of skull of Zysor ophus, to show the ephenethmam and “alisphenoid ” bones. A. Sphenethmoid: /'r., frontal; Pa.S., parasphenoid ; P.Fr., prefrontal ; H., pterygoid ; Sp. E. , sphenethmoid, Bp.“ Alisphenoid ” (CAB a Art., articulare of lower jaw ; Pa., parietal ; Qu., quadrate. Fig. 5. ie Vo- Three horizontal sections superposed. Vo., vomers; Pr.Ot., pro-otic ; Sg., squamosal; S.o., en E.o., exoccipital ; Tab., tabulare; Pr.At., pro-atlas, - Periodicals published by TAYLOR & FRANCIS. Published the First Day of every Month —2s. 6d. SERS LONDON, EDINBURGH, ‘and DUBLIN | PHILOSOPHICAL “MAGAZINE Bi AND F per JOURNAL OF ‘SCIENCE... = Conducted by SIR OLIVER JOSEPH LODGE, D.Se., LL.D., PRS, SIR JOSEPH JOHN THOMSON, 0.M., MA, Se. ‘D., FP. RS., JOHN JOLY, M.A., D.Se., E.RBS., E.GS., GEORGE CAREY FOSTER, B.A, LL.D., ERS, and WILLIAM FRANCIS, F. L. 8. Published the Pir st Da, ry of every Month.—2s. 6d. THE - ANNALS AND “MAGAZINE OF NATURAL HISTORY eadns: Zoology, Botany, and Geology. - Condueted by — WILLIAM CARRUTHERS, Ph.D., F.B.S., FL, s FAG ORS Phos _ ARTHUR E. SHIPLEY, M. A.; Se. D., P.R.S., F.Z.S., and - RICHARD T. FRAN ols, B. Z.8. ES Peuiden in 1877. Monthly, 1s, Annual Subser -iption Tue Companion * and 12 Monthly Numbers), 18s, ane, en Bf MONTHLY REVIEW OF ASTRONOMY. s Edited by A. S. EDDINGTON, M. A., M.Sce., F. R. S,, / Fr J. M. STRATTON, M:A., ahd H.S. JONES, M. Ay: B. Se. “Its Davieetritave include most of the leading English- -speaking Astronomers, ; aig its Articles give an interesting reflection of the progress of astronomical : ‘4 thought and achievement. : ' Published Monthly ys Price 1s. 8d. “Annual Subscription, 16s, post jr "ee. ‘ THE © “JOURNAL OF. _BOTANY pets AND FOREIGN. Kdited by — _ JAMES BRITTEN, K. 8. G, BF 1:8. ‘LATE SENIOR ASSISTANT, DEPARTMENT OF BOTANY, BRITISH MUSEUM, _ Articles by eae Botanists ; Announcements of Discoveries; Reviews; Ry ~ Bot Lp tani News and pases &e., See, Ried Lion Court, Fleet Street, London. CONTENTS OF NUMBER 9.—WNinth Series. z, XIX. Descriptions of New Pyralide of the Subfamily Py- re raustine. By Sir Grorce F. Hampson, Bart., F.Z.S., &e. XX. New Australian Hymenoptera of the Family Hwantidee in the British Museum. ‘By Rowzanp E. Turner, F.Z.8., F.E.S. .... 197 XXI. A revised Classification of the Otomyinw, with Descriptions of new Genera and Species. By Orprirnp THomas............ .» 203 XXII. The Hedgehog of Palestine and Asia Minor. By Oxprrenp Mntiaas ae ceo aig see ac S usta PRL UIE Melgar tan oe me 211 XXIII. Ona new Jumping Mite of the Genus Nanorchestes from. the Mendip Hills, By Sranuey Hirst ............... SRR ary et 213 XXIV. On some External Characters of Ruminant Artiodactyla.— Part III. The Bubaline. and Orygine. By R.1. Pocock, F.R.S, .. 214 — XXYV. Descriptions from the Joicey Collection of new Species of ‘ Syntomide, Nymphalide, and Hesperide, and Two Genera of Syntomide, By W.J. Kare, FBS. ....----, fechas ae XXVI. Observations on the Genus Lysorophus, Cope. By Roserr Broom. With a Note, by Prof. W. J. Soutas ....-. cece eee es -. 232 *,* Ttis requested that all Communie¢ations for this Work may be addressed, post-paid, to the Care of Messrs. Taylor and Francis, Printing Office, Red Lion Court, Fleet Street, London. “THE ANNALS AND eae “INOLUDING ZOOLOGY, ‘BOTANY, AND. GnoLoay. oo ty a. as Se ee NS re Oa GEASS By : CONDUCTED. ‘BY WILLIAM CARRUTHERS, Pu. DD; Saf BS., F.LS: E.G. S., ARTHUR E. SHIPLEY, MLA., Bo. D., F. R. 8., F, ZS8., AND --52 RICHARD ZT, FRANCIS, F. - LS. : BEING. A CONTINUATION OF THE * < girate” sedhesnank WITH MRESRS._ LOUDON AND CHARILESWORTH’S “MAG ALINE oF NATURAL HISTORY.” "TAYLOR AND FRANCIS, RED LION COURT, FLERE STRERD, Bz, Sold: by ‘Simpkin, hens shall, Hamilton, Rant & Co.; Ld: eS Paris ra ates and Hodges, Figgis, & Oo, Fedtas am Double Number. Price Five Shillings. WATKINS | & DONCASTER Raturalists, Ee 6 Keep. in stock every kind ot “APPARATUS and CABINETS required by ENTOMOLOGISTS, ORNITH- ~— OLOGISTS, BOTANISTS, &c. Also NESTING- BOXES, - Which should be fixed up in idloebnig ie ‘shrubberies ‘ before the breeding Season. % Botte A Large Stock of Butterflies, Moths, Birds, E - be ‘Full Oatalogue (84 pages) mailed free to ‘any address. 36, STRAND, LONDON, » WC. ENGLAND. SCIENTIFIC BOOKS AND SERIALS. JOHN WHELDON & CO. have the largest stock in the country: of Books in all Departments of Science and Natural History, also ‘Transactions and J ournals of Learned Societies, etc., in sets, runs, and single volumes or numbers. Ast 8%, Libraries or small pareels purchased for cash, SPECIAL CATALOGUES: Entomological, Omithoibeieal: Dosloriends Geolo- gical, War Economy (Agriculture, Husbandry, Gardening, Economic Botany, ete. ) | Botanical (General and Geographical), ete., 2d. each, post free. ; i 38 GREAT QUEEN STREET, ‘KINGSWAY, LONDON, WE. a - Telephone : ‘Gerrard 141 2. Super -Royal 8vo, pp. iv, 226. With 4 Plates and 16 Test: RO ee he gee A CONTRIBUTION * PHYTOGEOGRAPHY AND FLORA DUTCH N. w. NEW ‘GUINEA. L. S. GIBBS, FLS, FRMS. ‘Taytor AND FRANCIS, Red Lion Court, Fleet Streets Rates for Adverteenkaie) in ae Anais Hee Magazine of Natural History. es a * a Ole tet Se ‘Twelve Meee Insertion. 2 ‘Insertions. | Inaction. A pier PAGE - 450-0 - 32, Oe AB: O each 1427 each) ee ie HALF-PAGE - - 12 6 "4050 lit S47 G' ” pes QUARTERPAGE - 1226 U3,, ae 10 oO. All applications for space to be made igh ae Mr. H. A. COLLINS, 32 Birdhurst Ro r ae ™, . as ih So n a ie ; ‘ Uf THE ANNALS 5,. \ 9 cP] ; ) ty “199 3 AND ‘ “op a/ Busey MAGAZINE OF NATURAL HISTOR [NINTIL SERLES.) No. 10. OCTOBER 1918. XXVII.—On the Ries and Variation of the Hdible Frog, Rana esculenta, L. By G. A. BouLencer, RLS. (Published by permission of the Trustees of the British Museum.) Arter all I have written in the past on this common Batrachian, it may seem surprising that I should think it worth while to revert to the subject. The reason is that it is far frem exhausted; that I have never ceased accumu- lating material *, in the course of recently reviewing which I perceived characters hitherto overlooked; that it was desirable to test the value of certain differences appealed to within the last few years by advocates of the extreme multi- plication of species ; and that it is always useful to deal with individual variations, when large series of specimens are available, in order, by showing the instability of certain characters, to ensure a more correct appreciation of their importance when treating of allied species represented by less extensive material. Not that I think inconstancy in one case invariably follows in another, but such examples teach caution, and should be a warning to the inexperienced. Considering modern tendencies in zoography, it cannot be too often repeated that the method of describing so-called species and subspecies from single ‘specimens T or from at * About 800 specimens are now before me, selected from at least twice as many that have passed through my hands. f “On aura beau multiplier les espéces, on arrivera toujours ice , résultat que la description exacte d’un sujet pris au hasard, parmi soixante récoltés sur des points divers d'un méme rivage, ne pourra conyenir a aucun des cinquante neuf autres.” Duvyal-Jouve, Mém, Ac, Montpell. vii. 1871, p. 511. Ann, & Mag. N. Uist. Ser. 9. Vol.ii. 19 242 Mr. G. A. Boulenger on the Races and most a very few, when large series can be examined, or without reference to the data available through the labours of other investigators, is unfair to those who make use of works written on such lines. Systematics, if scientific, must take into full consideration the exceptional, aberrant, or annectant specimens, so often passed over without a word, though of so great an importance from the taxonomic an evolutionary points of view. It does not matter if thereby our definitions are obscured, the object to be attained is to depict the true state of things in Nature. To the four forms which | have previously * distinguished, as forma typica, var. ridibunda, Pall., var. lessone, Camer., var. chinensis, Osb., I have recently added a fifth, var. saharica t, founded on specimens obtained by Dr. I. Hartert in the far interior of the Algerian Sahara (E1 Golea, Tedikel oases), a small race nearly related to the var. r¢dibunda of the northern parts of Algeria but differing in the shorter tibiee, constantly less than half the length of head and body aud not overlapping when the limbs are folded at right angles to the body ; the membrane between the toes is very deeply notched, so much so that many specimens may be described as having the foot only three-fourths webbed. The Vomerine Teeth. I have never seen these teeth in two series on the round or elliptic bony bases that bear them, as described and figured by Fatio¢. They form a single series, composed of 3 to 8; in exceptional cases I find only 1 or 2 teeth (speci- mens of the typical form from St. Malo, Brussels, and Basle). Leydig § gives the number 3 as normal, but he caunot have examined many specimens, those on which he drew up his description being probably mostly of the var. lessone, as the figure of the foot given in his book indicates, and this number is very frequent in the variety in question, although it may rise to 5. In 8 frogsct the typical form from Basle I find only 2 to 4 teeth, whilst in 35 from other pa:ts of Switzer- Jand, from France, ard from Germany I count 3 to 7, 4 to 6 being the usual number; I] have also seen a_ toothiless specimen from Vienna. In about 30 specimens of the var. ridibunda from Germany and Austria I count usually 4 to 6 teeth ; 3 specimens have only 3, one has 7 on one side and * Proc. Zool. Soc. 1891, p. 374, and Taill. Batr, Eur. p. 270 (1898). + Noy. Zool. xx. 1918, p. 84. { Vert. Suisse, iii. p. 313, pl. v. fig. 7 (1872). § An, Batr. Deutschl. p. 112, pl. ii. tig. 20 (1877). Variation of the Edible Frog. 243 8 on the other, and one (from Vienna) has but a single tooth. 3 to 5 is the usual number in the var. chinensis, The series of teeth are usually nearer to each other than to the choanee, but they are sometimes equidistant in the typical form and the var. vidibunda and usually so in the var. saharica ; an arrangement such as is represented on the figures in Schreiber’s book * [ am sure never occurs. In a female from Cadillac, Gironde (var. ridibunda) the teeth form long, slightly curved series, extending almost right across the space between the choane. The series are sometimes hori- zontal, sometimes more or less oblique though seldom very Fio.1, S&S @2o @ @&@ oO @CDEDe a b Cc S®sca ® eee ®2n7°@ d e if Vomerine teeth in specimens fro.n St. Malo (a, b), Cadillac (ce), Basle (d), Oporto (e), aud Dead Sea (f). much so; a male from St. Malo has the series oblique on the right side, horizontal on the left. ‘lhe teeth are usually cxactly between the choanz, but they may extend backwards beyond a line connecting the posterior borders of the latter, or, more exceptionally, they may be ona line with their anterior borders (specimens from Oporto and Pekin). There is no difference whatever in the disposition of the vomerine teeth that could help in the definition of the various forms of R. esculenta, The Tongue. The tongue varies much in size: it may nearly cover the floor of the mouth or its width may be only about one-third that of the latter. Bedriaga Thas already mentioned that the posterior processes also vary much in length according * ‘Herpetologia Europea,’ 2nd ed. (1912).—So much in this book is merely careless compilatiun that I need noé further allude to it except to express amazement at the suggestion there made that the Spanish- Portuguese frogs named vars. d*spanica and perezt may be the same as the var. lesson@; also at reading that the males of R. greca and R. iberica are distinguished from those of allied species in having external vocal sacs. + Lurchfauna Europa’s, i. p. 36 (1891). F. 19 244 Mr. G. A. Boulenger on the Races and to individuals ; this is well shown by two specimens from Florence, representing the two extremes. A more or less distinct process between the two horns is sometimes present, as in a specimen of the var. lessonw from Noville, Switzer- land *, c | a e Showing the shape of the tongue in specimens from Berlin, var. ridi- bunda (a, b), Florence, f. typica (c, d), and Noville, var. lessone (e). The Head. According to Bolkay f, the three forms distinguished by him as species differ in the following points :— R. esculenta. Head comparatively narrow, tip of snout ending in a blunt point ; interorbital space equal to half, or frequently to three-quarters, the breadth of the upper eyelid. KR. ridibunda. Head broad, short, up of snout bluntly rounded ; interorbital space equal to one-third the breadth of the upper eyelid. R. chinensis. Head narrow, long, and very pointed at the - end ; interorbital space equal to half the breadth of the upper eyelid. There is no constant difference in the shape of the head between the two first, and although it is a fact that R. chi- nensis usually has a narrower head and a more pointed snout, * This process is usually distinct in the Indian R. heaadactyla, Less, It has been regarded as a specific character in a Central American frog (R. trilobata, Mocquard), which may be merely a young ZR. halecina, L, t+ Proc. Washingt. Ac. Se. xiii. 1911, p. 75. Variation of the Edible Frog. 245 this is by no means always so, and specimens are to be found in which the snout is much more rounded than in some R. ridibunda. I have selected three specimens, of which outline figures are here given, to show that the above definition of the three forms cannot be relied upon. . Fig. 3, 1 OW yi W b c Upper views of heads of typical form, g, St. Malo (a); var. ridibunda, @, Capljina, Herzegovina (b); and var. chinensis, 2, Broughton Bay, Corea (c), 3 nat. size, PX a The width of the head varies between 1 and 1 times its length in the typical form (28; 32 in 2 from Havre), be- tween 1 and 1} times in the var. ridibunda (= in some specimens from Herzegovina, France, Portugal, Algeria, Asia Minor, Persia, 36: 43 in ? from Kiev), between 1 and 1,4, in the var. chinensis. The width of the interorbital space is 4 to $ that of the upper eyelid in specimens of the typical form from St. Malo and Paris, 4 to 2 in others from Poitiers. In the var. ridibunda, taking only specimens from Germany aud Austria-Hungary into consideration, it is between 4 and 4, but it may be exceptionally % (¢ from Laaerberg near Vienna) ; 2 (in a large ? from Damascus) is another excep- tion. In the var. chinensés it varies between ? and 2. The head varies much in shape, and exceptionally may even be not unlike that of a typical 2. temporaria (9, var. ridibunda, from Crete). The canthus rostralis is always very obtuse ; I have never seen a specimen in which it may be said to be “ strongly marked,” as stated by Bolkay in his description of R. chinensis. The Hind Limb. That there are very considerable differences in the pro- portions of the hind limb, [ was the first to point out, and I have proposed to make use of these for defining varieties, with the necessary restrictions in the diagnoses imposed by 246 Mr. G. A. Boulenger on the Races and the many exceptions. The following figure shows how striking these differences between the extremes are :— Fig. 4. Hind limbs of var. ridzbunda, 9 from Astrakhan (a), and var, lessone, © from Stow Bedon, Norfolk (b). 3 nat. size. These differences reside in the length of the tibia compared to that of the head and body, to that of the thigh (causing the heels to overlap, to meet, or to fail to meet when the limb is folded at right angles to the body), and to that of the foot ; also in the size and shape of the inner metatarsal tubercle, its basal length being compared to the length of the inner toe (measured from the base of the tubercle). There is another character, not made use of before, derived from the thickness of the crural or tibial part of the limb ; this varies, like other characters, within certain limits, accord- ing to the actual leneth of the bone and the degree of plump- ness of the individual, but, comparing extreme forms, it will be found that the length of the tibia is usually over 3 times its width in the var. rtdibunda and under 3 times in the var, lessone. When a large material is carefully examined, it is found, however, that these differences break down for the sharp definition of the various forms; there is considerable over- lap between one form and the one next to it in the series, when the measurements are tabulated, thus precluding rigid definitions :— Variation of the Edible Frog. 247 ¥ t a 3. 4. 5. V.ridibunda ., 3-4 14-2+ 1-13 9-14 25-4 V.saharica .... 23-3 21-92 1-1} 9-13 23-43 I, typica ...... 3-4 18-2} 14-1 7-10 2-3 V. lessone...... 22-3 Qit,-22 ji Rib ol seeps) 1 2 V. chinensis .... 23-33 2-23 11-13 5-8 1-1? 1, Width of tibia in length.—2. Length of tibia in length from snout to vent.—3. Length of tibia in length of foot (measured from tarso-metatarsal articulation).—4. Length of metatarsal tubercle in length of tibia.—5. Length of metatarsal tubercle in length of inner toe. Bolkay gives the following characters for distinguishing his three species :— RF. esculenta. Heels never meet; tibio-tarsal articulation reaches space between tympanum and posterior corner of eye (2), or, at the utmost, space between anterior corner ot eyeand nostril ( ¢) ; inner metatarsal tubercle large, com- pressed, projecting, always longer than distance between it and subarticular tubercle of first toe. I. ridibunda. Heels always overlap ; tibio-tarsal joint just reaches back corner of eye (?), or end of snout (¢); inner metatarsal tubercle small, of flattish cylindrical form, not very projecting, always shorter than space between it and subarticular tubercle of first toe. Lt. chinensis. Heels never meet ; tibio-tarsal joint reaches posterior corner of eye or as far as space between anterior corner of eye and nostril; inner metatarsal tubercle very large, projecting, compressed, hard and sharp, always a good deal longer than its distance from subarticular tubercle of first toe, frequently equal to length of first toe. The proportion of the tibia to the thigh, expressed by the meeting or otherwise of the heels, is most useful for dis- tinguishing the races, but it varies like most other characters, aud we must not shut our eyes to exceptions to the rule, To take 2. ridibunda as an example, [ now find that the overlapping of the tibize is not so constant as I formerly believed. Ixceptions have already been noticed by Méhiely * in specimens from Southern Hungary, and I find the character to break down in 4 out of 13 examples from Angora and in 3 from Damascus which have lately been submitted to me by M. H. Gadeau de Kerville; besides, Tam now convinced that the var. susana, proposed by me for specimens from Persia J, in which the tibiee simply meet, * Zichy’s Zool. Forschungsr. p. 61 (1901). + Ann, & Mag. N. H. (7) xvi. 1905, p. 552. 248 Mr. G. A. Boulenger on the Paces and does not deserve to stand. ‘hese exceptions; occurring in Asia, cannot be disposed of by an appeal to hybridity, as in the Sone of critical specimens from Germany and Austria- Hungary, where the var. ridibunda occurs side by side with the typical form, which fact would render such an assumption legitimate, From. what I have myself observed in the Spree lakes near Berlin, I have no doubt the two forms cross in exceptional cases, notwithstanding the asyiig amy which maintains their segregation when living together, but we have no practical means of discriminating between such mongtels and truly annectant specimens. I may mention that the tibiae feebly overlap in one specimen of the typical form from Warsaw and in another from Mestre. As regards the J. chinensis, I am greatly surprised at Bolkay’s statement, which is contrary to the descriptions by myself and by Wolterstorff*, although supported by the description of one specimen by Stejneger T; the two first authors agree as to the heels meeting, Wolter- . o storff even adding that they sometimes slightly overkipe the only specimens in which I find the heels not to meet are from Kobe, Japan (two), and Pekin (6 out of 26), and they must be ood as exceptions to the rule, Although the lind limb is often shorter in the female than in the male, this is by no means generally the case; I can show no end of female specimens of the var. r¢dibunda from Central and Hastern Europe and Asia in which the tibio- tarsal articulation reaches beyond the eye, and even one, from Alemtejo, Portugal, in which it extends to the tip of the snout—that is, farther than in most males; in a male from Corunna it reaches the eye, whilst in a femalenon identical size and locality it reaches between the eye and the nostril. Bolkay’s way of expressing the length of the inner meta- tarsal tubercle as compared to the inner toe originates from me, with certain reservations, however t, but I have aban- doned it long ago, having found many specimens of the typical form in which the tubercle is not longer than its distance from the subarticular tubercle of the first toe, whilst, on the other hand, it may be as long in specimens of thie var. ridibunda, It has been pointed out by Bedriaga §, Wolterstorff, and Bolkay that the usually highly developed, shovel-shaped * Abh. Mus. Magdeb. 1. 1906, p. 140. + Herp. Japan, p. 97 (1907). t Proc. Zool. Soc. 1885, p. 668. § Wiss. Res. Przewalski Exped., Zool. iii, i. p. 1B (1899). Variation of the Edible Frog. 249 inner metatarsal tubercle of the var. chinensis is remarkable for a certain mobility, the distal part of its base being more less detached from the metatarsal of the inner toe with which it is connected by a web-like membrane. ‘This character is not only inconstant in this variety, as I was able to demonstrate to the second author by sending him for identification a cut-off foot of a specimen from Broughton Bay, Corea, which he returned to me named var. lessonee, but it is also found in some specimens of the latter (from Cambridgeshire and Norfolk) when the tubercle is very strongly developed, a fact also observed by Fejervary * in the case of his var. bo/kaya from Switzerland (=lessone). This character is correlative of the transformation of the tubercle into a fossorial organ, as already recognised by Wolterstorff, who fully admits the true state of things in the var. chinensis, from a diagnostic point of view, although unfortunately not acquainted with the amount of variation in the var. lessone. It has also been observed that the base of the tubercle of the var. chinensis does not run parallel to the axis of the longest toe, but is oblique to it; this is llowever only more or less so in the Chinese- Japanese frog, again in relation with the degree of development of the tubercle, and a similar dispos sition, varying in degree, is like- wise to be observed in the var. lessone. Although the metatarsal tubercle may be identical in the two varieties, I quite agree with Wolterstorff, and have always held the view that the var. chinensis cannot have been derived from the var. lessone, the two forms repre- senting independent extremes in the parallel evolution of the same adaptive character; but the var. /essone is there to illustrate the steps through which the cliaracter has been evolved out of a type such as the var. ridibunda, now so completely separated from the easternmost form of J. escu- lenta. Wolterstorff seems to look upon the typical form, or rather its hypothetical direct ancestor, as the origin of the races in question; Bolkay, in my opinion, is nearer the truth when he suggests R. ridibunda being nearer to ft. chinensis, but at the same time he inverts the drift of evolution in regarding the former as derived from thie latter, owing to theoretical considerations based on the now ex- ploded * preepollex ” and ‘ prehallux” theory. In his description of R. nigromaculata (chinensis), Stej- heger says the toes are about ? webbed. It is so in some eases, but rather the exception than the rule, and similar * Beitr. Herp. Rhonetal, p. 20 (1909). 250 Mr. G. A. Boulenger on the Races and exceptions occasionally occur in the typical form (Poitiers, Bologna) and in the var. ridibunda (Alemtejo, Majorca, Rahamna in Morocco). The outer metatarsals are separated nearly to the base in 2, esew/enta, only in their distal half in R. temporaria and FR. arvalis, the other Huropean species fillmg up the gap between the two extremes. Integument and Markings. The elongate glandules or interrupted longitudinal elan- dular folds on the back *, afford, generally speaking, a good distinctive character for the var. chinensis, but they may be very feebly marked or almost obsolete in some specimens (Kin Kiang, Yokohama), and they are occasionally fore- shadowed in the var. ridibunda (Beit Jenn, near Damascus), | so that the two formsare completely connected in this respect. I may point out another character, hitherto overlo>ked, which affords an absolutely constant distinction between the typical form and the var. chinensis. Sn ee aes este Posterior extremities of dorso-lateral folds in. specimens from Berlin, var. ridibunda (a), Cadillac, var. ridibunda (b), and Vienna, f. typica (Cc). Tie. 5. Tn the former, and also in the var. essone, the glandular dorso-lateral fold ends abruptly at some distance in front of the thigh, and it is often followed by a detached portion parallel with it but nearer to the mid-dorsal line and extend- ing on the base of the thigh. In the var. chinensis the fold extends uninterrupted to the hip, or, if broken up posteriorly, without any deviation from the straight line. Now, this striking difference is completely bridged over when we take the var. ridibunda,as well as the var. saharica, into con- sideration. Some specimens have the fold continuous and * A very variable feature in the American representative of FR: escu- lenta, R. halecina, L. Variation of the Edible Frog. 251 extending to the hip (fig. 5, a), others have a detached posterior part as in the typical form (c), whilst others again (b) con- nect the two conditions, the posterior part of the fold, though deviating, being confluent with the anterior and forming a bend before reaching the thigh. Bolkay mentions among the specific differences between R. esculenta, R. ridibunda, and R. chinensis, that the dorso- lateral fold is wider (as wide as the upper eyelid) in the second than in the two others. ‘his character is absolutely worthless, for in specimens of the typical form from France and Switzerland its width usually measures } to 2 that. of the upper eyelid, but may be equal to it (St. Malo, Havre, Basle, Zofingen), and in German and Austro-Hungarian specimens of the var. ridibunda $ to 2? that width is by no means unfrequent. The fold is always narrower than the upper eyelid in the vars, chinensis and lessone. In my previous descriptions of the var. ridibunda I have drawn attention to the fact that the dorso-lateral fold, though usually broader than in the other forms, is less prominent ; I should add that it is sometimes so flat that it cannot be traced without the use of a lens, when the pores with which it is studded indicate‘its course. It has not been pointed out however that these folds are rendered more inconspicuous still owing to the spots on the body being disposed quite irrespective of them, whilst in the typical form and the vars. lessone and chinensis they stand out on account of their lighter colour, hardly ever encroached upon by the spots, which may be arranged more or less in relation to them, especially when forming longitudinal bands. When a specinen of the var. ridibunda is seen at a short distance there is usually nothing to reveal the presence of the dorso- lateral folds, which strike the eye in the typical form and the vars. lessone and chinensis. These facts have a bearing on the question of the derivation of the forms which constitute the species A. esculenta, and confirm the view I have held ever since I took up the study of the subject that the var. rdibunda is the most primitive form, out of which the others have been evolved. In a paper recently published * on the derivation of characters in the genus ftana as a whole, the absence of the dorso-lateral fold is considered by me as the primitive condition, and the North American F. catesbiana, in which it is totally absent, is, for this and other reasons, regarded as nearest the hypo- thetical prototype among all the species of Hurasia and * Bull. Soc. Zool, France, 1918, p. 111. 252 Mr. G. A. Boulenger on the Races and America. Close to 2. catesbiana, there is another North American species, /?. septentrionalis, in my opinion derived trom it, in which the fold is either present or absent, according to individuals, but when present is short and very flat, with the spots and marblings irregularly distributed over the body. Such a type leads to the state of things in &. esculenta, var, ridibunda. Bolkay alludes to the transverse expansion of the dark spots on the back as an important character of . chinensis, but such transverse markings are by no means the rule in this variety, some specimens of which are, on the contrary, longitudinally streaked, as is often the case in the typical form and the var. lessone, but never in the vars. rédi- bunda and saharica. I may here mention that specimens with the black markings forming cross-bars on the back are exceptionally met with, not only in the var. ridibunda, but also in the typical form (females from Rivoli and Verona). ~ . The light vertebral streak or band is very frequent in the typical form and the vars. /essonee and chinensis, less 8) in the var. ridibunda, in which it is generally broader, and usually absent in the var. saharica. I do not think this light vertebral streak, which occurs in so many species, is to be looked upon as a primitive character ; the frequent cases of deviation of its course from the straight line (most strongly marked in specimens from Calcinaro and Cadillac) suggest a different interpretation, and, in the present state of our knowledge, its signification is highly problematic, as 1s that of a light line along the inner side of the upper surface of the leg which, in many Oriental and African species, often accompanies the vertebral streak, and exceptionally occurs in R. esculenta, var. chinensis (3 from Japan). Both streaks are absent in all American species, with the single excep-. tion of R. cantabrigensis, Baird, the representative of the European R. arvalis, Nilss. The Skull. The osteological characters appealed to by Bolkay are evidently derived from an examination of a very smail number of specimens; put to the test of a larger material they prove to be worthless for defining species. I am especially surprised at his statements concerning the nasal and fronto-parietal bones. Although usually in con- tact with each other in full-grown specimens of the typical form, as described and figured by Ecker, Fatio, and others, Variation of the Edible Frog. 253 the nasal bones are not always so; there are frequent ex- ceptions, as my own description implies *, but such exceptions occur as well in the var. ridibunda, even in large specimens (¢? from Vienna, 90 mm. from snout to vent, 9 from Prague, 120 mm.), and I have come across adults of the var. chinensis in which the nasal bones are completely separated from each other, as is usually the case in immature or small specimens of all the forms. As to the presence or absence of the anterior notch between the frontoparietals, tis is a mere individual peculiarity, usually dependent on the size of the specimen; yet I wish to draw attention to the figures given by Camerano + of small specimens of the var. lessone from Italy in which the anterior extremity of the frontoparietals answers to Bolkay’s definition of 2. chinensis. The Size. I append the measurements, from snout to vent, of the largest specimens of the different forms in the British Museum. According to Werner, the var. ridébunda may reach a length of nearly 150 mm, in Austria. Var, ridibunda ......s0+0%. 95 mm, 125 mm. Ware SaRarvem™® fin: sac ee 58 80 ICU! PAP SEI al 78 95 Warsilessoe@ sts chien tee ae 64 78 Vinita CH C10C71S28n sear eno 70 85 The Tadpole. I have examined large series of tadpoles of the vars. ridibunda and chinensis without succeeding in finding any characters by which to distinguish them from those of the typical form. The characters pointed out by Annandale § for the var. chinensis are not confirmed as regards the mouth- disc, the position of the spiraculum, or the length of the tail compared to that of the body. Conclusions, When dealing with polymorphic species, botanists often distinguish between forms (species, some term them) of first, second, third, and fourth rank, Applying this concept to Rana esculenta, the typical form representing of course the * Taill. Batr. Eur. p. 279. + Mem. Ace. Torin. (2) xxxy. 1883, p. 60. } Rept. Amph, Oesterr.-Ung. p. 88 (1897). § Mem. As, Soc. Beng. vi. 1917, p. 147. 254 Mr. G. A. Boulenger on the Races and first. grade, we may place the var. chinensis in the second, the var. ridihunda in the third, and the vars. saharica and dessone in the fourth. Looking at things from a practical standpoint, we must regard the var. sakarica as but a slight modification, a geographical race distinguishable from its nearest neighbour but impossible to define if specimens from the whole range of distribution of the var. ridibundu are taken into consideration. ‘The typical form is completely connected with the var. ridibunda, and where tle two co-exist in a locality, annectant individuals may be regarded as the result of crossing, such as undoubtedly must take place ; but this explanation fails when we have to deal with specimens from France, 8.E. Europe, aid Asia, where individuals of uncertain identification like- wise occur, although the discrimination of the two forms is in most cases quite easy. It is no longer so when we come to the typical form compared with the var. lesson, and in this case the naming of certain specimens 1s_ perfectly arbitrary, as those who have had to deal with a considerable material from places where the two forms co-exist fully admit *; yet, the extreme, what some would call the “ pure lessone,” such as it occurred in the Cambridge fens and is still found in a very few localities in Norfolk, is well entitled to varietal rank, its structural characters being fixed and so considerable in degree when compared with the typical form that it would undoubtedly be looked upon by many as a species were we not acquainted with the annectant examples from the Continent. ‘hese extreme specimens of the var. lessone represent the terminus of an uninterrupted series starting from the var. ridibunda and passing through what is called the typical form. Another terminus form, in which the principal characters of the var. lessone are repeated, is the var. chinensis, which in all probability is also derived from the var. ridibunda, but the connecting-links of which have disappeared or are still unknown. If we appeal to the existence of a hiatus between forms as a sole criterion for deciding on what is a species, then J. chinensis is entitled to stand as such ; however, considering the many points of agreement, and preferring to keep an eye on resemblances rather than on differences, the rank of variety or subspecies appears to me the more appropriate for this form, as it did to Lataste many years agot. Owing to the state of things in the * Of. Wolterstorff, Schr. Nat. Ges. Danzig, (2) xi. 1904, p. 46. + Bull. Soc. Zool. France, 1880, p. 61. ‘ Cette forme, que quelques auteurs regardent comme une espéce distincte, d’autres comme une Variation of the Edible Frog. 255 Kuropean forms, the course I have follewed is surely the better from a philosophical point of view, whilst the use of a varietal designation precludes all fear of the distinction being overlooked. The following diagram expresses the relationship between the five forms, as I conceive them :— Var. lessone. Var. chinensis. T°, typica. Var, saharica. | rks | Pa Var, ridibunda. + We cannot yet apply the test of crossing experiments in justification of the subordinate position assigned to R. chinensis, as Pfliiger was able. to do in the case of KR, esculenta and its var, ridibunda, but another physiological argument has been put forward by Wolterstorff: the large and often sharp-edged metatarsal tubercle of 2. chinensis is an adaptation to burrowing habits unlike those of A. esculenta. We are told that Dr. Kreyenberg observed the Chinese frog to dig and retire deep into the ground of dried-up rice-fields, and this habit is regarded as an important ethological differentiation from its Kuropean representatives. Curiously, however, Féjervary very shortly after redescribed the var. lesson under the name of var. bolkayi, from specimens living in marshes at the mouth of the Rhéne in Switzerland, and observed the behaviour of this frog on land to be different from that of the typical /?. esculenta, the large and somewhat movable metatarsal tubercle being used to burrow in the ground after the manner of Pelobates. It is interesting to note, in this connection, that Wolterstorff, who (1906) seemed to attach so great an importance to this peculiarity in the case of the Chinese frog, had (1904) only reluctantly recognised J’, lessona’s rank as a variety, a term which for him expresses mere individual variations, such as his colour- simple variété de ana esculenta, L., a des caractéres propres et con- stants qui lui méritent une description particuliére et un nom spécial ; elle me semble cependant assez voisine de Rana esculenta pour que je crois utile de ne l’en point séparer spécifiquement ..... Et si, aprés quelques hésitations, je me décide a classer [R. chinensis] comme sous- espéce de Lt. esculenta, @est par cette seule considération qu’elle me parait beaucoup plus yoisine de cette derniére que de toutes les autres vrenouilles,” 256 On the Races and Variation of the Edible Frog. varieties striata and nigromaculata in FR. arvalis *, and refused to admit it as a subspecies. These observations on the fossorial habits of the vars. chinensis and lessone should be borne in mind by those who appeal to the behaviour of the Indian 2. erassa, compared to that of the typical 2. t/gr/na, as an argument in favour of its specific distinction +. These supposed species offer a perfect parallel to the lines of evolution which can be traced in R. esculenta, as I have recently pointed out ¢. Although we must not expect to find among the species of the present day the actual types out of which their allies have been evolved, yet I think it legitimate speculation to look upon certain species, or certain small groups of species, as a sufficiently near approximation to help us towards an elucidation of the phylogenetic relationships, the expression of which should be the aim of taxonomy. In this sense, and with this reservation, I consider R. catesbiana, Shaw, aud AR. grylio, Stejn., as representing the most primitive forms of America and Eurasia; the species that cluster round them, R. septentrionalis, Baird, R. clamitans, Daud., R. onca, Cope, R. virgatipes, Cope, R. montezume, Baird, would be derived from the same stock ; they constitute a distinct section, waich is perfectly natural, though not susceptible of a very strict definition. From this section we may imagine the one of which ft. esculenta is the type to have been derived, and there is Jittle doubt in my mind that the Chinese &. plancy?, Lataste, is a connecting form, nearly allied to, but in most respects less modified than, R. esculenta, both having been evolved out of the same ancestor, possibly related to the Oligocene- Miocene R. meriant, H. von Mey. ‘The chief distinctive features of AR. esculenta compared to R. plancy? reside in a reduction of the nasal bones, the more obtuse fingers, and the very peculiar external vocal sacs. By what steps this last * Such modifications represent varieties only in the sense taken by horticulturists, and should not be given names in scientific nomenclature. Eliminating these cases, I apply the term varietas to every division of the system subordinate to the species, without any further consideration of hierarchy, in order to avoid complicating nomenclature by the use of tri-, quadri-, or even quinquenomials. In so doing, I simply adhere to the Linnean method which has so long been followed, and is still used by most of the botanists for whose work I have the greatest respect. “ Les variétés des systématistes sérieux sont les espéces de M. Jordan, au moins du Jordan des Observationes et du Pugillus..... Le mot variété employé par les botanistes pour désigner une race sauvage laisse peut-étre & désirer, mars a jowt de la priorité.” J. Briguet, Questions de Nomenclature, Bull. Herb. Boissier, 11. 1894, p. 84. +t Annandale, Rec. Ind. Mus, xv. 1918, p. 63. t Ree. Ind. Mus. xv. 1918, p. 51. On the Rhynchotal Family Ly geide. 257 character was reached, FR. halecina, L., is there to show us, for within the limits of this highly variable species, the vocal sacs may be said to be still in process of evolution ; situated behind the commissure of the jaws, as in R. esculenta, R. monte- zume, R. areolata, B. & G., and R, capito, Leconte, but unlike those of all other frogs, they are either internal or external, showing every degree of development, and when external they form folds which, in certain individuals, have a tendency towards the invagination characteristic of the sacs in RR. esculenta. We may well assume the direct ancestors of A. esculenta to have passed through such stages in the course of parallel evolution. XXVITI.— Contributions to a further Knowledge of the Rhynchotal Family Lygeide. By W. L. Distanr. [Continued from p. 179.] Lygeus degeni, sp. n. Head, pronotum, scutellum, corium, and body beneath griseo-fuscous, two small central spots on pronotum, two larger spots on clavus, and two still larger spots on corium— one on each side of claval apex—black; basal third of lateral margin to corium, connexivum beneath, and legs pale testaceous or ochraceous ; membrane pale fuscous, narrow base, lateral margins, and an irregular discal, transverse, angulated spot greyish white; antennz ochraceous, the apical joint fuscous, second joint a little longest, third and fourth joints subequal in length; pronotum and scutellum centrally longitudinally carinate ; the upper surface is more or less finely and obscurely very shortly pilose. Long. 8 mm, Hab. Abyssinia ; Taddecha, Mullka (Degen). E:XoPaMERA, gen. nov. Head robust, about as long as broad; eyes projecting beyond the anterior angle of pronotum but not reaching its anterior margin ; autennee with the basal joint stoutest and considerably passing apex of head, second joint longest ; rostrum with the basal joint not quite reaching base of head, its apex scarcely passing the anterior cox ; pronotum Ann. & Mag. N. Hist. Ser. 9. Vol. ii. 20 258 Mr. W. L. Distant on the elongate but very little longer than broad at base, lateral margins narrowly laminately carinate, anterior collar very narrow, subobsolete, anterior much longer than posterior lobe, convex, its lateral margins rounded, lateral margins of the posterior lobe obliquely straight ; scutellum longer than broad, subtriangular ; anterior femora thickened, spined beneath, anterior tibie distinctly curved, their apices dilated and inwardly a little angulate, intermediate and posterior tibiz moderately spinulose. . Type, E. ethiopica, Dist. Allied to Pseudopamera, Dist., from Central America. Exopamera ethiopica, sp. u. Head black, moderately shortly palely pilose; eyes darker black ; ocelli purplish red ; pronotum ochraceous, the lateral and anterior margins and the posterior lobe paler in hue, punctate, especially the posterior lobe, the basal margin and lateral basal angles more or less shining black ; scutellum ochraceous, basal and apical areas black, more or less coarsely punctate ; corlum ochraceous; clavus more or less closely blackly punctate, two prominent spots before claval area, an irregular transverse subapical spot, and the apical angle shining black, the whole corium more or less coarsely punctate ; membrane black, its apical margin ‘pale fuligi- nous ; head and abdomen beneath opaque black ; sternum shining black and coarsely punctate ; coxze, trochanters and legs, narrow lateral sternal margins and posterior sternal segmental margins ochraceous ; rostrum ochraceous ; mem- brane moderately passing the abdominal apex; antennz with the first, second, and- third joimts ochraceous, their apices black, fourth joint greyish white, it apical half black, second joint longest, third and fourth joints almost subequal in length. Long. 9-10 mm. Hab. Brit. E. Africa; Kibwesi (S. 4. Neave). Exopamera mirabilis. Aphanus mirabilis, Dist. Ann, Mag. Nat. Hist. (7) xii. p. 471 (1903). Hab. Fernando Po. ALBANYARIA, gen. nov. Body elongate ; head subtriangular, apical area distinctly narrowed and apex of central lobe distinctly prominent ; Rhynchotal Family Ly geide. 259 eyes moderately prominent and slightly passing the anterior angles of the pronotum; antenne moderately robust, basal joint only a little passing apex of head; rostrum with the basal joint almost reaching base of head ; pronotum a little longer than broad at base, and transversely constricted near base ; corium extending only to about three-fourths of the abdomen ; membrane absent; anterior femora incrassated and finely spined beneath; anterior tibiz a little curved but not centrally spined; scutellum elongate, longer than broad, Allied to Fontejus, Stal, but the pronotum much shorter, anterior tibize not centrally spined, &c. Albanyaria multicolorata, sp. n. Head, anterior lobe of pronotum, and the seutellum black ; the narrow posterior pronotal lobe and the extreme apex of scutellum greyish white ; antenne ochraceous, apex of third joint and more than apical half of fourth black ; corium ochraceous, the lateral marginal areas with three prominent black spots, the smaller near base, the largest near middle, and the third at apex, the exposed apical area of the abdo- men black ; body beneath: black ; posterior sternal segmental margins very pale ochraceous; legs reddish ochraceous, apical halves of the anterior femora and apices of the tibiz and tarsi black; antennz with the second joint slightly longer than the third and about subequal with the fourth ; scutellum more or less rugosely punctate ; clavus linearly somewhat coarsely punctate; rostrum ochraceous, the basal joint black, remaining joints imperfectly seen in carded type. Long. 54 mm. Hab, W. Australia; Albany (J. J. Walker). Genus Laryneopvs. Laryngodus, Herr.-Scheeff. Wanz. Ins. ix. pp. 191, 212 (1853). The short description given by Herrich-Scheffer and some imperfections in the figure given of the type of the genus render a fuller description of both necessary. Laryngodus australia, Herr.-Scheff. Wanz. Ins. ix. p. 212, fig. 967 (1853). Head fuscous brown; eyes black ; antenne dark casta- neous, apices of the first, second, and third joimts very , 20* 960 Mr. W. L. Distant on the narrowly black, fourth joint ochraceous with nearly apical third black; head and anterior lobe of pronotum fuscous brown, posterior pronotal lobe black, with two central spots and the lateral margins creamy white or very pale ochra- ceous; eyes black ; scutellum fuscous brown; corium dull ochraceous, darkly punctate, inner area of clavus more densely darkly punctate, disk of corium with an oblique longitudinal fascia—neither reaching base nor apex, a sub- central, transverse, very irregular Fascia, and the apical angle, black ; membrane fuscous, the veins, and some irregular suffusions and spots, pale dull ochraceous ; body beneath and legs dark castaneous, basal spine to antenne beneath, anterior margin of prosternum, cox, posterior angles of meso- and metasterna, and the greater part of basal joints of intermediate and poster ior tarsi pale ochraceous. Head elongate, longer than basal breadth including eyes, narrowed on apical area, and with a short spine at base of antenne ; eyes prominent, almost reaching base of head ; antenne with the basal joint shortest and. stoutest, second joint a little longer than third, which is again longer than fourth; pronotum punctate, with anterior lobe rane nar- rower, more globose, and about twice as long as the posterior lobe which is more strongly punctate, anterior lobe with a central, longitudinal, fasciate, flat impression; scutellum about as broad as long, subtriangular, thickly punctate, extreme apex ochraceous ; corium broadened on apical area ; membrane with the venation very prominent; anterior femora strongly thickened, narrowed at base and apex, distinctly spined beneath ; anterior tibiz flattened and sub- spinosely dilated at apices; rostrum about reaching the intermediate coxze. . Long. 10 mm. Hab. 8.W. Australia; Yallingup (R. EL. Turner). Bosbequius australis, sp. n. Head, anterior area of pronotum, scutellum and sternum, black or blackish; anterior margin and posterior area of pronotum and corium brownish ochraceous ; lateral pronotal margins and a spot near inner angle of apical margin to corium very pale luteous; abdomen beneath brownish ochraceous ; femora castaneous, their apices and the tibize and tarsi ochraceous; antennze dull ochraceous, second joint longest, first joint slightly passing apex of head*; head * In the typical Oriental species the basal antennal joint did not reach the apex of head. thynchotal Family Lygeide. 261 (including eyes) narrower than anterior margin of pronotum ; first joint of rostrum extending beyond base of head ; posterior area of pronotum, scutellum and corium coarsely punctate ; anterior femora strongly incrassate. Long. 8 min. Hab, Australia; Adelaide River (J. J. Walker). The type of the genus Bosbequius was from Tenasserim (Faun. Brit. Ind., Rhynch. ii. p. 65, fig. 1). Thebanus nigrinus, sp. 0. Dull ochraceous; head, anterior lobe of pronotum, and anterior area of scutellum black ; head beneath and sternum black ; legs ochraceous ; abdomen beneath dark slaty-grey ; antennee ochraceous, second joint a little longest, third and fourth almost subequal in length ; pronotum thickly, some- what coarsely punctate; scutellum punctate, black before the anterior branches of the cruciform carination, and dull ochraceous behind them ; posterior margin of pronotum concave before scutellum ; corium (excluding lateral mar- ginal areas) darkly punctate ; membrane slaty-grey, slightly passing the abdominal apex. Long. 35 mm. : _Hab. Burma; Karennee. Genus LacHNOPHOROIDES. Lachnophoroides, Dist. ‘ Nova Caledonia,’ Zool. i. p. 881 (1914). Type, L. ornatipennis, from New Caledonia (ibid. pl. xi. fig. 9 2). I am now able to amplify the description of this genus by sexual characters, having only seen a single 2 specimen previously. S. Pronotum distinctly longer than breadth at base ; anterior tibiz strongly sinuately curved and armed with a short robust spine near middle of under surface. 