‘ j i j wees 7 ‘ 1 § ye i “ ‘ ewe tet Ht : , i) Me hi Wi, id 1 ' 1 ry DS a Bult Fat Pay Wr er Vt ee 5,’ 949 ‘ ' : yi i : Platelet ds Woe 1 seats Oy parties speeder ea! a oe bif babes f j ut y } Ail tS Le A? iva fies {+ treme ied vi NW teTET Bec LE ete kk Peete ene eee Sele AW aieepitt ts actay rel { we ‘ i bape ‘ [ i adel Pave wate merry i ey ek PUR Ue Pe ‘ Fee ee vere kama ta Fr Heil settee lhe ee ' ‘ | " te “it f i : ] a wt meet dash Wy eth bde checks he Oe MAS Edema te Get te ting Prerer aie Mi Co were | dehy re v9} Aenearmiaed uty ' ‘ FH ‘ it ‘ a aretrhn | 4 waeta Pare rer ter erin Wier Grint het on ChE PD by Hataine te deers afr: S08 Merre rte Ei HoAemafia \ tut hail ' wy ' \ ' ‘ - Arent | Lad ' ' Pe VLE EG ied oh) uF ip T ;) Frew Wire Tn ou! Le see b ita tet ete rey “yi aradal 44 j Ww i aus : tie ty alas lade Rel, ween ah t he ee t ' 4 ti} Le i ) a) ; ‘ : vi 1 Weel a aed Hi tes ie rh Pe ereerrer re Tere a ty 49 jwsedgeaed a) t a! eye ra 4.0t ‘ \ ' ‘ f | : TAR “7 reer : fl ey ti 7 1 : we Sa harp rer ' 4 , ' ‘ hid ' ! ) J ’ 4,4 a4 n ‘ ubae bd a etait itaait ‘ a 4 “r Gtedetnd sear veut ‘ avevo? ip fit i : ’ ; t J java at Paver | ria a a iw ete a ‘4 ' ‘ (iit ja " ‘ a ta f ANY wa ' vA haw Ta Heawelave «(het 1edaes ian A et Aas ats Wee O98) \ i , ‘ : ’ 1% WV 4 vida he elay Guat) Lar bee Pee tirade ‘ Bi evaere ty kk ¢ aL 1 bey he (sil tek HER SA ad Wt 4 unead4 ‘ i : ‘4 ? 4 } ty aha % ; ei) i } ya | \ | it ote Pert eer Teen IG aves \etote Unrasemnangnaeonings Wty ; . ‘ ran nr] pris ‘ i A \ Weer! spuiaet waned Ga)! PER de hide 3, F . UF f biel Yel teases ? MERRIE TEL dae tea tedster te my fi ea i ' Wine : ‘ ‘ ‘ if ' ‘ ; And dat Bis 94 fone ner sgl gee Vile we teeet ‘ ‘ ' tied be } wr i : err raye er urt Pat at i i ithias Mit ENT Mab vse Hawes ant i. An , H ] h MEH IK Hern Te ATEER LITT Vite ni Fepa de tah Gada ‘Erinn bigvueieaeet ' ' 14 si ’ 4 yd a { v *} i \ 1a) a ish ‘el aaa AS i} TUR ese ba i Hit.) * i Pr Bh ta ‘ ¥) 5 J ‘ ih ed bi (aie lags aun tlepate te delet el eee ett Ay ite eek aay, ae eve r ait : wild ‘ ' 43 4 iM 0s Ve ig ten dati petit Na Fa ie ang beastee: hi Raa set "s ‘ ai aint 4) Wa ’ aA) Aaiheds a sisi | LACH abe titel yenerssy ae te 44 ; eg ae ey ' oe" tefilys lad Pree ya | satel te f ; mii 1g Mire kl aye " Ss rarernsd ae F iy Li , P ‘" a é Fee eee Wis Mh Asibe 1) tH em eld HiU ra onde ee a Cai Bia de gs mr rant eter Tun Leh. OC ia As bi ised aand da At ty er tbet ‘dak a yet . iY andes La. Le tee aang yey oh quinighantsnid inate shined WC edsag ae Mee eg AT ANT igs eh ai gar a: 7 ai wite een eed et ! ani deed ghey nit Eta m tedede ty WAC day? It ” be Be peal inate if ‘ aa ile iq Tesvay study annie varesneninad nig te ie ‘EL + ‘ al } } ACU ‘al mere tty spits Py een on 4 nee ae ya a CU ih ace the meugreereet nga qe a) y ; Nee A tat GMa dine i we geder Vie) WN iN tate \ ' ta) j ! « 4 iv’ . 'f 4 : tebes ” bt it 0 ibe ) ’ vy t mt ‘ Paes AD | ae { in ag ue v HN rn" sti diag Weis Oe Nia tate hea ta “N st ARE LRilc ¢ try a ‘ “it iT. ace 4 ; i ; : : teAy wo taw . ’ ’ “ . e, ‘e en i ! ‘ piile : ' 1 ath ’ ta HY Watane rye pred vant VW iP roe Meare 9 eyyee.5 “he ts LF om ; Rt i ; wana oo LY tr ae stad ter Nieutyr4 vhs ga Tit pate ' } Nee a | Vas als ae Hoy | sHisofer wae salsatey : , ! |. 144) ) “ 9 i? } What ey we Ni 4 eden etete baby ‘ty Uae eke P ins ad gar iot a hes \} pyaa | aa Wey) \f ah tr Rritaty Tey Oe pe detatads ledate ey we OU Het ye! A au , . ¥ “ yh re j phi PAPE TE TERY i’ lenin A Who) haanee ee magna ed bd CA “ue Mtr Pi 44 ‘ar aae : Wu! ‘ ome tedden! 1G 0G fee Mh al i pad dns i 1s : he ri a ‘ ; r ait ate ih, \; pindeae ‘ Wedee ' Handi vp ape 40 tly Hah j “4 “nf a “id ‘ t Asti) ‘ ie ely Yate Oe be Wad eA hore ye behets yt ‘ ,” ; tele S52 4 ny M Date} A! re q . ia \ i NAM Sad f fetes’ i pidheieie BN ge Be yey 3) | Ved } a Tee te ane ae ean ay othe nb ain aoe (yet eka ; ' } ‘ , “i Pe VEMTEPON BTUti ail ieee LL 4 aha ryerrry ‘ ty id Ky Pe 3 ‘ amg ty Gene ‘ mesh . periatn Rei hath bias \ ‘ ba yAVMV AY eae, ‘: vali rete abut fie seedy dep he “aia i | iL y ‘ gam Reet’ ‘ \4 Fa Vy , ’ f hy nett is aateasas tie apa sina a Hae i wut} ak | it A elit Gale ia detees fa Waletn ie oy 10) fl 7 fe ) ie i! ‘ ya eel i pany Pha yy la tems dys ‘) dere fh pt aa elapse $4) tiie } ¥ ae ‘ LT Ate backs (to A ya gote ty MW heyy ia tebe taetad tera Wee vy Hy iegit i . vn 1 ‘ "4 Tae | | : TNs plidedetaat ag i anes sete rey { a ' j J +) we 1 te 1 Waa ‘ ' rf Oa Pen Oi viehyes jatiy ee Ne : ' ' , | raat tinea e ane fede abe Wak Pr ‘ ( ° aa" eta Gs ceded Oey ee aa tenes de abe x P ya nig + ye P| Ardy AA i way i iiel See Se a hags! q ¢ ; i ativit We J ‘ * soliaaha’ i We 1 Ay . ; ) ‘ ae ae CHIR Ct Ry . mney 4 at r ay, tala sete Mas " j site . : wit ale} Wh) aly Masia hdl eit yy ¢ 1 NASH i ) ; i ‘ ‘ 4" { Dalek ola SN) nha asa teeat TW wae Aid tea | ay wii be wk ay tie min 1 9 of of . ! Hag \ Va ‘ hp Ape Poa, Hite hh Hd aM bts ay \ ity Myvi ese aedep. tos lopaaedemean 4 acgvan (198 Titctit Sepanitraeea ' ’ at vy vel | P bs 1 pa taeee 1 16a a dee OA iu Vahl te! it iene a NW f : ura if ati tan ‘ a atta gate ye lege Oi Re in i Ye Wy yay }3) OE ’ a ’ eds Ve ke by 4 erate REM Re 4 ‘ +8 oA OL, Velivis pat eatin git ante 4) PRR deg ’ ie eh hile 4 Ay " a ycd raege eh hh ba , ‘ ' ) ' Hud } 4 4 4 6 Alina My . Qasi erty ver H ektaetats qu Wrath) ‘y i aly i ‘ at VIL ey ! Beda Laat Pell Ae Nakitsites ) aw ; ee) vile «i SedeQem at! - “f } ’ { Piha te j Nj tigi eset dirham Wie ae ' ’ ; j Voda ns) " Nhs tome ME mH iat ay “he . War a) Dy eh a 4 Pry Peep usahl Ne aut way Mt ‘iy ! \ ‘ wieY de t) inf Oa fie ee shi LF : Ch ie 1 he any ij Agate inh Teglle } a } ' 1 wil ir ME nah let EACLAL Uk i Ase? ’ qa tat Pamaatadr enti te kerf ive iay Ree van nat Hits ot i“ : ‘ Lipa i eats ne! ioean Fh ‘ y ‘ { ' iy ut ye i dat ri hee +h nin ; MP entree the sit 7 } ef Adaya Ahr 1f , , j Wie hy at Nt aut ey : ’ Mas “i gun ’ ’ 1 4 ‘ \ | ole t ; i } ; ’ ; iF 2) 1 r ; ; } } h ' i " ' { ir) V i N mt | Ma Ra ; b ' ; i . , : iy ' ria ama ee ) j ' eee \ its Sy Tule sepe Nt: ; may eine! Hehe si dstavg! . j f h b) to ek 1 $i) , Lal , ‘Yt ' 4 j p ' yf J " iy \ val’ ayy i ' ae) ‘ ; johet ’ } pe iy $055 }, Loa lat . eet: +y ee H i 1 Aide f ey P ‘ ‘ dl ’ ¢ . ‘ i eT ’ } aa} ; t La ‘ ; aie nie WAAR ht ue ! ‘ Hay yeni G ‘| Biya Tren trey Uae ae st isily Niet Wei vis nd Pee is tive ty dad Prine Peron, eee 4 Wi ihatde: ena a fate iedh H aed Hoe it td ¢ Ay Sod Raider Hat tin ay UTS Wid Ne MRM ha nai (eek doe wed ht ere Aiea te ae \ataabraaid! prune ibe Guay disc teey rrnieT Pore. tk ane Perri ity prigent ide ey Bete 1 del anche tly: Tie as a Att fe ratio da a i PUL Gel Naqedeend ; : ae cing tah Ht NOs wis O Falniel v\ a Be aap Pay pratt Lidtde epantiatesls Pure rarate uk auth FI dys ba 1D) ' i) aerun ATT as Was Hea hk Gi) et ae te utile it aS bee aed a, dete NL taste Brey Cod are La Atta dethian § G SOUT RT EC Fei Coed Pn e) PAC bide be a Cert ae! i 14 oem dsal iv NYP Ate ot cee ey i ieee ithe isd he xf } HiGt ts ee 1 binned y de) sare i saad tein " 11} ated rr it, Hii gaa litea SO ae Fee aa eg ae midis pe iL +f rat Darth Wigs dada tae ts Aoitanly lost imeaa Gide ri rents 7 ees " i bap bags handed ead Past « it My Titty ited , agit errata pigede sabes ‘ deat ire Dike aap ae teaahede bt ee Wily as Vu WINDS as Use a lage hed vo we Ah) 48 ahs Vata! Pees fa) Hd hy Atiey ¥i Liha ih ae aia Lpedeate ay) 1 Hu jretl i] OTe TIP iG Ete eels ners LACHER CHEICE RIESE Wied ians wb ats nl Deb ptededaded Rat UN te aA erat batt e Jigs Videds sabe ha hadieedt TAL Co) ee ve ledon ee 4 poke rhe voce bet Hear Sia viee eird a gebna H Pon seewntee saree are iy ie eye vn enh 20 ed oP Oe ‘neat bese iger et ee ea apa dae et mrrenrrey HTC HEL ak oe. oe ay eae aa “cae ‘ie ne see ‘asada qt Rivied Hird adees ai Hat ii ih, Gene a OTC Phe ‘ a is it ert ane bade Cate ted nee! a PHN Shai bse pinueds sp edad bhp lee faded cies ea be ide Ae tice (dba beta Sei Mii (hewn ns ‘ ae wit dane fay Pers rey hs Sy LD ae PT Hae oh Th i. ‘ ier ied ited: aesenty BT : ded, ay rye au! #0) Pitas nd Aaa te Dodds Leite Ath PG he taake beth a ae ae YY AAAAAAAAAAASAASAAAAAAAAAAAAAAM? - saaeAiloaunanniesaay BAA AARAAA R AREA annnnnannannns ANAR Ae ARANAARY Lay AN RAIA. an i A AANA AA | ee aA eee AARARAAAAALY A. | l AY : AaAAAARARAABAD YY) |! Aa gar AAAAAAAAAAAAAAARARARAMAAAAIARAARRDNAA®AOAA ZS | NAYATAIA VAVAVAAAVAVAIAA lant AABAAAA AIA AK TrYyYV VY. 4 a fp AAaARAARRRARARAA sb Are : ARARAAAAAA CE & 7 COS. S.C CCe@ «Cc ce CQ Se c Cace “Ce CC a ae aaa < Cees ace . ie eee " | SOE af eel | i : oe ack Sie : ; uy | c eee — New and noteworthy mosses from Jamaica, * rae ee with notes on a new Pat 7 Comp _ Fenner A. CHACE, Jf... ... Soa <1 ee 5 ee | Zoouoey. -Nomenclatorial changes involving Calitornia a9 ot This Journal is indexed in the International Index to ae | = | | Feprvary 15, 1936 No, 2 - JOURNAL IGTON ACADEMY F SCIENCES BOARD OF EDITORS Roitanp W. Brown Espen H. Toour Uv. 8. GEOLOGICAL SURVEY BUREAU OF PLANT INDUSTRY ASSOCIATE EDITORS 2 oe 7. WENSEL Harotp Morrison PRILOSOPHICAL BEAST, ENTOMOLOGICAL SOCIETY : E. ‘A. GotpMAN | W. W. Rusey BIOLOGICAL Ss ipa GEOLOGICAL SOCIETY “AGNES CHASE : J. R. SwanrTon 5: BOTANICAL SOCIETY ANTHROPOLOGICAL SOCIETY R. E. Gipson CHEMICAL SOCIETY PUBLISHED MONTHLY BY THE i yi _ WASHINGTON ACADEMY OF SCIENCES 450 AHNAIP Sr. at Munasua, WISCONSIN This JouRNAL, the official organ of the Washington Senden ¢ (1) short original papers, written or communicated by members proceedings and programs of meetings of ‘the Academy and affiliated | so notes of events connected with the scientific life of Washington. The JouRNA monthly, on the fifteenth of each month. Volumes correspond to a cia y Manuscripts may be sent to any member of the Board of Hditaws Iti is 1 quested that contributors consult the latest numbers of this Journat and confor manuscripts to the usage found there as regards arrangement of title, subheads onymies, footnotes, tables, bibliography, legends for illustrations, and other ma Manuscripts should be typewritten, double-spaced, on good paper. Footnotes sh be numbered serially in pencil and submitted on a separate sheet. The editors assume responsibility for the ideas expressed by the a eee nor can they Wi correct other than obvious minor errors. : Illustrations in excess of one full-page halftone or two full-page line drax be paid for by the author. Line drawings are preferred. si es ea Proof.—In order to facilitate prompt publication one proof will eenerdlly to authors in or near Washington. It is urged that manuscript be submitted form; the editors will exercise due care in seeing that copy is followed. _ - : Author’s Reprints.—Fifty reprints without covers will be furnished gratis bearing the name of the author and title of the article, with inclusive ‘pagina date of issue, and reprints, will be furnished at cost when ordered, in | aecorda the following schedule of prices. he Copies 4 pp. 8 pp. 12 pp. 16 pp. - Covers- 50 as < es ihe 22 alae a 100 $75 $1.20 $1.50 — $2.00 Bh Py (ak 150 1 25 1.90 2.05 Si26 ; 3.50 — 200 1 ee 55 2.60 3.60 - 4.50 To A rete 250 2.00 3.30 4.65 6275... 5.00 : Envelopes for mailing reprints with the author’s name aaa address printed corner may be obtained at the following prices. First 100, $4.00; additional 100, ‘ Subscription Rates.—Per volume................ erpaene 55h he este e ; To members of affiliated societies; per volume.............. Fe ak oA: MSS Sinele numbers.) gab Aigo oA SE ee ee ae et Bed Remittances should be made payable to ‘‘Washington Academy of Science addressed to 450 Ahnaip Street, Menasha, Wis., or to the Treasurer, H. G. Coast and Geodetic shan i Washington, ey Ee of Exchanges.—The JouRNAL does not exchange with other publications. Missing Numbers will be.replaced without charge provided that élaim is is m /d the Pelcg ad within thirty days after date of following i issue. foo gummed OFFICERS OF THE ACADEMY ae : President: O. E. Mernzer, U. S. Geological Survey. Corresponding Secretary: N. R. Smiru, Bureau of Plant eden: Recording Secretary: CHARLES THom, Bureau of eee Bs and Soils. Treasurer: Henry G. AVERs, Coast and Geodetic Survey. JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES Vou. 26 FEBRUARY 15, 1936 No. 2 BOTANY .—Certain Cactaceae of Venezuela: New and old species of Opuntia and Melocactus.1. H. Pirrrer, Caracas, Venezuela. In the course of my botanical work in Venezuela, I have become closely acquainted with some of the Cactaceae which form one of the main elements of the thorn-bushes of this country. Besides, I have now in cultivation and under observation a few doubtful species. The following are the first results of my investigations. Opuntia metuenda Pittier, sp. nov. (Cylindropuntiae-Leptocaules) Horrida, insolite armata, trunco ramisque lignosis, teretibus, decumbenti- bus vel interdum suberectis; areolis parvis cano-villosis, parce vel haud glochidiatis, aculeis 1-3, nudis, porrectis, flavescentibus, uno longiore, ar- matis; articulis teretibus, dissitis, plus minusve alternis, caducissimis; foliis parvis, crassis, obovoideis, acuminatis, cito caducis; areolis articuli cujus- que 22-28, 3—4 inferioribus inermibus aculeatisve vel setis rigidis rubescenti- bus instructis, reliquiis superioribus l-aculeatis, aculeo nudo, longo, acicu- lari, flavescenti, interdum aliisque 1-2 obsoletis adjectis; glochidiis incon- spicuis vel nullis; floribus parvis, luteis; ovario obovato, areolis paucis, glochidiatis vel raro apicem versus 1-2 aculeis instructo; perianthio plus minusve rotato, segmentis exterioribus brevibus latisque, minute apiculatis, interioribus obovato-spathulatis, basi angustis, apice obtusis subacumina- tisve; staminibus numerosis, flavescentibus, filamentis tenuibus, antheris oblongo-linearibus basifixis; stylo flavo, crasso, apicem versus attenuato; stigmate magno, purpureo, 6—-8-lobulato; bacca obovoideo-globosa, matura rubra, areolis fere inermibus, glochidiatis, interdum armatis. Caudices 40-120 cm. longi, 1.2-1.5 cm. diam.; articuli 4-6 cm. longi, 0.8-1 cm. diam. Aculei 1.5-4 cm. longi. Folia 3 mm. longa. Flores plus mi- nusve 3 cm. longi, 1.5—2 cm. diam. Petala interiora 1.5-1.7 cm. longa, 1-1.5 cm. lata. Stamina plus minusve 1.5 cm. longa. Stylus (cum stigmate 3-4 mm. longo) 1.5 em. longus. Bacca 1-1.2 cm. diam. Lara: Savannas around Barquisimeto, June 28, 1913 (Pittier 6415); Barquisimeto, in ravines descending to Rio Turbio, flowers and fruits, Sep- tember 18, 1923 (Pittver 11176, type). The above description was drawn (flower and fruit) from this latter specimen and partly from living plants in my garden obligingly sent from Barquisimeto by Brother Nectario of the La Salle College, to whom I extend my best thanks. From recent information it appears that the species is far from being restricted to the vicinity of Barquisimeto. Dr. A. Jahn reports it as growing abundantly in the hills 1 Received October 24, 1935. 41 42 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 26, NO. 2 that surround Lagunillas (Mérida), at an altitude of about 1000 to 1200 meters. It is also said to be common in arid places through the State of Faleén and in eastern Venezuela around Cumana (Sucre). Thus, the altitudinal range of the species would be from sea level to about 1200 meters. Opuntia metuenda is known locally as guasabar, guasdbara, tuna de guasd- bara (Lara), guasdbana (Sucre). Opuntia bisetosa Pittier, sp. nov. (Platyopuntiae-Elatiores) Planta elata, e basi ramosa; articulis ovato-oblongis obovatisve, crassi- usculis, pallide viridibus; areolis fere 7-seriatis, 6 vel 7 in quaquam serie? (marginalibus exceptis) ; aculeis areolarum juvenilium 5 vel minus, primum lutescentibus demum albidis; aculeis areolarum maturarum 8-10, robustis, albidis, 1 centrali, erecto, longiore, compresso, distorto, 7-9 brevioribus, acicularibus, plus minusve obliquis, inferioribus pluries setis 2, longis, ri- gidis, suffultis; foliis tenuibus, elongatis, subteretibus, subtiliter apiculatis, apice purpurascentibus; floribus modice magnis; ovario obconico, areolis 8-seriatis, 5 quaquam serie, inferioribus glochidiatis setaceisque, superiori- bus 1-3-aculeatis, albo-glochidiatis, aculeis albis, pilis brevibus rufescenti- bus cinctis, basi foliis 1-2 deciduis suffultis; foliis marginalibus 5, brevis- simis, crassiusculis, apice spinulosis; perianthio cupuliformi, segmentis ex- terioribus late obovatis truncato-apiculatis, interioribus obovatis, apice rotundatis emarginatisve, basi virescentibus, apice incarnatis; staminibus numerosis, filamentis pallide roseis, antheris minutis, albis; stylo crasso, apicem versus leviter attenuato, albo; stigmate magno, subgloboso, pallide viridi, 8-lobulato; bacca ovata, omnino saturate rosea, superne aculeata, aculeis tenuibus brevibusque; seminibus albis. Planta, ut videtur, usque ad 1.2 m. alta. Articuli usque ad 36 cm. longi, 18 em. lati. Areolae 3-4 cm. remotae. Aculeus centralis usque ad 5 cm. longus. Folia 4-5 mm. longa. Flores 6.5 em. longi, 5 em. diam.; folia marginalia 5-6 mm. longa; perianthi segmenta interiora 2.8 cm. longa, 1.8 em. lata; stamina circa 1.8 cm. longa; stylus 2-3 cm. longus; stigma 5-7 mm. diam. Bacca 5 ém. longa, 2.2 cm. diam., aculeis tenuibus usque ad 11 mm. This description was drawn from a living plant obtained from a joint col- lected at Sanare, State of Lara, by Mr. Fr. Tamayo and cultivated in my garden. The species seems to be allied to Opuntia wentiana Britt. & Rose, from which it is easily distinguished by the two bristles at the base of the inferior spines in the areoles of the joints. Botanical specimens distributed under Pittier 13578 (type). Opuntia wentiana Britt. & Rose, Cact. 1: 116. 1919 Plant erect, much branched; joints ovate, obovate or elliptic, thickish, whitish green and covered in age with tiny white dots which give them a glaucous appearance; areoles 4 to each oblique row (marginal not included) ; spines on young joints mostly 3, later 4 to 6, to each areole, strong, terete, at first yellowish, turning to light gray; leaves early deciduous, thick, rounded, the apex spinulous, more or less purplish; flowers medium-sized ; ovary obconical with 5 oblique rows of areoles, these 4 to the row and leav- * According to my experience with the Venezuelan Platyopuntiae, the seriate dis- position of the areoles on each face of the joint and the number of areoles in each row are fairly constant and constitute a good specific character. The same can be said of the leaves on the upper margin of the ovary. Fes. 15, 1936 PITTIER: CACTACEAE OF VENEZUELA 43 ing the base of the ovary smooth and spineless; areoles spineless or with 1 minute spine, with a central tuft of white glochids surrounded by a narrow cushion of brown cellular hairs; at the base of the areole a short, thick, apiculate leaf; calycinal leaves 5-6; segments of the perianth broadly obo- vate, the outer ones shorter, more or less greenish, minutely apiculate, the inner ones larger, yellow, often tinged with pink near the apex, slightly emarginate; stamens numerous, the upper part of the filaments and the anthers pale pink; style whitish, its base thick, rounded, pointed at the base and attenuate towards the apex; stigma lobes 5-6, parted, cream colored, forming a subglobose head; fruit rather small, rounded at base, spineless but glochidiate, the umbilic very deep. Joints 20-25 cm. long, 12-13 cm. broad. Leaves 3-5 mm. long. Distance between areoles in row 4-5 cm. Major spine up to 7 cm. long (on the average 4-5 cm.). Flowers 6 cm. long, 6 cm. diam. when fully open. Calycinal leaves 2 mm. long. Inner segments of perianth 3 cm. long, 2.5 em. broad. Stamens 1.8-2 cm. long. Style (including stigma head 7 mm.) 2.7 cm. long. Fruit about 3 em. long, 2.2 cm. diam. But for the leaves, which are not subulate, and the inner segments of the perianth, of a darker yellow (turning to pink in age) and not acute, the above described plant agrees with the incomplete description of Opuntia wentiana as given by Britton and Rose and should at all events be keyed in the same group, which includes O. caracasana, O. wentiana and the new O. bisetosa. From the first one, the flowers and fruits of which are not known, it differs in the habit, the larger joints, the thick, turgid leaves and the spines, longer and mostly more numerous. The above description was made from living specimens in my garden, ob- tained from a joint brought from El Tocuyo, State of Lara, by Mr. Fr. Tamayo. Herbarium specimens are being distributed under P2ttzer 13577. Melocactus amoenus (Hoffmannsegg) Pfeiffer, Enum. Cact. 43. 1837 Melocactus caesitus Wendland in Miquel, Nov. Act. Nat. Cur. 18, Suppl. 1: 184. 1841. Melocactus griseus Wendland, 1. ¢., p. 185. Cactus amoenus Hoffmannsegg, in Preiss, Verzeich. ed. 7, 22. 1833. In their standard work on Cactaceae, Britton and Rose give brief descrip- tions, under the names Cactus amoenus and C. caesius, of two supposedly distinct species of Melocactus, the first growing on the northern coast of Colombia, the other on the coast of Colombia and Venezuela. A mere com- parison of the meager descriptions of these two species would incline one to the belief that they are identical. The given differentiating character “spines curved”’ and ‘‘spines straight’’ is too vague and inconstant, since perfect specimens of the Venezuelan M. caesius have always more or less curved spines. Britton and Rose give Colombia as the type locality of Melocactus amoenus but in the second edition of Férster’s Handbuch der Kacteenkunde we read? 3 Carl Friedrich Férster’s Handbuch der Kacteenkunde in ihrem ganzen Umfange, . . von Theodor Riimpler. Zweite ganzlich umgearbeitete Auflage, pp. 425-426. 1886. 44. JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 26, No. 2 [translated]: ‘‘Fatherland Venezuela, where Hd. Otto found it up to an alti- tude of 1600 m. in the vicinity of La Guaira, growing on a red clayey soil among Agave, large columnar Cerez and Opuntiae. It existed there in consider- able numbers and in all possible forms and sizes.’ Continuing, the Hand- buch gives a description of the plant, which is wanting only in the characters of the fruit and seeds, and which applies accurately to most of the specimens I have handled: ‘‘Body conical-depressed, at first globose-depressed, gray- ish green. Ribs 10-12 [10-15] obtuse, not very salient; areoles white [to gray], [2.5 cm.] distant, sunken, but at first roundish; spines straightish, stiff, subulate, spreading, at first reddish, later dark brown; radial spines 8, the upper ones [usually shorter, 1 cm. and up] very short, the inferior ones very long [up to 3 cm.]; central spine solitary, erect, almost always absent on young individuals. Cephalium vaulted, whitish [with red bristles and spines]. Flowers in July, [1.7 to] 2.5 em. in diameter when fully open, which happens in the afternoon only, the segments of the perianth pinkish red, long-lineal, spreading. [Fruit about 3 em. long, obclavate, red]. One of the prettiest and most freely blooming species. Reaches a height of 15 to 18 em., with a diameter slightly larger. The extirpation of the cephalium results sometimes in the production of new buds.”’ The additions between [ ] are mine and complete the description. Further on we have a description of Melocactus caesius, which has 10 ribs, instead of 10 to 12, on a globose-depressed body, the areoles grayish, 2.5 cm. distant, the spines strong, stiff, straightish, pale reddish, the radial ones 8, spreading, the central one shorter. The flowers completely developed measure 15 to 18 mm. in diameter, the pinkish red segments of the perianth are lineal, and obtuse or emarginate at the apex, the stamens and the 7 lobes of the stigma yellow. According to these descriptions, much more accurate than the ones of Rose and Britton, the two species would differ in the shape of the adult body, the number of the ribs, the color of the areoles, the length of the central spines and the diameter of the flowers, all characters very variable in any species. Moreover the Handbuch, after describing a variety griseus of M. caesius, with 15 ribs, adds that in all probability M. caesius and M. griseus are only forms of M. amoenus. My own observations fully confirm this last view. In a single station, in a group of a large number of individuals of the same ancestry, all the types included under the descriptions of M. amoenus, M. caesius and M. griseus were represented. The form of the body is far from constant, the number of ribs varies from 10 to 15, the color of the areoles is either gray or white, the spines are seldom straight, the radial ones almost always 8 and of vari- able length. The color of the flowers is almost uniform but their size is re- duced in the poorer forms, in which also the spines do not attain their full development. Fes. 15, 1936 BROWN: TEMPSKYA 45 For these reasons, I think that Melocactus caesius and M. griseus should be considered as synonymous with M. amoenus, which is the oldest name. It is even doubtful whether the forms described under the two first names ought to be considered as fixed varieties. The type of M. amoenus was probably from the vicinity of La Guaira, Venezuela, and not from Colombia. The original description dates from 1833, that is from a time when the secession of Venezuela from Great Colombia in 1830 had hardly become generally known. There is no mention of any bot- anist or collector having visited the arid parts of the coast of Colombia in the few years previous to the first publication of the species, while in Caracas several persons, stimulated by the example of the illustrious Vargas, were interested in the flora of the surrounding region and may have sent to Europe samples of such striking plants as the one in question. Melocactus amoenus appears almost everywhere in the arid parts of the coast of northern South America, including the islands, between Trin- idad and Santa Marta. It is very doubtful whether the specimens photo- graphed by Dr. Stahel in the interior of Dutch Guiana belong here. The cephalium of these plants, as it appears in the picture published in Brit- ton and Rose (Cactaceae 3: 234) is much larger and more prominent than in our species. I also think that the upper altitude of 1600 m. cited in the translation from Foérster’s Handbuch is exaggerated. To my knowledge, the species does not appear above 1000 m. and reaches its fullest develop- ment in the warmest stations, not far distant from the seashore. The plant also appears scattered in the semi-desert districts around Barquisimeto, El] Tocuyo and Carora, in the State of Lara. PALEOBOTAN Y.—Frield identification of the fossil ferns called Temp- skya.! RoLtanp W. Brown, U. 8. Geological Survey. Tempskya is a genus of fossil ferns whose remains, so far as now known and understood, are found in place exclusively in strata of Cretaceous age. Unlike the attractive black impressions of the fronds of Carboniferous ferns and seed-ferns commonly illustrated in text- books and exhibited in museums, specimens of T’empskya are rough, irregularly cylindrical or conical blocks (Fig. 5) resembling some oc- currences of fossil wood. The foliage of ferns, it may be noted, has been found in the same or contemporaneous strata as the Tempskya trunks, but the two have never been found in direct connection and consequently, no definite correlation between foliage and stems has yet been made. Because Tempskya has considerable value as a 1 Published by permission of the Director, U. S. Geological Survey. Received October 19, 1935. Helpful suggestions in the preparation of this paper are acknowl- edged to my colleague, Charles B. Read. 46 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES’ VOL. 26, No. 2 stratigraphic indicator, and because collections made in the past few years have revealed the fact of its occurrence at numerous Cretaceous localities in the western United States, it is deemed appropriate now to emphasize the importance of looking for and collecting specimens of this genus, and, as an aid to the collector, to show how Tempskya may be recognized in the field and distinguished from other fossil plant materials with which it might be confused. The first American specimens of Tempskya to be described were collected from the Patapsco (Lower Cretaceous) formation exposed in the valleys of Stony Run’‘and Deep Run, north of Severn, Anne Arundel County, Maryland, and were named 7. white: by E. W. Berry (1) in honor of Dr. David White, of the U. 8. Geological Survey. Although the cell structure in these specimens is silicified, it is poorly preserved, but enough is present to substantiate clearly the reference to the genus, which was first defined from European material. The original European tempskyas came from a number of horizons and localities. The first to be described were from Wealden (Lower Cretaceous) deposits in Tilgate Forest and the beach near Eccles- bourne, Sussex, England. They were called ‘‘arborescent ferns,’’ and later, Endogenites erosa, implying affinity with monocotyledons, pos- sibly palms. The name Tempskya, in honor of Friedrich Tempsky, the discoverer of several specimens in gravels along the River Elbe, east of Neupaka in Bohemia, Czechoslovakia, was given by Corda (2), who described four species in 1845, making 7’. pulchra the nominal type of the genus. However, the first fairly complete generic char- acterization was published by Kidston and Gwynne-Vaughan (3) in 1911, for it was based upon an excellently preserved specimen, which they called T. rossica, collected from Tertiary gravels (most probably derived from the erosion of nearby Upper Cretaceous deposits) in the basin of the Karaganda River on the west flank of the Mugodjar Mountains, Russia. Tempskyas have also been recorded from the Wealden in northern France and near Hanover, Germany; and from the Perucer beds (Cenomanian, Upper Cretaceous) near Lana, 35 kilometers west of Prague, Czechoslovakia. In 1908, A. C. Silberling, of Progress, Mont., found a specimen of Tempskya, 33.5 cm. long, in sediments then thought to be of Kootenai (Lower Cretaceous) age, but now known to be of lower Colorado (Upper Cretaceous) age, in the Musselshell Valley about 10 miles southeast of Harlowton, Mont. This well-preserved specimen was not described until 1924, when Dr. A. C. Seward (4), of Cambridge, Fes. 15, 1936 BROWN: TEMPSKYA 47 England, named it 7. knowltonz in honor of Dr. F. H. Knowlton, of the U. 8. Geological Survey. C. H. Wegemann, of the U. 8. Geological Survey, in 1910, collected a specimen from a sandstone in the Thermopolis shale (Upper Cre- taceous), 25 feet below the Mowry shale, in sec. 28, T. 41 N., R. 81 W., near Kaycee, Wyo. Although it posseses a number of well- defined stems embedded in a groundmass of roots, this specimen was not recognized as Tempskya, but was reported as ‘‘rootlets’”’ on page 64 of U. 8S. Geological Survey Bulletin 471, 1912. In recent years during the course of field work various members of the U. S. Geological Survey have collected additional characteristic material in the Cretaceous terrains of Idaho, Wyoming, Utah, and Nevada. The Idaho specimens were collected by G. R. Mansfield, W. W. Rubey, and J. 8. Williams, and were found at two horizons— the weathered uppermost beds of the Wayan formation, of Colorado age, in the Lanes Creek quadrangle to the southeast of Grays Lake; and an especially prolific locality in the Aspen shale, also of Colorado age, in sec. 19, T. 5 N., R. 44 E., of the Teton Basin. Most of the Wyoming specimens were collected by W. W. Rubey and J. 8. Wil- liams from the Aspen shale, in the northeast corner of the Afton quadrangle; but one water-worn specimen was found by the writer in a gravel terrace resting upon the Green River formation in the valley of Twin Creek, near Fossil. The only specimen from Utah was collected by Frank L. Hess, of the U. 8S. Bureau of Mines, from Upper Cretaceous deposits near the head of Pack Creek just west of Mount Peale in the La Sal Mountains. The Nevada specimens were found by W. W. Rubey and Eugene Callaghan in the lower part of the Overton fanglomerate about 6 miles southwest of Kaolin, Clark County. The material in the recent collections makes it possible to differ- entiate several new species, but the description, illustration, and naming of these species, together with a precise statement of their stratigraphic implications, is reserved for more elaborate treatment than is possible here. As an illustration of the stratigraphic signifi- cance of Tempskya, however, it may be stated that the Nevada specimens collected by Rubey and Callaghan have served, in con- junction with other plant fossils, to demonstrate satisfactorily that the portion of the Overton fanglomerate containing them is Creta- ceous in age, not Miocene (?) as now designated (5). Because the plant and animal fossils found in many continental sediments are, at best, 48 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 26, NO. 2 so fragmentary, scarce, and lacking in diagnostic features that, with the limitations of our present knowledge, they are of little value as stratigraphic indicators, it becomes a matter of some importance to search for and collect such outwardly unprepossessing objects as Tempskya trunks and other petrifactions, which have well-preserved internal structures and can be identified with certainty as belonging within a given geologic range. The aspect of Tempskya specimens depends somewhat upon the characteristics of the formation from which they originate. In external color or tone they may differ considerably, according to the weather- ing they have undergone, but internally, when well-preserved, they are usually dark. In size they may be as small as the specimen il- lustrated in Figure 5, or they may be a foot in diameter and several feet long. Their shape is generally roughly cylindrical or conical, with a surface that is irregularly corrugated, ridged, or furrowed, small holes sometimes terminating the broken ends of the curved rounded ridges. The transverse surface (see top of Fig. 5), especially when wet, usually reveals a number of scattered, irregularly circular or lobed bodies, from 0.5 to 1 or more centimeters in diameter, enclosing minute horseshoe-shaped figures and surrounded by a groundmass of thousands of closely packed, small, circular or elliptic structures. The former are cross-sections of a dichotomously branched stem with leaf traces; and the latter are sections of adventitious roots, forming a mat or felt around the stems. The whole constitutes in effect a trunk, called a false stem, that has become silicified. The corrugations on the external surface are thus seen to be portions of roots. When suffi- ciently magnified the stems are seen to consist of a prominent ring of xylem (wood) surrounded by a cortex composed of large-celled, thin-walled parenchyma; and an outer zone of thick-walled scleren- chyma. The conspicuous features of the roots are their almost uni- form size and the presence of large-celled xylem elements arranged in the form of a cross. In the older parts of Tempskya trunks, roots may sometimes be found invading the stems. It should be stated that not all Tempskya blocks show the presence of stems; some may represent the outer or lower portions of a large trunk and thus con- Fig. 1.—Transverse section of white ash, Fraxinus americana. Fig. 2.—Transverse section of western yellow pine, Pinus ponderosa, with three resin canals. Fig. 3.—A coal ball. Fig. 4.—Transverse section of Psaronius gutbiert. After Corda. Fig. 5.— Tempskya knowltoni, showing stems embedded in a groundmass of circular to elliptic adventitious roots. Fig.6.—Piece of Palmozylon sp. The fragment near the base, show- ing sections of two adventitious roots, is not a part of this trunk. All figures are some- what diagrammatic and lack some details not considered necessary of delineation for the purposes of this paper. 49 BROWN: TEMPSKYA Fes. 15, 1936 ) “4 ii Se ee wee om i ) v rN} v Or @ By SS, yout’ ~ OF. at For explanation of Figs. 1-6, see bottom of opposite page. 50 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 26, NO. 2 sist only of a mass of roots. Such specimens are practically worthless so far as specific determination is concerned, but are valuable as in- dicating the presence of a Tempskya in the formation. How may the tempskyas be distinguished in the field from other petrifactions of somewhat similar appearance? Excepting a few species of ferns, the confusion of Tempskya with the usual plant petrifactions encountered in the field is unnecessary if the collector is armed, as he invariably should be, with a hand lens of at least 10 to 14-power, and has learned a few elementary anatomical facts. The common fossils that genetically and anatomically simulate Tempskya are specimens of a tree-fern genus, Psaronius, sometimes called ‘‘starling stones,’’ found in upper Carboniferous and Permian strata. Figure 4 illustrates a transverse section of a Psaronius. It will be observed that the central portion of this section represents a single stem composed of a large number of concentrically arranged, arcuate vascular bundles, which, if viewed in longitudinal section, would pre- sent a system having a very complex pattern. Adventitious roots form a protective periphery around the stem. As a whole, this trunk, like that of Tempskya, could also be called a false stem. Those irregular, rounded or angular nodules called coal balls (Fig. 3), found in strata of the Pennsylvanian and other periods, and un- fortunately often overlooked and neglected by field geologists, may sometimes roughly resemble Tempskya. However, examination of them with a lens reveals that usually they are composed of a hetero- geneous mixture of stems, leaves, roots, and other organs derived perhaps from as many different species of plants. The plant petrifactions most commonly encountered in Mesozoic and Cenozoic formations are silicified or lignitized trunks of gymno- sperms (cycads, ginkgos, conifers) and angiosperms (monocotyledons, dicotyledons). Of the gymnosperms the conifers are most frequent, but they will be least likely to be confused with Tempskya. A typical living conifer with which fossil conifers may be compared is the western yellow pine (Pinus ponderosa), of which Figure 2 is a trans- verse section revealing parts of four annual rings, the darker portion of the rings being late wood. This section consists of radially disposed wood elements (tracheids) of nearly uniform size, and resembles in appearance a wall constructed of small, hollow tiles. The darker color and apparently smaller size of the tracheids in the late growth of the annual ring is caused by the thickening of the cell walls. At irregular intervals the rows of tracheids are separated by thin, radial, pith rays; and, scattered sparsely among the tracheids, generally in or near the Frs. 15, 1936 BROWN: TEMPSKYA 51 late wood, are a few, large, circular openings, denoting resin canals. Such a section showing tracheids and resin canals is normally char- acteristic of the wood of that group of conifers which includes the pines, spruces, larches, and Douglas firs. Presenting sections similar to that of a pine, but normally without resin canals, is the other coniferous group that includes the true firs, hemlocks, sequoias. junipers, cedars, cypresses, and bald cypress. Perhaps the petrifactions most likely to be confused with Tempskya are found among the monocotyledons, namely, fossil palm trunks or portions thereof. Figure 6 shows the weathered, ropy, external sur- face of a Palmozylon, a transverse section of a portion of the stem, and some external material near the base, but not part of this stem, showing adventitious roots. The distinctive feature that separates Palmozxylon from Tempskya is the presence, as seen in cross-section, of numerous oval or bottle-shaped, more or less radially directed, fibrovascular bundles, consisting of a small vascular portion with several large vessels (xylem) and a larger, usually darker, fibrous portion composed of sclerenchyma. The transverse sections of the adventitious roots are normally larger than those in Tempskya and under low magnification have the appearance of two concentric circles. Specimens of Palmozxylon consisting of roots are most likely to be mistaken for Tempskya with stems, and those showing fibro- vascular bundles for Tempskya with roots. Contrasted with conifers and monocotyledons, the woody dicoty- ledons are distinguished by the presence of numerous relatively large vessels distributed conspicuously among the smaller elements. These vessels may be concentrated in the spring wood or beginning of the annual ring, as in the white ash (Fig. 1), or they may be spread rather evenly throughout the annual ring. The former are the ring- porous woods and include the oak, chestnut, sassafras, elm, mulberry, ash, locust, and catalpa. The latter are the diffuse-porous woods and include the walnut, hickory, birch, willow, poplar, sycamore, red gum, beech, cherry, maple, tulip tree, buckeye, and linden. A word of caution here is necessary to the effect that gradations between ring-porous and diffuse-porous woods are not uncommon and that the vessels of some species, such as the willows and poplars of the diffuse- porous group, are sometimes so numerous and small that even with a 14-power hand lens these species when poorly preserved as fossils may be mistaken for conifers. It is regrettable that the study of fossil wood has not yet been as systematically and thoroughly organized as to be more generally 52 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 26, No. 2 applicable to the solution of stratigraphic problems. The reason for this backwardness lies neither in the lack of good material Gn the field) nor in the availability of equipment required forsuch studies, but principally in the failure of geologically-biologically educated students and acceptable materials to meet under favorable auspices. Both the geologist and the plant-anatomist are in part to blame for this state of affairs—the geologist for collecting uncritically and the anatomist for rejecting the poor material with unconcealed disdain— the whole performance resulting in general discouragement. It is hoped, however, that this unfortunate impasse can presently be over- come and that all good specimens of T’empskya, fossil wood, and other plant petrifactions may receive adequate attention. The collector should be advised that acceptable collecting will be seasoned with discrimination and will provide representative specimens, not neces- sarily bulky, but showing the essentials required for identification, such as the stems in Tempskya, the central core or pith in gymno- sperms and angiosperms, as the case may be, and a generous portion of the best development of the wood. Thus, because fossil wood usu- ally occurs as broken trunks or stumps of such large size as to dis- may the uninformed collector, it becomes a matter of exercising nicety of judgment in selecting, with the help of a hand lens, only typical portions that have well-preserved features of diagnostic value. LITERATURE CITED . Berry, E. W. Lower Cretaceous. Md. Geol. Survey. p. 298, pls. 37, 38. 1911. . Corps, A. J. Flora Protogaea. Beitrdge zur Flora der Vorwelt, p. 81. 1846. . Kinston, RoBrErRtT, and GwyNNE-VauGHAN, D. T. Ona new species of Tempskya from Russia. Russ. k. mineral. Gesell., Verh. 48: 1-20, pls. 1-8. 1911. . SEwArRD, A.C. Ona new species of Tempskya from Montana: Tempskya knowltont, sp. nov. Annals of Botany 38: 485-507, text figs. 1-3, pls. 16,17. 1924. . LONGWELL, C. R. Geology of the Muddy Mountains, Nevada. U.S. Geol. Survey Bull. 798. 1928. Or = Whe ZOOLOGY .—The histology of nemic esophagi. V. The esophagi of Rhabditis, Anguillulina, and Aphelenchus.!. B. G. Currwoop, Bureau of Animal Industry, and M. B. Cuirwoop. This is the fifth paper of a series dealing with the structure of the esophagi of various groups of nematodes. In the previous papers (Chitwood and Chitwood, 1934-1935) the histology of the esophagi of Rhabdias eustreptos, Oesophagostomum dentatum, Heterakis gallinae, and Metestrongylus elongatus has been described. The nomenclature in this paper is the same as that used in previous ones. 1 Received March 13, 1935. Frs. 15, 1936 CHITWOOD AND CHITWOOD: NEMIC ESOPHAGI D3 In the present paper, the esophagi of Rhabditis terricola and R. lambdiensis are described in full, while descriptions of the esophagi of Anguillulina dipsact and Aphelenchus avenae are given in brief and in the form of a comparison with those of the species of Rhabditis. THE ESOPHAGI OF RHABDITIS TERRICOLA AND R. LAMBDIENSIS Both of these species were studied in detail, the results being nearly, but not entirely, identical. The description is based on R. terricola, comparisons with FR. lambdiensis being made where differences were noted. Illustrations were made of the species of which the best sections were available. GROSS MORPHOLOGY. The esophagus of Rhabditis terricola is 144 to 172u long and consists of 3 major parts: A corpus distinctly subdivided into an anterior part, the precorpus, 58 to 70u long, and a posterior part, the post- corpus, 24 to 34u long; an isthmus 30 to 42u long; and a bulb 30y long. The precorpus is cylindrical, some of its tissueextending anteriorly and surround- ing the mesostom and telostom. The lumen of the precorpus is somewhat subtriangular at the base of the stoma; immediately posterior to this point it becomes triradiate, each radius terminating distally in a small, incomplete “tube.” In R. lambdiensis these ‘‘tubes”’ are much larger than in R. terricola. The “‘tubes”’ gradually become smaller in the posterior part of the precorpus and disappear in the postcorpus. The postcorpus takes the shape of a fusi- form swelling commonly called the ‘‘median bulb’’; its lumen is triradiate, the walls of the radii converging distally. The isthmus is a long, narrow re- gion, the lumen being similar to that of the precorpus. The bulb is pyriform and contains a well developed valvular apparatus which alters the shape of the lumen. The cuticle lining the anterior part of the value is relatively thin, the lumen subtriangular to triradiate according to the stage of contraction. This is followed by the esophageal valve which consists of thickened parts of the cuticular lining to which muscles are attached. NUCLEAR DISTRIBUTION. Precorpus. There are 18 nuclei in the precor- pus of which 12 are of nerve cells (ni_12) and 6 are of the radial muscle fibers (r;_.). The nerve cell nuclei are arranged in 4 groups of 3 nuclei each, 1 nu- cleus being in the center of each esophageal sector. The first group (ni_3) is situated a short distance from the anterior end of the corpus; the second, third, and fourth groups (n4., n7~9, and niy_12) are situated in series, one behind the other. The radial nuclei (r;_.) are arranged as a single group of 6 nuclei, 1 on each side of each esophageal sector, the group being situated near the base of the precorpus and between the third and fourth groups of nerve cell nuclei. In Rhabditis lambdiensis the same number of nuclei are present as in R. terricola, but the fourth group of nerve cell nuclei (ni1_12) is situated anterior to the radial nuclei (ri_s). Postcorpus. There are 28 nuclei in the postcorpus, comprising 6 radial 54 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 26, NO. 2 nuclei (r7_12), 8 marginal nuclei (mis), and 19 nerve cell nuclei (nj3_31). The marginal nuclei are situated near the anterior end of the postcorpus, 1 at the end of each esophageal radius; these nuclei are usually bilobed, the lobes being designated a and b (see mya, etc., in Fig. 1). This lobing of m1_; is discussed in the description of the nuclei. The radial nuclei (r7_12) are situated near the middle of the postcorpus and arranged as in the case of the first group (ri_s). The nuclei of the nerve cells are arranged as a chain in the dorsal sector (ni3,16,23,28,29) and similarily in each subentral sector (M14 ,17,19 21,24 ,26 ,30 and Ns 18,20 ,22,25 27,31). | he nuclei in each sector are situated near the center of the sector, but the distance from the lumen varies considerably; for the Fig. 1.—Rhabditis terricola; diagrammatic representation of the esophagus, a—b, precorpus; c—d, postcorpus; e-f, bulb. relative positions see Figs. 1-2. The 3 most posterior nerve cell nuclei (nz9_31) are situated just posterior to the orifices of the subventral esophageal glands 1 being external to each of the 3 gland masses. Isthmus. No nuclei are present. Prevalvar region. The prevalvar and valvar regions of the esophagus are not distinctly separate. Contraction of the valve changes the relative posi- tions of nuclei and for that reason the nuclei of the valvar region are in- cluded with those of the prevalvar region. There are a total of 17 nuclei in this region as follows: 6 radial nuclei (113_1s), 3 bilobed marginal nuclei (m4_¢), 6 nerve cell nuclei (nze_37), and 2 gland cell nuclei (g2_3). The marginal nuclei are situated near the anterior end of the prevalvar region at the ends of the esophageal radii, 1 lobe lying on each side of each radius (Fig. 1). The radial muscle nuclei (r3_1s) are situated 4 to 10u posterior to the marginal nuclei, 1 on each side of each sector. These nuclei are nearer the center of their sec- tors than those of the previous radial nuclei (r,;-, and r;_12). The nerve cell nuclei nze_34 are situated near the posterior level of the radial nuclei, 1 in the center of each sector. The nerve cell nuclei n3s_37 are situated 4 to 6u poste- rior tO Nge_34, 1 on each side of the dorsal sector, n3; being near the lumen at the Fras. 15, 1936 CHITWOOD AND CHITWOOD: NEMIC ESOPHAGI 55 right side and ns; near the margin of the sector on the left side. The gland nuclei (g2_3) are in the lateral parts of the subventral sectors or sometimes in the lateral parts of the dorsal sector; their position may be at the posterior level of the radial nuclei (Fig. 1) or at the level of ngz7. Postvalvar region. The postvalvar region contains 13 nuclei as follows 6 radial nuclei (19-24), 8 marginal nuclei (m;_9), 3 nerve cell nuclei (ngg_49), and 1 gland nucleus (gi). The radial nuclei are arranged in 2 groups of 3 nuclei each (ri9_e1 and 224), 1 nucleus in the center of each sector, the first group being 3 to 6u anterior to the second group and nearer the lumen than the second group (Fig. 2). The marginal nuclei (m;_9) are arranged as a single Fig. 2.—R. lambdiensis; longitudinal reconstruction of postcorpus. group, 1 nucleus being at the side of each esophageal radius; their relative position varies considerably. The nerve cell nucleus n3g is situated on the right side of the ventral esophageal radius, whereas nu is situated on the left side of the same radius and nzg on the dorsal side of the left subdorsal radius. The gland cell nucleus g; is near the posterior end of the postvalvar region on the right side of the dorsal sector. ESOPHAGO-INTESTINAL VALVE. The esophago-intestinal valve is compar- atively large and well developed in Rhabditis and consists of a trilobed in- ternal structure and a circular external structure. Seven nuclei have been observed in these structures, but it is difficult to determine with certainty which nuclei belong to each of the two parts. Their positions vary somewhat, but usually there are 2 subdorsal nuclei, 2 dorsolateral or lateral nuclei, 2 subventral nuclei, and 1 ventral nucleus. CHARACTER OF NUCLEI. The radial nuclei each contain 1 nucleolus or sometimes 2 nucleoli, moderately large in either case, surrounded by a nu- cleoplasm having little or no affinity for stains. The radial nuclei of the first group (rs) are oval in cross section, (Fig. 3a); they are very much com- pressed longitudinally and often quite long. The radial nuclei of the second group (r7z_12) are similar to those of the first group in being oval in cross sec- 56 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 26, NO. 2 tion (Fig. 3b), but they may not be quite as elongate longitudinally (Fig. 2). The radial nuclei of the third group (113_1s) are less oval in cross section (Fig. 3c), and not elongated longitudinally; there is considerable irregularity in shape. The radial nuclei of the fourth group (19-21) are nearly spherical and rest in a densely staining cytoplasm (Fig. 3e) which appears to be very simi- lar to the cytoplasm of the esophageal glands. The radial nuclei of the fifth group (Tee24) are usually somewhat compressed (Fig. 3f), their shapes de- pending on the state of contraction of their associated muscles. MM. i 3 S m 1 ne ey 3 VU - = alos fon ce Sy LO a r, CS S wt \ il vu . 'y ry Pg = Nag M32 ig cS {> > My y > Ny / ee \, co X) I as Ree © : J 17 VS Tu 4 oy Pie: Ng4 ft) Me0 Ss ‘ @ Mk Ne 2 C 5 d 36 Im 19/99 Qo) hae fy "35, fe, a ve knot Me ag S S / . s I 1. \d Fig. 3.—R. terricola; cross sections of the esophagus. a, precorpus at level of ni6; b, postcorpus, compare with Fig. la; R. lambdiensis c-f, bulb (serial sections). The marginal nuclei are similar to the radials in that the nucleoplasm has little or no affinity for stain and each nucleus contains 1 or 2 nucleoli. The marginal nuclei of the first and second groups (my_3,4-¢) are usually bilobed, each lobe containing a nucleolus. The lobes are joined anteriorly (Fig. 3 b-d). The third group of marginal nuclei (m7_,) are similar in character to those of the second groups but not lobed (Fig. 3e). The nuclei of the first and second groups of marginals are apparently not lobed in Rhabditis lamb- diensis (Fig. 2). The nuclei of the esophageal glands (g:3) are larger than other nuclei of the esophagus with the exception of the nuclei of the fourth group of radials (ri9_21); they are generally similar to rj9_2, being somewhat spherical and having 2 large nucleoli. The dorsal esophageal gland nucleus (gi) is the largest nucleus of the esophagus (Fig. 3f), while the subventral nuclei (g2_3) Fes. 15, 1936 CHITWOOD AND CHITWOOD: NEMIC ESOPHAGI 57 are of about the same size as the radial nuclei (rj9_21) (Fig. 3d). The nucleo- plasm contains a few coarse basophilic granules and there is often a dense basophilic margin adjacent to the inconspicuous nuclear membrane. The nuclei of the nerve cells of the esophagus differ from the other nuclei in that the nucleoplasm has a very definite affinity for basic stains. The baso- philic material is concentrated next to the nuclear membrane for the most part, but clumps of granular material are present also near the center of the nucleus. Nucleoli were not always observed in the nerve cells but quite often a bilobed nucleolus was seen. The nerve cell nuclei are spherical or slightly elongated, and are so consistent in appearance that they may be easily re- cognized by reference to Figures 2 and 3. The protoplasm surrounding the nerve Cell nuclei is relatively meager and no particular study of it was made. The neurones of the precorpus (ni_12) all appear to be bipolar, as do all ex- cept 3 (Nos_2s) of the postcorpus (Fig. 2). The exceptions (Nzs_2s) are neurones of the commissure which is situated in the posterior part of the postcorpus. The nerve cells which are situated posterior to no9_31 may be commissural cells also. The nerve cells of the prevalvar region (N3:_3.) are commissural cells of the prevalvar and valvar region, while the remaining nerve cell of this region (n37) and those of the postvalvar region (n3s_49) are probably a part of a postvalvar commissure; the later commissure has not been observed with certainty. ESOPHAGEAL GLANDS. There are 3 esophageal glands, 1 dorsal and 2 sub- ventral. The dorsal gland opens at the base of the glottoid apparatus through a very short transverse duct lined with cuticle. This duct dilates distally from the orifice, forming an ampulla which may or may not be lined with cuticle (facts uncertain). A narrow tube not lined with cuticle extends pos- teriorly from the ampulla through the middle of the dorsal sector. In the pre- corpus this tube is surrounded by a very meager amount of glandular proto- plasm. In the postcorpus the glandular mass becomes a little larger and branches from the tube have been observed, and in the isthmus the glandular mass is represented only by a delicate strand of protoplasm. The dorsal gland becomes larger and lobed posterior to the valve and is most concen- trated in the right side of the dorsal sector where its nucleus lies. The glan- dular protoplasm at the level of the nucleus is reticulate or facuolate. The subventral esophageal glands open into the lumen of the esophagus near the base of the corpus, between the levels of the nerve cell nuclei nes_2s and Ne9_31. Like the dorsal gland, each has a short transverse duct lined with cuticle which is connected with an ampulla; the protoplasm of the gland is rather extensive and reticulate in this region, the reticuli being apparently in direct communication with the lumen of the ampulla. The subventral gland masses become narrow and strand-like in the isthmus and remain of this character in the anterior part of the prevalvar region. In the posterior part of the prevalvar region and in the valvar region the gland masses are multilobed and reticulate. 58 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 26, NO. 2 THE ESOPHAGUS OF ANGUILLULINA DIPSACI The esophagus of Anguwillulina dipsaci (Fig. 4) cosists of the same gen- eral regions as the esophagus of Rhabditis, but differs in that the corpus and isthmus are proportionately thinner, and the bulbar region is cylindroid Fig. 4.—Angutllulina dipsaci; longitudinal reconstruction of esophagus. instead of pyriform and is entirely devoid of valves. This type of posterior swelling has been termed a pseudobulb although it is homologous with the bulb of Rhabditis. The lumen of the entire esophagus is greatly reduced (Fig. 5), being scarcely distinguishable except in the postcorpus (Fig. 5). The precorpus apparently contains either 15 or 18 nuclei the arrangement and nature of which were not determined. The postcorpus contains 25 nuclei (Fig. 5) which appear to consist of 2 sets of radial nuclei (11-6, rz_12), 1 set of marginal nuclei (m,_3), and 10 nerve cell nuclei (ni_1)). These designations, it should be noted, were made on the basis of comparison with Rhabditis and not on purely structural grounds. The isthmus is exceedingly minute in cross Fig. 5.—A. dipsaci; cross sections of esophagus. a—b, precorpus; c-f, postcorpus (serial sections); g—m, bulbar region (serial sections). section and, like other forms, contains no nuclei. The bulbar region is most interesting, for although 30 nuclei have been found as in Rhabditis, there is practically no similarity in other respects. The musculature is exceedingly degenerate, and the esophageal gland tissue massive. Small nuclei are grouped near the anterior and posterior ends of the bulbar region (Fig. 5) with no apparent symmetry. The dorsal and subventral esophageal gland nuclei stand out in contrast, being of much greater size than the other nuclei. As shown by Goodey (1929), the dorsal esophageal gland opens into the esophagus at the base of the stylet and the subventrals open at the base of the postcorpus. Fras. 15, 1936 CHITWOOD AND CHITWOOD: NEMIC ESOPHAGI 59 THE ESOPHAGUS OF APHELENCHUS AVENAE The esophagus of Aphelenchus avenae is similar to that of Angwillulina with several outstanding exceptions, the most conspicous being the more massive development of the postcorpus and the degeneration of the bulbar region, the latter feature allowing the esophageal glands to lie in the body cavity. We find but 10 nuclei in the precorpus and 33 in the postcorpus (Fig. 6), the total number being the same as in Angwillulina. Of the 33 nuclei in the postcorpus, 15 (ri_12, mi_3) appear to check with those in Anguillulina, but the different arrangement of the remaining 18 nuclei makes their rec- ognition impractical. The bulbar region appears to have 30 nuclei arranged without apparent symmetry (Fig. 6). Only the 3 esophageal gland nuclei Fig. 6.—Aphelenchus avenae; cross sections of esophagus. a—d, precorpus; e-g, postcorpus; h—o, bulbar region. are clearly recognizable. One point is notable, namely, that the esophageal glands, which extend posteriorly as a cylindroid mass of tissue, are sur- rounded by a nucleated covering. The fibrous tissue corresponding to these nuclei probably represents marginal or radial fibers. Goodey (1929) found that the dorsal esophageal gland opens in the anterior part of the postcorpus, whereas the subventrals open in the usual position; the writers were not able to verify or disprove this observation, but on the basis of sections it appears probable that he is correct. LITERATURE CITED Cuitwoop, B. G., and Cuirwoop, M.B. The histology of nemic esophagt. Parts I-II. Z. Zellforsch. u. mikroskop. Anat. 22: 29-37, 38-46. 1934. Part IIT. This Journau 24: 557-562. 1934. Part IV. This Journau 25: 230-237. 1935. Goovry, T. On some details of comparative anatomy in Aphelenchus, Tylenchus, and Heterodera. Jour. Helminth. 7: 223-230. 1929. 60 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 26, NO. 2 ZOOLOGY .—The occurrence of the terrestrial amphipods, Talitrus alluaudi and Talitrus sylvaticus, in the United States.1 CuarR- ENCE R. SHOEMAKER, U. 8. National Museum. (Communicated by Mary J. RaTHBUN.) The first record of the occurrence of a strictly terrestrial amphipod in the United States is that of Dr. J. Paul Visscher and Chester S. Heimlich (Science, 72: 560, 1930). In 1930 they reported Talitrus alluaudi Chevreux abundant in a greenhouse at Columbus, Ohio, where they say it had survived for more than two years. In 1931 specimens of this species from a greenhouse at Flemington, New Jersey were sent to the Nation Museum for identification. These are the only known occurrences of TJalitrus alluaudi in America. T. alluaudi was described from the Seychelles Islands in 1896 where it occurs in rotton trunks of coconut trees and in the humus of forests. In temperate countries, however, it has been reported only from greenhouses. In 1918 and 1919 Mrs. Kate Stevens, of the Natural History Museum, Balboa Park, San Diego, California, sent to the National Museum specimens of an amphipod which had been found upon a sidewalk in Balboa Park. Mr. Frank F. Gander in 1927 sent the Museum excellent specimens of the same species which he had found in Balboa Park. In December, 1934, and January, 1935, Prof. 8. F. Light of the University of California sent the Museum further speci- mens of this species from a garden in Pasadena, California. Upon careful study, this amphipod proved to be Talitrus sylvaticus Haswell, and Mrs. Stevens’ specimens of 1918 appear to be the first of this species to be taken in America. The specimens sent to the Museum by Professor Light were secured in October, 1934, at Pasadena by Mrs. Merwin, who, in a letter to him, says, ‘“The pests came during the recent rains in such quantities that they had to be cleared out of the gutters and literally shoveled from in front of the door. They even came into the house under the front door, but were dead when we found them. They seem to come up alive, and of a dark color, from between the bricks around the front door, though also found in garden gutters.” In a later letter to Professor Light, dated November 17, 1934, she says, ‘“‘The amphipods which you requested reappeared. In yesterday’s rain they were found, though not in the former quantity, in our garden and a friend also reports them in Arcadia, which is about five miles east of Pasadena.”’ 1 Published by permission of the Secretary of the Smithsonian Institution. Re- ceived October 3, 1935. Fra. 15, 1936 SHOEMAKER: TERRESTRIAL AMPHIPODS 61 _ Fig. 1.—Talitrus sylvaticus view, enlarged. c, maxilla 1. f, branchial vesicle of peraeopod 4. g, pleonsegment 2. h, pleon segment 3. 7, distal end of peduncle of uropod 1. k, uropod 3. 1, telson. Haswell, female. a, entire animal. 6, antenna 1, top d, maxilliped. e, branchial vesicle of gnathopod 2. j, spine at distal end of peduncle of uropod 1, enlarged. 62 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 26, No. 2 Recently there were sent to the National Museum for identification seventeen specimens of Talitrus sylvaticus, which were found in a dog’s water pan at New Orleans, Louisiana, July 25, 1935. T. alluaudi is a rather small species, which measures from 5 to 7 mm. in length exclusive of the antennae. The largest specimens of T. sylvaticus from California measure 13 mm. exclusive of antennae, while those from Louisiana measure from 5 to 11 mm. Both of these amphipods are widely distributed. 7. allwaudi has been recorded living in the open at the Seychelles Islands, Mada- gascar, Java, Gambier Archipelago, Taumotu Islands, and Man- gareva. It has been recorded from greenhouses in Belgium, France, Monaco, Switzerland, Germany, Denmark, Hungary, British Isles, Sweden (Stephensen, 1935), the United States, and there is in the collection of the U. 8. National Museum a specimen found in New York City in a shipment of celery from Bermuda. Talitrus sylvaticus has been recorded from New South Wales, Victoria, Hawaii, Marquesas Islands, and now from the United States. It has been found from just above high tide up to 2,520 feet altitude. As Calman, Stephensen, and others have pointed out, this species appears to be subject to considerable variation in some of its char- acters. The third pair of pleopods may be lacking altogether, or may be quite small with the branches vestigial. Chevreux said that speci- mens sent him by Chilton had third pleopods resembling the first, being biramous, though of smaller size. In the California specimens the third pleopods are very much as figured by Stephensen for the Marquesas specimens, consisting of a reduced peduncle and a single vestigial ramus. The telson also apparently is quite variable. Haswell, in his original description, says that the telson is cleft in the middle line. Thomson said it appeared quite entire to him. Sayce describes and figures the telson with margin entire. Stephensen states that it is cleft for one-third of the length, and figures it so. The specimens from California and Louisiana have the telson entire as I have figured it. Two specimens in the Museum collection from Hawaii, identified by A. O. Walker, agree in all characters with the specimens from the United States, and have the telson entire. It would appear then that the telson in 7’. sylvaticus may be either entire or partly cleft. The outer plate of the maxillipeds is rather broad, and rounding distally Fig. 2.—Talitrus sylvaticus Haswell, female. a, gnathopod 1. 6, gnathopod 2. c, peraeopod 2. d, peraeopod 3. e, peraeopod 4. f, peraeopod 5. g, pleopod 1. h, pleopod 2. 72, pleopod 3. j,k, l, distal end of pleopod 3, enlarged. Fess. 15, 1936 SHOEMAKER: TERRESTRIAL AMPHIPODS For explanation of Fig. 2, see bottom of opposite page. = A f 63 64 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 26, NO. 2 in the Californian and Hawaiian specimens, whereas Chilton (Jour. Proc. Royal Soc. of N.S.W. 50: 84, fig. 3) figures this plate as narrow and distally acute with the inside margin concave. This is a very peculiar discrepancy which I cannot account for. Stephensen (Bernice P. Bishop Mus. Bull. 142: 20, 1935) states that the palp of the maxillipeds has a small fourth joint. In the specimens which I have examined the palp bears a small distal fleshy lobe marked off by a row of spinules, but which does not appear to be separated from the third joint. ) Talitrus alluaudi may be expected to appear in greenhouses in other parts of the United States, as it is probably transported in the soil around the roots of plants. The occurrence of Talitrus sylvaticus in such widely separated localities as San Diego and New Orleans would seem to indicate that this species is much more common in the warmer parts of the United States than is now known. It is only when interest or curiosity prompts persons to have these creatures identified that their presence becomes known. MALACOLOGY.—Two new land shells from the Philippine Islands. Paut Bartscu, U.S. National Museum. In my paper on the land shells of the genus Obba from the Min- doro Province, Philippine Islands, published in Bulletin 100, volume 6, part 8, United States National Museum, I discussed Obba listers Gray and figured the type species, as well as a number of subspecies belonging to this group. To these I added several more races in a paper published in the Journal of the Washington Academy of Sci- ences 24: 318-323, 1934. A collection recently received from Mr. Frederick 8. Webber con- tains two unnamed races which are here described. Obba listeri webberi, n. subsp. Fig. i Shell small, rather elevated, the spire forming a regular, almost hemi- spheric cone, the lower surface being much less rounded. Nuclear whorls 2.1 horn colored, the last half of the last turn with a pale brown band below the suture and another one occupying a median position. The postnuclear turns are of pale buff ground color and bear interrupted bands of brown, of which the first is near the summit, and the other a little above the middle, while the anterior half of the whorls is flecked and blotched with brown. The under side has an interrupted band about one-third of the distance between the peripheral keel and the umbilicus anterior to the keel, and the region between this interrupted band and the keel is also marked with flecks and blotches of brown. The nuclear whorls are marked by faint lines of growth. On the 1 Published by permission of the secretary of the Smithsonian Institution. Re- ceived October 23, 1935. Frs. 15, 1936 BARTSCH: PHILIPPINE SHELLS 65 postnuclear whorls the lines of growth increase in strength and are strongly marked, very decidedly so on the last turn. The malleations are strongest anterior to the middle of the turns. The under surface is also strongly mal- leated and here the malleations extend almost to the umbilicus. The lines of growth here are even stronger than on the spire. The inner half of the base, including the umbilicus, is marked by well incised spiral striations. The um- bilicus is rather narrow and about one-third covered by the parietal lip. The last whorl is constricted behind the inner lip, while the outer lip is decidedly deflected. The aperture is oval, slightly angulated at the periphery; peri- stome moderately thickened and reflected, the inner lip bearing a strong median tooth within. Fig. 1.—Obba listeri webbert, new subspecies. Fig. 2.—Obba listert catanduanensis, new subspecies. The type, U.S.N.M. Cat. No. 314057, was collected near the lighthouse on Tres Reyes Island near Marinduque. It has 4.6 whorls, and measures: Height, 12.8 mm.; greater diameter, 25.5 mm.; lesser diameter, 21.0 mm. U.S.N.M. Cat. No. 314058 contains 3 paratypes from the same station, while 7 additional specimens are in Mr. Webber’s collection. The lot yields the following average measurements: Height, 12.8 mm.; greater diameter, 25.9 mm.; lesser diameter, 21.8 mm. This subspecies is much more elevated, with the spire more regularly conic and the base comparatively more flattened than any of the other named races of Obba listerz. Obba listeri catanduanensis, n. subsp. Fig. 2 Shell of medium size, lenticular, of flesh colored ground color, mottled and splotched with dark chestnut brown on the upper surface and also on the outer half of the lower surface. Nuclear whorls 2.3 in the type, horn colored. The first half unicolor, while the succeeding portion shows a brownish flush near the suture and a median, rather broad brown band. The postnuclear portion of the shell shows the brown markings, arranged in more or less regular spiral zones, an interrupted band near the summit, a more or less median broken band and the rest fulgurated and mottled with brown. The nuclear whorls are marked by fine lines of growth, which are a little stronger 66 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 26, NO. 2 near the summit and increase in strength as the mollusk adds to the sub- stance of the shell. The postnuclear whorls are strongly acutely keeled at the periphery, which flares slightly upward, and the surface is marked by malleations which are strongest on the outer half of the whorls, and some- what irregular impressed spiral lines, which are coarser on the upper portion of the whorls. Suture poorly impressed; periphery sharp. The base is moder- ately arched, the outer half of the last whorl is strongly malleated, the inner half finely spirally striated. The lines of growth here are of the same strength as on the spire. The umbilicus is moderately open, and the shell is constricted immediately behind the inner lip, while the upper lip is decidedly bent up- ward. The aperture is broadly oval; peristome moderately expanded, thick- ened and reflected. A low tooth is present on the inside of the middle of the basal lip. The type, U.S.N.M. Cat. No. 314056, is one of 3 specimens sent to us by Mr. Webber collected at Virac, Catanduanes. It has 4.5 whorls, and meas- ures: Height, 10.6 mm.; greater diameter, 32.1 mm.; lesser diameter, 25.7 mim. Two paratypes in Mr. Webber’s collection yield the following measure- ments: Height, 12.8 and 12.1 mm.; greater diameter, 31.7and 33.2 mm. ; lesser diameter, 25.8 and 25.8 mm., respectively. An additional specimen, U.S.N.M. Cat. No. 311069, received from Mr. Maxwell Smith, comes from Batu, Catanduanes, a station not far removed from Virac. It yields the following measurements: Height, 10.6 mm.; greater diameter, 32.8 mm.; lesser diameter, 24.7 mm. This subspecies in general form resembles the typical race, but is much smaller and of much darker coloration. ENTOMOLOGY.—Some butterflies from eastern Virginia.t AUSTIN H. Ciark and Leia F. Cuarx, U.S. National Museum. Since the days of Boisduval and Le Conte the butterflies of eastern Virginia have received little attention. Various collectors have visited the region, but only a few notes on some of the more unusual species have been published. We have made a preliminary reconnaissance of this area, visiting Accomac and Northampton Counties on July 20-27, 1935, Princess Anne County on September 23-24, 1934, and Norfolk and Nanse- mond Counties on September 1-3, 1935. Although our time was limited, we feel that we secured a fairly complete representation of such butterflies as were flying when we were in any given locality, and therefore that our list is sufficiently detailed to serve as a basis for future intensive work. Included in the list are records of nine species from Bayford, Northampton County, kindly given us by Dr. Florence Walker of Bayford, and of one from Lake Drummond which we owe to the courtesy of Dr. Paul Bartsch. With these included our list totals sixty-nine species. 1 Received October 26, 1935. Fes. 15, 1936 CLARK AND CLARK: BUTTERFLIES 67 We are under great obligations to Dr. Hugo Kahl, of the Carnegie Museum, Pittsburgh, Pa., who was so very kind as personally to bring to Washington the type specimens, male (Figs. 7, 8) and fe- male (Figs. 5, 6), of Atrytone dion W. H. Edwards, and the unique type of Atrytone dion race alabamae Lindsey (Figs. 9, 10) for com- parison with our material, and also to Mr. Ernest L. Bell, of Flush- ing, N. Y., who most courteously verified our determination of cer- tain skippers. Family NYMPHALIDAE: Subfamily Saryrinan: Enodia creola (Figs. 1, 2); western border of the Dismal Swamp and westward in wet woods, locally frequent, always with the following. Enodza portlandia; locally fre- quent to abundant in low wet woods throughout Princess Anne, Norfolk, and Nansemond Cos. Neonympha gemma; locally common in very wet woods in Princess Anne Co.; less common in Norfolk and Nansemond Cos. Neonym- pha areolatus var. septentrionalis (Figs. 3, 4); along Norfolk Southern Rail- way, about 13 miles north of the North Carolina line; frequent. Neonympha sosybius; common everywhere in woods in Princess Anne, Norfolk, and Nansemond Cos. Neonympha eurytus; Bayford, Northampton Co., common in spring (F. Walker). Cercyonzs alope alope; occasional throughout except in the eastern part of Princess Anne Co., where it is replaced by the follow- ing. Cercyonis alope pegala; Virginia Beach and Princess Anne; frequent. Subfamily NyMPHALINABE: Chlorippe clyton; Bayford, Northampton Co., sometimes common (F. Walker). Basilarchia arthemis astyanaz; occasional throughout. Basilarchia archippus; occasional throughout. Junonia coenia; everywhere common. Pyrameis atalanta; not seen in Princess Anne Co., elsewhere occasional to common. Pyrameis virginiensis; not seen in Princess Anne Co., elsewhere occasional to frequent. Pyrameis cardui; Wachapreague and Locustville, Accomac Co., occasional; Suffolk, occasional; Green Sea, frequent. Vanessa antiopa; Bayford, Northampton Co., October (F. Walker). Polygonia comma; Bayford, Northampton Co. (F. Walker). Phyciodes tharos; common throughout. Brenthis myrina; Bayford, Northampton Co., rare (F. Walker). Argynnis cybele; Bayford, Northampton Co., rare (F. Walker). Euptoieta claudia; occasional or locally frequent throughout. Subfamily DanartnaE: Danais plexippus; occasional throughout but no- where common. Subfamily LipyTHEINAE: Libythea bachmanni; Bayford, Northampton Co., one (F. Walker). Family RIODINIDAE: Charis virginiensis; common just south of Vir- ginia Beach. Family LYCAENIDAE: Subfamily LycaEninaE: Chrysophanus phlaeas hypophlaeas; Wachapreague, Accomac Co., and Jamesville and Kiptopeke, Northampton Co. Everes comyntas; generally frequent throughout. Lycae- 68 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 26, No. 2 nopsis argiolus pseudargiolus ; not seen in Princess Anne Co. PIs Wwisene local- ly frequent. Subfamily THEcLINAE: Atlides halesus; Lake Drummond, October 27 (P. Bartsch). Strymon cecrops; occasional to frequent in once throughout. Strymon melinus; generally frequent throughout. Mitoura gryneus; Bayford, Northampton Co. Family PAPILIONIDAE: Subfamily Prerinar: Kurema lisa; frequent to abundant throughout. Eurymus philodice philodice; New Church and Wachapreague, Accomac Co., and Virginia Beach; rare; Eurymus philodice eurytheme; abundant everywhere in Accomac and Northampton Cos., in- frequent in Princess Anne, Norfolk, and Nansemond Cos. Zerene caesonia; Bayford, Northampton Co., one (F. Walker). Catopsilia eubule; everywhere frequent in Nansemond, Norfolk, and Princess Anne Cos., becoming abun- dant near the sea; frequent along the western shore of Northampton and Accomac Cos., and at Chincoteague Island; one female of form sennae from Virginia Beach. Anthocaris genutia; Bayford, Northampton Co., common in spring (F. Walker). Pzeris rapae; frequent near farms throughout, though generally not very common. Subfamily PAPILIONINAE: Papilio polyxenes asterius; frequent through- out, and the commonest swallowtail in Accomac and Northampton Cos. Papilio cresphontes ; frequent at Kiptopeke and Bayford, Northampton Co., and also found at Wachapreague, Accomac Co., and Little Creek and Deep Creek, Norfolk Co. Papilio glaucus; not seen in Princess Anne Co., but elsewhere frequent. Papilio troilus; frequent throughout; least common in Princess Anne Co. Papilio palamedes ; the most abundant swallowtail in and near the Dismal Swamp; generally common in suitable situations in Princess Anne, Norfolk, and Nansemond Cos. Family HESPERIIDAE: Subfamily Pyrainaz: Goniurus proteus; Vir- ginia Beach, one. Hpargyreus tityrus; not seen in Princess Anne Co., oc- casional to common elsewhere. Thorybes bathyllus; occasional throughout. Thorybes confusis; Princess Anne, one. Pyrgus communis; Wachapreague and Harborton, Accomac Co., and Bayford, Northampton Co. Pholisora catullus ; not seen in Princess Anne Co., elsewhere locally frequent. Thanaos juvenalis; Wachapreague and Dahl Swamp, Accomac Co. Thanaos horatius; occasional to frequent throughout. Thanaos terentius; Wachapreague and Dahl Swamp, Accomac Co., and Kiptopeke, Northampton Co. Subfamily Hesprriinan: Ancylorypha numitor; frequent to common throughout. Hylephila phylaeus; occasional to frequent throughout. Atalo- pedes campestris ; the commonest skipper in Accomac and Northampton Cos.; occasional in Nansemond Co.; not seen in Norfolk and Princess Anne Cos. Talides manataaqua; on the line of the Norfolk Southern Railway, about 13 miles north of the North Carolina border; frequent. Talides themisto- cles; occasional throughout. Wallengrenia otho egeremet; Dahl Swamp and Cashville, Accomac Co.; Bayford, Northampton Co.; on the line of the Fes. 15, 1936 CLARK AND CLARK: BUTTERFLIES 69 14 Figs. 1, 2—Enodia creola, Dismal Swamp, Va., male, Sept. 1, 1935 (1), and female, Sept. 2, 1935 (2). Figs. 3, 4—Neonympha areolatus var. septentrionalis, near the Dis- mal Swamp, Va., Sept. 3, 1935, male (3) and female (4), under side. Figs. 5, 6.— Aitrytone dion Edwards, female, type specimen, Whiting, Indiana, upper (5) and under (6) sides. Figs. 7, 8.—Atrytone dion Edwards, male, type specimen, Whiting, Indiana, upper (7) and under (8) sides. Figs. 9, 10.—Atrytone dion race alabamae Lindsey, male, type specimen, from Mobile County, Albama, upper (9) and under (10) sides. Figs. 11, 12.—Atrytone dion alabamae, male, Dahl Swamp, Accomac Co., Va., July 23, 1935, upper (11) and under (12) sides. Figs. 13, 14.—Atrytone dion alabamae, female, Dahl Swamp, July 25, 1935, upper (13) and under (14) sides. 70 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 26, No. 2 Norfolk Southern Railway, about 14 miles north of the North Carolina border. Poanes zabulon; Bayford, Northampton Co.; near Adam’s Swamp, Nansemond Co., about 1 miles north of the North Carolina border and about 3 miles west of the Dismal Swamp. Poanes yehl; Green Sea, one male; Dismal Swamp, near Suffolk, one female. Atrytone dion alabamae (Figs. 11-14); Dahl Swamp, Accomac Co., common; Green Sea, Norfolk Co., one female. Atrytone ruricola; Dismal Swamp; Green Sea. Lerema accius; the commonest skipper in Nansemond, Norfolk, and Princess Anne Cos.; not seen in Northampton or Accomac Cos. Amblyscirtes textor; frequent every- where in wet woods in Nansemond, Norfolk, and Princess Anne Cos., and locally common and even abundant about the Dismal Swamp. Amblyscirtes carolina; western border of the Dismal Swamp about 8 miles south of Suf- folk, three, in company with large numbers of the preceding. Lerodea l’her- minier ; western border of Dismal Swamp, frequent: Virginia Beach, one. IT erodea eufala; Green Sea, Norfolk Co., one. Prenes panoquin; Wacha- preague and Chincoteague Island, Accomac Co., abundant on Borrichia fru- tescens; Hack’s Neck, Accomac Co.; Bayford, Northampton Co., common. Prenes ocola; Virginia Beach, one; Dismal Swamp, one. ENTOMOLOGY.—The bees of the genus Agapostemon (Hymenop- tera: Apoidea) occurring in the United States: GRack ADEL- BERT SANDHOUSE, Bureau of Entomology and Plant Quaran- tine. (Communicated by 8. A. ROHWER.) This study of the Agapostemon occurring in the United States was undertaken to facilitate the identification of these species. The re- sults presented in this paper are based on the examination of about four thousand specimens and many dissections of the male genitalia. The collection of the Agapostemon in the United States National Museum has served as a basis for this revisionary study. This was supplemented by loans from the American Museum of Natural History (through Dr. F. E. Lutz), the Academy of Natural Sciences of Philadelphia (through Mr. E. T. Cresson, Jr.), the Illinois State Natural History Survey (through Dr. T. H. Frison), Cornell Uni- versity (through Dr. J. C. Bradley), McGill University (through the late Mr. Albert F. Winn), South Dakota State College (through Prof. H. C. Severin), the Bureau of Biological Survey (through Mr. J. R. Malloch), the University of Minnesota (through Dr. C. E. Mickel), and the private collections of Drs. Joseph Bequaert, Harold Morrison and T. B. Mitchell, Prof. H. A. Scullen, Mr. C. N. Ainslie 1 Published by permission of the Chief of the Bureau of Entomology and Plant Quarantine. Received November 8, 1935. Fes. 15, 1936 SANDHOUSE: AGAPOSTEMON el and the late Mr. C. L. Fox (whose collection is now in the California Academy of Sciences). , I have also studied the types in the Academy of Natural Sciences of Philadelphia and wish to take this opportunity to express my ap- preciation of the courtesies shown me by Mr. E. T. Cresson, Jr., dur- ing my visit to that institution. Dr. T. D. A. Cockerell and the late Dr. James Waterston kindly compared specimens with the types of Smith’s species in the British Museum. To Mr. E. P. Van Duzee I am indebted for the opportunity of examining certain paratypes from the collection of the California Academy of Sciences. Genus AGAPOSTEMON Guerin Agapostemon Guerin, Iconog. Regne Animal de G. Cuvier, Insects 3: 448. 1844. Genotype, Apis (Andrena) femoralis Guerin. (Monobasic) Agapostemon F. Smith, Cat. Hymen. Ins. Brit. Mus. 1: 85-86, pl. 4, figs. 1-4. 1853.—Provancher, Natur. Canad. 13: 203. 1882.—Petite Faune Ent. Canad., Hymen., p. 703. 1883.—Robertson, Trans. Acad. Sci. St. Louis 7: 325. 1897.—Crawford, Proc. Nebr. Acad. Sci. 7: 159. 1901. HALICTI INTERMEDII, Groupe Agapostemon Vachal, Misc. Ent. 11: 89. 1903. —19: 12. 1911. The name Agapostemon was proposed by Guerin in 1844 for a subgenus of Apis with Apis (Andrena) femoralis Guerin the only species included, al- though he mentioned having seen other species with the characters which he uses to define the subgenus. His definition is as follows: ‘‘Nous connaissons plusieurs espéces a cuisses ainsi renflees. Ce sont des males. Peut-etre jugera-t- on a propos de les reunir en un sous-genre, que nous proposerions de nomer Agapostemon. Il serait aux Andrenes ce qu’est le genre Nomia parmi les Ha- lictes.” Frederick Smith (1853) was the first to give Agapostemon generic rank, in this being followed by Cresson, Robertson, Dalla Torre, Cockerell and Crawford; he included in it seven species, four of which were described as new. His discussion of generic characters was based chiefly on those of the head, especially of the trophi. Vachal (1903) treated Agapostemon as a group or subgenus of Halictus (Halictt agapostemones). This genus can be separated from the other genera of the Halictinae rep- resented in the nearctic fauna by the length of the posterior tibia, which is as long as, or longer than, the combined length of the tarsal joints, while in the others it approximates more nearly that of the metatarsus alone. The sexual dimorphism is so great that some characters will have to be given separately for each sex. The characters of the genus as here limited are as follows: Head, when viewed from the front, appearing rounded except where the clypeus extends below the lower margins of the eyes. Eyes large and bare, forming the lateral boundaries of the head for most of its length; their inner margins quite strongly emarginate. Front foveolate-punctate, when viewed laterally, level with the eyes, occupying about one-half the space from vertex 72 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 26, NO. 2 to apex of clypeus. Clypeus and postclypeus of equal length, strongly con- vex, with well-separated punctures; clypeus extending at least half its length below the lower margins of the eyes; apical margin truncate. Gena (malar space) very short. Postgenae declivous behind the eyes, striate-punctate. Labrum of female with basal portion as wide as truncate apex of clypeus, sides nearly parallel; apical portion about one-third as wide as basal, sides converging toward apex and fringed with curved bristles; labrum of male nearly triangular, sides slightly convex, basal portion with transverse eleva- tion. Mandible of female strongly curved, apically bidentate, inferior tooth much larger and extending beyond superior; mandible of male edentate, narrowing gradually to pointed apex. Antennae inserted about half way between apical margin of clypeus and postocellar line; scape of female a little more than one-third length of antenna, scape of male about as long as joints two to four. General contour of thorax more regular than that of honeybee (see Snod- grass, Anatomy and physiology of the honeybee, fig. 231, 1925.), that of female more robust than that of male. Prothorax showing no striking modifications; posterior margin of prothoracic lobe heavily fringed with hair. Mesoscutum of male with uniform, nearly contiguous punctures, varying little among the species; that of female closely uniformly punctate or with punctures of two distinct sizes, the punctation specifically distinct. Metatergum irregularly foveolate. Mesopleura foveolate. Metapleura of female with somewhat ir- regular, but principally horizontal low carinae, of male foveolate. Propodeum extending caudad to a distance about equal to length of metatergum and thence abruptly declivous to attachment of abdomen; posterior surface ir- regularly carinate, enclosed by a sharply defined carina, more strongly de- veloped in female; dorsal surface lacking the somewhat crescentic disk or enclosed space found in many of the other halictine bees; sculpturing con- sisting of low carinae variously arranged; lateral surfaces with nearly hori- zontal low carinae, between them rows of small but deep punctures. Tegula smooth, ovoid, about same color as basal wing-veins. Wings varying little from those of related genera and little within the genus; hyaline, slightly yellowish infumate, usually slightly darker apically; second cubital cell of — male distinctly narrower. Legs of female and front and middle legs of male not varying in structure within the genus; hind legs of male showing greatest modification and vary- ing with the species. In hind leg of female, posterior surface of femur with a single row of very long and strongly curled branched hairs; anterior surface with several rows of shorter branched hairs; hind tibia with knee-plate obsolescent, posterior margin with long simple hairs, anterior margin with variously branched hairs; inner calear pectinate, usually with from three to five broad spatulate teeth. Hind leg of male with femur usually thickened, distinctly wider than trochanter and toothed below near apex; tibia broader than metatarsus; metatarsus frequently enlarged and toothed; first and second joints of tarsus coalescent. Abdomen of female broadly ovoid, length as ordinarily extended about twice its width at apex of second tergite; tergites finely and uniformly punctured; fifth tergite with median rima, laterally densely pubescent; sixth tergite with well-defined pygidial area; second to fifth with basal fasciae of pale appressed hair. Abdomen of male more slender and frequently curved downward at apex; seven tergites, but only six sternites exposed; tergites uniformly punctured and pubescent, seventh with a distinct polished oval pygidial area bounded apically and laterally by a carina; sternites one to Fes. 15, 1936 SANDHOUSE: AGAPOSTEMON 73 six varying little within the genus, seventh and eighth very small and lying against ventral surface of basal ring, with slight specific differences. In both sexes head and thorax brilliant blue-green (as in Chrysis); ab- dominal tergites of female concolorous with thorax, or black or brown, ter- gites of male transversely banded black and yellow, the black sometimes obscurely tinged with blue-green. Legs of female brown or yellowish brown, of male yellow variously marked with black. The genitalia of the male are specifically distinct and a study of them has assisted greatly in determining the amount of variation within the species. Since the paper by Beck (Proc. Utah Acad. Sci. 10: 89-137, pls. I-VII. 1933) gives the results of rather extensive studies in homologies of parts of the male genitalia of bees, his terminology was used in labeling these parts. For a description of the genitalia of a species of Agapostemon, see page 109 and plate VII, figures 168-169 of his paper. In the present paper dorsal and ven- tral views of the entire genitalia are given for Agapostemon virescens (Fabri- cius), since they approach most nearly those of the genotype, femoralis (Guerin), and for each species a dorsal and slightly caudal view of only the distal portion of the coxopodite and stylus, since they present the greatest specific differences. KEY TO THE SPECIES OF AGAPOSTEMON OCCURRING IN THE UNITED STATES L) . EQTMAISS 0, ob Re ene Renn ieee Se 2 BES. 2 oc) eye Sug eo este ge Daag lage ek ae eae ee 10 2. Abdominal tergites not concolorous with the thorax, but black, brown 7 PESTA GE OTIS Bate eee Ne cet as Sala A 7 CRY Ota) ok 5) Abdominal tergites concolorous with the thorax, brilliant green or UEC C i i eee aes oe ae a ae eg 6 3. Apex of clypeus black. Abdominal tergites entirely black; base of first tergite with lateral patches of white hair; hair on apices of tergites black. Posterior surface of propodeum with oblique carinae. Front and middle legs dark brown. Tegular and wing-veins brown-testa- Apex of clypeus yellow, or black and yellow. Abdominal tergites testa- ceous, brown, or, if nearly black, always tinged with brown or blue- green, especially apically; base of first tergite with a wide band of dense white hair; hair on apices of tergites white, except on the fifth, where it is yellowish or brownish. Posterior surface of propodeum irregularly carinate. Front and middle legs largely or partly yellow. Tegula and wing-veins yellow-testaceous...................... 5 4. Base of mandible yellow. Postgenae laterad of the hypostomal carinae with several moderately coarse striae. Hair on posterior legs strongly infuscated. Head and thorax brilliant green, usually not at all bluish. Carinae on dorsal surface of propodeum not forming a median tri- angle. Punctures on abdominal tergites separated by twice the diam- eter of a puncture. Species smaller, 11 to 12mm. long.............. 5 Ss otk Ble ke CON 0m 8 2 sce a aa Oe a virescens (Fabricius) 74 10. JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES’ VOL. 26, NO. 2 Base of mandible reddish black. Postgenae laterad of hypostomal carinae very finely striate. Hair on posterior legs yellowish white. Head and thorax green, usually strongly tinged with blue. Carinae on dorsal surface of propodeum forming a median triangle. Punc- tures on abdominal tergites separated by the diameter of a puncture. Species larger, 14 to 15mm. long.............. coloradinus (Vachal) Scape largely yellow. Femora and abdomen testaceous. Yellow of cly- peus extending upwards in the middle to form a triangle; apical mar- ein yellow 20-4 =. Sasi hae ee ee melliventris Cresson Scape dark brown. Femora and abdomen dark brown, the apex of the latter faintly tinged with blue-green. Yellow of clypeus not extending upwards in the middle; apical margin black 7....-. 0...) Sa Ca pleats etna can bartt ke ne melliventris var. plurifasciatus (Vachal) Mesoscutum with punctures nearly uniform in size and nearly con- PEP UOUS seas Aeee a eweeew Spee lg ee tS ve Mesoscutum with well-separated punctures of two distinct sizes... .9 Mesoscutum not at all rugose between punctures, more finely and uni- formly punctured. Abdominal tergites appearing dull, with punc- tures separated by less than their diameter. Dorsal surface of pro- podeum dull, with irregularly anastomosing carinae. Pubescence strongly tinged with ochreous. Wings dusky, especially at the apices Me Ee es ee es ke splendens (Lepeletier) Mesoscutum rugose between punctures, more coarsely punctured, foveolate-punctate on the anterior and lateral portions. Abdominal tergites shining between punctures, which are separated by more than their diameter. Dorsal surface of propodeum shining, with longitu- dinal carinae. Pubescence and wings paler....................-. 8 Species larger, usually about 12 to 13 mm. long; blue-green. Pubescence white. Sixth abdominal tergite with hair on the basal third entirely PDE soa sles ARE eta eek wae ee ea ee ee cockerells Crawford. Species smaller, usually less than 10 mm. long; green, not at all bluish. Pubescence distinctly yellowish. Sixth abdominal tergite with hair fuscous, except for a small patch of yellowish hair at each side; no pale hair on the median basal portion.......... radiatus (Say) Smaller punctures of mesoscutum usually separated by about their diameter. Dorsal surface of propodeum irregularly carinate, usually with a distinct median triangular area. Species larger, about 12 mm. long. Pubescence slightly tinged with yellow...... texanus Cresson Smaller punctures of mesoscutum usually separated by at least twice their diameter. Dorsal surface of propodeum with longitudinal ca- rinae. Species smaller, about 10 mm. long. Pubescence pure white. . wa aioe telson a eh eee eee angelicus Cockerell Base of first abdominal tergite usually of a brownish tint, but never distinctly black; dark bands on the intermediate tergites scarcely Fes. 15, 1936 SANDHOUSE: AGAPOSTEMON 15 FL. 12. 13. 14, 15. one-third the length of a tergite. Legs distad of the trochanters pale, except for a brownish spot at the apex of the hind femur and one at base of hind tibia; hind femur hardly wider than the trochanter, the tooth near the apex weakly developed. Wings clear testaceous....11 Base of first abdominal tergite black; dark bands on the intermediate tergites fully one-half the length of a tergite. Legs distad of the tro- chanters conspicuously marked with black; hind femur distinctly wider than trochanter, the tooth near the apex strongly developed. Mer -uibe coronene) y iniumaueds: 6.0 2c eb ek ct 12 Trochanters of front and middle legs yellow, of the hind legs tinged with green. Scape entirely yellow, or with a small brownish dot on upper side near the apex. Dark bands on the abdominal tergites not fezenime the lateral mareins................ melliventris Cresson Trochanters of all the legs green. Scape with the upper side brown. Dark bands on the abdominal tergites reaching the lateral margins. . MIN rhe a Ses melliventris var. plurifasciatus (Vachal) Dark bands on the abdominal tergites strongly tinged with metallic blue-green, which is especially conspicuous laterally on the apical SELULEGGS 2: Gee 4 ee REE ee ee pe, aaa ae heen ae - 13 Dark bands on the abdominal tergites dull black, with no metallic GULLS - 2 2/2 cco: aka eR ps AGRReA as =p UO Sas Beas ee ne, Gear 14 Species larger, 11 to 12 mm. long, usually more yellowish green, with pubescence slightly yellowish. Dorsal surface of propodeum with a distinct triangle in the middle. Front and middle trochanters with varying amounts of black and yellow; if largely black, then there are marks of black on the bases of the femora; hind tibia always with a long black mark on the anterior surface, often also with one on the Pe METMOTMSMEEACE. <6 8 ord. Gael. Peg le 4 ee texanus Cresson Species smaller, about 9 mm. long, usually more bluish green, with pubescence pure white. Dorsal surface of propodeum without a median triangle, the carinae coarser. All the trochanters black, but no black on the bases of the femora; hind tibia with a long black mark on the posterior surface, but never with one on the anterior surface i ee Lajethed wich bol ei ed ee angelicus Cockerell Hind femur swollen so strongly that its width is distinctly more than eee ter inoretclemmpne eye bh eee Wel ey on. 15 Hind femur not so strongly swollen, its width less than one-third of its ERG 8 ek tie ds os SR Gane ae aan ooh «a a 16 Black bands on the abdominal tergites occupying more than half of their length; fifth and sixth abdominal sternites largely black. Hind femur a little more than one-half as wide as long. Dorsal surface of propodeum dull, with irregularly anastomosing carinae. Wings dusky ey Fe Ne ee ee eS splendens (Lepeletier) Black bands on the abdominal tergites occupying less than half of 76 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 26, NO. 2 their length; basal two-thirds of fifth sternite and most of sixth yel- low. Hind femur about three-fourths as wide as long. Dorsal surface of propodeum more polished, with carinae on the entire length on the middle and laterally on the basal fourth principally longitudinal. Wines clear hyaline 2 Uiy0 aa ee eee cockerella Crawford 16. Species smaller, about 9 mm. long or less. Dorsal surface of propodeum with longitudinal carinae. Punctures of mesoscutum contiguous, giving a dull, almost velvety appearance. Abdominal sternites more than half yellow; apical margin of the fourth with a median green spot, laterally with two stout bristles on each side. Front and middle trochanters largely yellow; bases of femora yellow. .radiatus (Say) Species larger, about 11 mm. long. Dorsal surface of propodeum with irregularly anastomosing carinae. Punctures of mesoscutum more widely separated, the interspaces shining. Abdominal sternites largely black; apical margin of the fourth lacking the green spot in the middle between the lateral bristles. All the trochanters black; bases of femora strongly marked with black. o....2...2.2.5 3) 4 4 17 17. Black markings on the posterior surfaces of the front and middle fe- mora confined to the basal halves, of the hind femur to the apical third; posterior surface of the hind tibia yellow, except for a small black spot near the apex; front and middle tibiae with basal spots of black. Abdominal tergites with much black hair apically. Second and third sternites with hair quite sparse; sixth with a low median carina. Head and thorax brilliant green, usually not at all bluish. Carinae on the dorsal surface of the propodeum not forming a median tri- anoular Vara: . epee) re one ee eae! virescens (Fabricius) Black markings on the posterior surfaces of all the femora and tibiae extending nearly the length of the surfaces. Abdominal tergites with hair largely white. Second and third sternites with conspicuous apical fringes of white hair; sixth without a median carina. Head and thorax distinctly bluish green. Carinae on dorsal surface of propodeum form- ing aA median trtanele. au se ee ee coloradinus (Vachal) Agapostemon virescens (Fabricius) Andrena virescens Fabricius, Syst. Ent., p. 378, n. 12. 1775.—Spec. Insect. 1:474, n. 16. 1781.—Mant. Insect. 1: 299, n. 18. 1787.— Olivier, Encyel. Method. Ins.; Hist. Nat. Ins. 1: 137, n. 23. 1789.—Fabricius, Ent. Syst. 2: 314, n. 28. 1798. Apis (Andrena) virescens Gmelin, Linné, Syst. Nat., Ed. 13a, 1 (pt. 5): 2792, n. 185. 1790. Apis virescens Christ, Natur. d. Insect., p. 154. 1791. Andrena nigricornis Fabricius, Ent. Syst. 2: 313, n. 28. 1793.—Coquebert, Illustr. leonogr. Ins. 2:63, 115; fie. 7: 180K, Megilla virescens Fabricius, Syst. Piez., p. 333, n. 23. 1804. Centris nigricornis Fabricius, Syst. Piez., p. 360, n. 33. 1804. Hylaeus nigricornis Klug, Magaz. f. Insectenk. 6: 222. 1807.—Magaz. Ges. naturf. Fr. Berlin 2: 57, n. 85. 1808. Fes. 15, 1936 SANDHOUSE: AGAPOSTEMON 77 Halictus nigricornis Say, Boston Jour. Nat. Hist. 1: 394, n. 1. 1837.—Le- conte, Writ. Thomas Say 2: 772, n. 1. 1859. Halictus dimidiatus Lepeletier, Hist. Nat. Ins. 2: 283, n. 24. 1841. Agapostemon nigricornis Smith, Cat. Hymen. Ins. Brit. Mus. 1: 86, n. 1. 1853.—Cresson, Trans. Amer. Ent. Soc. suppl., p. 293. 1887. ?Agapostemon tricolor Provancher (not Lepeletier), Natur. Canad. 13: 203. 1882.—Petite Faun. Ent. Canad. Hymen., p. 703. 1883. Augochlora radiata Provancher, Natur. Canad. 13: 205. 1882.—Petite Faun. Ent. Canad. Hymen., p. 705. 1883.—Dalla Torre, Cat. Hymen. 10: 96 (in part). 1896. Apis viridula Cresson (not Fabricius), Trans. Amer. Ent. Soc., suppl., p. 309. 1887. Agapostemon bicolor Robertson, Trans. Amer. Ent. Soc. 20: 148. 1893.— Dalla Torre Cat. Hymen. 10: 97. 1896. Agapostemon viridula Robertson (not Fabricius), Trans. Amer. Ent. Soc. 22: 118. 1895. Agapostemon virescens Dalla Torre, Cat. Hymen. 10: 98 (in part). 1896.— Cockerell, Ann. Mag. Nat. Hist. (9) 8: 363. 1921. Agapostemon viridulus Robertson, Trans. Acad. Sci. St. Louis 7: 325. 1897. —Crawford, Proc. Nebr. Acad. Sci. 7: 173. 1901. Halictus (Agapostemon) viridulus Vachal, Misc. Ent. 11: 90, 101. 1903. Halictus (Agapostemon) virescens Viereck, Conn. State Geol. & Nat. Hist. Survey Bull. 22 (pt. 3): 704. Type.—Female, in the Banksian Collection at the British Museum, where it was seen by Cockerell in 1921 and the identity of the species confirmed. The present locations of the types of nzgricornis and of dimidiatus are un- known to the writer. The types of bicolor are in Robertson’s collection. Distribution.—Insofar as is known, distributed throughout the United States from coast to coast north of 40 degrees latitude, extending down the Mississippi Valley and eastern slope of the Rocky Mountains to Louisiana and Texas. About 700 specimens have been examined from the following states: Maine, New Hampshire, Vermont, Massachusetts, Connecticut, New York, New Jersey, Pennsylvania, Maryland, Virginia, District of Columbia, North Carolina, Alabama, Kentucky, Tennessee, Ohio, Michigan, Indiana, Illinois, Iowa, Minnesota, Missouri, Louisiana, Texas, Colorado, Kansas, Nebraska, North Dakota, South Dakota, Montana, Wyoming, Idaho, Oregon, and Washington. Of the nearctic species, virescens and the closely related coloradinus are nearest to the genotype, femoralis. Virescens can be separated from colora- dinus by the characters given in the key. The male is readily distinguished by the median carina on the sixth sternite. The female is the only one having a black abdomen which is widely distributed throughout the United States. Agapostemon coloradinus (Vachal), n. comb. Agapostemon coloradensis Crawford, Proc. Nebr. Acad. Sci. 7: 163. 1901.— Cockerell, Ann. Mag. Nat. Hist. (7) 19: 532. 1907. Halictus (Agapostemon) coloradinus Vachal, Mise. Ent. 11: 90. 1903. (Pro- posed for Halictus (Agapostemon) coloradensis Crawford, not Halictus (Augochlora) coloradensis Titus.) 78 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 26, No. 2 Agapostemon tyler Cockerell, Ann. Mag. Nat. Hist. (8) 20: 241. 1917 (new synonymy). Agapostemon martini Cockerell, Pan-Pacific Ent. 3: 153, female only. 1927 (new synonymy). Type.—Female (lectotype selected by Crawford), southern Colorado, in the collection of the United States National Museum. The specimen on which Cockerell based the description of the male of coloradinus and the type and ‘‘cotype”’ (allotype) of tylerz are also in this collection. The type of martini is in the collection of the California Academy of Sciences. Distribution.—Apparently limited to the southern Rocky Mountain region from South Dakota and Colorado to western Texas, southern Arizona and Mexico. Only 38 specimens have been seen from the following states: Texas, Colorado, Nebraska, South Dakota, Utah and New Mexico; also from Mex- ico. This species is closely related to virescens and apparently replaces it in the southwestern part of the United States. The type of martini was not seen, but the description of the female agrees well with the type of coloradinus. Both sexes of tylert have been compared with coloradinus and found to be identical. Agapostemon melliventris Cresson Agapostemon melliventris Cresson, Trans. Amer. Ent. Soc. 5: 101. 1874.— Rept. Geogr. & Geol. Explor. & Surv. 100th Merid. 5: 721, pl. 33, fig. 4. 1875.—Trans. Amer. Ent.Soc., suppl. p. 298. 1887.— Dalla Torre, Cat. Hymen. 10: 97. 1896.—Cockerell, Trans. Amer. Ent. Soc. 24: 146. 1897. —Crawford, Proc. Nebr. Acad. Sci. 7: 164. 1901. Agapostemon digueti Cockerell, Proc. Calif. Acad. Sci. 12: 539. 1924 Git synonymy). Type.—Female, lectotype, Nevada, in the collection of the Academy of Natural Sciences, Philadelphia. The types of dzgweti are in the collection of the California Academy of Sciences. The synonymy of diguetz is based upon a study of paratypes of both sexes in the collection of the United States National Museum. Distribution.—Apparently limited to the extreme southwestern part of the United States and northern Mexico. In the United States extending from southern Texas to southern California and north to Utah and Oklahoma. The variety plurifasciatus replaces the typical form in northeastern Colorado northwestern Kansas and Nebraska. About 400 specimens have been ex- amined from the following states: Texas, Oklahoma, Colorado, Nebraska, Utah, New Mexico, Arizona and California; also from Lower California and Mexico. Agapostemon melliventris var. plurifasciatus (Vachal), n. comb. Agapostemon fasciatus Crawford, Proc. Nebr. Acad. Sci. 7: 163. 1901. Halictus (Agapostemon) plurifasciatus Vachal, Misc. Ent. 11: 93, 101. 1903. (Proposed for Halictus (Agapostemon) fasciatus Crawford, not Halictus fasciatus Nylander.) Frs. 15, 1936 SANDHOUSE: AGAPOSTEMON 79 Type.—Female and allotype, male (lectotypes selected by Crawford), from Lincoln, Nebraska, in the collection of the United States National Museum. Distribution.—I have seen specimens of this variety only from Lincoln, Nebraska; Sterling, Colorado; and Clay County, Kansas. This differs from melliventris only in color as given in the key. Since no morphological differences could be found, plurifasciatus 1 is considered to be a color variety of melliventris. Agapostemon splendens (Lepeletier) Halictus splendens Lepeletier, Hist. Nat. Ins. Hymen. 2: 283, n. 25. 1841.— Cresson, Trans. Amer. Ent. Soc., suppl., p. 293. 1887.—Dalla Torre, Cat. Hymen. 10: 85. 1896. Agapostemon aeruginosus Smith, Cat. Hymen. Ins. Brit. Mus. 1: 86, n. 3. 1853.—Cresson, Trans. Amer. Ent. Soc., suppl., p. 293. 1887.—Dalla Torre, Cat. Hymen. 10: 97. 1896. Agapostemon nigricornis Robertson (not Fabricius), Trans. Amer. Ent. Soc. 20: 147. 1893.—Dalla Torre, Cat. Hymen. 10: 97. 1896. Agapostemon splendens Robertson, Trans. Acad. Sci. St. Louis 7: 328. 1897. —Crawford, Proc. Nebr. Acad. Sci. 7: 161. 1901.—Howard, Insect Book, pl. 3, fig. 14. 1905.—Graenicher, Ann. Ent. Soc. Amer. 23: 158, 168. 1930. Halictus (Agapostemon) aeruginosus Vachal, Misc. Ent. 11: 95. 1903. Halictus (Agapostemon) splendens Vachal, Misc. Ent. 11: 95. 1903. Halictus (Agapostmeon) nigricornis Vachal, Misc. Ent. 11: 100. 1903. Type.—Female. ‘‘Carolina.’’ The present location of the type is unknown to the writer. The type of aerugznosus is in the British Museum where it was compared by Cockerell and Waterston with specimens of splendens and the synonymy confirmed. Distribution.—Insofar as is known, distributed throughout the eastern and central United States from southern New Hampshire to southern Flor- ida and west to Texas and eastern Colorado. No specimens have been seen from north of the 45th degree of latitude nor west of the 105th meridian. Over 300 specimens have been examined from the following states: New Hampshire, Massachusetts, Connecticut, New York, New Jersey, Mary- land, Virginia, North Carolina, Georgia, Florida, Alabama, Michigan, In- diana, Illinois, Iowa, Minnesota, Missouri, Arkansas, Louisiana, Texas, Oklahoma, Colorado, Kansas, Nebraska, and Arizona. Agapostemon cockerelli Crawford Agapostemon cockerelli Crawford, Proc. Nebr. Acad. Sci. 7: 161. 1901.— —Cockerell, Pan-Pacific Ent. 3: 155, female only. 1927. Agapostemon femoratus Crawford, Proc. Nebr. Acad. Sci. 7: 162. 1901.— Cockerell, Pan-Pacific Ent. 3: 157. 1927 (new synonymy). Agapostemon radiatus Cockerell (not Say), Ent. News 9: 27. 1898 (new synonymy). Agapostemon californicus Crawford, Proc. Nebr. Acad. Sci. 7: 164, female only. 1901 (new synonymy). 80 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 26, NO. 2 Agapostemon pulcher Robertson (not Smith), Canad. Ent. 34: 49. 1902 (new synonymy). ; ? Nomia cillaba Cameron, Trans. Amer. Ent. Soc. 28: 376. 1902 (new synonymy). Halictus (Agapostemon) ? pulcher Vachal, Misc. Ent. 11: 94. 1903. ? Halictus (Agapostemon) cockerelli Vachal, Misc. Ent. 11: 95. 1903. Halictus (Agapostemon) femoratus Vachal, Misc. Ent. 11: 100. 1903. ? Agapostemon cillaba Cockerell, Ann. Mag. Nat. Hist. (8), 4: 311. 1909. Agapostemon martina Cockerell, Pan-Pacific Ent. 3: 153, male only. 1927. (new synonymy). Type.—Female, holotype, from Mesilla Park, New Mexico, in the col- lection of the United States National Museum. The type (lectotype, selected by Crawford) of femoratus from Moscow, Idaho, and a paratype of martina are also in this collection. The type of cillaba is in the British Museum. The allotype (‘‘cotype’’) of martinzis in the collection of the California Academy of Sciences. Cockerell’s designation (1927) of a type locality for femoratus can- not be considered as a true type fixation, as it was based upon a selection of a locality from literature and not from a study of any of the type series. Distribution.—Insofar as is known, distributed throughout the western part of North America, west of the 100th meridian, from British Columbia to Mexico. It apparently replaces radiatus in the west. Over 300 specimens have been examined from the following states: Texas, Colorado, North Dakota, Montana, Wyoming, Idaho, Utah, New Mexico, Arizona, Califor- nia, Nevada, and Washington. Material was also seen from Alberta and British Columbia, Canada, and from Mexico. Agapostemon radiatus (Say) Halictus radiatus Say, Boston Jour. Nat. Hist. 1: 394, n. 2. 1837.—Leconte, Writ. Thomas Say 2: 772, n. 2. 1859. Halictus tricolor Lepeletier, Hist. Nat. Ins. Hymen. 2: 289, n. 33. 1841. Augochlora radiata Smith, Cat. Hymen. Ins. Brit. Mus. 1: 80, n. 22. 1853.— Dalla Torre, Cat. Hymen. 10: 96, in part. 1896. ? Agapostemon tricolor Smith, Cat. Hymen. Ins. Brit. Mus. 1: 86, n. 2. 1853. Agapostemon pulchra Smith, Cat. Hymen. Ins. Brit. Mus. 1: 87, n. 4. 1853. —Cresson, Trans. Amer. Ent. Soc. suppl., p. 293. 1887. Agapostemon radiatus Cresson, Trans. Amer. Ent. Soc. suppl., p. 293. 1887. —Robertson, Trans. Amer. Ent. Soc. 20: 147, in part. 1893.—Dalla Torre, Cat. Hymen. 10: 97. 1896.—Robertson, Trans. Acad. Sci. St. Louis 7: 327. 1897.—Crawford, Proc. Nebr. Acad. Sci. 7: 163, in part. 1901.—Howard, Insect Book, pl. 3, fig. 11. 1905.—Lutz, Fieldbook of Insects, pl. xciv. 1918 (1st ed.), 1921 (2d. ed.). Agapostemon tricolor Robertson, Trans. Amer. Ent. Soc. 20: 148. 18938. Agapostemon pulcher Dalla Torre, Cat. Hymen. 10: 97. 1896. Halictus (Agapostemon) radiatus Vachal, Misc. Ent. 11:95, 102, 104. 1903. ?Halictus (Agapostemon) cockerelli Vachal, Misc. Ent. 11: 94. 1903. Agapostemon sulcatulus Cockerell, Ann. Mag. Nat. Hist. (8), 4: 25. 1909 (new synonymy). Frs. 15, 1936 SANDHOUSE: AGAPOSTEMON 81 Fig. 1—Agapostemon virescens (Fabricius). Male genitalia, dorsal view. Fig. 2.— A. virescens. Male genitalia, ventral view. BR, Basal Ring; BB, Basal Bridge; GF, Genital Foramen; Cox., Coxopodite; St., Stylus; Vol., Volsella; Par., Paramere; Pen., Penis. Fig. 3.—A. virescens. Distal portion of coxopodite, dorsal view. Fig. 4.— A. angelicus Cockerell. Distal portion of coxopodite, dorsal view. Fig. 5.—A. splen- dens (Lepeletier). Distal portion of coxopodite, dorsal view. Fig. 6.—A. coloradius (Vachal). Distal portion of coxopodite, dorsal view. Fig. 7A. texanus Cresson. Dis- tal portion of coxopodite, dorsal view. Fig. 8.—A. melliventris Cresson. Distal portion of coxopodite, dorsal view. Fig. 9.—A. radiatus (Say). Distal portion of coxopodite, dorsal view. Fig. 10.—A. cockerelli Crawford. Distal portion of coxopodite, dorsal view. The illustrations were made by Mrs. Eleanor A. Carlin of the Bureau of Ento- mology and Plant Quarantine. 82 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 26, No. 2 Type.—Female, Indiana, probably destroyed. The location of the type of tricolor is unknown to the writer. The type of pulchra is in the British Museum, where it was seen by Cockerell and Waterston and its synonymy confirmed. They felt that the locality ‘“‘California”’ on the type of pulchra must have been erroneous. The type of sulcatulus is in the United States National Museum. Distribution.—Insofar as is known, distributed throughout the eastern part of the United States, east of the 105th meridian, from Maine to Georgia and from North Dakota to northern Texas. It is apparently replaced in the western states by cockerell:z. About 650 specimens have been examined from the following states: Maine, New Hampshire, Vermont, Massachusetts, Rhode Island, Connecticut, New York, New Jersey, Pennsylvania, Mary- land, District of Columbia, Virginia, North Carolina, South Carolina, Geor- gia, Alabama, Kentucky, Tennessee, Ohio, Michigan, Indiana, Illinois, Iowa, Minnesota, Wisconsin, Missouri, Louisiana, Texas, Oklahoma, Colo- rado, Kansas, Nebraska, North Dakota, South Dakota, and New Mexico. Agapostemon texanus Cresson Agapostemon texanus Cresson, Trans. Amer. Ent. Soc. 4: 255, in part. 1872.—Trans. Amer. Eng. Soc. suppl., p. 298. 1887.—Dalla Torre, Cat. Hymen. 10: 97. 1896.—Robertson, Trans. Acad. Sci. St. Louis 7: 325. 1897.—Crawiord, Proc. Nebr. Acad. Sci. 7: 160, (? in panty 1901.—Howard, Insect Book, pl. 14, fig. 2. 1905. Agapostemon texanus subtilior Cockerell, Ent. News 9: 27. 1898.—Craw- ford, Proc. Nebr. Acad. Sci. 7: 160. 1901( new synonymy). Agapostemon californicus Crawford, Proc. Nebr. Acad. Sci. 7: 164, male only. 1901 (new synonymy). Agapostemon borealis Crawford, Proc. Nebr. Acad. Sci. 7: 160. 1901.— Cockerell, Pan-Pacific Ent. 3: 156. 1927 (new synonymy). Halictus (Agapostemon) borealis Vachal, Misc. Ent. 11: 94. 1903. Halictus (Agapostemon) texanus Vachal, Misc. Ent. 11:94. 1908. Halictus (Agapostemon) subtilior Vachal, Misc. Ent. 11: 95, 102, 104. 1903. Agapostemon texanus towensis Cockerell, Ann. Mag. Nat. Hist. (8), 5: 363. 1910 (new synonymy). Agapostemon texanus vandyke: Cockerell, Proc. Calif. Acad. Sci. 14: 191. 1925 (new synonymy). Agapostemon cockerelli Cockerell, not Crawford, Pan-Pacific Ent., 3: 155, male only. 1927 (new synonymy). Agapostemon vandykei Cockerell, Pan-Pacific Ent. 3: 155. 1927. Type.—Female, lectotype, from Texas, in the collection of the Academy of Natural Sciences of Philadelphia. The type of subtilior is probably in Cockerell’s collection. The type of borealis is in the Academy of Natural Sciences of Philadelphia. The type of zowensis and the lectotype (selected by Crawford) of californicus are in the United States National Museum. The type of vandykez is in the collection of the California Academy of Sciences. Distribution.—Insofar as is known, distributed throughout the United Fess. 15, 1936 SANDHOUSE: AGAPOSTEMON 83 States from coast to coast north of the 40th degree of latitude, extending southward along the Appalachian Mountains-to North Carolina, along the Rocky Mountains to southern Texas and Mexico, and along the Pacific Coast Ranges to southern California. About 1000 specimens have been ex- amined from the following states: New Hampshire, Connecticut, New York, Pennsylvania, North Carolina, Michigan, Indiana, Illinois, lowa, Minnesota, Wisconsin, Louisiana, Texas, Oklahoma, Colorado, Kansas, Nebraska, North Dakota, South Dakota, Montana, Wyoming, Idaho, Utah, New Mexico, Arizona, California, Oregon, Nevada, and Washington. Agapostemon angelicus Cockerell Agapostemon angelicus Cockerell, Proc. Calif. Acad. Sci. 12: 537. 1924.— Pan-Pacific Ent. 3: 156. 1927. Agapostemon texanus Cresson, Trans. Amer. Ent. Soc. 4: 255, in part. 1872.—? Authors, in part (new synonymy). Halictus (Agapostemon) teranus Vachal, Misc. Ent. 11: 94, ? in part. 1903. Type.—Female, Pond Island, Bay, Angel de la Guarda Island, in the col- lection of the California Academy of Sciences. Distribution.—Apparently limited to the southern Rocky Mountain re- gion from southern Texas to North Dakota and Idaho and along the Pacific Coast from southern California to the 40th degree of latitude. Over 400 specimens have been examined from the following states: Texas, Colorado, Kansas, Nebraska, North Dakota, South Dakota, Wyoming, Idaho, Utah, New Mexico, Arizona, and California. This species is very similar to texanus and has undoubtedly been confused with it in most collections. The females may be distinguished by the sculp- turing of the dorsal surface of the propodeum; the males, by the markings on the hind legs. 84 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 26, NO. 2 PROCEEDINGS OF THE ACADEMY AND AFFILIATED SOCIETIES BOTANICAL SOCIETY 267TH MEETING The 267th regular meeting was held in the assembly hall of the Cosmos Club, October 1, 1935, President W. W. Dinuu presiding, attendance 70. Notes and reviews.—J. B. S. NorTON reviewed the recently issued volume 2 of Hutchinson’s The families of flowering plants. Program.—W. C. LowpERMILK: Notes on the Third International Con- gress of Soil Science, Oxford, July 30-Aug. 7, 1935. CHARLES DRECHSLER: Notes on the Deutsche Botanische Gesellschaft, Cologne, Aug. 28—Sept. 2, 1935. Reports on the Sixth "International Botanical Congress, at Amsterdam, Sept. 1-7 were made as follows: Phytopathology by CHARLOTTE ELLIOTT; physiology by A. M. Hurp-Karrer; taxonomy by Doris HayzEs; my- cology and bacteriology by CHARLES THOM. 268TH MEETING The 268th regular meeting was held in the assembly hall of the Cosmos Club, November 5, 1935, President W. W. DreHu presiding, attendance 92. G. H. Coutuinewoop, 8S. L. EMSwEeLurrR, JEWELL G. Guass, Max M. Hoover, M. T. JENKINS and Haroup H. SmitTH were elected to member- ship. Notes and reviews.—ROLAND W. Brown reported the finding of a speci- men of Hamamelis in Pennsylvania with rose-colored petals. This form has been previously mentioned by Sargent. M. B. Warts discussed the competitive effects of the outdoor flower business upon the cut-flower trade, as well as certain aspects of autumnal coloration. H. B. Humpurey reported on the differences in leaf-fall and coloration between the District and Maryland. Program.—L. W. Knpuart: The place of botany in weed research. J. A. Martin: The classification of American wheats. A review of 'Techni- cal Bulletin 459 of the United States Department of Agriculture by J. A. Clark and B. B. Bayles, with a further discussion of the work of the Di- vision of Cereal Crops and Diseases of the Bureau of Plant Industry along the same lines. H. B. Humeurey: The epiphytotic of black stem rust in 1935. Discussed by Messrs. CLARK, HASKELL, LEIGHTY, LUDwic and WalITE. 269TH MEETING The 269th meeting was held in the assembly hall of the Cosmos Club, December 3, 1935, President W. W. Dirt presiding, attendance 97. Irvine W. Drx and JosepH J. Stroup were elected to membership. Program.—W. W. DieHu: An outline of mycogeography. Distribution of fungi over the earth is only partly in conformity with the better known dis- tribution of seed plants in life zones,—and then chiefly when that is neces- sitated by a restricted obligate parasitism or saprophytism. There are other provinces of fungus distribution made up of endemics which overlap these life-zones. Recognition of endemic provinces and of provinces conforming to life-zones is complicated by the great number of species of fungi which Fes. 15, 1936 PROCEEDINGS: BOTANICAL SOCIETY 85 are cosmopolitan in distribution. This cosmopolitanism is being increased through the influence of man in aiding many species to surmount the bar- riers which formerly restricted their distribution. (A wuthor’s abstract.) 35TH ANNUAL MEETING The 35th annual meeting was held immediately following the adjourn- ment of the 269th meeting. The recording secretary reported that 34 new members had been elected during the past year, bringing the membership of the society to 249, of whom 243 were active and 6 honorary. Two mem- bers passed away during the year, Davip GriFFiTHs and W. W. EGGLEsTON. A biographical sketch of the former was prepared by H. B. HuMpHREY and read by E. O. WooTEN who gave in addition personal recollections of the deceased. A. B. CLAWSON read an obituary of Mr. Eggleston. Four mem- bers who had retired from professional work during the course of the year were elected to honorary membership, Lester H. Drwny, C. L. SHuEar, WILLIAM STUART and Merton B. WalIrTE. The following officers were elected to serve for the ensuing year: Presi- dent, JoHN W. Roperts; Vice-President, CHARLES F. SwWINGLE; Recording secretary; G. F. Gravatt; Corresponding secretary, ALICE ANDERSEN; Treasurer, NELLIE W. Nance. CHARLES DRECHSLER was nominated as vice-president of the Washington Academy of Sciences representing the Botanical Society of Washington. CHARLES F. SwINGLeE, Recording Secretary 86 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 26, No. 2 @Obituary ALBERT SPEAR HITCHCOCK, principal botanist in charge of systematic agrostology, Bureau of Plant Industry, U. 8. Department of Agriculture, died on shipboard, December 16, 1935, while en route home from the 6th International Botanical Congress at Amsterdam and a period of study in European herbaria. Doctor Hitchcock was born September 4, 1856, at Owasso, Michigan. He received the degree of Bachelor of Science in Agricul- ture from Iowa State College in 1884, and the Master’s Degree in 1886. The same institution awarded the doctorate in science in 1920. Following a brief experience as an assistant in chemistry, he turned definitely to a botanical career as an instructor in botany at Iowa College for the years 1886-89. After spending the years 1889-91 as a botanical assistant at the Missouri Botanical Garden he became Professor of Botany at the Kansas Agricul- tural College, remaining in that position until 1901 when he was called to the Bureau of Plant Industry, at Washington, as Assistant Agrostologist. Merited promotion brought him the title of Agrostologist in 1905, Senior Botanist in charge of systematic agrostology in 1924 and Principal Botanist in 1928. For many years he was Custodian of Grasses of the U. 8. National Herbarium, Smithsonian Institution. During 1919 he was a member of the National Research Council. Doctor Hitchcock was not merely an herbarium botanist, but had trav- elled and botanized in the field to an extent equalled by few botanists. His travels in quest of grasses and botanical information had taken him into all the states of this country, Canada, including the Arctic regions, Labrador, Alaska, the West Indies, South America throughout its length, Europe, Hawaii, the Philippine Islands, China, Indo-China, and Japan. His numer- ous and authoritative publications have covered the field of systematic agrostology, culminating in the recently issued monographic masterpiece, Manual of the grasses of the United States. Doctor Hitchcock was president of the Botanical Society of America in 1914. He was also a member of the Washington Academy of Sciences, the Botanical Society of Washington (President, 1916), the Biological Society of Washington (President, 1923), the Washington Biologists’ Field Club, and the Kansas Academy of Sciences. zo rte gare y Somat © But ge = ea CONTENTS Botany.—Certain aan. of Venezuela: New and old specie Opuntia and Melocactus. H. PITTIER. Sea 5 ole bari ee an PaLEoBoTANy.—Field identification of the fossil ferns called Tempsky Rouanp Wes BROWN fo Os tes oe Zoowoay. —The histology of nemic esophagi. V. The esophagi of Rh ditis, Anguillulina, and Aphelenchus. B. G. Currwoop and M. FE ty CHIT WOOD... Gla Soe Bee ee of Peete 2 eee eit alluaudi and Talitrus sylvaticus, in the United States. Ri SHOEMAKER vss 0 cchtly A te Bal ae cs hee aera aS Pehaaiats ENtToMoLoGcy.—Some butterflies from eastern Virginia. Aust N Crakk and. Geral GpAR Boi. ios via, hue eos EntromoLocy.—The bees of the genus Agapostemon (Hymenopt : Apoidea) occurring in the United States. Grace A. SaNDHOUS: PROCEEDINGS: BOTANICAL SOCIETY. ......0......... a ery sane “fee OpiruaRy: ALBER? 8. HircHCGOCK. |. 