Ynys wl aa is RH i i Ry oe MDS 1p ‘f oe ‘y y ee cor ii te Cane 4) pee) i PAG MC a Ye rah WE ‘ Me Wa AO Ha att a tliat it ne i Ih ; ND ant raat Hh AL WRN ly ‘ute HW hy hae A i iy) hi Wid Way B cee ae ts iM Hatemwtet Relay RI Aa Noi vii Vat PLN te ie ee iy ‘) y iy Wii hens i oe SMG j ite NING, Pituat| aia ale tah 4 bey a i WAR PR aD iit +e srt ate es on ia #4 i ead Asa d : 1 i a 4 rae aK sta hs Wat ; i ON, Sa af wt May H he) Hs Ash MeN oe fa eh ‘iy ohh Ma MA i Ra SO 1a bray J Rive wie okie eye OCU NM DS ON 4s) TPE Ne Lk alk b Pte as) HUNT aL ; UR ICRN, iis My ue ROKR . a wi i of ‘ i Mh POA WN ph Ny a . 4 ah RENN 8 44s) ya \} EAM RASA AALS bd 4 i 4 we Ny ana | ay He ,) at Pa rei, NS 4 4 Pah hw avn ie) ua Cont Dyn Sa Are hey irs atic Sibhele Hoe Sieh MY, Ate Nie Aih iN is A “Ag ane ty aah ' ‘ WG “ ee aN - tai i Wah! 4 AN LAG 4 ne EAN ADAIR Aa ih wilh ’ mane 4 Hy ein 5 \ 3 a cee ha iy a8, i UGS tet ‘ ins eh 4h x cn heii “i sti 6 4 EG N Ne RMT eit Peat vn » iN vite Mg hat) be ous a 1 fiytish ie eee ior \ t Cy sa Va Re ah Al . ae aie vig sie ih i peat be prunion ao ee) yy in a " tie did ne oh i sh ial ‘i y sf 4 bi aah iis! hos nt) ihe Cue. te rt Nt ait , ae a tae) a Ant Aa sees BES Lan } La ROCHA TORN ites Hasso detent ees ae ny ean THe a Why, ti inet HKD aa ieee = — ont i HI Wl fess nt nN) nn iy i rye he a we a A Bi OM; ‘i oy Ary oh ae o Wwe ye! MN Can a ‘i st a i pie neh Min vie si NN MN NY ot Ort) i ee ‘ it bet Hy aw Pee uN i Bee ec ee eS Se ae es ee) ss Dy Mie cat a + Wee, 3 oak Hani ae HAA aes ret; fares a, 2 te Se ew ae: = tee: A cit ‘ oe CO AAI aati b Bria ‘; OVA iar ua a 5 P Ap? anisina rae ‘ . : s mae mae : ; &, nivk Tie Ah Crs \ Aaa Nan An , ¢ y y moi Pe ok a tos oN x ' } f s ' { ink i 1 ee a . ‘ “yy M - {; , 1 SS if re” : ¥ 1 1 ‘ J if ‘ i i) if ine K ph c Kye t ai 1 i 1 , vi ‘ a) Pm Ay A) eee y an? i ay "| y hy, ay y i” t ii wT) , } ¢ } fhe ao ’ BART , ro 9 q : 4 i A : " i f my 4 y 1 » ae : ¥ ‘ i mm { an i } j { al iy, [hy SS | ’ ; / liad u iN » eg i et d { y , t; : i iy ARS ' Li} 7 , } ran ks ;/ f i i 5 a } i ut ‘ 1 : aA 5 asi a eee i Se 1 , =f * : i iy ue a LY, (AWAY ba he are" 5 TW AL Ay, uaa Mak , 1 oe 4 7 ¥ D t uy ‘ ' f UL LO ee De AN im ane ft late tt eg + a ype, Sayan Nea saa oak ¥ ie he i a SMITHSONIAN INSTITUTION UNITED STATES NATIONAL MUSEUM PROCEEDINGS OF THE UNITED STATES NATIONAL MUSEUM VOLUME 39 WASHINGTON GOVERNMENT PRINTING OFFICE 1911 ADVERTISEMENT. The scientific publications of the National Museum consist of two series—Proceedings and Bulletins. The Proceedings, the first volume of which was issued in 1878, are intended primarily as a medium for the publication of original papers based on the collections of the National Museum, setting forth newly- acquired facts in biology, anthropology, and geology derived there- from, or containing descriptions of new forms and revisions of limited groups. A volume is issued annually, or oftener, for distribution to libraries and scientific establishments, and, in view of the importance of the more prompt dissemination of new facts, a limited edition of each paper is printed in pamphlet form in advance. The dates at which these separate papers are published are recorded in the table of contents of the volume. The present volume is the thirty-ninth of this series. The Bulletin, publication of which was begun in 1875, is a series of more elaborate papers, issued separately, and, like the Proceedings, based chiefly on the collections of the National Museum. A quarto form of the Bulletin, known as the “Special Bulletin,” has been adopted in a few instances in which a larger page was deemed indispensable. Since 1902 the volumes of the series known as ‘‘Contributions from the National Herbarium,” and containing papers relating to the bo- tanical collections of the Museum, have been published as Bulletins. Ricuarp RATHBUN, Assistant Secretary, Smithsonian Institution, In charge of the United States National Museum. May 13, 1911. IL TABLE OF CONTENTS. Anprews, E. A. Sperm transfer in certain Decapods.— Wort Otc camueny (1) LOLVe cele. 2 ot Asuwortu, JAMES Hartiny. The Annelids of the family Arenicolide of North and South America, including an account of Arenicola glacialis Murdoch.—No. 1772. Sep- mememetae ONO Gi eee yh ye Bartscu, Paut. Descriptions of new Mollusks of the fam- ily Vitrinellide from the west coast of America.—No. MEAS Deel Ue yO 1 OM LG. 2 Ls ty. Neti wane Rt ead lh New species: Cyclostrema baldridgei, C. miranda, C. adamsi, Cir- culus liriope,C. diomedex, Cyclostremella dalli. The recent and fossil Mollusks of the genus Alabina from the west coast of America——No. 1790. SemmnygNlin el OI GAr! hee. Awe oe Cee a eke wie eg New species: Alabina barbarensis, A. hamlini, A. phanea, A. diomedex, A. ignati, A. io, A. monicensis. New subspecies: Alabina tenuisculpta diegensis, A. t. phalacra. The recent and fossil Mollusks of the genus Diastoma from the west coast of America.—No. 1802. LCE Ges gata) 20 1) as ae ea UNE New species: Diastoma chrysalloidea, D. oldroydx, D. stearnsi. The west American Mollusks of the genus Alaba.— Noteeic ¢652222 {Sena ees eee Antero-ventral view of the anterior portion of A. pusilla Quatrefages. In this view all the parts, except the prostomium, are seen somewhat foreshortened. x8. L, lateral lobes of prostomium; M, median lobe; M. Gr, metastomial groove; Mo, mouth; Nr’; first neuropodnim: ss-ce. tees s-e eee eee Dorsal view of the anterior end of A. pusilla Quatrefages showing the widely divaricated lateral lobes (Z) and the median posterior portion (P) of the prostomium (NV) first notopodium; (N.Gr)nuchalgroove. X8.......------ Antero-ventral view of the anterior end of a specimen of A. claparedii from Crescent City, Californias | C6. ceca uo ce Cust eee eee cee ee ee eee Dorsal view of the anterior end of thesamespecimen. X6...........-----+-+: Page. 12 14 15 16 16 LIST OF ILLUSTRATIONS. A. assimilis, anterior end, dorsal aspect, of specimen from Uschuaia. The prostomium is in.a state of normal extension...............2..-.---2255---- A. assimilis, var. affinis, anterior end, dorsal aspect of specimen from the Falk- land Islands. The prostomium is protruded to its fullest extent. x6. JL, lateral lobe of prostomium; M, median lobe; N’, first notopodium; P, median posterior portion of prostomium; Ph, pharynx 362.5222. 2 oal ee een sete eet A. cristata, anterior end, dorsal aspect of a specimen from Florida. x4. JL, lateral lobe of prostomium; L, N, lip of nuchal organ; M, median lobe of prostomium; NV’, first notopodium; Ph, pharynx..........-.....-.-.------ A. glacialis, ninth gill of No.8. Three of the axes are cut off. X 15........-- A. glacialis, third gill of No.6. Three of the axesare cut off. X 35.........-- A. glacialis, anterior end, dorsal aspect. X 6. L, lateral lobe of prostomium; LN, lip of nuchal organ slightly everted; 1, median lobe of prostomium; N’, first notopodium (the notopodium is retracted, leaving a slit on the sur- face) VAG re mich altoroOVvies- can ssace 2 ce skeen LeeGieeen eS aoeoe deceecseeeae A, glacialis, diagram representing the annulation of the anterior end. 4. Due bist OUOMOUHIMNMN <2 eicnc'. ss sales iie wins soho Seemesas oes Cete ees A, glacialis, neuropodial crotchets. -A and B were found together, the crotchet B was in use, A is one of the two small crotchets found in this neuropodium. - SE COSC PS REUNOLOStH eA A ws on AA Sein ahead Pe SA erate eae SRA Loe Wee see Mee Ne Neoseps reynoldsi. 2, top of head; 3, side of head; 4, underside of head; 5, fore Pe GIMME Sep enine eck. see Saeele weiner Pa ee ak Se Peete Mesotergum and base of right wing of Txniopteryx frigida (stonefly), showing the wing-bearing notum (JV) and the postnotum (PN): A, anal vein; ANP, ante- rior notal wing process; 1Az, first axillary; 2Az, second axillary; 3Az, third axillary; AxC, axillary cord; C, costa; Cu cubitus; M, media; m, median plate of wing base; NV, notum; PN, postnotum; PNP, posterior notal wing process; Pph, postphragma; Psc, precutal differentiation of notum; R, radius; Rd, posterior marginal reduplication of notum; Sc, subcosta; Tg, tegula.......... Mesotergum of Blatella germanica (cockroach), dorsal view, illustrating a tergum consisting of a notal plate alone; a, chitinous fold reflected upon posterior edge of protergum; ANP, anterior notal wing process; anr, line formed by anterior ventral notal ridge; AxC, axillary cord; PNP, posterior notal wing process; pnr, line formed by posterior ventral notal ridge; Rd, posterior redu- plication of the notum; vnr, line formed by median ventral V-shaped notal Ventral view of mesotergum of Blatella germanica: ANR, anterior notal ridge; PNR, posterior notal ridge; VVR, median V-shaped notal ridge, the ‘‘ento- LQ re MMM te syste wise oer cere ars bere eetsierisl Sica =\e-a¥e © essai INle econ toe aes Mesotergum of Holorusia grandis (cranefly), showing division of notum into three parts (Psc, Sct, and Scl), back of which is postnotum (PN): AxC, axil- lary cord; ANP, anterior notal wing process, PN, pn, postnotum; PNP, pos- terior notal wing process; Pph, postphragma; Psc, prescutum; Ad, posterior reduplication of notum; Sc/, scutellum; Sct, scutum; u, lobe of prescutum | SUSI WSTT SE | VST zi 0) ek Oto Oe ELE Sg ts a RE ge gO IRR ee oy oe one Metathorax of Txniopteryx frigida (stonefly), left side, wings removed: Cz, coxa; CzxP, pleural coxal process; Epm, epimerum; Eps, episternum; F', base of femur; WV, notum; P, episternal parapterum; PN, postnotum; PS, pleural suture; g, sternopleural suture; S, sternum; Yn, trochantin; 7’, trochanter; Deepal tetal MW PYOCERS ec ce. = Sie cease wine ieee toen sewtaya)s 5 Sowa someeaioa Left metapleurum of Txniopteryx frigida (stonefly) internal; c, sclerite connect- ing parapterum (P) with head of costal vein of wing; Epm, epimerum; Eps, episternum; P, episternal parapterum; PA, pleural arm; PN, lateral part of postnotum, continuous with epimerum; PR, pleural ridge; S, sternum; T7'n, irochantin 2b. pleural Win? PrOCeSS: .oc swimming Legs. oS cn toec Se one hee oe ee aoe ee ee el 273 First antenna of Ergasilus manicatus, showing the form found in the Ergasiline. 