2. Pronotum about as long as broad at base; anterior tibiz unarmed. Lachnophoroides crudelis. Pachymerus crudelis, Hag]. Ofv. Vet.-Akad. Forh, 1895, p. 462. Hab. W. Africa; Gaboon (fide Haglund). Lagos; Ondo (A. B. S. Powell). N.E. Rhodesia; Upper Luangwa R. (S. A. Neave). Uganda Protect. between Junja and Busia, E. Busoga (S. A. Neave). Abyssinia (Lake Rudolph Exped.—Ph. C. Zaphiro). 262 Mr. W. L. Distant on the Lachnophoroides rudolfianus, sp. n. Head and anterior lobe of pronotum dull, dark ochraceous, posterior pronotal lobe paler ochraceous with darker punc- tures in somewhat transverse series and with three central longitudinal darker series, lateral margins broadly and basal margin narrowly pale ochraceous, a black spot near each basal angle ; scutellum ochraceous, darkly punctate, a large castaneous spot at base and two linear black spots on apical area; corium very pale ochraceous, more or less darkly punctate, clavus with a black and greyish spot at base, beyond middle of corium a broad transverse dark castaneous fascia and the apical margin narrowly and irregularly of the same colour; membrane pale shining ochraceous; head beneath and sternum piceous ; abdomen beneath dull dark testaceous ; legs and rostrum ochraceous; antenne ochra- ceous, apices of the second and third joints and apical half of the fourth black, second joint a little longest, third and fourth joints subequal in length; anterior femora robust, strongly spined beneath near apex, anterior tibiz.in ¢ strongly curved, and with a prominent spine beneath near middle. Long. 8 mm. Hab. Soudan; Kaig (Lake Rudolph Exped.—C. Singer). Aphanus littoralis, sp. n. Head black or very dark castaneous, eyes griseo-fuscous ; antenne dull ochraceous, apices of first, second, and third joints more or less fuscous, fourth joint dark fuscous with a broad subbasal greyish annulation; pronotum ochraceous, prominently brownly punctate, the lateral margins almost impunctate, anterior half (excluding margins) dark easta- neous and almost impunctate, with a small central pale ochraceous spot at anterior margin; scutellum ochraceous, prominently brownly punctate, the basal area black; corium ochraceous, rather finely brownly punctate, extreme lateral margins almost impunctate ; membrane brownish ochra- ceous with somewhat paler mottlings ; body beneath casta- neous, the lateral margins, posterior sternal segmental margins, rostrum, and legs ochraceous, the lateral abdominal margin with large castaneous spots; second, third, and fourth, jomts of antenne gradually decreasing in length, the second a little longest; first joint of rostrum about reaching base of head; membrane about reaching abdo- minaLapex ; pronotum with a more or less distinct central Rhynchotal Family Ly geide. 263 longitudinal narrow carination ; scutellum a little foveately depressed at base. Long. 85-10 mm. Hab. Blue Nile (#. S. Crespin), nr. mouth of Dinder R. and Roseires (S. S. Flower). N.W. shore of L. Nyasa, from Florence Bay to Karonga (S. A. Neave). Aphanus ferrugineus, sp. n. Head black; antennz with the basal joint black, second and third joints ferruginous ; pronotum pale ferruginous, coarsely darkly punctate, the anterior area (excluding margins) black ; scutellum black, coarsely darkly punctate, becoming paler and more ferruginous on apical area, and with an ochraceous spot on each lateral margin near base ; corium brownish ochraceous, darkly punctate, with two small obscure black spots in oblique series on apical half, the lateral margins narrowly impunctate; body beneath black, the posterior sternal seginental margins, rostrum, and legs ferruginous; second joint of antenne considerably longer than third; apex of central lobe of head distinctly prominent ; in some specimens the femora are distinctly darker—almost black —than the tibiz ; basal joint of rostrum passing base of head ; membrane, a little paler than corium, reaching abdominal apex. Long. 8-85 mm. Hab. Nyasaland (Cotterell) ; W. shore of L. Nyasa between Domira Bay and Kotakota (S. A. Neave). N.i. Rhodesia ; Mid-Luangwa Valley (S. A. Neave). Aphanus apicalis. Rhyparochromus apicalis, Dall. List Hem. ii. p. 562 (1852). Rhyparochromus turgidifemur, Stal, Ofv. Vet.-Ak. Forh. 1855, p. 32. 1. Rhyparochromus nigromaculatus, Stal, Ofv. Vet.-Ak. Foérh. 18565, p- 32. 2. Beosus apicalis, Stal, Hem. Afr. ii. p. 165 (1865). Aphanus erosus, Dist. Ann. & Mag. Nat. Hist. (7) viii. p. 501 (1901), In describing my A. erosus I wrote, “ Allied to A. apicalis, Dall., differing by the exceedingly coarse punctuation on the lateral margins of the pronotum and corium, &c.” Compared with the type of Dallas, that held good at the time of writing, but since then a large number of species have reached the British Museum, and intermediate varieties occur. Hab. 8. Africa (Brit. Mus.). Ovampo L. (Eriksson). 264 Mr. W. L. Distant on the Transvaal; Pretoria (Distant), Lydenburg (Krantz). N.E. Rhodesia ; Mid-Luangwa Valley (Neave). Blue Nile; Roseires (Flower). Congo (Richardson). Aphanus albigera, sp. n. Head and anterior area of pronotum black, posterior pronotal area ochraceous, brownly punctate, and at the lateral marginal junctions of these two areas a somewhat large pale ochraceous spot; scutellum black; corium ochraceous, thickly darkly punctate, extreme lateral margin impunctate, inner claval margin for about half its length from base pale ochraceous and impunctate, from thence to . apex very thickly blackly punctate, a short elongate black line near outer claval margin, followed by a large black spot near and outside claval apex, the apical margin of corium narrowly black, the two last markings separated by a small pale impunctate spot; membrane brownish with the venation somewhat paler in hue; body beneath, rostrum, and legs black, coxal spots and narrow, irregular posterior margins to sternal segments pale ochraceous; antennze with the third joint shortest, second and fourth subequal in length. Long. 6-63 mm. Hab. South Africa; Grahamstown. Natal; Durban (F. Muir). , Allied to A. apicalis, Dall., but a smaller and narrower species, maikings of the pronotum and short third joint of antenne different. Aphanus nigrellus, sp. n. Head, antenne, pronotum, and scutellum black, lateral pronotal margins ochraceous ; corium dull ochraceous, two short claval lines and the apical area black, the latter con- taining a prominent, central, transverse greyish-white spot, and the extreme apical angle also of that colour ; membrane griseo-fuscous, with an apical white spot; body beneath, rostrum, and legs black; antennz somewhat robust, third joint a little shorter than second or fourth joints ; pronotal lateral margins distinctly, somewhat longly pilose. Long. 6 mm. Hab. Nyasaland; between Ft. Mangoche and Chikala Boma (S. A. Neave). Allied to both A. apicalis, Dall. and the preceding species here described—A. albigera, but differing by the colour of the pronotum and its longly pilose lateral margins, Xe. Rhynchotal Family Ly geide. 265 MAXAPHANUS, gen. nov. Allied to Aphanus, Lap., from which it differs by the longer and more elongate body; the longer and more robust basal joint of the antenne, which is as long as the head and projects considerably beyond its apex; anterior femora shortly spined beneath, with a long and very distinct spine before apex, anterior tibiz also shortly spined beneath beyond base. Maxaphanus africanus, sp. n. Dark castaneous, in some specimens almost piceous ; lateral margins (excluding basal areas) of pronotum and sometimes a small central spot to same, corium with about basal half of lateral margin, a small lateral spot beyond it and nearer apex, a small discal spot outside the apical claval area and a minute ‘spot before posterior margin, extreme apex of scutellum, rostrum and legs, ochraceous ; apical areas of femora and the tibiz and tarsi darker and more brunescent ; antenne dark castaneous, fourth joint (excluding apical area) pale ochraceous, second and third joints almost subequal in length and longest, fourth longer than first which considerably passes the apex of head; pronotum distinctly, broadly, transversely impressed near middle, the anterior area smooth, the posterior area finely wrinkled, lateral margins distinctly laminate ;-corium dis- tinctly punctate ; membrane pitchy-brown, the veins pro- minent, the two inner veins strongly curved at base. Long. 13-14 mm, Hab. Nyasaland ; Mlanje (S. A. Neave). N.E. Rhodesia; Upper Luangwa R. (S. A. Neave). Uganda; Tero Forest (C. C. Gowdey), Entebbe (C. A. Wiggins). Metochus holsti, sp. n. Head, anterior lobe of pronotum, and scutellum black, posterior pronotal lobe piceous, darkly punctate, and with a pale central longitudinal line ; corium ochraceous, clavus, a broad irregular transverse fascia connecting apex of clavus with lateral margin, and the apical margin black, the anterior area between the clavus and lateral margin is ochraceous, brownly punctate, the area between the trans- verse fascia and apex creamy-white; membrane fuscous with obscure paler mottlings; head beneath and sternum black ; abdomen beneath dark castaneous, with some lateral 266 Mr. W. L. Distant on the marginal ochraceous macular markings ; rostrum ochraceous, basal and apical joints piceous; femora black, their bases and the whole of the tibiz and tarsi more or less ochraceous ; antennz piceous, basal half of apical joint ochraceous, second joint longest, third and fourth almost subequal in length ; anterior femora robust, shortly spinose beneath. Long. 10 mm. Hab. Japanese Archipelago ; Tsushima Island (P. Holst). Dieuches \relatus, Dist. Ann. & Mag. Nat. Hist. (7) viii. p- 505 (1901). Hab. Mashonaland; Umfili River (G. A. K. Marshall). Nyasaland; Valley of N. Rukuru, Karonga District (S. 4. Neave). Uganda; Entebbe (C. C. Gowdey). Abyssinia ; Gibe River (Ph. C. Zaphiro). The type was from Mashonaland. Dieuches parvipicius, sp. n. Head, pronotum, and scutellum black ; anterior half of lateral margins and some small spots (usnally two but some- times four) on disk of pronotum, two spots near base and extreme apex of scutellum ochraceous ; antennee ochraceous, apex of third joint black, more or less mutilated in the twelve specimens now before me; corium ochraceous, brownly punctate, extreme lateral margins pale and im- punctate, a spot at base of clavus, a large spot near inner posterior angle, a very small spot in a line with it on lateral margin, and the apical margin black; body beneath black : rostrum and legs ochraceous, apex of rostrum and usually apical areas of the femora—more or less——black ; antennz with the second and third joints almost subequal in length ; scutellum with a more or less distinct, central, longitudinal carinate line. Long. 7-8 mm. Hab. Katanga; Kambove and Luffra River (S. A. Neave). Allied to D. patruelis, Stal, but a smaller species with both the pronotal lobes black. Dieuches consimilis, sp. n. Allied to the preceding species in general markings and coloration, but a larger species with the basal joint and apices of the remaining antennal joints black; posterior pronotal lobe more strongly and coarsely punctate ; scutellum Rhynchotal Family Lygeide. 267 without the central carinate longitudinal line which is always more or less pronounced in D. parvipicius. Long. 9-10 mm. Hab. Uganda; Entebbe (C. C. Gowdey). Katanga; Kambove (S. A. Neave). Abyssinia (C. Singer). Dieuches smitha, sp. n. Head and anterior lobe of pronotum testaceous, posterior pronotal lobe ochraceous, thickly darkly punctate, lateral pronotal margins pale, impunctate; scutellum testaceous, extreme apex pale ochraceous ; corium dark ochraceous or brownish ochraceous, lateral margins and a large irregular spot before apex pale ochraceous; membrane brownish ochraceous; body beneath testaceous ; lateral margins of sternum, posterior margin of metasternum, and lateral abdominal margins ochraceous; rostrum and legs ochra- ceous, apical areas of femora and apices of the tibiz piceous ; antennze ochraceous, the basal joint and apices of remaining joints dark testaceous or piceous, second and fourth joints longest and subequal in length; pronotum with a central longitudinal carinate line on posterior lobe ; first joint of rostrum about reaching base of head ; membrane not quite reaching abdominal apex in @, distinctly shorter in 2. Long. 10-11 mm. Hab. 8. Africa (Dr. Smith’s Coll.). Graham’s Town (fF. Pym). Allied to D. umbrifer, Stal. Dieuches sloggetti, sp. n. Black ; lateral margins of pronotum and corium, second joint and base of third joint of antenue (fourth joint muti- lated), tibiz and tarsi stramineous or pale ochraceous ; second joint of antennz much longer than third ; pronotum somewhat narrow and elongate, posterior lobe thickly punc- tate ; corium and clavus more or less thickly punctate; first joint of rostrum about reaching base of head. Long. 9 mm. Hab, $3. Africa; Deelfontein (Col. Sloggett). MeETADIEUCHES, gen. nov. Head robust, about as long as breadth between eyes, which almost reach anterior margin of pronotum or are not far removed from same, in front of eyes laterally strongly obliquely sinuate, the apex of the central lobe prominent ; 268 Mr. W. L. Distant on the antennz with the basal joint moderately stoutest, slightly apically curved, shorter than second joint which again is a little shorter than third, fourth almost subequal in length to first ; rostrum reaching the anterior coxe; pronotum elongate, longer than breadth at base, lateral margins of anterior lobe slightly oblique, those of the posterior lobe more prominently oblique, the posterior angles subnodulose, basal margin almost truncate, very slightly concave, anterior margin truncate ; scutellum moderately long and slender, slightly longer than broad at base, lateral margins straightly oblique; legs elongate, anterior femora finely spined beneath, anterior tibize slightly dilated at apex; membrane passing abdominal apex. Type, M. dispar, Hag). Metadieuches dispar. Dieuches dispar, Hagl. Ofv. Vet.-Akad. Forh. 1895, p. 460. Hab. Gaboon (Sjéstedt). Cameroons (/scalera). Uganda; Entebbe (Dr. C. A. Wiggins and C.C. Gowdey), Mwera, Kyanja, Mabira Forest, Katanga River (C. C. Gowdey), shores of L. Isolt or Wamala, 3800 ft., and S. of L. George (S. A. Neave). Poeantius variegatus, sp. n. Head and anterior lobe of pronotum black ; posterior lobe of pronotum dark castaneous and coarsely punctate, the anterior and posterior lobes separated by a transverse ochra- ceous fascia; scutellum black ; corium ochraceous, a longi- tudinal fascia in clavus, and nearly the apical half of corium black, the latter containing a narrow transverse pale ochra- ceous fascia a little beyond its middle; membrane dull greyish; head beneath and rostrum dull, dark castaneous, posterior margin of metasternum more or less greyish white ; abdomen beneath black; legs black, apices of anterior and intermediate femora and the anterior tibiz ochraceous ; (posterior legs mutilated in type) ; antenne with the basal joint ochraceous, second and third joints black, second a little longer than third (fourth joint mutilated in type) ; head deflected, immersed to eyes, a little longer than broad ; pronotum with a central longitudinal, ill-defined carinate line; scutellum a little longer than broad ; rostrum about reaching the intermediate cox. Long. 63 mm. Hab. Gaza Land; near Chirinda Forest (G. A. K. Mar- shall). Rhynchotal Family Lygeeide. 269 Letheus longirostris. Letheus longirostris, Reut. Ent. Tidskr. viii. p. 102 (1887). Hab. Madagascar (fide Reut.). Rodriguez ((Gullion). Natal (Bell-Marley). N.E. Rhodesia; Lower Luangwa River, near Petauke, N.W. shore of L. Nyasa (S. A. Neave). This species is variable in size; specimens now before me in length range between 9 and 12 mm. Letheus descriptus. Rhyparochromus descriptus, Walk. Cat. Het. v. p. 103 (1872). Rhyparochromus alienus, Walk. tom. cit. p. 105. Letheus signatus, Dist. Ann. & Mag, Nat. Hist. (7) viii. p. 506 (1901). Letheus descriptus, Dist. Faun. Brit. Ind., Rhynch. ii. p. 89 (1904). Hab. N. India. Ceylon. ‘Tenasserim. North Borneo. Sula Island. Natal; Durban (Bell-Mariey). N.E. Rho- desia; Upper Luangwa River (S. A. Neave). We are now able to record the distribution of this species (previously only known from the Indian and Malayan regions) to the southern Ethiopian habitats of Natal and Rhodesia. Bergroth (Phil. Journ. Sci. xiii. p. 95 (1918)) has devoted nearly three large octavo pages to the description of a species from the Philippine Islands (LZ. robustus) which is apparently to be separated by the longer rostrum, ‘‘ reaching middle of third ventral segment.” In deseriptus the rostrum only extends to about the posterior coxe as described by Walker. Genus ABANUS. Abanus, Dist. Faun. Brit. Ind., Rhynch. v. p. 81 (1910). In describing the type of this genus from specimens received from Bengal, I wrote “pronotum elongate, about as long as broad at base.” This character from an examina- tion of a series of specimens of another species received ‘from tropical Africa appears to be of a sexual (female) character only, while in the male the pronotum is consider- ably longer than broad at base. Abanus ugandensis, sp. nu. Head and anterior lobe of pronotum black, basal area of pronotum brownish ochraceous, blackly punctate, and with a central ill-defined pale lJevigate longitudinal line, lateral pronotal margins pale ochraceous ; scutellum black, punctate, elongate, with two small discal spots and the extreme apex 270 On the Rhynchotal Family Lygeide. ochraceous ; corium very obscure ochraceous, thickly black punctate, the lateral margins pale ochraceous, apical margin more distinctly black; membrane dark fuligimous, some- times with small ochraceous suffusions; apical area of abdomen above—as seen beyond membrane—black, the apical margin dull ochraceous ; body beneath black, narrow lateral sternal and abdominal margins, very narrow posterior margin of prosternum, coxal margins, rostrum and legs, ochraceous; antennze ochraceous, extreme apices of first and second joints, apical third of third joint, and fourth joint, excluding broad basal annulation, black, first joint passing apex of head, second longest, third and fourth sub- equal in length; rostrum reaching the intermediate coxe, first joint about reaching or slightly passing base of head ; prosternum thickly, coarsely punctate at base. Long. 9-10 mm. Hab. Uganda; Mabira Forest, Chagwe, Tero Forest (C. C. Gowdey), Entebbe (C. A. Wiggins), Mpumu (Miss M. Robertson). Katanga (S. A. Neave). Genus GoNATAS. Gonatas, Dist. Biol. Centr.-Amer., Rhynch. i. p. 219 (1882); Faun. Brit. Ind., Rhynch. ii. p. 89 (1904). This genus, originally described from Central America and subsequently received from the Oriental Region, is now also represented by a species from Natal. Gonatas natalensis, sp. 1. Head black; antenne with the first and second joints stramineous, remaining joints mutilated in type; pronotum with the anterior area black, posterior area ochraceous ; scutellum black; corium dull greyish white, clavus pale ochraceous ; membrane dull greyish white ; body beneath black ; rostrum and legs pale ochraceous, apical abdominal segment castaneous ; head including eyes scarcely narrower than anterior margin of pronotum, which is distinctly strongly darkly punctate on the pale posterior area, its lateral margins moderately ampliated and slightly sinuate at the junction of the anterior and posterior areas, its posterior margin distinctly moderately concave ; scutellum longer than broad, moderately elevated, and distinctly foveate on the basal area, basal and lateral margins punctate ; membrane reaching the abdominal apex. Long. 53 mm. Hab. Natal; Durban (Bell-Marley). The Myth of the Ship-holder. 271 XXIX.—The Myth of the Ship-holder*: Studies in Echeneis or Remora.—I. By E. W. Gupeur, State Normal College, Greensboro, N.C., U.S.A. [Plates XV.-X VII. ] ConTENTS. Introduction, The Myth of the Ship-holder. The Myth explained. First Explanation: Foul Bottoms. Second Explanation: The Adhering Remora acts as a Rudder. Third Explanation: Large Numbers of Adhering Remoras. Fourth Explanation: “ Dead- Water.” Bibliography. Explanation of the Plates. INTRODUCTION. Kver since the time of Aristotle, the ship-holder or sucking-fish, because- of its peculiar structure and habits, has greatly interested men both scientific and unscientific. Possessed of a suctorial disk on the head and the shoulder region, it is able to attach itself to whales, porpoises, turtles, rays, and sharks, or ta large fishes of any kind, and thus secure transportation and opportunity to obtain food without exertion. It likewise attaches itself to boats, ships, floating wrecks, or even logs in the same way and for the same purpose. From this it is an easy transition to the belief of the ancients that attaching itself thus to a vessel it might retard or even hold it back. Hence the name Hcheneis, one that holds back a ship, and Remora, a holding back. “There is scarcely a fish of the existence of which the ancients have been equally certain, and which has so much occupied their imagination—from a power thought to be inherent in the creature to counteract the strongest physical agencies,—as the Hcheneis of the Greeks or the Remora of the Latins.”’ ¢ * In gathering the material for this paper, I am under much obligation to Dr. C. R, Eastman of the American Museum of Natural History, New York City, and to Dr. H. M. Lydenberg, Reference Librarian of the New York Public Library. In his work for the American Museum on the great biblicgraphy of fishes, Dr. Eastman ran across and kindly trans- mitted to mea large number of the references made use of in this paper. Dr. Lydenberg has, as heretofore, been a court of last resort for obscure and seemingly unintelligible references, every one of which he has, by reason of his large knowledge of matters bibliographical, been able to clear up. My best thanks are hereby rendered to him and to Dr. East- man for their many kindnesses. + Ginther, ‘On the History of Echeneis,’ 1860. 272 Mr. E. W. Gudger on the The earliest references to this interesting fish are to be ~ found in Aristotle’s ‘History of Animals.” A fish having such an extraordinary structure as the sucking-disk and having such unusual habits could hardly be expected to have escaped the keen observation of the Father of Natural History. Yet there is nothing in Aristotle’s writings to indicate that he ever saw or at any rate that he ever examined the Echeneis with the care which he bestowed on the other animals of which he wrote. In Prof. D’Arcy W. Thompson’s scholarly translation (Oxford, 1910), one may read (Book II. 14, 505 6, 19-22): ‘ Of fishes whose habitat is in the vicinity of rocks there is a tiny one, which some call the Echeneis or ‘ship-holder’.... Some people assert that it has feet, but this is not the case: it appears, however, to be furnished with feet from the fact that its fins resemble these organs.” Again (Book V. 31, 557 a, 30-31): “In the seas between Cyrene and Egypt there is a fish that attends on the dolphin which is called the ‘dolphin’s louse.” This fish gets exceedingly fat from enjoying an abundance of food while the dolphin is out in pursuit of its prey.” In a footnote, Prof. Thompson identifies this fish as Naucrates ductor, a pilot-fish found in the Mediterranean. Now the term pilot-fish is applied rather indefinitely to a number of different fishes. The Echeneis or Remora is possibly the one best known, from its habit of sticking to dolphins, sharks, or any large fishes and swimming before their snouts. In our waters Seriola zonata and S. carolinensis, amber-fishes of the family Carangide, are found associated with sharks and are called pilot-fishes. They are likewise found around the rudders of vessels and hence are also called rudder-fishes. The Naucrates ductor of Prof. Thompson is a pilot-fish of the same family but of a different genus. Tt is found in warm waters throughout the world and has the same habits as the other pilot-fishes. Thompson’s footnote thus leads one away from the idea that the “ dolphin’s louse” is a sucking-fish, but it should be noted that this last reference comes in a section devoted to sucking insect parasites, lice, ticks, and fleas, and con- cludes with those crustaceans, “ sea-lice”’ so called, which live parasitically on fishes. So from this internal evidence it seems probable that the fish referred to is an Echeueid, a sucking-fish, which attaches itself in a louse fashion to the dolphin as these fish are known to do *. * In a short note published in ‘Science’ for September 1, 1916, the present writer endeavoured to show that Prof. Thompson’s identification Myth of the Ship-holder. 273 In corroboration of the foregoing, Hasselquist may be quoted. In his ‘Journey to Palestine’ (1757) he notes that the Arabs at Alexandria ealled the sucking-fish (Echeneis neucrates) “Chamell Ferrhun.” Dr. Frank R. Blake of the Johns Hopkins University has been good enough to pass on this Arabic name. He writes that Chamel means louse, and that ferrhun is probably—or, at any rate, possibly —an erroneous transliteration for theArabic ferzhun, meaning agile or nimble. And that this meaning fits the actions of the fish, anyone knows who has ever tried to catch with a dip- net a shark-sucker from off its selachian host—-it dodves as expertly as a squirrel around a tree. However, Dr. Blake says that there is an Ethiopic word ferihun, meaning terrible, and that Hasselquist’s name may mean “ the louse of the terrible oue,” and since this fish is found most frequently adhering to the shark, this translation seems the most logical one. In further corroboration of the contention that the “‘dolphin’s louse” is the Hceheneis, another eastern traveller, Forskal (1775), may also be quoted. At Dyjidda, a town on the eastern side of the Red Sea about midway between Suez and Aden, Forskal collected Mcheneis neucrates, and was at especial pains to note that the Arab fishermen there called it “‘ Keide” or “ Kaml el Kersh,” which he translates “ the louse of the shark ”; while at Loheia, a town on the same side of the sea, but further towards the south-east, it is called “ Keda.” Dr. Blake has further obhged me by passing on these terms also. He finds that “‘ Kaml el Kersh” means “ the louse of the fish of prey,” which fish Forskal tells us in the context was a shark belonging to the genus Carcharias. Keda, he thinks, is probably a transliteration of the Arabic Keide, a fetter or band, hence “ the attached one.’ Still other testimony may be adduced as to the even more recent use of this name. ‘he German traveller Riippell in his ‘Fische des Rothen Meeres’ (1835), published only some eighty years ago, says of Echeneis: “In the northern part of the Red Sea it is called Delka or else Gammel el Kersh, of the dolphin’s louse as Nauerates ductor is erroneous as is Aristotle’s calling the little fish which lives among rocks Echeneis. The latter was identified as a goby and the “ dolphin’s louse” was shown to be a sucker-fish. Prof. Thompson on receiving this short paper very kindly wrote me that, while there might be still some uncertainty about the rock-dweller, he agreed as to the identity of the “ dolphin’s louse.” And now it seems well to incorporate this note in these introductory paragraphs and to add certain other data which have come to hand since the above article was published. Ann. & Mag. N. Hist. Ser. 9. Vol. ii. 21 274 Mr. E. W. Gudger on the in the southern part Aied.” The latter names are, of course, variations of those noted above. Dr. Blake has not been able to throw any light on the word Delka. From all this we see that, in the near East where changes take place slowly, Echeneis was still called ‘louse’ some two thousand years after Aristotle. While to-day in our own waters, as well as in most tropical seas, there is a certain small Echeneid fish which Gill (1862) has named Phthier- ichthys lineatus, the striped louse-fish. To return now to Prof. Thompson’s “tiny fish whose habitat is in the vicinity of rocks.” It seems to me that this fish cannot possibly be an Echeneis. ‘The Kcheneis is not a “tiny” fish, since the adult forms generally range in length from ten inches to three feet ; likewise, so far as is known to naturalists, it does not dwell among rocks. In fish literature of the medieval and renaissance times, how- ever, we do frequently run across references to Echeneis as a dweller among rocks, but I take these accounts to be merely echoes of Aristotle, since they are in other respects mere copies of preceding writers. Furthermore, this fish is said to have feet or, at any rate, fins resembling such organs. To the present writer there is no doubt that the fish here referred to is a goby, for gobies are small fish, are found in or near rocks, and have their forwardly-placed pelvic fins transformed into hand-like or sucker-like prehensile organs*, Tue Myr or THE SHIP-HOLDER. It will be remembered that Aristotle (384-322 B.c.) calls our fish Echeneis, ship-holder, but that he nowhere refers to the miraculous power alluded to by other but later writers. So it is doubtful whether he knew of these alleged powers, but if that be true why should he have named it ship-holder ? His words are “which some call the Echeneis or ‘ ship- holder,’ ’’ and he is evidently quoting some previous writer, or giving the name in common or everyday use. One thing is clear, 7. e. he is not the originator of the term, nor is it very evident that he knew the fish by personal observation. Before bringing to the attention of the reader the various stories ascribing miraculous powers to our fishes, * Since writing the above I have found that Lowe, so long ago as 1848, expressed the belief that Aristotle’s Echeneis was a blenny or a goby or a Chironectes and that the dolphin’s louse was an Kcheneis. On both of these points Giinther (1860, 1880) likewise is in agreement with the author of the ‘History of the Fishes of Madeira,’ Day (1880-84) also has briefly expressed his belief in this identification. Myth of the Ship-holder. 275 figures of the fishes themselves are presented. Pl. XV. of this paper shows Leptecheneis naucrates (fig. 2) and Remora brachyptera (fig. 3), which are commonly found in our Atlantic waters. The essential external differences between the fishes are readily seen from the figures. Fig. 1 shows the sucking-disk of the Remora. Consideration of the structures of these fishes is reserved for a later paper. The first definite reference to the ship-retarding power of the Echeneis is in a poem on fishing, “ Halieutica,’ by the Latin poet, Ovid (43 8.c.-17 or 18 a.p.). Verse 99 reads : “ Parva Echeneis adest, mirum, mora puppibus ingens” ; which may be translated, ‘‘ The small Hcheneis is present, wonderful to say, a great hindrance to ships.” Pliny the Elder (23-79 a.p.) twice refers to the Echeneis. In Book 1X. Chapter 41 of his ‘ History of Animals’ he says: “It is believed that when it (Echeneis) has attached itself to the keel of a ship its progress is impeded, and that it is from this circumstance that it takes its name.” This (together with other data extraneous to our subject) is taken from Aristotle. Then Pliny quotes one Mucianus (about whom nothing has been obtained) that a murex, a kind of gasteropod mollusk, has a similar ship-retarding power, and gives from this writer an alleged instance of a ship being held by it. Pliny in the same chapter quotes one Trebius Niger that the fish is about one foot in length and that it can retard ships. I have been unable to find out anything about this writer; this reference, like the one to Mucianus, is entirely obscure *, In Book XXXII. Chapter 1, Pliny gives what is the first detailed account of the ship-holding power possessed by the Echeneis, and it seems well to quote him in extenso as given in Bostock aud Riley’s translation (1857). “And yet all these forces [winds, tides, &c.] .....a single fish, and that of a very diminutive size.... the fish known as the ‘ Echeneis’.... possesses the power of counteracting ..... A fish bridles the impetuous violence * Pliny also gives two other uses of the Echeneis, which though outside the scope of this paper, are of enough interest to appear in a footnote. The first (which he seems to have had from the Greeks) is . its use in love philters, and for the purpose of delaying judgments and legal proceedings ; all of which he justly says are evil properties, compen- sated for, however, by its use to stay the flow of blood in pregnancy and for the preservation of the foetus 7m wtero. The second use, quoted from Trebius Niger, is that when preserved in salt it is able to draw up gold from the bottom of the deepest well. These fictions are gravely repeated by many writers down to the middle of the seventeenth century .... at least as late as the time of Rabelais (ene 21 276 Mr. E. W. Gudger on the of the deep, and subdues the frantic rage of the universe— and all this by no effort of its own, no act of resistance on its part, no act at all, im fact, but that of adhering to the back ean ‘At the battle of Actium, it is said, a fish of this kind stopped the praetorian ship of Antonius in its course, at the moment he was hastening from ship to ship to encourage and exhort his men, and so compelled him to leave it and go aboard another. Hence it was, that the fleet of Caesar gained the advantage in the onset, and charged with re- doubled impetuosity. In our own time too, one of these fish arrested the ship of the Emperor Caius (Caligula) in its course when he was returning from Astura to Antium: and thus, as the result proved, did an insignificant fish give presage of great events; for no sooner had the emperor returned to Rome than he was pierced by the weapons of his own soldiers. Nor did this sudden stoppage of the ship long remain a mystery; the cause being perceived upon finding that, out of the whole fleet, the emperor’s five-banked galley was the only one that was making no way. The momeut this was discovered some of the sailors plunged into the sea, and on making a search about the ship’s sides, they found an Echeneis adhering to the rudder. Upon its being shown to the emperor, he strongly expressed his indignation that such an obstacle as this should have im- peded his progress, and have rendered powerless the hearty endeavours of some four hundred men. One thing too, it is well known, more particularly surprised him, how it was possible that the fish, while adhering to the ship, should arrest its progress, and yet have no such power when brought on board ”*. This full and circumstantial account by Pliny is of great - value, and the more so since everything leads one to believe in Pliny’s full credence in the wonderful power of the ship- stayer. In the paragraph following the above, our old Roman naturalist thus refers to its Latin name: “ Some of our own authors have given this fish the Latin name of ‘mora’ [delay],’’ another reading gives “ remora.” The next of the ancients to write of our fish is the famous historian, Plutarch (46 a.p.)._In his ‘ Symposiaes,’ Book II. * Bostock and Riley say in a footnote, “ And well might it surprise him. If there was any foundation at all for the story, there can be little doubt that a trick was played for the purpose of imposing on Caligula’s superstitious credulity and the rowers as well as the diving sailors were privy to it.” Later it will be shown how entirely erroneous is this conjectural explanation of Pliny’s translators. Myth of the Ship-holder, 270 question 7, he says: “ Cheremonianus the Thrallian, when we were at a very noble fish dinner, pointing to a little, long, sharp-headed fish, said the Echeneis (ship-stopper) was like that, for he had often seen it as he sailed in the Sicilian sea, and wondered at its strange force ; for it stopped the ship when under full sail, till one of the seamen perceived it sticking to the outside of the ship and took it off” But there was incredulity even in that day for Plutarch adds, “Some laughed at Chaeremonianus for believing such an incredible and unlikely story.” Then Plutarch offers for this phenomenon an explanation of his own which will be given later. Next we come to Oppian, who flourished late in 200 a.p. In his poem Halieutica—‘ On the Nature of Fishes and the Fishing of the Ancients’’—as translated by John Jones, there are some 38 lines in which in very poetical and effusive fashion the action of the ‘‘sucking-fish”’ is described. In short, he tells. how the fish clings to the keel of the swift ship and retards it, though the wind causes the sails to belly out. He seems, however, to have confused with the KEcheneis the lamprey ee] which has a round suctorial mouth. : The last of the ancients to catalogue the myth of the ship-detainer was Aelian, a Roman author contemporary with Oppian in the latter part of the third century a.p. In his ‘De Natura Animalium,’ Book I. Chapter 36, he refers to “that fish which all men call remora because it holds back and delays ships.” And, again, in Book III. Chapter 17, he tells us in very interesting fashion that: “ Kcheneis is a pelagic fish, black in appearance, equal in length to an average-sized eel, and named for the thing it does. For adhering with its teeth to the extreme stern of a ship driven by a following wind and full sails, just as an unmastered and unbridled horse is held in with a strong rein, so the fish overcomes the most violent onset of the winds and holds the ship as if tied fast to her wharf. In vain the middle sails belly out, in vain the winds rush forth, it holds steady the thing to which it adheres, The sailors’ know this indeed for the cause of this matter. Hence the name given to this fish, which, because of their experience with it, they call Echeneida (Remora).” We next hear of the ship-holder in the writings of the early Christian Fathers, and I am able, thanks to the kind help of Dr. Hastman, to quote herein from two. The first of these seems to have been Saint Basil, sometimes called the Great, bishop of Caesarea in Cappadocia. In his 278 Mr. E. W. Gudger on the Hexameron *, Homily VII. paragraph 56, he writes: “If now you hear say that the greatest vessels sailing with full sails are easily stopped by a very small fish, by the Remora, and so forcibly that the ship remains motionless for a long time, as if it had taken root in the middle of the sea, do you not see in this little creature a like proof of the power of the creator ?” St. Ambrose (840-397) in his ‘ Hexameron,’ the first edition of which bears the imprint Basilez, 1566, describes Echinus (probably a misspelling of Echeneis) as a foreteller of storms. “ At the approach of a tempest the fish lays hold of arock and sticks fast to it until calm weather returns. The sailors, noting this, govern themselves accordingly.” This is probably an echo of Aristotle’s little fish found among rocks, and seems to be the first of a long succession of similar stories, ascribing to this fish weather-forecasting powers. St. Ambrose, however, does not seem to give the ship-holding story. Jorath, who was probably an Oriental Christian of the twelfth century, speaks of a fish called Achandes which sticks fast to ships in the sea, thus making them to stand stock still f. About the year 1250, Bartholomew Anglicus wrote his encyclopedic work ‘De Proprietatibus Rebus,’ which was translated by John Trevisa in 1397, and printed at Win- chester in 1491. The following is his interesting account of the ship-holder, for which also I am indebted to the kindness of Dr. Eastman :— “ Enchirius is a little fish unneth [only] half a foot long ; for though he be full little of body, nathless he is most of virtue. For he cleaveth to the ship, and holdeth it still steadfastly in the sea, as though the ship were on ground therein. Tho’ winds blow, and waves rise strongly, and wood [violent] storms, that ship may not move nother [neither] pass. And that fish holdeth not still the ship by no craft but only by cleaving to the ship.” In 1475, Johann von Cuba (or Cube) published at Metz his ‘ Hortus Sanitatis.’ In the edition of 1536 on page 78 of chapter 34 he discourses of Echeneis or Echinus. This, * “ Hexameron is the title of nine homilies delivered by St. Basil on the cosmogony of the opening chapters of Genesis...... Basil read the book of Genesis in the light of scientific knowledge of his day.” He was born in 829 and died in his fiftieth year. + For this reference I am indebted to Dr. Kastman, who ran across it on page 71 of Von Cuba’s ‘ Hortus Sanitatis,’ to which reference will be made later. - Myth of the Ship-holder, 279 he says, is a little foot and a half long fish which lays hold of ships and causes them to stand still as if rooted in the sea, being held by nothing save the little fish. His story adds nothing to what we already know, but he does one thing which is of great interest, he gives us a quaint figure, which so far as I have been able to find, is the first and only effort to illustrate the myth. It is reproduced as fig.4(P]. XV.). And in this connection one is led to wonder why this story, so interesting to these old-time writers, was not also a favourite theme for illustrators, why it has come down to us with but one picture. In the ‘ Annotationes’ of Francisco Massari, published at Basiliz in 1537, there are in chapter 35 some three or four pages of data on the Echeneis, but careful perusal shows that this is but a revamping of the ancients with not a single new legend added, so Massari may be passed without further comment. In the year 1550 there was published at Lugduni ‘ Liber I. De Sympathia et Antipathia Rerum’ by Hieronymous Frascatorius, on page 24 of which is the statement that, “ Furthermore it seems to be beyond all doubt that Echeneis is that little fish which we call Remora, which causes to stand still in mid-ocean the ship moved by the force and impetus of the wind” *. According to both Gesner and Aldrovandi, there is to be found an account of the ship-holding power of Echeneis in Adam Lonicer’s ‘ Naturalis Historiz Opus Novum in Quo Tractatur de Natura,’ etc., Frankfurt, 1551. The only edition found in New York is the German translation, which appeared as ‘ Kreuterbuch’ in 1560. Dr. Lydenberg kindly looked through the 1682 edition of this in the New York Public Library, but could not find any reference to Echeneis. I have not been able to locate another copy. However, in Gesner’s ‘ Historia Animalium,’ IV. (1558), aud also in Aldrovandi, there is a considerable quotation from Lonicer with reference to Echeneis. Careful study of this, however, shows that no new data are given. The account of Edward Wotton (1552) is but a rehash of Aristotle, Pliny, and the other Greek and Romau writers. His one statement worthy of repetition reads ‘ Let the winds rush and tle tempests rage, the Remora dominates the furor, overcomes these great forces, and compels the vessels to stand still, which no chain and anchor have been * For a transcript of Frascatorius [ am indebted to the courtesy of Mr. Charles Perry Fisher, Librarian of the College of Physicians, Philadelphia. 280 Mr. E. W. Gudger on the made heavy enough to do.” This, however, seems to be taken from Pliny. In the sayings of Pantagruel, Rabelais (1553), in Book IV. Chapter 62, has the following:—‘....an Echeneis or Remora, a silly, weakly fish, in spite of all the winds that blow from the thirty-two points of the compass, will m the midst of a hurricane make you the biggest first-rate remain stock still, as if she were becalmed, or the blustering tribe had blown their last.””? And again, in Book V. Chapter 26: “....there (in the country of Satin) I saw a Remora, a little fish called by the Greeks Ecleneis, near a big ship which was motionless although under full sail, on the high sea.” We now come to Rondelet (1558), who attempts to show that the retardation of ships might have been effected by the Echeneis of Pliny, the great shell-fish of Mucian, or the eel of Oppian. Indeed, he asseverates (page 313) that he has known a lamprey to thus hold. back a boat: “. . . it [Oppian’s eel] stops it and holds it [a boat] back ; a thing which corresponds to our lamprey, and which IT have known through experience, for if it puts its mouth against a boat it stops it, and I have seen it thus.” ‘hen he adds, “‘ There is no need to marvel that various fishes are called by different authors by the same name, nor that the same fish be called by many and divers names, for that often happens.” For the rest, Rondelet quotes and comments on the accounts of Pliny and others on the true Echeneis (pp. 334-5), but adds nothing of himself. More might be expected of this great ichthyologist ; but it seems that he never saw the fish (he gives no figure of it) and knew nothing of it at first-hand. Conrad Gesner was the greatest of the encyclopedic writers of natural history, and his ‘ Historia Animalium,’ Books I.-LlII., was published Basel, 1551-1558*. In Book IIIL. he discourses at considerable length ‘ Con- cerning Hcheneis or Remora,” but there is nothing in his writings to indicate that he ever saw the fish. He adds no new data ; but this section of his book is of value because iin it he quotes a large number of the writers previously eited in this paper. However, even here his value to the student of ichthyological archeology is crippled by the fact * Tt will be noted that the works cited of both Gesner and Rondelet are dated 1558, and yet Gesner quotes Rondelet at considerable length. However, the apparent discrepancy disappears when it is remembered that Rondelet’s ‘ L’ Histoire Entiére des Poissons’ is but a translation into his native French of his original work first published in Latin in 1554. Ball Myth of the Ship-holder. 281 that he quotes his predecessors by name only, rarely by book or chapters He adds nothing to our knowledge of the Myth. Gesuer, however, is the first writer since the ancients to attempt a description of Echeneis. This description, which is found in the last paragraph of his section on the Echeneis, is evidently that of a goby, and is quoted here that the reader may judge for himself, and uot be led into the error of er editing Gesner with the first description. “There is a little fish found in the ocean at Emda in Frisia (so a certain friend has related to me) four digits long, of very slimy skin, without scales, having a head “large in proportion to its body, eyes small, the rest of the body cone- shaped. Under its chin it had the form of a sucker by which it probably adheres to rocks, for when he pressed this cavity with his finger (so my friend narrated it) it adhered to it so that it could be carried about.” In Chapter XX XVII. of Liber X. of his ‘Operum,’ pub- lished at Lugduni in 1564, Jerome Cardan writes of the action of the Remora as if it were a settled fact, but adds nothing of value to detain us here. He will be referred to later as offering an explanation of the ship-staying powers of the fish. Departing from the beaten track of repeating what some previous writer had copied, the Dutchman, Jan Huygen van Linschoten, or, as his name is Latinized, Joannes Hugo Linscotanus (1596),- gives the following interesting and detailed account of the ship-holding power of the Remora :— *¢ And because I am now in hand with the Fishes of India, IT will here declare a short and true Historie of a Fish, although to some it may seeme incredible, but it standeth painted in the Viceroyes Pallace in India, and was set downe by true and credible witnesses that it was so, and therefore it standeth there for memorie of a wonderful thing; together with the names and surnames of the ship, Captaine, day, & yere when it was done, and as yet there are men living at this day, that were in the same shippe and adventure, for that it not long since, and it was thus. That a ship sayling from Mosambique into India, aud they having faire weather, a good fore winde, as much as the Sayles might brave before the winde, for the space of fourteene dayes together, directing their course towards the Equinoctiall line, every day as they tooke the height of the Sunne, in stead ‘of diminishing or lessening their degrees, according to the Winde and course they had and held, they found themselves still contrarie, and every day further backwards then they were, to: the 282 Mr. E. W. Gudger on the great admiration and wondering of them all, and contrarie to all reason and man’s understanding, so that they did not only wonder thereat, but were much abasht beeing stead- fastly perswaded that they were bewitched, for they knew very well by experience that the streame or course of the water in these countries did not drive them back, nor with- holde them contrarie to all Art of Navigation, whereupon they were all in great perplexity and feare, standing still and beholdinug each other, not once knowing the cause thereof. *“* At ye last the chiefe Boteson, whom they call the masters mate, looking by chance overbord towards the beakhead of the ship, he espied a great broad taile of a Fish that had winded itselfe as it were about the beakehead, the body thereof beeing under the keele, and the heade under the Ruther, swimming in that manner, and drawing the shippe with her against the wind and their right course : whereby presently they knewe the cause of their so going backe- wards: so that having at last stricken long with staves and other weapons uppon the fishes taile, in the ende they stroke it off, and thereby the fish left the ship, after it had layne 14 dayes under the same, drawing the ship with it against wind and weather: for which cause the Viceroy in Goa caused it to be painted in his pallace for a_perpetuall memory, where | have often read it, with the day and the time, and the name of both shippe and Captaine, which I cannot well remember, although it bee no great matter” *, Ferrante Imperato, a pharmacist of Naples, having a taste for natural history, formed a collection of such objects, and made the description of these the basis of his book ‘ Historia Naturale,’ published at Venetia, 1599. In this he writes : ‘‘ Although the Remora of the ancients has by many been described under the forms of different fishes, there is, how- ever, no description that fits except the one proposed by us. It has on the upperpart of the head tentacles similar to the vibratile combs [cirri, literally ringlets] of the polyps by which it attaches itself to ships or the bodies of large whales and other fishes.” With the above description Imperato published a figure of * Linschoten’s book was first published in Dutch at Amsterdam in 1596, but was translated into English and published in London in 1598, while in the following year (1599) a Latin version appeared at Amsterdam, The above account is taken literally from the Hnglish edition. For photostats of it and of the original Dutch edition I am indebted to the kindness of Dr. Lydenberg, who not only sent these, but who had pre- viously in a most skilful manner run down Linschoten from an exceed- ingly indefinite and obscure reference in Nieremberg to the ‘ Pro-Rex of Joanues Hugo,” Myth of the Ship-holder. 