5).c0 jee ee This Journal is indexed in the International Index to Periodicals | No 3 V ACADEMY UNCES ‘BOARD OF EDITORS --—- Rotanp W. Brown _ -Exsew H. Toour ss U. 8. GHOLOGICAL SURVHY BURHAU OF PLANT INDUSTRY Haroup Morrison ENTOMOLOGICAL SOCIETY W. Ww. RUBEY GEOLOGICAL SOCIETY J. R. Swanton ANTHROPOLOGICAL SOCIETY D R. E. Grsson CHEMICAL SOCIETY PUBLISHED MONTHLY 3TON ACADEMY OF SCIENCES 450: ABNAIP Sr. HA, Wisconsin we iz ee Se Journal of the Washington Academy of Sciences — This JouRNAL, the official organ of the Washington Academy ‘of Sciences, publigeae fn (1) short original papers, written or communicated by members of the Academy; (2). proceedings and programs of meetings of the Academy and affiliated societies; (3) _ notes of events connected with the scientific life of Washington. The JouURNAL is issued monthly, on the fifteenth of each month. Volumes correspond to calendar years. Manuscripts may be sent to any member of the Board of Editors. It is ursentia re- i quested that contributors consult the latest numbers of the JourNAt and conform their — manuscripts to the usage found there as regards arrangement of title, subheads, syn- — onymies, footnotes, tables, bibliography, legends for illustrations, and other matter. — Manuscripts should be typewritten, double-spaced, on good paper. Footnotes should — be numbered serially in pencil and submitted on a separate sheet. The editors do not — assume responsibility for the ideas expressed by the author, nor can they undertake OR correct other than obvious minor errors. ie! i; istration: in excess of one full-page halftone or two full-page line drawings may yu be paid for by the author. Line drawings are preferred. pt Race ae oe Proof.—In order to facilitate prompt publication one proof will generally et send a a to authors in or near Washington. It is urged that manuscript be submitted ip final — form; the editors will exercise due care in seeing that copy is followed. Author’s Reprints.—Fifty reprints without covers will be furnished gratis. Covers bearing the name of the author and title of the article, with inclusive pagination and date of issue, and reprints, will be furnished at cost when ordered, in accor with | the following schedule of prices. py Copies 4 pp. 8 pp. 12 pp. 16 pp. Covers — a, OO) nea tele Roe Mp eee $2.00. 100 § 275 $1.20 $1.50 $2.00 2.70 150 1,25 1.90 y Fes |" 3.25 3.50 200 1.75 2.60 3.60 4.50 Ps ea 250 2.00 3.30 4.65 ey 5.00 Envelopes for mailing reprints with the author’s name and address printed i in n the ‘ corner may be obtained at the following prices. First 100, $4.00; additional 100, $1. 00: Subsertption Rates:—Per volume’... 00) 4 oped ss 0 hon ae eae ee ene ie To members of affiliated societies; per volume............... pce tai a 50 “Vi ‘Single numbers... -- 1... 11s eee ee eee eect ene tenet eet e eee 3 Remittances should be made payable to ‘‘Washington Academy of Sciences’”’ ae addressed to 450 Ahnaip Street, Menasha, Wis., or to the ogetinee H. G, Bele 3 ; Coast and Geodetic Survey, Washington, DC: Exchanges.—The JouRNAL does not exchange with other publications. Missing Numbers will be replaced without charge provided that claim is made to the Treasurer within thirty days after date of following issue. OFFICERS OF THE ACADEMY President: O. E. Mutnzer, U. S. Geological Survey. Corresponding Secretary: N. R. Smita, Bureau of Plant Industry. — Recording Secretary: CHARLES THOM, Bureau of Chemistry and Soils. Treasurer: Hunry G. AVERS, Coast and Geodetic Survey. JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VoL. 26 Marcu 15, 1936 No. 3 MEDICINE.—Comings and goings of epidemics... GrorGE W. Mc- Coy, M.D., National Institute of Health.’ Although the study of disease as manifested in the individual pa- tient is one of the most interesting vocations for those whose tastes run along such lines, of far greater interest, and obviously of greater importance, is the study of disease as it affects a group, or groups, of people. The group may be small or it may be so large as to include all or most of the race. Everyone knows that the manifestations of the same disease differ with individuals and an acquaintance with this variance is often the secret in making correct diagnoses. In a similar manner, outbreaks of disease at different times and in different places show variations, the recognition of which leads to accurate epidemiological diagnoses. I shall endeavor to illustrate some of the problems in epidemic dis- eases that have been solved and to consider others that remain to be solved. By observation and study of disease as manifested in a population it is often possible to learn the factors governing the prevalence of the particular disease and, this learned, to take effec- tive measures of prevention or suppression. Epidemic is defined as ‘‘common to, or affecting at the same time, a large number in a community.’’ Generally, the term is used in refer- ence to communicable diseases but we have it applied to such miscel- laneous non-medical occurrences as traffic accidents and bank failures. In this presentation, I have had occasion to consult several stand- ard works, particularly the Bible, Hirsch’s Handbook of Geographic and Historical Pathology, Delmage’s work entitled The Nation’s Health, and, finally, Vaughan’s Epidemiology and Public Health. The latter, an exceedingly valuable work and especially useful to anyone interested in the history of epidemics, was the last important literary contribution of the late Dr. Victor C. Vaughan, a very distinguished physician, long a resident of Washington and a member of the Cosmos 1 Address of the retiring president of oe Washington Academy of Sciences delivered January 16, 1936. Received January 17, 1936. 87 WAR G6 iese 88 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 26, NO. 3 Club. I have drawn freely on all of these works and here make ac- knowledgment of these sources of information. | Delmage (p. 65) states that the principal diseases prevalent during the middle ages were “famine pestilences, plague, leprosy, ergotism, scurvy, dancing mania, syphilis, malaria and various skin diseases.” Omitting skin diseases as being too indefinite for comparison, only two of the remaining diseases he mentions are of great importance today, malaria and syphilis. I am not informed as to the exact nature of the ‘‘dancing mania”’ of those days but I have a suspicion that it is still on the decline after a recrudescence of the post war period and the era of national prohibition. Delmage makes no mention of any of the diseases we recognize as heading the list of causes of death at the present time—cancer, heart disease, pneumonia, and tubercu- losis—though doubtless they were just as prevalent then as now, but on account of the greater prevalence of other diseases did not earn for themselves a prominent place as major scourges. Epidemics may be due to unusual and rarely occurring combina- tions of circumstances. For instance, a consignment of psittacine birds is shipped from an aviary in California to an eastern department store; the birds are displayed and admired; some are purchased and taken home; then, perhaps ten days later, a few of the employees of the store, admirers, or purchasers of the birds become ill with a peculiar type of pneumonia that puzzles attending physicians. These cases are scattered in various parts of the community. At first sight there is no suspicion as to relationship between cases. Then, quite possibly by chance, it develops that the illness began a week or ten days after the arrival of the birds in the store or after bringing the new bird into the home—the bird, or birds, perhaps sickening in the meantime. The solution of the epidemic from that time forward be- comes a routine matter for the health department staff. Perhaps many similar outbreaks of this type come and go without being recog- nized. An excellent example of how a disease ordinarily not known to prevail in epidemic form may, by some peculiar combination of cir- cumstances, become epidemic is to be found in the outbreak of amebic dysentery centering in Chicago in the summer and fall of 1933. Here was a disease which the very elect among special students in the field had said never was, and indeed never could be, epidemic; and we all subscribed to the dictum. Then came one of the unforeseen, and humanly speaking unforeseeable, tragedies in public health. Through peculiar circumstances there occurred an opportunity for Marcu 15, 19386 MCCOY: EPIDEMICS 89 the drinking water supplies of two hotels in Chicago to become in- fected with the protozoal organism causing one of the types of dysen- tery. There resulted about 1200 reported cases, scattered all over the country, and of these about 100 died. There is good reason to believe that the total number of cases and deaths exceeded this considerably but it was not practicable to secure full reporting. It took quite a while even for those of the medical profession who were especially interested in the problem to admit that there could be, and was, an epidemic of this type of disease. In addition to being the first wide- spread epidemic of this disease in a civil population it made us all conscious of the possibility of impure water being involved in the causation of the disease as it ordinarily occurs. To illustrate that our concept of epidemics need not be limited to communicable diseases let me briefly discuss scurvy. We first hear of it in the 13th century in connection with the Crusaders, although doubtless it existed long before that time. According to Vaughan, Joinville (1260, p. 37) wrote: ‘“‘The barbers were forced to cut away large pieces of flesh from the gums, to enable their patient to eat. It was pitiful to hear the cries and groans of those on whom this opera- tion was performing; they seemed like the cries of women in labor.”’ The disease was a special scourge of navigators and explorers. Scarcely an expedition but suffered cruelly from the ravages of what we now know to be an easily preventable disease. Thus, Vasco da Gama’s voyage around the Cape of Good Hope, in 1497, resulted in the loss of about two-thirds of the men, mostly from scurvy. Magellan’s crews on the voyage around the world, begun in 1519, the first in history, were harrassed by the disease. Cartier’s expedition to Canada, 1535, has yielded us a clinical description of such noteworthy fidelity that I forbear reading it to you so gruesome are the details. Apparently the disease was previously unknown to Cartier and his - medical staff but they found that the juice of the bark and leaves of a certain spruce was an effective remedy, a clear hint as to the rela- tion to diet now so well established. As early as 1564, orange juice was prescribed in the treatment of scurvy, the first mention of citrus fruits in connection with the disease. Lind, a British naval surgeon, whose view was about 150 years in advance of his time, carried out (about 1750) what we would now call a feeding experiment to determine the diet most effective in the treatment of scurvy, but (and it shows that bureaucracy was pretty much the same then as now) it took another 50 years for the British Admiralty to take the necessary administrative action to profit by 90 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 26, No. 3 his findings. Lind added nothing essential to what had been known for 200 years, but he was a true scientist and furnished proof of what up to that time had been only belief. Captain Cook, in his voyage around the world from 1772-1775, lost not a man. He carried an ample supply of sauerkraut. Take your choice, sauerkraut or scurvy! The sailor of that day was not fastidious and had no difficulty in making the choice. To show you that I have a little historical data of my own, and lest someone criticize the ac- curacy of my statement about Cook’s voyage, I want to say that I know quite well that on a later voyage, 1779, Captain Cook was killed by the natives of Hawaii. I have visited the place where he is said to have been killed, and, in order to show still further my first hand information on this subject, let me say that I have visited the monu- ment erected to Captain Cook’s memory by the British government, and am aware of the fact that every few years a British cruiser is sent to Hawaii to “‘police up’’ the surroundings of the monument. Yet in spite of these experiences, several years after Captain Cook’s voyage, a 10 weeks’ cruise of the British fleet in the Bay of Biscay re- sulted in 2400 cases of scurvy—so slow is official authority to recog- nize and adopt new measures. This disease has played an important role in land military operations as well as at sea, prolonged sieges al- ways bringing it to the fore among the besieged and besiegers, always, of course, due to the same essential cause, deficiency in diet. This disease, which at one time was so prevalent, affecting civil as well as naval and military populations, has now become practically unknown in the civilized world save in what we might call the rudimentary form that occurs among infants. “Sweating sickness” is the only disease entity on record that seems to have vanished completely centuries ago. According to Delmage, this disease prevailed in England for a period of about 65 years in the latter part of the 15th and the early part of the 16th centuries. Ac- cording to the same author, ‘‘it was characterized by a very sudden onset, acute symptoms, and profuse sweating” or, as a contemporary observer puts it, by a ‘‘grete swetyng and stynking and a contynual thirst with a grete hete and headache.” It seems to have sought out by preference the highly placed. Thus, we find mentioned as victims two lords mayor and four aldermen of London; also Wolsey, the Vene- tian ambassador, and Anne Boleyn. The King, Henry VIII, appears to have escaped the affliction by keeping on the move. It sounds like the history of an infectious disease but against this is evidence that it produced no immunity as repeated attacks in the same individual Marcu 15, 1936 MCCOY: EPIDEMICS 91 are referred to. This, of course, does not absolutely exclude the in- fectious nature. The disease invaded the Continent of Europe and is said to have broken up the Diet of Worms in 1521, perhaps fortu- nately for intellectual progress and the liberalization of the minds of men. Medical historians are not able to identify this epidemic visita- tion with any earlier or later outbreak of any sort. Having given an example of a disease that has disappeared, let me now give you examples of new, or relatively new, diseases—though it is doubtful whether any disease is really new; often it is new only in the sense that it is newly recognized. We usually regard tularaemia as a new disease since not even the name was known more than a decade ago, but my colleague, Doctor Francis, who has contributed so much to the knowledge of the disease, has unearthed cases that occurred long before we had any scientific knowledge of the infection, which, as you know, usually is derived from dressing infected animals, nearly always rabbits. A friend with a much better knowledge of the Bible than I have calls my attention to the fact that among the ancient Hebrews the rabbit (coney) was forbidden as food, and argues that this prohibition could come only from a knowledge of a disease derived from the forbidden food (Leviticus, chapter 11, verse 5). The injunction in the 8th verse of the same chapter reads: ‘“‘Of their flesh ye shall not eat and their carcasses shall ye not touch.” This disease, not communicable from person to person, would hardly have been recognized as epidemic before the days of bacteri- ology and for practical purposes may be regarded as a newcomer but it is recognized now in nearly every State in the Union and in several foreign countries. For a century or more physicians in the Mediterranean area have recognized a disease of long duration and peculiar clinical character- istics which many years ago was found to be contracted usually by the drinking of goat’s milk. So firmly established was the relation be- tween the goat and the sick man that no one thought of Mediterranean fever, undulant fever, or Malta fever, as it was variously known, Save in association with possible sources of infection among goats. In this country, the occasional case recognized was traceable to goats and the distribution was limited almost entirely to the southwestern part of the country where the goats were known to be infected. Then came the brilliant work of Miss Evans and others, who showed that undulant fever might be derived from cattle infected with contagious abortion. A new disease was created, or discovered, almost overnight. Cases began to be recognized nearly everywhere in the United States 92 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 26, NO. 3 and in many other parts of the world. Whether the many cases now reported by physicians really represent a new disease in this country or whether similar cases have been with us a long time but have not been correctly diagnosed, can not be answered in a satisfactory -« manner. There would seem to be valid arguments for both views. For practical purposes, we may regard the infection derived from cattle as a new disease in this country, and indeed this holds good for the temperate parts of the world generally. Among the epidemic diseases relatively new to the race, none is more interesting than the group affecting the nervous system. I shall discuss briefly two of these. Cerebrospinal fever, perhaps more often called cerebrospinal, or epidemic, meningitis, appears to have been quite unknown until 1805, when it was reported in Switzerland. Within a few years it has been identified in many places in Europe and America. This is a disease of very pronounced manifestations in the sick individual and of marked peculiarities in its epidemic qualities. The great Osler thought that the keen medical minds of the 17th and 18th centuries would not have overlooked it, and said: ‘‘In cerebrospinal fever we may be witnessing the struggle of a new disease to win a place among the great epidemics of the world.” A year after the occurrence of the Geneva outbreak (the first re- ported) the disease was discovered, independently, in New England, and from that time down to the present most countries have had anywhere from a few scattered cases to disastrous epidemics. For a long time after the disease became established as an entity quite apart in origin from any other disease there was much questioning as to whether or not it was contagious. For one of the most convincing experiences in this connection let me quote from Vaughan: For many years Boudin stood almost alone not only among French ob- servers, but among those in all parts of the world, in contending for the in- fectious nature of this disease. One of the most convincing instances cited by Boudin in his contention for the infectious nature of the disease is the experience of the 18th infantry (of the French army). Cerebrospinal menin- gitis had been epidemic among the civil population in the Department of Landes in the Lower Pyrenees adjacent to Spain for some years prior to 1836. This regiment was stationed at Bayonne in this part of France and was in part recruited from the population of this section. Cerebrospinal men- ingitis appeared in two garrisons in the Department of Landes, Bayonne and Dax, in which this regiment was quartered. The regiment moved first to Bordeaux and then to Rochefort, but the disease not only continued with the regiment, but spread to some slight extent outside the garrison, espe- cially at Rochefort where the inmates of a prison became involved. This regiment was stationed at Bayonne in 1836. It left Rochefort at the end of Marcu 15, 1936 MCCOY: EPIDEMICS 93 1838 and went to Versailles. In February, 1839, six men of the regiment, all occupying the same room, came down with cerebrospinal meningitis at in- tervals of a few days. At Versailles the disease extended to other regiments, but out of a total of 160 cases at that place 116 occurred in the Eighteenth, while the remainder were distributed through four other infantry regiments and three troops of cavalry. The Eighteenth regiment was subsequently transferred to Chartres and later distributed to Metz, Strasbourg, Nancy, Schlestadt, and Colmar, in all of which places cerebrospinal meningitis not only continued in the Eighteenth Infantry, but especially attacked the re- cruits to this regiment, and spread to other military organizations with which the Eighteenth came in contact and, in some places, to the civil population. There is some reason for believing, though the proof is not clear, that the disease was spread by troops from the Department of Landes towards the Mediterranean. From 1841 to 1849 nearly every garrison in France from Lille in the North to Marseilles in the South and from Brest on the Atlantic Coast to Strasbourg on the Rhine reported small epidemics of this disease. Not only was this true, but cerebrospinal meningitis was earried by French soldiers to Algerian garrisons. Finally, in this group, comes epidemic encephalitis, the history of which is recent. I think there is much question as to the identity of various outbreaks of the encephalitis group of diseases. For practical purposes we may consider the outbreak in St. Louis in 1933 as a unique event in medical history. Coming apparently from nowhere, progressing for three months, and vanishing as mysteriously as it be- gan, the disease could not be identified beyond doubt with any previ- ously recognized clinical and epidemiological entity. It attacked about 1,000 people and took the lives of about one-fifth of that number. It was tentatively grouped, by those studying it, with an epidemic condition that has been prevailing in Japan occasionally for perhaps half a century, but there were certain features of dissimilarity which had to be admitted. One of the most important of the questions we are called upon to consider is this: To what extent are man’s efforts, aimed directly at the control and suppression of disease, effective? I am compelled to confess that those of us concerned especially in preventive medicine are likely to claim too much. Often we have failed to take into ac- count that epidemics may come and go without much regard for what we do or fail to do. There is coming into existence a more sane attitude on this point and we find ourselves adopting a more critical attitude towards our efforts at disease prevention, and when we at- tempt to appraise the results of our own efforts we take into account as well the operation of natural causes which we are not able to con- trol. Let me give you as an example of a disease in the control of which I think there is no question but that purposeful measures have 94 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 26, NO. 3 been highly successful. I have in mind yellow fever. Many of us re- member when it occurred all too frequently in epidemic form in the United States, and was practically always present in tropical America. Then came the brilliant studies of the army commission made up of Reed, Carroll, Lazaer, and Agramonte, work which showed that in a certain species of mosquito lay the secret of the spread of the disease. This was followed by the work of the administrative authorities who showed that by attacking the mosquito, attacking it especially at the most vulnerable point, the breeding places, the chain of events lead- ing to the epidemic could be broken and the disease prevented or promptly eliminated if it already had gained a foothold. Here was a disease, the elimination of which from the world seemed readily feasible, and one of the great international public health agencies set for itself the task of freeing the world of yellow fever. The task was begun perhaps 20 years ago and it may be said without doubt that great good has been accomplished but it may be said, equally without doubt, that the goal appears no nearer now than when the work was begun. True, the disease has been readily suppressed in the important ports and other cities of the world but it has been found to lurk in the more remote places, places in which the attack on it is much more difficult. It has been found, too, that more than one species of mosquito may carry the disease, and finally, that atypical clinical cases may serve as foci of spread. In spite of these facts, let me say that on the whole the results of the warfare on yellow fever have been brilliant, though falling short of our expectations. Now let me give you an example of a disease in which the problem of elimination seemed to be even more easy of solution than that of yellow fever. I refer to pellagra, a nutritional, or dietary, deficiency disease very prevalent in our Southern States and in certain other parts of the world. The brilliant experimental work of Goldberger and his associates showed that the only thing needed to prevent the de- velopment of pellagra in an individual and even to cure it if already developed was a suitable diet. When it came to practical application of the experimental findings on a large scale, difficulties were en- countered. It was not, and is not, easy to modify the dietary customs of large groups of people and, of course, the economic factor usually is the predominant one since a “‘suitable’”’ diet costs more than the relatively inexpensive diet which often leads to the development of pellagra. So we still have a disgraceful amount of the disease, if we think in terms of how much we know about it and how easy is its prevention, but probably we have no more than is inherent in dietary Marcu 15, 1936 MCCOY: EPIDEMICS 95 customs and the economic status of the parts of the world in which pellagra prevails. In this day when smallpox is so rare that many physicians go through a busy professional career without seeing a case, it is diffi- cult to credit the statement that there was a time, not so long ago, when English employers of domestic servants made it a condition for acceptance into service that the prospective employee should have had smallpox in the natural fashion. Perhaps the prevalence of no other disease, has been so much influenced by direct attempts at con- trol as this one. Contrast the experience of Boston, 1721, when about 6,000 were attacked in a population of 10,500 with the experience of today when the disease is so readily controlled by vaccination that it has ceased to be a major scourge. While not yielding to anyone in honoring Jenner for having given us vaccination, I must remind you that before his day there was a crude method of prevention of small- pox based on the same principles used by Jenner. Purposeful inocula- tion of smallpox itself usually gives rise to a mild form of the dis- ease which protects against an attack acquired in the natural way. This method antedates the Christian era. It is significant of the changed attitude of the population as to its social responsibility that inoculation, the precursor of vaccination, was chiefly practiced among the well-to-do, that is among those who could pay for it. Nowadays a child has a poor chance to escape vac- cination, be he poor or rich. I will now discuss briefly two diseases that have become relatively rare, due, in my opinion, not so much to man’s direct attempts to control them but to influences concomitant with the advance of civilization. The “black death” of the 14th Century usually is identified with plague of more recent times, but the accounts that have come down to us give no sure grounds for insisting on the identical nature of the two. When we read of the disease in those days being so virulent as to infect swine (animals never attacked by plague as we know it), our skepticism seems well founded. On the other hand, some of the clinical descriptions seem to leave little choice but to call the disease plague. Whatever the nature of that early epidemic may have been there is no doubt as to its destructiveness. Thus, three out of every 5 in Florence are said to have died; in Italy as a whole and, indeed, in England, half of the population of that day died from this disease. This outbreak gave us one notable, crude advance in preventive medicine, the genesis of quarantine systems. In the early form it con- 96 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 26, NO. 3 sisted of arbitrary detention of ships at ports still uninfected for 40 days after their arrival from an infected country. I have said it was crude, but no more crude than the measures taken by some govern- mental authorities within the past few months in connection with the spread of the epidemic of poliomyelitis which occurred last sum- mer, and it is to be said that the earlier attempts to quarantine and the later ones are about equally ineffective. The London plague of 1665 is said to have killed 98,000 among a population of one-half million. We have had what we consider pretty bad outbreaks in modern times (notably British India) but there is nothing to even remotely approximate the London experience. Happi- ly, itis now recognized that plague will not become a menace in any modern community. We owe this fortunate state of affairs not so much to any intensive improvement in sanitation but rather to changes in our environment which have put us on less friendly terms with the rodent population. It is suggested that the complete ob- literation of plague in England was due to the fact that the black house rat, which came in close contact with man, was driven out in the 18th century by the brown rat, which ordinarily is not found in such close association with man. I have spoken of the relative freedom the world now enjoys from plague but I might possibly be charged with too much optimism did I fail to tell you that once before, for a period of about 7 centuries, England remained free from plague, that is, between the Justinian plague of the 7th century and the other outbreak of the same disease known as the “black death” in the 14th century. The fact that leprosy is the only definite clinical condition men- tioned in the Bible, and the space given to it there, indicates how im- portant a problem it was in those days. This disease was so prevalent in Europe in the Middle Ages that leprosy isolation hospitals were numbered by the thousands while today in the whole civilized world they may be numbered only as a few dozen. It may be stated here (though not necessary to the present discussion) that there is good authority for the statement that the leprosy hospital often formed the nucleus of the general hospital after the former ceased to be used for the special purpose for which it was constructed, and, further, that many of the great European hospitals of today go back to such humble origin. It is interesting to observe how the ideas of the race as to the cause of leprosy have varied from time to time. Thus, in biblical times the injunction for sanitary administrative control such as Moses laid Marcu 15, 1936 MCCOY: EPIDEMICS 97 down spell out a clear recognition of the fact that the disease was considered communicable, or, as often stated, that “leprosy begets leprosy.”’ Then came a period of several centuries in which it was looked upon as not communicable. By the latter part of the middle ages hard experience had forced into the minds of man the fact that the disease did spread, and those who suffered from it were a menace to the health of those around them, hence the large number of asylums referred to as having existed at that period. This view prevailed until well into the 19th century. In 1867, the Royal College of Physicians (London) declared leprosy to be non-contagious. A little later came the studies of the great Norwegian, Hansen, who established the in- fectious and communicable nature of the malady, a fact so generally accepted at the present time that we have difficulty in appreciating that for long periods the minds of men rejected the communicability of the disease. The facts at hand do not justify any other conclusion as to the spread and decline of leprosy at different periods and in different countries. I venture the suggestion that leprosy is to be regarded as an epidemic disease which rises and falls due to circum- stances beyond our knowledge and largely beyond our control. It differs from other infectious diseases, however, 1n that the cycle covers one to several centuries whereas with most other epidemics the cycle is completed in a few weeks, or a few months at the outside. I can make this suggestion with much safety to my professional reputation since no one can disprove it today, and it will take several centuries to prove it if it chances to be well founded. In conclusion, I will mention problems presented by an epidemic disease that has not been influenced in respect to prevalence either by purposely directed efforts or by the onward march of civilization and progress. I am a member of a group charged with certain responsibilities in connection with researches on influenza and other respiratory dis- eases and rarely do we meet without some discussion, often disputes, as to just what influenza is. The questions most frequently asked, and never answered, are: 1. Are the various pandemics that occur at intervals of 20-30 years due to the same virus? Tentatively, the answer is yes. 2. A much more difficult question is as to the inter- epidemic prevalence of the disease. Are these relatively mild out- breaks due to the same virus that causes the pandemic outbreaks or are they due to different viruses? There is no agreement on this point and we must keep an open mind. Until we have some new method of approach enabling us to identify the causative agent with certainty, 98 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 26, No. 3 speculation is rather futile. In the meantime, the term “‘influenza”’ is likely to continue to be used by physicians and laymen for miscel- laneous respiratory disorders to which no other diagnosis is readily applicable. Perhaps what I have just said may be made somewhat clearer by the explanation that the diagnosis of true influenza is based more on the characteristics of the epidemic prevailing at the time than on the manifestations in the individual patient. In 1918, one could say with much confidence that any group of patients had influenza but in 1936 any small group or any single patient is not so certainly disposed of from the point of view of diagnosis. In the epidemics, particularly in the pandemics, the outstanding features are widespread prevalence, high incidence, and lack of preference for any age, sex, social or racial group. It is true there are some varia- tions in the attack rates for different age groups in different epidemics; nevertheless, what I have said is substantially correct. Another feature is the general progress of the disease from east to west. Always there is an interval of a few months between the appear- ance of the disease in, let us say, Moscow and in San Francisco. The outbreak of 1918 and that of 1920 were the first to be studied by modern methods and yet I fear that the intensive and extensive studies carried on then, and later, have not resulted in any really helpful information so far as treatment of the disease and control of outbreaks are concerned. We did learn from the few experiences that strict quarantine measures would protect well isolated communities but that this was impracticable on a large scale. It was impracticable even for military camps, where requirements designed to protect could be made more rigid than for a civil community. Some com- munities, in attempting to control the disease, did things that in retrospect appear amusing. One west coast city required everyone to wear gauze masks over the mouth and nose, closed the churches, but, for readily understood reasons, did not close the saloons. Perhaps a story is worth telling to illustrate the situation. It chanced that the decline of influenza was about synchronous with the Armistice celebration in San Francisco and a physician friend of mine was ac- costed on the streets in the early morning of Armistice Day by a bibulous individual celebrating in true San Francisco style, and that means, to say the least, with enthusiasm. The bibulous individual was laughing and very voluble and my friend remarked to him that, of course, winning the war was wonderful but it scarcely seemed to justify the extreme kind of celebration in which he was indulging. The tipsy one replied hilariously that he was not celebrating the end Marcu 15, 1936 COVILLE: ROCK MIDGET 99 of the war but how wonderful it was to have a city with every church closed, every saloon open, and every woman muzzled! |The speaker concluded his remarks by showing a series of slides depicting events and phenomena associated with past epidemics. Ed. | BOTANY.—Rock midget, a new species of Mimulus from Death Val- ley, California.! FREDERICK V. CoviLuE, U. 8. National Herba- rium. The plants of Death Valley, California, are of special interest be- cause they exist under conditions of extreme heat and aridity. This interest has become much more widespread since Death Valley was made a national monument, in 1933. More than 42,000 persons visited the valley in 1935. My own Botany of the Death Valley Expedition, published in 1893, is still the principal work of reference on the plants of Death Valley. In the process of preparing a new account of these plants, several species new to science have been found and published. Another is described in the present paper. Its specific standing has been confirmed by Dr. Adele Lewis Grant, author of A Monograph of the Genus Mimulus, 1924. The illustration (Fig. 1) is from a photo- graph made by Mr. B. Anthony Stewart, of the National Geographic Society. To him and to the Society I am indebted for the privilege of publishing it. Mimulus rupicola Coville & Grant, sp. nov. Annuus, Mimulo tricolori affinis, sed corollis brevioribus et pallidioribus, foliorum caulinorum apicibus et calycium lobis acutissimis, nec obtusis; pedicellis fructiferis recurvatis, deinde porrecte curvatis, nec rectis; capsulis faleatis, acutis, puberulis, a sutura convexa mox dehiscentibus, nec, modo M. tricoloris, ovatis vel oblongis, obtusis, glabris, et tardissime dehiscen- tibus. Fissuras rupium calcariarum montibus eremis Vallis Mortis, nec, ut M. tricolor, locos aquosos exsiccantes Californiae intramontanae, habitat. Plant belonging to the section Oenoe, annual, 2 to 15 em in height, begin- ning to flower when very young and with only a few leaves, all basal, the plants later becoming much branched from the base; stems densely glan- dular-pubescent to glandular-villous; basal leaves 2 to 6 cm long, 2 to 10 mm wide, the elliptical-lanceolate to linear-lanceolate or oblanceolate blades narrowed at the apex to a usually blunt point, and tapering very gradually at the base into a petiole often as long as the blade and margined for most of its length, entire or rarely with an irregular tooth, glandular- pubescent, often rather sparingly, the margins and petiole often glandular- villous; stem leaves opposite, up to 15 mm in width, elliptical-oblong, acu- minate to a usually very acute tip, narrowed below to a short margined petiole, glandular-pubescent, usually densely so, and often glandular-villous; flowers occurring singly in the axils of the leaves; pedicels glandular-hairy, 1 Received December 21, 1935. 100 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 26, NO. 3 at first commonly 1.5 to 2 mm long, and straight, in fruit sometimes increas- ing in length to 4 mm and bent sharply downward and to one side at the base, then, above, forward and outward, twisted half way round, becoming hard and woody and solidly attached to the rigid stem and to the base of the capsule; calyx usually 10 to 12 mm long, sometimes 13 mm, in depau- perate plants sometimes only 6 mm long, plicate, scarious below the sinuses, glandular-pubescent, and sometimes with a few longer, soft, gland-tipped Rock midget, Mimulus rupicola Coville & Grant, sp. nov. Natural size. hairs, the throat oblique, the teeth very unequal, ciliate, triangular-ovate to triangular-lanceolate, attenuate to slender, often filiform, very acute tips, the posterior tooth 2 to 4 mm long, about twice the length of the others, reaching 2.5 to 4 mm farther than the two anterior teeth; corolla about two and a half times the length of the calyx, usually 20 to 30 mm long, in depau- perate plants sometimes only 15 mm, the tube, throat, and lower part of the lobes sparingly glandular-puberulent, tube slender, yellow, about twice as long as the calyx tube, expanding rather abruptly to a funnelform throat, this yellow, with small purple spots and densely clothed within, on the Marcu 15, 1936 COVILLE: ROCK MIDGET 101 yellow anterior areas at the mouth, with large, yellow, club-shaped hairs, lobes rose-colored, sometimes almost white, approximately equal, commonly 2.5 to 5 mm in length and a little more in breadth, each with a conspicuous maroon spot at the base, or the spot sometimes wanting on the lower lobe, or even on all the lobes; stamens 4, included, the filaments arising from the corolla tube at a distance of about one-third the whole length of the corolla from the base, glandular-hairy for a millimeter or two at the base, smooth above, the posterior pair about 4 mm long, the anterior pair 2 to 3 mm longer; anthers smooth, connivent in pairs, the four anther cells of each pair about 1 mm long and nearly as broad, forming a cross, and dehiscent by slits for their whole length; ovary about 2 to 2.5 mm long to the end of the acuminate tip, 2-celled, with many ovules; style included, a little longer than the longer pair of stamens, sparingly hairy, the hairs short and gland- tipped; stigma peltate-funnel-shaped, about 1.5 mm high and 2 mm or more wide, the margin fringed with glandular hairs; capsule cartilaginous, lanceolate, somewhat sickle-shaped, somewhat compressed, the convex mar- gin toward the back (the longer-toothed side) of the calyx, up to 9 mm in length, acute, puberulent, with a groove on each side opposite the placentae, dehiscent by a narrow slit from apex to base on the convex margin and a third or less its length from the apex on the concave margin, the convex, originally posterior margin standing outward and below at maturity and the concave margin inward and above, because of the twist in the fruiting pedicel; seed light brown to blackish brown, narrowly elliptical in outline, terete, abruptly contracted at the apex into a distinct beak, the body 0.7 to 1 mm long, 0.3 to 0.4 mm thick, minutely reticulated at full maturity, the areolae rectangular. Type specimen in the United States National Herbarium, no. 1,630,865, collected on the foot of Nevares Peak, Funeral Mountains, Death Valley, California, altitude about 2,500 feet, April 6, 1935, by M. French Gilman (no. 1251), with both flowers and fruit. Collected also at the same locality, which is near Cowcreek, April 23, 1932, in flower (Coville & Gilman 400), March 18, 1935, in flower (Gilman), and in October, 1935 (Gilman), with an abundance of fruit on the dead stems. It was first found by Mr. Gilman in Cottonwood Canyon, Panamint Mountains, April 21, 1932 (Coville & Gil- man 360), in flower, altitude 900 feet. The species has been collected also in Titus Canyon, Grapevine Mountains (Gilman 1196, April 29, 1935, past flowering). In all these localities the plants grew in the shaded crevices of limestone rocks. ‘The conspicuous flowers, rose-colored to almost white, with a large maroon spot at the base of each corolla lobe, and a golden throat; the habitat of the plant, in the shaded crevices of rocks; and its small size, all combine to give the species charm. The plant sometimes begins to flower when it is less than an inch in height and has only basal leaves. The larger specimens, up to 6 inches in height, are much branched from the base. A curious characteristic of the species is the position of the fruit. When the flowers open, the flower- stalk is straight and the long tooth of the calyx stands uppermost in the flower. Later the flower-stalk bends abruptly downward and to one side, and then upward and outward, in a sigmoid curve, and at the same time it twists about 180 degrees, until the long tooth of the calyx is on the lower side, the result being that the capsules, which are somewhat sickle-shaped, 102. JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 26, NO. 3 stand out from the stem like the rustic coat-hooks of a log-cabin camp. The narrow slit in the ripe cartilaginous capsule is then on the lower and outer side, a position that presumably facilitates the scattering of the seeds. The fruits of some of the later (upper) flowers assume this position only partially, as if drouth and death had come too soon. In such cases the capsule is often straight and the split imperfect. For more than three hundred years it has been generally admitted that a rose by any other name would smell as sweet. Without ques- tion, too, a rose would smell as sweet without a name. But it is my belief that any plant of sufficient interest to attract public notice should have a popular name as well as a Latin name. Plants of the genus Mimulus are commonly called monkeyflower, because their two-lipped corollas suggest a monkey’s face. The present species, however, belongs to a group in which the corollas are not definitely two-lipped. This group of species has sometimes been separated from Mimulus as a distinct genus, Hunanus. The name Hunanus means a dwarf. I suggest for these dwarf species of Mimulus the common name, midget, and for this new species from Death Valley the com- mon name, rock midget. The accompanying illustration will, I hope, confirm the propriety of this suggestion. Not all the plants of the des- ert are protected from heat and drouth by special adaptations, such as the absence of leaves, and the internal storage of water. In the present case the habitat itself lends protection to an annual that is tiny, thin-leaved, and delicate. Like the conies of ancient Palestine, the plants of this new species ‘‘are but a feeble folk, yet make they their houses in the rocks.”’ BOTAN Y.—Leaf venation as a means of distinguishing Cicuta from Angelica.: Miriam L. Bomuarp, U.S. Forest Service. (Com- municated by FREDERICK V. COVILLE.) The close resemblance of poisonous waterhemlocks (Cicuta spp.) to certain harmless umbellifers, particularly wet-habitat species of Angelica, indicates the importance of ascertaining diagnostic char- acters by which the waterhemlocks may be distinguished. Numerous and sudden fatalities have been caused by Cicuta, not only among domestic livestock but among human beings, especially children. The extremely toxic character of the underground parts ranks this genus as one of the most virulently poisonous groups of flowering plants na- tive to the North Temperate Zone. Cicuta is chiefly a North American 1 Received January 27, 1936. Marcu 15, 1936 BOMHARD: LEAF VENATION 103 genus, being widely distributed in wet places practically throughout the continent. Only one species, C. virosa, occurs in Europe and Asia; it is considered such a menace in some parts of northern Germany that its extermination is ordered by law.’ Because of the general resemblance of their compound leaves and white-flowered umbels, waterhemlocks are most often confused with species of Angelica in the western United States and Canada where both genera are well represented and widely distributed, and where angelicas characteristically grow in moist or wet situations.*? Water- hemlocks are also sometimes mistaken for species of Osmorhiza, or for the woollyhead-parsnip (Sphenosciadium capitellatum), and some other members of the parsley family, many of which are excellent forage plants.* The fruits, it is true, provide a dependable means of recognizing all the genera mentioned. The flower clusters of Osmorhiza and Sphenosciadium are also sufficiently distinctive to serve for un- mistakable recognition, but those of Cicuta and Angelica can scarcely be distinguished with certainty by persons not technically trained in botany. When neither the fruits nor flowers are present and the foliage alone must be depended upon for identification, accurate de- termination in the case of Cicuta and Angelica is often impossible on the basis of the leaf characters now used, and, as will be discussed later, the internal structure of the root crown or rootstock has proved to be unreliable as a guide in recognizing waterhemlocks. The writer recently had occasion® to make a critical comparison of herbarium specimens of Cicuta occidentalis, the most widely distrib- uted waterhemlock in the West, and Angelica lyallii, one of the most common and valuable forage species of the angelicas, and was im- pressed by the striking contrast in the venation of their leaflets. As is usual in serrate leaves, the secondary veins proceed toward the middle of the marginal teeth in the leaflets of A. lyallii, but in C. occidentalis they apparently end in the notches between the teeth. Such a definite difference in the venation of this representative species of a poisonous genus suggested the possibility of its usefulness as a diagnostic generic character. 2 THELLUNG, ALBERT. Umbelliferae, in Heci1, Gustav. [Illustrirte Flora von Mittel- Europa 5: 1166. 1926. Munich. ’ This was strikingly illustrated on a California range a number of years ago where Angelica breweri was eradicated at considerable expense in the belief that it was a waterhemlock. The writer’s attention has also been called to a recent toxicological investigation made on a species of Angelica under the misapprehension that it was a species of Cicuta. 4 The fact that waterhemlocks are also confused with toxic umbelliferous genera, such as Berula, Orypolis, and Sium, need not be considered here. ° In connection with the preparation of the Range Plant Handbook, a forthcoming publication of the United States Forest Service. 104 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 26, NO. 3 A review of the literature relating to Cicuta brings to light that Bigelow,® in the first volume of his American Medical Botany, pub- lished in 1817, was probably the first to note that, in the eastern spotted waterhemlock (C. maculata), ‘The veins end in the notches, and not at the points of the serratures’’; his excellent colored plate clearly shows the leaf venation. Rafinesque’s description of this species also mentions ‘‘with veins ending at the notches, which is very unusual.’’’ Torrey and Gray® refer to this characteristic venation of C. maculata, but credit Darlington® (instead of Bigelow) with first having observed this feature. In the first five editions of Gray’s Manual of Botany, the termination of the veins in the notches of the leaflets is given in the description of the genus Cicuta which included only C. maculata and C. bulbifera. In 1868, Gray published his new species, C. californica; in the original description, he states, ‘“‘venis primariis tenuibus in dentes disinentibus,’’!® and definitely points out that this differs from the venation of C. maculata. Gray’s observation that the veins run to the teeth in C. californica no doubt accounts for the omission of reference to leaflet venation in generic descriptions of Cicuta in subsequent editions of his Manual, even though this native California species is quite outside the geographic range of that work. However, the characteristic venation of the leaflets of the eastern spotted waterhemlock has long been recognized as specific and is re- ferred to in several of our standard manuals which describe the plants of Canada and the southeastern or eastern United States and in various publications on poisonous or medical plants. It is remarkable that none of the current plant manuals and floras of western North America mention this similar peculiarity of venation for any of the western waterhemlocks. Chesnut and Wilcox" pub- lished, in 1901, a very good figure of western waterhemlock (C. oc- cidentalis) clearly showing the termination of the veins in the notches of the leaflets but not calling attention to this feature in the text. Cer- tain other investigators of poisonous plants have used illustrations showing the venation of this species; Jacobson” mentions it when 6 BiagELOow, Jacos. American Medical Botany 1: 127, pl. 12. 1817. Boston. 7 RAFINESQUE, C.S. Medical Flora; or, Manual of the medical botany of the United States of North America 1: 108, fig. 22. 1828. Philadelphia. < Sen: JoHN, and Gray, Asa. A Flora of North America 1:610. 1840. New ork. ® DARLINGTON, in his Flora Cestrica (1837), does call attention to the termination of the veins in the notches of the leaflets of C. maculata. 10 Gray, Asa. Characters of new plants of California and elsewhere. Proc. Am. Acad. Arts & Sciences 7: 344-345. 1868. 11 CuEesnutT, V. K., and Witcox, E. V. The stock-poisoning plants of Montana; a preliminary report. U.S. Dept. Agr. Div. Bot. Bull. 26: 80-86, pl. 7. 1901. 122 JACOBSON, C. ALFRED. Water Hemlock (Cicuta). Univ. Nevada Agr. Exp. Sta. Tech. Bull. 81:15. 1915. Marcu 15, 1936 BOMHARD: LEAF VENATION 105 describing the “‘Cicuta occidentalis-vagans group.’’ Greene’s original descriptions of C. ampla and C. arguta refer to the trend of the veins but, unfortunately, the meaning is ambiguous." No statement as to the possible importance of leaflet venation as an aid in recognizing other species of waterhemlocks has come to the author’s attention. In order to discover whether this character could be used in distinguishing Czcuta from Angelica, the writer has ex- amined all the available herbarium specimens (over 1,800) of these two genera in the collections of the United States Forest Service, the New York Botanical Garden, and the United States National Her- barium. Even without the aid of a hand lens, the principal veins are, for the most part, clearly visible on the under side of the leaflets, especially in Cicuta. It was found that the secondary veins of the leaflets are definitely directed toward the notches in all the species of Cicuta except C. californica. In those species in which this character is most clear-cut and easily recognized, the veins apparently end in the notches of the leaflet margins. The actual termination of the veins may be seen upon closer examination; enlarged cuts showing the vena- tion of several of the marginal teeth of the leaflets of five species of Cicuta (Fig. 1.—A, B, C, D, E) make clear that it is the direction or trend of the veins toward the notches which must be recognized as diagnostic, since the veins or their branches continue beyond the notches along the edges of the teeth, either coinciding with the margin or running more or less parallel to it, terminating in the frequently spinulose tips of the teeth. This is in direct contrast to the condition in Angelica, where the secondary veins proceed directly toward the middle of the teeth, notwithstanding the fact that subsidiary veins branch off to the notches and continue along the margin, as can be seen in the enlarged illustrations of the marginal teeth in two species of Angelica (Fig. 1—F, G). This distinctive venation of the waterhemlocks is obvious in Cicuta maculata, C. curtisti, C. occidentalis, C. bolanderi, C. douglasi, and in the European C. virosa. Particularly in the first four of these species, the veins are more or less elevated and widely spaced; their trend in the direction of the notches of the coarse, evenly spaced serrations may be determined at a glance. The fact that the leaflets of some specimens of C. douglasii tend toward an incised-serrate margin (Fig. 1—E) must be taken into consideration when noting the veins in this species; in some of the deeper serrations, the veins proceed to 13 GREENE, EpwaRp L. New species of Cicuta. Leaflets 2.: 241, 238. 1910-1912. Washington, D. C. 106 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 26, No. 3 the middle of the teeth (behaving much as midribs), although they are directed toward the notches in the majority of the serrations pre- cisely as in other specimens in which the margin is not incised but evenly and coarsely serrate. In C. vagans, the veins proceed toward the notches but tend to become much less prominent near the margin, especially in the upper portion of the leaflets. C. bulbifera is a dis- tinctive species bearing clustered bulblets in the axils of the upper re- duced leaves; its slender leaflets, remotely serrate above and often Fig. 1.—Leaflet venation in Cicuta (A—-E) and Angelica (F, G). A, C. occidentalis; B, C. curtisi1; C, C. bolanderi; D, C. bulbifera; E, C. douglasii; F, A. lyallit; G, A. bre- wert. Leaflet in A, B, C, E, F, and G, one-half nat. size; D, nat. size. Detail of mar- ginal teeth to the right of each leaflet, one and one-half nat. size; D, twice nat. size. Drawn by Leta Hughey. lobed below, may be so delicate that the direction of the veins cannot be clearly seen without the aid of a lens (Fig. 1.—D). The only exception to the rule that in Cicuta the veins apparently end in the notches of the leaflets is C. californica, but this species is exceptional in other respects as well, and is certainly not a typical Cicuta. California waterhemlock is practically an aquatic and has a rather restricted range in coastal California. There is no possibility of confusing it with either of the two species of Angelica (A. hender- soni and A. tomentosa) which occur within its range because these Marcu 15, 1936 BOMHARD: LEAF VENATION 107 angelicas may be recognized by distinctive leaf differences other than venation. In all the species of Angelica examined, the veins proceed toward the middle of the teeth. A. lyalli (Fig. 1—F) may be considered typical of the species occurring in the United States and Canada. A. brewerz is illustrated (Fig. 1.—G) to emphasize the fact that, even though a branch vein is sent off to the notch, the secondary vein continues directly to the tip of the tooth. In some exotic species of Angelica which possess delicate, much divided, or otherwise distinc- tive leaves, recognition of the characteristic form of the leaflets renders the traditional vein-to-tooth character relatively unimportant. Since Thellung™ reports that, in Europe, Crcuta virosa is sometimes confused with Angelica silvestris, which in the earlier stages of growth furnishes forage for cattle, particular attention was given to an ex- amination of available herbarium specimens of these two species. It was found that the veins are easily traced toward the notches in the narrow, lanceolate to linear, sharply serrate leaflets of C. virosa and directly to the middle of the teeth in the ovate leaflets of A. silvestris. This difference in venation has apparently been overlooked by re- cent European investigators who mention only the narrowness of the leaflets and the usual chambered character of the root crown as aids in identifying the poisonous waterhemlocks. Attention is called, how- ever, to the fact that the root crown is occasionally solid. Recognition of the presence of cross-partitions in the underground parts of waterhemlocks has been strongly advanced as an infallible guide in identifying these plants. This is, however, by no means a trustworthy criterion, because, although it is true that transverse partitions are characteristic of the root crown or rootstock of older plants of all the waterhemlocks, these partitions are sometimes ob- scure, and, in younger plants, are often absent. Moreover, many species of Angelica (A. lyallii1, A. ampla, A. breweri, etc.) sometimes have well-marked cross-partitions in their roots. In fact, the develop- ment of chambers in the thickened underground parts has been ob- served in various members of the Umbelliferae. The distinctive venation of the waterhemlocks should serve as a convenient field character for the recognition of these dangerous plants and should prove particularly useful in the identification of young plants early in the spring when there is most danger of live- stock-poisoning. 14 THELLUNG, ALBERT. Op. cit., p. 1338. 18 THELLUNG, ALBERT. Op. cit., p. 1164. 108 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 26, NO. 3 PALEOBOTAN Y.—A fig from the Eocene of Virginia. Epwarp W. Berry, Johns Hopkins University. The Eocene of Maryland and Virginia comprises a series of ex- clusively marine and often highly fossiliferous sediments segregated into an older Aquia and a younger Nanjemoy formation which were described in detail in 1901 by Clark and Martin.” Traces of the land vegetation of the time are represented by occasional drift fruits and fragments of wood. | The rather extensive investigations leading up to the report cited above resulted in the discovery of only a single plant—fruits de- termined by Hollick (Idem., p. 258, pl. 64, figs. 11, 12) as two varieties of a botanically unidentified species of Carpolithus. During the last 35 Fig. 1.—A probable fossil fig from Pense, Saskatchewan. Fig. 2.—Ficus aquiana, n. sp., from Belvedere Beach, Prince George County, Va. years fairly continuous collecting has been prosecuted at the more prolific Eocene outcrops but it was not until the summer of 1933 that an identifiable land plant was discovered. This was a fairly complete cone of a pine’ from near the type locality of the Aquia at Belvedere Beach, Prince George County, Va. Continued exploration of the ex- cellent section at Belvedere Beach resulted in the discovery by R. L. Collins, during the summer of 1935, of a fruit of Ficus. This is much the most convincing fossil fig that I have ever seen, being more char- acteristic than the remarkable Ficus ceratops Knowlton* which is so abundant in the Lance formation of Montana, Wyoming, and western Canada. The new species may be named in allusion to the horizon which is not many miles from the type locality of the formation. 1 Received November 11, 1935. * CuaRk, W. B., and Martin, G.C. Eocene. Md. Geol. Survey. 1901. 3 Berry, KE. W. This JouRNAL 24: 182, 1934. * KNow.ton, F.H. Torrey Bot. Club Bull. 38: 389, figs. 1-4, 1911. Marcu 15, 1936 BERRY: EOCENE FIG 109 Ficus aquiana, n. sp. Fig. 2 In life this fig must have been approximately spherical. As a flattened fossil, there is a slight narrowing in the region known as the neck in the pyriform species, but the scar of attachment is invaginated with the body of the fig pushed forward on one flat side and backward on the other. The eye is open and relatively large, about 9 x 5mm, with its edges somewhat thick- ened. The interior of the specimen was filled with sediment and no traces of seeds were observed, indicating that the plug or scales had rotted away while the fruit was floating and that the sediments which filled the interior entered through the eye, as the walls were perfectly intact when the speci- men was collected. In modern figs of comparable size the eye is closed to a greater or less extent by scales of various forms. That this was the case in the fossil is indicated by the prominence of the vascular bundles on the thickened lip. The ribs are numerous and well shown in the photograph. The walls of the fruit (properly the receptacle) are relatively thin and the cavity large and filled with sediment. This feature as well as the fact that this fruit floated from the land (probably stream borne) and became filled and cov- ered by sediment and was lignified, is rather conclusive evidence that in life it was dry and woody instead of fleshy, as otherwise it would have rotted and failed of preservation. Its dimensions are, as preserved, about 4 cm in length, about 3.7 cm in diameter, and about 1.5 cm in thickness. I have, unfortunately, no complete collection of the fruits of existing species of Ficus. One unidentified woody specimen is identical in surface ornamentation, but has a smaller eye and a shape like Ficus ceratops. The most similar recent fig seen is a specimen of Ficus repens Rotel collected by F. N. Meyer near Soochow, China; but I do not know if this specimen is typical of the species. It is about the same size as the fossil with similar thin woody walls. The ribbing is slightly less prominent on the outer surface. The actual eye is smaller but the apex is invaginated around the latter for about the area of what I have called the eye in the fossil and has the appear- ance of offering the possibility of macerating out to match the fossil. The general form is similar, i.e., roughly globular and inflated, but is produced into a short and slender neck. If the latter were softened by maceration and pushed during a compression similar to that undergone by the fossil, the result would not be very different. There is a possibility, although I regard it as remote, that my interpretation is wrong, and that what I regard as the base and the scar of attachment is the plugged eye, and what I regard as the eye is the broken neck. The lack of any reasonable explanation for such a break where the vascular strands would be concentrated, or for the opening to be symmetrical and lipped, seems to invalidate such an inter- pretation. Occasional fruits of Ficus have been encountered and described from strata ranging in age from Upper Cretaceous to late Pleistocene. Except for casts in Pleistocene tuffs or travertine in France and Italy, these are mostly lignified, structureless and unsatisfactory objects for study. Exceptions to this statement are the present specimen and the late Upper Cretaceous or early Eocene form from the West referred to above and described by Knowlton in 1911. 110 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 26, NO. 3 In this connection I am figuring an object sent me by B. F. Baxter of Pense, Saskatchewan, from approximately the same horizon as Ficus ceratops Knowlton. This is a chalcedony pebble with faint in- dications of ribs and a pronounced umbilicus which might very well represent a fig ‘“‘eye,’”’ since the eye-like plug is surrounded by a nar- row cavity into which a fine needle can be inserted a distance of 4 mm inside the rim or lip. In color and form it bears a striking re- semblance to a modern fig, but it would be unsafe to press such an identification without sacrificing the single specimen, and even this might not yield results. It seems to me that this probably represents a fig which has been replaced by chalcedony since it is hard to conceive of a pebble with an umbilicus and radial markings. On the other hand if it is of organic origin one would expect more than a single specimen. If additional specimens are found I would not hesitate to consider its identity as proved. I have records, doubtless not complete, of 402 named species of fossil figs. The majority of these are based upon leaf impressions rang- ing in age from mid-Cretaceous to Recent. Doubtless a considerable number of these may be wrongly identified, but the majority can scarcely be disputed especially since they are occasionally associated with fruits. It would be an interesting speculation as to whether the setting of seeds in an Eocene fig required the intervention of an- cestral Blastophaga wasps as is the case with some of the best culti- vated figs. All of the several hundred existing species of figs do not however require caprification, but certainly the form of the fruit- multiple flowered on the inside of a hollow receptacle is to be cor- related with insect pollination of some kind and if this be granted the habit was already established at the dawn of the Tertiary. Much has been written of the place of origin of fig culture, Solms- Laubach suggesting southern Arabia, which seems doubtful to me. None who have written on this fascinating subject have had any knowledge of the fossil record or of the presence of F2cus carica (the edible fig) in Interglacial deposits (2nd and 8rd Interglacial) in both France and Italy, both fruits and leaves being found and both in this case highly characteristic. Fruits supposed to represent this species have even been described by Engelhardt and Kinkelin from the Pliocene of south Germany, but their identification is not so certain. A great many recent species other than Ficus carica are eaten occa- sionally or regularly in different parts of the world. Sturtevant (1919) names 18 species which are eaten regularly or in times of scarcity of food, and doubtless many more are occasionally eaten. I would not be Marcu 15, 1936 GOLDMAN: POCKET GOPHERS ft. surprised if it were eventually shown that wild figs were present throughout the Mediterranean basin and were an article of diet of Paleolithic man, which is quite apart from the question as to where dried figs first became an article of commerce, which may well have been in eastern Mediterranean countries. The genus Ficus, as afore- said, is a large one in modern floras with between five and six hundred species adapted to a variety of habitats, but confined to the warmer parts of the world, although by no means confined to the tropical zone. The Eocene has yielded a prolific flora in the Mississippi embay- ment region and in western North America but practically nothing in eastern North America except the isolated lignite basin at Brandon, Vermont,’ from which a large number of fruits and seeds have been described. The latter, as well as the present species of Ficus, indicate climatic conditions more genial than those indicated by the plants known from the Chesapeake Miocene, and probably somewhat more genial than the corresponding latitudes of the present day, but cer- tainly to be denominated temperate, rather than tropical or even sub- tropical. ZOOLOGY.—New pocket gophers of the genus Thomomys.' E. A. GOLDMAN, Bureau of Biological Survey. Twenty years have passed since the pocket gophers of the genus Thomomys were revised by Bailey (North Amer. Fauna, No. 39: p. 1-136, Nov. 15, 1915). Much additional material has become avail- able for study, and the tendency toward the recognition by name of less and less strongly marked stages of differentiation has resulted in the description of many new forms, especially in the plastic Thomomys bottae group. These extremely sedentary animals inhabit regions of highly diver- sified topography and climate. Populations consist of colonies which appear to be loosely and more or less intermittently in contact with neighboring colonies, or continuity of range may have become par- tially or completely interrupted by barriers associated with geological history. In such a setting full play has been given to forces that bring about localized modification, in response to environmental and genetic factors. The result has been the production of a profusion of forms, varying greatly in degree of minor differentiation, and yet maintain- ing the same pattern of more essential characters with remarkable 5 Berry, HE. W. Am Jour. Sci. 47: 211-216. 1919. 1 Received December 16, 1935. 112 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 26, NO. 3 fidelity. How many should be segregated, and whether particular ones are best designated by binomial or trinomial names are problems de- pending for ultimate solution on coérdinated studies in the field and laboratory. The criterion—presence or absence (observed or assumed) of in- dividuals possessing intermediate characteristics—for distinguishing between species and subspecies is excellent in theory but not always satisfactorily applicable in practice. Uncertainty must exist where the number of specimens is insufficient to determine the range of in- dividual variation, and where detailed knowledge of field conditions is lacking. A complicating factor is evidence that some forms may intergrade in one region and occur in close juxtaposition without ap- parent intergradation in another. Until the codrdinated field and laboratory studies can be completed the choice of binomial or tri- nomial names must depend on individual judgment of the evidence in each case. In the treatment of forms I have endeavored to be as consistent as possible. Experience in classifying pocket gophers, and many other mammals, has shown that trinomials may usually be applied with safety to populations exhibiting differential characters in combina- tions known to be of subspecific value only, elsewhere in the group. Thomomys muralis, sp. nov. Grand Canyon Pocket Gopher Type.—From lower end of Prospect Valley, Grand Canyon, Hualpai Indian Reservation, Arizona (altitude 4,500 feet), No. 202580, o adult, skin and skull, U. S. National Museum (Biological Survey collection) ; col- lected by E. A. Goldman, October 3, 1913. Original number 22264. Distribution.—Isolated on terraces along the inner gorge, far below the outer rim in Prospect Valley, a lateral pocket within the Grand Canyon, near the eastern end of the Hualpai Indian Reservation, Arizona. General characters——A diminutive, ochraceous buffy species, separated from the ranges of the similarly-colored forms of the region by an interposed arm of the range of the darker subspecies, Thomomys bottae fulvus. Differing from fulvus in much smaller size, lighter color, and cranial proportions. Very similar in size and color, and bearing a close general resemblance to Thomomys bottae desertorum, but geographically isolated, and cranial details indicating no intergradation. Color.—Type (acquiring fresh pelage): Upper parts between light ochra- ceous buff and ochraceous buff (Ridgway, 1912), purest along sides, thinly mixed with black on top of head and median dorsal area; outer sides of fore- arms light buff; thighs nearly pure white; under parts overlaid with white; muzzle blackish; ears entirely black along with the black postauricular patches usual in the group; feet and tail white. One of the topotypes is near tawny in general color of upper parts, with under parts overlaid with light ochraceous buff. Marcu 15, 1936 GOLDMAN: POCKET GOPHERS 1f3 Skull.—Very similar in general to that of T. b. desertorwm, but braincase more rounded and inflated, the basicranial region tending to bulge more prominently posteriorly; frontal region broader; premaxillae usually less extended posteriorly, the ends more nearly conterminous with nasals; upper incisors more strongly recurved (slightly more procumbent in desertorwm). Compared with that of T. b. fulvus the skull is much smaller and more deli- cate in structure, and differs in detail in about the same characters as form desertorum. Measurements.—Type: Total length, 194 mm; tail, 64; hind foot, 26. Two adult female topotypes: 182-190; 57-56; 24.5-25.5, respectively. Skull (type): Occipitonasal length, 33.2; zygomatic breadth, 20.5; breadth across squamosals (over mastoids), 17.7; interorbital constriction, 7; length of nasals, 11.3; maxillary toothrow (alveoli), 7.2. Remarks.—The geographic isolation of Thomomys muralis in the Grand Canyon appears to be complete. In places it was found inhabiting strips of soil on ledges only a few feet wide, bounded above and below by vertical cliffs hundreds of feet high. Specimens examined.—Four, all from the type locality. Thomomys bottae desitus, subsp. nov. Big Sandy River Pocket Gopher Type.—From Big Sandy River, near Owen, Mohave County, Arizona (altitude 2,000 feet). No 227802, o adult, skin and skull, U. 8. National Museum (Biological Survey collection); collected by E. A. Goldman, Sep- tember 21, 1917. Original number 23332. Distribution.—Big Sandy River Valley and desert region southeastward to Wickenburg; probably also including the valley of the Bill Williams River, Arizona. General characters.—Very similar in color and form of cranium to Thom- omys bottae desertorum of Detrital Valley, but decidedly larger. Differing from Thomomys bottae chrysonotus of the Colorado River Indian Reservation, and Thomomys bottae cervinus of the Salt River Valley, in smaller size and more tawny coloration. About equal in size to Thomomys bottae fulvus of the Mogollon Plateau region, but color much clearer tawny, the back less mixed with black, and cranial details distinctive. Color.—Type (acquiring fresh pelage): Upper parts in general between tawny and ochraceous tawny (Ridgway, 1912), only slightly darkened on top of head and middle of back by black-tipped hairs; outer sides of forearms light ochraceous buff; lower part of sides and thighs whitish; under parts overlaid with white, tinged with buff across throat and chest; muzzle black- ish; ears black, except anterior margin which is buffy, the confluent black postauricular markings prominent; feet white; tail buffy above, whitish below, becoming white all around near tip. Skull.—Essentially like that of 7. b desertorwm, but much larger. Com- pared with those of T. b. chrysonotus and T. b. cervinus the skull is much smaller, and less angular, the supraoccipital region more fully inflated (lacking the deep, median supraoccipital excavation usually present in chrysonotus and cervinus); audital bullae much smaller, Very similar to that of T, b. fulvus, but bullae larger. 114. JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 26, NO. 3 Measurements.—Type: Total length, 230 mm; tail, 70; hind foot, 30.5. Two adult male topotypes: 219-238; 72-75; 30-31, respectively. An adult female topotype: 210; 62; 29.5. Skull (type): Occipitonasal length, 39.5; zygomatic breadth, 24.8; breadth across squamosals (over mastoids), 20; interorbital constriction, 6.8; length of nasals, 14.7; maxillary toothrow (alveoli), 8.2. Remarks.—At the type locality T. b. desitus inhabits the loose sand along the broad alluvial river bottom, habitat in marked contrast with the harder upland soils perforated by its geographic neighbor, 7’. b. desertorum. The skull of desitus indicates close relationship to fulvus, and it is obviously a desert representative of the same group. Specimens exramined.— Total number, 22, all from Arizona as follows: Big Sandy River (near Owen), 5; Big Sandy River (Neale’s Ranch, at 2,000 feet altitude), 7; Wickenburg, 10. Thomomys bottae hualpaiensis, subsp. nov. Hualpai Mountain Pocket Gopher Type.—From Hualpai Peak, Hualpai Mountains, Mohave County, Ari- zona (altitude 7,000 feet). No. 227796 ,o adult, skin and skull, U. 8. Na- tional Museum (Biological Survey collection); collected by E. A. Goldman, October 6, 1917. Original number 23351. Distribution —Known only from the Hualpai Mountains, Arizona. General characters.—A light ochraceous buffy subspecies of medium size. Closely allied to Thomomys bottae desitus, of the adjoining valley of the Big Sandy River; size about the same, but color distinctly paler; skull differing in detail. Similar in general to Thomomys bottae desertorum of the desert plains of the Detrital Valley to the north, but considerably larger and paler colored; cranial characters also distinctive. Color.—Type: Upper parts near light ochraceous buff (Ridgway, 1912), clearest along sides, thinly mixed with black on top of head and over back; forearms and thighs pale ochraceous buff; under parts in general overlaid with pale ochraceous buff, varying to a deeper tint on throat and chest; a pure white spot on chin; muzzle blackish; anterior borders of ears invaded by buffy tone of head; rest of ears and postauricular spots deep black; feet white; tail brownish above on basal two-thirds, whitish below, and white all around on terminal third. Skull.—Closely resembling that of 7. b. desitus, but braincase somewhat lower; zygomata usually more slender; nasals more wedge-shaped, narrower posteriorly; audital bullae slightly smaller, dentition about the same. Com- pared with that of T. b. desertorum the skull differs mainly in decidedly lar- ger size. Measurements —Type: Total length, 245 mm; tail, 78; hind foot, 31.5. Skull (type): Occipitonasal length, 40; zygomatic breadth, 25; breadth across squamosals (over mastoids), 20; interorbital constriction, 6.7; length of nasals, 14.8; maxillary toothrow (alveoli), 7.8. Remarks.—This high mountain form most closely resembles its near geographic neighbor, desztus, but is distinguished by paler color. It requires Marcu 15, 1936 GOLDMAN: POCKET GOPHERS 115 no close comparison with chrysonotus which is much larger and still paler, or with fulvus which is much darker. Specimens examined.—Seven, all from the type locality. Thomomys bottae internatus, subsp. nov. Upper Arkansas River Valley Pocket Gopher Type.—From Salida, Chaffee County, Colorado (altitude 7,000 feet). No. 150997, & adult, skin and skull, U. 8S. National Museum (Biological Survey collection); collected by Merritt Cary, November 10, 1907. Original number 1247. Distribution.—High valleys along the eastern side of the Rocky Moun- tains from the Upper Arkansas River Valley, Colorado, south to north- eastern New Mexico. General characters——A large, ochraceous buffy or tawny subspecies. Closely resembling Thomomys bottae pervagus of the Upper Rio Grande Val- ley, New Mexico, but smaller; color very similar; cranial details distinctive. Similar in general to Thomomys bottae rucdosae of south-central New Mexico, but smaller; color much lighter, more uniform, not strongly mixed with «black as in ruwzdosae; skull also different. Color —Type: Upper and under parts between ochraceous buff and tawny (Ridgway, 1912), the top of head and back faintly darkened by black- tipped hairs; muzzle blackish, except lips which are whitish; ears encircled by black; feet white; tail brownish above, white below, becoming whitish all around at tip. In some specimens the under parts vary to light ochraceous buff. Skull.—Very similar to that of T. b. pervagus, but smaller and lighter in structure; premaxillae usually less extended posteriorly beyond ends of nasals; audital bullae smaller; dentition lighter. Compared with that of T.b. ruidosae the skull is larger and more angular, the temporal ridges more prominent; zygomata less strongly bowed outward posteriorly (widest an- teriorly); maxillary arm of zygoma relatively heavier; dentition similar. Measurements.—Type: Total length, 233 mm; tail, 74; hind foot, 32. Five adult female topotypes: 231 (220-239); 76 (73-80); 32 (31-34). Skull (type): Occipitonasal length, 39.1; zygomatic breadth, 24.2; breadth across squamosals (over mastoids), 19.7; interorbital constriction, 6.7; length of nasals, 13.9; maxillary toothrow (alveoli), 8.1. Remarks.—The range of T. b. internatus seems to represent an extension of the T. bottae group northward along the east side of the Rocky Moun- tains. In general characters the present form closely approaches T'. b. pervagus from which, however, it appears to be completely isolated by the high mountains inhabited by 7. fossor to the westward. Specimens from northeastern New Mexico grade toward T’. b. ruidosae. Specimens examined.—Total number, 19, as follows: Cotorapbo: Gardner, 2; Salida, 8. New Mexico: Folsom, 2; Oak Canyon (near Folsom), 2; Sierra Grande, 4; Trinchera Pass (mouth), Colfax County, 1. 116 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 26, NO. 3 Thomomys bottae howelli, subsp. nov. Grand Junction Pocket Gopher Type.—From Grand Junction, Mesa County, Colorado (altitude 4,600 feet). No 75684, 9 adult, skin and skull, U. 8. National Museum (Biological Survey collection); collected by A. H. Howell, November 7, 1895. Original number 493. Distribution.—Known only from the type locality in the Grand River Valley, western Colorado. General characters ——A rather large, pallid subspecies with a broad, flat- tened cranium. Similar to the palest specimens of Thomomys bottae aureus of the San Juan River Valley, southeastern Utah, in color, but under parts more thinly overlaid with buffy white, and cranial characters, especially the broad, flat braincase, distinctive. Approaching Thomomys bottae osgoodi of the Fremont River Valley, Utah, in color, but much larger and skull widely different. Color.—Type (winter pelage): Upper parts in general between tilleul buff and pale olive buff (Ridgway, 1912), somewhat darkened on head by a mix- ture of cinnamon buff and brown; a few inconspicuous dusky-tipped hairs along median line of back; muzzle dusky; ears and postauricular spots deep, contrasting black; under parts thinly overlaid with buffy white, the hairs becoming pure white to roots on inguinal region; thighs pure white to roots © all around; feet white; tail buffy whitish, slightly paler below than above. Skull—Similar in general to that of 7. b. aureus, but braincase conspicu- ously broader and flatter; zygomata more widely spreading; nasals shorter; premaxillae more attenuate posteriorly; interparietal larger; audital bullae more rounded and fully inflated anteriorly; incisors short, as in aureus, but less strongly recurved. Compared with that of 7. b. osgoodi the skull is much larger, with flatter braincase, shorter nasals, and posteriorly narrower pre- maxillae. Measurements—Type: Total length, 219 mm; tail, 71; hind foot, 29. Skull (type): Occipitonasal length, 36.8; zygomatic breadth, 23.7; breadth across squamosals (over mastoids), 20; height of braincase (over audital bullae), 12.1; interorbital constriction, 7.1; length of nasals, 10.8; maxillary toothrow (alveoli), 7.4. Remarks.—T. b. howelli is based on a single specimen exhibiting characters which, in view of geographic isolation, seem to warrant subspecific recogni- tion. It is more closely allied to 7. b. aureus than to any other known form. Thomomys bottae optabilis, subsp. nov. Naturita Creek Valley Pocket Gopher Type.—From Coventry, Naturita Creek Valley, Montrose County, Col- orado (altitude 6,500 feet). No. 149962, @ adult, skin and skull, U. S. Na- tional Museum (Biological Survey collection); collected by Merritt Cary, July 31, 1907. Original number 1105. Distribution.—Cultivated flats between Naturita Creek and the San Mig- uel River, southern Montrose County, Colorado; limits of range unknown. General characters.—A large, cinnamon-buff subspecies. Size about as in the allied form, Thomomys botiae aureus of the San Juan River Valley, south- eastern Utah, but color decidedly darker and cranial details, especially the more widely spreading zygomata, distinctive. Also similar in size to the Marcu 15, 1936 GOLDMAN: POCKET GOPHERS 117 higher mountain form, Thomomys bottae apache, of the Jicarilla Indian Reservation, northwestern New Mexico, but color richer, more buffy, less dusky, and skull combining differential features. Color.—Type (acquiring summer pelage): Upper parts near cinnamon buff (Ridgway, 1912), purest along sides, finely and evenly mixed with black on top of head and over back; under parts in general overlaid with pinkish buff; hairs on chin and middle of throat pure white to roots; fore- arms and thighs like under parts; muzzle and middle of face blackish; ears and postauricular areas black; feet white; tail buffy grayish above, some- what lighter below to near tip which is silvery white all around. Skull—tbLarge, but rather light in structure. Similar to that of T. b. aureus, but less robust; frontonasal region more depressed along median line; zygomata more slender, but more widely and squarely spreading, the antero-external angle weakly developed; nasals broader posteriorly, less wedge-shaped, the sides more nearly parallel; premaxillae narrower poste- riorly; anterior nares higher, the nasals less flattened above; exposed portion of upper incisors longer, more procumbent. Very similar in general to that of T. b. apache, but lighter in structure; rostrum more slender; premaxillae narrower; frontal region narrower, more constricted; exposed portion of incisors longer. Measurements.—Type: Total length, 250 mm; tail, 76; hind foot, 32. Skull (type): Occipitonasal length, 42.9; zygomatic breadth (at antero- external angle), 27.2; breadth across squamosals (over mastoids), 21.4; in- terorbital constriction, 6.4; length of nasals, 14.2; maxillary toothrow (alveoli), 8.2. Remarks.—T. b. optabilis inhabits high valley areas, above the range of T. b. aureus, along the western side of the Rocky Mountains. In cranial characters it approaches TJ. b. apache, but is readily separated by richer coloration. It requires no close comparison with the pallid subspecies, 7’. b. howell, of the Grand River Valley. Specimens examined.—Two, from the type locality. Thomomys bottae guadalupensis, subsp. nov. Guadalupe Mountains Pocket Gopher Type.—F¥rom McKittrick Canyon, Guadalupe Mountains, Texas (alti- tude 7,800 feet). No. 109225, @ adult, skin and skull, U.S. National Museum (Biological Survey collection) ; collected by Vernon Bailey, August 22, 1901. Original number 7821. Disiribution.—Guadalupe Mountains of southern New Mexico and west- ern Texas. General characters.—A light colored, medium-sized subspecies; pectoral mammae two pairs as usual in forms of bottae; very similar in general to Thomomys bottae texensis of the Davis Mountains, but color usually lighter; skull more massive, and differing in detail. Contrasting strongly in lighter color, compared with Thomomys bottae ruidosae of south central New Mexico, and cranial characters also distinctive. Color.—Type (summer pelage): Upper parts light ochraceous buff (Ridg- way, 1912), slightly darkened on top of head and over back by black-tipped hairs; under parts near pale ochraceous salmon, this tone extending upward over lower part of sides and including forearms and thighs; ears black, 118 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 26, No. 3 invaded by buff anteriorly; black patches behind ears small; muzzle brown- ish; feet white; tail with a brownish tinge above, white below. Color varying in topotypes to between ochraceous buff and tawny above, and light ochra- ceous buff below. Skull.—Similar to that of T. b. texensis, but broader and heavier; frontal region broader; nasals broader; premaxillae usually less extended beyond ends of nasals posteriorly; incisors shorter, decurvature about the same. Size of cranium about as in T. b. rwidosae, but rostrum and nasals shorter; frontal region broader; zygomata less strongly bowed outward posteriorly; incisors shorter, decurvature about the same. Measurements.—Type: Total length, 218 mm; tail, 64; hind foot, 29. An adult female topotype: 200; 65; 29. Skull (type): Occipitonasal length, 37.1; zygomatic breadth, 23.7; breadth across squamosals (over mastoids), 19.0; interorbital constriction, 6.9; length of nasals, 12.3; maxillary tooth- row (alveoli), 6.9. Remarks.—T. b. guadalupensis is distinguished by pale coloration com- bined with cranial features unlike those of any of its geographic neighbors. The upper incisors are remarkably short in the older adults, especially males. Specimens examined.—Total number, 6, as follows: New Mexico: Guadalupe Mountains (Dog Canyon, 6,800 feet), 2. Trxas: Guadalupe Mountains (McKittrick Canyon), 4. Thomomys lachuguilla limitaris, subsp. nov. Big Bend Pocket Gopher Type.—From four miles west of Boquillas, Brewster County, Texas. No. 110339, o adult, skin and skull, U.S. National Museum (Biological Survey collection); collected by Vernon Bailey, May 28, 1901. Original number 7659. Distribution.—Northern side of the Rio Grande River Valley, from the ae Bend” east at least to Devils River, and north to Castle Mountains, exas. General characters—A small, pale buffy subspecies. Closely allied to Thomomys lachuguilla lachugutlla of the El Paso region, but still smaller, throat and chest white instead of buffy, and skull differing in detail; mam- mae, pectoral two pairs, inguinal two pairs, total eight as in lachugwilla. Similar in size and color to Thomomys wmbrinus perditus, a geographic neighbor south of the Rio Grande, but pectoral mammae two pairs instead of one pair, and cranial characters different. Color.—Type (unworn summer pelage): Upper parts near pinkish buff (Ridgway, 1912), the top of head and back somewhat darkened by black- tipped hairs; under parts overlaid with white; forearms, thighs, and feet white; muzzle brownish; ears black, except anterior margins which are buffy; black postauricular patches rather small; tail thinly clothed with whitish hairs above and below. Skull.—Similar to that of T. 1. lachuguilla in form, but still smaller, more delicate in structure; zygomata relatively more slender; premaxillae less extended beyond ends of nasals posteriorly; audital bullae relatively small- Marcu 15, 1936 GOLDMAN: POCKET GOPHERS 119 er; dentition lighter; upper incisors decurved about as in lachuguclla. Com- pared with that of T. wu. perditus the skull is similar in size and delicate structure, but the zygomata are less strongly bowed outward, the sides more nearly parallel; nasals less wedge-shaped, the posterior ends truncate instead of emarginate; lachrymals articulating less broadly with maxillae, as viewed from above; upper incisors more decurved, less procumbent Measurements.—Type: Total length, 200 mm; tail, 67; hind foot, 27. An adult female topotype: 208; 66; 28. Skull (type): Occipitonasal length, 32.5; zygomatic breadth, 20.2; breadth across squamosals (over mastoids), 17; interorbital constriction, 7; length of nasals, 10.8; maxillary toothrow (alveoli), 6.8. Remarks.—This new form is based upon a few specimens from several localities that exhibit too great a departure in cranial details for satisfactory reference to typical lachuguzlla. It bears a superficial resemblance to Thomomys umbrinus perditus, but a summation of characters indicate specific distinction. Specimens examined.—Total number, 9, all from Texas, as follows: Boquillas (type locality), 2; Castle Mountains, 1; Comstock, 3; Devils River, 13 miles below Juno, 1; Marathon, 1; Samuels, 19 miles west of Langtry, 1. Thomomys lachuguilla confinalis, subsp. nov. Rock Springs Pocket Gopher Type.—From 35 miles east of Rock Springs, Texas (altitude 2,450 feet). No. 117571, & subadult, skin and skull, U. S. National Museum (Biological Survey collection); collected by Vernon Bailey, July 11, 1902. Original number 7910. Distribution.—Known only from the type locality in the upper part of the Nueces River Valley, central southern Texas. General characters—A small, cinnamon form, with weakly developed skull. Similar in general to Thomomys lachuguilla lamitaris, but color richer, near cinnamon instead of pinkish buff, and skull more delicate in structure. Color.—Type (acquiring fresh pelage): Head and anterior part of back near cinnamon (Ridgway, 1912), moderately mixed with black; rest of upper parts in worn pelage dull grayish; under parts white, the hairs white to roots on throat, under sides of forearms and inguinal region; forearms tinged with buff along outer sides; muzzle blackish; ears black, becoming buffy near anterior base; forefeet, thighs, and hind feet white; tail light brownish above, white below. Skull.—Similar in size to that of T. 1. limitarzs, but more slender; zygomata narrower and tending to converge anteriorly, the sides less nearly parallel; premaxillae more prolonged beyond ends of nasals posteriorly; interparietal quadrate and nasals nearly truncate posteriorly as in limitaris; palate nar- “lee dentition lighter; incisors thinner and narrower; maxillary toothrow shorter. Measurements.—Type: Total length, 200 mm; tail, 60; hind foot, 28. Skull (type): Occipitonasal length, 32.8, zygomatic breadth, 19.4; breadth across squamosals (over mastoids), 16.5; interorbital constrictions, 6.2; length of nasals, 10.8; maxillary toothrow (alveoli), 5.9. 120 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 26, NO. 3 Remarks.—T. |. confinalis is based upon a single specimen representing the extreme eastern limit of the known range of the genus in Texas. The combination of color and cranial details appear to be well beyond the range of individual variation in T. I. limitarzs, the only form with which it requires close comparison. Thomomys pectoralis, sp. nov. Carlsbad Pocket Gopher Type.—From vicinity of Carlsbad Cave, Carlsbad Cave National Monu- ment, Eddy County, New Mexico. No. 244372, o adult, skin and skull, U.S. National Museum (Biological Survey collection) ; collected by Vernon Bailey, March 17, 1924. Original number 10222. Distribution.—Known only from the type locality in the Pecos River Valley, southeastern New Mexico. General characters —A small, pinkish buffy species, closely resembling Thomomys lachuguilla of the Rio Grande Valley region near El Paso, Texas, but smaller; pectoral mammae apparently limited to a single pair, instead of two pairs as in lachugwilla; skull smaller, less massive, and differing in detail. Size much smaller and color paler than in Thomomys bottae guadalu- pensis of the neighboring Guadalupe Mountains; pectoral mammae appar- ently one pair, instead of two pairs as in guadalupensis; cranial characters quite different. Color—Type: Upper parts near pinkish buff (Ridgway, 1912), slightly darkened on head and over dorsum by admixture of black-tipped hairs; under parts overlaid with pinkish buff, extending upward to include outer sides of forearms and thighs; muzzle blackish; ears encircled by black; feet white; tail brownish above, white below. Skull.—Similar in general to that of T. lachugwilla, but smaller, less massive; rostrum narrower; zygomata more slender; interparietal more tri- angular; premaxillae less extended beyond ends of nasals posteriorly; bullae smaller; upper incisors thinner and slightly more procumbent. Compared with that of 7. 6. guadelupensis the skull is smaller, less angular, and lighter in structure; interparietal more triangular; bullae more smoothly rounded; dentition similar. Measurements —Type: Total length, 186 mm; tail, 48; hind foot, 25. An adult female topotype: 188; 55; 26. Skull (type): Occipitonasal length, 34; zygomatic breadth, 21.8; breadth across squamosals (over mastoids), 18; interorbital constriction, 6.7; length of nasals, 11.3; maxillary toothrow (alveoli), 6.8. Remarks.—This little pocket gopher presents a departure from the neigh- boring forms in the apparent reduction of the pectoral mammae to a single pair. At least I have been able to find only one pair in a topotype in which the mammae had been functional and are clearly visible. In this character the present form agrees with Thomomys uwmbrinus of Mexico, but differs in important cranial features and requires no close comparison. Specimens examined.—Three, all from the type locality. Marcu 15, 1936 FOX: CHINESE SPIDERS 121 ZOOLOGY .—Chinese spiders of the families Agelenidae, Pisauridae, and Sparassidae.1 Irnvinc Fox (Communicated by C. F. W. MUESEBECE. ) In the course of identification of Chinese spiders which have ac- cumulated in the United States National Museum, several new species were found. These new species, together with others noted below, were collected by Dr. D. C. Graham chiefly in Szechwan Province, China, during the years 1923, to 1930. _ I wish to thank the authorities of the United States National Mu- seum, especially Dr. E. A. Chapin, for placing these collections in my hands, and for their helpfulness and courtesy while this work was in progress. Family AGELENIDAE Agelena difficilis, n. sp. Fig. 5 Male.—Total length, 9.40 mm. Carapace, 4.85 mm long, 3.56 mm wide. Abdomen, 4.95 mm long, 3.07 mm wide. Carapace brown with a lighter median band which widens from the eye rows backward for about a fourth of the length of the carapace; it then abruptly narrows to a point and ends at the longitudinal furrow. Sides of the carapace with wide submarginal white stripes, and much narrower dark brown marginal ones. Clypeus reddish brown, chelicerae lighter, clothed with long white hairs. Sternum, labium, and endites reddish, darker than the coxae. Legs uniform light yellow, clothed with white hairs. Dorsum and sides of the abdomen dark brown, venter with a broad brown stripe extend- ing from the epigastric furrow to the spinnerets. Spinnerets light brown, the last pair about twice as long as the first pair. The distal joint of the last pair of spinnerets about twice the length of the basal. Eyes characteristically in two strongly procurved rows. Anterior median eyes closer to the laterals than to each other and slightly larger than the laterals. Eyes of the second row equal in size, equidistant, spaced about a diameter apart. The lateral eyes of both rows subequal and subcontiguous. Median ocular quadrangle about as long as wide, slightly wider in front than behind (15/14), the posterior median eyes about five-sixths the size of the anterior median. Clypeus equal in height to one and one-third times the diameter of an anterior median eye. Chelicerae with five teeth on the lower margin and three on the upper. Tibia I armed below with a submedian and a basal pair of spines; tibiae II, III, IV, with a single submedian and a single basal spine. Tibia and patella I, 5.94 mm long. Tibia and patella IV, 5.84 mm long. Patella of the palpus about five-sevenths the length of the tibia, both together about one-half the length of the femur. Distal edge of the patella with two sharp black spurs. Tibia hook-like inwardly. Type locality China: Male holotype and 2 male paratypes from Suifu, Szechwan Province, October, 1930. U.S.N.M. Cat. No. 1149. This spider resembles Agelena japonica Karsch in the possession of an embolus which is robust and projects forward. The embolus differs from 1 Received November 11, 1935. 122 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 26, NO. 3 that of the latter species, however, in being spiral-shaped basally and form- ing a distinct coil distally, as is shown in Figure 5. Agelena injuria, n. sp. Fig. 3 Male.—Total length, 11.78 mm. Carapace, 5.74 mm long, 4.85 mm wide. Abdomen, 6.14 mm long, 3.17 mm wide. The carapace of this specimen is damaged, and therefore somewhat dis- torted. It is reddish brown with a faint lighter median band that is as wide as the eye rows anteriorly, but converges posteriorly to a narrower portion which disappears in the pronounced longitudinal groove. Sides of the cara- pace with wide submarginal bands of whitish hairs. Clypeus dark, chelicerae lighter, sparsely provided with white hairs. Sternum, labium, and endites dark red; labium and endites with much lighter edges. Legs light reddish or orange, patella, and the proximal ends of the femur and tibia darker. Dorsum of the abdomen dark brown, tinged with red at the base. Sides of the abdomen brown, the venter much paler, furnished with whitish stripes, one on each side, which more or less outline it. Spinnerets light brown in color, with the last pair about twice the length of the first pair. Distal joint of the last pair of spinnerets much longer than the basal (14/6). Eyes characteristically in two strongly procurved rows. Eyes of the first row about the same size, the medians closer to the laterals than to each other. Eyes of the second row with the medians smaller than the laterals, equidistant, about one diameter of a posterior lateral eye apart. The lateral eyes of both rows subequal and subcontiguous. Median ocular quadrangle slightly longer than wide (19/17), wider in front than behind (17/16), the posterior median eyes about five-sevenths as large as the anterior median eyes. Clypeus slightly more than the diameter of an anterior median eye in height. The metatarsi of the posterior legs bear numerous long hairs. Tibia and patella IV, 8.32 mm long. Patella of the palpus less than twice as long as the tibia, both together about three-fifths the length of the femur. Patella provided with a black lateral spur extending outward; tibia with a ventral distal process that does not touch the posterior border of the bulb. Type locality China: Male holotype from Yao-Gi, Mupin, Szechwan Province, 8,000 feet, July 14, 1929; male paratype from Tatsientu, July 20, 1923. Type: U.'S.N.M. Cat. No. 1150. This spider bears considerable resemblance in its general structure to Agelena labyrinthica (L), but is immediately distinguishable from that species by the structure of the palpal organ which has an embolus that is distinct and separated from the process of the bulb. Agelena opulenta L. Koch Agelena opulenta L. Koch. Verh. Zool.—Bot. Gesell. Wien 27: 757, pl. XV, fig. 20. 1877. Records.—China: Yunnan Border, 6,000 ft., October, 1928, female; Szech- wan Province, Suifu, 1,000 ft., October, 1930, two females. Coras luctuosus (L. Koch) Coelotes luctuosus L. Koch. Verh. Zool.—Bot. Gesell. Wien 27: 752, pl. XV, figs. 14, 16, 1877. Record.—China: Szechwan Province, Suifu, 1,000 ft., November 17, 1930, emale. Marcu 15, 1936 FOX: CHINESE SPIDERS 123 Family PISAURIDAE Dolomedes chinesus duoformus, n. subsp. Fig. 2 _ Female.—Total length, 17.33 mm. Carapace, 7.92 mm long, 6.44 mm wide. Abdomen, 9.20 mm long, 5.15 mm wide. Carapace yellowish brown, lighter behind the eyes, darker about the longitudinal groove. Sides of the carapace presenting a mottled appearance having irregular lighter spots and streaks. Posterior lateral eyes situated on black spots. Clypeus with two dark lines that converge at the anterior median eyes; each of these lines extend down a chelicera as a median stripe. Labium and endites reddish yellow, sternum with a central light leaf-like design. Legs with coxae, femora, and patellae blackish mottled with yellow beneath, but reddish above as are the other joints. Abdomen brown, darker than the carapace; venter reddish, with indications of a darker median stripe that extends from the epigynum to the spirfherets. First row of eyes longer than the second (30/26), recurved, the median eyes closer to the laterals than to each other and much larger (7/4). Second row narrower than the third (26/52). The eyes about their diameter apart, each eye seven-ninths its diameter from an anterior lateral eye. Clypeus equal in height to the length of the area formed by the first and second eye rows. Legs moderately long, anterior tibiae with 2—2—2—2 spines below, the last pair apical. Tibia and patella I, 12.18 mm long. Tibia and patella IV, 11.88 mm long. Type locality.—China: Female holotype from between Suifu and Yachow, Szechwan Province, 1,000 ft., June 5, 1929; two female paratypes from Yachow, Szechwan Province, 2,000 ft., July, 1928; one female paratype from Ningyuen Fu, Szechwan Province, 6,200 ft., July 31, 1928. Type: U.S.N.M. Cat. No. 1151. This subspecies differs from D. chinesus Chamberlin in being much smaller in size, and in lacking a dorsal dark band outlined by white stripes. The epigyna, although apparently identical, differ slightly in that the median piece of D. chinesus duoformus, new subspecies is wider than that of D. chinesus Chamberlin, although the latter spider is as a whole larger in size. Dolomedes insurgens Chamberlin Dolomedes insurgens Chamberlin. Proc. United States Nat. Mus. 63 (art. 13) 25-26, pl. 6, fig. 41. 1924. Records.—China: Szechwan Province, Yachow, May 1928; between Suifu and Yachow, 1000 ft., June, 5 1929, two males; Mt. Omei July, 1921, male; Suifu 2,000 ft., April 18, 1926, male. Pisaura lama Bésenberg and Strand Pisaura lama Bosenberg and Strand. Abh. Senckenb. Naturf. Gesell. 30: 306-307, pl. 13, fig. 340. 1906. Record.—China: ee Province, Yao-Gi, near Mupin, 8,000 ft., July 14, 1929, female. 124 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 26, NO. 3 Fig; 1.—Heteropoda amphora, n. sp., epigynum. Fig. 2.—Dolomedes chinesus duo- formus, n. subsp., epigynum. Fig. 3.—Agelena injuria, n. sp., palpus. Fig. 4.— Heteropoda grahami, n. sp., epigynum. Fig. 5.—Agelena difficilis, n. sp., palpus. Fig. 6.— Heteropoda virgata, n. sp., palpus. Marca 15, 1936 FOX: CHINESE SPIDERS 125 Family SPARASSIDAE Heteropoda amphora, n. sp. Fig. 1 Female.—Total length, 18 mm. Carapace, 6.93 mm long, 6.43 mm wide. Abdomen, 11.00 mm long. Carapace reddish yellow, discolored by deeper red bars or short stripes, and dark, sparsely distributed hairs. Spaces between the eyes red. Clypeus light yellow, with a reddish brown edge. Chelicerae reddish brown with darker longitudinal stripes. Coxae, sternum, and endites clear yellow, labium somewhat darker and edged with brown. Legs and palpi yellowish, femora punctuate with red beneath, bearing three large red spots, each at the base of a spine, on the prolateral surfaces; patellae, tibiae, and metatarsi darker than the femora above. Dorsum of the abdomen deep red, with indications of a lighter mark anteriorly. Venter much lighter, abruptly differentiated from the sides; it bears some indication of a darker median stripe which extends from below the genital fold nearly to the spinnerets. First row of eyes slightly procurved, two-thirds as wide as the second row, which is slightly recurved, with the laterals on protuberances. Anterior median eyes more than a diameter apart, three-fifths of a diameter from the anterior lateral, and much smaller than these (6/13). Posterior median eyes separated from each other by more than a diameter, from the posterior lateral by almost two diameters and also much smaller than the latter (2/3). Median ocular quadrangle longer than wide (15/14), narrower in front than behind (9/14). Clypeus higher than the diameter of an anterior lateral eye (15/12). Chelicerae with four teeth on the lower margin, of which the one furtherest from the claw is the least robust, and three teeth on the upper margin, of which the middle one is the most robust. Labium slightly wider than long, and much shorter than the endites (22/55). Posterior tibiae and metatarsi with 2-2-2 spines below, the last pair apical, metatarsi with 1-1 spines on the lateral surfaces. Anterior tibiae with 2—2—2-—2 spines below, the last pair apical, anterior metatarsi also with 1—1 spines on the lateral surfaces. Trochanters notched. Epigynum chiti- nous, wider than long, flattened below, with the median piece oval, vase- like being constricted anteriorly. Type locality.—China: Female holotype from between Suifu, and Yachow Szechwan Province, 1,000 feet, June 5, 1929. Type: U.S.N.M. Cat. No. 1152. This spider resembles the Ceylonese species, H. Kandiana Pocock in having an epigynum whose median piece is linguiform, and a pale clypeus which is more or less crescentic in shape. It differs in being smaller by six milli- meters, and in possessing an anterior row of eyes which is slightly procurved rather than strongly so. These similarities and differences are derived not from an examination of specimens of H. Kandiana Pocock themselves, but from a study of Pocock’s brief descriptions.? Heteropoda grahami, n. sp. Fig. 4 Female.—Total length, 12.00 mm. Carapace, 4.55 mm long, 4.45 mm wide. Abdomen, 7.52 mm long, 4.95 mm wide. * Jour. Bombay Nat. Hist. Soc. 12: 752. 1899; The fauna of British India, Arach- nida, p. 261, 1900. 126 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 26, NO. 3 Carapace yellowish with red markings at the pars cephalica, and a darker median line which extends from between the posterior median eyes to the rear slope. Sides of the carapace with reddish streaks, and dark margins. Eyes on a red background. Clypeus light yellow; chelicerae also yellow, each with short reddish bars followed by several reddish spots. Sternum yellow with four dark marks on each side, and a larger diamond-shaped central one. Labium and endites yellowish. Legs yellow beneath, all the joints except the metatarsi and tarsi richly provided with small red spots. Femora with three large red spots on the prolateral surfaces. Dorsum of the abdomen having a yellow ground densely provided with irregular red mark- ings, which become more dilute at the sides and venter. The eyes in two slightly recurved rows, with the first row three-fourths as wide as the second. Anterior median eyes about two-thirds of a diameter apart, one-third of a diameter from the anterior lateral, and smaller than these (3/4). Posterior median eyes about a diameter and a third apart, about two diameters from the posterior lateral, and smaller than these (6/7). Median ocular quadrangle slightly longer than wide (20/19), nar- rower in front than behind (15/19). Clypeus higher than the diameter of an anterior lateral eye (5/4). Chelicerae with four teeth on the lower margin, of which the one furthest from the claw is the least robust, and two teeth on the upper margin with strong indication of a third one. Labium slightly wider than long, and much shorter than the endites (15/35). Anterior tibiae with 2—2—2-2 spines below, the last pair apical, 1 subme- dian spine on each lateral surface, and 1—1 above, the first spine basal; an- terior metatarsi with 2—2 spines below, 1—1 on the lateral surfaces and one above. Tibia of fourth leg with 2-2-2 spines below, the last pair apical, 1-1 on the lateral surfaces, and 1 above; metatarsus of fourth leg with 1-1-1 spines below, the last apical and reduced, 1—1—2 on the lateral sur- faces, the last pair apical and reduced, none above. The third legs are like the fourth except that they lack the reduced apical apines on the metatarsi. Legs 2134. Trochanters notched. TABLE 1.—LEG MEASUREMENTS IN MILLIMETERS Femur Patella Tibia |Metatarsus| Tarsus Total I 6.63 2.48 alae 6.04 2.08 | 23.86 — a Lea 7.23 2.67 6.73 6.34 2.18 “boeae (ite Ee are 5.05 4.55 1.58 18.71 ee am 2.18 5s 4.74 | 1.98 || aoe Epigynum not chitinous in front, and open; behind expanded into two projections, which are contiguous at the bases, but widely separated distally. Type locality.—China: Female holotype from Ningyuen Fu, Szechwan Province, July 31, 1928; immature female paratype from Yachow district, Szechwan Province, 1,800 ft., May, 1928. Type: U.S.N.M. Cat. No. 1153. This spider seems to be most closely allied to the Japanese Heteropoda aulicus L. Koch, but differs from that species in the relationship of the eyes and in the structure of the epigynum. Marcu 15, 1936 FOX: CHINESE SPIDERS 127 Heteropoda virgata, n. sp. Fig. 6 Male.—Total length, 10.49 mm. Carapace, 5.45 mm long, 4.75 mm wide. Abdomen, 4.95 mm long, 2.97 mm wide. Carapace clear reddish brown with darker streaks at the sides. Sides of the carapace, clypeus, and chelicerae lighter. Sternum, labium, and endites orange. Legs clear yellow. Dorsum of the abdomen reddish brown, sides lighter. Venter pale, outlined by a dark line on each side which extends from the genital fold almost to the spinnerets. First row of eyes straight or slightly procurved, three-fourths as wide as the second row, which is slightly recurved, with the laterals on protuber- ances. Eyes of the first row about equidistant, the median smaller than the lateral (3/4). Posterior median eyes separated from each other by about a diameter, from the posterior lateral by more than a diameter (11/7), and smaller than these (6/7). Median ocular quadrangle longer than wide (18/21), narrower in front than behind (15/18). Clypeus about as high as the diameter of an anterior lateral eye. Chelicerae with four teeth on the lower margin, of which the one furthest from the claw is the least robust, and three teeth on the upper margin, of which the middle one is the most robust. Labium wider than long (18/15), and much shorter than the endites (15/38). Anterior tibiae with 2—-2—2-2 spines below, the last pair apical, 1 spine laterad, 1-1 above, the first spine basal; anterior metatarsi with 2—2 spines below, 1—1 laterad. Tibia of the fourth leg with 2-2-2 spines below, the last pair apical, 1-1 laterad, 1 above; metatarsus of the fourth leg with 1—1—1 spines below, 1—1—2 laterad. The third leg like the fourth except that it lacks the reduced distal spines on the metatarsus. Legs 2143, trochanters strongly notched. TABLE 2.—LEG MEASUREMENTS IN MILLIMETERS Tr.&Femur| Patella Tibia |Metatarsus| Tarsus Total -_ I 6.83 2-51 ic 6.24 6.92 2.48 25.04 oa II eat 2 2.67 6.83 7.03 2.57 26 .82 _ III 6.53 2.18 5.34 6.14 1.88 i; 22.07 IV Eee | 8 iss ple) Wace SS nme le Sa eines esr The tibia of the palpus bears a hooked projection which is almost as long as the tibia itself. Type locality —China: Male holotype from between Suifu and Yachow, Szechwan Province, 1,000 ft., June, 1929. Type: U.S.N.M. Cat. No. 1154. The unusually small size and glabrous condition of this spider together with the structure of the palpus, which is shown in Figure 6, will immedi- ately distinguish it from other oriental species. Heteropoda forcipata (Karsch) Sarotes forcipatus Karsch Berliner Entom. Zeitschrift 25: 38. 1881. Records.—China: Szechwan Province, near Mupin, 3,000 ft. July 1, 1929, 128 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 26, No. 3 female; Suifu, 1922, female; Washan, 2,000 ft., July 18, 1925, female; Ning- yuen Fu, July 31, 1928, female. Heteropoda venatoria (Linnaeus) Aranea venatorza Linnaeus Syst. Nat., 12th ed., p. 1035. 1767. Records.—China: Szechwan Province, Washan, 2,000 ft., July 18, 1925, female; Suifu, 1922, female; April, 1924, female; April, 1928, two females; March 15, 1929, female. ICHTHYOLOGY.—Description of a new flatfish, with notes on related species.' Isaac GINSBURG, Bureau of Fisheries. (Communi- cated by Waupo L. ScumirTT.) An extensive study of the American species of flatfishes belonging to the genus Paralichthys and the closely related genera Hippoglossina and Pseudorhombus, was carried out by me recently. Since the publi- cation of the final report is likely to be delayed for some time, some of the more interesting results of that study, as well as a description of a new species, are published here separately. Certain errors in the systematics of the species under consideration have been repeated for years and have acquired the weight of tradi- — tion. Nevertheless, the morphological facts compel me to break away from tradition. In a recent valuable book on the systematics of the flatfishes by Norman,’ which is bound to be used as a standard book of reference by students of flatfishes in the years to come, a number of these errors are included. The most important of such current errors are here corrected. The three genera under consideration were incompletely separated heretofore, and as a consequence some of the species were referred to genera to which they do not belong. The only substantial character which has been used for distinguishing Pseudorhombus from Para- lichthys and Hippoglossina was the presence or absence, or the relative development, of the anterior branch of the lateral line. This char- acter was used by previous authors and its use is continued by Norman. However, this character only incompletely separates the groups of species. In the smaller Indo-Pacific species, which belong to Pseudo- rhombus, this accessory branch is generally better developed, reaching the dorsal profile, while in the American species it is usually not as 1 Published by permission of the U. 8S. Commissioner of Fisheries. Received De- cember 30, 1935. ? Norman, J. R. A systematic monograph of the flatfishes (Heterosomata). Vol. 1. Psettodidae, Bothidae, Pleuronectidae. British Museum, London, 1934. Marcu 15, 1936 GINSBURG: FLATFISH 129 well developed; but this is not always true. In some species of Pseudorhombus the accessory branch is not better developed than in most American species; and in californicus, the genotype of Para- lichthys, it is often well developed, reaching or nearly reaching the dorsal profile. Although these facts are shown in part by the outline figures published by Norman, he continues the use of this character in his key which is thus not entirely in accord with his own figures. As a matter of fact, the use of this character as the basis for the major division of the three genera leads to false conclusions. During my studies it was determined that the presence or absence of acces- sory scales constitutes a valuable character for the major division of the species into natural groups. Although this character was neg- lected heretofore, its use leads to a more natural classification of the species. KEY TO THE GENERA a. Accessory scales absent. b. Eyeball and orbit very large to moderately large, interorbital reduced to a mere ridge and origin of dorsal more or less behind anterior margin of eye; the three characters always occurring together. Anterior acces- sory branch of lateral line often rather poorly developed.............. ee ees Sew ee os Hippoglossina Steindachner bb. Eyeball and orbit usually small or moderate, sometimes moderately large; interorbital usually wider than a mere ridge, sometimes reduced to a narrow ridge; origin of dorsal usually in front of anterior margin of eye, sometimes behind its anterior margin; sometimes approaching Hippoglossina in one or another of these characters, but the three usually not occurring together. Accessory branch of lateral line usually but not always reaching dorsal profile...... Pseudorhombus Bleeker aa. Accessory scales always present, except in small specimens, their appearance with respect to size differing with the species.............. ss. Dice itch RIE ESI et a eee Paralichthys Girard By the use of the above synopsis Paralichthys may in practice be distinguished satisfactorily, except for the smaller specimens; but much remains to be desired with respect to the separation of Hzip- poglossina and Pseudorhombus. The characters given in the key are the best known at present for separating those two genera, and judging solely by these characters, the best course would seem to be to unite them. However, each one of the two groups of contained species has a distinctive physiognomy and it seems highly probable that further study will reveal more satisfactory characters for sepa- rating them. Current usage is therefore continued and the two genera 130 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 26, No. 3 are recognized as distinct for the time being. No matter how limited the synopsis may be for use in practice, it evidently groups the species in accordance with their true natural relationship. Hippoglossina mystacium, n. sp. Fig. 1 Description.—On eyed side scales ctenoid on body; mostly cycloid on head, but many weakly ctenoid scales present. On blind side, ctenoid scales present on posterior part of body, the ctenoid scales extending on midline to a distance behind arch about equal to 1/2 of its chord; scales on head and on body anteriorly cycloid. Scales in 52 rows over straight part of lateral line to end of hypural; 28 perforate scales in arch; 26 rows in a chord sub- tending the arch. Three cycloid, embedded scales on maxillary. Accessory scales absent. Gill rakers, 3 comparatively long ones on upper limb, with pty ry EM z & 4 iN ay Pag rai ea) ey if: HOO NSH ye SO A db) 9 Win) ) PUG iy ye FIBA ISI JOY). FENG OOS DD Ay dyed FOE PAN AR I) Bey ey hay) Py ase a) LNRM MD Nie ean MII? III) Ye) YY) 9/y2)y?y 4 x : y) yr Paty!) )-).) \ ) yy) yy). , MIDI )G Boia Be Hi nae A WAY M WD N HAY Ary), =)=) =) =)=) 3) 9)'-))2) 9 ©) =)<)=) 2) 32)2)2)2) 229) LAKIN RRP LLLIDAAMARRAE PRP TDPRI pps yar eH ) eer > MK AVERT PERN) ) is Nk ), 7}. y3 y iy A HN PAAR nt PUNT BORON \ Fig. 1.—Hippoglossina mystacium, n. sp. Length of specimen 18.3 cm. Drawn from the type by Louella E. Cable. j Pi) ie) Oey, At ye ) es) ) ) os 2 widely spaced and very small ones above; 12 on lower limb (same on both sides). Anal rays 55; dorsal rays 66; pectoral rays 11 on eyed side, 10 on blind side. Origin of dorsal nearly over middle of eye. Eye notably large; interorbital a mere ridge. Anterior teeth but slightly enlarged. Maxillary extending to a vertical through posterior margin of pupil, conspicuously narrow posteriorly. Sinistral. Color nearly faded, traces of 6 spots in 2 lengthwise rows present, like in the other species of the genus, somewhat nearer to upper and lower pro- files than to straight part of lateral line, every spot in either row placed on a transverse line with its fellow in the other row; first pair of spots on a transverse line through about middle of arch in lateral line, second pair on a vertical somewhat nearer to head than to base of caudal, third pair not far from ends of dorsal and anal fins; traces of smaller spots on caudal peduncle, one each near upper and lower posterior angles, at base of caudal rays, these two spots being continued to a slight extent on blind side. Measurements of type.—Total length 183 mm. Standard length 146.5 mm. Greatest depth 39.8, least depth of caudal peduncle 9.5, head 30.7, length of maxillary 13.6, greatest width of maxillary 3, orbit 10.4, eyeball 8.5, snout (to margin of upper orbit) 6.5, left pectoral 17.9, right pectoral 13.1, —— Marcu 15, 1936 GINSBURG: FLATFISH 131 left ventral 9.1, right ventral 8.8, caudal 24.9, and straight part of lateral line 57.1% of standard length. Length of chord subtending arch in lateral line 3.6 times in straight part; length of a vertical from chord to apex of arch 3.2 times in arch. Holotype.—U.S.N.M. 77393, near Taitao Peninsula, Chile; Albatross Station 2787; lat. 46°47/30” S., long. 75°15’ W.; 61 fath.; Feb. 9, 1888. Comparison.—The nearest relatives of this species are Hzippoglossina stomata EKigenmann and Eigenmann, H. bollmani Gilbert and H. macrops Steindachner. The specimen described was directly compared with speci- mens of the former two species, including their types; but for its comparison with macrops, I had to rely on the original account. It is most closely related to stomata, differing in having a shorter and narrower maxillary, and a shorter head. It further differs in that the maxillary is almost bare of scales, while that of stomata has a small patch of ctenoid scales. The present species is more remotely related to bollmanz. It differs strikingly from bollmani in having more numerous gill rakers, and the anal rays are also more numerous. The number of scales and dorsal rays falls near the upper end, but outside the frequency distribution of that of bollmant. The ctenoid scales on the blind side extend more forward in bollmani. The body is deeper than in bollmani; but the two species approach closely in the length and width of the maxillary and the length of the head. As compared with the description of macrops, the present species has a shorter head, a more slender body and the ctenoid scales on the blind side extend more forward. Hippoglossina oblonga (Mitchill) This species from the east coast, which is common enough to enter the commercial catch, although its market possibilities are limited by its com- paratively small size, has been placed universally, except by the early authors, in Paralichthys. However, unlike all species of Paralichthys it lacks accessory scales. In this it agrees with the species of Hippoglossina. Also, it always has some ctenoid scales on the blind side, a character normally present in most species of Hippoglossina but not in those of Paralichthys, except to some slight extent in infrequent individual variants. Furthermore, this species has a very narrow interorbital, a comparatively large eye and relatively small teeth, nearly agreeing with the species of Hzppoglossina in these respects and unlike all species of Paralichthys. It is evident that this species is congeneric with the other species of Hzppoglossina. 