274 First antenna of Bomolochus eminens, showing the form found in the Bomolochi- mee and Peniacanthine o> ios eee ve ee are ae ae 274 Second ‘antenna ‘of Hrgasilus centrarchidarum = 12... * 22022 2. ee ee 275 Second antenna of Bomolochus nitidus 2: 28555 ~ 2 2 cs eee ee 275 Mouth-parts of Ergasilus mugilis; la, labrum; 1b, labium; md, mandible; mz’, first maxilla: @2”, second maxilla = .022 5522) jo) al atace nae ey rea 275 Mouth-parts of Bomolochus exilipes. an, second antennee; la, labrum; /b, la- bium; md, mandible; mz’, first maxilla; mz’’, second maxilla; mxp, maxilliped. Between the terminal joints of the mandible and second maxilla can be seen a hairy structure similar to the paragnaths in free-swimming forms.......-... 276 Muscular system of Ergasilus versicolor in dorsal view. a, muscles moving first antenna; b tof, muscles of enlarged second antenne; g to /, flexors of head and first two thorax segments; m, 0, and 7, muscles moving swimming legs; m, b, transverse muscle band; n, flexor muscle of third and fourth thorax segments; p, muscles controlling the opening of the oviduct; q, flexor muscles of fifth and genital segments; s, flexor muscles of abdomen; t, accessory muscle Of SECON Antennas. Fae. Lec see alaak belempeisiea’e crane sya hele Siac ee eee 286 Muscles controlling opening of oviduct in Ergasilus versicolor. a and c, closing muscles; -0 and.d, opening muscles’: 2 2c /2 2s 3-6 ot seco cise ess see eee ee 288 Ventral surface of female Ergasilus centrarchidarum. a, median lobe of stomach; b, diagonal lobe; ¢, lateral lobe; d, excretory tubes; 2, intestine; 7, rectum; s, SEOMACH Sis ais 2L Pee are cee ee re eet ets ase Reece Toeret enti SS tee 291 Excretory tubes of Ergasilus centrarchidarum. a and 6, from different indi- WAG alg 2) 2 2 ce on ce csaleiale te os slate ciniele = wicie e's Srmmicle Gate arate seieege ee 293 Median longitudinal section of Zrgasilus centrarchidarum. a, anus; am, muscles of second antenne; dg, digestive gland; dm, dorsal muscles; e, eye; g’ to g’’, ganglia of thoracic and genital segments; gc, gland cells; 7, intestine; J, lacu- ne; m, mouth; n, ventral nerve cord; 0, anterior end of ovary; @, cesophagus; p, protractor muscles of stomach; sbg, post (sub) esophageal ganglion; spq, pre (supra) esophageal ganglion; sr, semen receptacle; st, stomach; up, MLELING PTOCESS OL OVALY sci. cic als ale tre sla aceite weal ats ore Sie ee teva eee ee eee 295 Transverse section through the carapace of Ergasilus centrarchidarum behind the mouth. dvm, dorso-ventral muscles; 7, stomach; lm, longitudinal muscles; n, ventral nerve cord; 0, undeveloped egg cells of ovary; up, uterine proc- LIST OF ILLUSTRATIONS. Transverse section through third thoracic segment of Ergasilus centrarchidarum. eg, cement gland; g, ganglion of third swimming legs; 2, intestine; lm, longi- tudinal muscles; m, dorso-ventral muscles of third legs; n, nerve of third leg; MEIBeMICT TECEP LACIE. oy i sf) = yar bse ajo sae Sales etn ons seas selene oie es Dorsal view of carapace of Ergasilus centrarchidarum, showing position of first fwo eee cells that issue from the ovary<---...2-22--2s---/a-+225+-2-2--% The first uterine processes forming posteriorly through the first thorax segment. - The second uterine processes forming anteriorly in the head; posteriorly can be seen the oviducts connected with the first uterine processes. ............-.- The third and fourth oblique uterine processes, extending forward and down- ward from the outer sides of the first processes. Anteriorly can be seen the fused ovary and posteriorly the cement glands............-..-..----+------- Fully matured female of Ergasilus centrarchidarum, showing relative size and shape of completed processes in dorsal view.........-..-------+--+---+---+--- The relative size and shape of completed processes as seen in ventral view... . The relative size and shape of completed processes as seen in lateral view... - . Cement gland of Ergasilus centrarchidarum. a, dorsal view; 6, longitudinal Serre eer LLOU I saree ee wera ante state Si Es cralciava Sisierarnfurecche ate miners arene ae Genital segment of male Ergasilus chautauquaénsis, showing sperm receptacles. Mouth-parts of female Ergasilus centrarchidarum. 1b, labium; md, mandible; md. p., mandibular palp; mz’, first maxilla; mx’’, second maxilla. ..-.-.... Mouth-parts of male Ergasilus chautauquaénsis. 1b, labium; md, mandible; mz’, first maxilla; m’’, second maxilla; mxp, maxilliped...........----.-.- Three successive stages in the growth of the egg cases of Ergasilus centrarchidarum, showing that the eggs ripen in groups and not singly...........-.....------- Newly hatched nauplius of Ergasilus centrarchidarum, dorsal view. ..-.------- First antenna of nauplius of Hrgasilus centrarchidarum........-----------+-+-+- Second antenna of nauplius of Ergasilus centrarchidarum.......--.---+-+y---+-+- Mandible of nauplius of Ergasilus centrarchidarum.....--..-------+-++02+-++5- First metanauplius larva of Ergasilus centrarchidarum. The first maxillee in this figure are concealed beneath the mandibles...........--..------------- ACU Ob MES MCLANAUPILUSIARVAS , Bay of Paita. L. K. Scumarpa, Neue wirbellose Thiere, vol. 1, pt. 2, p. 52, Leipzig, 1861. A. piscatorum ¢, Callao. E. Gruss, Vid. Medd. naturh. For. Kjébenhavn for Aaret 1858, p. 120. Kjébenhavn, 1859. A. marina®, Puerto Montt, Chile. E.Enurrs, Festschr. K. Ges. Wiss. Géttingen, 1901, p. 176. Rathbun? and Tauber ¢ also mention the occurrence of this spe- cies on the shores of the United States. Remarks on the foregoing records.—The specimens on which the spe- cies A. natalis Girard was founded are very shortly and insufficiently a A. claparedii, see pp. 7, 8, 9. b Probably A. claparedit, see p. 8. ¢ See p. 9. @R. Rathbun. The Worms in: The Fisheries and Fishery Industries of the United States. Section 1. Washington, 1884, p. 833. e¢ Annulata Danica, p. 110. no.1772. ANNELIDS OF THE ARENICOLIDH—ASH WORTH. 7 described; no single character is mentioned by means of which the species with which Girard was dealing can be absolutely fixed. Never- theless, it is, I consider, practically certain that the specimens in question were ordinary examples of A. marina. It is evident from the description that Girard mistook the ventral for the dorsal surface, for he speaks of the dorsal region as being marked by a conspicuous smooth line, which, upon the cephalic region, subdivides into right and left branches, which unite again anteriorly; this dorsal line is given as one of the principal specific features. This smooth band or line is, however, really ventral in position, and is seen in nearly all specimens of A. marina (and, indeed, in other caudate Arenicolidee) ; it marks the position of the ventral nerve cord, and in front it is con- tinuous with the metastomial grooves, which mark the course of the cesophageal connectives. I have endeavored to procure the type- specimen of this species, but without success; the curator of the museum of the Boston Society of Natural History informs me (letter dated, November 4, 1908) that he has not been able to locate the type and that there is no record of its having been given to the Society. There can be little doubt that the specimens were examples of A. marina, which species, as may be seen from the list of records given above, has been found at other stations in the immediate neighbor- hood of Chelsea. There are two specimens in the Museum of Comparative Zoology, Cambridge, Massachusetts, which were collected at Grand Manan and bear the label ‘‘ Arenicola natalis Girard.”’ (Mus. No. 87.) Ihave re-examined these and have no hesitation in referring them to the species A. marina, with which they agree in every respect, both in regard to external characters and internal organs. The records from Vancouver Island, Puget Sound, the Bay of Paita, Callao, and Puerto Montt can not be passed without emen- dation. Von Marenzeller® states that he examined examples of A. marina from Vancouver Island. As his diagnosis was reached from a con- sideration of the external characters only it seemed to me that a re-examination, which should include the internal organs, was desir- able. Professor von Marenzeller kindly lent me a specimen for this purpose. The specimen is about 70 mm. long and looks at first sight like a small example of A. marina, except that the lateral lobes of its prostomium are proportionately larger than those of A. marina. Although I had a suspicion that it should not be referred to this species, it was impossible to finally determine this point by an examination of the external features only. Accordingly, an incision was made along the greater part of the worm and the flaps of the body wall turned aside. It was then seen that instead of a single pair of a Zool. Jahrb. Abth. Syst., vol. 3, 1888, p. 12. 8 PROCEEDINGS OF THE NATIONAL MUSEUM. vou. 39. cesophageal glands, as in A. marina, this specimen has five pairs, the first being 9 to 10 mm. long and the others 1.5 to 3.5 mm. long. There are no muscular pouches on the first diaphragm (these struc- tures are present in A. marina), and statocysts could not be found after careful search.? Five pairs of nephridia are present, opening on the fifth to the ninth segments. These internal characters and the nature of the prostomium indicate that the worm is to be referred to the species A. claparedu Levinsen. Child’s record of A. marina from Puget Sound is founded on a misapprehension. He was con- cerned with the cytology of the ova and probably did not make any examination of the systematic characters of the worms. Johnson® states that the specimens examined and recorded by him as A. claparedii were given to him by Child; moreover, subsequent records of specimens from Puget Sound and all the specimens from that area which I have myself examined belong to the species A. claparedi. We need therefore have no hesitation in transferring Child’s record from A. marina to A. claparedi. Schmarda records A. piscatorum from the Bay of Paita, but a statement in his description shows that he was not dealing with this species, for he mentions the presence of twenty glandular sacs just anterior to the stomach, that is, the specimen had twenty cesophageal glands, whereas A. marina has only two. I have tried to find Schmarda’s specimens, but have failed todoso. Prof. Dr. K. Grobben has been good enough to look through the catalogue of the collection in the Zoological Institute of Vienna, where I thought the specimens might possibly be, but there is no example of Arenicola in that col- lection from the Bay of Paita and no specimen of this genus collected by Schmarda. His specimens from Paita were examples either of A. claparedi or A. assimilis, probably the former, judging from the distribution of these species (see p. 17 and p. 20). At any rate they can not be examples of A. marina, as is definitely shown by the number of their cesophageal glands, and we may provisionally regard them as belonging to the species A. claparedi.° @Jn order to definitely establish the absence of statocysts, it would be necessary to cut serial sections of the anterior end of the worm, but this was impossible in the present instance. 6H. P. Johnson, Proc. Boston Soc. Nat. Hist., vol. 29, 1901, p. 422. ¢It should be borne in mind that, at the period when the records by Schmarda (1861) and Grube (1859) were published, A. marina, or, asit was then almost universally called, A. piscatorum, was the only known caudate species of Arenicola with nineteen segments and thirteen pairs of gills. It was not until 1883 that Levinsen pointed out the characters which distinguish A. claparedii from A. marina. It is easy to under- stand that up to that time all specimens of Arenicola with nineteen segments and thirteen pairs of gills would be at once referred to the species A. marina, a practice which prevailed, with one or two exceptions, until little over ten years ago. No.1772. ANNELIDS OF THE ARENICOLIDZ—ASHWORTH. 9 Grube’s record of ‘‘Arenicola piscatorum Cuy.” from Callao is a mere mention of the name without any comment whatever. In the hope of finding this specimen I wrote to Professor Levinsen, of the Museum of Copenhagen (in the publications of which museum the record was published), but he informs me that he can not find any corresponding specimen in the museum collection. I have therefore no means of verifying or amending this record by reference to the actual specimen, but I consider that, as a record of A. marina, it should be only provisionally given. It is equally, if not more, probable that Grube was dealing with one of the species externally closely similar to A. marina, for instance, A. claparedw.* Ehlers records A. marina from Puerto Montt, Chile. Fortunately, the two specimens on which the record is based have been kept. Iam indebted to Professor Ehlers and to Doctor Michaelsen of the Natur- historisches Museum, Hamburg, to which institution the specimens belong, for the opportunity and permission to examine them. The worms are both dark colored, nearly black and badly preserved, 123 and 110 mm. long, respectively; the longer one is apparently incomplete, the posterior end of the tail bemg absent. In the number of their segments and the number and position of their gills, almost the only external characters available for reference, these specimens agree with A. marina, but on examining the internal organs of one of them which had already been partially dissected, it was evident that the specimen could no longer be referred to this species. ‘Ten ceso- phageal glands are present on each side; there are no pouches on the first septum, and statocysts could not be found in spite of most careful search. The internal organs of the other specimen were also examined; ‘seventeen cesophageal glands are present, but no septal pouches and no statocysts. The anterior region of one of the specimens was cut into serial sections, but statocysts could not be found. Had these vesicles been present in the living animal they would have been recognizable even though the tissue was so badly preserved. We may conclude that these specimens do not posses statocysts and that they belong to the species A. claparedu, the only species in which statocysts are absent. The gills, though not in a good state of preser- vation, are of the pinnate type and the neuropodia of the posterior branchial region are broad cushions not nearly reaching the mid- ventral line; both these characters confirm the diagnosis previously reached. The lobes of the prostomium are not sufficiently well preserved to be of service in diagnosis. Both these specimens are remarkable in that they possess six pairs of nephridia, while typical specimens of A. claparedw have only five pairs. Each of the first a See footnote ¢ on page 8. 10 PROCEEDINGS OF THE NATIONAL MUSEUM. VOL. 39. nephridia, in both specimens, is in contact at its anterior end with the third septum, but in none of them could a nephridial funnel be seen on the anterior face of this septum. Possibly these nephridia were, in life, not provided with funnels, i. e., they were incomplete or reduced, as not uncommonly happens in the case of the corresponding first nephridium in A. marina, but the very defective preservation of the specimens does not permit me to determine this point decisively. The vesicles of some of the other nephridia are strongly dilated; a similar condition has been observed in the nephridial vesicles of other species of Arenicola in the breeding season. The condition of the nephridial vesicles suggests that the specimens were sexually mature. They were collected on June 17, 1900. The records of A. marina from Vancouver Island, Puget Sound, and Puerto Montt are shown above to be invalid; the specimens on which the records were founded have been re-examined and shown to belong to the species A. claparedvi, under which they are now to be recorded. (See p. 12, footnote b, and p. 14.) Schmarda’s record of A. piscatorum from the Bay of Paita should also no longer be credited to this species, but probably to A. claparedi, and Grube’s record of A. piscatorum from Callao should be accepted with reserve. There is therefore no certain record of the occurrence of A. marina on the west coast of North or South America. I have examined specimens of A. marina from the following stations on the eastern coast of North America: Rigolet, Labrador. [83.] Cape Breton, Nova Scotia. [23662.] Halifax, Nova Scotia. [8931.] Eastport, Maine. [546, 587.] Gloucester, Massachusetts. [9365, 9368.] Provincetown, Massachusetts. [219.] ¢ Barnstable, Massachusetts. [No number.] Woods Hole, Massachusetts. Buzzard’s Bay, Massachusetts. [9367.] All the specimens, except those from Woods Hole, are in the col- lection of the U. S. National Museum. The numbers in brackets are the registration numbers. Before leaving the consideration of this species of Arenicola, I wish to refer briefly to a specimen in the collection of the Zoological Insti- a Webster and Benedict (U. S. Comm. Fish and Fisheries, part 9, Rept. Com- missioner for 1881, p. 725, Washington, 1884) state that A. marina was not found at Provincetown or at Wellfleet although it was carefully looked for, but the former author evidently made a further and successful search, for the bottle No. 219, which contains nineteen specimens, bears the note ‘‘From H. E. Webster.”’ no.1772. ANNELIDS OF THE ARENICOLIDU—ASHWORTH. 11 tute of the University of Vienna, for the opportunity of examining which I am indebted to the kindness of Prof. Dr. K.Grobben. This is a single specimen labeled, ‘‘.A. piscatorum, Chile, No. 253,’’ and added in pencil are the words ‘‘var. carbonaria.”’ The specimen is not in a good state of preservation, its muscles are very relaxed, and the tail region is broken into two pieces. Its total length is 300 mm., of which the tail, only a portion of which is present, re- presents 50 mm. The external characters and internal organs of this specimen agree absolutely with those of A. marina; it certainly belongs to this species. There is unfortunately no information available either as to the history or the exact place of capture of this worm. If it be really from Chile, it is, so far as I am aware, the only known specimen of Arenicola marina from the west coast of America. Summary of the distribution of Arenicola marina on American shores.—Arenicola marina has been recorded from a considerable number of stations on the eastern coast of North America, from Rigolet, Labrador, in the north to Noank, Connecticut, in the south. Although there have been a few records of A. marina from the west coast of North and South America, a re-examination of all the recorded specimens still in existence has shown that they are examples of A. claparedw. 'The only specimen of A. marina which I have seen from the west coast of America is one in the Museum of the Zoological Institute of Vienna, said to be from Chile, but no information regard- ing the history of the specimen or the exact station where it was captured is available. Further distribution of A. marina.—The shores of the White Sea, Siberia, Norway, Sweden, Denmark, Germany, Holland, Belgium, the British Isles, France, Portugal, the Mediterranean (a few stations only, for instance, Trieste), the Faroe Islands, Iceland, Greenland, the Marquesas and Kingsmill (Gilbert) Islands. ARENICOLA CLAPAREDII Levinsen. Nineteen chetigerous segments; thirteen pairs of gills, the first, which is on the seventh segment, may be small or absent; gills usually of the pinnate type but may be bushy; lateral lobes of the prostomium very large, much larger than the median lobe, and often folded in their anterior portion (fig. 2); neuropodia clearly visible in each segment, those of the posterior branchial region are wide antero- posteriorly, forming cushion-like pads, but are not so long as those of A. marina, so that neither the muscular ridge nor the groove which contains the crotchets, approaches the mid-ventral line; five 12 PROCEEDINGS OF THE NATIONAL MUSEUM. VOL. 39. pairs of nephridia which open on the fifth to the ninth segments?; several (four to sixteen) pairs of cesophageal glands, the anterior pair long and slender, the others shorter and more or less pear-shaped; no pouches on the first septum; statocysts absent. This last is a most noteworthy feature, and is diagnostic of the species, for all other known species of Arenicola possess statocysts. This species has been hitherto re- corded from only three stations on the American coast, namely: A. claparedii, Crescent City, California. F. W. Gamez, and J. H. AsnworrH. Quart. Journ. Micr. Sci., vol. 43, 1900, p. 423. A. claparedii, Crescent City, California. J. H. AsHWoRTH, Quart. Journ. Micr. Sci., vol. 46, 1908, pp. 773-774. A.claparedei, Puget Sound,> Washington. H. P. Jounson, Proc. Boston. Soc. Nat. Hist., vol. 29, p. 421, 1901. A. claparedii, Puget Sound, Washington. J. H. AsHwortH, Quart. Journ. Micr. Sci., vol. 46, Bae 1903, p. 74. Fig. 2.—A. CLAPAREDM, ANTERIOR END, A. claparedii,¢ Falkland Islands. E. M. DORSAL ASPECT OF SPECIMEN FROM CRES- Pratt, Mem. Proc. Manchester Lit. Philos. Soc., CENT CITY, CALIFORNIA, TO SHOW THE vol 45 no 13 p 12: no 14 i) 15 1901 PROSTOMIUM. L,LATERAL LOBE OF PRO- ek aeRO No rey A 9 = STOMIUM; M, MEDIAN LOBE; Mo, WoUTH: A. claparadii,¢ Falkland Islands. R. Vat- N}, FIRST NOTOPODIUM. x6. LENTIN, Mem. Proc. Manchester Lit.° Philos. Soc., vol. 48, no. 23, p. 9, 1904. The records from Crescent City and Puget Sound hold good, but those from the Falkland Islands must be transferred to the species A. assimilis. Miss Pratt’s record is founded on certain post-larval specimens of Arenicola taken by Mr. Vallentin in Stanley Harbor. These post-larval specimens were subsequently re-examined, two of them were sectioned, with the result that they were conclusively shown to be young stages of A. assimilis, var. affinis.4 Mr. Vallen- tin’s record is undoubtedly based on Miss Pratt’s. All the specimens of Arenicola which he collected in the Falkland Isiands were examined by me and it was from them that I described the new variety affinis of A. assimilis in the paper cited 4 (pp. 768-772). All the specimens belong to this variety; there is no example of A. claparedit among them. The specimens, five in number, from Crescent City, are in the Museum of Comparative Zoology, Cambridge, Massachusetts (register @ Out of 100 specimens examined I have seen only three which depart from this con- dition; in each of these there was also a nephridium opening on the fourth segment. These three specimens were from the west coast of South America, namely, from Co- quimbo (A. pusilla’? Quatrefages) and from Puerto Montt, Chile. See also p. 16. 6 This species was also obtained in Puget Sound by C. M. Child, but was erroneously stated to be A. marina. (Trans. N. Y. Acad. Sci., vol. 16, 1898, p. 387.) See p. 8. ¢ A. assimilis, var. affinis. See the lower part of this page. @J. H. Ashworth, Quart. Journ. Micr. Sci., vol. 46, 1903, pp. 764-768. no.1772. ANNELIDS OF THE ARENICOLID®—ASHWORTH. 13 number 91); there is another, doubtless from the same batch, in the Museum of the Zoological Institute, Géttingen (no. 27a). In four of those in the Harvard collection and in the Géttingen example the seventh segment is abranchiate, the first gill being borne on the eighth segment; the fifth Harvard specimen has a pair of very small gills on the seventh segment. These specimens vary in length from 105 mm. (of which the tail forms 13 mm.) to 207 mm. The tail of the latter specimen reaches the extraordinary proportion of 117 mm. A specimen from Puget Sound, from the collection of H. P. John- son, is preserved in the Harvard Museum (no. 956) and another is in the Department of Biology, University of California (no. 1066). The former is 94 mm. and the latter 60 mm. long, of which in each case the tail forms 13 to 14 mm., but the posterior end is apparently incomplete in both cases. In both these and in two other specimens from the same collection, given to me by Doctor Johnson, the first gill is borne by the seventh segment, but in one specimen there is a gill only on the right side of that segment, the corresponding gill of the left side being absent. There are examples of this species in the Smithsonian Collection from Constantine Harbor, Amchitka Island, Aleutian Islands (1047); Atka Island, Aleutian Islands (no number); Sand Point, Humboldt Bay, California (no number). The single specimen from Humboldt Bay, the one from Amchitka Island, and one of the four from Atka Island were opened in order to see the internal organs. All are typical examples of the species in regard to both their external and internal features, prostomium, neuropodia, nephridia, cesophageal glands, absence of septal pouches; statocysts could not be found on dissection of the anterior region of these specimens. In the case of the dissected specimen from Atka Island, the region in which the right statocyst is situated in those species which possess these organs, was excised and cut into serial sections, an examination of which proves that a statocyst is not present in this worm. I have recently examined seven specimens from Dutch Harbor, Unalaska (Harriman Expedition), and a dozen specimens collected at San Juan Island, in the Strait of Juan de Fuca. All are typical examples of A. claparedit, and in all the first true gill is present. The