283 Echeneis or Remora which, so far as I have been able to find, is the earliest portrayal of the sucking-fish. This is reproduced herein as fig. 5 (Pl. XVI.). It correctly shows the projecting lower jaw, the position and general make-up of the sucking-disk, and the position of all fins, especially thé long dorsal and ventral ones. ‘The tail is not good. It is probably a Remora, since there is no effort to portray the lateral stripe of Echeneis. The crudity of the figure is, of course, apparent, but it is the first, and it is a fair ‘portrayal. The disk is clearly shown, and in the description its function is definitely indicated for ‘the first time in history * We come now to another original story of the wonderful power of the Remora. It is quoted from Hkman (who will be referred to later), who says that it was told by Bartolomeo Crescentio Romano in his book ‘ Nautica Mediterranea’ published at Rome in 1607. ‘This book | have not seen. “....and I must tell you about another deed of the devil, because you must know in how many ways this enemy of mankind works against poor seamen. “On a voyage from Gaeta to Napoli, the galley ‘8. Lucia,’ wheu sailing before a fresh wind and being two miles from port, stopped quite immovable in spite of her sail being strained. The steersman examined the rudder to see whether there was some rope or net fastened to it, and as nothing was found, he commanded the oars to be got out and the gailey slaves to be forced on with hard blows. But the galley did not move from the spot, and when she had been lying motionless for a quarter of a hour cor more, the other gallevs, which had sailed on, shortened sails, waiting. Then a man named Catelano told the captain .... to have three monks removed from the deck of the galley, and averred that the galley would then immediately begin to move ; and when the captain had them removed, the galley certainly did begin to speed lke an arrow. “hen all the men were about to throw these three poor fellows into the sea, saying that they were excommuni- cated; but the same man Catelano helped them saying, that this was a strategem of the devil to the detriment of the monks ; and he obtained permission that they should only be taken from the vessel. “This occurrence would have caused scientific men to suppose that a very small fish, resisting the progress of the * The above figure and description are taken from the 1599 edition of Imperato’s book found in the library of the Academy of Natural Sciences of Philadelphia. or it 1 am indebted to the kindness of Dr, Edward J, Nolan, Librarian. 284 Mr. E. W. Gudger on the vessel, had got the better of the force of the sails and oars and made the vessel stop.” We next come to another great ichthyological encyclo- pedist of the Renaissance, Ulyssis Aldrovandi, whose huge folio, ‘ De Piscibus et de Cetis,? was published in 1613 at Bononie. This author devotes to the Remora some five pages, which are taken chiefly from Gesner. He discourses at considerable length of the ship-holding power of the Remora, and quotes Aristotle, Pliny, Rondelet, and several others of the authors previously considered in the present paper. However, it seems probable that he never saw the fish—at any rate, a careful translation of his very difficult Latin nowhere reveals any definite statement that he had seen it. However, he does the one good thing of giving us a figure and description which adds materially to our know- ledge. A photographic reproduction of his drawing’is given here as fig. 6 (Pl. XVI.). Note that it is labelled the “Remora of Imperato and the author.’ Aldrovandi ex- pressly says “... my drawing corresponds with that one’s,” but his figure looks like an Echeneis, and his description below confirms this idea. He says :— “The color of the whole body almost inclines to violet, its sides are glistening, the body is cut into two in the middle by a sub-green line, and its tail verges to blue. There are six fins to the body, three on the belly, two each in the region of the stomach and one at the anus. Likewise there is one on the back, and the tail ends in another.... Its mouth is not unlike a dog’s except that the lower jaw projects beyond the upper jaw contrary to that which we see in the shark. I think that this is a truer figure [than Im- perato’s| 7 *. This description seems to have been made from the fish rather than from the drawing, since the latter does not show the median line. it is to be regretted that Aldrovandi does not give us a definite statement on this point. Aldrovandi, in his discussion of the Remora, gives this interesting incident :—“ Within the memory of our parents, it is said that the ship of Franciscus Turonensis, the Cardinal, when he was once upon a time going from Gaul by maritime journey into Italy, according to the narrative of Peter Melara of Bologna, a very brave knight and at the * For the scholarly translation of Aldrovandi, I am indebted to Mrs. S. P. Ravenel, and to Miss Julian Dameron, associate professor of Latin in the College. Miss Dameron has also been so kind as to help me with a number of the other Latin articles herein referred to. Myth of the Ship-holder. 285 same time a very learned man, was delayed by a very small fish in the midst of its course ”’ *. The reference made to this same incident by John John- ston, in his book ‘ A History of the Wonderful Things in Nature,’ London, 1657, on page 301, is probably taken from Aldrovandi. At Geneva, in 1614, Bartholomew Keckermann published his works, and in his ‘ Disputationes Physice’ he discusses the ship-staying power of the Remora. He adds nothing to our knowledge of the myth, but does offer an interesting explanation, which will be considered later. . We next come to Rochefort, whose interesting and in- structive book on the Antilles was published at Rotterdam in 1665, who says that certain fish bear the name Remora *“‘ because they adhere to vessels as if they wished to arrest them in their course.” Note the clause “as if they wished.” The old order is passing away, men are beginning to seek a rational explanation of the retardation of ships, and doubt is being cast on the efficacy of the Remora as the agent. So more explicitly writes Du Tertre, whose valuable natural history of the Antilles was published but two years (1667) after Rochefort’s work. In the course of his descrip- tion of the Remora and explanation of its activity, he writes :-— “For myself I hesitate to submit my judgment to that which some authors assure us concerning the Remora, saying that it brings to a full stop a ship which sails before the wind with canvas stretched on a full sea. Since there is so great a quantity of Remoras around the Western Isles, one could scarcely find a ship that would not have several attached to her, yet nevertheless during the century or more that these islands have been frequented, it has never been noted that a single ship has been thus arrested by the Remoras, ‘his has caused me to think that the two or three vessels, which have been said to have been arrested by the Remoras, have been detained by some miracle or charm, and since at the time some Remoras have been attached to them * Being unable to do anything whatever with this reference, I veferred it to Dr. Lydenberg, who very kindly went into the matter fully. He finds that there was a Peter Melara of Bologna who left certain MSS. which are or were to be found in the “ Biblioteca dell’ Instituto ” of that city. He suggests that Aldrovandi had access to this particular MS, This conjecture is strengthened when one remembers that Aldrovandj lived, wrote, and published his book in Bologna, Note, further, that he prefaces his statement by saying “ within the memory of our parents,” 286 Mr. E. W. Gudger on the in their usual fashion, to these have been falsely attributed the cause of their detention.” It will be shown later how closely Du 'Tertre came to a true explanation, and it is to be regr etted that in substituting one mythical explanation for another he narrowly missed the truth. Therein he was better churchman than natu- ralist. Le Maire (1695) writes “ Le Sucez [Echeneis] is so called because it attaches itself by sucking. It is in size about equal to asole. When it attaches itself to the rudder, it retards the vessel, but does not stop it as the Remora is falsely said to do,” In the face of what has just been quoted there is now to be presented from one of the most remote corners of the world another and much later story of the Myth. Faber, in his ‘Natural History of the Fishes of Iceland’ (1829), gives the following circumstantial account :— “In Jau Olsen’s MS. it may be read [that]: ‘In the year 1720, by chance it happened on the strand before Hunevand’s- Harde (in Nordisland) with a boat which had been rowed out for the autumn fishery, that when the fishermen wished to return they could not move the boat, although they rowed with all their might. Then there was noticed behind on the rudder a short stumpy fish, blackish-gray in color, which moved itself a little and adhered so solidly to the boat that one could scarcely pull it loose with the hand. It left behind on the boat a mark of its body, and when it was pulled loose the boat went forward. The fishermen burned it on the shore whereby a great stench was produced. This animal appears to have been a Remora, and through this account the matter seems to be confirmed that there are really such living fish which can bring a ship to a standstill.’ ” Faber then concludes : “The exaggeration of the account being allowed for, it is not to be doubted this was a sucking fish.”’ There is now to be given the latest and most modern account of retardation by the Remora that has come to light. In 1778 there was published in London, “ Translated under the author’s inspection,” the ‘Travels in Dalmatia’ of the Abbé Alberto Fortis. The locality, it should be noted in passing, is not very far removed from the countries Greece and Rome, in which the legend originated. In a letter to Signior Marsili, Professor of Botany in the University of Padua, Fortis writes :—‘‘ I will finish this letter by relating a fact, to which you may give that degree of faith which you think it merits. You have often read in ancient natu- Myth of the Ship-holder, 287 ralists, of wonderful things done by the Remora, or Echeneis and not without some surprise will have learnt Pliny’s story, who after having told us, on the faith of another, how Anthony was retarded on bis voyage by means of this fish, asserts positively, that a ship with Caligula on board and four hundred rowers, was actually stopped by one of these fishes, while the rest of the fleet went on at a great rate. When I read this, I contented myself to shrug my shoulders, without perplexing my brain to find out by what natural processes, or matter of fact, such an opinion could become so generally received, that a man of sense as Pliny certainly was, should affirm it in positive terms. But chance led me to the discovery. We were sailing in a small bark between Vruillia and Almissa with a fresh equal gale, in the afternoon. The mariners were all at rest, and the steersman only was awake, and attended alone in silence to the direction of the bark ; when, on a sudden, we heard him call aloud to one of his companions, ordering him to come and kill the Paklara. Our learned friend Signior Guilio Bajamonti was with me, and understanding what the man meant, desired him to show him the fish that he wanted killed, but the fish was gone, Having interrogated the steersman, who did not want sense, and was a fisherman by profession, why he had ordered the Paklara killed, and what harm it had done; he answered, without hesitation, that the Paklara used to take hold of the rudder with his teeth, and retarded the course of the bark so sensibly, that not only he, but every man who sat at the helm felt it there without seeing it. He added, that many a time he himself had catched the Paklara in the act and had frequently killed and eat it. That it was often met with in the waters of Lissa. ‘That in shape it resembled a conger eel, and in length did not usually exceed a foot and a half. That if I had a mind to see, and catch one of them I needed only to go in a fishing boat, in the warm season, between the islands of Lessina and Lissa, where he had never failed to meet with them every year. [ will not desire you to believe everything my pilot said; but confess that I should be very glad to see the Paklara when it had taken hold of the rudder of the bark under sail. ‘ihe wonderful strength of the muscles of some little marine animals, such as the Lepades, that so obstinately resist any attempts to disengage them from their rocks, the stroke proceeding with such rapidity from the Torpedo, known at Venice by the name of pesce tremolo aud in the sea of Dalmatia by that of Trnak; the vigor shewn by the Dentici in their convulsive motions even when out of their own element, not to mention 288: Mr. EK. W. Gudger on the the larger fish, such as, Tunny, Dolphins, ete., give me ground to suspect, that if all that the ancients wrote con- cerning the Remora be not just literally true, it is not alto- gether false. It certainly is a thing worthy of some reflection, that Pliny speaks so diffusely concerning this phenomenon, as a known fact that could not be called in question. ‘The Greeks adopted the.notion of this extravagant faculty, by superstitiously hanging the Remora about women with child, to prevent abortion. I am not, however, so ready to credit these extravagances or in the least persuaded of the wonderful retarding force of this little fish ; and think it sufficient to believe that the force of the Paklara may be felt at the rudder of a small bark, without troubling myself further about the Remora. “The Remora of the ancients, and the Paklara of our days, have this remarkable difference, that the first is almost. always of the testaceous kind, and the second is of the genus Murena.”’ From this we see that the Abbé was half convinced of the correctness of the sailor’s belief as to the power of the Paklara. However, he thinks this fish to be a lamprey eel, while the Remora of Pliny is in his opinion a shellfish, This is confirmed by a further reference on page 325, which reads as follows :—‘‘ Among the curious fishes found in those waters [of Lissa] the Paklarais the most remarkable: I did not see it, but the description given me by the fishermen, agrees with the Hcheneis of Artedi, and Gouan, though, in my opinion, not with the Hcheneis or Remora of the ancients.” Before going into an explanation of the Myth of the Ship-holder, it may be of interest to show that the term Remora has attained a place in literature. Among the Romans we find Lucilius saying “ A certain voice Sounding forth made for you a Remora in your progress.” Again, Plautus says “‘Those things are distasteful which obstruct many undertakings and they make for a Remora both in public and private affairs.” However, since the word Remora is a common Latin term for a delayer or retarder, we cannot be sure that its use above is a reference to the fish; more probably it is a use of the term in its original and ordinary sense. Probably not such, however, is the use of the term by St. Basil (329-379). He affirms that “ Life is a voyage and in our life’s ways, countries, courts, towns, and rocks are remoras.”’ In English literature, however, more direet allusions are “ Myth of the Ship-holder. 289 to be found. Thus Spenser, in his ‘ Visions of the World’s Vanity,’ i. p. 108, writes :— “ Looking far forth into the ocean wide, A goodly ship, with banners bravely dight, Through the main sea making her merrie flight. All suddenly there clave unto her keel A little fish that men call Remora, Which stopt her course, and held her by the heel, That wind nor tide could move her thence away.” And Ben Jonson says (‘ Poetaster, III. 1) :— “TI say a remora, For it will stay a ship that’s under sail.” And again, in his Act TIT. Scene 1, he makes Horace say to Fuscus Aristius of Crispinus, a great bore, who had nearly talked him to death :-— “ Aristius. What ails’t thou man ? Horace. *Death, I am seized on here, By a land remora: I cannot stir, Nor move but as he pleases.’ Maundrell, in his ‘ Aleppo to Jerusalem’ (p. 46) writes :— “We had his promise to stay for us, but the remoras and disappointments we met with in the Road had put us back- ward in our journey.” And again, Jeremy-Taylor quaintly says :—‘‘ A gentle answer is an excellent remora to the progresses of anger, whether in thyself or others.” Before leaving this part of the subject, the following story may be added as of interest. In David Livingstone’s ‘ Missionary Travels and Researches in South Africa’ (New York, 1858), on page 556, in writing of the Barotse valley on the Leeba River, one of the headwaters of the Zambesi, he says :—‘ The Barotse [people or tribe] believe that at certain parts of the river a tremendous monster lies hid and that it will catch a canoe, and hold it fast and motionless, in spite of the utmost exertions of the paddlers.” ln the Indian Ocean around Zanzibar the Remora abounds in great numbers, and is used, as I shall show in another paper, for the purpose of catching turtles by virtue of its propensity for clamping itself fast to any floating object. At first 1 was inclined to think that the Barotse myth was a Ann. & Mag. N. Hist. Ser. 9. Vol. ii. 22 290 Mr. E. W. Gudger on the far distant echo of the Zanzibar stories; but Livingstone shows very conclusively that the mhabitants of the upper Zambesi in his day had uo communication whatever with the coast. Such communication may have existed at an earlier day, and, at that time the story may have been brought inland, or it may have arisen spontaneously. At any rate, it is very curious and is worth repeating in this connection. Tue My? EXPLAINED. First Explanation: Foul Bottoms. In giving the explanations of the Myth of the Ship-holder, it seems best to take them up chronologically, for, as might be expected, even in ancient days there were men whose minds sought a rational explanation. The first person who attempted to clear up this matter seems, so far as can be found, to have been Plutarch (46 a.p.). On page 277 his account of the statement of Cheeremonianus the Thrallian has been given, and it will be recalled that the latter was laughed at for believing such an extraordinary thing. However, Plutarch, entering into the conversation, said :— “Therefore as those things mentioned are but conse- quences to the effect, though proceeding from one and the same cause, So one and the same cause stops the ship, and joins the Echeneis to it ; for the ship continuing dry, not yet made heavy by the moisture soaking into the wood it is probably that it glides lightly, and as long as it is clean, easily cuts the waves ; but when it is thoroughly soaked, when weeds, ooze, and filth stick to its sides, the stroke of the ship is obtuse and weak; and the water coming upon this clammy matter, doth not so easily part from it; and this is the reason why they usually scrape the sides of their ships. Now it is likely that the Echeneis in this case, sticking upon the clammy matter, is not thought an acci- dental consequence to this cause, but the very cause itself.” Now it must be conceded that this is a reasonable explana- tion, and we will find that until the middle of the sixteenth | century it was repeated as explanatory of ship-retardation. Gesner (1558) quotes Piutarch at length, insists on the retarding effect of mosses and alg (‘‘ multa alga & musco innascete”’), and plainly shows that he regards these (among which the Echeneis is found) as an efficient cause in the slowing up of the speed of ships rather than the action Myth of the Ship-holder. 291 of the fish itself, although nowhere he expresses a disbelief in this power of the Echeneis. Levinus Lemnius * (1559), in discoursing of “ Sea-weed and Sea Fucus,” apparently only amplifies Plutarch when he . Says :— ** But Mosse must be held to be a thing different from these: one kind whereof grows not only on the shores, but upon the sterns of the ships, when they come home from long voyages, to whicii not only Mosse and Sea-weeds, but shell-fish and a little fish called Echeneis stick so fast, that they will stop Ships, and hinder their courses, therefore our men use to rub them off with sharp brushes, and scrape them away with irons that are crooked for the purpose, that the ship being tallowed and careened well and smoothly may sail the faster.” Aldrovandi, Gesner’s great successor and copier (1613), devotes several pages of his huge folio to “ Occultane an Manifesta Vi Naves Remoretur,” most of his data being taken from Gesner. He gives at length Plutarch’s explana- tion of the retardation as due to growths of marine algze among which the Echeneis clings, thus being “‘ not the cause of the retardation of the ship but an accident of the effecting cause.” Aldrovandi is the Jast of those who allege the growth of sea-weeds as a cause of the retardation. It began to be seen that, while such marine growths would slow up a ship, they did not explain the remarkable instances of retarda- tion in which the speed of the vessel was checked for a while but which was presently regained. However, another attempt had been made to explain these erratic movements of vessels, and this will now be given. Second Explanation: The Adhering Remora acts as a Rudder. This seems to have been first advanced by Rondelet (1558) in these words :-— “Pliny and others are greatly astonished that it is possible for this fish to have the power to stop a moving vessel propelled by sails and oars; but, as Aristotle says, one wonders at many things of which one does not understand the cause .... which we will give concerning the effect of * Lemnius’s book ‘ De Occultis Naturze Miraculis’ was first published at Antwerp in 1559. ‘The above quotation is from the English edition, ‘ Concerning the Secret Miracles of Nature,’ Book III. Chapter 9, pp, 218- 219, published at London, 1658. 22* 292 Mr. It. W. Gudger on the this fish taken by itself in the place it requires. Because the rudder is small and placed at one end of the boat it is managed by one man who does not exert himself greatly. In the same way it is easy for that which moves one end to move the whole, for as the force and swiftness of those , things which are thrown or moved finally ceases, so at the end of a continuous thing in motion the movement is weak and feeble, and because it is weak it is easily disturbed and overcome. As a boat, which is a continuous thing, goes very swiftly when driven by the winds, the first end called the prow goes more rapidly, and the rear end called the stern goes not so rapidly for in this latter place is the rudder which, moved here and there, makes the prow move easily also, for the reason above mentioned, and consequently the vessel as a whole moves. In this way, if a vessel is lightly driven straight ahead, and if the Echeneis or Remora, having put its mouth against the rudder, moves it here and there, it is necessary that this movement through the con- tinuity of the vessel be communicated also to the prow and that it stop in its first course to waver in this direction or that according as the fish moves it ; for it is a thing proved by reason, and certified by experience, that however little one of the ends is moved, the other also and indeed the whole of any continuous body is moved in the same way.” In this Rondelet seems to have taken from Aristotle’s treatise on Mechanics the latter’s explanation of how a rudder causes a ship to change her course, and to have adapted it as seen above to try to show how the Echeneis causes a ship to change her course and be delayed. The above is a good translation of Rondelet’s old and very difficult French *. In another place, speaking of Oppian’s Remora, which he identifies as the lamprey eel, and which is said to stop and hold back vessels, Rondelet affirms that this is “a thing which corresponds to our lamprey and which I have known through experience, for if it puts its mouth against a boat it stops it, and J have seen it thus.” Here for the first time we have an eye-witness account of the ship-retarding power of a fish. ‘The lamprey has around suctorial mouth by which it transports stones to make its “nest” at the breeding-season, and by which it fastens itself to fishes. That it should tius fasten on to a vessel is by no means improbable, nor is itimprobable that by violent motions it could slow up the speed of a small boat. The ‘ De Subtilitate Rerum, Liber X.’ of Jerome Cardan * For this translation I am indebted to Miss Hinda Hill, head of the Department of French in this College. Myth of the Ship-holder. 293 seems to have been first published in 1550; however, it was included in his complete works published in 1564 at Lugduni. On page 117 of this edition he has a column devoted to the Remora and its activities. He describes at some length and in bad Latin how the Remora by adhering to the rudder and waving its tail to right and left, turns the ship in first one and then the other direction, thus causing it to waver and lose speed. He compares its action to that of the steersman of a boat, who, using an oar over the stern, influences her course more than all the rowers who are pulling hard. Gesner (1558) quotes Rondelet at length, but somewhat simplifies the explanation of the latter, saying that when the Echeneis affixes itself to the stern or rudder, and when it moves body or tail it causes the vessel to stand still, or, at any rate, to waver in its course, “just as when in a calm the helmsman turns the ship in her prosperous and swift course over to a more inexperienced steersman who is not able to hold the tiller straight,” and hence the ship has a wavering movement and does not make good progress. Imperato (1599), who, as previously noted, was the first to explain how the Remora fastens itself to vessels or fishes, says :—‘“‘It has on the upper part of the head tentacles, similar to the vibratile combs [cirri, literally ringlets] of the polyps, by which it attaches itself to ships or to the bodies of whales and other large fishes and retards their course and restrains them at will; not otherwise than the rudder, while projecting but little from the vessel, has the power of directing its course.” The next writer to proffer the explanation we are discussing is Aldrovandi (1613). However, he starts by quoting Aristotle on the use of the rudder in changing the motion of a ship. He then advances tie same arguments which we have found in Gesner and which the latter expanded from Rondelet. However, Aldrovandi argues at considerable length and somewhat ingeniously, but the gist of his argu- ment is that the Remora sticking fast to the stern or rudder by moving its tail or body moves this continuous thing, the ship, causing it to hesitate or even pause in its course. It must be said, however, that Aldrovandi’s Latin is so im- perfect, and hence so hard to translate, that it is hard to say how much of this is Gesner and how much Aldrovandi. With the rise of the Renaissance, and the freeing of men’s minds from many old-time superstitions, it began to be seen that it was an absurd impossibility any longer to think that one small fish could retard, much less cause to come to a 294 Mr. EK. W. Gudger on the standstill, a large vessel. And so we find Rochefort (1665) remarking (as noted heretofore) that Remoras “adhere to vessels as if they wished to arrest them in their course.” Du Tertre, who was a contemporary of Rochefort, and whose book was published but two years later (1667), had seen a number of Remoras attached to ships m the West Indies, but had never known of a vessel which had been brought to a standstill by them. So he preferred to think that such vessels “had been detained by some miracle or charm.” Third Explanation: Large Numbers of Adhering Remoras. Dampier, whose ‘ Voyages’ was published in 1697, tells us that he found great numbers of Remoras in the Caribbean Sea and the Gulf of Mexico, and goes on to say with regard to their retarding power :— ‘“‘ Any knobs or inequalities at a Ships bottom are a great hindrance to the swiftness of its sailing; and 10 or 12 of these [Remoras]| sticking to it, must needs retard it, as much in a manner as if its bottom were foul.” And in this con- clusion Catesby (1754) fully agrees. Le Maire (1695) remarks that ‘‘ Le Sucez,” if it attaches itself to the rudder, may retard the vessel but cannot stop it, as the old legend falsely had it concerning the Remora. While Leguat (1721) emphatically says that “It is very certain that these fish attach themselves often to vessels in the water, and when the number is sufficiently great, one cannot doubt that they are an obstacle to the course of these floating edifices, since they prevent their easy movement over the waves.” John Barbot (1732) is also very emphatic on this point. Referring to the common notion that the Remora by sticking to a ship can retard it, he says, “....some part whereof might be possible, if a sloop or small vessel had a thousand or more sticking to its sides and stern, they being commonly, at full length, about 8 foot long or better, for then they might considerably retard the sailing of such a vessel; but it is ridiculous to say that they can have any power over great ships under full sail, as is pretended.” In close agreement with Barbot is the great French naturalist Lacépéde (1829), who in turn is probably quoting from the naturalist Commerson, from whose manuscripts most of Lacépéde’s information with regard to foreign fishes seems to have been obtained. After discussing the various Myth of the Ship-holder. 295 myths concerning the “ ship-holder,” the French ichtlyolo- gist goes on to say “Tn the midst of these ridiculous suppositions, one truth however stands out; that is that on the instant when the keel of the vessel has adhere to it, so to speak, a great number of echeneises, it would experience in moving through the water a resistance comparable to that which a great number of shelled animals [barnacles?] would make if attached equally on its surface, when it glides with less speed through a fluid which erating on the asperities brings it about that the vessel does not possess the same ‘ liveliness.’ But one does not fail to think that the circumstances under which the echeneises would find themselves thus accumulated [in such numbers] against the timbers and exterior of a ship would be extremely rare in all latitudes.” On this matter Lowe, in his ‘ Fishes of Madeira’ (1843), after reviewing many of the Greek and Roman legends, makes the following conservative statement :— “*.,.. there is much doubtless of mere fiction or exag- gerated fancy ; yet, on the other hand, it would be rash altogether to deny the truth. Like most popular accounts or vulgar errors, they may probably be founded on some real circumstances, or “natural occurrence, distorted by exaggeration into the wonderful. There would be nothing marvelous, that a Lamprey, of even ordinary size, fixed to the keel or rudder of a boat, suspended by one end and struggling in the water should, as related by Rondelet upon his own experience, greatly retard such vessel’s progress, render its course unsteady, and baffle the exertions of the rowers. ‘‘ Again it is remarkable that the Dalmatians at this day, as Sclineider in his note on Aelian, II. 17, mentions on the authority of the Abbé Fortis, possess the same idea regarding a fish they call Paklara, which the ancients held regarding their Echeneis or Remora. So strange a notion is not likely to have originated from communication with others amongst a wild and illiterate population; or, again, to have sprung up spontaneously and independently without some real ground. Without recourse, therefore, to the marvelous or extraordinary on the one hand, or to mere fiction on the other, it does not seem unreasonable to suppose that the accidental attachment to the rudder of a smail sized vessel of some fish like Rondelet’s Lamprey may have originated an impression, which has subsequently been generalized and transferred to other sucking-fishes, in themselves incapable of producing like effects.” 296 Mr. E. W. Gudger on the The soundness, the reasonableness of the conclusions reached by the various writers in the immediately preceding pages will appeal to every reader, but it must be remarked that these are all conjectures, not facts observed and recorded by scientific men. However, just here I am fortunate in being able to give the following quotation from one of the most eminent ichthyologists of the present day, Mr. David G. Stead. In his ‘ Fishes of Australia’ (1906), pages 190, 191, we read :— “Now, though it would be altogether impossible and out of all reason to suppose that one individual [Echeneis | could exert sufficient power to delay or retard a vessel’s progress, still an instance has actually come under my notice, in which a sailing-vessel was considerably delayed while in tropical seas through a shoal of ‘ Suckers’ attaching themselves all round its sides and bottom.” Unfortunately, I have had no experience of my own as to the retarding powers of this fish, but in the summer of 1915 I carefully questioned (avoiding all leading queries) one of the most experienced fishermen at Key West, Fla. We had just caught a large shark, and were vainly attempting to hook its sucking-fish attendant, when I related the story of the ship-holder, cast some doubts on it, and asked Griffin what he thought of it. He replied about as follows :—“ They sure will hold a boat. I have seen ten or twelve under a boat at one time. This was while king-fish fishing at Bahia Honda. ‘The king-fish were in big schools and were followed by hundreds of sharks. The ‘suckers’ on the boat came from the sharks. My brother and me had boats just like each other in size and build, but his was a little better sailer than mine. The first day he beat me, both sailing before the wind, but the second day I beat him. He said, ‘No wonder 1 am losing, too many “suckers”? hanging on her bottom,’ All the Key West fishermen know that ‘ suckers’ will sure hold a boat.” This was corroborated from his own experience by my captain, au educated young Englishman from the Bahamas. Aud both men agreed that of two fishing-boats of equal size aud speed, the one having behind it a “trolling squid” for mackerel will be retarded and will lose in a close race. In order that the reader may get a elear idea of the “brake” which a good-sized sucking-fish may put on the movements of its host, figure 7 (Pl. XVI.) is introduced just here. ‘This is from a photograph of a model in the United States National Museum of a shark with its adhering Hcheneis. The fish is about half the size of the shark—say, Myth of the Ship-holder. Zoe 3 feet to the shark’s 6. Argument is not needed to establish the idea of a “ brake.” ‘The figure is from a note by R. I. Geare in ‘Scientific American’? for 1902. Mr. Geare remarks that the shark often becomes ‘emaciated from the strain of pulling these uninvited guests around.” However, it should be stated that in the fizure here given the Echeneis is much larger in proportion to the size of the shark, so far as my experience goes, than is the case ordinarily. Echeneis is known to attain a length of 3 feet. A Remora half that size would be extraordinarily large. On the other hand, however, mention should be made of the fact that, while these semi-parasites are small, not infrequently several may be found on one shark. Ou ashark taken at Tortugas I found three, while one at Key West was infested with four, the largest about 30 inches long. Scattered throughout ancient and medieval literature are a number of more or less isolated explanations of submarine cliffs, of maguetic rocks, and of supernatural aud inexplicable forces which held vessels as if anchored. These are widely scattered and little emphasized, and it does not seem worth while to go into them. A fair example is that of Kecker- mann (1614), who alleges that the Remora sticking to the stern of the vessel pours out a very viscid and cold humour which causes the water around the rudder to be congealed, making the vessel to lose steerage. Again, Johnston (1657) notes that the lodcstone has the power of attracting things, and thinks that the Remora has some such non-understand- able power. Fourth Explanation: ‘* Dead-Water.” From the foregoing accounts no one can doubt that a school of Remoras attaching themselves to a small vessel. can seriously arrest it in its course, but that they could noticeably retard a large sailing-vessel or a steamer is absurd. ‘However, there is not lacking evidence from the days of Pliny to the present time that large sailing-craft and in our times even steamboats have been mysteriously checked in their courses and even stopped almost or quite still. These being facts, it is necessary to find an explanation for them. ‘This is to be found in the “ Dead-Water ” of sailors. The phenomenon of “ Dead-Water,” in which a sailing- vessel loses velocity and in a light wind may even come to a stop, and in which even a steamer may be retarded, has long been known to seamen. Probably the earliest notice of this is to be found in Chapter X. of the ‘Agricola’ of Tacitus, where, 298 Mr. E. W. Gudger on the in speaking of the geography of Britain, he says :—‘ Thule | Norway ?] was also seen, previously hidden by snow and winter ; but the seais said to be tough and hard for the rowers and to be little stirred by winds.” Nansen, in his Norwegian North Polar Expedition (1893- 1896), repeatedly noticed this phenomenon. On his return he turned over this problem to V. Walfrid Ekman for explanation. Ekman’s paper may be found in the ‘ Scientific Results’ of the expedition, volume v. (1904), and from it the following interesting data are taken. In order to ascertain the prevalence of this phenomenon, Ekman published appeals for information in thirty-six foreign and in all available Scandinavian newspapers. From the former he received nine answers citing the appearance of ‘“‘ dead-water”’ in ten different localities, while from Seandi- navian waters no less than thirty-two regions are reported to abound in this phenomenon. From this data Ekman concludes that ‘*.... From some reason or other it (dead- water) is comparatively seldom met with beyond Scandinavia or appears in a less decided manner than in the Norwegian Fjords.” Foreign reports give dead-water as occurring off Taimur Island on the coast of northern Silesia, also in Kara Sea and Bay in the same region, on the Murman coast of north- west Russia, as very “‘ troublesome .... . off the great river mouths of South America,” while off the mouth of the Orinoco a ship had to anchor to prevent drifting out of her course, ‘I'his phenomenon is reported from the Gulf of Mexico and it has been experienced off the Baffin Bay coast of Labrador, while the Saint Lawrence mouth is designated by one Norwegian captain as one of the worst regions in the world for dead-water. Two circumstantial accounts are cited for this phenomenon off the mouth of Fraser River ‘and another near Vancouver Island, in which localities it bears the familiar name used by Ekman. There are two reports of its occurrence in the mouth of the Congo, one for the mouth of the Loire River, and two for the Garonne River and the basin of Arcachon near Bordeaux. These last instances, however, are not of such pronounced dead-water as in the following report of its oceurrence not merely in the Mediterranean but between the island of Cerigo and the southern part of Greece. ‘This very cirecum- stantial account is, because of its pertinence to the Myth, given verbatim :— “On January 2, 1858, we were between Cape Matapan and Cerigo and sailed eastward for the Archipelago. The Myth of the Ship-holder. 299 wind was W.N.W., a gentle breeze and water quite smooth. We had all sails set and made about 34 knots. At 10 a.m., when we were about 12 naut. miles 8.W. of Cerigo, the brig no longer answered her helm and began to go up northward to the wind. We worked the helm but to no avail. We backed the yards and shivered the braces and made all conceivable manceuvres, but the ship only turned a little and went back again. The little wind we had, seemed to be the same as before, and there were many ships in company both to port and starboard of us, which sailed away, whilst we were lying as if at anchor. Yet’ there was one sail about 3 miles to port of us in the same predicament. “In this manner we lay for 1# hours, when the ship began to glide and fall to leeward a little. Wethen got the head sails filled and had the aftersails shivering, and without any command of the helm the vessel got down into its course. The most remarkable thing was, however, that when I stood afore, I sawa long stripe stretching from the bow far over the water on each side dividing the water into two parts. The water around the ship was light gray, but ahead of the stripe it was wholly dark. These stripes seemed by and by to move aft... . of course it was the ship that began to glide slowly onward ox > and after 5or-6 minutes when the stripes had passed along the ship and had left the stern and the rudder, then, at that same moment, the ship again answered her helm and made head-way. The wind was about the same—W.N.W. by W. a gentle breeze. We made 3 knots, but no more, in the afternoon. ‘‘ When we approached Cerigo, the ship was about to get into dead-water again, but by working the rudder to and fro, we steered again, and after that, we did not feel the (lead-water any more. “The ship, during its long voyage, had become very dirty and overgrown with barnacles of 10 or 15 cm. in length, which may have had some effect.” From Ekman’s quotations from his correspondents as to the occurrence of dead-water around Scandinavia, the following short excerpts are taken. In perusing them the reader is asked to bear in mind the very words of the quotations concerning the actions of ships found in the first section of this paper. The ‘ Fram’ being in dead-water off Taimur Island.... “It may therefore be supposed that the speed was reduced to about a fifth of what it would otherwise have been” : and when steam was cut off at 100-150 metres from the buoy, the speed was so reduced that the engine had to be 300 Mr. E. W. Gudger on the started to reach it. ‘ Sailing vessels may . . . be seen stuck fast in spite of a breeze brisk enough to keep the sails fully BEVAINE 2): Sona Sometimes it happened that one vessel gets into dead-water and another not, though it is impossible to discover any reason for this.” “..... we already had good speed, when all at once the ship took. dead-water... . she stopped so quickly that it looked as if she had dropped anchor.” ‘Che vessels being becalmed, ‘‘ One of them was towed away without any difficulty, while the other, though of similar size, got into dead-water, and an extraordinary amount of work was required to get this vessel from the spot.” Another ship in dead-water drifted back four miles with the current “ against the direction of the steady fresh breeze, although they had all sails set.”’ Another observer writes that in dead-water it “.... feels as if something were fastened to the ship and holding it back.””. ‘In such cases, one or more vessels might suddenly lose their steering and remain on the spot, while others pass freely through the midst of them at a distance as short as two or three ships’ length. After a while it was the turn of the other vessels to get into dead-water.” ‘ We scarcely glided along and were forced to have all sails set, until we were quite near our anchorage. ‘Then the dead-water suddenly let go its hold. Believe me, they were both in a hurry, the ship and the pilot. Braces and falls ran a race together, and we only just got the anchor dropped without any misfortune.” “The brig got into dead-water..... The speed was lost, and the ship was as if nailed to the spot.” When the dead-water let go with the sails drawing, “.... it all at once appeared as if the vessel had cut loose from a mooring aft.” An 8-knot steamer in dead-water “.... according to the pilot’s own phrase, hardly moved from the spot.” Other descriptions might be quoted, but, save the one now to follow, these are the most typical. The one now to to be given, with a sketch showing the appearance of the water around the vessel, is from the pen of Kommandor- kaptein Joh. Kroepelien of the Norwegian Navy. He writes that the ship with all sails set, heeling over rather stiffly before a fresh breeze “..... all of a sudden, lost her headway without any perceptible external cause, and the turning power of the rudder became nil. “We then perceived that the ship had taken dead-water. From about amidships and outwards on both sides and to a considerable distance aft she was surrounded by a mass of dead-water, smooth as glass, as if the surface were covered with oil. The line betweeu this smooth surface and the Myth of the Ship-holder. 301 water farther out, looked like boiling ‘rips’ and was quite distinct, the outer surface being strongly rippled by the breeze. ‘The roar caused by the ‘dead mass of water which, clinging to the ship, was dragged along through the water outside, was so loud that it might well have been deemed we were in the vicinity of a rapid, I do not remember the appearance of the wake, nor, I believe, was there anything remarkable about it. The rudder was of no use; we were forced to handle the ship by means of the sails and our two boats towing from the bow, and thus we proceeded at a speed of one or two knots. “In this manner we went on for a couple of hours. All of a sudden, without any known cause, we were set free from the dead-water. The wind had been very steady the whole time, and we had constantly endeavored to keep the ship in the same course. After being freed from the dead-water the ship got headway, and after a while we logged 7 knots, going close to the wind. Captain Kroepelien’s sketch is reproduced herein as fig. 8 (P]. XVII.). Concerning such an appearance as is here shown, Ekman writes: ‘As the boundary waves (to be described and explained later) follow the vessel, their wave erests and wave hollows remain in an invariable position relative to the vessel. Ifthe wave motion gives to the water at a particular spot a velocity with the vessel, it would appear as though a bulk of water were being dragged along with her, although it is really always a new mass of water which follows the vessel for a short distance. It is exactly analogous to a boat sailing before the wind with just the same speed as the breaking waves at her side. In the case of dead-water, on the other hand, the illusion will be more complete, because the vessel moves at a slow velocity, and the waves causing the motion of the water are themselves not visible.” In perusing the foregoing accounts, the reader cannot have failed to be struck by the capriciousness of the pheno- menon of dead-water, its sudden and seemingly inexplicable appearance, its equally sudden aud mysterious disappearance. It may cause a ship gradually to lose speed, or suddenly to be stopped stillas if ‘ nailed,”’ “ moored,” or “anchored ” to the spot. ‘The ship may gradually rejain her speed or may suddenly speed away ‘Sas if a mooring had been cut.” Again, a ship may fall into dead-water while a near neigh- bour but a few cable lengths away may sail on her course without ‘ let or hindrance.” The instances just quoted, closely, almost precisely, parallel 302 Mr. E. W. Gudger on the the accounts from the old writers given in the first part of this paper, and there can be no doubt that their phenomena were bona jide occurrences of dead-water. One cannot wonder then that when a ship was thus checked and an Echeneis found, as it was not unlikely to be, sticking to rudder or hull, that the fish was deemed the cause of the checking of the speed of the vessel, and that the myth grew and became widespread. Thus far we have been occupied with Ekman’s accounts of dead-water, now let us consider his explanation of this strange phenomenon. After a study of some 42 accounts and descriptions, foreign and domestic, he generalizes as follows: “..... IT conclude that dead-water may occur in every place where fresh water flows out over the sea, but that for some reason or other it is comparatively seldom met with beyond Scandanavia or appears in a less decided manner than in the Norwegian fjords. ... . Dead-water only appears near to coasts, in those places where a suitable layer of fresh or brackish water rests upon the heavier sea- water. 9? 12. In his ‘ Researches on the Fossil Remains of the Extinct Mammals of Australia,’ &c. (1877) another interchange was made, thus :—— For pl. x. fig. 11 read pl. ix. fig. 12. 99 39 9 >) 99 bb) 13. There remains the smaller “ distal phalanx ” referred to by Krefft in the ‘ Caves and Rivers Report.’ This specimen is 20 mm. long by 14 mm. in breadth, inclusive of the sheath or hood, which is complete proximally, but brokeu away towards the distal end of the bone. It is similar in shape to the nuail-bone called Mylodon by Krefft, but with a greater degree of curvature, and less size. The articular surface, just as in that previously referred to, occupies nearly the whole of the proximal end, and is divided into two subarticular surfaces by a median longitudinal ridge for adaption to the convexities at the distal end of the penultimate phalanx. The tuberous process for the flexor tendon attachment is remarkably prominent and stout in comparison to the size of the entire phalanx ; on the plantar surface of this tuberosity are the two arterial foramina. Krefft considered this to belong “to either a dog or cat-like creature.” With this last exception such are the phalanges described by Krefft as Mylodon australis, a supposed Australian Edentate, and referred by Owen to his Thylacoleo carnifex by deduction. In considering the affinity of these bones, the following general conclusions may, 1 think, be fairly arrived at :— 1. The law of probabilities is decidedly adverse to Krefft’s view. Had an Kdentate existed in Australia in Post-Tertiary times, some more definite trace would have been met with ere this. 2. Aright calcaneum, referred to this genus by Lydekker, is all we know of the feet of Thy/acoleo, and this determination is problematical *. * Lydekker, loc. ett. p. 195. some Ungual Phalanges. 313 3. The reference of Krefft’s Mylodon phalanges to Thyla- coleo on the part of Owen was purely ‘‘ conjectural ”’ (to use his own expression), but at the same time a clever piece of deduction based on his view of the carnivorous habits of the ‘*‘ Marsupial Lion.” 4. If we accept for the time being, the phalanges called Mylodon? australis as those of Thylacoleo, such acceptance will not in the least strengthen the views held either by Owen on the one hand, or Flower and his followers on the other, as to the gastronomical habits of Thylacoleo, hooded phalanges occurring amongst both herbivorous and carnivorous animals. 5. As possibly referable to Thylacoleo Owen figured two entirely distinct types of ungual phalanges. We are now acquainted with the pedal bones of Dipro- todon through the researches of Prof. E. C. Stirling, and it can be legitimately surmised that those of its second cousins Nototherium and Euowenia were similar. None of the Macropodide can put in a claim ; amongst the flesh- eaters, Sarcophilus and Thylacinus, and the Dasyures, with the non-marsupial Warrigal, the osteological structure is too well known to require comment. Finally, in all probability, although “ conjectural” Owen’s view of the nature of the hooded nail (eliminating those without a sheath) bones will in the long run prove to be correct ; reduction of other genera by elimination supports it. If such be the case, then what is the claw referred to Thylacoleo by Lydekker? This will be investigated immediately. The following is the synonymic bibliography of Krefft’s ungual phalanges :— Mylodon? australis, Krefft, Guide Austr. Foss. Remains, 1870, p..4.) - “ australis, Krefft, Austr. Vert. Foss. & Recent (Industrial Progress of New South Wales), LS7 is parls. Spelean Animal or Unguiculate Mammal, Owen (pars), Phil. Trans. 1871, pt. i. pp. 262, 263, pl. xiii. figs. 11-12 (non 138, 14). Megatheroid Animal, Krefft, Ann. & Mag. Nat. Hist. (4) x. 1872, p. 180. Spelean Animal .... Thylacoleo, Owen, Foss. Remains Kixtinct Mamm. Austr. i. 1877, pp. 182-183, i. pl. ix. figs. 11-12. 314 Mr. R. Etheridge, Jnr., on Mylodon? australis, Krefft, N.S. Wales Votes & Pro- ceedings, v. 1882, p. 558, 14th numbered pl., figs. 7-9. Il. THe Uneuar Poaranx ProvisionaLty CaTALoGuED AS THyLACOLEO BY LYDEKKER. Many years ago a plaster replica of another of Kreifft’s specimens *, described in MS. as the ‘‘ nail-bone of the hind foot of a gigantic Phalanger, probably a small Zygomaturus, Nototherium, or Diprotodon” was forwarded to the Geologi- cal Department of the British Museum (Natural History). I surmised this might be No. M. 1526 f of the ‘ British Museum Catalogue of Fossil Mammals,’ part v. (p. 195) catalogued by Lydekker as “ cast of an ungual phalangeal ” provisionally of Thylacoleo; by correspondence Dr. Arps. Woodward confirmed this. The original bone is preserved here and is slightly imperfect ; it is from the Wellington Caves, and bears the number A. 13320 (Pl. XVIII. fig. 2). It is manifest, if the sheathed nail-bones (“ Mylodon”’) are referable to Thylacoleo, following Owen, such an arched, laterally compressed and naked bone, one of those spoken of by Krefft as “ large nail-bones ... evidently those of a Phalanger” {, cannot. One of these§ is probably the original of both Owen’s illustrations of his non-sheathed Thylacoleo ungual phalangeal. Our collection contains five of these bones, four from the Wellington Caves ossiferous breccia (Pls. XVIII.—XIX. figs. 2-9), the fifth from Cope’s Creek, probably froma thermal mud-spring deposit (Pl. XX. figs. 10-12). These vary much in size and degree of dorsal curvature, and for tlhe convenience of description may be taken separately. Type 1.—-The phalanx in question || (Pl. XIX. figs. 8 & 9) is highly arched, compressed laterally, the dorsal edge thin, sharp (trenchant), the degree of curvature almost equai to the quadrant of a circle, the general appearance of the bone being decidedly hook-like. ‘lhe proximal end is imperfect, the articular surface and the plantar tuberosity gone; it is 35 mm. wide, and in thickness 8 mm. The second example never seen by Krefft or Owen (Pl. XX. figs. 10-12) is a more perfect specimen, one in * Krefft, ‘ Caves and Rivers Report,’ pl. 14 (numbered), fig. 12. + Dr. A. 8. Woodward informs me this should read 1536, { Wrefft, doc. cit. pl. 14. (numbered), tigs. 11 and 12. § Krefft, Joc. cit. MS. description of pl. 14. fig. 11. || Krefft, foc. cit. pl, 14. (numbered), fig. 2. some Ungual Phalanges. 315 which the proximal articular surfaces, allowing for wear and tear, ave perfect. The lateral surfaces (at the. point of disruption in Pl. XIX. figs. 8 & 9) suddenly bulge outwards to form an expanded proximal end with a concave articular surface divided by a longitudinal central ridge, and below a very strong and comparatively large cushion for the attach- ment of the flexor tendon. Immediately above the centre of the tendon tuberosity on either side, are the foramina of the digital arteries. The surface of both specimens is pitted aud roughened. Length of complete bone 51 mm.; breadth 45 mm. approximately ; thickness 13 mm. Type 2.—The phalanges of the second type (Pls. X VI[I.— XIX. figs. 2-7) differ from those of the first by a greater length in proportion to width, a much less arched dorsal edge, and a slightly less lateral compression, otherwise the same features characterize both. The following are the dimensions of the largest :— Length 45 mm.; breadth 29 mm.; thickness 11 mm. In the sheathed, or hooded terminals of Owen, although the nail-bone is compressed laterally (Pl. XVIII. fig. 1) the dorsal edge is only sharp or trenchant distally, the proximal eud is truncate-flattened forming an elongately triangular surface. The articular surface for union with the distal end of the penultimate digit is highly concave, and much overhung above, as figured both by Owen and Krefft. The sheath is one with the core, or nail-bone, at the proximal end around the articular coucavity, and along the plantar surface as far as it extends; the tuberosity is to some extent Hattened. ‘lhe arterial foramina pierce the sheath through the plantar surface of the tendon tuberosity, and then appear to enter the nail-bone as im the preceding type. Immediately below the dorsal truncate surface at the proxi- mal end are two other arterial foramina. Now, to what type of Marsupial do these ungual phalanges (Pls. XVIII.~XX. figs. 2-12) belong? It will be more satisfactory to consider Types 1 and 2 separately. Type 1 (Pls. XIX.—XX. figs. 8-12) is the ‘‘ nail-bone of a gigantic Pialanger of Krefft,” but this form appears to have been quite unknown to Owen. In the Macropodide the nail- bones are elongate, non-trenchant, more or less oval in section, and very feebly arclied, if at all. The nail-bones of the Peramelide are double, more or less circular, and non- trenchant. In the Phascolomyide, or Wombats, these terminals are again rounded above, roughly oval in section, and not hooked. The nail-bones of the Diprotodontide, 316 Mr. R. Etheridge, Jnr., on guided by Prof. E. C. Stirling’s reconstruction of Diprotodon, resemble to some extent those of the Kangaroos, plano- convex, slightly curved, broad plantar surface, and the proximal concavities occupying the whole articular surface, instead of about two-thirds as in Types 1 and 2; moreover, the position of the foramina of the plantar artery branches is markedly different. What is true of the nail-bones of Diprotodon is possibly equally true of those of Nototherium and EHuowenia. , There remain the Dasyuride and Phalangeride. In the first, taking the Tasmanian Wolf ( Thylacinus cynocephalus, Harris) as an example, the nail-bones are long, more or less oval in section, rapidly decreasing in size from the proximal to the pointed distal end. The latter are more particularly accentuated in the Tasmanian Devil (Sarcophilus ursinus, Harris, Pl. XIX. fig. 14), in which the distal ends of these nail-cores are to all intents and purposes, acicular ; hence, I dismiss the Dasyuride from consideration. This reduces comparison to the Phalangeride, the family in which Krefft placed * these remains. The resemblance of the large complete specimen (PI. XX. figs. 10-12) from Cope’s Creek to similar bones of some members of this family is very striking. For the purpose of comparison I have selected two, the Great Flying Phalanger (Petauroides volans, Kerr) and the Koala, or “ Native Bear” (Phasco- larctos cinereus, Goldfuss) +. In the Flying Phalanger it is the 4th and 5th digits which terminate in nail-bones so remarkably like the Cope’s Creek fossil (Pl. XX. figs. 10-12), but in the Koala the resemblance is not so strong (Pl. XIX. fig. 13), in consequence of the much greater length in proportion to. width; this, however, only partially holds good for the pollices t. With these facts before me I can come to no other conclusion than that the subjects of Pls. XIX.-XX. figs. 8-12 are the terminal phalanges of an euormous Phalanger, following Krefft in this opinion, but in a more restricted sense than he employed the term. We may now pass to the second type (Pls. XVIII.—XIX. figs. 2-7). The two specimens are Krefft’s ‘“ large nail-bones * Bearing in mind that Krefft included Diprotodon, &e., in this family. i Gh fact in connection with the terminal phalanges, or nail-bones, of the Phalangers in general is very obvious, the stouter and stronger build of those of the fore feet, accompanied with a greater degree of curvature. { One of the most noticeable features in Type 1 is the remarkable slab-sided, or straight-walled appearance. some Ungual Phalanges.. 317 ... evidently those of Phalangers,” and one (Pl. XVIII. fig.5) is Owen’s Thylacoleo “ungual phalangeal”’ (his fig. 13) and Lydekker’s Thylacoleo “ ungual phalangeal.””’ By the same method of elimination as observed in the case of Type 1, I reduce consideration in this instance to the Phalangeridz alone. There is no greater degree of variation between Types 1 and 2 than there is in the forms of the terminals of the same foot of. many species of Phalangeridz. I, there- fore, again support Krefft’s views of the affinity of these bones, but to what genus of the family the animal possessing them was most nearly allied only time can prove. For my own part [ am rather in favour of a gigantic Koala. The following table explains the relative identity of the various figures referred to :— Krefft’s figs. Owen’s figs. Owen’s figs. Tene Austr. Foss. Phil. Trans. Extinct Mamm. ie Remains. 1871 (1872). of Australia, Bs: MMe eet ee |), a seed bes Figs. 8 & 9. RA a tae hs ee) ats eulgn Pe ts seals 8 PL. xiii. figs. 11 & 12.| Pl ix. figs. 11 & 12. | Fig. 1 Salt Oe PSE ea tere ee a aie ces cee een ce 11] ? PL xiii. figs. 13 & 14.| ? Pl. ix. figs. 13 & 14. |? Figs. 5-7. MOL eee, Geert Wid. viet crit Figs. 2-4. In these notes I have sought to show that :— 1. Owen figured as the possible ungual phalanges of Thylacoleo two entirely distinct nail-bones—a ‘‘ hooded” form, and an unhooded or unsheathed one; both cannot belong to the same kind of animal. 2. If the hooded bone be accepted for the time being as of Thylacolev, then the bone catalogued as “ cast of an ungual phalangeal ” by Lydekker cannot possibly be so. 3. The non-sheathed terminals (Types 1 and 2) were never claimed by Krefft as appertaining either to his Mylodon australis, or to Thylacoleo. 4. Thylacoleo is regarded by the advocates of its herbi- vorous nature as a member of the Phalangeride. If it be so, then the phalanges of Types 1 and 2 may, perhaps, be those of it. 5. If the suggestion contained in the last paragraph should * This is the original of the replica called by Lydekker Thylacoleo (A.M, 13320, B.M, 1526 (86)). 318 On some Ungual Phalanges, prove correct, it follows that the identity of the hooded bones (*‘ Mylodon australis ’’) has yet to be discovered. The suggestion of an extinct Koala may possibly be not so speculative as would at first sight appear when it is remembered that Mr. C. W. de Vis described * a portion of a fibula that he believed represented “ a progenitor of the Koala.” The further discovery of a premaxillary with its palatal process was held to strengthen this view. Said Mr. de Vis :— The Koala, or Native Bear, is now one of the few types of Australian life which has not been recognized as a part of its ancient economy: yet it is one of which no one could be surprised to find an ancestral form among the past modifications of marsupial structure.’ He proposed to distinguish the former owner of this fibula by the name of Koulemus ingens. Portion of a shoulder-blade was referred to another extinct Pha'anger (Archizonurus securus). EXPLANATION OF THE PLATES. Fig. 1, The original of Krefft’s “ungual or terminal phalanx of a creature allied to Mylodon,” with “its peculiar protecting bone partly broken.” The original of Krefft’s figs. 7 and 8, and Owen’s 11 and 12. Wellington Caves. Xx 2 diam. Fig. 2. Ungual phalange “equalling or surpassing those of a Lion r (Owen). Thisisthe original of Krefft’s tig. 12, and Lydekker’s Catalogue (M. 1526 (36)). Wellington Caves. x 2 diam. Fig. 8. Dorsal view of the bone, fig. 2. x 2 diam. Fig. 4. Plantar o Es is Fig. 5. Another phalange similar to Fig. 2, but with the dorsal surface straight, or even a little concave. This is probably the original of Owen’s figs. 13, 14. Wellington Caves. x2 diam. Fig. 6. Dorsal view of fig.5. x 2 diam. ee lpia’ ees 3 . Highly compressed ungual phalange with the proximal portion broken away. Original of Krefft’s fig. 2. Wellington Caves. *lebasi, 36. *correptus, 28. lineatus, 20. *curvidens, 31. longicornis (24). External Characters of Ruminant Artiodactyla. 367 *luteicauda, 17. rubripennis (9). *luteoguttatus, 16. rubrofasciatus (24). nigricollis (19). rugosus (1). ¥nigrolimbatus, 11. sexguttatus, 15. notatus (18). sexmaculatus, 3, octopustulatus, 35. *sexpustulatus, 14. pallipes, 22. speciosus (24). pietus (3). sptnosus (34). porrectus (2). splendidus, 6. quadrilineatus, 27. subgriseus, 23. quadriteniatus (26). trifasciatus, L, quadrivittatus (26). variegatus, 18. revolt (18). vittaticollis, 26. riveti, 13. vittatus, 21. rubripennis, 8. Horsell, Aug, 1918. XXXIII.—On some External Characters of Ruminant Artio- dactula.—Part IV. The Reduncine (Cervicaprine) and Aipycerine. By R. I. Pocock, F.R.S. As in the previous papers of this series published in the ‘Annals’ for June, August, and September of this year, the pagination subjoined to the specific headings refers to my treatise on the Cutaneous Glaads of the Ruminants printed in the Proc. Zool. Soc. for 1910. Subfamily Repvycrvz (olim Cervicaprine). Genus Peusa. Pelea capreolus, Bechst. (p. 911). A second specinien of this species, which came into my hands since 1910, enables me to confirm in every particular the characters of the genus, based on external features, which I pointed out in that year. Since this specimen, like the first, had no trace of inguinal glands, I think it may be assumed that Owen’s statement as to their presence was false. . The only fact I have to add to my original description is that the false hoofs on both the fore and hind feet are united across the middle line. Genus Exzorracus, Gray. Eleotragus arundinum, Bodd. In 1910 I was not in a position to incorporate an account / 368 Mr, R. I. Pocock on some of this species in my paper. The examination, however, of an adult female specimen in 1911 revealed some interesting features connected especially with the rhinarium and inguinal glands. The rhinarium (fig. 1, C), as in all the Reduncine, has a narrow philtrum, but it recalls that of Peleain the backward extension of its upper surface a long way beyond the poste- rior angle of the nostrils. This area of it, however, is not so inflated as in Pelea. iui D ND A, Extremity of penis of Kobus defassa from the left side. B. The same of Redunca redunea. C. Rhinarium of Eleotragus avundinum from the right side. x 2, - The same of Redunca redunca from the front. xX 2. . The same from above. x 3. F. The same from the left side. x 2. As in Pelea, there is no trace of preorbital glands, as Owen stated. In the feet the interdigital web is naked, as in Pelea, but there is no trace of pedal glands, and the false hoofs are not united, but separated by a narrow tract of naked skin. The feet, indeed, resemble those of Adenota and. of most examples of Redunea. External Characters of Ruminant Artiodactyla. 369 Owen correctly recorded the presence of inguinal glands in this species, but gave no particulars. They are, as a matter of fact, peculiar. On each side of the mamma, which are arranged in a quadrilateral, and rather far out from them, is a large orifice opening backwards and inwards, not outwards, and this leads into a pouch about 3 inches deep which runs obliquely forwards and outwards along the depression between the thigh and the abdomen. The area round the mamme and the glands is naked, and the secre- tion of the glands has a starchy smell, like flour-paste. For information as to the structure of the penis, see under Redunca (q. v. infra). On the strength of the information regarding the rhinarium and inguinal glands I gave him in 1914, Mr. Lydekker (Cat. Ung. Mamm. ii. p. 203) granted subgeneric rank to Eleotragus. But, as I pointed out to him at the time, the characters which distinguish the type-species of Eleotragus from that of Redunca (olim Cervicapra) are quite sufficient for generic admission. ‘The structure of the rhinarium affiliates Eleotragus with Pelea, aud distinguishes it from Redunca. On the other- hand, the absence of pedal glands and the presence of inguinal glands show affinity to Re- dunca and departure from Pelea. In the direction of the inguinal glands and in the presence of only a single pair, representing the shallow anterior pair of Redunca, Eleotragus is distinct from that genus. Genus Repvunca (olim Cervicapra) *. Redunca redunea, Pall. (p. 913). A male example of this species from the Sudan (G. Blaine), and probably referable to the race described as cottoni, re- sembles in every particular, so far as the characters under discussion are concerned, the examples of the typical race of the species from Senegambia which | described in 1910. The rhinarium (fig. 1, D, E, F), viewed from the front, has a convex upper margin ; the nostrils are about as widely separated as in Eleotragus, and, as in that genus, there is scarcely a trace of naked skin below them ; the philtrum is as wide above as the internarial septum, narrow inferiorly, and expands slightly where it passes into the gum of the upper lip; it is mesially grooved up to the level of the lower * On the evidence supplied by Palmer, I follow Lydekker in adopting Redunca tor Cervicapra, the latter being a synonym of Antilope. Ann. & Mag. N. Hist. Ser. 9. Vol. ii. 27 370 Mr. R. I. Pocock on some border of the nostril, but there is no depression on the antero-superior surface of the rhinarium ; the posterior edge of the upper surface of the latter is only slightly angular, the hairs of the muzzle extending in a nearly straight line - across between the posterior angles of the nostrils. It is in this respect that the rhinarium differs so markedly from that of Hleotragus. There is a bare patch of skin below the ear *. Of the two pairs of inguinal glands, the anterior consists on each side of a wide but shallow pouch, and the posterior of a subcylindrical but dilatable pouch about 2 inches deep, the yellow secretion having a starchy smell. Of the pedal glands no vestige remains; on the fore foot the false hoofs are united at the base, on the hind foot they are separated by a narrow strip of naked skin. The glans penis (fig. 1, B) is slightly thickened .towards the extremity, then gradually narrowed to a blunt point ; the urethral canal is produced into a short slender tube overlapping the tip of the penis to a small extent. This penis is very like that of Hleotragus arundinum described and figured by Lounberg (Ark. Zool. Stockholm, (5) v. no. 10, p. 6, fig. 5, 1909), except that the urethral process appears to be a little longer, Genus ADENOTA, Gray. Adenota kob, Erxl. (p. 915). I have nothing to add to the description of this species published in 1910 ; but it is important to recapitulate the characters upon which the genus should be sustained, although Mr. Lydekker regarded it merely as a subgenus of Kobus. It resembles Kobus in the structure of the rhinarium -(q. v. infra) and in possessing a tufted instead of a bushy tail like that of Pelea and Redunca. It differs from Kobus in having a preorbital gland, consisting of a thickened area of skin, and a single pair cf inguinal glands. In one of the specimens described in 1910 I recorded the presence of an additional vestigial or rudimentary inguinal gland, lying far out away from the mamme, on the right side. This * This patch was absent in the two examples of the typical race of this species described in 1910. This statement was evidently overlooked by Mr. Lydekker in 1914, when he cited the presence of this patch as one of the features distinguishing Fedunca trom Kobus. The naked patch is not glandular, but consists of very thin skin. Its function is unknown. External Characters of Ruminant Artiodactyla, 371 gland clearly represents one of the anterior pair present in Redunca. This very interesting fact shows that in Adenota representatives of the posterior pair of inguinal glands seen in Redunca are retained, whereas Eleotragus retains the homologues of the anterior pair of Redunca, Genus Kosus, Smith. Kobus defassa, Riippell (p. 916). In 1910 I was unable to publish reliable information as to the cutaneous glands of any species of the genus Kodus, having only the dried skin of the head of K. defassa and dried fect of K. marie for examination. Since that date I have been able to examine an adult male and female of K. defassa and an adult male hybrid between K. defassa and K, ellipsiprymnus. Preorbital gland.—Although I was unable in 1910 to discover a trace of this gland on the dried head-skin of K. defassa, I suggested the probability of the existence of a gland resembling that of Adenota kob. This suggestion, however, very clearly furnished no justification for Mr. Lydekker stating, on my authority, that rudimentary face-glands are present in the genus (Cat. Ung. Mamm. pp. 199 & 225, 1914). Fresh material proved my guess to be erroneous. Kobus resembles Eleotragus and Redunca in having no preorbital glands, as Owen long ago stated. The rhinarium (fig. 2, A, B) was described by Mr. Ly- dekker as “ normal.’ By this epithet he clearly meant unlike that of Pelea and EHleotragus. But, as a matter of fact, there are certain features about the rhinarium of Kobus which, according to my conception, are distinctly abnormal in the sense that, within the limits of the Reduncine, they are peculiar to the genera Kobus and Adenota, the rhinarium which most nearly approaches the normal in the Reduncine being found in Redunca. In Kobus the anterior surface of the rhinarium is bilobate, owing to the presence of a wide median depression up which the median groove of the philtrum extends as high as the summit of the anterior portion of the nares. ‘There is also a wide area of naked skin passing beneath the nostrils to their posterior extremity laterally. Finally, on the dorsal side the hair of the summit of the muzzle encroaches as an angular field to a point nearly on a level with the anterior extremities of the nostrils, and on each side of this field the upper rim of the nostrils is elevated. The encroachment of this hair gives a biconvex 27* 372 Mr. R. I. Pocock on some aspect to the upper edge of the rhinarium from the front aspect, the corresponding edge in Redunca being evenly convex from side to side. A rhinarium of this structure is found only in Kobus and Adenota within the limits of the Reduncine. Inguinal glands are absent, as Owen stated, and there is no trace of pedal glands. s “ee BYE Sy id Waray Wi : A A. Rhinarium of Kobus defassa from the front. i. B. The same from above. The extremity of the penis (fig. 1, A) is much more bulbous than in Redunca, with a downbent rounded apex, and the urethral canal is of unusual length, recalling that of Ovis in the extent to which it overlaps the end of the penis. The figures of the penis of this genus published by Lénnberg (Nova Acta R. Soc. Upsal. (3) xx. pl. il. fig. 4, 1904) and by Gerhardt (Verh. Deutsch. Zool. Ges. xvi. p. 153, 1906) represent the urethral prolongation as curling up on the left side of the termination of the glans and External Characters of Ruminant Artiodactyla. 378 closely applied to it. It is also much shorter than in the specimen I examined. By the characters described in this paper the genera of Reduncinze may be distinguished as follows :— 1. a. Rhinarium swollen above and extending ; back far beyond posterior angle of nostrils. Pelea, Eleotragus. a’. Rhinarium otherwise. b. Rhinarium not deeply and widely grooved in front, extending as a narrow strip below nostrils laterally ; its posterior border nearly straight between the MGISELUIS c guetye ota ieee eas ex la, wal ue oie ats Redunca. b'. Rhinarium deeply grooved in front, a wide naked strip below nostrils late- rally; its posterior border acutely an- gular between the MORONS Say mone ted Adenota, Kobus. 2. a. Preorbital gland absent ...........0006- Pelea, Eleotragus, Redunca, Kobus. a’. Preorbital gland a thickened area of skin.. Adenota. 3. a. Inguinal glands absent ...........+.05. Pelea, Kobus. a’. Inguinal glands present. b. Two pairs of inguinal glands .......... Redunea, b'. One pair of inguinal glands. c. Anterior pair of inguinal glands of Re- dunca retained as long anteriorly directed pouches ..............505> Eleotragus. c'. Posterior pair of inguinal glands of Redunca retained as short inwardly directed pouches. ...........---05- Adenota, 4, a. Pedal glands retained as flask-shaped sacs with short duct and small orifice ...... Pelea. a’. Pedal glands aborted ............ 138 aes Eleotragus, Redunca, Adenota, Kobus. 5. a. Penis with urethral tube short, slightly surpassing attenuated end of glans .... Hleotrugus, Redunca. a'. Penis with urethral tube very long, far surpassing bulbous end of glans ...... Kobus. Subfamily Mprcrrzivz*. Genus /ipyceros, Sund. Aipyceros melampus, Licht. (p. 918). The feet of a specimen of this species from British Kast Africa, brought home for me by Mr. F. C. Selous, enables * L instituted this subfamily under this name in 1910; but Lydekker, while adopting the group in 1914 (Cat. Ung. Mamm. iii. p. 4), emended the title to Avpycerotinz, but quite unwarrantably, ABpycerine being, I believe, correctly formed and having the advantage of brevity. 374 Dr. K. Andersen on new Bats of the me to confirm my description of the metatarsal glands and to substantiate the correctness of my supposition as to the structure of the fore feet, published in 1910. The fore feet are exactly like the hind feet, except for the absence of the metacarpal glands. Pedal glands are absent. A piece of the skin of the inguinal region of the same specimen showed two pairs of mamme, but no trace of inguinal glands, thus agreeing with the dried skins in the British Museum. Hence it may be concluded that Owen’s statement that inguinal glands are present in the genus is erroneous; and since he affirmed at the same time the existence of large preorbital glands, which, according to universal testimony, are absent, it seems obvious that the specimen he examined did not belong to the genus Apyceros at all, but was probably some large form of Gazella. XXXIV.— Diagnoses of new Bats of the Families Rhino- lophidee and Megadermatide. By KNuD ANDERSEN. [AT the request of Dr. Knud Andersen, who expects to be absent from his scientific work for some time, the following diagnoses are published, mostly in the form of extracts from the synopses of species prepared by him for the second volume of the ‘ Catalogue of Chiroptera.’ By this method the exact relationship of the species to their nearest allies is readily seen, together with the cha- racters distinguishing them. The “ groups” in which the species of AAinolophus are placed are those recognized (though under different names) in Dr. Andersen’s “ List of the Species and Subspecies of the Genus Rhinolophus” *, 1905.—O. T.] Genus RHINOLOPHUS. Rh. megaphyllus group. (Called simplea group in the ‘Annals’? paper, 1905.) a’. Connecting process higher posteriorly than anteriorly (at junction with sella). a. Ears longer, 16°5-21 mm, (inner margin). General size larger; forearm 40-49 mm. a®, Nose-leaves larger: breadth of sella at base 25-3 mmm., of horseshce 9 10°5, * Ann. & Mag. Nat. Hist. (7) xvi. p. 648 (1905). Families Rhinolophide and Megadermatide. 375 Constriction at middle of sella always distinct. b°. Nose-leaves smaller: breadth of sella at . base 2-2°3 mm., of horseshoe 7:7-9. Constriction of sella often obsolescent. c*, Lancet cuneate or subcuneate. d*, Lancet hastate or subhastate (constric- tion of sella obsolescent or absent). e°. Nasal swellings 5:2-5°5 mm.; c-m* * US 2 eT ae Ra Bae borneensis f°. Nasal swellings 4:9-5:°2 mm.; cm? 6:2-6'7. Lancet peculiarly short- ened (prebably nearest hastate), looking as if broader at base than long. Forearm 40-40°5 mm. (S. ren geicere fo states ais tha cise at javanicus, sp. Nn. 6°, Ears shorter, 15-16°5 mm. on inner margin. General size smaller; forearm 37-39 mm. e*, Connecting process as usual. Nasal swellings 46-48 mm.; c-m® 63-65. Forearm 38-39, (Madura.).......... madurensis, sp. 0. d°®, Connecting process rather more pro- nounced than usual. Nasal swellings 4-3 mm.; e-m? 59-63. (Luzon.) .... virgo. 6'. Connecting process broadly rounded off, as low posteriorly as anteriorly (at junction with sella). Sella distinctly expanded at middle, narrower at base than across expansions, constriction (at or above middle) very distinct. ce’. Forearm 46 mm.; tibia 20. Sella broader. (Bandon, Lower Siam.) ........5....0.% robinsont, sp. n. d’, Forearm 40-44 mm.; tibia 16-17. Sella narrower. (Pulo Tioman; P. Pemangil.) Alossi, sp. n. Types :— javanicus. Female. B.M. no. 9.1. 5.174. Original num- ber 1655. Collected 18th March, 1908, by G. C. Short- ridge at Pangandaran, Dirk de Fries Bay, S. Java. Presented by W. E. Balston. madurensis. Female. B.M. no. 10.4.7.9. Original num- ber 2164. Collected 4th November, 1909, by G. C. Shortridge at Soemenep, E. Madura. Presented by Oldfield Thomas. robinsont. Female. B.M. no. 18.8. 2.1. Original num- ber 527/13. From Kao Nawng, Bandon, Lower Siam, 13th June, 1913, Presented by the Federated Malay States Museum. klossi. Female. B.M.no.18.8,2.2. From Pulo Pemangil, June 1915. Presented by the Federated Malay States Museum. * c-m’=front of canine to back of m?, 376 Dr. K. Andersen on new Bats of the Rh. pusillus evoup. (Called lepidus group in 1905.) w. Connecting process like an erect (nearly equi- lateral) triangle, its front margin practically Pe (uon-coucaye). . Smaller; forearm 33:5-43 mm. ....,...4. ( pusillus subgroup.) - a’, Skull and teeth larger; skull to front of : ; canine 16:5-18°'7 mm.; cond.—can,* 14°4- 16°9; mandible 11-132; c-m 6'2-7°5.. (lepidus series.) a, Base of fur of back paler, ora with the darker tips ....:...--se0. lepidus. ce‘. Skull and teeth averaging ‘larger ; total length to front of canine 16:8- 18°7 mm.; cond.-can. 15-169; e- m®> 65-75. Forearm 38-42°5. (Upper Burmat) ies ee thee eet l. shortridget, subsp. n. 6°, Fur of back uniform from base to tip . refulgens. J+. Sella subacute, its tip forming an equilateral triangle in front view. : (Subartra ee cretiejae ninth arian 7. cuneatus, subsp. n. 6, Skull and teeth smaller; skull to front of canine 15°3-16°7 mm.; cond.-can. 13:5-14'8; mandible 98-11; c-m*5:5-6'4. (puszllus series.) (Fur of back pale at base. Sella conspicuously constricted at middle, markedly narrower at tip than at base.) a, Smaller, with relatively shorter tibia and smaller foot. Skull 153-16 mm. ; cond.-can. 13°5-14:2; forearm 35:5- 39:7 ; tibia 14-16; foot (c. u.) 7-8. a‘, Canines, p' and ps; unmodified; ps ; sometimes external, but generally half or wholly in row. Forearm 35'5-39'7 mm..... faa heea noe . blythi, sp. n. a’, Fur conspicuously pale above and below. (Kumaon.) .......... b. blythi. 6°. Fur conspicuously darker above [subsp. n. and below. (Darjiling to China. ) b. szechwanus, b‘, Canines much heavier than in a*; p' and ps; conspicuously reduced” in size; ps generally external. General size as in a’, ce’. Teeth markedly larger; c—m 6:4 mm.; c-m;, 67. (8S. Liu-Kiu; Tshig@ki: ea. sae toe perditus, sp. 0. d’, Teeth not larger than usual; c-m* 55-57 mm.; c-m, 5°8-61. (Middle Liu-Kiu; Okinawa.) .. pumilus. b°, Larger, with relatively longer tibia and larger foot. Tibia 16°5-17°5 mm. (Japan)... . Groner Onantes Caamrion, BVZiS): does. St Gis ee enOOt XXXIII. On some External Characters of Ruminant Artiodactyla. —Part IV. The Reduncine (Cervicaprine) and Aipycerine. By Ret: Pasoox, CRS. ea ea eee aes png Se ea niece 367 Ld : > +3 “XXXIV. Diagnoses of new Bats of the Families Rhinolophide and Megadermatide. By Knup ANDERSEN ..............-. ; hie 374 XXXV. Descriptions and Records of Bees. —LXXX. By T.D. A. Cocxrrett, University of Colorado Pea brates) Geo wep cae 4» 384 -BIBLIOGRAPHIOAL NOTICE, i Life and Letters of Sir Joseph Dalton Hooker, O.M., G.C.S.1, Based. on Materials collected and arranged dy Tay Hookets By LEONARD ELUXERE 6 Uo ins he oe tees ee es See my i ag *,* Itis requested that all (iciuiu iste teatn for this Wo ork may he ian, post-paid, to the Care of Messrs. Taylor and pd Printing caine Red Lion Court, Fleet Street, London, WITH SIX PLATES. » Volk: s ae ‘NINTH SERIES. ; No. 11. ; ae a= | ; ere beees . mM “THE. ANNALS MAGAZINE OF NATURAL HISTORY, ck ENCLUBING§: 3. 22% *) Shee Konica : % 200L0GY, sit ee AND G@EOLO ; a; ‘CONDUCTED. BY . WILLIAM C ARRUTHERS, Pa. D., Ly R. S., FuUILS:, F: Gs ARTHUR pis eee MA, So.D., F.R.S., F.Z.8 . AND RICH HARD. ab FRANCIS, PLS. BEING A CONTINUATION or THE. 0 ANNALS” ‘COMBINED With ‘MESSRS, LOUDON AND CHARLESWORTH’S “MAGAZINE OF NATURAL HISTORY.’ ¢ ou gah LONDON: Seat ‘TAYLOR AND FRANCIS, RED LION COURT, FLERY STREET. | Sold by Simpkin, Marshall, Hamilton, Kent, & Co, Ld., Bailliére, Paris and Hodges, F Figcis, & Co., att oy Raturalists, bb con in stock every | kind of OLOGISTS, BOTANISTS, &¢. Also ee : which should be fixed up in parse or Shruppany a before the breeding Season. ae Ma 10 A Large Stock of Butterflies, Woths, Birds, er tc. pops Full Oatalogue (84 pages) mailed free to any aadiad! ‘ 36, STRAND, LONDON, W.C., ENGLAND. WATKINS & Meee he. : “a Price 198) 6d. foarte cy ne wat ey FAUNA. OF ERITISH. INDIA (Including Ceylon “andl Burma). ie she ee Published under the authority of the Secr etary of State for India i in Couneil, Medium 8vo, with Teat Illustrations. Ay ao é RHYNCHOTA, ‘Vol. wae Taytor & Francis, | Red Lion Court, Eicee Street, EC. As Price 12s, 6d. A CONTRIBUTION Super- Baie 8vo, pp. iv, 226. With 4 Plates and 16 Test gure 8. 'PHYTOGEOGRAPHY AND FLORA DUTCH N.W. ‘NEW GUINEA. be 8: GIBBS, F.LS., PRMS. TAYLOR AND FRANCIS, Red Lion Court, Fleet Streot. ees Rates for Advertisements 1 in ‘he Annals and Magazine of Natural History. Ones Gik:: ee Pyiclve ih Gage Insertion. -- Insertions. We tions Ne PAGE - - - - Lig 2 Oey @ Yates a7 C}, O each f. +2 O cach HALF-PAGE - fete i ents oo 4 O° ¢) Soha aber Wf 6. : Net. QUARTER-PAGE - 126 AS 10° Oe as All applications for space to be nutes to ge : Mr. H. A. COLLINS, 3A, Birdhurst Road, s, Croydon, ee oes a THE ANNALS AND MAGAZINE OF NATURAL HISTORY. [NINTH SERIES. ] No. 1l. NOVEMBER 1918. XXXVI.—Descriptions of New Pyralide of the Subfamily Pyraustine. By Sir Georee F. Hampson, Bart., F.Z.S., &e. [Concluded from p. 196.] Genus Pronea will stand as Tp Honaiide EMAb ia Clits Da GION CLONE vasnusWemasccsugvevarsay says fulvalis. (1h) Hapalia bifossata, sp. n. Antennz of male laminate with ridge of scales above ; fore wing with depressed streaks beyond the cell above and below vein 6. 3. Head and thorax pale red-brown; abdomen whitish with diffused brown bands except at base leaving whitish segmental lines; frons with white lines at sides; palpi red-brown, white at base ; pectus, legs, and ventral surface of abdomen white tinged with ochreous brown, the fore tibize with dark brown band at extremity. Fore wing whitish tinged with red-brown, the costal and terminal areas broadly suffused with red-brown and the latter irrorated with darker brown, the costal edge dark brown to the postmedial line, then whitish with two minute dark spots on it; antemedial line red-brown, oblique to submedian fold and incurved below vein 1; a brown point in the cell towards extremity and obliquely curved discoidal striga; the depressed streaks beyond the cell whitish; postmedial line dark brown, slightly incurved below costa, then excurved and minutely waved to below vein 3 where it is retracted to below end of cell, excurved below sub- median fold; a terminal series of minute dark spots, rather bar- shaped below vein 4; cilia dark brown, chequered with whitish at Ann. & May. N. Hist. Sr. 9. Vel, ii. 8, 394 Sir G. F. Hampson on new tips. Hind wing white, the inner area tinged with red-brown, the terminal area suffused with red-brown to vein 2; a blackish point at lower angle of cell; postmedial line brown, excurved from discal fold to vein 2 where it terminates; a terminal series of minute dark spots to vein 1; cilia red-brown with white tips to submedian fold, then wholly white ; the underside white with the costal area tinged with red-brown, black points at the angles of cell, the-post- medial line black, punctiform, and extending to vein 1. Hab. Peru, Carabaya, Oconeque (Ockenden), 1 3 type. Exp. 20 mm. (17) Hapalia lobibasalis, sp. n. Antenne of male laminate with ridge of scales above; hind wing with the costa lobed near base. 6. Head and thorax whitish suffused with red-brown ; abdomen white slightly suffused with red-brown; palpi dark red-brown, white in front to extremity of 2nd joint; pectus, legs, and ventral surface of abdomen white slightly suffused with red-brown. Fore wing whitish suffused with red-brown, the costa darker brown ; a faint oblique sinuous brown antemedial line; postmedial line brown, waved, excurved from vein 6 to 4, then oblique ; a terminal series of minute black-brown spots to vein 2; cilia with a brown line at middle. Hind wing white tinged with red-brown; a terminal series of minute black-brown spots to vein 2; the under- side with indistinct curved brown postmedial line from costa to vein 2. Hab. Ectvapor, Zamora (Abbé Gaujon), 1 S$ type. Hap. 20 mm. (4a) Hapalia magnifovealis, sp. n. 3. Head, thorax, and abdomen orange-yellow ; frons with white lines at sides; palpi white at base; pectus, legs, and ventral surface of abdomen white tinged with yellow, the fore tibiz orange- yellow in front. Fore wing orange-yellow, the costal area more fulvous orange, the terminal area rather narrowly suffused with red-brown and glossed with silvery blue from below apex to sub- median fold; antemedial line indistinct, orange, very oblique; the fovea beyond the cell large, white with a brownish white boss in it ; postmedial line orange, excurved to vein 3, then retracted to below end of cell and oblique to inner margin; a punctiform red- brown terminal line from below apex to submedian fold slightly defined on inner side by orange; cilia fulvous orange, whitish at tips. Hind wing orange-yellow, the cell and costal area to near apex and the inner area white; an orange postmedial line from vein 5 to submedian fold; the terminal area narrowly suffused with red-brown and glossed with silvery blue from below apex to Pyralidee of the Subfamily Pyrausti xe. 395 vein 2; a dark brown terminal line from apex to submedian fold ; cilia orange-yellow with a deeper orange line through them and some whitish at tips to submedian fold, then white tinged with yellow. Hab. Perv, Yahuarmayo, 1 d type. ZHvp. 18 mm. (15 b) Hapalia endotrichialis, sp. n. 3. Head, thorax, and abdomen orange-yellow ; palpi with the basal joint white ; pectus and ventral surface of abdomen at base with some white. Fore wing orange-yellow; a faint brownish antemedial line, oblique to median nervure, then erect; minute reddish brown spots in the cell towards extremity and on disco- cellulars ; postmedial line reddish brown, excurved and slightly waved to below vein 8, then retracted to below end of cell and waved to inner margin; a fine dark brown terminal line. Hind wing orange-yellow ; postmedial line brown, arising at vein 6, oblique to vein 2, then slightly incurved and ending at submedian told; a fine dark brown terminal line except towards tornus. Hab. Formosa (Wileman), 1 3 type. Exp. 28 mm. (24a) Hapalia glaucostigmalis, sp. n. @. Head and thorax rufous; abdomen greyish suffused with red-brown; antenne red-brown; palpi red-brown, white below to near extremity of 2nd joint; pectus, legs, and ventral surface of abdomen white mixed with red-brown, the fore legs red-brown, white on inner side. Fore wing rufous, the inner area paler ; small obliquely placed dark brown spots in the cell, in submedian fold, and on inner margin; a small grey-white spot in middle of cell and discoidal lunule defined by dark brown; postmedial line dark brown, excurved from below costa to vein 4, then oblique and slightly sinuous ; a brown subterminal shade. Hind wing ochreous white; an indistinct curved brown postmedial line; a slight terminal brown shade from apex to vein 2; cilia whitish. Hab. Cotompta, Rio Derg, 1 2 type. Earp. 24 mm. (346) Hapalia nigristriatalis, sp: n. 3g. Head, thorax, and abdomen white tinged with red-brown, the last darker towards extremity ; palpi suffused with red-brown below, white above ; pectus, legs, and ventral surface of abéomen white tinged with red-brown. Fore wing white tinged with red- brown, the inner half whiter to beyond middle, the terminal area broadly suifused with red-brown; a rather diffused red-brown fascia through the cell; the veins beyond the cell slightly streaked with red-brown and veins 7, 6 streaked with black defined below by white streaks towards termen ; five black points oa terminal 29 396 Sir G. F. Hampson on new part of costa which is white ; some blackish scales on lower disco- cellular; a terminal series of minute blackish points on an ochreous white line ; cilia dark brown, chequered with white at tips. Hind wing white, the termen narrowly suffused with red-brown to sub- median fold; a terminal series of black points to vein 2; cilia white. Hab. Cotompt1a, San Antonio (Palmer), 1 3 type. ap. 22 mm. (85a) Hapalia tristigmalis, sp. n. Head white tinged with cupreous brown; thorax pale cupreous brown ; abdomen whitish banded with dark brown except towards base; antenne dark brown ringed with white; frons with white lines at sides; palpi dark brown, white in front towards base and with some white at tips; pectus, legs, and ventral surface of abdo- men white mixed with some black-brown, the fore tibiz black- brown, the tarsi ringed with black. Fore wing whitish suffused with cupreous brown with a slight purplish gloss, the costal edge black with alternating white marks towards apex; a curved black antemedial line, defined on inner side by white and with a small triangular creamy white spot beyond it below the costa; a small conical creamy white spot defined by black except above in upper part of cell towards extremity; a sinuous black line defined on outer side by white from lower angle of cell to inner margin; a creamy white postmedial patch defined by black from costa to vein 5, its outer edge angled outwards at vein 6, then reduced to a bar; cilia creamy white, chequered with brown at base and with brown line at middle. Hind wing white faintly tinged with brown ; slight dark spots in upper part of cell towards extremity and at upper angle; a faint brown postmedial line, incurved between discal and submedian folds ; the terminal area suffused with pale purplish brown except towards tornus; cilia chequered with brown at base and with brown line at middle to vein 2; the underside with black spots in the cell near base and before and at end of cell, a postmedial series of black spots, incurved at discal fold and ex- curved above inner margin. Hab. Cotomsta, Sierra del Libane (H. H. Smith), 4 3, 2 2 type, Bonda (H. H. Smith), 1 6,192. Hxp. 18 mm. (36 a) Hapalia distictalis, sp. n. @. Head, thorax, and abdomen dark cupreous brown; palpi white below to near tips ; pectus, legs, and base of ventral surface of abdomen with some white, the tarsi creamy white. Fore wing dark cupreous brown; a faint oblique dark antemedial line; a small triangular creamy white spot defined by blackish except above in upper part of cell towards extremity ; a faint sinuous dark line from lower angle of cell to inner margin; a _postmedial Pyralide of the Subfamily Pyraustine. 397 creamy white bar defined by blackish between veins 8 and 4, its outer edge slightly angled outwards at veins 6 and 5; cilia white at tips with some brown scales mixed. Hind wing dark cupreous brown ; a faint dark discoidal bar; a faint slightly sinuous dark postmedial line from vein 4 to tornus; cilia white at tips; the underside with waved dark postmedial line, incurved at discal fold ; both wings with white line at base of cilia. Hab. ConomBra, Don Amo (4H. H. Smith), 1 2 type, Bonda (H. H. Smith),1 2. Hap. 20-22 mm. (86d) Hapalia flavipartalis, sp. n. 3. Head and thorax yellow mixed with red-brown, the frons whitish, the antennz whitish tinged with brown; palpi red-brown, white below towards base and with some whitish at tips; abdomen white mixed with red-brown ; pectus, legs, and ventral surface of abdomen white, the fore tibize yellowish, the mid femora with minute brown spot at extremity. Fore wing with the basal half orange-yellow, the base suffused with red-brown, the terminal area red-brown ; an oblique sinuous brown antemedial line; a brown annulus in middle of cell ; a curved brown medial line confluent with the inner side of a reddish brown discoidal spot defined by dark brown and with dark brown striga in centre, the spot confluent on outer side with the brown terminal area; a conical orange- yellow postmedial patch from costa to vein 5, defined by dark brown and its inner edge confluent with the yellow basal area at costa; cilia white at tips from below apex to vein 4 and with some white at submedian interspace. Hind wing white, tinged with red-brown except the cell and costal area to beyond middle; the cilia white; the underside white, the terminal area tinged with brown to vein 2. Hab. Cotomsta, Choko, R. Siato, 1 ¢ type. Hwvp. 20 mm. (386) Hapalia umbriferalis, sp. n. ¢. Head and thorax rufous, some white on vertex of head and on metathorax behind; abdomen dark red-brown with white segmental lines; palpi with some white at base; pectus, legs, and ventral surface of abdomen white mixed with rufous, the femora, tibie, and tarsi banded with black. Fore wing rufous suffused with dark brown, the costal area bright rufous except towards base, with three small black spots on the costa towards apex; antemedial line black-brown, angled outwards below costa, excurved below the cell and angled inwards above inner margin, defined on inner side by whitish ‘below the cell; a small black annulus in upper part of middle of cell and discoidal figurefof-eight shaped mark, its upper and lower parts filled in with rufous, the rufous from costa ex- tending into the cell before it ; postmedial line black-brown defined 398 Sir G. F. Hampson on new on outer side by whitish, strong and obliquely downeurved to vein 6, then excurved and minutely dentate to vein 2 where it is retracted to below angle of cell and bent outwards below submedian fold ; a terminal series of minute black spots with whitish striae between them; cilia dark red-brown, whitish at tips. Hind wing red-brown, rather darker at termen on which there is a series of minute blackish points; cilia white at tips; the underside pale rufous slightly irrorated with dark brown, a minute black spot in middle of cell and small spots at the angles, postmedial line black, maculate, excurved to below vein 38, then retracted and ending in a small spot below vein 2, a terminal series of black points to vein 2 and some dark brown at submedian fold. Hab. Prru, San Domingo (Ochkenden), 1 & type. Exp. 22 mm. (506) Hepalia conisanalis, sp. n. @. Head, thorax, and abdomen red-brown mixed with some ereyish, the last with white segmental lines except towards base ; frons with white lines at sides ; palpi rufous, white at base; pectus legs, and ventral surface of abdomen white tinged with red-brown. Fore wing whitish suffused with red-brown and thickly irrorated with dark brown, the terminal area rather more strongly suffused with red-brown ; antemedial line rather diffused, brown, slightly waved; a pean brown spot in upper part of cell towards extre- mity and discoidal striga; a brown shade beyond the cell from costa to vein 2; postmedial line brown, minutely waved, excurved from below costa to vein 3, then retracted to below end of cell; a rather punctiform dark brown terminal line to submedian fold ; cilia with a brown line through them, the tips whitish. Hind wing whitish suffused with red- brown and irrorated with dark brown ; postmedial line indistinct, brown, slightly excurved from diseal fold to vein 2 where it terminates; a rather punctiform dark terminal line to 2; cilia with a brownish line near base and the tips whitish to vein 2, then wholly whitish. Hab. Br. C. Aprica, Shiré Valley, Mwanza R. (Weave), 1 2 type, Mt. Mlange (Neave),1 2. Exp. 20 mm. (10la) Hapalia lunilinealis, sp. n. S$. Head, thorax, and abdomen rufous, the genital tufts white ; palpi below towards base and pectus in front white; tarsi white tinged with rufous. Fore wing rufous; antemedial line indistinct, brown, oblique, and slightly sinuous to vein 1, then incurved ; a slight dark discoidal lunule; postmedial line formed by minute dark lunules, excurved from below costa to below vein 3, then retracted to below end of cell and erect to inner margin; a brown terminal line; cilia whitish tinged with rufous and with brown line near base. Hind wing whitish tinged with rufous; a curved Pyralidze of the Subfamily Pyraustine. 