7. oblonga is evidently most nearly related to Lioglossina tetrophthalmus Gilbert from the west coast. After placing it where it properly belongs in the system, the boundary hitherto drawn between the genera Lioglossina and Hippoglossina largely breaks down, although they may be recognized as subgenera. Pseudorhombus isosceles (Jordan) This species likewise lacks accessory scales and has ctenoid scales on the blind side, and consequently must be removed from Paralichthys. It fairly agrees with the other species of Pseudorhombus and is the only known American representative of that genus. This species, the genotype of Tarphops, and four other Indo-Pacific species have ctenoid scales on the blind side. Judging by the morphology of the species of these three closely related genera, this character forms a more adequate and natural basis for separation than the number of scales. It would seem, therefore, desirable on grounds of morphology to rearrange the species of Tarphops and Pseudo- 132 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 26, No. 3 rhombus and place them in two subgenera based on the presence or absence of ctenoid scales on the blind side. In that case, zsosceles will fall in the subgenus Tarphops. Paralichthys triocellatus Miranda Ribeiro, judging by the original account, is probably a synonym of zsosceles. Four specimens of a flounder obtained at Cape Frio, Brazil, by the Terra Nova and recorded by Regan as Para- lichthys oblongus are described and figured by Norman (op. cit., p. 80), and referred to triocellatus. Norman describes these specimens as having cycloid scales on the blind side. If trzocellatus is in fact a synonym of zsosceles, the four specimens forming the basis of Norman’s account must represent an unnamed species. Even if trzocellatus is distinct from isosceles, Norman’s specimens evidently have fewer scales than Miranda Ribeiro’s fish and they probably represent a new species anyway. Paralichthys patagonicus Jordan and Goss This species is placed by Norman in the synonymy of voraz (his brasil- vensis). However, it has ctenoid scales on the eyed side and is entirely dis- tinct from that species. P. bicyclophorus Miranda Ribeiro which Norman recognizes, is possibly a synonym of patagonicus. At least, the original ac- count of bicyclophorus fails to show how it differs from this species. Paralichthys brasiliensis (Ranzani) Something may be said about the nomenclature of this and another spe- cies. Norman resurrects a name out of the synonymy and designates this species as P. orbignyana (Valenciennes), and uses the name of braszliensis for an entirely different species, although he apparently did not examine the types on which either one of those two names were based. This shifting about of the names of species, one of them well established, seems unfortu- nate. This is a common species on the coast of Brazil which has been de- scribed and recorded a number of times by American authors to whom it was known for more than half a century under the name of brasiliensvs. There are those biologists who claim that a well established name should be retained regardless of priority and there are some cogent arguments which may be advanced in favor of that contention. Without discussing the pros and cons of this proposition, I think that it will be generally admitted that, at least, a well established name is not to be changed unless good and sufficient reasons are advanced for the change. In the present case the only way of definitely determining the question is by a reexamination of the types of both, brasiliensis and orbignyana, since the original accounts are not sufficient to identify the particular species. Pending such study I continue the use of the well established name brasiliensis for this species. Judging by the material in the U. 8. National Museum and that recorded by Norman in the British Museum, the present species is much more common than the following. Considering probabilities alone, therefore, the chances are much greater that Ranzani had specimens of the present species. Paralichthys vorax (Giinther) American writers have generally placed the name voraz in the synonymy of their brasilzensts; while Norman who designates the brasiliensis of Ameri- can authors as orbignyana, describes this species under the name of braszl- zensis. We thus have a nice, and possibly unnecessary confusion of names. Norman, who studied the types of voraz, correctly distinguishes this species from the brasiliensis of American authors. Marcu 15, 1936 GINSBURG: FLATFISH 133 It is remarkable that in the structural characters which I studied in de- tail, such as the number of gill rakers, fin rays, scales and proportional measurements, this species agrees or very nearly agrees with albigutta from the east coast of the United States, although the geographic ranges of the two species are widely discontinuous. The only substantial difference is found in the color, voraz lacking the ocellated spots characteristic of albz- gutta. There is only one specimen of this species in the U. 8. National Museum. Norman records only two specimens, the types, in the British Museum. That author also lists with a query one large, stuffed specimen from Fort Famine, Magellan Strait. This is apparently the same specimen which Ginther described as Pseudorhombus dentatus, stating that the scales are “minutely ciliated.’’ The presence or absence of ctenoid scales in large speci- mens is always a good specific character in Paralichthys. This specimen, therefore, represents either patagonicus or an unnamed species. Paralichthys lethostigma Jordan and Gilbert Norman evidently did not well separate his material of Paralichthys from the east coast of the United States. The specimen from Beaufort, North Carolina, which he records under this species is an unusually slender ex- ample of albigutta. The specimen which Norman lists from Tobago can hardly be a lethostzgma considering the comparatively limited geographical distribution of the species of Paralichthys. It may likely prove to be an example of P. tropicus Ginsburg which is very near in its structural char- acters to lethostigma. The counts of the dorsal and anal rays given by Nor- man for lethostigma range too low. This is evidently due to his inclusion of some albigutta material in his account. The frequency distributions of these counts are very nearly the same in dentatus and lethostigma. The species of Paralichthys are not easy to distinguish. Neverthe- less, if frequency distribution tables of the more important specific characters are prepared, and the several characters of any given specimen are compared with such tables, it becomes a relatively easy matter to refer with assurance individual fish to their proper species. Such tables of the scale, gill raker, and fin ray counts, and compara- tive tables of proportional measurements are included in my manu- script. 134 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 26, No. 3 PROCEEDINGS OF THE ACADEMY AND AFFILIATED SOCIETIES THE ACADEMY 38TH ANNUAL MEETING The 38th Annual Meeting of the Washington Academy of Sciences was held in the Assembly Hall of the Cosmos Club, January 16, 1936, with 45 members present. President McCoy called the meeting to order at 9:05 P.M. The minutes of the 37th Annual Meeting were read and approved. The Corresponding Secretary, Dr. Paut E. Hown, submitted the following report on the membership and activities of the Academy: Membership: During the year 1935, 18 persons were elected to regular membership, 15 of whom had accepted and qualified for membership before the end of the year. One was elected to honorary membership: Dr. JuLEs ScHOKALSKY, Leningrad, U.S.S.R., in recognition of his distinguished con- tributions to physical geography, oceanography and meteorology. Sixteen resignations were accepted, of which 11 were resident and 5 non- resident members. The Academy lost by death 14. The net loss in member- ship was therefore 15 or 2.7 per cent. The members of the Academy stood in respect as the Secretary read the names of those lost by death. Members: Wiuuiam H. Burr, New York, New York, December 138, 1934. JosEPH H. Bryan, Washington, D. C., February 3, 1935. Epwarp 8. Dana, New Haven, Connecticut, June 16, 1935. Lron DomintAn, Montevideo, Uruguay. Marion Dorset, Washington, D. C., June 14, 19385. Ernst G. Fiscnpr, Washington, D. C., September 22, 1935. DanieL R. Harper, Washington, D. C. ALBERT 8. Hitcucock, Washington, D. C., December 16, 1935. WattTeR Hovau, Washington, D. C., September 20, 1935. A. Bruck Macauutum, London, Ontario, Canada, April 4, 1934. Davin K. SuHuts, Washington, D. C., October 21, 1935. (CHARLES) Davin Wuits, Washington, D. C., February 6, 1935. Patron: Henry C. Perkins, Washington, D. C. Honorary Member: JOHANNES SCHMIDT, Copenhagen, Denmark. On January 1, 1936, the membership consisted of 14 honorary members, 3 patrons, and 522 members, one of whom was a life member. The total membership was 539, of whom 387 reside in or near the District of Colum- bia, 131 in other parts of the United States, and 21 in foreign countries. The numerical index which the Secretary in previous reports has called the “index of rate of loss of interest” in the organization on the part of its Marcu 15, 1936 PROCEEDINGS: THE ACADEMY 135 members as measured by the number per year per hundred who resign or are dropped, has the value of 2.7 per cent for 1935. The Board of Managers held five meetings with an average attendance of 12. The 29th edition of the Red Book was published during the year. The Academy was represented by Dr. A. 8. Hitcucock at the Inter- national Botanical Congress held in Amsterdam in September, 1935. The Recording Secretary presented the following report: The 38th year of the Academy began with the 262nd meeting and ended with the 268th meeting. Of these seven meetings, two were joint meetings, one with the Medical Society, one with the Philosophical Society. All were held in the Assembly Hall of the Cosmos Club. The 262nd meeting was held on Friday, February 15, 1935, with about 70 persons present. President McCoy introduced the speaker, retiring president L. B. TuckERMAN, who addressed the Academy upon Fiction in measurement. The 263rd meeting was held March 8, 1935, with about 40 persons present. Mr. W. M. Corssz presided and introduced Principal CHarLes A. Epwarps of the University College, Swansea, Wales. The title of his address was Science, education and industry: whither drifting ? The 264th meeting was held on Thursday, March 21, 1935, with about 125 persons present. Past-president L. B. TucKERMAN introduced Professor RAYMOND PEARL, of Johns Hopkins University, who spoke upon Biology and human trends. The 265th meeting was held Thursday, April 18, 1935, with 42 persons present. Vice-president J. F. Coucn introduced RussEtu 8. McBripe#, who spoke upon The service and failures of chemistry in the advancement of civil- ization. The 266th meeting was held jointly with the Medical Society of the District of Columbia on November 21, 1935, with about 100 persons present. President McCoy introduced Col. P. M. AsHBurN, M.D., who spoke upon The medical history of the conquest of the Americas in the 16th and 17th centuries. The 267th meeting was held jointly with the Philosophical Society on Thursday, December 19, 1935, with 70 persons present. President McCoy introduced Dr. Sanrorp V. Larxery, Librarian of the Welch Memorial Library of Johns Hopkins University, who spoke upon The role of sczentists in the Elizabethan Government. The 268th meeting was held on January 16, 1936, with about 110 persons present. Vice-president O. H. GisH introduced the retiring president, Dr. GzrorGEe W. McCoy, who spoke upon the Comings and goings of epidemics. At the end of the meeting, the vice-president declared a recess preceding the 38th Annual Meeting. The report of the Treasurer, H. G. Avers, was read by Mr. Howarp 8. RAPPLEYE: 136 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 26, NO. 3 CASH RECEIPTS AND DISBURSEMENTS Receipts Krom! Back Ditesti. nee. eee eee oe $ 265.00 Krom: Dies torelSor wes a. eee ee 2,390 .00 From, DuestoraliO36. 5. sce. ac aa. ee ee re 30.00 From Subsertptions for l935. >= 4... ee 743 .60 Krom= Subscriptions torelOse. . 20-45. 4a 281.40 Krom Sales of Jouwnmals.go% 2. ge 40 2 ee 102.87 Erom: Payments for veprimts...)0.4.-e2 oo eee 238 .83 From: Salessor 1932 -Directory:.. ../. bese -e. a 1.80 Kron: Sales of 1935: Directory>: <2. 2.29 2625s 41.70 Brom Interest-on Depesifss o6. s..- = eae ees 25.27 From Interest-oneinvestments 2: ...5..5 6 ee 1,459.70 ‘Rotal -neceiptse -. ssaenc: sh « Slane eee eee oes ae $5,580.17 Cash; Balancerianuathy 1 ehOS5k 7951. Goin oe one, eer ean ae 2,526.54 To becaccounted tors... sos. ink. Cl 2 etn ae ee ae ee ee $8,106.71 Disbursements Hor. pecretary s Once iat <2 2 naan: 4. cee sire ee $ 290.35 Bor Preasurer's:Oftice 19342. 7.2. Os le? oe ge 225 Kor Ereasuners Ofhee 195500 2. 24 fis. Lean ne 226.88 For Journal’ Othiee, V9S40" 2 eae - so ae eee 46.94 Kor JourmalsOiice W935... 2. a5 as a 324.72 For Journal Printing 1934.9 0, eee. fe ee 384.65 Hor Journal Printing, 1oSon80 2 90. os se i eee 2,753 .24 MorcRe prints’. 26). ous ele oe ke nee 495.83 For lilstrations< 2 ceca) oo oe ee 360 .46 For Meetings Committee, 1934. ......5..2554 20.00 For’ Meetings Committee, 1935.0... 30..5 229: 201.50 RoruDireetory ey ss ee ee ee eee oe 629 .37 Bank Debit Memos, as follows: DUS Pe ir ees 21 aed ea Ber ame $0.35 SUDSCRIPUON Se, <8. wae eet ee es ee wal Saleslon Journals: / ea0t es pees .50 FUGPTIMNS 5 Seka oor barre ose ee ars 9) intron Investments. a2). 5 auc sein 44 {7 Total Disbursements... ose tee oe ee ee $5,737 .94 Cash: Balance. December ole 1935" 45h. ne a ae aie 2,368.77 Totals ©... fee Aue, Cee re. ee RRO hese eee See $8,106.71 Note: Of the above expenditures, the amount of $453.84 was paid covering bills chargeable to 1934. He listed the investments of the Academy as totalling $21,096.37. The Auditing Committee, Howarp 8. RappLtnyn, W. H. Brapiey and JAMES F’. Coucu, reported: “The Treasurer’s records of receipts and expenditures as shown in his account books and included in his report have been examined and found correct. All vouchers have been examined and found to be correct and properly approved. The balance sheets submitted by the bank and the securities listed in the Treasurer’s report have been examined. The state- ment of the assets of the Academy was found correct. The records of the Treasurer’s Office have been carefully and systematically kept, thus greatly facilitating the work of the Auditing Committee.” Makcu 15, 1936 PROCEEDINGS: THE ACADEMY 137 The Board of Editors, Jonn A. Stevenson, F. G. BrRickWEDDE and Roxanp. W. Brown, submitted the following report covering the publica- tion of Volume 25 of the Journal for the year 1935: 7 “Volume 25 consisted of 588 pages, including an eight-page index. This compares with 576 pages in 1934, 588 in 1933, 572 in 1932, and 552 in 1931. It contained 75 original papers as contrasted with 78 in 1934, 77 in 1933, and 79 in 1932. Thirty-seven papers were by members of the Academy, and 38 were communicated of which latter number several were by authors who became members of the Academy after the date of receipt of their papers. Original papers were illustrated by 73 line cuts and 10 half-tones. Excess cuts illustrating several papers were paid for by the respective authors. Space in the volume was distributed among the different sciences, as follows: Pages 8 papers on Physics, including Geophysics and Seis- OP LESTE 2g cup sey eRe ae ie eee 106.8 Rees COMO MEMISHTY. 2.2.00. oe ee ee 15.20 Meesecio on O@rystalloeraphy...... 2.62... ee ee 14.20 Peemeetsron narmacolory .. 0. ee e.. 11.6 9 papers on Geology, including stratigraphy and hy- Ser! eee tah Soe do ban ays 6 a 8 40.55 10 papers on Paleontology, including paleobotany...... 34.15 Eapapers on general Biology. ®...................6.: 24.7 reer Ol mOtLANY...... +. Woe... ee ee ee 87.20 2 EES CYA TOS CGS, | ie a 83.40 9 papers on Entomology, Ornithology and Conchology.. 35.20 MeeeeeIsemUrsSVEnOlogy. ... 0... 26s. ee eee lea ce 14.35 Proceedings of the Academy and affiliated societies occupied 54.7 pages, as follows: (as DE TIEIDI lee Ee ee 2.65 Pmapropolorical SOCIeLY............ 2.0... ee eee eee eee 235 Pe MIIEIBO SOIC 5a oe ck eo i ee 1110 PaO ESOC 8! 5026 es. Suns De fk Se he bes 24.45 memmeraraiCal SOCICLY. .. 2... eee ee 1 a Scientific notes and news, and obituaries occupied the remaining 47 pages. As in 1934, scientific notes and news as supplied by Science Service were used to the extent of approximately four pages per number, space given to original papers permitting. Acting upon the recommendation of the Board of Editors, the use of news notes was dispensed with by the Board of Managers at the end of the calendar year. The number of papers in the biological field presented for publication was approximately the same as in the preceding year so that further effort has been made to obtain papers in the field of the physical sciences. The JouRNAL is, relatively speaking, up- to-date with manuscripts submitted to it. The cost of printing, wrapping, mailing and postage was $4.59 per printed page; illustrations 70 cents per page; alterations 10 cents per page; and re- prints 84 cents per page. Of this latter amount approximately one-half was paid for by the respective authors. The cost of the News Service was $300. Exclusive of the cost of the News Service and reprints, the total cost per page for printing the Journal was approximately $5.20, a slight reduction over last year’s cost.” 138 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 26, NO. 3 The Board of Tellers, Grorcr A. GREENBANK, W. H. BrapuLeEy and L. V. Jupson, reported the election of the following officers: President, O. EK. Mrrnzer; Non-resident Vice Presidents, Paut D. Foorn, Pittsburgh, Pennsylvania, and P. C. STANDLEY, Chicago, Illinois; Corresponding Secre- tary, NATHAN R. SmitH; Recording Secretary, CHARLES THom; Treasurer, H. G. Avers; Board of Managers, PAuL E. Hows and H. L. Drypsn. The Corresponding Secretary read the list of nominations for vice-presi- dent submitted by the affiliated societies as follows: Philosophical Society of Washington, Francis B. SILSBEE Anthropological Society of Washington, F. H. H. Rosprerts Biological Society of Washington, Cuas. E. CHAMBLISS Washington Section, American Chemical Society, James H. HIBBEN Entomological Society of Washington, A. H. Cuarx National Geographic Society, FrepERIcK V. CovILLE Geological Society of Washington, W. T. ScHALLER Medical Society of the District of Columbia, H. C. MAcATEER Columbia Historical Society, ALLEN C. CLARK Botanical Society of Washington, CHARLES DRECHSLER Archaeological Society of Washington, ALES HRDLICKA Washington Section, Society of American Foresters, S. B. DETWILER Washington Society of Engineers, Pau C. WHITNEY Washington Section, American Institute of Electrical Engineers, H. G. DorsEY Washington Section, American Society of Mechanical Engineers, H. N. EATON Helminthological Society of Washington, Emmett W. Pricz Washington Branch, Society of American Bacteriologists, H. W. ScHOEN- ING Washington Post, Society of American Military Engineers, C. H. Brrps- EYE Washington Section, Institute of Radio Engineers, Louis CoHEN By vote of the Academy, the Recording Secretary was instructed to cast one vote for the list as read and the vice-presidents were declared elected. President McCoy appointed Past Presidents W. J. HumMpHreys and L. B. TucKERMAN to escort President-elect Mr1nzmr to the chair. President MEINzER took over the gavel and addressed the Academy briefly. Adjournment followed at 9:45. CHARLES THom, Recording Secretary CONTENTS Mepicine.—Comings and goings of epidemics. Groren W. McCoy. . Botany.—Rock midget, a new species of Mimulus from Death Valley, California, Freperick V. COVILUM, <7 92 eee. vg ee Botany.—Leaf venation as a means of distinguishing Crouta from Angelica. Mrrtam L.BoMaARD. 2 0-2. 00. oe a ee ti PaLropotany.—A fig from the Eocene of Virginia. HEpWwarpD vs BYRRY 7 a. 0 ee de Sk se ee ee Slot eee Zootoey.—New pocket agen of the genus Thomomys. E. AL GOLDMAN . 0 6 Sg 8 ss eee BY Oe ee ee eee es Zootogy.—Chinese spiders of the families Agelenidae, Piseuridae, and Sparassidae. Irvine Fox.. Pfr beh ae ieee IcutHyoLoay.—Deseription of a new tae with notes on related _ species. Isaac GINSBURG........ My ate ‘ Procrepines: The ACADEMY. oy sis. 04. 1s dh This Journal is indexed in the International Index to Periodicals Aprit 15, 1936 | No. 4 Ee ea \SHINGTON ACADEMY _ OF SCIENCES BOARD OF EDITORS BrickwepDE = =~—‘ Rowanp W. Brown Esren H. Too.e AU OF STANDARDS _ U. 8. GHOLOGICAL SURVEY BURBAU OF PLANT INDUSTRY ASSOCIATE EDITORS H. T. WENSEL Harotp Morrison PHILOSOPHICAL SOCIETY ENTOMOLOGICAL SOCINTY _E. A. GotpmMan W. W. Rusey BIOLOGICAL SOCIETY GEOLOGICAL SOCIETY AGNES CHASE J. R. SWANTON BOTANICAL SOCIETY ANTHROPOLOGICAL SOCIETY R. E. Grpson CHEMICAL SOCIETY PUBLISHED MONTHLY BY THE WASHINGTON ACADEMY OF SCIENCES 450 Aunarp Sr. at MenasHa, WISCONSIN | Entered as second class matter under the Act of August 24, 1912, at Menasha, Wis. Authorized January 21, 1933.) Journal of the Washington Academy of Sciences This JOURNAL, the official organ of the Washington Academy of Siielieas, publighies (1) short original papers, written or communicated by members of the Academy; (2) proceedings and programs of meetings of the Academy and affiliated societies; (3) notes of events connected with the scientific life of Washington. The JouRNAL is issued monthly, on the fifteenth of each month. Volumes correspond to calendar years. er Manuscripts may be sent to any member of the Board of Editors. It is uyeontiaal re- quested that contributors consult the latest numbers of the JouRNAL and conform their manuscripts to the usage found there as regards arrangement of title, subheads, syn- onymies, footnotes, tables, bibliography, legends for illustrations, and other matter. Manuscripts should be typewritten, double-spaced, on good paper. Footnotes should | be numbered serially in pencil and submitted on a separate ace The editors do not assume responsibility for the ideas expressed by the author, nor can they undertake to correct other than obvious minor errors. Illustrations in excess of one full-page halftone or two full-page line drawings may be paid for by the author. Line drawings are preferred. Proof.—In order to facilitate prompt publication one proof will generally be ee ; to authors in or near Washington. It is urged that manuscript be submitted in final form; the editors will exercise due care in seeing that copy is followed. Author’s Reprints.—Fifty reprints without covers will be furnished orulee Coser bearing the name of the author and title of the article, with inclusive pagination and — date of issue, and reprints, will be furnished at cost when ordered, in accordance with the following schedule of prices. * < Copies 4 pp. 8 pp. 12 pp. 16 pp. Covers | 50 Se ie a ieee eae ene,” shi REL eS $2.00 100 $ .75 $1.20 $1.50 $2.00 248 150 125 1.90 2.55 3.25 3.50 200 L:75 2.60 3.60 4.50 4.25. - 250 2.00 3.30 4.65 5.75 5.00 Envelopes for mailing reprints with the author’s name and address printed in ties ae corner may be obtained at the oles prices. First 100, $4. 00; additional 100, $1.0 oe Bes Subscription Rates—Per volume... ..-2..:00eeceeeeeeeeeneceneese e+ 16. 00 To members of affiliated societies; per volume. Dette cette eee t teen e ees ee 50 | Simele NuMYEFS «055 7S is a ee Oe res TEN pane e (60. Remittances should be made payable to ‘‘Washington Academy of Sciences” and. ae addressed to 450 Ahnaip Street, Menasha, Wis., or to the Treasurer, H. G. Avers, oa Coast and Geodetic Survey, Washington, 1G: 5) ee Exchanges.—The JouRNAL does not exchange with other publications. Missing Numbers will be replaced without charge provided that claim is made to ; the Treasurer within thirty days after date of following issue. OFFICERS OF THE ACADEMY President: O. E. Mrinzzr, U. S. Geological Survey. Corresponding Secretary: N. R. Smiru, Bureau of Plant Industry. Recording Secretary: CHARLES THOM, Bureau of Chemistry and Soils. Treasurer: Hpnry G. Avers, Coast and Geodetic Survey. JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VoL. 26 AprRIL 15, 1936 No. 4 ASTRONOM Y.—Szmon Newcomb, 1835-1909.1 EpGar W. Woo.arp, United States Weather Bureau. Two and a half centuries ago, Sir Isaac Newton presented convinc- ing evidence that the force which holds the moon in its orbit is identi- cal with the force that causes unsupported bodies near the surface of the earth to fall, and that this force conforms to the inverse square law. On the basis of this evidence and Kepler’s empirical laws of planetary motion, Newton proposed the hypothesis of Universal Gravitation. From the general Laws of Motion (also formulated by Newton on the basis of his own and Galileo’s work) and the Law of Gravitation, developed Celestial Mechanics, one of the greatest achievements of the human intellect. The problems of celestial mechanics have always been a constant challenge and a source of inspiration to the mathematician and to the astronomer; and among those who have contributed to this difficult and intricate subject, will be found some of the most illustrious names in the history of the mathematical sciences. Among the select few who have been outstanding masters of the field, none ranks higher than Simon Newcomb, the one hundredth anniversary of whose birth occurred in 1935. Simon Newcomb was one of the most notable men of science that America has ever produced; and his life cannot fail to be an inspira- tion to others—it is a story of boyhood dreams and ambitions, ful- filled by patient and persevering self-effort. He early formed the con- viction that one should choose that sphere in life to which he was most strongly attracted and for which his faculties best fitted him. He himself was irresistibly attracted to celestial mechanics; it seemed to him to embody the highest intellectual power to which man had ever attained—from merely the positions of the celestial bodies in the sky, as observed from night to night, and on the basis of the single fact that each of these bodies is attracted by all the others, as stated 1 Presented before the Maryland-District of Columbia-Virginia Section of the Mathematical Association of America, College Park, Md., December 7, 1935. Re- ceived December 17, 1935. 139 acl 140 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 26, No. 4 by the law of gravitation, the astronomer is to weigh these mighty bodies, determine the vast orbits in which they are moving in space, predict their majestic motions during future time, determine what changes of form and position their orbits will undergo through count- less ages, make maps showing exactly over what cities and towns on the surface of the earth the shadow of the moon, during an eclipse of the sun, will pass fifty years hence or over what regions it passed centuries ago. To thousands of men who could achieve success and wealth in almost any walk of life, to hundreds who could wield em- pires, there would be but one who could succeed with this impressive problem—those who have done it are the select few of the human race—‘‘Nearer the gods no mortal may approach.”’ Simon Newcomb was born March 12, 1835, at Wallace, on the north coast of Nova Scotia; the centenary of his birth was appro- priately commemorated on August 30, 1935, when a cairn erected by the Historic Sites and Monuments Board of Canada was unveiled there by his daughter, Mrs. Joseph Whitney, of Washington, D. C., in the presence of many notable officials. Newcomb’s father was a country school teacher, an occupation which in those days meant an almost nomadic life, and Newcomb’s childhood and boyhood were spent in various parts of Nova Scotia and Prince Edward Island. He was unusually strong of body, mind, and character, rather precocious, and had a conquering power of mental concentration. He received very little formal schooling; when he attended his father’s school at all, he came and went with entire freedom, but he knew the alphabet at the age of 4, began arithmetic at 5, and geography at 6. He avidly read a work on astronomy before he was 10 years old, and was greatly attracted by the subject. With- out the guidance of an instructor, he eagerly studied algebra, natural philosophy, and navigation, mostly from old books that had belonged to his grandfather; and he became enraptured with Euclid at the age of 15. He was uncommonly deficient, however, in skill at any kind of manual labor, particularly that required on a farm; and because of the conditions of life at that time and place, this shortcoming, com- bined with the lack of appreciation by the people around him of his taste for learning, made his boyhood one of sadness. From reading, he gradually formed a vague conception of a different world—‘‘a world of light,’ as he expresses it in his Reminiscences,” ‘“where dwelt men 2 Simon Newcoms. Reminiscences of an astronomer. Boston. 1903. Simon NEWCOMB 142 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 26, NO. 4 who wrote books, and people who knew men who wrote books, where lived boys who went to college and devoted themselves to learning’’— and he longed to get into this world. No possibility of doing so pre- sented itself, however; and circumstances led him to apprentice him- self at the age of 16 to a physician of Moncton, New Brunswick. Newcomb soon discovered that this physician was only a dishonest quack, from whom he could obtain nothing; and after two wasted years, he made his escape by running away. After a painful and diff- cult month, he finally reached Salem, Massachusetts, where he was met by his father who had meanwhile sought his fortune in the United States after the death of Newcomb’s mother. His father finally located in eastern Maryland; and here at the beginning of 1854, Newcomb, at the age of 19, became a country school teacher. He seems to have been successful in that work, though he writes that, looking back, he is deeply impressed with the good nature of the people in tolerating him in the position. While teaching, he spent every spare hour studying all the books on mathematics and astronomy that he could secure or to which he could gain access, inclusive of Newton’s Principia. He frequently visited the city of Washington during the next few years. Here he went as far as the gate of the Naval Observatory, and looked wistfully in, but feared to enter. He heard for the first time of the Smithsonian Institution, and its library proved to be a rich attraction; in May, 1856, he got permission to climb into the gallery of this library and see the mathematical books. ‘‘Here I was delighted to find the great- est treasure that my imagination had ever pictured—a work that I had thought of almost as belonging to fairyland. And here it was right before my eyes—four enormous volumes—‘ Mécanique céleste, by the Marquis de Laplace, Peer of France; translated by Nathaniel Bow- ditch, LL.D., Member of the Royal Societies of London, Edinburgh, and Dublin’.’’ He secured special permission to take with him the first volume of this great work of which he had so long been dreaming, and he carried it in triumph to his little schoolhouse; he found most of it quite beyond his powers at the time, but this fact served only as an incentive to continued effort. In 1855, when 20 years old, Newcomb published his first paper: A correspondent of a newspaper had written a long letter to refute the Copernican theory; the arguments, though wholly fallacious, seemed to Newcomb to appear so plausible that he was much alarmed lest the world’s belief in the Copernican doctrine be severely weak- ened, and so he hastened to the rescue with an answer pointing out ApriL 15, 1936 WOOLARD: SIMON NEWCOMB 143 the fallacies. His name in capital letters printed at the bottom of his article filled him with a sense of temerity at having perhaps intruded where he might not be wanted; but it brought him the presentation of a book of Tables and formulae from Col. Abert, and a letter from J. Lawrence Smith (afterward a member of the National Academy) transmitting a copy of a pamphlet on a theory of the origin of meteor- ites, and asking Newcomb’s opinion on the subject. ‘“‘I had not yet gotten into the world of light. But I felt as one who, standing outside, could knock against the wall and hear an answering knock from within.”’ Newcomb soon became acquainted with Joseph Henry, Secretary of the Smithsonian Institution, and with J. E. Hilgard, assistant in charge of the Coast Survey; he had never before looked upon a real live person of eminence in the scientific world, and had often won- dered whether there were any possibility of making the acquaintance of so great a man as Joseph Henry. His reception by Henry and Hil- gard was most delightful; and it was through their interest that he secured an appointment, when not quite 22 years old, as computer in the office of the American Ephemeris and Nautical Almanac, then located at Cambridge, Mass. Newcomb writes in his Reminiscences: “T date my birth into the world of sweetness and light on one frosty morning in January, 1857, when I took my seat between two well- known mathematicians [Joseph Winlock and John D. Runkle] before a blazing fire in the office of the ‘Nautical Almanac’.’’ His im- pressions of Henry and Hilgard, and of Winlock and others in the Almanac Office, were fully up to the most sanguine of his boyhood conceptions of men of science. His duties required five hours a day, and his salary was $30 a month. He enrolled as a student of mathematics in the Lawrence Scientific School in Harvard College, studying under Benjamin Peirce, and received the B.S. degree the following year. Newcomb was not satisfied at the prospect of doing nothing more than make routine calculations with formulae prepared for him by others; indeed, not yet having mastered the Mécanique céleste, he was almost disappointed to find that he was considered qualified for a position in the Nautical Almanac Office, but he consoled himself with the reflection that the ease of the work would not prevent him from working his way up. From this beginning he rose rapidly to a position of commanding eminence in mathematical astronomy, per- haps the most difficult field of human knowledge; and except for the one year at Harvard, he was self-taught. His original contributions 144 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 26, No. 4 to celestial mechanics had already begun when his paper On a method in dynamics, dated April 2, 1858, appeared in Gould’s Astronomical Journal; and his prdduGtinity continued uninterruptedly from that time to his death, 51 years later. On October 7, 1861, Newcomb was appointed professor of mathe-. matics in the U. 8S. Navy, for service at the Naval Observatory in Washington. His tastes and talents were in mathematical, rather than observational, astronomy; and he had applied for this position (on the recommendation of B. A. Gould, the first leading astronomer whose acquaintance he had made) only because it was desirable to provide for the future. He was without any experience or knowledge of astronomical observing; but he applied himself with diligence, and the results of his work with several different instruments during the next 10 years give abundant evidence of his energy and ability as an observer, and his remarkable breadth of view and power in planning and executing systematic and effective observing programs and in reducing and discussing the observations. There is no doubt that his experience in observational astronomy was of great value in his later theoretical work; he was unequaled in the comparison of theory with observation, and in the deduction of valuable results from large masses of data. Newcomb’s observational work did not wholly cease until about 1875; but several years previous to that time, the discrepancies be- tween the observed positions of the moon and the positions computed from Hansen’s Lunar Tables had become a serious matter; and at Newcomb’s request, he was relieved from regular observing duties about the end of 1869, in order to conduct an investigation into the motion of the moon. The Lunar Theory developed into the leading interest of his life, and it received his best efforts during much of the time from 1870 to his death. In 1877, Newcomb was appointed Director of the Nautical Al- manac, and held that position until his retirement, devoting himself to mathematical astronomy which he had most at heart. It is im- possible here even to mention all of his leading contributions to celestial mechanics and his many invaluable services to astronomy, to say nothing of his comparatively minor contributions, and his interests and activities in other fields; the complete bibliography of his writings*® contains 541 titles, including 318 pertaining to astron- omy, 35 to mathematics, 42 to economics, and 146 on miscellaneous 3 R. C. ARcHIBALD. Simon Newcomb: cee ae of his life and Work. Mem. Nat. Acad. Sci., xvii, First Memoir, Pt. II. 1924 Aprit 15, 1936 WOOLARD: SIMON NEWCOMB 145 subjects that comprise even fiction! Attention must here be confined to his greatest project, the construction of new theories and tables for the motions of all the principal bodies of the solar system; this undertaking was one which he had long cherished, and upon assuming charge of the Nautical Almanac was enabled to put into execution. A reconstruction of the lunar and planetary theories and tables was then badly needed, because the precision of astronomical obser- vations had come to exceed the accuracy of the existing mathematical theories of the celestial motions. The computed motions did not repre- sent the actual motions exactly, and the deviation of prediction from observation was constantly increasing. It is true that the discordance between theory and observation was not large enough to be of prac- tical importance in the applications of astronomy; but if the celestial motions are produced solely by gravitational forces that act in con- formity with Newton’s Law, then a mathematical theory of the mo- tions developed from this hypothesis should represent the observa- tions exactly, over any period of time, and it is of the utmost im- portance to determine whether any observed discrepancies are due to some inadequacy of that hypothesis, or merely to imperfections of the mathematical theory. If the solar system were composed of only the sun and a single planet, then the orbit of that planet would be an exact ellipse that would never change its form, size, or position. Actually, however, each of the planets and satellites moves in an exceedingly irregular and ever-changing path, because of the disturbances produced by the attractions of all the other bodies in the system. None of the de- partures from regular elliptic motion is very great; but the longer and the more accurately the motions are observed, the more com- plicated they are found to be. Now, there long had been no question, of course, that these irregular motions of the planets around the sun, and those of the satellites around the planets, are at least very nearly in accordance with Newton’s law of gravitation; but the important question is whether the mutual gravitational attractions given by Newton’s Law completely explain all the manifold irregularities in the actual motions that can be detected by observation. To answer this question, long continued series of observations of the utmost attainable accuracy are first necessary, to determine with as much precision as possible how each body does move; and next, it is necessary to make a theoretical calculation of how each would move under only the Newtonian gravitational attractions of all the others, to the same degree of accuracy as the available observations. 146 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 26, NO. 4 Serious difficulties are encountered in both of these steps; and it is not easy to determine with certainty the significance of differences between theory and observation. In the first place, accurate observations are difficult to obtain with even the finest instruments—untiring industry and great skill are required. Some allowance must always be made for inevitable errors of observation; and to combine observations made at different times and places, by different observers with different instruments, into a homogeneous whole for comparison with theory, is a task that calls for rare ability. In the second place, the mathematical theories of the celestial motions are of a complexity almost beyond conception: It is well known that the general solution of the problem of even as many as three bodies moving under their mutual gravitational attractions cannot be obtained in any usable form. In the case of the planetary system, the very special circumstance that one of the bodies—the sun—dominates the entire system, permits an approximate solution of any required degree of accuracy to be obtained, valid over a more or less limited interval of time and in a form suitable for numerical calculation, but only at the cost of tremendous labor when the approx- imation is to be pushed to the accuracy of modern astronomical ob- servations. Single formulae that fill dozens of printed quarto pages, and require months or even years to derive, are not unusual; the final equations that represent the motion of the moon contain over 1400 terms, and to obtain them is almost the work of a lifetime. Only three times in the entire history of astronomy has anyone had the courage to undertake the systematic construction of complete theories for all the principal bodies of the solar system: The first attempt was by Laplace; his results are contained in the immortal Mécanique céleste, and upon them were founded the tables by Lin- denau and Bouvard which remained in use for over 50 years. Laplace, however, developed the equations to only a low order of approxima- tion; they were sufficient at that time, and demonstrated the agree- ment of the celestial motions with Newton’s Law to a first approxi- mation, but during the first half of the nineteenth century the de- velopment of precision astronomy under the leadership of Bessel, and the accumulation of further observations, led to a need for more accurate theories. The task was undertaken by Leverrier, one of those who had calculated the position of the planet Neptune prior to its telescopic discovery. During the 22 years from 1855 to his death in 1877, Leverrier succeeded in publishing new theories for all the Aprit 15, 1936 WOOLARD: SIMON NEWCOMB 147 major planets; his work marked an epoch in gravitational astronomy, and resulted in greatly improved planetary tables, but it still left much to be desired, and deviations of the planets from their tabular positions soon began to appear. Newcomb therefore set himself, at the age of 42, the task of again developing improved theories of the celestial motions, free from the defects of the previous investigations. The existing tables of the sun, moon, and planets, the catalogues of star positions, and the values of the fundamental astronomical constants that must be determined from observation, were then in a chaotic state; uniformity, and even consistency, were largely lacking. Newcomb planned the construction of theories and tables for all the major planets, the moon, and the other satellites, with an accuracy comparable to that of modern ob- servations, and based on uniform and consistent values of the funda- mental constants that themselves should be accurately determined from all existing observational data. It is practically impossible to effect any adequate conception of the appalling magnitude of this monumental project; and even an examination of the published re- sults gives no adequate impression of the intricacy and complexity of the work, or of the immense labor involved in the execution of the program of lunar and planetary investigations which Newcomb mapped out. The determination of the fundamental constants in- volved the discussion of all observations of worth ever made on the positions of the sun, moon, and planets at the 13 leading observatories of the world since 1750; the number of meridian observations of the sun, Mercury, Venus and Mars alone was 62,030, and merely this part of the work was probably of greater magnitude than any single astronomical investigation ever before undertaken by one individual, although the final results are embodied in a modest octavo volume of only 200 pages—one of the classics of astronomy.’ For twenty years, driven by untiring energy and singleness of pur- pose, and sustained by an unusual power of continuous work, New- comb devoted himself, solely in the interests of astronomical science, to this colossal task. Needless to say, it could not be performed with- out aid; not only is a corps of computers necessary, but also assistants with technical ability are needed. Newcomb himself emphasizes this; and he gives due credit, in many formal acknowledgments in his papers and on the title pages of his monographs, to his assistants, 4Srmon Newcoms. The elements of the four inner planets and the fundamental con- stants of astronomy. Supplement, American Ephemeris and Nautical Almanac for 1897. Washington, 1895. 148 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 26, No. 4 while in his Reminiscences he pays high and generous tribute to those who aided him: The most difficult part of the whole program, the theory and tables of the giant planets Jupiter and Saturn, was as- signed to George William Hill, whom Newcomb refers to as easily the greatest master of mathematical astronomy during the last quarter of the 19th century, and who labored incessantly for 10 years on the intricate calculations required by reason of the large masses and mutual proximity of these two great planets. Newcomb writes: ‘“‘And here was perhaps the greatest living master in the highest and most difficult field of astronomy, winning world-wide recognition for his country in the science, and receiving the salary of a department clerk. I never wrestled harder with a superior than I did with Hon. R. W. Thompson, Secretary of the Navy, about 1880, to induce him to raise Mr. Hill’s salary from $1200 to $1400.”’ Hill himself took little interest in this matter—‘‘He did not work for pay, but for the love of science. ... That I could not secure for him at least the highest official consideration is among the regretful memories of my official life.’ Of Cleveland Keith, Newcomb says: ‘“‘Without his help, I fear I should never have brought the tables to a conclusion.”’ The completion of Newcomb’s great program was imperiled by his automatic retirement on March 12, 1897, at the age of 62, accom- panied, unfortunately, by circumstances that were not to the credit of those responsible for them and which forced him to cease work on the uncompleted project. Eventually, however, the planetary tables were completed and published, and a star catalogue issued, by the Nautical Almanac Office, partly under Newcomb’s supervision. His investigations of the motion of the moon had to be left unfinished, but were later completed during the last six years of his life under the patronage of the Carnegie Institution; the manuscript was finished less than a month before he died. For several months before his death on’ July 11, 1909, at the age of 74, Newcomb was aware that his days were numbered, and he devoted all his energy, often while in great suffering, to the completion of the lunar work. No lunar tables were constructed, however; and little was ever accom- plished with’ the other satellites in the solar system. The huge task of discussing the comparison of his planetary tables with observation, and all the problems arising therefrom, were left untouched. An enor- mous mass of material that had been gathered for future work was left, and has remained unused. A great volume of essentially com- pleted work was never published—in particular, a complete exposition Apri 15, 1936 WOOLARD: SIMON NEWCOMB 149 of the investigations and theories on which the planetary tables were based has not appeared. Newcomb wrote in the preface to the tables: “The question to what extent these and other researches shall be completed and published is one the decision of which must be left to others.’’ After a lapse of nearly 40 years, it is only too painfully evi- dent what that decision was. During the 15 years from 1882 to 1897, eight and a half quarto volumes of the Astronomical Papers of the American Ephemeris, a series established by Newcomb expressly for the publication of re- sults connected with his program, were issued under his supervision; during the next 15 years, following his retirement, nothing further was issued, until in 1912 his last memoir on the motion of the moon was posthumously published; in the 23 years since then, two brief monographs and a star catalogue are all that have been added to this series. Except for three brief papers by various authors, the first eight and one-half volumes are made up of twenty-five great mono- graphs by Newcomb himself, three extensive memoirs by Hill, and the theory and tables of Jupiter and Saturn, also by Hill. It is evident that, despite the invaluable assistance rendered by others, the credit for the work that was accomplished belongs to Newcomb; it was due to his ability and vigor in planning the investigations, developing methods and procedures, supervising and checking, and pushing the program ahead, that it was carried as far as it was. The volumes of the Astronomical Papers that contain the results are among the most priceless treasures of astronomical literature; they are now in use in the preparation of the annual emphemerides issued by the principal countries of the world. In his prefaces to the volumes of tables, however, Newcomb writes regretfully: ‘‘On sur- veying the completed work the author is painfully conscious that in several points it fails to reach the standard he had set for it.’’ He enumerates the principal causes of the defects, and calls attention to the fact that a future generation must reconstruct the work. Newcomb had a facile pen; he wrote clearly, in excellent English; and not the least of his services to astronomy were his many superb popular books and articles. His commanding position in astronomical science was universally recognized, and he was the recipient of a long list of honors, including degrees from 17 universities, and numerous medals and prizes from learned bodies. He was somewhat reserved and unapproachable, and inclined to be gruff, which sometimes led to an erroneous impression of him on the part of those who had not 150 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 26, No. 4 come to know him; but his personal characteristics, as well as his scientific achievements, won many high tributes at the time of his death.° CHEMISTRY.—A pH conversion chart... W. H. Goss, Washington, D. C. (Communicated by F. G. BRICKWEDDE.) A conveniently arranged chart or graph is useful to biologists and chemists in saving much time when converting emf values to pH readings in measurements on culture media and other solutions with hydrogen, glass, or quinhydrone electrodes and 0.1N, N or 4.1N (saturated) KCl-HgCl reference standards. The stoichiometric and mathematical equations correlating pH and emf values are rather complicated with certain types of electrodes and with special re- actions involving, for example, oxidation-reduction. Several con- stants and variables may be involved in some cases, which must be placed in the equations correctly both as to sign and magnitude. The temperature-emf coefficient of each electrode must be applied. Fre- quently there is temporary instrumental or electrode hysteresis aris- ing, for example, from electric leaks through moisture on the surface of the potentiometer or leads, or from irregular changes in tempera- ture and activity of reference standards. Recognition and interpreta- tion of these errors require detailed theoretical background and ex- tensive experience beyond the scope of this paper. Such errors ob- viously cannot be corrected by a single graph or any other simple method. The mathematical expressions correlating emf and pH in- volve exponential functions which can be solved by use of logarithms or a slide rule. As many scientists in biology and chemistry and fac- tory workers find such calculations foreign to their daily experience, it is believed that a chart will be helpful in work with some of the more commonly used electrode systems when free from the above men- tioned errors. Fig. 1 gives the variation in emf and pH values with temperature for several of these electrode systems. When made about 18 inches wide and 14 inches high the pH values from 0 to 14 can be read to 0.01 with a maximum error of +0.02 between 15° and 30°C. An 8” X10” photograph can be read to 0.02 pH. This covers the ranges of temperature and the precision needed in most routine measure- > See the Memorial Addresses read before the Philosophical Society of Washington. Bull. Phil. Soe. Wash. 15: 133-167. 1910. 1 Received July 15, 1935. 151 HART GOSS: CONVERSION C 15, 1936 Apri 15, ° ¢ “we *o Nee “o e % ¢ *«, © ‘o Nie wy OR Ne —39 “e a) 2 We ean Oh Ae ey ae i == RAFFLE wil == jase 74 ala/[2 Op AODRRRAOURE aL wilt 2S ii UA uu mm ui == Rot tT] 1 H —— ‘ia ce Ea Ha == | WT git ==F Hina TT On fH nin Hi Hil == i cS Ea he a == 5 AR HR fa == 5 ES Ela HH == Ss inn Hh Mt iil Hy = = 1 TIT) 1S Ninny Hit 1 mn tu H sul = = ima Mt HI mi iit ly = lain iN MEH mul HT = = (Hin TH ill msi i 4 : iN ITLL mutt = = i Hl ull = = HIT iM mul ii = = HH or Ht mt = = TTT HOUUNANUUN quill qu! =f = nn TH ull = S HAITI il ti = 5 t CH wT Ht = A) EAC HT wal = = TT TH my Hu il wt = = nn in Ht fla rt mt = = nH nH HAH A rau ni = ee Ae Ct = | HUUTEAUENONAN TIE I mn Hi = 2 a A mt = il i B => a HHH Ee a ois et Mee nt | = 2 dé HA Re qi u _ =I: I HUN tll TTI ut Ht == S58 | Re ae —_—== ily Ht Tc A = ty Ha Cn HHA ra Hr ils == na Fa HR CTT el il == aa Han a i nn HAT HAL HT i == HOT nT i UHHAy eg UVTI mL ult => a A Ha aR == H a mn ma it =e i a tobe mn ill == ! OH Ae A junnnl ul 2S | THIN TS HL nn ul —— A A Ha ==. HH CA ULE == HHL i == on HR iu Hl ul == TT ER i 2 ne Ee CH a = | TTT mn nm = Mu Mi = i I te na il Ht = A Rey Hee He HH 4 i Mit a nH ee HH = Tee a GHEE aR HR = iT rT? J = aA nu AER mau i = Mi rae {at i fu itl =o TT IM HTT MALAI peal wut it = nu nM mH i mann wan = Mth INTHE HHH HM unl HIM = Mittin mui it mn (TT ill mn HT wt — ait HHH TNATNOHT HTH Hl =a) ce ae =) HH AH il i 5 HTT il AH TI nut Ee Ei HE HR HTT TTT TOT TT Hutt rrT Mumia i alts lelsls lal HEV raaneyg et EULA tal 1% | % Ne No “os “Sto “b Bbtovese / 3 A = athe Ny ® ye “Oo 2 = bale ls > NG @ ios = a 6s 7 Be ee Ry See My Se a Se se, oa ae 58 85 8 i S = Ss Sie S00 Seas ’ ° a LuO my) = See iage # oS Ose HA = = Bas es rae: dz = pedis Gene) oe ues = Os Jj = QZ x<@ ° jer a Q x5 = z|e A O x ° Eo oO x Zz z|° © i E ia a * < ae H 0) ee N ii ae . t. 152 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 26, NO. 4 ments. Although the saturated KCl-HgCl reference electrode is per- haps most widely employed, because it has practically no contact potential against most unknowns, the 0.1N and N KCl-HgCl stand- ards are frequently used with a saturated KCl solution or agar gel bridge between the calomel and hydrogen electrodes. The graph therefore includes all these systems with no corrections necessary for contact potentials. The entire chart is made up of ten small horizontal graphs with their designations at the left side. The top and third scales numbered 1 are merely duplicates of the fundamental emf scale. They permit the accurate placing of a straight edge vertically at the same voltage on both to get the reading on the pH scale (second) corresponding to this emf or to any other emf in the lower graphs. The lower scales, No. 2 to 8 inclusive, give the total emf readings of the indicated electrode pairs corresponding to pH values which can be read directly on the pH scale. The left-hand upper or reference electrode is positive (+) to the lower or hydrogen ion electrode and the emf readings in the graphs are therefore posztive in all cases ex- cept in graphs 5, 6 and 7 where the emf to the left of 0 (zero) are negative (—) because the quinhydrone electrode is the positive one. (See note at the bottom of the graph.) The chart is first placed on a small drawing board or table with a straight top edge, preferably under glass, cellophane or celluloid sheets when cleanliness is desired. By use of an accurate T-square with transparent straight edges both ends of No. 1 scale are brought into positions perpendicular to the top edge of the board and the chart is then fastened down with thumb tacks. When the T-square is moved along the top edge of the board to any intermediate position such as at 0.5 volt on the No. 1 scale the pH corresponding thereto and to any emf on the lower graphs at the desired temperature can be easily read. The second or pH scale gives the pH readings from 0 to 14 cor- responding to the emf between a hydrogen (or glass or quinhydrone) electrode in the unknown solution and a duplicate electrode in a standard solution having unit hydrogen ion concentration or activity when at the temperature of the unknown. The contact potentials are assumed to be entirely eliminated. This enlarged pH scale is there- fore especially useful in obtaining the pH of the unknown as follows: The standard emf (positive or negative) of the reference electrode toward any hydrogen ion electrode (Pt, glass, quinhydrone, antimony, etc.) in a N H* solution is algebraically subtracted from the total Aprit 15, 1936 GOSS: CONVERSION CHART 153 emf (positive or negative) given by the unknown solution when measured with the same reference and hydrogen ion electrodes; the difference is an emf found on the No. 1 scale or its extension (to e.g. pH of —1 or 15). The vertical straight edge is run through this point and cuts the horizontal line in the pH scale corresponding to the temperature of the unknown solution at a point which is the pH of the unknown. For example if a saturated calomel reference electrode is +0.245 volt toward a Pt-H,. electrode in N Ht solution and +0.745 volt toward the Pt-H, electrode in an unknown solution, the differ- ence is 0.500 volt. If the saturated calomel electrode is —0.453 volt toward a quinhydrone electrode in N H+ solution and +0.047 toward the quinhydrone electrode in an unknown solution, the difference is +0.047 —( —0.453) = +0.500 volt. Such a difference of +0.500 volt on the No. 1 scale corresponds to pH 8.66 if the temperature of the unknown is 18°C. and 8.32 if it is 30°C. The slopes of the emf lines in graphs Nos. 2, 3, 4, correspond to the temperature coefficients of the 4.1N (saturated), 1N, and 0.1N KCl- HgCl electrodes, —0.00076, —0.00025, and —0.00008 volts per de- gree, respectively, referred to the hydrogen electrode having zero temperature coefficient. In graph No. 8 the slope of the emf lines represents the temperature coefficient of the quinhydrone electrode, —0.00074 volts per degree, and in graphs Nos. 5, 6, 7 the slopes represent the difference between the temperature coefficient of the quinhydrone electrode and that of the particular KCl-HgCl electrode used. Since the temperature coefficients of the saturated KCl-HgCl and the quinhydrone electrodes are almost equal, —0.00076 and —0.00074 respectively, a cell composed of these has a very small temperature coefficient as shown by the nearly vertical emf lines in graph No. 5. Graphs Nos. 2 to 8 are used as follows: The reference and hydrogen ion electrodes are assumed to be at the same temperature. Suppose that a total emf of 0.300 volt is measured at 18°C. in the electrode system used in graph No. 2. The straight edge is run from the point of intersection of the 18°C. and 0.3 volt coordinates of graph No. 2 vertically through the pH scale, note being taken that the straight edge gives the same reading on the two No. 1 graphs. The straight edge intersects the 18° line of the pH scale at pH 0.84; if the tem- peratures of both electrodes are at 30° the pH reading for 0.3 volt is 0.96. Likewise a total emf of 0.6 volt at 18° and 30°C. corresponds to pH 6.04 and 5.95 respectively; and 1.0 volt at 18° and 30°C. gives pH values 12.97 and 12.61. 154. JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 26, NO. 4 It will be noted that in some of the different electrode systems Nos. 2 to 8 the pH increase with rise in temperature from 15° to 30°C. may be positive at one end of the pH scale, zero near the center, and negative at the other end. In other words we should be able to calculate for each electrode system the pH value at which the dH /dT slope for the total emf of two electrodes with different temperature coefficients is equal to the slope dH/dt = —0.000198 pH for the ab- solute pH scale. This pH reading remains the same whether both electrodes are at 15° or 30°C. This condition is satisfied when the single electrode potentials L, and EH, for graphs Nos. 2 to 8 give dE ,/dT —dE,/dT =dExz/dT = —0.000198 pH or | pH =(dE,/dT —dE,/dT)/(—0.000198) or dpH/dT =0.0 By use of the temperature coefficients chosen above for various elec- trodes and the above equation it is calculated that for graphs Nos. 2 to 8 the pH values are 3.84, 1.26, 0.40, 0.1, —2.98, —3.33, and 3.74 respectively, at which the pH readings cannot change with equal rise in temperature of both electrodes. To the left of these pH values in- crease in temperature causes rise in the pH reading for the same emf and the reverse to the right. These pH points for graphs 6 and 7, namely pH —2.98 and —3.383 respectively, correspond to roughly 1000 N hydrogen ion concentrations which are beyond practical realization. For graphs 6 and 7, then, the pH always decreases for the same emf with rise in temperature. It should be noted that in graphs Nos. 2 to 8 the slopes correspond to the difference between the temperature coefficients of the reference and hydrogen ion electrodes used. These are generally kept at the same, preferably constant, temperature such as 25°C., but even if they are at different temperatures the correct pH value can be read from graphs Nos. 2, 3, 4 and 8 by placing the vertical straight edge at the intersection of the observed emf and the horizontal ordinate for the temperature of the reference electrode used and reading off the pH scale at the intersection of the vertical straight edge with the horizontal ordinate for the temperature of the hydrogen ion electrode. When the two electrodes are at different temperatures and neither has a zero temperature coefficient, as is the case when quinhydrone is used with a calomel electrode in graphs Nos. 5, 6 and 7, a correction for any difference in their temperatures must be applied to the emf reading before using the chart. The correction to be added to the emf reading is the algebraic change in the normal emf of the reference APRIL 15, 1936 GOSS: CONVERSION CHART ip calomel electrode when brought to the temperature of the quin- hydrone electrode, and is easily applied as follows: The magnitude of the correction is the temperature coefficient of the calomel elec- trode multipled by the difference between its temperature and that of the quinhydrone electrode. If the calomel electrode is at a higher temperature, this correction should be subtracted from the emf read- ing measured, before consulting the graph, because the calomel elec- trode temperature coefficients have negative values. If the calomel electrode temperature is lower than that of the quinhydrone elec- trode, the correction should be added to the reading. As an example assume that a quinhydrone electrode is at 28° and a 0.1N calomel electrode is at 18°, and that the emf reading of this cell is —0.136 volts. The temperature coefficient of the 0.1N calomel electrode is —0.00008 volts/degree. Then (—0.00008)(10) = —0.0008 volts or practically —0.001 volt. Then —0.136+(—0.001) = —0.187 volts for use on the chart. The corresponding pH reading on the 28° line of the pH scale and scale No. 7 is 3.74. The glass electrode is used with so many combinations of internal and external calomel, AgCl-HCl and quinhydrone reference elec- trodes that the only general rule is to make all corrections for the reference standards (e.g. 0.1N HCl-HgCl, 0.1N HCl-AgCl, or 0.1N HCl-quinhydrone or Thompson’s? metal coating) necessary to get the emf between the unknown solution on one side of the glass mem- brane and the pH standard (H+ =1 or 0.1N HCl with pH 1.07) on the other side and use this emf with graph No. 1 and the pH scale to get the difference in pH at the temperature of the glass electrode. This general procedure applies to the use of the chart with the antimony or any other hydrogen ion electrode or buffer standard. The emf of the reference electrodes toward the normal hydrogen ion electrode and the pH values used in making the chart are those common today: namely, +0.2496 at 20°C. and +0.2420 at 30°C. for the saturated KCl-HgCl-Hg electrode; +0.2860 at 20°C. and +0.2835 at 30°C. for the N KCl-HgCl-Hg electrode; +0.3379 at 20°C. and +0.3371 at 30°C. for the 0.1N KCl-HgCl-Hg electrode; and +0.7029 at 20°C. and +0.6955 at 30°C. for the N H+-quinhy- drone electrode. Graph No. 9 is a log scale added to make easy the calculation of hydrogen ion concentration from the pH reading. It gives only the mantissa (fraction) of the log as the characteristic (integer) is used by inspection as follows. A pH value such as 7.25 is the log (on base 2 THompson, M. R. Bur. Standards J. Research 9: 852. 1932. 156 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 26, NO. 4 10) of the dilution of the hydrogen ions, i.e., of the number of liters containing one gram equivalent H+. As concentration (Cy) is the inverse of dilution (Vz in liters), for pH 7.25 Cy =1/V_=1/10'* = 10°-°/108 = 10°.” x 10—-° = 5.63 X10-®. A graph such as No. 9 would ordinarily read from left to right in making the above calculation, familiar to those regularly employing logarithms, but by construct- ing the linear scale to read from right to left, the above calculation can be eliminated and the Cy, corresponding to any pH value, or vice versa, can be read directly as follows: For pH 7.25 read off the fractional pH value 0.25 from right to left to get the corresponding number, 0.563 in this case. The H-ion concentration is this number multiplied by 10 to a minus power equal to the integer or whole number of the pH value, e.g., Cy =0.563 X10—’ or 5.63 X10-°. A pH value in integers alone (e.g., pH 7.0) gives directly a Cy value equal to 10 to the negative integer power (e.g., 10~"). A 14X18 inch enlargement of the pH conversion chart described can be used with a maximum error of +0.02 pH unit. It should, therefore, be a valuable aid to biologists, chemists, and students en- gaged in pH measurements, both as a tool for conversion of routine emf measurements into pH values and as a guide toward a clear con- ception of the inter-relationships between the electrode systems com- monly used in electrometric pH measurements. ACKNOWLEDGMENT The writer is indebted to Dr. S. F. Acree of the National Bureau of Standards for suggesting the preparation of such a general purpose pH chart, and to Mr. G. H. Lovins for aid in its design and con- struction. PALEONTOLOGY.—WNomenclatorial notes on fossil and recent Bry- ozoa.t R.S. Bassuer, U. 8. National Museum. Taxonomic studies of the Bryozoa have interested the writer to such an extent that he recently published a bibliographic index of fossil and recent genera, giving the classification, genotypes, prin- cipal citations and other information useful to the specialist on this group.” Further researches have shown the necessity for additional changes in nomenclature which are recorded in the present paper. 1 Published with the permission of the Secretary of the Smithsonian Institution. Received January 17, 1936. 2 Part 67 of the Fossilium Catalogus, Bryozoa Generum et Genotyporum Index et Bibliographia. pp. 1-229, 1935. APRIL 15, 1936 BASSLER: BRYOZOA 157 Order TREPOSTOMATA Ulrich Family BATOSTOMELLIDAE Ulrich, 1890 Stenoporella, n. gen. Like Stenopora Lonsdale, 1844, save that instead of diaphragms numerous spines project from the walls into the zooecial cavity, and the beaded struc- ture of the walls is nearly obsolete. Stenophragma Munro, 1912, has semi- diaphragms projecting from one side of the wall only and has regularly beaded wall structure. Genotype.—S. romingerz n. sp. Mississippian of Arkansas. Stenoporella romingeri, n. sp. Figs. 1-3 Zoarium an explanate mass about 100 mm in diameter and 25 mm at the thickest portion, consisting of several superposed layers of zooecia, with surface smooth and clusters inconspicuous, the zooecia being nearly uniform in size. Zooecia angular, 8 in 2 mm, thick walled, with mesopores practically wanting. Large acanthopores of the type found in Stenopora oc- cupy only the junction angles and average 2 to each zooecium. Diaphragms wanting, their place being taken apparently by semidiaphragms, which in this case appear as spines projecting into the zooecial cavity. These spines resemble similar projections in such genera as Chaetetes and Favosites, but the presence of Stenopora-like acanthopores and wall structure seems to justify the reference of the genus to the Bryozoa. Occurrence.—Found by Dr. Carl Rominger at Cave Creek, Arkansas, in strata said to be of Chester age. Holotype.—No. 53833, U. 8. National Museum. Order CYCLOSTOMATA Busk Family CERAMOPORIDAE Ulrich, 1882 Haplotrypa, n. gen. This new genus is proposed for various parasitic or discoidal species which have the ceramoporoid wall structure, namely of irregularly laminated tissue, but entirely lack the lunarium characteristic of most other genera of the Ceramoporidae. The apertures are direct, and externally bear a re- semblance to the trepostomatous genus Monotrypa Nicholson, 1879. Spati- opora Ulrich, 1882, consisting of thin parasitic expansions with oblique cells and blunt acanthopores, is a related genus. Genotype.—Haplotrypa typica, n. sp. Range, Ordovician to Devonian. Haplotrypa typica, n. sp. Figs. 4, 5 Zoarium a lamellate expansion of superposed layers several centimeters wide and 4 or more mm in thickness. Surface smooth with inconspicuous maculae of larger zooecia, of which there are 3 in 2 mm while 4 of the ordinary ones occur in the same space. Zooecia angular, thin walled, some- times in contact but often separated by narrow interspaces. In thin sections the laminated, ceramoporoid structure with apparent perforations in the walls is quite evident. Diaphragms are practically absent in both sets of tubes. Occurrence.—Niagaran group (Osgood), Osgood, Indiana. Holotype.—No. 92132, U. S. National Museum. 158 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 26, No. 4 Y, (0 gC. TN TL tT ICN = Se Wl EY ; NNT My For explanation of Figs. 1-12, see bottom of opposite page. Apri 15, 1936 BASSLER: BRYOZOA 159 Order CRYPTOSTOMATA Vine Family RHABDOMESONTIDAE Vine, 1883 Rhomboporelila, n. gen. Solid, ramose Rhabdomesontidae with the zooecial tubes in the axial region regularly rhombic or quadrate in cross section. Superior hemiseptum and two sets of acanthopores, a large set at the end of the zooecia and a small one entirely surrounding them, well developed. Genotype.—Rhomboporella typica, n. spo. Carboniferous of Bolivia. The discovery of this type of structure in the Cryptostomata illustrates an interesting case of parallel development in two different orders of the Bryozoa, as a similar occurrence of rhombic or quadrate zooecia in trans- verse section is found in the Silurian genvs Rhombotrypa Ulrich and Bassler 1904, and the Middle Carboniferous Rhcmbotrypella Nikiforova, 1933, be- longing to different families of the order Trepostomata. End views of the axial part of the branches in these three genera are so similar as to lead to confusion, but the cryptostomatous characters of the present form are so evident that there can be no thought of close alliance. In each case the similarity is caused by the fact that the zooecia of the immature zone simultaneously at definite intervals develop new tubes and undergo a change in orientation of their sides. This process has been explained in detail by the authors of Rhombotrypa,’ and it will suffice to say here that the rhombic or quadrate cross section is maintained by this concurrent development of new tubes. Rhomboporella typica, n. sp. Figs. 9-12 Zoarium, a cylindrical, solid, smooth branch, 3 mm in diameter, dividing at intervals. Surface without distinct maculae but with areas of slightly larger zooecia having thicker walls. Zooecia elongate, polygonal, with walls thin to thick according to age, bearing two sets of acanthopores, one of distinetly larger size at the ends of the orifices, and the other of smaller granules ornamenting each wall. Zooecia in irregular quincunx, 6 to 7 in 2 mm measuring along the longer diameter. Tangential sections reveal the thickened walls, the large and small sets of acanthopores and the few intervening mesopore-like areas. The vertical 3 Smithsonian Miscellaneous Collections 47: 44. 1904. Figs. 1-3.—Stenoporella romingeri n. sp. 1, A tangential section through the mature region, X18, showing the large acanthopores and the semidiaphragms in the form of blunt spines. 2, portion of the same, X21. 3, vertical section, 18, exhibit- ing the aspect of the semidiaphragms in this direction. Chester at Cave Creek, Arkansas. Figs. 4, 5.—Haplotrypa typica n. sp. Tangential and vertical thin sections, X18, illustrating the ceramoporoid wall structure and the absence of lunaria. Niagaran group (Osgood), Osgood, Indiana. Figs. 6-8.—Cliotrypa ramosa Ulrich and Bassler, n. sp. 6, Tangential section, X18, near the surface showing the normal zooecia and one of the ovicell-like forms, as well as the granose interspaces. 7, vertical section, X18, of half of a branch with the ovicell-like expansions and the hemiphragms developed. 8, transverse thin section, X18. The thin walled immature region and the mature zone with hemiphragms and expanded zooecia, are evident. Mississippian (New Providence shale), King’s Moun- tain, Kentucky. Figs. 9-12.—Rhomboporella typica n. sp. 9, zoarium, natural size. 10, tangential thin section, X18, through the mature zone. 11, vertical section illustrating the origin of new tubes at regular intervals, and the superior hemiseptum. 12, transverse thin seetion, X18, showing the rhombic form of the immature zooecia. Carboniferous of Chulpapampa, Bolivia. 160 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 26, No. 4 section is with one exception that of a typical member of the Rhabdome- sontidae with the superior hemiseptum well developed. The exception, as explained in the generic remarks, is the development of a rather wide im- mature or axial zone in which the zooecia originate new tubes simultane- ously at regular intervals. In vertical fractures of a branch this is shown by alternating smooth and uneven spaces which when cut by the thin section give the aspect exhibited in Fig. 11. Transverse sections are unusually in- teresting as it is here that the quadrate or rhombic form of the immature zooecia is best exhibited. The end of a branch moistened and viewed under a hand lens shows this character equally well, and gives a ready clue to the species. The large acanthopores originate in the axial region for they are distinctly visible in transverse thin sections. Occurrence.—Carboniferous of Chulpapampa, Bolivia. Holotype.— No. 68813, U.S. National Museum. Family FistuLiroripa& Ulrich, 1882 Cliotrypa Ulrich and Bassler, n. gen. Fistuliporidae like Fistulocladia Bassler, 1927, that is, narrow, solid, cylindrical, smooth branches with ovicell-like inflations in the tubes which develop in addition well defined semidiaphragms projecting into the zooecial cavity in the mature region. This genus originally distinguished and named by Dr. Ulrich and the writer in 1897 when specimens were distributed to various students, was defined by the junior author in the Paleontology of Timor 16: 49, 1929, but remained invalid because the genotype had not been figured. This is herewith corrected with the following description and figures. Genotype.—Cliotrypa ramosa n. sp. Range, Mississippian and Permian. Cliotrypa ramosa Ulrich and Bassler, n. sp. Figs. 6-8 Zoarium of small, solid, smooth, branching cylindrical stems, 2.5-4 mm in diameter, bearing oval zooecial apertures with strongly marked lunaria, separated by solid, granose interspaces and exhibiting large solid maculae at intervals of about 4mm. Measuring lengthwise, 4 zooecia occur in 2 mm. In vertical thin sections, the zooecial tubes are thin walled in the solid axial region becoming thick walled in the mature zone and developing at intervals rather thick incomplete plates from alternate sides of the wall as hemi- phragms in place of the ordinary diaphragms, and occasionally expanding into spherical, ovicell-like structures which then contract to normal size, or may appear as swollen prominences at the surface. The subsolid interspaces and the maculae are separated by vesicles and towards the surface are traversed by numerous small tubuli. Occurrence.—Mississippian (New Providence shale), King’s Mountain, Kentucky. Holotype.—No. 92133, U. S. National Museum. Order CHEILOSTOMATA Busk The following changes in family, generic, and specific names are suggested in this order. Family URCEOLIPORIDAE, new name Proposed in place of Euthyridae Levinsen, 1909, invalid name, since Huthyris Hincks, 1882, is preoccupied by the fossil brachiopod genus Huthyris Quenstedt, 1869. ApRIL 15, 1936 BASSLER: BRYOZOA 161 Genus Euthyrisella, n. gen. Named in place of Euthyris Hincks, 1882, preoccupied by FHuthyris Quenstedt, 1869. Genotype.—Euthyris obtecta Hincks, 1882. Recent of North Australia. Family CHEILOPORINIDAE, new name The genus Hippopodina Levinsen, 1909, was described as possessing an endotoichal ovicell and the family Hippopodinidae was founded by him in 1909, based upon this character. However, the ovicell in the genotype, Hippopodina feegensis Busk, 1884, is hyperstomial, and the genera with endozooecial ovicell must be classified otherwise. The new family Cheilo- porinidae is, therefore, proposed, based upon Cheiloporina Canu and Bassler, 1923, a genus with numerous species, ranging from the Eocene to the Recent. The family Hippopodinidae may be retained for the single type genus or future researches may show it to be related to the Schizo- porellidae. Adeona joloensis, new name Proposed for Adeona porosa Canu and Bassler, 1929, from Jolo, Philip- pines, preoccupied by Adeona porosa Canu and Bassler, 1923, from the Miocene of Santo Domingo. Escharoides erectoides, new name Proposed for Peristomella erecta Canu and Bassler, 1920, from the Tertiary of South Australia, preoccupied by Perzstomella erecta Canu and Bassler, 1920, from the Vicksburgian of Alabama, both species now being referred to Escharovdes. Callopora horniana, new name Proposed to replace Callopora crassospina Canu and Bassler, 1923, from the Pleistocene of California, preoccupied by Callopora crassospina Canu and Bassler, 1920, from the Eocene of North Carolina. Cellaria elongatoides, new name Proposed in place of Cellaria elongata Canu and Bassler, 1928, from Morocco, preoccupied by Cellarza elongata Canu, 1908, from the Patagonian of Argentina. Floridina voigti, new name Name proposed for Floridina bifoliata Voigt, 1930, from the Danian drift of Anhalt, Germany, preoccupied by Floridina bifoliata Canu and Bassler, 1920, a Tertiary species from Mississippi. Dacryonella minuta, new name Proposed for Dacryonella minor Canu and Bassler, 1920, from the Jack- sonian of Florida, preoccupied by Dacryonella (Membranipora) minor Hincks, 1885, a recent species. Gemelliporina, new genus Gemellipora Smitt, 1872, by the rules of nomenclature, is a synonym of Pasythea Lamouroux, 1871, so that the second group of species typified by Gemellipora glabra Smitt, 1872, retained under this name by Canu and Bassler, must be classified elsewhere. The new name Gemelliporina is, there- 162 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 26, No. 4 fore, here proposed for species with keyhole-like aperture, hyperstomial ovicell and tremocyst frontal, with Gemellipora glabra Smitt, 1872, a common species of the Gulf of Mexico, as the genotype. Figularia duvergieri, new name Proposed for Figularia carinata Duvergier, 1924, from the Helvetian of Salles, France, preoccupied by Figularza (Figulina) carinata Waters, 1923, a recent species from the East Indies. To the Fossilium Catalogus the following generic citations should be added: Spirillopora Giirich, 1896. Lower Devonian Girich, Verh. d. Russ.-Kais. Mineral, Gesell. zu St. Petersburg, (2) XOX pe 213 3 96, Genotype.—S. anguillula Girich, 1896, idem, p. 2138, pl. X, fig. 17. Poland. Unrecognizable. Figure shows only a twisted stem with cells in spiral rows. Vetofistula Etheridge, Jr., 1917. Devonian Etheridge, Jr., Geol. Surv. Queensland, Pub. No. 26, p. 17, 1917. Genotype.—V. mirabilis Etheridge, Jr., 1917. Not recognizable. Zeapora Penecke, 1893. Devonian Penecke, Jahrb. d. kk. geol. Reichsanst. XLIII, p. 610, 1893. Genotype.—Z. gracilis Penecke, 1893. Alps. Unrecognizable. A trepostome but illustrated by only a poor section. PALEONTOLOGY.—A new Allagecrinus from Oklahoma. Epwin Kirk, U. 8. Geological Survey. The U. 8. National Museum has recently acquired a number of Pennsylvanian crinoids from Mr. H. L. Strimple of Bartlesville, Oklahoma. Some of them are of considerable biologic and strati- graphic interest, and Mr. Strimple deserves much credit for discover- ing and calling attention to this material from a hitherto barren field. The most interesting crinoids collected are a suite of Allagecrinus preserving the arms. Allagecrinus and allied genera have been known for a long time from many parts of the world, and several hundred specimens have been collected. Up to the present, however, none has been found with the arms attached. The species itself proves to be new and is here described as Allagecrinus strimpler. Altogether 10 dorsal cups and 18 complete crowns, as well as several sets of dis- sociated arms, have been available for study. 1 Published by permission of the Director, U. S. Geological Survey. Received January 29, 1936. Aprit 15, 1936 KIRK: ALLAGECRINUS 163 Allagecrinus strimplei, n. sp. The species is comparable in size with the larger known species of Al- lagecrinus from the Pennsylvanian and Permian. The largest dorsal cup has a diameter of approximately 5 mm. The smallest individual is about one- half as large. The largest complete crown has a height of 12 mm. The other Fig. 1.—Allagecrinus strimplei, n. sp. Young individual in three-arm stage. Fig. 2—Four-arm stage. Fig. 3—Three-arm stage with incipient fourth arm appearing. Fig. 4.—Four-arm stage, primary arm in center and secondary arm to left. Figs. 5, 7.— Basal views of two dorsal cups showing variation in lobation. Fig. 6.—Proximal portion of column lying on a set of arms. Fig. 8.—Five-arm stage; primary arm in center; secondary arms to right and left. Fig. 9—Largest complete crown; primary arms in center and to right; secondary arm toleft. All figures are X3. They have been drawn in pen and ink from photographs and are somewhat diagrammatic. specimens are intermediate in size and yield a fair series of growth stages that are especially interesting as regards the development of the arms. The dorsal cup is depressed bowl shaped, with a diameter approximately twice the height. The radials are tumid, giving the dorsal cup a distinctly pentalobate to substellate outline as seen from above or below. The basals form a pentagon approximately two-fifths the diameter of the cup. What 164 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 26, NO. 4 appear to be sutures within the basal circlet may be seen at times, but the almost complete coverage of the basals by the proximal columnal makes it impossible to identify and orient the basal elements with certainty. If sutures are present, as seems probable, one would expect three unequal ele- ments. The radials are higher than wide. In very young specimens the radials are moderately convex. With increasing age the radials become more and more tumid. In the largest specimens each radial bears a large protuberance. In most cases the upper surface of this knob is flattened to concave as seen in profile. This is well shown in the drawings. The tumidity of the radials varies in amount and character as between individuals of nearly the same stage of development, and even between different radials of the same speci- men. Based on the relative convexity of the radials the series of specimens in hand could readily be separated into two species, the younger specimens in one species and the older in another. The right and left posterior radials are but slightly notched by the first anal plate, so little in fact that it is diffi- cult to identify the posterior interradius except in very well-preserved and carefully cleaned specimens. The anal plate notches more deeply into the right posterior radial than the left. The articulating face of each radial is produced inward as a shelf. On this shelf the position of each arm is marked by a depression bounded on the sides by converging ridges. The shelf sup- porting the arms is continuous except in the posterior interradius. In the posterior interradius the shelf is interrupted by a parabolic notch which in well-preserved specimens is clearly shown. : In the youngest specimens which have the arms preserved, three large arms are present. Later, additional arms appear. In the oldest specimens preserving the arms, five arms of approximately equal size are found. In the largest dorsal cup the converging ridges on the superior faces of the radials and the articulating facets indicate at least eight arms, with a pos- sibility that a ninth arm is just beginning to appear. If so, this ninth arm would give three arms to one ray, which is the left posterior. The radials bearing two arms each are the left posterior, anterior, and right anterior. Owing to the fact that we do not have a large number of dorsal cups, and that where the arms are preserved they are often partially detached from the radials and shifted from their original positions, it is not possible to determine accurately the orientation of the three original arms. The arms are variable in structure. The three primary arms seem to agree in being made up of a short first brachial followed by a very long second brachial. Additional brachials are added subsequently. The secondary arms have the short first brachial, but the second brachials are not so dispropor- tionately long, and several shorter brachials make up the length attained by the two brachials of the three primary arms. An average of many meas- urements of the first brachials gives a length of 0.9 mm. In the primary arms the second brachials have an average length of 6.7 mm. In one ray where several brachials are preserved, presumably a secondary arm, the brachials give the following measurements: 1, 0.9 mm; 2, 1.7 mm; 3, 2.1 mm; 4, 2.6 mm. Another arm of the same type gives the following measure- ments for the brachials: 1, 0.9 mm; 2, 1.3 mm; 3, 1.6 mm; 4, 1.8 mm; 5, 2 mm. In the secondary arms the greatest variation in length seems to be found in the third brachial, measurements varying from 1.3 mm to 3.4 mm. The mature brachials have an average maximum breadth of 1.4 mm. The backs are strongly convex, and the surface is covered with fine granulations. The secondary arms as they first appear have flat or slightly convex backs, which gradually increase in convexity with age. The union of the first APRIL 15, 1936 COTTAM: FOOD HABITS 165 brachial with the radial is weak, as is also the union between the first and second brachials. As a result the arms are seldom in true alignment and in their original positions. Fortunately fragments of column have been found in such relationships with the Allagecrinus crowns as to leave no doubt as to their belonging to- gether. The column figured lies on a set of arms and is incomplete in its proximal portion. The proximal columnals are thin and become narrower distad. This tapering proximal portion of the column is similar to that commonly found in the Flexibilia. Below the tapering columnals the char- acter of the column changes completely. The columnals are beadlike, and the nodals are relatively large. The general aspect of the column is very like that of one of the young Flexibilia from the Devonian or Mississippian. There is no described species of Allagecrinus with which A. strimplei may be confused. Horizon and locality —The specimens were collected by Mr. H. L. Strimple in the Dewey limestone (Pennsylvanian) near Dewey, Oklahoma. Types.—The cotypes are in the Springer collection in the U. S. National Museum, No. 8. 4126. BIOLOGY.—Food of Arctic birds and mammals collected by the Bart- lett Expeditions of 1931, 1982, and 1933.1 CLARENCE CoTTaM, U. 8S. Biological Survey. (Communicated by Watpo L. SCHMITT. ) On Captain R. A. Bartlett’s three arctic expeditions in the summers of 1931, 1932, and 1933, alimentary material, mostly gullet and stom- ach or gizzard contents of 115 birds representing 21 species and one additional subspecies, was collected. This material, subsequently sub- mitted to the Biological Survey for analysis, forms the basis for this paper. Fifty-three birds were secured on the first expedition during July and August from northwestern Greenland in the vicinity of Clavering Island, northward to slightly beyond the 74th parallel north latitude and between the 13th and 29th meridians west longi- tude. The second expedition returned with 20 birds from western and northern Greenland, northward to latitude 76° 33’. These birds, like- wise, were taken during the months of July and August. Localities mentioned were Parker Snow Bay, Dalrymple Island, Cape York, Walstenholme, and North Star Bay in latitude 76° 33’. During the 1933 expedition, which extended from July to September, one weasel and five ground squirrels were collected in addition to 42 birds. Collections were made in the area between northern Hudson Bay and western Greenland. Localities recorded included Melville Peninsula; Duckett’s Cove, Hudson Strait; Igloolik Island near Fury and Hecla Strait; Cape Frigid, Southhampton Island; and the open seas at latitude 61° north and longitude 64° 20’ west. 1 Received February 11, 1936. 166 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 26, NO. 4 Even though the number of individuals and species in the collec- tions is not so large as one might wish, they are of more than ordinary importance as they represent a far northern area in which but little critical study of the food habits of birds has previously been made. The majority of these species were poorly represented, or even en- tirely wanting in the ‘“‘stomach files’’ of the Biological Survey. While the writer examined the majority of the stomachs submitted, generous assistance in this regard was received from A. L. Nelson, Cecil 8S. Wiliams, Leon H. Kelso, and F. H. May, all of the staff of the Biological Survey food-habits research laboratories. Specialists who aided by making identifications of questionable material include: Dr. Waldo Schmitt and Clarence Shoemaker, of the U. 8. National Museum, on crustaceans; F. M. Uhler, of the Biological Survey, fishes; Wm. B. Marshall, of the U. 8. National Museum, mollusks; J. R. Malloch, of the Biological Survey, two-winged flies (Diptera) ; and Dr. R. V. Chamberlin, of the University of Utah, spiders. A summary of the food percentages by volume as determined from the analysis of stomachs, gullets, or crops of the birds collected on these expeditions follows: i Gavia stellata (Pontoppidan). Red-throated Loon. Five adult and two juvenile stomachs of the red-throated loon provide convincing evidence that the birds’ piscivorous tendencies dominate. Two adult stomachs were taken from Clavering Island, northeast Greenland, two from Igloolik Island, northwest Greenland, and one from Cape Frigid, Southhampton Island, Northwest Territory. Four of these adult birds had made their entire meals on small coarse fish, largely Gadidae (? Boreogadus saida), sculpins (Cottidae), and tomcod (Microgadus tomcod), while the other had made 80 percent of its meal on sculpins (Cottidae), tomcod (Microgadus tomcod) and sand launce (Ammodytes sp.). The remaining 20 percent of the meal of this bird consisted of 54 polychaete (marine) worms, apparently a species of Nereidae. The food of these five adult birds would be summarized as follows: Cottidae, 34.2 percent; tomcod (Microgadus tomcod), 4.8 percent; Ammodytes sp., 2 per cent; codfish (Gadidae, part probably Boreogadus saida), 30 per- cent; miscellaneous fish, 25 percent; polychaetes, apparently Nereidae, 4 percent; copepoda and undetermined crustacea (which may have come from the stomach of a party-digested fish), trace. From the size of the stomachs of the two juvenile birds, taken at Igloolik Island, it would appear that the birds were probably three-fourths grown. It is evident that the food of adults and juveniles varies considerably. Perhaps because the young are less agile and therefore less expert as fishermen, they consume fewer fish. In the two birds examined a greater variety of food was noted than in the adults. It was surprising to find that a moss of the family Hypnaceae made up 43 percent and 33 percent respectively of the two meals. It was noted that all the moss was in the gizzard and from appearance it had APRIL 15, 1936 COTTAM: FOOD HABITS 167 been there for some time. Perhaps in this group of birds the effects of digestive action are less rapid on plant than on animal substances. Feathers are often found in quantity in the stomachs of loons and grebes, yet their function is not definitely known. Some have believed that they serve to protect the stomach walls against the sharp fish bones. It is not known whether the moss in these juvenile birds’ stomachs was serving as a source of food, or, similarly, as a protection to the tender, growing gastric organ against the sharp fish-bone fragments. The stomachs and gullets of both juveniles were well-filled and may be thus summarized: Tomecod (Microgadus tomcod), 59.5 percent; sculpin (Cottidae), 5 percent; polychaetes, apparently Nereidae, 2 percent; squid (Loligo sp.), 0.5 percent; moss (Hypnaceae) plant fiber, 38 percent; gastro- pods, trace; crustacea, including amphipoda and schizopoda (which may have been taken by the fish), trace. Fulmarus glacialis glacialis (Linnaeus). Atlantic Fulmar. Because the Biological Survey has no other record of a stomach examina- tion of the Atlantic fulmar it is to be regretted that only one well-filled stomach of this bird was submitted for examination. It was collected at 61° north latitude and 32°10’ west longitude. This stomach contained the rather indigestible fragments of the lenses of fish eyes from a preceding meal, amounting to 1 percent of the total food content. Fragments of two squid (Loligo pealer) made up 11 percent of the food consumed, and finely ground fragments of an unidentifiable mollusk, 1 percent. The remains of 52 marine worms (Nereidae) formed the bulk of the meal, amounting to 86 percent of the total stomach content. Wood fibers and vegetable debris constituted the remaining | percent. Many small parasitic gizzard worms (Nematodes) were also found in the stomach. Two bird lice collected from the bird’s plumage were identified by Dr. H. E. Ewing of the Bureau of Entomology as An- cistrona vagelli Fab. Branta leucopsis (Bechstein). Barnacle Goose. Only one stomach of the rare barnacle goose was included, and it was only about one-fourth filled with finely ground vegetable fiber. Of this, 40 per- cent was sedge (Scirpus sp.), 35 percent grass fiber (probably Poa sp.,), and the remaining 25 percent was unidentifiable. Sixty-one percent of the total stomach content consisted of fine sand. Lgppenthin (6, p. 42) in his studies of Greenland birds states that he found grass, leaves, and stems of serpent- grass (Polygonum viviparum) and mountain sorrel (Oxyria digyna) in stom- achs of this species. Clangula hyemalis (Linnaeus). Old Squaw. One stomach of the old squaw, or long-tailed duck, was obtained from the far north (Cape Frigid) and unfortunately it was too nearly empty to shed much light on the normal food preferences. The following items, however, were noted: Fragments of midge (Chironomidae) larvae; soft- bodied crustacea too finely comminuted for identification; sessile barnacles (Balanidae); limpet shells (Acmaea sp.); bivalves (pelecypoda) too finely broken to indicate species; seed fragments and plant fiber of sedge, prob- ably Carex sp.; moss; and undetermined plant fiber. A fairly large series of stomachs of this attractive bird, examined in the Biological Survey, in- dicates that it feeds largely on crustaceans and mollusks. 168 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 26, No. 4 Somateria mollissima borealis (Brehm). Northern Eider. Information obtained from examination of five well-filled stomachs of the northern eider, collected at Clavering Island, northeast Greenland, on August 6, and compared with two stomachs collected on the New England coast in February and March of earlier years, suggests that this northern duck consumes a higher percentage of fish in summer than it does in winter. Furthermore, these five examinations indicate that this species probably eats more fish than do other eider ducks. Fishes of the family Cottidae (sculpin) were found in each of the five stomachs examined and ranged from 2 percent in one stomach to 79 percent in another, with an average of 26 percent of the total food content for the five stomachs. The number of individual fishes eaten varied from 1 to 25, with two birds each consuming the latter number. As would be expected, mollusks constituted the largest percentage of the food, averaging 65 percent of the total. In bulk this was nearly equally divided between bivalves and univalves, the former averaging 33.4 percent and the latter 31.6 percent. A species of soft-shelled clam (Mya sp.) occurred most frequently and amounted to 16 percent of all food taken. Other bi- valves, occurring in about the order of their importance, were the arctic rock-borer (Sazicava arctica), the common edible mussel (Mytilus edulis), and another mussel (Crenella sp.). Of the univalves, the whelks (Buccinwm sp. and T'ritonofusus sp.) occurred in each of the five stomachs and averaged 5.6 percent and 3 percent respectively of the total food content. A species of moon shell (Natzca sp.) occurred in four stomachs and a bubble shell (Bulla sp.) occurred in one. Gastropods too thoroughly disintegrated to be accur- ately named made up 20 percent of the total food consumed. Soft-bodied crustacea formed 7.2 percent, amphipoda (mainly Atylus carinatus and Corophium sp.) being most important. Fragments of cuma- ceans and isopods were recognizable in two of the stomachs. Fragments of the following minor items made up 0.4 percent of the total food: Protozon (Foraminifera); worms (Annelida); starfish (Asteroidea) ; and beetles (Coleoptera). Vegetable food, consisting of sedge and moss plant fiber and algae, formed 1.4 percent. Gravel is found in practically all duck stomachs, and in these five northern eiders, it made up 11.6 percent of the total stomach content. This, however, was not figured in the food percentages. A few feather fragments—most probably from the bird’s own plumage—were found in one stomach. Feather fragments are frequently found in the stomachs of many birds, and particu- larly is this true of most waterfowl. The five stomachs examined contained an average of more than 12 items per stomach. In contrast to this, the two winter birds collected along the New England coast contained only mollusks. One of these was filled with the common edible mussel (Mytilus edulis) and the other, which was nearly empty, contained fragments of the common mud snail (Nassa obsoleta). Falco rusticolus candicans (Gmelin). White Gyrfalcon. Five stomachs of the white gyrfaleon were obtained during the third expedition on different dates in August and September at various points on the Melville Peninsula. The stomach and small intestines of one bird were entirely empty of food, but contained many tapeworms. Of the other four birds, one had made its entire meal on 2 collared lem- ming mice (Dicrostonyx rubricatus richardsoni) and another had made 97 per cent of its meal on 3 of these rodents, a horned lark (Otocoris alpestris) making up the remaining 3 percent of the food. The Melville Peninsula is APRIL 15, 1936 COTTAM: FOOD HABITS 1€9 regarded as representing the northern border of the range of this rodent. The third gyrfalcon had in its stomach the remains of one red-backed mouse (Hvotomys sp.) but, unfortunately, digestion had proceeded too far to permit accurate specific identification. Hvotomys gapperi, according to Preble (8, pp. 50-51), is the species extending north to the Fort Churchill area. His brother, Alfred E. Preble, according to the report (Preble, 1902, p. 51), col- lected one specimen 15 miles north of this locality. It is not known what race occurs in the Melville Peninsula area, a region from 600 to 900 miles north of the record cited by Preble, which is the northernmost record in eastern North America. It is not improbable therefore that the specimen obtained from the faleon’s stomach, which marks a notable extension of the range of this genus, is a new race or species of Hvotomys. The fourth gyrfaleon had made its entire meal of a luckless willow ptarmigan (Lagopus lagopus albus). While it is evident from the limited data available that this raptor does not shun bird food, the facts also indicate _ that rodents likewise are taken in numbers and probably are equally accept- _able as an article of diet. Most published comments of the food habits of this bird would lead us to believe that only game and other birds enter into its diet. Bendire (1, p. 282) writes that ‘‘The natives assured me that they [the falcons] repair to the rugged mountains... . to breed, and that they fed their young on the rock ptarmigan, which also seek that region for the same purpose. As a rule, these rocky cliffs are the summer homes of innumerable waterfowl, on whose young, as well as on ptarmigan, they prey to a great extent during the season of reproduction.’”’ Hagerup (3, pp. 292-293) writes of ptarmigan serving as the source of food for the bird in southern Greenland. Kumlien (5, p. 84) in his account of arctic American birds states that this falcon subsists wholly on ptarmigans and hares during the winter, and that while on his frequent excursions upon Disko Island he often had oppor- tunity of witnessing the hawk preying upon jaegers, kittiwakes, and other birds. He added that he was surprised that this predator does not possess swifter powers of flight and stated that “‘their success seems to depend more upon a stubborn perseverance than alacrity of flight.’? Macmillan (7, p. 409) concludes that it feeds on eider ducks, Mandt’s guillemots, ptarmigan, dovekies, and arctic hares. Other writers also mention that birds form its principal source of food. Stomach examination, however, indicates that rodents and other small mammals enter prominently into the species’ bill of fare. Falco peregrinus anatum (Bonaparte). Duck Hawk Two young birds taken on the Barrow River, Melville Peninsula, August 29, 1933 were available for food habits study. One bird had made its entire meal on a Richardson collared lemming (Dicrostonyx rubricatus richardsont) while the other had its stomach filled with bird bones and feathers, 40 per- cent of which was from a phalarope, apparently Phalaropus fulicarius and 60 percent from a red-backed sandpiper (Pelidna alpina sakhalina). Lagopus rupestris reinhardti (Brehm). Reinhardt’s Ptarmigan. The contents of three stomachs with crops of Reinhardt’s ptarmigan, col- lected August 6, at Clavering Island, were on hand for stomach analysis. All these apparently were fairly well filled, the contents indicating that this subspecies has food habits much in common with other varieties of ptarmigan. They are all highly vegetarian and seem to choose the vegetative 170 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 26, NO. 4 and succulent parts of many herbs and shrubs. It is interesting to note that each ptarmigan collected for this study had taken a trace of insects or spiders, yet of the total food content, animal matter equalled only 0.3 percent, with 2 percent as a maximum for one stomach. This bird had taken three species of flies, dungflies (Scatophaga furcata), dance flies (Rham- phomyia sp.) and Spilogona sp. and one spider (Dictyna sp.). One other bird had taken one ant (Formica sp.) and one fly (Spilogona sp.). Bulblets of serpent-grass (Polygonum viviparum) were the dominant item, averaging 44.8 percent of all food consumed. One crop contained no fewer than 2,300 of these bulblets, which formed 100 percent of the food, while another had consumed more than 1,200, making it 87 percent of the meal. The flowers with mature akenes and leafy fragments of White Mountain avens (Dryas octopetala) ranked second in importance, averaging 19.4 per- cent of the total food of the three birds and amounted to 48 percent of the meal of one. The scales, spikes, and akenes of short-leaved sedge (Carex misandra), constituting 18.9 percent of the total average, made this species only slightly less important. One stomach contained about 500 akenes and fragments of probably 20 spikes of this species of sedge, which was 86 per- cent of that meal. The other contents were: Buds, leaves, and stems of an artic willow (Salix sp.), 5.1 percent; seeds of alpine bearberry (Arctostaphylos alpina), 3.8 percent; leaves and stems of purple saxifrage (Antiphylla op- positifolia, 3.1 percent; seeds and fruiting bodies of sedge (Carex nardina), 2.8 percent; fruiting and leafy portions of spiked wood-rush (Juncozdes spicatum), horsetail (Hquisetum sp.), sandwort (Arenaria sp.), crowfoot (Ranunculus sp.), twisted whitlow grass (Draba incana), hairy lousewort (Pedicularis hirsuta), lousewort (Pedicularis sp.), speedwell (Veronica sp.), 2.3 percent. There was an average of 12 species of food items per bird. While 17.7 percent of the total stomach and crop content was gravel, this was not com- puted as a part of the food contents. Arenaria interpres interpres (Linnaeus). European Turnstone. From an examination of three stomachs, collected August 7 at Hudson Land, northeastern Greenland, it would seem that the arctic individuals of the European turnstone are more or less omnivorous in their feeding habits. These three birds contained an average of twelve species of items per stom- ach with a total of twenty-one. The three stomachs contained 74.67 percent animal tissue and 25.33 percent plant fiber. Gravel (not included in food percentages) amounted to 22 percent of the total stomach contents. Those food items found in excess of 1 percent of the total, were: Crustaceans (mainly amphipods), 23.3 percent; diptera (two-winged flies, mainly crane- flies Tipulidae and Anthomyiidae), 20.9 percent; wood and other plant fiber, 13 percent; algae, 12.34 percent; sea-squirt, or ascidian, 9.3 percent; spiders and salt-water mites, 6 percent; brittle-stars (Ophiuroidea), 5 per- cent; bees and wasps (hymenoptera, mainly Ichneumonidae), 3.67 percent; beetles (coleoptera), 2.67 percent; butterflies (Lepidoptera), 1.7 percent; snails (gastropoda), 1 percent. Fragments of fish and a marine worm made up the remaining 1.12 percent. Lgppenthin (6, p. 55) found midges, snakeflies, the larvae of a cranefly (Tipula sp.), larvae of caddice flies (Trichoptera), a few ichneumon flies (Ichneumonidae), a spider, fragments of a phalangid (Mitopus morio) which previously had not been recorded so far north in eastern Greenland and gvravel in the stomachs of a number of birds examined. APRIL 15, 1936 COTTAM: FOOD HABITS iy! Calidris canutus rufus (Wilson). American Knot. An examination of the stomach content of four adult and two juvenile American knots collected on the second expedition at Parker Snow Bay, west Greenland, revealed that this bird has a varied diet. The four adults had subsisted largely upon plant fiber and rootlets, which they had finely pul- verized. Each bird had made most of its meal on this material. It was some- what surprising that vegetable substance entered so prominently into the bill-of-fare, forming 93 percent of the adults’ and 24 percent of the juveniles’ food. Dipterous forms, largely larvae, were the principal animal food of the young birds, while snails (gastropoda) and undetermined fragments of mol- lusks comprised the chief source of protein food for the adults. A summary of the food of these birds follows: Four adults—Fragments of gastropods and unidentified mollusks, 6.25 percent; midges (Chironomidae) and other flies (Diptera), mostly larvae, 0.75 percent; spider, trace; crustaceans, trace; fish (taken by one bird), trace; bird louse (Degeerzella sp.), trace; moss plant fiber, 1.25 percent; Phippsia grass (Phippsia sp.), 16.25 percent; undetermined plant fiber and rootlets, 75.50 percent. Two juveniles—Fragments of midges (Chironomidae), mostly larvae, 27 percent; salt flies (Ephydridae), 17.50 percent; two-winged flies (Diptera), too broken to permit more specific identification, 11.50 percent; bees and wasps (Hymenoptera), 6 percent; spiders, 2.50 percent; univalves (gastro- poda) and undetermined mollusca, 4.50 percent; amphipoda, 4 percent; moss, 1 percent; meadow grass (Poa sp.), foxtail grass (Alopecurus sp.), cinquefoil (Potentilla sp.), and willow (Salix sp.), each 0.50 percent; frag- ment of undetermined grass (Gramineae), 10 percent; sedge (Carex sp.), 4 percent; and undetermined plant fiber and rootlets, 10 percent. K. Henriksen, according to Léppenthin (6, p. 58) found the birds in northeast Greenland feeding on midges and stems of serpent-grass (Poly- gonum viviparum). Phalaropus fulicarius (Linnaeus). Red Phalarope. Two red phalaropes were collected by the third expedition on July 20, 1933 at latitude 61° north and longitude 24°20’ west, but unfortunately the stomachs of both birds were only partly filled, although several items of food were noted. The following items in their approximate percentages were recorded for the two separate birds: No. 1. Fragments of lace-winged flies (Chrysopidae, probably Chrysopa sp.), two-winged flies (Ephydridae, Chi- ronomidae, Syrphidae, Asilidae, Megaselida sp., Dolichopodidae, and un- determined Diptera), squash bugs (Coreidae) and other Hemiptera, sawflies (Tenthredinidae), grasshoppers (Orthoptera) and beetles (Coleoptera), 90 percent; pulp of some seeds, probably sedge (Carex sp.), and vegetable de- bris, 10 percent. No. 2. Fragments of waterboatmen (Corixidae), 25 per- cent; fragments of midges (Chironomidae) and robber flies (Asilidae), 25 percent; seed pulp and one seed of sedge (Carex sp.), 50 percent. Stercorarius parasiticus (Linnaeus). Parasitic Jaeger. The stomach of one July specimen, taken in the Greenland Sea, at 73°32’ north latitude and 17°10’ west longitude, contained the remains of five or more tomcod (Microgadus tomcod), forming 41 percent of the meal. The re- mains of many crustacea, mainly amphipods (Themisto libellula) made up the remaining 59 percent. It is not known whether this food or the meals of 172 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 26, NO. 4 the two succeeding species were captured direct or resulted from forced re- gurgitations by other birds. | L¢ppenthin (6, pp. 84-85) states that in an adult female’s stomach a lemming mouse was found. A juvenile bird had 3 small fishes in its gullet, while its stomach contained fragments of fish and a lemming mouse besides pieces of grass, stems, and gravel. Stercorarius longicaudus Vieillot. Long-tailed Jaeger. Two specimens of the long-tailed jaeger were collected on the same date and in about the same part of the Greenland Sea as the preceding species. Like the parasitic Jaeger, these birds also had their stomachs filled with fish and crustaceans, suggesting that their food habits, when the birds occur in the same general region, are very similar. One stomach contained the remains of ten fish (seven of which were tomcod) besides considerable crustacean material. Fish (tomeod and an undetermined species) formed 29.5 percent of the total, while the following crustaceans made up the remaining 70.5 percent: schizopods (Thysanoessa inermis), 37.5 percent; amphipods, including Themisto lzbellula, 17.5 percent and Gammarus locusta 2.5 percent; and unidentifiable soft-bodied crustaceans, 13 percent. Lgppenthin (6, p. 93) comments that numerous items are taken as food by these birds. In good lemming years he says the birds subsist largely on these creatures. In other years as was the case in 1930, he states that they are forced to search for other food. He reports that the contents of one stomach of a bird taken near a lake was full of insects, lemmings, and gravel. A whelk (Buccinum sp.) was found in the stomach of another. Still another contained a cutworm (noctuid) larva, 1 adult and 1 larva of the brush-footed fly (Argynnis sp.), 2 Dasychira groenlandica, and a few crane- flies (Tipulsa sp.), along with the remains of a lemming. One bird was ob- served to contain a green berry in its bill. He cites an example of this jaeger pursuing a snow bunting and comments that these and other small birds are occasionally taken. Larus hyperboreus Gunnerus. Glaucous Gull. Ludwig Kumlien, in his report on the Howgate polar expedition of Arctic America (5, p. 97), noted that the glaucous gulls are extraordinarily greedy and voracious, that ‘‘nothing in the animal kingdom seems to come amiss to them. ... Eggs, young or disabled birds, fish, and crustaceans are their common fare.’’ He states further that they are very fond of feeding upon seal carcasses. Results obtained in the present investigation of six well-filled, and one other stomach collected in July between 73° and 75° north latitude and 15° and 17° west longitude, and three stomachs taken in the same month at Cape York, northern Greenland, along with two taken in August at Dal- rymple Island, northern Greenland, bear out this bird’s reputation of being a scavenger and quite an omnivorous feeder. The results also suggest that depredations upon other bird life may at times be serious. Of the number of birds collected, seven of the full stomachs were from adults and four from juveniles. Characteristic of most other species, the juvenile birds had consumed a much larger variety of foods than had the adults. Remains of other birds were also more numerous in the juvenile stomachs, suggesting that these birds had been hatched in the immediate neighborhood of other seafowl. 173 COTTAM: FOOD HABITS APRIL 15, 1936 d0eI], y OF et €°0 | 2°0 | 80 z 29 | 89 Le ouen Geer Z°0 ¢°0 Th see Steg 22, 14 144 | % ‘Ay ZI ¢ I g z Z 88 I 14 E8661 II z I 9 c% a | Z 8e I 14 14 | 286261 6 66 I 14 | 218261 P GL qT | 4 O01 144 | IL8261 sopluoAns PP eo 1y 6 T On| eer GPT | 9°ez | 892 | 29 Fal Ve 14 6° 4 | % ‘aay I 001 [8661 7 14 G8 GT 14 144 | 086261 l I 14 14 I 86 14 14 14 | 626261 b I 96 Z I LL8Z61 g 66 I 918261 g 9 91 g 14 81 13 | $18261 g 0g G 13 8I 14 vt =| SL8261 ESR eee ee sy[NPy SIGE IS | ELE SEI ES |EE lee] BIBS) 2 BS 1EB) SEF EP 2 | BS ES) RS Ee) Ss | ES | ge] ce Re.) FS 5 g S b@ Bue en iy = Bo & a. & Qo53 Es a8 2g 0g * foe ae @ ot a oO a. 6 ESE = ahaee o8 | = = ual ese 8 me | ot oa | 68 ee ae eo 28 ee | QU erie lnce Bet met vice B.s = fee Ep HL = a sas a ie r= = Bgml|L 5p ae = = tic) OF A.08 EX) ae +8 E an ss = 5 ae =S a) ES : ea aaa SR: | ® BS : S Vecle gal" nal o> ) — [ ' % ~ BS eS < © = B = TTOD SQOONVIN THL AO GOO T TH]—']T WIA J, 174 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 26, No. 4 The accompanying table summarizes the food of the birds as ana- lyzed in the Biological Survey Laboratory. Where genus only is given species could not be determined. From the tabulation it is noted that coarse fish were first in im- portance with the adults, forming more than half (57.14 percent) of the total food consumed; while birds were second, amounting to 28 percent. It was somewhat surprising to find that these two items were almost of equal value with the young birds, fish (tomcod) total- ing 43.50 percent, and birds comprising 45 percent. The presence of decayed seal (Phocidae) flesh confirms the bird’s reputation as a scavenger and carrion feeder. The kelp and other plant substances taken by the young birds were taken incidental to the process of learning to feed and forage for themselves. Pagophila alba (Gunnerus). Ivory Gull. Only one heavily gorged stomach and gullet of the ivory gull, taken at latitude 74°05’ north and longitude 17°15’ west, was submitted for examina- tion. Fragments of a tomcod made 1 percent of the meal, while the remaining 99 percent consisted of fragments of soft-bodied crustaceans as follows: 115 schizopods (Thysanoessa inermis), 95 percent; 5 amphipods (Apherusa glacialis), 1 percent; copepods, trace; other undetermined crustaceans, 3 percent. Rissa tridactyla tridactyla (Linnaeus). Atlantic Kittiwake. The single stomach of the Atlantic kittiwake, taken June 22, at latitude 62°30’ north and longitude 29°07’ west was fairly well filled with fish bones, predominantly tomcod and a species resembling the Engraulidae. Ninety- nine percent of the total food content was fish, while fragments of one un- determined damsel-fly (Zygoptera) made up the remaining 1 percent. In ad- dition there was a trace of plant fiber. Sterna paradisaea Linnaeus. Arctic Tern. Only three stomachs of the Arctic tern from the north Greenland seas were submitted for examination. These were collected in July between 73° and 74° north latitude and 15° and 17° west longitude. These contained 33.67 percent fish, mainly tomcod (Microgadus tomcod) ; 65.67 percent soft- bodies crustaceans, mainly schizopods (Thysanoessa inermis); and a frac- tion of one percent of vegetable debris. One of these 3 birds was a juvenile that had fed solely on crustaceans, mostly Thysanoessa inermis. Uria lomvia lomvia (Linnaeus). Brunnich’s Murre. Of seven stomachs collected in July 1931, between 72° and 73° north latitude and 13° and 18° west longitude, only three were well-filled. On the second expedition two birds of this species, collected at Parker Snow Bay, western Greenland, were submitted—only one being full. On the third arctic trip, 34 additional birds were collected near Hudson Strait, Resolution Island and Salisbury Island. Of these, 31 were sufficiently full to be used in the computation of food percentages. From an analysis of these stomachs it is evident that Brunnich’s murre APRIL 15, 1936 COTTAM: FOOD HABITS 75 subsists principally on but few species of items. One amphipod species (Themisto libellula) occurred in all but one stomach and comprised 90.37 percent of the entire food consumed. Nearly two-thirds of the birds had taken this large amphipod to the extent of 100 percent of their meals, a number consuming no fewer than a hundred individuals. In a few of these stomachs other items occurred, but amounted to only a trace in the food. Other amphipods amounting to 1.40 percent included Gammarus locusta, Pseudalibrotus nanseni, and Gnathia mazillaris. The schizopod (Thysanoessa inermis) was apparently acceptable when available, as one murre had con- sumed more than 125 of these creatures, comprising 87 percent of its meal. But four of the birds had fed on these crustaceans and in total they formed 5.51 percent of the entire food intake. Undetermined soft-bodied crustaceans formed 1.20 percent, and in the aggregate, crustaceans formed 98.48 percent of all food consumed by the 35 birds. Fish, mostly sculpin (Cottidae), were fed upon by three birds and formed 1.06 percent of the food. The remain- ing 0.46 percent of the average meal consisted largely of polychaete worms, probably Nereidae. While their mandibles occurred in 9 stomachs, they usually formed but a trace in the food. Plant fiber occurred as a trace in two stomachs. The principal items noted in the stomachs but not figured in the computa- tions included squid (Loligo sp. which formed the principal item in one stomach one-tenth full), undetermined crustaceans, and fish. Alle alle (Linnaeus). Dovekie. Seven well-filled stomachs with five full gullets from dovekies collected on ice floes early in July from the north Greenland polar seas (between lat. 72° and 74° N. and long. 13° and 18° W.) were available for the present study. They reveal that this species shares the well-known propensity of many northern sea birds to feed extensively on soft-bodied crustaceans, as all but a mere trace of food in each of two stomachs consisted of these small aquatic creatures. One stomach contained a few fragments of fish bones and another a trace of an unidentifiable bone. A common and apparently an easily obtainable schizopod (Thysanoessa inermis), constituted the dominant item in all the gullets and in five of the stomachs, averaging 57.86 percent of the total food consumed by the seven birds. One bird had devoured no fewer than 133 of these reddish shrimp-like creatures while other stomachs showed remains of 75, 56, 53, and 33, respectively. Amphipods (Hyperiidae, Gam- marellus sp. and Themisto libellula) amounted to 16.57 percent of the total, while fragments of unidentifiable crustaceans, probably consisting largely of the schizopods and amphipods herein listed, made up the remaining 25.57 percent of the total food consumed. It was interesting to note that a trace of feathers was found in four of the seven stomachs. Cepphus grylle grylle (Linnaeus). Black Guillemot. If the two available July stomachs of the black guillemot can be taken as a criterion, its summer food in the North Atlantic consists of an unusually high percentage of fish. The two stomachs contained 79.5 percent tomcod (Microgadus tomcod), 12 percent undetermined fish bone fragments, and 8.5 percent fragments of amphipods and undetermined crustaceans. Cepphus grylle mandti (Mandt). Mandt’s Guillemot. Five stomachs of Mandt’s guillemot were obtained from North Star Bay (lat. 76°33’ N.) and Dalrymple Island, northern Greenland and Wolsten- 176 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 26, No. 4 holme, western Greenland. Of these, three were juvenile birds two of which contained but little food. The other juvenile stomach being but one-fifth full, had made 10 percent of its meal on fish, 10 percent on the gastropods (Margarites sp.), 8 percent on undetermined gastropods, 15 percent on the crustacean, Harpacticus uniremis, 25 percent on amphipoda, including Pseudalibrotus litoralis and Oedicerotidae, and 32 percent on undetermined crustaceans, probably mostly amphipods. The other juvenile birds had taken fish, squid, crustaceans, gastropods, annelid worms, and plant fiber. The two adult birds, while consuming much of the same kinds of food, had subsisted more heavily on fish. One bird had consumed 14 sculpins (Cottidae), representing 3 species. Undetermined fish comprised 4 percent of the total while crustaceans, including Spirontocaris polaris, S. fabrici, S. garmardi, Harpacticus uniremis, and Amenophia peltata, constituted 37 per- cent. Bivalves, squid (Lolzgo sp.), and annelids each formed 0.50 percent of the total. 3 As would be expected, the two races of guillemots are similar in their food abits. Hantzsch (4, p. 90) states that “‘six stomachs [of Mandt’s guillemot] contained in four cases fish remains, one digested crustacean remains, unmis- takable prawn remains, one Gammarus and one small snail.’”’ Bent (2, p. 164) asserts that “‘the food of Mandt’s guillemot seems to consist mainly of small fishes, crustaceans, and other soft-bodied sea animals.”’ Corvus corax principalis Ridgway. Northern Raven. The stomachs of two northern ravens, one taken near Parker Snow Bay and the other at Dalrymple Island, furnish evidence that this northern race is as predacious and omnivorous in its food tendencies as any of its relatives farther south. The stomach contents from the two meals are as follows: No. 1. Parts of 1 chinch bug (Nysius groenlandicus), trace; fragments of 5 horsefly pupae (Tabanidae), 1 percent; bird egg fragments, 2 percent; bird feathers and bone fragments not positively identified, but possibly from the dovekie (Alle alle), 10 percent; carrion and hair fragments of reindeer (Rangifer groenlandicus), 40 percent; other (?) carrion, 6 percent; moss plant fiber of 2 species, 1 percent; 88 berries and 125 additional seeds, fruit pulp, floral parts, and leafy fragments of the cowberry (Vaccinum [? vitis- idaea]), 26 percent; 20 berries, fruit pulp, and 8 additional seeds of the crowberry (Hmpetrum nigrum), 8 percent; plant fiber and floral parts of bent grass (Agrostis sp.), 1 percent; plant fiber and floral parts of meadow grass (Poa sp.), 1 percent; plant fiber of undetermined grass (Gramineae), trace; seed fragments and plant fiber of sedge (Carex sp.), 1 percent; leafy frag- ments of a heath, possibly Casszope sp., trace; undetermined plant fiber, 3 percent. No. 2. Parts of bird, undetermined, 15 percent; fragment of bird egg, about the size and color of that of a duck, 1 percent; fragments of mam- malian carrion, 10 percent; fragments of an undetermined insect, trace; frag- ment of mollusks, trace; 9 berries and 11 additional seeds of the crowberry (E'mpetrum nigrum), 60 percent; moss plant fiber, 1 percent; undetermined plant fiber, 13 percent. Mustela arctica arctica (Merriam). Weasel. But one partly-filled stomach of this weasel was available for laboratory study and it contained fragments of a single item—a lemming mouse (prob- ably Lemmus trimucronatus). APRIL 15, 1936 SWALLEN: NEW GRASSES 177 Citellus parryii parryii (Richardson). Hudson Bay Ground Squirrel. Five stomachs of the Hudson Bay ground squirrel show that, like its more southern congeners, it is primarily a vegetarian. While a fairly large number of plant foods were consumed, two items—vetch (Astragalus sp.) and sedge (Carex sp.)—appeared in each stomach examined, and formed 73.40 percent and 14.20 percent respectively of all food taken. Smartweeds (Polygonum sp.) formed 8 percent, while wood rush (Luzula sp.), rush (Juncus sp.) and rose (Rosaceae) formed 0.6, 0.2, and 0.2 percents respec- tively. Unidentified plant debris aggregated 1 percent. Other vegetable foods occurring as a trace included: Moss, grass (Gramineae), waterbuttercup (Ranunculus sp.), cinquefoil (Potentilla sp.), Labrador tea (Ledum groen- landicum), and blueberry (Vaccinium sp.). Four of the five ground squirrels had fed to a limited extent on the larvae of craneflies (Tzpula sp.) which averaged 2.40 percent of the food. One animal also had the remains of fish from a preceding meal in its stomach. This, however, formed but a trace of the total food. LITERATURE CITED 1. BENDIRE, CHARLES. Life Histories of North American birds. Smithsonian Institution, U.S.N.M. Special Bull. 1: 282. 1892. 2. Bent, A. C. Life histories of North American diving birds. Smithsonian In- stitution, U.S.N.M. Bull. 107: 164. 1919. 3. Hagerup, A. Some account of the birds of southern Greenland. Auk 6 (4): 292-293. .October, 1884. 4. Hantzscu, B. Contribution to the knowledge of the avifauna cf northeastern Labrador. Canadian Field Naturalist 42 (4): 90. April, 1928. 5. Kumuien, Lupwie. Contribution to the natural history of Arctic America, made an connection with the Howgate Pclar Expedition 1877-78. Dept. of Interior, U.S.N.M. Balle et 5. 1879. 6. L¢gppENTHIN, BERNT. Die Vogel Nordostgrénlands zwischen 73°00’ und 75°30’. N. Br. K¢gbenhaven. C. A. Reitzels. Forlag. 1932. 7. Macmiuuan, D. B. Four years in the white North, p. 409, 1918. 8. PreBLE, E. A. A biological investigation of the Hudson Bay region. North . American Fauna. 22: 50-51. 1902. BOTANY.—Three new grasses from Polynesia... Jason R. SwWALLEN, Bureau of Plant Industry. The following new species were in a large collection of grasses re- cently received for study from the Bernice P. Bishop Museum, Hawaii. One was collected on the island of Rapa, one on Raiatea, Society Islands, and the other on Aiwa, Fiji Islands. Aristida aspera Swallen, sp. nov. Perennis; culmi dense caespitosi, erecti, nodis geniculati, asperi, 40-60 cm alti, ramis dense fasciculatis; vaginae internodiis elongatis breviores, scaber- ulae; ligula 0.2 mm longa; laminae planae vel involutae, flexuosae vel falcatae, 2-8 cm longae, 1-2 mm latae, glabrae; paniculae 3-10 cm longae, ramis appressis paucifloris remotis; spiculae appressae vel divergentes; Received February 14, 1936. 