absence of statocysts was determined in a large specimen from Unalaska by dissection, and in the case of those from San Juan by examination of serial sections of the anterior end of one of the worms. The specimens from Unalaska are the largest examples of this species I have seen. Their length, 132 to 160 mm., is not specially remarkable, but they are of massive build, the four largest specimens have a girth (measured at the fifth segment) of 50 to 60 mm. 14 PROCEEDINGS OF THE NATIONAL MUSEUM. VOL. 39. Arenicola claparedw has been taken at three other points on the western seaboard of America, but was recorded under other names. To the records given above the following should therefore be added: A, marina, Vancouver Island. E. von MARENZELLER, Zool. Jahrb. Abth. Syst., vol. 3, 1888, p. 12. A. pusilla, Coquimbo, Chile. A. pE QuaTREFAGES, Histoire naturelle des An- nelés, p. 266, Paris, 1865. A. marina, Puerto Montt, Chile. E. Enters, Festschr K. Ges. Wiss. Gottingen, 1901, p. 176. I have shown above that a re-examination of one of von Maren- zeller’s specimens from Vancouver Island (see p. 7) and of the two specimens recorded by Ehlers from Puerto Montt (see p. 9) clearly proves that they do not belong to the species A. marina, but to A. claparedw. The specimen from Coquimbo is the type-specimen of A. pusilla, de Quatrefages, the characters of which were thus defined: ‘‘Annuli ebranchiati 9. Branchie magne ramosissime.’’ These diagnostic features are so inadequate that the position of this species with regard Fia. 3.—ANTERO-VENTRAL VIEW OF THE ANTE- s 0 RIOR PORTION OF ‘‘A. PUSILLA’”’? QUATREFAGES. to other species of Arenicola has IN THIS VIEW ALL THE PARTS, EXCEPT THE been quite indeterminable, and in- PROSTOMIUM, ARE SEEN SOMEWHAT FORE- ° . . SHORTENED. X8. 1, LaTerat Loses or eed it has been impossible to de- aMASNOMDL GabovE, Mo uooen: We weer (cide whether)or not) this ispecies. NEUROPODIUM. a valid one. I have recently made an exhaustive examination of this specimen as far as is possible without damaging its diagnostic fea- tures. It is small, slender, and incomplete, only the anterior region, as far back as the eleventh chetigerous annulus, being preserved. It is about 35 mm. long and 3 to 3.5 mm. in diameter. The first gill is borne on the eighth segment (not the tenth, as stated by de Quatrefages), but is small. The other gills are larger and tend toward the pinnate type. The first seven neuropodia are feebly developed, those of the succeeding segments are larger, and those of the tenth and eleventh segments form well-developed cushion-like ridges. The setz are very similar to those of Californian examples of A. claparedw. The prostomium is very fully everted, carried for- ward, and, as it were, disjiayed over the anterior end of the worm so that, in order to obtain a view of its lobes, an antero-ventral view is necessary (fig.3). The lateral lobes of the prostomium are in the form of two flattened disks, the edges only of which are visible in a NO. 1772. ANNELIDS OF THE ARENICOLIDH—ASHWORTH. 15 dorsal view of the worm; their flat faces are well seen on looking at the worm from the anterior aspect. The lateral lobes envelop the smaller median lobe posteriorly and laterally. As seen from the dorsal aspect (fig. 4) the lateral lobes of the prostomium are widely divari- cated, and immediately behind their point of union there is a small median structure (P) which lies in the nuchal groove, its posterior and slightly narrower end being hidden in the median portion of the nuchal organ. This little structure seems to have been regarded by Fauvel, who examined the external characters of the specimen about eleven years ago, as the median lobe of the prostomium,* but, as we have already seen, the true median lobe is anterior to the point of union of the lateral lobes (as it is in other species of Arenicola) while the rae lobe under discussion is posterior to this junction. This structure is the most posterior portion of the prosto- mium, only a very small portion, or none, of which is usually visible, be- cause it is generally almost or entirely hidden in the nuchal organ; the ex- treme protrusion of the prostomium in ATT) KWH this specimen has brought this posterior AT, raion the erent al) median portion into view e exami aan | aenony ys nation of the internal organs of this worm showed that there are six pairs of nephridia opening on the fourth to pe RR te care Pima epi: . QUATREFAGES the ninth segments, twelve cesophageal SHOWING THE WIDELY DIVARICATED Pianosen(cmaon each side), and that , - Ts lores (2) sx Te meen POSTERIOR PORTION (P) OF THE PRO- septal pouches are not present. Most STOMIUM (WN!) FIRST NOTOPODIUM; Corer examination of the region. in. 742) NvcHAD GEoovE. xs which statocysts should be looked for failed to reveal their presence ; to definitely establish their absence it would be necessary to make serial sections of a portion of the anterior end of the worm, but that is, of course, precluded in this case. I can only say, therefore, that, having very carefully searched for the statocysts, as far as was pos- sible in such a valuable specimen, I believe them to be absent. We have now the information at our disposal to enable us to deter- mine whether A. pusilla de Quatrefages is henceforward to be regarded as a valid species or whether it should be merged with one of the better- known species. It is clear that the diagnosis given by de Quatre- fages is erroneous, for gills are borne on the eighth and ninth segments. The comparatively high grade of development of the gills mentioned by de Quatrefages is not a definite character on which to found a a‘*Petit lobe médian triangulaire.’’ Mem. Soc. Nation. Sci. Nat. Math. Cher- bourg, vol. 31, 1899, p. 177. 16 PROCEEDINGS OF THE NATIONAL MUSEUM. VOL. 39. species; at least two other externally similar species of Arenicola have gills as highly developed as those of A. pusilla. The characters given in the original diagnosis of the species A. pusilla, one of them erroneous, the other insufficient, can not be held to establish it as a valid species, nor is there any feature of the internal anatomy which marks this specimen as unique. The most striking of the external features of A. pusilla is undoubt- edly its prostomium, the high grade of development of the lateral lobes of which is par- alleled, among known species, only in A. 2 claparedw. Figs. 5 and 6 represent the ante- y Aol igQ a) Wl ‘ rior end of a specimen Fig. 5.—ANTERO-VENTRAL VIEW OF fFiG. 6.—DORSAL VIEW OF of A. claparedii from THE ANTERIOR END OF A SPECIMEN THE ANTERIOR END OF THE i J | OF A. CLAPAREDIT FROM CRESCENT SAME SPECIMEN. X6. FoR Crescent City, Califor- f CITY, CALIFORNIA. X6. For LET- LETTERING SEE FIG. 4. nia (Harvard Collec- TERING SEE FIG. 3. 4 tion), which has a prostomium of the same type as that of A. pusilla, although owing to contraction in the oral region at the time of preservation the ventral portions of the lateral lobes of the Californian specimen have been brought nearer together and have unduly compressed the median lobe. Nevertheless it is obvious that the Californian specimen and A. pusilla have prostomia of an identical type, and that the latter specimen, judged by its prostomium, is to be referred to the species A. claparedi. This diagnosis is confirmed by the examination of the internal organs, the results of which are given above (p. 15), particularly by the absence of statocysts, which is a feature of special significance. The only point of difference in regard to the internal organs between A, pusilla and A. claparedvi is that in the former there are six pairs of nephridia and in the latter typically only five pairs, there being usually no nephridium opening on the fourth segment. But this sole difference is not important; certainly it is not one which would justify the separation of two otherwise identical forms. The two examples of A. claparedvi from Puerto Montt (see p. 10) have also six pairs of nephridia; this may be a character of the Chilean race of the species. We may conclude that the specimen of A. pusilla de Quatrefages is a fragment of a small example of A. claparedit Levin- sen; had it been complete it would probably have been about 80 to 90 mm. in length. A more detailed account of “A. pusilla” and dis- cussion of its affinities will be found in a forthcoming paper, by the present writer, in Annales des Sciences Naturelles, Zoologie, ser. 9, vol. 10, Paris, 1910. In his account of the specimens of A. claparedw from Puget Sound Johnson ® states that “the only notable points of difference between ® Proc. Boston Soc. Nat. Hist., vol. 29, p. 1901, p. 421. no.1772. ANNELIDS OF THE ARENICOLIDA—ASHWORTH. 17 the Puget Sound specimens and those from Naples are the vastly greater size—at least eight times as great—of the former and the smaller number of cesophageal cceca or pouches in the latter.’ The difference in regard to the number of cesophageal coca in speci- mens from Naples and the western seaboard of America seems to be a clear one, and is sometimes even striking. Neapolitan specimens seldom have more than four pairs of cceca, but I have seen only one American example with as few as five pairs; the others had six, eight, nine, and ten pairs, and Johnson records specimens with fifteen and sixteen pairs, and one with sixteen cceca on the right and eighteen on the left side. Johnson’s remark regarding the comparative size of Neapolitan and American specimens is not in agreement with my experience; possibly he had been supplied with very small Neapolitan examples. There is not so great a difference in the length of specimens from the two regions. I have seen sixteen specimens from Puget Sound, including four from Johnson’s collection, the largest of which is 103 mm. long (of which the tail forms 28 mm.), and 11 of them are less than 50 mm. long. I have recently had a Neapolitan example 97 mm. long (of which the tail forms 20 mm.), and the average length of nine specimens which have just passed through my hands is 80 mm. (of which the tail forms 17 mm.). Lo Bianco states that examples of this species from the Bay of Naples attain a length of 150 mm.* The longest American specimen of A.claparedw I have seen is one from Crescent City, California, in the Harvard collection, which is 207 mm. long, but this great length is largely accounted for by the unusual extent of the tail, which meas- ures no less than 117 mm. American specimens have generally a thicker body wall than Neapolitan ones, and are stouter; among the scores of living and preserved Neapolitan examples of this species which I have examined I never saw any whose girth approached that of the massive specimens from Unalaska. (See p. 13.) Summary of the distribution of Arenicola claparedu on American shores.—Arenicola claparedi is now known from several stations on the western seaboard of America, namely, the Aleutian Islands (Amchitka, Atka, Unalaska), Vancouver Island, San Juan Island, Puget Sound, Crescent City and Humboldt Bay, California; Coquimbo and Puerto Montt (Chile). Further distribution—Naples, Ossero (Adriatic), Angra Pequena (southwest Africa), and North Japan. ARENICOLA ASSIMILIS Ehlers. Twenty chetigerous segments; thirteen pairs of gills, the first, which is on the eighth segment, may be small or absent; the gills may be bushy or may tend toward the pinnate type, but are seldom clearly and typically pinnate; median lobe of the prostomium @ Atti R. Accad. Sc. fis. mat., Napoli, vol. 5, ser. 2, 1893, no, 11, p. 9, Proc. N.M.vol.89—10——2 18 PROCEEDINGS OF THE NATIONAL MUSEUM. vou. 39. large or moderately large, lateral lobes in the form of a V, the limbs of uniform width, not dilated anteriorly, though they may be bent (figs. 7, 8); neuropodia similar to those of A. claparediw; six pairs of nephridia, which open on the fourth to the ninth segments; several (6 to 15) pairs of cesophageal glands, the anterior pair long and slender, the others smaller and more or less pear-shaped; no pouches on the first septum; a pair of large statocysts in the peristomium, which open to the exterior; statoliths numerous, spherical or rounded secreted chitinoid bodies. Fic. 8.—A. ASSIMILIS, VAR. AFFINIS, ANTERIOR END, DORSAL ASPECT, OF SPECIMEN FROM THE FALKLAND ISLANDS. THE PROSTOMIUM IS PRO- Fig. 7.—A. ASSIMILIS, ANTERIOR END, DORSAL TRUDED TO ITS FULLEST EXTENT. 6. JD, LAT- ASPECT, OF SPECIMEN FROM USCHUAIA. THE ERAL LOBE OF PROSTOMIUM; MM, MEDIAN LOBE; PROSTOMIUM IS IN A STATE OF NORMAL EXTEN- 1, FIRST NOTOPODIUM; P, MEDIAN POSTERIOR SION. FOR LETTERING SEE FIG. 8. X6. PORTION OF PROSTOMIUM; PH, PHARYNX. ARENICOLA ASSIMILIS, var. AFFINIS Ashworth. Nineteen chetigerous segments; thirteen pairs of gills, the first, which is liable to reduction, on the seventh segment. Statoliths numerous, consisting of either foreign bodies, such as quartz grains, or of rounded secreted chitinoid bodies. Other characters as given above for A. assimilis Ehlers. A. assimilis has been recorded from— California.e E. Exuers, Polychaeten, p. 104, in Hamburger Magalhaensische Sammelreise, Hamburg, 1897. Punta Arenas (Strait of Magellan), Lapataia Nueva (Beagle Channel) and Uschuaia (Tierra del Fuego). E.Exuers, Polychaeten, p. 104, in Hamburger Magalhaensische Sammelreise, Hamburg, 1897; also Festschr. K. Ges. Wiss. Géttingen, 1901, p. 178. Susanna Cove (Strait of Magellan).o E. Exuers, Polychaeten, p. 104, in Ham- burger Magalhaensische Sammelreise, Hamburg, 1901; also Zool. Jahrb., Suppl. 5, Fauna Chilensis, vol. 2, pp. 265, 269, 1901. South Georgia. E. Enters, Polychaeten, p. 104, in Hamburger Magalhaensische Sammelreise, Hamburg, 1897; also Festschr. K. Ges. Wiss. Gottingen, 1901, p. 178. @ Transfer to A. claparedii, see p. 19. 6 Transfer to var. affinis, see p. 19. no.1772. ANNELIDS OF THE ARENICOLIDE—ASHWORTH. 19 Schmarda’s record of A. piscatorum from the Bay of Paita (see p- 8) is included by Ehlers under the species A. assimilis (Festschr. Géttingen, p. 178). The variety afinis of A. assimilis is recorded from the ee stations on or near the American coast: Falkland Islands. J. H. AsHworru, Quart. Journ. Micr. Sci., vol. 46, 1903, pp. 764-772. Susanna Cove (Strait of Magellan). J. H. AsHwortu, Mitth. Kongl. Zool. Mus. Berlin, vol. 4, 1910, p. 352. The specimen on which the record of A. assimilis from California is based is in the Géttingen Museum and, by the courtesy of Professor Ehlers, I have been permitted to examine it. Professor Ehlers in- formed me that this is a duplicate from Professor Agassiz’s collection, which was sent to Géttingen to be examined. The remaining speci- mens were returned to Professor Agassiz and are doubtless those which [ have at present in my hands from the Harvard collection (bottle No. 91). I have compared the Géttingen specimen with the Harvard examples and find that they agree in every respect. In the former there are nineteen chetigerous segments, the first of the twelve pairs of gills is on the eighth segment; there are five pairs of nephridia which open on the fifth to the ninth segments; there is no nephridiopore on the fourth segment; the lateral lobes of the prosto- mium are well developed, and in pigmentation and general appear- ance this specimen agrees absolutely with the Harvard specimens of A. claparedii from California (see p. 12). I have no hesitation, after making this direct comparison, in confirming the opinion which I expressed in 1903 ¢ that Ehlers is in error in recording A. assimilis from California; the record should be transferred to A. claparedit. It is very probable that Schmarda’s specimen, regarded by Ehlers as A. assimilis, was really A. claparedii; the point at which the specimen was obtained (the Bay of Paita) is within the known range of A. claparedvi, but is over 3,000 miles north of the nearest station from which A. assimilis has been recorded. Through the kindness of Doctor Michaelsen, of the Hamburg Museum, I have been enabled to examine specimens of A. assimilis from Punta Arenas, Lapataia Nueva, Uschuaia, and South Georgia, and the specimens from Susanna Cove, now in the Kéniglische Zoolo- gisches Museum, Berlin, were recently entrusted to me for examina- tion by Director Brauer. The examples from Punta Arenas, Uschuaia, and South Georgia have twenty chetigerous segments and are typical specimens of A. assimilis, but those from Lapataia Nueva and Susanna Cove—the original examples determined by Ehlers, as the accompanying labels testify—have only nineteen segments and are referable to the variety affinis. @ Quart. Journ. Micr. Sci. vol. 46, 1903, p. 774. 20 PROCEEDINGS OF THE NATIONAL MUSEUM. VOL, 39. A post-larval specimen with nineteen segments and therefore referable to the variety affinis, was recorded by Ehlers from Uschuaia among the “‘roots’’ of Fucus. Summary of the distribution of Arenicola assimilis on American shores.—Typical examples of A. assumilis have been found at Punta Arenas (Strait of Magellan), Uschuaia (Beagle Channel), and South Georgia. . Examples referable to the variety affinis are known from Uschuaia (a gill-less post-larval specimen), Lapataia Nueva (Beagle Channel), Susanna Cove (Strait of Magellan), and the Falkland Islands. Further distribution.—Typical examples of A. assimilis have not yet been recorded from any other stations than those mentioned above. The variety affinis is, however, found also at Otago Harbor, New Zealand, and Stewart Island,* and at Kerguelen. ?