399 postmedial series of brown points on veins 5 to 2; a red-brown terminal line and line near base of cilia to vein 2. Hab. Ecuavor, Zamora (Abbé Gaujon), 6 3 type. Euzp. 28 mm. (104¢) Hapalia rubritactalis, sp. n. 3. Head, thorax, and abdomen ochreous yellow tinged with rufous ; palpi rufous, the basal joint white ; pectus, legs, and ventral surface of abdomen white, the fore legs tinged with rufous ; a faint diffused brownish antemedial line from subcostal nervure to inner margin ; a small brownish spot in upper part of cell towards extre- mity and discoidal bar; postmedial line indistinct, diffused, brownish, excurved to vein 3, then retracted to below angle of cell and erect to inner margin, slightly defined on outer side by yellow; the costal area yellower towards apex. Hind wing ochreous yellow suffused with rufous, the inner margin whitish ; postmedial line brownish defined on outer side by diffused yellow, erect to vein 2 towards termen, then retracted and again erect to termen above tornus; the terminal area suffused with rufous to vein 1, leaving some yellow on termen; cilia white. Hab. “Germ. E. Arnica,” Ruaha R., Kilossa Rd. (eave), 1 ¢ type. Exp. 20 mm. (127) EHapalia carbonifusalis, sp. n. Head fuscous brown mixed with some ochreous ; thorax fuscous brown ; abdomen greyish suffused with fuscous brown; antennze fuscous brown; palpi black-brown with some white below ; pectus, legs, and ventral surface of abdomen grey suffused with fuscous brown, the fore tibie with black band at extremity. Fore wing fuscous brown mixed with grey-white; antemedial line blackish, oblique to median nervure, then erect; a slight white discoidal lunule defined by fuscous brown; postmedial line rather diffused blackish, slightly excurved at vein 7, and bent outwards between veins 5 and 3, then retracted to below end of cell and erect to inner margin; a blackish terminal line ; cilia chequered with blackish at tips. Hind wing fuscous brown tinged with grey. Hab. Br. C. Arrica, Mt. Mlanje (Weave), 3 $, 3 Q type. Exp. 16-20 mm. (127d) Hapalta conistolalis, sp. n. 3. Head, thorax, and abdomen dark brown mixed with grey- white; antenne dark brown; palpi black-brown; fore tibiz at extremity and the tarsi banded black and white. Fore wing thickly irrorated with dark brown and grey-white; antemedial line black, slightly waved, oblique to submedian fold, then erect; a small rather diffused blackish spot in middle of cell; a small white 400 Sir G. F. Hampson on new discoidal lunule irrorated with brown and defined at sides by black ; postmedial line black, excurved at vein 7 and between veins 5 and 3, then retracted to below angle of cell and excurved below submedian fold; a terminal series of small black spots; cilia white mixed with brown. Hind wing grey-brown irrorated with fuscous; a dark terminal line except towards tornus ; cilia white mixed with brown and with brown line at middle. Hab. N. Nreerta, Zingeru (Simpson), 1 3 type, Minna (Macfie), 1 ¢. Hep. 20 mm. (127 e) Hapalia pulverulenta, sp. n. 3. Head and thorax reddish brown mixed with grey-white ; abdomen whitish tinged with red-brown ; frons with white lines at sides; palpi red-brown, white at base; pectus, legs, and ventral surface of abdomen white, the fore legs suffused with red-brown. Fore wing reddish brown mixed with some white ; a brown ante- medial line in submedian interspace, angled outwards to a slight spot at submedian fold; slight brown spots at middle of cell and on discocellulars; postmedial line formed by small brown spots, defined on outer side by slight white marks and with some white before it at discal fold, excurved from discal fold to vein 3, then ineurved ; a terminal series of minute blackish spots. Hind wing pale reddish brown ; a terminal series of black points to vein 2; cilia white tinged with red-brown. Underside of fore wing grey- brown, the costal area white to near apex ; hind wing white. Hab. Cryton, Ambalangoda (Jlackwood, Green, Pole), 3 3 type. Hap. 20-22 mm. (128 a) Hapalia poliostolalis, sp. n. @. Head, thorax, and abdomen grey-brown with a leaden gloss, the last with white segmental lines; palpi white below; pectus, legs, and ventral surface of abdomen white tinged with brown. Fore wing grey-brown with a leaden gloss; a faint erect brown antemedial line; a faint dark discoidal bar; postmedial line rather diffused dark brown, very slightly waved, excurved from costa to below vein 8, then retracted to below angle of cell and erect to inner margin ; cilia white tinged with brown, with dark line near base and slight spots near tips. Hind wing grey-brown with a leaden gloss, the cilia white with a dark line near base; the underside white mixed with brown, obliquely placed small black spots at the angles of cell, a curved punctiform dark postmedial line, and terminal series of black points. Hab. Formosa, Kanshirei (Wileman), 1 9 type. Hap. 16 mm. Pyralids: of the Subfamily Pyraustine. AOL (la) Pyrausta pectinalis, sp. n. Antenne of male bipectinate with long fine branches to two- thirds length. 3. Head and thorax pale red-brown ; abdomen whitish suffused with red-brown; antennz ringed with black towards base; palpi black-brown, white below to near extremity of 2nd joint; pectus, legs, and ventral surface of abdomen white, the fore cox dark brown towards base, the femora and tibize suffused with red-brown. Fore wing glossy red-brown ; a faint dark discoidal bar; cilia with pale line at base and some whitish at tips. Hind wing glossy red- brown ; a faint dark mark at upper angle of cell; cilia with some whitish at tips; the underside whitish tinged with red-brown, a faint rather diffused brown postmedial line from costa to vein 4. Hab. Perv, Chanchamayo, 1 ¢ type. Hap. 26 mm. - (81b) Pyrausta fulviflavalis, sp. n. Q. Head whitish tinged with fulvous ; thorax fulvous ; abdomen whitish suffused with fulvous; palpi rufous, white below; throat white ; pectus, legs, and ventral surface of abdomen pale rufous, the mid tibize on outer side and all the tarsi white. Fore wing fulvous, the costal edge brown to middle, then white ; antemedial line indistinct, brown, eblique and waved to above vein 1 and angled inwards above inner margin ; a brown point in upper part of middle of cell and curved discoidal striga ; -a diffused brown spot beyond lower angle of cell ; postmedial line brown, dentate, oblique to vein 5, then inwardly oblique and incurved above inner margin ; cilia rufous. Hind wing semihyaline whitish tinged with orange- yellow, the terminal area orange-yellow to submedian fold, angled inwards at vein 2 to below end of cell; a curved series of slight red-brown lunules on veins 4, 3, 2; a red-brown terminal line and the cilia rufous from below apex to vein 2. Hab. Argentina, Puerto Aguirre (Betton), 1 Q type. Exp. 32 mm. (88a) Pyrausta violascens, sp. n. Q. Head and tegule fulvous ; thorax very pale purplish; abdo- men white with a violaceous grey tinge; palpi rufous, white below towards base ; pectus, legs, and ventral surface of abdomen white faintly tinged with brown. Fore wing very pale purplish, the costal area fulvous to beyond middle; a faint oblique brownish antemedial line ; a small fulvous spot in the cell towards extremity and discoidal bar; a faint brownish postmedial line, excurved and slightly waved between veins 5 and 2, then retracted to below angle of cell and oblique to inner margin; cilia whitish. Hind wing very pale purplish, the inner area whitigh; a faint brownish 402 Sir G. F. Hampson on new postmedial line, excurved and slightly waved between veins 5 and 2, where it terminates ; cilia whitish. Hab. Gow Coast, Kumasi (Sanders), 1 9 type. Hxp.28 mm. (55a) Pyrausta fulvilinealis, sp. n. 3. Head and thorax white mixed with some fulvous ; abdomen white; antenne pale fulvous; frons with black bars at sides; palpi fulvous mixed with some blackish, white below towards base ; pectus, legs, and ventral surface of abdomen white, the fore femora red-brown above, the tibize black on inner side and the tarsi ringed with black. Fore wing creamy white, the costal area tinged with fulvous and the costal edge black-brown to end of cell; antemedial line fulvous, oblique, slightly excurved below costa ; a small fulvous spot in the cell towards extremity and discoidal lunule defined by fulvous; postmedial line fulvous, interrupted, angled outwards below costa, then incurved to vein 5 where it is interrupted, oblique to vein 2, then represented by a bar below angle of cell and oblique line from vein 2 to inner margin; subterminal line fulvous, rather interrupted, oblique to vein 5, excurved between veins 5 and 4, and angled inwards at vein 2 to near the postmedial line; the costa fulvous towards apex; a fine fulvous terminal line. Hind wing ereamy white; a fulvous discoidal bar; postmedial line fulvous, slightly bent outwards between veins 5 and 2, then retracted and obsolete to lower angle of cell, then oblique to inner margin; sub- terminal line fulyous, slightly excurved between veins 6 and 2 and ending at tornus; a fine fulvous terminal line and slight line near base of cilia. Hab. Ucaxrpa, Mbale-—Kumi Rd. (Neave), 1 3 type. Exp. 32 mm. (58a) Pyrausta distictalis, sp. n. 3. Head and thorax whitish suffused with fulvous ; abdomen creamy white faintly tinged with rufous ; pectus, legs, and ventral surface of abdomen creamy white, the fore legs tinged with rufous, the femora, tibiz, and base of tarsi blackish above. Fore wing very pale yellow, the base suffused with fulvous, the costal edge blackish ; a minute black spot in the cell towards extremity and another at lower angle. Hind wing uniform very pale yellow. Underside of fore wing tinged with brown except on inner area. Hab. Br. C. Arrica, Mt. Mlanje (Weave), 2 5 type. Exp. 24 mm. (6le) Pyrausta leucoplacalis, sp. n. 3d. Head and thorax cupreous brown with some white on meta- thorax ; abdomen white indistinctly banded with cupreous brown ; antennz whitish tinged with cupreous brown; sides of frons and Pyralidee of the Subfumily Pyraustine. 403 palpi black-brown, the latter white below; pectus, legs, and ventral surtace of abdomen white, the fore femora and tibize suffused with cupreous brown and the mid tibiz with cupreous brown spots at extremity. Fore wing cupreous brown, an ochreous white fascia below costa from the antemedial to beyond the postmedial line ; antemedial line dark brown defined on inner side by ochreous white, arising at median nervure and slightly angled outwards above inner margin, an ochreous white patch beyond it at inner margin; a small semihyaline white spot in middle of cell and discoidal spot defined by dark brown except above where it is con- fluent with the subcostal fascia ; postmedial line dark brown, waved and defined on outer side by a waved ochreous white band, with a semihyaline white patch before it beyond the cell and spots below veins 4, 3, 2, excurved from costa to vein 4, then oblique; a narrow terminal ochreous white band and a terminal series of small brown spots to vein 2; cilia white. Hind wing semihyaline white t@ the postmedial line, then ochreous white; small black-brown subbasal spots below the cell and above inner margin ; a black discoidal bar ; postmedial line black-brown, arising below costa, curved and waved between veins 5 and 2, where it is retracted, then sinuous to inner margin ; a wedge-shaped cupreous brown subterminal patch with waved edges from below costa to vein 3, then a rather diffusec interrupted sinuous line ; a terminal series of small brown spots to vein 2; cilia white. Hab. Cotomsta, Sierra del Libane (H. H. Sinith), 2 3 type. Exp. 26 mm. (103 6) Pyrausta xanthyalinalis, sp. n. @. Head and thorax pale yellow tinged with rufous ; abdomen pale yellow; frons with blackish bars at sides; palpi black-brown above and white at base; pectus, legs, and venta surface of abdo- men white, the legs tinged with yellow, the fore legs with dark brown mark at femoro-tibial joint. Fore wing pale yellow, thinly sealed, the costal area tinged with rufous and the costal edge dark brown to the postmedial line; antemedial line brown, slightly curved; adark brown discoidal lunule; postmedial line dark brown, curved inwards and obsolescent between veins 5 and 2 and slightly excurved above inner margin; a terminal series of brown striz from apex to vein 4. Hind wing pale yellow, thinly scaled; a brown discoidal striga ; postmedial line brown, curved inwards and obsolescent between veins 5 and 2. Hab. Ecvapor, R. Pastaza, El Topo (Palmer), 3 ¢ type. Exp. 24 mm. (106 6) Pyrausta microdontalis, sp. un. 2. Head and thorax whitish tinged with red-brown ; abdomen white faintly tinged with brown; palpi red-brown, white below ; 404 Sir G. F. Hampson on new pectus, legs, and ventral surface of abdomen white, the fore legs and mid femora streaked with brown. Fore wing whitish suffused with pale reddish brown and slightly irrorated with fuscous; a eurved blackish antemedial line; a black discoidal bar; postmedial line blackish, curved and minutely dentate to vein 2 where it is retracted to below end of cell and oblique to inner margin; a faint rather diffused dentate brown subterminal line; a fine black terminal line; cilia whitish at tips. Hind wing whitish suffused with pale reddish brown and irrorated with fuscous, the inner margin white; an oblique blackish discoidal bar; postmedial line rather diffused blackish, waved to vein 2, then retracted to below angle of cell and ending at tornus; a blackish subterminal shade with slightly waved outer edge to vein 2, then oblique; a fine black terminal line; cilia with dark line near base, the tips white. Hab. Br. EB. Arrica, N. Kavirondo, Maramas Distr., Lala (Weave), 1 2 type. Hap. 20 mm. (107 b) Pyrausta pulvereiumbralis, sp. n. 3S. Head and thorax ochreous tinged with rufous; abdomen whitish suffused with red-brown and with white segmental lines towards extremity, the anal tuft tinged with rufous; palpi white below towards base; pectus, legs, and ventral surface of abdomen white slightly tinged with brown. Fore wing ochreous tinged with rufous, irrorated with brown from before the antemedial to beyond the postmedial line except on costal area, the medial area whitish except towards costa; antemedial line fulvous yellow with a brownish line on it, curved; a small brown spot defined by . fulvous yellow in upper part of cell towards extremity and brown discoidal bar defined by fulvous yellow; an oblique brown shade from beyond upper angle of cell to inner margin beyond the post- medial line, which is fulvous yellow with a brownish line on it, excurved from vein 7 to 5, then rather oblique to vein 3, then bent inwards to lower angle of cell, then again rather oblique and bent outwards to inner margin, a brown shade with waved outer edge beyond it from below costa to vein 3; a curved rather diffused fulvous yellow subterminal line, arising below the costa; cilia white with a faint ochreous brown line at middle. Hind wing white; a brownish postmedial line, bent inwards at vein 2, then oblique to tornus ; a brownish subterminal line. Hab. Axysstyta, Diré Daroua (Kristensen), 1 5 type. Eup. 24 mim. (107 e) Pyrausta fulvitinctalis, sp. n. Q. Head and thorax fulvous; abdomen red-brown with fine white segmental lines on medial segments ; antenne dark brown ; frons with white lines at sides; palpi yellow with a fulvous tinge ; pectus, legs, and ventral surface of abdomen white tinged with Pyralide of the Subfamily Pyraustinz. 405 rufous. Fore wing red-brown, suffused with fulvous to middle and on costal area to apex; a faint dark antemedial line, oblique to submedian fold, then inwardly oblique; postmedial line dark, oblique towards costa, then excurved and minutely waved to vein 3, slightly angled inwards at vein 2 and erect to inner margin; a fine dark brown terminal line ; cilia with a fine pale line at base followed by a brown line. Hind wing red-brown; an indistinct curved dark postmedial line; a dark brown terminal line; cilia with a fine pale line at base followed by a brown line ; the underside paler, the costal area ochreous white to the postmedial line. Hab. Ecuavor, Zamora (Abbé Gaujon), 1 2 type. Exp. 20 mm. (1086) Pyrausta wanthocepsalis, sp. n. 3. Head yellow tinged with rufous; antenne, thorax, and abdomen glossy fuscous brown; palpi dark brown, yellowish above and white below to near extremity of 2nd joint; pectus, legs, and ventral surface of abdomen white tinged with brown. Fore wing glossy fuscous brown slightly irrorated with whitish; a diffused whitish spot in end of cell; postmedial line whitish, somewhat dilated at costa, incurved at discal fold, excurved to vein 3, then retracted to below angle of cell and excurved below submedian fold ; a terminal series of slight black points and fine white line at base of cilia. Hind wing pale brown with a slight cupreous tinge ; cilia with a fine white line at base followed by a brown line, the tips with some whitish. Hab. Mexico, Guerrero (H. H. Smith), 2 3 type, Godman- Salvin Coll., Guadalajara (Goldsmith),1 $. Hwp. 20 mm. (1186) Pyrausta infuscalis, sp. n. Q. Head, thorax, and abdomen dark reddish brown; palpi dark brown, white below to near extremity of 2nd joint; pectus, legs, and ventral surface of abdomen white tinged with brown. Fore wing dark reddish brown slightly irrorated with whitish; a faint dark antemedial line, oblique towards costa and defined on outer side by whitish below the cell; postmedial line indistinct, rather diffused dark brown slightly defined on outer side by whitish, somewhat angled outwards below costa, incurved at discal fold, excurved to vein 2, then retracted to below angle of cell and erect to inner margin; a terminal series of small rather triangular blackish spots ; cilia whitish mixed with brown. Hind wing pale reddish brown, the costal area whitish to beyond middle; cilia whitish with a brown line near base; the underside whitish mixed with brown, a dark postmedial line excurved below vein 7 and between veins 5 and 2. Hab. Sixutm (Méller), 1 2 type. Hxp. 20 mm 406 On new Pyralide of the Subfamily Pyraustine. (139 b) Pyrausta auricinctalis, sp. n. 3. Head and tegule orange-yellow, the latter with some dark brown dorsally ; thorax dark brown mixed with yellow; abdomen dark purplish brown, the two terminal segments orange-yellow, the genital tufts paler yellow ; antenne brown, orange-yellow towards base; peetus and legs yellowish tinged with brown. Fore wing dark purple-brown ; the base orange-yellow ; the costal edge orange- yellow to a medial orange-yellow band from costa to above vein 1, rounded below, the costa beyond it orange-yellow; an orange- yellow terminal band with curved inner edge; cilia orange-yellow at base, whitish at tips. Hind wing dark purplish brown; an orange- yellow terminal band, the inner edge slightly incurved at submedian fold ; cilia orange-yellow, whitish at tips. Hab. Br. E. Arrica, N. Kavirondo, Maramas Distr., Lala (Neave), 1 3 type. Herp. 18 mm. (8a) Pegostoma subterminalis, sp. n. $6. Head, thorax, and abdomen white mixed with reddish brown; antenne brown; palpi dark brown, white below; pectus, legs, and ventral surface of abdomen white mixed with dark brown. Fore wing white, the basal area and costal area to apex tinged with red-brown; a red-brown subterminal band, its inner edge incurved below vein 5 and slightly angled outwards above vein 1; cilia pale red-brown. Hind wing pale red-brown. Under- side white suffused with red-brown. é Hab. Orange R. Cotoxy, Bloemfontein (Hekersley), 1 3 type. Hep. 16 mm. (4a) Noctuelia anartalis, sp. n. 9. Head, thorax, and abdomen dark brown mixed with some white; antenne dark brown; palpi dark brown, the basal joint white; pectus, legs, and ventral surface of abdomen white mixed with dark brown. Fore wing red-brown mixed with white and slightly irrorated with dark brown; an oblique black-brown line defined on inner side by white from upper part of cell towards extremity to inner margin; some diffused blackish beyond upper angle of cell; postmedial line white defined on inner side by a fine slightly dentate black-brown line, incurved below vein 4 and slightly angled outwards below submedian fold ; a diffused sinuous whitish subterminal band indented by a wedge-shaped dark mark from termen above vein 1; cilia white with a brown line near base and some brown at tips. Hind wing orange-yellow, the costal area white; some dark brown irroration along vein 1; a narrow red-brown terminal band, ending in a point at submedian fold, its inner edge slightly waved; cilia brown, white at tips. Underside On the Synonymy of some European Diplopods. 407 of fore wing white, tinged-with yellow on disk, the costal area irrorated with red-brown ; hind wing orange-yellow, the costal area white, irrorated with red-brown except towards base, the terminal band formed by red-brown irroration. Hab. KE. Turxestan (Avinof’), 1 2 type. Exp. 22 mm. (7a) Noctuelia josialis, sp. n. 3S. Head and tegul orange-yellow, the latter with black-brown patches at tips glossed with blue, with orange-yellow stripes at sides and the patagia with some orange-yellow scales; abdomen black-brown with a cupreous gloss and orange-yellow subdorsal stripes, the genital tufts white; antenne black: frons with black patch; palpi black, the basal joint and base of 2nd joint yellow; femora whitish tinged with brown; ventral surface of abdomen with white stripe except at extremity. Fore wing black-brown with a cupreous gloss; an orange-yellow fascia along median nervure to near termen where its extremity is rounded; an orange- yellow streak on inner margin. Hind wing black-brown with a cupreous gloss; a broad orange-yellow stripe in and below the cell to near termen, extending to inner margin at base and narrowing somewhat with its lower edge oblique beyond the cell. Hab. VENEZUELA, Esteban Valley, Las Quiguas, 1 ¢ type. Exp. 80 millim. XXXVII.—On the Synonymy of some European Diplopods (Myriapoda), with Special Reference to Three Leachian Species. By RicHarD S. BAGNALL, F.L.S. One of the drawbacks to students of British Myriapods undoubtedly lies in the unsatisfactory state of the nomenclature. When one remembers that, amongst the Diplopods, there are so many instances of two (or more) species being so closely related as to be practically indistinguishable, except by a dissection and study of the male, one at once realizes how difficult it must be for a discoverer of a species so closely allied to one already known to decide which of the two was the one described by an older naturalist at a time when present-day methods were not used. A case in point: Brachyiulus pusillus, a graceful little Julid with a pair of yellowish stripes. down the back, was described by Leach from Edinburgh and London more than a hundred years ago. In recent years Verhoeff showed that there were two species, externally alike but abundantly 408 Mr. R. 8. Bagnall on the distinct in the structure of the male gonopods ete., describing one of them as new under the name of Brachyiulus littoralis, The dissection of male examples, however, from an abun- dance of British material proves that all our examples are referable to Verhoeff’s species. Surely, by deduction, one must refer the British material to Leach’s species, and so sink Verhoefi’s name as asynonym. And, further, another name must be found for the puszllus of Verhoeff (non Leach). The present memoir is an attempt to show my deductions as to the true synonymy of three of Leach’s species, from which it will be seen that new names will have to be found for Craspedosoma rawlinsi, Verhoeff (non Leach), and Brachyiulus pusillus, Verhoeff (non Leach). As existing names (now sunk as synonyms) may be found applicable, I leave this question to more capable hands. I have, however, suggested a new name for Craspedosoma simile, Attems (non Verhoeff), the issue in this instance not being complicated by old synonymy. Of four of Leach’s memoirs on Myriapods containing practically the same subject-matter, I have perused the following :— Leach, W. E. 1814-15. ‘A Tabular View of the Ex- ternal Characters of Four Classes ef Animals, which Linné arranged under Insecta ; with the Distribution of the Genera composing Three of these Classes into Orders &c., and Descriptions of several new Genera and Species.” In Trans. Linn. Soc. Lond. vol. xi. (1815) pp. 306-400 (Class II. Myriapoda, pp. 376-386). Leach, W. E. 1817. ‘The Characters of the Genera of the Class Myriapoda, with Descriptions of some Species.” In the ‘ Zoological Miscellany,’ ii. pp. 36-45 (with 10 plates). The following extract is from the first of these references :— [p. 379 | “Spec. 7. Julus pusillus. “ J. Segmento ultimo submucranato, corpore cinerascente nigro aut fusco-brunneo lineis duabus rufescentibus. “ Long. Corp. 5 ad 6 lin. “Habitat prope Kdinburgum sub lapidibus ; in Battersea fields, Londinum prope, inter graninum radices. “‘ Copulatione observavi. a ; Synonymy of some European Diplopods. 409 [p. 380] “8. Corpus rufescens lateribus lineaque lengitudinale dorsali fuseus brunneis. “ Dorsum lineis fortioribus exaratis, distantibus rectis sub- lveequalibus, Antennee fuse articulis dilutis. Pedes lutes- centes, - “Gen. 3. CRASPEDOSOMA.T [ Footnote ] This genus was proposed by my much lamented friend Richard Rawlins, Hsq., who discovered tlie first species, “Corpus lineare, depressum, segmentis lateraliter com- pressis, marginatis. Antenne articulo secundo tertio breviore. “* Segmentis lateribus medio prominulis. Spec. 1. Craspedosoma Rawlinsii. “C. dorso fusco-brunneo lineis quatuor punctorum albi- dortin, ventre pedibusque rufescentibus. « Long. Corp. 7 lin. “ Habitat inter muscos et sub lapidibus propre. Edin- burgum vulgatissima. Detexit R. Rawlins cujus nomen gerit. «e*® Segmentis lateribus postice productis. “ Spec. 2. Craspedosoma polydesmoides. “ C. dorso rufo griseo, ventre pallido, pedibus rufescentibus basi pallidis, angulo segimentorum postico setigero. “ Habitat in Danmonid prope Plymouth, sub lapidibus passim. Detexit Dom Montagu. “Corpus rufo-griseum, pedibus pallidioribus. Dorsum lineé longitudinaliter impressum. Segmenta valdé promi- nentia angulo antico rotundate ; postico retrorsum producto, setifero seté conic alba. Facies saturate rufo-grisea. Oculi atri, Antennee rufo-griseze sub-pilosula. Venter pallidus, ulbidus. Pedes rufescentes, basi pallidi.” [End of p. 380. ] Brachyiulus pusillus (Leach), non Verhoeff. Syn. Brachyiulus (Microbrachyiulus) littoralis, Verhoeff. Julus pusillus, Leach, pie Trans. Linn. ei Lond, xi. p. 379 ; 1817, Zool. Mise. iii. p. 8 In 1917 I brought forward B. B. (Microbrachyiuts) littoralis, Verhoeff, as British on the strength of a large number of Ann. & Mag. N. Hist. Ser. 9. Vol. ii. 30 410 Mr. R. 8. Bagnall on the examples found at Ainsdale, near Southport, in April 1916, which were kindly identified by Brélemann. Since then I have taken examples of the same species in the Forth Area of Scotland, in the counties of Northumberland and Durham, both inland and on the coast, and in other localities, including the Soufh Coast at Swanage. In every case expert examina- tions of the males were made by Mr. and Mrs. Brade-Birks, proving the species to be Verhoeff’s littoralis. Leach described J. pustllus from Edinburgh and London, and as I have secured material from one of these localities, and no British examples as yet dissected have been found to be referable to pusillus as diagnosed by Verhoeff, one is forced to the conclusion that when he demonstrated that there were two allied species, Verhoeff unfortunately gave the name Utioralis to what was in reality Leach’s species. I may have the opportunity this winter of going into the question of how far Verhoeft followed previous continental authors as regards B. pusillus ; in any case, a new name must be found for B. pusillus of Verhoeff (non Leach), but as the names boleti, Am Stein (1857) and stuwbergit, Fanzago (1875), are given as synonyms of pusil/us by Latzel, and might be referable to either species, I dare not go further in the matter just now. Craspedosoma rawlinsii, Leach. Syn. Craspedosoma simile, Verhoeff, non Attems. Craspedosoma rawlinsn, Leach, 1814, Trans. Linn. Soc. Lond. xi. p- 880; 1817, Zool. Mise. ili. p. 36, pl. exxxiv. figs. 1-5. Craspedosoma raulinsii, Samouelle, 1819, The HEntomologist’s Useful Compendium, p. 114. Craspedosoma rawlinsti, var. stmile, Verhoeff, 1891, Berl. Ent. Zeitsch. xxxvi. pp. 129-180. Craspedosoma simile, Verhoeff, 1910, Sitzungsber. Ges. Naturf. Freunde, no. 1, pp. 19-62, figs. Verhoeff first described his s¢mele in 1891 as a variety of vawlinsi, but later raised it to specific rank, and in 1910 (reference above cited) he reviewed the genus Craspedosoma (pp. 80-55) and gave the tables of his subdivisions, species, and subspecies. ‘That the species he regards as rawlinsid and simile are well characterized is distinctly demonstrated, but here again I contend that Verhoefi’s species should be referred to the species Leach described. In-1912 I sent Verhoeff specimens of Craspedosoma from Gibside, County Durham, which he returned as C. simile and Synonymy of some European Diplopods. —~ 411 C. simile rhenanum, and as such [ recorded them *, Examples identified by Ellingsen from Norway (a large series) were all referred to simile (and subspecies and varieties thereof ) by Verhoeff (Zool. Anz. xxxix. pp. 499-511, May 1912), whilst the C. raw/insii: recorded trom Holland in moles’ nests by Father Heselhaus, 8.J. (Tijdschrift voor Ent. lvi. 1913, p. 240), was later (/. c. lvil. 1914, p. 80) referred by Verhoef to simile. It therefore seems that no examples of what he regards to be rawlinsti have been examined by Verhoeff from our faunal area, all so named being referred to simile, and until the reverse is proved I consider it distinctly advisable to regard Verhoeff’s sime/e as a synonym of raw- finsu, Leach. in the meantime, it is to be hoped that more British examples may be secured for study. Thus a new name is necessary for the rawlinsti of Verhoeff (non Leach), but as Latzel gives the names marmoratum, C. Kk. (1847), and gibbosum, Am Stein (1857), as synonyms, it would not be wise to suggest a new name without further research. Craspedosoma leachi (nom. nov.), Bagn. Syn. Craspedosoma simile, Attems (non Verhoefi), 1895, Sitz. k. Akad. Wiss. Wien, math.-naturw. Cl. civ. pp. 75-76. A species allied to mutabile, Latz. When Attems described it he was aware of Verhoeff’s var. simile of rawlinsit, but the raising of this form to specific rank rendered it necessary to give another name to Attems’s species. Polymicrodon polydesmoides (Leach). Syn. Polymicrodon latzeli (Verhoeff ). Craspedosoma polydesmoides, Leach, 1814, Trans. Linn. Soe, Lond. xi. p. 880; 1817, Zool. Misc. iii, p. 36, pl. cxxxiv. figs. 6-9 ; Samouelle, 1819, The Entomologist’s Useful Com- pendium, p. 114. ¢ Atractosoma polydesmoides of later British authors. Atractosoma latzeli, Verhoeff, 1891, Berl. Ent. Zeitsch. xxxvi. pp. 127-128, figs. 4-6. Polymicrodon latzeli, Verhoeff, 1897, Berlin. Archiv. f. Natur- gesch, i. pp. 129-188; 1912, Trans. Nat. Hist. Soc. North- umberland & Durham, n. s., iv. pp. 159-166, pl. x. figs. 4-7. Also Polymicrodon latzeli of recent autbors. Atractosoma latzeli was described by Verhoeff in 1891 from the south of England, his description being based upon * “Brief Records of Chetechylene vesuviana, Newp., and other Myrio- pods new to the British Fauna,” The Zoologist, July 1912. 30* 4 * Mr. L. B. Prout on new a solitary poorly preserved male example, and six years later the same author instituted the genus Polymicredon for that species. In 1911 I submitted numerous examples of P. lat- geli to Verhoeff from the north of England, who (1912) wrote at some length upon this material. Nowhere have I seen any attempt to show how Jatzeli differs from Leach’s species polydesmoides, described somewhat over a hundred years ago (and figured) from South Devon, of which Samouelle says ‘inhabits Devonshire under stores. It is common all along the borders of Dartmoor and on the southern coast. It was once taken by Dr. Leach in the garden of the British Museum.” I have twice stated that there appeared to be two allied species, referring the commoner to datzeli and the rarer to polydesmoides ; but in recent years I have made a closer study of the Diplopoda, and I am convinced that the so-regarded rarer species is in reality the later larval stages of latzeli. Verhoeff states (1912, p. 165) that the occurrence of P. latzelt in the north of England is very noteworthy from the zoogeographical point of view “ since this is the first time that a Craspedosomid of ‘ Atractosoma-habit’ has been recorded from the northern region affected by the Ice Age. This is by far the most northerly record for any such Craspe- dosomid.” Asa matter of fact, the species is not uncommon in Scotland and is one of the commonest Diplopods in the northern counties of England ; it is probably as common in the midlands and the south, where I have collected it in North and South Devon, Bath, Oxford, Swanage, Ports- mouth, Isle of Wight, and in the. London district. T see no grounds whatever for the retention of the name latzeli, which I consider must fall as a synonym of poly- desmoides. XXXVITI.—New Lepidoptera in the Joicey Collection. By Louis B. Provt, F.E.S. Family Zygenide. 1. Caprima chrysosoma. ? —31 mm. Head and body orange-ochreous ; antennal shaft blackish, with blue irroration (tips lost); tarsi blue-blackish on upper side; tibial spurs almost entirely atrophied. Lepidoptera in the Joicey Collection. 413 Fore wing long and narrow, more recalling Aphanto- cephala, or even Docleopsis, than Caprima; SC* wanting, R* just stalked, DC acutely inangled; black, irrorated with blue ; a small ochre-yellow patch at base, produced on the space between costal edge and vein C to a length of nearly 2 mm. ; a narrow ochre-yellow streak from SC at4or5 mm. from base, running very obliquely in direction of termen but not quite reaching SM’. Hind wing black with blue irroration ; abdominal margin ochre-yellow for a width of over 1 mm. At termen appear- ing to widen on account of some yellow irroration. Underside similar, but in part with stronger blue and purple reflections, the yellow markings somewhat extended, the fore wing with some additional yellow scales in and distally to the posterior angle of the cell and at distal end of abdominal margin. Aru Is., March-May 1916 (W. J. C. Frost). Family Geometrida. Subfam. Srzrrazivz. 2. Semeopus subtranslucens. ? .—33 mm. Head and body nearly concolorous with wings; antennal joints not projecting ; ciliation fully as long as diameter of shaft ; pectus not densely hairy. Fore “wing with apex acute, termen rather irregularly subcrenulate; proximal areole ample, distal minute, SC’ arising well down on the stalk of SC’~°’; subdiaphanous whitish, with slight pink reflections and with some some- what olivaceous* irroration; costal margin and _ base olivaceous *; markings olivaceous *, antemedian line before one-third, excurved in cell and in submedian area; cell- mark ocelloid ; median line dentate, from five-eighths costa, oblique outwards to SC°, somewhat incurved between the radials and strongly behind middle, reaching hind margin about middle; a duplicating line just beyond the median commences about R’, feeble at first but becoming distinct and thickening, almost connected with median by olivaceous shading in posterior part ; postmedian line dentate, placed midway between this and termen or slightly nearer the * “ Buff with a tinge of olive” would perhaps better describe this shade. 414 Mr. L. B. Prout on new latter, very oblique outwards between SC* and SC’, where it is acutely angulated, incurved and thickened into two spots between the radials and again (though less strongly) behind M?*; terminal line olivaceous, accompanied by tri- angular interneural dots (pointing proximad). Hind wing with termen irregular, dentate, the teeth at R' and R* longest and sharpest; R? very shortly stalked, M°* arising rather nearer R° ; Irroration in proximal half in part fuscous; first line wanting; cell-spot round, black, without pale centre; the other ‘markings corresponding to those of fore wing. Underside paler ; fore wing with costal margin somewhat olivaceous ; both wings with cell- spot ocelloid, median and postmedian and terminal markings nearly as above. Sierra del Libane, Colombia, 6000 feet (H. H. Smith). Rither recalls S. trygodata, Warr. (Nov. Zool. xi. 36), but distinguishable by the relatively long antennal ciliation and longer teeth of termen of hind wing, as well as by the venition. These two species together with ‘ Trygodes” pertumna, Schaus, so far bridge over the supposed gap between Semeopus and Tr ygodes that I doubt whether the latter can be regarded as more than a section. 3. Anisodes (Brachycola) clandestina. 3 .— 32 mm. Structure of antenna, palpus, legs, areole, etc., approxi- mately as in absconditaria ; palpus with secoimd joint beneath perhaps clearer whitish and more appressed-scaled ; abdominal cavity enormously developed, the sternal tuft less developed. Smaller, wings shorter, irroration fairly strong, purple-reddish (in absconditaria extremely weak, browner), underside more strongly marked, including some rather noticeable pink irroration at middle of costa of hind wing. Khasis, type in coll. Joicey; 1 g in coll. L. B. Prout (genitalia examined by Rev.C. R. N. Burrows). Pundaloya, Ceylon (coll. ‘ving Mus.). Penang and Gunong IJjau (coll. Tring Mus.)—ocelloid form of central spot persisting (in type giving place to punctiform).—Larut Hill, Perak, 4360 {t., 21st April, 1898 (S. S. Flower), 1 3; Singapore (fH. N. Ridley), a good series; Sarawak, 1 ¢ 2 (Wallace) «coll. Brit. Mus.). This is essentially the obrinaria of Hampson’s ‘Fauna of British India, Moths,’ iii. p. 446, although, on account of Lepidoptera in the Joicey Collection. 415 shortage of material and preponderance of ? ? in the British Museum collection at that time, he mixed in some very heterogenous elements. A. odrinaria, Gn.=caligata, Walk.=similaria, Walk., and A. pallida (bon. sp.?) belong to the typical section Anisodes and have no areole. A. odli- viaria, Walk.=:suspicaria, Snell., to the section Perizera, Meyr. (nec Hamps.), also with no areole, but with hind femur tufted. I should have considered this a local form—more rufes- cent—of niveopuncta, Warr. (Nov. Zool. iv. p. 48), but the genitalia show that it has reached full specific rank. In niveopuncta the uncus is more long and slender, the valves very different, the penis has a very distinct cornutus (or perhaps bunch of cornuti), and’ there is a better developed pair of hair-brushes on the 4th (?) abdominal segment. 4. Flavinia allogaster. 3 .—30 mm. Closely similar to circumdata, Maassen (Stiibel’s Reisen, Lep. pp. 101, 130, t. iv. f. 22). Abdomen with a pale dorsal line as in alcidamea, Druce (Proc. Zool. Soc. Lond. 1890, . 498). ' rae wing with the apical black border broadened, its proximal edge on the upper surface at R* being over 4 mm. from the apex, at R* fully 3 mm. from termen, on the under surface very slightly less broad; black on hind margin slightly broadened. Hind wing with the black distal border above less narrowed between R' and M’. Peru, without more exact locality. Type in coll. Joicey (ex Schaus) ; three in coll. Brit. Mus. from the same source, mixed with true circumdata. Family Drepanide. 5. Cyclidia substigmaria, Hbn. It has been unaccountably overlooked that this species was described and figured by Hubner (‘ Zutriige,’ iii. 29, figs. 519- 520) from ‘ China,” 7. e. no doubt S. China, and represents unmistakably the form later described by Walker (List Lep. Ins. xxiv. 1121) from Hong Kong as “ Abravas”’ capitata, though the last-named author neglects to describe the underside. The common Indian race, which has for so long passed as substigmaria (see, for instance, Hampson’s ‘ Fauna 416 Mr. L. B. Prout on new of British India, Moths,’ vol. i. pp. 327, 828, fig. 225, Strand in Seitz ‘ Macrolepidoptera,’ vol. 11. p. 196, pl. 23f), therefore remains without a name and J propose to call it Cyclidia substigmaria superstigmaria, subsp. nov. Ground- colour whitish, markings fawn-brownish, always more or less shadowy, subtornal spots at inner margin of fore wing well defined, cell-spot of hind wing above black. Dharmsala, Kulu, Sikkim, Burma, etc.; type ¢ (Dar- jeeling, ex coll. Lidderdale) in coll. Joicey. From Vrianatong, Tibet, comes a greyer, more suffused race, with the cell-spot of the hind wing above generally less deep biack than in the form superstigmaria, the sub- tornal brown markings of fore wing not, or scarcely, more strongly developed than the posterior end of the line which precedes them proximally. I name this substigmaria inter- media, subsp. nov. Type in coll. Joicey. Typical substigmaria from China and Formosa (also, in Tring Museum, from Tonkin) is very similar to subsp. intermedia, but less dark grey, the cell-spot of the hind wing above still weaker, the subterminal dots generally connected by stronger grey shading, the subtornal markings of the fore wing frequently confluent with the preceding line so as to form a brownish pyramid, the cell-spots generally less intensely black. The Japanese representative, nigralbata, Warr. (Nov. Zool. xxi. p. 401), may possibly be a separate species, though most collections have mixed it with “ capitata”’ (i. e., substigmaria substigmaria), not even recognising the marked distinctions as racial. Family Arctiidae. Subfam. Lrrvosrane. 6. Caprimima esthla. gS 2? .—31-32 mm. Similar to C. ca/ida, Walk., but larger. The yellow on patagia aud tegule more extended. Fore wing with the yellow area broad, the black at base rather broad, especially in the 2, where it curves outwards along costal margin, the black costal margin in middle very narrow in ?, wanting in ¢. Hind wing rather more produced in tornal region than in calida, the black along abdominal margin broad, at apex moderately broad, at distal margin between M’ and tornus, on the other hand, quite narrow (recalling zsabelle, Rothsch.) ; Lepidoptera in the Joicey Collection. 417 apical area wanting the “cupreous-red”’ cloud which in calida is always present beneath and generally also above. Goodenough I., 2500-4000 ft., Apr. 1913 (A. 8. Meek). Type 3,2 2 2 in coll. Joicey. Also in Tring Museum, Possibly a local form of calida, though very different from Hampson’s “ ab, 1.” a Subfam. Arcruynz. 7. Helactinidia tornensis. 3 .—30 mm. Similar to chiguinda, Druce. fore wing slightly more rounded, rather blacker brown; streak behind cell longer, crossing base of M’?; outer band broader, not indented at posterior extremity of cell. Hind wing without the black costal area ; the streaks on submedian fold and in abdominal area wanting. Torné, Cauca Valley, Colombia, August 1907. Type in coll. Joicey. Family Hypsida. 8. Phegorista bisignibasis. ? .—58 mm. Head and thorax above black ; face marked with white at lower extremity, occiput and front of thorax narrowly marked with white; breast and palpus beneath (to near end of second joint) orange ; abdomen orange with narrow black anterior rings ; legs orange marked with black, tarsi mostly black ; antennal joints not projecting. Fore wing light reddish orange, along costal and hind margins narrowly and irregularly black; a small black patch at base, with its outer edge convex and containing a pure white basal spot, close to costa; apical region black, its boundary rather straight from proximal end of areole in direction of tornus but narrowly interrupted at submedian fold, followed by a black subtornal and a small whitish tornal spot between SC* and M? placed in the apical patch near its proximal edge, slightly broader than in agaristoides, Bdv., but proximally indeuted in the middle; fringe spotted and tipped with white. Hind wing scarcely more reddish ; a black distal border about asin agaristoides. Underside similar, fore wing without white tornal spot. Tanga, German H. Africa, February. Type in coll. Jeicey. 418 Mr. T. D. A. Cockerell—Descriptions and 9. Phegorista trialbata. 3 .—85 mm. Akin to agaristoides, differing as follows :—Palpus with third joint shorter ; second joint beneath narrowly marked with white (in agaristoides less narrowly with orange). Fore wing above with the oblique streak behind cell larger and narrower, pinkish white; a small long-oval pinkish- white spot in front of it, beyond middle of cell 3 subapical patch white, as in some agaristoides, but considerably broader and somewhat longer, reaching vein M?’, its distal edge irregularly curved; no supplementary spot on sub- median fold ; fringe not white at apex. Hind wing with the border narrower than in agaristotdes ; orauge ground-colour less reddish than in most agaristoides. Fore wing beneath orange as far as the black apical area, only with the costal margin narrowly black. Uganda (H. 8. Gledhill). Type in coll. Joicey. XXXIX.—Descriptions and Records of Bees —LXXX1. By T. D. A. Cocxrretzt, University of Colorado. Augochlora (Odontochlora) lyoni, sp. n. ? .—Length about 8°5 mm., anterior wing 6. Robust, black, with strong metallic tints as follows: clypeus (which is smooth, with well-separated large punc- tures) green in middle and purplish at sides ; cheeks blue- green next to orbits, otherwise purplish; region on each side of antennze obscurely purplish ; vertex greenish ; tubercles bright green; mesothorax with disc obscurely green, margins purple; scutellum greenish; postscutellum and area of metathorax purple; mesopleura dark purple edged with blue; first abdominal segment suffused sub- laterally with bright green and purple; second with similar colours, but less distinct, the remaining segments black. Flagellum ferruginous beneath ; front dull and granular ; ocelli not enlarged; process of labrum broadly truncate, slightly bigibbous; mesothorax densely punctured, except the posterior middie, where the punctures are sparse on a shinmg ground; area of metathorax with numerous very fine more or less wrinkled striz; posterior face with no Records of Bees. 419 sharp margin; tegule reddish. Wings dusky, stigma and nervures pale yellowish brown ; first r. n. meeting second t.-c. Legs reddish piceous, with pale pubescence; hind spur simple. Abdomen shining, thinly hairy, with very small punctures; first dorsal segment with a low tubercle on middle of disc ; first ventral segment with a long slender spine ; last dorsal segment with fuscous hair. San Julian, Venezuela, July 19, 1900 (MW. W. Lyon, Jr.). U.S. Nat. Museum. Nearest to the Mexican A. zophodes (Halictus zophodes, Vachal), but distinguished by the smooth and shining surface of clypeus, with well-separated punctures. The tubercle on the first dorsal segment of abdomen recalls the Australian Halictus mirandus, Cll. Agapostemon viequesensis, sp. n. ? Length about 8 mm., anterior wing 6. {fead and thorax brilliant bluish green ; lower margin of elypeus broadly black ; labruin and mandibles red, the latter black subapically ; sides of face and front suffused with purple-blue ; flagellum dull ferruginous beneath, but the last joint bright ferruginous on both sides ; clypeus and supraclypeal area shining ; mesothorax dull, minutely granu- lar ; scutellum rather yellowish green, shining, somewhat bigibbous; area of metathorax purple, poorly defined, with ohscure rugse; posterior truncation bright green, with a sharp edge; tegule light ferruginous. Wings dusky hyaline, stigma clear honey-colour ; second s.m. receiving first r.n. a considerable distance from its end. Legs light ferruginous, with pale yellowish hair, that on outer sides of tibiae more or less fuscous. Abdomen mainly yellowish green, with blue-purple shades on apical half, but the first three segments have transverse median bands of reddish brown, where the surface is not metallic ; bases of segments with pale tomentum ; venter mainly pale fulvous. Vieques Island, Porto Rico, West Indies, Feb. 1899 (dug. Busck). U.S. Nat. Museum. In Vachal’s table it runs out at 14, and it is scarcely to be compared with any described species. The extreme bases of the abdominal segments are testaceous, but the apical margins sliding over them are not noticeably dis- coloured. Neocorynura discolor (Smith). Augochlora tisiphone, Gribodo, is a synonym. Smith’s 420 Mr. T. D. A. Cockerell—Descriptions and type was from Oajaca, and Gribodo’s was marked ‘“€Oajuca? ” (sic). The following species are now recorded from new localities : — Augochlora radians (Vachal). Cacao, Trece Aguas, Alta Vera Paz, Guatemala, April 25 (Schwarz and Barber). This is probably the same as the so-called A. vesta from Mexico in the British Museum, but it is not true vesia. A. fervida, Smith. Tlahualilo, Durango, Mexico, at peach blossoms (4. W. Morrill). A, illustris (Vachal). Colombia, from C, F. Baker collection. A. phemonoé (Schrottky). Sapucay, Paraguay, March (W. T. Foster). A. nigrocyanea, Ckll. Tampico, Tamaulipas, Mexico, Dec. 6 (F. C. Bishopp). A. esow {(Vachal). Paraiso, Canal Zone, Panama, Jan. 18 (Aug. Busck). A. seminigra, Ckll. Cordoba, Mexico, Jan. 20 (F. Knad). The A. nigrocyanea females from Tampico are variable ; one has strong purple tints on. apical part of abdomen, which the other lacks ; the latter has the mesothorax black. Xenoglossa howardi, sp. n. 3. (Type.)—Length about 12 mm., anterior wing 9. Black, including the clypeus and antenne; mandibles fulvous apically, bidentate, but with a slight notch on inner side indicating the rudiment of a third tooth; labrum brown at sides, covered with appressed pale hair; maxillary palpi 5-jointed ; hair of head long and creamy white, with some fuscous hairs on vertex and below antenne ; hair of thorax above clear reddish fulvous, without black; a large patch on middle of mesopleura, and tubercles, with dark fuscous hair; tegule ferruginous. Wings dusky. Legs black, the spurs stramineous, and tarsi at apex ferruginous ; hair of middle and posterior tibiz and tarsi dark brown, but the femora and anterior legs with pale hair. Abdomen shining black, minutely punctured, hind margins of segments 2 to 4 suffusedly reddened; no hair-bands, but base of second segment at sides with thin greyish hair; venter with thin whitish hair. 2? .