178 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 26, No. 4 glumae mucronatae, l-nerves, prima 4.5 mm longa, secunda 6.5 mm longa; lemma 7-8 mm longum, glabrum, callo 0.5 mm longo, dense piloso; aristae 7—9 mm longae, teretes, scabrae. Perennial; culms in dense hard clumps, erect from short knotty rhizomes, sometimes geniculate at the nodes, 40-60 cm tall, somewhat flattened, minutely nodulose roughened, freely branching from all the nodes, the branches often in dense fascicles; sheaths much shorter than the elongate internodes, scaberulous; ligule minutely lacerate, 0.2 mm long; blades flat, becoming involute, especially toward the tip, firm, more or less flexuous or faleate, 2-8 cm long (usually less than 5 em), 1-2 mm wide, glabrous; panicles 3-10 cm long, the branches rather distant not overlapping, few- flowered, appressed; spikelets appressed to somewhat spreading, the pedicels 2-7 mm long; glumes mucronate, 1l-nerved, the first 4.5 mm, the second 6.5 mm long, glabrous; lemma 7—8 mm long slightly narrowed toward the tip, glabrous; callus 0.5 mm long, blunt, rounded; densely short-pilose; awns 7-9 mm long terete, scabrous, equally spreading. Type in the Herbarium of Bernice P. Bishop Museum, collected on grassy slopes of Atanui Valley, altitude 100 m, Rapa, Polynesia, July 24, 1934, by F. R. Fosberg (no. 11620). Duplicate type in the U. S. National Herba- rium. Aristida aspera is related to A. caput-medusae Domin which has a shorter panicle (3-6 em long) with short, approximate, abruptly divergent branches, and smooth culms. Locally called ‘“‘tine.”’ Garnotia mucronata Swallen, sp. nov. Perennis; culmi dense caespitosi, 10-30 cm alti, glabri; vaginae glabrae internodiis longiores; ligula truncata 0.3 mm longa; laminae planae, lance- olatae, pungentes, glabrae, 2-6 cm longae, 2-6 mm latae, marginibus scabris; panicula 4-12 ecm longa, stricta, ramis fasciculatis appressis, ad 3 cm longis; spiculae binae 4-5 mm longae basi breviter pilosae; glumae 4 mm longae; 3d-nerves, acutae vel acuminatae; lemma 4 mm longum, 3-nerve, glabrum mucronatum vel breviter aristatum; palea hyalina lemma aequans. Perennial; culms densely tufted, erect, 10-30 cm tall, glabrous; leaves crowded toward the base in a dense cluster; sheaths rounded on the back, glabrous, the lower ones short, overlapping, the uppermost elongate; ligule hyaline, truncate, 0.3 mm long; blades flat, or the margins inrolled, firm, lanceolate, sharp-pointed, 2-6 cm long, 2-6 mm wide, glabrous, the margins scabrous; panicle 4-12 cm long, strict, the branches in rather distant fascicles, appressed, the lowermost as much as 3 cm long; spikelets paired, 5 mm long, including the awn, the hairs at the base short and inconspicuous; pedicels triangular, unequal, the lower about 1 mm long, the upper 2-3 mm long; glumes equal, 4 mm long, acute or acuminate, 3-nerved, the nerves of the first glume scabrous, those of the second glume nearly glabrous; lemma firm, 4 mm long, 3-nerved, glabrous, awn-pointed or awned, the awn not more than 1.5 mm long; palea thin, hyaline, equaling the lemma. Type in the Herbarium of Bernice P. Bishop Museum, Hawaii, collected on a high moor, Temibani Plateau, alt. 700 m, Raiatia, Society Islands, Oct. 5, 1934, by Harold St. John (no. 17295). Duplicate type in the U. S. National Herbarium. Endemic to Raiatea. AprRIL 15, 1936 SWALLEN: NEW GRASSES 179 Two other endemic species of Garnotia have been described from Raiatea, G. raiateensis Moore and G. depressa Moore. In the first of these the blades are as much as 8 em long and not more than 4.5 mm wide, and in the second the blades are involute, not more than 3.5 em long, while the blades in G. mucronata are flat, comparatively short and broad, not more than 6 cm long and as much as 6 mm wide. Furthermore the lemmas in G. mucronata are awnless or nearly so whereas those of G. raiatensis and G. depressa bear slender awns 4—6 mm long. Another collection, St. John 17298, from the type locality, also belongs to this species. Eragrostis scabriflora Swallen, sp. nov. Perennis; culmi erecti dense caespitosi, graciles, simplices vel ramosi, 20- 60 em alti, glabri; vaginae internodiis breviores, glabrae, in ore dense pilosae; ligula truncata lacerata 0.3 mm longa; laminae erectae vel adscendentes, subinvolutae, 5-15 em longae (inferiores reductae), infra glabrae supra seabrae; paniculae 3-15 em longae terminales et axillares ramis solitaribus appressis, ad basin dense floriferis, ad 2 cm longis; spiculae 6—18-florae appressae; glumae acutae 1.5 mm longae; lemma 2 mm longum subacutum, minute scabrum; palea lemma aequans carinis breviter ciliatis. Perennial; culms erect, densely tufted, slender, wiry, simple or branching from the middle nodes, 20—60 cm tall, glabrous; sheaths much shorter than the internodes, glabrous with a conspicuous dense tuft of hairs at the mouth; ligule truncate, finely lacerate, 0.3 mm long; blades firm, erect or ascending, 5-15 em long, the lowermost much reduced, flat or loosely folded, smooth below, scabrous and with a few long hairs at the base on the upper surface; panicles 3-15 cm long, the branches solitary, appressed, densely flowered from the base, rather distant in the lower part of the panicle, becoming approximate toward the summit, not more than 2 em long; spikelets short- pediceled, appressed to the branches, 6—18 flowered (usually 10-12 flowered), the rachilla continuous; glumes acute, l-nerved, about 1.5 mm long, scabrous on the keel; lemmas 2 mm long, subacute, minutely scabrous, more so on the nerves; palea persistent, obtuse, a little shorter than the lemma, shortly ciliate on the upper half of the keels. Type in the Herbarium of Bernice P. Bishop Museum, Hawaii, collected on “‘bare spots in wooded basin and on limestone ridges, alt. 25-40 m, Aiwa (E),”’ Fiji Islands, August 30, 1924, by E. H. Bryan, Jr. (no. 528). Duplicate type in the U. 8. National Herbarium. This species has also been collected ‘‘on bare hillsides (dry) and moist hollows (luxuriant), alt. 10-60 m,... Olorva,” Fiji Islands (Bryan 520). This grass is apparently the one which Summerhays and Hubbard? refer to E. elongata (Willd.) Jacq. In that species, however, the lemmas are smooth and abruptly acute, the panicle branches are longer, scarcely densely flowered, and usually spreading. The sheaths also lack the tuft of hairs at the mouth which is a conspicuous character of EL. scabriflora. _? SUMMERHAYES, V. S. and Hussarp, C. E. A supplement to the grasses of the Fiji Islands. Kew Bull. Misc. Inf. 1930: 262. 1930. 180 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 26, No. 4 PLANT PHYSIOLOGY .—Inhibition of arsenic injury to plants by phosphorus.:| ANNIE M. Hurp-Karrer, Bureau of Plant In- dustry. (Communicated by M. A. McCa tt.) The inhibiting effect of sulphur on the absorption of selenium by wheat, and the quantitatively reproducible relation between the two elements at any given degree of injury to the plant,” suggested a search for other pairs of elements similarly related. It appeared from the quantitative nature of the interrelationship that the phenomenon might be a simple mass effect resulting from the inability of the plant to differentiate between the two elements, because of their chemical similarity, in the processes of absorption, translocation and synthesis. The possibility of a generalization covering other similarly related pairs of elements was then apparent.® Reasoning by analogy, from the positions of selenium and sulphur in Group VI of the periodic table, the first pair of elements chosen for investigation were arsenic and phosphorus, which occupy cor- responding positions in the adjacent Group V. The experiments were conducted with wheat seedlings germinated on blotters and transferred to flasks containing 600 cc of nutrient solution. Three different nutrient solutions were used, containing equal concentrations of calcium, magnesium, potassium, sulphur, ni- trogen, and iron but with differing concentrations of phosphorus, viz., 10, 60, and 120 parts per million as calctum monophosphate. The calcium was equalized by varying the amounts of calcium chloride. Control plants grown in these solutions were alike, showing that the differences in chlorine and phosphorus were insufficient to visibly affect growth over the period of an experiment—about two weeks. In several successive experiments it was found that arsenic added to these solutions at a rate of 30 p.p.m. as sodium arsenate killed the plants where only 10 p.p.m. of phosphorus was present, injured those with 60 p.p.m. of phosphorus, but hardly affected those with 120 p.p.m. of phosphorus. These results are in conformity with the hypoth- esis proposed to explain the selenium-sulphur antagonism, namely, inhibition of toxicity by the presence of an excess of a chemically similar non-toxic element as a simple mass effect. __ In the first two experiments the pH values of the nutrient solutions Received March 2, 1936. 2 HuRD-KARRER, ANNIE M. Selenium injury to wheat plants and its inhibition by sulphur. Jour. Agr. Research 49 (4): 348-357. 1934. Factors affecting the absorp- tion of selenium from soils by plants. Jour. Agr. Research 50 (5): 413-427. 1935. 3 Hurp-Karrer, ANNIE M. Plant physiology involved in the problem of the selentum disease of livestock. This JouRNAL 25: 335-336. 1935. AprRIL 15, 1936 BARBER: FLEABEETLE 181 were 4.2, 3.9, and 3.6, respectively, for the low-, medium- and high- phosphorus levels. In the third experiment, the reactions were ad- justed to pH 7.3, 6.2, and 6.5, respectively, by using one to two drops of strong sodium hydroxide. Acidity of the solutions was therefore not a factor in the antagonism. Repeated attempts to inhibit arsenic injury to plants in the local clay loam by additions of phosphate have been unsuccessful. How- ever an inhibiting effect was obtained in a sandy loam. These results suggest that phosphate applications will reduce or prevent arsenic injury to plants where the type of soil is such as to permit the phos- phate to remain available. Such an effect of phosphorus might be of considerable importance in areas having soils contaminated with in- jurious concentrations of arsenic from sprays. Preliminary results indicate a corresponding relationship between rubidium and potassium, the second pair of elements selected for study from their positions in the periodic table. Into the picture might possibly be fitted also the well-known calcium-magnesium an- tagonism. Other combinations such as calcium with strontium and with barium are being investigated. ENTOMOLOGY.—A new Ecuadorian fleabeetle injuring crucifers (Coleoptera: Chrysomelidae).1 H. 8. Barser, Bureau of En- tomology and Plant Quarantine. Prof. F. Campos R. recently submitted a sample of an apparently new species of green Disonycha with the statement that its larvae, in great numbers, cause damage to various kinds of cruciferous plants at Guayaquil, Ecuador. An earlier sample from the same place and observer, received in January 1918, and another sample labeled as from Chira, Peru, May 1928, G. N. Wolcott, are believed to be the same species, but host-plant records do not accompany these speci- mens. No description applicable to these samples has been found, and in order that a name may be available for the species a brief diagnosis is here offered together with notes from comparison with its near relatives. Disonycha camposi, n. sp. Length 5 mm; width 2.4 mm. Black with bluish reflection; the elytra metallic green, the occiput, pygidium, last ventral abdominal segment, distal parts of femora, and lower surface of the basal two (or three) antennal segments, yellow. Habitat: Ecuador and Peru. Similar to D. laevigata Jacoby in shape, size, sculpture, and color of 1 Received January 15, 1936. 182 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 26, No. 4 elytra but differs in genitalic details as well as in the melanic suppression of the bright yellow color of the body and appendages, vestiges of this yellow color remaining on the basal antennal joints and the apices of all femora as well as in a transverse oval spot on top of the head and another involving the pygidium and most of the last visible sternite. The pronotum and propleurae as well as the metasternum are shining black; the first four visible sternites together with the meso- and meta-pleurae are black, opaque and clothed with fine appressed whitish pubescence. The aedeagus is of similar form and curvature to that of laevgata but darker brown in color of integument, shorter and more explanate apically, and the concave orificial plate is broadly truncate at apex in the type, with deceptive brown coloration suggesting apical bifurcation, whereas in laevigata this concave orificial plate is more attenuate and acutely pointed at apex with a brownish median area which is narrowly produced and rounded apically. The aedeagus of the also very similar D. collata (Fab.) resembles the above in profile, curvature, and out- line, and the orificial plate displays a similar apical bifurcation of the brown sclerotization but its apex is laterally compressed, elevated and prolonged into an erect lamella. In a paratype from Peru the apex of the orificial plate is less truncate and suggests a slight unfolding of the adjacent part of the normally invaginated internal sac, the structure of which seems not to have been investigated in any species of Disonycha. Without such investigation the current standards of specific, subspecific, or synonymic status of avail- able specific names can be little more than temporary opinion. D. campost may also be closely related to the two Colombian species, D. exima Harold 1876, which, it has recently been suggested, may be a prior name for laewgata Jacoby 1897, and to D. steinheclt Harold 1876, but these two forms appear, from the original descriptions, to be larger and to have the pronotum and undersurface ferrugineous or testaceous. Type o’, allotype 2, and eleven paratypes in the collection of the United States National Museum. The selected holotype is one of ten specimens submitted from Guayaquil by Prof. Campos, eight of them about the end of 1917 and two in April 1935, the latter as adults of larvae injuring crucifers. Three other paratypes are labeled ‘‘No. 25-28 Chira, Peru. May 1928. G. N. Wolcott Collector.” It is a pleasure to select for this species the name of the zealous Ecuadorian who has contributed much to the knowledge of the insects of his country. CONTENTS AsTRONOMY.—Simon Newcomb, 1835-1909. EpGar W. WoouarD... CuHEmistry.—A pH conversion chart. W.H.Goss................ PALEONTOLOGY.—Nomenclatorial notes on fossil and recent Bryozoa. Res -DASKLER.. | 223. x Wand Surlele Gots WRG Chea erat ee een res ee PALEONTOLOGY.—A new Allagecrinus from Oklahoma. Epwin Kir« BioLtocy.—Food of Arctic birds and mammals collected by the Bart- lett Expeditions of 1931, 1932, and 1938. CiLarmeNcE CoTTaM Botrany.—Three new grasses from Polynesia. JASON R. SWALLEN... PLant PuysioLocy.—Inhibition of arsenic injury to plants by phos- phorus... ANNIE M) HuRp-KARRER. 0.0) EntTomMo.Locy.—A new Ecuadorian fleabeetle injuring crucifers. H.S. BARBER ;) site. seis eh ee Ree i eter. S) This Journal is indexed in the International Index to Periodicals May 15, 19386 No. 5 s JOURNAL OF THE ye JASHINGTON ACADEMY = OF SCIENCES Bet BOARD OF EDITORS FF, G. Brick wepp3 - Rozanp W. Brown Esren H. Toour 3 BUREAU OF STANDARDS U. 8S. GHOLOGICAL SURVBY BUREAU OF PLANT INDUSTRY ASSOCIATE EDITORS H. T. WEenseEL Haroup MorRIsoNn PHILOSOPHICAL SOCIETY ENTOMOLOGICAL SOCINTY EK. A. GoLDMAN W. W. Rusby BIOLOGICAL SOCIETY GHOLOGICAL SOCIETY Aa@nes CHASE J. R. Swanton BOTANICAL SOCINTY ANTHROPOLOGICAL SOCIBTY E. E. Gipson 2 CHEMICAL SOCIETY PUBLISHED MONTHLY iy BY THE | WASHINGTON ACADEMY OF SCIENCES 450 Aunarp S87. at MrenasHa, WISCONSIN Nesamuers” Entered as second class matter under the Act of August 24, 1912, at Menasha, Wis. Acceptance for mailing at a special rate of postage provided for in the Act of February 28, 1925. Authorized January 21, 1933. Journal of the Washington Academy of Sciences This JouRNAL, the official organ of the Washington Academy of Sciences, publishes: (1) short original papers, written or communicated by members of the Academy; (2) | proceedings and programs of meetings of the Academy and affiliated societies; (3) — notes of events connected with the scientific life of Washington. The JouRNAL is issued — monthly, on the fifteenth of each month. Volumes correspond to calendar years. | Manuscripts may be sent to any member of the Board of Editors. It is urgently re- quested that contributors consult the latest numbers of the JouRNAL and conform their manuscripts to the usage found there as regards arrangement of title, subheads, syn- onymies, footnotes, tables, bibliography, legends for illustrations, and other matter. Manuscripts should be typewritten, double-spaced, on good paper. Footnotes should be numbered serially in pencil and submitted on a separate sheet. The editors do not assume responsibility for the ideas expressed by the author, nor can they undertake to correct other than obvious minor errors. Illustrations in excess of one full-page halftone or two full-page line drawings may be paid for by the author. Line drawings are preferred. Proof.—In order to facilitate prompt publication one proof will generally be sent | to authors in or near Washington. It is urged that manuscript be submitted in final form; the editors will exercise due care in seeing that copy is followed. Author’s Reprints.—Fifty reprints without covers will be furnished gratis. Covers bearing the name of the author and title of the article, with inclusive pagination and date of issue, and reprints, will be furnished at cost when ordered, in accordance with the following schedule of prices. Copies 4 pp. 8 pp. 12 pp. 16 pp. Covers 50 ag pean ee ae $2.00 100 $.¢5 $1.20 $1.50 $2.00 ea As) 150 1:25 1.90 2.55 3.25 3.50 200 1:75 2.60 3.60 4.50 4.25 250 2.00 3.00 4.65 5.75 5.00 Envelopes for mailing reprints with the author’s name and address printed in the corner may be obtained at the following prices. First 100, $4.00; additional 100, $1.00: Subscreplion. Wates.—T er. WGWIME 2 oy sin cc wen Wi pais 'c a oa eae a eh nde ana $6.00. To members of affiliated societies; per volume............cccceceeeccees 2.50 Single numbers s... 5S wales ae wn lees tel Uhnke Woe ae ta ae .50 Remittances should be made payable to ‘‘Washington Academy of Sciences’ and addressed to 450 Ahnaip Street, Menasha, Wis., or to the Treasurer, H. G. Avers, Coast and Geodetic Survey, Washington, D. C. Exchanges.—The JouRNAL does not exchange with other publications. Missing Numbers will be replaced without charge provided that claim is made to the Treasurer within thirty days after date of following issue. . OFFICERS OF THE ACADEMY President: O. E. Mretnzer, U. S. Geological Survey. Corresponding Secretary: N. R. Smitu, Bureau of Plant Industry. Recording Secretary: CHARLES THOM, Bureau of Chemistry and Soils. Treasurer: Henry G. Avers, Coast and Geodetic Survey. JOURNAL | OF THE WASHINGTON ACADEMY OF SCIENCES Vou. 26 | May 15, 1936 No. 5 COSMOLOGY.—The physical universe... HiRBERT DINGLE, Im- perial College of Science and Technology, London. (Com- municated by Hucu L. DrypEn.) The word universe is used in so many senses, even within the field of scientific discussion, that it is the duty of anyone who uses the word to state what he intends it to represent. For the purpose of this lec- ture, the universe will mean the totality of physical existence; that is to say, the matter and energy existing in space considered as a whole. In using this expression, I wish at present to leave open the questions whether space is infinite, and whether if so, it contains an infinite amount of matter. Whatever the answers to these questions, we can still discuss the universe because even infinity can be treated as a whole; the infinite is not necessarily the indefinite. A simple example of this may be helpful. The series of numbers is infinite and it is there- fore impossible for all numbers to be thought of individually in a life- time: nevertheless, when a word is uttered we know immediately whether it represents a number or not. 6,425,814 is a number; a horse is not: it is not necessary to go through the series member by member to verify this. Hence infinity is not inapprehensible: we can make assertions about it and distinguish its contents from what lies outside it. It is the character of the whole, and not the catalogue of the parts that will be the subject of our discussion. As we shall see, the modern investigation of the universe is sometimes inconsistent in this re- spect: it professes to speak of the whole, but does so in terms incom- patible with its profession. That, however, is an anticipation. There is another preliminary remark to be made. A complete dis- cussion of the physical universe would include an account of its mechanical and thermal properties, for both of these are physical in character. Time will not permit of this, however, and, as a matter of fact, in the present state of science, these two aspects of the subject are quite sharply divided. We can describe the mechanical properties of a system (whether or not it is the universe I will not presume to 1 The sixth Joseph Henry Lecture delivered before the Philosophical Society of Washington on Saturday evening, February 1, 19386. Received February 11, 1936. 183 184 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 26, NO. 5 say), including its history from a hypothetical beginning to a hypo- thetical end, without any reference at all to the exchange of heat be- tween its parts which is so important from the physiological point of view, for instance; and in such descriptions we usually speak of “‘the universe”? without pausing to remark that it is only the mechanical aspect of the universe that we are considering. It is to that aspect that I shall confine myself in this lecture. Problems concerning the al- leged running-down of the universe, important as they are, will be out- side our scope. ) The universe has come prominently to the fore in astronomical discussions in the last few years, for reasons which we shall see, but it would be a mistake to regard the study of the universe as essen- tially a modern subject. It is, in fact, almost the most ancient of sub- jects, and has undergone many vicissitudes in the course of scientific history; and one of the chief objects of this lecture is to place the modern developments in a historical setting. There is a common idea that science, in its gradual progress, has at last, for the first time, caught a glimpse of the universe, its previous history having been concerned entirely with the partial aspects of things presented to us by the contents of our own neighborhood. That is emphatically not the case. On the contrary, there is exhibited throughout history a re- markable proneness among investigators to regard their horizon as the boundary of existence, and the annals of astronomy show us that no sooner does man realize that what he once thought was the whole is only a part, than he immediately invests the larger conception to which he has attained with the same illusory finality that formerly distorted the old. This is not a characteristic of the past only. Many modern investigators are as sure as ever their forerunners were that they have at last discerned the universe. It is largely in order to com- bat this attitude that I have elected to treat the subject historically, and if I say some things that have been familiar to you from child- hood, you will understand that I do not say them in order to convey information, but to indicate the proper background against which to see more modern ideas. Early astronomy was simply cosmology—no more and no less. The Earth was fixed at the centre of the universe, and the heavenly bodies were points on a celestial sphere or spheres which revolved regularly round the Earth. Certain complications were introduced into this mechanism to account for the details of planetary motions, but the essential point for our purpose is that the only subject of study was the whole system of heavenly motions, and not at all the char- May 15, 1936 DINGLE: PHYSICAL UNIVERSE 185 acteristics of any part of the system. The time had not arrived when the Sun, for instance, could be the object of individual study. Man contemplated the whole universe, beyond which was the heaven of heavens. The scheme in principle was very simple, and all that astronomy was concerned with was its adaptation to the observations of planetary positions and its application to the practical needs of man. ee These, however, proved to be surprisingly troublesome problems, and by the time of the Renaissance the elaboration of the scheme had become so complex that thoughtful minds began to suspect that some fundamental! reconstruction was needed. As everyone knows, this was begun by Copernicus. The Earth was removed from its central posi- tion and placed among the planets which revolved round the Sun. The change from a geocentric to a heliocentric system is, of course, one of the greatest advances in the history of thought, extending far beyond the particular cosmological example which introduced it, but even more important than a change of viewpoint was the introduction of the idea of infinite space. With a central Earth and a spherical firmament, cosmology came to a natural end at the frontiers of theology: there was no need of infinity, and the idea was not enter- tained. But when the firmament was destroyed, and when, shortly afterwards, it appeared that the number of visible stars could be multiplied indefinitely by increasing telescopic power, infinity forced itself on men’s attention, and the conception of an unlimited popula- tion of stars became the natural successor of the homely, limited universe of previous centuries. With this change, astronomy, in the narrower sense of the word, was born. The universe had apparently passed beyond the reach of human thought. At the same time, its separate units had entered the reach of human thought. The telescope provided the means of study- ing individual bodies, and so, by force of necessity, the contemplation of the whole gave way before the examination of the part. For a brief spell cosmology ceased to exist. It was reestablished by Newton, though in a new and previously unknown form. Newton made no attempt to grasp the aggregate of stellar existences, but he showed how to discover characteristics of the whole, even though that whole could be only partially observed. His conception of universal law restored cosmology on a new basis. He introduced the idea of principles inherent in the whole and fully ex- pressed in every part—principles which could be deduced from partial observation and then applied to the whole. His laws of motion and 186 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 26, No. 5 gravitation were of this character. They were statements, not of the actual, material structure of the universe, but of something which characterized the universe whatever its material structure might be. Cosmology therefore changed from a description of the body of nature to a search for the soul. This, to be sure, involved assumptions. Before you can extend a generalization from a few observations to the whole of material existence you must postulate that the whole of material existence is in fact a universe, having universal characteristics. If that is not true, your laws of motion or gravitation or what not may hold in your own neighborhood, but not elsewhere. Furthermore, you must assume that your observations of a limited region are accurate enough to enable you to apply your deductions throughout the whole—possibly in- finite—extent of space; you must assume, not only that there are universal laws, but that you have actually found them. These as- sumptions are worth noting because they have perhaps been too generally ignored. When, by wniverse, you mean, as in the old days, a particular system of material bodies, there are no such assumptions; cosmology is simply a statement of what you observe. But when you are forced to give up the attempt to observe the whole, but still pre- sume to speak of a wniverse because the part you observe shows mu- tual relations between its parts, it is well to bear in mind that you have no longer the solid ground of primary fact beneath your feet, but an idea which, however, plausible, is still of an essentially dif- ferent character from the basis of the cosmology of old. There is no harm in assumptions, however, provided their existence is not forgotten. Accordingly, Newton’s laws were accepted and their implications examined. One would hardly expect that laws which, by hypothesis, are held to be true whatever the material contents of space, would be able to give any information about those contents, yet this turned out to be possible. It became clear that if Newton’s laws were actually universal, there could not be an infinite space in which stars were uniformly distributed, no matter how coarse the scale considered or how thin the distribution postulated. Since, by that time, the idea that space was infinite had become almost axio- matic, this meant that as one receded to distant places one would at last pass beyond the region of matter into emptiness: the universe of matter would be an island in an infinite ocean of vacuity. This was de- duced from the fact that in a region of uniform material density, the mass of matter in a sphere centred at the Earth would be proportional to the cube of the radius, whereas the effect of distance on the gravi- May 15, 1936 DINGLE: PHYSICAL UNIVERSE 187 tational attraction of this mass on the Earth would be inversely pro- portional to the square of the radius. Hence Newton’s law of gravita- tion would require that the gravitational effect at the Earth of the matter in such a sphere would, on the whole, increase with the radius, and so, with an infinite distribution of matter, would become infinite which was held to be contrary to observation. I am not going to discuss this argument, because I am now simply describing history, not criticising it. It might be well, however, to say a word or two about the meaning of the coarseness of the scale of density considered. It is obvious, of course, that on the fine scale, the distribution of matter is far from uniform. The density of the Earth is much greater than that of water, and the density just outside it is almost zero. Two volumes of, say 1 cc. each may therefore contain very different amounts of matter, depending on the particular cubic centimeters chosen. If, however, we compared two volumes of, say, a billion cubic light years, we might find that they contained approxi- mately equal amounts of matter; and if not, it might still be that much larger units of volume would give us a uniform distribution. Two volumes of 1 cubic millimetre in a sponge may differ widely in material content, for one may be in a hole and the other in the ma- terial of the sponge: two volumes of 1 cubic inch, however, would con- tain approximately equal amounts, so that there is a scale on which the sponge is a uniform distribution of matter. Since, in the cosmologi- cal problem we are, by hypothesis, dealing with infinite space, all finite units are infinitesimal, so we are justified in considering units of any finite size at all, and if we can find a size which will give us uni- formity of density, the argument applies. Now the conclusion of the argument was held to be unsatisfactory. It was not philosophically satisfying to think that a particular bit of space was occupied by stars while the rest was empty. Minds which had awakened from a geocentric illusion were apt to regard with sus- picion any idea which tended to assign them a unique position in space, and so it was felt that somehow space ought to be filled uni- formly—in other words, the universe should be homogeneous. Ap- parently, however, this could not be secured without abandoning the universality of Newton’s law of gravitation, and this also was repug- nant. No solution of the difficulty presented itself, and so the matter rested. : Meanwhile, telescopes and spectroscopes had been busy increasing the extent and accuracy of our knowledge of the contents of our own neighborhood. The solar system belonged to a vast organization of 188 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 26, No. 5 stars which appeared to be one of many such organizations—the spiral nebulae. The evidence for this was rather slender at first, but increased later, and at the time which we have reached in the de- velopment of theoretical ideas, it had captured the imaginations rather than the intellects of astronomers. Despite definite indica- tions that our stellar system was thinning out with distance from the Sun, as the gravitational argument required, the spiral nebulae were held to be island universes which, on the coarse scale, might provide the homogeneity of material distribution which instinct demanded. Insofar as the unsettled ideas of the time can be said to have pre- sented a definite aspect, it was either that of a single system of ma- terial bodies in a surrounding infinite void, or that of a multitude of such systems with an unknown distribution throughout that void. Whichever idea was held, however, it was regarded as an idea of the universe. The possibility that what we observed was too small to be typical in any degree of the whole was scarcely accorded even lip- service. Such was the position at the advent of relativity. This affected our problem in two ways. In the first place, it showed that we could not adequately consider the distribution of matter in space without considering its distribution in time also—a fact which is often falsely expressed by the statement that time and space are identical. And, in the second place, it displaced Newton’s law of gravitation by another which did not prohibit the possibility that space was completely filled by a homogeneous distribution of matter. I am not going to discuss the principle of relativity itself but I want to say a few words on each of these points. The inseparability of time and space is easier to realize in relation to the universe than in any other relation, because of the greatness of the scale. It all depends on the fact that we know of no physical means of communication between separated points of space which is more rapid than light. Between terrestrial events light passes practi- cally instantaneously, so that we can say with sufficient exactness that a thing happens when we see it. But when we see an event on even the nearest spiral nebula, we know that that event happened nearly a million years ago. We can theoretically, if we wish, deter- mine what terrestrial events were simultaneous with it, but when we are dealing with such intervals, simultaneity loses its interest. We are much more concerned with what we can see now in the nebula than with what is happening there now, which will not be revealed to us for nearly a million years. Hence the structure of the universe May 15, 1936 DINGLE: PHYSICAL UNIVERSE 189 at any one time is no longer the object of our investigation. Instead, we take history and structure together in a union which relates our present observations with one another and relegates to the back- ground the results of calculations concerning conceptual simultane- ous happenings. That is all that the amalgamation of time and space means, and if one thinks of it in terms of these large-scale phenomena instead of terrestrial events it ceases to be mystical and one is saved from the illusion that somehow time and space are the same, despite our knowledge that they are very different. The possibility that space might be filled homogeneously with matter without violating known facts arises from the possibility, afforded by the new law of gravitation, that space might be finite, though unbounded. This is another of the so-called mysteries of rela- tivity, but it is not at all incomprehensible I think, if it is approached in the right way. The difficulty which one feels is that if the matter of the universe is contained in a finite space, one can always imagine an infinite space outside that finite region, so that it is not accurate to say that space is finite. Very well, let the surrounding infinite space be imagined. We then have a homogeneous sphere, let us say, of matter surrounded by in- finite emptiness. This sphere, of course, consists of the stars and nebulae that we see around us, together with all that might be beyond the present reach of our telescopes; and we are somewhere in the midst of it. Now let us try to send some matter outside this region into the enveloping void. We find this impossible, because the law of gravita- tion is such that the matter cannot escape;? it always returns like a stone thrown into the air. It is possible, of course, to project a stone fast enough to escape from the Earth altogether, but the laws of mo- tion and gravitation in the Einstein universe are such that it is im- possible for the stone to get free into our imaginary void. There is nothing difficult to conceive in that: it simply means that nature pro- vides no method of projecting a body fast enough to escape from this system of bodies which we call the universe. We might imagine, however, that we can see outside the system. But that is impossible also. If we project a beam of light upwards, it will be attracted by the rest of the universe and bent back so that it cannot escape. Furthermore, if we were situated at the very boundary of our sphere we could still look upwards and see only light from stars beneath us. The stars would appear to be above us because their light * I am now describing the original static Einstein universe; we will consider later the more recently conceived expanding universe. 190 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 26, NO. 5 would enter our eyes in a downward direction, but the stars them- selves would not be there. If we travelled along their light we should go upwards a little, and then turn round until at last we should enter the star from which the light came, at a place somewhere inside the system. This also is not inconceivable; it is simply an example of a fact very familiar to everyday experience. When we see the Sun on the horizon in the evening, it has already set: its light is bent by the atmosphere of the Earth so that we see the Sun although it is already below our level. In precisely the same way, though from a different cause, the stars which our boundary observer would see above him would be below his feet. Now let us notice two things. Since nothing—neither light nor matter—can escape from our universe, and we get no indication of anything outside, we can dispense with the notion of a surrounding emptiness. We can go on thinking about it if we like, but if so we are confusing our thought, because we can get no knowledge of it: every- thing we can see or know belongs to our own system of stars. Hence, since astronomers do not wish to load their minds with ideas which serve no purpose, they discard it, and call space simply the region which is accessible to observation. The surrounding infinity has no scientific existence: it is simply an illusion, necessary in an immature stage of thought, just as the propositions of arithmetic have at first to be incarnated in problems about apples or pears. When we know the principles of the problem we can dispense with the part which does not enter into the solution. The second thing to notice is that since our hypothetical observer at the boundary of the system sees stars above him just as an inside observer does, he cannot tell that he is at the boundary. Detailed considerations show, in fact, that his view would be indistinguishable from that of the inside observer, and therefore there is nothing to be gained by saying that he rather than the other is on the boundary. The statement previously made that the stars were ‘beneath his feet”’ would mean nothing to him: that would be the appearance to a hy- pothetical observer outside the universe. We have already dispensed with the outside space which is all that could give a meaning to a boundary, and so it follows that there is no boundary. Wherever you are in the universe you see the same kind of spectacle and have the same freedom and limitation of movement. Hence one part of the universe is just hke another: the universe is homogeneous and un- bounded. It is nevertheless finite because you can, in principle, count the stars in it and get a finite number. Also, you can travel about May 15, 1936 DINGLE: PHYSICAL UNIVERSE 191 and, in principle, survey the whole. The whole difficulty of conceiving of finite unbounded space is therefore that of discarding artificial aids to thought: when the lesson is learned, the parable is no longer neces- sary. The universe I have described is the famous Einstein universe of the early days of relativity. At that time the remarkable phenomenon of the recession of the spiral nebulae*® was not known. It has since been discovered that these objects are receding from us (i.e., from our whole stellar system, for we are now speaking on the largest scale consistent with our knowledge) at speeds proportional to their distances: the most distant nebula now known to take part in this recession is some 230 million light-years away, and it is moving with a speed of 40,000 kilometers a second. Of course, there is much uncertainty about these measurements—not only with regard to the actual figures, but also with regard to the interpretation of the phenomena which we call distance and velocity. The velocities are measured by the displace- ments of spectrum lines, and these may have some unknown cause in objects so exceptional and unfamiliar as those which we are consider- ing. Nevertheless, there is no valid reason for rejecting the ordinary interpretation. Many feel inclined to do so because of the enormous velocities involved, but that is pure prejudice. There is no reason at all why nebulae should not move at such speeds, and indeed we ought to expect that-the qualities they exhibit will differ in important re- spects from those to which our provincial minds are accustomed. When we have, as here, to choose between exceptional behavior of bodies and exceptional laws of nature, the former is certainly the choice which is more in accord with the general practice of science. If, then, we reject unscientific expectation as a sufficient reason for doubting the motions of the nebulae, we are left with only the healthy scepticism which does not forget that it is employing speculative building material, but nevertheless goes on building. Relativity mechanics is quite consistent with a universal nebular recession; indeed, there is a sense in which it predicted the phe- nomenon, though only as one among other possibilities. The mutual support of observation and theory therefore gives considerable plausi- bility to the idea that we are at last in touch with the universe itself, for the mechanical system which relativity describes as expanding is one which is completely self-contained. Like the rejected Einstein universe, it has no space outside itself which has any scientific mean- 3 The term, spiral nebulae is used to denote the extra-galactic nebulae generally: many of them, of course, do not show a spiral structure. 192 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 26, NO. 5 ing, notwithstanding the fact that we find it difficult to conceive how a system can expand when there is nothing for it to expand into. This difficulty appears to be at bottom a play upon the word nothing, for actually nothing is just what we want to make the idea credible; if there were something there, expansion might be prevented. To put it in another way, space becomes significant only when the system has expanded so as to include it: whatever is not included at any instant has no significance, for it is quite without influence on the system at that instant. At a later instant, when the system contains more space, we are inclined to think of the added amount as having previously existed outside, but we cannot identify that added amount. It is not something which we can label and say, ‘‘This part has been in the universe all the time, and this part has only just come in.’’ All we can discover from observation is that if we measure the distance be- tween two nebulae at two different instants, the distance will be greater at the later instant. We do not observe a boundary constantly moving outwards; we simply observe nebulae receding all around us. Consequently the expansion of the universe is only a picturesque expression of a characteristic of our measurements. If we try to pic- ture the universe as an isolated system from which we can withdraw and observe it as a whole, we feel compelled to insist on a surrounding space in which it can be isolated. If we give this up and keep to the actual facts of the case, which are simply concerned with relations be- tween our measurements, our difficulty vanishes. I think the difficulties of understanding the modern views of space are enhanced by the idea of space curvature, which is so commonly talked about. The fact is that this is a purely mathematical expres- sion. When we are dealing with a simple two-dimensional surface, like that of the Earth, situated in our ordinary, familiar three-dimen- sional space, a certain mathematical function of our survey of the surface happens to represent our ordinary idea of its curvature, and for that reason the function 1s called the curvature. Now our imagina- tions will not extend to athree-dimensional ‘surface’ situated in a four- dimensional continuum, but our mathematics can easily be general- ized to include any number of dimensions, so that mathematically we can speak of the curvature of this three-dimensional “‘surface”’ which is our ordinary space. But it is a great mistake to try to picture this, for it is quite impossible. All we can imagine is the set of measurements we make and the relations between them, and it would be far better to speak of those imaginable things than to transfer a quite legitimate mathematical metaphor into a sphere of thought where it has no May 15, 1936 DINGLE: PHYSICAL UNIVERSE 193 place. Space is curved only in a mathematical sense; in terms of what we ordinarily mean by curvature, the idea is meaningless. The present idea of the expanding universe thus owes its existence to the impact of observation on a set of possibilities offered by rela- tivity mechanics. Theoretically, the static Einstein universe is pos- sible, provided that space is finite, though unbounded. The expanding universe, however, is subject to less limitation. Either finite or infinite space is permissible if it is expanding, for there is then no imperative necessity for every ejected body to return to the system; it all de- pends on the conditions of the expansion. We may therefore have a finite amount of matter homogeneously distributed in a region in such a way that its parts are continually getting further and further apart; or we may have an infinite amount of matter similarly distributed. Which we actually have we do not know. Possibly we have neither, for the distribution of matter may not be homogeneous, and in that case, many other possibilities are open. With our present limited knowledge, however, we prefer to think that there is no unique region of space, such as the place of greatest density would be, and so far as observation has extended, our preference is justified. We must not for- get, however, that all that we can yet observe may be not only a very small part of the whole, but a part which is not large enough to show anything but local peculiarities. A surveyor of a mountain range would hardly discover the true spheroidal form of the Earth by generalizing his observations. It therefore remains quite possible that the universe is static on the large scale, and that the expansion we observe is local—not in space alone, but also in time. In that case it may later cease and turn to a contraction—the whole universe, while showing these minute disturbances, remaining majestically quiescent on the grand scale. There is a tendency to suppose that this cannot be because, as Eddington has shown, the static Einstein universe is unstable, and a small disturbance would necessarily make it expand or contract. This argument, however, seems to me to work in just the opposite direction. If we are actually considering the whole universe, then there is no difference between stability and instability. A mechanical sys- tem is stable when, if it is very slightly disturbed by some external force, it tends to recover its former state; and it is unstable when, in the same circumstances, it tends to depart further from its former state. Now what Eddington has shown is that if we have an Einstein universe, one such disturbance would necessarily destroy its static state and set it expanding or contracting, and he concludes that such 194 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 26, NO. 5 a disturbance occurred sometime in the past. But if he is actually deal- ing with the universe, such a disturbance cannot happen because there is, by hypothesis, nothing outside to cause it: if there is some- thing outside (i.e., not outside a region of space but outside the con- tent of the definition of the mechanical system), then the system is not the universe. Insofar, then, as relativity mechanics or astronomical observation indicates that the system of spiral nebulae has expanded through the disturbance of a primordial Einstein system, it indicates also that that system was not the universe but only a part thereof which was affected by the rest. It therefore argues in favor of the idea that our expanding system of nebulae is merely a local unit in a larger universe. This is one of several examples of the difference between the mechanical problems of the universe and those of any partial system. The whole of physical nature has unique characteristics arising sim- ply from the fact that it is the whole and not a part, and nothing is easier than to overlook this difference and apply ‘established’ mechani- cal laws where properly they are inapplicable. Another example is the inapplicability of the ordinary process of abstraction. We can ab- stract laws of motion from moving planets, falling stones, rolling billiard balls, swinging pendulums, and so on, by ignoring everything that is not common to all such things, but we cannot ignore anything in the universe. There is only one universe, and therefore there is no difference between essential and accidental characteristics: they are all one. Again, all considerations of probability are irrelevant. It is quite illegitimate to seek the most probable behavior of a universe by considering many universes, as in statistical mechanics we seek the most probable behavior of an assembly of molecules by considering many such assemblies. Many universes are not merely actually non- existent; they are logically non-existent, for the universe is by defini- tion the whole. This fact also has been conspicuously overlooked, I think, in certain considerations of the character of the universe. And, as a final example, the universe is the only mechanical system which has no boundaries, and therefore the problem of the universe differs from all other mechanical problems in having no boundary conditions. For these reasons, I do not feel greatly concerned about the most prominent enigma in the present position of the universe problem— namely, the very short time-scale by which we appear to be faced. The recession of the nebulae can be extrapolated backwards to a time when all the matter forming them was more or less together in one place, and that time is no longer ago than about 210° years—which May 15, 1936 BARTELS: GEOPHYSICAL CYCLES 195 is smaller than is convenient to explain certain characteristics of stel- lar astronomy. Certainly the problem is a very pertinent one, and de- serving of all consideration, but it seems to me that whenever the re- quirements of the theory of the universe come into conflict with the requirements of special studies, it is the former that is far more likely to yield. The universe itself is a postulate—it is not something we observe: and the conditions of its study are so peculiar and so little apprehended as yet compared with those of partial problems that it seems fantastic to allow wide extrapolations from the most difficult of observations to dominate the detailed studies on which scientific laws themselves are founded. The study of the universe is helpful if it lifts our minds above the limitations imposed by undue specialization, but it becomes harmful if we allow it to fetter us still further in our consideration of the actual. GEOPHYSICS.—Some aspects: of geophysical cycles.1 J. BARTELS, Department of Terrestrial Magnetism, Carnegie Institution of Washington. (Communicated by L. R. Harsrap.) The common geophysical cycles of everyday experience can be con- nected with the rotation of the earth (solar and lunar diurnal-varia- tions) or its revolution (annual variations), with the sun’s rotation (27-day recurrence in terrestrial-magnetic activity, etc.), or with some physical resonance-effect depending on the form of the essential differential equation (water-waves, seiches, seismic waves, etc.). The familiar up and down in the records of many geophysical and other phenomena (atmospheric pressure, temperature, rainfall, tree-ring widths, business indices, mortality, etc.) has challenged many efforts to trace cycles of other periods than a day or a year. The outcome of this extensive search, made with analytical, graphical, optical, or mechanical procedures, is a bewildering variety of cycles claimed, but all are of a controversial nature. This curious result seems to be due to inadequate methods of research. A study of typical cases is pro- posed, leading to an outline of a statistical theory of geophysical and cosmical periodicities. Before the cyclic effect has been established in the series of observations by statistical methods, a physical ex- planation cannot be attempted. The usual mathematical analysis gives at the same time too much 1 The subject of this paper formed the basis of a more extended address under the above title by the author at the 1087th meeting of the Philosophical Society. Received January 10, 1936. 196 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 26, NO. 5 and too little. The theorem of Fourier series, or the recently developed theory of almost periodic functions, assert namely that every finite series of observations can be satisfactorily approximated by a sum of sine-waves of different periods and amplitudes, and even an infinite number of such sums can be found. This purely mathematical fact does not lend any significance to each individual sine-wave. A con- tinuation of the sum of sine-waves into the future ceases in general to fit and is therefore of no use for forecasting (example: H. Kimura’s? analysis of relative sunspot-numbers 1750-1911; the 29 sine-terms, continued, disagree with the observed numbers since 1912’). A nota- ble exception is furnished by tidal theory, in which the periods of the essential sine-waves in ocean-tides are derived from the astronomical - data on the movements of earth, moon, and sun; the amplitudes and phases of these sine-waves are computed by harmonic analysis of at least one year of tide-gage observations and then used for prediction. The same holds for tides of the solid earth (experiments of Michelson and Gale).* Such a complete representation of the data as sums of interfering sine-waves, with periods of submultiples of a master- period, is the aim of many workers on cycles, but it can be misleading (example: Seasonal change of amplitude in diurnal variation of at- mospheric temperature at Berlin can be interpreted as spurious stidereal-time variation). The problem of cycles is only part of a more general one, namely that of the morphology of geophysical time-curves, which should comprise an adequate mathematical description of the essential fea- tures of such curves. For simplicity, only values given at equidistant time-intervals (hourly values, ete.) are considered. Such a group of values for the same quantity, regarded as a statistical population, can be submitted to the ordinary procedures for computing averages, frequency-distributions, standard deviations, etc. These statistical conceptions are mostly developed in biometry. There is, however, a fundamental difference between lists of data in biometry and the time-series of geophysical observations; successive data in the former case are mostly independent (random), while geophysical data mostly show positive conservation, that is, high values are more . likely to be followed by other high values, etc. Thus, for example, if one examines the differences of consecutive random numbers he finds in them negative conservation, whereas a similar examination of 2 Mon. Not. R. Astr. Soc. 73: 543-548. 1913. 3 Terr. Mag. 39: 231-236. 1934; 40: 215-217. 1935. 4 Astroph. J. 50: 3380-345. 1919. May 15, 1936 BARTELS: GEOPHYSICAL CYCLES 197 daily sunspot-numbers will reveal positive conservation. The original ordinates may be y, and the averages of h successive ordinates may be y(h); the standard deviations may be m=m(l) for the y, and for the y(h). For random-ordinates, m(h) =m//h; for ordinates with conservation, this is measured by a(h) [m(h)/VWh) ].2 Because m(h) =m/V/h/o(h), c(h) is called equivalent number of identical ordi- nates, or h/o(h) is called equivalent number of random-ordinates among h successive ordinates. [¢(2h)/ca(h)|—1 is the ordinary correlation- coefficient between successive averages. Example: Daily relative sun- spot-numbers, near sunspot-maximum; for h=365, h/c(h) is only 2.6. If c(h) approaches a limit o, this might be called the natural time-unit for the quantity. The usual computation of cycles is demonstrated for the lunar atmospheric tide. It begins with arranging the data in rows (individ- ual sets) of r ordinates, and forming average sets, which consist of the averages of the r columns of hf individual sets. The standard devia- tion of the individual sets may be ¢, that of the average sets may be ° c(h). In the random-case, in which there is no significant cycle of a period comprising r ordinates (these ordinates themselves may be random or not!), ¢(h) =¢/Wh; otherwise, o(h) can be computed as above as the square of the ratio ¢(h) to ¢/Wh, and, if significantly different from unity, denotes quasi-persistence (the analogue of con- servation), measured by a(h), the equivalent length of sequences. If o(h) approaches a limiting value o(«), the series shows asymptotic quasi-persistence (example: Twenty-seven-day recurrences in terres- trial-magnetic activity, o(%) between 3 and 4). If o(h) is asymptoti- cally proportional to Wh, a persistent wave is found. Harmonic analysis allows a refinement of this method by taking into account the conservation in the original ordinates. Individual sets of r ordinates, in the time-interval t=0 to 27, are represented by sums of terms (a,cos vt+b,sin vt) =c¢sin(vt+a,). In the harmonic dial for fixed frequency v, an individual wave is represented by a plane vector with the rectangular coordinates a,, b,, and the polar coordinates c,, a,, or only by the endpoint of this vector; in the generalized harmonic dial in (r—1) dimensions, the vector has the coordinates ai, 61, dz, bs, ..., and the square of its length is c?+c,” +---=2¢. For random-ordinates, the mean c,? is about 4¢£?/r, independent of the frequency; in ordinates with conservation, the mean ¢,’ is smaller than that value for higher frequencies, and greater for low frequencies. The mean c¢, plotted against v gives the mean periodogram. 198 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 26, No. 5 Cycles with fixed period can be classified by considering clouds of points in the (plane or generalized) dial. The dimensions of these clouds are given by the mean harmonic amplitude, c for individual sets, c(h) for averages of h sets. For random-cycles, c(h) =c/ Wh, that is, the clouds for averages shrink in the ratio 1/./h with respect to the origin. In case of quasi-persistence, the clouds shrink more slowly than in the ratio 1/-/h, namely, in the ratio 1/Wh/c(h), with [c(h) /(c/Vh)\2?=c(h). o(h) =0() denotes asymptotic quasi-persistence. These conceptions may be illustrated nicely by examples dealing with terrestrial-magnetic activity; a number of such are treated by the author in his paper Random fluctuations, persistence, and quasi- persistence in geophysical and cosmical phenomena® where random- walk and summation-dial are used for illustration instead of shrink- ing clouds. A complete analysis should comprise the whole spec- trum, but the actual amount of necessary calculations is reduced be- cause of the infective property of quasi-persistence in cycles of adja- -cent periods. Correlations between cycles in different phenomena are traced by setting the clouds in one dial to the same phase and turning the phases of coordinated vectors in the other dial by the same phase- angle. An example: Suicides in Berlin 1906 to 1914, compared with terrestrial-magnetic activity, full or half solar-rotations (27- or 13.5- day cycle); no appreciable correlation, contradicting the conclusions of IT. and B. Diull.® This statistical aspect of cycles is imposed by the fact that, instead of the carefully planned experiments in the laboratory, geophysics often provides us, so to say, with a number of large-scale experiments which all go on simultaneously, and the results of which have to be disentangled. The methods described need not be applied in detail in each case, because, after some experience, the examination of some samples will suffice for estimating the significance and the character- istics of a cycle. As to the mathematical side, the methods may be considered as a first attempt to amalgamate the ideas of A. Schuster’ on the statistical aspect of hidden periodicities with those of Lexis? on super-normal and subnormal dispersion in ordinary statistical series; there is room for considerable refinement, because the formulae 5 Terr. Mag. 40: 1-60. 1935. 6 Bioklim. Beibl. 2: 24-31. 1935. ’ Terr. Mag. 3: 138-41. 1898; Cambridge Phil. Trans. 18: 107-135. 1899; Phil. Trans. R. Soc., A, 206: 69-100. 1906. 8 H. L. Rimrz, Mathematical statistics, 146-155 (1927). May 15, 1936 GAZIN: MUSTELID CARNIVORE 199 were derived roughly as an emergency measure not only for detecting significant cycles, but even more for checking the exuberant produc- tion of cycles claimed without adequate tests, or, worse, after mis- leading applications of distorted periodogram tests. PALEONTOLOGY .—A new mustelid carnivore from the Neocene beds of northwestern Nebraska! C. Lewis Gazin, U. S. National Museum. In the spring of 1935 the U. 8. National Museum purchased from Mr. Ted Galusha of Hay Springs, Nebraska, a fossil mustelid skull from the later Tertiary of Nebraska. The specimen was found in a draw to the south of Antelope Valley, near the center of the southern half of Sec. 30, T. 31 N., R. 47 W., Dawes County, Nebraska. The deposits at this locality are believed to be equivalent to the Snake Creek beds, but the horizon represented in this series is not definitely known. The age is apparently upper Miocene or possibly lower Pliocene. The skull is unique in comparison with living forms and is distinct from known skulls of fossil mustelids, and although comparisons with fossil forms known only from lower jaws are difficult, and perhaps unsatisfactory, an analysis of the requirements for dental occlusion apparently eliminates it from previously described genera. Craterogale,’ n. gen. Type.—Craterogale simus. Generic characters.—Skull short and broad; strong postorbital processes and rugged, separate temporal ridges; prominent lambdoidal crests and mastoid processes; zygomae deep, powerful, and widely expanded; infra- orbital foramen of moderate size and nearly circular; basicranial region short with large bullae firmly fused to surrounding elements; bullae project mark- edly downward, forward, and slightly inward, with antero-externally di- rected foramen through projected portion; audital tube short; foramen ovale, foramen lacerum medius, and eustachian foramen closely grouped; dental formula; teeth small and stout; P* relatively small; P* with, small parastyle, deuterocone well forward, and without tetartocone; M! with elongate external portion, conspicuous parastyle, and moderately large heel which is expanded anteroposteriorly to about the same extent as the ex- ternal portion; protocone of M! large and hypocone less prominent. 1 Published by permission of the Secretary of the Smithsonian Institution. Re- ceived February 17, 1936. 2 Derived from the Greek krateros, strong, gale, weasel. 200 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 26, NO. 5 Craterogale simus,? n. sp. Type.—Nearly complete skull, U.S.N.M. no. 13801, including P? to M! on both sides. Horizon and locality—Snake Creek (?) beds, Upper Miocene or Lower Pliocene, Dawes County, Nebraska. Specific characters.—Only known species; size of skull approximately one- fifth larger than that of Mephitis nigra, but much more rugged; P? to M! about 24 mm. Other specific characteristics are not clearly discernible in the absence of additional species. Description of skull—The skull of Craterogale stmus is intermediate in size between that of a fisher, Martes pennantz, and of a martin, Martes amerv- cana, but relatively much broader and more rugged than either. The skull is proportioned somewhat more as in Mephitis, though larger, and with a breadth suggestion of the felids. ——— SIL) eS — yi SA K SS Zp $ Vi a / ———— ous Fiz 1.—Craterogale simus, n. gen. and sp. Skull, type specimen, U.S.N.M. no. 13801, lateral view, natural size. Upper Miocene or lower Pliocene, Dawes County, Nebraska. Drawing by Sydney Prentice. The rostrum is short and broad with prominent pre-orbital or lachrymal fossae. The orbital cavities, though open posteriorly, are well defined by prominent postorbital processes of the frontals and noticeable, but less de- veloped, processes on the jugals. The anterior margin of the orbit is inter- rupted by a marked outward and upward projection of the lachrymal bone. Laterally the maxillae join deep and powerful zygomae. The relative de- velopment of the zygomatic arch is not approached in any of the living mustelids and its depth is actually equal to or greater than that in much larger individuals of the wolverine. On its anterior and ventral surface the zygoma shows a marked excavation, presumably for attachment of the an- terior masseter lateralis. The infraorbital foramen is of moderate size, nearly circular, and terminates anteriorly above P*. On the dorsal surface of the rostrum the premaxillae project backward between the nasals and maxillae to a point of contact with the frontals. In some mustelids, as in the Mephi- tinae the premaxillae apparently do not extend so far back, but in the Lutrinae, for example, both conditions are found. The cranial portion of the skull is moderately elongate and not greatly inflated, much less so than in Lutra or Taxidea. The dorsal surface of the cranium is characterized by strong, outstanding temporal ridges, more 3 Derived from the Greek szmos, snub-nosed. May 15, 1936 GAZIN: MUSTELID CARNIVORE 201 rugged than in Helictis, extending from the prominent postorbital processes and remaining separate to the occipital crest. The ridges most closely ap- proach one another at a point slightly posterior to the postorbital constric- tion and are most outstanding near the occipital crest. A weak sagittal crest is present from about the postorbital constriction to the occiput. The posterior or occipital portion of the skull exhibits heavy lambdoidal erests which extend from the temporal ridges at the top of the occiput in diverging, nearly straight lines to the prominent mastoid processes. These ridges project outward and backward to a marked degree. The paroccipital processes, just posterior and median to the mastoid processes, are not com- pletely preserved but do not appear to have been developed so strongly as the mastoid processes. The surface of the occiput shows a moderately in flated median ridge from the foramen magnum below to the crest of the occi- put. The foramen magnum and the occipital condyles are of moderate size and the surface of articulation extends entirely around the lower margin of the foramen. The palate is broad and short, though relatively a little longer than in Mephitis, and shows a prominent groove on each side extending forward from the principal posterior palatine foramina. These foramina are located inward from about the middle of the carnassials, not so far forward as in the otters and skunks, nor so posteriorly placed as in the badgers. The portion of the palate extended posterior to the dentition is noticeably concave transversely and projects backward to a point about equivalent to that in Lutra or Helictis, more than in Mephitis, and distinctly less than in many forms such as TJ'ayra, Grisonella, Taxvidea, Meles, and the fisher, Martes pennanti. The width of the posterior narial opening is about as in Mephitis or Conepatus, but with no indication of the partition conspicuous in the mephitines. Between the molar tooth and the posterior narial opening, on each side of the extended portion of the palate, is a small though conspicu- ous process, apparently defining the antero-ventral limit of the origin or attachment of the pterygoid muscles. A similar process is present on the ex- ternal surface of the pterygoid plate, on the antero-dorsal margin of the pterygoid fossa opposite the hammular process. These appear well defined in comparison with living mustelids. Perhaps the most striking character of the skull is the extent to which the bullae project below the basis cranii. The bulla appears well inflated, rela- tively perhaps as much as in Tazidea, with a more transversely compressed portion extending downward, forward and slightly inward. Just below the inflated portion the tympanic exhibits a large foramen directed outward and forward from the medial surface. The osseous portions extending downward behind and in front of this aperture are tightly closed below but not coossi- fied. Apparently this remarkable development on the tympanic is due to a muscular attachment along the ventral margin, possible extending back nearly or entirely to the base of the paroccipital process. In all probability this muscle was the digastricus. Such a marked development at its origin would be entirely in keeping with the unusual development indicated for the masseter and temporalis. The digastricus usually originates at the paroc- cipital process, but according to Windle and Parsons‘ in terrestrial carni- vora it also arises ‘‘often from the contiguous paramastoid and bulla tym- pani.”’ Reighard and Jennings? note that in the cat the digastricus originates 4 WINDLE, B. C. A., and Parsons, F. G. Proc. Zool. Soc. London, pp. 376-377. 1897. 6 REIGHARD J., and JENNINGS, H.S. Anatomy of the Cat, p. 107, New York, 1929. 202 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 26, NO. 5 “from the outer surface of the jugular (paroccipital) process of the occipital bone, and by tendon from the tip of the mastoid process and from the ridge between the mastoid and jugular processes.” This latitude for attachment in carnivora leads one to suppose that in this instance the origin of the digastricus extended well forward on the bulla. Other muscles originating in this region, with forward insertions, such as the styloglossus and stylohyoid, are not eliminated but it seems unlikely that these would be developed to the extent indicated by the process on the bulla. SS Fig. 2.—Craterogale simus, n. gen. and sp. Skull, type specimen, U.S.N.M. no. 13801, dorsal view, natural size. Upper Miocene or lower Pliocene, Dawes County, Nebraska. Drawing by Sydney Prentice. The foramen enclosed by the ventral projection from the tympanic is relatively large. Its direction and position suggest that it may have carried the external carotid artery. After leaving the side of the trachea this artery turns outward, gives off various branches including the internal carotid and occipital arteries, passes deep to the digastricus and between the digastricus and styloglossus. Assuming that the muscle which has its origin on the bulla was the digastricus, the external carotid in maintaining its normal relations would have either passed through the foramen or anterior to the bulla and thence external to the styloglossus and the bones of the hyoid. If the styloglossus muscle in part originated at or near the mastoid process and the*stylohyoid bone was attached in a normal position near the stylo- mastoid foramen then the external carotid in passing through the foramen in question would be in a position to meet these conditions with the least deviation from its normal course. The inferred course of the external carot- id through the foramen which is approximately in its normal path seems likely inasmuch as the position of the artery would have been established in the tissue before ossification had surrounded the vessel. If it could be May 15, 1936 GAZIN: MUSTELID CARNIVORE 203 demonstrated that the carotid ascended external to the bulla then the fora- men might have been occupied by the posteriorly directed internal carotid and occipital arteries. The jugular veins are apparently eliminated as possibilities since the in- ternal jugular passes posteriorly from the foramen lacerum posterius and the external jugular descends superficial to the position in question. Of the nerves which pass forward from the vicinity of the foramen lacerum poste- rius only the ninth and twelfth, the glossopharyngeal and hypoglossal, are not entirely eliminated, but these would apparently descend to the vicinity of the pharynx and then forward to the tongue, and not outward enough to have gone through the foramen in the projected portion of the bulla. The bullae in Craterogale simus are of additional interest in being so com- pletely fused with the adjacent bone elements, apparently a strengthening associated with the muscular attachment to the bullae. Medially the bullae join the basioccipital and basisphenoid forming a nearly smooth, trans- versely concave surface. Postero-externally and posteriorly the prominent mastoid process and incomplete paroccipital process are separated from the bulla by an acute notch, but at the roots of these processes the mastoid and exoccipital bones are firmly joined to the bulla. Anteriorly and antero- externally the bulla is strongly united to the alisphenoid and squamosal. The anterior margin of the bulla is not curved backward dorsally to meet the alisphenoid and squamosal, and does not exhibit a transverse fissure as ob- served in forms having inflated bullae, such as Taxzdea, Gulo, and Martes, but is extended forward, lapping out on the postero-ventral surface of the postglenoid almost or quite to its anterior margin (a condition noted by Matthew’ in Leptarctus). The external audital tube is short due to the inflation of the bulla, and is solidly joined to the adjacent squamosal and periotic bones, quite filling the space between the zygomatic and mastoid processes. The meatus is moder- ately large, nearly circular and directed outward about perpendicular to the median vertical plane of the skull, not forward as in so many of the muste- lids. The glenoid surface is well developed, with a relatively broad postglenoid process as in Mephitis and with the anterior margin of the fossa extended forward uniformly about as in Martes. The outer portion of the anterior mar- gin is prominent but does not recurve to form a locking joint as in Lutra and ulo. As general in mustelids the alisphenoid canal is absent, unless its openings _are entirely concealed in the recesses of the foramen rotundum and foramen ovale, which seems unlikely. The opening of the optic foramen occupies a position farther forward from the foramen lacerum anterius than in Mephitis but relatively not so far forward as in Martes. The foramen lacerum anterius and foramen rotundum are separate but placed very close together. Posterior to these and at the antero-median margin of the bulla the foramen ovale, eustachian foramen, and foramen lacerum medius are very closely grouped together. The foramen ovale opens unusually close to the eustachian fora- men, closer than in Mephitis, and is separated from this opening and the foramen lacerum medius by a thin plate of bone which extends forward as a ridge to form the posterior margin of the pterygoid plate. The small pos- terior opening of the carotid canal is forward of the foramen lacerum pos- terius about one third of the distance to the foramen lacerum medius. Be- 6 MatrHew, W. D. Bull. Amer. Mus. Nat. Hist., 50: 188-146. 1924. 204 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 26, No. 5 cause of the obliterated contact between the bulla and the basioccipital and basisphenoid the elements forming the canal cannot be certainly determined but the position of the canal opening on the bulla suggests that it was prob- ably entirely within the bulla. The small condylar or hypoglossal foramen is situated close to the postero-median margin of the large foramen lacerum posterius. In this respect its position is similar to that in Martes and some of the other genera of mustelids. In specimens of Mephitis examined the condy- lar foramen is much closer to the condyles. The stylomastoid foramen is partially obscured in the deep cleft between the bulla and the ridge connect- ing the paroccipital and mastoid processes. In the case of the postglenoid foramen a somewhat anomalous condition exists in which the postglenoid or temporal canal opens in two positions on both sides. In one position the opening is approximately as in Lutra, Taxidea, and Gula, antero-ventral to Fig. 3.—Craterogale simus, n. gen. and sp. Skull, type specimen, U.S.N.M. no. 13801, ventral view, natural size. Upper Miocene or lower Pliocene, Dawes County, Nebraska. Drawing by Sydney Prentice. the audital tube, but is very close to the postglenoid process. In the second position, antero-dorsal to the audital tube and along the posterior wall of the zygomatic process of the squamosal, the opening is similar to that in Mephi- tis. The two openings though widely separate at the surface presumably join beneath the included segment of the audital tube which has fused to the squamosal. A similar condition was observed in a recent skull of Tazidea taxus. Description of teeth—The type of Craterogale semus exhibits the 3rd and 4th upper premolars and the Ist molar on both sides, and alveoli for the canine and 2nd premolar. In all probability the completed formula was as follows: [3, C+, P3, M3. The combination of cheek teeth is common among mustelids, occurring typically in such genera as Mustela, Mephitis, Spilo- gale, and T'axidea. May 15, 1936 GAZIN: MUSTELID CARNIVORE 205 The teeth, though sturdy, are small considering the muscular develop- ment indicated for the lower jaw. P?, as shown by the alveoli, was two rooted, with the anterior root small and placed inward from the canine alveolus. P? is simple, without accessory cusps or styles, and exhibits a slight cingulum around the medial and posterior margin of the tooth. This tooth is separated by a short diastema from the posterior alveolus of P?. The carnassial is moderately low crowned and in occlusal view is nearly triangular in outline. The shearing blade is relatively broad transversely, and blunt, although the bluntness may be largely due to wear. The deutero- cone (or protocone) is a distinct conical cusp placed well forward at the an- tero-lingual angle of the tooth, much as in Martes and Mustela, farther for- ward than in Mephitis. The lingual margin of the tooth posterior to the deuterocone is almost a straight line, with no appreciable indentation to set off the deuterocone such as observed in Martes. This inner portion of the tooth is without talon or tetartocone, but exhibits a weak cingulum for a short distance posterior to the deuterocone. The anterior margin of the tooth, extending from the deuterocone to a weak parastyle at the antero- external angle of the tooth, is only slightly indented, much less so than in Martes or Mustela. On this portion of the tooth margin the cingulum is conspicuous, and continues around on the outer surface, becoming weaker posteriorly. The molar tooth is relatively large, with the outer portion distinctly long anteroposteriorly and the summits of the paracone and metacone well spaced. The portion of the tooth extending outward from the ridge connect- ing the paracone and metacone projects prominently forward and outward from the paracone, forming a conspicuous style at the antero-buccal angle of the tooth. The lingual portion of the tooth is well developed, but the heel is not so markedly expanded as in Lutra, Taxidea, or Conepatus, though al- most as much as in Mephitis. The fossil molar, however, does not show the median anteroposterior constriction seen in Mephitis and many other mus- telids. The protocone is prominent and rather centrally placed as compared with Mephitis and Lutra. It may be noted that the antero-ventral surface of the ridge connecting the protocone to the paracone is considerably worn. The hypocone is moderately developed but is not expanded so prominently as in Mephitis and Lutra. A moderate cingulum extends forward from the hypocone, around the antero-lingual side of the protocone, but its develop- - ment is not comparable to that in Martes and Mustela. The ridge extending buccally from the posterior extremity of the hypocone is also noticeably worn, presumably through occlusion with Mo. Comparisons with fossil forms.—Of the comparisons which may be made with other fossil mustelids, that with Leptarctus is among the most pertinent. Craterogale stmus shows a striking resemblance in skull structure to Leptarc- tus primus, as illustrated by Matthew,’ but exhibits a distinctly different type of dentition. Points of similarity between the two are seen in the short and broad skull, deep and widely expanded zygomae, prominent postorbital processes, separate and rugged temporal ridges, heavy lambdoidal crests, short basicranial region with large bullae which are markedly fused with the postglenoid processes, very short audital tube, and in the absence of P! and M?. No detailed comparisons with the bullae can be made since these ele- ments are incomplete in the Leptarctus skull. Craterogale simus differs from Leptarctus primus in a narrower postorbital constriction, less widely sepa- 7 MatruHew, W. D. Bull. Amer. Mus. Nat. Hist., 50: 138-146, fig. 37, 1924. 206 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 26, NO. 5 rated temporal ridges (perhaps a matter of age), a less convex profile antero- posteriorly along the temporal ridges and dorsoventrally along the lamb- doidal crests, P* without tetartocone (hypocone), and M' shorter and rela- tively wider transversely. The dental structure in Craterogale is more nearly that of a typical mustelid, whereas that in Leptarctus makes a closer ap- proach to certain of the Melinae. Among the other mustelids known from the Snake Creek beds, Mvonictis makes the nearest approach to the type of lower dentition presumed for Craterogale, and comparisons between the two forms are not entirely satis- factory. The species of Mionicts are of relatively small size, though actually larger than Craterogale simus. The premolars are three in number and do not have accessory cuspules, but the premolar series is relatively longer and the premolars relatively larger and more elongate than would be expected in C. sumus. 'The lower carnassial in species of Mzonzctis has a large talonid, as seems evident for C. semus, but the length of this tooth in the smallest of the two forms, Mionictis elegans, is 11 mm. The length of this tooth in C. simus apparently could not have been greater than about 91/2 mm and was prob- ably about 8 1/2 mm. Phionictis is one of the better known upper Miocene mustelids, having as its type P. ogygia® which is represented by both skull and mandible. The skull of Craterogale semus is quite unlike that of Plionictis ogygia in the de- velopment of the zygomae, and in having rugged temporal ridges. More- over, the upper molar in the Plzonzctis skull is much shortened anteropos- teriorly. This and other characters in both upper and lower dentitions of Plionictis are modifications in the direction of Mustela. Craterogale is distinct from Brachypsalis in having a more reduced denti- tion. The maxillary portions referred by Matthew to species of Brachypsalis show alveoli for a second molar, and the upper carnassials and molars in these specimens belong to much larger species than C. semus. The upper carnassials in the species referred to Brachypsalis are illustrated® as having more medially extended deuterocones, and the first upper molars show a less prominent antero-buccal angle, a stronger hypocone, and a more outstanding cingulum antero-medial to the protocone. The species of Sthenictis are characterized by a less reduced lower premolar series as are species of Martes, unlike the shortened type of jaw which is evident for Craterogale. The lower carnassial in Sthenictis bellus and S. do- lichops from the Sheep Creek and lower Snake Creek horizons respectively, and in the similar Cernictis hesperus’® from the Pliocene of California, has a relatively smaller talonid than is indicated for Craterogale. Plesictis' from the Oligocene of Europe exhibits marked temporal ridges, and the upper carnassial, though relatively longer, resembles that in C'ra- terogale. However, the skull of Plesictzs is more elongate, has weaker zy- gomae, four instead of three premolars, and the bullae, though inflated, do not project ventrally as in Craterogale. M‘ in Plesictis genettoides is rela- tively short and transversely wider, with a somewhat less expanded heel than in Craterogale. Remarks.—The relationships of Craterogale sumus are obscure, due largely 8 MatruHew, W. D. Mem. Amer. Mus. Nat. Hist. 1 shes 383-384, figs. 8-9, 1901. 3 MarrHew, W. OD. Ibid., pp. 131-134, figs. 29-83, 1924. 10) FLAT Lehi em re OUice Mammalogy 16: 137-138. 1935. 1 See H. Hewpine, Abh. schweiz. palaeont. gesells. 50: 1-21, pls. 1-4, 19380; and J. Viret, Ann. Univ. Lyon, n. s. fasc. 47: 166-183, pl. 15: pl. 16, ‘figs. 1-2; pl. 30, figs. 1-4, 1929. May 15, 1936 SWALLEN: NEW GRASSES 207 to the incomplete nature of so much of the fossil mustelid material which has been described from North America. It is possible that Craterogale is related to. Mzonictis, but this can not be demonstrated from the known ma- terial. On the other hand the structural similarity between this skull and that of Leptarctus as far as known seems more than mere coincidence, al- though the teeth are very distinct. However, the differences between the dentitions may not be much greater than between some mustelid genera which have been referred to the same subfamily, such as between Meles, Helictus, Arctonyx, and Tazidea, or as between Lutra, Lataz, and Enhy- driodon. The skull characters exhibited by Craterogale, which possesses a truly mustelid dentition, furnish additional evidence for mustelid affinities of Leptarctus. Moreover, the cranial characters of Leptarctus, as described and figured by Matthew, seem to warrant recognition of a separate subfamily, the Leptarctinae, to which Craterogale simus may be referred tentatively. MEASUREMENTS (IN MILLIMETERS) OF SKULL OF CRATEROGALE SIMUS Length from anterior margin of nasals to occipital condyles (9.2 Length of nasals 1723 Distance from anterior margin of nasals to line between postorbital processes of frontals 27.3 Distance from postorbital processes to posterior margins of lambdoidal crests 49.6 Width between orbits Blas Width at postorbital constriction 15.5 Width across zygomatic arches 61.0 Greatest depth of zygomatic arch 13.28 Width across mastoid processes 41.2 Width across occipital condyles Uae Depth from temporal crests to ventral extremities of bullae 41.3 Distance from posterior margin of palate to foramen magnum 38 .0 Width of palate between molars 14.3 bo ING a Length of cheek tooth series, P2-M! P’, anteroposterior diameter P’, transverse diameter P4, anteroposterior diameter parallel to outer wall P4, greatest length over deuterocone P4, transverse diameter perpendicular to outer wall M!, anteroposterior diameter perpendicular to anterior margin M}, greatest diameter M!, transverse diameter perpendicular to outer margin ® Approximate “TCO Or 01 CO NT CO Or NOON NO oO BOTANY.—Three new grasses from Mexico and Chile Jason R. SwWALLEN, Bureau of Plant Industry. Among the grasses in a recent collection of plants made by Francis W. Pennell in Mexico, were two new species of Muhlenbergia. One of these was found in the Sierra Gazachic, 35 kms. southwest of Minaca, Chihuahua, and the other in the Sierra Madre Occidental, near E] Salto, Durango. The third species here described was col- lected at Cajon de los Pelambres, Dept. Illapel, Chile, by G. Looser. 1 Received February 14, 1936. 208 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 26, NO. 5 Muhlenbergia lucida Swallen, sp. nov. Perennis; culmi caespitosi, erecti, 60 cm alti, pubescentes; vaginae pu- berulae vel scaberulae, internodiis longiores; ligula obtusa, 3-6 mm longa; laminae 15-30 cm longae (superiores 2-5 cm longae), involutae, flexuosae, puberulae vel scaberulae; panicula 15 cm longa, ramis filiformibus, flexuosis, implicatis, basi nudis; pedicelli 10-15 mm longi, filiformes, flexuosi; glumae 3 mm longae, obtusae, hyalinae, pilosae; lemma dense villosum, 4 mm lon- gum, bifidum; arista 12 mm longa, obscure geniculata, basi tortilis; palea villosa lemma aequans. Perennial; culms caespitose, erect, 60 cm tall, appressed-pubescent; sheaths longer than the internodes, minutely scabrous or appressed-pubes- cent, sometimes nearly glabrous; ligule firm, obtuse, 3 to 6 mm long; blades 15 to 30 cm long (the uppermost 2 to 5 cm long extending beyond the base of the panicle), flexuous, firm, involute, minutely pubescent or scabrous; panicle 15 cm long, the branches scabrous or pubescent, filiform, flexuous, implicate, naked at the base, the lowermost 5 to 6 cm long; spikelets distant, the pedicels filiform, flexuous mostly 10 to 15 mm long, gradually enlarged upward toward the summit; glumes equal, 3 mm long, obtuse, hyaline, pilose, nerveless; lemma densely villous, 4mm long, bifid, the teeth 1 mm long, sub- acute, the midnerve prominent, red, extending from between the teeth into an awn 12 mm long, more or less geniculate, twisted below the bend; palea equal to the lemma, hyaline, villous. Type in the U. 8. National Herbarium no. 16143880 collected on ledges of igneous rock, ‘‘Barranca Colorad,’’ Sierra Gazachic, alt. 2300-2500 m, 35 km. southwest of Minaca, Chihuahua, Mexico, September 16-17, 1934, by Francis W. Pennell (no. 18955). Muhlenbergia lucida is probably most closely related to M. argentea Vasey, the only other perennial species which has bilobed lemmas. There are several well marked differences, however, which easily distinguish the latter, the most conspicuous being the narrow panicles with appressed or ascending branches, the short pedicled, appressed spikelets, and the nearly glabrous lemmas. Muhlenbergia subaristata Swallen, sp. nov. Perennis; culmi dense caespitosi, erecti, 120 cm alti, glabri vel minute pubescentes; vaginae internodiis longiores, scabrae; ligula truncata, 1-2 mm longa; laminae 15-80 cm longae, 1-3 mm latae, involutae vel planae, attenuatae, flexuosae, scabrae; panicula 25 cm longa, ramis adscendentibus basi nudis, usque ad 10 cm longis; pedicelli filiformes, flexuosi, 1-5 mm longi; glumae 1-1.3 mm longae, obovatae, obtusae, erosae, scabrae vel pubescentes; lemma 4-4.5 mm longum, scabrum vel pubescens, muticum vel arista usque ad 2 mm longa praeditum; palea acuta, lemmate paulo longior, scabra vel pubescens. Perrenial; culms caespitose, erect, 120 cm tall, glabrous or minutely pubes- cent; sheaths rounded on the back, overlapping, or rarely a little shorter than the internodes, scabrous; ligule truncate, 1-2 mm long; blades loosely involute, becoming flat with age, flexuous, 15-30 cm long, 1-3 mm wide, attenuate to a fine point, scabrous; panicle narrowly pyramidal, 25 cm long, the compound branches, rather distant, ascending, naked at the base, the lowermost 10 cm long; pedicels filiform, flexuous, 1-5 mm long, densely May 15, 1936 COVILLE: GILMANIA 209 pubescent below the spikelets; glumes 1-1.38 mm long, obovate, obtuse, erose, scabrous, pubescent at the tip; lemma 4—4.5 mm long, purple, scabrous or pubescent, awnless or with an awn as much as 2 mm long; palea acute, equaling or slightly exceeding the lemma, scabrous or pubescent. Type in the U. 8. National Herbarium no. 18572 collected along river, above Arroyo de Agua, El Salto, Sierra Madre Occidental, alt. 2600-2650 m, Durango, Mexico, September 1, 1934, by Francis W. Pennell (no. 18572). This species of Muhlenbergia belongs in the group with M. capillaris (Lam.) Trin., M. rigida (H.B.K.) Kunth, and M. reverchoni Vasey & Scribn., but differs from all in the short, usually obovate glumes, and from the first two in the awnless or short awned lemmas Festuca panda Swallen, sp. nov. Perennis; culmi dense caespitosi, 15-40 cm alti, nodo unico geniculati, glabri; folia basi aggregata; vaginae rubrae, glabrae, membranaceae; ligula ciliata 0.3-0.5 mm longa; laminae foliorum ‘basalium firma, arcuatae, 2-5 em longae, infra glabrae supra scabrae; lamina folii culmi 1 cm longa ap- pressa; panicula 3-7 cm longa, ramis brevibus simplicibus appressis vel adscendentibus 1—3-spiculis; spiculae 8-10 mm longae, 4-5 florae, appressae; gluma prima acuta I-nervia, 2.5-3 mm longa, glabra vel marginibus scabe- rula; gluma secunda obtusa, 3-nervia, glabra, marginibus scabris; lemma 6 mm longum, acutum, muticum, glabrum, marginibus scabris; palea acuta, lemma aequans, carinis scabris. Perennial; culms densely caespitose, 15-40 cm tall, somewhat flexuous, usually geniculate at the single node, glabrous; innovations numerous; leaves crowded in a dense basal clump; sheaths reddish, glabrous, becoming membranaceous; ligule short-ciliate 0.2-0.56 mm long; basal blades dis- tichous, firm, stiffly arcuate spreading, 2-5 cm long, glabrous on the lower surface, scabrous on the strongly nerved upper surface; culm blades one, 1 em long, appressed; panicles 3-7 cm long, the short, simple branches appressed or ascending, bearing 1-3 spikelets; spikelets 8-10 mm long, 4-5 flowered, appressed; first glume acute, l-nerved, 2.5-3 mm long, glabrous, or sparsely scabrous on the margins, the second obtuse, 3-nerved, glabrous, scabrous toward the tip and on the margins; lemmas 6 mm long, acute, awnless, scabrous on the tip and margins, otherwise glabrous or nearly so; palea acute, about equaling the lemma, scabrous on the keels. Type in the U. S. National Herbarium, no. 1614378 collected at Cajon de los Pelambres, alt. 2900 m, Dept. Illapel, Chile, January 1932, by G. Looser (no. 2151). The geniculate single noded culms with one short appressed culm blade are characteristic. BOTAN Y.—Gilmania, a new name for Phyllogonum, a very rare genus of plants from Death Valley, California, apparently in process of extinction.! FREDERICK V. CoviLuE, Bureau of Plant Industry. My attention has been called by Mr. C. V. Morton to the existence of a genus of mosses named Phyllogonium, older than the very rare 1 Received March 14, 1936. 210 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 26, NO. 5 genus of flowering plants from Death Valley, California, named by me in 1893 Phyllogonum. Since the latter name therefore is not valid, I propose in place of it a new genus name, as follows: Gilmania Phyllogonum Coville, Contr. U. 8. Nat. Herb. 4: 190. pl. 21. 1893. Not Phyllogonium Brid. Bryol. Univ. 2: 671. 1827. The genus is now renamed in honor of Mr. M. French Gilman, of Ban- ning, California, ornithologist and botanist, who has devoted years of ob- servation and study to the flora of Death Valley and whose intelligent and persistent search for this seemingly lost plant has resulted in its rediscovery. The original description of the genus is reprinted as follows: Polygonacearum genus Eriogoneis affine. Nodi foliis tribus herbaceis petiolatis instructi. Flores lutei, pedicellati, ad nodos fasciculati, sine invo- lucro sine bractea, facie in axilla folii, inserti, fasciculis superis brevitate internodorum adjacentibus, demum confluentibus.—Planta annua, pros- trata, divergente ramosa, omnino luteola, ramis e nodo singulo tribus aut quinque, inaequalibus. Nomen genericum e ¢vAdav et yovu derivatum est, nodi enim folia tria plene expansa, nec ut plerumque in generibus propinquis bracteis reducta, gerunt. The genus has but one species, which is given a binomial name as follows: Gilmania luteola Phyllogonum luteolum Coville, Contr. U. 8. Nat. Herb. 4: 190. pl. 27. 1893. For an illustration and detailed description of the species and a discussion of the relationship of the genus to other genera of the family Polygonaceae, botanists are referred to the original publication. The collections now available show that the plant when fully developed sometimes attains a diameter of more than 10 inches, the irregular rosette of stems lying flat on the ground. This habit and the yellow color of the stems and of the innumerable small flowers have suggested the common name goldcarpet. Even the leaves are yellowish. The specimens collected by Mr. Gilman show that when the plant reaches the fruiting stage the stems stand up from the ground, a position which undoubtedly facilitates the dis- persal of the seeds. The recorded history of Gilmanza is as follows: On April 7, 1891, I found two specimens of an unknown plant in the lower part of Furnace Creek Canyon, Death Valley, California. They were growing in the dry gravelly wash of the canyon. No other specimens were seen. In the Botany of the Death Valley Expedition, published in 1893, these two specimens were made the basis of a new genus and species, Phyllogonum luteolum. On May 18, 1915, Mr. 8. B. Parish collected a few specimens of the plant May 15, 1936 COVILLE: GILMANIA 211 in Death Valley and in 1918, in Notes on Some Southern California Plants,’ he said: Very sparingly scattered among the pebbles covering the dry bed of the stream [Furnace Creek], immediately above the small marsh [Furnace Creek Springs] from which the stream rises, probably the exact spot where Coville, on April 7, 1891, collected the two specimens on which he founded the genus, since which time the plant has not been rediscovered. Two small specimens were also seen in a dry wash between Furnace Creek and Saratoga Springs. So far as known, the species is an endemic of Death Valley, and very rare even there. The plants are prostrate, and the largest found [by Mr. Parish] had stems hardly 3 cm long. In his book, Death Valley, published in 1930, W. A. Chalfant, on page 85, gives the following statement by Professor Willis L. Jepson, who collected plants in Death Valley in 1917: When on a long tramp down the Furnace Creek wash, my first trip down it, watching eagerly for the rare plants of the region, I saw a single individual of Phyllogonum luteolum about two inches high, a species known only from Death Valley. I took it and put it in my press, expecting to find enough of it, and went on down the wash. Within perhaps two hundred or two hundred and fifty yards I suddenly woke up to the fact that I was not finding any more, and so returned to the original spot, locating it with much difficulty. Whereupon I set up my plant press as a marker and carefully cast around the spot at an ever increasing distance for nearly an hour in the hope of finding other individuals, but failed. Nor did I on subsequent days find any more either there or elsewhere. When I was again in Death Valley at the season for this evanescent an- nual, in April, 1931, and in April, 1932, accompanied on both journeys by Mr. M. French Gilman, we both searched very carefully for this plant, but found no trace of it, nor had Mr. Gilman been able to find it on previous journeys. He developed the theory that the few individual plants of this genus that had been collected in Furnace Creek wash had grown from oc- casional seeds that had been brought down the wash from some higher ele- vation which was the real home of the plant. Mr. Gilman was again in Death Valley in 1934 and 1935. In 1934 he looked carefully for the plant but did not find it. In 1935 he renewed his search, and was rewarded by getting it in four different places, all within a few miles of the original locality. The first plant of Gilmania found by Mr. Gilman, on April 14, 1935, was from the floor of Death Valley close to the foot of the mountains, two miles south of Furnace Creek Inn, a few rods south of the entrance to Golden Canyon, and about 100 feet below sea level. There were fourteen individual plants within a radius of 150 feet. Mr. Gilman afterward watched them carefully, but only three became large enough to flower and produce seed. The others dried up and died when they still had only a few leaves and no branches. 2 Bot. Gaz. 65: 336. 1918. 212 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 26, NO. 5 The next locality found by Mr. Gilman, on April 20, 1935, was near Artists Drive, a road in the mountains, entered from the bottom of the valley at a point about six miles south of Furnace Creek Inn. Most of the plants were at an elevation estimated at a little over 1,000 feet, along the crests and slopes of a ‘“‘whitish hill formation’’ which ‘‘seemed rather highly chemical.” Farther down Artists Drive, in a larger canyon leading out to the floor of the valley, Mr. Gilman found three well-developed plants at the edge of a sandy wash at about 400 feet elevation. Altogether, about forty plants were seen in the two localities along Artists Drive. Most of them were very small, and later died from lack of rain. On April 27, 1935, Mr. Gilman, starting from Teck Springs, about half a mile northeast of Furnace Creek Springs, went north of east into some “low yellowish-white hills” and found there about thirty-five plants of Gilmanza, five or six of them of mature size, the others very small. They were scattered over several of the small hills at an altitude of about 1,000 to 1,200 feet. Part of the drainage of this area goes into Furnace Creek, and these hills may well have been the source of the plants that have been found hereto- fore in the dry gravel of that stream bed by Parish, Jepson, and myself. Samples of the soil in which Gilmania plants were growing, obtained by Mr. Gilman, were examined at the Rubidoux Laboratory, Riverside, Cali- fornia, through the courtesy of Mr. C. 8. Scofield. Mr. Scofield has reported on these soils as follows: The three soil samples were tested for total soluble salts by digesting for 24 hours, with shaking, in water, at the ratio of 1 part of soil to 5 parts of water. The solutions thus obtained were analyzed. Sample No. 1, taken near the mouth of Golden Canyon, contained 0.816 per cent of total salts, chiefly calcium sulphate. Sample No. 2, from a yellowish-white ridge on Artists Drive, contained 1.042 per cent of total salts, also chiefly calcium sulphate. Sample No. 3, from the edge of a sandy wash on Artists Drive, at the base of a bank of yellowish-white material, was less saline, containing only 0.167 per cent of salts, chiefly sodium bicarbonate and sodium sulphate. The boron content of all three samples, 0.44, 0.24, and 1.24 parts per million, was so low that this constituent is not to be considered as limiting plant growth in these soils. As compared with agricultural soils, the salinity of samples 1 and 2 would be regarded as high but not too high to prevent the growth of many species of plants. Sample No. 3 would be considered not saline. These analyses do not disclose either the presence or the absence of sub- stances that explain the occurrence of Gilmania on these areas. The seeds of some of the desert annuals germinate so promptly when the ground is moistened that the young plants sometimes blossom and produce seed from a single good rain, before the ground has become dry enough to kill the plants. Occasionally one of these annuals, a Cryptantha, for example, produces seed when it is less than half an inch in height. From Mr. Gilman’s observations in’ 1935 it is his tentative opinion that Gzlmanza can May 15, 1936 DUCKE: MYRISTICACEAE 213 not do this, but that at least two rains, properly timed, are necessary to enable the plant to produce seeds. In the season of 1935 the rainfall in the bottom of Death Valley was unusually heavy. The autumn and winter rains up to January 5, 1935, had totaled 0.89 inch. On January 5 there was a rain- fall of 0.27 inch, on February 4, 0.72 inch, and on March 2, 0.09 inch. Mr. Gilman believes that a few seeds of Gilmania germinated after the rain of February 4 and that the rain in March carried these plants through to ma- turity. Some of the plants reached a diameter of more than 10 inches and produced an abundance of seeds. Most of the plants found by Mr. Gilman in 1935, however, were tiny, consisting of only a small basal rosette of leaves, and they died without producing either branches or flowers. These small plants, Mr. Gilman believes, grew from seeds that germinated after the rain of March 2 and dried up in the rainless period that followed. From these observations we have a fairly clear view of the reason for the rarity with which flowering specimens of Gilmania appear, because good rains, suitably timed, are not common in Death Valley. For example, up to February 6, 1936, there had been only 0.21 inch of rain in the bottom of Death Valley since March 2, 1935. This total of less than a quarter of an inch in nearly a year was made up of 0.10 inch in May, and 0.11 inch in December. Apparently Gilmanza is a plant in process of extinction through the ex- treme dryness of Death Valley. Its seeds, like those of many desert annuals, evidently are able to lie dormant in the ground for several years. Some of them germinate after a good rain if the temperature conditions are suitable for germination, but these germinated seeds do not produce fruiting plants, apparently, unless the seedlings are boosted to a suitable size and vigor by a second rain, adequate in amount and properly timed. In most years any little Galmanza plants that have been able to start will die before they pro- duce seeds, from lack of a second rain. The continued existence of this species apparently depends on the dormancy of a sufficient number of seeds to carry it over unfavorable years to years of adequate and properly timed double rains. If Death Valley becomes drier and drier, and years with suitable double rains become more and more infrequent, the vitality of the old Gil- mania seeds in the soil will ultimately be insufficient to span these longer periods of years when no new seeds are produced, and extinction, which is now a menace, will become a fact. | BOTANY.—WNotes on the Myristicaceae of Amazonian Brazil, with descriptions of new species.: I. ApouPpHO Ducks, Jardim Botan- ico, Rio de Janeiro. (Communicated by E. P. K1u.1p.) During my botanical trips in Amazonia I assembled a good number of plants of that interesting but not sufficiently studied family, ! Received February 15, 1936. 214 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 26, NO. 5 Myristicaceae, which must be considered one of the most important elements of the hylaea-flora, principally in the western half of this immense plain. The types of the new species are preserved in the herbarium of the Jardim Botanico of Rio de Janeiro; cotypes have been distributed among the principal institutions of Europe and America. Of these last, special mention may be made of the U.S. National Herbarium in Washington. The material sent to that her- barium was compared with the Myristicaceae in the herbarium of the New York Botanical Garden by Dr. A. C. Smith, for which I thank him very kindly. Wood samples accompanied by herbarium material of several species have been deposited at the Yale School of Forestry. A part of the material examined was gathered by me when I was in the service of the late Dr. J. Huber, the eminent botanist, Director of the Para Museum (Museu Goeldi); duplicates of that material are deposited in the Jardim Botanico, Rio de Janeiro. COMPSONEURA COMPSONEURA DEBILIS (A. DC.) Warb. ‘‘Catinga’’ of Camanaos, Upper Rio Negro, State of Amazonas (Ducke, Herb. Jard. Bot. Rio 24445, distrib- uted as a new species. This very small tree with pure white flowers is more like certain species of Casearza (Flacourtiaceae) than a Myristicacea. Adult leaves thick-coriaceous; ripe fruits orange-red; arilloid entire, purplish; seeds spotted, as in C. sprucei and C. ulez. | CoMPSONEURA RACEMOSA Ducke. Sao Paulo de Olivenga (Rio SolimGes) ; observed once (male tree). COMPSONEURA CAPITELLATA (A. DC.) Warb. To this species, described from Eastern Peru, probably should be referred a small tree of the upland forest of SA0 Paulo de Olivenga, a Brazilian village near the Peruvian frontier (Herb. Jard. Bot. Rio 19803 and 23693, male trees, and 19576, female tree with flowers and fruits). The fruits, ellipsoid or less frequently nearly globose are up to 5 cm long and 3 em thick, though not yet mature. They are green, glabrous, with a longitudinal keel at one side; their hard, ligneous pericarp is 3mm thick when dry; the seed is not yet developed. These fruits are surely glabrous, with a longitudinal keel at one side; their hard, ligneous pericarp is 3mm thick when dry; the seed is not yet developed. These fruits are surely very different from those of all other known Myristicaceae. COMPSONEURA ULEI Warb. This species must be placed, by its androe- cium, in the section Coniostele Warb.; its fruit resembles externally and in- ternally that of C. debzlzs but is a little larger and of a pale yellow color, and its arilloid is white, instead of red as in all other Myristicaceae of which I have seen fruits. This very small tree, with long branches and yellowish May 15, 1936 DUCKE: MYRISTICACEAE 215 green flowers, is not very rare in the undergrowth of the upland forests of the Middle Amazon (western part of the State of Para: Santarem, Lower and Middle Tapajoz, Obidos, Lower Trombetas; eastern half of the State of Amazonas: Maués, Manaos, Porto Velho; northwest of Matto Grosso: Santa Cruz of the Rio Jamary). OsTEOPHLOEUM OSTEOPHLOEUM PLATYSPERMUM (A. DC.) Warb. One of the most fre- quent and widely distributed Myristicaceae of the upland rain forest of Amazonian Brazil, from the mouths of the Amazon and the neighborhood of the capital of Para to Sao Paulo de Olivenga, not far from the Peruvian frontiers. It furnishes here, for this family, the largest trees, these some- times more than 40 meters high, but only in the thickest stems do we find a very thin red heartwood. IRYANTHERA Warb. The species of this very natural genus are more difficult to classify than those of Virola, being nearly as numerous but much more uniform in their characters. Indument is always scarce, the leaves and the adult fruits being glabrous; the structure of the androecium is less variable than in Virola. Probably the fruits furnish the best characters to establish a natural arrange- ment of the species, but unfortunately most of these are only known in the male form. This genus is apparently restricted to the Amazonian hylaea (including the Guianas and the northwestern part of the State of Maranh4o), where it is represented by a rather considerable number of species, though much less abundant in individuals than is Vzrola; it is one of the most char- acteristic elements of the hylaea-flora. All the species grow in upland virgin forest, where they prefer the neighborhood of small streamlets. All are known by the vernacular name ‘ucuhtiba-rana”’ (false ucuhtiba), those which furnish wood of good quality also as “‘pundn.”’ Iryanthera dialyandra Ducke n. sp. Speciei J. ulez Warb. primo ad- spectu valde similis, at foliorum basi anguste rotundata vel subcordata, inflorescentiis, saepius e ramulorum parte inferiore ad folia delapsa, elon- gatis, usque 70 mm longis; ab ipsa omnibusque aliis speciebus hucusque notis differt antheris 3 ad columnae apicem sessilibus e basi liberis divergentibus. Planta feminea ignota. Arbor parva, partibus vegetativis glabra; foliorum petiolus brevis vel ad 18 mm longus, lamina vulgo ad 22 cm rarius 27 cm longa et ad 7-9 rarius 11.5 cm lata, saepius subobovato-elliptica colore et nervatione ut in specie citata; inflorescentiae tenuiter ferrugineo-puberulae vix ramosae, rhachi robusta; flores virescentes, e nodulis (vel ramulis mini- mis) fasciculati, pedicellis 3-6 mm longis, perigonio basi bracteolato 1.5-3 mm longo in alabastro plus minus obovato, anthesi tripartito, androecei columna gracili quam antherae multo longiore. Habitat circa Mandos silva non inundabili ad ripas paludosas rivulorum, locis Estrada da Raiz inter Cachoeirinha et rivum Mindt (Herb. Jard. Bot. Rio 19578) et Estrada do Aleixo (Herb. Jard. Bot. Rio 24446), mensibus julio et augusto florifera, leg. A. Ducke. 216 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 26, No. 5 The free anthers give to this species an isolated position within the genus. Many species are similar in leaves and sometimes also in the form of the in- florescences, but their anthers are entirely connate with the column of the androecium. IRYANTHERA ULEI Warb. (J. macrophylla Warb., male, not female.) I. macrophylla Warb. (1897) is a mixture of Myristica macrophylla Spruce ex Benth. (a plant which, according to the descriptions, would be remark- able by its very large leaves but which is known only in female herbarium material with anomalous fruits) and of Schwacke 595 (=3532), a male plant of which I examined herbarium specimens in the Museu Nacional of Rio de Janeiro. A cotype of I. ulez (Rio Jurua-Miry, Acre Territory, Ule 5724) cor- responds exactly with Schwacke’s plant, collected at Manaos. I found, near this city, a third individual evidently conspecific with both the last men- tioned plants. It was a small upland forest tree, growing at the swampy margins of a streamlet near the Estrada do Aleixo (Herb. Jard. Bot. Rio 24456). The leaves of J. ulez reach rarely 25 cm in length, their average size being only 20 cm; they are thick but very fragile. The inflorescences are remark- ably short; the column of the androecium, though variable in length, always exceeds the length of the anthers. This species may be mistaken for many others, as for instance I. dialy- anthera, but the elongate inflorescences and the form of the androecium of the latter are distinctive. Iryanthera polyneura Ducke, n. sp. Speciebus J. ulez, dralyandra, longiflora, paraensis et lancifolia affinis, ab omnibus differt foliorum costis secundariis utrinque 28-32, inflorescentiis masculis (solis notis) 15-25 mm longis, simplicibus, pedunculo (brevissimo vel ad 6 mm longo) et rhachi crassis, florum fasciculis multifloris in inflorescentia dense agglomeratis, dense rufo-velutinis. Arbor mediocris ramulis superne tenuiter cano-tomen- tellis, foliis adultis glabris petiolo 15-20 mm longo, lamina 20-29 cm longa et 6-11 cm lata, basi obtusa vel anguste rotundata, apice vulgo abrupte acute acuminata, basi et apice saepe complicata, crasse coriacea fragili, utrinque granulosa, costis secundariis supra impressis subtus prominentibus, ante marginem arcuato-conjunctis, venulis non conspicuis. Flores viridi- ferruginei, ante anthesin subglobosi vix ultra 1 mm diametro, anthesi pro- funde trilobi intus glabri, androecei columna brevi quam antherae (6, perfecte adnatae, breves) vix longiore. Habitat silva non inundabili circa Fontebéa (Rio Solimées, civ. Ama- zonas), 4-9-1929 leg. A. Ducke (Herb. Jard. Bot. Rio 24454). Very similar, at the first glance, to many other species with thick but very fragile leaves; it differs, however, from all others by the more numerous lateral ribs of the leaves. The very short and dense inflorescences may be anomalous; in J. lancifolia we find similar inflorescences beyond the more numerous inflorescences of elongate form which evidently represent the nor- mal type. I. ulec Warb. has very short and dense male inflorescences, but its leaves have only 14 to 18 lateral ribs. May 15, 1936 DUCKE: MYRISTICACEAE 217 Iryanthera longiflora Ducke, n. sp. Speciei J. ule: partibus vegetativis simillima, solum ramulorum tomento diutius persistente. Inflorescentiae masculae (solae notae) in ramulorum parte inferiore infra folia numerosae, binae, 6-12 cm longae, sat flexuosae, super basin pauciramosae vel (praeter nodulos floriferos interdum longe pedunculatos) simplices, partibus omnibus tenuiter rufo-ferrugineo-puberulis, nodulis vel ramulis floriferis modice dis- tantibus multifloris; flores brunnescenti-virides, pedicello 5-8 mm longo tenui, bracteola parva pilosa, perianthio ante anthesin obovato-oblongo ad 3 mm longo vix 2 mm lato, anthesi oblongo-ureolato vix ad 1/3 ab apice trilobo, intus glabro, androecei columna cylindrica tenui quam antherae perfecte adnatae multo longiore. Arbor mediocris ramulis crassis superne cano-tomentellis, foliorum glabrorum petiolo 1-2 cm longo, lamina 20—30 cm longa et 5.5—9.5 cm lata, oblongo-elliptica vel lanceolato-oblonga, basi ob- tusa vel acutiuscula, apice acuta vel brevissime acuminata, crasse coriacea fragili, siccitate supre fusca subtus rufo-ferruginea, supra parum subtus vix nitida, utrinque (subtus fortius) granulosa, costis secundariis utrinque 15-20, longe ante marginem arcuato-conjunctis, supra impressis subtus fortiter prominentibus, venulis non conspicuis. Habitat silva terris altis secus flumen Purtis inter Boca do Acre et Monte- verde, civitate Amazonas, 10-3-1933 leg. A. Ducke (Herb. Jard Bot. Rio 24457). The leaves are like those of J. wlez, but the inflorescences and their in- sertion, as well as the flowers, are very different. In comparison with the other species with large and thick but very fragile leaves, I. polyneura has the nervures of the leaves much more numerous and the inflorescences very short and dense; I. dialyandra has the anthers not connate, divergent; I. paraensis has very elongate and thin inflorescences and very small flowers; I. lancifolia has the nervures of the leaves very faint; and all these species have smaller and more or less globose flowers. Finally, in I. densiflora the flowers have a somewhat elongate form, but they are shorter than in the others and the inflorescences are simple and dense; the leaves of this species are, however, very different. Iryanthera lancifolia Ducke, n. sp. Affinis speciebus aliis foliis magnis crassis coriaceis fragilibus fultis; differt ab J. dialyandra antheris 6 perfecte adnatis; ab J. paraensis inflorescentiis multum minus elongatis, robusti- oribus; ab J. longiflora inflorescentiis minus elongatis, non ramosis, floribus parvis; ab J. wlez inflorescentiis sat longis et floribus minoribus; ab I. poly- neura costis secundariis foliorum multum minus numerosis et inflorescentiis sat longis; a speciebus citatis omnibus foliis magis lanceolatis costis secun- dariis in utraque pagina tenuissimis. Arbor 20-25 m alta, praeter innova- tiones glabra; foliorum petiolus 10-18 mm longus robustus canaliculatus, lamina 16-24 cm longa et 4-7 cm lata, saepissime oblongo-lanceolata vel lanceolata, basi obtusa vel subacuta, apice acuta vel breviter acuminata, crasse coriacea fragilis utrinque granulosa, supra fuscescens nitida, subtus ferruginescens subopaca, costa primaria valida subtus prominente, costis secundariis utrinque 14-18 supra tenuissime immersis subtus tenuissime prominulis. Inflorescentiae masculae fere omnes infra folia e ramuli parte vetustiore, binae, vulgo 3—7 cm longae rarius breves (abnormes?), simplices, nodulis floriferis vulgo sat approximatis, rhachi crassiuscula tenuiter puber- 218 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 26, NO. 5 ula; flores viridi-ferruginei, in nodulo plures fasciculati, cum pedicello (brevi vel usque 5 mm longo, tenui) et bracteola (parva) dense rufo-velutini, perianthio ante anthesin subgloboso vix ultra 1 mm in diametro, anthesi profunde trilobo intus glabro, androecei columna brevi quam antherae 6 parvae perfecte adnatae parum longiore. Arbor feminea ignota. Habitat prope Mandos silva non inundabili inter Estrada da Raiz et Igarapé Mindi, 138-7-1932 leg. A. Ducke (Herb. Jard. Bot. Rio 24553). IRYANTHERA PARAENSIS Huber (J. elongata Huber). Nos. 429 and 542, male and female, of Dr. Sandwith’s British Guiana collection are not at all different from the plants of Brazilian Amazonia. This species is distributed in the hylaea from the vicinity of the Atlantic as far as the Upper Amazon (Fonteb6a and Tonantins). Kuhlmann found it in northwestern Matto Grosso between Pimenta Bueno and Riozinho (Herb. Comm. Rondon 1974). It is distinguished from I. hostmannii by the frequently larger leaves, the very elongate thin and flexuous male inflorescences, the very small flowers, and a very short androecium column. IRYANTHERA HOSTMANNII (Benth.) Warb. This species from the Guianas and the Upper Rio Negro of Venezuela has not been found in Brazil. I have seen a fruiting specimen in the Museu Nacional (Maroni, French Guiana, Melinon 1861, without number); its leaves correspond exactly with the drawing in Warburg’s monograph, plate 4; one of its two fruits resembles the fruit of the same plate, but the other has an almost triquetrous form, reminding one of the fruit of I. tricornis. IRYANTHERA DENSIFLORA Huber. Readily recognized by its simple and dense inflorescences with relatively large flowers. A tree of the very moist upland forest from the Amazon estuary (islands of Breves and Gurup4), the region of the railway between Belem do Parad and Braganga (Peixeboi), and the Middle Tapajoz (S. Luiz, at the foot of the lowest rapids). Iryanthera coriacea Ducke, n. sp. Arbor parva, praeter innovationes tenuiter rufo-tomentellas glabra. Foliorum distichorum petiolus 12-17 mm longus robustus profunde canaliculatus; lamina 13-20 cm longa 4.5—-7 cm lata, obovato-lanceolato-oblonga vel -elliptica, basi anguste rotundato-ob- tusa complicata, apice sensim vel subabrupte modice longe acuminata, | crasse et firme coriacea, utrinque nitida, subtus aliquanto pallidior et granu- loso-rugulosa, costa mediana utrinque prominente subtus basi crassa, costis lateralibus utrinque 15-18 ante marginem arcuato-anastomosantibus supra fortiter impressis subtus tenuiter prominulis vel subimpressiusculis, venulis reticulatis impressis supra bene conspicuis. Inflorescentiae masculae super petiolorum insertiones vulgo binae, usque ad 8 cm longae, simplices, sub- strictae vel arcuatae rarius subflexuosae, cum floribus tenuiter rufo-sericeae; flores e nodulis vel ramulis brevissimis fasciculati, in fasciculo vulgo plurimi, pedicellis 2-4 mm longis tenuibus, perianthio basi bracteola squamiformi munito, ante anthesin subtriquetro-ovato vel subgloboso diametro ad 1.5 mm, anthesi aperto 3 mm lato usque ad medium obtuse trilobo, intus glabro, androecei columna robusta superne dilatata, antheris 6 perfecte adnatis plus minus aequilonga. Planta feminea ignota. May 15, 1936 DUCKE: MYRISTICACEAE 219 Habitat cirea Mandos loco Estrada do Aleixo silva humosa humida non inundabili, 16—5—1933 leg. A. Ducke (Herb. Jard. Bot. Rio 24451). Remarkable by its thick and hard but resistant leaves (not fragile as those of I. ulei, dialyandra, paraensis etc.), with the lateral ribs deeply immersed on the upper side and weakly prominent beneath. The inflorescences are rather elongate, simple, with very small flowers. Iryanthera elliptica Ducke, n. sp. Arbor 30-metralis trunco robusto eylindrico, praeter innovationes rufo-sericeas glabra. Foliorum petiolus 8-14 mm longus robustus canaliculatus; lamina 7.5-12.5 cm longa et 4-5.5 em lata, plus minus elliptica rarius oblongo- vel obovato-elliptica, basi rotundata vel obtusa complicata, apice abrupte breviter vel modice longe acuminata, sat crasse et dure coriacea, utrinque nitidula, subtus granulosa et pallidior vel ferruginescens, costa mediana subtus mediocriter crassa, costis secundariis in utroque latere tenuissime impressis plus minus obso- letis, venulis inconspicuis. Inflorescentiae masculae praesertim in ramulo- rum parte vetustiore supra axillas foliorum jam delapsorum binae, usque ad 8 em longae, vulgo parum vel modice flexuosae, sat dense rufo-sericeo- puberulae, simplices vel supra basin ramosae; flores flavido-virides e nodulis vel ramulis brevibus modice distantibus fasciculati, sat numerosi in fasciculo, pedicellis ad 3 mm longis tenuibus, bracteola dense rufo-pilosula, perianthio ante anthesin subtriquetro-obovato vix ad 1.5 mm longo demum profunde trilobo extus parce puberulo intus glabro, androecei columna cylindrica quam antherae 6 perfecte adnatae multo longiore. Planta feminea ignota. Habitat prope Manaos in silvis terris altis ultra Flores, 1-7-1932, leg. A. Ducke (Herb. Jard. Bot. Rio 24450). This species resembles at the first glance J. sagotzana but may be recog- nized by its coriaceous leaves, which are nearly as thick and as hard as those of I. corvacea (through a little more fragile) and are of elliptic form and very much smaller. It is distinguished from both by the much longer androecium column. IRYANTHERA JURUENSIS Warb. I have in hand a great number of her- barium specimens, among them a cotype (Ule 5460). From these I cannot distinguish no. 7103 of the Herb. Amaz. Mus. Para, which is the type of I. grandiflora Huber for the staminate flowers. Huber’s species would seem to differ from I. juruensis chiefly by its longer flowers, but the same twig has numerous smaller (normal) flowers, which do not exceed the dimensions of true J. juruensis. The column of the androecium in this species is at least as long as the anthers; in this particular Huber’s diagnosis of I. grandiflora is incorrect. The fruit is cylindric with rounded extremities, transversal in relation to the peduncle, and is slightly larger than that of J. paraensis. The species J. juruensis is spread through the upland forests of the Middle and Upper Amazon, from Santarem, Obidos, and the middle courses of Tapajoz and Trombetas to the Lower Japura and the Upper Jurua (Acre Territory). Krukoff collected it at the headwaters of the Rio Machado (northwestern Matto Grosso). To this species belong nos. 8553 and 8749 of the Herb. Amaz. Mus. Para (Faro and Serra de Parintins) and no. 23682 of the Herb. 220 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 26, NO. 5 Jard. Bot. Rio (Lower Madeira), all distributed under the name I. sagotiana by Huber and by myself. Iryanthera grandis Ducke, n. sp. Arbor ultra 30 m alta, trunco cylin- drico robusto, ligno interiore amplo rufo-brunneo bono, glabra, innovationi- bus rufo-tomentellis. Foliorum petiolus usque 2 cm longus supra canalicu- latus, lamina 13-25 cm longa, 5-7.5 cm lata, obovato- vel rarius sublanceo- lato-oblonga, basi anguste rotundata vel breviter obtusa saepe complicata, rarius subcordata, apice brevissime et vulgo obtuse acuminata, adulta sub- coriacea fragilis, supra nitidula, subtus subopaca et parum pallidior quam supra, costa mediana subtus crassa, costis lateralibus utrinque 16—20 ante marginem plus minus evanescentibus, subtus tenuiter prominulis, venulis nullis vel in utraque pagina tenuissime immersis. Inflorescentiae masculae super petiolorum insertiones saepius binae, vulgo 5-12 cm longae, rhachi apicem versus attenuata in vivis molli et flexuosa, tenuiter rufo-ferrugineo- puberula; flores circa apices nodulorum vel ramulorum brevium modice distantum fasciculati, cum pedicello (ad 2 mm longo, tenui) et bracteola (parva, sub ipso flore inserta) densius rufo-ferrugineo-sericei, perianthio intus glabro, in alabastro subtriangulari-globoso, anthesi aperto vix ultra 1 mm diametro late campanulato profunde tripartito, androecei columna brevissima infra angustata, antheris vulgo 6 albidis adnatis subaequali vel parum breviore. Inflorescentiae femineae fructiferae solae visae, e trunci nodis vix ultra 3 cm longae; fructus maturus breviter crasse stipitatus, circa 4 cm altus et crassus, circa 5 cm latus, demum in valvas duas semiglobosas dehiscens, pericarpio lignoso durissimo circa 1 cm crasso, glabro, arillodio purpureo integro, semine 3.5—-4 cm lato et fere 2 cm alto et crasso, testa laxe sulcato-reticulata, albumine non ruminato. Habitat circa lacum José-Asst: prope Parintins (civ. Amazonas), silva non inundabili, arbor mascula, 15-9-1932 leg. A. Ducke cum ligno no. 151, (Herb. Jard. Bot. Rio 24447); in silvis collinis prope cataractas Mangabal medii fluminis Tapajoz (civ. Para), arbor feminea fructifera, 10-12-1919 leg. Ducke (Herb. Jard. Bot. Rio 2896). ‘‘Ucuhtiba-rana” vel ‘‘punan’’ appellatur. This species is allied to J. tricornis Ducke, which furnishes also a very valuable wood. The leaves, however, are much larger, glabrous, with more numerous lateral ribs; the male inflorescences are shorter and stouter, with less distant floriferous nodes; the fruit has a very different form, being re- markable for its size and its ligneous, exceedingly thick pericarp. IRYANTHERA TRICORNIS Ducke, Tropical Woods No. 31, 11. 