® Remarks on Arenicola assimilis—The three caudate species of Arenicola—marina, claparedi, and cristata, present a remarkable con- stancy in the number of their cheetigerous segments and in the seg- ments on which gills are borne. Only in rare cases is an additional chetigerous segment found at the posterior end of the branchial region, and very rarely does this segment bear gills. Among the thousands of specimens of A. marina which have passed through my hands during the last few years I have seen only three with a com- plete chxetigerous and branchiate twentieth segment. Ehlers separated A. assimilis from A. marina because in the former the first gill was not on the seventh but on the eighth or ninth seg- ment. But the number of chetigerous segments is a still more strik- ing character, and, in view of the constancy in the number of these segments in the other caudate species, it seemed to me that the occurrence of examples of A. assimilis with twenty and others with nineteen chetigerous segments called for recognition. It was on these grounds that, in 1903, I established the variety afinis for those examples of A. assimilis with only nineteen chetigerous segments. The varietal and typical specimens are, however, very closely related; the only differences between them are in regard to the number of segments and the position of the first gill. A. assimilis (in the wider sense, that is, including the variety) is clearly the characteristic species of the southern regions. It extends from Tierra del Fuego, by way of the Falkland Islands, South Georgia, and Kerguelen, to New Zealand. The most northerly station from which it has been recorded is Otago Harbor, about 46 degrees south. @ Unpublished record, from the manuscript of the present writer. 6 This specimen from Kerguelen, recorded by Grube (Monatsb. K. Preuss. Akad. Wiss. Berlin, aus dem Jahre 1877, 1878, p. 511) as A. piscatorum Cuvier var., is now in the Kéniglische Zoologisches Museum, Berlin. I have recently examined it, and ~ conclude that it is to be referred to the species A. assimilis var. affinis. no.1772. ANNHLIDS OF THE ARENICOLIDZ—ASHWORTH. 21 The distribution of this species presents a remarkable parallel to that of certain Oligochetes. Beddard® has pointed out that the characteristic earthworms of New Zealand are Acanthodrilide, and that the same family is equally characteristic of Patagonia. This close resemblance, in regard to their earthworms, between Patagonia and New Zealand is accentuated by the fact that the only earthworms known from the intervening localities, the Falklands, South Georgia, Marion, and Kerguelen islands, belong to the genus Acanthodrilus. Beddard regards these facts as evidence in favor of a former greater extension northward of the cirecumpolar antarctic continent, and he is inclined to believe that this region did not include the Cape of Good Hope. The evidence afforded by the distribution of the earthworms points to a more recent communication between Patagonia and New Zealand than between either of these countries and the Cape of Good Hope. . The occurrence of Arenicola assimilis in the southern extremity of South America, the Falklands, South Georgia, Kerguelen, and the southern portion of New Zealand supports the view that there was formerly a more extensive antarctic continent. It is noteworthy that the only species of Arenicola known from South Africa is A. lovent Kinberg,? which is widely different from A. assimilis in almost every character, a fact which sug- gests that the antarctic con- tinent was not continuous with the Cape of Good Hope, to which conclusion, as we have already seen, Beddard was inclined to come from a study of the earthworms. ARENICOLA CRISTATA Stimpson. Arenicola antillensis LUTKEN. ra gab Fic. 9.—A. CRISTATA, ANTERIOR END, DORSAL ASPECT, OF A SPECIMEN FROM FLORIDA. X4. Ty LATERAL LOBE OF Seventeen cheetigerous BES PROSTOMIUM; L. N,LIP OF NUCHAL ORGAN; M, MEDIAN ments; eleven. pairs of gills, LOBE OF PROSTOMIUM; NN}, FIRST NOTOPODIUM; PH, * HA the first situated on the sev- "™“**** enth segment; gills large, pinnate, their axes generally joined basally by a web-like membrane; the median lobe of the prostomium (fig. 9) a ¥. E. Beddard. A Monograph of the Order of Oligocheta. Oxford, 1895, p. 154. See also, by the same author, A Text-book of Zoogeography. Cambridge, 1895, pp. 60, 170. 6 This species, the anatomy and characters of which have not been described, is at present under investigation by the writer. 22 PROCEEDINGS OF THE NATIONAL MUSEUM. Vou. 39. larger than the lateral ones; the neuropodia of the first few seg- ments are very small, in fact, it is often impossible to see them on the first three segments and that of the fourth segment is a very short slit, the neuropodia of the posterior branchial region are long dorso-ventrally, they and their grooves reach almost to the mid- ventral line; six pairs of nephridia which open on the fifth to the tenth segments; one pair of esophageal glands, conical, cylindrical, or clavate and comparatively short; a pair of well-developed, muscular finger-shaped pouches projecting backward from the first septum; a pair of statocysts, closed, spherical or ovoid sacs, each containing a single, large, secreted, statolith. This species has been recorded from the following stations on the American coast: A. cristata, Woods Hole, Massachusetts. C. M. Cup, Arch. Entwickelungsmech, vol. 9, 1909, p. 587. A. cristata, Woods Hole, Massachusetts. R.S. Linum, Mitth. Zool. Stat. Neapel, vol. 17, 1905, p. 344. A. cristaia, North Falmouth, Massachusetts. C.M. Camp, Arch. Entwickelungsmech, vol. 9, 1900, p. 587. A. cristata, New Jersey. H. E. Wesster, 32d Ann. Rept. New York State Mu- seum, p. 117, Albany, 1879. A. cristata, Anglesea, New Jersey. J. E. Ives, Proc. Acad. Nat. Sci. Phila. for 1890, p. 73, Philadelphia, 1891. A. cristata, Beaufort, North Carolina. E. A. ANpREws, Proc. U. S. Nat. Mus., vol 14, p. 289, Washington, 1892. A. cristata, Charleston Harbor, South Carolina. W. Strmpson, Proc. Boston Soc. Nat. Hist., vol. 5, 1856, p. 114. A. antillensis, Florida, Captiva Key, Florida. E. Enters, Mem. Mus. Comp. Zo6l. Harvard, vol. 15, 1887, p. 173. A. cristata, Captiva Key, Florida. F. W. Gametr and J. H. ASHWORTH, Quart. Journ. Micr. Sci., vol. 43, 1900, p. 423. A. cristata, Manatee River, Florida. J. E. Ives, Proc. Acad. Nat. Sci., Phila. for 1890, p. 74, Philadelphia, 1891. A. cristata, Bermuda. H. E. Wessrer, Bull..U. 8. Nat. Mus., No. 25, p. 523, Washington, 1884. A. cristata Bermuda. A. E. Verriin, Trans. Conn. Acad. Arts and Sci., vol. 10, pt. 2, 1900, p. 599; vol. 11, 1902, p. 39; vol. 12, 1907, p. 147. A. cristata, Bluefields, Jamaica. F. W. Gampir and J. H. Asnworrs, Quart. Journ. Micr. Sci., vol. 43, 1900, p. 423. A. antillensis, St. Croix (Santa Cruz). C. Litrxcmn, Vid. Medd. Naturh. For. Kjébenhavn, Aaret 1864, 1865, p. 121. This is one of the most readily recognized species of Arenicola by reason of its seventeen chetigerous segments, eleven pairs of large gills, and the beautifully pinnate character of the gills which is so marked a feature of most of the specimens. The character of the gills of Liitken’s specimens was so striking that he held it to be a sufficient basis for the delimitation of a new subgenus for which he proposed the name Péeroscolex. no.1772. ANNELIDS OF THE ARENICOLIDZ—ASHWORTH. 20 I have recently examined specimens from the stations given below; to those in the Smithsonian collection the register numbers are affixed: Beaufort, North Carolina. [149, 4861, 42896.] Florida. [211.] Key West, Florida. [4531.] Pensacola, Florida. [27229.] Bermuda. [34478.] Curacao. [38798.] San Pedro, California. Monterey Bay, California. The specimen recorded from Pensacola, Florida, extends the distribution of this species westward along the coast of Florida to its extreme western limit. The one from Curacao is the first example to be recorded from the coast of South America. Of special interest is the finding of this species on the western seaboard of America, which greatly extends its known range, for this species has hitherto been found only at Naples, on the eastern coast of the United States from Woods Hole southward, in Bermuda, and in the West Indies. In one of the specimens from San Pedro there is a small gill on the right side of the sixth segment. This is the onlyspecimenof Arenicola, out of thousands examined, in which I have seen a gill on the sixth segment. Arenicola cristata is a giant among Polycheta. One of the examples from Beaufort has a length of 385 mm. (of which the tail forms 115 mm.) and the specimen from Pensacola reaches a length of 460 mm. (of which the incomplete tail forms 80 mm.) and its girth at the sixth segment is about 60mm. The largest example of this species which I have had is one from Woods Hole, which attains the great length of 515 mm. (of which the tail is 190 mm.) and a girth of 75 mm. In this, as in other caudate species of Arenicola, the tail is obscurely segmented, at segmental intervals there is a larger annulus and upon this, in most American examples, there are, on the ventro-lateral or lateral region, hollow outgrowths of the body wall usually in the form of thumb-shaped processes. The processes at the base of the tail, 1. e., in the first tail segment, are usually the largest, and occasionally one of them, generally the most dorsal one, is branched at its distal end, resembling a small gill; this branched process corresponds in position, and seems to be serially homologous, with the gills on the preceding segments. Neapolitan examples of this species do not possess these long caudal processes; slightly larger epidermal papille are present on the larger annulus of each tail segment, but they are of the same order as the papille of the other annuli. 