—Length about 12°5 mm. Similar to the male, but all the legs with dark brown Lecords of Bees. A421 hair ; dark brown hair on sides of thorax more extensive ; second and third abdominal segments with a thin transverse band of greyish tomentum, not conspicuous. Type (male) from the Federal District, Mexico (J. R. Inda, 56). U.S. Nat. Museum. Female from Oaxaca, Mexico, Sept. 18 (L. O. Howard). Related to X. assimilis (Smith), but without the black patch of hair on thorax above in female. The male antennz are formed as in X. pruinosa (Say). The species belongs to the subgenus Peponapis of Robertson, though differing from his type-species in the black clypeus of male and reiluction of pale hair on female abdomen. Allodape candida, Smith. ?.—Mkonumbi, near Lamu, Tana River, E. Africa, Sept. 1892 (Chanler Exped.). This differs slightly from Smith’s description, and from a specimen from Abyssinia, sent by Gribodo, in that the light band on clypeus is not at all widened at the lower end. Leptergatis globulifera, sp. nu. 3d .—Length 6-6°5. mm. Black, with the long flagellum dull ferruginous beneath, tegulze rufo-piceous, legs more or less suffused with reddish, the tarsi and tibiz at apex ferruginous. Close to L. armata (Smith), differing thus: scape dark ; ocelli closer together ; clypeus and labrum entirely black, mandibles mainly dark reddish ; tegule darker ; abdominal hair-bands less distinct; wings a little more dusky. The hind legs are practically as in ZL. armata. The co-type has the mandibles paler, with a large pale yellowish spot, beyoud which they are ferruginous. Venezuela; type from Aroa, Dec. 12, 1910 (M. A. Carriker). U.S. Nat. Museum. Another is from Lagunita de Aroa, 2000 ft. alt. (47. A. Carriker). Prosopis holomelena, sp. n. ? .—Length about 6 mm., anterior wing 4°5. Entirely black, without light markings; robust, with no depression between first and second dorsal abdominal seg- ments ; clypeus long, dull, the punctures very indistinet ; apical part of flagellum bright ferruginous beneath; punc- tures of mesothorax and scutellum excessively minute, the 422 Mr. T. D. A. Cockerell— Descriptions and surface between them microscopically rugulose; area of metathorax with irregular ruge; tegule black. Wings slightly dusky, stigma and nervures very dark ; second s.m. long; recurrent nervures meeting the transverso-cubitals ; tibie and tarsi with some pale hair. Abdomen shining, impunctate, the surface with a delicate microscopical tessellation. Buitenzorg, Java, March 10, 1909 (Bryant and Palmer). U.S. Nat. Museum. Nearest to P. impunctata, Friese, but easily separated by the entirely black face. Prosopis coroicensis, sp. n. 3 .—Length about 7°5 mm., anterior wing 6°2. Black, robust, without yellow markings on thorax or legs ; face long, eyes very long; clypeus (except a narrow dark stripe on each side), large supraclypeal mark (rounded above), lateral face-marks (extending along orbital margins halfway up frent, where they end obtusely, shaped like feet on tip-toe, with very long tapering toes), all bright chrome- yellow ; antennz piceous; scape very short; mandibles stout, suffused with reddish ; front dull, very ’ densely and finely punctured ; mesothorax and scutellum dull, with very large well-separated punctures; mesopleurz with large sparse punctures ; area of metathorax with coarse transverse and longitudinal ridges; posterior truncation. very coarsely sculptured, flat, with well-defined margins ; tegule piceous. Wings deep fuliginous ; first r.n. joining first s.m. con- siderably before its end. Legs more or less reddish, the auterior tibisee dusky ferruginous in front. Abdomen shining, without hair-bands; first two segments quite strongly punctured, third with minute punctures ; first ventral segment emarginate at apex. Coroico, Yungas, Bolivia, May 1, 1899. U.S. Nat. Museum. No collector’s name is given. “ By the venation this resembles P. petroselini, Schrottky, but it is easily separated by the fuliginous wings and other characters. Prosopis tricolor, Schrottky. ? .—Differs from the male thus : clypeus with an elongate- cuneiform rufo-fuscous mark on each side ; antennz entirely ferruginous; yellow band on prothorax interrupted in Records of Bees. 423 middle ; marks at bases of tibie cream-colour. Schrottky only described the male. : San Bernardino, Paraguay, Oct. 21 (K. Fiebrig). U.S. Nat. Museum. Prosopis flavohumeralis, sp. n. ? .—Length about 6 mm., anterior wing 4°5. Black, with yellow markings ; mandibles ferruginous ; labrum black ; clypeus yellow except narrow lower margin and a stripe on each side, failing above ; supraclypeal mark broadly subtriangular, while above it, on front, are two narrow yellow marks close together; lateral face-marks extending nearly to summit of eye, where they are broadly but very “obliquely truncate, and diverge a little from the orbital margin; scape and flagellum: dusky ferruginous beneath, darker above ; front very densely and minutely punctured, vertex more coarsely ; tubercles and the sharp projecting anterior lateral angles of prothorax yellow, but no other yellow on thorax; mesothorax and scutellum perfectly dull and coarsely punctured ; area of metathorax with raised lines in the form of a square, but without ve sculpture, except a microscopical cancellation all over; posterior truncation distinct; tegule with a yellow spot. Wings dusky; recurrent nervures meeting transverso- cubitals ; marginal cell broad (deep). Legs with anterior tibiz yellow in front, the others at base ; tarsi more or less reddish. Abdomen shining, without hair-bands ; first seg- ment distinctly though minutely punctured, second and third extremely sparsely and indistinctly. San Bernardino, Paraguay (K. fMiebrig). U.S. Nat. Museum. In Schrottky’s tables of Paraguay species this runs to P, itapuensis, Sky., but differs by the dusky wings and spots on angles of prothorax. It seems to closely resemble P. lychnis, Vachal, differmg in the punctuation of the abdomen. Prosopis howardiella, sp. n. g .—Length about 3°5 mm. Head all black except a large obtusely trilobed (the sides concave) pale yellow patch on clypeus; scape black; fiagellum thick, ferruginous beneath ; thorax entirely black ; mesothorax and scutellum with sparse very minute punctures on a microscopically tessellate surface; area of metathorax 424 Mr. T. D. A. Cockerell—Deseriptions and large, with a few small irregular basal plice, and a median raised line continuous to hind margin; posterior truncation ’ of metathorax not clearly defined as usual, its upper lateral corners not defined at all, but its upper middle separated by a short ridge from the basal area, while an oblique ciliated ridge limits it on each side; abdomen impunctate, microscopically transversely lineolate, first segment nar- rowed. Wings clear, very faintly dusky apically ; recurrent nervures ending a little before the transverso-cubitals ; second submarginal cell nearly square, its inner and outer sides parallel; bases of tibiz, and anterior tibiz in front, cream-colour ; tarsi pale ferruginous. Oaxaca, Mexico, April 30 (L. O. Howard). U.S. National Museum. . Looks like some small Pemphredonid wasp, but is a true bee, with many plumose hairs on body. It is more or less related to Vachal’s P. recisa, P. puerula, P. fissa; &c., but much smaller and very distinct. Prosopis subgrisea, sp. n. 9 .—Length about 7 mm., anterior wing 5°3. Black, with yellowish-white or brownish-white markings ; mandibles and labrum black ; clypeus long, black, the lower margin suffusedly reddish, but with a cream-coloured stripe running down its middle (not quite reaching upper end), not quite so broad as the area on either side ; supraclypeal mark small, roundish; lateral face-marks linear, extending along orbital margins nearly halfway up front; scape and base of flagellum ferruginous, rest of flagellum black above and faintly reddish below; front appearing granular ; upper part of prothorax with linear light margin, and greater part of tubercles light ;.a light band covering anterior half of scutellum, a band on postscutellum, and axille light ; mesothorax dull, coarsely punctured; area of metathorax with coarse ruge ; posterior truncation and sides of meta- thorax densely covered with pale grey tomentum ; pleura sparsely punctured; tegule with a light spot. Wings brownish hyaline, with the costal field, including marginal cell and beyond, fuliginous ; hind tibie with rather more than basal half white. Abdomen dullish, the punctures excessively minute and close; first and second segments with yellowish-white marginal hair-bands, third to fifth with hind margins obscurely pallid; apex with dark fuscous hair. Records of Bees. 425 San Rafael, Jicoltepec, Mexico. U.S. National Museum. From the Ashmead collection ; no doubt collected by C. H. T. Townsend. Resembles P. mexicana, Cresson, but easily separated by the linear lateral face-marks, and other characters. It is evidently closely allied to P. maculipennis, Smith, known only in the male, but that has yellow markings and the first abdominal segment rather strongly punctured. Prosopis knabi, sp. n. 3 .—Length about 3°75 mm., anterior wing 3. Black, with yellow markings ; scape black, broadly red at end, and largely in front; flagellum entirely bright ferruginous, a little darker above; clypeus entirely, sub- triangular supraclypeal mark (broader than long), and lateral face-marks ail light yellowish, the latter ending obtusely on orbital margin about halfway up front (former practically as in P. episcopalis, Ckll.) ; pale marks of thorax confined to tubercles and a broadly interrupted line on prothorax above ; tegule testaceous, hyaline in front, with a yellow spot ; mesothorax closely and strongly punctured, scutellum rather more sparsely, the surface between the punctures smooth ; base of metathorax with strong longitudinal and transverse rugz, but the sculpture is mainly and essentially . transverse ; whole sides of thorax strongly punctured, the metathorax at sides bare (without grey tomentum) ; knees, anterior tibiz (except a large patch behind), middle and hind tibie very broadly at base and narrowly at apex, and the tarsi all pale yellow. Wings clear; stigma and nervures sepia ; first recurrent nervure joining first submarginal cell a short distance before its end. Abdomen appearing im- punctate under a lens, but the microscope shows minute punctures on first segment. Champerico, Guatemala, Aug. 4,1905 (frederick Knab). U.S. National Museum. This minute species recalls some of those of the United States, such as P. modesta, Say, but it will be readily known by the red flagellum and transverse rugs at base of metathorax. The foliowing localities are new :— Prosopis mexicana, Cresson. Tampico, Mexico, Dec. 15 (EZ. A. Schwarz) ; Frontera, Mexico. Prosopis azteca, Cresson. San Rafael, Jicoltepec, Mexico (L. O. Howard). Ann. & Mag. N. Hist. Ser. 9. Vol. ii. 31 426 On some Fishes from the Shari River. XL.— On some Fishes from the Shari River, with Descriptions of Two new Species. By G. A. BOULENGER, F.R.S. (Published by permission of the Trustees of the British Museum.) M. A. Baupon, Administrator of the Ubanghi-Shari Colony, French Equatorial Africa, has kindly sent me, for the British Museum, a little series of small fishes from the Shari River, containing examples of two species not included in Dr. Pelle- grin’s excellent book ‘Les Poissons du Bassin du Tchad,’ and of two others that are undescribed. The genus Barbus, as yet unknown from that Basin, is represented by two species: B. pleuropholis, Blgr., pre- viously recorded from the Congo, the Aruwimi, and the Uelle, and B. baudoni, sp. n. The Cyprinodonts belong to two species: Haplochilus acuticaudatus, Pellegr., and I, hutereaut, Blgr., the latter recently discovered in the Uelle. Other species are Anabas petherici, Gthr., Tilapia melanopleura, A. Dum., Hleotris nana, Blgr.*, and Ander- sonia brevior, sp. n., belonging to a very remarkable genus of Siluride, of which a single species was known: A. leptura, Blgr., from the Upper Nile and the Bahr-el-Gebel. Barbus baudoni. Depth of body equal to length of head, 32 to 32 times in total length. Snout rounded, shorter than the eye, which is 22 times in length of head and equals interorbital width ; mouth small, terminal, with thin lips; no barbels. Dorsal IIL 8, equally distant from centre of eye and from caudal, border very feebly concave; last simple ray not enlarged, not serrated, a little shorter than head. Anal II] 5, not reaching caudal. Pectoral about ? length of ‘head, not reaching ventral; base of latter below middle of dorsal. Caudal peduncle 13 times as long as deep. Scales radiately striated, 23-2472, 2 between lateral line and ventral, 8 round caudal peduncle. Yellowish brown above, silvery beneath ; a band of crowded black dots from the gill-opening to the base of the caudal ; on this band, three round black spots, the first just in front of the dorsal, the second just behind the latter, * These specimens connect the Nile fish with £. wellensis, Bler., which is probably not entitled to stand as a distinct species. On new South-American Batrachians. 427 the third at the base of the caudal; a fourth black spot above the anterior rays of the anal. Total length 30 mm. Allied to B, trisptlomimus, Blgr., from the Ogowe and Lower Congo. Andersonia pellegrint. Depth of body 9 times in total length, length of head 6 times. Head 14 times as long as broad; snout obtusely pointed, as long as postocular part of head, 3 times as long as diameter of eye, which is ? interorbital width, “Maxillary barbel twice as long as inner mandibular, and 2 length of head. Median occipital process 34 times as long as broad, narrower than and 1} times the length of the laterals. Dorsal I 6, twice as distant from end of snout as from caudal, first ray as long as head. Anal 9. Pectoral 2 length of head. Caudal peduncle a little more than 4 of the total length, 24 dorsal and 21 ventral scutes, the last 9 on caudal peduncle. Greyish above, with four rather indistinet dark bars across the back; dorsal blackish in the distal third. Total length 42 mm. Closely allied to A. leptura, Blgr. Distinguished by the smaller eye and the different proportions of the occipital processes, Named in honour of the distinguished author of the ‘ Poissons du Bassin du Tehad.’ XLI.— Descriptions of new South-American Batrachians. By G. A. BOULENGER, F.R.S. (Published by permission of the Trustees of the British Museum.) Phyllobates kingsburyt. Head slightly longer than broad. Snout rounded-sub- truncate, projecting beyond the mouth, as long as the orbit ; loreal region vertical ; nostril nearer the tip of the snout than the eye; interorbital space broader than the upper eyelid; tympanum very distinct, half the diameter of the eye, 3 to 4 {imes its distance from the latter. Fingers moderate, first and second equal, or first slightly the longer ; ae rather 3 428 Mr. G. A. Boulenger on new small; subarticular tubercles feebly prominent. 'Tibio- tarsal articulation reaching the eye; tibia half the length of head and body. Toes moderate, perfectly free, the disks larger than those of the fingers but smaller than the tym- panum ; subarticular tubercles feebly prominent ; two small metatarsal tubercles, inner oval, outer round; an oblique fold along the distal half of the tarsus. Skin of upper parts finely shagreened, of lower parts smooth. Brown above, with a paler dorso-lateral streak ; a black streak round the snout, continued, as a broad band, on the side of the body ; usually a white streak along the upper lip, continued along the body to the groin, edged below, on the body, by a black streak or series of spots; limbs brown, with dark brown spots, arm and thigh lighter, with a dark brown streak in front and behind ; lower parts white, uniform on throat and breast mottled with greyish brown. From snout to vent 28 millim. Four specimens from El Topo, Rio Pastaza, Hastern Ecuador, altitude 4200 feet; from Mr. M, G. Palmer’s collection, 1912. Named in pious memory of my late Attendant, Frederick Kingsbury, killed in action in Palestine, Feb, 25, 1918. Dendrobates ranoides. Head slightly longer than broad. Snout truncate, very feebly projecting beyond the mouth, longer than the eye ; loreal region vertical ; nostril nearer the tip of the snout than the eye ; interorbital space broader than the upper eyelid ; tympanum very distinct, 3 the diameter of the eye, 3 times its distance from the latter. Fingers rather slender, first and second equal; disks small, not much wider than the finger ; subarticular tubercles very indistinct. Tibio-tarsal articulation reaching the eye; tibia half the length of head and body. Toes slender, perfectly free, the disks larger than those of the fingers but only about half the diameter of the tympanum}; subarticular tubercles feebly prominent; two small metatarsal tubereles, inner oval, outer round ; a curved fold along the distal half of the tarsus. Skin granulate, finely on the upper parts and belly, more coarsely on the sides. Reddish brown above, marbled with dark brown on the head and back and with blackish cross-bars on the limbs ; a pale dorso-lateral streak ; a black streak round the snout, continued, as a broad band, on the temple and along the side South-American Batrachians. 429 of the body ; tympanum reddish brown; lower parts white with numerous small black spots and vermiculations. From snout to vent 22 mm. A single specimen from Villavicencio, Quatiquia River, Colombia, altitude 400 feet. Presented by the Wellcome Bureau of Scientific Research. Alylodes roseus. Tongue oval, slightly nicked behind. Vomerine teeth in short transverse series considerably behind the choane. Head as long as broad ; snout rounded, not projecting beyond the mouth ; canthus rostralis indistinct ; loreal region very oblique, concave ; nostril twice as far from the eye as from the tip of the snout ; interorbital space as broad as the upper eyelid; tympanum hidden. Fingers moderate, first a little shorter than second ; disks large, a little broader than long ; subarticular tubercles moderate. Tibio-tarsal articulation reaching the eye; tibia half the length of head and body. Toes moderate, perfectly free ; disks as large as those of the ‘fingers ; subarticular tubercles small, feebly prominent; a single metatarsal tubercle, rather large and prominent. Skin smooth above, granular on the belly; three subconical tubercles on the upper eyelid. Grey above, with dark brown variegations ; loreal region dark brown; a white streak on the canthus rostralis and on the edge of the upper eyelid, and a broader, dark-edged one from the eye to halfway down the side of the body ; dark oblique bars on the sides of the head and body and on the limbs ; upper eyelids and sides of body with deep pink spots ; groin, sides of thigh, lower surface of arm, forearm, and tibia, and upper surface of tarsus and metatarsus deep pink ; throat, belly, and lower surface of thighs grey, marbled with brown. From snout to vent 27 mm. A single specimen from Andagoya, Choco, Colombia. Presented by Dr. H. G. F. Spurrell in 1916. Fylodes trachyblepharis. Tongue oval, entire or slightly nicked behind. Vomerine teeth in small groups just behind the choane. Head as long as broad; snout rounded, not projecting beyond the mouth ; canthus rostralis distinct; loreal region oblique, concave ; nostril nearer the tip of the snout than the eye; interorbital space as broad as the upper eyelid ; tympanum distinct, half 430 Mr. G. A. Boulenger on new the diameter of the eye. Fingers moderate, first a little shorter than second ; disks rather large, round, smaller than the tympanum ; subarticular tubercles rather small, feebly prominent. Tibio-tarsal articulation reaching the nostril or the tip of the snout; tibia 12 times in length of head and body. ‘Toes moderate, perfectly free ; disks a little smaller than those of the fingers ; subarticular tubercles small, feebly prominent ; two metatarsal tubercles, inner oval, rather large and prominent, outer round and small. Upper parts with small glands, belly granular; upper eyelids with several subconical tubercles. Brown above, back and sides of head yellowish ; a >—<-shaped black marking behind the back of the head, the antero-lateral branches of which extend to the eyes ; a dark canthal streak, and two dark bars from the eye to the edge of the mouth; an oblique dark temporal streak ; limbs with dark cross-bars; sides of thighs deep pink ; lower parts, throat, and breast finely speckled with brown. From snout to vent 20 mm. Three specimens from El.Topo, Rio Pastaza, EK. Ecuador, 4200 ft.; from Mr. M. G. Palmer’s collection, 1912. L-ptodactylus hololius. Tongue oval, slightly nicked behind. Vomerine tecth in long, slightly oblique series behind the choane, not extending outwards beyond the vertical of the inner borders of the latter. Head as long as broad; snout rounded, scarcely projecting beyond the mouth; canthus rostralis indistinct; loreal region oblique, slightly concave; _ nostril equidistant from the eye and from the tip of the snout; interorbital space as broad as the upper eyelid ; tympanum very distinct, half the diameter of theeye. Iingers moderate, obtuse, first a little shorter than second; _ subarticular tubercles rather large and very prominent. Tibio-tarsal articulation reaching the eye; tibia a little less than half the length from snout to vent. Toes slender, obtuse, perfectly free, not margined ; subarticular tubercles moderately large, very prominent ; two small metatarsal tubercles, inner oval, outer round ; no tarsal fold. Skin perfectly smooth ; no dorsc-lateral feld. Pale brown above, with dark brown spots; a dark cross-bar between the eyes, followed by a rhombic spot; a A-shaped dark marking between the shoulders ; limbs with rather indistinct dark cross-bands ; lower parts white. South-American Batrachians. 431 From snout to vent 26 mm. A single specimen from Pebas, R. Maraiion, Peru ; from the collection of Mr. J. J. Mounsey, 1913. Leptodactylus diptychus. Tongue oval, rather strongly nicked behind. Vomerine teeth in long transverse series behind the choanz, not ex- tending outwards beyond the vertical of the inner borders of the latter. Headas long as broad; snout rounded, projecting considerably beyond the mouth; canthus rostralis indistinct ; loreal region oblique, slightly concave; nostril a little nearer the end of the snout than the eye ; interorbital space a little narrower than the upper eyelid; tympanum very distinct, two-thirds the diameter of theeye. Fingers moderate, obtuse, first much longer than second ; subarticular tubercles large and very prominent. ‘Tibio-tarsal articulation reaching between the eye and the nostril; tibia half the length from snout to vent. Toes slender, obtuse, perfectly free, not margined; subarticular tubercles rather large, very pro- minent ; two metatarsal tubercles, inner oval and about half as long as the inner toe, outer round and very small; a tarsal fold. Skin smooth above, with small warts on the sides of the body; a glandular fold above and behind the tympanum and another, narrow but prominent, from behind the upper eyelid to the hip; throat and belly smooth, with a groove defining a ventral disk; lower surface of thighs granulate. (greyish brown above, the dorso-lateral folds lighter; tym- panum reddish brown; a dark brown canthal streak ; temporal fold edged with blackish; lips with dark brown spots; a brown bar between the eyes and a A-shaped marking between the shoulders; limbs with narrow dark brown cross- bars ; a white streak, edged on both sides with dark brown, along the back of the thighs ; lower parts white. From snout to vent 44 mm. A single specimen from the Andes of Venezuela. Leptodactylus laticeps. Tongue roundish, entire. Vomerine teeth in very long, slightly curved transverse series behind the choane, extending outwards to below the centre of the latter, Head much broader than long, much depressed ; snout broadly rounded, scarcely projecting beyond the mouth; canthus rostralis indistinct ; loreal region very oblique, slightly concave ; 432 On new Southe American Batrachians. nostril nearer the end of the snout than the eye ; tympanum very distinct, nearly as large as the eye. Fingers rather short, very obtuse, first much longer than second; subarticular tubercles large and very prominent. Tibio-tarsal articulation reaching the posterior border of the eye ; tibia 2} times in length trom snout to vent. ‘Toes rather short, obtuse, perfectly free, not margined ; subarticu'ar tubercles small, prominent ; two metatarsal tubercles, inner elliptic and two- thirds the length of the inner toe, outer round; no tarsal fold. Skin smooth; no folds on the back. Pale brown above, with large roundish black spots on the back and sides and on the upper surface of the head ; five very regular vertical black bars on each side of the head, traversing the mouth, separated by narrower whitish bars; tympanum blackish, whitish in the centre; limbs with black cross-bars 5 whitish beneath, spotted with black. From snout to vent 85 mm. A single specimen from Santa Fé, Argentina, received from Mr. Falkland Ricketts in 1898. Hyla leptoscelis. Tongue circular, entire and slightly free behind. Vomerine teeth on a level with the posterior borders of the very large choane, in slightly curved oblique series forming a chevron pointing forwards. Head as long as broad, very strongly depressed ; snout rounded, not projecting, as long as the eye ; canthus rostralis obtuse ; loreal region very oblique, feebly concave; nostril near the tip of the snout; interorbital space a little broader than the upper eyelid; tympanum distinct, half the diameter of the eye. Fiugers moderate, with moderately large disks, outer with a slight rudiment of web; no projecting rudiment of pollex. Hind limb extremely slender ; tibio-tarsal articulation reaching a little beyond the tip of the snout; tibia eight times as long as broad, 3 the length of head and body. Toes 3 webbed ; a feeble tarsal fold. Skin smooth, granular on the belly and under the thighs ; heel with a pointed dermal appendage, which is half as long as the eye. Yellowish above, with purplish-brown markings ; a large spot on the snout, two V-shaped bands between the eyes, two cross-bars on the back, a V-shaped band on the sacral region, and angular cross-bars on the limbs. From snout to vent 26 mm. A single specimen from Lago do Iachy, above Sao Paolo On some Sawflies from the Australian Region. 433 de Clinenca, R. Solimoens, Brazil; from the collection of Mr, J. J. Mounsey, 1913. Hylella ocellata. Tongue cireular, entire, and slightly free behind. Head broader than long, very strongly depressed ; snout rounded, not projecting, as long as the eye, which is obliquely turned forward ; no canthus rostralis, loreal region feebly concave ; nostril near the tip of the snout ; interorbital space broader than the upper eyelid ; tympanum distinct, 2 the diameter of the eye. Fingers rather long, with moderately large disks, outer one-fourth webbed. Hind limb very slender; tibio- tarsal articulation reaching beyond the tip of the snout; tibia seven times as long as broad, 2 the length of head and body. Toes 2 webbed. Skin smooth, belly granular. Violet-blue above (in spirit), with round white spots, which are small and crowded on the sides of the head and on the limbs and large and scattered, and surrounded by a blackish ring, on the back ; the blue colour forms a very narrow band on the thigh ; upper lip with a white edge; sides and lower parts white. From snout to vent 29 mm. A single specimen from Huancabamba, EH. Peru, above 3000 feet (coll. E. Boettger, 1912). XLII.— Notes on and Descriptions of some Sawflies from the Australian Region. By S. A. Rouwer, Forest Insects, U.S. Bureau of Entomology, Washington, D.C. Tus short paper, which is a contribution from the Branch of Forest Insects, United States Bureau of Entomology, contains the descriptions of four new species of sawflies. One of these species is especially interesting, because it represents a new genus which is the basis of a new subfamily. The material upon which this paper is based was submitted for study by the British Museum (Natural History), and all the types will be returned to that institution. Xiphydria obtusiventris, sp. n. In Konow’s table of Xiphydria this runs to fumicornis, . Konow, but it differs from the description of that species in 434 Mr. S. A. Rohwer on some a number of ways and does not seem to be closely allied. The unusual short ovipositor and ninth tergite cause the abdomen to be rounded, not tapering, apically, and gives this new species a distinctive appearance. Female-—Length to end of abdomen 8 mm.; anterior margin of clypeus rounded, medianly depressed, but with a median protuberance, which at first sight gives the impression that there is a small median tooth ; malar space about half as long as the width of mandibles at the base ; surface of clypeus with dorsad-ventrad striz; face and front reticulate ; middle fovea small, indistinct ; ocelli in a low triangle, the postocellar line longer than the ocellar line; vertex and posterior orbits finely aciculate ; antennz distinctly tapering apically, 18-jointed, the third joint distinctly longer than fourth but not as long as 4 plus 5; pedicellum not half as long as third joint; scape subequal in length with third joint ; proscutum broad, well defined by foveolate notauli, but the median longitudinal furrow is feeble; surface of scutum and prescutum reticulate, with a more sparsely sculp- tured area at the anterior middle of prescutum and lateral middle of scutum; scutellum finely granular anteriorly, smooth and shining posteriorly; sides of pronotum granular, but with many longitudinal raised lines in addition ; anterior part of mesepisternum reticulate, the posterior portion smooth, polished ; abdomen finely granular, but the depressed apical margins of the tergites are almost without sculpture ; ninth tergite short, rounded apically, giving the end of the abdomen somewhat the same appearance as in Oryssus ; . ovipositor broad ; straight above, obtusely pointed apically and tapering from a broad base, not extending much beyond the apical margin of tergites; legs normal ; venation usual, the intra- radius joins the radius about one-fourth the length of the intraradius from the end of the second cubital. Black ; antennz and legs ferruginous ; wings hyaline, with a faint yellowish tinge; venation pale brown, stigma dark brown ; mandibles and sheath piceous. Type-locality. Kuranda, N. Queensland, Australia. Described from a single female collected May 3-June 2, 1913, by R. E. Turner at an altitude of 1100 ft. Type. British Museum (Natural History). ZENARGINE, Subfam. nov. Based on the genus Zenarge described below, and belongs to the family Argide, where it may be readily separated from either of the subfamilies by the following key :— Sawflies from the Australian Region. 435 Subfamilies of Argide. Anal vein complete and separate for its entire length ; first and second anal cells separated by an oblique interanal vein ; anella and recurrentella wanting. Zenargine. Anal vein either partly or entirely wanting ; first anal cell wanting or small and separated from the second by the submedian vein; anella and re- “currentella present, « <5 (+ wgbivass 4 ois sd aig ink Yas le A, TIntercosta Present ve t,2 ts, q7se\aais ade weiss aid Argine. Tntercoste Wanting Yoss cn ao ce tp wee So ee alee Sterictiphorine. The Argids, largely because of their three-jointed an- tenne, have long been considered as a distinct group, but most classifications have failed to show any relationship between them and such groups as the Perreyiid#, Loboceride, or Pterygophoridee. A study of these four families shows, however, that they have much in common, and it is not unlikely that they had a common origin and are phylo- genetically closely allied. The subfamily Zenargine adds some evidence to this assumption, because it has certain characters which suggest an affinity with the Perreyiidse and certain others which suggest Loboceride. The venation in the Zenargine is different from all other sawflies. The anterior wing probably represents a generalized Argid, because, with the exception of the complete anal vein, it presents nothing remarkable. The apex of the radial cell and the form of the radial and cubital cells, especially at the base, however, suggest Loboceras. The hind wing is much more specialized than the hind wing of the Argids, because of the loss of anella and recurrentella, and is not unlike Perreyia. The shape and foveolation of the head is not typical of the Argids, but recalls more the head of some of the Perreyiidee. y In MacGillivray’s classification the genus Zenarge runs to the subfamily Lophyrinz, but it has but little in common with this group, and does not even resemble it closely in venation. ZENARGE, gen. nov. Genotype. Zenarge turneri, Rohwer. Clypeus long, the dorsad-ventrad length nearly half as great as the apical width, the anterior margin rounded laterally and emarginate medianly, the dorsal margin com- posed of three sections, the lateral sections half the length of the median section, the entire dorsal margin sharply defined ; 436 Mr. S. A. Rohwer on some. labrum short, nearly truncate apically; malar space about one-third as long as the width of mandibles at base ; inner margin of eyes slightly converging towards the clypeus, the area between them wider than high and the distance between them at the clypeus greater than the length of the eye ; ocelli in a low triangle, the posterior ones distinctly in front of the supraorbital line ; width of posterior orbits about two-thirds the cephal-caudad length of eye ; antennze 3-jointed, the third thickened apically in female, but nearly of a uniform thick- ness in male; pronotum well developed laterally ; prescutum well defined and with a faint median longitudinal depression ; anterior margin of the scutellum subangulate, the posterior marin rounded, the surface convex ; first parapteron present, but in specimens in which the pronotum fits close it is con- cealzd by a lobe-like projection of the pronotum ; sternauli present but not sharply defined; mesepimeron large, with a cephal-caudad suture at about the middle; second pleural suture straight; third pleural suture straight ; the metepi- sternum and metepimeron of equal height ; propodeal spiracle large, elongate-oval, and placed near the base on the dorsal surface; metascutellum distinct ; metapostnotum much re- duced, hardly visible ; propodeum completely chitinized and without a median suture ; abdomen cylindrical ; ninth tergite not especially large laterally ; cerci distinct ; sheath with the lower margin much thickened, the ventral surface sculp- tured and with some long hair; basitarsi distinctly shorter than the following joints ; claws simple; intermediate tibie armed with a pair of spines at the apical third; posterior tibiee armed with a single spine at the apical third ; costal cell rather narrow; intercostal vein present; radial cell without a cross-vein or a distinct appendage, pointed at apex ; three closed cubital cells; the second and third each receiving a recurrent near the base; basal vein joining the subcosta a short distance before the origin of the cubitus, longer than the first recurrent, therefore not parallel with it ; first dis- coidal cell similar in outline to that of Caloptilia ; nervulus received at about its length from the basal vein ; anal vein complete, the first and second anal cells very much the same as in Pseudosiobla ; radiellan cell without an appendage ; one closed cubitellan cell; recurrentella wanting ; anella wanting. Zenarge turnert, sp. n. Female—Length 10 mm. Anterior margin of the clypeus arcuately emarginate medianly ; supraclypeal area convex, Sawflies from the Australian Region. 437 triangular in outline; median fovea rather large, deep, with sloping walls, nearly cireular in outline ; antennal furrows very poorly defined but present ; ocellar basin shallow, rather large, triangular in outline but only poorly limited below ; postocellar line distinctly shorter than the ocellocular line, subequal with the ocelloccipital line; postocellar furrow present; postocellar area poorly limited laterally, much wider than long ; head shining, front with rather spare punctures; thorax shining, with small scattered punctures ; stigma three times as long as wide, of nearly uniform width for basal two-thirds, then gradually tapering to metacarpus; third cubital cell narrowed above, the third intercubitus subequal in length with the third abcissa of the radius; abdomen shining ; sheath seen from the side with the apex rounded. Black ; clypeus, labrum, mandibles (except tips), face, inner orbits narrowly above antenne, posterior orbits, margin (anterior, posterior, and lateral) of pronotum, tegule, apical two-thirds of scutellum, metascutellum, a broad band of mesoepisternum, and metepisternum yellowish white ; abdo- men ferruginous, propodeum and apical two tergites black ; legs black, four anterior coxee, trochanters, apices of femora, entire tibia, and tarsi yellowish white ; hind coxe except a large spot on upper lateral surface, trochanters, basal fourth of hind tibiz, and four apical joints of hind tarsi yellowish white ; wings sublyaline, venation including stigma dark brown. Male.—Length 9mm. Agrees very well with the cha- racters given for the female; differs in colour from the female in having the mesosternum ferruginous, in having all of the black of the legs (except hind tibie and basitarsus) replaced by ferruginous ; apex of abdomen black ; tergites with distinct punctures which become so close on the basal segments that the surface is coriaceous; hypopygidium very deeply arcuately emarginate apically, Type-locality. Killara, Sydney, N. S. Wales, Australia. Described from two females (one type) and one male collected at an altitude of 400 feet on August 17, 1913, by R. E. Turner, after whom the species is named. Type and allotype. British Museum (Natural History). Paratype. U.S. Nat. Mus. Genus ANCYLONEURA, Cameron, The genus Ancyloneura, Cameron, belongs to the tribe Euriini, and falls close to Neoeurys, Rohwer, but may be 438 Mr. S. A. Rohwer on some separated from the last-mentioned genus by the obsolete antennal furrows and by having the hind basitarsus shorter than the following joints. The species which belong here have not been fully described, and seem to be closely related. The following key, which is based on literature, may aid in distinguishing the forms described :— Key to the Species. Hind femora black; antennz 15-jointed (Kirby’s CTH G ON Ep atanire Sates dearer Tas oe eae nigripes (Smith). Hind femora reddish ; antennz with less than 15 POUR Seite. ie cals woo aie atcl> ars Ae ave la A Gls wi ie euebare fs is 1. Markings of the fore legs ‘sordid white” ; antenne 13-jointed. (Aru.) ........000s varipes, Cameron, Markings of the fore legs ferruginous; an- tenn 12-jointed. (New Guinea.) ...... wollastont, Rohwer. Ancyloneura wollastont, sp. 0. In the absence of the first intercubitus this species differs from the recognized generic characters, but in all other ways it agrees with my notes and with the description. Female.—Length 455 mm. Shining, without apparent sculpture; median fovea rather deep, elongate, linear; post- ocellar line slightly shorter than the ocelloccipital line; post- ocellar area not defined anteriorly and defined laterally by rather broad depressions ; antennz 12-jointed, the third joint slightly longer than the fourth and fifth ; from the third joint the joints gradually decrease in length until the eleventh, which is subequal in length with the twelfth ; eleventh joint a little more than twice as wide as long; stigma about three times as long as greatest width, angulate near base and tapering to a narrow apex ; first intercubitus wanting; third cubital cell as long on the radius as the combined first and second; second recurrent about two- thirds the length of the second intercubitus from the base of the third cubital cell; sheath concealed ; lower apical margin of lancets with regular rounded teeth. Black ; apical part of femora (more extensively on posterior pair), anterior tibiz, base of anterior tarsi, basal two-thirds of intermediate tibie, and basal half of hind tibiee ferruginous; wings brown apically, hyaline basally ; venation dark brown. Type-locality. Iwaka River, New Guinea. Described from one female, collected February 1911 by A. F. R. Wollaston. Type. British Museum (Natural History). Sawfltes from the Australian Region. 439 Genus PoLycionus, Kirby. In ‘Genera Insectorum,’ fase. xxix. 1905, p. 40, Konow places the genus Polyclonus, Kirby, asa synonym ot Ancylo- neura, Cameron. This seems to the author to be wrong, and as very little is known concerning the genus the tollowing notcs, taken from specimens in the British Museum, and made in 1909, may be of value :— ‘A female of Polyclonus atratus, Kirby (genotype), from Melborne, Victoria, ‘C. F. 8. 00, No. 1164, proves the genus is a good one. It may be briefly described thus: Length 5 mm. ; expanse 12°5 mm. Clypeus truncate ; malar space very narrow, practically wanting ; antennal furrows indis- tinct but complete; a distinct furrow from the anterfor ocellus to between bases of antenne ; head strongly granular ; antenne wanting beyond 12th joint, each joint beyond the second with a ramus like Pterygophorus ; scutum and scu- teilum shining, sparsely punctured; tarsal claws simple ; venation like Perreyia (fig. 80, plate 39, Proc. U.S. Nat. Mus. vol. 29,1906), except that the third cubital receives the second recurrent and the third cubital cell is longer than the second. Black; labrum, mandibles, tibia, and tarsi pallid ; wings hyaline, iridescent; venation black.” From these characters and others gained from an incom- plete generic synopsis the author is of the opinion that the genus belongs to the tribe Kuriini, where it is easily distin- guished by the ramose antenne of both sexes. Neoeurys tasmanica, sp. n. This new species is closely allied to metallica, but may be separated by narrower sheath, darker stigma, and shorter distance between the second recurrent and second intercubitus. Female—Lenugth 5 mm. Antennal furrows complete to occiput ; middle fovea shallow, wedge-shaped ; postocellar furrow wanting; postocellar line subequal with the ocell- - ocular line; antennze 13-jointed, the third joint but slightly longer than the fourth; scape but slightly longer than the pedicellum ; sculpture of the head fine and close; stigma slightly angled ‘at base, then regularly tapering to apex; second recurrent received by the third cubital cell half the length of the second intercubitus from the base of the cell; prescutum and scutum medianly finely granular and some- what opaque; sides of the scutam and scutellum shining ; mesepisternum with small rather close punctures ; sheath 440 Mr. R.I. Pocock on some slightly concave above, rounded apically, and tapering to the rather narrow base. Blue-black, with a faint bronzy tinge to head ; palpi, apices of anterior femora, and all of the tibize rufo-ferruginous ; wings dusky hyaline, venation (including stigma) dark brown. Male—Length 3 mm. The male assigned here agrees closely ; the middle fovea is somewhat deeper and the apices of all the femora are pale; the lower margin of the stigma is pale, and the second recurrent joins the third cubital cell somewhat further from the base. Hypopygidium narrow and truncate apically, Type-locality. ‘Tasmania. Described from one female (type) collected on the summit of Mt. Wellington, 1904, by A. M. Lea, and one male (allo- type) from Haglehawk Neck, S.H. Tasmania, Feb. 12- Mar. 3, 1913, collected by R. Ki. Turner. Type and allotype. Collection British Museum (Natural History). XLITI.—On some External Characters of Ruminant Artio- dactyla.—Part V. The Tragelaphine. By R. I. Pococg, F.R.S. Subfamily TraerraPHine. The only fresh material available in 1910 for examination of the cutaneous glands of this group belonged to the genera Tetraceros, Boselaphus, and Tragelaphus. For the rest dependence had to be placed upon the inspection of dried skins and living examples, which yielded unsatisfactory results, Since that year additional material of those genera, as well as fresh examples of Strepsiceros, Limnotragus, and Taurotragus, have come into my hands, and these have enabled me to clear up some doubtful points. Genus Tetraceros, Leach. Tetraceros quadricornis, Blainv. (p. 921). I have nothing to add to my description of the glands of this species published in 1910, except to say that an adult female had the glands of the false hoofs of the hind legs as External Characters of Ruminant Artiodactyla. 441 well developed as in the male. Their secretion had a decidedly pungent and unpleasantly musteline odour. The rhinarium is well developed and ‘ bovine.” From the anterior aspect the upper margin is strongly convex and the area beneath the nostrils is mesially grooved and very wide—wider, in fact, than the area above those orifices—and visible to a considerable extent in profile view. From the dorsal sid@ the anterjor margin is convexly truncated, and the posterior margin is straight between the posterior angles of the nostrils, the hair of the nose not extending forwards beyond that line. Genus Bosetaruus, Blainv. Boselaphus tragocamelus, Pall. (p. 926). In a male example the preorbital gland had a much shallower pit than in the female described in 1910, and was without definite lids. The gland itself, moreover, was not regularly heart-shaped, but was longer than thick and of irregular form. The rhinarium (fig. 1, A, B,C) is large and “bovine,” closely resembling that of Tetraceros, but more protuberant in front, and, beneath the nostrils, laterally and with a wider internarial septum. On its dorsal side the hair advances a little way between the nostrils, so that the poste- rior border of the rhinarium is concave. In 1910 I briefly described the glandular nature of the skin between the ialse hoofs of the hind feet in the female. The same feature is present in the male where the skin between the widely separated false loois is clothed with longish hair, is very thick and glandular, and mesially folded, In the fore foot there is no trace of the gland, the false hoofs being larger and the hair restricted to the uarrow area between them. ‘this gland (fig. 3, B) on tle hind foot of Buselaphus clearly represents an earlier stage of the evolutiou of the pair of pouch-hke glands present in Tetra- ceros. The presence of similar glands in Taurotragus and Strepsiceros (cf. infra) serves to link Boselaphus with the African Tragelaphines, and refutes, if refutation be needed, Riitimeyer’s claim that Bosedaphus belougs to a differcit group. Inguinal glands are absent and there are two pairs of MANIne. The penis (fig. 1, D, E) agrees, generally speaking, with the sketch and description published by Gerhardt (op. cit. Aun. & Mag. NS Hip Ser, os Kel. ie on 442 Mr. R. I. Pocock on some viive al ~ AN 1 X <7 Rhinarium of Boselaphus tragocamelus from the front. x 2. ‘The same from above. The same from below. The extremity of the penis of B. tragocamelus from below. The same from the lett side. External Characters of Ruminant Artiodactyla. 443 p. 153). It ends in an elongated subovate portion defined by a shallow constriction. The urethral canal, however, reaches the extremity of this, lying rather upon its right than on its left side. Genus Tracextapuus, Blainv. Tragelaphus scriptus, Pall., and its subspecies (p. 929). The only specialized cutaneous glands which occur in this species and its numerous affiliated forms, of which sylvaticus is the commonest in our Zoological Gardens, are the inguinals, which, according to my examination of a large number of specimens, are invariably present as a pair of small pouches lying far out in front of the four teats, the orifice being in the fold between the thigh and the abdomen. The only other genera of Tragelaphines which possess these are Limnotragus aud Strepsiceros. Asinall the African Trage- laphines preorbital and interdigital pedal glands are absent. The glands between the false hoofs of the hind legs, found in Tetraceros, Boselaphus, Strepsiceros, and Taurotragus are also absent. The rhinarium is variable with respect to the width of the area between the edge of the lower lip and the nostrils. Sometimes there is a definite narrow philtrum as in S¢repsi- ceros and adult examples of Taurotragus, but not infrequently the hair of the upper lip does not encroach so far towards the middle line, leaving a broader irregularly shaped naked space. This variation may be a matter of age, or it may prove to have aracial significance. Otherwise the rhinarium seems to resemble that of Tauwrotragus and _ Strepsiceros, except that the posterior edge between the angles of the nostrils is straight from side to side. The penis, as described and figured by Lonnberg (Ark. Zool. Stockholm, (5) v. no. 10, p. 7, fig. 6, 1909), is distally attenuated, wish a terminal sigmoid “flexure, the urethral canal not being prolonged beyond the tip of the glans penis. Genus Limnorracus, Scl. & Poc.* Limnotragus spekei, Scl. (p. 930). Examples of the two races gratus and selousi resemble * Although this genus is of very doubtful value, it may be explained that, at the request of Mr. Thomas, who in 1900 was compelled by ill- health to abandon temporarily all zoological work, I took his place in the completion of vol. iv. of the ‘ Book of Antelopes,’ Strictly speaking, therefore, although the matter is of no great moment, this peneng name should be ascribed to Sclater and myself. AAL Mr. R. I. Poeock on some Tragelaphus with respect to the cutaneous glands, the inguinais being present and similarly placed and the glands between the false hoofs absent. The rhinarzum also is like that of Tragelaphus, except that the area between the nostrils =o the edge of the lower lip is usually at all events wider. It is as wide as the inter- narial septum in a specimen ‘of selousi and wider in au example of ¢ gratus. I have never seen it narrower, as 1s sometimes the case in Tragelaphus. Tenney appears merely to differ from Trayelaphus in the length ot the loots and the nakedness of the posterior surface of ithe pastern and sae But,as Meinertzhagen has pointed out (P. 4. 