1932; Archiv. Jard. Bot. Rio 6: 9. 1933. This species, the ‘‘pundn”’ of the Solimoées river (Fontebéa, 8. Paulo de Olivenga) is a tree 18 to 30 meters high, with cylin- dric stem which furnishes a highly esteemed reddish brown, afterwards dark brown heartwood. It grows in the humid upland forest. It can be recognized by the medium-sized obovate leaves, the thin and flexuous male inflores- cences, and the three-horned fruits, the form of which is more exaggerated than that of the fruits of I. hostmanni. IRYANTHERA SAGOTIANA (Benth.) Warb. A small or scarcely medium- sized tree, frequent in the upland forests from the environs of Belem do May 15, 1936 DUCKE!: MYRISTICACEAE 221 Parad to Santa Izabel at the Braganca railway, but not yet observed else- where in Amazonia. Nos. 8553 and 8749 of the Herb. Amaz. Mus. Para (from Faro and Serra de Parintins), as well as Herb. Jard. Bot. Rio 23682 (of the Lower Madeira), are evidently I. juruensis Warb. The present spe- cies seems to be limited to that part of the hylaea near the Atlantic. IRYANTHERA LAEVIS Mef. A tree of medium size, collected by myself in the upland forests in the State of Amazonas near Mandaos (Herb. Jard. Bot. Rio 24459), Fonteb6a (Herb. Jard. Bot. Rio 24458), and Parintins (Lake Uaicurapa, Herb. Jard. Bot. Rio 24455). This species was previously collected in Amazonian Peru by Tessmann (type) and by Killip and Smith. Our herbarium specimens were distributed as a new species, but Dr. Smith has established their identity with J. laevis, an insufficiently described spe- cies of which he has seen a typical specimen. The leaves of I. laevis resemble in color and nearly also in texture those of I. sagotiana, but in many other characters this species is more like J. juruen- sis, from which it differs, however, by the very insignificant nervures of the leaves and by the ample male inflorescences. The flowers though variable in size, are among the largest in this genus. A very good character is the abundance of the pale lenticels, which are never completely absent on the young branchlets and sometimes are present even on the rachis of the in- florescences; I never observed them in any other species of this genus. Iryanthera obovata Ducke, n. sp. Arbor mediocris, partibus vegetativis praeter innovationes tenuiter puberulas omnino glabris. Foliorum petiolus 1-1.5 em longus validus supra canaliculatus; lamina 8-11 cm longa et 3.5-6 cm lata, obovata vel oblongo-obovata, basi obtusa vel subacuta, apice vulgo late obtusa rarius rotundata vel brevissime obtuse acuminata, sub- elastice coriacea, margine subtus lineiformi saepe revoluto, concolor et vix nitidula (adulta subglauca), praesertim subtus dense granulosa, costa mediana utrinque prominente subtus basi crassa, costis lateralibus nullis vel supra tenuissime et obsolete impressis, venulis nullis. Inflorescentiae masculae (solae notae) super axillas foliorum (interdum delapsorum) binae vel solitariae, breves vel saepius usque ad 8 cm longae, simplices vel super basin pauciramosae, graciles, tenuiter ferrugineo-puberulae, florum fascicu- lis e ramulo brevissimo vel nodulo; flores in fasciculo bini vel pauci, vires- centes, pedicellis 5-9 mm longis tenuissimis, perianthio basi bractea squami- formi breviter et longius pilosula, in alabastro subgloboso-ellipsoideo, an- thesi circa 2.5 mm diametro, ad medium obtuse trilobo, extus appresse puberulo intus glabro, columna androecei e basi dilatata anguste cylindrica, antheris 6 omnino adnatis longiore. Habitat in silvis ‘‘catinga’”’ in regione Rio Negro superioris (civit. Ama- zonas) prope Camandaos (Herb. Jard. Bot. Rio 24452) et circa affluentem Curicuriary (Herb. Jard. Bot. Rio 24462), Octobre florens, leg. A. Ducke. The present species can be recognized by its obovate and coriaceous leaves, nearly destitute of nervures, and by its rather elongate but few dense in- florescences with long and very thin pedicels. It resembles, at first glance, Osteophloeum platyspermum, but the leaves are shorter, more coriaceous, and 222 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 26, NO. 5 without nerves, and the inflorescences are different; these, in our new species, resemble a little the inflorescences of Compsoneura ulet. IRYANTHERA PARADOXA (Schwacke) Warb. This species seems to have been found only once (Schwacke 3736=575, Museu Nacional Rio de Ja- neiro). It must be rare, because I have never observed it near Mandaos, the type locality, where I collected no less than six of its congeners. It should be easily recognized from the diagnosis and illustrations in Warburg’s monograph. @Obituary ANDREW NELSON CAUDELL, entomologist of the Bureau of Entomology and Plant Quarantine and custodian of Orthoptera at the U. 8. National Museum, died at his home, 605 Keefer Place, Washington, D.C., March 1, 1936. He was born August 18, 1872, at Indianapolis, Indiana. After receiving the Bachelor of Science degree at Oklahoma Agricultural and Mechanical College in 1897, and a year’s study at Massachusetts Agricultural College, he joined the Bureau of Entomology of the U.S. Department of Agriculture in 1898. Mr. Caudell’s scientific work was concerned chiefly with the natural history and classification of the Orthoptera about which he wrote numerous papers, one of the last (unpublished) being in joint authorship on the Orthoptera of the District of Columbia and vicinity. His hobby was philately to which he also contributed important papers. Mr. Caudell was a member of the American Association for the Advance- ment of Science, Association of Economic Entomologists, Entomological Society of Washington, Washington Academy of Sciences, and the American Philatelic Society. CONTENTS CosmoLocy.—The physical universe. HERBERT DINGLE........... GropuHysics.—Some aspects of geophysical cycles. J. BARTELS...... PaLEONTOLOGY.—A new mustelid carnivore from the Neocene beds of northwestern Nebraska. C. Lewis GAZIN...............-..+-- 199 ce Borany.—Three new grasses from Mexico and Chile. Jason R. _ PWALUERR AS bce ss Baas bua Soace eigen 1b ool ata ent onan Borany.—Gilmania, a new name for Phyllogonum, a very rare genus _ of plants from Death Valley, California, apparently in process of _ extinction. FREepERIcK V. CovILLE...... sees cee peste see setae Botany.—Notes on the Myristicaceae of Amazonian Brazil, with de- scriptions of new species. I. ApoLtpHo DucKE............... OBITUARY: ANDREW N. CAUDELL........... Fee apc ee yee ES > ar This Journal is indexed in the International Index to Periodicals - JUNE 15, 1936 No. 6 JOURNAL “2-K.+ TASHINGTON ACADEMY OF SCIENCES ee BOARD OF EDITORS _ F.G. Brickweppe Rozanp W. Brown Esgn H. Toor e 10 OF STANDARDS U. 8. GHOLOGICAL SURVEY BURHAU OF PLANT INDUSTRY ASSOCIATE EDITORS H. T. WensEL HarotD MorRIsoNn PHILOSOPHICAL SOCINTY ENTOMOLOGICAL SOCIHTY E. A. GOLDMAN | W: W. Russy BIOLOGICAL SOCINTY GEOLOGICAL SOCIETY - Aanrs CHASE J. R. Swanton BOTANICAL SOCIDTY ANTHROPOLOGICAL SOCIETY E. E. Gipson CHEMICAL SOCIETY PUBLISHED MONTHLY BY THE WASHINGTON ACADEMY OF SCIENCES 450 Aunaip St. ‘ut Menasua, WISCONSIN oe Entered as second class matter under the Act of August 24, 1912, at Menasha, Wis. eae Bpoeptanee for mailing at a special rate of postage provided for in the Act of February 28, 1925. ie vo lee . Authorized January 21, 1933. This JouRNAL, the official organ of the Washington Abbie of Bocce ub 1es (1) short original papers, written or communicated by members of the Academy 2 proceedings and programs of meetings of the Academy and affiliated societies; (3. notes of events connected with the saat ee fie, of pease: The J OURNAL is issue Manuscripts may be sent to any member of the Board of Edie iti is ae res quested that contributors consult the latest numbers of the JourNat and conform their manuscripts to the usage found there as regards arrangement of title, subheads, syn- — onymies, footnotes, tables, bibliography, legends for illustrations, and other matter Manuscripts should be typewritten, double-spaced, on good perce Footnotes shoul be numbered serially in pencil and submitted on a separate sheet. The editors do not assume responsibility for the ideas expressed by the author, nor can they nndartalee te: correct other than obvious minor errors. ey Illustrations in excess of one full-page halftone or two full-page line drawings m may be paid for by the author. Line drawings are preferred. aes ae Proof.—In order to facilitate prompt publication one proof will Gres be ee to authors in or near Washington. It is urged that manuscript be submitted in final form; the editors will exercise due care in seeing that copy is followed. ae ae 1 Author’s Reprints.—Fifty reprints without covers will %e furnished gratis. Cover: bearing the name of the author and title of the article, with inclusive pagination an date of issue, and reprints, will be furnished at cost when ordered, in Bovordeteay the following schedule of prices. aT , Copies 4 pp. 8 pp. P42 pp. =: eS pps 50 Resa eeeue , + emi ob cata Tene 100 $ .75 $1.20 $1.50. $2.00 150 1.25 1.90 2.55 3.25 200 175 2.60 3.60 4.50 250 2.00 - 8.30 4.65 ~~ 5.75 Envelopes for mailing reprints with the author’s name and address printed it corner may be obtained at the following prices. First 100, $4.00; additional 100, Sad . Subscription Rates.—Per volume......... oh sean ee To members of eye & pestaces 7 per volume. . Loe Single num bers7 3. oars sae Steerer es 960 Remittances should be made payable to ‘‘Washington Academy of Sciences” and addressed to 450 Ahnaip Street, Menasha, a or to the al H. G. Aver Coast and Geodetic Bete A Washington, D. Exchanges.—The JouRNAL does not exchange with other publications. Missing N umbers will be replaced without charge provided that claim is s made to . the Treasurer within thirty days after date of following issue. ; OFFICERS OF THE ACADEMY President: O. E. Mreinzer, U. 8. Geological Survey. a Corresponding Secretary: N. R. Smiru, Bureau of Plant aeciee. Recording Secretary: CHARLES THoM, Bureau of Chemistry and Koil ae Treasurer: Henry G. AVERS, Coast and Geogeue jab fe slice es ae Peal JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES Vou. 26 JUNE 15, 1936 No. 6 GENERAL SCIENCE.—The new Aristotle... FREDERICK Barry, Columbia University. (Communicated by C. E. CHAMBLISS.) To those students of the history of science whose interests are—so to say—more scientific than historical, whose experience in the pro- duction of new knowledge has made particularly vivid the realization that discovery is far less frequently the result of happy accident than the consequence of systematically methodical investigation, and not so often the reward of mere industry as the fulfillment of theoretical anticipations—in a word, to ourselves, the most significant of all past scientific labors are those which have contributed either to the deter- mination of effective ways and means in research or to the develop- ment of that conceptual representation of nature which enables us in some degree to understand the complicated interrelations of diverse phenomena, to organize our knowledge as it grows, and to discern the most promising pathways of further investigation. It is common knowledge that we owe to the ancient Greeks, uniquely, not only the first inculcation of the scientific temper among thoughtful men and the first studies of nature that were exclusively guided by the disinterested spirit of free inquiry, but also the per- manent establishment and extensive development of the basic prin--— ciples of scientific method and the first theoretical correlations of positive knowledge. Among them the scientific habit of thought was born; and, during the eight hundred years that fate allowed for its vigorous development, was completely matured, both by introspec- tive analyses of the natural processes of inference which yielded the principles of rigorous logical and mathematical procedure and by critical studies of the limitations of natural knowledge which yielded our characteristically pragmatic philosophy of suspended judgment. On the other hand, the problems presented to the ancient theorists by natural phenomena commonly observed were sufficiently difficult to tax the acumen of their keenest and most subtle understanding. a Sah delivered before the AcapEmMy, January 30, 1936. Received February 223 224 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 26, NO. 6 The record of their scientific achievement, therefore, does not reveal much evidence of the investigation of those hidden regularities in natural processes that now interest us most; which, as we know, are revealed only by the experimental disentanglement of superimposed effects. Their theoretical physics was practically restricted to the mathematical representation of the most patent phenomena—those of geometrical configuration, apparent celestial movements, mechani- cal equilibria and the reflection and refraction of light,—and the rest of their natural philosophy remained in the condition of preliminary qualitative generalizations little more than descriptive. In compari- son with the subtlety of their critical and methodical work, their theory, consequently, seems to us for the most part surprisingly simple and even naive. We have so completely assimilated their mathematics, which remains the foundation of our own, that even when we remem- ber its origin we forget that to them it was not at all a purely con- ceptual technique, but on the contrary a natural science; and since the rest of their physical theory, based on a necessarily meagre knowl- edge, has now become largely obsolete, it is very easy for us, naively on our own part, to dismiss it from serious consideration with refer- ence to the progressive development of science, by pronouncing it wrong rather than incomplete. If, nevertheless, curiosity leads us to examine somewhat critically the content of ancient theory, it is not long before we realize that its falsity is precisely like that of all later scientific representations of nature that have been successively in- validated by the growth of knowledge; that the suggestive conceptions it embodied have assisted rather than impeded the development of later theory, and that they have been by no means so frequently dis- carded as progressively modified in conformity with the advance of knowledge,—in other words, that, contrary to first impressions, ancient theory was incomplete rather than wrong. That such is the case is clearly due to the methodical predilection of the Greek natural philosophers. Their original motivation was pure curiosity: they desired, as Aristotle said, to know; not like their pred- ecessors to seek security and advantage by magical participation in the mysterious potencies that seemed to control an external world more feared than loved, nor to find the satisfaction of desire and long- ing in poetical and religious imaginings of its nature, but in accordance with their cheerfully confident and rational predisposition to discover its immanent reasonableness, which they never doubted, and to de- pend on that. For men of this temper it was nothing more than obedi- ence to the dictates of common sense to turn, when baffled in their JUNE 15, 1936 BARRY: ARISTOTLE 225 first attempts to discover this reasonableness by simple unguided observation of the confused and restless world about them, to the more careful study of ways of knowing. Nothing more strikingly illustrates their acuteness than that, having once confirmed their instinctive expectation by the discovery of certain regularities of interrelation among natural configurations and recurrent processes, they devoted their later efforts primarily to the problems of under- standing; for they appear to have appreciated even then, as clearly if not as vividly as we do ourselves, that valid theoretical knowledge depends unconditionally upon the practise of sound method, and therefore that the determination of effective and dependable proce- dures alike in research and in conceptual correlation are necessarily prior to the productive investigation of phenomena in detail. These philosophers thus became the pathfinders of natural science; blazed the passable trials that penetrated her rough domain, cleared them, and erected upon the secure foundations thus discovered the first watchtowers from which that domain could be surveyed, or, as they themselves expressed it, could be theorized. This was their great achievement; and the work was so well done—with a discriminating care so cautious that often our superficial critics even blame them for it—that though we have since widened and extended the ways they chose and rebuilt their theoretical structures, we have, after centuries of further exploration, neither abandoned those ways nor disturbed the foundations that they laid. The most productive and influential of all these ancient methodical theorists was Aristotle, whom our ancestors for centuries venerated as The Philosopher, and whom we as well must still accord, I think, this uniquely distinctive title. It is probable that no individual thinker has ever so stimulated and developed the critical thought of other men. His conceptions have become a part of our intellectual heritage and, variously interpreted, affect profoundly—whether or not we are con- scious of the fact—not only the character of our unconsidered precon- ceptions, but in considerable degree the philosophic temper which determines the criteria of our reasoned judgments. This circumstance makes unusually important all studies which yield us clearer insight into the original meaning of his philosophy, accentuating their his- torical value which is great on its own account. I wish this evening to discuss,—not of course at all thoroughly but, if I can, suggestively,—certain implications of one such study recently made which appears to me to be quite unusually significant, particu- 226 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 26, NO. 6 larly with reference to its scientific bearing. Accept these remarks, if you please, not as a critique—I make no pretence of scholarly com- petence in the matter—but rather as a brief commentary, of the na- ture of, let us say, a report such as an observer with scientific interests might be expected to submit on the findings of recent research in a field not his own. These findings will supply my text; the sermon, for better or for worse, must be an interpretation essentially personal, even though it develops only immediate inferences, more or less ob- vious and in probable accord with most well-considered opinion. The work to which I refer is Werner Jaeger’s admirable study, Aristoteles, which appeared a year or so ago and is now available in a very pleasantly readable translation. This book, there can be no doubt, must be read by everyone who is seriously interested in the larger and more important issues of the history of thought—and par- ticularly, so I think, those of the history of scientific thought. For myself it had peculiar interest, since its thesis, convincingly supported by scholarly arguments based upon the results of protracted and minute research and coodrdinated by an unusually scrupulous con- structive imagination, brought into clear definition the substance of suspicions I had long entertained but could not, for lack of sufficient historical and scholarly knowledge, justifiably express: namely, that the Aristotle with whom the professional philosophers have made us familiar was not, any more than the Aristotle of the scholastics, the veritable ancient thinker, but rather a legendary figure—the product of a persistent theological—or, if you prefer, a metaphysical—pre- dilection in speculative thought, which, having its roots in our com- mon background of Christian belief, has, as everybody knows, deter- mined the motive and contributed largely to the substance of all systematic philosophy from Spinoza to Hegel, and provided the guid- ing undercurrent of its most influential dialectical criticism from Berk- ley to the present time. This philosophical tendency has, throughout the modern period, predominated in our institutions of learning, and is, I think, quite properly, and not at all accidentally, called Academic; for its spirit and method, essentially unmodified by the successive invalidation of particular dogmas and unaffected by the scientific habit of thought which antagonizes it, remain essentially the same as those originally inculcated by Plato; and as common thought becomes progressively more critical, its doctrines acquire an increasingly vivid pre-Christian coloring, reverting more and more definitely to the abstract concep- tual idealism of this first philosophical supernaturalist. I can now a JUNE 15, 1936 BARRY: ARISTOTLE 227 do no more than suggest in this way, by reference to its derivation, the general character of that type of philosophical thought which until very recently has wholly prevailed among the learned, and through the control of higher education by conservative institutions in their development from monastery and cathedral to university has, by perpetuating and continuously fortifying the idealistic atti- tude and its doctrinal presuppositions among scholars, practically determined the point of view from which the philosophy of the past is almost universally interpreted and evaluated. It has required no conscious effort to accomplish this: our common religious feeling, our moral aspiration, our necessary belief in the real significance of human life itself, inevitably seeks and usually demands the sanction of a transcendent reality beyond ourselves and beyond the world we know to sustain our labors and to guide them. The idealistic philoso- phy responds to all such needs. Even its coldest intellectual formula- tion focuses attention on the eternal, the changeless, the universal; and in one manner or another conceives it necessarily as mind, which alone can be thought of as immaterial, and so incorruptible and abid- ing. Plato separated such an ultimate reality from the perceptible world of change and turmoil and imagined it as prior, self-subsistent and dominated by an immanent Good; Aristotle conceived it as immate- rial Fcrm, of which the physical world was an embodiment, developing in successive stages through particular forms from the gross and transitory to the ethereal and eternal under the control of a divine Purpose. Is it to be wondered at that in Christian Europe the names of these great men, who at one time were master and student, should be associated in thought and interpreted with reference to those most abstract and general of all their ideas, which refer to the ultimate problem which not Christians alone, but earnest men of all peoples in all ages have considered the gravest and most profoundly significant? Is it to be wondered at that endless labor should have been expended in efforts to reconcile their explicit contradictions; that in the failure to effect this reconciliation, those rationalistic philosophers who, un- like the theologians, could not escape their dilemma through denial of the final adequacy of reason and acceptance of revelation in its place, should seek refuge in some new variant of the Platonic realism, evading the arguments by which Aristotle had invalidated it by arbi- trarily rejecting the very premises of his contradictory theory of knowledge? And finally therefore, is it to be wondered at that, in the ensuing turmoil of dialectical disputations, the distinctive character 228 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 26, NO. 6 of his philosophy—which, after all, is based on common sense and with respect to its fundamental principles is as clearly understandable as one would expect an eminently logical naturalist to make it—has been somehow lost to common view? Whether or not it is to be wondered at, such is the case. In a word, the traditional Aristotle is not the portrait of any possible philoso- pher. It is a composite photograph of a single man facing opposite ways. One figure, rather faintly outlined, is that of a man who studies the actual world, writes elaborate works about the earth and the stars, about the forms and habits of animals, their anatomy and physiology, about man and his society, his art and the history of his thought and institutions; who derives all knowledge from experience, identifies the real with the actual and constructs a hesitant theology inductively from a logical interpretation of the phenomena. The other, much more distinct, is a transcendentalist who perceives that all that is incorruptible and permanent in nature is immaterial, that the essence of reality is disembodied form, the particular manifestations of which change perpetually in a process of development toward the changeless and eternal under the direction of divine power immanent in the whole, who therefore conceives the universal to be prior to and deter- minative of the particular, and reasonsaccordingly, inferring by means of an irrefutable deductive logic which he designs and subtly elabo- rates for the purpose, the essential character of particular realities from necessary universal premises: his only significant divergence from the thought of his great master Plato being occasioned by a less penetrating insight which permits him still to indulge a youthful ardor in studying the imperfect and distorted images of sense impres- sion, and—perhaps under the influence of a personal ambition to be original—to devise an empirical theory of knowledge, the inductive logic of which he never perfects for the simple reason that this is impossible. For the sake of clearness I have sharpened the actual outlines of this contrast; as a graphic pattern, however, it is correct. Both men pictured are lovers and students of nature, and both are empiricists, though one is earnestly and emphatically so, clearly conscious of the full import of this theory of knowledge, while the other is either weakly self-deceived by early predisposition or guilty in some sense of sophistry. On the other hand, both identify the permanent and changeless in experience as the ultimate reality, and conceive it as im- material; but to the one its immateriality is conceptual, to the other it is transcendental. In other words, both see that the elements of JUNE 15, 1936 BARRY: ARISTOTLE 229 all knowledge are general ideas, but to the one these universals are generated in human thought by the perception of similarities and differences among particular sensory impressions, to the other they are preéxistent and are only discovered in this way; to the one, consequently, they are derivative, to the other prior. In general, therefore, according to the one true knowledge is to be arrived at only through a carefully laborious inductive process checked at every stage by the criterion of consistency, while to the other this process is unnecessary beyond a certain point, since true knowledge finally arrives through immediate intuitive recognition of the preéxistent universal, and by the deductive elaboration of its implications. I think that the fundamental contrast is most clearly illustrated by this last antithesis. The one Aristotle reasons on the basis of empirical postulates, deriving concepts of increasing generality from the study of particular phenomena, inductively,—a posteriori; the other reasons from transcendental postulates, discerning by direct intuition im- mediately the general ideas that are only indicated by the final proc- ess of induction, and deriving from them implications of increasing particularity, deductively—a priorz. In the philosophy of one the transcendent metaphysics is a superstructure, in the other it is a foundation. The one calls his general theory—which we call the Metaphysics—First Philosophy, which is practically equivalent to what Herbert Spencer called First Principles; the other calls it The- ology. The faint Aristotle is both temperamentally and methodically a scientist, the more vivid Aristotle is a dogmatic metaphysician. It must not be supposed that either of these portraits is a misrepre- sentation: evidence enough can be found in Aristotle’s treatises to make both of them plausible. Until the time—toward the end of the nineteenth century,—when the influence of scientific positivism began to make itself felt in the world of philosophical scholarship, the pic- ture of Aristotle the metaphysician almost completely obscured the other. When Zeller wrote, however, and afterward, that of the scien- tist emerged more clearly. It remained, nevertheless, in the back- ground; and excepting in the thought of investigating scholars, re- mains there still. I think that we may account for this persistence of an old impres- sion by ascribing it not only to the force of habit but to two other causes at least: first, to the fact that, even among philosophers other than idealists—probably among all of them excepting a few radical empiricists, pragmatists and sceptics—the basic purpose of philoso- phy, if it is to be distinguished from science, still remains the Quest 230 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 26, NO. 6 of the Absolute; second, to the fact that scientists, who among all men would be most likely to grasp the full import of Aristotle’s theory of knowledge and reality, offer little or nothing to the discussion, for the adequate reason that usually they accept without question the traditional view of him, and as a natural consequence seldom read his works at all. This will be evident to anyone who glances at the earlier chapters of our typical summaries of the history of science. Here one always finds some reference to the great Stagirite, for his historical importance cannot be ignored; but this reference is brief, and its usual implication is to this effect: that he was a subtle dialec- tical thinker who attempted on the basis of reasonable but arbitrary presuppositions to excogitate once and for all the ultimate nature of things, and succeeded in constructing a cosmology so impressive that it imposed itself like an incubus on the thought of men for nearly two thousand years, making impossible any significant progress of natural knowledge until its hold was broken by the destructive criti- cism of Bacon and the experimental investigations of Galileo; that this repressive influence was due to its perpetuation of old Greek representations of nature which were the final results of a succession of naive hypotheses quite unsupported by experimental evidence, subtly elaborated by deductive reasonings a priori, and supported by dialectical arguments purely logical in the characteristic manner of the Greeks, whose scorn of vulgar labor inhibited their scientific development by restricting their thought to the domain of pure ratioc- ination; that Aristotle is to be credited with an effort to promote the study of nature and particularly that of animal life, but that, sharing these limitations and guided by an unusually intense desire and apti- tude for synthetic conceptual generalization, he ultimately produced little more than a transcendental metaphysics, the only positive achievement involved in which was the development of formal logic; that this is sufficient to secure his fame, but that his philosophy as a whole cannot properly be called science at all, involving, as it does, conceptions of immanent potency, vital force and abstract notions of causality derived from the idea of a guiding mind and purpose in nature, and culminating, as it does, in a theology; and that, finally, his simply descriptive biology is vitalistic, his psychology animistic, his chemistry merely an empty occult theory of subsistent forms and qualities, and his physics, which at least is positive, arbitrarily dog- matic and wholly wrong, especially with reference to falling bodies. Now, even this estimate cannot be said to be wholly unjust. It is JUNE 15, 1936 BARRY: ARISTOTLE 231 that which a scientist must in candor make, when, accepting as is reasonable the expositions and commentaries of the historians of philosophy, he evaluates the work of Aristotle from his own point of view. If, however, curiosity leads him to examine these presentations in his customary way, separating explicit statements of particular conceptions and doctrines from their argumentative context, com- paring them with reference to their diverse implications, and thus inferring, without reference to past opinion, their common ground, he will then certainly begin to suspect that, false though Aristotle’s pic- ture of nature may be, there is the possibility that its falsity may not after all be that of a synthesis based on premises and guided by habits of reasoning fundamentally wrong, but rather that of a genuinely scientific theory of nature that has been invalidated by new knowl- edge, like Copernican astronomy or Cartesian mechanics, like the corpuscular and caloric theories of light and heat, or Stahl’s phlogis- ton theory, or Lamarck’s theory of evolution, or—dare I say—Bohr’s theory of the atom; concerning all of which one may assert, as the curate at the Bishop’s breakfast table said of the eggs served him, that parts of them are excellent. If this suspicion engenders sufficient courage, he will then read the original treatises. This will be difficult, even in translation. The Metaphysics, to which he will have been invited first to turn, is not consistently sequential, and its form in detail is imperfect; the diction is very often turgid and elliptical, there are reiterations and cross- weavings of slightly variant arguments and criticisms which cannot be understood without further knowledge of the antecedent and con- temporary philosophy; and often the parts cannot be made clear with- out a knowledge of the whole. Each in its own way, the other theoreti- cal treatises offer like difficulties. The subtle elaboration of the An- alytics is intolerable; the Physics is a mass of discussions wholly con- cerned with abstract generalizations. Throughout, the arguments by appeal to reason wholly predominate, and it is the logician who passes judgment, though the grammarian is always at his elbow appealing to common sense, and the observer contributes to the evi- dence adduced. Altogether, the first impression gained from this read- ing is such as to discourage the hope of finding here much of science as we know it, and to concede to the philosophers the reasonableness of their traditional evaluation. When the broad outlines of the system become clear, however, when the whole becomes in this sense known, the scientist whose presence we first suspected again emerges; but 232 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 26, No. 6 still he is far from dominating the metaphysician, and it becomes even clearer why the traditional figure has always been composite and in- consistent. But now, suddenly and convincingly, this age-long difficulty has been resolved: not by further rational analysis of the content of the treatises as they stand, but by an historical investigation which has at last presented to us clearly outlined, in place of the static sculp- tured figure of tradition, the picture of a living man; and in place of the conglomerate collection of his writings, which scholars heretofore have been forced to interpret all at once as if they were the product of a single mind, at least two groups of writings composed at different times and guided by the changing basic conceptions of a rapidly ma- turing critical thought. This is the work of Jaeger. I cannot here and now, of course, stop to present his arguments: granting the facts he adduces, as we always must and do grant them in discussing the results of highly reputable scientific research, I shall only say that his inferences, based on a considerable mass of data both historical and literary, seem to me to involve a surprisingly slight and inconsequential amount of conjec- ture and to be wholly convincing—the more so since I can imagine no conclusions opposed to those he draws to be consistent with the his- torical evidence that refers to the thought and work of Aristotle’s im- mediate disciples. In broadest outline, then, let us survey in the light of Jaeger’s findings, the course of their master’s intellectual develop- ment. When as a youth of seventeen,—born of a family of physicians and bred under the influence of the naturalistic habit of thought which the Ionian medical schools had long inculcated among men of that profession—he left his father’s home in Stagira and came to Athens, the Academy of Plato had already been solidly established as the stronghold of the most conservative philosophical tradition of Greece. It was not a centre of important scientific activity: indeed it had never been that. To the Ionian philosophy, the first science properly so-called—the first disinterested, objective, exclusively rational and freely critical investigation of nature, which had culminated in the atomic theory of Leucippus and Democritus and the physical astron- omy of Anaxagoras—Plato, like his venerated teacher Socrates, re- - mained completely indifferent. To the great sophists of the preceding generation, contemporaneous with Socrates, who had laid the founda- tions of that empirical and critical epistemology which later produced JUNE 15, 1936 BARRY: ARISTOTLE 233 the sceptical philosophy of suspended judgment that we now call scientific, he was sharply antagonistic. Their humanistic interest he shared, and his original motivations were wholly ethical and political ; but his intentions were reactionary, and his earlier philosophical work was guided by the purpose of establishing a supernatural sanction for morals and an aristocratic state as the agency by which his teach- ings, thus supported, were to be imposed upon humanity. In Aris- totle’s time these motives had already found expression in the famous transcendental theory according to which the ultimate reality was an incorporeal world of subsistent Ideas apart from the world of appear- ances, perceptible phenomena being merely the distorted images of this reality and worthy of study only for the suggestions they con- veyed to the imprisoned souls of men of the final Truth and eternally dominant Good which in their free immortal state these souls had known by immediate intuition. The Academy, when Aristotle entered it, was no longer engaged in discussing the ethical and religious import of this theory, but was attempting the more difficult task of giving it clear conceptual defini- tion. Discussion among the younger members of the school was eager, lively and doubtless keenly dialectical. It would not be a bad guess to ascribe the logical trend of Aristotle’s later thought primarily to this stimulation. What else he gained from it can only be inferred from the content of Plato’s later treatises. The first of these, the Theaetetus, shows clearly that its author had already found in the abstract con- cepts of mathematics a clue to the more subtle definition of the Ideas. Mathematics, then, was becoming increasingly important in the dis- cipline of the school; but Academic interest in it was metaphysical, and appears to have contributed little or nothing of significance to its development. The great geometers of the time lived and worked else- where, as had their predecessors: Archytas at Tarentum in Italy, and Eudoxus of Cnidus, the outstanding scientific genius of the age, at Cyzicus. There Eudoxus had established a scientific school the activity of which profoundly influenced the Academy, who assimilated what they deemed important of its discoveries, just as they had assimilated those of the now dispersed Pythagoreans. From certain refugees of this strange brotherhood, who had de- veloped the first systematic mathematics out of their quasi-Orphic religious rites of purification, Socrates had already borrowed a not inconsiderable amount of mystical doctrine; and Plato, following him, was fascinated by it. His poetical temper, which responded so spon- taneously to the elusive suggestions of ancient myth and legend, found 234 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 26, NO. 6 not only in these but also in the Pythagorean numerology a:vague significance which appeared to him profound; and at this time, the new knowledge of Babylonian astralism which the widely-travelled Eudoxus had brought to Greece accentuated this wondering interest in things mysterious. Plato’s later work in this manner became even more deeply tinged with mysticism, and toward the end verged strongly toward the occult. His theory of Ideas, originally the imagi- native product of a moral aspiration, but now rationalized, except- ing for its Orphic and Pythagorean content, as an idealistic represen- tation of the world of ultimate reality, became more definitely reli- gious. With its further syncretic development, Chaldean astrology appeared for the first time as an element in Greek speculative thought; the descriptive names of the planets were replaced by the names of those Hellenic deities which most closely resembled the Babylonian; Zoroastrian elements entered to accentuate its blending of morals with the worship of the heavens, and so on—until finally its metamor- phosis into a system of dogmatic beliefs was complete. This stage of Plato’s thought is best represented by the Tz:maeus, an indefinite and poetically suggestive work which scholars have struggled for centuries and still struggle to interpret rationally, and by the Laws in which, among many expressions of arbitrary intolerance, he condemns as atheistical and punishable by death all those who still venture to study astronomy otherwise than in his own manner as an orthodox religion. The younger generation in Plato’s school did not accept all this, and it seems that not afew of them were becoming destructively crit- ical of certain tenets of his theory even at the time when he wrote the Parmenides, though all of them, probably, subscribed to his fun- damental doctrine. Among these was Aristotle. He remained at the Academy for twenty years; wrote dialogues of literary excellence in the later didactic manner of his master, and meanwhile, though still under the compelling influence of Plato’s personality, gradually de- veloped his own ideas. But, so far as we know, he wrote nothing in contravention of the school’s accepted teachings before Plato died. For three hundred years thereafter only his immediate followers knew his later treatises; for shortly after his own death these disappeared, and were rescued from oblivion only when Andronicus of Rhodes recovered them in the time of Cicero. Aristotle’s ancient reputation, therefore, was that of a disciple of Plato; and his dialogues were ac- cepted by the Neoplatonists themselves as pure Platonic doctrine. When Plato died Aristotle left Athens, and lived for a few years in JUNE 15, 1936 BARRY: ARISTOTLE 235 association with his friend Hermias, the tyrant of Atarneus in Asian Aeolia, lecturing, either occasionally or regularly, in the school that had been founded at Assos under this ruler’s protection. Then, having accepted an invitation from Philip of Macedonia to become his diplo- matic adviser and the tutor of his son, the young Alexander, he spent another few years at the Macedonian court. These independent occu- pations, in part philosophical, in part political, occupied the next twelve years, which were the most productive of his life. It was during this time that he first formulated the principles of his own philosophy, and broke awayfrom the intellectual tradition of the Academy, giving up the opinions that formerly held him, gladly, as Plutarch says, be- cause he considered this a necessary sacrifice to the truth. The dialogues written during his last preceding years in Athens, though they lack completely the allusive indefiniteness and mystical obscurity of Plato’s writings, emphasize a demand for exactitude in scientific thought and give expression to that love of nature which he seems never to have lost, still assert the primacy and sufficiency of a purely theoretical knowledge, such as is revealed only to the creative intellect that reflects upon immaterial being alone and which discovers thus and contemplates only what is perfect, changeless and eternal in nature; and they reiterate the belief that the human soul, immortal though imprisoned in the corruptible flesh, is part of this divinity, and advocate a rigorously ascetic code of morals which recognizes the utter worthlessness of earthly things. Such is the tone of the Aristotle of the Academy as he is revealed to us in the extant fragments of his dialogues. That shown in a later short treatise entitled On Philosophy, which Jaeger assigns to his very earliest years at Assos, is strikingly different. In this work, which appears to be a statement written expressly to define his first depar- ture from the Platonic doctrines, his whole mental attitude appears to be changed. The religious feeling remains; its object, however, is no longer the imperceptible world of eternal Forms, but the living world of nature. Cicero quotes from it the following eloquent passage: If there were men who had always lived beneath the earth in good and shining habitations, adorned with statues and pictures and supplied with all the things possessed in abundance by those who are considered happy, and if, however, they had never gone out above the earth, but had heard by rumour and report that there is a certain divine presence and power, and then if at some time the gorges of the earth were opened and they were able to escape out of those hidden places and to come forth into these regions which we inhabit, then, when they suddenly saw the earth and the seas and the sky, when they had learnt the greatness of the clouds and the power of the winds, when they had gazed on the sun and recognized his greatness 236 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 26, NO. 6 and beauty and the efficacy with which he causes day by spreading his light through the whole sky, when moreover, night having darkened the lands, they perceived the whole sky laid out and adorned with stars, and the variety of the lights of the moon, now waxing now waning, and the risings and settings of them all and their courses ratified and immutable to all eter- nity—when they saw this they would straightway think that there are gods and that these are the mighty works of gods. Evidently enough, it was no ascetic transcendentalist who wrote these words. They are those of a man who is wholly alive, and alert to the beauty and wonder of the sensible world about him—the Ionian naturalist come to life again. Even his discussion of theory has this background: it is historical; successive philosophical ideas are pre- sented, as well as may be chronologically, as phases of a continuous development; and the treatment of religion is the same. This natural history of thought—the first ever written—is followed by a destruc- tive criticism of the theory of Ideas. Aristotle’s original arguments are not preserved, but, being of necessity logical, are not improbably the same as those of the first book of the Metaphysics, which was written, according to Jaeger, at this time or very little later. These show that the Ideas are wholly gratuitous, that their existence would imply an infinite regress of like ideas-of-ideas, that whenever their nature is defined they become mutually inconsistent, that the notion of a participation of phenomena in the Ideas is meaningless, and that the latter, being changeless, offer no explanation whatever of the processes of nature; from all of which he concludes that they are mere words. It is the physical world itself, he says, which is eternal; a world of things in process of change, being continually generated and destroyed, but in definable ways which produce, disintegrate and reproduce, endlessly, the same forms out of an indefinite material substratum; forms which exhibit such similarities that they can be classified by types and groups of types increasingly inclusive but always definable by abstraction. It is this very corporeality, this end- less series of recurring forms—shapes and lustres, densities, odors, viscosities, tones and colors, these specific potencies which always yield the same substances by like processes of growth and transmutation, and these sharply definable relations between bodies that are char- acterized by such attributes, which permit their grouping by classes in generic order—it is these elements of experience that constitute the eternal and changeless. For, whether they are masses, or potencies or qualities, or species and genera, and though their distribution per- petually changes, as such they are always the same, and occur in the same associations. They are incorporeal and universal, that is to say, JUNE 15, 1936 BARRY: ARISTOTLE 200 they are ideas; but they do not exist apart from the physical world; they are, on the contrary, in it and of it; and further, since this is so, they become known, not at all by some reminiscent illumination of the immortal soul, but by observation and study of the actual world, as general concepts which define its underlying structure and order. This minimal amount of Aristotle’s final First Principles may con- fidently be assigned to the period of his treatise On Philosophy. Whether or not it was further developed at this time does not matter; for the basic conceptions of his final philosophic doctrine are here pre- sented. It is evident that together they signify, not a modification, revision and development of the ideas of Plato, but their unequivocal repudiation. It is still contended, and with justice, that Aristotle’s recognition that all knowledge is conceptual, the terms necessary in any reasoning whatever being abstract and general, is due to his training in the Academy, where all discussion concerned itself not ingenuously with the nature of things without reference to the process of knowing, but rather, introspectively and self-consciously, with ideas. But this was the character of all critical thought from the time of the sophists down. Aristotle himself acknowledged his debt to Socrates by saying that two things in fairness must be ascribed to him, inductive argu- ments (about the meanings of terms) and universal definition. The school of Plato had developed this analysis, and brought clearly to light the fact that general concepts, in contrast with the changing world to which they originally referred, were changeless forms of thought; but hypnotized by this idea just as the Pythagoreans before them had been hyponotized by the idea of changeless number, and the Eleatics by the idea of the Absolute, and infected with the mys- ticism of these older schools which Socrates had transmitted, they shared his complete indifference to natural philosophy, forgot that the changeless entities on which they meditated were ideas conceived by the human mind, sought no generating cause behind them and in- stead of developing a rational epistemology pronounced them abso- lute. Without further reference to the vagaries of mystical thought to which this dogmatism led, it is easy to see why Aristotle rejected it. His spontaneous love of the world of nature, which animates the striking passage I have just quoted, forbade him to accept as valid any theory of reality which could not explain its incessant change, and which ignored the immanent power within it—the compulsion of 238 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 26, NO. 6 fact, which obviously determines both will and waking thought. In addition, his genetic habit of thinking, which may, perhaps, be traced to that particular interest in living nature which was the interest of his ancestors, compelled him tirelessly to seek out causes for every- thing, and to rest dissatisfied with any doctrine which left anything causeless save the totality of Being. He thus discovered the generating cause of the ideas in perceptions of the natural world, remembered, compared, discriminated, generalized and arranged in generic order; and thus outlined the first psychological epistemology, that which refers all knowledge to physical experience. This is made evident in the first paragraph of the Metaphysics, from which I shall presently quote. And finally, his treatment of the Ideas shows even more clearly that to his logical mind, the keenest that had yet appeared in the world, the Platonic conceptions were altogether too incoherent and by implication too absurd to be credible even as reasoned dogma. I must emphasize this point. Without reference to his own theory of the nature of things, and by logic so pure that even the purest ra- tionalist could find no flaw in it, he utterly destroyed the Platonic theory. If its like survives today this can only be by grace of an act of faith. In its place Aristotle erected a structure of general conceptions, explicitly derived from sensory experience of the physical world. I venture to say that no mutation in the whole history of philosophy— save that of its very origin in ancient Ionia—was so uncompromis- ingly abrupt and complete. What, now, was the final character of this new philosophy of nature? Already we have surveyed its basic con- ceptions. Time presses, and gives no opportunity for discussion of its subtle elaboration and of the arguments adduced to support it; but this is not necessary. Its elaboration is largely a minute analysis in- tended to make its postulates and the logical consistency of its derived conceptions unequivocally clear; its arguments are those presented to contemporary philosophers, principally Platonists, who are met usu- ally on their own ground. The analysis is minute beyond all present necessity, the arguments are equally wearisome and their purport not infrequently obscure excepting to one fully versed in the conflict of opinion in Aristotle’s own day. I wish, however, to call attention to certain points that would be of particular interest to a scientist whose curiosity might lead him to ask how far this ancient thinker advanced toward that view of the world, familiar to us, which is the actual out- come of the philosophical empiricism he sought to establish. Everything essential to this inquiry will be found in the final text of JUNE 15, 1936 BARRY: ARISTOTLE 239 the Metaphysics,—which is a late compilation of lecture briefs and texts, in large part written at the time of his treatise On Philosophy but amended and amplified in his later years—and almost enough for the purpose will be found in its fourth book, called Gamma. The very first words of this treatise state, simply, both its purpose and the predisposition of its author. ‘‘All men,’’ says Aristotle, ‘“‘by nature desire to know. An indication of this is the delight we take in our senses; for even apart from their usefulness they are loved for themselves. ... By nature animals are born with the faculty of sen- sation, and from sensation memory is produced in some of them... . The animals other than men live by appearances and memories and have but little of connected experience. . . . But in men, from mem- ory experience is produced; for many memories of the same thing pro- duce finally the capacity for a single experience. ... And art arises when from many notions gained by experience one universal judg- ment about a class of objects is produced.” This is the utterance of a man who not only wishes to know, but asks immediately how we know; and answers the question briefly by sketching the natural process of knowing in fewest words as if it were a matter commonly understood. There is not here—nor, indeed, else- where—any suggestion of the possibility of knowledge derived other- wise than by inductive inference from experience of the natural world. This world Aristotle recognizes as independent of and prior to the act of knowing, just as we do, and for the same reason,—namely, that it is a world of necessity in which compulsions constrain both will and thought; and he says further that ‘‘it is impossible that the substrata which cause sensation should not exist even apart from the sensation. For sensation,’’ he says, “‘is surely not the sensation of itself; but there is something beyond the sensation which must be prior to it, since that which moves is prior to that which is moved... .”’ Not only, therefore, do we derive our knowledge from experience: the character of our knowledge is predetermined by natural necessity. Conse- quently, with respect to its sensory foundation at any rate, it is true knowledge: a direct reflection of fact. ‘‘For,’’ he says, “‘it is not possi- ble to be in error with respect to the question as to what a thing is save in a accidental sense,’ that is, by inconsistent identifications due to ignorance of attendant circumstances and consequently erratic inference, which further and more extensive observation will correct. Against the sceptics who emphasized the diversity of different indi- vidual sensory impressions of the same thing, and insisted that a choice between the conflicting ideas derived from such impressions 240 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 26, NO. 6 must be arbitrary, he argued that ‘“‘the truth is not that what appears exists, but that what appears exists for him to whom it appears, and when and in the sense in which and in the way in which it appears.” Here is an admission of the relativity of knowledge; but a denial of its indeterminacy, since its relativity is defined as the mutual inter- dependence of impressions. Know the whole Gestalt, and you know truly. It is but a step from this position, in the direction of Aristotle’s own genetic tendency of thought, to admit, in recognition of the in- completeness of our knowledge, the necessarily tentative character of all conceptual representations; but Aristotle will not go this far, for reasons that will appear. He knows, like everybody else, that we do possess a store of perfectly dependable knowledge derived from the correlation of diverse impressions which within human experience is final: in a word, that we know a multiplicity of undoubted facts, some of them of inclusive generality. He expressly postulates, there- fore the possibility of a similar positive knowledge of everything, by asserting a priori the minimal necessary assumption, which is that a thing cannot be and not be in the same sense at the same time. This postulate, called the law of contradiction by logicians, might equally well be called, by one more interested in the theory of knowledge, the law of exclusive discrimination; for it is derived from the conviction that things can be unequivocally identified. He justifies it, conclu- sively enough, by the observation that to reason at all we must start with postulates of some sort, since otherwise our thought would be involved in an infinite regress of conceptions; and remarks that no postulate could be more simple than this one, which is, furthermore, consistent with all experience. He challenges the sceptics to dispute it, and promises to entangle them in hopeless confusion of thought if only they will say something and not take refuge in silence, expressing their doubt, as Cratylus was forced to do, by merely wiggling their fingers. We are all familiar with the general outlines of the conceptual structure that he builds on this foundation: that of a finite corporeal world which is continuous—for there is something everywhere even if it is no more than warmth and cold or the movement of the aether which men call light—and yet heterogeneous, separated into parts distinguishable by different attributes. These distinguishable some- things (or, more briefly, things), which Aristotle calls substances, are generated and destroyed, come to be and pass away; but while they exist, there is that in each of them which, quite apart from those JUNE 15, 1936 BARRY: ARISTOTLE 241 accidental attributes that may occasionally and indiscriminately affect them, is always the same,—persists,—remains unchanged; which upon its regeneration reappears unchanged; which, indeed, con- tinuously exists unchanged as its incorporeal essence, the object of pure conception. There still attaches to the word essence—thanks to the persistent bad habit of many scholars who continue to think about Aristotle in Latin—a not inconsiderable flavor of the occult. It is worth while, therefore, to remark concerning this word (one among several in the translated Aristotelian vocabulary which are similarly affected) that the suggestion of transcendental mystery it conveys is quite gratui- tous. This we realize at once when we say instead that this suggestion of mystery is not essential. There is nothing mystical about these es- sences: they are simply physical properties or, in Aristotelian ter- minology, forms which uniquely characterize different kinds of sub- stances, and thus define them. They are, of course, not tangible things but concepts—which, when they refer to substances, are qualifying attributes expressed by adjectives such as combustible, univalent, triclinic, vertebrate, parasitic; and when they are made the abstract objects of discussion become substantive and are expressed by nouns such as combustibility, univalence, triclinicity, backbonyness, para- sitism. Further illustrations of this sort will make it quite clear that the mystery of the world of essence is merely the product of a naive confusion of thought precisely like that which is nowadays evident in the ingenuous or whimsical imaginings of those who try to picture a burglar entering a bank-vault through the fourth dimension. Those who are familiar with alchemical theory will recognize the same sort of confusion in the thought of those sooty spagyrists who attempted to manipulate the three alchemical essences or Principles, their phil- osophical mercury, sulphur and salt—which were fluid metallicity, combustibility and vitriosity—as if they were corporeal things. No such confusion affects the thought of Aristotle. He perceives, simply, that by their physical properties, or essences, all substances may be unequivocally discriminated and compared, and identified as individuals of definite species, or substances of a second conceptual order, which chemists who think of copper and caustic soda and su- crose call substances, without qualification, but which biologists who think of chimpanzees and roses and pneumococci still call species. These species, thus defined by their essences (and not otherwise truly defined), are now seen to fall into natural classes themselves definable by that which is common to particular groups of essences, namely 242 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 26, NO. 6 the essence of a more inclusive genus. The biologist still finds it con- venient, if not necessary, to identify the multifarious species of ani- mals and plants with reference to these genera, as homo sapiens, drosophila melanogaster, solanum tuberosum, and so on; and chemists do likewise, though more confusedly, when they speak of potassium cyanide or diparaamidothiodiphenylamine. And the genera may like- wise be grouped in successive classes of higher order. Note that all these classes are real, but incorporeal and definable only diagram- matically or symbolically, that is, by abstract ideas which are essences and that they persist, though individual organisms die and compounds are transformed. This is enough to illustrate the grosser outlines of Aristotle’s en- visagement of the natural order. The Academics before him had taken the first steps toward such classification, but their taxonomy was con- fined to the futile arrangement of pure ideas as such. It was Aristotle who first discerned this generic type of relation among things per- ceived, recognized its universality, realized its significance as the basis of all knowledge, and pointed out the possibility of its develop- ment to extreme exactitude by increasingly minute and discriminat- ing observation. He developed its implications subtly and elaborately, and identified it with the order of deductive thought. Species and genera, once determined by observation and inductive inference, typified in their complicated relations the relations of all ideas. The syllogism is nothing other than the statement of the simplest generic relation: The individual Socrates is of the species man. The species man is of the genus mortal being. Therefore Socrates is of this genus and is mortal. In the Analytica he reduced the most complicated of deductive proc- esses to this simple type, the ‘‘first figure of the syllogism.” J. 8. Mill properly criticised this sort of reasoning as circular, since all general premises presuppose a knowledge which the conclusion merely makes explicit. I hardly think that Aristotle would have disputed this: but he might well have pointed out that latent knowledge is merely po- tential, and does not become actual until it is made explicit: there are » very few of us who can determine by mere inspection the real roots of a cubic equation. He developed no fixed method of inductive logic, for the simple reason that there can be no such thing; since, as he knew, the process is one of observation and comparison, and effective pro- cedures in this sort of thinking are many. JUNE 15, 1936 BARRY: ARISTOTLE 243 Aristotle does not, in his theoretical works, actually apply this scheme of thought, and so develop it. He does, however, call attention to the several aspects of phenomena with reference to which, inevita- bly, they are described and become by conceptual definition known. These are the Categories: we might call them conditions of actual existence. They are also ways of knowing and primary forms of asser- tion. To know a thing we must know where it is, when it occurs, how it affects our senses, how much it is, how it acts or is affected by ac- tion, and what its relations are to other things; or, to use the custom- ary formula, we must know it as Substance, and with reference to Space, Time, Quality, Quantity, Activity, Passivity and Relation. These categories overlap, but they are exhaustive. In physics today we have reduced their number—sometimes unit- ing Space and Time in a four-dimensional continuum, considering time Gf I may be forgiven) an imaginary distance; wholly eliminating Passivity by the law of action and reaction; and determining Quality by measurement, thus representing it by, if not actually reducing it to Quantity. But these are all theoretical innovations, the philosophi- cal import of which appears not to be settled yet. There are those who still believe that there is something essentially different in Space and Time: that being large and being old are not in any sense the same; though growing up and growing old are related, being different aspects of a single process of change; and Aristotle himself said that all change involves and may be determined by motion in space (a pregnant re- mark!) from which motion, he also said, we derive the conceptions of space and time. In practise, further, we are still compelled to define many passivities, without reference to specific reacting forces, as hard- ness or elasticity or breaking strength, and so on. The definition of qualities by quantitative measurements, finally, is not yet universal, and if compulsory would be sometimes very awkward in chemical, biological, and other scientific practise. In short, when we alter these categories we depart from the actual sensory world which Aristotle studied and which remains the only complete reality; our conceptual representations being, in fact, diagrammatic and imperfectly so at best. Now, all this representation of substances, essences, species, genera and logical categories refers only to the permanent structure of the world—or rather, to its structural plan, which is the same as that of our reflective thought. But Aristotle was interested, and intensely so, in the phenomena of change: of movement, generation and corruption, 244 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 26, NO. 6 material transformation. Beneath all his detailed criticism of the theories of his predecessors there frequently appears the fundamental objection that they are defective in not attempting to explain occur- rences as well as things. This, he observes, is the primary defect of the whole Pythagorean-Eleatic-Platonic complex of doctrines. Those of other philosophers, notably Heraclitus and the Atomists, show deeper insight, for they are dynamic. Even in them, however, theory remains inadequate in not taking into account the actually determinate char-- acter of all processes of change. Heraclitus’ conceptions of cyclic process and dynamic equilibrium are incidental features in a theory of universal flux, whereas actual changes—if we except the motion of the stars—have beginning and end in fixed substantial forms: water changes to air, bog-ore to iron, acorn to oak. The doctrine of the Atom- ists is even less acceptable; for the random motion they imagine would produce not a limited but an infinite number of types of change, which is contrary to fact, and this with no regularity of process, which is again contrary to fact. To meet these objections, Aristotle proposes his fowr causes, which may be thought of as analogous to the categories in that they refer to the determining conditions of all change, these being at the same time necessary inferences from the phenomena—aspects of nature again—and also ways of knowing. There must be something to change —a material cause; an action or potency to start it—an efficient cause; a necessity which makes it follow a definite course always the same in similar circumstances—a formal cause; and finally a purpose, defined by its end—a final cause. The material cause is matter, Aristotle’s indeterminate substratum actually existent only in substances, which we quantify as mass; the efficient cause we call energy,—mechanical, chemical or vital; the formal cause, natural law. The final cause we ignore in science, not because there is no purpose in nature, for there is purpose in nature, at least in men and higher animals; but because it does not assist us to understand phenomena. To Aristotle, however, the phenomena of life, in which he was most intensely interested, were inexplicable without it: he had to include it, or leave his own system in one aspect defective like the others. Also, the reasonableness which all Greeks from first to last imagined as latent in nature, that imma- nent intelligibility which accounted for the possibility of understand- ing, was doubtless very vivid to the thinker who had discovered the exact correspondence of the forms of nature and the forms of thought, and worked to the same end. He conceived all nature, therefore, as JUNE 15, 1936 BARRY: ARISTOTLE 245 purposeful, and thought of formal causes as evidences of intelligent guidance determined by the final cause of all—a divine intention, not supernatural but, like the physical world which was its embodiment, eternal. This was his entelechy. His system, unlike our own, was teleo- logical; and like those that preceded it culminated in a conception that we call extra-scientific, that is, the conception of the divine. In this sense it became a theology: Aristotle himself gave his First Phi- - losophy this alternative title. This introduces us to that part of the philosophy we are considering upon which the interest of metaphysicians is still focused. Jaeger himself devotes the greater part of his criticism to the development of Aristotle’s theological doctrine; not only because of its historical importance, I think, but also because it still appears to him, as a philosopher, the most important aspect of his great predecessor’s thought. You will not blame this distinguished scholar for any of the tedium you may have felt during the preceding discussion: its famil- iar matter he doesn’t mention at all, not even to refer to its scientific implications. What, then, is this theology? Let me, first of all, refer once more to the text of the Metaphysics, in order to make clear its philosophical basis, which, as I have already indicated, is the general conception there presented of the world of nature. The first step toward a con- sideration of the transcendental is taken when we meditate upon the most general of all ideas, that of existence. What, then, transcends this world of nature and its reflection in reason? In the most compre- hensive meaning of the words, What zs? At this point Aristotle’s meditation carries him to what will at first seem, even to a scientist, a radical empirical extreme. He answers, in effect —‘‘Nothing trans- cends the world of nature; indeed there zs nothing, really, but what is known.”’ In short, he identifies that which is with that which is known; which means for him, we must remember, known exclusively through experience. There is always the possibility of new experience: the natural world has this potency. But Aristotle means by potency ex- actly what we do when we refer to gravitational or thermodynamic potential or potential energy in general: not a mystical pervasive complex of indeterminate influences, the primitive Mana, a participa- tive feeling of which is the beginning of religion, but merely a blanket term which designates a number of specific potentials (let us say), each definable by its actual effect. Aristotle criticises sharply a philosopher 246 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 26, NO. 6 he rather admires, Anaxagoras of Clazomene, for ascribing the origin of things to an indeterminate mixture. There is no such thing, he ar- gues, because it cannot be defined. Nothing really 2s but the content of knowledge; for the word “‘is’’ has no meaning unless it refers to some- thing: no predication can be made about the unknown. Things may come into being; but before they are known they are not, and can be sald to be possible only by inference from an actual event. We might try to amend this statement by saying that all this is so, but only so far as we are concerned; that other animals with senses dif- ferent from ours may know things not known to us, which are there- fore actual to them; and that we all, including Aristotle himself, believe that the world existed before there were any human beings to know it. But note that these statements all express possibilities only: it cannot be categorically asserted that animals do know what is unknown to us, much less what they know, until we are able to infer from our own knowledge of their behavior that they do perceive directly something (like sounds of high pitch or odors) which we perceive otherwise, or from other knowledge infer to be existent. Then we may assert not only that they do sense what we cannot, but that they did formerly sense what we could not. In the same way we may assert that the earth existed before there were human beings to know it, but only after we can infer from a mass of present knowledge that this must have been so; and then, this fact is a part of knowledge. When we speak of Being, says Aristotle, we mean the known; all else is not- Being. Note that this not-Being includes all that is indeterminate in any way: space unoccupied or unbounded by substance; or, in any sense whatever, the infinite, since only the finite is actual; or potency not specifically designated by some actual effect; or matter deprived of form,—that is, substance without attributes—but not pure form, for this, though incorporeal, may be conceived with reference to the actual—not only as some common abstract idea, but as figure and number, arithmetic form, type of logical relation, or mode of behavior. In short, anything definable may exist; but nothing indefinable, for the indefinite cannot be thought about, excepting by negation. Aristotle’s final theology is consistent with these premises. Jaeger describes its development most interestingly, at length; our own divergent interest permits us to summarize his findings briefly. In the treatise On Philosophy, referring to religious rituals, Aristotle says: ““Those who are being initiated are not required to grasp any- JUNE 15, 1936 BARRY: ARISTOTLE 247 thing with the understanding (uafety), but to have a certain inner experience (zafetv), and so to be put into a particular frame of mind, presuming that they are capable of this frame of mind in the first place.’’ He thus sharply separates the religious from the theoretical interest; and his theology becomes a part of his physics, the super- structure of his theory of nature, or, as our ancestors expressed it, “the Queen of the Sciences’’—in a word, a natural theology. To find his evidences of the divine he contemplates the heavens, like the old Pythagoreans and like multitudes of other men, before and after. Up there, the changeless stars at immeasurable distances move eternally in their circular courses, measuring endless time. This motion is natural to the heavens and proves them to be of other sub- stance than any we know in this thick vapor which surrounds us on Mother Earth, where everything when unconstrained moves up or down. The purer air above us becomes transformed beyond the moon, where only the purest exhalations penetrate; it becomes celestial aether, an essence sui generis. Within these silent realms all movement is spontaneous, for here it originates: the stars are living beings, there- fore, of purest substance, nourished by the finest exhalations; and since this is so, their conscious minds are more intelligent than it is possible for us to conceive; for did not Hippocrates show that the purest foods and airs and waters produce in men the finest minds? The stars’ spontaneous movements, however, are all the same, and even the wandering planets below them share in their ceaseless daily revolution. This proves the presence in the furthest celestial realm of a single Potency that guides them all: a first cause of movement, it- self quiescent Form; the divine Essence, which is God. Not the Crea- tor, for the cosmos is uncreated, eternal: not the Idea of the Good, un- less one sees supreme good in supreme intelligence and the perfection of its actualization in celestial movement; for there is no Providence in Aristotle’s heaven, only power. His God is the divine potential energy that animates the world. In the twelfth book of the Metaphysics, called Lambda, Aristotle develops these conceptions and defends them by arguments to prove the actuality of the divine in nature; that is to say, the existence of God as thus conceived. Of this book, the first seven chapters are coher- ent doctrine; but the eighth, which superimposes fresh arguments that are clearly intended to clarify the preceding by a more minute analysis of the phenomena, introduces in so doing remarkable incon- sistencies and corresponding doubt, as if the scientist were driven by 248 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 26, NO. 6 force of detailed evidence to question his prior conclusions and recon- struct his theory. Jaeger shows that the coherent books are old, dating from the period of the treatise On Philosophy, but that the eighth is much later, the product of further studies finally matured. Its recon- sideration of the earlier doctrine is based on the mathematical analysis of celestial motions made by Eudoxus, from which this astronomer deduced the first purely scientific planetary theory—the system of homocentric spheres which Aristotle had accepted as the basis of his own physical cosmology. This showed that although all planetary movements could be conceived as if they were the resultants of uni- form circular revolutions variously compounded, each one of these composite movements was peculiar; and Aristotle found himself un- able to account for their striking differences by the assumption of a single physically definable cause. In the terms of his doctrine this meant that the question was left open as to whether the primary potency, the divine essence, were one or many; in short, whether or not it was possible to infer from the phenomena the presence of one Omnipotent God. Thus Aristotle’s natural theology broke down. Confessing failure, he abandoned it, leaving to his successors, in place of a definitive theology, an unsolved problem. His last words upon the matter imply a mood which closely approaches to agnosticism. He writes, referring to his own physical model of the cosmos: ‘‘Let this, then, be taken as the number of the spheres (the carrying spheres of the planets), so that the unmovable substances and principles may probably be taken as just so many: the assertion of necessity must be left to more power- ful thinkers.”’ And again in the Physics: ‘‘Motion, then, being natural, the first mover, if there is but one, will be eternal also; if there are more than one, there will be a plurality of such movers. We ought, however, to suppose that there is one rather than many, and a finite rather than an infinite number... and in this case it is sufficient to assume only one mover.” This expresses his final attitude toward the problem. The question is open; but for methodical reasons, and in accord with the principle of parsimony, it will be better to assume one ultimate essence. These statements do not at all suggest the temper which we associ- ate with theology. In Aristotle’s thought religious feeling is com- pletely dissociated from his theory of the cosmos; he clings instinc- tively to a monistic view of the world as the most of us do today; but perceives, as frequently we do not, that this view is not a neces- JUNE 15, 1936 BARRY: ARISTOTLE 249 sary inference from phenomena, but a postulate, which is to be justi- fied on methodical, that is to say, on pragmatic, grounds. I think you will agree with me that this final doctrine looks much more like theo- retical physics than anything else. A man of science is inclined to won- der why, even in its original form, it was ever called by Christians a theology. In reality this was an historical accident: a falsification of its original tone, due to the reintroduction by scholastic commentators of that religious feeling which Aristotle had excluded from it. The God he conceived,—a universal Potency actualized in celestial mo- tion, the pure Form that defines the nature of cosmic Power—what would this have been if a miracle had revealed it to his searching thought? Would it not have been what it is to us: the Form, or as we say, the Law, of Universal Gravitation? The critical revision of Aristotle’s theoretical works, which pro- duced the final texts of the Physica, the Metaphysica, the Analytica, the de Caelo and the De Generatione et Corruptione, was made after Alexander succeeded to the throne of Macedonia, and left the philoso- pher free to return to Athens, there to organize his own philosophical school, the Lyceum. Perhaps the most significant result of Jaeger’s critical work is his definitive reconstruction of this establishment. It was not at all a school of philosophy of the familiar Greek type, but something entirely new: an institute of scientific research, where stu- dents assembled, not to sharpen their wits and stimulate their crea- tive thought bylively dialectical disputation about the nature of things in general, but instead to listen to formal lectures: some on the biology and medicine of the day, illustrated by specimens of animals and plants, anatomical charts and similar demonstrative apparatus; some on meteorology or geography, or the history of events; others on mathematics, theoretical music and astronomy, or on the critical history of philosophy. Thus the natural sciences were taught each as a particular discipline and were coordinated by historical and critical studies. This education led directly to the prosecution of specialized research by those whose taste and aptitude encouraged it, each in a chosen field, and made possible certain coéperative in- vestigations in historical and biological studies directed by Aristotle himself. From the beginning, the new work undertaken by the school was exclusively scientific. Aristotle himself, with the probable assistance of his students, then wrote those works that mark the beginning of 250 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 26, NO. 6 scientific scholarship—the histories, based on records and inscriptions, which include his encyclopaedic survey of the constitutions of the Greek city states. Then also he compiled the Historia Animalium, and wrote his extensive treatises on the gross anatomy, the move- ments and the reproduction of animals, and on physiological psychol- ogy, together ‘with many shorter monographs on particular biological phenomena. In consideration of the fact, which Aristotle himself asks us to remember, that this was the first work ever done in com- parative biology, its results are admirable: its method is altogether predominantly inductive, its findings being based not so much on the evidence of common report as on the results of patient and careful observations, in part experimental and frequently remarkably accu- rate; and its critical tone, particularly in the discussion of prevalent beliefs concerning dreams, divination, prophecy and so on, is whole- somely sceptical. This work marks the dominance of the philosopher’s maturest thought by unequivocally scientific interests, and is con- vincing evidence of his ability in scientific research, even when this is unfairly judged by present standards. His critical syntheses of work in other fields is based, of necessity, on the findings of others, and only too frequently reflects the ignorance of his time, especially in me- chanics, where it is developed (probably in greater part by his suc- cessors) from reasonable but untested and erratic premises. Such defects, however, whenever the works in which they occur may be considered his own, are obviously to be attributed not to a persistent predilection for a priori reasoning quite inconsistent with the whole character of his later writings, but rather to the impossibility that one man could check the data in all fields of research. It seems to me, moreover, that, important as these investigations were, they are of less significance as contributions to knowledge than as evidence of the complete emancipation of critically philosophical thought at the beginning of the Alexandrian Age from the insidious influence of those mystical vagaries which, through the teachings of the Academy, had threatened and very nearly accomplished the total extinction of Greek natural science. That this calamity was averted was due to the unaided effort of a single great man whose penetrating acumen and conceptual grasp and whose uncompromising intellectual integrity were powerful enough not only to disencumber his own mind of these hypnotic influences, even against the persuasions of a vener- ated teacher reiterated during twenty years, but to turn the whole tide of philosophical inquiry back again through the original channel JUNE 15, 1936 BARRY: ARISTOTLE 251 of its flood to the disinterested investigation of nature. The achieve- ment was magnificent. But Aristotle not only turned the tide: by the organization of his school he ensured its continued flow. After him, his friend and succes- sor Theophrastus, criticizing destructively the metaphysical super- structure of his earlier philosophy, pronounced all theology tran- scendental and excluded it from scientific consideration; in the next generation Strato of Lampsacus likewise discarded the conception of final cause. A multitude of scientific works, inspired by the spirit of the master’s own researches, were produced by the school: the critical history of philosophy of Theophrastus, the history of mathematics of Eudemus, the botany of Theophrastus and his work on minerals, the theoretical music of Aristoxenus, the geography of Dicaearchus, and many others. It is often remarked that the Peripatetic school produced no men of great talent after Strato. This, I think, means that it produced no speculative philosophers of distinction; and this is true. But if its influence on the progress of thought be thus in- cautiously estimated, the judgment is quite fallacious. Those who in- herited the Peripatetic tradition and practised its methods were not philosophers, but scientists—it was the school itself which established this distinction—and among them were many of the very greatest thinkers of antiquity. The Alexandrian Age was the ancient Golden Age of Science. The empirical tendency of all Greek philosophy in the three centuries that followed Aristotle’s death is a very striking fact, and its association with an increasingly sceptical spirit no less so. The Museum at Alex- andria, founded by Ptolemy Philadelphus and organized in accord- ance with the sought advice of the Peripatetics Demetrius of Pha- lerum and Strato of Lampsacus, was something very like an enlarged Lyceum—another great institute of research. It became the most im- portant center of that renewed intellectual stimulation which ulti- mately gave to the world the finest products of Greek scientific genius: the technology of Ctesibius and Hero, the mathematics of Archimedes and Apollonius; the astronomy of Aristarchus of Samos, Eratosthenes, Hipparchus and Ptolemy; the anatomy of Herophilus and Erasistra- tus—which came in part directly from that Metrodorus who had taught in the Lyceum—and the experimental physiology of Galen; the cultural anthropology of which we have fragments in the works of Callimachus; the history of Polybius and of the doxographers who handed down the Peripatetic history of science; the geography of 252 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 26, NO. 6 Strabo; the literary criticism of Aristarchus of Samothrace and other distinguished scholars who edited and preserved the ancient classics and standardized the Greek language; and the sceptical philosophy of Pyrrho, Carneades, Aenesidemus and Sextus Empiricus whose criti- cal labors long perpetuated, even against the rising tide of super- naturalism which finally overwhelmed it, that methodical philosophy which, almost completely disregarded until our own time, we are forced to recognize as the explicit formulation of the principles of sci- entific judgment which we now universally—though not always con- sciously—accept as valid. In the entire range of human experience up to the time of Kepler and Galileo there is no intellectual achievement that can bear com- parison with this. Without doubt it was stimulated in particular ways—especially in mathematics and in medicine—by the work of Aristotle’s predecessors: by the latter Pythagoreans and the school of Eudoxus, by the medical schools of Cos and Cnidus, and otherwise; and doubtless also it was encouraged by the practical spirit that uni- versally characterized the early Alexandrian temper, and by the ag- nosticism that then prevailed among the intellectual classes. There is, however, the most conclusive evidence that no single influence af- fected it as significantly as that of the Peripatetic school, the school of Aristotle; which, guided by his undying spirit, determined the char- acter, and fixed the intellectual standards of the most powerful educa- tional establishments of the age. Such was the work of Aristotle. Interpreted by the medieval scho- lastics and by their spiritual descendants among modern speculative philosophers—whose predispositions have naturally led them to value most highly the early philosophical conceptions that linked his thought with that of the greatest religious thinker of antiquity—he has become almost universally known as the father of transcendental metaphysics. But the labors of an acutely critical scholar, clarifying at last the history of his intellectual life, now leave no room for doubt that, judged by the final outcome of his thought and labor, he be- longs not to the theologians but to us: the greatest of ancient meta- physicians who became in maturity the founder of critical natural philosophy, and of systematic scientific research. JUNE 15, 1936 | DUCKE: MYRISTICACEAE 253 BOTAN Y.—Notes on the Myristicaceae of Amazonian Brazil, with descriptions of new species. II.1 AnpoutpHOo Ducks. (Communi- cated by E. P. Killip.) Vriroua Aubl. Synopsis of the species (principally from Amazonian Brazil) according to the characters of the fruit and of the androecium. A. Fruit densely covered with soft-velvety subpersistent hairs which are ramified in numerous short lateral spurs; pericarp rather thin. a. Length of these hairs up to 8-9 mm Anthers connate, commonly subobtuse at the apex, slightly longer than the androecium column. 1. V. loretensis A. C. Smith. b. Length of these hairs up to 3-4 mm Anthers free in terminal part, a little divergent at the apex, very much longer than the androecium column. 3. V. divergens n. sp. The following probably should also be included in group A: ce. 2. V. mollissima (A. DC.) Warb. Female plant unknown. Indumen- tum nearly like in V. loretensis? Anthers connate, very much longer than the androecium column. Leaves very large. Eastern Peru. Not seen. d. V. urbaniana Warb. Female plant unknown. Anthers like V. diver- gens, according to the description. Goyaz. Not seen. B. Fruit covered with easily deciduous tomentum composed of very small (0.1-0.2 mm long) stellate hairs; pericarp thin. Anthers very much longer than the column, lineal or narrowed to the apex. a. Fruit ellipsoid, 11-16 mm long and 10-12 mm broad. Trees of upland forests and of ‘‘campos”’ woods. 4. V. sebifera Aubl., 5. V. mocoa (A. DC.) Warb., and 7. V. rufula Warb. b. Fruit almost globose, 7-12 mm in diameter. Tree of the periodically inundable forest. 9. V. cuspzdata (Benth.) Warb. ec. Fruit (according to Warburg) globose-ovoid, up to 27 mm long; not seen. Pozuzu (Peru). A male plant I found at Rio Purts (State of Amazonas, Brazil) belongs perhaps to this species. 6. V. peruviana GA] DC.) Warb. C. Fruit with indument as in B, ellipsoid, 15-18 mm long, 12-14 mm broad, with thin pericarp. Anthers nearly as long as the column. Forest tree of the hylaea. 10. V. venosa (Benth.) Warb. The two species V. sessilis (A. DC.) Warb. and V.. subsessilis (Benth.) Warb. are ‘‘campos”’ shrubs of central Brazil, remarkable for their very small size and for their sessile or subsessile leaves; their fruits are covered with a more or less deciduous tomentum. I have seen fruits of V. swbsessilis (State of Bahia, Zehniner 516=3054); they are oblong-obovoid, 15-20 mm long and 8-9 mm thick; their pericarp is a little thick but hard, covered with very minute stellate hairs. The fruits of V. sesszlzs are, according to Warburg, subglobose. D. Adult fruits glabrous, pericarp thick. Anthers as long as or shorter than the androecium column. | a. The thinner but more rigid coriaceous pericarp keeps its original form when dry. Fruits more or less globose or ovate-ellipsoid. 1 Notes on the Myristicaceae of Amazonian Brazil, with descriptions of new species. I. This JOURNAL 26: 213. 1986. Received February 15, 1936. 254 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 26, NO. 6 I. Fruits 15-22 mm thick. From Ceara to the Guianas and Lesser Antilles. 11. V. surinamensis (Rol.) Warb. II. Fruits 20-35 mm thick. From Rio de Janeiro and Minas Geraés to Santa Catharina. V. gardnerx (DC.) Warb. and V. bicuhyba (Schott) Warb. b. The very thick but more fleshy pericarp twisted and wrinkled ir- regularly and roughly when dry. Fruits generally ellipsoid. I. Fruits (of our specimens) 20-30 mm long. 12. V. carinata (Benth.) Warb. II. Fruits 30-40 mm long. 13. V. albidiflora n. sp. III. Fruits about 30 mm long. Male plant unknown. Northwestern Matto Grosso. Not seen. V. elliptica A. C. Smith. IV. Species of the affinity of the three last, but the female plant is unknown. Eastern Peru. Not seen. V. weberbauerz Mef. E. Fruit densely hispid-velvet with persistent hairs (ramified in numerous short lateral spurs), ovate or obovate-ellipsoid, about 30 mm long and 20 mm broad, with hard and thick ligneous-coriaceous pericarp. Anthers shorter than the androecium column. a. Length of hairs of fruits 0.5-1.5 mm 17. V. multinervia n. sp. b. Length of these hairs 2—2.5 mm 18. V. decorticans n. sp. ce. 16. V. multicostata n. sp., 15. V. crebrinervia n. sp., 14. V. minutiflora n. sp., V. rugulosa Spruce ex Warb., and V. mycetis Pulle may be- long in this group; the female plants are, however, unknown. I have not seen the two last species. F. Fruits very minutely rugose, covered with a thin but persistent tomen- tum of microscopic stellate hairs, oblong or obovate-ellipsoid, 15—20 mm long, about 10 mm thick; pericarp rather hard. Anthers about as long as the column, or shorter. 19. V. calophylla Spruce and 20. V. calophylloidea Mef. G. Fruit unknown. Anthers shorter than the column. The aspect of this plant is very peculiar and shows no affinity with the other species. 21. V. parvifolia n. sp. 1. VIROLA LORETENSIS A. C. Smith, 1931, floriferous (V. villosa Ducke, 1932, fructiferous). A small tree, remarkable for its few ramified, hanging, flexuous, often elongate inflorescences (up to 30 em when fructiferous), and for its partly long and rather woolly indument. The subglobose-ellipsoid fruits 2 cm long are densely velvety with soft, articulate and shortly spur- branched hairs, up to 8 or 9 mm long. This indument seems to be more or less deciduous after the fruits mature. The staminate flowers correspond rather to the section Amblyanthera of Warburg; the connate anthers have the apex slightly obtuse, their length (when completely developed) equaling or slightly exceeding that of the column. It grows in the inundable forest along the rivers and sometimes in very moist upland forests, in argillaceous soils. Its geographic area includes Amazonian Peru, where it was found near Iquitos by Killip and Smith, by Kuhlmann, and by myself, and the western- most region of the Brazilian Amazon (Lower Japura and Lower Javary), where, in the service of the late Dr. Huber, I collected specimens which are now in the Para Museum (Herb. Amaz. 6792 and 7419). 2. VIROLA MOLLISSIMA (A. DC.) Warb., of Eastern Peru, seems to be ~ eS Fs ae... JUNE 15, 1936 DUCKE: MYRISTICACEAE 255 allied to V. loretensis, but is easily distinguished by its very large leaves and by its anthers, which according to all authors are like those of the common V. sebifera. I have not yet found it. 3. Virola divergens Ducke, n. sp. Speciebus V. sebsfera (in Amazoniae dimidio orientali communi) et V. urbaniana (mihi solum e descriptione nota) affinis. A prima differt: partibus omnibus aliquanto maioribus et robusti- oribus, indumento densiore et aliquanto longiore, antheris parte superiore non connatis subdivergentibus, fructibus maioribus et longe densissimeque rufovillosis; a secunda (fructu ignoto) differt praesertim bracteis parvis caducissimis. Arbor parva vel mediocris; folia vulgo 20-80 cm longa, in individuis masculis saepius angustiora quam in femineis, saepe magis ob- longa et marginibus magis parallelis quam in specie V. sebsfera; costae secundariae vulgo utrinque 15-22, ut venulae transversae subtus pilositate longiore densiore et saturatius rufa notatae; inflorescentiae utriusque sexus saepe longiores et ampliores quam in specie citata; androeceum breviter stipitatum antheris 3 parte superiore non connatis subdivergentibus apice acutis; fructus maturus brevissime pedicellatus, globosus vel elliptico- globosus 2.5-3 em diametro, undique pilis mollibus detergibilibus pulchre rufis vulgo 3-4.5 mm longis articulatis et brevissime spinuloso-ramosis densissime subvilloso-velutinus, pericarpio mediocriter tenui sat fragili, seminis arillodio purpureo a basi laciniato, testa distincte sulcata. Habitat in silva primaria non inundabili circa Manaos, leg. A. Ducke: loco Estrada do Aleixo, florifera, arbor mascula 27-4-1932, Herb. Jard. Bot. Rio 24550, feminea 15-5-1932, Herb. Jard. Bot. Rio 24548; loco Estrada do Taruma, fructifera, 2-12-1932, Herb. Jard. Bot. Rio 24549). Prope Porto Velho fluminis Madeira leg. J. G. Kuhlmann 8-9-1923 (arbor feminea flor., Herb. Jard. Bot. Rio 24547). This species, when floriferous, may be mistaken for luxuriant and densely tomentose individuals of the common JV. sebifera; the anthers are, however, free at their apex, as in the central Brazilian species V. urbaniana Warb. It may be at once distinguished by its fruits, which are densely covered with rather long, articulate and spur-branched silky hairs of a beautiful red brown. V. loretenszs has the fruits rather similar but with much longer artic- ulate hairs; in all other characters it is very different. 4. VIROLA SEBIFERA Aubl. This widely distributed species is common from the Guianas through the whole State of Para to the State of Maranhao (Anil, near S. Luiz, Herb. Gen. Mus. Para 517); it grows principally in sec- ondary forest and in rather dry woods of the ‘“‘campos”’ regions. In the State of Amazonas it is less frequent; I have not found it around Mandaos. It is, however, mentioned by Warburg for the SolimGes region (Tiffé, the ancient Ega) and for eastern Peru (Tarapoto). I think I must refer to this species a tree of the upland rain-forest of Seringal Iracema (Rio Acre, Territorio Acre), Herb. Jard. Bot. Rio 24551, the leaves of which are more membranous and less tomentose than in the typical form. The geographic area of V. sebsfera includes the whole hylaea, the central part of Matto Grosso, the State of Goyaz, and the northern half of the State of So Paulo; I have seen herbarium specimens from all these regions. 256 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 26, NO. 6 5. Vrroua Mocoa (A. DC.) Warb. Our herbarium has now an excellent representation of this species, which is allied to V. sebsfera. It differs only in its non-cordate leaves, which have more distant nerves and more dense and uniform indument beneath. The fruits are not different from those of V. sebifera, their form being only a little more subglobose; they have the same ferruginous deciduous tomentum of exceedingly small stellate hairs. The tree, according to the collector, attains no more than 8 meters. Our specimens come from the type locality: Yurimaguas, Eastern Peru, Herb. Jard. Bot. Rio 23688 (male) and 24546 (female tree, with flowers and adult fruits), coll. J. G. Kuhlmann. 6. VIROLA PERUVIANA (A. DC.) Warb. A medium-sized tree of the scarcely inundable forest along the banks of the Rio Purtis above the mouth of the Rio Acre (State of Amazonas) corresponds rather well to the descrip- tion in Warburg’s monograph, but the leaves are membranous. Out of many floriferous individuals I have found only male trees (Herb. Jard. Bot. Rio 24552). 7. VIROLA RUFULA Warb. Our herbarium specimens, viz. 24495 (male), 24497 (female, floriferous) and 24496 (fructiferous), from Sao Paulo de Olivencga (Rio SolimGes), 24498 (male), from 8. Gabriel (Upper Rio Negro), correspond very well to the description in Warburg’s monograph, the type coming from Coary (Rio SolimGes). This species differs very little from V.. sebifera, but its leaves are a little more oblong-lanceolate, membranous, pale or glaucous, and very scantily hairy beneath; the fruit resembles that of V. sebifera, having the same dense but deciduous ferruginous tomentum of minute stellate hairs. It grows to medium size in upland forest, especially where the growth is less dense or partly secondary. We have also nos. 24541, 24542, and 24544, all male, from the upland forest northeast of Manaos, where the trees grow in the same conditions as above mentioned, which probably belong to the same species. Their leaves vary greatly in width, sometimes on the same branch; many of them have the broad form of those of V. sebifera, while others are narrow and ob- long-lanceolate as in V. cuspzdata: texture, color, and the faint tomentum are as in true V. rufula. I have not seen fructiferous material. No. 24545, male, from the Upper Rio Negro at the limit between the upland and the inundable vegetation, differs from the others only in its still longer leaf-apex. No. 24505, female, floriferous, from the Rio Ouro Preto, affluent of the Rio Pacanova, northwestern Matto Grosso, has more elongate leaves and more developed tomentum. | 8. VIROLA THEIODORA (Spruce ex Benth.) Warb., from Manaos, may not differ from V. rufula, but according to the description the leaves are broader and thicker. We may, however, note that the type consists of fructiferous twigs. Warburg considers the rugose or undulate testa as the best differential character, but the fruits he had seen were not ripe, and their seeds may have become rugose when dry. The tea-smell, referred to by Spruce, can be JUNE 15, 1936 DUCKE: MYRISTICACEAE 257 noted in dry leaves of many of the Vzrola species; it is not restricted to V. therodora. 9. VIROLA CUSPIDATA (Benth.)Warb. (V. elongata (Benth.) Warb.) Very similar to certain forms of V. rufula, but the trees are always small; the leaves are extremely variable in the width of the base but are more distinctly lanceolate-oblong or ovate-lanceolate, terminating often in a very long point; they are always less tomentose than those of V. sebzfera, but more strongly tomentose than those of V. rufula; the fruits are a little smaller than in either, and are globose rather than ellipsoid, having the same tomentum. This species is limited to the margins of lakes and rivers of the Middle and Upper Amazon, periodically subject to a long and deep inundation; in many parts of the Rio Negro, Trombetas, and Tapajoz it is a common and char- acteristic element of the flora. I have not as yet observed any other Myris- ticaceae in identical conditions except V. surinamensis, which is confined, in Amazonia, to the estuary and coastal region. I have observed V. cuspidata also in Upper Amazonia, at the Lower Javary, and Ule collected it at the Lower Jurua (5024) and at the Juruaé Miry, Territory of Acre (5709). Our herbarium specimens come from the environs of Manaos (Rio Taruméa, Herb. Jard. Bot. Rio 24467, and Uypiranga, 24770, male plants; Lake Marapata, 24469, fructiferous plant), from the Upper Rio Negro above the mouth of the Curicuriary (24446, male), from Boa Vista do Arary down to Itacoatiara (24465, male), and from Itaituba, Rio Tapajoz (24468, fructifer- ous). I fail to find any distinctions between V. cuspidata and V. elongata; both leaf forms are very often found in the same tree. The varieties pwnctata and membranacea are only insignificant forms. I do not know whether V. elon- gata var. subcordata should be included here, as I have only seen an old male specimen from the Cassiquiare region (Spruce 3172). 10. VrrRoLA vENOSA (Benth.) Warb. This species, widely distributed through the hylaea from the coastal region to the Upper Amazon, is an up- land forest tree, of small or middle size when in secondary formations, but rather large when in virgin growth. Frequent near Mandos (Herb. Jard. Bot. Rio 24500, male; 24499, female tree, floriferous and fructiferous); these specimens are exactly like the type, which I have seen. The ripe fruits are 15-18 mm long and 12-14 mm thick; they resemble in form and in the tomentum (which is ferruginous, deciduous, and composed of minute stel- late hairs) those of V. sebifera, having a little thicker pericarp; the red aril- loid is laciniate. These fruits are not described in Warburg’s monograph; they have nothing to do with the fruits of V. venosa var. pavonis Warb. (of Andean Peru), which evidently does not belong to our species, having axil- lary inflorescences (the true V. venosa seems to have always pseudo-terminal inflorescences, sometimes at the summit of leafless lateral twigs). V. venosa var. marta Warb., from the Japura, has also much larger fruits than our species and must be separated from it. 258 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 26, NO. 6 I have also gathered male plants of V. venosa: near Belem do Para (Herb. Amaz. Mus. Para 15849); at Barganca near the eastern Atlantic coast of Para (Herb. Jard. Bot. Rio 19569, slightly different from the species type in its leaf base, which is nearly as acute as in V. surinamensis); at Itaituba and along the lower cataracts of Rio Tapajoz (Herb. Jard. Bot. Rio 18630 and 2702); in the Lower Trombetas region (Herb. Amaz. Mus. Para 12042); near Tocantins, Upper Amazon (Herb. Jard. Bot. Rio 19570). Certain leaves of the last specimen have nearly parallel sides as in V. carznata, but the in- florescences are surely those of V. venosa. I have also seen a specimen col- lected at the Rio Marmellos, tributary of the Lower Madeira (Ule 6115, male). 11. VIROLA SURINAMENSIS (Rol.) Warb. This widely distributed and well known species grows in some of the Lesser Antilles, Trinidad, the Guianas, southern Venezuela, and the northernmost part of the Brazilian State of Amazonas (Rio Surumt, Ule 7988), the coastal region of Para in- cluding the whole Amazon estuary, the northern part of Maranhdo, and northwestern Ceara (Comarca de Granja, foot of the Serra Ibiapaba, in marshy ground near water, M.A. Lisboa, Herb. Mus. Para 2436). It is ex- tremely abundant in the low islands of the great estuary, inundable by the Atlantic tide; in some of these it represents the majority of the rather large trees up to 20 meters high. The enormous quantities of ‘‘ucuhtba”’ seeds yearly exported from Para, or there consumed in industries, come from this species. By its rather large globose, glabrous fruits, it is distinguishable easily from other species, certain forms of which may resemble it (e.g. V. carinata, V. venosa). 12. VIROLA CARINATA (Benth.) Warb. E STz = « ~ x J 4 -2.4h, ferent places conveyed a fairly consistent story but the more tranquil messages were not in general agreement, only those received at Berlin and those received at the Ebro Observatory being in fair accord.” 2 General references: (a) J. Bosuer, Les courants telluriques, Traité d’électricité atmosphérique et tellurique, publié sous la direction de E..M athias, 453-498 (1924); (6) JULY 15, 1936 GISH: TERRESTRIAL ELECTRICITY 271 Such was the status in 1922 when the Department of Terrestrial Magnetism of the Carnegie Institution of Washington, in order to further its program for investigating the electrical and magnetic phe- nomena of the earth, installed an earth-current measuring system at its magnetic observatory near Watheroo, Western Australia.*:* Since then this activity of the Department has been extended, first by es- tablishing another system at its observatory near Huancayo, Peru, in 1925.° Later through cooperation with the United States Coast and Geodetic Survey and the American Telephone and Telegraph Com- pany the registration of earth-currents was begun in 1931 at the Coast and Geodetic Survey magnetic observatory near Tucson, Arizona.® Registrations were also obtained at College, Alaska, from August 1932 to June 1934, through cooperation with the United States Coast and Geodetic Survey and the University of Alaska,’ and at Chesterfield, Inlet, Canada, through cooperation with the Meteorological Service of Canada.* The data from the two latter stations are of special sig- nificance because these places are both close to the Arctic Circle and because these projects were a part of that remarkable international cooperative program known as the Second International Polar Year. ~ Telephone and telegraph organizations have naturally been interested in these electric currents for a long time, but as a rule they have made no extended investigation of this class of phenomena. However, the Bell Telephone Laboratories have made a notable exception to the rule by conducting registrations of earth-currents during recent years at a number of places in the United States.° Some systematic meas ments have also been made during the past decade in Sweden.!® Dr. G. C. Southworth of the Bell Telephone Laboratories, in consultation with members of the Department of Terrestrial Magnetism, planned J. E. Burpank, Terr. Mag., 10: 23-49 (1905); (c) A. Nippoupt, Hrdmagnetismus, Erdstrom, und Polarlicht, Sammlung Géschen No. 175 (1921); (d) J. Bartets, Handbuch der Experimentalphystk, 25:1, 645-647 (1928); (e) B. Gutenberg, Lehrbuch der Geophystk, 429-433 (1929); (f) Harth-currents, Encyc. Brit., 14th ed., 7: 837-841 (1929); (g) O. H. GisH, Natural electric currents in the Earth’s crust, Sci. Mon., 32: 5-21 (1981); (h) O. H. Gisu, Les courants électriques naturels de l’écorce de la Terre et leur rapport avec le magnétisme terrestre, Cong. Internat. d’Electricité, Paris, 11¢ Sec., Comm. No. 1-C-2 (1932). References (a) and (b) contain thorough bibliographies up to the dates of their publication. Reference (h) cites original articles for the years 1925 to 1931, inclusive. 3 O. H. Gisn, Terr. Mag., 28: 89-108. 1923. 40. H. Giso and W. J. Rooney, Terr. Mag., 33: 79-90. 1928. 5 O. H. Gisu and W. J. Rooney, Terr. Mag., 35: 213-224. 1930. 6 W.J. Rooney, Terr. Mag., 40: 183-192. 1935. 7W.J. Rooney and K. L. SHerman, Terr. Mag., 39: 187-199. 1934. 8 B. W. Curriz, Terr. Mag., 39: 293-297. 1934. ®°G. C. SoutuwortH, Terr. Mag., 40: 2387-254. 1935. 10 T). Stenquist, Htude des courants telluriques Mém. Direction Gén.Télégr., Stock- aes Fasc. 1, (1925). Fasc. 2 (1930); Terr. Mag., 32: 148-145 (1927), 33: 205-209 272 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 26, No. 7 the program and devised the means by which long-distance telephone- lines could be used satisfactorily for this work without interfering with the use of the lines for telephone-service. The tranquil messages received in these more recent endeavors are in reasonably good agreement among themselves and also with those obtained at Berlin and at the Ebro Observatory. This outcome in- stills confidence in the technique which has been developed for receiv- ing them and enables one now to better grasp the broader significance of these messages. The features of this technique which deserve mention here pertain to the elimination of distorting effects of local origin, especially those which depend upon the manner of making the electrical contact with the Earth. In the installation at Greenwich, it will be recalled, that was accomplished by soldering the connecting wires to water-pipes. Plates of metals or coils of wire have been variously used instead of the water-pipes. Sometimes these electrodes are surrounded by char- coal or clay obtained from a common source, or a metal is suspended in a solution of one of its salts, the whole being contained in a porous jar, which is imbedded in earth. These are some of the means used in an attempt to avoid the battery-effect which generally exists between two plates of metal or other material when placed in earth. Even when all practical precautions are exercised, these effects may still be great enough to introduce important error in measurements of the character here considered. These contact electromotive forces or more briefly contact-poten- tials must be recognized for what they are and constantly kept in mind by investigators in order that interpretations be not confused by this extraneous feature. Although it has not been found possible to entirely eliminate such effects, yet it is feasible to control them within certain limits and to an extent such that their effect may not confuse the measurements of what seem to be the more important aspects of the true earth-currents. That this is possible has been clearly shown by results obtained at the Watheroo Magnetic Obser- vatory.! There, as in all installations which have been made under the guidance of the Department of Terrestrial Magnetism, pure lead wire or tubing is used as the material for making the contact with earth. This in the form of a flat spiral, or web, is placed in the earth five or ten feet below the surface and compactly imbedded in clay, 11Q, H. Gisu, Carnegie Inst. Wash. Year Book 27, 253-254 (1928); also C. R. Assemblée de Prague, 1927, Union Géod, Geéophys. Internat., Sec. Mag. Electr. Terr., Bull. No. 7, 247 (1929). JuLy 15, 1936 GISH: TERRESTRIAL ELECTRICITY 273 which in some cases is carted to the place for that purpose. The wire which leads from this electrode to the conducting line is very care- fully insulated from the surrounding earth so that the entire contact with earth is confined to the lead web at a depth where no appreciable short-time changes of temperature, humidity, soil-moisture, etc., oc- eur, and where mechanical action resulting from operations on the surface can exercise only a minor influence on the electrode. However, even after all precautions are taken, the slowly changing environment of the electrode may give rise to conspicuous changes in the contact- potential of the electrodes. The net effect of these upon the data de- rived from the measurements may be reduced by placing the points of contact farther apart, thereby increasing the part contributed to a given measurement by the true earth-current without a corresponding increase in that extraneous effect which is brought forth by the elec- trodes. The question may arise as to how these extraneous effects can be detected in the measurements or registrations. The answer to this calls for a description of another feature first used systematically at the Watheroo Magnetic Observatory and later at the Huancayo Mag- netic Observatory. The points for contact with the earth at Watheroo were selected so as to lie on true north-south and true east-west lines, respectively. Two pairs of points for each of these directions are now in use there (earlier there were three). Their positions and the dis- tances they span differ. Thus two independent records for each com- ponent of the earth-current gradient are obtained. If it is found by comparing these that the variations in the gradients are nearly equal, then it is almost certain that they are free from important spurious effects arising from the electrodes. For it is not to be expected that the conditions which give rise to variations in the electrode-potentials vary with position in just the manner required to yield the same varia- tions in gradients derived from two different pairs of points which represent different spans on the earth. The potentials which charac- terize the earth-currents may, however, be expected to be such a func- tion of position. Thus on the average one might expect the electrode- effects to be a smaller part of the measured gradient when the span between points is great. The nature of some spurious electrode-effects and the manner in which they may be detected are illustrated by graphs constructed from data obtained at Watheroo.'? These, which appear in Fig. 2, show variations during the day which are different for three pairs of points so situated that they should yield the north- 2 W.J. Rooney, Terr. Mag. 37: 363-374. 1932. l2 16 | 120TH EAST MERIOIAN TIME NORTHWARD COMPONENT OF Pe 1. 25-MILE "0; 2./S-MILE aro EASTWARD COMPONENT Q” Nz 2.05-MILE ot he aera 1929 JAN FEB MAR APR MAY JUN JUL AUG SEP OCT NOV DEC JAN F f WET SEASON ORY SEASON v/km 1 ° - 1 SCALES IN m 1930 EB MAR APR MAY JUN JUL AUG SEP OCT NOV DEC | wer season ORY SEASON ORY SEASON +75 ‘A Oy 991MM EASTWARD 0,0, 3.92 KM NORTHWARD SCALE IN mv OyPy 2.0/ KM NORTHWARD Oy My 3.29 KM EASTWARO Fig. 4.—Consistent messages. These were obtained at Watheroo in December 1927 after improving the receiving system. They show that inconsistencies such as are ex- hibited in Fig. 3 may be avoided and a high degree of consistency realized when suitable care is exercised. Fig. 5.—Unawvoidable irrelevant messages. The daily means of potentials registered at Watheroo during 1929 and 1930, shown in these graphs, have no common character- istic such as should be noticeable if they were related to a general flow of electricity in the earth. JuLY 15, 1936 GISH: TERRESTRIAL ELECTRICITY 275 ward component of the earth-current gradient. Those drawn in broken lines are obtained between points one mile apart, whereas that drawn solid is for a span of two miles. Since in computing the gradients, here expressed in millivolts per kilometer, the measured differences of potential are divided by the distance between the points, any effect which may not depend on distance would tend to be less in the case of the longer span. For this and other reasons the data for the longer span are regarded as relatively free from extraneous effects. If then one subtracts the values represented by the solid curve from corresponding values on the broken curves the differences should, in a measure, represent the extraneous effects. The differences thus ob- tained are shown graphically in Fig. 3. One striking feature of these difference-graphs is their consistent nature and the similarity in char- acter. They show a remarkable resemblance to the diurnal change in temperature at that place with, however, some lag in phase.'? Certain possible sources of these effects as suggested by this observation were examined and it was found that by improving the insulation of the buried portion of the copper conductor, which joins the line with the lead electrode, this feature could be completely eliminated. However, the resistance of the original electrodes was greater than desired, it being at times as much as 2000 ohms per electrode. This necessitated that the insulation between the copper conductor and the adjacent earth be better than with electrodes having lower resistance. Because of this and other considerations, an attempt was made to install new electrodes in such manner and position that they should have much lower resistance. The agreement obtained after these improvements were effected is shown in Fig. 4. While the evidence here shown dem- onstrates the possibility of obtaining satisfactory measurements of the more quiet aspects of earth-currents, yet it should also indicate the caution which must be exercised if gross errors are to be avoided. It is, however, not generally feasible to eliminate slower changes of the electrode-potentials. The character of these slower changes as observed at Watheroo are shown in Fig. 5. The ordinates in these graphs are daily means of the measured potential-differences (not gradients); positive values signify that the reference-point is at a higher potential than one to the east or to the north of it. When one observes that the absolute values as well as the changes in these po- tentials are generally independent of the distance and also of the di- rection of a point with respect to the reference-point, he necessarily concludes that the more constant part of these measured values, the 13 QO. H. Gisu, Carnegie Inst. Wash. Year Book 24: 214-215. 1925. 276 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 26, No. 7 daily mean, arises mainly from electrochemical effects of local origin and that this part of the measured quantity should therefore not be regarded as significant. These data give no indication of earth-cur- COLLEGE-FAIRBANKS, ALASKA CIW—C &GS—UA GREEN MICS MEAN BOERS NORTHWARD EARTH-CURRENT LOCAL NOON LOCAL NOON = ea i W he ie EASTWARD EARTH-CURRENT TUCSON, ARIZONA C&CGS—CIW—AT ET a | | 60 L AA ic =n (\ Nauta fe, NMI Rl | ii hil Ay I | V nih a Ty YT I Wi ‘mad ATT, i TANT Vid Ml ca Oe so a \ Ss 20 LOCAL NOON NORTHERLY EARTH-CURRENT i LOCAL NOO. rip a WP Pc Wee luca ri H I a = 7H ‘mall Wad WY ae) EASTWARD EARTH - CURRENT HUANCAYO, PERU ciw i+ MH LOCAL NOON NORTHWARD EARTH-CURRENT LOCAL NOON zo sine cs EASTWARD EARTH-CURRENT WATHEROO, WESTERN AUSTRALIA CiIW 15 Red NY. Nn Wr ADA, 10 > < S NS NORTHWARD EARTH-CURRENT LOCAL NOON pee NOON net rae EASTWARD EARTH-CURRENF Fig. 6.—Impulsive messages. Such as these, which were received April 30 to May 2, 1933, at four places ranging in latitude from 65° north to 30° south, announce events which occur simultaneously over the entire earth. JuLy 15, 1936 GISH: TERRESTRIAL ELECTRICITY 277 rents which maintain a constant direction and intensity at a given place. It may also be mentioned that the results obtained at the Huancayo Magnetic Observatory bear this out, but stronger evidence bearing on this point is obtained at the Tucson Observatory, where the lengths of lines are 57 km in the northerly direction and 94 km in the easterly. There the steady part never exceeds 20 per cent of COLLEGE-FA/IRBANKS, ALASKA CIW—CEGS—UA GREENWICH MEAN HOURS 16 Local 9 8 16 Local 9 1600 NOON NOON 1200 oe ly | ih Ah | \ % ys | iF 800% J ‘fi | | \ HY fl ‘ey oo = w~ Wit VA iN 400$ Ht o HORIZONTAL MAGNETIC FORCE TUCSON, ARIZONA €&GS—CIW—AT ET LOCAL ft is) S ~ 8 GAMMAS 9S HORIZONTAL MAGNETIC FORCE HUANCAYO, PERU OCAL LOCAL 400 NOON NOON F300, 5 ee . Ml 2008 a ae (2 Sean es S Ul Yan /” er, 1/00 l 2 HORIZONTAL MAGNETIC FORCE WATHEROO, WESTERN AUSTRALIA cay LOCAL LOCAL z NOON h, NOON 100% os HORIZONTAL MAGNETIC FORCE (7) ZA Fig. 7.