24 PROCEEDINGS OF THE NATIONAL MUSEUM. VOL. 39. The type-specimen of A. cristata Stimpson is apparently no longer in existence. I have made inquiry for it from the curator of the museum of the Boston Society of Natural History, but I am in- formed that it is not in that museum; that it was probably in Stimp- son’s own collection, and, if so, was probably destroyed when the Chicago Academy of Sciences was burned in the great Chicago fire. The specimens on which Liitken based his new species A. antillensis are preserved in the Universitets Zoologiske Museum, Kjébenhayn. — I have recently examined two of them. They are typical American specimens of A. cristata. Summary of the distribution of Arenicola cristata on American shores.—A. cristata is found on the eastern coast of the United States, from Woods Hole, Massachusetts, southward. It extends along the western coast of Florida to Pensacola; it is also recorded from Bermuda, Jamaica, Santa Cruz, and Curacao, and from two stations, San Pedro and Monterey Bay, on the Californian coast. Further distribution.—Naples, Suez,* Barrow Island (northwest Australia) ,¢ Misaki (Japan).¢ ARENICOLA GLACIALIS Murdoch. This species was founded to contain specimens of Arenicola taken on the International Polar Expedition to Point Barrow, Alaska. Mur- doch’ remarked that the worms in question are closely allied to A. marina, but that they have only six setigerous segments anterior to the gills and eleven branchiferous segments, instead of seven and thirteen, respectively, as in A. marina. (He should have said six and thirteen, for there are normally six anterior abranchiate seti- gerous segments in A. marina.) He described each gill as a cluster of about fifteen simple cirri and noted that the tail of each worm, which forms about a third of the length, is without tubercles or other appendages. Five specimens were picked up on the beach on September 12 and 13, 1882, after a fresh westerly gale, and two mutilated ones were obtained from the gullet of an eider duck which had been diving on one of the sandy patches, in about 3 fathoms, just above the station. There are no figures of this species, and the above constitutes the whole of the information regarding it. Consequently the systematic position of A. glacialis with regard to other species is in doubt, and indeed its validity as a species has been questioned. In the number of its anterior abranchiate segments and of its branchiferous seg- ments this species agrees with A. cristata, and on this ground some @ Unpublished records from the manuscript of the present writer. ® Report Intern. Polar Exped. to Point Barrow, Alaska, p. 155, Washington, 1885; also in Proc. U.S. Nat. Mus., vol. 7, p. 522, Washington, 1885. no.1772. ANNELIDS OF THE ARENICOLIDA—ASHWORTH. 25 writers? have considered A. glacialis to be a synonym of A. cris- tata. Von Marenzeller,> however, places it under A. marina. The specimens, which are in the collection of the Smithsonian Institution, are now in a very fragile condition; only one of them is complete, another is in two parts, held together by a strand of muscle, and the other three are fragmentary. I have made as complete an examination of these worms as is possible in their present condition, and having regard to their value as types. The inspection of the internal organs was made on one of the imperfect examples (No. 3 in the list below), which consists of the anterior end as far back as the seventh chetigerous annulus and was almost dropping to pieces. The examination of the specimens shows that A. glacialis is a distinct and valid species. As Murdoch’s description is totally inad- equate, I propose to give an account of the species as far as I have been able to investigate it on the material at my disposal. The material consists of: 1. A complete specimen 90 mm. long, of which the tail, which is strongly contracted, forms 11 mm. 2. A specimen 105 mm. long, of which the tail forms 45mm. This specimen is broken, about the ninth segment, into two pieces, held together by a strand of muscle. 3. A portion of an anterior region, 26 mm. long, as far back as the seventh chetigerous annulus. 4, A portion of an anterior region, 14 mm. long, as far back as the third cheetigerous annulus. 5. A fragment similar to No. 4, 12 mm. long. 6. A posterior region, consisting of eleven segments and tail, x Bes longing to either No. 4 or No. 5. 7. A posterior region, consisting of eight segments and tail. 8. A branchial region. 9. A portion of a tail. All the specimens are dark brown to nearly black in color. Each of the specimens, No. 1 and No. 2, possesses seventeen che- tigerous segments, the last eleven of which are provided with gills. It is evident, from an examination of all the specimens, that the gills of this species are small. The axes of the largest gills are not more than 2 mm. in length, measured from their origins to the tips of their terminal gill filaments. The first gill may be very small, as in specimen No. 2, and the last gill is smaller than the preceding ones. @P. Fauvel, Mem. Soc. Nation. Sci. Nat. Math. Cherbourg, vol. 31, 1899, pp. 169, 171. F. W. Gamble and J. H. Ashworth, Quart. Journ. Micr. Sci., vol. 43, 1900, p. 428. b Zool. Jahrb. Abth. Syst., vol. 3, 1888, p. 15. 26 PROCEEDINGS OF THE NATIONAL MUSEUM. VOL, 39. The gills are best studied in No. 8. The most fully expanded gill of this specimen is shown in fig. 10. It consists ef nine or ten axes, which arise from a short, curved, common basal structure situated immediately posterior to the notopodium. The largest axis (see fig, 10, C) bears five branches, each of which divides once, i. e., forks. The longest of the resultant gill filaments are thumb or finger shaped structures, not more than about 0.7 mm. in length, and the shortest are mere tubercles. This axis, measured from its origin to the tip of the fila- ments, is 2 mm. in length. The smaller axes of this gill bear fewer branches, only two or three, and these may be simple, 1. e., undivided, distally (see fig. 10, A,B). Another gill, from fragment No. 6, and one of the most fully expanded gills present in the whole of the specimens, is shown in fig. 11. This gill is smaller than the Fig. 10.—A. GLACIALIS, NINTH GILL OF NO. 8. : f ° THREE OF THE AXES ARECUTorr. x15. One just described. It is composed of seven axes, the longest and largest of which (A) bears three branches, each of which bifurcates, and one of the two so formed again divides, but the other does not. In this way three groups, each of three gill filaments, are clustered at the end of the axis. From the base of this axis to the tip of its filaments is a distance of about 1 mm. In all the gills examined the branches of the gill axis are not given off right and left, but are clustered at the end of the axis, so that each axis and its branches look almost like a hand and fingers (see, for instance, figs. 10,11,B). This character distinguishes the gills of A. glacialis from those of any other species, and especially from those of A. cristata, the only other Fig. 11.—A. Gracias, THIRD GILL OF No. 6. THREE OF species in which there are eimai lati cas eleven pairs of gills, for in A. cristata the branches borne on the gill axes are much more numerous and are regularly arranged along the sides of each axis, forming a typically pinnate gill. The gills of A. glacialis are also much smaller than those of A. cristata. The gill axes of a specimen of the latter 110 mm. long are from 4 to 5mm. in length, compared with 1 to 2 mm. in examples of A. glacialis of about the same size. no.1772. ANNELIDS OF THE ARENICOLID#—ASHWORTH. 27 The prostomium is well seen in only two of the specimens. It con- sists (see fig. 12) of two large lateral lobes of almost uniform width (i. e., they are not markedly dilated anteriorly), which unite pos- teriorly to form a wide median portion of the prostomium. Between the lateral lobes is situated the me- dian one, which is small. Its trans- verse diameter is less than one-third that of the whole prostomium and its antero-posterior diameter is also N. Gr; short (about .5 mm.). The nuchal organ exhibits the usual relations to the prostomium. In the specimen figured (from No. 5) the lip (L. N.) of the nuchal organ is well everted. The an- J nulation Flt aa ‘