5. T9TG. 1, p. 377), there is sometimes a patch of hair in the middle of the pastern between the false hoofs and the hoots themselves. But in two examples which came together from the Congo to the Zoological Gardens the feet of the male were Hated behind, while those of the female had the patch in question. Genus Srrupstceros, H. Smith. Strepsiceros strepsiceros, Pall. (p. 931). The fresh carcase of a hornless male, three or four months old, from South Africa, is all the material of this have seen. The rhinarium has a narrow grooved philtrum and the hair upoa the upperside of the nose spreads forwards some distance between the nostrils. Otherwise the resembles that of Tragelajhus. ‘There is no trace of preorbilal gland. fnguinal glands also are absent. Possibly their absence in this specimen was due to immaturity, smee both Owen and Ogilby agree as to their presence in the species. When present they probably resemble in size and position those of S. inberbis, of Trayelaphus, and Limnotragus. Pedal glands of the interdigital type are absent, but upon the hind feet there are glands associated with the widely separated false hoofs as in Taurotragus. On the inner side of each false hoof there is a fringe of long black hair growing from a glandular thickening of the skin, the secretion of which is discharged amongst the roots of the hairs and intoa hairless cleft between the thickening and the false hoof. ‘The skin of the middle of the area between the false hoofs is clothed with short hair aud is thin and not specially glan- species L rhinarium Evternal Characters of Ruminant Artiodactyla, 445 dular. On the fore feet no such fringes exist, the false hoofs being small, close together, and overlapping *. Strepsiceros imberbis, Blyth. Of this species I have seen one fresh specimen, an imma- ture castrated male from Somaliland, and the feet and inguinal area of an adult female from British Kast Africa, kindly brought home for me by Mr. F. C. Selous. These specimens resemble in nearly every particular the example of S. strepsiceros, above described. The upperside of tne rhinarium, however, was not overgrown with hair to quite the same extent, and there was a single pair of inguinal glands, each consisting of a narrow sack 2 inches deep, with a small circular orifice, and lying far out in advance of the two pairs of mammie, as in Trayelaphus aud Limnoirayus. The glands close to the false hoofs (fig. 3, D) of the hind feet were exactly as described in S. strepsiceros, aud on the fore feet the false hoofs were smaller than on the hind feet and separated by a narrow strip of naked skin, horny ia one of the specimens. The penis of the castrated male was very small and simple, with a bluntly rounded termination. ‘The urethral canal was not produced beyond the end of the glans. Strepsiceros has iitherto been distinguished from Trage- laphus merely by small differences in the horns of very little systematic value. Particularly satisfactory, therefore, is the discovery of the difference between the two genera supplied by the glands adjoining the posterior false hoofs. Genus Tavrorracus, Wagn, Taurotragus oryx, Pall. (p. 932). To the description of the cutaneous glands of this species published in 1910 1 have to make one important addition. ‘Uhis is the presence of glands close to the false hoofs of the lind legs, precisely resembling those described above under Stirepsiceros. ‘These are as well developed in a calf’ one day old as in the adult, and they are the only specialized cuta- beous glands present in the genus, so far as my observations go (tig. 3, A, C). 1 have never succeeded in finding a trace of the preorbital gland described by Mr. W. L- Sclater, und am compelled to disbelieve in its existence. ‘The rhinarium (fig. 2, A, B, C) in the adult is not “ bovine,” .* Nyala angasi resembles Strepsiceros and differs from Trayelaphus in possessing the glandular fringes by the false hoofs of the hind legs, 446 Mr. R. I. Pocock on some like that of Boselaphus, It is much less protuberant both in front of and beneath the nostrils laterally, and the septum between the expanded nostrils is narrower. Beneath the nostrils in front the rhinarium spreads somewhat to right ge Oe “Ny c < A. Rhinarium of Tawrotragus oryx from the front. 3. The same from above. C. The same from the side. D. Extremity of penis of 7. oryx from the left side. HK. The same from below, and left, being nearly as broad here as just above the nostrils ; but beneath this it rapidly narrows to form a mesially grooved philtrum which is about as wide as half the internarial septum. The upper edge from the front view is External Characters of Ruminant Artiodactyla, 447 lightly convex; the posterior edge from above is lightly concave, the hairs of the upper side of the nose spreading Fig. 5, = \ 2IZLE AZIZ > eld im Hes at TAN Sr 2 1,1, VAY onl wy Ws Ny J val \\ 1 Te ue WL A Vigh 7! 1) \ et le i Bp is Nj ; hn) ne \ a ti i} iH iy edd A z : A. Transverse section through the false hoofs and glands of the hind foot of Taurotragus oryx. B. The same of Boselaphus tragocamelus. C. Lower view of hind foot of Tawrotragus oryx, showing the glandular fringes encircling the false hoofs on the inner side. D. The same of Strepsiceros imberbis. forwards a little in advance of the posterior notch of the nostrils. The width of the philtrum appears to vary sometimes 448 Laternal Characters of Ruminant Artiodactyla. with age in an interesting manner. Thus in a calf one day old it is wider than in the adult, being about three-fourths the width of the internarial septum, whereas in a foetus about three months developed the naked area beneath the nostrils is very broad, broader even than in the adult Boselaphus, giving the rhinarium a strictly bovine appear- ance. This suggests that the bovine type of rhinarium is tle primitive type in the Ruminantia *. As I recorded in 1910, inguinal glands and interdigital pedal glands are absent, but the hind feet possess glandular thickenings of the skin surmounted by a fringe of black hairs (fig. 8, A, C) precisely as in Strepsiceros. The penis (fig. 2, D, HE) of an old male has an elongated, undulating, attenuated terminal portion, much longer than in Boselaphus, and, as in that genus and others belonging to the Tragelaphinge in which this organ has been described, the urethral canal is not produced beyond the tip of the glans. The points of interest connected with the characters above enumerated may be summarized as follows :— (1) Preorbital gland present.......... weeees Letraceros, Boselaphus. . pipe ROSEN wd Cie af aen etme teen Traygelaphus (Ltmno- trugus), Nyala, Strep- siceros, Taurotragus. (2) Inguinal glands present . ...)c0i. sa. oe ee Tragelaphus (Limno- tragus), Strepsiceros (Palwaysin thelatter). i 5g oy MAISON hytics. ote Nes Se: weaee. YLetraceros, Boselaphus, Laurotragus. (8) Glands between posterior false hoofs absent. Yvrayelaphus (Limne- iragus). present. Consisting’ "of definite pockets within false “hoofs afiht Wi fous hanne & ee eee Tetraceros. Consisting of a thickening of is skin only. Thickened skin extending across fet- lols iy ic ZA icc eae pee eee Deoselaphus. Thickened skin restricted to area close to false hoofs and surmounted by DAH G SOL Maas i a hace leek e .. Nyala, Strepsiceros, Laurotragus. * It may be added that in the foetal specimen above alluded to the faciat vibrissee were well developed, consisting of shoit scattered mystacials and submentals, a row of superciliaries and suboculars, an upper and a lower genal tuft arising from the white spots on the cheelk and interramals. It is singular that the Artiodactyla and the Carnivora are the only orders of mammals known to me which possess as a primi- tive character two genal tufts—an upper and a lower—on each cheek. ie “Published ihe Br st Dig of ever ‘y Month. ary 6d. Ss : AVS MR THE 4 LONDON, EDINBURGH, and DUBLIN a PHILOSOPHICAL MAGAZINE ine AND | JOURNAL OF SCIENCE, Conducted by - aa SIR OLIVER JOSEPH LODGE, D.Se., LL.D., F.R.S, ~ SIR JOSEPH JOHN THOMSON, 0. M., M. A., Se. DB R. S., we JOHN JOLY, M.A. D.Sc., E.RS, E.GS., Sed GEORGE CAREY FOSTER, B. A., LL. Ds, F, R. S., ane. Wet, WILLIAM FRANCIS, BULB ots Re oa : : _ Published the First Day of every Month. Dd sad THE NWALS. AND MAGAZINE OF NATURAL HISTORY ke including Zoology, Botany, and Geology. | a * Conducted by : WILLIAM CARRUTHERS, Ph.D., FR. ae E. SHIPLEY, M.A., Se. D., FB. yh ee OMAR D ch. FRANCIS, S., F.LS., nas, L. sere Syst eee F.Z. y . ty Wouuded in 1877. Monthly, 1s. Annual Subscription BES oe and 12 Monthly Numbers), 13s. TH B OBSERVATORY, =e eee REVIEW OF ASTRONOMY. Edited by -. A, S. EDDINGTON, M.A., M.Sc., F.RB.S., a eel Es M. STRATTON, M.A., and H. $8. JONES, MA. B. Sa. Its Contributors include most of the leading English- speaking Astronomers, and its Articles give an interesting reflection of the BLOgTess of astronomical thought and achievement. Published Monthly. Dorie 1s. 8d. Annual Subscription, ie 16s. post free. ; gi | THE ; oe bas ks JOURNAL OF BOTANY it, - BRITISH AND FOREIGN. . t fae: 5 = Kdited by _ Nh oe JAMES BRITTEN, K.S.G., E.LS,, a ¢ Be _. LATE SENIOR ASSISTANT, DEPARTMENT OF BOTANY, BRITISH MUSEUM, 7 “Articles by prominent Botanists ; Announcements of Discoveries Reviows Si es Botanical News and eugene &e., &e. he : “Red Lion Court, Fleet Street, London. oe c mr : P. XXXVI. Descriptions of New Pyralide of the Subfamily Py- — | raustine. By Sir Groree F, Hampson, Bart., F.ZS.,&. ..:., ... 893 | . > es oy 2 ait XXXVII. On the Synonymy of some European Diplopods _ (Myriapoda), with Special Reference to Three Leachian a By 2 Ricuarp 8. Baewann, F.L.S. 2.5. XXXVIIL. New Lepidoptera in the Joicey Collection. By. Lo B, Prove, PRGA tee ie Pees Ce ee oe oe +. 412 XXXIX. Descriptions and. Records of Bees —LKXXI. By T. D. A. Cocxurrtt, University of Colorado......... Volecwees 418 XL. On some Fishes from the Shari River, with Descriptions. of Two new Species. By G. A, Bournyenr, F.RB. ....... ee ace XLI. Descriptions of new South-American Batrachians G.- A. Bourunerr i RS2 oe eee ee as ey acy ce a oe XLII. Notes on and Descriptions of some Sawflies from Australian Region. By 8..A. Ronwer, Forest: Insects, U.S. Bu ef Entomology, Washington, D.C. ........ PS ay SR Ben i XLIII. On some External Characters of Ruminant Artiod ot —Part V. The Tragelaphine, By R.1. Pocock, F.R.S. ..:sys5 Red Pisin Court, ves Street, Teadaw e 4 gS 8, NINTH SERIES. No. 12. a THE ANNALS” | MAGAZINE OF NATURAL HISTORY. L | ZOOLOGY, BOTANY, ann GEOLOGY. CONDUCTED BY WILLIAM CARRUTHERS, Pu.D., F.R.S., F.LS., F.G.S., ARTHUR E. SHIPLEY, M.A., Sc.D., F.R.S., HZ.8:3 : AND RICHARD T. FRANCIS, F.Z.8. : - ; * BEING A CONTINUATION OF THE “‘ANNALS” COMBINED WITH é MESSRS, LOUDON AND CHARLESWORTH’S “‘ MAGAZINE OF NATURAL HISTORY.” = = LONDON: TAYLOR AND FRANCIS, RED LION COURT, FLEET STREEY. ¥ _ Sold by Simpkin, Marshall, Hamilton, Kent, & Co., Ld: , Bailliére Paris: b ; nerd 1 Hodges, Figgis, & Co., Dublin, . Raturalists, ret ae genus” Keep in stock every kind ot APPARATUS. ae CABINETS required by ENTOMOLOGISTS, ORNITH- OLOGISTS, BOTANISTS, &c. Also NESTING- BOXES, which should be fixed up in gardens or * shrubberies before the breeding Season. Set een \A Large Stock of Butterflies, Moths, Birds, Fags, ae. ; — Full Oatalogue (84 pages) mailed free to any address. 36, STRAND, LONDON, W.C., ENGLAND. | ST A TSS I SE IE SS tek Peni ese Sidecar Rises oe DS ES oes ie De A SCIENTIFIC BOOKS AND SERIALS. | JOHN WHELDON & CO. have the largest stock in the country of Books | in all Departments of Science and Natural History, also Transactions and J ournals of Learned Societies, etc., in sets, runs, and single volumes or meee Libraries or small parcels purchased for cash. SPECIAL CATALOGUES: Entomological, Onutielogitat Zoological, Geolo- sical, War Economy (Agriculture, Husbandry, Gardening, Economic BOuay ete. 1ae Botanical (General and Geographical), etc., 2d, each, post free, 38 GREAT QUEEN STREET, KINGSWAY, LONDON, W.C. 2. Telephone: Gerrard 1412. Super-Royal 8vo, pp. iv, 226. With 4 Plates and 16 Text-figures. _ Price 12s. 6d. A CONTRIBUTION PHYTOGEOGRAPHY AND FLORA DUTCH N.W. NEW GUINEA. L. S. GIBBS, F.L.S., F.R.MS. TayLor AND Francis, Red Lion Court, Fleet Street. et Rates for Advertisements in the fae and ~ Magazine of Natural Be 2 One Six = Pwelve Insertion. Insertions. ~ Oe eertions. some PAGE: #2 soap esae = 20-0 1216 O each =! 12° 0 each) 3s HALFsPAGE 3-88 952 Ge AO gO Gs all QUARTER-PAGE - 12 6 1 a a ?< 10: 0-4, Jae All applications for ener to be made to ta aN THE ANNALS AND MAGAZINE OF NATURAL HISTORY, (NINTH SERLES.] No. 12. DECEMBER 1918. XLIV.—On some Hexternal Characters of Ruminant Artio- dactyla.— Part VI. The Bovine. By ik. I. Pocock, F.R.S. Subfamily Borvivz. IT retain this subfamily as a matter of convenience only, being unacquainted with a single character of importance by which it may be disting uished from the Tragelaphine. On the other hand, close affiliation between the two is attested by a large number of common characters. Indeed, Anoa depressicornis, the most primitive form of Bovine, quite commonly shows the typically Tragelaphine white spots and patches on the face, throat, and feet, which must be regarded as strong-evidence of near affinity with the Tragelapline stock, as I pointed out in 1910. For close upon a century there has been great divergence of opinion regarding the status of the groups into which the species of the Bovine naturally fall. In 1827 Hamilton Smith split up the Linnean genus Bosinto a number of sub- -genera—Bison, Bibos, etc. By Gray, who added Poephagus to the series, these were granted generic rank. In this opinion he was followed by Riitimeyer, and more recently by Matschie. English authors, like Blanford, Flower, and Lydekker, on the contrary, retained the genus Bos in a comprehensive sense, giviug subordinate rank to the others. In 1910 I followed that course, being unable to find evidence from the characters I was then working at for defining the 9 Ann. & Mag. N. Hist. Ser. 9, Vol. ii. a3 450 ° Mr. R. J. Pocock on some alleged genera and subgenera. Since that year, however, study of certain other external features—notably the rhina- rium and penis—have supplied additional characters to those derived from the skull, horns, tail, distribution of hair, and outward form, which, I think, justify Gray’s claim that the groups are worthy of generic recognition. Probably other characters bearing out this view will come to light with the examination of further material. So far as the cutaneous glands are concerned, the genera have the following mainly negative features in common :— Preorbital glands, as in all African Tragelaphines, are absent. Inguinal glands are invariably absent, as in the Trage- laphine genera Vauwrotraqus, Boselaphus, and Tetraceros. Pedal glands of the interdigital type are also invariably absent, as in all Tragelaplines. Glands ou the false hoofs are absent, as in Tragelaphus. ‘Two pairs of mamme are present, as in all Tragelaphines. Genus Bos, Linn. Bos, Linn. Syst. Nat. ed. 10, p. 1758: type, taurus. Rhinarium (figs. 1, A, B; 3, C) large; viewed from the front its upper margin is evenly convex from side to side and the median area below the line of the widely separated expanded nostrils is wider than the internarial septum throughout its extent, the hairs of the upper lip extending inwards neither beneath the nostrils above nor along the edge of the upper lip below ; above the edge of the lip there ruus upwards a short shallow median groove, which is present in all genera, and thus disproves Lydekker’s state- ment (Cat. Ung. in Brit. Mus. i. p. 11, 1912) that the rhinarium in the Bovine is undivided. A few scattered hairs arise from the rhinarium inferiorly, and its surface is sculptured and reticulated. The anterior portion of its dorsal surface is exposed to a varying degree in accordance with the extent to which the hair of the upper side of the muzzle spreads forwards between the nostrils; but the naked upper edge of the nostrils is always of considerable width and depth, and not narrowed as in Bison and Poephagus. The extension of the hair between the nostrils above varies according to the breed, being greater, for instance, in British park cattle (B. taurus) than in Indian humped cattle (B. in- dicus) ; but intergradation between these two forms seems to be supplied by other breeds of B. taurus. External Characters of Ruminant Artiodactyla, 451 The penis of B. taurus, as figured by Garrod (Proc. Zool. Soe. 1877, p. 10, fig. 19) is well known. It ends in an ovately rounded knob or cushion, on the lower side of which the orifice of the urethra terminates without running out into a definite tubular prolongation. In B. indicus (fig. 4, B, C) the penis is of a similar type. ; Rhinarium of zebu (Bos indicus) from above. X 3. The same from the front, A. B. The only existing members of this genus, as here recorded, are the numerous domesticated breeds of cattle referred to B. taurus and B. indicus. Apart from these there are a certain number of extinct species, of which the aurochs 33* 452 Mr. R. I. Pocock on some (B. primigenius) is the best-known form. In domesticated cattle the skull is so variable in structure that it would 2) is A. Rhinarium of American bison (Bison bison) from the front. x 2 B. The same from the side. C. The same of African buffalo (Syncerus caffer equinoctialis) from the side. require the examination of a long series of specimens to formulate a generic diagnosis based upon cranial characters. 2 en External Characters of Ruminant Artiodactyla. 453 But the success of such an undertaking would be doubtful, seeing that the skulls of some domesticated breeds differ more from aurochs-like breeds than the latter differ from other genera of Bovine. To this variability is probably to Lo a? Mir. ler >>, . V7 4% YR0 Ns Si A. Rhinarium of yak (Poephagus grunniens) from the front. x 3. B, The same from the side. C. The saine of zebu (Los indicus). X d. be attributed in a great measure the prevalent admission of subgeneric rank to the groups into which the existing species of Bovine fall. The ears are no less variable in size and shape than the skull and horns, even in closely related breeds. 454 Mr. R. I. Pocock on some Genus Brsos, Hodgson. Bibos, Hodgson, Journ, Asiatic Soc. Bengal, vi. p. 499 (1837): type, gaurus, H. Smith. Gaveus, Hodgson, op. cit. xvi. p. 706 (1847): type, frontalis. Gauribos, Ur ibos, "Bubalibos, Heude, Mém. Hist. Nat. Chin. v. pt. i. p. 8 (1901): types (now selected) respectively laosiensis, platycerus, annamiticus, Heude. The rhinarium of the two forms I have examined—namely, frontalis, which is almost certainly a domesticated breed of B. gaurus, and banteng—does not differ in any important Vig. A; ag * ae: ee aaa ———~__ iB) m ——— F . End of penis of African buffalo (Syncerus caffer equinoctialis?) from the left side. The same of zebu (Bos indicus). The same from below. . The same of banteng (Bibos banteng) from left side. . The same of gayal (Bibos, frontalis) from below. . The same of American bison (Bison bison) from the side. . The same from below. Qamyban Pp respects from that of Bos, although the dorsal surface seems to be less overgrown with hair than even in B. indicus. The hair encroaches only to a slight extent between the posterior angles of the nostrils, so that the posterior border of the upper side is lightly concave. This feature may, however, prove to be variable. In the feet the interungual integument is naked as in Bos, not hairy as in Bison. The penis (fig. 4, D, ) in both the above-mentioned species differs from that of Bos in that the urethral canal is External Characters of Ruminant Artiodactyla, 455 produced into a short tube free from the terminal cushion- like thickening of the glans, as in Poephagus (ef. infra). Genus Bison, H. Smith. Bison, H. Smith, Griffiths, An. King. v. p. 373 (1827): type, bison, Linn, Bonasus, Wagner, Schreb. Saiug., Suppl. iv. p. 515 (1844): type, bonasus, Linn. The rhinarium (fig. 2, A, B) differs from that of Bos and Bibos in being more overgrown with hair both above and in front. In front the hair of the upper lip spreads towards the middle line along the lower margin of the nostrils and even penetrates the inner portion of those orifices. Hence at this level the rhinarium is not wider than the internarial septum. Inferiorly, however, it expands, and is broad where it passes into the edge of the upper lip. Dorsally the hair of the nose spreads over the upper surface of the rhinarium almost to its anterior margin, leaving a comparatively narrow naked rim bordering the nostrils above, so that from the anterior aspect the upper edge of the rhinarium does not present the evenly convex upper margin seen in Bos and Bibos. ’ The feet also differ from those of the two last-mentioned genera in having the interungual web overgrown with hair, which is sometimes stuck together with secretion. This hairy clothing has been observed in two pure-bred specimens, male and female, which died at different seasons of the year. Hence it may be inferred that the growth of hair on this part of the foot is not a seasonal character, as it appears to be in some of the Caprine Ruminants—e. g., Ammotragus lervia and Ovis musimon*. The penis (fig. 4, F, G), like that of Bos, has no free prolongation of the urethral canal. Although I have cited Bonasus as a synonym of Bison, it must be explained that that course is justified mainly by inference, since I have had no opportunity of examining fresh material of the European species, B. bonasus, which is * Some of the American bisons that have been imported into England as pure-bred stock appear from the higher carriage of the head, higher quarters, longer horns, and other points to have taurus-blood in their veins. They are hybrids known as cattaloes in the United States. One of these had the interungual integument of the hind feet naked as in Bos taurus, whereas the ‘interungual skin of the fore feet was covered with a growth of short hairs, being intermediate in this respect between the naked condition seen in B, taurus and the long-haired condition seen in Bison bison. 456 Mr. R. I, Pocock on some a very distinct species from its American ally B. bison, and so far as external appearance is concerned, especially as regards the higher, flatter hind-quarters, serves to connect the type of Bison with Bos. Not! ving 18 known of its feet or penis. Nevertheless, judging from living examples, the vhinarium seems to be shaped like that of Bison bison. Genus Porruacus, Gray. Poephagus, Gray, List Mamm. Brit. Mus. p. 153 (1843); id. Cat. Ung. Brit. Mus. p. 39 (1852): type and only species, grunniens, Linn. The rhinarium (fig. 8, A, B) is low and depressed and the whole of the upper surface is covered with short hair except for a comparatively narrow strip running along the upper margin of the nostrils. Beneath the inner edges of the nostrils in front the rhinarium is a little wider than the internarial septum, but the lower portion of its anterior surface is largely overgrown by the hairs of the upper lip, which encroach towards the middle line, leaving a median naked philtrum which is narrower than the internarial septum. In this last-mentioned particular the rhinarium of Poephagus differs from that of all other genera of Bovine. The penis, as recorded by Lonnberg (Ark. Zool. Stockholm, (5) v. no. 10, 1909), has a short tubular urethral prolonga- tion free from the terminal glandular thickening, apparently exactly as in Bibos fronialis and banteng. Genus Anoa, H. Smith. Anoa, H. Smith, Grifiths, Anim. King. vy. pp. 855, 827, as subgenus of Antilope : type, depress ye nis, H. Smith. Bubalus, id. op. cit. p. 871: type, bubalis (= bubalus, Linn.). Buffelus, Riitimeyer, Verh. Ges. Basel, (2) iv. p. 884 (1865): type, now selected, bubalus, Linn. (= ndicus, Riit.). Probubalus, id. loc. cit. : ‘type depressicornis (=celebensis, Riit.). The rhinarium of the two very distinct species I have examined—namely, depressicornis and bubalis—seems to resemble that of Bos and Bibos in all essential characters, exhibiting a large naked dorsal area and a nearly parallel- sided area below the level of the nostrils in front, which is wider than the internarial septum. The feet have the interungual integument naked. The penis I have not examined, but according to en (Nova Acta Soc. Upsal. (3) xx. p. 60, pl. 1..fig. 16, 1903) there is no definite tubular urethral S Aeamee in A. de- pressicornis. Tis figure, nevertheless, suggests the presence External Characters of Ruminant Artiodactyla. 457 of a short urethral process. The statement, however, must be accepted in preference to the figure. Genus Syncerus, Hodgson. Syncerus, Hodgson, Journ, Asiat. Soc. Bengal, xvi. pt. 2, p. 709 (1847) : type, brachyceros, Gray. Planiceros, Gray, Cat. Rum. Brit. Mus. p. 10 (1872), as subgenus of Bubalus: type, planiceros, Blyth (=centralis, Gray). peat, id. op. ctt. p. 12, “as subgenus of Bubalus: type, caffer, parm Apart from the shape of the head, horns, and the size of the ears, I am not acquainted with any important external characters by which the African buffaloes may be distin- guished from their Asiatic allies. My examination, how- ever, 18 restricted to one example—a young bull—of S. caffer equinoctialis? In this specimen the penis was thinner than in other Bovines, and there was no trace of a tubular pro- longation of the urethral canal free from the terminal thickening of the glans (fig.4, A). A side view of the large rhinarium is duces iene, C. Riitimeyer long ago pointed out some of the cranial differences between the African and Asiatic buffaloes, and, admitting them as distinct genera, adopted the name Bu- balus for the former and introduced Buffelus for the latter. For no very good reasons, apparently, he severed the anoa (A. depressicornis) from the Asiatic forms and proposed Probubalus for its reception. In 1901 Loénnberg (K. Svenska Vet.-Akad. Hand]. xxxv. no. 8) adopted Riitimeyer’s opinion as to the generic status of the two types of buffalo, and backed it by the addition of other cranial features. At the same time he showed that the anoa falls into line with the big buffaloes of India, the link between the two being supplied by mindorensis. He followed Riitimeyer also in the matter of nomenclature, with the exception that Probubalus lapsed as a synonym of Buffélus. Nevertheless, in 1903 (N. Acta Soc. Upsal. (3) xx. pp. 55-61) Lénnberg writes on the soft anatomy of Anoa as if it were a genus apart from other Asiatic buffaloes. ‘The reason for this course is not clear, In 1911 Wollister (P. Biol. Soc. Wash. xxiv. p. 191) adopted the views of Ruitimeyer and Loénnberg regarding the buffaloes of Africa and India, without, however, being aware, so far as can be judged, of their publications upon this - subject. Not possessing a skull of depressicornis for exami- nation, he left Anoa alone, adopting the name Budalus for 458 External Characters of Ruminant Artiodactyla. the Asiatic forms and Syncerus for the African. In this matter he was perfectly correct, if Anoa be left out of con-" sideration. But if, as seems to be the case, depressicornis is not generically, or even subgenerically, distinguishable from bubalis, the name Anoa must supersede Bubalus for the Asiatic buffaloes by virtue of page priority. In view of the distinguishing cranial characters between the African and Asiatic buffaloes pointed out by the above- quoted authors, it seems impossible to escape from the conclusion that the two groups deserve generic separation. From lack of material for examination I am unable to add any new external features to those that have been already published. Hollister’s statement, however, that the ears of African buffaloes (Syncerus) are distinguished from those of Asiatic buffaloes (Anoa) by being heavily fringed is not always true. ‘The ears, nevertheless, as I pointed out in 1912 (‘ Field, Aug., p. 396), are very different in shape, those of the Asiatic buffaloes being narrower and much more pointed than of their African allies. Setting aside the characters derived from the shape of the head, the horns, the height of the withers, the length and bushiness of the tail, the distribution of hair on the body, and others that have been made use of by previous workers who have adopted subgeneric or generic titles for the Bovine groups, the incidence of the external features to which attention has been particularly directed in this paper to support the generic recognition of these groups may be briefly summarized as follows :— (1) a. Rhinarium reduced inferiorly by the en- croachment of the hair of the lower half of the upper lip to form a distinct phil- trum which is narrower than the intef- narial septum; its upper surface ovyer- grown with short hair up to the anterior margin, leaving a narrow naked rim above Che MOSiTNSs AL). 5 «cites balers ee a Centos Poephagus. b, Rhinarium very wide inferiorly above the edge of the upper lip, wider than the inter- narial septum, and forming no distinct philtrum ; the hairs of the muzzle spread- ing inwards beneath the nostrils and entering the inner angles of those orifices, reducing the width of the rhinarium at this level; its upper surface covered with hair almost to the anterior edge, so that only a narrow naked rim borders the MOSS AHOVEs 05 Vas ceca dele Quip elll a Bison. Mr. R. EK. Turner on Fossorial Hymenoptera, 499 e. Rhinarium large and naked, everywhere wide below the level of the nostrils in front, its dorsal surface overgrown poste- riorly between the nostrils to a varying extent, but never sufficiently to reduce the upper edge of the nostrils to a narrow NAK CG TUN: ists ss Sha te eh arg tit ee iomers Bos, Bibos, Anoa, Syncerus. (2) a. Feet with the interungual integument evergronvn with hair. :....5.-sfa0 ss eubype.« Bison. b. Feet with the interungual integument naked ..... sh cian nbheonice len ateerstar eters Bos, Bibos, Poepha- gus, Anoa, Syn- cerus. (8) a. Penis with a short tubular urethral pro- cess free for a short distance from the terminal thickening of the glans........ Bibos, Poephagus. 6. Penis without tubular urethral process .. Bos, ison, Anoa, Syncerus. XLV. — Notes on Fossorial Hymenoptera. — XXXVI. On new African Philanthine. By Rowranpo E. Turner, EZ:S.,, F.ES. Philanthus fossulatus, sp. n. 9. Nigra; clypeo, mandibulis basi, scapo subtus, facie usque ad emarginationem oculorum, fronte macula; femoribus anticis subtus, femoribusque intermediis macula parva apicali flavis ; pronoto margine postico, callis humeralibus, tegulis, mesopleuris antice, postscutello, tergito primo macula utrinque, secundo fascia obliqua utrinque, tertio, quarto quintoque fascia apicali, sexto macula magna utrinque, sternitis 3-5 fascia undulata antice bisinuata, secundo fascia lata postice emarginata, sexto fere toto, tibiis tarsisque albidis; flagello, coxis, trochanteribus, femoribus, segmentis abdominalibus primo, secundo, sextoque, tertio apice quintoque basi ferrugineis ; alis hyalinis, venis fuscis, stigmate costaque testaceis. Long. 10 mm. ?. Clypeus very broadly rounded anteriorly, with a few scattered and shallow punctures ; antenne inserted nearer to the eyes than to each other, the front between them distinctly swollen. Front very closely and finely punctured- rugulose, the vertex much more strongly punctured. Antenne not very stout; second joint of the flagellum slender at the base, gradually thickened to the apex, about 460 Mr. R. KE. Turner on Fossorial Hymenoptera. as long as the third and fourth joints combined, third joint a little broader at the apex than long. Ocelli in a broad triangle, the posterior pair fully half as far again from each other as from the eyes. Pronotum as broad as the meso- notum, smooth and shining, the mesonotum shining, with large and rather sparse punctures ; scutellum and _ post- scutellum shining, the former with a few small punctures. Tergites shining, rather closely covered with large and very deep punctures, on the fourth tergite the punctures become sparser and shallow at the apex, those on the fifth tergite are small and scattered, sixth tergite almost smooth ; sternites shallowly and sparsely punctured. Median seg- ment finely and closely punctured; the basal triangular area large, covering almost all the dorsal surface, smooth and shining with a well-marked median sulcus and without marginal carine. Cubitus of the hind wing interstitial with the transverse median nervure, the fore wings with a small fuscous cloud at the extreme apex. Hab. Bohotle, Somaliland (A. F. Appleton). Easily distinguished by the very coarse puncturation of the tergites. “Nearly allied to the group of P. venustus, Rossi. Philanthus flagellurius, sp. n. ©. Nigra; mandibulis, apice exce pto, clypeo, facie infra antennis tegulisque macula basali pallide flavis; tibiis tarsisque anticis femoribusque anticis infra fiavo-testaceis ; tibiis tarsisque inter- mediis posticisque, femoribusque intermedils posticisque apice extremo testaceis ; abdomine rufo-testaceo, basi fiavescente ; alis fusco-hyalinis, venis nigris, stigmate testaceo; antennis crassis- simis. Long. 12 mm, @. Clypeus rounded at the apex, shining, shallowly and very sparsely punctured; front very finely and closely longitudinal ly rugulose, vertex punctured, the punctures more or less confluent transversely ; posterior ocelli as far from exch other as from the eyes. Antenne very stout ; second joint of the flagellum rapidly broadened from the hase, almost as broad at the apex as long, scarcely longer than’ the third joint ; the third to tenth joints broader than long. Mesonotum and mesopleure closely and rather coarsely punctured, scutellum and _ postscutellum more closely and finely punctured ; median segment irregularly rugulose on the sides and on the apical slope; the triangular dorsal area rugose, margined by distinct grooves. The two ee ae ee Mr. R. E. Turner on Fossorial Hymenoptera. 461 basal tergites subopaque, without distinct punctures ; the apical tergites shining, with a few small and scattered punctures ; sternites shining, sparsely but more strongly punctured ; the second sternite smooth, except at the apex. Cubitus of the hind wing originating just beyond the transverse median nervure, Hab. Usangu District, German East Africa, 8500 to 4500 ft. (S. A. Neave), December; Lilongwe District, Central Angoniland, 4000 to 5000 ft. (S. A. Neave), May 28-—June 2, 1910. Somewhat resembles P. dolosus, Kohl, but is easily dis- tinguished by the very stout flagellum and the sculpture of the scutellum and median segment. 'Philanthus fuscipennis, Guér. Philanthus fuscipennis, Guér. Iconogr. regn. anim. iii., Insect. p. 443 (1845). Philanthus consimilis, Kohl, Ann. Naturh. Hofmus. Wien, vi. p. 349 (1891). dQ. Philanthus reticulatus, Cameron, Sjéstedt, Kilimandjaro-Meru Exp., Zool. ii. p. 270 (1910). Hab. The whole Ethopian region. A very variable species in colour; the yellow markings on the scutellum and postscutellum are usually obsolete, as in Guérin’s description, Philanthus nigrohirtus, sp. n. Q. Nigra, mandibulis macula basali, clypeo, facie, macula parva pone oculos, vertice macula obliqua utrinque oculos attingente, pronoto margine postico, tegulis, callis humeralibus macula parva, mesopleuris antice, scutello, postscutello, femoribus anticis intus, tibiisque supra flavis; abdomine fulvo-flavidulo, segmento primo basi nigro; fronte inter antennas dense nigro-hirsuto ; alis fuscis. gd. Feminz similis; fronte supra antennis bimaculata (spe transverse fasciata), vertice immaculato, scutello postscutelloque nigris, nonnunquam flavo-maculatis, clypeo apice macula minuta nigra. Long., 2 12 mm., ¢ 10 mm. ¢. Clypeus very broadly rounded at the apex, very sparsely punctured, with a long black hair springing from each puncture ; front very closely and finely punctured, with delicate longitudinal striz, and rather thickly clothed with long black hairs, which are especially dense between the antenne ; vertex shining, rather closely punctured ; the 462 Mr. R. E. Turner on Fossorial Hymenoptera, ocelli in an almost equilateral triangle, the posterior pair almost as far from each other as from the eyes. Antenne stout, the second joint of the flagellum not as long as the third and fourth combined, the fourth as broad as long. Pronotum smooth ; mesonotum shining, closely punctured, more closely anteriorly than posteriorly, clothed with black hairs ; scutellum and postscutellum almost smooth, pleure closely punctured. Median segment closely and finely punctured, the sulci defining the basal area almost obsolete, a broad longitudinal depre-sion on the middle of the dorsal surface not quite extending to the base. Abdomen smooth and shining, sixth tergite delicately longitudinally striated ; sternites sparsely punctured. Fore metatarsus with seven spines. Cubitus of the hind wing originating distinctly beyond the transverse median nervure. 3. The sculpture throughout rather stronger than in the female, scutellum sparsely punctured, median segment finely punctured-rugose; tergites smooth and shining, the seventh tergite with large scattered punctures. Fourth jomt of the flagellum distinctly longer than broad. Distance between the eyes on the vertex about equal to the length of flagellar joints 2-4. Hab. Mt. Kokanjero, 8.W. of Elgon, Uganda Protectorate, 6400 ft. (S. A. Neave), August 1911 ; Ruwenzori, 7000- 8000 ft. (Scott Elliot). Males with the black pubescence somewhat shorter are in the collection from Ankole-Toro Border, E. of Lake George (S.A. Neave), October 1911; Nandi Escarpment, 5800 ft. (S. A. Neave), May 1911 ; and Uchwezi Forest, British E. Africa (S. A. Neave), March 1912. Philanthus niyrohirtus, subsp. calvus, subsp. n. Specimens of both sexes from the Luangwa Valley, N.E. Rhodesia, are without the long black hairs on the head and thorax, but do not differ appreciably otherwise. For this form I suggest the above subspecific name. The female is without yellow marks on the vertex. This approaches P. stecki, Schulz, but the eyes are a little further apart on the vertex, the posterior ocelli in stecki being distinctly nearer to the eyes than toeach other. Specimens apparently not distinct specifically from calwus from W. Africa (Gambia, Gold Coast, Togo, and N. Nigeria) often have eight spines on the fore metatarsus. These seem to be distinct from P. camerunensis, Tullgr., in which the posterior Mr, R. E, Turner on Fossorial Hymenoptera. 463 ocelli are much further from the eyes than from each other and the clypeus more narrowly rounded. Phiianthus loeflingii, Dahlb. Philanthus loeflingii, Dahlb, Hymen. Europ. i. p. 495 (1845), 9. Philanthus innominatus, Bingh. Ann, & Mag. Hist. (8) x. p. 212 (1902). Hab. The whole Ethiopian region from Harar and the Gambia to Natal. Philanthus triangulum, Fabr. Vespa triangulum, Faby. Entom, Syst. p. 378 (1775). Crabro diadema, Fabr. Spec. Intect. i. p. 471 (1781). Philanthus frontalis, Gerst. Monatsber. Akad. Wiss. Berlin, p. 509 (1857)! ° & Hab. The whole Ethiopian region. Philanthus histrio, Fabr. Philanthus histrio, Faby. Syst. Piez. p. 801 (1804). Philanthus formosus, Sm. Cat. Hym. B.M., iv. p.471 (1856). ¢. Philanthus flavolineatus, Cameron, Sjostedt, Kilimandjaro-Meru Exp., Zool. ii. p. 271 (1910). Philanthus trichocephalus, Cam. Ann, Transyaal Mus, ii. p. 146 (1910). Hab. E. Africa from Harar to Natal; Angola. Philanthus ugandicus, Magy. Philanthus ugandicus, Magy. Bull. Mus. Hist. Nat. Paris, xiv. p. 188 (1908). Q. Philanthus pilifrons, Cameron, Sjostedt, Kilimandjaro-Meru Exp., Zool, ii. p. 271 (1910). ¢. Hab. Fi. Africa, Transvaal to Harar. I think that these, although differing much in colour, are only sexes of one species; but in specimens from Mombasa the males are coloured as the females, with the abdomen wholly testaceous red on the second and third tergites and a yellow spot on each side of the first tergite, the fourth and fifth tergites are marked with black at the base. This appears to be the usual colouring of the species from Harar to Johannesburg. I have seen no females with the colouring of P. pilifrons, but several males from the Nandi plateau and Usanga. Philanthus limatus, Bingh., is allied to this species, but not identical. 464 Mr. R. E. Turner on Fossorial Hymenoptera. Philanthus strigulosus, sp, n. Q. Nigra; clypeo, facie, macula curvata inter antennas, fascia transyersa frontali, orbitis externis anguste tegulisque flavis; tergitis primo macula magna utrinque, secundo, apice excepto, tertioque lateribus fulvo-ferrugineis; tergitis quarto quintoque lateribus anguste, sternitis 2-5, basi nigris, femoribus posticis apice, anticis intermediisque fere totis, tibiis tarsisque flavo- testaceis; alis flavo-hyalinis, apice leviter infuscatis, venis fulvis. 3. Femine similis; fascia frontalilatissima; tergito quarto etiam fulvo-ferrugineo, apice in medio nigro, sexto lateribus flavo- maculato. Long., 9 18 mm., ¢ 17 mm. ?. Clypeus broadly rounded anteriorly, sparsely and shallowly punctured ; front between the antennz convex, very tinely and closely punctured, the front above the antenne very finely and closely longitudinally striated, punctured between the striz ; vertex shining, coarsely, but not closely punctured; ocelli in a broad triangle, the posterior pair a little further from the eyes than from each other ; pubescence dark fulvous on the front, black on the vertex and thorax ; second joint of the flagellum as long as the third and fourth combined, eaeh of the two latter a little longer than broad. Pronotum closely punctured ; mesonotum closely and strongly punctured anteriorly, much more sparsely in the middle and at the apex; scutellum shining, coarsely but sparsely punctured; postscutellum more closely punctured. ‘Triangular area of the median segment. very coarsely obliquely striate-rugose, margined by a very broad smooth and shining space; the sides and apex of the segment very closely, but not coarsely, punctured rugulose. Tergites rather sparsely punctured; the sixth tergite very delicately longitudinally striolate towards the apex; sternites with very sparse large punctures. Basal joint of the fore tarsi with eight spines on the outer margin. Cubitus of the hind wing originating a little beyond the transverse median nervure. 3. Clypeus, face, vertex, mesonotum, and scutellum much more closely punctured than in the female. A bunch of long black hairs springing from just above the base of the mandibles on each side and reaching more than halfway to the middle of the margin of the clypeus. The two basal tergites more closely punctured than the others; seventh tergite coarsely but sparsely punctured. Hab. Near Johannesburg, Transvaal (4. J. Cholmley); f Mr. R. E. Turner on Fossorial Hymenoptera, 465 Basutoland, between Matsekuwa and Mafeteng (R. Craw- shay), March 30, 1902. In the sculpture this approaches P. rugosus, Kohl, which I have not seen, but is a larger species, very differently coloured. There are only seven spines on the fore tarsus of the female in rugosus, instead of eight, and the clypeus of the male rwgosus is armed with three small teeth, which are absent in strigulosus. There is also no mention in Kohl’s description of the tufts of long hairs near the base of the mandibles. The puncturation of the second and third tergites of the female is as close as on the first, though the punctures are smaller. Cerceris bagandarum, sp. u. 2. Nigra ; capite ferrugineo, fascia lata frontali nigra; clypeo, facie, carina interantennali, tergitisque primo, basi uigro, secundoque flavis; pronoto, mesonoto lateribus anguste, tegulis, pleuris, scutello, postscutello, segmento mediano, tergito sexto basi, sternitis primo dimidio apicali, sextoque, pedibusque ferrugineis; coxis supra, femoribusque posticis supra nigris; alis flavo-hyalinis, apice late infuscatis, venis testaceis ; ely peo apice porrecto; mesopleuris subtuberculatis; sternito secundo area elevata basali nulla. 3. Femine similis; pleuris nigris, segmento mediano nigro macula magna ferruginea utrinque, sternitis secundo, sexto, septinioque, tergitisque sexto septimoque ferrugineis ; tergitis tertio, quarto quintoque fascia angusta transversa angulis apicalibus flava; alis subhyalinis, haud flavescentibus; clypeo haud_ porrecto apice angustato et obtuse tridentato; mesopleuris haud tuber- culatis. Long., 9 16 mm., ¢ 11 mm. ?. Mandibles with a large triangular tooth on the inner margin at about one-third from the apex. Clypeus gradually raised from near the base, strongly convex and porrect at the apex, but without a free lamina. Antenne inserted about half as far again from the anterior ocellus as from the base of the clypeus; interantennal carina stroug ; second joint of flagellum about two and a half times as long as the first. Posterior ocelli nearly twice as far from the eyes as from each other and as far from the hind margin of the head as from the eyes. Clypeus aad face subopaque almost impunctate, front and vertex closely punctured- rugose; thorax and median segment more coarsely punc- tured- rugose ; mesopleure with a small tubercle ; triangular basal area of the median segment strongly and regularly Ann. & Mag. N. Hist. Ser. 9. Vol. 11. 34 466 Mr. R. le. ‘Turner on Fossorial Hymenoptera. transversely striate, the strie very feebly arched. Abdomen almost smooth, finely aciculate, the basal segment distinctly broader than long, with a few scattered punctures ; sixth tergite strongly narrowed from the base to near the middle, thence narrowly produced with almost parallel sides and narrowly rounded at the apex. Sixth sternite deeply tri- angularly emarginate at the apex, with tufts of golden hairs springing from just beneath the apical angles, the sixth tergite margined laterally with golden hairs, springing from beneath the segment. S$. Mandibles with a blunt ill-defined tooth near the middle of the inner margin; clypeus and front minutely punctured, sparsely clothed with short sericeous pubesceuce ; the clypeus. longer than broad, narrowed auteriorly, the apical margin with three obtuse teeth. Antenne inserted nearly as far from the base of the clypeus as from the anterior ocellus ; second joint of the flagellum ‘twice as long - as the first. First tergite broader than lone; sixth sternite with an acute spine and a tuft of long golden hairs at the apical angles ; seventh sternite shallowly emarginate at the apex; seventh tergite parallel-sided, truncate at the apex, half as long again as broad. Hab. Kafu River, near Hoima-Kampala Road, Uganda Protectorate, 8500 ft. (S. 4. Neave), December 29- aie 190i, 2 9 ¥; Siroko River, near W. foot of Mt. Elgon, 3600 ft. Uganda Protectorate (S. A. Neave), Aug. 12-14, 1911, 12. Very uear C. diodoata, Schlett., though differing much in colour, The structural points in both sexes correspond closely, but the striation of the basal area of the median segment is more oblique in diodonta and the puncturation of the second tergite is quite, distinct, not obsolete as in the present species; the second tergite is also broader in diodonta, being rather sharply broadened just behind the base. Cerceris sodalis, sp. 0. 2 g. Very close to C. bagandarum and practically identi- cal with that species in the structure, colour, and sculpture of the head, thorax, and median segment, the female, however, has the posterior margin of the pronotum and the post- scutellum yellow. ‘The colour of the abdomen is ferruginous in both sexes, the sternites at the base and the middle of the second tergite black ; the first tergite with a narrow apical band, second very broadly at the sides and narrowly at the apex, tergites 83-5 in the female and 3-6 in the male rather Mr, R. Ki. Turner on Fossorial Hymenoptera. 467 less broadly at the sides and narrowly at the apex yellow. The sixth tergite of the female is very narrow at the apex, more so than in bayandarum, and the second tergite is more distinctly punctured in both sexes than in that species, though less closely than in diodonta. The second tergite of the female is broader than in bagandarum, though scarcely as broad as in diodonta. Hab. 30 miles from Magadi Junction, British E. Africa (F. G. Hamilton), May 1912; Marsabit, British E. Africa (C. A. Neave), October 1911; east shore of Victoria Nyanza, near Karungu (S. A. Neave), April 1911; Kibwezi, British K. Africa, 3000 ft. (S. A. Neave), April 1911. It is quite possible that this and bagandarum may prove to be a subspecies of diodunta, but they are quite easily distinguished, and until large collections are available may conveniently stand as distinct species. C. severini, Kobl, is also very near in structure. Cerceris bicolor, Sm. Cerceris bicolor, Sm. Cat. Hym. B.M. iv. p. 447, no, 52 (1856). 9. Cerceris fossor, Sm. Cat. Hym. B.M. iv. p. 447, no, 54 (1856). ¢. Cerceris andersoni, sp. n. ¢@. Nigra; mandibulis, apice excepto, flagello, articulis apicalibus supra infumatis, tegulis, segmento abdominali sexto, pedibusque, coxls exceptis, ferrugineis ; clypei lamina macula magna, carina inter antennas ad clypel basin, facie fascia lata longitudinali utringue, postscutello, tergitis primo, tertio, quarto quintoque fascia angusta apicali, sternitoque tertio macula transversa apicali utringue flavis; alis sordide hyalinis, apice cellulaque radiali infuscatis, venis fuscis, stigmate testaceo ; clypeo lamina porrecta libera; mesopleuris haud tuberculatis ; sternito secundo area basali elevata nulla. Long. 10 mm. @. Clypeus with a porrect lamina, free from near the base, the lamina coarsely punctured at the sides, the apical margin very shallowly and broadly emarginate and nearly equal to the distance from the base of the clypeus to the apex of the lamina; the clypeus below the lamina smooth and shining, truncate at the apex. Antenne inserted about twice as far from the anterior ocellus as from the hase of the clypeus, the second joint of the flagellum less than half as Jong again as the third. Inner orbits of the eyes almost parallel ; posterior ocelli further from the eyes than from each other. Face sparsely punctured ; head and thorax 34* 468 Mr. S. H. Haughton on a new very closely rugosely punctured, the postscutellum more’ sparsely punctured; pronotum about two-thirds as long as the scutellum. Median segment rugosely punctured; the basal area triangular, almost equilateral, obliquely striated, with a median longitudinal groove, the apex irregularly transversely striated. Tergites strongly but not closely punctured, first tergite broader at the apex than long ; pygidial area rugulose, elongate, fully twice as long as its greatest breadth, and more than three times as long as its apical breadth, the apex subtruncate. Second sternite shining, sparsely punctured, Hab. Eastern edge of forest of Aberdare Mountains, 7300 ft. (7. J. Anderson), February 1911. This belones to the group of the European C. labiata, and is rather closely related to that species, but is not very near any other Ethiopian species. The interantennal carina is less elevated than in- Jadiata, and is flattened to- wards the base of the elvpeus. Two females from Mlanje Plateau, Nyasaland, 6500 ft. (S. 4. Neave), December 1912, have the postscutellum black and the lamina of the clypeus much reduced in size.. These may represent a subspecies, but I cannot regard them as specifically distinct. XLVI.