—Impulsive magnetic messages. The magnetic effects shown here corre- spond to the electrical messages exhibited in Fig. 6. These appear to be magnetic and electric versions of the same narrative. the amplitude in the daily change and the absolute magnitude of that part of the measured potentials (always less than 0.3 volt) is such that it may be reasonably attributed to local electrochemical effects. This is the justification for omitting the daily means from the numeri- cal data and from further consideration in this discussion. The definition of earth-currents given in a well-known modern dic- tionary may be of interest here and will help to stress a point. It is as follows: ‘‘Light electric currents apparently traversing the earth’s surface, but which in reality only exist in a wire grounded at both ends, due to small potential-differences between the two points at 278 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 26, NO. 7 JANUARY 28=29 EARTH-CURRENT DISTURBANCES AND AURORAE, COLLEGE-FAIRBANKS, ALASKA, JANUARY 27 TO MARCH 22, 1933 20 LEGEND AND SCALES (ARB/TRARY) 0 ° (8) EARTH-CURRENT DISTURBANCE go [ CLOUDINESS: AURORA, QUIET lr [ MOVING z [ 7) s Ss JAN. 9, 1933 Jt | Le a 8 72 16 20 4 a. 16 20 = | 105? WEST MERIDIAN MEAN HOURS 105" WEST. MERIDIAN MEAN HOURS, ir 4Y=FIRST DIFFERENCES IN DIURNAL +/07 +20 TX VARIATION OF EASTWARD MAGNETIC . NORTHWARD COMPONENT COMPONENT ZAIN | iN +10 of af “0 ‘ \ DECEMBER ~~ \ v ee — se 2 i = es ANNUAL f R _OECEMBE: -/07- MAGNE T/C HORIZONTAL INTENSITY 1933 EARTH-CURRENTS 1932-1934 CALM DAYS ALL DAYS Lp 4A X=DIURNAL VARIATION IN____ sy. —' “| \ NORTHWARD MAGNETIC / ’ COMPONENT. e —/07 = T x cp ~ / i \ Ye \ DECEMBER sx, | EASTWARD COMPONENT L ie ye? DECEMBER For explanation of Figs. 8, 10, and 12, see bottom of opposite page. JuLY 15, 1936 GISH: TERRESTRIAL ELECTRICITY 279 which the wire is grounded.” Although this is not a model definition either as regards content or form of expression, yet the inaccuracy of the content may perhaps be excused on the ground that all so-called earth-current measurements contain an irrelevant element which may be dominant unless recognized for what it is. If then such nondescript grist is innocently put through a crude statistical mill, the meal which comes out may indeed spoil the pudding. One who is about to in- vestigate these phenomena of the earth would do well to start with the attitude of skepticism expressed in this definition, an attitude also recommended for more general application by Francis Bacon when he wrote, “‘If a man will begin with certainties, he shall end in doubts; but if he is content to begin with doubts, he shall end in certainties.”’ The information now accumulated enables one to view some of the broader aspects of the system of electric currents which circulate in the earth. One sees that most earth-current storms which are ob- served in the middle latitudes occur simultaneously everywhere on the earth as illustrated in Fig. 6. Comparing these with the corre- sponding magnetic records reproduced in Fig. 7, one notices a pro- nounced similarity in the character of the magnetic and the electric records. When one is disturbed, the other is also disturbed. This, as well as the similarity in the character of the disturbances, is obviously not a mere coincidence. The correspondence between the occurrence of aurorae and dis- turbances in earth-currents is represented by Fig. 8. Here the dark blocks represent by their height the degree of disturbance in the earth-currents. The vertical lines above these indicate the occurrence of aurorae, their length being proportional to the brilliance of the aurora. The zigzags indicate moving forms.’ Further evidence of a relationship between earth-currents, aurora, and possibly solar ac- tivity is presented in Fig. 9. This is the result of a statistical study which was designed to determine whether there is a tendency for earth-current storms or polar lights to recur after a period of time, a sort of return engagement.'* The curves in the right-hand side of 144 W. J. Peters and C. C. Ennis, Terr. Mag., 31: 57-70. 1926. H. U. Sverdrup, Res. Dep. Terr. Mag., 6: 510-512. 1927. Fig. 8.—Electrical messages pertaining to polar lights. Displays of polar lights and certain electrical effects observed at College, Alaska, January 27 to March 22, 1933, appear to be related. Fig. 10.—Tranquil messages. This specimen received at Tucson, Arizona, January 9 to 11, 1933, represents a type of message which is repeated day after day, year after year. Such messages vary in intensity with the seasons and may be obscured on occa- sions when impulsive messages are interjected. Fig. 12.—Character changes with season. This is especially pronounced at Tucson in both the electric and the magnetic elements, and is equivalent to a shift of latitude. 280 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 26, NO. 7 Fig. 9 indicate the relative frequency at which a storm may follow another storm at intervals of 25 to 33 days. It is noticed that this fre- quency is greatest at about 27 days after the first occurrence in the case of earth-currents. Although for the aurora the indications are not so positive, yet there appears to be some evidence of a similar recurrence. It should be recalled in this connection that 27 days is the time required for a sunspot to rotate with the sun. A relationship is also found between the variations in earth-currents, the activity of ae tolsea | ||| Bees eer [7 Feed ale Pelee a POLAR LIGHTS, “MAUD” EXPEDITION ane ciea OFF SIBERIAN COAST. 1922 TO 1925 basa Sema kd | | | | seconoary | || PURSE ieimsasiiyo |b. ia a a | Fig. 9.—Impulsive messages repeat. The evidence presented in this chart shows that if an impulsive message (an earth-current storm) is received on a certain day, then there is a high probability that another will be received 27 days later. A similar tend- ency is found for terrestrial magnetism and for polar lights. MILLIVOLTS PER KILOMETER AURORAL CHARACTER-NUMBERS the earth’s magnetism, and the occurrence of spots on the face of the sun. They all run through a cycle which has a period of roughly eleven years. When sunspots are numerous, magnetic changes are greater and more frequent, and the earth-currents undergo more in- tense and more frequent fluctuations than at times when sunspots are less plentiful. From such observations one concludes that these earth-current disturbances must arise out of an influence which is ca- pable of acting directly on the whole earth at once and that the activity on the sun in some way influences the electric currents in the earth. When one examines the more quiet aspects of earth-currents, he finds there regular changes during the day such as appear in Fig. 10. These undergo some modifications from season to season, and they 16 QO. H. Gisu, Sci. Mon., 32: 5-21. 1931. JuLY 15, 1936 GISH: TERRESTRIAL ELECTRICITY 281 wax and wane during the years. The average character of the more. tranquil changes observed at Tucson, Arizona, are shown by the graphs in Fig. 12. If the amplitudes of these wave-like graphs are chartered for different places and different times of the year, graphs such as those shown in Fig. 11 are obtained. It will be seen from this that the amplitude is a minimum in midwinter whether the stations be north or south of the equator and that in general the values for summer tend to be high, yet there appears to be a tendency for large { SaNGee anual ° WAT By - 1924 -1927 SES BERLIN 1864 - 1887 NOES Zen GREENWICH 1865, 1866 ee Al ies [SS Fig. 11.—Intensity changes with season. The tranquil electric effects are generally more intense in summer than in winter. values to occur near the time of the equinoxes. There is also evidence that the amplitude of daily change varies with a period of about eleven years and that this corresponds approximately with the varia- tions in sunspot number."* The diurnal changes in earth-currents and terrestrial magnetism at Tucson, Arizona, are compared in Fig. 12. Several points of correspondence will be noted in these. The rate of the diurnal change in the westward magnetic component closely re- sembles the diurnal change in the northward earth-current com- ponent. The manner in which both the magnetic and the electric effects change with season at this station is of special interest in that it reveals a further relationship between these two classes of phe- nomena. When one attempts to ascertain from the data just what relation- 16 Carnegie Inst. Wash. Year Book 34: 234. 1935. 282 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 26, NO. 7 ship exists between earth-currents and terrestrial magnetism, he is confronted with some difficulty. This is especially pronounced in the case of earth-current and magnetic storms. During these storms the changes in the earth-currents are sometimes of the same character as those in the corresponding component of terrestrial magnetism, the two increasing or decreasing in unison (see Figs. 6 and 7). Then again they differ considerably in character although the duration of the dis- turbed periods corresponds. If the magnetic changes were due to elec- tric currents in the earth, then they should be roughly proportional to the electric changes; thus the graphs which represent the magnetic changes (Fig. 7) should be about the same shape as those for the earth- currents (Fig. 6). However, if the magnetic changes produce the earth-currents, the relationship would be quite different. The earth- currents would then be roughly proportional to the rate of the mag- netic changes. Thus, even though the magnetic disturbance be large, if it is changing but little the earth-current at the corresponding time would be small. A comparison of the observed storm-changes in earth- currents and in the earth’s magnetism therefore seems to indicate that sometimes one relationship holds, sometimes the other. Viewed superficially this may be taken to indicate that part of the time the earth-currents are the cause of the magnetic changes and part of the time the result of those changes. This apparent duplicity of character, together with the inconstant nature of earth-current storms, are ob- stacles which stand in the way to a comprehension of them. Since it is not proposed here to venture far into the free and airy realm of speculation, we shall leave this aspect and turn to a further examina- tion of the relation between the more quiet aspects of earth-currents and the corresponding changes in terrestrial magnetism. When one compares the diurnal variations in these two phenomena, he finds some evidence for the view that the slow periodic changes in earth-currents may be in the main induced by magnetic changes simi- lar to those observed. Thus in Fig. 12 it is seen that the rate of change of the eastward magnetic force at Tucson is of a very similar character to the variations in the northward component of the earth-currents.°® Such simple comparisons are of course inadequate for more than one reason. Any electric current must produce some magnetic effect so that it is reasonable to expect that the magnetic changes are not wholly in- dependent of the earth-currents which seem in some way to arise from the magnetic changes. The mathematical relations which take into account such reactions between the inducing magnetic changes and JuLy 15, 1936 GISH: TERRESTRIAL ELECTRICITY 283 the induced current were worked out by Chapman and Whitehead." That relation enables one to calculate the components of the earth- currents, or rather the gradient which is proportional to them, from /2 LOCAL MEAN HOURS a 8 ARBITRARY UNITS NORTHWARD |\COMPONENT fone ® i“ > as | NN a 1 Ss a WATHEROO AND EBRO in Ss v/kM) SCALE OF CALCULATED VALUE S(MV/ocM} SCALE OF OBSERVED VALUES (M' / ie ae i ef NZ OBSERVED VALUES N — CALCULATED VALUES Fig. 13.—A prediction after the event. The solid curves represent the average diurnal-variation in the tranquil electric effects. The broken lines are those predicted wholly from the tranquil diurnal-changes in terrestrial magnetism. 17 Cambridge, Trans. Phil. Soc., 22: 463-482. 1922. 284 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 26, No. 7 the observed magnetic changes. In Fig. 13 these calculated values of the earth-potential gradient are compared with those which have been observed.!® There seems to be fair qualitative agreement for the north- ward component but not for the eastward component. The values calculated for Watheroo are six times those observed. For Ebro the calculated values are one-fifth those observed. No quantitative com- parison is possible in the case of Berlin because the observed values are expressed in arbitrary units.‘ Although the agreement is not as good as one should like, yet there seems to be evidence here to sup- port the view that the regular daily changes in earth-currents are in the main induced by corresponding daily changes in terrestrial mag- netism. According to Chapman and Whitehead, the difference be- tween calculated and observed values of the eastward component may perhaps be attributed to an inadequacy of the magnetic data. Re- ferring to the formula used in the calculations they say “‘the higher harmonics in the magnetic potential are perhaps hardly sufficiently well determined to bear the weight given them in this formula.” Irregularities in the structure of the earth’s crust are not considered in making the calculations, yet these present contrasts in electrical conductivity which certainly distort the electric flow. Perhaps the most pronounced large-scale contrast of this nature is that between land and sea, the conductivity of sea-water being several orders of magnitude greater than that of land.!® Other currents of more or less local extent and of quite different origin may at places be superim- posed upon the more general system, thus adding to the complexity (see reference 15). Furthermore, one should not neglect to question the reliability of the earth-current data provided those data were ob- tained in such a way as to give no criteria by means of which it may be ascertained that extremely local phenomena, especially such as are produced by the method of measurement, have been eliminated. Such modifying influences may account for some of the disparities which have been noted. In any case, the induction theory is the only one now in sight that can claim attention. One may therefore tentatively entertain the view that the earth- 17a Before sending this to press the values of these data were converted to conven- tional units on the basis of information found in a paper by von Stephan. Berlin. SitzBer. Ak. Wiss. No. 39: 553-561 (1886); see also O. H. Gisu, Terr. Mag., 41: 87-88 (1936). The Berlin data now appear to be only 13 per cent greater than those derived by CHAPMAN and WHITEHEAD. 18 Calculations for Greenwich and Berlin are given in reference 15, those for Ebro by S. Cuapman and T. T. Wurreneap in Terr. Mag., 28: 125-128 (1923), and those for Watheroo by GisH and Roongy have not been published in detail. 19 H. ErTEt gives a helpful theoretical discussion which takes account of contrasts in the electrical conductivity of the earth. Berlin, Veréff. Met. Inst., No. 391 (1932) . JuLy 15, 1936 GISH: TERRESTRIAL ELECTRICITY 285 currents which are observed are in the main induced by magnetic variations, but that their strength and direction are modified in a manner which varies from place to place and which depends upon the distribution and configuration of oceans and continents as well as upon other structural features of the earth. Modifications produced in the earth-currents by the deep structure of the earth’s crust may thus constitute electrical messages which contain information about conditions in that little-known region. LOCAL MEAN HOURS s) EG Beet N ete Lia JSS Sle SoMa ( ASER NILA? mee Ts AGH Mgesanaee 60° ai. 7 , = ~ ’ ? ~ s - ~ ¢ x ‘ Me OLN hae SELON + 4 A , Nain lt) . - ~ \ ’ - . Lu) reer a (ter U ties ’ Le al) fh CN t ‘ ' ‘ ‘ ¥ = 1 1 § ‘ ' ‘ 1 SP Fig. 14.— Electric currents beyond the stratosphere. Electric currents in the high atmosphere flowing in extensive eddies like those depicted in this world map by Bartels would account for the average tranquil diurnal-changes in terrestrial magnetism. This current-system is fixed relative to the sun and hence the earth rotates within it. The magnetic field of different parts of this system passes over a given place at different times of day, thus bringing a regular daily change in the earth’s magnetism. The magnetic forces which, on the view just outlined, induce the earth-currents have their immediate origin in the high atmosphere in about the same region which reflects radio waves. If a system of elec- tric currents having the character represented by the diagram of Fig. 14 circulates in that region of the atmosphere, it would be capable of producing the daily magnetic changes which are observed at the sur- face of the earth. To justify the assumption that there exists such a system of currents or any equivalent which may produce the corre- 286 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 26, NO. O 4 8 NOON 16 20 | | | | | | | | 90° 120° 150° /80° 150° 120° 90° 60° POLE UNIFORM | WARY Ke MAGNETIZATION if IZ WL, OE I \_¢ 7g) wy Ro be WG Bh ood at? ee ar N Z g YoU [on C7 , ; q Be ey Ta ) AA OLIAZ YY HOKE 1Z Yj | if x NS \ \ i fo “4 MM come 4 EBC\ | SS [iN ean Nl Ze a SI SS SSN S — t A @=C/W AND COOPERATING | STATIONS @ = OTHER STATIONS H| - ~ mC aed FuicAtel eA ene BS ae ia) Mbt po a Se 90 160 che ° uz. gt” POLE UNIFORM MAGNET/ZAT/ION 8 YU G7 ¢ Ey ka ty < ® ist Tee : > x iiP a Z | fe GAS Ys Ay ~ "YY YyY if “Ze UY Ca LP ay! or GS: oT /V/ } Ak ; ES: yi a vi is Oo ay) Y H 3 St ‘ 2) br aco eS \R im AN ‘LGA eG iD at | if j NE Ie Na eae | WZ VARS i? @ =C/W AND COOPERATING i STATIONS --4@ = OTHER STATIONS _|___ a ls Slee 9 Figs. 15 and 16.—An interpretation: electrical eddies in the earth. The tranquil electric messages as interpreted here refer to electric currents which circulate in ex- tensive eddies in the earth’s crust. These, as viewed from the sun at about the time of the equinoxes, are shown here diagrammatically—(a) for daytime over the Western Hemisphere (Fig. 15) and (b) for daytime over the Eastern Hemisphere (Fig. 16). mE, Jouy 15, 1936 GISH: TERRESTRIAL ELECTRICITY 287 sponding magnetic effect would carry us beyond the scope of this dis- cussion, since it is here desired to simply point out somewhat of the mechanism by which the electric currents in the earth may be in- duced. The portion of this system of currents which appears at the center of the diagram is always directly under the sun and therefore the whole system moves relative to the earth, making a rotation once each day. The magnetic field of this electrical circulation as viewed from outside of the earth is one which in its principal features does not change appreciably with time, but, since it is moving relative to an observer on the earth, it appears to him to undergo a regular diurnal variation. This magnetic field, together with the earth which rotates within it, constitute the electric machine which generates the electric currents in the earth. Thus one might expect to find in the earth a general system of electrical circulations which is related to that repre- sented for the higher atmosphere. It has recently become possible to construct, on the basis of ob- served data, a world picture of the electric currents which circulate in the earth. This picture shows a number of great electric eddies (see Figs. 15, 16, 17, and 18). Eight of these are located in the middle latitudes, four in the northern hemisphere and four in the southern hemisphere, symmetrically placed on either side of the equator with centers about equally spaced in longitude and lying along a parallel of latitude near the tropics of Cancer and Capricorn (see Figs. 15 and 16). Four more such eddies with centers in the Arctic are also dis- closed (see Figs. 17 and 18). Although there are no data to establish . the fact, it seems likely that there are also four corresponding eddies in the Antarctic. All of these eddies follow the sun in such a way that there are eight on the sunlit side of the earth and eight on the dark side. The curves which outline the eddies are constructed in such a way that two adjacent curves indicate the boundaries of a tube of flow. Those tubes which are bounded by solid lines all contain the same amount of current except that in the case of the innermost curves the flow is sometimes less than that for a full tube. In order to _ show some of the weaker eddies it was necessary to subdivide some of the tubes. These are outlined by broken lines. The direction of flow is that of the arrows. Current-systems corresponding to the charts which are exhibited here would completely account for the average diurnal variations ob- served in earth-currents at Watheroo, Tucson, and Chesterfield Inlet. Similar charts have been constructed from other sets of data; in fact, all the principal series of earth-current data have been examined in JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 26, NO. 7 288 208! | “(ST “SIq) roy dsiue yy ULOISVY IY} Ul oUITJAVpP 10] (q) pue (ZT “SIq) oLoYdstUIOF] U19}Sa A OY} Ul OUITZABP JO} (B) —o10ydsiuleyyT UlOyJION 94} UI SoIpps oyy, “yoodse ivjod ey ,— sg pur JT ‘“Sssiy NOON | oy «il (i) OE L PHY Uy, My i . 1/ iy SS WT Hl te pi ae HM ) : Dg ap Y LK : } iiss is) Ny yi ee) My fis Ss me pay Ti i SX bs pee ay , SN i (fay: bp j i Oram: PAB NOILYZ/ Xv Wt b aa p NOILEZ/L INV WHOJINA F10d oo CATT TTY i ie IK Xe af / ! DOS | ARN a WMA Mifyeg: NN TN, il Tan ae ATAU NPY hed Nui HE eS Wir ate Co : Ore h hs DN AN s 2 fll is i es 1) he a Monee cay aninisting Uy Yi SNO/LYLS YFHLO = @ <> SNO/LVLS | ON/ILVYISOOD ONY M/D = @ / MIDNIGHT MIONIGHT JuLY 15, 1936 WOOD AND FERGUSON: WUSTITE 289 the same manner and without exception they are consistent with a scheme having the general features here depicted. In order that this picture may the better represent the observed facts, the tubes of flow must be regarded as very flexible, easily deformable, so that the shape _ of the tubes may be readily distorted and even the centers of the ed- dies displaced in such a way as to conform with the distribution of the electrical properties of the Earth, especially of that pertion which con- stitutes the more immediate environment of a given eddy. It also seems likely that the development and the orientation of the eddies should be in some relation to the magnetic axis of the earth. When the general aspects of pertinent earth-current data are viewed in such a perspective they will, it is believed, be seen to be consistent with the principal features of the interpretation which the charts are designed to convey. Returning again to an examination of the charts, it will be noticed that the current in the daylight eddies is considerably greater than that in the others; at least this is true for the eddies located in the middle latitudes. The centers of the forenoon eddies of the middle latitudes are approximately on the meridian for which the time of day is 9 a.m., while the afternoon eddies center on the meridian for which the time is about 3 p.m. Considerable flexibility must be allowed for this feature. The circulation in the forenoon eddy of the northern hemisphere and that in the afternoon eddy of the southern hemisphere are clockwise, whereas in the other two daylight eddies the circulation is counter-clockwise. A similar description applies to the nighttime eddies. The circulation in the eddies of the Arctic region is in the same sense as the corresponding eddies of the middle latitudes in the north- ern hemisphere. As these eddies move relative to the earth the direc- tion and intensity of the earth-current at a given place change, those changes depending upon the position that place may occupy in the eddies and hence depending also upon the latitude of the place. This then is a world view of the gross aspects of the quiet-period earth- currents—the most comprehensive interpretation thus far made of the tranquil electric messages from the earth. It is, however, but a beginning; much deciphering remains yet to be done. CHEMISTRY.—Notes on the preparation and composition of wustite phases! E. E. Woop and J. B. Fera@uson, University of To- ronto. 7 1 Received March 25, 1936. 290 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 26, NO. 7 THE DEHYDRATION OF FERROUS HYDROXIDE Ralston (12) has suggested that ferrous hydroxide might be a suit- able material from which to prepare ferrous oxide. Pure white ferrous hydroxide was prepared from aqueous solutions of ferrous sulphate and potassium hydroxide in an atmosphere of nitrogen. It was filtered, washed with water and then dehydrated in vacuo, the whole opera- tion being conducted in a closed apparatus. Great care was taken to exclude extraneous oxygen and the ferrous sulphate solution was pre- pared by the method of Benson (2). The relative amounts of ferrous and total iron were determined on the same sample without weighing by the electrometric method of Hostetter and Roberts (8). The dehydration experiments were made at temperatures ranging from 100 to 180°C. and for periods of time varying from 95 minutes to 7.5 hours. We considered the reaction velocity too small at tem- peratures below 100°C. to warrant work at lower temperatures. The final products always contained ferric iron which was larger the higher the temperature of dehydration. It also increased with time up to a certain point and then remained constant at constant temperature as though the oxidation were directly related to the production of water vapour which was removed by a suitably protected vacuum pump. Welo and Baudisch (13) seem inclined to attribute the formation of ferric iron, at least in part, to the decomposition of ferrous oxide to form magnetite and iron. We found no trace of gas when a sample was dissolved in acid and doubt if free iron were present in our prod- ucts and would prefer to regard a wustite as a primary product. Since our most reduced materials only contained 80.4 percent ferrous oxide, this method of preparation is no better than many others and the technique is very much more difficult. THE DECOMPOSITION OF FERROUS OXALATE Andrew, Maddocks and Howatt (1) claim to have prepared ferrous oxide, 99.5 percent pure or better, by heating the oxalate in vacuo at the proper temperature. Jette and Foote (9) were unable to dupli- cate their work. We made a number of experiments and although we varied our procedures in many ways, we were also unable to duplicate the work of the former. One of the stumbling blocks in such work is the analytical method and Andrew et al. may not have obtained the pure ferrous oxide which their analytical results seem to indicate. But the work of Jette and Foote and of ourselves shows that their method is not a procedure for JULY 15, 1936 WOOD AND FERGUSON: WUSTITE 291 the preparation of pure ferrous oxide which can be employed with any reasonable assurance of the production of a pure oxide. THE REACTION BETWEEN IRON AND MAGNETITE The early observations of Chaudron (4) indicated that iron and magnetite, formed as decomposition products of wustite, would read- ily re-combine at higher temperatures. Mulligan (6a) also observed in our laboratory a similar re-combination. But the later work of Chaudron and Forestier (5), as Ralston (12) points out, does not seem to confirm these observations. In order to be sure that our results might not be determined by a particular wustite sample, we selected a number of materials which differed in preparation and composition for this study. The total and free iron values for these were respectively as follows: A, 75.8 and 0.26; B, 79.2 and 9.1; C, 77.4 and 5.1; and D, 78.7 and 9.5 percentages respectively. These samples were sealed in Pyrex glass tubes in vacuo and heated for five hours at 515°C. (D, decomposed) or for five hours at 610°C. (R re-combined). The tubes were removed from the furnace and chilled as rapidly as possible. The products were analysed by the electrometric method for total iron using potassium dichromate and by the mercuric chloride method (7) for free iron. Mulligan had pre- viously used the copper sulphate method for free iron. The results are given in Table 1. There are certain irregularities in this table. The TABLE 1.—THE FREE [RON CoNTENT OF HEAT-TREATED WUSTITE. SAMPLE. A. B. C. D. —S— weight percent a Initial 0.26 eal yell Deh D 12.0 23.65 13.75 20.75 R 0.1 10.2 D.9 7.8 D Jou) 16.4 Geo 11.2 Rk 0.35 walk i.4 6.1 second decomposition and re-combination did not yield the same products as the first. In C and D, the final products contained less free iron than the original materials showing that some of the mag- netite and iron formed during the original preparations of these sam- ples had also re-combined. But there is ample evidence that the re- combination of such iron and magnetite is easily obtained and the earlier observations are thus confirmed. The further question, whether or not iron and magnetite from ex- traneous sources will readily combine, was also investigated. A finely 292 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 26, NO. 7 ground electrolytic iron and a sample of Kahlbaum’s magnetite were selected for this study. The former had been reduced in hydrogen at 800-900°C. and was originally from 100 to 200 mesh material. The latter was an extremely fine powder containing 24.07 percent of iron in the ferrous condition, the theoretical value being 24.12. Various mixtures were heated in vacuo in sealed Pyrex glass or silica glass con- tainers which were removed from the furnace and quenched as rapidly as the containers would permit without breakage. Gas usually formed from the charge in the initial heat treatment and a slight oxidation occurred. We attributed this to air adsorbed on the fine magnetite powder. Because of this gas production and oxidation, the composi- tion of the charge was adjusted by the addition of free iron after this first treatment in cases in which a definite total composition was re- quired. The samples were also ground in an agate mortar in nitrogen between heat treatments in order to uncover residual iron and also prior to analysis and to magnetic separation when this was made. In some experiments with silica tubes, it was found advisable to hold the charge in a container of black sheet iron inside the silica tube to prevent contamination by flakes of silica when the charge was quenched. Complete combination was not observed in samples high in iron. A sample which initially contained 19.4, percent free iron contained 1.63 percent after a heat treatment of 7.75 hours at 691°C. Complete com- bination was obtained at 1097+10°C. with samples which initially contained 10.9; and 8.4, percent free iron respectively, and with the former sample (10.97) at 926+3°C. A serious attempt was made to prepare pure ferrous oxide by the complete combination of iron and magnetite in the proper proportions but without success, the prod- ucts always containing free iron and ferric iron. Since pure ferrous oxide could not be prepared in this manner, attempts were made to find the limits of the wustite field, next to iron, on the iron-oxygen diagram through the use of samples containing initially an excess of iron. The final products were separated by means of a strong electro- magnet. A sample which contained initially 80.1 percent total iron was heated for 8.5 hours at 683°C., 5.5 hours at 693°C. and 6 hours at 703°C. The non-magnetic portion contained no detectable free iron, 77.03 percent total iron and 70.6, percent iron in the ferrous con- dition. This result agrees fairly well with the value of 76.9 given by Pfeil (11) and also with the value indicated for this temperature on the diagram given by Mathewson, Spire and Milligan (10). Pfeil indi- cates a vertical boundary line at 76.9 percent and Jette and Foote (9) JULY 15, 1936 WOOD AND FERGUSON: WUSTITE 293 found evidence of free iron in samples containing 77.4 percent iron which had been heated respectively at 600, 795, 915 and 1035°C. From the ferrous iron content of samples prepared by us at 1097 and 926°C. we obtained points at 77.0; and 76.9 percent total iron which would agree with Pfeil’s observations. Since the invariant point, Fe + liquid=wustite at 1380°C. determined by Bowen and Schairer (8), lies at 76.8; percent total iron, the boundary line, if vertical in the upper range, must lie nearer iron than this percentage. Additional experiments at 727°C. indicated that the wustite field extends, at this temperature toward magnetite to at least 76.0 percent total iron and probably as far as 75.2 percent. Since the weight of evi- dence at present available seems to favour the opinion that pure fer- rous oxide is unstable at and below the invariant point at 1380°C. and observers have disagreed upon the exact extent of the wustite field, the question as to whether or not the most stable wustite was obtained in many experiments is a pertinent one. It was our reference (6) to such wustites which has been misinterpreted by Ralston (12) to mean different allotropic forms of ferrous oxide.” But there can be no doubt that in many cases unstable wustites must have been formed if the compositions which have been reported can be considered at all reliable. We cannot say that our wustites were the most stable solutions, we can only say that, with this new procedure, we have been able to prepare products which compare favorably with those pro- duced by other methods and that the procedure is relatively simple and convenient. REFERENCES ANDREW, Mappocks and Howartrt, J. Iron and Steel Inst. 124: 285. 1931. Benson, Univ. of Toronto Studies, No. 25. 1902. BoweEn and ScHarrEerR, Am. J. Sci., 24: 177. 1982. Cuaupron, Ann. Chim., 16: 221. 1921. CHAUDRON and ForestiER, Comp. Rend., 178: 2173. 1924. FERGUSON, (a) This JouRNAL, 13: 275. 1923. (b) Trans. Faraday Soc., 21: 240. ee ees 1925. GROEBLER and OBERHOFFER, Stahl u. Eisen., 47: 1984. 1927. HostettTer and Roserts, J. Am. Chem. Soc., 41: 1337. 1919. JETTE and Foote, J. Chem. Phy., 1: 29. 1933. 10. MatTHEwson, Spire and Miuuiean, Am. Soc. of Steel Treating, 19: 66. 1931. 11. Pret, J. Iron and Steel Inst., 123: 237. 1931. 12. Rauston, U.S. Bureau of Mines, Bull. No. 296. 1929. 13. Weto and Baupiscu, Chem. Rev., 15: 45. 1934. al ei 2 Misquotation admitted in private communication. 294 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 26, NO. 7 ZOOLOGY .—Echinoderms collected by Capt. Robert A. Bartlett in the seas about Baffin Land and Greenland. Austin H. Cuark, U.S. National Museum. Capt. Robert A. Bartlett’s investigations of the waters about Baffin Land and Greenland carried on during the past few years have re- sulted in bringing together a noteworthy collection of echinoderms which greatly increases our knowledge of the distribution of these animals in that section of the Arctic. Especially noteworthy are the records from no less than eleven stations south and west of Baffin Land, previously unknown terri- tory so far as the echinoderms are concerned. Scarcely less interest- ing are the records from northwestern and eastern Greenland, regions where, because of the formidable difficulties to be overcome, little collecting ever has been done. The localities at which he collected echinoderms were the following: LABRADOR: |. Saglak Harbour. West GREENLAND: 2. Disko Island. 3. Northwest of Upernivik (lat. 74° 21,’ IN:, long. 602307 W.): BaFrFrin LAND AND VIcINITY: 4. Three miles south of Salisbury Island, Hudson Strait. 5. Fox Channel. 6. Fox Channel (lat. 66° 30’ N., long. 80° W.). 7. Fox Channel (lat. 66° 438’ N., long. 80° 07’ W.). 8. Southern part of Fox Basin. 9. Fox Basin. 10. Center of Fox Basin. 11. Off the northwestern end of Vansittart Island, Frozen Strait, north of Southampton Island. 12. Sturgis Bourne Strait, eastern end of Hurd Channel, Melville Peninsula. 13. Duckett’s Cove, Hurd Channel. 14. South of Cape Marti- neau, Melville Peninsula. 15. Off Cobourg Island, northwestern Baffin Bay (lat. 75° 40’ N., long. 78° 40’ W.). 16. Off the eastern end of Cobourg Island (lat. 75° 40’ N., long. 78° 50’ W.). 17. Off the eastern end of Co- bourg Island (lat. 75° 40’ N., long. 78° 53’ W.). 18. Off the eastern end of Cobourg Island. 19. Off the eastern end of Cobourg Island (lat. 75° 40’ N., long. 78° 55’ W.). 20. Off the eastern end of Cobourg Island (lat. 75° 40’ N., long. 78° 56’ W.). 21. Off the southern end of Cobourg Island (lat. 7° 40” N~ Jone: 78° 58” W_). NorTH GREENLAND: 22. Cape York. 23. Kerkotak, Salvo Island, Mel- ville Bay, just east of Cape York. 24. Thule, North Star Bay, north of Cape York. 25. North Star Bay. 26. Parker Snow Bay (lat. 76° 07’ N., long. 68° 20’ W.). 27. Parker Snow Bay. 28. Saunders Island, Westen- holm Sound. 29. Off Dalrimple Rock, Westenholm Sound. 30. Five miles south of Cape Chalon, Westenholm Sound. 31. Herbert Island, Whale Sound. 32. Northumberland Island, Murchison Sound. 33. Murchison — Sound. 34. Inglefield Gulf. 35. Karnah, Inglefield Gulf. 36. Nerke, near Morris Jessup glacier. 37. Cape Alexander, Smith Sound. East GREENLAND: 38. Angmagsalik. 39. Between Greenland and Jan Mayen (lat. 72° 21’ N., long. 16° 30’ W.). 40. East of Scoresby Sound (lat. 70° 21’ N., long. 16° 30’ W.); 110 fathoms. 41. Off Wollaston Foreland 1 Published with the permission of the Secretary of the Smithsonian Institution. Received March 10, 1936. Juuy 15, 1936 CLARK: ECHINODERMS 295 (lat. 74° 04’ N., long. 17° 58’ W.). 42. Off Wollaston Foreland (lat. 74° 20’ N., long. 16° 30’ W.). 48. Clavering Island. The species collected by Captain Bartlett (exclusive of the holo- Figs. 1—4.—-Strongylocentrotus drébachiensis, two young specimens with the test about 17 mm (Figs. 1, 2) and 13 mm (Figs. 3, 4) in horizontal diameter, aboral (Figs. 1, 3) and oral (Figs. 2, 4) views. Collected by Capt. R. A. Bartlett on the west side of Clavering Island, east Greenland, in 50-15 fathoms, on August 3, 1930. thurians) were the following. The numbers refer to the localities listed above: ASTEROIDEA: Crossaster papposus, 6, 8, 9, 138, 16, 18. Henricia san- guinolenta, 9, 10, 40, 43. Pteraster militaris, 9, 10, 43. Stephanasterias albula, 9, 10, 14, 18, 24, 28, 31, 32, 34, 35, 37, 38. Leptasterias polaris, 4, 43. Lep- tasterias groenlandica, 4, 18, 24, 30, 31, 32, 35, 37, 48. OPHIUROIDEA: Ophiacantha bidentata, 1, 3, 5, 6, 8, 9, 10, 12, 13, 14, 296 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 26, NO. 7 15, 16, 30, 31, 34, 40, 43. Ophiopus arcticus, 6, 10, 39, 48. Amphiura sunde- valli, 43. Ophiopholis aculeata, 4, 15, 16, 26. Ophiocten sericeum, 13, 15, 17, 19, 22, 25, 28, 32, 36, 38, 48. Ophiura robusta, 5, 6, 7, 8, 9, 10, 14, 15, 16, 19, 21, 25, 28, 30, 32, 37, 43. Ophiura sarsi, 17, 22, 25, 27; 28)°34, 3Onam Stegophiura nodosa, 13, 22, 29, 43. ECHINOIDEA: Strongylocentrotus drébachiensis, 1, 2, 4, 10, 11, 16, 20, 21, 22, 23, 25, 28, 29, 30, 32, 37, 40, 41, 42, 43. CRINOIDEA: Helcometra glaczalis, 4, 10, 16, 17, 32, 33, 48. Poliometra prolixa, 40, 41. Note.—Worthy of special mention are two small specimens of Stron- gylocentrotus drébachiensis (Figs. 1-4) from Clavering Island on which, on the aboral surface, the pedicellariae are so very numerous as to be more conspicuous than the spines. ZOOLOGY.—A new pocket gopher from New Mexico: K. RAYMOND | Hau, University of California. (Communicated by E. A. GOLDMAN.) In the spring of 1985 Miss Annie M. Alexander accompanied by Miss Louise Kellogg collected for the Museum of Vertebrate Zoology a series of Thomomys from the Rio Grande Valley near Albuquerque, New Mexico. Specimens from this general region had been referred to Thomomys aureus Allen, but this was at a time when that name was used in a more inclusive sense than it is at present. It was, therefore, no surprise to find that the gopher from Albuquerque could not be referred to aureus; indeed it was knowledge of this probability and curiosity as to the true identity of the pocket gopher there which led Miss Alexander to make special effort to obtain the specimens. Com- parisons reveal that the animal from Albuquerque pertains to an un- named race which is larger, and different in other respects, from fulvus, tularosae, pervagus, opulentus and aureus, the subspecies of Thomomys bottae (see Goldman)? whose ranges adjoin that of the new form. For the privilege of making direct comparisons with opulentus I am obliged to Major E. A. Goldman and Dr. H. H. T. Jackson of the United States Bureau of Biological Survey. To Dr. H. E. Anthony of the American Museum of Natural History I am similarly obliged for lending the original series used in naming T’homomys toltecus. Thomomys bottae connectens, new subspecies Type.—Male, adult, skull and skin; no. 66627, Mus. Vert. Zool.; Clawson Dairy, 5 miles north of Albuquerque, 4,943 feet elevation, Bernalillo County, New Mexico; May 6, 1935; collected by Annie M. Alexander, original no. 2981. 1 Received January 3, 1936. 2 Proc. Biol. Soc. Washington 48: 135, 150. 1935. oo - JuLy 15, 1936 HALL: POCKET GOPHER 297 Range.—Valley of the Rio Grande in central New Mexico, probably from. northern Socorro County northward to Bernalillo. Diagnosis.—Size: Large (see measurements). Color: In fresh summer pelage cinnamon buff (color terms after Ridgway)*; whitish below with plumbeous areas and with cinnamon buff extending onto pectoral region, and sometimes to other sections of the underparts; insides of cheek pouches and usually nose, blackish; postauricular patches small. Winter pelage with a reduced amount of cinnamon buff on the upper parts producing a “‘gray”’ coat. Skull: Large; rostrum broad and its length amounting to more than 67 percent of zygomatic breadth; nasals posteriorly truncate; hamulus of lacrimal large; interpterygoid space V-shaped and provided with median spine. TABLE 1.—MEASUREMENTS IN MILLIMETERS OF ADULT TOPOTYPES AND Types oF T. B. CONNECTENS AND T. B. TOLTECUS - ~ Connectens Toltecus Catalogue number ee 66638 |66642 |66635 |66636 |66634 |66627 | 66628 sae —_ | ——_ & 4305 | 4304 | 28° Sex Q Q (eh ot of ot of of rot of ot Total length 240 232 257 270 267 256 256 261 — — — Length of tail 72 66 76 73 74 68 70 a2 — — — Length of hind foot 32 31 36 35 35 35 35 35 29h | 27> | 2gb Basilar length See MOD SOIR Wee 4 | AOSOn EAP a alinoieo posse lmoueo Length of nasals Ae ails |muleypesalalia 242) NGealmelaeOu) 16SSeleulicele ina: | TS aia 5 Zygomatic breadth 2D ROMO Laa Polen ole ts | SOLe Ie2oeo — SOs mIPAsee, | 2S aca 2oco Mastoid breadth 20 29FI 2020s eZ baa 24y. S242 beeen 2228 | 24 RIG |e 2828S 2227 Breadth of rostrum 8.8 ele aOR ON Osan) 1020) Oke 8.4 9.1 8.8 Interorbital constriction 6.9 hew Thor UU eve 7.4 hae: hats he 6.7 6.5 6.6 Maxillary tooth-row 9.1 Sao 9.8 9.4 Omit 9.4 9.3 9.4 SAD 8.0 8.3 Extension of premaxillae posteriorly to nasals 2.1 2.9 2.2 Dee, 3.8 3% 26 PAT 33.2 3.4 3.4 Depth of skull LG2Sml eeu LOLOR mt Onl sIS ouils Onis | Leeda 1520) | T6561 6ce Length of rostrum‘ ie OROn | SLE Sh leat ce Ae Otel! | Des Teo Vis | 1758 ® Estimated. b Measured from the dried skin. : a pigeured between the anterior border of the nasals and the maxilla at the lateral end of the haruius of the lacrimal. Comparisons.—Compared with Thomomys bottae aureus, connectens is larger in external measurements, darker colored above, and below has the pectoral region strongly marked with cinnamon buff rather than white, and it is larger in every cranial measurement taken. The interpterygoid space is V-shaped rather than U-shaped; the exoccipital extends farther laterally, revealing an inverted V-shaped rather than inverted U-shaped, face of the mastoid portion of the auditory bulla; the rostrum, relative to the basilar length, is longer and wider; and the skull is more than a fourth heavier as shown by the crania of adult males without lower jaws which average 4.66 grams as against 3.33 grams. _ Compared with specimens of Thomomys bottae opulentus from Las Cruces (1), Garfield (5), Las Palomas (1) and San Marcial (6), T. b. connectens is slightly less reddish above and especially below, is larger in external and cranial measurements—in many parts of the skull constantly so. The skull of connectens is by actual weight much heavier: males average 4.6 grams as against 2.5 and females average 2.6 grams as opposed to 1.9. In connectens the premaxillae extend farther behind the nasals, the external meatus is pro- longed into a distinct tube, the temporal ridges approach one another more closely, the hamulus of each lacrimal bone is as large again, the exoccipital 3 Color standards and color nomenclature. Washington, D.C. 1912. 298 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 26, NO. 7 extends farther laterally revealing less of the mastoid portion of the auditory bulla, and the skull is generally more angular with better developed processes and ridges marking areas of muscle attachment. Compared with 5 topotypes and the type of Thomomys bottae toltecus, T. b. connectens is slightly lighter colored, has much longer hind feet (on dry skins, adult males measure 33 mm. as against 29 mm.) and is constantly larger in every part measured, except for the mastoid breadth and extension of the premaxillae posteriorly to the nasals. In the two parts of the skull indicated there is a slight overlap in measurements. Relative to the basilar length each of the five adult males of connectens has longer nasals and ros- trum, broader rostrum and a lesser width across the zygomatic arches than has either of the two adult males of toltecus. In connectens the length of the rostrum amounts to more, rather than less, than 67 percent of the zygomatic breadth. Also, in connectens the hamulus of each lacrimal bone is as large again and the inferior margin of the anterior opening of the infraorbital canal is continued anteriorly as a distinct ridge rather than curved upward to form part of an ellipse. Measurements, for the most part not previously available, for toltecws are offered above as facilitating comparison with related races. Remarks.—Among named subspecies of 7. b. bottae whose ranges approach nearest to that of connectens, probably greatest similarity is shown to T. b. aureus. The range of connectens, as known to me, however, is separated from that of aureus by a large area from which no specimens have been examined though the species 7. bottae doubtless occurs in suitable environ- ments there. The northwestern limits of range of connectens, then, remain to be determined. Specimens from Socorro are variously intermediate in color and to a certain extent in external measurements between connectens and opulentus but cranially they agree well with the latter. Specimens examined.—Total number, 19 as follows: Type locality, 14; 4.5 miles south Albuquerque, 4,943 feet elevation, Bernalillo County, New Mexico, 4. ZOOLOGY.—A note on Dictyocaulus from domestic and wild rumi- nants... G. Dixmans, Zoological Division, Bureau of Animal In- dustry. (Communicated by Mauricr C. HALL.) Chapin (1925) described as Dictyocaulus hadweni n. sp. a nematode collected from the lungs of the American bison, Bison bison, at Wain- right, Alberta, Canada, by Dr. Seymour Hadwen. Following the de- scription, Chapin devoted a short paragraph to a comparison of D. hadweni with D. filaria, and differentiated D. hadwenz from D. filaria by the more abrupt termination of the dorsal rays, the complete fu- sion of the medio-lateral and postero-lateral rays, and the longer spic- ules. Apparently the use of the name Dictyocaulus filaria should be considered as a lapsus calami for Dictyocaulus viviparus, or else Chapin really intended to compare D. —— with D. flarta 1 Received March 17, 1936. Se JuLY 15, 1936 DIKMANS: DICTYOCAULUS 299 hadweni resembles D. viviparus in the characters mentioned by the - author and there would, therefore, have been some reason for at- tempting to differentiate between them, but D. filarzva differs so mark- edly from D. hadweni in the conformation and size of the spicules alone that comparisons on any points other than spicules would be the only ones in order. Chapin stated that D. hadweni differs from D. filarzva “‘in the longer spicules,’ and from this statement the reader would infer that the spicules of D. hadweni are longer than those of D. filaria whereas, as a matter of fact, the spicules of D. hadwenz are shorter than those of D. filarza. Since this fact could not have escaped ONn. yy ual Fig. 1.—Dorsal rays of bursa of Dictyocaulus viviparus showing (1-6), variations in position and, (7-13), variations in termination. Chapin’s attention if he had made a comparative study of these two nematodes, it might be inferred that the use of the name D. filaria was actually a lapsus calamz. However, this inference also meets with difficulties because had the author intended to compare D. hadweni with D. viviparus and had he made a comparative study of these two forms he would have noted their similarity in the very points on which he attempted to differentiate them, viz., the fusion of the medio- lateral and postero-lateral rays, the position and termination of the dorsal rays, and the conformation of the spicules, the spicules of D. hadweni differing from the spicules of D. viviparus in size only. For some time the writer has been making all identifications of lungworms from domestic and wild ruminants as these have been re- ferred to the Zoological Division for determination, and during that time considerable difficulty has been experienced in differentiating D. 300 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 26, No. 7 hadweni from D. viviparus on any valid morphological grounds. Dr. W. E. Swales of the Institute of Parasitology, McDonald College, Quebec, Canada, under a codperative agreement for exchange of specimens in various parasite groups, has referred specimens of Dictyocaulus collected from cattle and various wild ruminants in Canada to the Zoological Division of the Bureau of Animal Industry for study. Since the females of species of Dictyocaulus are, for all practical purposes, indistinguishable, the writer has based his study of the members of the genus occurring in ruminants in North America on a comparison of male specimens, with special reference to the char- acter and relative position of the rays of the bursa, the termination of the dorsal rays, and the size and morphology of the spicules. These studies give the following results: The ventro-ventral ray is always shorter than the latero-ventral ray, but there is no fixed ratio between the lengths of these 2 rays, and their relative lengths cannot be con- sidered as a diagnostic character; the externo-lateral ray is always single and in position is widely separated from both the latero-ventral and the fused medio-lateral and postero-lateral rays; the externo- dorsal ray is single and somewhat shorter than the rays on either side of it; the dorsal rays are doubled, their positions and terminations vary considerably (fig. 1), and no specific value can be attached to either their position or their termination. The conformation of the spicules of Dictyocaulus from cattle, D. viviparus, is similar to that of the spicules of available specimens of Dictyocaulus from wild rumi- nants. In general the spicules of Dictyocaulus from wild ruminants are somewhat longer than the spicules of D. viviparus from cattle, but there is considerable overlapping (table 1). So far as the females of D. hadweni and D. viviparus are concerned there are no definite morphological characters on which they can be separated from each other. It appears, therefore, that D. hadweni can- not be separated from D. viviparus on any valid, morphological grounds and, since D. viviparus is the older name, D. hadweni must fall into synonymy so far as its morphology is concerned. Only care- fully controlled feeding experiments could establish whether we are dealing with biological varieties capable of infecting only cattle or only deer or whether these nematodes are biologically as well as mor- phologically identical. Hsti (1935) described as Dictyocaulus khawi a nematode stated tog have been collected from the lungs of swine in Tonkin, French Indo- China. The only point in which this nematode is said to differ from — ih ee — oe Juny 15, 1936 DIKMANS: DICTYOCAULUS 301 D. viviparus is that one of the 3 digitations of the dorsal ray has an externo-lateral position. In view of the fact that the terminations of the dorsal rays may be extremely variable in position and appearance (fig. 1) this minor difference does not appear to be sufficient for the creation of a new species. On morphological grounds, therefore, Dictyocaulus khawi would also be considered asasynonym of Dictyo- caulus viviparus, but in view of the fact that over most of the world Dictyocaulus is not found in pigs in spite of evident opportunities for TABLE 1.—LENGTH OF SPICULES OF DICTYOCAULUS VIVIPARUS COLLECTED FROM DIFFERENT Hosts ELGai, No. of male specimens | Minimum length of Maximum length of oe examined spicules in microns spicules in microns Cattle Bos taurus 29 220 AID American bison Bison bison 14 220 295 Moose Alces americana 1 255 «255 Elk Cervus canadensis 10 255 SAD, Reindeer Rangifer tarandus i 220 315 Black-tailed deer Odocoileus columbianus 3 — 235 300 White-tailed deer Odocoileus virginianus + 176 220 Mule deer Odocoileus hemionus 3 220 220 Red deer, probably Cervus canadensis 4 255 295 infection from cattle, there is the possibility that under special con- ditions existing in French Indo-China a strain of D. viviparus has be- come adapted to swine, as Necator americanus of man appears to have become adapted to swine in the West Indies, with the development of a new biological species which may show, as does N. suillus, certain fixed morphological characters, however slight, differentiating it from the parent species. Since Hsii did not collect the specimens which he describes there is, of course, also the possibility of error in labelling to be considered. Skrjabin (1931) described as Dictyocaulus eckerti a worm found in the lungs of reindeer in the U.S.S8.R. He stated that this worm differs from D. viviparus as follows: a, In the presence of cervical papillae 302 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 26, No. 7 (these are not mentioned in the part of the paper describing the nema- tode); b, in the presence of a mouth capsule; c, in the length of the spicules; d, in the structure of the terminal portion of the dorsal ray. The length of the spicules is given as 290 to 310u and the figures ac- companying the description show that they are morphologically simi- lar to the spicules found in D. viviparus. The figure of the bursa of D. eckerti shows that the termination of the dorsal rays is similar to that of D. viviparus. With reference to the presence of a mouth cap- sule, the presence of a small, shallow, mouth capsule is a character of the genus and an examination of a number of specimens of D. viviparus collected from cattle at Bethesda, Md., shows that this char- acter is present in D. viviparus from cattle as well as in D. eckerti from reindeer in the U.S.S.R. With reference to the presence of cervical papillae in D. eckerti, there appears to be no other record of the pres- ence of these structures in members of the genus Dictyocaulus, and a careful examination of a number of specimens from both cattle and reindeer failed to reveal their presence. No information is furnished as to the appearance and location of the cervical papillae in D. eckertt. From Skrjabin’s description and from the figures accompanying it, it appears that the lungworm collected from reindeer in the U.S.8.R. does not differ materially from Dictyocaulus collected from the lungs of reindeer in Alaska, aside from the rather surprising occurrence of cervical papillae in the former, and aside from this one item D. eckerta appears to be identical with D. viviparus. DIcTYOCAULUS VIVIPARUS (Bloch, 1782) Railliet and Henry, 1907 Synonyms.—Gordius viviparus Bloch, 1782; Ascaris vituli Bruguiére, 1791; Strongylus vitulorum Rudolphi, 1809; Strongylus micrurus Mebhlis, 1831; Dictyocaulus hadweni Chapin, 1925; .(?) Dictyocaulus eckertt Skrjabin, 1931; (?) Dictyocaulus khawi Hsii, 1935. Hosts.—Cattle, Bos taurus; American bison Bison bison; Moose, Alces americana; Elk, Cervus canadensis; Reindeer, Rangifer tarandus; Black- tailed deer, Odocoileus columbianus ; white-tailed deer, Odocotleus virginianus ; Mule deer, Odocotleus hemionus; Red deer, probably Cervus canadensis; (?) Swine, Sus scrofa domestica. DICTYOCAULUS FILARIA (Rudolphi, 1809) Railliet and Henry, 1907 Dictyocaulus filaria is the common lungworm of sheep in North America, and has been collected also from the black-tailed deer, Odocoileus colum- bianus, and from the white-tailed deer, Odocoileus virginianus. In 24 male specimens from both deer and sheep, the length of the spicules varied from 330 to 580u, spicules 330u long and 580y long each being found once and the lengths of all other spicules varying from 365 to 515u. The ventro-ventral ray was found to be shorter than the latero-ventral ray in all the specimens Juty 15, 19836 COVILLE AND MORTON: ERIOGONUM INTRAFRACTUM 303 examined. The dorsal rays terminate in 3 processes but the position of these processes varies to such an extent that no diagnostic significance can be at- tached to it. Bhalerao (1932) described as Dictyocaulus unequalis a nematode collected from the large bronchi of a Tibetan sheep. He stated that this nematode differs from Dictyocaulus filarva in having shorter spicules, only one male 28 mm long, with spicules 280 to 290u long, being available for examination, and also in that the ventro-ventral ray of the bursa was shorter than the latero-ventral ray, and he gave the specific name wnequalzs because of this inequality of the ventral rays. Since the ventro-ventral ray in Dictyocaulus is usually shorter than the latero-ventral ray, this supposed difference be- tween D. filaria and D. unequalis disappears and, in the absence of any no- ticeable morphological difference, the difference of 40 to 50 microns in length in spicules based on the examination of a single specimen does not appear to be sufficient reason for the making of a new species when an otherwise iden- tical species has a range of 250 microns. Dictyocaulus unequalis is, therefore, considered as probably a synonym of D. filaria. LITERATURE CITED BHALEeRAO, G. D. On some nematode parasites of goats and sheep at Muktesar. Indian Jour. Vet. Sci. and An. Husb. 2(8): 242-254. 1932. CuHapin, E. A. New nematodes from North American mammals. Jour. Agr. Res. 30: 667-681. 1925. Hst, H. F. A study of some Strongyloidea and Spiruroidea from French Indo-China and of Thelazia chungkingensis Hsti, 1933, from China. Zeitschr. f. Parasitenk. 7: 579-600. 1935. SxRJABIN, K. 1. Worm infestations of the reindeer. For veterinarians, technical workers and students. State Publishing House. Moscow, Leningrad. (In Russian). Pp. 1-86. 1935. BOTAN Y.—Eriogonum intrafractum, a new species and new sub- genus from Death Valley, California.t FREDERICK V. COVILLE AND C. V. Morton, U. 8. National Museum. In the spring of 1932 the senior author and Mr. M. French Gilman, while studying the flora of Death Valley, California, under the auspices of the National Geographic Society, found a peculiar-looking perennial species of Eriogonum, which because of its immaturity re- mained unnamed. The plants were found in Titus Canyon, Grape- vine Mountains, six miles below Leadfield, at an altitude of about 2,000 feet, growing in the crevices of a blue limestone ledge sloping to the north. The dead stems of the preceding year were about three feet high. Mr. Gilman revisited the locality and obtained additional specimens in 1934. These also were immature, and it was not until October, 1935, that Mr. Gilman was able to obtain good material, 1 Published by permission of the Secretary of the Smithsonian Institution. Re- ceived February 26, 1936. 304 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 26, NO. 7 this indicating at once a new species which can not be referred to any of the recognized subgenera of Eriogonwm. The new subgenus de- scribed below must therefore be added to the already rather long list of plants endemic to this peculiar region. Eriogonum subgenus Clastomyelon? Coville & Morton, subg. nov. Herbae perennes, caulibus erectis, altis, infra inflorescentiam non ramosis, fistulosis, demum interne transverse articulatis; folia radicalia, longe- petiolata, pilosa; involucra sessilia, in verticillis disposita, late campanulata, non angulata, mox fissa et explanata, bracteolis nonnullis suffulta; bracteolae involucrorum saepe virides, crassae; pedicelli florum pubescentes; fructus basi bulboso-expansus, perspicue 3-cristatus, medio abrupte contractus. TYPE SPECIES: Hriogonum intrafractum Coville & Morton. Three subgenera of Hriogonum have commonly been recognized, namely, Ganysma, Oregonium, and Eueriogonum, from all of which the new subgenus Clastomyelon differs in the following characters: 1. Articulate stems. The main stem (which is erect, leafless, and un- branched) is divided into many transverse articulations. These, averaging about 5 mm in length, do not become obvious until the plant is past ma- turity, at which time they are easily visible as undulations beneath the outer stem covering. The latter at length weathers into shreds, and the monil- iform articulations fall off one by one. Concerning this feature M. Gilman wrote: ‘‘When examining the Eriogonums notice the very peculiar structure of the larger parts of the stalks. Under the bark the stems are made up of little rings from one-fourth to three-fourths inch long. When old the bark falls away, leaving the stems made up of a series of white sections or joints looking like tiny napkin rings. The rings easily unjoint when weathered enough, and on the ground around old plants the fallen stems indicate the vintages of the different years’ growth of stalks.’’ These facts are obvious on several of the herbarium specimens cited. 2. Whorled involucres. Except on the ultimate branches of the stem the involucres are sessile and borne in whorls. They are usually three in number, each corresponding to an outer bract; but inasmuch as each is voluminously filled with flowers and bractlets, the general appearance is that of a series of heads with the stems going through the middle. Such an inflorescence is un- known in any of the other subgenera, in which the involucres are borne singly in cymes, or in true terminal heads. 3. Harly ruptured involucres. The involucre itself is broadly campanulate, unangled (a further distinction from the subgenus Oregonium, to which Clastomyelon is most nearly allied), and deeply 5-partite. The expanding flowers soon rupture the involucre, however, so that at maturity it is flat and platelike, or in old material often appears as though composed of distinct lobes, thus masking the identity of the plant with Eriogonum. 4. Hairy pedicels. The pedicels of the individual flowers within the involu- 2 From xdaorés, broken, and pveddv, pith. JuLY 15, 1936 COVILLE AND MORTON: ERIOGONUM INTRAFRACTUM 305 ere are definitely pilose. Although no general statement has been made in the literature concerning the pedicels in the various subgenera of Hriogonum, we find by study of the almost complete representation of recognized species in the National Herbarium, that they seem to be uniformly glabrous, except for a very few species, in which they bear minute glands. The hairs in Clasto- myelon are often somewhat deciduous after maturity. 5. Flask-shaped, crested fruits. The fruit of Clastomyelon is very peculiar in form, being conspicuously bulbous-dilated at base and trilobate, each lobe bearing a conspicuous rounded crest, filled with pithlike tissue. At the middle the fruit is abruptly contracted, the upper portion being slender and sharply triangular. Although it is not possible to say at present that such fruits are not found in species of other subgenera, it seems likely that that is true. At least, in none of the species that we have examined has there been any approach to this condition. The above mentioned characters by which F. intrafractum differs from all previously known species seem to us amply sufficient ground for the creation of a new subgenus, to take a place coédrdinate with the three other recognized subgenera. Additional characters of merely specific importance, which will separate the present species from others of the genus, are the tall leafless stems, usually more than one meter high and unbranched up to the inflores- cence, the relatively large, long-petiolate basal leaves, uniformly short- pilose (but not at all tomentose) on both surfaces, the numerous accessory bractlets present in the inflorescence (including an involucre of three basally united bractlets present at the base of each involucre), the minutely pubes- cent perianth (ochroleucous when young, yellow and rose in age), deeply constricted at the middle and with flaring limb, and the unequal perianth segments. In most species of Hriogonum with unequal segments, the inner are shorter and narrower than the outer, but in H. intrafractum the reverse is true. Eriogonum intrafractum Coville & Morton, sp. nov. Herba perennis, caulibus erectis, altis, solitariis, teretibus, non ramosis, glaucis, fistulosis, demum interne perspicue articulatis; folia basalia, longe- petiolata, late oblonga obovatave, apice rotundata, basi cuneata, utrinque breviter pilosa; involucra in verticillis circum ramos inflorescentiae dis- posita, sessilia, aperte campanulata, 5-lobata, pubescentia, non angulata, mox fissa; florum pedicelli pilosi; perianthium ochroleucum, demum basi roseum apice luteum, basi rotundatum, medio constrictum, segmentis externe pubescentibus, exterioribus quam interioribus brevioribus et an- gustioribus; fructus basi bulboso-inflatus, trilobatus, lobis cristatis, medio abrupte contractus, apice angustus, triangularis. Perennial herb with a thick woody taproot; stems erect, 0.9-1.2 m long, usually solitary, terete, up to 11 mm in diameter, unbranched except in the inflorescence, there two or three times subdichotomously (or rarely tri- chotomously) branched, glabrous, glaucous-green, fistulous, internally transversely articulate into short segments 3-9 or even 16 mm. long, the outer layer at length exfoliating, thus allowing the stramineous, moniliform articulations to fall off separately; leaves all basal, long-petiolate, the 306 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 26, NO.7 petioles pale yellow, densely short-pilose, exceeding the blade in length, the latter broadly oblong or obovate, up to 7 cm long and 3.5 cm wide, rounded at the apex, cuneate at the base, yellowish or grayish green, membraneous, entire, plane, densely and uniformly short-pilose on both surfaces, the midvein and 3 or 4 prominent secondary veins pale yellow, prominently elevated beneath; secondary rosettes of leaves often borne at the base of the stem; bracts at the primary furcation of the stem whorled, about 5, ovate, acute, densely short-pubescent, deciduous; inflorescence subspicate in appearance, consisting of densely flowered involucres borne in whorls up the branches, or these solitary on the ultimate branches; bracts subtending each whorl of involucres three, free, oblong, about 3 mm long and 1.4 mm wide, sharply acute, pilosulous on both surfaces, fleshy, green; internal accessory bractlets of two kinds: (1) 3 bractlets at the base of each involu- cre, similar in shape and texture to the outer bracts but smaller and united at base and forming a single tripartite bract, (2) supernumerary bractlets (usually narrowly oblong) irregularly disposed between the outer bracts and the involucres; involucres sessile, openly campanulate, 5-lobed, about 3 mm long, pubescent on both surfaces, fleshy, not angled, the lobes ovate, erect, about 1.5 mm long, the tube ruptured in one or several places at maturity by the expanding flowers, the entire involucre then becoming flat and plate-like; pedicels of the flowers pilosulous, interspersed with numerous linear-oblanceolate pubescent bractlets, these sometimes fleshy and green; perianth ochroleucous, in age becoming rose at the base and yellow toward the apex, 3 mm long, rounded at the base, definitely constricted at the mid- dle, there about 1.7 mm wide, flaring and about 2.2 mm wide at the apex, the segments minutely pubescent externally, the outer broadly oblanceolate, shorter and narrower than the inner, rounded at the apex, concave, the inner carinate below, plane above; stamens much shorter than the perianth segments, the filaments subulate, glabrous, adnate at the base to the perianth segments, the anthers oval, white; styles 3, slender, free, recurved; stigmas capitate; fruit triangular and flask-shaped, the much inflated base being expanded into three prominent lobes, abruptly contracted at the middle into a sharply triangular upper part, sparingly short-hairy when young. Type in the U. S. National Herbarium, no. 1,631,285, collected in Titus Canyon, Grapevine Mountains, Death Valley, Inyo County, California, | at about 2,000 feet altitude, Oct. 13, 1935, by M. French Gilman (no. 2120). The following additional specimens, all from Death Valley, have been ex- amined: Same locality, Apr. 25, 1932, Coville & Gilman 442; May 8, 1934, Gilman 1194; June 11, 1935, Gilman 1690. Death Valley Canyon, July 9, 1935, Gilman 1917. Also several unnumbered specimens collected in Titus Canyon by Mr. Gilman to show the articulations of the old stems. CONTENTS ZooLocy.—A new pocket gopher aon New ia E. R FALL eee epee ences eee eee from Death Valley, California. FREDERICK v. Cove Vi MORTON . 5088 . 20h (es: x < Pe a 3) PY. "Pie nics : 7% Paes y \ f | Ce RY. S Aw wd AGN \\ “y C| Fig. 1.—Showing distribution of subtropical and temperate forest units during the Cenozoic of western America. A, Eocene; B, Miocene; C, Recent. In all maps, oe spaced continuous lines indicate a subtropical, and broken lines a temperate of Mexico and Central America where it is now found. From Alaska, extending across into Siberia, in Eocene time, the temperate forest was shifted to middle latitudes during the Miocene. It is now broken up into various units, of which the dominant coast redwood forest has survived only. along the coast from Oregon to central California, with elimination of many of the broad-leafed deciduous genera. Many genera of this temperate forest have survived also in the forests of the Cordillera, especially in the mountains of Mexico and Central America. A study of the distribution of this modern vegeta- tion throws much light on the climatic history and the stratigraphic sequence of the Cenozoic rocks in western North America. | LITERATURE CITED 1. CHanegy, R. W., and Sansorn, E. I. The Goshen flora of west central Oregon. Carnegie Inst. Wash. Pub. 439: 48. 1933. | 2. Stock, Curester. Evidence of changing climates during the later Eocene and Oligo- cene of California. In Program and Abstracts, 35th Annual Meeting, Cordil- leran Section, Geological Society of America, 1936. 324 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 26, NO. 8 3. CuHanery, R. W., and Sansorn, EK. I. The Goshen flora of west central Oregon. Carnegie Inst. Wash. Pub. 439: 1-103. 1933. Sanporn, E. I. The Comstock flora of west central Oregon. Idem 465: 1-28. 1935. Potsury, 8. S. The La Porte flora of Plumas County, California. Idem 465: 29-81. 1935. MacGinitigz, H. D. The flora of the Weaverville beds, Trinity County, California. (MS.) Conpit, C. The San Pablo flora of central California. (MS.) CHANEY, R. W. Quantitative studies of the Bridge Creek flora. Amer. Jour. Sci., 5th ser. 8: 127-144. 1924. CuHanEy, R. W. A comparative study of the Bridge Creek flora and the modern redwood forest. Carnegie Inst. Wash. Pub. 349: 1-22. 1925. MacGrinitig, H. D. The Trout Creek flora of southeastern Oregon. Idem 416: 21-68. 1933. LaMottr, R. 8. The upper Cedarville flora of northwestern Nevada and adjacent California. Idem 455. In press. OuivEeR, E. A Miocene flora from the Blue Mountains, Oregon. Idem 455: 1-27. In press. 8. Dorr, E. Pliocene floras of California. Idem 412: 1-112. 1980. AXxELROD, D. I. A Pliocene flora from the Eden beds. Amer. Mus. Novitates 729. 19384. 9. GaRDNER, J. 8. A monograph of the British Eocene flora. Vol. 1. Palaeont. Soc. London, 1886. 10. CHangy, R. W., and Exias, M. K. Late Tertiary floras from the High Plains. Carnegie Inst. Wash. Pub. 476. In press. 11. Merriam, J. C. A contribution to the geology of the John Day Basin. Univ. Calif. Bull. Dept. Geol. 2: 269-314. 1901. Babee Aan ZOOLOGY.—WNotes on the Crustacea, chiefly Natantia, collected by Captain Robert A. Bartlett in Arctic Seas... MarGarret E. Van WINKLE, Wellesley College, and Waupo L. Scumirt, U. S. National Museum. Nearly every year for more than a decade Captain Bartlett has brought back a veritable zoological treasure trove to the United States National Museum. Some portions of his extensive collections have been reported upon by various specialists in this Journal, in the Proceedings of the National Museum, and in the Smithsonian Mis- cellaneous Collections. Here for the first time, however, is assembled a complete list of his decapod crustacea, together with some casual notes on a few other forms. The specimens listed were taken in each of the years 1924 to 1935, inclusive, with the exception of 1928 and 1934, when Captain Bartlett and his schooner, the Morrissey, were under exclusive contract to other parties. Of particular interest are the collections made in Fox Channel, Fox Basin, and the Straits of Fury and Hecla north of the Melville Peninsula. The only decapod crustacea previously recorded from these waters were five in number, secured by the second Parry Ex- pedition in the Fury and Hecla, for which the Straits were named, in the fall of 1822. These species were taken in nets off Igloolik 1 Received April 2, 1936. Aue. 15, 1936 VAN WINKLE AND SCHMITT: CRUSTACEA 325 Island, the winter quarters of that year, where they were said to have been ‘“‘found abundantly,’’ or ‘“‘taken in considerable numbers.”’ These particular decapods are Spirontocaris groenlandica, S. polaris, S. spinus, Sabinea septemcarinata, and Sclerocrangon boreas.? Captain Bartlett also obtained the same species at several stations in the same general neighborhood, with the exception of Sabinea septemcarinata. Indeed, two of the species, Spirontocaris groenlandica and S. spina, were caught in the entrance of the Straits at a point a little farther north than Igloolik Island, together with two other deca- pods not seen by the earlier expedition: Spirontocaris phippsi and S.gaimardit. S.gaimardii belcheri and S. fabricii, which were dredged by Captain Bartlett at several stations, had also been missed by the Parry Expedition, as had Pagurus kréyert, which Captain Bartlett collected off Igloolik Island. The bulk of the decapod material of the 1929, 1930, and 1931 ex- peditions, amounting to over a thousand specimens, was determined by Mrs. Van Winkle, of the Department of Zoology of Wellesley College. The second author undertook the balance of the material not otherwise credited. We are indebted to Dr. Mary J. Rathbun for naming the several species of brachyura, while the identifications of the amphipods cited in connection with the crustacea found in seal and codfish stomachs were furnished by Clarence R. Shoemaker, whose complete report upon the amphipods is reserved for a later date. In order to add. to our knowledge of the distribution of the species discussed in this paper, brief reference has also been made to crus- taceans that were named for the Biological Board of Canada and which had been secured under its auspices during a series of biological and fisheries surveys carried on in Hudson Strait and Hudson Bay in the years 1927, 1928, and 1930. For conciseness, the localities from which the Bartlett crustacea listed below were collected are referred to by number in the manner suggested by Austin H. Clark in his report upon the Bartlett echino- derms (this Journal 26: 294). For more ready reference, these lo- 2 Ross, JAMES CLARK. Marine Invertebrate Animals, in Appendix, Jour... . Third Voyage... Discovery ... Northwest Passage... , p. 120, London, 1826; and in Ap- pendix, Narrative... Second Voyage... Search... Northwest Passage..., pp. Ixxxi-lxxxiv, London, 1835. Nils von Hofsten, in his masterly treatment of the crustacea of EHisfjord, Spitz- bergen (Kungl. Sven. Vet. Akad. Handl. 54 (No. 7). 1916), depicts the Ross finds, among others, on a series of charts showing the polar distribution of each of these five species, with the exception of S. groenlandica. On the distribution chart of this species the Igloolik record established by Ross appears to have been inadvertently omitted, 326 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 26, No. 8 calities are arranged in five groups lettered A to E. Within their respective groups the localities have been arranged, so far as possible, in order of latitude from south to north, without regard to longitude or date. Localities at which crustacea were dredged, unless otherwise indi- cated, are as follows: A. LABRADOR COAST. 1. Saglek Bay, September 25, 1925. 2. Kamak- torvik Bay, 59° 31’ N., 68° 50’ W., July 12, 1929. 3. Coast of Labrador, August and September, 1925. B. Fox CHANNEL AND Fox Basin, BArFrin LAND, AND MELVILLE PENIN- SsuLA. 4. Four to five miles east of Cape Dorchester, August 8, 1927, 25 fathoms. 5. West end of White Island, Comer Straits, August 10, 1933. 6. Hurd Channel, between Bushman Island west of Vansittart Island and Melville Peninsula, August 10, 1933. 7. Between two unnamed islands south of Cape Martineau, east of Duckett’s Cove, August 19, 1933, 7-15 fathoms. 8. Three stations in Duckett’s Cove, Hurd Channel, August 11— 13, 1933, 1-14 fathoms. 9. Cove, north shore of Lyon Inlet, August 24, 1933, shore collecting, low tide. 10. Four stations in Fox Basin between 66° 30’ N.-66° 46’ N. and 79° 15’ W.-80° 07’ W., August 10-13, 1927, 32- 37 fathoms. 11. Bight, Cape Penrhyn, August 31, 1933, 11 fathoms. 12. Fox Basin, 67° 45’ N., 79° 09’ W., August 24, 1927. 138. Center Fox Basin, August 24, 1927, 25 fathoms. 14. Between Ooglit Islands and Eskimo village at Pingitkalik, September 5, 1933. 15. Two short dredge hauls in entrance to Straits of Fury and Hecla, September 3, 1933, 20-380 fathoms. | C. Barrin Bay, ELLESMERE LAND, AND WEST GREENLAND. 16. Three stations, southeast end of Cobourg Island, 75° 40’ N., and between 78° 50’ W. and 78° 56’ W., August 3-4, 1935, 24-18 fathoms, gravel-rocky bottom, surface temperature 38-39° F. 17. Two stations south end of Cobourg Island, 75° 40’ N., 78° 58’ W., August 4, 1935, 8-20 fathoms, very rocky bot- tom surface temperature 36°F. 18. Hudson Bay Company’s Post, Ponds Inlet, Jones Sound, August 29, 1926, 10 fathoms. 19. Craig Harbour, Jones Sound, August 26, 1926, 7-20 fathoms. 20. Cape York, 76° 00’ N., August 21, 1926 and August 28, 1932, 7-15 fathoms. 21. Kerkotak, Salvo Island, Melville Bay, August 28, 1932. 22. Parker Snow Bay, 76° 07’ N., 68° 20’ W., July 22, 1926 and July 24, 1935, 5-12 fathoms, muddy bottom. 23. Off Dalrymple Rock, Wolstenholm Sound, July 22, 1926. 24. Saunders Island, Wolstenholm Sound, July 22, 1926, 10-12 fathoms. 25. Two sta- tions, Karnah, Inglefield Gulf, August 14, 15, 1926, 5-20 fathoms. 26. Four stations off Northumberland Island, Whale Sound, August 15-17, 26, 1926, 7-30 fathoms. 27. Two stations off Herbert Island, Whale Sound, July 25, 1926, 4-25 fathoms. 28. Hackluyt Island, Whale Sound, 77° 26’ N., 72° 30’ W., July 30, 1935, 11-20 fathoms, bottom small stones, surface temperature 37°F. 29. Three stations, Murchison Sound, August 19-21, 1926, 17-20 fathoms. 30. Five miles south of Cape Chalon, July 27, 1982. 31. Cape Alexander, Smith Sound, August 26, 1932. D. East coast OF GREENLAND. 32. Four stations, Angmagssalik, August 30, 1930, August 27-29, 1931, 20 fathoms. 33. Off Cape Stosch, Hudson Land, 74° 04’ N., 17° 50’ W., July 30, 1931, 120 fathoms. 34. Be- tween Clavering Island and Hornes Foreland, July 31, 1930. °35. Clavering Aue. 15, 1936 VAN WINKLE AND SCHMITT: CRUSTACEA O20 Fiord, August 2, 1930. 36. Pendulum Island, July 20, 1930. 37. Bight Shannon Island, July 29, 1930. E. Berrinac SEA AND ALASKA. 38. One and a half miles southeast of Cape Cheerful, Unalaska, August 4, 1924, from stomach of fish taken in 3 fathoms. 39. About 15 miles north of Big Diomede Island, June 14, 1924. 40. Twenty-two miles off Shishmaref Inlet, June 27, 1924, 18 fathoms, ship stationary in ice. 41. Thirty miles off Devils Mountain, June 20, 1924, 16-18 fathoms, mud bottom. 42. Two stations, mouth of Kotzebue Sound, July 10, 1924, 10-17 fathoms, mud bottom, ship drifting in ice. SPECIES COLLECTED Pandalus borealis Krgyer. DandC. Taken twice by Captain Bartlett from the stomach of codfish at Angmagssalik, east Greenland, August 27, 1931, remains of two specimens; and at Laxebugt, Disko Island, 69° 19’ 14’”’ N., 54° 14’ W., fragments of one specimen. Pandalus goniurus Stimpson. E 39, 41, 42. Spirontocaris groenlandica (J. C. Fabricius). B7,15. C 16, 17, 20, 23, 25, 26, 27, 29, 31. D 32. At Angmagssalik, east Greenland, this species was also found in the stomach of a cod, August 27, 1931. The Biological Board of Canada had specimens obtained along the western shore of Hudson Bay, north of Churchill in 19 and 42 fathoms, off Mansel Island in 75 fathoms; in Hudson Strait at Nottingham Island and Sugluk Creek: and at Port Burwell, Ungava from cod stomachs. Spirontocaris polaris (Sabine). A 2,3. B 4, 5,7, 8,9, 10, 11, 138, 14. C 16-23, 25, 26, 27, 29, 30, 312 D 32, 33, 35, 36. Also found in cod stom- achs, Angmagssalik, east Greenland, August 27, 1931. In one lot of speci- mens (B 7) there were no lateral spinules on the epimera of the fourth abdom- . inal somite of one specimen, while another had a spinule on the left side only. The Biological Board of Canada had received specimens from the southern part of Hudson Bay, 57° 19’ N., 85° 32’ W., from 52 to 54 fathoms; from Hudson Strait at Nottingham Island and Sugluk Creek; and from Wakeham Bay and Port Burwell, Ungava, both dredged and from cod stomachs. Spirontocaris microceros Krgyer. A 1. Two specimens of good size were collected by Captain Bartlett from Saglek Bay, Labrador, September 1, 1925. The larger, 42 mm in length, had four rostral teeth, two on the cara- pace and two on the rostrum proper; the rostrum reached to the middle of the cornea and fell a little short of the first segment of the antennular peduncle. The smaller specimen, about 34 mm in length, was very typical of the species; its rostral teeth were five in number, two on the carapace and three on the rostrum proper; the rostrum was about as long as the eye, but a little short of the first segment of the antennular peduncle. The Biological Board of Canada had a specimen of this species from a cod stomach from Port Burwell, Ungava, and two others determined as S. zebra Leim from the same source. Stephensen,? with Miss Rathbun,‘ believes that these two species are probably identical. In his paper Stephensen lists five localities on the southwest coast of Greenland where undoubted S. mzcroceros has been obtained, Leim' records S. zebra from three localities in New Bruns- 8’ STEPHENSON, K. Crustacea Decapoda, Godthaab Expedition. Meddel. Gr¢gn- land, 80 (No. 1): 81. 1935. 4 "RATHBUN, M. J. Decapoda. Canadian Atlantic Fauna, 10 m: 12. 1929. 5 Lerm, A. H. A new species of Sptrontocaris with notes on other species from the Atlantic Coast. Trans. Royal Canadian Instit. XIII (No. 4): 187. 1921. 328 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 26, No. 8 wick and one in Nova Scotia, but makes no mention of S. microceros. Miss Rathbun contributes another locality for S. microceros: Misaine Bank, off Cape Breton Island, 45° 19’ N., 58° 51’ 15” W., 45 fathoms. Spirontocaris phippsii (Kroyer) (= Hippolyte turgida Krgyer). A2. B 5, 6, 18, 14, 15. C 20-31. D 32. In occasional specimens of S. phippsit the lower or smaller of the two supra-orbital spines may fail to develop. This species has also been taken by the Biological Board of Canada at Nottingham Island in Hudson Strait; and at Wakeham Bay and Port Bur- well, Ungava, at the latter place from cod stomachs. Spirontocaris spina (Sowerby). A 2,3. B77, 10, 14, 15. C17, 21, 23, 25, 26, 27, 29, 30, 31. D 32. E42. Also found in stomach of cod, Ang- magssalik, east Greenland, August 27, 1931. The Biological Board of Canada has taken this species in Hudson Bay, just south of Mansel Island in 88 fathoms; in Hudson Strait at Sugluk Creek; and in Wakeham Bay and Port Burwell, Ungava; at the last named locality, however, only from cod stomachs. Spirontocaris fabricii (Krgyer). A2,3. B11,14. C 22, 23, 24,25. One large ovigerous female about 65 mm long out of a lot of five specimens from Parker Snow Bay (C 22) carries an adventitious supraorbital spine on the right orbital margin. This abnormality is of some interest, as the identity of the specimen is otherwise beyond question. This species, like Chionoecetes opilio, mentioned below, is found in Arctic Alaska, the Bering Sea, and Siberia, as well as from Casco Bay, Maine, to west Greenland. Miss Rathbun’ has already reported S. fabriciz from Port Burwell, Ungava, and the east side of Hudson Bay, where the Biological Board of Canada secured a specimen in 1930 north of Churchill in 30 fathoms; additional specimens were obtained from Wakeham Bay and from cod stomachs from Port Burwell, Ungava. Spirontocaris gaimardii (Milne Edwards). Al, 2,3. B 4, 6,38, 105i C 16, 22, 25-28, 31. D 32,34. E39, 40,42. Also found in cod stomachs at Angmagssalik, east Greenland, August 27, 1931. The Biological Board of Canada has taken S. gaimardi at two stations along the southwestern shore of Hudson Bay in from 20 to 38 fathoms; and again from cod stomachs at Port Burwell, Ungava. Spirontocaris gaimardii belcheri (Bell). A2. B7,10,18. © 20, 23, 25, 27, 29. D 32. Also found in cod stomachs at Angmagssalik, east Green- land, August 27, 1931. Von Hofsten (op. cit., p. 29) does not believe that the subspecies of S. gaimardi can be sustained as distinct entities, and gives his distribution records as for the species proper. Spirontocaris stoneyi Rathbun. E 39. The single specimen that Cap- tain Bartlett dredged north of Big Diomede Island, Bering Strait, consti- tutes the third known record for the species. Originally described from the Bering Sea,’ it has since been found at Shoal Tickle, southeast of Nain, Labrador. Crago dalli(Rathbun). E39. Sabinae septemcarinata (Sabine). A3. A single specimen was obtained from the stomach of a Ringed Seal secured off the Labrador coast in 1925. I cannot explain its absence from the dredge hauls made by Captain Bart- “s x Ratuswun, M. J. Decapod Crustaceans. Canadian Arctic Expedition, 7 (Pt. A): . 1919. Ae Ratuswun, M. J. Decapod Crustaceans. Harriman Alaska Expedition, X: 103. Ava. 15, 1936 VAN WINKLE AND SCHMITT: CRUSTACEA 329 lett, for it was taken by the second Parry Expedition off Igloolik Island near the Straits of Fury and Hecla and occurs on both coasts of Greenland. It is a more or less circumpolar species. In Hudson Bay it was taken by the Biological Board of Canada at nine stations in the southeastern central part of the Bay in depths ranging from 30 to 87 fathoms. Sclerocrangon boreas (Phipps). B 6,8, 10,13. C16, 18, 20, 22, 23, 25, Maer, 29:30, 51. D32,33, 35, 37.. 139) 42. Captain Bartlett found Sclerocrangon boreas in cod stomachs as well as in his dredgings from Angmagssalik, east Greenland, August 27, 1931. This shrimp grows to good size and is one of the principal articles of food of the square-flipper seal, Phoca barbata. One to several dozen could be recognized from among the stomach contents of four different specimens of this seal. The two largest Sclerocrangons taken from these seal stomachs were 45 and 5 inches long. One of these square-flipper seals was captured in the Straits of Fury and Hecla, another in Lyon Inlet, and two out in Fox Basin at approximately 66° 12’ N., 78° 59’ W. The depth of the water in this vicinity is 34 fathoms. Sclerocrangon boreas, says von Hofsten (op. cit., p. 80), is a panarctic form, also penetrating the boreal region in favorable localities. It prefers very cold waters and ground overgrown with algae. The matted, fibrous contents of the stomachs of two seals from Fox Basin, which contained numerous S. boreas substantiates this. In Hudson Bay the Biological Board of Canada obtained this crustacean at four stations more or less in the same latitude across the middle of the Bay at depths ranging from 30 to 72 fathoms; specimens were also collected at Nottingham Island, Hudson Strait; and at Port Burwell, Ungava from a cod stomach. Sclerocrangon ferox (Sars). D338. A truly high arctic form, found only in waters of very low, usually negative, temperature and at depths of ninety fathoms or more. We find but a single specimen in Captain Bartlett’s collec- tion, from 120 fathoms, near Cape Stosch, Hudson Land, east Greenland, 74° 04’ N., 17° 50’ W., July 30, 1931. This is the third specimen of the species ever to come to the National Museum and the first we have had from the western hemisphere. Argis lar (Owen). E39,40,41,42. With this species Stephensen unites the next, A. dentata Rathbun. Though known from Greenland to Nova Scotia, and from the Bering Sea to British Columbia and east to Siberia, in Captain Bartlett’s collection it is represented in hauls made only between Bering Strait and Kotzebue Sound in 1924. Argis dentata (Rathbun). A2. C20, 22,29,31. D382. Also found in cod stomach from Angmagssalik, east Greenland, August 27, 1931. The Biological Board of Canada has specimens of this species from eight stations well scattered throughout Hudson Bay, with depths ranging from 30 to 80 fathoms; and both free swimming and from cod stomachs from Port Burwell, Ungava. Pagurus krgyeri Stimpson. B15. A single specimen taken off Igloolik Island at the entrance to the Straits of Fury and Hecla. As European workers still merge this species with P. pubescens Krgyer, it is impossible to define the distribution of either species.-In the National Museum there is an extensive series of P. krgyerz from along the east coast of America as far south as Virginia, 37° 19’ 45’’ N., 74° 26’ 06” W., from 120 fathoms; several Labrador localities are represented: off Narak, Nain, Port 330 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 26, No. 8 Manvers, Cape Mugford, and Hebron, 7-60 fathoms; and Greenland: God- havn Harbor, Disko Island, and Hare Island, 70° 20’ N., 56° W., 90 fathoms. Further, there are specimens from the Firth of Clyde, 10-15 fathoms, and from Varanger Fiord, East Finmark. For the specimens of true P. pubescens in the collection of the Museum, the southern limit is off Cape Hatteras, 35° 42’ 00” N., 74° 54’ 30” W., 43 fathoms, while the northern limit is Kgg Harbor, Labrador, 7 fathoms. The Biological Board of Canada got a specimen from each of two Hudson Bay stations on the west side of the Bay north of Churchill, in depths of 30 and 63 fathoms; and another specimen from a cod capture at Port Burwell, Ungava. Pagurus trigonocheirus Stimpson. E 39, 40, 41, 42. Pagurus capillatus (Benedict). E 42. Pagurus splendescens Owen. E 39, 40, 41, 42. Pinnixa occidentalis Rathbun. E38. A single specimen from about the northern limit for this species, from the stomach of a fish caught 14% miles S.E. of Cape Cheerful, Unalaska. Chionoecetes opilio (O. Fabricius). E 40, 41. In its distribution, C. optlio is like Spirontocaris fabriciz, occurring in the Pacific boreal and arctic regions, as well as on the northeast coast of America and west Greenland, but not in east Greenland. Unlike S. fabric, its presence in Fox Basin can- not be established. Captain Bartlett’s three specimens of this species are from the arctic coast of Alaska. Hyas araneus (Linn.). A 3. Hyas coarctatus alutaceus Brandt. A 3. C. E 39, 40, 41, 42. Cap- tain Bartlett got one specimen of this species along the Labrador coast in 1925; fragments of 10 small individuals from a cod caught at Laxebugt, Disko Island, Greenland, in 1935; and 10 at four localities in the Bering Sea and Alaska in 1924. The Biological Board of Canada obtained this crab in James Bay; at Fort Churchill and Churchill River; and at six dredge stations in the eastern and northwestern parts of Hudson Bay from depths of 19 to 82 fathoms; in Hudson Strait from Charles Island, Nottingham Island, Sugluk Creek, and Eric Cove; and at Cape Wolstenholme, Wakeham Bay, and Port Burwell, Ungava. CRUSTACEA IDENTIFIED FROM THE STOMACH CONTENTS OF WHALES, SEALS, AND FISH Sperm whale. Decapoda: Chionoecetes opilio (O. Fabricius). Finback whale. Euphausiacea: Thysanoessa inermis (Krgyer), T. raschit (M. Sars). Sulphur-bottom whale. Euphausiacea: Thysanoessa raschit (M. Sars). Amphipoda: Themisto compressa forma bispinosa Boeck. Ringed or Floe-rat seal. Decapoda: Sabinea septemcarinata (Sabine). Euphausiacea: Thysanoessa inermis (Krgyer), T. raschii (M. Sars). My- sidacea: Mysits oculata (O. Fabricius). Amphipoda: Themisto libellula (Mandt), Gammarus locusta (L.). Bearded or Square-flipper seal. Decapoda: Sclerocrangon boreas (Phipps). Isopoda: Arcturus baffina (Sabine). Harp seal. Euphausiacea: Thysanoessa inermis (Krgyer). Amphipoda: Themisto libellula (Mandt). | Unidentified seal. Decapoda: Spirontocaris gaimardii belcheri (Bell). Mysidacea: Mysis oculata (O. Fabricius). Ava. 15, 19836 CHITWOOD AND CHITWOOD: NEMIC ESOPHAGI dol Codfish. Decapoda: Pandalus borealis Krgyer, Sprrontocarts groenlandica (J. C. Fabricius), S. polaris (Sabine), S. spina (Sowerby), S. garmardiz (Milne Edwards), S. gaimardii belcherz (Bell), Sclerocrangon boreas (Phipps), Argis dentata (Rathbun), Hyas coarctatus alutaceus Brandt. Amphipoda: Themisto libellula (Mandt), Anonyx nugax (Phipps), Pseudolibrotes nansent Sars, Gammarus locusta (L.), Gammaracanthus loricatus (Sabine). Unidentified fish. Decapoda: Pinniza occidentalis Rathbun. ZOOLOGY.—The histology of nemic esophagt. VI. The esophagus of members of the Chromadorida.t. B. G. Cuirwoop, Bureau of Animal Industry, and M. B. Cuirwoop. This paper is the sixth of a series (Chitwood and Chitwood, 1934- 1936) dealing with the structure of esophagi in representatives of various groups of nematodes. Previous papers in the series have covered representatives of the Rhabdiasidae, Strongylidae, Meta- strongylidae, Heterakidae, Rhabditidae and Anguillulinidae. The present paper covers representatives of the order Chromadorida, namely: Plectidae, Camacolaimidae, Axonolaimidae, Comesomati- dae, Cyatholaimidae, Tripyloididae, Desmodoridae, Chromadoridae, Monhysteridae, Linhomoeidae, and Siphonolaimidae. In representa- tives of the Chromadorida as in the other aphasmidian order, Eno- plida, absolute identifications of nerve cell, radial and marginal nuclei are often not possible, as there is too little distinction between the characters of these 3 types of nuclei, and the cell bodies of “nerve cells” are seldom observable. However, the distribution of nuclei is sufficiently similar in the various genera for homologies to be ascer- tained. In the following text, the authors have identified nuclear types to the best of their abilities. In some instances it has been possible to determine cytologically the identity of a given nucleus, while in other cases the position indicated that the nucleus in question was homolo- gous to one definitely identified in another form although they might differ cytologically in some respects. Future papers will include repre- sentatives of the orders Enoplida (EKnoplata, Dorylaimata, and Di- octophymata) and Spirurida (Camallanata and Spirurata). The data given in this paper are, for the most part, presented in tabular form (Figs. 3, 5, 10) and in illustrations, since the essay form of presentation would result in extended descriptions requiring much more space than the present form. The text calls attention to the major features given in the tables and illustrations, and presents some data not immediately obvious in the latter. Previous papers in 1 Received June 22, 1936. 332 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 26, NO. 8 this series supply extended descriptions of other forms, sufficient to orient the reader of the present paper. PLECTUS GRANULOSUS (Plectidae) Pigs.45 2.8 The esophagus of Plectus resembles in a general way that of Rhabditzs, as does the mouth cavity; it consists of an anterior cylindrical part, the corpus, connected with a posterior swelling or bulb by an indistinctly set off isthmus. The lumen of the esophagus is similar to that of Rhabditis in that in the precorpus it terminates marginally in well developed ‘‘tubes”’ (Fig. 1, B), while in the remainder of the esophagus the lumen is simple, i.e., without particular modification, except in the bulb where it is hexalobate at the valve. The precorpus contains 29 nuclei (Fig. 1, A-E) as follows: One group of 3 bilobed or 6 marginal nuclei, (mia_p—M3._»); 3 groups of 6 radial nuclei each (ri_¢, 7-12, and ri3-13); and 2 groups of nerve cell nuclei (mi_3 and n;_.). The marginal nuclear pairs, mj,» etc., may or may not be connected in such a way as to represent lobes of a single nucleus rather than individual nuclei; however, it is not possible to determine this on the basis of present material. A similar appearance is given by the marginal nuclei in Rhabditis, in which case the lobes were found to be joined anteriorly (Chitwood and Chitwood, 1936). The postcorpus contains 26 nuclei (Fig. 1, F-L); 6 marginal nuclei (m4._»- Mega_b), Similar to those of the margins of the precorpus, 6 radial nuclei (T1904), and 14 nerve cell nuclei (nj,7-19). The isthmus is too indistinct to be recognized as a unit. The most posterior nerve cell nuclei of the post- corpus might be considered as belonging to the isthmus. The prevalvar region of the bulb contains 14 nuclei (Fig. 1, O—-P), of which 6 are marginal nuclei (m7._p—Moa_p), 6 radial nuclei (125-30), and 2 nerve cell nuclei (ngo0_21). The postvalvar region of the bulb contains 24 nuclei as follows (Fig. 1, Q-S); 6 marginal nuclei or nuclear lobes (mioa—p—Miza_p), 6 radial nuclei in 2 groups (113-33, T3436), 9 nerve cell nuclei (nze-30), and 3 gland cell nuclei (g:_3). The marginal nuclear pairs of the postvalvar region of the bulb appear in all probability to represent lobes rather than individual nuclei. In the series illustrated (Fig. 3) mu is not double; ng3_24 were not observed. The orifices of the esophageal glands were not determined with absolute certainty. The dorsal gland appears to open into the lumen at the base of the stoma, while the subventrals appear to open at or near the level of ni7—19. The esophago-intestinal valve of Plectus is extremely well developed (Fig. 2, A-C) and consists of 23 nuclei. (Some of these nuclei are probably intestinal—compare Figs. 2, B—C and 2,1). The valve is laterally elongated and rather flat. Two other representatives of the same subfamily, very closely related to Plectus, were studied, these being Chronogaster gracilis and Wilsonema bacil- Ave. 15, 19836 CHITWOOD AND CHITWOOD: NEMIC ESOPHAGI 300 livorus, in both of which the nuclei of the esophagus appear to be similar in number and distribution to those of Plectus. ANONCHUS MIRABILIS (Plectidae) Figs. 2, D-F; 3 The esophagus of Anonchus mirabilis is cylindrical in the adult stage, but corpus, isthmus and pseudobulb are faintly recognizable in larval stages. Fig. 1—Plectus granulosus. A-F, precorpus; F—L, postcorpus; M-P, prevalvar region; Q-S, postvalvar region (See Fig. 2, B, for no9_30). The lumen is simple and triradiate, without marginal ‘‘tubes’’ (Fig. 2, E). The most striking pecularity of the esophagus is the presence of large chro- midial bodies in the marginal regions (Fig. 2, D). The nuclear number and distribution is nearly identical with that of Plectus granulosus, the following differences being noted: neop_21 and ne3_24 334 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 26, NO. 8 map Fig. 2.—A-C, Plectus granulosus. A, Longitudinal section through bulb and esoph- ago-intestinal valve; B—C, Cross sections, through esophago-intestinal valve (also includes very small part of bulb with no9_30). D-F, Anonchus mirabilis. D, corpus at base of stoma; E, corpus somewhat further posterior; F, bulbar region. G—I, Cama- colaimus prythercht. G, corpus; H, base of bulbar region showing g, and nos_30; I, esoph- ago-intestinal valve. J, Axonolaimus spinosus, corpus. K—M, Sabatierta vulgaris K, corpus; L-M, esophago-intestinal valve. N-—P, Paracanthonchus sp. N, anterior part of corpus O, corpus somewhat more posterior, P. esophago-intestinal valve. are in the dorsal parts of their respective sectors; nos and No. have not always been found; ne7 has not been observed; the marginals have been inconsist- ently observed owing to the confusion caused by the marginal chromidial bodies. Ava. 15, 1986 CHITWOOD AND CHITWOOD: NEMIC ESOPHAGI o00 The marginal nuclei are very faint and nothing definite can be said about them. The gland cell nuclei differ from the other nuclei in that several small nucleoli are present instead of a single large one (see Fig. 2, F). The eso- phageal glands are very large and there is a vesicle near the orifice of each gland. The dorsal gland opens into the lumen of the esophagus anterior to the level of the first radial group, while the subventrals open anterior to the level of ni7_19. ! The esophago-intestinal valve is similar to that of Plectus but the dorso- ventral form is more pronounced. Eleven large nuclei and possibly 4 smaller ones were observed in this structure. CAMACOLAIMUS PRYTHERCHI (Camacolaimidae) Figs. 2, 3 The esophagus of this species is cylindroid, slightly constricted at the nerve ring, and gradually enlarged posteriorly. The lumen is normally closed, simple or triradiate, and minute marginal ‘‘tubes’’ are present only in the anterior part of the esophagus (Fig. 2, G). There is a total of 81 nuclei, the same number as in Plectus if the marginals are considered as lobed. nuclei rather than double nuclei. The following differences in nuclear distribution in this species, as compared with Plectus granulosus, have been noted: No and ne are in the ventral or marginal regions of their respective sectors; Neg is anterior to mu; the fifth group of radials is subdivisible into 2 groups, T5, 27, and 23 being posterior tO 130, 26, ana 29 respectively; Tis ana 22 Were not al- ways observed. The esophageal glands are much more highly developed in Camacolaimus than in Plectus and occupy the greater part of the bulbar region (Fig. 2, H). The dorsal esophageal gland appears to open into the lumen slightly anterior to mi_3, while the subventrals open near the level of nis_ic. The esophago-intestinal valve (Fig. 2, I) is elongated, dorso-ventrally flattened, and contains 11 nuclei (only 8 shown in figure). The rudimentary stoma is surrounded by esophageal tissue and contains a dorsal tooth which is apparently not connected with the dorsal esophageal gland. APHANOLAIMUS sp. (Camacolaimidae) Fig. 3 The esophagus of Aphanolaimus is narrow, gradually enlarged posteriorly and without visible modified regions; the stoma is completely rudimentary and esophageal tissue extends to the anterior extremity. The esophageal lumen is simple, as is also the esophageal lining. The first, second and fourth groups of marginal nuclei are simple, while the third group is lobed as in Plectus. There are a total of 80 nuclei, corre- sponding to those of Plectus with the following exceptions, m7. and Mgax were inconsistently observed as lobes of m7, and mga, respectively; nz7 is posterior tO 21} Te5 ana 2g are nearly marginal in position; no3_2, were not ob- served but 1 nucleus, x1, marginal in position, and in the right subventral sector, may correspond to ngs. 336 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 26, NO. 8 The esophago-intestinal valve is similar to that of Camacolaimus. The esophageal glands are similar to those of Camacolaimus except that the dorsal gland nucleus is near the left side of the dorsal sector. AXONOLAIMUS SPINOSUS (Axonolaimidae) Pies. 2, 35.3 The esophagus of Axonolarmus is clavate, relatively short, and muscular throughout. This form is very close to Plectus in the character of the stoma and also in the presence of well developed marginal ‘‘tubes’’ in the corpus (Fig. 2, J). The nuclear distribution of the corpus is closer to that of Plectus than to any of the forms previously mentioned. The first marginal group appears to consist of 6 separate and distinct nuclei (mia_1»—Msa_sp); this is followed by the first 3 radial groups (1T1_¢,7_12 13-18) in series. The second mar- ginal group appears to be somewhat variable. In the series illustrated it con- sists of 7 separate nuclei, 4 (M41 442 ANd Moai oa2) being present in the dorsal sector, 2 (m4 and msz,) in the left subventral sector, and 1 (ms5,) in the right subventral sector. The fourth group of radial nuclei appears to consist of 6 or 8 nuclei (T1924) with the ventral nucleus in each subventral sector some- times doubled (reia,21, and Yee.,22»). The nerve cells of the corpus agree in number and position with those of Plectus. The nuclei of the bulbar region are difficult to identify, and the labelling given in Fig. 3 is to some extent arbitrary. The nuclei labelled c:_3; are large and bilobed; n23 ana 24 Also appear to be bilobed in some instances. The dorsal esophageal gland orifice is situated just anterior to the first group of marginal nuclei, while the orifices of the subventral glands appear to be near the level of nig_i9. The esophago-intestinal valve is well developed but not so elongated as in Plectus; it is distinctly triradiate anteriorly and rather circular posteriorly, containing 10 to 11 nuclei. It is very similar to the valve of Subatierza vul- garts. For purposes of comparison with Plectus, the total number of nuclei or nuclear lobes in the corpus is 58 (55 in Plectus) and in the bulbar region 35, 38 or 40 (38 in Plectus). DORYLAIMOPSIS METATYPICUS (Comesomatidae) Fig. 5 The stoma and esophagus of Dorylaimopsis are closest, among the forms thus far studied, to those of Axonolaimus. The marginal ‘‘tubes”’ are well developed, and there is a very slight thickening of the cuticular lining ex- tending throughout the corpus. The first and second marginal groups appear to consist of 3 bilobed nuclei. Data regarding the radial nuclei are not en- tirely satisfactory, but there appear to be 22 or 24 radial nuclei (in 4 groups) in the corpus, r7 and rip of Axonolaimus being the ones sometimes absent. The nerve cell nuclei (ni_19) of the corpus agree with those of Axonolaimus. The marginal nuclei of the bulbar region (m;_») and m1o_12) are simple. As in Axonolaimus, Ye; and Yes are posterior tO reg and reg, but res and ro are at the Ava. 15, 1936 CHITWOOD AND CHITWOOD: NEMIC ESOPHAGI ool level of re7 and res. The last group of radials presents the unusual arrange- ment of r32 and rz5, being anterior to the level of r33 and rs, which are in turn anterior to rs: and rs; the closest counterpart of this grouping is seen in Theristus. The nerve cells of the bulbar region correspond in number to those of Plectus, ngo_21 being just posterior to re7_23 and slightly ventral to the mid- L.SV. LISV-TV. J a PY-TSV-IDCILD TS ICD IDC ISV Tv} LE SS Ea PY. TSV-IDE EO TO ICD TDI Nee Se 24| Myo L.SV. TV TSV- TOC TOT DILDO JOLISV- TV. "34 Nee Ne Ga "ze o N2g = SELES KeATIER AC | G £7) AXONOLAIMUS Fig. 3.—Table of nuclear distribution. sector region; No3_»4 are subventral, posterior to 133-34; Noe 25,30 are distributed approximately as in Plectus; gi_3 are all in the center of their sectors between the levels of r33_34 and rz: and rz5. The esophageal glands of Dorylaimopsis are highly developed; the dorsal gland orifice is at the base of the stoma, and the subventral gland orifices are at the level of mis49. The glandular tissue is relatively much greater in this form than in Axonolaimus or Plectus but not as great as Camacolaimus. The esophago-intestinal valve is small and dorsoventrally flattened, and contains 12 nuclei. 338 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 26, No. 8 Fig. 4.—A-D, Theristus setosus. E-I, Terschellingia pontica. E, corpus; F, bulbar region; G-~—I, serial sections through esophago-intestinal valve. J, Desmolaimus zeelandicus v. americanus, longitudinal section through bulb, and esophago-intes- tinal valve. SABATIERIA VULGARIS (Comasomatidae) Figs. 2, K-M; 3 The esophagus of Sabatieria is like the esophagi of Axonolaimus and Dorylaimopsis in that well developed marginal ‘‘tubes” are present. In the gross morphology, shortness and relatively great development of the eso- phageal glands is more nearly like Dorylaimopsis than Axonolaimus. Un- Ava. 15, 1936 CHITWOOD AND CHITWOOD: NEMIC ESOPHAGI 309 like both of the mentioned forms, the posterior part of the stoma is collapsed and completely surrounded by well developed esophageal tissue. The nuclei of the esophagus agree substantially with Dorylazmopsis both in number and distribution (See Figs. 3, 5). The lumen of the esophago- intestinal valve is triradiate anteriorly, laterally elongated posteriorly; 12 nuclei are present in the valve. The dorsal esophageal gland has its orifice near the level of mi_3, i.e., posterior to the base of the collapsed stoma. The subventral glands have their orifices at the level of n1z, i.e., the level of the nerve ring. PARACANTHONCHUS sp. (Cyatholaimidae) Fig. 2, N—P The esophagus of Cyatholaimus presents several interesting modifications. The stoma is completely surrounded by esophageal tissue, and a dorsal tooth projects into the anterior part of the lumen. This tooth probably corresponds to one of the three present in Dorylaimopsis. There is no evidence of its connection with the dorsal esophageal gland which appears to have its orifice at the base of the stomatal region. The slightly clavate esophagus presents no external peculiarities except that the lumen (Fig. 2, N) is rather unusual, the general appearance being that of an axonolaimoid or plectoid form (Figs. 1, 2, J) in which the termini of the margins have become converging after having once been tubular; this is best visualized by comparison of the illus- trations. Both the marginal and radial nuclei are practically marginal in position; the marginal nuclei of the corpus are double. Since our series are incomplete, no attempt to give nuclear distribution will be made. However, the general impression is very similar to that given by Dorylaimopszis. The esophago-intestinal valve is short, markedly triradiate (Fig. 2, P) and consists of 8 nuclei which definitely belong to the valve, and 5 additional nuclei which also may belong to that structure; there are also 4 nuclei dorsal to the esophagus in a mass which apparently connects with the body wall. BATHYLAIMUsS sp. (Tripyloididae) The esophagus of Bathylaimus is cylindrical, and surrounds the conoid stoma. The lumen is nearly of a simple triradiate character; the esophageal radii extend nearly to the external surfaces in Theristus but the margins are very faintly rounded showing a similarity to Sabatzeria. The general outline of the esophageal cross-section is subtriangular, not unlike that of Theristus (Fig. 4, A) instead of nearly round as in the forms previously described. The position and form of the radial and marginal nuclei are also more like that of Theristus than of Cyatholaimus. The radial muscles are diffuse in attachment to the lining. Incompleteness of our series prevents us from giving further data on the structure of this form. The esophago-intestinal valve is triradi- ate. 340 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 26, No. 8 THERISTUS SETOSUS (Monhysteridae) Figs. 4, A-D; 5 The esophagus of Theristus is cylindrical, its lumen simple, and the lining unmodified. The dorsal esophageal gland opens near the base of the conoid stoma, while the subventral esophageal glands open near the level of nis_ie. The radial and marginal nuclei differ greatly in size as well as in position (Fig. 4, A). When clearly observed, the first 2 groups of marginal nuclei are GLE ss) ESS a) IEEE PV_TSV [Otic OT Dt fOLjsv.] V_| ae ES RE ee TC) PVT SVT OUCOT DT UD DLISV Tv n, 10 e Nzal TRIPYLIUM TERSCHELLINGIA Fig. 5.—Table of nuclear distribution. double, while the fourth is not doubled. The third group of radial nuclei con- sists of only 3 instead of 6 present in other forms. The esophago-intestinal valve is partially triradiate (anteriorly), but at the junction with the intes- tine it is definitely flattened dorsoventrally, practically identical with Ter- schellingia (Fig. 4, H); it contains 23 nuclei. MOoONHYSTERA CAMBARI (Monhysteridae) Fig. 5 The esophagus of Monhystera cambari is similar to that of Theristus, with the following exceptions, The representatives of the fourth group of radial nuclei (rig_21) are all at the same level or nearly so; no3_24 are ventrally sub- Avg. 15, 19836 CHITWOOD AND CHITWOOD: NEMIC ESOPHAGI 341 marginal just posterior to the level of 130-31}; Te9-32 and rgo_s1 are at levels in reverse of the levels at which they are, respectively, in Theristus; all of the marginals are simple except my: (not doubled). The esophago-intestinal valve is similar to that of Theristus and Terschellingia; 19 nuclei were observed. TRIPYLIUM CARCINICOLUM V. CALKINSI (Linhomoeidae) Figs. 5, 6 The esophagus of Tripylium is cylindroid; anteriorly it surrounds the prismoidal stoma; posteriorly it is connected with the intestine through an enlargement commonly termed the “bulb” though this structure is not a Fig. 6.—Tripylium carcinicolum v. calkinsi. A, stomatal region; B—C, corpus; D, bulbar region; E—F, esophago-intestinal valve. part of the esophagus. The lumen is modified throughout due to the thick- ened cuticular attachment points of the lining (Fig. 6, C). This type of lumen appears to be derived from a plectoid or rhabditoid type in which there has been a disappearance of the marginal tubes and a development of attach- ment points for the radial muscles; such a phenomenon occurs in the develop- ment of forms such as Oesophagostomum. The nuclei are distributed as in Theristus, with the following exceptions, The marginal nuclei are not doubled and are directly marginal in position (Fig. 6, B); reo and ro; are at the same level as 13 and ro; reg and 133 are POs- terior to Yeo and 39; No3_94 are ventral submarginal and immediately anterior to the level of nog_30, nos being just anterior to Me3_94. The esophago-intestinal valve is dorsoventrally flattened (Fig. 6, E-F), but some explanation regarding this region is necessary. Baylis (1915) and 342 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 26, NO. 8 Cobb (1920) have considered the swelling as an esophageal “bulb,” and ac- tually this structure is formed by the esophago-intestinal valve and the in- testine. At least 12 nuclei are ectodermal in origin; the external tissue forming the ‘“‘bulbar’”’ enlargement, probably endodermal in origin, contains 4 nuclei. The cytology of the latter is identical with that of the intestinal cells following it. In the related linhomoeid, Desmolaimus, there is a cylindrical elongated structure (Fig. 4, J) between the esophagus and intestine; the definite ecto- dermal valve tissue makes up only a small part of the esophago-intestinal cylinder. However, in this case the tissue is peculiar in being basophilic and differing both from the esophagus and the intestine. The esophagus is similar in all respects to that of T’rzpyliwm except for a well developed chromodoroid swelling at its base. On the basis of comparison with the postesophageal structure of linhomoeids, this structure must be considered as a new organ. TERSCHELLINGIA PONTICA (Linhomoeidae) Figs. 4, E-I; 5 The esophagus of Terschillingia consists of a cylindrical corpus, following a rudimentary stoma, and a well developed bulb; the latter is actually a part of the esophagus, not a homologue of the so-called bulb of Tripylium but a homologue of the bulb of Desmolaimus. The lumen is subdistally dilated (Fig. 4E, F); the musculature is concentrated, but no cuticular thickenings of the lining are present. The nuclear distribution (Fig. 5) is rather similar to that of Tripylium, Theristus, and Monhystera, but only three groups of radial nuclei were observed in the corpus; no3_24 and Neg were not observed. The dorsal gland orifice is slightly anterior to the level of the first group of marginal nuclei (mi_3) while the subventral gland orifices are at the level of Nis_i6. The esophago-intestinal valve (Fig. 4, G—I) is very strongly dorso- ventrally compressed and contains at least 19 nuclei, probably more. It is surrounded by intestinal epithelium as in Tripylium. SIPHONOLAIMUS CONICUS (Siphonolaimidae) Fig. 5 Stphonolaimus is a peculiar form concerning the stoma and esophagus of which there has been much discussion. Being limited to a study of a single series of sections, the writers present the results of their study of this form with as little interpretation as possible. The minute stoma seems to be in the form of a stomatostyle surrounded by muscular tissue which consists of 6 strands passing posteriad and closely applied posteriorly to the anterior end of the esophagus, then passing to the body wall where they are inserted sublaterally. The stomatostyle overlies the anterior end of the esophagus in which 8 distinct cavities are present. Whether or not these cavities represent the 3 esophageal glands is not known; only the dorsal cavity is traceable posteriorly past the isthmus. In the corpus a total of 37 nuclei were observed, 16 corresponding to n1i_16¢ of other forms, while the remainder are marginal and radial nuclei. The lat- ~ Ava. 15, 1936 CHITWOOD AND CHITWOOD: NEMIC ESOPHAGI 343 ter probably correspond to m1_3, 1s, 7-12 and Yi9—24 Other forms, i.e., the third radial group and second marginal group are apparently absent. The bulbar region contains 32 nuclei. The esophago-intestinal valve is dorsoven- trally flattened and contains 6 nuclei which definitely belong to the valve. MonopostHIA HEXALATA (Desmodoridae) fies: 7, 10 The esophagus of Monoposthia is, in general, typical of the Chromo- doroidea and Desmodoroidea. The prismoidal stoma (Fig. 7, A) is sur- rounded by esophageal tissue which protrudes anteriorly into the lumen in Fig. 7.—Monoposthia hexalata. A—C, corpus; A, in stomatal region; B, just pos- ag ie orifice of dorsal gland; C, near base; D—G, bulbar region; H, esophago-intes- tinal valve. the form of a large dorsal tooth. The post-stomatal region of the esophagus consists of a cylindroid corpus gradually enlarged posteriorly and joined directly to the short thick bulb. The lumen of the corpus is very minute, showing but faint indications of terminal ‘‘tubes.”’ The corpus contains 55 nuclei as follows: 2 groups of 3 marginal nuclei (mi_3 and m4_¢); 4 groups of 6 radial nuclei (ri_¢, 7-12, 18-24); 19 nuclei (mi_i9) presumably of nerve cells; and 4 nuclei (si_s) possibly of nerve cells. The bulb contains 40 nuclei as follows: 2 groups of 3 marginal nuclei (m7_9, Mio-12); 12 radial nuclei in 3 groups (re5_30, F3i—33, T3436); 8 gland nuclei (gi-3); 10 presumptive nerve cells (nz2-39); and 8 possible nerve cells (ss_19) ; and 1 nucleus of uncertain type (x1). The dorsal esophageal gland appears to open into the lumen of the esoph- agus near the level of ni, while the subventrals appear to open near the anterior end of the bulb (around the level of noo_21). 344 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 26, NO. 8 The esophago-intestinal valve is short, more or less triradiate, and con- tains about 12-15 nuclei (Fig. 7, H). Metachromadora, a closely related form of the same family, has an esoph- agus similar in a general way to that of Monoposthia but the esophageal lining has well developed thickenings for the attachment of radial muscles, particularly well developed in the bulbar region. Fig. 8.—Ethmolaimus rivaliensis. A, stomatal region; B, corpus; C—G, bulbar region; H-I, esophago-intestinal valve. ETHMOLAIMUS RIVALIENSIS (Chromadoridae) Figs. 8, 10 This form is nearly identical in esophageal structure with Monoposthia. The stoma, lumen and gross morphology are similar. The nuclei of the corpus (55) appear, to be similar in character and distribution, there being 6 margin- al nuclei, 24 radial nuclei, and 4 questionable (si_s). Most of the nuclei (total 39) of the bulb likewise correspond to those of Monoposthia, with the following exceptions, s;_, were not recognized but 2 additional nuclei were sometimes observed in the left subdorsal sector. The esophago-intestinal valve is short, dorsoventrally elongated, and. con- tains 13 nuclei. Aug. 15, 19836 CHITWOOD AND CHITWOOD: NEMIC ESOPHAGI 345 CHRoMADORA sp. (Chromadoridae) Figs. 9, D-K; 10 The esophagus of this form is similar to that of Hthmolaimus and of Monoposthia, with the following exceptions: Grossly the stomatal region is not set off from the remainder of the esophagus and the bulbar region is relatively much shorter and smaller; the distinction between marginal and radial regions is more marked; 6 additional nuclei were observed in the corpus (xi_s); the third group of radial nuclei is double (hence 6); 85, S¢, Neea Fig. 9.—A-C, Microlaimus sp. A, corpus; B—C, esophago-intestinal valve. D-K, Chro- madorasp. D-I, serial sections through bulbar region; J—K, esophago-intestinal valve. (of Monoposthia) were not observed; additional nucleus (x7) right laterodorsal was observed near the level of g1, otherwise the nuclei in the bulbar region are as in Monoposthia. The esophago-intestinal valve is dorsoventrally elongated, consisting of 12 nuclei. Chromadora exhibits a pair of subdorsal pigment spots which consist of masses of brown granules in the subdorsal marginal and submarginal areas of the esophagus just posterior to the stoma- tal region; no special cells were observed in association with the spots. Microuaimus sp. (Microlaimidae) Figs. 9, A—-C; 10 The esophagus of Microlaimus resembles that of Chromadora more closely than any other of the forms studied. The subtriangular stoma is surrounded by esophageal tissue, the corpus is cylindrical, the bulbar region quite en- larged and the esophago-intestinal valve elongated. The lumen (Fig. 9, A) is without marginal tubes and the lining without thickened cuticular attach- ment points though the radial fibers are highly concentrated. The number of nuclei and their disposition is clearly most like Monoposthia, Ethmolaimus 346 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 26, No. 8 and Chromadora (compare Figs. 7-10). However, in addition to the nuclei S1_1o, there is a group of 6 nuclei (¢i_s) apparently resulting from further division of the second radial group (r7_-12). The elongate form of the esophago- intestinal valve grossly recalls the form of that structure in Camacolaimus, Plectus, or Terschellingia, but the cross section (Fig. 9, B—C) clearly indi- cates other relationships since it is plainly triradiate. It contains 11 nuclei. GBa Se a Pe BSS RoE Se TDCICOT IED LTS LW Gil Gg Ee a ES SV isV.[OLTLO] OLD} Dt] sv.T Vv] Fig. 10.—Table of nuclear distribution. LITERATURE CITED Bayuis, H. A. Two new species of Monhystera (nematodes) inhabiting gill chambers of land crabs. Ann. & Mag. Nat. Hist. 8s., 16: 414-421. 1915. Cuitwoop, B.G.and M. B. The histology of nemic esophagi. Parts I-II. Z. Zellforsch. 22: 29-37, 38-46. 1934. Part III. This Journat 24: 557-562. 1934. Part IV. This JouRNAL 25: 230-237. 1935. Part V. This JouRNAL 26: 52-59. 1936. Cops, N. A. One hundred new nemas. Contrib. Sc. Nemat. IX: 217-348. 1920. . i RR 5 rhe aye? th mas : xf CONTENTS PALEOBOTANY.—Plant distribution as a guide to age doterinstiiea Rawr W. CHANEY... 2.60660 e eee eee tee teen ence ees : ZooLocy.—Notes on the Crustacea, chiefly Natantia, collected. aah Captain Robert A. Bartlett in Arctic Seas. Marcarer E. Van WINKLE and Watbo L: SCHMITT... 202..-.4.5 74a en eee ZooLocy.—The histology of nemic esophagi. VI. The esoph: members of the Chromadorida. B. G. Currwoop and M. B. CHITWOOD. «6-10 eee eee eee ene eter eett ee et teen ee This Journal is indexed in the International Index to Periodicals ry , : 2a &y hd h He i SEPTEMBER 15, 1936 No. 9 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES BOARD OF EDITORS F. G. BricK WEDDE Rotanp W. Brown Espen H. Tooie BUREAU OF STANDARDS U. S&S. GEOLOGICAL SURVEY BUREAU OF PLANT INDUSTRY ASSOCIATE EDITORS H. T. WENSEL Harotp Morrison PHILOSOPHICAL SOCIETY ENTOMOLOGICAL SOCIETY E. A. GoLDMAN W. W. Rusey BIOLOGICAL SOCIETY GEOLOGICAL SOCIETY AGNES CHASE J. R. SWANTON BOTANICAL SOOIETY ANTHROPOLOGICAL SOCIETY R. E. Grsson CHEMICAL SOCIETY PUBLISHED MONTHLY BY THE ) Va f WASHINGTON ACADEMY OF SCIENCES ™ NAL ie: 450 Aunatp Sr. es aT MpnasHa, WISCONSIN Entered as second class matter under the Act of August 24, 1912, at Menasha, Wis. Acceptance for mailing at a special rate of postage provided for in the Act of February 28, 1925. Authorized January 21, 1933. Journal of the Washington Academy of Sciences This JouRNAL, the official organ of the Washington Academy of Sciences, publishes: _ (1) short original papers, written or communicated by members of the Academy; (2) _ proceedings and programs of meetings of the Academy and affiliated societies; (3) notes of events connected with the scientific life of Washington. The JouRNAL is issued monthly, on the fifteenth of each month. Volumes correspond to calendar years. Manuscripts may be sent to any member of the Board of Editors. It is urgently re- quested that contributors consult the latest numbers of the JourRNaAt and conform their — manuscripts to the usage found there as regards arrangement of title, subheads, syn- onymies, footnotes, tables, bibliography, legends for illustrations, and other matter. Manuscripts should be typewritten, double-spaced, on good eng Footnotes should — be numbered serially in pencil and submitted on a separate sheet. The editors do not assume responsibility for the ideas expressed by the author, nor can they undertake to correct other than obvious minor errors. Illustrations in excess of one full-page halftone or two full-page line drawings may be paid for by the author. Line drawings are preferred. Proof.—In order to facilitate prompt publication one proof will generally be sent to authors in or near Washington. It is urged that manuscript be submitted in final form; the editors will exercise due care in seeing that copy is followed. Author’s Reprints.—Fifty reprints without covers will be furnished gratis. Covers — bearing the name of the author and title of the article, with inclusive pagination and date of issue, and reprints, will be furnished at cost when ordered, in accordance with the following schedule of prices. : Copies 4 pp. 8 pp. 12 pp. 16 pp. Covers 50 a oes es C ekna ai hee a $2.00 100 $3. .99 $1.20 $1.50 $2.00 Ne 150 1,25 1.90 2.55 3.25 3.50 200 1:75 2:60: :* 3.60 4.50 4.25 250 2.00 3.30 4.65 5.75 5.00 Envelopes for mailing reprints with the author’s name and address printed in the corner may be obtained at the following prices. First 100, $4.00; additional 100, $1.00: | Subscription Rates.—Per volume. ...........0se0ceereuse vivian 4 Obie te bike . -$6.00 To members of affiliated societies; per volume. .............eceeeeeceees 2.50 Single NUMIbeIS . Fs CF Ea TE ow seni oe re ack Cie at Oe Oe soe ee ot Cee Remittances should be made payable to ‘‘Washington Academy of Sciences” and addressed to 450 Ahnaip Street, Menasha, Wis., or to the Treasurer, H. G. Avers, Coast and Geodetic Survey, Washington, D. C Exchanges.