—A new Dinosaur from the Stormberg Beds of South. Africa, By 8S. H. Haveuron, B.A., F.G.S., Assistant Director, South Afiican Museum. (Published by permission of the Trustees of the South African Museum.) Thecodontosaurus minor, sp. 0. The specimens forming the type of this new form were presented to the South African Museum by the late Dr. M. Ricono. ‘They consist of a left tibia, a cervical vertebra, aud a portion of the left ilium. Left Trbia.—The tibia is 109 mm. long. The proximal articular surface is 81 mm. long and 18 mm. broad. This surface for the most part slopes obliquely backwards and laterally, the inner border being convex from front to back aud higher in front than behind. ‘The tuberositas-tibie is almost the highest point of the bone; it is prolonged ante- viorly and turned slightly outwards. ‘he lateral condyle is — Dinosaur from South Africa. 469 strongly developed. Below the head the shaft thins rapidly until at its middle it has an antero-posterior thickness of 12 mm. and a width of 10 mm. Thence it thickens towards the distal end, ‘The anterior face is flat, with a prominent edge on the lateral side and a rounded edge medially. ‘The outer sharp edge is continued down to the anterior distal process. ‘lhe posterior border of the shaft is rounded. The distal surface is trapezoidal in-form. The inner ante- rior border is 20°5 mm. long, the posterior outer border 16 mm. long, while the posterior inner border is 12 mm. long. The anterior process lies 7 mm. above the posterior process. Between the two on the outer surface of the bone is a shallow groove. Cervical Vertebra.—The length of the body is 31 mm, The anterior articular surface is slightly larger than the posterior. Both are considerably higher than broad. The body is pronouncedly amphiccelous. There is a prominent median ventral keel, sharper in its anterior half. The whole body is strongly compressed laterally, having a width at the middle of 5 mm. and at the anterior end of 8mm. The eanal has a height and breadth anteriorly each of 5 mm. The ends of the zygapophyses are missing. The dorsal spine was low and fairly long, with a somewhat convex upper border. Ischium.—A portion of what is probably the left ischium is preserved, including the proximal articular surface. The bone is bent strongly backwards, more so than in Uhecodonto- saurus antiquus as figured by von Huene, so that the ischium must have been directed very strongly backwards. At the broken distal end the bone is 12 mm. thick and 6°5 mm. broad. The inner border of the proximal surface is straight, the lateral border has a prominent outward projection, the maximum width of the surface being 9 mm. The nature of the tibia and the ischium mark these remains off from the Plateosauride, and place them among the Theco- dontosauride. They indicate a member of this family smaller than any hitherto described from South Africa, and which cannot be exactly identified with any Huropean species. I propose, therefore, to give it a new specific name, Thecodontosaurus minor. Type. S.A.M. Cat. no. 3451. Locality. Pitsing, Maclear, C.P. Cutting in road to Naude’s Nek. Llorizon. Red Beds, just below halfway from base. 470 The Re-discovery of Cylindroiulus parisiorum. XLVII.—WNotes on Myriapoda.—XI1V. The Re-discovery of Cylindroiulus parisiorum (Brélemann et Verhoef’). By Hinpa K. Brape-Birks, M.Sc., M.B., Ch.B., L.R.C.P., M.R.C.S., and the Rev. 8S. GRAHAM BRADE-Birks, M.Se. We hope to deal before very long with some centipede and millipede material from the English Midlands, but we think the present brief note advisable, owing to the exceptional interest of the species it records. Mr. 8S. Priest, F.G.S., with Mr. and Mrs. F. J. Epps (all members of the Dartford Naturalists’ Field Club) visited Upper Arley, Worcestershire, on 22. vii. 1918, and took a number of millipedes and centipedes between the bark and trunk of fallen timber in a meadow next to the churchyard there. This material, which was kindly submitted to us by the collectors, included a species of Julus (s. 1.), which upon dissection we found to be referable to Cylindroiulus parisi- orum (Brélemann et Verhoeff, 1896). SS == 2S SSS Anterior and posterior gonopods in profile. x 100, H. XK. B.-B. del. We sent our drawing of the gonopods to M. le Dr. Henry W. Brélemann, who agrees with our diagnosis, and informs us, in litt., that nobody appears to have identified the species since its first description (1). ‘hus some doubt had arisen in Dr. Brélemann’s mind as _to the validity of the species. The English rediscovery of the animal is therefore of some Importance. Externally C. parisiorum is practically indistinguishable 7 “ene cue On the Pectoral Fin of Eusthenopteron. 471 from C. britannicus, Verhoeff, and C, fristus, Verhoeff, both of which are not uncommon English species. However, the gonopods, which are figured by Brélemann and Verhoeft (loc. cit.), are quite definite diagnostic characters, and so there is no doubt about the record. Our material bears these numbers :—1379, 1380, 1381, 1382, Brade-Birks collection. REFERENCE. (1) BrétemMann, H. W., and C. W. Vernorrr. “ Matériaux pour servir & une faune des Myriapodes de France.” Feuille des Jeunes Naturalistes, Sept. 1896, no. 311, pp. 214 e¢ seg., with 10 text-tigs. XLVITI.—Note on the Pectoral Fin of Eustheuopteron. By Dr. BRANISLAV PETRONIEVICS. THE pectoral fin of Husthenopteron was figured and described for the first time by Whiteaves (comp. J. F. Whiteaves, 1889, p. 87, & pl. v. fig. 5), whose description was improved by Traquair (comp. R. H. Traquair, 1890, p. 19). Two other specimens of the same fin were figured by A. 8. Wood- ward (1898, p. 25) and W. Patten (1912, p. 391). During my stay in London this year the pectoral fin in the British Museum specimen P. 6796 of Lusthenopteron, figured by A. 8. Woodward (whose figure was republished by E. 8. Goodiich in 1902, pl. xvi. fig. 1), was somewhat newly prepared by Mr; FO: Barlow. a give here a new figure of 1t (comp. text-fig. 1) and a brief description. The pectoral fin in our specimen is composed (1) of an axis, (2) of preaxial radials, and (3) of postaxial processes. — The axis consists of four pieces. The first or basal piece is situated behind the displaced cleithrum, of which the inferior edge lies near to its superior edge in the specimen. It is not possible to decide whether this elongated and somewhat obscure bony matter is to be identified wholly with the basal piece of the fin, or whether it does not comprise also the coraco-seapular ossification. Should this latter be the case, then the front edge of the postradial process of the basal would mark the limit between the basal and coraco-scapula. The second piece of the AXIS is expanded and slightly bifurcated posteriorly. he third ‘piece is somewhat longer than the second and expanded still more posteriorly, where it has not only a large postaxial process, but is also more distinetly bifurcated. 472 Dr. Branislav Petronievics on the Fig. 1. Pectoral Fin of Zusthenopteron, British Museum specimen P. 6796. Nat. size. cl., cleithrum ; cosc., the possible coraco-scapula; J.azt., the first axonost or the basal; 2.azt., second axonost; 3.azt., third axonost; 4.azt., fourth axonost; J.prar., first preaxial radial; JZ.pra.r., second preaxial radial; IZ/.pra.r., third preaxial radial; pa.pr., postaxial process ; dermal rays are represented by lines, Peetoral Fin of Kusthenopteron. AT} Finally, the fourth piece of the axis is somewhat con- stricted in the middle, and quite distinctly bifurcated poste- riorly (a feature not marked in the figure of A. S. Woodward, i868). When looked at with a maguifying-glass, these two posterior branches seem to continue in two separate ossifica- tions, so that the composition of this fourth axonost of two separate parts is not improbable, although not to be affirmed with certainty, the separating line between the two being perhaps due to a crack, One sees also with the magnifying- glass the clear attachment of a dermal ray to the left of these two bifurcations, while a fragment of somewhat crushed bony matter attached to the right bifurcation also probably represents dermal rays. There are three preaxial radials in our specimen. The uppermost radial is attached to one of the two articulating surfaces of the basal axonost; it is bent inwards in the middle and constricted posteriorly. The new preparation shows the attachment of the dermal rays to this radial very -clearly. The second radial, attached to the smaller of the two articulating surfaces of the second axonost, is also con- stricted posteriorly, but not sufficiently preserved in its poste- rior part. The third radial, better preserved than the second, is constricted in the middle, but the limit of its posterior part is indeterminable. It is attached to the smaller of the two articulating bifurcations of the third axonost. There are only two postaxial processes in our specimen, and no postaxial radials at all. The first process is a large pro- longation of the basal axonost (this prolongation is not well visible in the figure of A. S. Woodward, 1898), and the second a prolongation of the third axonost, while the second and the fourth axonosts are devoid of similar processes (on the left side of the second axonost some bony matter is visible in our specimen, but it is evidently a crushed scale). Having finished the description of the fin in question, I will add some remarks concerning the problem of the origin of the tetrapod limb. The resemblance of the internal skeleton of the pectoral (and also of the pelvic) fin in Husthenopteron to the internal skeleton in the tetrapod limb has been empha- sized by several authors (by Patten, Watson, Broom, Gregory), and Watson especially has tried to point out in detail the homologies of both (comp. Watson, 1913, p. 25 seg. and figs. 1 & 2). But his restoration of the pectoral tin of Kusthenopteron (l. ¢. fig. 2) is wrong, inasmuch as he takes no account of the posterior bifureation of the fourth axonost (in this respect the restoration of Broom, 1913, p. 460, fig. 1 is more accurate) and represents the postaxial process of the AT4 Dr. Branislav Petronievics on the basal axonost as a separate postaxial radial (in this respect the restoration of Broom is exact). . Now I consider the posterior bifureation of the fourth axonost in our specimen as of exceptional importance for the question of homologies. As tlhe pelvic fin of Husthenopteron is far more reduced than its pectoral fin (comp. fig. 1 of pl. xvi. in Goodrich, 1902, which shows that there is no fourth axonost in the pelvic fin—British Museum specimen P. 6794—and no postaxial processes), we must infer that the paired fins of Husthenopteron represent a stage far in advance of that stage of the paired fins in its ancestors, which was the starting-point for tle evolution of the paired limbs in tlie primitive ancestors of the Tetrapoda*. If this inference is a right one, then it is not improbable that the posterior bifurcation of the fourth axonost in our specimen is a remnaut of a more primitive stage when the fourth axonost was com- posed of two separate ossifications, the paired fins of Hustheno- pteron being evidently the reduced archipterygium-type of Gegenbaur (a resemblance recognized by Woodward, Tre- quair, and others). So that we have to conclude from this evolution that the axis of the tetrapod limb runs along the humerus, ulna, ulnare, and between the fourth and fifth finger T (comp. text-fig. 2, in which some further hypothetical homo- logies have been indicated). This conclusion, as one secs, * * This conclusion is confirmed also by the skull, which in Hustheno- pteron is simpler than in the more primitive Osteolepide, whose paired fins are also less reduced (comp. the tins of Megalichthys figured by Ed. D. Wellburn in his paper “On the Genus Megalichthys,” in Proc. Yorkshire Geol. & Polytechnic Soc. vol. xiv., 1900), I may add in this connexion that the skull of Osteolepis may be considered to approach nearer to the Stegocephalian skull than is shown by the restoration of Pander (comp. Chr. H. Pander, ‘ Ueber die Saurodipterinen, &c.,’ 1860, pl. i. figs. 8 & 9), lately reproduced by Gregory (comp. Gregory, 1915, tig. 2, A, B). Pander’s restoration was founded on the specimen of Osteolepis microlepidotus figured by him in pl. i. fig. 1; but tig. 4 on the same plate represents a specimen in which all the three characteristic bones of the Stegocephalian skull (supratemporal, intertemporal, post- orbital) are present. + The pectoral fin of Swuripterus taylort (figured and restored by Gregory, 1915, plate iv. and fig. 9) does not militate against this supposi- tion. ‘This fin, less reduced than that of Lwsthenopteron, has three elements attached to the third axonost, so that these three elements may correspond with the three digits on the ulnar side of the tetrapod limb. As the two outer of these three elements have almost the same length, it may well be supposed that the axis runs between the two (and not along the outer one alone, as Gregory hypothetically supposes—comp. Gregory, 1915, p. 860). I should mention that the first to emphasize the resemblance of the Swwzpferus-fin with the tetrapod limb was its discoverer, James Hall himself (comp. J, Hall, ‘Geology of New York,’ part iv, 1843, p. 282). Pectoral Fin of Husthenopteron. 475 does not entirely confirm the theory of Gegenbaur, according to which the tetrapod limb is derived from a reduced uniserial archipterygium (comp. Gegenbaur, 1898, p. 520), but never- theless it is more in conformity with this theory than with the other (also advocated by Watson), which takes a reduce biserial archipterygium for the base of the tetrapod limb. Fig. 2. The internal skeleton of the Pectoral Fin of Eusthenopteron, showing homologies with the tetrapod limb, Nat. size, hu., humerus; u., ulna; 7., radius; w/., ulnare ; p., pisiform; ca., three distal carpalia; I—V., digits; av., axis of the tetrapod limb, In conclusion, I desire to express my thanks to Dr. Smith Woodward for the loan of the new preparation and_ for valuable help. LITERATURE, 1. J. F. Wurrraves. “TIllustrations of the Fossil Fishes of the Devonian Rocks of Canada,” in Trans. Roy. Soc. Canada, vol. vi. 1889, p. 77 seg. (on ELusthenopteron, p. 78 seq.). 2. R. H. Traquartr. “Notes on the Devonian Fishes of Scaumenac Bay and Campelltown in Canada,” in Geol, Mag. vol. vii. 1890, p. 15 seg. (on Husthenopteron, p. 18 seq.). 476 Mr. T. D. A. Cockerell—Descriptions and . A. S. Woopwarpb. ‘Catalogue of the Fossil Fishes ‘in the British Museum,’ pt. ii. 1891 (on Lusthenopteron, p. 361 seq.). 4, ——. ‘ Vertebrate Paleontology,’ 1898 (on Lusthenopteron, p. 25 seq. & p. 76 seq.). 5. E.8. Goopricu. ‘On the Pelvic Girdle and Fin of Eusthenopteron,” in Quart. Journ. Mier. Soe. vol. xlv. 1902, p. 311 seq. 6. ——. ‘ Cyclostomes and Fishes,’ Part IX. Vertebrata Craniata of Sir Ray Lankester’s ‘ A Treatise of Zoology,’ 1909. . L. Hussaxor. ‘Notes on Devonie Fishes from Scaumenac Bay, Quebec,” New York State Museum, Bulletin 156, 1912, p. 127 seq. (on Lusthenopteron, p. 131 seq.). 8. W. Patren. ‘The Evolution of the Vertebrates and their Kin,’ 1912 (on Eusthenopteron, p. 391). 9. W. K. Gregory. “ Present Status of the Problem of the Origin of the Tetrapoda, with special reference to the Skull and Paired Limbs,” in Annals N.Y. Acad. Sci. vol. xxvi. 1915, p. 317 seg. (on Eusthenopteron, p. 353 seg. & p. 364). 10. C. Gecensaur. ‘ Vergleichende Anatomie der Wirbeltiere,’ i. Bd., 1898. 11. D. M.S. Watson. “On the Primitive Tetrapod Limb,” in ‘ Anato- mischer Anzeiger,’ vol. xliv. 1918, pp. 24-27. 12. R. Broom. “ On the Origin of the Cheiropterygium,” in Bull. Amer. Mus. Nat. Hist. vol. xxxii, 1918, pp. 459-464. ioe) ALIX.—Descriptions and Records of Bees —UXX XIL. By T. D. A. Cockrreti, University of Colorado. Exomalopsis mellipes, Cresson. The male, not before known, has been collected by H. H. Hyde at Medellin, Vera Cruz, Mexico (Baker coll., 1785). It runs in Friese’s table of males to H. planiceps, Sm., but is larger, with red legs. Exomalopsis vincentana, Cockerell. The male, previously unknown, was collected by HL. H. Smith on the windward side of St. Vincent. It is hardly 5 mm. long, and there is much black hair on mesothorax, scutellum, and legs. It is nearest to H, globosa, but dis- tinguished at once by the ochreous-yellow tarsi. There is a series of small Hxomalopsis (including Antho- phorula), which are superficially similar and easily confused. They may be separated by the following table, based ‘on females :— Second abdominal segment with oblique stripes of light hair at sides, but no apical band ie es © 8b 0M 8 Oe wn 2 a6 te Sew W108. e)6 euviione Liecords of Bées: * ee ATT ' Second abdominal segment with an ’ apical Weeeee ranged “5a eter sigs eMemadev Meth, aft Oh cks 3 1. Dise of scutellum with black hair .......... globosa (Fabr.). Dise of scutellum with fulvous hair ........ 2. 2. Basitarsus with much black hair ..,....... pulchella, Cresson. Basitarsus with pale hair ................ similis, Cresson. 5. Second segment of abdomen with a narrow ; apical band of snow-white hair...... eoeee. verbesine, Ckll. Second segment with a broad band.......... 4. Abdominal hair+bands . clear white; sia DREIRG rr aN rele a: 7.0\0 9.50 056, wk. aalahhaeceene chlorina, sp. n. Ab Jominal bands greyish or yellowish ; ey es HLGOVONCOMS sfa7 Fels, « cht e\vecs’ oh v whe daecarelatalehoraieient 5. 5. Hind legs with much black hair............ 6. Hind legs with hair mainly or nearly all pale ; species of Anthophorula: .......cccceeees ts 6. Flagellum ferruginous beneath, abdomen TOR SI Ward vr sec ey Ahicte rere meemattAG wo Srape Cate .. mtens, Ckll. Flagellum dark coffee-brown beneath. .... albovittata, sp. n. ‘¢ Text ule rufo-testaceous; stigma larger, pale enemies eM EN. ox. or Bee texana, Friese. Tegule dark; stigma smaller .)............ 8. 8. Dise of mesothorax polished and smooth .... coguillett?, Ashmead. Dise of mesothorax punctured ............ .. morgan, Cll. Ezomalopsis albovitiata, sp. n. Closely allied to the Californian £. nitens, but less robust ; flageilum dark ; hair of face pure white ; disc of mesothorax with fine but distinct punctures; hair of scutellum shorter and greyish instead of yellowish; hair on base of first abdominal segment pure white, apex of first segment with only a rather small patch of white hair on each ‘side. The loose scopa of hind tibiz and tarsi is black behind (above) and white in front; the wings are dusky, and the tegule are piceous. Oaxaca, Mexico (Crawford). U.S. Nat. Museum. There is some resemblance to Leptergatis globulifera, but the front is smooth and shining in the Hzomalopsis, densely punctured in the Lepteryatis. Exomalopsis chlorina, sp. n. ? .—Length about 6 mm. ; Eyes bluish green; hair at sides of face dense and pure white; flagellum red beneath, dark above ; hair of thorax white ; tegulz rufo-piceous; wings clear, stigma and nervures pale amber ; stigma much smaller than in H. tevana ; bands on abdominal segments 2-5 broad and pure white; scopa of hind legs on outer side white, blackish at base of tibia, 478 Mr. T. D. A. Cockerell— Descriptions and dark fuscous on inner side of basitarsi ; mesothorax very distinctly punctured; tarsi red at apex. Las Cruecs, New Mexico, at flowers of Spheralcea in garden of my house, Aug. 24 (Cockerell). I had confused this with E. texana, hut, having received a topotype of the latter, I find it is quite distinct. Exomalopsis thermalis, sp. 1. 9 —Length about 9 mm. Very robust, black; hair of head and thorax long and white, with a slight creamy tint; head very broad; eyes olive-green ; labrum black ; mandibles chestuut-red im middle; clypeus flattened, shining, sparsely punctured ; flagellum chestnut-red beneath; mesothorax closely and strongly punctured; scutellum shining, with very fine punctures ; tegule bright rufo-fulvous. Wings yellowish, the large stigma and the nervures clear ferruginous ; small joints of tarsi red; hair on inner side of tarsi ferruginous ; middle tibiz with short fuscous hair on outer side beyond middle; middle basitarsi with long white hair on outer side ; scopa of hind legs long and plumose, largely black on outer side, that on basitarsus of three colours—black, white, and red. Abdomen very broad, with a glaucous tint; first two segments closely punctured as far as the narrow arched pale hair-band, beyond that smooth and shining, the second segment with excessively minute punctures ; segments 3 to5 with broad bands of yellowish tomentum, the fifth broadly fringed with fuscous hair apically. Aguascalientes, Mexico, Dec. 1, 1909 (F. C. Bishopp). U.S. Nat. Museum. Exomalopsis crucis, sp. 0. 9 .—Length about 8°5 mm. Closely allied to the last, differing thus : scape more or less reddish, especially at base ; flagellum pale ferrugimous beneath ; labrum clear red, with pale reddish hair ; hair of thorax above strongly tinged with yellowish; scutellum closely and very distinctly punctured; first abdominal segment reddish basally. Medellin, Vera Cruz, Mexico (H. H. Hyde ; Baker coll., 1785). U.S. Nat. Museum. : These two species are related to EH. mellipes, Cress. (which has red legs) ; and more especially to H. frederici, Ckll., which has the tarsi, and tibiz at apex, ferruginous—at a ee Records of Bees. 479 least, in the male (female unknown). I questioned whether E. thermalis might be the female of frederici, but the fine short pile on basal part of third abdominal segment in thermalis is pale greyish ochreous, in frederici it is black. The hind spurs of thermatis and crucis are strongly curved at end, as in frederici. A second specimen of £. crucis comes from San Juan Allende, Mexico, Nov. 29 (C. H. T. Townsend). Leptergatis globulifera, Cockerell. The female, not before known, was taken by M. A. Carriker at Aroa, Venezuela, Dec. 12, 1910. It is much like L. armata, Sm., but has redder antenne. From the female alone, I should have regarded the insect as a local race of urmata. Tetrapedia diversipes, Klug. Manaos, Brazil (Miss H. B. Merrill); San Bernardino, Paraguay (4. Mebrig). Nomada calloptera, sp. n. $.—Length about 10°5 mm.; expanse about 18°5. Ifead and thorax black, densely punctured, with long and abundant pale fulvous hair; lower corners of face broadly (with a sharply pointed extension upward along orbit), broad band aloug lower margin of clypeus, base of the simple mandibies, labrum (which is not dentate) and the rather stout scape in front, all yellow; eyes pale grey; flagellum thick, simple, black above (except the sutures), ferruginous beneath; third antennal joint brighter red, about half as long as fourth; scutellum bigibbous, very coarsely punctured; tubercles red and polished, but no other light marks on thorax ; tegule red. Wings clear, the apex fuscous ; stigma clear bright ferruginous, nervures fuscous ; b. n. going a short distance basad of t.-m.; first and second t.-c. nervures convex outwardly. Legs red, anterior tibiz with an apical yellow spot ; middle trochanters black above, with a red spot, and lighly polished ; middle femora black beneath Lasally ; hind femora black behind except at apex. Abdomen red with rather pale yellow markings, hind mar- gins of first three segments broadly fuscous, first segment with more than basal half black, and small yellow marks sublateraily ; second segment black at base, and with a very large yellow patch (not pointed mesad) on each side ; third 480 Mr. T. D. A. Cockerell— Descriptions and with a very’ broadly interrupted yellow band, excavated behind sublaterally; fourth to sixth with yellow bands, interrupted by a red spot on each side; apical plate broad, notched ; venter red with yellow bands. Tokyo, Japan, April 12, 1909 (Sasaki). U.S. Nat. Museum. It is also labeiled Yamada. In the table of Palearctic species it runs near N. manni, Moraw., differing by the black scutellum. It is quite dis- tinct from all those described from Japan. It is a large species of Nomada, s. str. Nomada pyrifera, sp. 0. ? .—Leneth about 10 mm. Head and thorax red with black markings, closely punc- tured, the hair white ; labrum pale yellow, with no distinct tooth ; malar space pale yellowish; mandibles simple, red, black at apex; lower part of clypeus, and lower part of supraclypeal area, suffusedly yellowish; middle of front, extending to occiput, black, and cheeks black with a broad red band behind eyes ; antenne entirely red, long, reaching to base of abdomen; third joint scarcely half as long as fourth (this at once separates it from the superficially similar N. japonica, Sm.) ; mesothorax with three black bands, confluent in front; scutellum strongly elevated, entirely red; area of metathorax black in middle and red sublaterally; pleura nearly all red; no yellow on thorax ; tegule pale red. Wings clear, dilute fuscous at apex; stigma ferruginous; nervures fuscous ; b. n. going far basad of t.-m.; second s.m. very broad, receiving first r. n. about middle. Legs bright ferruginous, hind femora with a black stripe behind. Abdomen smooth and polished, ferruginous ; basal half of first segment black, second segment with a very large pyriform (pointed mesad) spot on each side; fourth and fifth segments with yellow bands, failing laterally ; venter with broad yellow bands. Japan (presumably Tokyo), May (Sasaki), U.S. Nat. Museum. This also runs near N. manni in the Palearctic fauna, but is readily distinguished by the pattern of abdomen and the red scutellum. Sasaki collected two males, of different species, which looked like N. pyrifera. One I have described as N. calloptera, as it differs from pyrifera in the colour of the stigma and the basal nervure going less basad ; the other, collected at ‘'okyo in April, I suppose to be the true male of pyrifera. It is unfortunately in very bad condition, but Records of Bees. 481 the following characters can be made out : mandibles largely yellow ; face densely covered with white hair ; scape swollen, yellow in front ; mesothorax all black ; tubercles yellow ; scutellum with yellowish or reddish spots ; metathorax and pleura all black ; venation and colour of stigma as in pyri- fera; first abdominal segment with basal half black, apical half red, and two large yellow spots, not far apart, on the red ; second segment with pyriform marks larger, meeting in the middle line; segments 3 to 6 with entire yellow bands ; apical plate feebly notched; venter with yellow bands. Andrena melanospila, sp. u. 9? .—Length 10 mm. Black, the head and thorax with copious moderately long hair, dull white on face, cheeks, and pleura, pale fulvous on occiput and dorsum of thorax (brightest on scutellum), but black on mesothorax posteriorly, and on front and vertex ; malar space linear; process of labrum rather narrow, obtuse; clypeus brightly polished, with sparse small punc- tures ; facial fovee broad, dark brown, not extending below level of antenne ; antenne dark; third joiut much longer than fourth, but not quite as long as fourth and fifth ; mesothorax dull and granular, shining posteriorly ; scutellum shining, without evident punctures; area of metathorax dull and finely granular; tegule piceous. Wings dusky, the large stigma and nervures dull reddish; b. n. meeting t.-m.; second s.m. receiving first r. n. distinctly beyond middle ; scopa of hind tibiz white in front and black behind. Abdomen dull, not punctured; second segment depressed scarcely a fourth; hind margins of segments 2 to 4 with narrow pure white hair-bands; caudal fimbria purplish black. Soochow, China (N. Gist Gee). U.S. Nat. Museum. In the Palearctic fauna this falls near to A. denticulata (Kirby), from which it is easily separated by the narrow white abdominal bands and the black and white hair of hind tibie. It is not like any of the species described by Strand from Tsingtau. The abdominal bands are as in A. wilkella, but that has an entirely different clypeus. Andrena delicatula, sp. n. | ¢ .—Length 8 mm. Black, superficially exactly hke A. albicrus, but running in tables of Palearctic species to A. lapponica, which is a Ann. & Mag. N. Hist. Ser. 9. Vol. it. 35 A482 Mr. O. Thomas on larger insect. Hair of head and thorax long and white, very faintly yellowish on scutellum, a little blackish hair at sides of face ; mandibles long and curved ; process of labrum weakly bilobed ; clypeus dull, covered with long white hair; antenne entirely dark ; third joint about equal to fourth ; mesothorax and‘ area of metathorax dull and granular tegule piceous, reddish posteriorly. Wings slightly dusky ; the large stigma and nervures dull ferruginous ; b. n. falling some distance short of t.-m.; second s.m. broad, receiving ‘first rin. at middle: Legs black, tarsi reddish at apex. Abdomen shining, not punctured, segments 2 to 4 with thin white hair-bands at sides only ; apex emarginate. Soochow, China (N. Gist Gee, 121). U.S. Nat. Museum. The abdomen has little of the long loose hair so con- spicuous in A. albicrus. Among the Japanese species, this falls nearest to A. precociformis, Ckll., which is larger, with shining clypeus and chestnut-red stigma. ‘The cheeks are broader and flatter in A. delicatula. From Soochow also comes Nomia chalybeata, Smith (N. Gist Gee, 140). * * Agapostemon cockerelli, Crawford. TReeoRe Colorado, Sept. 7, 1918 (Cockerell). New to Colorado. Colletes sieverti, Cockerell. Gregory Canyon, Boulder, Colorado, July 13 (Cockerell). Trigona ruficrus corvina, Cockerell. Chagres River, Panama Canal Zone, Oct. 9, 1917, “ chew- ing on the leaves of young citrus plants ” (Harold Morrison). L.—A new Species of Kligmodontia from Catamarca. By OLDFIELD ‘l'HOMAS. (Published by permission of the Trustees of the British Museum.) THE British Museum has recently received a small collection of mammals from Chumbicha, Catamarea, collected by Sr. HE. Budin, and among them there occur specimens of the following new species :— a new Species of Eligmodontia. — 483 << "Hligmodontia marica, sp. n.. Size smaller than’ in other species. Fur soft and fine, hairs of back about 7 mm. in length. General colour above pale sandy buff, darker along the back, paler on the sides where it is nearly “ pinkisli buff.’ Whole of under surface pure sharply defied white, all the hairs, even laterally, white to ‘their bases, Middle of face and crown darker buffy like the back, aréa‘ between eyes and ears, and a patch above each ‘eye piler like the sides. Ears large, the usual piebald arrangement of their colour strongly marked ; a whitish patch at base of proectote, middle part of proectote nearly black, terminal part and whole of metentote greyish buffy, the fine hairs along the edge white. Limbs wholly white, the buffy body-colour not or scarcely encroaching on the white of the upper arms; palins and soles with the structure characteristic ot Hligmodontia, but the hairy covering quite thinly spread. Tail ‘longer’ than head and body, dull buffy above, whitish below, the contrast not so marked as it is in the southern species, ‘ Skull markedly smaller than that of the other species, especially as compared with that of the forms geographically nearest. 2 Dimensions of the type (measured in the flesh) :— Head and body 65 mm.; tail 93; hind foot 20; ear 15. j Skull: greatest length 21°4; zygomatic breadth 12 ; nasals 8; interorbital breadth 3°8 ; breadth of brain-case 11 ; palatrlar length 9°33 palatal foramina 4°$; upper molar series 370. Hab. Chumbicha, Catamarca. Alt. 600 m. Type. Young adult male. B.M.no.18.11.11.1. Original number 311. Collected 30th July, 1918. Presented by Oldfield ‘Thomas. This beautiful little mouse is the smallest species of the genus and is readily distinguishable by size from L. hértipes aud morent, occurring north and south of it respectively. BH. typus, with which the Bahia Blanca elegans is always assumed to be synonymous, is also larger, and the belly-hairs are broadly slaty at base. ‘he more southern /. morgand has a proportionally shorter tail. Sr, Budin says of HH. marica :—“ This pretty mouse has been the one which has most pleased and interested me of all the rodents. It was caught among the prickly pears [‘pencas’”] in one place only, in a space some bik square . 30 484 Mr. O. Thomas -on metres in area, where I obtained four specimens, but saw none anywhere else, and it is evidently very rare.” [As an indication of the extent to which our British National Museum has participated in the general advance in the systematic knowledge of Mammalia, and the corresponding accumulation of typical specimens, I may perhaps be per- mitted to record that, so far as I am able to calculate, this ig the two-thousandth mammal to which, as the official mammalogist of the Museum, I have had occasion to give a name. And many hundreds more have been described and named by other workers. The vastness of the collection— especially of types—indicated by these figures is due mainly to the patriotism of our countrymen all over the world, many of whom have been proud and pleased to contribute to their National Museum merely because it is the National Museum, without pay or return, and often in climates where mere existence is a burden. Having possessed for forty years the great privilege of working on this wonderful collection, I feel I cannot too strongly express my appreciation of the generosity and public spirit shown by its many contributors—whether those who at home have provided funds for making expeditions, or abroad have made collections to be added to the National treasures. My own share in the woik, carried on as it has been under the most favourable conditions, has been a continuous pleasure. And in appreciation of one important element in this pleasure, the sympathetic and ever-ready help of my wife, I have given to this attractive little animal the above specific name. | LI.—Two new Forms of Leggada. By OLpFIELD ‘THOMAS. (Published by permission of the Trustees of the British Museum.) Leggada bella sybilla, subsp. n. Near L. b. induta, but with much shorter fur. Hairs of back about 4°0-4'°5 mm. in length. General colour buffy, not so bright as in induta, and broadly darkened on the back, the flanks clear buffy. Belly pure sharply defined white. A very small subawal white spot. Hands and feet white. ‘ail pale greyish above, white below. Two new Forms of Leggada. 485 Skull about as in induta, smaller than in minutoides, slightly larger than in marica. Posterior nares of normal shape. Dimensions of the type :— Head and body 55 mm.; tail 46 ; hind foot 13. Skull: greatest length 18; condylo- incisive length 16°3 ; nasals 6° 8; breadth of brain-case 8°5 ; palatal foramina 4; upper molar series 3:0. Hab. Benguella, Angola. Type from the Usolo River. Type. Adult female. B.M. no. 5.5.9. 70. Original number 7. Collected 18th July, 1904, by Dr. W. J. Ansorge. Seven specimens. The type of sydilla was captured at the same time of year as that of induta, so that the difference in the fur is not seasonal. Dr. Ansorge also obtained examples of this pretty mouse in November and December. In ZL. 6. marica the molars are only 2°6 mm. in length. Leggada paulina, sp. n. Intermediate between the two West-African species LL. museuloides and setulosa. Size markedly less than in setwlosa, rather greater than in musculoides, (General colour greyish mouse-colour above, with a wash of drabby or buffy along the cheeks, shoulders, and flanks. Under surface pure white, not so sharply defined as in musculoides. HKars small, as in musculoides. Forearms tinged with buffy, legs greyish; hands and feet white. Tail so thinly haired as to appear naked to.the unaided eye, the fine hairs brown above, whitish below ; the scales brown throughout. Skull intermediate between those of setulosa and musculoides. Brain-case rounded, not so flattened as in musculoides. Masseteric knob of zygomatic plate near its anterior border. Dimensions of the type (measured in flesh) :— Head and body 67 mm.; tail 48; hind foot 13°7; ear 9°95. Skull: greatest length 18:2; condylo-incisive length 16°5 ; zygomatic ; brenden 9; nasals 6°7; interorbital breadth 3°6 ; breadth of brain-case 84; palatilar length 7:9; palatal foramina 3°9; upper molar series 3. Hab. Bitye, Ja River, §.E. Cameroons. 2000’. Type. Adult female. B.M. no. 14. 1. 24.27. Original ‘number 694. Collected 15th September, 1913, by Mr. G. L. Bates. Though evidently allied to L. musculoides, of which it may 486 Mr. W. T.. Distant on ¥he° ~ be a Gereronns representative, this mouse is distinguishable by its larger skull and darker coloration, in ‘whieh latter it nearly resembles the common Cameroons fs setulosa, in whose company it was captured, and for whose young it might readily be mistaken. Contributions to a further Knowledge of the Rhynchotal Family Lygeide. By W. L. Disranr, [Continued from p. 270.] Astacops tigrinus, sp. 1. Head, pronotum, scutellum, and corium pale ochraceons ; poeta black, basal joint ochraceous; apices of the stylated eyes black ; boty beneath pale ochrac ‘eous with prominent transverse, somewhat broad, black fascize, the most premi- nent bei g at the anterior margins of the meso- and metasterna, and ai the po-terior.margins of the abdominal segments, there is also a small black spot on each side of the anterior marginal area of the prosternum and a central black longitudinal fascia on the apical abdominal segment; legs black, anterior and intermediate femora (excluding bases), apical third. of posterior femora, and extreme bases of tibiw ochraceous ; tarsi mostly black; antennee with the second and. fourth joints subequal in length, each a little longer than third; scute!lum transversely subeonvex on basal area, ceutrally thence to apex strongly carinate ; membrane black, apical margin pale and passing the abdominal apex. Long. 12 mm. Hab. Philippine Islands; Mindoro Is'and, Baco Base (J. J. Mounsey). Scopia:tes nigripes. Scopiastes nigripes, Dist. Ann. & Mag. Nat. Hist. (7) vil. p. 533 (1901). Astacops melampus, Bergr. Phil. Journ, Sci. xiii. p. 57 (1918). Hab. Queensland. Macropes simoni, sp..n. Head, pronotum, scutellum, body beneath, aud legs black; anteune piccous, apical joint black ; ; hemelytra pale creamy yellow, clavus brown, vein outside clavus also brown, uearly Rhynchotal Family Ly geide. 487 apical half of corium black; membrane with the base black, and with a large discal ‘spot fuscous with the veins black ; antennz with the first and second joints subequal in length, each a little shorter than fourth ; rostrum passing the anterior coxee ; pronotum with the auterior lobe smooth, shining, black, punctate anteriorly and laterally, with two finely impressed central longitudinal lines, posterior lobe more opaque and thickly punctate, anterior lobe not promi- nently broadened asin M. philippinensis, Dist., but gradually somewhat convexly narrowed to apex ; membrane reaching or very slightly passing the anterior margin of the apical abdominal segment; scutellum centrally, longitudinally carinate. Var. Abdomen beneath and the legs brownish ochraceous. Long. 5-53 mm. Hab. Philippine Islands (#. Simon). A species readily distinguished from M. philippinensis, Dist., by its small size and structure of the pronotum, &c. Bergroth has recently described another small species, M. lacertosus, from the same habitat, but, as he states ‘‘ pro- notum in the male with the greatest width before the middle ” and with different colour-markings to the “ elytra,” it cannot be confused with his specific creation. . ~ . Dinomachus marshall, Dist. Ann. & Mag. Nat. Hist. (7) vii. p- 473 (1901). Bergroth, my constant but by no means infallible critic, has recently (Medd. Mus. Zool. Afd., Gottenborg, p. 6, 1914) referred to my very short and quite misleading “description of the genus.” He states that I have ‘‘ omitted the most important character of D. marshalli, viz., the extraordinary length of the rostrum, which reaches the middle of the abdomen.” As I had only an imperfect specimen before me when I wrote my description (I described the imperfect condition of the antennze), | could not describe a mutilated rostrum. However, tew regard Bergroth’s animadversions too seriously. ddd. Hab. Mashonaland ; Sal'sbury (Marshall). Mozam- bique; Bazi River, Zululand (Bell-Marley and Warren). Transvaal ; Lydenburg (Kraniz); Natal; Durban (Bell- Marley )—Brit. Mus. In the above series the length varies from 8 to 11} mm. I have already described species of Dinomachus from the Oriental Region, and I now add another two species from Australia. 488 Mr. W. L. Distant on the Dinomachus .kurande, sp. n. Head black with a basal spot between the ocelli and the apex of the central lobe ochraceous ; pronotum ochra- ceous, somewhat thickly, coarsely, darkly punctate; narrow Jateral and anterior margins, a slender central longitudinal carination, and two similar but oblique carinations on posterior lobe dull ochraceous; scutellum very coarsely darkly punctate, a central longitudinal carination on pos- terior half, which apically bifurcates on each side, ochraceous; corium ochraceous, thickly, coarsely, darkly punctate, the lateral margins very narrowly ochraceous, apical angles ochraceous with a small black spot; membrane bronzy brown ; body beneath imperfectly seen in carded type; legs very pale ochraceous, subapical areas of the femora and annulations to the tibie and tarsi castaneous ; antenne pale ochraceous, apex of the second joint and nearly the whole of the third and fourth joints pale brownish, second joint much the longest, third and fourth joints almost sub- equal in length, first joint distinctly passing apex of head ; rostrum imperfectly seen in carded type. Long. 7 mm. Hab. Queensland; Kuranda (F. P. Doda). Dinomachus doddi, sp. n. Head castaneous, coarsely punctate, apex of central lobe and a central longitudinal line between ocelli ochraceous ; pronotum ochraceous, somewhat darkly punctate, a broad, subanterior, transverse fascia, two central longitudinal spots at base, and a submarginal ]ine on posterior lobe castaneous; Seuhe thant castaneous, coarsely punctate, a central longi- tudinal carinate line obliquely branching on each side of apex castaneous ; corium ochraceous, coarsely punctate, its extreme apical margin piceous ; membrane pale bronzy ; body beneath castaneous; rostrum, coxe, legs, disk, apex and segmental marginal spots to abdomen beneath ochra- ceous ; rostrum about reaching the intermediate coxe; sternum very coarsely punctate ; antennze ochraceous, apices of the first, second, and third joints and nearly the whole of fourth joint pale castaneous, second joint longest, third a little longer than fourth ; pronotum with a central longi- tudinal carinate line and with the subanterior transverse fascia slightly globose and very sparingly punctate. Long. 8 mm. Hab. Queensland ; Kuranda (#. P. Dodd). Rhynchotal Family Why eecide. 489 Masoas transvaaliensis, Dist. Ann. & Mag. Nat. Hist. (7) xviii. p. 290 (1906). The type of this species was from the Transvaal (Pretoria) ; the Brit. Mus. now contains two other specimens from Angola which are slightly larger, measuring in length 4;mm. The type has only a dimension of 3} mm. Oxycarenus collaris, Muls. & Rey. Aun. Soc. Lin. Lyon, 1852, p. 102 ; Oshan, Verz. Pal. Hem. Bd. 1, Heteropt. p- 300 (1906). This Palearctic species, as hitherto understood, must now be also included in the Oriental fauna, as the British Museum has recently received specimens from the Agricultural Col- lege, Poona. It was found “infesting in large numbers the capsules of the safflower plant grown in Poona” (Harold Mann). Maruthas bicolor. Maruthas bicolor, Dist. Nov. Caledon. 1, L. iv. p. 379, pl. xi. fig. 5 (1914). Oxycarenus bicoloratus, Bergr. Phil. Journ. Sci. xiii. p. 73 (1918). Hab. New Caledonia.° Clerada apicicornis, Sign. in Maillard, Notes sur l’Ile de la Réunion, Ins. p. 28, pl. xx. fig. 8 (1862). This very widely distributed species can now be recorded from Queensland ; Kuranda (F. P. Dodd). Pamera tricolorata, sp. n. Head, pronotum, and scutellum black; corium dark cas- taneous ; apex of scutellum and lateral marginal area of ecorium to beyond middle ochraceous, on apical area of corium two pale ochraceous or greyish spots in transverse series, in some specimens these spots are united and in others they are practically absent; membrane brownish ochraceous ; body beieath and legs black ; apices of femora, basal areas of intermediate and posterior femora, and the whole of the tibize and tarsi ochraceous ; antennz piceous, second joint paler, fourth joint with basal half pale ochra- ceous, second joint a little longest, third and fourth almost subequal in length; auterior lobe of pronotum with a distinct anterior collar, convex, a little longer than posterior lobe but narrower, the posterior lobe somewhat coarsely punctate; seutellum centially longitudinally carinate, the carination 49() Mr. W. L. Distant o2 the bifureate towards, base; corium, excluding lateral marginal area, more or less thickly punctate ; membrane not passinz abdominal .apex ; rostrum reaching or slightly passing anterior Coxe. Long. 6—7 mm. Hab. Queensland; Kuranda Gite sa R. E. Turner ; April, F. P. Dodd). Adelaide River (J. J. Walker). Tenim- ee Island: (W. Doherty). Pamera vincta, Say. This very widely distributed species has now been received from Queensland (Townsville), where it was taken by Mr. F. P. Dodd. AUSTROPAME RA, GEN. NOV. Head long, anteocular portion abeut as long as postocular, but the anteocular portion acuminately apically produced ; eyes moderately prominent ; ocelli situate ‘a little behind a line between the posterior margins of the eyes; antenne inserted a little in front of eyes, first joint about as long as head, second longest; pronotum with a narrow anterior collar about as long as broad at base, strongly laterally sinuate, the anterior lobe subglobose and shorter than the posterior lobe; ro s j= a ee A. W. Waters. del. MEDITERRANEAN BRYOZOA. gS GV Ole sat é > f ; G ) } : L / : , } an 4 “eS =, “ | * 8% Sq Ops ad Ve Ke 2) 2H, aes ase er eres ie - (= t i ah 2 ee L ; | ) | fe ey Ann, & Mag. Nat. Hist. S. 9. Vol. 1. Pl. XT. FISHER. Type of Asterina coronata cristata (Fisher). BAYLIS. Ann. & Mag. Nat. Hist. S. 9. Vol, 11. Pl. XIV. Dicrocelium lanceatum, var. symmetricum. GUDGER. Ann. 5 Mag. Nat. Hist. S. 0. Fig. 1. yD TS KY Sul. wy, : SESE Vx 4 Nite ae pr bf SN ree Be coe aout: poate aw tor nie ate yg oa pe GUDGER, Ann. § Mag. Nat. Hist. 8.9. Vol. IT Pl. XVIs Fre. 5. eam LUE ee = Mm ST — A | S. Tv ' TH] A in) vy \ oo / 6 gr eto! f30:4.4 905s papi tetee Nip peet rere ce ae 4 Bree ne ‘He ; s ; Pee” he by, oarg Wa arto, 4 —, rnnsersenonanteseet A peer bbe p 04, Li « MAay “Ur a ) € ern, lls | phar apPappp wy lt Sel Ie agpane? Ce ~Fr ile lal ae: A sty Pap» tte nei WA “Ramee as TLbbel ts i f AT roy Pe Unng Ty yy eile eis rhe \ ap at HA. a > at’ on. Sagal: dad dN Qe ’ vs The “fa, ap Ne-sas, arp WRT ARTY prea COP tat a anny . Ni ln'al i flys : OTRAS mr, rane, an anne An-. waa ty fT da dadaasoaelineny pAar-= PET ORR, Sr RAMI iil TiduailT pamcat Lilt cette SS aia a scan RRS Aaa \ Portes acttnnet wih man os OO aRAA” a aannemanarya , - 4 Waa, Att h sh PRP ange sa nae LUD notin woah Pm,» : : LpaPate.o, "Ye y WA 6 ry y ee ry an siren tee meprnnd® IN ‘| nay Re Ll gyrrenmonnnnenarnasneet Salter: ~~ Gg. ian j | AER e soe a go TEE dH PALARLE iT} | eartcitsantis,. Pee Aas YY Pe oe A Shar pan OM VARAAT RAC cell f pantent ARe if itd Nall oP hr BOC nA Gabi EAA wWNn Cate Np Nica - “ is BY Vs AP rp mat pel Ty | ete GF Wopaas tare sMine tense ett tO aay Manatnan Aan nat ac RAN = 9.0.0, TUPRORRRD espana cee eenneaad PAneatnproneanapa a as eSa 2 2 2 i ie ASW SR, . BACAR Adan art Pty ANP ORIRR Siainicer anh SAARI aapaat monnaanhasatat Ala ARMA 4 yrA\nha . eh 7 esaanale ‘a of fhopedely ces RAR RRR TON ROO Thre TT TS F a AML staan 66 ney atts ed May, 28 i PaManbirater MAAR PPA asain A i NAP a Clas ma i: 2. Yang iy PT a BER TEA ey ‘ pea’ ea feat ay a seaMaiLenis PRL iM: om =a | an i} “VEO hel al af ak al al uh eee ees) Pe. «ag mn Tae ara’ oe wr igai er P14 “re y Ta | A Add Ve rhesseud Wideideeooass ey en) TNA Ada ae oe ey r, . fl MATT nat the... ''\ Mol ‘ce ay ioe | ||! a! anne | pon. gt a 4 A co y bm % a 4a, Sie £3 Ms mp Wid hha ser -. 2 sna ret! | r VFAg y? Prae r arth ae ee Sastulus 4 Nee My, Na " yet titon ahi Nraahs TUN pil aii aT Ne Sr wh =s << \ AB a { anaes 1ae, Bol . Mian, ge'4 Pitty _) Beeeee por gee“. rer AY ahd Bi e r! ye i a Salp MMppe rr uPhree? ve pear ‘dibe,> P : Be en r Ww be rx) ¢ & ef r ~~ mn ile aa { habit isla ANA Pps EO Oke enn of Vid . =. 4 J 4 ° yv “hate, oe .4 fe Se | ae ABA A AAR nee ¢s ees prin an. ‘\ . 4 ae | yo | HL {{ een. tal a aa ‘ Eee Mia aaa aha dt ter: renal’ ony ‘ceecaee ae AMARA VNR, y RA 64e.,"8 he yas real es all Marta Aen a (Yyy aes ee Riviere insee AP +, t, oe dae ta. Lal N “An, ay yt] Ay: Hp, ryt Le) AY Pra pe s Wes Van site ere Spm - ; | So ee BRD uM Qyy |, pape rN > a p/\p. > i ‘A aed a ¥ _ ~s Nee %y ely We TP’ Nears, itt ypenr ADA “ir Nopape antPat itt nas ee Vay, ae a. a. PTT | Ai & oy v Nias és Benen he be i ugg an Ny, Gor ea, Vipv Aa 1 wd - fs | mens SA nee > pie nant Ph Fe EOS - Pee orAre ; PA APRP MA, ADA vey se ul Labs. | eeeenMpapuan... 1 A . ening orn A Aili sang g A mr, a6 es ARe freer a Mn mane | ati Bead dalaier ts Maa-a © POUR TBRAIgAAne : papriprrececcnnine 5 Op naee nhs ah Ny fle SrA ne Vela ol ala 4 ee mhan” ar. Vv pprasata” PA rutrce? | ‘ gaan 8A ORBARAA a-Rexeu i oy & mae f AA ahawcal a al

rat meeeer ee so ee é Per ote wearers ibd. A Bide ne SADDAVANADRRB LD SERED YD 2 eh BAA Dw a Bata OR eae aa b eRe noe ; eee | othe ee et RAG LE AROS OS ee ee ee oe et FREAD RAS Ne PTO RS MAD Am 3 Rah: POR et Ah SRA KG BRAG RRA MP VRRS © 2A: yh ie ANE A Moth 6% ME a Bat : H.R sol a tah, ald inet pty 8 ot Plath tyes > Pies pete ta Rvye re tr ar Lakes ed Oh IRSAAD A eee erie) BR HGRA Rao AGRA S ery | eery SOR RAR Bada. at4 : +549 bE. M54 925 250 8904 8a b-e 342k 4aNs 4s iene tt et) Stab oe ane “4 2s eA, ‘ Dhb,0, $A 4) : frat A0, soe, ae eer SP Hidve. MIME 3 9088 01314 0793