—The Journat does not exchange with other publications. Missing Numbers will be replaced without charge provided that claim is made to the Treasurer within thirty days after date of following issue. OFFICERS OF THE ACADEMY President: O. E. Mrrnzpr, U.S. Geological Survey: Corresponding Secretary: N. R. Smiru, Bureau of Plant Industry. Recording Secretary: CHARLES THOM, Bureau of Chemistry and Soils. Treasurer: Henry G. Avurs, Coast and Geodetic Survey. JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VoL. 26 SEPTEMBER 15, 1936 No. 9 OCEANOGRAPHY.—Simple portable tide-gages..1 CHESTER K. WENTWORTH, Board of Water Supply, Honolulu, T. H. (Com- municated by W. W. RuBEY.) The tide-gages described below have been devised in connection with the study of bench-forming processes on the island of Oahu. They are described here in order to indicate the convenience and ac- curacy of simple apparatus of this sort for use by various naturalists who may have occasion to make moderately precise local measure- ments of sea level or tidal fluctuations. Their chief value is for those who require determinations within 0.1 or 0.2 foot and who wish to avoid the bias of off-hand estimating of sea-level which may result from use in estimating, of just those features whose level is to be determined. Even for such rather rough estimates, an instrumental method applied directly to the water level has superior validity, espe- cially when used by various persons. THE MANOMETER TYPE. This consists of a reservoir, such as a gallon syrup jug, to which are connected, (a) a syphon and 25 to 50 feet of 4 inch rubber hose, (b) a mercurial manometer made up of glass tube with rubber hose bends bound to a section of meter stick, or prepared scale and, (c) a short vent tube. By applying suction to the jug, a water level may be maintained in it which is above the average of wave fluctuations in the sea to which the long tube reaches, by an amount indicated by the reading of the mercurial manometer. In actual use the jug is nearly filled with sea water and the sea tube tossed into the sea after starting the syphon. After the syphon has run long enough to fill the tube with water, the vent tube is closed and the syphon continues to run until rarification of air in the chamber has reached a balance with the water-level difference. This occurs within one or two minutes and the amount of water level differ- ence can be computed from the mercury difference shown on the manometer, or a scale can readily be constructed so that the manom- eter reads directly in feet of sea water. The effect of three- or four- 1 Received March 16, 1936. 347 SEP 18 1938 348 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 26, No. 9 foot waves in the sea on the water surface within the jug is negligible and the pulsation of the manometer is comparatively slight, owing to the damping effect of the long tube. The general form of the ap- paratus and its relations when in use are shown in Figs. 1 and 2. O Vent Tvbe Taromerer Fig. 1.— Manometer type of portable tide-gage. Glass parts in outline, rubber parts in black, supports and scale omitted. OTHER TYPES OF TIDE-GAGE. ‘Two other principles were applied in devising simple gages. In a dilatometer type, a small air chamber is closed off with air at atmospheric pressure above a column of sea water extending down the sea tube and rising to a point marked ona gage tube. Then a lower valve is opened and the water level drops to a new position of balance in the gage tube. The increase in volume of the air is a measure of decrease of pressure, by Boyle’s law, and it is easy to compute, or to read on a calibrated scale, the indicated difference between final water level in the gage tube and the mean » of sea level pulsations. A capillary or some other form of damping is necessary in this type to secure accurate readings. An orifice type was also used, based on the principle that rate of flow through a given orifice is a function of head. The procedure is to measure the time required for a fixed amount of water, as from a sort of bulb-pipette, to flow through a small glass jet under the pull Sept. 15, 1936 WENTWORTH: TIDE-GAGES 349 due to pressure difference between the mean levels in the pipette and in the sea, as connected by the water column in the sea tube. This device achieves an automatic integration and measurement of mean differences through the half minute or more of flow, as built and Lif? = 13.22 Ll yg Soe) OB aa /Weas7 Water Leve/ Fig. 2.—Diagram showing relations of apparatus to sea level in operation. The value of lift of sea water and difference of mercury level in the manometer are in the ratio of densities of sea water and mercury, respectively. calibrated. A few measurements will readily determine enough points on a curve to achieve such a calibration, and determine the value of B in the formula 1 B . t 350 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 26, NO. 9 where h is difference of head, ¢ is time of flow, and A a constant fixed by the volume and size of the jet. Both the dilatometer and the orifice types are See tae casual single measurements by the operator from a standing position, whence a hand level sight can then be made. Naturally neither of them is so accurate or so convenient for a longer series of measure- ments in determining a semi-permanent bench mark as the manom- eter type. The manometer type also has the advantage that from 15 to 30 successive measurements may be made with a single filling of the jug, by venting the air chamber after each reading and waiting 1 to 2 minutes for a new balance to be reached. : PERFORMANCE OF THE MANOMETER TYPE. It will already have occurred to the reader that the results of any single measurement in- dicate only the difference between the water level in the reservoir and the contemporary level in the sea. Such a measurement has to be corrected by the amount by which such contemporary sea level lies above or below mean sea level. This is readily done by means of a table incorporated in the standard tide tables, or by graphic methods, given the time and height of adjacent high and low tides. The following tables indicate the performance obtained with the manometer type of gage. TABLE 1.—AccuRAcY OF WATER LEVEL MEASUREMENTS® Average No. Maximum Absolute Probable error Probable error Selection readings deviation, maximum single reading, single reading, in group average deviation average extremes Whole series 4.5 Ont3 030" 0.057 0.1438 50 groups O20u8 Worst 10 4.2 0.24 0.30 0.110 0.143 groups? 0.092 Best 10 3.9 0.045 0.07 0.021 OL02% groups 0:01 * 226 individual determination, 50 groups. Figures in feet. Includes several series taken when waves 4 and 5 feet high were running past the sea tube. ° Omitting a series of four identical readings, giving probable error of 0.000. In view of the fact that most of the groups contain four readings, the probable error of the mean of readings in a given group is about half the probable error of a single reading.? Thus it appears that while Da? 2 Probable error of single reading = .6745 ye n Probable error of mean of n readings = .6745 —. nN VN Sept. 15, 1936 WENTWORTH: TIDE-GAGES 351 exceptional single readings in a long series may vary from the con- temporary mean water level by 0.3 feet, the expected error of a single reading (P.E.) is under 0.06 and the probable error of an average of four readings will commonly not exceed 0.03 feet. These are instru- mental and manipulative errors; the value to be measured is the exist- ing local contemporary water level, with such abnormalities as may result from winds, coastal configuration and the like. It is recognized, especially through the recent work of Johnson? and others that mean sea level, or any other position of sea level fluctuations, is not at all a regular or uniform level or surface and that in its measurement in a standard or regional sense, or for comparisons with past or future positions, requires selection of a site as free as possible from local or temporary peculiarities. On the other hand physiographic and biologic features are formed in relation to the existing average or extremes of water level at a given time and place and it appears appropriate to measure such local datum surfaces. Indeed such locally measured sea levels may in places be more signif- icant than a regional mean sea level established by precise level from a remote bench mark. Aside from the instrumental errors made at a given time, the other chief source of error in fixing mean sea level locally from a short series of measurements is in computing the tidal stage at the time of meas- urement. In the writer’s studies it has been assumed that the tide stages between adjacent high and low positions followed a sine curve. Thus the stage of tide can be computed by the formula [Page 7 oe fae: DA asl ll +3R.( —— |) where H, is the contemporary tidal stage to be determined, H., the stage midway between adjacent high and low tide stages, R. the difference between the adjacent extremes, and 7, Tn, Tx:, T1, the clock times of the contemporary observation, the time midway be- tween adjacent extremes and times of high and low tide respectively. If algebraic signs are used strictly this formula applies; in most cases the proper sign of the last term will be apparent by inspection.‘ In order to determine the validity of mean tide as computed from single short series of measurements made at random tidal positions, a total of 9 series of measurements was made at one readily accessible 3 Jounson, D. W., Studies of Mean Sea Level. National Research Council, Bull. 70. 1929 * A table for making these computations is also carried in the annual Tide Tables published by the Coast and Geodetic Survey. 352 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 26, NO. 9 station on the Oahu coast, just west of the Koko blow hole. Results of these measurements are shown in the following table. TABLE 2.—SuUCCESSIVE SEA LEVEL MEASUREMENTS AT ONE STATION Tidal time (fraction Elevation No, | ofinterval beforgor | ‘Tidalstage | Tift “Above tidal | Abovemean | from mann datum level 1 +0.29 a7 3.92 5.49 4.69 10% 2, —0.70 .18 Heo 5.48 4.68 .08 3 —0.23 1 yk 3.92 5.64 4.84 .08 4 +0.16 1.45 4.07 5.52 4.72 .04 5 +0.22 Ik 35 7% 4.16 5.53 4.73 .03 6 +0.27 1.29 4.20 5.49 4.69 .07 7 —0.07 1.88 3.78 5.66 4.86 10 8 —0.03 1.90 Sat 5.61 4.81 .05 9 0.00 1.90 Sneha 5.65 4.85? .09 ® Average for this column is 4.763. Each of the determinations in the above table is the average of several individual measurements, in most instances, 12, and may be regarded as not over 0.01 or 0.02 in error as related to mean con- temporary water level. Extreme variation in the determination of sea level is 0.10 feet; the standard deviation is 0.071 and the probable error of an individual determination is 0.048 feet. Both theory and practice indicate that errors are likely to be greater in measurements made in narrow, shallow inlets and in those made at low stages of tide. At any rate it appears that with the low tidal range of Oahu (extremes, 3.0 feet) and with the moderate variations of water and air temperatures found in the tropics, mean water level may be de- termined within three or four. hundredths of a foot in not over ten minutes with a readily portable apparatus. On coasts of greater tidal range errors in such brief measurements will be proportionately greater, but it may be presumed that here, with greater difficulty of making estimates, such measurements will be equally acceptable in this type of work. Also in higher latitudes more careful calibration at different temperatures may be required. If more precise measure- ments are required, the manometer type of gage is easily adapted to a long series of readings, which will reduce observational errors as desired and may be used to throw much light on the form of the tidal curve. Sept. 15, 1936 BROWN: GLYPTOSTROBUS 353 PALEOBOTAN Y.—The genus Glyptostrobus in America. ROLAND W. Brown, U. 8. Geological Survey. Glyptostrobus is a coniferous genus represented today by a single species, G. penszlis Koch, called water-pine, occurring only a few feet above sea-level near Canton and Foochow in the coast region of southeastern China. Like the maidenhair tree, Ginkgo biloba Lin- naeus, the water-pine apparently no longer exists in the wild state, but is cultivated and protected. It is planted as a binder along the banks of canals and watercourses and as a windbreak on the out- skirts of villages. Although somewhat smaller and less hardy than the swamp cypress, T'axodium distichum (Linnaeus) Richard, of the southeastern United States, the water-pine resembles the latter in general appearance, form, and habit. Both species produce deciduous annual branchlets, heterophyllous foliage, and, when growing in ex- tremely wet situations, send up from their roots those conical, woody, breathing organs called ‘‘knees,”’ upright in Tazxodium but bent over in Glyptostrobus. The resemblances between Glyptostrobus and Taxodium led to a confusion of at least one living species of Taxodiwm, T’. adscendens Brongniart, with Glyptostrobus pensilis; but the differences between the two genera were clearly stated in 1926 by Henry and McIntyre,? so that now they may be readily distinguished as follows: Glyptostrobus.—Leaves cupressoid on branchlets bearing flowers and cones; flowers and cones solitary and terminal on erect branchlets; cones persistent after ripening, pyriform, with overlapping, elongated, non-peltate scales attached to a disc at the base, without resin glands on inner face; seeds small, ovoid, with a long terminal wing; tracheids narrow; bordered pits on radial wall of tracheids, mostly uniseriate; bordered pits on tangen- tial wall of tracheids in autumn wood, few and widely spaced; simple pits on radial wall of parenchyma cells, few, on tangential wall, none; ter- minal wall of parenchyma cell, thin; simple pits in each crossfield 2 to 6, mostly 3 or 4, oval, in one or two horizontal rows; wood rays 2 to 14 cells high, with thin terminal walls; transverse section shows resin cells arranged in 1 to 3 irregular bands. Taxodium.—Leaves acerose on branchlets bearing flowers and cones; flowers and cones numerous and lateral on pendulous branchlets; cones dis- integrating after ripening, globose or ellipsoid, with peltate scales fitting to- gether at the edges, exhibiting a rhomboidal or diamond-shaped exposed face and having resin glands on inner face; seed large, triangular, without wing; tracheids almost twice the width of those in Glyptostrobus; bordered pits on radial wall of tracheids, mostly multiseriate; bordered pits on tan- 1 Published by permission of the Director, U. 8S. Geological Survey. Received April 3, 1936. 2 Henry, AuGustTingE, and McIntyre, Marion. The swamp cypresses, Glypto- strobus of China and Taxodium of America, with notes on allied genera. Proc. Royal Irish Acad. 37, sect. B, no. 138: 90-116. 8 pls. 1926. 354 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 26, NO. 9 gential wall of tracheids in autumn wood, more numerous than in Glypto- strobus and some closely spaced; simple pits on radial wall of parenchyma cells more numerous than in Glyptostrobus, on tangential wall may occur in autumn wood; terminal wall of parenchyma cell conspicuously thickened or nodose; simple pits in each crossfield 1 to 5, mostly 3, circular, in one Fig. 1.—Glyptostrobus oregonensis, branch with cones and cupressoid foliage. Figs. 2, 3.—Cone-scales, outer and inner faces, respectively, of G. dakotensis. Fig. 4.— Cryptomeroid foliage of G. dakotensis, found with cone-scales, Figs. 2, 3. Fig. 5.— Cone of swamp cypress, Taxodium distichum. Fig. 6.—Cone of water-pine, Glypto- pyle pensilis. Fig. 7—Cone of pond cypress, Taxodium adscendens. All figures natural size. horizontal row; wood rays 2 to 20 cells high, with thick terminal walls; trans- verse section shows resin cells usually arranged in one band. Although the living species, Glyptostrobus pensilis, can now be dis- tinguished readily from species of Taxodium and other coniferous genera, the detection and separation of the fossil representatives of Glyptostrobus is fraught with considerable uncertainty. This is par- ticularly true when foliage only is available. In the living species this Sept. 15, 1936 BROWN: GLYPTOSTROBUS 355 may be of three kinds—cupressoid, taxodioid, and cryptomeroid, in allusion to the typical foliage, respectively, of Cupressus, Taxodium, and Cryptomeria. A given fossil shoot or twig within this range might therefore merit any one of four interpretations, let alone being con- fused with other genera, such as Sequoia, Cunninghamia, Torreya, Juniperus, Tsuga, etc. Because of the uncertainty concerning the identity of such twigs as are ordinarily preserved in shale and sand- stone, those recorded fossil species of Glyptostrobus based upon foliage alone will not be discussed here but will be regarded as doubt- ful identifications. Attention will be centered on those species purport- ing to be based upon authentic, characteristic cones or cone-scales. The origin and early history of Glyptostrobus is shrouded in ob- scurity. A few species, based upon foliage, have been recorded from the Cretaceous of North America, Europe, and Australia, but these must be considered as of doubtful status until cones are discovered in association or connection with the foliage. From the Eocene onward, however, remains of cones or cone-scales in association with foliage have substantiated the presence of species of Glyptostrobus in the northern hemisphere, circumpolar in the Tertiary but later reduced to one species restricted to southeastern Asia. Two Tertiary European species have been described—G. europaeus (Brongniart) Heer and G. ungerz Heer—but because the latter appears to be only a phase or variety of the former, the tendency now is to synonymize both, thus leaving G. europaeus as the only species to represent the genus during the Tertiary in the eastern hemisphere.* In North America the first remains strongly suggesting Glypto- strobus are scattered cone-scales and cryptomeroid-cupressoid foliage, described from the Fort Union (Eocene) formation of North Dakota as G. europaeus by J. 8S. Newberry. For reasons that will be given im- mediately, I shall rename these specimens. Glyptostrobus dakotensis, new name Figs. 2-4 Glyptostrobus europaeus (Brongniart) Heer. Newberry, J. 8., U. 8. Geol. Survey Mon. 35: 24, pl. 26, figs. 6-8a; pl. 55, figs. 3, 4, 1898. Remarks.—This species may be described as being different from G. europaeus in several specific respects: 1. The cone-scales are, on an average, shorter and broader. 2. The cone-scales occur scattered and detached, a characteristic apparently not shown by G. europaeus or by the living G. penstlis, whose cones remain intact and do not disintegrate readily after ripening. 3. The species occurs in the lower part of the American Eocene, a considerable time interval from typical G. europaeus in the European Mio- cene. 3 See synonymy given by E. W. Berry in Lower Eocene floras of southeastern North America. U.S. Geol. Survey Prof. Paper 91: 169. 1916. 356 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 26, No. 9 Occurrence.—Numerous localities in the Fort Union formation in North Dakota and Montana. Figured specimens in U. S. National Museum. The collections made by I. C. Russell, in 1902, from light-colored, tuffaceous Miocene beds near Beulah, Malheur County, Oregon, contain numerous leaves of a chinquapin, Castanopsis convexa (Les- quereux) Brooks, and the coniferous branch with cones to be de- scribed here as Glyptostrobus oregonensis, hew name Fig. 1 Glyptostrobus linguaefolia (Lesquereux) Brooks. Brooks, Betty W., Annals Carnegie Mus. 24: 281, pl. 4, fig. 6, 1935. Remarks.—This is the specimen referred to by Brooks in her discussion of G. linguaefolia, cited in the synonymy, and said to be in the U.S. National Museum collections. It carries a number of small annual branchlets covered with cupressoid, lingulate foliage and terminated by characteristic pyriform, glyptostroboid cones averaging 2 cm in length, with elongated, imbricated scales. Similar twigs, but no cones, have been reported from approximately contemporaneous deposits on Sucker Creek near the Oregon-Idaho bound- ary. The similarity of this specimen to the figures of G. europaeus (Brongniart) Heer* and G. ungerz Heer’ is striking, but there are differences that, in my opinion, justify the segregation of it as a species distinct from the American Eocene G. dakotensis on the one hand and the European Miocene G. euro- paeus on the other: 1. The cones are longer and narrower, more pyriform. 2. The edges of the cone-scales are less crenulate, almost smooth. A reexamination of the better preserved types of the Florissant species called Sabina linguaefolia (Lesquereux) Cockerell shows that the leaves are decussate or opposite in alternate pairs, a character that renders untenable their assignment to Glyptostrobus as proposed by Brooks. Lesquereux,® how- ever, has recorded G. ungerz from the Florissant lake beds and is emphatic in saying that ‘“‘the cones of the species . . . are not those of a Sequoza but of a Glyptostrobus.’’ The drawing (Lesquereux’s Fig. 4), showing a branch with three cones, having imbricated cone scales, would seem to confirm his identification. At the close of his discussion Lesquereux says: ‘‘Very com- mon at Florissant. The specimens figured are mostly those of the Princeton Museum.” The specimen from which Fig. 4 was drawn does not, upon in- quiry, seem to be in the collections either of the Princeton Museum or those of the American Museum of Natural History, where some of Lesquereux’s types were deposited; and without examination of this specimen I am un- willing to affirm or deny that it is a species of Glyptostrobus. The possibility that the artist may have stretched a point, as artists sometimes do in re- sponse to wishful thinking (abetted perhaps by their employers), must be ‘Heer, Oswatd. Flora tertiaria Helvetiae 1: 51, pl. 19; pl. 20, fig. 1, 1855. 6 Ibid., 1: 52, pl. 18; pl. 21, fig. 1. 5 LESQUEREUX, LEo. The Cretaceous and Tertiary floras. U.S. Geol. Survey Terr. Rept. 8: 187, pl. 22, figs. 1-6a, 1883. Sept. 15, 1936 BROWN: GLYPTOSTROBUS B00 considered, for only the uppermost cone purports to have imbricated scales; the other cones resemble those of Sequoia affinis Lesquereux.’ The state- ment, ‘‘Very common at Florissant,” is not confirmed by the large Florissant collection of the U. 8S. National Museum or by that of the Princeton Museum, which do not contain a single such cone with imbricated scales; but do contain numerous examples of Sequoza affinis, whose cones resemble in size and shape those of the alleged Glyptostrobus ungeri of Lesquereux’s Fig. 4. I, therefore, shall await the rediscovery of the type specimen of Fig. 4, or the finding of specimens equivalent to it, before passing final judgment. Theoretically, Glyptostrobus would probably not be out of place in the Florissant flora. Associated with G. oregonensis in the same beds is a chinquapin, Cas- tanopsis convexa, whose living relatives are found in western California and Oregon, particularly in the humid coast valleys; and in southeastern Asia. At elevations of 5,000 feet or more on the west slopes of the Sierras the species become shrubby. Does the fact that species of Castanopsis and a species of Glyptostrobus now survive at or near sea-level in southeastern Asia mean that eastern Oregon, where the fossils were found, was, at the time the plants were entombed, at or near sea-level? Or were those species adapted to higher elevations and a more rigorous climate? The statement that ‘“‘the plant-bearing beds at Beulah are younger than the Payette, and, although the evidence is not all that could be wished, are referred to the upper Miocene (corresponding with the Mascall beds of Merriam, in John Day Valley, Oregon)’’’ will have to be taken for what it is worth until it is learned definitely by careful mapping and paleontologic comparisons what the terms Payette and Mascall now include and to what they should be limited. My recent tentative opinion? was to the effect that a part, at least, of the Payette is lower Miocene and that some localities in the Mascall are upper Miocene. To this should be added that other parts of what is now called Payette may be middle or upper Miocene and that some localities now called Mascall may be middle Miocene. Occurrence.—Miocene beds near Beulah, and on Sucker Creek, Malheur County, Oregon. Figured specimens in U. S. National Museum. According to present information, the foregoing species, Glypto- strobus dakotensis from the Eocene, and G. oregonensis from the Miocene, appear to be the only species that can be clearly differenti- ated as Glyptostrobus among the plant fossils of the western hem- isphere. Apparently the genus became extinct in North America sometime late in the Tertiary. 7 LesQquEREUX, Leo. The Tertiary flora. U.S. Geol. Survey Terr. Rept. 7: 75, pl. 7, figs. 8-5; pl. 65, figs. 1-4, 1878. : "RUSSELL, I. C. Notes on the geology of southwestern Idaho and northwestern Ore- gon, U.S; Geol. Survey Bull. 217: 63. 1903. * Brown, Rotanp W. Miocene leaves, fruits, and seeds from Idaho, Oregon, and Washington. Jour. Paleont. 9: 586. 1935. 308 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 26, No. 9 BOTANY.—WNew plants mainly from western South America—V.! ELLswortH P. Kiuuip, U. 8. National Museum. | The present paper contains descriptions of three new species, one from the upper Amazon bacin, collected by the National Geographic Society’s expedition to the headwaters of the Orinoco River, and two from the historic Mutis Herbarium? at the Jardin Botanico, Madrid, which has recently become available for study. In addition, four new combinations of names are made, and a new name is substituted for an invalid one. Piratinera mollis Killip, sp. nov. Arbor, ramis demum glabris; folia oblonga, apice acuminata, supra sub- scaberula, in costa sparse puberula, subtus dense et molliter pilosula, pilis divaricatis, nervis lateralibus 8—10-jugis; pedunculi axillares, puberuli; re- ceptaculum depresso-hemisphaericum, bracteis orbiculatis vel subreniformi- bus, peltatis, ciliolatis; flores 9 2, stigmata 2. Tree; young branches very slender, brown, sparingly or densely pilosulous, the older glabrous; petioles 1 to 2mm long, densely pilosulous; leaves oblong, 3 to 7 cm long, 1.5 to 38 cm wide, sharply and subabruptly acuminate at apex, rounded-cuneate and unequal at base, coriaceous or subcoriaceous, above dark green and lustrous, slightly scaberulous, sparingly puberulous on midnerve, beneath pale, densely and softly pilosulous throughout, with spreading, curved or straight hairs, the principal lateral nerves 8 to 10 pairs, united near margin, prominulous above, slightly elevated beneath; peduncles axillary, solitary, 2 to 3 mm long, erect or recurved, puberulent; receptacle depressed-hemispheric, 4 to 5 mm in diameter, covered through- out with orbicular or subreniform, peltate, ciliolate, glabrescent or sparingly puberulent bracts 0.5 to 1 mm in diameter; staminate flowers not seen; pistil- late flowers 2, the stigmas 2, about 1.2 mm long, elevated above the scales. Type in the U.S. National Herbarium, no. 1,563,014, collected in Colom- bia, between 1760 and 1808, by José Celestino Mutis (no. 365). Duplicate at Madrid. This species is also represented by Mutis 362 (U. S. N. H. and Madrid). In Blake’s key to the species of this genus* the proposed new species would come nearest P. acutifolia because of the spreading hairs on the under sur- face of the leaves and the long sharp leaf tips. It differs from that plant in having much smaller leaves with fewer lateral nerves, a much smaller re- ceptacle, and larger bracts. Inga caudata Killip, sp. nov. Ramuli et folia juvenilia rufo-hirtella, demum glabra; petiolus et rachis anguste alata, glandulis sessilibus; foliola 3-juga, oblanceolata, caudato- acuminata, subauriculata, membranacea; flores brevispicati; calyx anguste 1 Published by permission of the Secretary of the Smithsonian Institution. For preceding parts see this Journal 16: 565-573. 1926; 19: 191-195. 1929; 21: 347-353. 1931; 24: 42-52. 1934. Received April 30, 1936. 2 See Killip: A scientific resurrection: the Mutis Herbarium at Madrid. Bull. Pan Am. Union, March 1933, pp. 162-171. 3 This JOURNAL 12; 395. 1922. SEPT. 15, 1936 KILLIP: NEW PLANTS 309 tubulosus, glaber, lobis late ovatis; corolla tenuis, adpresso-flavo-hirsuta ; tubus staminalis exsertus. Branchlets and foliage sparingly rufo-hirtellous when very young, soon glabrous; petiole and leaf-rachis narrowly winged, the wings up to 2 mm wide, the glands sessile, concave, 1 to 1.5 mm wide; leaflets 3 pairs, oblanceo- late (uppermost) or ovate-oblong (lower), up to 15 cm long and 4 cm wide, caudate-acuminate, gradually narrowed to a subauricular base, short- petioluled, membranous, the nerves rather prominent beneath; peduncles and rachis of inflorescence about 6 cm long, slender, glabrous or very spar- ingly pilosulous, the flowers short-spicate; calyx narrowly tubular, about 1 em long and 2 mm in diameter, glabrous, striate, the lobes broadly ovate, obtuse; corolla slender, 2.5 cm long, 1.5 mm wide, appressed-flavo-hirsute, the lobes ovate-lanceolate, 3 mm long, acute; stamen tube exserted about 8 mm, the free filaments 2 cm long. Type in the U. 8. National Herbarium, no. 1,517,914, collected on the Rio Maturaca, below Salto de Hua, State of Amazonas, Brazil, December 10-12, 1930, by E. G. Holt and E. R. Blake (no. 531). This species is related to I. longiflora Spruce and J. micradenia Spruce. From both it differs in having caudate leaflets; the uppermost leaflets are oblanceolate, not oblong or ovate-elliptic as in its two relatives. The flowers are smaller and slenderer than in I. longiflora; and the stamen tube is long- exserted, whereas in J. micradenia it is included. Parosela carthaginensis (Jacq.) Killip, comb. nov. Psoralea carthaginensis Jacq. Enum. Pl. Carib. 27. 1762; Stirp. Amer. 206. 1763. Psoralea enneaphylla L. Sp. Pl. ed. 3, 1076. 1764. Psoralea emphysodes Jacq. Coll. 4: 144. 1790. Dalea enneaphylla Willd. Sp. Pl. 1338. 1800. Psoralea emphysodes Rydb. N. Amer. Fl. 24: 113. 1920. Jacquin’s first description of Psoralea carthaginensis, in the Enumeratio is brief, ‘‘Psoralea foliis pinnatis; spicis axillaribus. Pluk. Phyt. t. 166. f. 2.” A much amplified description was published the following year (1763) by Jacquin, Plukenet’s description and figure of ‘‘Colutea enneaphyllos... ”’ being cited. Under Psoralea enneaphylla L. (1764) Linnaeus quotes the first Jacquin description and cites the Plukenet reference. So far as can be de- termined Plukenet’s figure represents a plant identical with the one common in northern South America, so that Psoralea enneaphylla is a mere substitu- tion for Psoralea carthaginensis. Specimens recently collected by Heriberto (no. 283) from the type local- ity, Cartegena, by Killip and Smith (no. 14448) from the vicinity of the neighboring village Turbaco, and by Pennell (no. 4060), from the same de- partment, agree perfectly with Jacquin’s description, and I can see no reason for longer postponing the transferal of Jacquin’s name. The typical form of P. carthaginensis has essentially glabrous branches and leaves, the leaves having 4 to 6 pairs of leaflets. Other specimens from Colombia and Venezuela have up to 9 pairs of leaflets, which, with the 360 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 26, NO. 9 branches, are more or less pilosulous. These, I believe, are mere variants of P. carthaginensis, and I am confident that some, if not all, of the following will eventually have to be reduced to synonyms of this species: Psoralea phymatodes Jacq., Dalea phymatodes Willd., Dalea vulneraria barbata Oerst., Parosela barbata Rydb., Dalea domingensis DC. and Parosela domingensis Millsp. I am purposely transferring this species to Parosela rather than to Dalea in the hope that attention will be called to the unfortunate inclusion of Dalea in the list of Proposed Conserved Generic Names. The genus is wholly a New World one, and the choice of a name is primarily a matter of concern to American botanists. Macbride has well stated’ the case for Parosela. It is the name adopted by Rydberg in the North American Flora (178 species), by Standley in the Trees and Shrubs of Mexico (106 species), and by Mac- bride in a revision of the perennial South American Paroselas® (33 species). If Dalea is conserved, it will be necessary to make new combinations for 80 to 90 species, assuming that all are valid. If the species in the small genera segregated from Parosela by Rydberg are placed in Dalea, 10 will require new combinations; if in Parosela, only four. It is sincerely hoped that the Committee on Nomenclature, which is scrutinizing the proposed conserved names, will reject Dalea. Tephrosia carpinteri (Rydb.) Killip, comb. nov. Cracca carpintert Rydb. N. Amer. Fl. 24: 172. 1928. Tephrosia gracillima (Robinson) Killip, comb, nov. Cracca angustissima Vail, Bull. Torrey Club 22: 32. 1895. Not Kuntze (1891). Tephrosia ambigua gracillama Robinson, Bot. Gaz. 28: 201. 1899. Cracca gracillima Heller, Cat. N. Amer. Pl. ed. 2, 7. 1900. Derris pterocarpus (DC.) Killip, comb. nov. Deguelia scandens Aubl. Pl. Guian. Frang. 2: 750. pl. 300. 1775. Lonchocarpus? pterocarpus DC. Prodr. 2: 260. 1825. Derris guianensis Benth. Journ. Proc. Linn. Soc. Bot. 4: Suppl.: 106. 1860. Derris scandens Pittier, Contr. U.S. Nat. Herb. 20: 41. 1917. Not Derris scandens Benth. (1860), an Asiatic species. Derris being a conserved name, the above change in nomenclature for the American plant, sometimes known as Derris guianensis or Dequelia scandens, is necessary. In monographing Dalbergieae, Bentham proposed the name Derris guianensis for this American species, citing Lonchocarpus pterocarpus DC. as a synonym. Although De Candolle’s description is rather brief, there can be little doubt but that it applies to this plant, and his specific name is the earliest one available. This is one of the important fish poisons in northern South America. 4 Nomina generica conservanda proposita. Int. Rules of Bot. Nomencl. p. 135. 1935. 5 Field Mus. Bot. 4: 100. 1927 6 Field Mus. Bot. 4: 99-113. 1927. SHpt. 15, 1936 GUNTER: GASTROPODS 361 Gouania podocephala Killip. Gouania ulmifolia Tr. & Planch. Ann. Sci. Nat. V. Bot. 16: 382. 1872. Not Hook. & Arn., 1833. Voyria macrantha Killip, sp. nov. Caulis crassus, prope apicem ramis 1—2 brevibus; squamae ovatae, apice rotundatae, per partem tertiam connatae; calycis tubus cylindrico-campanu- latus, lobis 5, late ovatis, rotundatis; corolla rubro-purpurea, extus ad basin glabra, ad apicem puberulenta, intus pulverulenta, limbo rotato, 4 to 5 cm lato, ad faucem lobato, lobis 5, obovatis vel rhombeo-obovatis, obtusis; antherae ovatae; capsula ovoidea. Stem stout, 6 to 7 cm high, about 2 mm thick, erect, once-branched near apex, glabrous; scales opposite, 6 to 12 mm apart, ovate, 3 to 6 mm long, rounded, connate in lower third, the sinus acute; peduncles up to 13 mm long, stout, gradually widening to calyx; calyx tube cylindric-campanulate, about 1.5 cm long, 5 to 6 mm wide at throat, the lobes 5, broadly ovate, 1.5 to 2 mm long, 3 to 4 mm wide, rounded, minutely ciliolate, glabrous other- wise; corolla bright red-purple, the tube cylindric, 3.5 to 4.5 cm long, gla- brous at base without, puberulent on upper half without, pulverulent within; the limb rotate, 4 to 5 cm wide, lobed to throat, the lobes 5, obovate or rhombic-obovate, 2 to 2.4 em long, 7 to 11 mm wide, obtuse; filaments in- serted about 5 mm below throat of tube, 0.5 to 1 mm long; anthers ovate, 1 to 1.5 mm long; style slender, about 3.7 cm long; stigma capitate, papillose on upper surface, smooth beneath; ovary sessile; capsule ovoid, 1 cm long, 6 mm in diameter. Type in the herbarium of the Jardin Botanico, Madrid, collected in Co- lombia, between 1760 and 1808, by José Celestino Mutis (no. 3054). Repre- sented also by Mutis 2566. Known also from the following Colombian collec- tions: Dept. Sur de Santander: ‘“‘ Camp Carare II,’’ Haught 1621 (U.S.N.H.). “Camp Aguila,’’ Carare Valley, Haught 1871 (U.S.N.H.). The flowers of V. macrantha are much larger than in other species of Voyrza or in the closely related genus Leiphaimos. Apparently the species comes nearest V. caerulea Aubl. but in addition to having larger flowers than V. caerulea, the scales and the calyx lobes are rounded, not acute. ZOOLOGY.—Radular movement in gastropods... GORDON GUNTER, Bureau of Fisheries. (Communicated by Pau 8. GALTSOFF.) Cuvier? in 1817 described the radula in gastropods as a passive in- strument moved secondarily by its supporting cartilages. Huxley’ (1853) disagreed with him and maintained that the radula executes movements independent of its cartilages and moves over them ‘‘chain- saw-like.”’ He believed that this type of movement prevailed “‘through- out the Cephalopoda and Gasteropoda.’’ Most of his argument was 1 Printed by permission of the Commissioner of Fisheries. Received February 20, 1936. 2 Memoire pour servir a l’histoire et a l’anatomie des mollusques. Paris. 1817. 3 Phil. Trans. Roy. Soc. London 143: 29-65. 1858. 362 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 26, No. 9 based on anatomical structure, but he stated that the apparatus could ‘‘be seen working in Firoloides and Atlanta.”’ Most writers in describing the odontophoral apparatus in gastro- pods have not mentioned the issue, but there have been adherents of both theories. Lacaze-Duthiers* (1856), Geddes: (1879), Amaudrut® (1898) and Simroth’ (1901) rejected Huxley’s hypothesis for that of Cuvier, either by direct statement or by implication in their descrip- tions. Geddes studied under Huxley and took up the subject at the latter’s instigation. He rejected the chainsaw idea and concurred with Cuvier. His objection to Huxley’s theory was based principally on the close attachment of the radula to the radular sac by the radular mem- brane and his belief that the radula could not slide over the acute angle formed by the apex of the cartilages. Wegman® (1884) in describing Haliotis stated that the radular membrane slides over the cartilages, which implies that the radula does also. Herrick® (1906) in describing the anatomy of the odonto- phore of Busycon canaliculatus sided with Huxley, and gave a very clear exposition of his conception of its mode of action. Dakin!® (1912) working on Buccinum also followed Huxley. He stated that the radula could be made to slide over the cartilages in narcotized animals by pressing the proboscis between the thumb and forefinger. Both he and Herrick stated as further evidence for their belief, that radular movement could be felt when animals were induced to rasp the finger tip. The writer has repeated this experiment with Melongena corona and Thais floridana, but has found it impossible to tell by such means whether the radula slides or is passively borne by the cartilages. Because of the internal position of the radula in the radular sac within the proboscis, its small size, and the fact that when in use the whole proboscis is often covered by the foot and its open end is placed on the food, observation of the living animal in the natural state seems to be almost impossible. These obstacles are probably the rea- sons why such observations, which would have settled the question, have been few and relatively incomplete. Some workers have stated that they have observed living animals, but most of their conclusions appear to be drawn from the anatomy, and the writer has not found 4 Ann. Sci. Nat. 6: 225-281. 1856. 5 Trans. Zool. Soc. London 10: 485-491. 1879. § Ann. Sci. Nat. 7: 1-291. 1898. 7 Bronn’s Klassen und Ordnungen des Thier-Reichs. Band 3. Leipzig. 1901. § Arch. Zool. Exper. et Gen. 2: 289-378. 1884. * Amer. Nat. 40: 707—737. 1906. 10 Liverpool Marine Biology Committee Memoirs 20: 1-115. 1912. Sept. 15, 1936 GUNTER: GASTROPODS 363 a complete description of radular movement based on direct observa- tions of the odontophoral mechanism at work in the living animal. Fasciolaria gigantea of the Florida coast is an ideal animal for study and observations were made on it at the Indian Pass Laboratory of the United States Bureau of Fisheries. It is one of the largest gastro- pods in the world and has a radula 2 mm. wide. When stimulated by a bit of oyster meat it gives the usual food reaction of carnivorous gastropods which culminates in the extrusion of the proboscis. This has been described in detail by Copeland"! for Nassa obsoleta and Busycon canaliculatum. If the food is withdrawn the odontophore may be seen through the proboscis opening, with the naked eye, to con- tinue working in the following manner. The cartilages bearing the radula move forward and upward with a licking motion. At the same time the radula moves upward and over the cartilages like a chainsaw or belt over a pulley, with a motion so rapid that it gives the illusion of rapid rotation. When the odontophore reaches the end of the for- ward movement it begins to move downward and backward. At the same moment the radula reverses its direction, almost too speedily for detection, and slides downward and under the cartilages. As the forward movement of the cartilages is started again the radula re- peats the movement first described. The whole process proceeds at an even rate so that the radula has the appearance of a rapidly spinning wheel being carried back and forth on a frame. As Herrick (op. cit. p. 721) has pointed out the teeth are folded together as they pass back and forth under the cartilages and rasping can only be done during the upward stroke. The mouth is situated so that food can be drawn into it only by this movement and not by the downward one. In Thais the food may be seen to progress down the proboscis in peristaltic waves, which are apparently timed and initiated by the piston-like strokes of the odontophore. The objections of Geddes and others to the theory that the radula has motion independent of that of the cartilages, were based on anatomical studies and cannot hold in the face of observations to the contrary on living animals. He seemed to disregard the significant fact that the radula is a long, ribbon-like, jointed apparatus, borne on a very flexible membrane; the whole being eminently fitted for the band-over-pulley type of function. The plates bearing the teeth are loosely joined together so that the whole radula can bend back upon itself. This jointed structure allows the radula to move over the apex 1 Jl, Exp. Zool. 25: 177-227. 1918. 364 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 26, NO. 9 of the cartilages as described for Fasciolaria. The radula is not too closely attached to the radular sac, as Geddes stated, or too firmly attached to the cartilages for independent movement, as shown by the fact that in Thais floridana and Thais sp. the radula of an amputated proboscis can be made to move over the cartilaginous support while the latter is perfectly stationary. These animals are very similar in structure to Buccinum, one species upon which Geddes worked and arrived at the opposite conclusion. Fig. 1—Diagrammatic side view of radula. These studies apparently prove the hypothesis of Huxley and his adherents concerning the band-over-pulley movement of the radula. Yet Huxley (op. cit. p. 31) stated that ‘‘the buccal cartilages take no part in the movement of the tongue-plate.’’ As shown in the above observations this is incorrect, at least for the forms described here. Von Siebold (quoted by Huxley, loc. cit.) had previously recognized that the cartilages play a part for he stated that by protrusion and retraction this organ is used by the Cephalophora as an ingestive ap- paratus. Oswald” (1893) seems to have first realized that both move- ments were concerned, although he said the band-over-pulley motion was very limited. The illusion that a radula in operation is a small wheel is very sig-. nificant, for it can be shown by reference to the description above and the accompanying diagram that the radula is functionally a small, rapidly rotating wheel or drum covered on the outside with small spikes. As the cartilaginous frame moves forward and upward its trajec- tory is that of an arc as shown by arrow number 1 of the diagram. At 22 Jen. Zeitschr. f. Naturw. 28: 119-162. 1893. Sept. 15, 1936 WILSON: COPEPODS 365 the same time the radula is pulled up and over the apex of the car- tilages by the radular muscles in the direction shown by arrow 2. This imposes a second speed on the first. The teeth sliding over the apex are held at right angles to the ribbon and pulled in the arc shown by arrow 3. From the simple law that the speed of a turning wheel is greatest at its circumference, it is seen that the barbs of the teeth at this point are moving at a greater speed than the ribbon itself or the teeth at any other point of the radula. This superimposes a third speed upon the other two, and it is easily seen how the illusion of a rotating wheel is created. This simple mechanical adaption moves the functional part of the radula at a rapid, but undetermined speed, and probably accounts to a large extent for the ability of many gastro- pods, such as Thais, Murex, Eupleura, Urosalpinx and Buccinum to bore through the hard shell of other mollusks. It is also possible that the radula is assisted by an acid secretion which softens the shell. The distinctive structure of the odontophoral apparatus and its similarity in prosobranchiate gastropods leads the writer to believe that this mode of radular movement is typical for most prosobranchs and probably many other gastropods. Observations on Thais flori- dana, Thais sp., and Melongena corona substantiate this hypothesis. The writer is indebted to Dr. A. 8. Pearse for the specimens of Fasciolaria. 7 ZOOLOGY.—Copepods from the far north collected by Capt. R. A. Bartlett.1 CHARLES B. Wiuson, Westfield, Massachusetts. (Com- municated by Mary J. RaTHBUN.) For several years Capt. R. A. Bartlett has been gathering plankton from the coasts of Labrador, Canada and Greenland, the last cruise taking place during the summer of 1935. The samples thus accumu- lated have been submitted for examination by the National Museum and the copepods found in them are here listed. The localities from which the plankton was obtained may be conveniently divided into four groups according to geographical location. The first group ex- tended along the entire coast of Labrador from 52 to 60 degrees North Latitude, and included several fishing and whaling grounds. The second group began at the mouth of Hudson Strait, just north of Labrador and extended northwest up the strait into the northern end of Hudson Bay, and thence into Fox Channel, a northern arm of the bay reaching into the Arctic Zone south of Baffin Island. In this 1 Received March 18, 1936. 366 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 26, NO. 9 group a surface haul taken in the Bay of Gods Mercy on the south shore of Southampton Island, between Hudson Bay and Fox Chan- nel, was remarkable for the number of southern species found in it. Other surface hauls were made while at anchor off the mouth of Fury and Hecla Strait during a furious gale of wind and snow. In these hauls the number of harpactids, usually found only close to the bottom, suggests that the violent agitation of the water during such a gale thoroughly mixes plankton that otherwise might be arranged more or less in layers. The third group embraced Davis Strait, Baffin Bay, the west coast of Greenland, and the straits which separate ‘Greenland from Ellesmere Island from 76 to 78 degrees North Lati- tude. The fourth group ran along the east coast of Greenland north to the polar ice cap, with one or two stations between Greenland and Spitzbergen. During the early cruises hauls were made with a dredge at or near the bottom and these seldom contained development stages. During the summer of 1935 a fine-meshed net was used, nearly always at the surface, and all of these hauls contained abundant development stages of copepods, crabs and Euphausiids. An effort was made to take these hauls at regular intervals during the entire twenty-four hours. As a result the plankton contains not only species regularly frequenting the surface during the daytime but also the diurnal migrants which come up from below during the night. All the localities here recorded are in regions which have been hitherto examined but little for plankton. The lists of copepods here given furnish an interesting supplement to Willey’s Hudson Bay copepod plankton (Contrib. Canadian Biol. and Fisheries, n.s., vol. 6, no. 25), since they reach farther north into the Arctic Circle. For this reason the lists are well worth publishing though they do not contain any new species. Again, our knowledge of the geographical distribution of all but a _very few species is extremely limited, and new localities are here re- corded for nearly every species listed. Some found hitherto only on the coast of Norway or around Franz-Josef Land or Spitzbergen are here revealed far to the west and may well become circumpolar eventually. Willey has called attention to the presence of Arctic copepods in Passamaquoddy Bay (Amer. Acad. Arts & Sci., vol. 56, no. 5). In these lists we find the exact reverse, the presence of temperate species far within the Arctic Circle. Evidently we cannot as yet fix with com- plete accuracy the north and south limits of very many species. Sept. 15, 1936 WILSON: COPEPODS 367 The results of this study are presented as a simple faunal list in which the species are arranged alphabetically in each locality. Follow- ing this list are notes on the several groups of copepods represented in the collection, Calanoida, Harpacticoida, Cyclopoida, and Noto- delphyoida. STATIONS ALONG THE COAST OF LABRADOR Hawkes Harsor. 53° North. Shallow water within the harbor. Dredge tow at the bottom. August 26, 1929. Calanus finmarchicus, Dactylopusia signata, D. vulgaris, Ectinosoma neglectum, Metridia longa, Orthona similis, Oithonina nana, Paracalanus parvus, Pseudocalanus elongatus, Tisbe furcata. Hawkes Isuanp. 53° North. Whaling ground 30 miles off shore. Dredge tow at bottom, net tow at surface. August 31, 1929. Calanus finmarchicus, C. hyperboreus, Harpacticus chelifer, Oithona similis, Oithonina nana, Para- calanus parvus, Pseudocalanus elongatus, Temora longicornis, T. stylifera. Development stages of copepods and crabs at surface. Rep Isuanp. 53° North. Shallow water near shore. Net tow at the surface. August 26, 1929. Calanus finmarchicus, C. hyperboreus, Oithona similis, Pseudocalanus elongatus. Crab zoéas and megalops very abundant. Brn’s Cove, Care Aruuik. 55° North. Shallow water within the cove. Surface tows during the day and at night. September 17, 1929, August 16, 17, 1985. Acartia clausii, A. longiremis, Calanus finmarchicus, C. hyper- boreus, Farranula carinata, F. rostrata, Laophonte elongata, Oithona similis, O. spinrostris, Oncaea borealis, O. venusta, Paracalanus parvus, Pseudocalanus elongatus, Temora longicornis, T. stylifera. Development stages, especially Calanus, Oithona, Temora. KaiaG-La-PAIT Bay. 56° North. Shallow water within the bay. Surface tows during the day. September, 1929. Calanus finmarchicus, C. hyper- boreus, Orthona similis, Paracalanaus parvus, Pseudocalanus elongatus, Temora longicornis, T. stylifera. Muerorp Bay. 58° North. Shallow water within the bay. Dredge tows at the bottom. September 4, 5, 1929. Calanus finmarchicus, C. hyperboreus, Paracalanus parvus, Pseudocalanus elongatus. Movurts or Hupson Strait. 59° North. Deep water off shore. Dredge tow at the bottom. July 25, 19383. Cyclopina schneidert, Dactylopusia vulgaris, Ectinosoma neglectum, Parathalestris jacksoni, Pseudobradya minor, Tisbe furcata. CANADIAN STATIONS Bay or Gops Mrrcy, SouTHAMPTON IsLanp. 64° North. Shallow water in the bay. Both day and night tows while at anchor in the bay. August 5, 1933. - Acartia clausi, A. longiremis, Bradypontius magniceps, Calanus finmarchi- cus, C. tonsus, Clausocalanus arcuicornis, Corycaeus anglicus, Dactylopusta tisboides, D. vulgaris, Farranula carinata, F. rostrata, Harpacticus uniremis, Laophonte elongata, L. perplexra, Otthona similis, O. spinirostris, Oncaea borealis, O. venusta, Paracalanus parvus, Pseudobrada minor, Pseudocalanus elongatus, Pseudocyclops obtusatus, Robertsonia tenuis, Tisbe gracilis, T. fur- cata, Undinula darwini, Zaus abbreviatus, Z. spinatus, Zosime typica. Devel- opment stages. Mout oF Frozen Strait. 66° North. Surface tows in the strait and in Fox Channel. August 14-16, 1933. Strait half frozen across. Acartia clausit, 368 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 26, NO. 9 A. longiremis, Ameira longipes, A. tau, Calanus finmarchicus, Cyclopina schneidert, Danielssenia typica, Dactylopusia signata, EKctinosoma curticorne, E. neglectum, Laophonte elongata, Mesochra pygmaea, Orthona similis, On- caea borealis, Paracalanus parvus, Pseudobradya similis, Pseudocalanus elongatus, Pseudothalestris pygmaea, Robertsonia tenuis, Rhynchothalestris helgolandica, Temora longicornis, Thalestris gibba, Tisbe furcata, T. minor, Zaus abbreviatus, Z. spinatus. Abundant development stages of crabs and copepods. , Fox CHANNEL. 66° to 67° North off Cape Penryhn, Melville Peninsula. Dredge tows at bottom and net tows at surface. August 27-30, 1933. Acar- ta clausit, Alteutha depressa, Arctopontius expansus, Arctotrogus orbicularis, Ascomyzon intermedium, Botryllophilus inaequipes, Bradypontius caudatus, B. groenlandicus, B. magniceps, Calanus finmarchicus, C. hyperboreus, C. tonsus, Clausocalanus arcuicornis, Cyclopina schneidert, Dactylopusia signata, Dermatomyzon nigripes, Doropygella thorellii, Dyspontius striatus, Hemicy- clops purpureus, Laophonte elongata, L. horrida, Lichomolgus agilis, Metridia longa, Myzopontius pungens, Orthona similis, Parartotrogus arcticus, Para- thalestris jacksoni, Pseudobradya minor, Pseudocalanus elongatus, Pseudo- molgus groenlandicus, P. leptostylts, Robertsonia tenuis, Thalestris gzibba, Tisbe furcata. Fury anD Hecua Strait. 70° North. Anchored off Esquimaux village. Surface tows, midnight to 8 a.m. in a furious gale of wind and snow. Septem- ber 7, 19383. Acartia clausit, A. longiremis, Amallophora typica, Ametra longipes, A. tau, Amphiascus minutus, Bradypontius caudatus, Calanus finmarchicus, C. hyperboreus, C. tonsus, Clausocalanus arcuicornis, Cletodes tenutpes, Cyclopina elegans, Dactylopusia signata, Ectinosoma curticorne, E. neglectum, Farranula rostrata, Laophonte perplexa, L. similis, Mesochra pyg- maea, Microsetella norvegica, Orthona similis, Oncaea borealis, Paracalanus parvus, Parathalestris jacksoni, Pseudobradya minor, P. similis, Pseudoca- lanus elongatus, Pseudomolgus groenlandicus, Pseudothalestris nobilis, P. pygmaea, Robertsonia tenuis, Thalestris gibba, Tisbe finmarchica, T. furcata, Zaus abbreviatus, Z. spinatus. Development stages of crabs, copepods, and Euphausiids. DAVIS STRAIT, BAFFIN BAY, AND WEST COAST OF GREENLAND Davis Strait. 64° 47’ North. Mouth of the strait. Surface tow taken amid floating ice, 3:15 p.m. September 12, 1935. Acartia clawsi1, one male; Calanus finmarchicus, copepodid stages, few adults; Ozthona similis, all adults, the females with ovisacs; Paracalanus parvus, two adults. Davis Strait. 69° 18’ North. Surface tow, 6 a.M., raining. September 4, 1935. Calanus finmarchicus, adults and copepodid stages; Oithona similis, adults, the females without ovisacs; Paracalanus parvus, adults and copepo- did stages. Davis Strait. 70° North. Surface tow, 10 p.m., in floating ice. September 8, 1935. Calanus hyperboreus, adults, the only tow containing them in abundance; Halithalestris cront, two adults; Oithona similis, adults, a few ae with ovisacs; Paracalanus parvus, adults; Pseudocalanus elongatus, adults. Disko IsLAND, GREENLAND Coast. 69° 20’ North. Surface tow taken at 10 a.m. July 17, 19385. Acartia clausii, A. longiremis, Calanus finmarchicus, Halithalestris cront, Paracalanus parvus, Pseudocalanus elongatus. CoxsoureG ISLAND, OF ENTRANCE TO JONES SOUND, SOUTH OF ELLESMERE Lanp. 75° 40’ North. Surface tow 2:30 a.m. September 4, 1935. Calanus Sept. 15, 1936 WILSON: COPEPODS 369 jfinmarchicus, Halithalestris croni, Oithona similis, Paracalanus parvus, Pseu- docalanus elongatus. A few nauplii. CaPE YORK AND MELVILLE Bay, GREENLAND Coast. 75° North. Surface and vertical hauls during the daytime. July 18-21, 1926; August 4-6, 1932; July 22,1935. Calanus finmarchicus, C. hyperboreus, Ectinosoma curticorne, Metridia longa, M. lucens, Oithona brevicornis, Paracalanus parvus, Pseudo- calanus elongatus, Temora longicornis. HERBERT ISLAND, INGLEFIELD Bay. 77° North. Surface and vertical hauls during the daytime. July 25, 1926; July 25, 1985. Augaptilus glaczalrs, Calunus finmarchicus, C. hyperboreus, Ectinosoma curticorne, E. neglectum, Metridia longa, Oithona similis, Parathalestris jgacksoni, Pseudocalanus elongatus, Tegastes falcatus, Thalestris gibba, Tisbe furcata, Zaus aureli, Z. goodsirt, Z. spinatus. Development stages of crabs and copepods. MURCHISON SOUND AND WHALE Sounp.. 77° North. Dredge tow at bottom and net tow at surface. August 17-21, 1926; July 25, 1935. Calanus fin- marchicus, Dermatomyzon nigripes, Diosaccus tenuicornis, Harpacticus chelafer, Oithona similis, Parathalestris jacksoni, Paracalanus parvus, Pseudo- calanus elongatus, Tisbe furcata. NORTHUMBERLAND ISLAND. 78° North. Dredge tow at bottom. August 16-17, 1926; July 30, 1935. Altewtha depressa, Amphiascus minutus, A. nasutus, Bradypontius magniceps, Calanus finmarchicus, C. hyperboreus, C. tonsus, Cyclopina schneidert, Dactylopusia signata, D. vulgaris, Daniels- senta fusiformis, D. typica, Ectinosoma curticorne, E. neglectum, Halithalestris cront, Harpacticus superflecus, H. uniremis, Metridia longa, M. lucens, Para- thalestris gacksont, Pseudobradya minor, Pseudocalanus elongatus, Stenhelia aemula, Tachidius brevicornis, Tegastes falcatus, T. nanus, Tisbe furcata, Zaus aurelit, Z. goodsiri, Z. spinatus. SmitH SounD, ELLESMERE LAND. 78°10’ North. Surface haul at 6:20 P.M. July 31, 1935. Calanus finmarchicus, Orthona similis, Paracalnus parvus, Pseudocalanus elongatus. Development stages, especially of Ozthona. EASTERN COAST OF GREENLAND ANGMAGSSALIK, 66° North. Dredge tow at the bottom. August 30, 1930; August 28, 19381. Calanus finmarchicus, Herpyllobius arcticus, Microsetella norvegica, Paracalanus parvus, Pseudocalanus elongatus, Tisbe furcata. Orr Cape Simpson. 73° North. Surface tow amid pack ice. August 13, 1930. Calanus finmarchicus, C. hyperboreus, Pseudocalanus elongatus. De- velopment stages of copepods, crabs, and Euphausiids. PrnpDuLuUM Istanp. 74° North. Dredge tow at the bottom. July 20 and 30, 1930. Augaptilus glacialis, Cyclopina schnetdert, Diosaccus tenutcornis, Ectinosoma curticorne, E. neglectum, Laophonte elongata, Mormonilla polaris, Parartotrogus arcticus, Parathalestris jacksont, Pseudobradya minor, Pseudo- molgus groenlandicus, Scolecithrix brevicornis, Temorttes brevis, Tisbe furcata, Undinella oblonga, Zaus goodsiri, Z. spinatus. BETWEEN CAPE BISMARCK AND THE NORTHWEST SIDE OF KoupEway Is- LAND. 76° North. Temperature of water 34°F. July 28, 1930. Calanus fin- marchicus, few adults, but numerous development stages, together with those of crabs and Euphausiids. NOTES ON THE CALANOIDA Acartia clausii and A. longiremis. These two species are cosmopolitan in distribution and both have been reported from the Hudson Bay plankton by Willey. The males are better differentiated than the females and both sexes were found as far north as the 70th parallel. 370 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 26, NO. 9 Augaptilus glacialis G. O. Sars. This is the only polar representative of the genus and was first obtained during the Norwegian North Polar Expedi- tion. It has since been found much farther south and thus is not exclusively arctic in its distribution. The new localities here given carry it far to the west and suggest that it may be circumpolar. Calanus finmarchicus (Gunner). Present at almost every locality visited, but nowhere in any abundance except at Inglefield Bay. It was usually found in company with one or both of the other species of the genus, and development stages, probably including all three species, were abundant in the surface tows. Calanus hyperboreus Krgyer. This large arctic form was also widely distributed but even less abundant than the preceding. The Davis Strait tow from 70° North was the only one containing more than one or two specimens. Calanus tonsus Brady. ‘This species lacks the serrated inner margins of the basipods of the fifth legs and the genital segment of the female is con- siderably swollen. This is probably the first record within the arctic circle and the number of specimens is very small. Metridia longa Lubbock. A hardy arctic species but nowhere found in any abundance, although Nordenskiold has claimed that it is able to live in immense numbers in water-drenched snow at a temperature below Zero C. Metridia lucens Boeck. Sars reported this species from within the Arctic Circle on the west coast of Norway, but added that he had never found it in other samples from the Arctic Ocean. The present records from Baffin Bay enable us to regard the species as an arctic form. Mormonilla polaris G. O. Sars. Found only at Pendulum Island on the east coast of Greenland and obtained by Sars in the Norwegian North Polar Expedition as far north as the 81st parallel of latitude. Paracalanus parvus (Claus). This is another common species very widely distributed here in the north and sometimes occurring in large numbers. It may be distinctly southern in its range, as reported by Sars, but it is evi- dently not prevented by the cold from breeding also in the Arctic Ocean. Pseudocalanus elongatus (Boeck). Very widely distributed throughout the entire area and often the most numerous single species obtained. It was somewhat more abundant in shallower water than in deeper water, and evi- dently descends to the very bottom. The new localities here added make it circumpolar in its distribution. Pseudocyclops obtusatus Brady. In spite of its generic name this is a peculiar small calanid, readily recognized by the structure of the fifth legs in both sexes. It was obtained by Sars on the Norway coast nearly as far north as the single female found in the plankton from the Bay of Gods Mercy. Scolecithrix brevicornis G. O. Sars. The specimens described by Sars were found north of the 81st parallel of Latitude and east of Franz-Josef Land. Capt. Bartlett’s specimens came from Pendulum Island on the east coast of Greenland and a little farther south. Temora longicornis G. O. Sars. Cleve has reported this species from the Atlantic Ocean as far north as the 72nd parallel of Latitude. With one exception the specimens of the present list all came from the Labrador coast considerably farther south. Temora stylifera (Dana). This is primarily a tropical species, but has been reported by Giesbrecht as far north in the Atlantic Ocean as the 60th Supt. 15, 1936 WILSON: COPEPODS vl parallel of latitude. The present specimens came from the Labrador coast considerably farther south than that parallel. Temorites brevis G. O. Sars. This is a very small species originally ob- tained during the Norwegian North Polar Expedition considerably to the east of the present localities. Here it was found only at Pendulum Island on the east coast of Greenland. Undinella oblonga G. O. Sars. This small copepod looks very similar to a cyclopid, but an examination of its appendages shows it to be a true Calanid. It has not thus far been found anywhere except in the Arctic Ocean. Undinula darwini (Lubbock). This calanid is usually found much farther to the south, but two males were present in the plankton taken in the Bay of Gods Mercy on Southampton Island. The structure of the fifth legs of these males is so peculiar as to leave no doubt of their identity. NOTES ON THE HARPACTICOIDA As already stated many of the present specimens were taken in dredge hauls at or near the bottom. Since there are more Harpactids at that depth than other kinds of copepods it would be expected that the species of this suborder would outnumber those of any of the other groups. The following lists shows that this really happened, the number of Harpactids equalling that of all the other suborders combined. Alteutha depressa Baird. This is not a true arctic species, but it is a hardy copepod and has been found by Sars about as far north on the west- ern coast of Norway. Ameira longipes Boeck. This copepod has been reported by Sars from the polar islands north of ‘‘Elsemer”’ (Ellesmere) Land, taken in the second “Fram” expedition. Thus far the species has been confined to polar seas. The present specimens came from the mouth of Frozen Strait. Amphiascus minutus (Claus). Found only near Northumberland Island in Smith Sound in company with the following species. It has been reported by T. Scott from Franz-Josef Land at about the same latitude farther east. Amphiascus nasutus (Boeck). Twice the size of the preceding species, in - company with which it was found, and so readily separated by size alone. It is distinctly arctic in distribution but comes down into the temperate zone along the west coast of Norway. Cletodes tenuipes T. Scott. This harpactid has been reported from Franz-Josef Land by Sars, but has not before been captured in American waters. In the present plankton it was confined to the single peculiar tow taken near the mouth of Fury and Hecla Strait. Dactylopusia signata Willey. Found on the Labrador coast, in Fox Channel, and also at Northumberland Island, where it was in company with the following species. Willey’s original specimens came from farther west and were captured in a net let down through the ice. Dactylopusia vulgaris G. O. Sars. Found in company with the preceding and about the same size but easily distinguishable by the structure of the fifth legs in the female. This is one of the commonest harpactids and was found by Sars along the entire Norwegian coast, even those portions which lie in the Arctic Ocean. Danielssenia fusiformis (Brady). Confined to the single locality of 372 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 26, NO. 9 Northumberland Island in shallow water. This is a temperate rather than an arctic species and is here reported from Greenland for the first time.. Danielssenia typica Boeck. Found in company with the preceding and also at the mouth of Frozen Strait, these being the first American localities. It has been reported from Nova Zembla and Franz-Josef Land farther east. Diosaccus tenuicornis (Claus). A few specimens were taken between Greenland and Ellesmere Land. Although it is not an arctic species it has been found nearly as far north on the Norwegian coast. Ectinosoma curticorne Boeck. Found in several of the present lists and has been reported from Spitzbergen by T. Scott. Its small size and very short first antennae, each with a dusky patch inside the basal segment, will serve as distinctive features. Ectinosoma neglectum G. O. Sars. Found on the Labrador coast and at several of the Canadian localities both in Capt. Bartlett’s plankton and in that obtained by the Canadian Arctic Expedition. It was also reported by Sars from the north coast of Norway and is probably circumpolar. Halithalestris croni (Krgyer). This is one of the largest harpactids and may be recognized by its size and by the long divergent caudal rami. It is a pelagic species but drifts in with the currents into various bays and straits and has been reported before from the west Greenland coast by Walker and Miers. Harpacticus chelifer (Miller). Found only on the Labrador coast and in Murchison Sound, but has been reported by Sars from Egedesminde on the west coast of Greenland and by T. Scott from Franz-Josef Land. Harpacticus superflexus Willey. Northumberland Island in the strait between Greenland and Ellesmere Land furnished the only specimens of this species. It was originally obtained by the Canadian Arctic Expedition considerably farther west and nearly as far north. Harpacticus uniremis Krgyer. Found in company with the preceding species and has been reported from Behring Sea by Poppe, from Spitz- bergen by T. Scott and from the whole Norwegian coast by Sars. It may thus be considered as an arctic form, coming down at times into the tem- perate zone. Laophonte elongata Boeck. Found in three of the four regions here listed and reported by T. Scott from Franz-Josef Land. It may be distin- guished by the structure of the fifth legs and by the long parallel caudal rami. Laophonte horrida Norman. Found only in Fox Channel in company with the preceding species but has been reported from both the eastern and western coasts of Greenland. When obtained the specimens are usually so densely covered with mud that the body spines are invisible. If the mud is washed away the spines will serve to identify the species. Laophonte perplexa T. Scott. Originally obtained by Scott from Franz- Josef Land the species was subsequently reported by Sars from Norway. Its presence in the plankton from the Bay of Gods Mercy is the first record from American waters. Mesochra pygmaea (Claus). This minute harpactid is a dwarf form whose body segments have the appearance of being telescoped together. It has been reported from Franz-Josef Land and from the polar islands north of Grinnel Land. Microsetella norvegica Boeck. Found at only two of the present sta- tions but given a wide distribution in the Arctic Ocean by Cleve and Sars. It is a very small species but often found in large numbers. Sept. 15, 1986 WILSON: COPEPODS 373 Parathalestris jacksoni (T. Scott). Found in the last three regions but not along the Labrador coast and reported by Sars from the northern coast of Norway and by T. Scott from Franz-Josef Land. Pseudobradya minor (T. & A. Scott). Like the preceding species this is well distributed everywhere except on the Labrador coast, but only one or two specimens in any locality. It was obtained by the Canadian Arctic Expedition from Bernard Harbor, Northwest Territories far to the west. Pseudobradya similis (T. Scott). Found only in the plankton from the mouth of Frozen Strait, almost up to the Arctic Circle, and originally ob- tained by Scott from Franz-Josef Land. Pseudothalestris pygmaea (T. Scott). This is another of the dwarf species whose body segments appear to be telescoped together, giving it a peculiar stunted appearance. It was found only at the mouth of Frozen Strait. Rhynchothalestris helgolandica (Claus). This was originally obtained by Claus from Helgoland, but has been reported from Greenland by Stephensen and from various localities in the Arctic Ocean. It was here con- fined to the plankton from the mouth of Frozen Strait. Robertsonia tenuis Brady. Only a few specimens of this species were found in the second group of localities, but it has been obtained by Scott from Spitzbergen and Franz-Josef Land, and from Greenland by Wesen- berg-Lund. Stenhelia aemula (T. Scott). A few specimens were found at North- umberland Island and this is the first record of the species in polar regions. Tachidius brevicornis Lilljeborg. Another species captured only at Northumberland Island in the bottom dredge. It is a widely distributed species chiefly confined to temperate localities but taken by the Canadian Arctic Expedition well within the Arctic Circle. Tegastes falcatus Norman. Also taken in the bottom dredge at North- umberland Island, it has been reported by T. Scott from Nova Zembla and Franz-Josef Land. It has also been found around the British Isles and near Ceylon and hence is not confined to arctic seas. Tegastes nanus G. QO. Sars. This was found in company with the preced- ing species and is even smaller in size. It has previously been reported only from the Norwegian coast by Sars, at about 63° North Latitude. Thalestris gibba (Krgyer). Two specimens were obtained from Fox Channel and three from Inglefield Bay; it has also been recorded by Scott from Franz-Josef Land and by Sars from the north coast of Finland. Tisbe finmarchica (G. O. Sars). Sars reported this as a true arctic species from the northern coast of Finmark and the polar islands north of Grinnell Land. In the present lists the species is confined to two specimens from the peculiar plankton taken at Fury and Hecla Strait. Tisbe gracilis (T. Scott). This species is confined to the plankton from the Bay of Gods Mercy. It was originally found on the Scottish coast, but was reported by Sars from the Finmark coast and the west coast of Norway as far north as the present locality. Tisbe minor (T. Scott). Found in the present plankton only at the mouth of Frozen Strait, but reported by Scott from Franz-Josef Land, and by Sars from the west coast of Norway as far north as Frozen Strait. Tisbe furcata (Baird). Found at a majority of the localities here listed, often in considerable numbers, and recorded as widely distributed through- out the Arctic Ocean. Zaus abbreviatus G. O. Sars. This is the smallest species of the genus, 374 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 26, NO. 9 and has been found on the west coast of Norway and north of Grinnell Land. Its appearance at two of the present localities is new to American waters. Zaus aurelii Poppe. Found at Northumberland Island, and recorded by Scott from the northern coast of Finmark. This is one of the larger species and has a very small rostrum and a very narrow urosome. Zaus goodsiri Brady. Found on both the east and west coasts of Green- land and recorded from Spitzbergen by Scott and from the polar islands north of Grinnell Land by Sars. It is half as large again as the preceding form and shows a huge angular rostrum. Zaus spinatus Goodsir. Found in many of the present localities except those along the Labrador coast, and reported to be widely distributed in the arctic zone. This is the smallest of the four species and only two-fifths the size of goodsirz, with a slightly prominent semicircular rostrum. Zosime typica Boeck. This species was confined to the plankton from the Bay of Gods Mercy, its first appearance in American waters, but it has been recorded by Scott from Franz-Josef Land and Nova Zembla. NOTES ON THE CYCLOPOIDA Some of the cyclopids here listed probably live semiparasitically upon or within the bodies of other marine animals. But having been captured while swimming freely there is nothing to connect them with any host. Some of them, however, appeared in the plankton taken by the Danish Ingolf Expedition, and in the account of that expedition were assigned by Hansen to definite hosts. Arctopontius expansus G. O. Sars. Occurring in these lists only in Fox Channel, it was obtained by the Ingolf Expedition in Davis Strait, and is probably more or less well distributed in this region. Neither in the Ingolf material nor in that examined by Sars was any host discovered. Artotrogus orbicularis Boeck. Found only in Fox channel; Boeck’s original specimen came from the south coast of Norway, where other speci- mens were later found by M. Sars. It has also been reported from the Kara Sea by Hansen, and in all three instances was taken from nudibranch mol- lusks. Ascomyzon intermedium Hansen. Found only in Fox Channel, but was also obtained by the Ingolf Expedition from Davis Strait, where it was dredged at a depth of 582 fathoms. All the specimens thus far obtained have been females with no hint of their host. Bradypontius caudatus G. O. Sars. Associated with the preceding species in Fox Channel, but not obtained by the Ingolf Expedition. Sars stated that it was the largest siphonostomous cyclopid (2.90 mm. long) with which he was acquainted and nothing is known of its host. Bradypontius groenlandicus Hansen. A second species of the same genus from Fox Channel and reported by the Ingolf Expedition from Davis Strait. Hansen’s material included both sexes but the present specimens were all females, with nothing to suggest their host. Bradypontius magniceps (Brady). A third species of the genus from Fox Channel, not obtained by the Ingolf Expedition but reported by Sars from western Finmark a little farther north, without data as to the host. Cyclopina schneideri T. Scott. Found at several localities in Fox Chan- nel and on both coasts of Greenland and recorded by Scott from Spitz- Sept. 15, 1936 WILSON: COPEPODS 375 bergen and Franz-Josef Land and from the Norwegian coast by Sars. This is very likely a free swimming form and not semiparasitic. Corycaeus anglicus Lubbock. Sars has designated this as a pronouncedly pelagic species which is occasionally swept into the Norwegian fjords and similar locations by strong currents. It has not been reported previously from any American locality. Dermatomyzon nigripes (Brady). Found in Fox Channel and Murchison Sound and reported by Scott from Spitzbergen and Franz-Josef Land. It is not, however, exclusively arctic but is also found in the temperate zone. Dyspontius striatus Thorell. Found only in Fox Channel but reported by Sars from the Finmark coast considerably farther north. It is also widely distributed in the temperate zone and is even found in the Bay of Naples. Sars regarded this species as in all probability free swimming. Farranula carinata (Giesbrecht), and F. rostrata (Claus). These minute cyclopids seem out of place in plankton from so far north, but their identi- fication is easy and certain. They were present on the coast of Labrador, in the Bay of Gods Mercy and at Fury and Hecla Strait. Hemicyclops purpureus Boeck. Found in Fox Channel which is a little farther north than previous records of the species. When alive the oviducts and ovisacs are bright red in color and this is not wholly lost in formalin. Lichomolgus agilis (Leydig). Found in Fox Channel and reported by Scott from the British Isles and by Sars from the Norway coast. Sars was able to determine that the Norway specimens were parasitic upon nudi- branch mollusks. Myzopontius pungens Giesbrecht. Found only in Fox Channel and re- ported by Scott from Franz-Josef Land and by Sars from the Norway coast although Giesbrecht’s types came from the Bay of Naples. Oithona similis Claus. Found in most of the localities here listed and very widely distributed in all oceans and zones. To judge from the larvae found in the surface tows it breeds extensively in these northern latitudes. Oithona spinirostris Claus. Common along the Norwegian coast as far north as the present localities, but has not before been reported from any American plankton. Usually found in company with the preceding species. Oithonina nana Giesbrecht. Found in company with the preceding along the southern portion of the Labrador coast, and this record is apparently a little farther north than any previous American report. Oncaea borealis G. O. Sars. Found at the Bay of Gods Mercy and in Fox Channel and reported from the Norway coast and the Polar Sea. It is somewhat more strictly an arctic form and not found much farther south. Oncaea venusta Philippi. This is another southern form found on the coast of Labrador and in the Bay of Gods Mercy. It is very widely dis- tributed in temperate and tropical localities but this is the farthest north that it has been recorded. Parartotrogus arcticus T. Scott. Found in Fox Channel and at Pendulum Island and has been reported from Spitzbergen and Nova Zembla by Scott, - and from the east coast of Greenland by Hansen. Pseudomolgus groenlandicus Hansen. Found only at Pendulum Island on the east coast of Greendland but reported by Hansen to occur also on the west coast along the shore of Baffin Bay. NOTES ON THE NOTODELPHYOIDA Botryllophilus inaequipes Hansen. A single female was taken in Fox Channel while Hansen’s types were obtained from Davis Strait. It will prob- ably be found in other localities upon further investigation. 376 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 26, NO. 9 Doropygelia thorellii Aurivillius. Found only in Fox Channel but taken by the Ingolf Expedition from Davis Strait and by Schmidt from southeast of Iceland. It is found in the branchial cavity of ascidians. Herpyllobius arcticus Steenstrup & Liitken. A single female was found on the annelid, Harmothoé imbricata, on the east coast of Greenland. The Ingolf Expedition reported the species from Davis Strait and the west coast, all the annelid hosts belonging to the genus Harmothoé. ICHTHYOLOGY.—A new polynemid fish collected in the Sadong River, Sarawak, by Dr. William T. Hornaday, with notes on the genera of Polynemidae.! GrorGE S. Myers, United States Na- tional Museum. (Communicated by A. WETMORE.) Fifty-nine years ago, William T. Hornaday, now the honored di- rector-emeritus of the New York Zoological Park, travelled in India, the Malay Peninsula, and Borneo to collect natural history specimens for Ward’s Natural Science Establishment of Rochester. His book dealing with that trip, Two Years in the Jungle (New York, 1885), is now one of the classics of zoological exploration in Asia. The fishes collected by Dr. Hornaday became the property of the United States National Museum, in which institution he remained for several years as chief taxidermist. Some of the fishes were identified by Dr. Tarleton Bean and the rest have recently been determined by the present writer. One species, a remarkable Polynemus, appears to be unnamed more than half a century after its collection. Polynemus hornadayi, n. sp. Holotype.-—U.S.N.M. 100632, a specimen 195 mm standard length and 260 mm including caudal fin, obtained by W. T. Hornaday on October 2, 1877, while fishing with poison, in the Ensengi River, a large creek emptying into the Sadong River from the west about six miles below Simujan, south- western Sarawak, Borneo. Dr. Hornaday described the Ensengi as a stream 40 feet wide and 8 to 10 feet deep, with murky water and swift current. The holotype is the identical specimen illustrated on the plate facing page 386 of Two Years in the Jungle, from a pen and ink drawing of the late Dr. Frederic A. Lucas. The figure is reproduced here. Paratypes.—U.S.N.M. 35719, ten smaller specimens, obtained at the same place and on the same date. One of these is now in the British Museum. Diagnosis.—A species of Polynemus allied to P. hilleri Fowler, P. para- diseus Linnaeus, and P. dubius Bleeker in the presence of but seven dorsal spines, but differing in the very tiny scales, 94 to 97 in the lateral line to caudal base and 31 to 35 in transverse series from origin of first dorsal to pelvic origin. Description.—Dorsal VII-I, 153. Anal II—114. Pectorals 16 to 18, with 7 filaments. Pelvics I-5. Lateral line scales with pores, from upper end of gill opening to end of hypural fan, 94 to 97. Transverse scales from origin of first dorsal to pelvic base 8 to 9/1/21 to 25. Gill rakers fairly long, 12 on upper and 15 on lower limb of first arch. one Published by permission Secretary, Smithsonian Institution. Received April 17, Sept. 15, 1936 MYERS: POLYNEMIDAE 377 Greatest depth approximately equal to head length, 4.8 in standard length. Eyes minute, 9.5 in head length, 2.9 to 3.1 in the convex interorbital space; approximately 2 in snout, and nearly 7 in postorbital part of head. Mouth rather large, reaching far behind eye. Maxillaries scaly, 1.8 in head. Upper jaw not emarginate at symphysis. Anterior and posterior nostrils close together, immediately in front of eye. Preopercular edge finely denticu- lated for some distance above its rounded posterior angle. Upper lip scarcely evident; lower well developed, but interrupted at symphysis. A narrow band of extremely fine, villiform teeth, narrowly interrupted at the symphysis, in each jaw; the band is wider towards the front but the teeth do not extend to the outside of the jaws. Vomer with a narrow, transverse, crescentic (or angled) patch of similar teeth. An oblong patch, rounded in front and acu- Fig. 1—Polynemus hornadayi, new species. Holotype. After Hornaday, from a drawing by the late Dr. Frederic A. Lucas. Reproduced by permission of Charles Scrib- ner’s Sons, New York. minate behind, of similar teeth on each palatine, and a very small, narrow patch on each pterygoid. Origin of first dorsal very slightly behind that of pectorals. Spines of first dorsal slender and somewhat flexible, the first one nearly three fourths as long as the second. Second spine longest, a little shorter than head, minus snout. Origin of second dorsal slightly in advance of that of anal. Soft dorsal, anal, and caudal densely and finely scaled for more than half their lengths. A few rows of fine scales along and behind each spine of first dorsal, and along pelvic spine. Caudal long and very deeply forked, with pointed lobes. Pectoral fin pointed, straight, the upper part of its base above mid-line of body depth, its rays all simple; the fin is approximately 1.5 times as long as head and reaches the posterior third of the soft dorsal base. The two upper pectoral filaments exceed the caudal tip by more than one and two-thirds the total length of the fish (from snout tip to caudal tip). The length of the longest (upper) filament on the holotype is equal to more than 2.5 the total length of the fish including caudal. The third filament is nearly as long as the first two, the fourth nearly reaches to caudal tip, the fifth to the anal origin, the sixth to vent, and the seventh to beyond middle of the appressed pelvics. Pelvic fins longer than postorbital part of head. Distance between origins of anal and pelvic fins approximately equal to head length. Vent far in advance of anal fin, between tips of appressed pelvic fins. Air-bladder absent. 378 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES’ VOL. 26, NO. 9 Color of alcohol specimens dirty brownish. No black marks visible on pectorals. Measurements.—Measurements in millimeters of four specimens are given. These are taken from point. to point, as indicated, with dividers, not re- duced to ideal axial measurements. Under each category the first measure- ment refers to the holotype, the second to the British Museum paratype, and the third and fourth to two other paratypes. Standard length 195, 142, 141, 44. Depth at first dorsal origin 53, 34, 35, 26.5. Head length (minus opercular membrane) 53, 37, 36, 28.5. Maxillary length (snout tip to its end) 28, 22, 20, 16. Snout (middle of tip to eye) 12, 7, 8.5, 7. Eye diameter (horizontal) 5.5, 4, 3.5, 3. Interorbital (between fleshy orbital rims) 16, 12, 12, 9.5. Postorbital 38, 28, 24.5, 19.5. Width end maxillary 9, 6, 6.5, 4. Predorsal (snout tip to origin first dorsal) 67, 46, 48, 36. Snout tip to second dorsal origin 113, 82, 83, 63. Least depth caudal peduncle 21, 15, 14.5, 11. Length caudal peduncle (base of last anal ray to middle of end of hypural joint) 46.5, 37, 37, 26. Pelvic origin to anal origin 58, 42, 38, 27.5. Length second spine of first dorsal 35.5, 25.5, 27, broken in smallest example. Length spine of soft dorsal 12.5, 10, 11.5, 9. Length second anal spine 13, 14, 13, 7.5. Length pectoral fin 85, 58.5, 58, 48.5. Length first free pectoral filament of holotype 670 (filaments broken in most of other specimens). The relationships of P. hornadayi are discussed below, under the genus Polynemus. NOTES ON THE GENERA OF POLYNEMIDAE” In attempting to place the new species described above, I have en- countered considerable confusion in the generic classification of this family. After examining the material in the National Museum, I have prepared the following synopsis of the genera, which, while merely tentative, seems to express the general phylogenetic lines in the family better than most systems. The last general review of Polynemid genera was given by Gill (Proc. Acad. Nat. Sci. Philadelphia, 1861, pp. 271-282) and since his revision there has been little improvement in the system and few new characters have been brought forward. Although the time is not ripe for an exhaustive discussion of the phylogeny of the genera, it appears to me that Polydactylus contains the most generalized species, and all of the other genera are more or less direct derivatives of Polydactylus-like ancestors. Eleutheronema has specialized in the dentition and the elongate, regular body form. Filamanus has developed a peculiar physiognomy, long filaments, and a very compressed form, while it has lost the pectoral fold. Penta- nemus has something of the appearance of Filimanus, but has de- veloped a very long anal. Galeoides is much like the plebejus type of Polydactylus but it has a narrow maxillary and has developed numer- 2 It should be noted that in the case of Polynemids with very long filaments, such as Pentanemus and Polynemus, the filaments are brittle and easily broken when the specimens were originally preserved in alcohol. Specimens originally fixed in formalin have the filaments pliable and tough, and not easily broken. | Spr. 15, 1936 MYERS: POLYNEMIDAE 379 ous filaments extending far forward, with a long pectoral fold to cover them. Polynemus is a very distinct branch, which has a-spe- cialized shoulder girdle architecture, very long filaments, loss of air- bladder, a slender peduncle, long caudal, sharp head, and small scales. It, too, has lost the pectoral fold in changing the position of the girdle elements. Polistonemus is purely an offshoot of Polynemus, differing only in the much greater number of filaments. If intermedi- ate species are found, it should not be recognized as a distinct genus. Synopsis of the Genera la. Anal fin approximately twice as long as soft dorsal fin, and its origin more anterior than that of the latter; preopercular edge entire; pectoral fin inserted rather high, without a fleshy fold extending from the lower part of its base to cover the bases of one or more of the pectoral filaments; maxillary widened behind, the mouth rather oblique and the snout projecting very little beyond it; pectoral rays all simple; pectoral filaments very long, 5 in number; teeth not extending to outer part of jaws. Pentanemus Giinther. 1b. Anal fin of approximately the same length as soft dorsal fin, or shorter; preopercular edge more or less denticulated. 2a. Pectoral fin inserted low, the upper part of its base much below mid-line of body; lateral line straight or with a faint curve anteri- orly. 3a. No sharp fold of skin extending from lower part of base of pectoral fin to cover the bases of one or more of the pectoral filaments; maxillary very wide at end; mouth very oblique; snout squarish and very blunt and projecting little or not at all beyond mouth; scales large; pectoral filaments 7 in num- ber, very long; lateral line angled, with the forward part high though but little curved. Filimanus, new genus. 3b. A sharp fold of skin projecting downward or forward from lower end of base of pectoral fin and covering the bases of one or more of the pectoral filaments; mouth chiefly horizontal; snout projecting considerably beyond mouth; pectoral fila- ments rather short; lateral line nearly straight. 4a. Lower lip absent except towards rictus, the teeth extend- ing on the exterior part of the jaws; elongate, small- scaled species with a very large mouth, long maxillary, and only 3 or 4 pectoral filaments. Hleutheronema Bleeker. 4b. Lower lip extending far forward and no teeth on outside of jaws. 5a. Maxillary distinctly widened at its end; pectoral fold little developed. Polydactylus Lacépéde. 380 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 26, No. 9 5b. Maxillary scarcely widened at its end; pectoral fold very well developed, covering the bases of most of the 9 or 10 pectoral filaments. Galeoides Ginther. 2b. Pectoral fin inserted high, the upper part of its base at mid-depth of body, or higher, without a sharp pectoral fold extending down from lower part of pectoral base to cover the bases of one or more of the pectoral filaments; base of upper pectoral filament inserted higher than lowest ray of pectoral fin; lateral line with its anterior part rising in a long, low curve; caudal peduncle elongate; head small; pectoral filaments very long; maxillary widened behind; no air-bladder. 6a. Seven pectoral filaments. Polynemus Linnaeus. 6b. Fourteen pectoral filaments. Polistonemus Gill. PENTANEMUS Giinther Pentanemus (Artedi) Giinther, Cat. Fish. Brit. Mus., 2: 331. 1860 (type by monotypy Polynemus quinquarzus Linnaeus). A single species of this genus. Pentanemus quinquarius (Linnaeus), from West Africa, is known. I have not seen specimens, but have derived my data from Boulenger’s excellent figure and description (Cat. Freshwater Fishes Africa, 4: 100. 1916. fig. 61). Filimanus, n. g. This new genus is based upon a single species, the genotype, Polynemus melanochir Cuvier and Valenciennes (Hist. Nat. Poiss., 7: 513. 1831), from the Malay Archipelago. I have examined only a single specimen of Fili- manus melanochir, from Java. (U.S.N.M. 72742). ELEUTHERONEMA Bleeker Eleutheronema Bleeker, Versl. Akad. Amsterdam, 14:110. 1862 (type by monotypy Polynemus tetradactylus ‘‘C.V.”’ = P. tetradactylus Shaw) ; Bleeker, Versl. Akad. Amsterdam, vol. 14, 1862, p. 128 (description, with inclusion of two species, P. tetradactylus and P. tridactylus Bleeker). In the first citation of Bleeker given above he merely lists “Hleutheronema tetradactylus Blkr. = Polynemus tetradactylus C. V.’”’ Since the mere citation, prior to January 1, 1931, of a recognizable, previously described species in conjunction with a new generic name is sufficient under the International Rules (Article 25), this reference validates the generic name Hleutheronema. He mentions only one species, tetradactylus of Cuvier and Valenciennes, which zpso facto becomes the genotype. Only two species of Eleutheronema are currently recognized, H. tetra- dactylum, with four pectoral filaments, ranging from India to Northwestern Australia, and E. tridactylum, with three pectoral filaments, from the Malay Peninsula and Archipelago. Of the former, I have examined three specimens in the National Museum, from Rangoon (44726), Java (72740), and Formosa (85481). Of the latter species, I have studied one example (72737) from Java. PoutypactTyLus Lacépéde Polydactylus Lacépéde, Hist. Nat. Poiss., 5: 419. 1803 (type by mono- typy Polydactylus plumierii Lacépéde = Polynemus virginicus Linnaeus). SEPT. 15, 1936 MYERS: POLYNEMIDAE o8l Trichidion (Klein) Gill, Proc. Acad. Nat. Sci., Philadelphia, 1861, p. 274 (type by original designation Polynemus plumieriz Lacépéde = Polynemus virginicus Linnaeus). Klein’s name T'richidion was originally published in 1749. It is thus pre- Linnaean and can not be accepted. Walbaum, in 1792 (Artedi Gen. Pisc., pt. 3, p. 585), republished Klein’s Trzchidion but the International Commission on Zoological Nomenclature, in Opinion 21, has declared these reprinted names of Klein to be unavailable under the Code. This large genus is, as I define it, a very heterogeneous assemblage, but the species all seem to be more similar to each other than to any members of other genera. The range is practically co-extensive with that of the family. Very likely a careful study of adequate material will show a need for breaking it up into several more well knit groups. For example, Polydactylus quadri- filis is very different in appearance from the ordinary types such as wr- ginicus, approximans, and plebejus. I have examined a number of species in the National Museum, including P. virginicus (64063), P. octonemus (48883), P. opercularis (41054), P. ap- proximans (65621), P. sexfilis (55555; 65990), P. andicus (76627), P. plebejus (58048), and P. heptadactylus (72741). GALEOIDES Giinther Galeoides Giinther, Cat. Fish. Brit. Mus., 2: 332. 1860 (type by mono- typy Polynemus polydactylus Vahl=Polynemus decadactylus Bloch). This genus contains two species. Of the West African Galeotdes decadac- tylus I have examined several specimens from Sierra Leone and Ashantee (U.S.N.M. 42196; 42213). Steindachner, in his Ichthyologische Notizen, No. 8 (Sitzb. Ak. Wien, 60 (1): 137. 1870), has described Galeoides microps from China. I have not seen this fish and I find no reference to any examples obtained since the original description. PoLyNEMUS Linnaeus Polynemus Linnaeus, Syst. Nat., ed. 10, 1758, p. 317 (type as fixed by the International Commission on Zoological Nomenclature, under suspension of the Rules, Opinion 93, 1926, Polynemus paradiseus Linnaeus). The confusion that has reigned in regard to the application of the generic name Polynemus has happily been settled (whether on adequate premises or not I do not attempt to say) by the International Commission in Opinion 93 (Smithsonian Misc. Coll., 73: 5-10. 1926.) In Weber and de Beaufort’s useful synopsis of the Indo-Australian Poly- nemids (Fishes of the Indo-Australian Archipelago, 4: 196-218. 1922), the generic affinities of several of the species, according to the system proposed here, is not evident. In their key to the species of ‘‘Polynemus’’ (pp. 201- 202) it would seem that all of the species in sections 1 and 2 belong to Polydactylus, in addition to heptadactylus in section 3. Polynemus melanochir is referred above to a new genus, Filimanus. Polynemus longipectoralis, P. dubius, and P. borneensis (from which Trichidion hillerz of Fowler is distinct) certainly are congeneric with the Indian P. paradiseus. Polynemus macroph- thalmus also probably belongs to this genus. Including hillerz and hornadayz there seem, then, to be only seven species certainly referable to Polynemus. I have seen specimens of none of them save hornadayi, but in order to in- dicate the characters of the new form I have compiled the following key. My data for paradiseus is from Day (Fishes of India, p. 176, pl. 42, fig. 4). 382 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES’ VOL. 26, No. 9 Synopsis of the Species la. Dorsal spines 7, the first more than half as long as the second, which is the longest. 2a. Scales 94 to 97; pectoral fin plain, reaching last third of soft dorsal base; upper pectoral filaments more than twice as long as body and caudal fin; snout blunt and broad; eye 9.5 in head, about 3 in interorbital. (Sadong R., Borneo). P. hornadayt Myers. 2b. Scales 65 to 70 (or slightly more?). 3a. Pectoral fin with most of its terminal portion black; upper pectoral filaments only slightly longer than body and caudal fin. (Baram R., Borneo.) P. hillert (Fowler). 3b. Pectoral fin plain; pectoral filaments exceeding tip of caudal by nearly the length of head and body. 4a. Eye diameter about 3 in interorbital; pectoral fin with 15 rays; upper lobe of caudal longer than lower; scales about 70. (Bay of Bengal and coast of Burma). P. paradiseus Linnaeus. Ab. Eye 1.8 in interorbital; pectoral fin with 17 rays; caudal lobes equal; scales 65 to 70. (Sumatra, Borneo, and Siam). P. dubius Bleeker. 1b. Dorsal spines 8, the first minute, the third one longest. 5a. Pectoral fins about as long as head without snout; eye large, 4.5 to 6 in head; distance between origins of pelvics and anal less than length of head; scales 88 to 98; anal spines 3. (Sumatra and Borneo. ) P. macrophthalmus Bleeker. 5b. Pectorals longer than head; eye small, 5.5 to 7 or more in head; distance between origins of pelvics and anal more than length of head. 6a. Scales 84 to 87; anal spines 2. (Banjermassin, Borneo.) P. longipectoralis Web. & de Bfrt. 6b. Scales 65 to 66; anal spines 3. (Sumatra and Borneo.) P. borneensis Bleeker. POLISTONEMUS Gill Polistonemus Gill, Proc. Acad. Nat. Sci. Philadelphia, 1861, p. 277 (type by original designation Polynemus multifilis Schlegel). This genus is in all respects a Polynemus excepting for the increased num- ber of pectoral filaments. I have examined two fine examples of Polzs- tonemus multifilis (U.S.N.M. 53409; 53410) collected in the Kapoeas River, Borneo, by Dr, W. L. Abbott, This is the only known species. Sept. 15, 1936 PROCEEDINGS: GEOLOGICAL SOCIETY 383 PROCEEDINGS OF THE ACADEMY AND AFFILIATED SOCIETIES GEOLOGICAL SOCIETY 531ST MEETING The 531st meeting was held in the Assembly Hall of the Cosmos Club, November 13, 1935, President W. T. ScHALLER presiding. Informal communications.—ANNA I. Jonas presented new evidence for the age of the limestones of Frederick Valley, Maryland. She stated that the valley contains on its borders blue slaty limestone, called Frederick lime- stone, and in the center of the syncline purer, more massive, Grove lime- stone. Study of the fossils from these limestones has shown the Frederick limestone is Upper Cambrian in age, the Lower Ozarkian of Ulrich. This conclusion is based on the determination of brachiopods and trilobites by Cooper, Raymond, and Bridge. The overlying Grove limestone, on the basis of its fossils, cephalopods and brachiopods, as determined by Foerste and Cooper, is of Upper Ozarkian age comparable to the Chepultepec of Ten- nessee and Gasconade of Missouri. There is no Canadian or Middle Ordovi- cian in the valley as was thought by Bassler. The Conestoga limestone of Pennsylvania is similar to the limestones of Frederick Valley in lithology. Both the Conestoga and Frederick limestone unconformably overlie Lower and Middle Cambrian formations. Brachiopods believed to be from the Conestoga limestone were previously called Chazyan in age and considered the same as those from the Frederick limestone. These brachiopods have been recently determined by Cooper and Resser to be Nisusia festinata, characteristic of the Lower Cambrian Kinzers formation. The quarry from which the brachiopods were obtained is on the south face of a hill of Kinzers shale. At present no fossils except crinoid stems have been found in the Conestoga and it may be Upper Cambrian or lowest Ordovician in age. (Author’s abstract.) Program.—M. R. CAMPBELL: The origin of the material forming the alluvial fan of Potomac river. Discussed by Mr. STose. Ernst Cioos and H. GarLtanp HERsHEY, Johns Hopkins University: Structural age determination of Piedmont granites in Maryland. A direct age determination of Piedmont intrusives is not possible where they are sepa- rated from known Paleozoic sediments by a wide belt of metamorphics of disputed age. Therefore, structures which are recognizable over a large area were used. Flow cleavage and fracture cleavage are Paleozoic (post- Conestoga) structures that can be recognized everywhere in the Piedmont irom Maine to Alabama. The Maryland intrusives—Port Deposit Granite Gneiss and Baltimore Gabbro—follow, transgress, and engulf these struc- tures and destroy them by recrystallization through metamorphism. The writers suggest that these intrusives are Paleozoic and not pre-Cambrian as previously assumed. Flow cleavage and fracture cleavage may be useful tools in other portions of the Piedmont, and it may be necessary to change the age determinations of other intrusives. (Authors’ abstract.) Discussed by Mr. Srosz and Miss Jonas. 532ND MEETING The 532nd meeting was held in the Assembly Hall of the Cosmos Club, November 27, 1935, President ScHALLER presiding. 384 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 26, NO. 9 Informal communications.—C. 8. Howarp showed graphs of suspended matter, sulphates, and total discharge of the Colorado river at Grand Can- yon and at Willow Beach. Program.—TaistA STADNICHENKO: Petrography and microstructure of coal. While the first microscopic examinations of coal were made by Witham and Hutton in England in 1833, the systematic studies attempting to correlate the observed structures with the physical characteristics and the chemical composition date back not more than twenty-five years. The voluminous literature on the ‘‘petrography”’ of coal deals mostly with the descriptions of the observed structures and gives some botanical and morphological in- terpretations. Comparatively few works are devoted to the broader aspects of the origin, composition, conditions of accumulation and the effects of local and regional metamorphism on the various plant components con- tributing to the coal bed. At present several systems of nomenclature are in use and a number of terms are either synonymous or overlap one another. TABLE 1.—SystTEMsS OF NOMENCLATURE Common terms Stopes (1919) Thiessen (1920) used for over Fayal (1887 ) Jeffrey (1915) Expanded in Expanded a century 1932 1929 Bright Houilles Vitrain Anthraxylon claires Lignoid Attritus Dull Houilles Clarain foliares Canneloid Hard Houilles ternes Durain Also splint, synonymous with durain Mother of coal | Fusain Fusain Mineral charcoal (or fusain in France) The significance of the new terms is often exaggerated, as little has been added to our knowledge of coal through the introduction of ‘“‘clarain”’ or “attritus” in place of plant micro-debris, or ‘‘durain” in place of dull or hard coal—terms used for almost a century. The improved technique in the preparation of thin sections and polished surfaces of coal and the examination of the columnar samples in addition to an increase in the definite knowledge of the plant components of a coal bed contributed also the better understanding of the various problems such as origin, conditions of accumulation, coking, characteristics of ash, washing of coal, and fractional separation of the various components of a coal bed. (Author’s abstract.) Discussed by Mr. SCHALLER. W.H. Braver: Faulting of unconsolidated beds. Certain lithologic units of Upper Devonian age that crop out in western Steuben County, New York, show structural features indicating that the sea floor was warped and faulted during the depositional epoch. The warping is indicated most clearly by one sandstone unit 350 feet thick that changes rather abruptly into shale where it goes into certain synclines and back to sandstone again where it comes out Sept. 15, 1936 PROCEEDINGS: GEOLOGICAL SOCIETY 385 the other side and yet maintains a uniform thickness. Apparently the syn- clines where the shale accumulated sank enough during deposition to accom- modate a thickness of mud that, when compacted, equaled the thickness of the sand deposited adjacent to the troughs in the same time interval. The faults that occurred while the beds were still soft have throws ranging from a few feet to more than 100 feet. They are accompanied by bands of contorted, folded, and faulted beds of shale and fine grained sandstone that range in width from 100 yards to nearly two miles. The bands of deformed rocks are restricted vertically to a stratigraphic interval estimated to be 100 to 200 feet thick. Farthest away from a fault the beds in a deformed zone are gently folded but toward the fault the folding increases in intensity and cul- minates in a zone along the fault line where individual beds lose their iden- tity by flowage and rupture. Layers of sand, clay, and coquina flowed to- gether and commingled in a way that could only have happened when the material was nearly fluid. Layers of sand were plastically deformed and in the lower part of the deformed zones failed along the crests of sharp folds by microfaults. Microfaults in layers of unconsolidated sand apparently indicate that the deforming stress was applied rapidly. Many features of these deformed zones resemble features produced in modern unconsolidated deposits by earthquakes. Hence, they are inter- preted as the result of faults that tilted the sea floor and produced earth- quakes. (Author’s abstract.) Discussed by Messrs. Ruspry, Mispr, and Miss JONAS. Puiuir B. Kine: Permian of the Guadalupe mountains. This paper describes the complex stratigraphic variations of a group of rocks of Permian age exposed in the Guadalupe Mountains, a range lying in western Texas near the New Mexico boundary. The rocks are wonderfully exposed in steep, bare escarpments and deep canyons. They are all of marine origin, and con- sist chiefly of various sorts of limestone and sandstone. Fossils occur in great numbers in certain beds. The stratigraphic variations consist of: 1. A prominent unconformity in the lower part of the sequence, by which over 1,000 feet of the sandstones of the lower part of the Delaware Mountain formation overlap northward against the uplifted surface of the preceding Bone Spring limestone. 2. Abrupt replacement northward of sandstones and thin-bedded limestones of the middle and upper part of the Delaware Mountain formation by massive limestones of the Dog Canyon and Capitan formations. 3. Replacement of the massive Capitan limestone still farther north by the thin-bedded Carlsbad limestone. The massive Capitan and Dog Canyon limestones are found to occupy northeast-trending belts a few miles in width, and they apparently stood as barriers between the unlike deposits of Carlsbad and Delaware Mountain type. These features are related to the margins of a subsiding area, the Delaware Basin, which existed in Permian time, and the massive limestones have many features in common with the modern barrier reefs. (Author’s abstract.) Discussed by Messrs. R. C. WELLS, J. 5S. WILLIAMS, WoopRING, GILLULY, SEARS. 533RD MEETING The 533rd meeting was held in the Assembly Hall of the Cosmos Club, December 11, 1935, President SCHALLER presiding. By vote of the Society Mr. J. C. ReEp received a cash prize of ten dollars for the best presentation of a paper by a member during the past year. Messrs. Trask, KNEcHTEL, G. A. CoopmrR, BRADLEY, ANDREWS, and HEWETT were voted honorable mention. 386 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 26, No. 9 Vice-President H. D. Miszr took the chair during the presentation of Dr. Schaller’s Presidential Address entitled: A muzneralogist ventures in geology. 43RD ANNUAL MEETING The 43rd Annual Meeting was held in the Assembly Hall of the Cosmos Club after the adjournment of the 533rd regular meeting, President W. T. SCHALLER presiding. The annual report of the Secretaries was read. The Treasurer then presented his annual report showing an excess of assets over liabilities of $1,158.18 on December 11, 1935. The Auditing Committee re- ported that the books of the Treasurer were correct. The results of the balloting for officers for the ensuing year were as follows: President, M. I. GotpMaNn; Vice-Presidents, H. D. Mismr and R. C. WELLs; Treasurer, A. H. KoscHMann; Secretaries, Gho. TUNELL and G. A. COOPER; Members-at-large-of-the-Council, C. L. Gazin, J. F. Scoarrer, F. C. Cat- KINS, C. F. Park, and L. G. HenBrest; Nominee for Vice-President of the Washington Academy of Sciences representing the Geological Society, W. T. SCHALLER. W. D. JoHNsTON, JR., and G. TUNELL, Secretaries. 534TH MEETING The 534th meeting was held in the Assembly Hall of the Cosmos Club, January 8, 1936, President M. I. GoLpDMAN presiding. Informal Communications —M. M. KNECHTEL gave new information on the Gila formation of the Gila and San Simon valleys of Arizona. The dis- covery of fossils in lake beds of the Gila formation allows dating of the for- mation as Upper Pliocene. C. M. Miuron described A foraminiferal—analcite shale from Texas. In his study of the Terlingua mercury region in Texas in 1934 Mr. C. P. Ross obtained some specimens which showed rather exceptional features. One of them, this rock, is a highly fossiliferous tarry shale, the fossils being fora- minifera of the Globigerina and Tezxtularia types. These of course are too small to be seen with the naked eye; but prominent in the black tarry rock are great numbers of white analcite crystals, each the size of a pinhead or larger. Analcite is a common mineral in the igneous rocks of the region. One such amygdular rock examined in the laboratory, has analcite both as part of the groundmass, and in the amygdules, where it is associated with quartz, chalcedony, and calcite. It is of interest that the formation of analcite apparently occurred over a range of physical-chemical conditions: first, as a primary constituent of the igneous rock, second, under conditions of amygdular formation; and lastly in a calcareous shale. Discussed by C. P. Ross. Program.—Watson Monroe: Upper Cretaceous and lower Tertiary history of the Jackson area, Mississippi. The Jackson anticline is in south-central Mississippi about 40 miles east of the Mississippi River at Vicksburg. The core of the anticline is a deeply buried hill of Paleozoic sedimentary and Cretaceous igneous rocks. The rocks penetrated only in wells range in age from Carboniferous to Claiborne (Eocene). Those exposed at the surface range in age from Claiborne to Recent. Upper Cretaceous rocks overlie the Carboniferous. On the flanks of the Jackson anticline are several thousand feet of Cretaceous rocks not present on the top. Unconformably overlying the Cretaceous is the lower part of the Tertiary Sept. 15, 1936 PROCEEDINGS: GEOLOGICAL SOCIETY 387 system consisting of alternating marine and nonmarine beds with discon- formities at the top of each nonmarine bed. These disconformities may not represent uplift, but rather submergence of old deltas and coastal swamps. As there is no evidence of uplift of the area except for the disconformities, the lower Tertiary history may be one of subsidence and deposition only. The only deformations of the bedrock during lower Tertiary time were downwarpings at the beginning of deposition of each marine formation. The structure of the top of the basal Eocene formation (the Clayton lime- stone) appears in large part to be the result of compaction of small masses of Upper Cretaceous clay on the crest of the anticline and of thick beds of clay on the flanks. The structure gets progressively less complex upward in the section. Two main factors have influenced the geologic history of the area: First, there was not necessarily any uplift of the bedrock surface in the Jackson area from the beginning of the Eocene until the end of the Oligocene, and second, the structure of the rocks is the result of differential compaction of ey and Tertiary sediments over and around a buried hill. (Awthor’s abstract. Davin A. ANDREWs: Suggested Lance-Fort Union correlations in adjacent parts of Montana and North and South Dakota. With the completion of field work for the Rosebud coal field in 1930 and the Mizpah coal field in 1932, in southeastern Montana, the United States Geological Survey has com- pleted the detailed mapping of the strip that extends eastward from the Big Horn River in Montana to the headwaters of the Cannonball River in North Dakota and the Grand River in South Dakota. These reports describe the Big Horn County, Tullock, Forsyth, Rosebud, Mizpah, Miles City, Baker, Terry, and Ekalaka fields in Montana and the Marmarth and Sentinel Butte in North Dakota and northwestern South Dakota. Other fields join- ing or lying near this strip which have been mapped in detail are the Bull Mountain, Ashland, Northern Sheridan, Little Sheep Mountain, and Glen- dive fields in Montana. Early correlations of the Lance and Fort Union formations in this area, made before completion of this detailed mapping, do not agree in detail with the correlations suggested by compilations from these reports. The Cannonball marine member and the demonstrably equivalent portions of the Ludlow lignitic member, at the top of the Lance formation in North Dakota and South Dakota, are now classified as Cretaceous by the U. S. Geological Survey. Lateral tracing through the Ekalaka, Marmarth, Baker, Terry, Miles City, Mizpah, and Rosebud fields suggests that the Ludlow is equivalent to the basal 200-300 feet of the Tongue River member of the Fort Union as mapped in southeastern Montana. In the Rosebud and Miz- pah fields the so-called Lebo shale member at the base of the Fort Union and the Tullock member at the top of the Lance merge eastward into somber colored shales equivalent in large part to the Hell Creek member as mapped in Marmarth and Ekalaka fields and the lower unit of Lance in northwestern South Dakota. Inasmuch as the Tullock and Lebo overlie the Hell Creek near its type locality, the Hell Creek of Ekalaka and adjoining fields prob- ably contains beds younger than type Hell Creek. G. 8. Rogers, L. H. Woolsey and others originally correlated the beds in southeastern Montana with the type Lebo of the Crazy Mountains on the basis of high volcanic and andesitic content. Later published material of Coleman Renick and un- published work of M. N. Bramlette, F. S. Parker, and W. G. Pierce, of the U.S. Geological Survey, who found no microscopic evidence for differentiat- 388 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 26, NO. 9 ing Lebo from the overlying Tongue River and the underlying Tullock, do not confirm the earlier correlations. It is suggested that the name Lebo should be dropped in southeastern Montana, and Hell Creek should be dropped in extreme southeastern Mon- tana, adjoining parts of North and South Dakota and that some other name be substituted for the somber shale unit of the Lance of that region which does not imply correlations that now appear uncertain or even improbable. Although the basal 200 feet + of the Tongue River member appears to be equivalent to the Ludlow, it is impracticable and impossible, with data now available, to separate the part equivalent to the Ludlow from the Tongue River. Consequently, it is suggested that the contact between the yellow beds or Tongue River member of the Fort Union and the somber colored beds below (the so-called Lebo shale of Bull Mountain, Tullock, Forsyth, and Rosebud fields), should be taken as the best mappable boundary now known between the Lance and Fort Union in southeastern Montana. Dis- cussed by R. W. Brown, KNEcHTEL, W. C. MANSFIELD, STANTON, Huss. JAMES GILLULY: Pediments of the Ajo region, Arizona. The smoothly sloping carved-rock plains that commonly front the mountains of the south- western deserts—to which the name pediment has been applied—have occa- sioned much discussion among physiographers. There is difference of opinion as to almost every aspect of the pediments, from their shapes to the peculiar causes that determine their formation. The study of these surfaces has been greatly handicapped in the past by the dearth of good base maps showing them. When Bryan set out to illus- trate them, he was forced to fall back on maps of a scale of 1/125000. Part, at least, of the divergent views of these forms is probably due to the fact that quantitative data have been lacking. In the Ajo district an excellent map has recently become available, showing the topography of a considerable area on a scale of 1/48000 and a smaller but representative pediment area on a scale of 1/12000. Studies carried out on these maps furnish the basis for this paper. Ajo lies in the middle of the Papago country of southwestern Arizona, and the surfaces here discussed have been cited by Bryan as typical pediments. The region is part of the Sonoran desert section of the Basin and Range prov- ince. The bedrock structure is dominated by faulting; the latest movement was long enough ago so that many of the faults are represented by fault-line scarps and none of them are directly represented in the topography. The integration of the drainage of this very arid county is additional evidence that the faults are old. The bulk of the mountains are composed of granitic and gneissic rocks, with some steeply tilted old fanglomerates, andesitic tuffs and breccias and some thick, massive andesite flows. All the rocks except the massive ande- site lavas have been carved into sierra topography (mature); the massive andesites form mesas, with sharply cut canyons (young). The pediments are restricted to the fronts of the mature mountains and extend into them for considerable distances along the streams. The andesite mountains are fronted with more or less continuous talus piles that merge gradually into the bahadas—no pediments intervene. The pediments are especially well developed on the softer formations. In plan, the pediment is convex outward as it must be, of course, to con- form to the front of a nearly circular mountain mass. Most of the pediment is concave upward, but along a few of the streams it is convex upward, probably because of stream capture, for adjoining areas on both sides are Sept. 15, 1936 PROCEEDINGS: GEOLOGICAL SOCIETY 389 concave upward. In transverse profile, the pediment is concave upward with respect to all but one, (possibly two) of the streams. The fan-formed pedi- ment at the mouth of one of the streams occurs within a reentrant of the mountain front. It cannot therefore be explained in the same way as the gen- eral convexity of the pediment as a whole nor can it be due to special litho- logic factors, as far as known. However, it is at the debouchure of only one (2 ?) of twenty or more comparable streams and must be regarded as ex- ceptional. The pediment that extends well into the mountains along Darby arroyo has been mapped on the scale of 1/12000 on the Ajo Mining District Special map. Transverse sections of this are uniformly concave upward. Slight addi- tion to the pediment laterally would isolate several of the mountain spurs so that fan form is again not characteristic of the pediment. The drainage is dendritic high on the pediments and the streams are there sunk about 40 feet into the surface marked by the summits of the inter- fluves, but the transition from stream to divide is rather gentle. Further down the pediment, the drainage is more or less parallel and the relief of the surface somewhat less in feet but more abruptly concentrated at the channel edges. Near the bahada, the channels are divergent, and sunk abruptly for 10 to 15 feet. Gravel is there widespread, but higher on the pediments it is confined to narrow belts near the streams. The streams meander only in a few places. It is inferred that the upper parts of the pediments are being lowered by weathering and rill wash. Lateral planation is there negligible as shown by the restriction of the gravel. Farther downstream lateral planation is increasingly important and may eventually dominate in lowering the pedi- ment at the bahada slope. The absence of the pediments from the fronts of the mountains made up of massive andesite shows that weathering is a dominant factor in pediment formation, for although these hard rocks would of course obstruct lateral planation, one might justifiably expect some nar- row pediments on even the hardest rocks were lateral planation the domi- nant factor in their making. The obvious effectiveness of rill wash and weathering to lower the surface high on the pediments raises the tentative question as to whether pediments may not be born with drainage incisions. That is, this may be the form that pediments normally have when growing headward. Of course the incision of the channels at the lower ends of the pediments is sharp and must indicate a change in stream regimen, but it seems worth while to question whether such an interpretation is required for the more gently transitional incision of the streams below the interfluves that seems characteristic of the upper parts of the pediments. Discussed by Messrs. MatrHrs, ANDREWS, RUBEY. 535TH MEETING The 535th meeting was held in the Assembly Hall of the Cosmos Club, January 8, 1936, President M. I. GoLpMAN presiding. Informal communications —O. EK. M&INzER read extracts from a letter written by H. T. Stearns, U. 8. Geological Survey, Spreckelsville, Maui, T. H., January 3, 1936, describing the recent eruption on Mauna Loa. ‘All work in the months of November and December has been over- shadowed by the eruption of Mauna Loa. Friends woke me up on the night it broke out (November 21) and I drove to the Lahaina side of West Maui, where I saw a magnificent view of the eruption. The glowing fume cloud rose about 10 miles into the air and even from this distance (more than 100 390 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 26, NO. 9 miles) I could definitely make out the streams of lava. The glow was so bril- liant that it was successfully photographed from Oahu. -“T went to Hilo by plane early the next morning, thence to the Kilauea Military Camp, where I got animals to go as far as the Rest House at alti- tude 10,000 feet. From there up it was a hard hike to the source fountains at 12,500 feet. Reached the fountains at 10 P.m., and kept warm by them even though snow lay on the ground only a mile away. When | arrived 14 foun- tains were shooting very liquid lava 150 feet into the air, and three rivers of pahoehoe united within 25 feet of where I stood to form one river flowing at the rate of about 25 miles an hour. During the night the temperature of the lava fell slowly so that spatter cones began to form. By morning the fountains had partly walled themselves in and the pahoehoe rivers had slowed down to about 5 miles an hour. The rivers then began to overflow their levees and spread lava at my feet. By 10 a.m., the fountains barely shot above the cones except one lone one away up the rift. By nightfall November 24 all the fountains had ceased except one. “T then drove around to Hummuula, on the slopes of Mauna Kea, and spent November 26, 27, and 28 at the aa flows. Associated Press broadcasted the news that I was lost but instead I was just having a good time at the lavas. I slept at night with my feet in the aa laid down the day before. The aa advanced at various rates during these days—a mile during one night and the next day only 100 feet an hour. The rate varied chiefly with the slope of the ground over which it flowed. I was fortunate enough to see the fluid aa spread a quarter of a mile in less than an hour through a wooded area. It was only 1 to 3 feet thick and so hot that logs covered by the lava sounded like blast furnaces as they burned and some exploded throwing up hot lava into the air. At night some of the trees burned up in a flash like great Roman candles 100 feet high. In spite of this tremendous heat some of the large trees managed to stand upright in the lava flow with their foliage shrivelled but intact.” Program.—M. N. BRAMLETTE: Origin of Arkansas bauxite deposits. Dis- cussed by Messrs. SPENCER, Kine, Mismr, BuRCHARD, RuBrY and Hewert. CHARLES V. THEIS: Possible effects of ground-water on the Ogallala forma- tion of Llano Estacado. The Ogallala formation, of Pliocene age in the area considered, forms the cap of the Llano Estacado, the part of the High Plains south of the Canadian River in Texas and New Mexico. This variable con- tinental formation is made up of four elements: (1) crossbedded sand and gravel occurring principally in old broad valleys in the sub-Ogallala floor, but also occurring in places at every stratigraphic level in the formation; (2) lithologically structureless silt and fine sand making up about 75 per- cent of the total mass of the Ogallala; (3) lime-cemented beds forming ‘“‘caliche’’ at many horizons in the formation, differing from the previous two elements only in being cemented; and (4) a limestone, several feet thick, capping the remainder of the formation, also generally referred to as ‘‘ca- liche.”’ Overlying red silt up to possibly 100 feet thick, not previously sepa- rated from the Ogallala in this area, is considered the probable southern equivalent of the Pleistocene loess occupying the same stratigraphic position northward in Kansas and Nebraska. The petrographic and lithologic char- acter of the gravel deposits indicates that they are channel deposits of streams originating in the mountains to the west. The structureless material making up the bulk of the formation is fine-grained, well sorted, uniform in texture, and makes up the entire mass of the Ogallala throughout outcrops tens of miles long and up to 100 feet thick. These features strongly suggest Sept. 15, 1936 PROCEEDINGS: GEOLOGICAL SOCIETY 391 an eolian emplacement of material originally derived from alluvial deposits in the channels represented by the gravel in the Ogallala of today and also from similar deposits in the portion of the original High Plains to the west of the present remnants. The limestone at the top of the Ogallala formation is distinct from any other ‘“‘caliche’”’ bed in the Ogallala and forms the one key bed in the Ogallala that can be traced practically throughout the High Plains from the southern edge of the formation northward at least to Nebraska. The gnarly bedding of the limestone, the inclusion of breccias made up of limestone fragments cemented in a like matrix, the inclusion of widely spaced sand grains and pebbles, and the apparent presence of algal cells, reported by Elias from Kansas, suggest an open-water origin for much, at least, of this limestone. It appears probable that the water table must have stood at or near the surface in this area throughout at least much of Ogallala time. The thickness of a water body in a permeable mass resting on a sloping impermeable floor is proportional to the distance across it in the direction of dip of the floor, and approximately proportional to the ratio between recharge and permea- bility of the formation. At present the water table in the Ogallala stands about midway in the formation throughout much of the central part of the Llano Estacado. This position is maintained through rainfall penetration amounting to probably no more than 1/2 inch of water a year, or about 3 percent of the total precipitation. Rainfall penetration at the present time is inhibited by the cover of surficial loessial silt of probable Pleistocene age and by the limestone capping the Ogallala. In Ogallala time these were not present. At that time the Ogallala formation extended much farther both east and west. It was receiving flood run-off from mountain areas to the west, and there were no transecting streams tending to lower the water table. It therefore seems most probable that unless the climate was unduly arid the water table in Ogallala time must have stood so near the surface as to dispose of the excess water by evaporation and transpiration. The presence of grass seeds, discovered by Elias in the Ogallala of Kansas, and of hack- berry seeds as far south as the Llano Estacado, seems to indicate that the cli- mate was not severely arid. The area would be visualized as a gently un- dulating surface with pools perhaps in the lower portions, and the water table within the reach of grass roots over most of the area. The climate would not necessarily be more humid than at present. The application of this concept to the sediments entails some difficulties in detail, partly occasioned by the absence of detailed information about the deposits, but appears to explain the larger features fairly well. Under such conditions vegetation would probably be encouraged, which would aid in trapping fine sand and silt from the air and at the same time inhibit the for- mation of dunes, which are apparently absent. The accumulation would tend to build up parallel to the water table and to the underlying impervious floor. In cooler or more moist periods the movement of sand by the wind would be lessened and the water table would rise closer to the surface, thus giving opportunity for greater evaporation and therefore cementation of the sur- ficial sediments. In this way, the sand and silt would be cemented to form “‘caliche’’ at certain horizons. As Pleistocene time was approached or entered, with continued cooling and probably greater precipitation the water table would rise closer to the surface, forming more extensive pools in which limestone would be formed by algae or by inorganic means. At the same time, for the same reasons, soil moisture in the areas furnishing the sand to the air would be increased and deflation checked, thus giving an opportunity 392 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 26, No. 9 for the accumulation of limestone relatively free of sand. A discussion of how large such pools must have been or how much they could have shifted must be deferred until more detailed data are available regarding the lithol- ogy and petrography of the limestone. With continued cooling and greater rainfall, as the Pleistocene was well entered, a further rise of the water table would cause overflow of the pools and integration of surface drainage, with consequent rapid dissection and lowering of the water table, eventually to its present position. Discussed by BrapLEy, RuBEY, MEINZER, GILLULY, K. E. Lonman, GOLDMAN. ParKER D. TRASK: Proportion of organic matter converted into oil in Sante Fe Springs field, California. All the oil that was in the Sante Fe Springs field at the time the field was discovered is equivalent to a yield of 0.053 percent of the weight of the prism of sediments from which the oil seems to have been derived. As the organic content of these sediments at time of dep- osition is calculated to be 3.0 percent of the weight of the sediments, the yield of oil therefore would be 0.053/3.0 or 1.8 percent of the organic content of the sediments. This represents the minimum production of oil by the organic matter. The actual quantity that was generated by the source beds presumably was larger, as some oil that was formed may have failed to reach the reservoir. When the possible sources of loss of oil, such as retention by source beds, escape to the surface of the ground, destruction by bacteria and trapping and adsorption while migrating, are considered, the conclusion is reached that the most probable yield of oil by the organic constituents is of the order of magnitude of 4 percent, though it may have been as high as 15 percent. Discussed by SpENcmR, J. 8. W1LLIAMs, HEN- DRICKS. 536TH MEETING The 536th meeting was held in the Assembly Hall of the Cosmos Club, February 12, 1936, President M. I. GoLpMan presiding. Informal communications —Mr. E. T. ALLEN reviewed a new book by F. A. Perret, Eruption of Mt. Pelee, 1929-1932. Discussed by M. I. Goxp- MAN. Program.—AvoupH KnopFr: The igneous geology of the Spanish Peaks region, Colorado. Discussed by Messrs. Mirtin, GILLULY, SHENON, R. C. WELLS, SToOsE, GODDARD. F.E.Marruss: Erosional processes in the Alpine Zone of the Sierra Nevada. Discussed by Messrs. Murtiz, Capps, Brapuey, R. F. Grices, FAHEY. 537TH MEETING The 537th meeting was held in the Assembly Hall of the Cosmos Club, February 26, 1936, President M. I. GoLpMAN presiding. Informal communications.—L. G. Hrnpest described and illustrated peculiar odlite grains from the lower part of the Brentwood limestone, lower Pennsylvanian, near Fayetteville, Arkansas. The majority of these odlites have quartz nuclei of which the greater number are unabraded, euhedral crystals. By darkfield illumination particularly, these nuclei are divisible into two parts—an inner, more or less rounded grain of crystalline quartz and an investment which bears the crystal faces and has simultaneous ex- tinction with the inner grain. The investment contains inclusions which clearly were originally a part of the concretionary structure of the calcareous zone in the odlite; furthermore, the inclusions conform to the contour of the ghost grain in the center of the crystal. Other features also indicate that Sept. 15, 1936 PROCEEDINGS: GEOLOGICAL SOCIETY 393 the investment is secondary. The investment appears to have formed very slowly and that the silica passed through the walls of the odlite grains after the walls had reached a solid state and probably after the spaces between the grains had been filled with matrix. No reason has been found for sup- posing that the secondary silica was added during Recent weathering. (Author’s abstract.) E. B. Eckert described odlites forming about gas bubbles in a swimming pool at Pinkerton Hot Springs, Colorado. Discussed by Messrs. FAnEy, GOLDMAN, TRASK, HENBEST. Program.—A. H. Koscumann: Structure of the Magdalena mining Dis- trict, New Mexico. Discussed by Messrs. Kine, HENpDRiIcKs, LOUGHLIN. G. H. Lovueuuin: Ore deposits of the Magdalena mining district, New Mezico. Discussed by Messrs. Park, McKnicut, WiLuiAMs, Kine, HEN- BEST, SHENON, GODDARD. 538TH MEETING The 538th meeting was held in the Assembly Hall of the Cosmos Club, March 11, 1936, President M. I. GoLDMAN presiding. Program: R. C. Cavy: Distribution of Thermal Springs in the United States. E. T. ALLEN: Thermal Springs: criteria of their origin and factors in their differentiation. All hot springs studied by the writer are regarded as mag- matic, (1) because they contain relatively large amounts of chemical ele- ments, especially carbon, sulphur and chlorine, found in rocks only in traces, though all are of common occurrence in hot fumeroles; (2) because their heat is readily accounted for as the result of rising steam, chief among the mag- matic gases, while dry rock is low in heat capacity and a poor conductor of heat. The theory that hot springs are bodies of circulating ground water, heated by magmatic steam and enriched by other magmatic emanations which leach from the adjacent rock important amounts of mineral matter, agrees well with a great mass of observed fact. That some thermal springs of low temperature and dilute character derive their mineral matter from rock and atmosphere is possible, but that they are magmatic springs in the last stage of development seems equally plausible. Differentiation of ther- mal springs into the distinctive types actually found in the field is held to be due (1) to contact of the hot waters with limestone in the case of traver- tine springs; and (2) where limestone is absent to topography on the one hand, and differences in the volatility of magmatic emanations and their products on the other. Topography controls water supply, and the depth to which water descends in hot ground should be limited by its volume. Thus the nature of the dissolved mineral matter in the waters and the deposits they form are accounted for. (Author’s abstract.) Discussed by Messrs. GILLULY, R. C. WELLS, SCHALLER, BURBANK, RuBEY, HENBEST, HEss, MeErnzerR, Trask, G. R. MANSFIELD, INGERSON. D. F. Hewett: The problem of the warm springs of Georgia. Discussed by Messrs. ALLEN, BowLEs, WENZEL, S. LoHMAN, CapDy, FIEDLER, HEND- RICKS, MEINZER. 539TH MEETING The 539th meeting was held in the Assembly Hall of the Cosmos Club, March 25, 1936, President M. I. GoLDMAN presiding. Informal communications.—_W. W. Rusey called attention to experiments on stream dynamics soon to be undertaken at the National Hydraulic Labo- ratory, Bureau of Standards. The effects of damming a graded stream will 394 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 26, NO. 9 be studied. The results of this investigation should be of considerable inter- est to geologists and physiographers. G. A. Coorrr described a forthcoming monograph on Ozarkian-Canadian Brachiopoda by E. O. Uuricu and G. A. CoopEr. Program.—F. E. IncERson: Late glacial and post-glacial history of western Newfoundland. H. G. Fercuson and 8. W. Mutisr: Jurassic thrust faults in west central Nevada. In the Hawthorne and Tonopah quadrangles in west-central Nevada a belt of thrust faulting is approximately coextensive with the area of Triassic sediments which outcrop in the northeastern part of the Haw- thorne and northwestern part of the Tonopah quadrangle. The Mesozoic sediments involved in the thrusting include a thick volcanic series of Middle Triassic age (Excelsior formation), overlaid unconformably by Upper Triassic (Luning and Gabbs formations) and Lower Jurassic sedi- ments (Sunrise and Dunlap formations). The oldest and thickest of these, the Luning formation, probably did not extend far to the south or east of its present outcrop. The youngest Lower Jurassic formation (Dunlap for- mation) consists dominantly of fanglomerate, conglomerate and sandstone and is unconformable on the older formations. Its principal distribution is in an irregular belt overlapping the eastern and southern contacts of the Luning and Excelsior formations. Granitic intrusives, regarded as outliers of the Sierra batholith and presumably of late Jurassic age, cut the Mesozoic sediments. Apparently the first folding resulted in the formation of an irregular trough in which the thickest sections of the Dunlap were deposited: Folding and thrusting continued during the period of Dunlap deposition. The thrusts are older than the granitic intrusives. The date of thrusting there- fore, may be placed as about the end of early Jurassic time (end of Lias). The thrust faults show a noticeable lack of continuity and are irregular in attitude and distribution. In parts of the area there is evidence of movement | in nearly opposite directions on different thrusts. As far as evidence is obtain- able the movement on individual thrusts was relatively small, possibly of the order of magnitude of one to three miles and without great stratigraphic displacement. Evidence of movement under light load is shown by the com- binations of thrust and tear fault characteristic of many of the thrusts and there is direct evidence that several of the thrusts moved over an erosion surface. A small complex area in the Pilot Mountains shows that locally folding went on contemporaneously with faulting and two of the surface thrusts were folded into overturned anticlines. The general relations of the structural features of the area are explained on the assumption that a basin-like area of incompetent Upper Triassic and Lower Jurassic sediments suffered compression contemporaneously with the deposition of the youngest member of the series. This compression, prin- cipally from northwest, was succeeded intermittently by eastward pressure exerted by the rising Paleozoic land mass bordering the Mesozoic area. Irregularities of the surface of Middle Triassic volcanics on which the later sediments were deposited may have been effective in localizing and causing minor irregularities of the thrusts and folds. The field work was carried out in part with the assistance of grants from the Geological Society of America. (Authors’ abstract.) G. TUNELL and G. A. Coopmr, Secretaries. Pajiosoreey. —The pene ‘Glypiostrobus. in ie ies Rotax PROWAT) 02000220 ts ee ZOoLoey. eee from the far north collected Gee apt: | Bartlett. Cuarues B. WILSON... - 0. eee eee eee eee ice. —A new polynemid fish collected in the Sador : Sarawak, by Dr. William T. Hornaday, with notes on of Polynemidae. GEORGE S. MYERS. oe eg ee PROCEEDINGS: GEOLOGICAL SOCIETY...............-. — ci. 26 OcroBER 15, 1936 No. 10 JOURNAL Seed, oes SSA OF THE ASHINGTON ACADEMY OF SCIENCES BOARD OF EDITORS F. G. BricK WEDDE Roxtanp W. Brown Esen H. Toore BUREAU OF STANDARDS U. S. GEOLOGICAL SURVEY BUREAU OF PLANT INDUSTRY ASSOCIATE EDITORS H. T. WENSEL Haroitp Morrison PHILOSOPHICAL SOCIETY ENTOMOLOGIOAL SOOIETY HE. A. GoLDMAN . W. W. Rusey BIOLOGICAL SOCIETY GEOLOGICAL SOCIETY AaGNEs CHASE J. R. SwANTON BOTANICAL SOCIETY ANTHROPOLOGICAL SOCIETY R. E. Grsson CHEMIOAL SOCIETY PUBLISHED MONTHLY BY THE WASHINGTON ACADEMY OF SCIENCES 450 Auwnarp Sr. AT MrenasHA, WISCONSIN Entered as second elass matter under the Act of August 24, 1912, at Menasha, Wis. Acceptance for mailing at a special rate of postage provided for in the Act of February 28, 1925. Authorized January 21, 1933. Journal of the Washington Academy of Sciences This JouRNAL, the official organ of the Washington Academy of Sciences, publishes: (1) short original papers, written or communicated by members of the Academy; (2) — proceedings and programs of meetings of the Academy and affiliated societies; (3) notes of events connected with the scientific life of Washington. The JourNaAt is issued ea monthly, on the fifteenth of each month. Volumes correspond to calendar years. Manuscripts may be sent to any member of the Board of Editors. It is urgently re-_ me quested that contributors consult the latest numbers of the JouRNAL and conform their manuscripts to the usage found there as regards arrangement of title, subheads, syn- | onymies, footnotes, tables, bibliography, legends for illustrations, and other matter. Manuscripts should be typewritten, double-spaced, on good paper. Footnotes should be numbered serially in pencil and submitted on a separate sheet. The editors do not — assume responsibility for the ideas expressed by the author, nor can they undertake to correct other than obvious minor errors. Illustrations in excess of one full-page halftone or two full-page line drawings may . be paid for by the author. Line drawings are preferred. Proof —In order to facilitate prompt publication one proof will generally be sent to authors in or near Washington. It is urged that manuscript be submitted in final © form; the editors will exercise due care in seeing that copy is followed. Author's Reprints.—Fifty reprints without covers will be furnished gratis. Coutts bearing the name of the author and title of the article, with inclusive pagination and date of issue, and reprints, will be furnished at cost when ordered, in accordance with the following schedule of prices. Copies 4 pp. 8 pp. 12 pp. 16 pp. | Covers | eer eve ate aks be $2.00 . 100 Bry Seah $1.20 $1.50 $2.00 $2.75 150° = 1.25 1.90 2.55 3.25 3.50 200 1.76: 2.60 3.60 4.50 4.25 250 2 .00- 3.30 4.65 5.75 5.00 Envelopes for mailing reprints with the author’s name and address printed in the ‘ corner may be obtained at the following prices. First 100, $4.00; additional 100, $1.00: Subscription Rates.—Per volume........... AES a eee Se Rtg > $6.00 To members of affiliated societies; per volume.................... (eee ice 2.50 © Single numberseiee i cs sel as CAS Cea Slt © Wh eae each ee aie Remittances should be made payable to “Washington Academy of Sciences” and addressed to 350 Ahnaip Street, Menasha, Wis., or to the ee Gls Boe. Avers, Coast and Geodetic Survey, Washington, D.C Exchanges.—The JouRNAL does not exchange with other publications. Missing Numbers will be replaced without charge provided that claim is made to ‘ the Treasurer within thirty days after date of following issue. OFFICERS OF THE ACADEMY President: O. E. Mrtnzir, U. 8. Geological Survey. Corresponding Secretary: N. R. Smitu, Bureau of Plant Industry. Recording Secretary: CHARLES THOM, Bureau of Chemistry and Soils. Treasurer: Hpnry G. Avers, Coast and Geodetic Survey. i get JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VoL. 26 OcToBER 15, 1936 No. 10 PALEONTOLOGY.—A new species of ‘“‘Crassatellites” from the upper Miocene of Florida.1 W.C. MANSFIELD, U. 8. Geological Survey. “‘Crassatellites’”? (Hybolophus?) leonensis Mansfield Crassatellites (Crassatellites) gibbesit Mansfield, Florida State Geol. Survey. Bull. 8, pp. 80, 81, pl. 16, fig. 9, 19382. (Not “‘Crassatellites”’ gibbesit Tuomey and Holmes, 1856.) In the above publication, I identified specimens from the Ecphora zone and the Cancellaria zone (upper Miocene) of Florida as Crassatellites gibbesii (Tuomey and Holmes), a species described from the Waccamaw formation of South Carolina, but stated therewith that the Florida Miocene form was more elongate than the specimens figured by Tuomey and Holmes under their species, C. gibbeszz. After collecting more specimens of C. gibbesiz from the Waccamaw formation and the Caloosahatchee marl, I now believe that the upper Miocene form should not be specifically united with the Pliocene form and here refer the Miocene form to a new species ‘‘Crassatel- lites’ leonensis. < The adult shell of C. leonensis differs from C. gibbesii in having a much more elongate shell and coarser undulations over the beak. The concentric sculpture on the body of the Miocene shell is also more closely spaced than on the Pliocene shell. The Miocene species probably is a precursor of the Pliocene species. ’ Dimensions of the holotype (U.S.N.M. 371160).—Length 68 mm, height 47 mm. Type locality Borrow pit, Jackson Bluff, Leon County, Fla. Outside occurrence.—Two immature specimens collected at Raysors Bridge, Colleton County, 8. C., probably belong to the new species here described. The new species probably does not occur at the Natural Well (Duplin marl) of North Carolina—the species C. undulatus Say occurring here; nor at Wil- mington, N. C., the Pliocene species, C’. gibbesii occurring here. ‘‘Crassatel- lites’ densus Dall, a species described from the Oak Grove sand of Florida, has a more prominent posterior dorsal ridge and a smoother area over the middle of the shell than the new species here described. 1 Published by permission of the Director, U. S. Geological Survey, Washington, D.C. Received June 11, 1936. 399 396 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 26, No. 10 BOTANY.—WNew species of Arundinaria from Southwestern China! Y. L. Kenc, Academia Sinica, National Research Institute of Biology, Nanking, China. (Communicated by Agnes Chase.) Arundinaria violascens Keng, sp. nov. Culmus teres, internodio primo in specimine circ. 14 cm longo, 4.5 mm crasso, dense fasciculato-ramosus, ramulis fere omnibus racemos termi- nales gerentibus; vaginae ad ramulum ultimum 4-6, glabrae vel saepe puberulae, pleraeque setis fuscis fimbriatae; ligula 0.83-1 mm longa, trun- cata; laminae 2.5-5 cm longae, 3-5 mm latae, pleraeque involutae, tessel- latae, basi in petiolum brevissimum attenuatae, margine utroque scaberu- lae; racemus basi inclusus, 4-7 cm longus, 3—7-spiculatus, pedicellis erectis, laevibus, 4-14 mm longis; spiculae 5—9-florae, 2.7-4 cm longae, fusco-pur- pureae; glumae acuminatae, superne puberulae vel interdum glabrae, prima 5-7 mm longa, 3—5 nerva, secunda 7-11 mm longa, 7—9-nerva; lemmata ob- longo-lanceolata, acuminata vel in acumen producta, omnino puberulo- scaberula, 9-nervia, venulis transversis reticulata, infimo 12-15 mm longo, callo pilis circ. 1 mm longis dense barbato; palea 9-10 mm longa, bidentata, puberula, ad carinas versus apicem rigide ciliata; lodiculae 3, hyalinae, ovatae, 2-2.5 mm longae, fimbriatae, inferne nervosae; antherae 3, 5-6 mm longae, flavidae; ovarium fusiforme, circ. 2 mm longum, apice in stylos 3 brevissimos divisum, stigmatibus plumosis, circ. 3 mm longis; rhachillae articuli 4 mm longi, versus apicem incrassati pubescensque. Type in the Herbarium of the U. 8S. National Herbarium, no. 1214328, collected on mountains south of Likiang, near Hochin and Chiuho, Yunnan Province, May 25-28, 1922, by J. F. Rock (no. 4082). This species is probably related to A. Forrestii Keng (Yunnan: Forrest 14127 in Herb. Kew), from which it differs, however, in having smaller florets, shorter anthers, and shorter but stouter rachilla-joints. Arundinaria parvifolia Hack., sp. nov. Culmus (cujus pars superior floriferus tantum adest) teres, tenuissimus, internodiis in specimine 25-33 mm longis, circ. 0.75 mm crassis, ramis soli- tariis, simplicibus, 8-15 cm longis, racemos terminales gerentibus; vaginae ad ramum floriferum 6—9, imbricatae, 1-2 cm longae, glabrae; ligula 0.3— 0.5 mm longa, truncata; laminae (in vagina suprema) circ. 18 mm longae, 3 mm latae, nervo secundario utrinque 1, reticulato-venulosae, tenuiter acuminatae, basi breve angustae, margine altero scaberulae, altero cartila- gineo-laeves; racemi basi inclusi, 4-5 cm longi, 5—7-spiculati, pedicellis erectis, laevibus, 1-5 mm longis, infimo spatha cire. 1 cm longa subtento; spiculae 4—6-florae, 15-25 mm longae, viridulae; glumae acuminatae vel in acumen productae, tenues, stramineae, 3—5-nerves, prima 5—6 mm, secunda 6-7 mm longa; lemmata oblongo-lanceolata, acuminata, 7—nervia, trabe- culosa, praeter callum pubescens glabra, primo 8-9 mm longo; palea biden- tata, lemma aequans vel paulo superans, praeter carinas saepe 6-nervis, laevis vel ad carinas versus apicem scaberula; lodiculae 3, ovatae, 2.56 mm longae, superne ciliatae; antherae 3, 5 mm longae, pallidae; stigmata 2, circ. 2 mm longa, tenuissime plumosa; rhachillae articuli tenues, glabri, 3-4 mm longi. 1 Received May 27, 1936. Ooer. 15, 1936 DRECHSLER: DACTYLELLA 397 Type in the U. 8. National Herbarium, no. 1126301, collected in Yunnan Province but without precise locality, 1910, by E. E. Maire (no. 7532). This species is characterized by the glabrous rachilla-joints and the palea equaling or exceeding the lemma. Described from a single specimen ex herb. Vienna which bears the name Arundinaria parvifolia Hack. Arundinaria pauciflora Keng, sp. nov. Culmus in specimine circ. 40 cm longus, 3 mm crassus, internodiis in parte superiore teretibus, 4.5—-9.5 em longis, nudis, glabris, dense fascicu- lato-ramosus, ramis primariis usque ultra 20 cm longis et iterum ramulosis, racemos terminales plerisque gerentibus; vaginae ad ramulum floriferum ultimum pleraeque 3-6, imbricatae, glabrae vel ad collum et prope margines pubescentes, apice setis fuscis scaberulis saepe fimbriatae; ligula cire. 0.5 mm longa, truncata; laminae 1—3 cm longae, 3-6 mm latae, acuminatae, basiin petiolum brevissimum attenuatae, firmae, tessellatae, subtus glaucae, supra pubescentes, nervis secundariis utrinque 2-3, margine altero scabrae, altero fere cartilagineo-laeves; racemus inclusus vel demum breve exsertus, 2-3 cm longus, plerusque 3-spiculatus, pedicellis erectis, laevibus, 2-4 mm longis, bractea glumacea 2-3 mm longa saepe subtentis; spiculae 4—5- florae, 16-21 mm longae, leviter purpureae; glumae inaequales, glabrae vel interdum versus apicem ciliolatae, prima ovata, acuta, 3-4 mm longa, 1-3- nerva, secunda abrupte acuta, 6—7.5 mm longa, 7—-9-nerva; lemmata ovato- lanceolata, acuminata, 7—9-nervia, reticulato-nervosa, glabra vel ad nervos puberula, infimo 8-12 mm longa, callo albo-pubescente; palea angusta, 7-8 mm longa, ad carinas superne ciliata; lodiculae 3, ovatae, 1.5-2 mm longae, fimbriatae; antherae 3, 5 mm longae, terminate exsertae; stigmata 2-3, plumosa, 2-3 mm longa; rhachillae articuli incrassati, 2.5-4 mm longi, margine apiceque ciliati, dorso item adpresso-pubescentes. Type in the U. 8. National Herbarium, no. 1128976, collected at Shao- shan, Ningyuen region, Szechuan Province, altitude 2700 meters, April 15, 1914, by Handel-Mazzetti (no. 1365). This species is probably related to A. brevipaniculata Hand.-Mazz., but is distinguished by the racemes of a few spikelets. BOTAN Y.—A Fusarium-lIvke species of Dactylella capturing and con- suming testaceous rhizopods.!|_ CHARLES DRECHSLER, Division of Fruit and Vegetable Crops and Diseases, Bureau of Plant Indus- try. Agar plate cultures, started from diseased rootlets or other decay- ing plant materials that previously have been in prolonged contact with the soil, usually allow the gradual multiplication of various species of testaceous rhizopods, some of which often attain large numbers in the course of two or three weeks. Once a population of these animals has become established, it is in general less subject to rapid decline than the populations of animal types multiplying more 1 Received August 31, 1936. 398 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 26, No. 10 Scale-u O 5 1015 2025 Fig. 1—Dactylella passalopaga. For explanation, see opposite page. Oct. 15, 1936 DRECHSLER: DACTYLELLA 399 rapidly in agar cultures, including, for example, many species of nematodes and many of the smaller amoebae. This lesser tendency toward abrupt reduction in number of living individuals, appears in considerable measure to be due to infrequence of destruction by parasitic or predacious fungi. In view of the disasters often overtak- ing populations of amoebae from the development of members of the Zoopagaceae (4, 5, 7) it might be expected that shelled rhizopods, whose protoplasmic bodies are necessarily partly protruded for feed- ing and locomotion, would suffer serious depredations from the same group of conidial Phycomycetes. Yet only two members of the Zo- opagaceae, which will be described elsewhere, have so far been found to subsist on testaceous rhizopods. Difflugia globulosa Duj. and Trinema enchelys Ehrenb. were shown in an earlier paper (3) to be captured and consumed by Pedilospora dactylopaga Drechsl., a fungus referable to a quite different predacious series. This series, consisting of hyphomycetous forms belonging mostly to the genera Tricho- thecium, Arthrobotrys, Dactylaria and Dactylella (including Monacro- sporium), is destructive mainly to nematodes. The biological rela- tionship of P. dactylopaga would seem, therefore, to be a somewhat unusual one; as would also the similar relationship of an apparently allied fungus recently observed to subsist by the capture of two other species of shelled rhizopods. The scanty vegetative mycelium of the fungus in question might readily be overlooked in a mixture of microorganisms but for the conspicuous alignment of captured prey in two ranks, with oral ends directed toward one another (Fig. 1, A-D). Because of the large numbers in which it develops in old agar plate cultures, the smaller Fig. 1.—Dactylella passalopaga, drawn from material developed in mixed culture on maizemeal agar, with the aid of a camera lucida, at a uniform magnification; X 1000. A.—Portion of hypha on which fifteen specimens of Geococcus vulgaris, a—o, have been captured; the two slight protuberances p and q, possibly representing predacious modi- fications. B.—A short portion of hypha with two captured specimens of G. vulgaris; one, a, having been taken without positional disturbance of the filament; the other, b, having drawn the hypha partly into its mouth as in feeding. C.—Portion of hypha with three specimens of Huglypha laevis, each held by means of a lobed predacious gag; two of the animals, a and b, being shown mainly in optical section, the other, c, mainly in surface view. D.—Portion of hypha with (a) a captured specimen of EF. laevis, within which are shown assimilative hyphae arising from the lobes of the predacious gag; and (6) an enlargement representing probably an outgrowth from which a rhizopod man- aged to escape. E.—Portion of mycelium showing an old, partly evacuated, prostrate conidiophore, a, that has given rise to four younger erect conidiophores, three of which, b, cand e, have each produced a single terminal conidium, and the other, f, in addition, a second conidium on the oblique distal prolongation; all the conidiophores being shown denuded except c, whereon conidium d is shown in position. F.—Portion of hypha with a conidiophore, a, bearing a conidium, 6. G. H, I, J.—Detached conidia. L.—Two germinating conidia, a and b, the single germ tube from each having fused with a seg- ment of the other conidium. M.—A conidium showing early predacious development following partial ingestion by a specimen of G. vulgaris. 400 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 26, No. 10 rhizopod having a broadly ovoid, smooth test slightly thickened about the circular mouth, and otherwise corresponding well to Francé’s description (8) of his Geococcus vulgaris, provides an espe- cially striking display. Frequently the individual hyphae are beset for long stretches with the ovoid animals, here and there at such close intervals that the test of one will be in contact with that of its neigh- bor on either side (Fig. 1, A). A somewhat less abundant but still impressive display of predacious activity is furnished in the capture of an ovoid or often somewhat unsymmetrically bursiform species of Euglypha, measuring usually 35 to 50u in length and provided with aperture scales noticeably thickened at their slightly incurved apices (Fig. 1, C, D). In morphology the animal evidently corresponds most closely to the description of EL. laevis Perty as given in the ac- counts of Penard (13) and of Cash and Wailes (1); for though in some individuals a few rudimentary bristles may be present at the aboral end, a completely glabrous condition is by far more common. Examination of tracts of mycelial filament free of the two species of rhizopods mentioned, has not revealed any distinct structure to which a special function in the capture of prey could be assigned. Some slight protuberances to be found now and then on vegetative hyphae (Fig. 1, A, p,q), or even on germinating conidia (Fig. 1, K), may possibly represent predacious modifications, but their meager differentiation is hardly at all suggestive of any important role. The swollen parts, sessile or stalked (Fig. 1, D, b), that are occasionally seen and that in some degree resemble the well defined predacious organs of Dac- tylella tylopaga Drechsl. (6) and of Pedilospora dactylopaga, are very similar to the processes formed after prey has been engaged, and in all probability represent outgrowths from which animals have managed to escape. An absence of definite organs of capture previous to en- counter with prey, is unusual among the predacious Hyphomycetes, the only other example of such absence known in this series being found in the nematode-capturing species of Dactylella with quadrilo- bate conidia figured in another paper (2: Fig. 9, A, C). The necessity for special organs to initiate capture is apparently obviated by the feeding habits of the animals on which the fungus preys. As Geococcus vulgaris is a relatively small testaceous rhizopod with a proportionately small aperture, it might be expected that feeding would be restricted to objects like bacteria and the more minute of fungus spores. Such limitation, however, does not actually prevail, for often the animal obtains its nourishment from objects as formidable as the oospore of Pythium ultimum Trow, which not only Oct. 15, 1936 DRECHSLER: DACTYLELLA 401 exceeds it in size, but is surrounded, moreover, by the thick and thor- oughly substantial oospore wall. Oospores of P. ult¢emum and of many congeneric forms are, of course, consumed by other rhizopods flourish- ing in agar plate cultures: some of the larger species of Amoeba often enveloping a specimen until the durable wall is broken down and the protoplasmic material assimilated; while the robust testaceous Arcella vulgaris Ehrenb., with its capacious oral aperture, often ‘‘imports”’ one specimen after another, so that three or four oospores in various stages of digestion may be seen inside. Neither of these more usual modes of ingestion is followed by G. vulgaris, which instead, applies its mouth flush to the oospore wall, calks the zone of contact with a yellow secretion apparently identical with the substance closing up the test during periods of encystment, and gradually perforates the delimited portion of spore wall, probably by some sort of digestive action. Once communication is established with the interior of the oospore, the granular contents, now visibly degenerating, are drawn into the test of the animal,—the movement of material appearing much the same as in the sucking of an egg. When mycelium is attacked by Geococcus vulgaris the delay inci- dent to the resistance of the thick oospore wall is obviated, and the sucking action becomes evident while the filament is still intact. No exception is made of the hyphae belonging to the predacious fungus herein under consideration, as a filament of this species is often to be seen drawn into the mouth of an animal (Fig. 1, A, 7; B, b). To such undiscriminating voracity the fungus responds by rapidly proliferat- ing from the partly ingested portion a bulbous outgrowth slightly larger than the oral aperture, so that the rhizopod is securely held. Indeed, more generally, the fungus meets the animal half way, by putting forth the expanded outgrowth before suffering any physical change itself. In many instances the expanded part is nearly sessile on the filament (Fig. 1, A, a, b, c, f, g, h,j, k, 1, n, 0; B, a), but in other instances it is formed on a short branch (Fig. 1, A, d, e, m). The rangier connection apparently is brought about when the animal, after making contact with the filament, moves away, and is pursued thiough elongation of the outgrowth until the expanded part has at- tained a width making further movement impossible. The same sequence of events is followed also in the capture of Euglypha laevis. Because of the larger mouth of this rhizopod, a cor- respondingly bulkier gag is required to effect capture; the additional requirement being supplied through the proliferation of a number of expanded lobes (Fig. 1, C) in place cf the simple distended part. Like- 402 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 26, No. 10 wise, as might be expected, the assimilative hyphae thrust into the interior of the larger animal (Fig. 1, D, a) are longer and more numer- ous than those of the meager haustorial apparatus discernible, mostly with difficulty, in some specimens of Geococcus vulgaris. The exhaus- tion of materials from either rhizopod takes place without causing any sudden conspicuous change in appearance of the sarcode, the protoplasmic contents merely becoming more and more tenuous, much like the protoplasm of amoebae attacked by members of the Zoopagaceae, or like the contents of Amoeba verrucosa Ehrenb. at- tacked by Dactylella tylopaga. After a mycelium has been nourished for some time, erect conidio- phores arise singly (Fig. 1, F, a) or in small groups (Fig. 1, E, 5, c, e, f) from the rangy prostrate filaments. They show unmistakable simi- larity to the homologous structures of Pedilospora dactylopaga, though noticeably stouter and shorter in stature. When growing in mixed cul- ture in the presence of bacteria, the conidiophore usually concludes its development with the production of a single rather massive terminal conidium (Fig. 1, E, d; F, b). In some instances, however, it grows out from below the attachment of the first spore to produce a second farther on (Fig. 1, E, f); or, fairly often, it gives rise through lateral branching to one or more secondary conidiophores, under the increasing weight of which it is pressed down into a prostrate position. (Fig. 1, E, a). Though the conidium in its septation and elongate- fusoid shape bears a suggestive resemblance to the conidia of some species of Fusarium, it lacks the distinctive basal modification fre- quent in that genus, and is never borne on a sporodochium (Fig. 1, G-—J). Germination takes place readily, by the production usually of two germ tubes, one from each end (Fig. 1, K). Anastomoses of germ tubes with detached conidia (Fig. 1, L, a, b) or with mycelial fila- ments, and, indeed, vegetative fusions generally, are frequent in this species, as in other members of the predacious series of Hyphomy- cetes. Occasionally newly detached conidia are partly ingested by Geococcus vulgaris (Fig. 1, M), so that at times vegetative germina- tion may be preceded by predacious development. The morphology of its conidial apparatus makes the fungus clearly eligible for inclusion in the genus Dactylella Grove. Of the several forms described in this genus, D. minuta var. fusiformis Grove (9) offers apparently the closest resemblance, though the greater diameter (7 to 9u) and more abundant septation (9 to 12) ascribed to the conidia of the British species, would seem to exclude any strong like- lihood of identity. Similarity in varying degree is recognizable also Oct. 15, 1936 DRECHSLER: DACTYLELLA 403 when comparison is extended to the several species with narrow conidia that have been compiled in the genus Monacrosporium Oud. Thus M. subtile, as described and figured by Oudemans (12), ap- proaches the predacious form under consideration in diameter of conidium, but presents fairly decisive differences in the clavate shape and more abundant septation of that structure. The pronouncedly clavate shape illustrated in Harz’s original publication (10), together with the inferior length (35 to 38u) and less abundant (3 to 5) septa- tion, also sufficiently distinguishes the conidia of M. sarcopodioides (Harz) Berl. et Volg. On the other hand, the conidia of M. oxysporum Sacc. and March. (11) are described as being symmetrically spindle- shaped; yet their very acutely pointed ends emphasized in Marchal’s drawings, their greater length (96 to 105u) and width (9 to 10.5), and their more frequent (10 to 12) septation, not to mention the greater length (120 to 170u) and width (4 to 5u) of the supporting conidiophores, constitute details not reconcilable with the asexual re- productive apparatus found in my cultures. The fungus so curiously adapted to prey on small-mouthed testa- ceous rhizopods is believed, therefore, to represent a new species, for which a name having reference to the gaglike predacious structure would seem appropriate. Dactylella passalopaga sp. nov. Mycelium sparsum, repens, parce ramosum; hyphis sterilibus 1.2—2.7u crassis, hyalinis, mediocriter septatis, ex latere in ores animalium ramulum trudentibus qui intus tumet et sic animalia tenet; hyphis fertilibus paucis, solitariis vel parum aggregatis, hyalinis, plus minusve erectis, saepe 2—5- septatis, 40—110y altis, basi 2.8-4y crassis, sursum attenuatis, apice 1.2—1.8u crassis, unicum conidium vel interdum secundum post incrementum fer- entibus. Conidia hyalina, elongato-fusoidea, utrimque obtuse rotundata, 60—80u longa, nee crassa, 6—8-septata. Habitat in radicibus plantarum putrescentibus, in terra, saepe in humo silvarum, Geococcum vulgarem et Euglypham laevem capiens et consumens, prope Beltsville, Maryland. Mycelium sparse, creeping, rather scantily branched; the vegetative hyphae 1.2 to 2.7u wide, hyaline, septate at moderate distances, capturing some species of small-mouthed testaceous rhizopods by thrusting into the mouth of the individual animal a short lateral branch that expands inside to form a simple or lobed enlargement wider than the aperture; conidio- phores usually few in number, scattered singly or in small groups, hyaline, more or less erect, mostly 2 to 5 (average 3.3) times septate, 40 to 110u (average 69u) in height, 2.8 to 4u (average 3.3u) wide at the base, tapering upward to an apical diameter of 1.2 to 1.8u, bearing usually a single conidium and sometimes a second one on a distal prolongation originating immediately below the base of the first. Conidia hyaline, elongated spindle-shaped, 404 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES’ VOL. 26, No. 10 bluntly rounded at both ends, 60 to 80y (average 69x) long, 4.5 to 6u wide, and divided usually by 6 to 8 (average 7.7) cross-walls. Capturing and consuming Geococcus vulgaris and Huglypha laevis, it occurs in decaying plant roots, in soil, and especially often in leaf mold, near Belts- ville, Maryland. LITERATURE CITED 1. Casu, J., and WarLsEs, C. H. The British fresh-water Rhizopoda and Heliozoa. 5 vols. London, 1905-1921. 2. DrecusuER, C. Morphological diversity among fungi capturing and destroying nema- todes. This JOURNAL 23: 128-141. 1933. 3. Drecusuir, C. Pedilospora dactylopaga n. sp., a fungus capturing and consuming testaceous rhizopods. This JoURNAL 24: 395-402. 1934. 4. DrecusteR, C. Some conidial Phycomycetes destructive to terricolous amoebae. Mycologia 27: 6-40. 1935. 5. Drecusurr, C. Some non-catenulate conidial Phycomycetes preying on terricolous amoebae. Mycologia 27: 176-205. 1935. 6. DrecHsLER, C. A new mucedinaceous fungus capturing and consuming Amoeba verrucosa. Mycologia 27: 216-223. 1935. 7. DrecusuEr, C. New conidial Phycomycetes destructive to terricolous amoebae. Myco- logia 28: 363-389. 1936. 8. Francn, R. H. Das Edaphon. 99 p. Munich, 1913. 9. Grove, W. B. New or noteworthy fungi: Part III. Jour. Bot. 24: 129-137, 197-206. 1886. 10. Harz, C. O. Einige neue Hyphomyceten Berlin’s und Wien’s nebst Beitragen zur Systematik derselben. Bull. Soc. Imp. Nat. Moscou 44: 88-147. 1871. 11. Marcnwat, E. Champignons coprophiles de Belgique. Bull. Soc. Roy. Bot. Belgique 24: 57-77. 1885. 12. OupEMans, C. A. J. A. Aanwinsten voor de flora mycologica van Nederland, IX en X. Nederl. Kruidk. Arch. II 4: 203-278. 1885. 13. Pmenarp, E. Faune rhizopodique du bassin du Léman. 714 p. Geneva, 1902. ZOOLOGY .—New millipeds of the American family Striariidae.t H. F. Loomis, U. 8. Plant Introduction Garden, Coconut Grove, Florida. (Communicated by O. F. Cook.) The milliped family Striariidae is composed of a single genus con- taining species found only in the United States. The first species was described in 1888 by C. H. Bollman, who recognized it as the type of the genus Striaria, later designating a new sukfamily for its ac- commodation. In 1895 O. F. Cook elevated the subfamily to full family rank, and in 1896 gave it position as a suborder of the Coelo- cheta, on the same footing as the Lysiopetaloidea and the Chordeu- moidea, like the Chordeumoidea in the short body of 30 segments, and like the Lysiopetaloidea in having the segments longitudinally carinate, but contrasting with both these groups in the slow move- ments and heavily armored structure, with the head reduced and protected by the greatly expanded first segment, the lower carinae strongly developed, and the last segment broadly three-lobed, afford- ing protection when the animals are coiled, as in several specialized families of the order Merocheta. 1 Received June 18, 1936. Oct. 15, 1936 LOOMIS: NEW MILLIPEDS 405 In 1899 Cook formulated descriptions of the order, its three sub- divisions, and of the family, genus and type species, Strzarzva granulosa Bollman, and also added two new species, one from the District. of Columbia, while the other, based on a female specimen, extended the distribution of the group to California. The next and last addition to the genus was made by R. V. Chamberlin, in 1910, of another species, from Portland, Oregon, also based on a female specimen. In the material which led to the preparation of the present paper were two new species, both from California, giving a preponderance v Ve VU ae 8 Cernen Uys ay Ue v Vy Fig. 1.—Structural features of Striaria. A. S. nana, last segment. B. S. nana, lateral view of apical two-thirds of anterior portion of a gonopod. C. S. nana, left posterior gonopod from behind. D. S. imberbis, second leg of male. E. S. californica, labrum of male. F. S. californica, lateral view of apex of anterior gonopod. to the Pacific Coast. Also there were many specimens of S. californica Cook, from the males of which it has been possible to amplify the original description of the species. The characters shown by the new species, and additional speci- mens of some of those previously known, force the modification of several statements which have been made pertaining to the characters of the genus Striaria or higher classification units. Dorsal setae, in the same number and position as in the Chordeumoidea, were found in four species examined and are inferred to be present in those not seen. Only in specimens of S. nana were these setae found in com- plete series, but careful inspection of examples of the other species us- 406 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 26, No. 10 ually showed a few setae still remaining on some of the segments, especially those near the ends of the body, although most of the setae had been broken off. The disposition of these setae is as follows: On the first segment six setae are found a little back of the front margin, in front of the dorsal crests. The ensuing segments also have six dorsal setae, three on each side of the dorsum, the first located be- tween ridges 1 and 2, the second between ridges 3 and 4, and the third between ridges 5 and 6. Segment one and the last segment are without a median ridge, while the other segments have a pronounced ridge on the anterior subsegment but none on the posterior subsegment, its place being occupied by a fine distinct groove. The shape of the male labrum is reduced in value to a specific character with the dis- covery that in at least one species, and possibly another, the lateral corners are not developed into spines. In one species the pregenital legs of the males are not conspicuously heavier than the other legs. Characters for the recognition of the six members of the genus are given in the following key. KEY TO THE SPECIES OF STRIARIA Size small, not over 7 mm long; ventral crests beyond segment 12 or 13 re- dueed or obsolete: oe ie cee eee nana N. sp. Size larger, at least 10 mm long; ventral crests present on all but two or three of thelast segments... 00 0.4)0..6 2.) 49 e245 seo a eee First segment with 12 dorsal crests or ridges......... nazinta Chamberlin First segment with 10 dorsal ‘crests... 2.00 2. 5. ee Body strongly flattened, especially in front; labrum similar in both sexes; second male legs with a very large lobe projecting from beneath the Second jolts: 312: Meola ae ree aah eee imberbis n. sp. Body cylindrical or only slightly flattened; labrum of males with lateral corners produced into long spines; second male legs with lobes small or WANTING 5 oh. cn desc 8 Ree ol) Wien Se te er Body reaching 15 mm in length; males with the anterior margin of the la- brum nearly straight between the median teeth and the produced spine on. .erther7side.. 04:5 ssc eee ee ea ee californica Cook Body less than 12 mm long; male labrum with anterior margin strongly con- vex between the median teeth and the lateral spines................ Dorsal crests very coarse; lobes of the last segment separated by alistingte open sinuses; apex ie anterior gonopods terminating in about six short Spinesor teeta. . yb) ee So es ce ere eee granulosa Bollman Dorsal crests finer; lobes of the last segment separated by shallow notches rather than by deep and slender sinuses; apex of anterior gonopods terminating in about a dozen sharp teeth....... ...columbiana Cook Oct. 15, 1936 LOOMIS: NEW MILLIPEDS 407 Striaria nana n. sp. Six mature specimens, including the male type, and several young, col- lected at Altamont Pass, above Niles, Calif., December 1, 1926, by O. F. Cook, who found a single female south of Pescadero, Calif., February 21, 1929. Male type in U.S. National Museum. Diagnosis: The small size of the body; presence of all or nearly all dorsal setae; the lack of ventral crests on the posterior half of the body; and the shape of the gonopods distinguish this species. Description: Body small, the dorsum somewhat depressed on the first few segments, but cylindrical thereafter; dorsal setae generally all present; length 6 to 7 mm, width .7 to .8 mm. Head with not over 7 ocelli each side. Labrum of male broad, outer cor- ners sharply angled but, unless broken from the two specimens examined, without spines. First segment with 10 dorsal crests; surface elsewhere finely and densely granular as are the ensuing segments, including their lateral surfaces. Median ridge of the anterior subsegments fine but well elevated, the depres- sion on either side of its posterior part much less extensive than in S. californica or S. wmberbis. Second segment with the surface bearing the ventral crest scarcely at all produced outward beyond the line of descent of the side, not forming a nearly horizontal projecting shoulder as in S. imberbis. Interval between the ventral crest and the lowest dorsal crest a little wider than the next interval above. Large ventral crests present on the segments in front of the middle of the body, the anterior corner of each crest slightly produced forward and bluntly rounded; on the posterior segments the ventral crests are reduced to small low elevations or are entirely lacking. Last segment with the lateral lobes smaller in proportion to the median lobe than in the other species, and all lobes more acute; dorsal surface covered with slender suberect tubercles (Fig. 1, A). Males with first pair of legs stouter than the second pair. Second legs lacking a conspicuous lobe on the under side of joint 2, the other joints also normal. Third legs with each coxa produced into an erect subcylindric lobe relatively as long as found in S. granulosa, the inner face glabrous and shining, the outer side finely hispid; remaining joints normal. Legs 4 to 7 with the four basal joints very greatly expanded horizontally but not thick- ened, their surface covered with short, stout, fusiform, almost scale-like hairs, in contrast to the common type of hairs found on the other legs. An- terior gonopods with the apex bilobed, not at all spinose (Fig. 1, B). Posterior gonopods with the distal joint more greatly produced inward than in the other species, the lobe thinner (Fig. 1, C). This small species was at first thought to represent a new genus because of the presence of dorsal setae in the same position as they occur in the Chordeumoidea, but careful inspection of other species of Striaria showed nearly all individuals to have a few dorsal setae still in evidence, although it was apparent that most of the setae had been broken off. Immature speci- mens of these species show the setae in more complete series. 408 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 26, No. 10 STRIARIA NAZINTA Chamberlin Ann. Ent. Soc. Amer., 3: 242-248. 1910. A single female, discovered at Portland, Oregon, is the only known speci- men. The presence of 12 dorsal ridges on segment | is the species’ most striking character, 10 ridges being the usual number in the other species, with oc- casionally an additional, small, rudimentary ridge on one or both sides of this segment in some specimens. Although the description states that the labrum has only two teeth, and the illustration shows this condition, it is doubtful if other specimens exhibit- ing this feature would be found, since a three-toothed labrum is an ordinal character of the Coelocheta. Striaria imberbis n. sp. A mature male and female and a number of young, the eldest of which have 26 segments, collected south of Atascadero, Calif., January 1, 1928, by O. F. Cook. Male type in U.S. National Museum. Diagnosis: This species is distinguished by the flattened dorsum; the similarity of the labrum in both sexes; the unusually prominent lateral pro- duction of each side of segment 2 supporting the ventral crest; and the very large lobe of the second joint of the second male legs. Description: Length of male 10 mm, female 12 mm, width 1 and 1.2 mm respectively. Head narrower in front than in the other species; labrum of the male simi- lar to that of the female, the lateral corners broadly rounded; ocelli 5 to 7 arranged in two irregular rows. Second segment with the lateral surface which supports the ventral crest much more strongly projecting than in the other species, the interval be- tween the marginal crest and the lowest dorsal crest double the width of the next interval above. Dorsum of the segments depressed throughout, but the anterior ones more strikingly so. Posterior subsegments relatively longer than in the other species. Dorsal crests high but much thinner than those of S. columbiana. Surface between the crests with numerous granules on the anterior segments but caudally the granules decreased in size and number; on the lateral surfaces only the anterior segments have a few gran- ules; however, the last segment is densely covered with fine granules. Ven- tral crests extending to the posterior end of the body, those in front with the anterior corner projecting forward as a large acute tooth. Median ridge very prominent on the anterior subsegments, especially those in front, the depression on either side of its back half is more extensive than that in S. californica. Last segment with the lobes very narrowly, although deeply, separated, the median lobe much the largest. Gonopods of the type not dissected but the apex of the anterior portion does not appear to be spinose; the shoe-like posterior joints are of the same general shape as in the other species. First legs of the male stouter than the second pair; the latter with a very large, bluntly conic lobe projecting out- ward and backward from the under side of the second joint, the posterior side and apex with long hairs (Fig. 1, D); other joints normal. Coxal lobes Oct. 15, 1936 LOOMIS: NEW MILLIPEDS 409 of the third male legs considerably shorter than in the other species of which males are known. Legs 4 to 7 of the male with none of the joints conspicu- ously: enlarged as compared with the post-genital legs, their surfaces in- vested with similar hairs. STRIARIA CALIFORNICA Cook Proc.-U.S. Nat. Mus., 21: 675. 1899. Specimens collected by O. F. Cook in the following California localities have been examined: Santa Cruz Mts., Jan. 2, 1928; Cordelia, Feb. 20, 1929; Davenport, Feb. 21, 1929. Several males are included which allow more tangible characters to be given than from the single female on which the species was founded. Body attaining a length of 15 mm, making the species the largest of the genus. Eyes with 7 to 9 ocelli. Males with the labrum produced at each corner into a long spine, the margin between the spine and the median teeth al- most straight in contrast with the strongly convex margin of the other species (Fig. 1, E). Segment 1 and a few segments succeeding it rather thickly and coarsely granular, the granules thereafter decreasing in size and number and, except those bearing the dorsal setae, none are evident on the segments immediately preceding the anal segment, the surface of which is densely but finely granular. Sides of body without granules after the first few segments. Anterior subsegments, particularly those at the front of the body, with a median ridge which is especially pronounced on the posterior part of each subsegment, and the surface on each side is depressed to the level of the posterior subsegment which has a fine median furrow but no ridge. Second segment with the lateral surface supporting the ventral crest scarcely at all produced outward to form a strong shoulder as in the other species; the in- terval between the ventral crest and the lowest dorsal crest little broader than the interval between crests 5 and 6. Ensuing segments, to near the posterior end of the body, with thick ventral crests, those on the anterior segments bluntly produced forward. Gonopods with the apex of the anterior portion more simple than in either eastern species (Fig. 1, F); shoe-like joint of the posterior portion thickest near its mesial extremity. The second male legs have joint 2 bearing a much smaller, more inconspicuous lobe on the under side than that in S. emberbis, but with longer apical hairs; joint 3 greatly swollen except at base; outer joints normal. The third male legs have the coxal lobes longer than in the other species, their tips exceeding the distal end of joint 3. Male legs 4 to 7 with the inner joints quite strongly crassate, the outer joints more normal. STRIARIA GRANULOSA Bollman Ann. N. Y. Acad. Sci., 4: 108. 1888. Locality : Tennessee. STRIARIA COLUMBIANA Cook Proc. U. S. Nat. Mus., 21: 674-675. 1899. Localities: District of Columbia and Maryland. 410 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 26, No. 10 ZOOLOGY.—Anguillulina askenasyi (Biitschli, 1873), a gall forming nematode parasite of the common fern moss, Thuidium delicatulum (L.) Hedw.: G. STEINER, Bureau of Plant Industry. There are quite a few records of the occurrence of nematode galls on mosses from Europe, but, as far as known to the writer, there are no such records from this country. All previous observations were Fig. 1.—Thuidium delicatulum (L.) Hedw. with galls caused by the nematode Anguillulina askenasyi (Butschli). A, stem and branches with terminal galls. x6. B, detail of two galls. x24. summarized in 1906 by Schiffner and further additions reported later by Horn (1909). Through Mr. J. L. Sheldon of Morgantown, West Virginia, a sample of Thuidium delicatulum with galls was received with the inquiry as to the name of the nematode species present. The moss had been collected in a wood across the road from the old Iron Furnace, near 1 Received June 5, 1936. Fig. 2.—Anguillulina askenasyi (Biitschli). A, female. 370. 8B, head end of female; amph, amphid; ppl, cephalic papilla; owt, outlet of dorsal esophageal gland. <1900. C, front view of head of male. X1900. D, striated lateral field in adult; str, striae. X1900. HH, proximal end of ovary; cap, cap cell. X730. F, egg. X250. G,. variations in terminus of tail. 330. Oct. 15, 1936 STEINER: GALL FORMING NEMATODE 411 pl _cnph 9 da jo it é / Hate rasta a | oa Fee Ve LAN Vee? ea) WA a ae {? Wisi ¢) a he oS ah Be For explanation of Fig. 2 see bottom of opposite page. 412 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 26, No. 10 Albright, Preston County, West Virginia, on September 21, 1935. Thuidium delicatulum is a member of the family Thuidiaceae (Musct) and is a species occurring in moist woods in Europe, Asia and on this continent from Labrador to the Rocky Mountains, south to the West Indies and South America. The galls, measuring .680—.716 mm by .868—.877 mm, are terminal galls, found on the stem as well as on the branches. They closely re- semble those described by Massalongo (1898) as caused by T'ylenchus (Anguillulina) in Zierva gulacea Schimp., and are of the artichoke type (see Fig. 1), having no side shoots or any resemblance to the witches broom type of galls as described from some other mosses, e.g. Dicranum. The leaves on the galls are enlarged to a size greater than the branch leaves or even the stem leaves. Their color is the same green as that of the plant. They are densely set on the oval gall, the outline of which may be recognized through them. From all appear- ances it is concluded that the galls render the branches and stems blind; they probably drop in a later stage. Each gall usually contains a small number of adults (7 to 12) with some 200-300 larvae and eggs. Although the moss had been dried for 8 months, the larvae and adults revived after a short period of soaking. Description of the nematode (Figs. 2-3). The present parasite markedly resembles Anguzllulina dipsaci (Kiihn, 1858) Gerv. & v. Ben., 1859, with which it was formerly synonymized in spite of the clear-cut statement by Butschli that it is different (J. Ritzema Bos and al.). The body, however, is much stouter than in the true A. dzpsaciz, although the present specimens were not quite as stout as those described by Biitschli under the name A. askenasyt (Bitschli, 1873) from the moss, Hypnum cupressiforme Hooker. Cuticle with rather fine annulations, lu wide; tail pointed, sharply in adults, much less in the larvae. Lateral fields about 1/5 of body diameter in width; larvae with two edging striae and a middle longitudinal stria; adult with about 10 much finer longitudinal striae; annulation interrupted on lateral fields. Head sharply set off, somewhat rounded, with rather plain submedial papillae. Framework light. Buccal stylet short, finer and smaller than in A. dipsaci. Terminal esophageal bulb distinctly set off from intestine, with esophageal glands still within the bulb. Intestine with large cells, seemingly in double alternate series. Rectum and anus very fine, often even obscure, especially in the adult. Excretory pore in adults ventrad of terminal eso- phageal bulb; in larvae ventrad just back of nerve-ring. Female apparatus with postvulvar sac-shaped uterus usually extending more than half the distance vulva-anus. Ovary most often straight, rarely reflexed, with termi- nal cap cell (Fig. 2, KE). Outlets of ovary well differentiated, with oviduct leading into receptaculum seminis; latter differentiated from uterus proper by distinct narrow duct. Eggs deposited unsegmented. Male testis straight, not reflexed. Spicula much like those of A. dipsaci, with lineate gubernacu- lum of about 1/3 spicula length. Bursa rises suddenly a short distance behind, or level with, proximal spicula end, in most specimens reaching only about halfway down the tail, rarely a little farther; in a male with a tail Oct. 15, 1936 STEINER: GALL FORMING NEMATODE 413 of 103, it extended to a distance of 64u back of the anus. A. askenasyi is considered a good species and is herewith reestablished. The original form was observed only once, namely by Biitschli, in Germany, on the moss, Hypnum cupressiforme, where, according to this author, it induces certain transformations in terminal buds, without, however, producing galls. It lives between the leaves of the buds of this moss. Although resembling dipsaci, it differs from it by being stouter, by having a longer esophagus, in \ \, \\. sex a \ \ SAN Fig. 3.—Anguillulina askenasyi (Biitschli). A, anterior end. 420. B and C, two different male tail ends; in C the bursa beginning farther back than in B. 420. D, proximal end of testis; cap, cap cells. X640. JZ, larva of .648 mm length; sex an, sexual anlage. X330. fF, terminus of larval tail. «640. both sexes a longer tail, an intestine of brownish color, and a shorter bursa in the male which reaches only about halfway, or slightly farther, down the tail. In addition the vulva has a moreanterior position. Adults of the pres- ent species usually take a ‘“‘more or less rolled-up”’ spiral position. Gall forming nematodes in mosses in the past have been referred to Angutllulina davainii (Bastian, 1865); whether correctly or not cannot be stated at present, the published descriptions being rather incomplete. Measurements of specimens from Thuidium delicatulum: 2 (n 10) total length = .980-1.2 mm; a=23-31, @=8-11.5, y=8.5-12.8, »=73%-78%. 414. JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 26, No. 10 Distance vulva-anus=about twice length of tail; eggs 36-41u 80-100,; stylet 8-10u. o(n 5) total length = .92-1.2 mm; a=24-31, B=8.1-10.7, y =8.5-14.5; spicula about 304; gubernaculum 8—10y; stylet 8-10y. Measurements as given by Biitschli for the type form: 9 =1.7 mm; a=19, B=10, y=13; v=80%; eggs =50u X95y; stylet =13u. 7 =1.4 mm; a=20, B=10, y=12. Emended diagnosis of A. askenasyi: Resembling A. dzpsaci but stouter, vulva more cephalad, bursa shorter, reaching only half, or slightly more than halfway, down the tail; eggs much thicker, deposited unsegmented ; intestine of brownish color; female usually assuming shape of a sickle or a more or less rolled-up spiral. Type host: Hypnum cupressiforme Hooker. Type locality: Feldberg, Taunus (Germany). In view of the difference in symptoms as caused by Biitschli’s type form (ectoparasite, not producing galls) and the present form (entoparasite, pro- ducing galls), it is thought best to consider the two as different host varieties, a conception also supported by slight differences in dimensions. LITERATURE CITED Bastian, CHartton H. Monograph on the Anguillulidae, or free nematoids, marine, land, and freshwater, with descriptions of 100 new species. Trans. Linn. Soc. XXV: 73-179 Pls. 9-13. 1865. Bos, J. RITZEMA. L’anguillule de la tige (Tylenchus devastatrix Ktihn) et les maladies des plantes dues a ce Nematode. Arch. Musee Teyler Ser. II, Vol. III, 2° partie: 161-348, pl. 1-10. 1888. Bitscuui, O. Beitrage zur Kenntniss der freilebenden Nematoden. Nova Acta Ksl. Leop.—Carol. Deutsch. Akad. Naturforscher 36(5): 1-144, pl. 1-11. 1873. Horn-WarEN, Pauu. Beitrag zur Kenntniss der moos bewohnenden Tylenchus-Arten. Arch. Ver. der Freunde d. Naturgesch. in Mecklenburg. 63: 67-77. 1909. Massatoneo, C. Nuovo Elmintocecidio scoperto sulla Zieriajulacea Schimp. Rivista Patologia Vegetale 7(1): 87-89. pl. IV. 1898. SCHIFFNER, Victor. Neue Mitteilungen tiber Nematoden-Gallen. Hedwigia 45: 159— 172. 1906. ZOOLOGY .—The histology of nemic esophagi. VII. The esophagus of Leidynema appendiculatum (Lezdy, 1850.) B. G. Cuirwoop, Bureau of Animal Industry, and M. B. Currwoop. This paper is the seventh of a series dealing with the histology of nemic esophagi. In the earlier papers (1934-1936) the esophagi of Rhabdias eustreptos, Oesophagostomum dentatum, Heterakis gallinae Metastrongylus elongatus, Rhabditis terricola and Rhabditis lambdiensis have been described. The nomenclature used in this paper was ex- plained in the first paper of the series. GROSS MORPHOLOGY The length of the esophagus of Lezdynema appendiculatum usually ranges from 396 to 448. in the gravid female, but in the gravid female upon which subsequent measurements are based, it is 364u long. It consists of a eylindri- 1 Received March 13, 1935. Oct. 15, 1936 CHITWOOD AND CHITWOOD: NEMIC ESOPHAGI 415 cal precorpus, 126u long, a subcylindrical postcorpus, 108 long, a very short narrow isthmus, 36yu long, and a pyriform bulb, 64u long. The lumen is triradiate; in the corpus and isthmus each radius terminates in an incom- plete subcylindrical ‘‘tube.”’ The ‘‘tubes”’ are very wide in the precorpus (Fig. 2a), very narrow in the postcorpus (Fig. 2c), and scarcely distinguish- able in the isthmus. In the bulb region the sides of the radii converge; in this region the lumen is, of course, modified by the valves. Returning to the lumen of the corpus, we find that in the precorpus the central part of the lumen is open, subtriangular with thickened walls (Fig. 2a—b), while in the postcorpus and isthmus the lumen is closed, not subtriangular and the walls not thickened (Figs. 1b and 2c-—d). Fig. 1.—Cross section of esophagus. a, Region of precorpus in adult; b, region of postcorpus in adult. ; The esophagus of the female fourth-stage larva in the specimen studied is 218u long and consists of the same parts as in the adult, i.e., precorpus, postcorpus, isthmus and bulb, but the postcorpus is not as wide and the isthmus is very slightly longer. In addition we find that the ‘‘tubes’’ at the ends of the esophageal radii are not as large, the lumen of the precorpus not subtriangular, and its walls not thickened as in the adult. NUCLEAR DISTRIBUTION The distribution of nuclei in the esophagus differs slightly in the fourth- stage larva and in the adult. In general, the nuclei are closer together in the larva than in the adult, which is due to the esophagus being shorter; a few of the nuclei differ also in relative position. The nuclei are identical in both stages and for that reason the esophagus of the adult is described, differ- ences in relative position being noted as they occur. Precorpus.—There are 25 nuclei in the precorpus as follows: 6 radial nuclei (r1-¢) and 19 nerve cell nuclei (ni_19). The radial nuclei are situated near the base of the precorpus, and arranged in a single group (riz), 1 nucleus on each side of each sector. The nerve cell nuclei are arranged in groups of 3, one in the center of each sector. The first group (ni_3) is situated near the anterior end of the precorpus, the second, third, and fourth groups, (nas, n7_9, and Nio-12), respectively, follow one another rather closely in series. These are followed by the fifth group (ni3_1;) which immediately precedes the radial nuclei of the precorpus (ri_s) and the sixth group (nisg_1s) which is situated 416 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 26, No. 10 at about the same level as the radial nuclei. The nuclei nj3 and nig may be identical but in the majority of sections they appear to be separate. In addition to these nuclei there is a single dorsal nerve cell nucleus (nig) situ- ated posterior to ni7. In the fourth-stage larva the last 2 subventral nerve cell nuclei are situated at the level of nig instead of at the level of ni,. The radial nuclei (ri_¢) are situated near the level of nio_15 instead of ni3_18. Postcorpus.—There are 22 nuclei in the postcorpus as follows: 3 bilobed marginal nuclei (mi_3a ana b), 6 radial nuclei (re_i2), and 13 nerve cell nuclei (Neo_32). The radial nuclei (r712) are situated anterior to the middle of the postcorpus and arranged in the same manner as those (ris) in the Su ly Uo ro) LU we Bey Fie On, as 0 Nay ey Fig. 2.—Diagrammatic representation of groups of nuclei of esophagus. a-—b, Pre- corpus; c—d, postcorpus; e, prevalvar region; f—g, valvar and postvalvar regions. precorpus. The marginal nuclei (mi_3) are situated at approximately the same level, 1 lobe of a nucleus being on each side of each esophageal radius (Fig. 1b, mi, and my). The nerve cell nuclei (neo_32) are arranged symmetrically in the subventral sector but not in the dorsal sector and for that reason they will be located according to sector. The dorsal sector contains 3 nuclei (No4,27,30) the first (na) being at the level of the radial nuclei (1712), the second (ne7) some distance posterior to the radial nuclei, and the third (ngo) at the base of the postcorpus. Each subventral sector contains 5 nerve cell nuclei (n2o_21,22-93,25~96,28-29,31-32). Ihe first pair (N2o_21) is situated slightly anterior to the level of the radial nuclei, near the lumen of the esophagus, and the second pair (nz2»3) is situated at the same level or slightly posterior to the first pair and nearer the external surface of the esophagus. The third subventral pair (e525) is situated Just anterior to the level of ne7; the fourth pair (neg_29) is situated posterior to the level of Ne7, being near the external surface of the esophagus; the fifth pair (nsi_32) is at approximately the same level as, or slightly posterior to, the fourth group (Nes_e9). The nerve cells in the fourth-stage larva differ in that neo_23 are situated at the level of r712, Nes_o5 at the level of ne7, and ngi32 at the level of ngo. Oct. 15, 1936 CHITWOOD AND CHITWOOD: NEMIC ESOPHAGI 417 Isthmus.—The isthmus contains no nuclei. Prevalvar and valvar region.—This region contains 20 nuclei as follows: 9 radial nuclei (13.1), 3 marginal nuclei (mss), and 8 nerve cell nuclei (n3340). The radial nuclei are arranged in 2 groups, an anterior group of 6 nuclei (131s) arranged as are those in the corpus (r1_12),and a posterior group of 3 nuclei (19-21), 1 near the center of each sector. The nerve cells are ar- ranged as follows: n33 in the middle of the dorsal sector just posterior to the level of Tizanais} N3¢ana35 Near the middle of each subventral sector at the same level as n33; N3s_37 are Situated just posterior to n33, 1 on each side of the © AGS 2 Hoe o @ My Mg Az My Mz The Myg Mog Yo7 Myo eeacos §§ gd Ma Ns Mg My My5Myq Moy Mya Mog Mog Nao Fig. 3.—Nuclei of esophagus. Marginal, radial, gland nuclei and nuclei of nerve cells. dorsal sector; nss is situated near the ventral radius in the right subventral sector at approximately the same level as re1; n39_49 are Situated posterior to T1417, respectively, 1 in the lateral part of each subventral sector. Postvalvar region.—The postvalvar region contains 10 nuclei as follows: 3 radial nuclei (12.1), 3 marginal nuclei (m;_9), 3 gland nuclei (gi_3), and 1 nerve cell nucleus (nz). The marginal nuclei (m;_9) are situated near the anterior end of the postvalvar region, 1 being posterior to the other 2. One is situated near each esophageal radius, the side at which each nucleus lies being inconstant. The radial nuclei (122s) are situated slightly posterior to the marginal nuclei, 1 near the lumen and in the center of each sector. The gland cell nuclei (gi_3) are similarly arranged except that they are posterior to the radials and near the external surface of the esophagus. The single nerve cell nucleus (n4) is situated beside the left subdorsal radius of the esophagus at about the level of the gland cell nuclei. THE ESOPHAGO-INTESTINAL VALVE The esophago-intestinal valve consists of an internal trilobed mass of tissue and an external circular layer of tissue. Within the trilobed mass are 418 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES’ VOL. 26, No. 10 2 nuclei, 1 dorsal (e1) and 1 ventral (e2). The number of nuclei in the external layer could not be determined with accuracy. Three nuclei, 1 left and 1 right dorsolateral (e3_4) and 1 right or left subventral (e;), were consistently found. Two additional dorsolateral nuclei (esx_7x) were found external to e3_4in one specimen. However, it is possible these nuclei might be those of intestinal cells. CHARACTER OF NUCLEI The radial nuclei all contain a moderately large karyosome. The nuclei of the precorpus (ri_s) are laterally compressed, and have a clear, lightly baso- philic nucleoplasm (Fig. la). The second group of radial nuclei are similar in shape to those of the first group but larger (Fig. 1b). The nucleoplasm is slightly less basophilic, the karyosome’‘larger than the karyosomes of ri, and a smaller plasmosome is often visible (Fig. 1b). The nuclei of the third group of radials (13-13) are similar to those of the second group but more or Fig. 4.—Nerve cells. less ovoid; the nucleoplasm is lightly basophilic, nearly homogenous and each also contains a moderate-sized karyosome and a smaller plasmosome (Fig. 3). The nuclei of the fourth group (ro_21) are rounded in shape, dis- tinctly larger than the others, and each contains a very large karyosome as well as a small plasmosome (Fig. 3). The nuclei of the fifth group (te2_24) are similar to those of the fourth group, but tend to be slightly flattened (Fig. 3). The marginal nuclei are similar to the radials in general character but the nucleoplasm appears to be more reticular (Fig. 3). Each of the marginal nuclei of the postcorpus (ni_3) consists of a posteriorly bilobed body which in cross section appears as 2 nuclei (Fig. 3), each containing a moderate-sized karyosome and a small plasmosome. The marginal nuclei (mz») are not lobed but tend to be irregular in shape due to pressure from the marginal fibers; each contains a large karyosome and a small plasmosome (Fig. 3a). The marginal nuclei (m;_9) are ovoid and similar to mas. The nuclei of the esophageal glands (gi_3) are similar in general character to those of the radial nuclei rig, in having a large karyosome, small plasmo- some, and delicately basophilic nucleoplasm (Fig. 3a); in some specimens Oct. 15, 1936 CHITWOOD AND CHITWOOD: NEMIC ESOPHAGI 419 small clumps of granules occur in the nucleoplasm. The dorsal gland nucleus (21) is the largest nucleus of the esophagus, the subventrals (g2_3) being but slightly smaller. The nuclei of the nerve cells (ny_4:) differ from all other nuclei of the esoph- agus and are similar to one another in that the nucleoplasm contains large basophilic clumps (Fig. 3a) which make them much darker than other nuclei; in general neither karyosome or plasmosome may be observed. The nerve cell nuclei fall into 3 general groups as follows: Small nuclei (n1_19,24,27), about 1.15 to 3.3u long by 1.2 to 2.9u wide in cross section, with rather coarse basophilic deeply staining material; medium-sized nuclei (no3,95~26 ,28-38,41), about 3 to 5.9u long by 1.87 to 2.64 wide in cross section, with fewer baso- philic clumps in proportion to the size of the nucleus; large nuclei (n39_40), about 5 to 5.9u long by 4.3 to 4.5u wide in cross section. The nuclei of the first type belong to bipolar neurones of the 3 anterior esophageal nerves. The cell bodies of these neurones are extremely narrow. The neurones of the sec- ond and third types appear to be commissural nuclei of the postcorpus and bulbar regions. These neurones of the second type have moderate-sized cell bodies (Fig. 3b) and those of the third type have massive cell bodies (Fig. 3b). ESOPHAGEAL GLANDS The dorsal esophageal gland opens into the lumen of the esophagus at the anterior end of the precorpus. Its duct and finer structure are similar to those of the dorsal esophageal gland of Rhabditis (see Chitwood and Chit- wood, 1936). It is difficult to trace the glandular protoplasm in the bulb and particularly to distinguish between it and the sarcoplasm of muscle cells. The large nucleus (g,) apparently lies within the gland tissue, but it may not; we have attributed the nucleus rig to radial muscle, although there is the possibility that it should be attributed to the dorsal gland, and g; to the radial muscle; further study is necessary to determine this point with cer- tainty. The subventral esophageal glands open into the lumen at the posterior end of the postcorpus near the level of n39. They also are similar to the eso- phageal glands of Rhabditis except that they are situated near the middle of the subventral sectors in the bulbar region and not in the lateral part of these sectors. The same difficulties were encountered regarding the nuclei of the glands as were encountered in the case of the dorsal gland. It is possible that the nuclei reo_s1 should be attributed to the glands and that go 3 should be attributed to the radial muscles. However, the interpretation which the writers have given to these nuclei seems to be correct. LITERATURE CITED Cuitwoop, B. G., and Currwoop, M. B. The histology of nemic esophagi. Parts I, II. Z. Zellforsch. u. mikroscop. Anat. 22: 29-37, 38-46. 1934. Part III. This JOURNAL, 24: 557-562. 1934. Part IV. Ibid. 25: 230-237. 1935. Part V. Ibid. 26: 52-59. 1936. Part VI. Ibid. 26: 331-346. 1986. 420 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 26, No. 10 ZOOLOGY.—The_ systematic position of Indostomus paradoxus Prashad and Mukerji, a fresh water fish from Burma.! Rour L. Bourn, Hopkins Marine Station of Stanford University. (Com- municated by WauLpo L. ScHMITT.) The interrelationships between the sticklebacks, tubemouths, pipe- fishes, and their relatives have long mystified systematists. The families have been shuffled and reshuffled by Starks,? Regan,’® Jung- ersen,’ Gregory,’ and others into various groups, orders, suborders, and superfamilies in an effort to express their evolutionary history. In spite of careful and painstaking research, the problem still re- mains unsettled and the natural system obscure. The most doubtful point is the question of whether the Gasterosteoidea (including the families Gasterosteidae and Aulorhynchidae) show affinity to the Scleroparei or to the Hemibranchii and Lophobranchii. The Indostomidae of Prashad and Muker]i,® based upon their new species, Indostomus paradoxus, from Indawgyi Lake, Myitkyina Dis- trict, Upper Burma, is the most recently described family to be al- located to this systematic complex. Although of extreme interest as a possible indicator of the mutual relationships of the other families of the group, it has as yet been discussed only by its authors who state, ‘‘This new family is closely allied to the family Solenostomidae and to a certain extent to the Syngnathidae of the order Solenichthys Regan, but differs from either in several important characters.” Through the courtesy of Dr. G. 8. Myers of the United States Na- tional Museum, I have been able to examine a cotype of Jndostomus paradoxus. It displays many interesting features and I am convinced that Prashad and Mukerji erred in considering it closely related to the Solenostomidae and Syngnathidae. If we analyze the characters used by these authors to define the family Indostomidae and to differentiate it from its relatives, it appears that the family’s affinities are to be sought in more primitive groups than those suggested. The general body form, although possibly of minor significance, more closely approximates that of the Aulorhynchidae than it does that of any of the other families, certainly far more closely than it approximates that of the Solenostomidae or Syngnathidae. The de- ! Received September 5, 1936. 2 SrarKs, E. C. Proc. U. S. Nat. Mus. 25: 623-625. 1902. ’ Re@an, C. T. Biologia Centralia Americana: x—xi. 1908. ‘ JUNGERSEN, H. F. E. Kgl. Dankse Vidensk. Selsk. Skrift. (7) Naturv. & Math. 8: 329-334. 1910. 6 Gregory, W. K. Trans. Amer. Philos. Soc., N.S. 23: 228-229. 1933. 6 PrasHaD, B. and D. D. Muxerut, Rec. Indian Mus. 31: 219-220. 1929. Oct. 15, 1936 BOLIN: INDOSTOMUS 421 pressed head and caudal region of the Aulorhynchidae are very sug- gestive of Indostomus, the main proportional differences, though of minor significance, being the slightly greater depression and more robust build of the thoracico-abdominal region of Jndostomus. It should also be noted that, except for the greater caudal attenuation and the depressed instead of compressed body, /ndostomus rather closely approximates the hemibranchiate family Aulostomidae. When the fins are considered, we find the same relationships sug- gested. Indostomus has two dorsal fins, the first one composed of short spines unconnected by membrane, the second having its origin im- mediately behind the last dorsal spine. This condition is duplicated in the Aulorhynchidae and the Aulostomidae and is far different from the two complete and widely separated dorsals of the Solenostomidae and the single dorsal of the Syngnathidae. The anal fin of Indostomus differs from that of the Aulorhynchidae only in lacking a small spine at its anterior end, and from that of the Aulostomidae only in having its rays branched instead of simple. The latter difference also char- acterizes the second dorsal of these forms. The pectorals and pelvics of Jndostomus are similar in size and position to those of the Aulorhynchidae. From the Aulostomidae, Indostomus differs in having its pelvics in a more anterior position. This difference appears to be of relatively minor importance when compared to the marked differences existing between the normal pelvics of Jndostomus and the inordinately enlarged fins of the Sol- enostomidae or the totally absent fins of the Syngnathidae. Further, the pelvics of Indostomus are composed of four rays, not one spine and three rays as stated in the original description, the outer ray being enlarged and unbranched, but jointed. This is the condition found in the Aulostomidae and differs from that found in the Aulo- rhynchidae as well as the Solenostomidae. The two latter families have a well developed spine in the pelvic fin. The armature of the body in Indostomus is very similar to that of the Syngnathidae and, in all probability, strongly influenced Prashad and Mukerji to consider it a close relative of the pipefishes. It must be remembered, however, that such armature has been developed in many. widely separated families. We find it in the Loricariidae, the Ostraciidae, and the Agonidae, and while the importance of the bony scutes should not be minimized, neither should their importance be unduly stressed because of the conspicuous nature of the character. Its significance as an indicator of close relationship in the case under discussion is somewhat diminished by the contradictory evidence of 422 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 26, No. 10 the body form and is much overshadowed by the evidence-of the fins. Finally, the Aulorhynchidae are also equipped with bony scutes, although they are deeply imbedded and restricted to narrow median and lateral bands. Prashad and Mukerji state that Indostomus is without teeth. A careful examination of the cotype reveals moderately broad bands of minute, villiform teeth on the premaxillary and dentaries, another feature in which the species in question is similar to both the Aulo- rhynchidae and Aulostomidae and one in which it differs from the Solenostomidae and Syngnathidae. The nostrils of Indostomus I find to be single on each side, appearing as an elongated slit. This is clearly shown in Prashad and Mukerji’s excellent figures, although they state that there are two nasal open- ings and indicate in their table that these are similar to the double nostrils of the Syngnathidae. The single opening is somewhat more extensive than that of the Aulorhynchidae and is clearly different from the double openings of the Aulostomidae and Syngnathidae, but is hardly to be compared to the open nasal organ of the Soleno- stomidae. The Indostomidae are said to have ‘‘four complete lobate gills.” The Syngnathidae, however, of all the fish which I have been able to examine, are the only ones in which the gills are so sharply modi- fied in form and structure that they deserve the special designation lobate. The gills of the Solenostomidae, although equipped with com- paratively few filaments, represent only one extreme in a very wide but even numerical variation, and do not differ in basic form from the gills of other teleosts. In the number and shape of the filaments Indostomus is intermediate between the Aulorhynchidae and the Aulostomidae on the one hand and the Solenostomidae on the other. The lateral line system of Indostomus is much reduced. Small pores in the interorbital space, behind the eye, on the occiput, and just anterior to the upper end of the gill opening indicate that the supra- orbital, infraorbital and supratemporal canals are present. This con- dition is similar to that found in the Centriscidae, and is intermediate between that of the Aulorhynchidae and the Aulostomidae with their well developed lateral line systems and that found in the Soleno- stomidae and Syngnathidae without any lateral line system at all. The final analysis of the relationships of Indostomus must depend upon osteological investigations. Unfortunately, the species is so small (the only available specimen is 26 mm in standard length) that osteological investigation of such diagnostic characters as the pres- Oct. 15, 1936 BLAKE: MONOXIA 423 ence or absence of some of the pterygoid or branchial elements is impossible without macerating. This I have been unable to do, as the cotype which I have examined is apparently the only specimen of the species in this country and is too valuable to destroy. Of the known osteological characters, the sutural connection of the post-temporal with the cranium is indicative of relationship to the Hemibranchii and Lophobranchii. On the other hand, the fact noted by Prashad and Mukerji that none of the anterior vertebrae are fused indicates that Jndostomus is not closely related to these groups, but belongs instead with or near the Gasterosteoidea. The branchiostegals are 5 in number on the cotype, not 6 as re- corded in the type description. This number closely approximates that found in the Aulorhynchidae, 4; equals that found in some of the Aulostomidae, 4—5; and is markedly different from the much re- duced number found in the Solenostomidae, 1; and also the Syngnath- idae, 1-3. From the available evidence it seems that the Indostomidae can claim no very close relationship to any known family. The only char- acter tending to link it to the Lophobranchii is the nature of the arma- ture. The majority of characters, the body form, fins, teeth, lateral line system, anterior vertebrae and branchiostegals, indicate that its relatives should be sought among the Gasterosteoidea or Hemi- branchii. Of the families comprising these two groups, the Aulorhynch- idae and Aulostomidae are by far most similar to the Indostomidae. The latter family appears in many respects to occupy an intermediate position and serves as additional evidence of the relationship of the Gasterosteoidea to the Hemibranchii. While this relationship may not be close enough definitely to validate the questionable order Thoracostei, it is much closer than the relationship of the Gastero- steoidea to the Scleroparei which was suggested by Jungersen. ENTOMOLOGY .—A redisposition of Monoxia puncticollis and allied species.1. Doris H. Buaxe. (Communicated by Austin H. CLARK.) LeConte, in his treatment of Galeruca in 1865, divided the genus into five groups, the fifth group consisting of two species, G. maritima and G. morosa, both described by him. In 1885 he added a third species, G. erosa. These three species have been synonymized by Horn with Monozia puncticollis (Say). LeConte had never been able to 1 Received May 22, 1936. 424 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 26, No. 10 a. Erynephala glabra n. Sp. b, Evynephala maritima (Lec,) c Erynephala morosa (Lec.) d.Erynephala puncticollis (Say) Fig. 1.—a. Erynephala glabra Blake, type, Sierra de Durango, Mexico. b. Hryne- phala maritima (Lec.), type in LeConte collection; Gulf States. Genitalia from specimen from Wollaston, Mass. c. Erynephala. morosa (Lec.), type in LeConte collection; San Francisco, Calif. Genitalia from specimen from Alameda, Calif. d. Hrynephala puncticol- lis (Say), type of G. erosa Lee. in LeConte collection. Genitalia from specimen from Salt Lake, Utah. —Habit, about 5x; front of head, about 10x; Beonanoe in dorsal and later- al view, much enlarged. Oct. 15, 1936 BLAKE: MONOXIA 425 determine what species Say had before him in describing G. puncti- collis, and at one time ‘‘was disposed to think” it might be related to Monozia debilis. Horn used more than a page in attempting to show that the three species described by LeConte are in reality all forms of Say’s species G. puncticollis. The fact that these species are all twice as large as Say’s measurements of puncticollis he dismissed as a slip of the pen on Say’s part. Even if this measurement was an error by Say, there is still some objection to placing these very unlike species in LeConte’s homogeneous group of Monoxia, and still more to re- ducing to one species three that have very definite and distinctive characters. These three species agree with the majority of the species of Monozxia in one respect,—the claws of the male are toothed and of the female simple. Otherwise the two groups are not closely related, and I agree with Dr. Boving,? who found their larval characters entirely different from those of the smaller species of Monoxia and with LeConte, who never did incorporate them in that genus, that they should not be put with Monozia. Although Dr. B6ving does not find much difference between the larva of puncticollis and that of Galerucella notata, the mature beetles do not bear any close resem- blance to species of the latter genus. Therefore it has seemed best to set this homogeneous group aside in a separate genus. Erynephala, new genus The genus Erynephala is separated from Monozia (1) by its longer and differently shaped head, (2) by its differently shaped prothorax, which is widest near the base and is not obliquely truncate at the basal angle, and the disc of which is not channelled in the middle or depressed on the sides as in Monozia; (3) by its pygidium, which is not deflexed, as is usually the case in the male of Monozxia; (4) by its much longer and differently shaped aedeagus; (5) by its different larval habits (the larvae of Monozia are leaf miners, and the larvae of Erynephala feed in the open); (6) by its different larval characters, as shown by A. G. Béving. The genus Hrynephala is separated from Galerucella (1) by its shorter antennae; (2) by its differently shaped prothorax; (3) by its claws, which are simple in the female and toothed in the male; (4) by its longer and quite differently shaped aedeagus. Erynephala includes four closely related species, three of which occur in the United States, and one which is known only from the Sierra de Durango in western Mexico. Although EF. pwncticollis (Say) was the first to be de- scribed, EH. maritima is designated as the type of the genus because of the somewhat doubtful application of Say’s name. Of the United States species, maritima is known to occur from Halifax, N.S. to Texas, and there is one species labeled Jamaica in the Bowditch collection at Cambridge. The second species, pwncticollis, is known from Texas (inland) to Manitoba in 2 Boving, A. G., Proc. U.S.N.M., 75:29. 1929. 426 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 26, No. 10 the Great Plains and Rocky Mountains, and has also been collected in the Federal District near Mexico City, Mexico. The third species, morosa, is known only from the California coast. The three United States species are feeders on Chenopodiaceae and inhabit alkaline or saltmarsh regions. In ’ the western United States they are called ‘‘alkali bugs.” They are all considerably larger than the species of Monozia, ranging from 6 to 9 mm. All are dull brownish in color, sometimes piceous, and sometimes the elytra are marked by two rather indefinite vittae, one lateral and the other subsutural. The head is densely punctate and more or less pubescent above, with the median line not always well defined. The lower front is not produced, but broad, flat, and glabrous. The antennae are not half as long as the body, with the third joint longer than the fourth, the fourth longer than the fifth, and the remainder approximately the same length, and longer than broad. The prothorax is not twice as wide as long, and the sides are only slightly rounded, not at all angulate, and are nar- rowed anteriorly. The basal angles are not prominent and are without nodules. The disc is somewhat uneven, with a small depression in the middle and one on either side, not nearly as marked as in Galerucella or Monozia, and the surface is more or less coarsely punctate and nearly glabrous. The elytra are elongate, with parallel sides, slightly convex, without depressions, densely punctate, and either glabrous or only moderately covered with short pubescence. The epipleura are visible nearly to the apex. The anterior coxal cavities are open. The tibiae are not sulcate, and the first tarsal joint is as long as the next two. The claw in the male has a fine tooth on the inner side, not as long as the outer, and in the female the claw is untoothed. The aedeagus is very long, slender, and flat, and because of its length lies in a somewhat bent position in the abdomen. It is quite unlike any aedeagus that I have seen in species of Galerucella or Monozia. KEY TO THE SPECIES OF ERYNEPHALA 1. Upper surface nearly glabrous. Mountains of Western Mexico........ glabra n. sp. @: 0) o> m+ o> Se) vo “es ol tet ime Ye) oe) tee: Ses \e) ce: Ye ie ie ©) e- (ee 1s) (0) @ (wel) qeyse (ee, (ee fe) (a: <6, ei fe ee isis, eye) iene ing Upper surface more or less conspicuously pubescent................. 2 2. Elytra markedly wider than prothorax and covered with short, fine pubescence somewhat obscuring the punctation; prothorax usually some- what depressed with deep coarse punctures. Inland species (Federal District, Mexico; Texas; Great Plains; Rocky Mountains to Manitoba) sei a Yacca ab cel a Wi eeu AME ROMs a1) SRV eidiire, Bina ail oe A puncticollis (Say) Elytra not much wider than prothorax; pubescence only moderately dense with punctation not at all obscured; prothorax not depressed and more shallowly punctate. Maritime species......................95 3 3. Pubescence on head and elytra distinct and rather long; frontal tubercles on head well marked. Pacific coast..................... morosa (Lec.) Pubescence on head and elytra short, not at all conspicuous; frontal tubercles on head indistinctly marked. Atlantic coast. ..mariztima (Lec.) ERYNEPHALA PUNCTICOLLIS (Say) Galeruca puncticollis Say, Journ. Acad. Nat. Sci. Phila., 3: 458. 1824. Galeruca erosa LeConte, Trans. Am. Ent. Soc., 13: 28. 1885. Monoxia puncticols Horn, Trans. Am. Ent. Soc., 20: 83. 1893, in part. Ocr.:15, 1936 BLAKE: MONOXIA 427 As already stated, Horn included under Say’s name puncticollis LeConte’s three species, G. maritima, morosa and erosa. It is by no means certain that Say made a mistake in describing puncticollis as three-twentieths of an inch long, or that he did not have in hand a beetle quite different from any of LeConte’s species and about half their size. However, since no other species appears to answer his description, I am unwilling to change the name of this well known and economically important beetle. Of the three species described by LeConte, Say’s name can be applied only to the inland species, G. erosa, since Say wrote that puncticollis was taken on the Mississippi and on the Arkansas near the mountains. LeConte described G. erosa as ‘‘dull yellow, finely pubescent. Head strongly, densely punctured, prothoraxcribrate. Elytra finely, very densely punctured, outer joints of antennae and the tarsi fuscous. Length 8 mm. Utah. Quite different from our other species (maritima and morosa) by the coarsely sculptured thorax which has also four shallow discoidal impressions. The third joint of the antennae is a little longer than the fourth, whereby it differs from Trirhabda, which it greatly resembles in form.”’ In the LeConte collection are three specimens, all labelled Utah. The one bearing the label G. erosa is a female and the other two are males. These correspond with LeConte’s description and are without doubt the specimens he had before him in describing the species. The head is pale, with short appressed yellow pubescence, not entirely obscuring the punctation below; the lower front is broad, smooth and shining, and is without the wide depres- sion below the antennal base found in morosa. The first four basal joints of the antennae are pale and the rest darker brown. The prothorax is narrowed anteriorly and greatly depressed, with large, coarse punctures, each punc- ture bearing a short, pale yellow hair. The elytra are wider in proportion to the prothorax than in either morosa or maritima, and covered with a fine, dense, yellow pubescence, and the punctation is not as coarse as in morosa. The body beneath is pale. This is the most distinctive of the three United States species of EHry- nephala. It is the largest (sometimes as much as 9 mm.), has the densest elytral pubescence, and the most coarsely punctate and depressed prothorax. The aedeagus, resembling that of morosa in its tip, differs from both morosa and maritima in having the opening on the dorsal side situated farther from the tip. As in the other two northern species, there is great variation in color, specimens varying from pale to piceous. Usually the lower front of the head, prothorax and margin of the elytra are pale in the darkest specimens. The elytra in the majority of the specimens examined are yellowish brown, but occasionally they are vittate. Distribution: Texas (Del Rio, El Paso, Barstow, New Castle, Pecos), Kansas (Meade Co., Clark Co., Wichita), Nebraska (Lincoln), New Mexico (Hagerman, Albuquerque, Artesia, Maxwell), Utah (Provo, Bear River, Thatcher, Garfield, Saltair, Salt Lake), Colorado (Colorado Springs, Rocky Ford, Ft. Collins, Longmont, Greeley), Idaho (Sugar City), Montana (Billings), Manitoba (Winnipeg, Stonewall, Baldwin). ERYNEPHALA MOROSA (LeConte) Galeruca morosa LeConte, Rept. Pacific Survey, p. 70, 1857. Monoxia puncticollis Horn, Trans. Am. Ent. Soc., 20: 83. 1893, in part. LeConte’s Latin description of G. morosa may be translated thus: elon- gate, piceous, lightly covered with cinereous pubescence, the head finely 428 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 26, No. 10 and densely punctate, with two smooth callosities over the antennal bases, the prothorax strongly punctate, uneven, with a deep median and two more indefinite lateral foveae; the elytra a little wider than the prothorax, convex, densely and not finely punctate, the suture elevated, flat towards the scutellum; length .25 inch. The description is founded on a single specimen collected in a salt marsh at San Francisco. LeConte states that it resembles a black individual of G. maritima, but differs from the eastern species by having the prothorax less flattened in front and less rounded at the sides, and the hind angles not flattened, the elytra more coarsely punctured, and the pubescence longer and nearly white. In the LeConte collection is the specimen from which LeConte drew up his description, bearing the label G. morosa and also San Fr. and a round gilt label. It is a male and entirely dark except for the reddish brown mouth- parts. The head is without a deep median impression and has two well marked frontal tubercles over the antennal bases. There is also a pronounced depression directly below the antennal base, such as does not occur in either maritima or puncticollis. The head above is densely, shallowly and coarsely punctate with long white pubescence, longer than in maritima. The antennae are entirely dark. The prothorax has nearly straight sides, is slightly convex with a small median depression and two lateral ones, and is shiny and cov- ered with coarse, shallow, sometimes confluent punctures, not so dense in the middle. There is a slight inconspicuous pubescence on the sides. The elytral humeri, as in maritima, are not prominent, and the depression within is very slight. The elytra, not much wider than the prothorax, are shiny and have dense, coarse punctation and a white and not very conspicuous but long pubescence, slightly longer and more erect than in maritima. LeConte points out plainly the differences between morosa and maritima. His statement that the hind angles of the prothorax of morosa are less flattened than in maritima does not hold in all cases, but generally the pro- thorax of morosa is not so depressed as that of pwncticollis. Morosa differs from both maritima and puncticollis by the deep depression below the anten- nal bases on the lower front, as well as by the differently shaped aedeagus. It differs from puncticollis by its smaller size, less deeply punctate and de- pressed prothorax, less pronounced humeri, with the elytra not as wide in proportion to the prothorax as in puncticollis and longer white pubescence, in contrast with the short, yellowish pubescence of puncticollis. Like both maritima and puncticollis, morosa has several color forms. It may be pale, or the elytra may be vittate. A. T. McClay writes that he has always collected this species in the salt marshes along the California coast. Distribution: California (Lake Merritt, Alameda Co.; Millbrae, Los Angeles Co., Oakland, San Diego, Seal Beach). ERYNEPHALA MARITIMA (LeConte) Galeruca maritima LeConte, Proc. Ac. Nat. Sci. Phila., 17: 218. 1865. Monoxia puncticollis Horn, Trans. Am. Ent. Soc., 20: 83. 1893, in part. LeConte’s Latin description of G. maritima may be translated thus: elongate, testaceous, fuscous or black; head coarsely punctate, prothorax short, narrowed anteriorly, with broadly rounded sides and not at all prom- inent angles, at base on either side obliquely subtruncate; the disc some- what convex, strongly punctate, shortly canaliculate, and on either side vaguely foveate; the posterior angles flattened and obtuse; the elytra densely Ocr..15, 1936 BLAKE: MONOXIA. 429 and rather finely punctate and covered with short, not dense, pale pubes- cence. Length .30 inch. There are color varieties in which the prothorax is partly testaceous and the elytra black with the margin and suture pale. No type locality is given, but LeConte states that the species is abundant on the seacoast from New York to Florida. In the LeConte collection the specimen bearing the label G. marztima has also a round bright red label indicating the locality as the Gulf states. It is a mutilated male specimen without eyes, with only a portion of one antenna, and lacking two legs and part of a third. Except for the yellowish brown labrum it is entirely dark. The head is not at all conspicuously pubes- cent, in contrast with the long white pubescence of morosa, and is coarsely and confluently but shallowly punctate, not as densely punctate as in morosa and puncticollis. The antenna, of which only the first six joints remain, is dark. The prothorax is not depressed except for a faint median anterior spot and two faint lateral impressions and flattened hind angles, and has only slightly arcuate sides. It is. shining and with scattered coarse and rugose punctation, but is not as densely punctate as in morosa nor excavated with deep coarse punctures as in puncticollis, and is nearly smooth in the middle. There is little trace of pubescence. The elytral humeri are not prominent, as they are in puncticollis, and there is only a slight trace of intrahumeral depression. The punctation is dense, coarse and distinct, but not as coarse as in morosa, and there is rather sparse short pubescence, in contrast to the long pubescence of morosa and the thick, fine pubescence of puncticollis. Besides this specimen, there are eight others, four males and four females, labeled ‘‘Del.’’ All are somewhat paler, some yellow brown without vittae, others with entirely dark elytra and particolored prothorax, others vittate. This species is distinguished from morosa by the less conspicuous and shorter pubescence, the more finely punctate elytra, the distinctly flattened hind angles of the prothorax (see notes on morosa), and the quite differently shaped tip of the aedeagus. It is distinguished from pwncticollis by its gener- ally smaller size, less coarsely and deeply punctate and less depressed pro- thorax, less prominent elytral humeri, more sparsely pubescent elytra, and the differently shaped tip of the aedeagus. Some specimens from Florida and Texas, possibly representing M. puncti- collis var. texana Schaeffer’ are usually pale with brown elytral vittae, and show a more pronounced swelling near the tip of the aedeagus than is found in the northern specimens. Often in the northern specimens there is a one- sided swelling near the tip, so that this variation in the southern specimens seems to be only a matter of degree. The beetles do not present any other structural differences. E. maritima is not known to be injurious to beets. It is found only in the salt marshes on the eastern coast feeding on Salicornia, Dondia, and Salsola, although in breeding cages I have reared it with no difficulty from egg to adult on beet leaves. Distribution: Nova Scotia (Halifax), Maine, New Hampshire (Rye Beach), Massachusetts (Ipswich, Marblehead, Wollaston, North Cohasset, Cambridge), Connecticut (Milford, Lyme), New York (Long Island, New York City, Coney Island), New Jersey (Boonton, Avalon, Longport), Maryland (Ocean City), Virginia (Ft. Monroe, Virginia Beach, Wacha- 3 Schaeffer, Can. Ent., 64 (10): 237. 1932. According to Schaeffer this variety is more closely related to maritima than to morosa or to ‘‘typical puncticollis.”’ It would appear that he regards LeConte’s species morosa and maritima, which he calls ‘‘forms,”’ equal in rank with his variety terana. 430 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 26, No. 10 preague), South Carolina (Charleston), Georgia, Florida (Sand Point), Louisiana (Cameron), Texas (Corpus Christi, Brownsville, Galveston), Jamaica (Kingston, Liguana Plain). Two specimens labeled Kansas are probably mislabeled. Erynephala glabra, n. sp. In size, shape and coloring similar to H. maritima, but nearly glabrous on upper surface. Head reddish brown deepening to darker brown on occiput, coarsely and confluently punctate on upper half with a slight trace of fine pubescence. Tubercles not pronounced and depression below antennal base not marked. Antennae reddish brown and like the other species. Pro- thorax not twice as wide as long, narrowed anteriorly and with slightly arcuate sides; disc with trace of central and two lateral depressions, hind angles not as distinctly flattened as in erosa or maritima; punctation coarse and scattered, slightly more distinct than in maritima; only a slight trace of pubescence visible under high magnification; color deep reddish brown. Scutellum pubescent. Elytra with humeri no more developed than in mari- tima and morosa, densely and coarsely punctate, nearly glabrous excepting a slight trace of pubescence near lateral margin which disappears at apical angle; reddish brown with two darker vittae, one near suture, the other lateral, these broadening and coalescing at apex. Body beneath dark brown deepening to piceous on metasternum and first abdominal segments, lightly pubescent. Length 6.5 mm; width 3 mm. Type: male in Bowditch collection, Museum of Comparative Zoology, Cambridge, Mass. Type locality: Sierra de Durango, Mexico. Remarks: This species from the mountains of western Mexico is very similar to the eastern maritime species. The only external differences be- tween the two are the slightly deeper punctation and the nearly glabrous upper surface of the Mexican species. All the other species of the genus have distinctly pubescent elytra. The aedeagus, too, is different, having a tip similar to the eastern species, but being much wider behind the tip. ENTOMOLOGY.—Some new leafhoppers related to Thamnotettix.1 E. D. Bau, University of Arizona, Tucson, Arizona. The writer is working on the tree and shrub inhabiting division of the old genus T’hamnotettix and has recently divided the group into a number of genera. As there are requests for determinations in some of these divisions the following species are described. Gloridonus spatulatus Ball n. sp. Resembling gloriosus, smaller, golden with less green on the clavus. The female segment with a narrower and shorter notch. Length 9 5.5. mm. Vertex shorter and more obtusely angled than in gloriosus, scarcely half longer on middle than against eye, female segment long, rounding poste- riorly with a slightly wedge shaped notch reaching one third of the way to the black marked base. Male plates longer and roundingly narrowing, al- most oval instead of very broad and almost truncate as in glorzosus. The finger like tips curved up around the tips of the smaller and shorter styles. 1 Received July 11, 1936. Oct. 15, 1936 BALL: NEW LEAFHOPPERS 431 Pygofers longer than the plates with their apices extending as white spatu- late tips. These tips much exceeded by a pair of long black spines that lay alongside the anal tube. In gloriosus the pygofers are much shorter and blunter and do not equal the plates while the spines are short and curved. Holotype? @ and three paratypes Riverside, California, June 10, 1908, allotype o and five paratypes Ontario, California, June 12, 1908, all col- lected by the writer, and six paratypes Mint Canyon, California, June 7, 1935, taken by P. W. Oman and returned to him. Allygianus clathratus Ball n. sp. The male resembling a small and slender guitturosus, the female a large golden limbatus Van D., both sexes heavily reticulate. Length 9 7 mm.; o 5.5 mm. Vertex resembling that in limbatus, only slightly more than a right angle, almost as long as its basal width instead of very obtuse and nearly twice wider than long as in gutturosus. Elytra very long, parallel margined, not appressed, venation even more heavily reticulate than in gutturosus, several cross nervures between the sectors and the central anteapical, constricted and several times divided. The female uniform pale cinnamon, almost golden, the male smoky brown with definite arcs on front and a broken crescent above. A pair of dashes anteriorly and a pair of dots on the median line of vertex, four spots on disc, a broken line on pronotum. The nervures in both sexes broadly white. Holotype o& allotype 2 and seven paratypes Pine Valley, July 6, 1931, con two Beaumont Aug. 1, 1912. All taken by the writer in southern Cali- ornia. Pasadenus chicanus Ball n. Sp. Resembling pasadenus Ball, bigger, broader, with widely flaring elytra and semi-circular male plates. Length 2 7 mm. Vertex wider than in pasadenus but equally pointed and elytra longer, broader, with shorter apical cells. Color similar, the ivory saddle broader, the disc of the pronotum light brown with the anterior margin bearing an irregular row of black dots interrupted in the middle with yellow. Female segment with a quadrangular notch twice as wide as in depth, bordered by a black line while in pasadenus the notch only half as wide and is set off by a black area to the base. Male plates short, together semicircular with re- curved finger like tips, pygofer hooks large turned down with the curved tips almost touching the plates. In pasadenus the male plates are twice as long, narrowing to almost truncate tips while the pygofer spines are long, slender, curved outward and upward, and extend well beyond the plates. Holotype @ allotype & Chico, California, Aug. 11, 1912, and one para- type male, Big Bar, California, July 24, 1912, all taken by the writer, 4 paratypes Redding, California, June 28, 1935, and eight Paynes Creek, Cali- fornia, June 27, 1935, all taken by P. W. Oman and returned to him. Pasadenus margaritae Ball n. sp. Resembling pasadenus but much paler, the brown band reduced. Male plates long, spatulate. Length 2 5.5 mm. Structure of pasadenus nearly, more reticulations on the posterior half of corium. Vertex and face creamy shading to straw color below. Pronotum pale brown, subhyaline, the anterior submargin with irregular brown dots 2 Types in the author’s collection unless otherwise indicated. 432 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 26, No. 10 broken on the median line. Elytra with the ivory saddle of pasadenus very faintly margined with brown, a definite apical band. Female segment scarcely emarginate but with a broad black median stripe. Male plates long spatulate, three times the length of the valve, their apices turned up into chitinized hooks. Pygofer hooks just visible at the sides of the plates at about two thirds their length. Holotype & allotype @ and three male paratypes, Alpine, July 5, 1931, one female Pasadena, July 31,1912, and one female Santa Margarita, August 6, 1921. All taken in California by the writer. Doleranus atascasus Ball n. sp. Resembling perspicillatus O. & B. but much longer and narrower. The head markings even more definite while the long elytra have only traces of the dark markings. Length 9 4 mm. Vertex, pronotum and scutellum as in perspicillatus nearly with the dark markings in front and the brown ocellate spots very definite. Face longer and narrower with ares on clypeus and dark lines around lorae. Elytra much longer and narrower, one half longer than abdomen, the anteapical and apical cells long, the central anteapical but little inflated posteriorly. The nervures on clavus milky, those on corium except the outer sector smoky widened on the cross nervures but with only faint markings in the cells. Female pygofers longer and narrower; the segment rounding and elevated posteriorly with a slightly bifid projection more than twice as long as in perspicillatus. The male plates with long narrow fingerlike projections stronger than in that species. Holotype @ Atascasa Mts., Nov. 3, 1935, Allotype o Santa Rita Mts., Apr. 5, 1931, and one female paratype Huachuca Mts., Oct. 13, 1931. All taken by the writer in the mountains of Arizona. Doleranus kinonanus Ball n. sp. Resembling longulus in form and color, shorter with a shorter head and rounding male plates. Pale cinnamon with smoky clouds behind the cross nervures. Length 9 4.5 mm. Vertex obtusely conical, length less than two-thirds the width at base, shorter than in longulus. Pale cinnamon above and below, traces of brown ares on the sides of the clypeus, faint indications of usual color pattern on dorsum and definite smoky clouds margining the cross nervures. Female segment but little over one half as long as in longulus with the posterior margin slightly angularly excavated with a definite slightly dark margined triangular projection. Male valve and plates apparently fused into a con- vex shield as broad as long, the tips of plates distinct and separated by a ‘““V’’ shaped notch. Holotype @ allotype o and 16 paratypes taken by the author at Kino Bay, Sonora, Mexico, Dec. 9, 1931. Ollarianus rubianus Ball n. sp. Large, size of and superficial resemblance to T. languidus Ball, pale creamy with four black spots on vertex; elytra smoky inside of broad white, almost parallel, margins. Length 9 6 mm. Vertex very short, much shorter than in balli, over two and one-half times wider than long, almost parallel margined, pale, with two large black spots above the ocelli and two smaller ones between, sometimes elongate; prono- tum two and one-half times as long as the vertex, pale with indefinite sub- Oct. 15, 1936 BALL: NEW LEAFHOPPERS 433 marginal spots and obscure smoky stripes. Elytra, long flaring subhyaline, the nervures and apices smoky, the dark tergum and under wings showing through give the insect a smoky appearance with the broad flaring white costal margin in sharp contrast. Below creamy with a smoky crescent under the vertex and a few brown arcs well down on clypeus. Female segment twice as wide as long, posterior margin truncate with the angles rounding (arched so as to appear excavated), with a broad triangularly rounding median projection, male valve obtusely triangular, plates together long tri- angular three times the length of vale, the margins concave to the broad finger-like apices which almost equal the pygofers. Holotype 92 allotype o and four paratypes Nov. 3, 1935, two paratypes Apr. 17, two May 15, and one May 23, 1936, all taken from the aa of Mt. Atascasa, Arizona, by the writer. Ollarianus bullatus Ball n. sp. Size and form of ballz nearly. Pale creamy, with the inner pair of spots on vertex the largest, instead of those above the ocelli as in that species, and the face of the male black. Length 2 4mm. Vertex about twice as long as wide, almost parallel margined, over two- thirds the length of the pronotum. Elytra as in ballz, inclined to be flaring with similar venation. Female segment twice wider than long, parallel mar- gined but elevated so as to appear slightly angularly excavated, a broad roundingly triangular median projection. Male valve very broad and short, the plates together broad, roundingly narrowing with long filamentous pro- jections which are sometimes orange as in strzctus Ball. Pale creamy with a slight smoky cast to the elytra, a pair of small spots above the ocelli, a large pair between these some distance back from the margin, a pair just behind the eyes, a pair on the disc of the scutellum and another pair on the margins before the apex. Any or all of these spots may be missing but the face of the male has a black or smoky cloud in all examples. Holotype o& allotype 9 and three paratypes Aug. 6, 1934, one Aug. 5, 1934, one June 8 and one June 23, 1929, all taken by the writer in the Baboquivari Mts., Arizona except the last two which were taken in the desert nearer Tucson, Arizona. Ollarianus ollus Ball n. sp. Form of sirictus nearly, larger, longer and more slender than bullatus, with a short uniformly rounding vertex and faint markings, no black on face. Length 2 4.6 mm. Vertex slightly narrower than in bullatus parallel margined with four fine points or none, no dark on face. Markings on pronotum and scutellum small or wanting. Elytra long and inclined to be flaring. Slightly smoky in the males with the cross nervures emphasized. Female segment more than half as long asits basal width, roundingly narrowing, the posterior margin nearly truncate with the median third roundingly produced. Male valve small triangular, plates broad, long spoon-shaped with the tips elongated equalling the pygofers, the concave portion of the margins before the tips heavily black marked. Holotype & allotype 2 and 4 paratypes together with two nymphs taken by the writer from the east slope of the Santa Rita Mts., Arizona, Aug. 18, 1935. 434 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 26, No. 10 Ollarianus rudiculus Ball n. sp. Size and form of ollus nearly, the elytra long but not as inclined to be flar- ing. Pale smoky, heavily marked. Length of 2 5 mm. Vertex short parallel margined, creamy with four black spots in a reversed crescent. Pronotum smoky or cinereus with a submarginal row of spots, the pair behind the eyes usually round and definite. Scutellum creamy with a smoky pattern, a pair of round black spots on disc and another pair before the apex, elytra smoky subhyaline, the nervures pale brown. Face and be- low pale, in dark examples a pair of dashes on the upper part of the clypeus, another pair at apex and irregular fine dots outline the arcs. The female seg- ment, as in ollus, the male valve short and broad, triangularly narrowing and then produced into short, broad, divergent apices longer than the pygofers. Holotype o and allotype 9 and one paratype taken in Bear Canyon (Labeled Tucson) March 22, 1931. Three paratypes Sabino Canyon, eight from Atascasa Mt. and two Nogales, all taken by the writer in the mountains of Arizona. CONTENTS PAaLEONTOLOGY.—A new species of “‘Crassatellites’” from the upper Miocene of Florida. W.°C. MANSPIBLD. , 2 «...-. 9.40550 ee Botany.—New species of Arundinaria from Southwestern China. Y. L. Kaine 700). 2s fe ge en Botany.—A Fusarium-like species of Dactylella capturing and con- suming testaceous rhizopods. CHARLES DRECHSLER.......... ZooLocy.—New millipeds of the American family Striariidae. H. F. LOOMIS... 6 faye RR a ee Oe nae ZooLocy.—Anguillulina askenasyi (Biitschli, 1873), a gall forming nematode parasite of the common fern moss, Thuidium delicatu- lum (L.) Hedw.. 'G. STEINER 7.0. 2. eee oe ZooLtocy.—The histology of nemic esophagi. VII. The esophagus of Leidynema appendiculatum (Leidy, 1850.) B. G. Cuirwoop and M. B.-Carrwoob... 3) fo eae ae ee ee ZooLtocy.—The systematic position of Indostomus paradoxus Prashad and Mukerji, a fresh water fish from Burma, Rour L. Bouin... EntTomMoLocy.—A redisposition of Monozxia puncticollis and allied species. Doris Gy BLAKE 050 Ce ee ae ee BALI. 5b a ee ge This Journal is indexed in the International Index to Periodicals - Von. 26 NovEMBER 15, 1936 No. 11 age \' JOURNAL © * \ # iret way, 2 a OF THE ~~ “WASHINGTON ACADEMY __OF SCIENCES BOARD OF EDITORS __F.G. Brickweppr Rotanp W. Brown Espen H. Tooir BUREAU OF STANDARDS U. S. GEOLOGICAL SURVEY BUREAU OF PLANT INDUSTRY ASSOCIATE EDITORS H. T. WENSEL Harotp Morrison PHILOSOPHICAL SOCIETY ‘ ENTOMOLOGICAL SOOIETY ~ E. A. GotpMANn W. W. Rusey ‘ BIOLOGICAL SOCIETY GEOLOGICAL SOCIETY | _ Agnes CHASE J. R. Swanton rie ~ BOTANICAL SOCIETY ANTHROPOLOGICAL SOCIETY R. E. Gisson CHEMICAL SOCIETY PUBLISHED MONTHLY BY THE WASHINGTON ACADEMY OF SCIENCES ie 450 Aunatp Sr. a ; AT MeNASHA, WISCONSIN ‘Entered as second class matter under the Act of August 24, 1912, at Menasha, Wis. Acceptance for mailing at a special rate of postage provided for in the Act of February 28, 1925. Authorized January 21, 1933. Journal of the Washington Academy of Sciences oe This JourNAL, the official organ of the Washington Academy of Sciences, ouuiene (1) short original papers, written or communicated by members of the Academy; at proceedings and programs of meetings of the Academy and affiliated societies; (3) notes of events connected with the scientific life of Washington. The JourNAL is issued monthly, on the fifteenth of each month. Volumes correspond to calendar years. Manuscripis may be sent to any member of the Board of Editors. It is urgently re-_ quested that contributors consult the latest numbers of the Journat and conform their © manuscripts to the usage found there as regards arrangement of title, subheads, syn-- onymies, footnotes, tables, bibliography, legends for illustrations, and other matter. _ Manuscripts should be typewritten, double-spaced, on good paper. Footnotes should — be numbered serially in pencil and submitted on a separate sheet. The editors do not assume responsibility for the ideas expressed by the author, nor can they undertake to” correct other than obvious minor errors. f Illustrations in excess of one full-page halftone or two full-page line dr awings may be paid for by the author. Line drawings are preferred. Proof.—In order to facilitate prompt publication one proof will generally be per to authors in or near Washington. It is urged that manuscript be submitted in 2 form; the editors will exercise due care in seeing that copy is followed. Author's Reprints —Fifty reprints without covers will be furnished gratis. Covatke bearing the name of the author and title of the article, with inclusive pagination and date of issue, and reprints, will be furnished at cost when ordered, in accordance with - the following schedule of prices. ae Copies 4 pp. 8 pp. 12 pp. 16 pp. Covers 50 ae Bah ah ea es $2.00 : . 100 $ 95 $1.20 $1.50 $2.00 $2.75 Ye 150 1.25 1.90 2. De a. a0 3.50 + ad 200 176 2.60 3.60 4.50 4.25 Be Ne, 250 2.00 3.30 4.65 tg Cs al 5.00 Envelopes for mailing reprints with the author’s name and address printed in ‘thee age corner may be obtained at the following prices. First 100, $4.00; additional 100, $1. 00: bei Subseripiion’ Rates:.— Per volume. 60 bs 5 se ee ee Rp bones ae $6.00. To members of affiliated societies; per volume..................0.05 Ja. oes Sintgle MURDERS). 5 set pee ek a ate ee er ae