Co Ivl5 WITH THE COMPLIMENTS THK AUTHOR SmifPS INSTITUTION FOR KlOl/OGICAL ^f *jf^, . SMITHSONIAN INSTITUTION ()p i;s| VKRSITV OF CAL1F0J UNITED STATES NATIONAL MUSEUM LA JOLUA, C Bulletin 100 VOLUME 1, PART 4 CONTRIBUTIONS TO THE BIOLOGY OF THE PHILIPPINE ARCHIPELAGO AND ADJACENT REGIONS REPORT ON THE CHAETOGNATHA COLLECTED BY THE UNITED STATES FISHERIES STEAMER "ALBATROSS" DURING THE PHILIPPINE EXPEDI- TION, 1907-1910 By ELLIS L. MICHAEL Of the Scripps Institution for Biological Research, La Jolla, California WASHINGTON GOVERNMENT PRINTING OFFICE 1919 SMITHSONIAN INSTITUTION UNITED STATES NATIONAL MUSEUM Bulletin 100 VOLUME 1, PART 4 CONTRIBUTIONS TO THE BIOLOGY OF THE PHILIPPINE ARCHIPELAGO AND ADJACENT REGIONS REPORT ON THE CHAETOGNATHA COLLECTED BY THE UNITED STATES FISHERIES STEAMER "ALBATROSS" DURING THE PHILIPPINE EXPEDI- TION, 1907-1910 By ELLIS L. MICHAEL Of the Scripps Institution for Biological Research, La JoJla, California (Oft WASHINGTON GOVERNMENT PRINTING OFFICE 1919 /*) y LIBRARY UNIVERSITY OF CALIFORNIA 6/1 SANTA BARBARA CC, MS" REPORT ON THE CHAETOGNATHA COLLECTED BY THE UNITED STATES FISHERIES STEAMER "ALBATROSS" DURING THE PHILIPPINE EXPEDITION, 1907-1910. By ELLIS L. MICHAEL, Of the Scripps Institution for Biological Research, La Jolla, California. INTRODUCTION. This paper is based upon the chaetognatha collected by the United States Bureau of Fisheries steamer Albatross during the Philippine expedition of 1907-1910. Chaetognatha were taken at 46 stations scattered between the parallels of 21 31' north and 5 36' south latitude, and between the meridians of 117 53' east and 127 44' east longitude. The collection is represented by 12 species of Sagitta, of which one, Sagitta philippini, is apparently new; one of Pterosagitta; two of Eukrohnia; and one of Rrofinitta. The species of Sagitta, in order of the number of specimens obtained, are: S. enflata, (2,800); S. Tiexaptera, (700); 8. ferox, (600); S. pulchra, (550); 8. neglecta, (425); S. bedoti, (350); S. decipiens, (160); S. serratodentata, (100); S. planktonis, (85); 8. minima, (2); and one each of S. macrocephala and 8. philippini. Pterosagitta is represented by 32 specimens of P. draco; EukroJinia by 6 specimens of E. Jiamata and 5 of E. rich- ardi; and KroTinitta by 3 specimens of R. subtilis. Most of the material was preserved in formalin and is in excellent condition. In some cases, however, alcohol was used. Specimens preserved with it are distorted and the tannin extracted from the corks of the containers has turned most of them quite black, ren- dering identification uncertain and, in some cases, impossible. But, in so far as the collection permits, tables of diagnostic measurements are given for each species, enough measurements being made on each individual to enable reconstruction of its outline. Otherwise, the species are not further described except for those concerning which need of description is indicated by the literature. In lieu of descrip- tions, however, references are given to those published in other reports, particularly to Ritter-Zahony's (1911) revision of the group, which, with two or three exceptions, is adopted as my basis of classification. In order to make the report as serviceable as possible, keys are supplied for identifying not only those species obtained during the Philippine expedition but all the species in the group. Several new species and five new genera have been described since my last (1911) report. In addition, Ritter-Zahony (1911) has called attention to and admirably illustrated specificities in the presence and absence 235 236 BULLETIN 100, UNITED STATES NATIONAL MUSEUM. of rays in the lateral fins. In devising the keys these specificities are used and, except for Sagitta maxima (Conant), which I still find impossible to separate from. S. lyra, every species recognized as valid by Bltter-Zahony (1911) and several others described subsequent to his report are included. There is great need of keys for identifying poorly preserved material, but the minute structure of seizing jaws and other skeletal parts of the head, upon which such identification depends, is still undescribed in nearly half the species. The keys included in this report are therefore adapted only to the identification of well-preserved specimens. The area covered by the expedition is too large and the hauls were too scattered to yield definite information concerning the dis- tribution of the various species obtained. As pointed out elsewhere (1916, p. XVIII), variability in plankton distribution is enormous, and hydrobiological relations are too complex to be revealed without frequently repeated collections in very restricted areas and searching hydrographic observations corresponding in time and place to each net haul. Even though thousands of individuals of one species and none of another be obtained by a single haul, no conclusion is justi- fied other than that the former species was obtained and that the latter was not obtained. Such data afford no adequate evidence for concluding that the former species is more abundant in or more typical of that particular locality than is the latter. Had 20 or 30 hauls been made at each station with nets of similar filtering capaci- ties, sufficient evidence for such conclusions might have been obtained. But rarely more than one haul was made at each station, so the data 3 r ield little else than records of occurrence. These records are given for each species. In comparing the species occurring in the Philippine region with those obtained from the vicinity of San Diego (the only coastal region of the Pacific off either of the American continents from which chaetognaths have been described), two interesting and suggestive facts come to light: First, those species obtained in largest numbers from the Philippines are those which, as a rule, occur rarely, if at all, in the San Diego region, and the opposite. Second, of those species common to both regions, the number of teeth is greater in Philippine specimens. This is contrary to what might have been expected. For, judging from the fact that chaetognatha collected under the auspices of the Scripps Institution from as far south as 23 north, off Lower Cali- fornia, are essentially like those within the San Diego region proper, and, realizing that this is but two degrees north of the northern boundary of the Philippine area from which chaetognatha were obtained, one would naturally infer a close relationship between the Philippine and San Diego faunas. To find it quite the reverse is therefore suggestive of a fundamental and far-reaching difference in CHAETOGNATHA COLLECTED BY STEAMER ALBATROSS. 237 the other faunas, and so of the whole economic and fisheries situa- tions of the coastal waters on opposite sides of the Pacific at corre- sponding latitudes. Extensive exploration of the Pacific, particu- larly of the coastal waters of Central and South America, is needed, however, to discover the full significance of what is here so clearly indicated, and it is regretted that no chaetognatha are described from these regions. But, in spite of this, it seems probable, from the meager data at hand, that conclusions reached through explora- tions in the western Pacific are largely inapplicable to the waters of the eastern Pacific, and the opposite. Some space is therefore taken at the close of the paper in briefly comparing the Philippine and San Diego chaetognatha. It is hoped this will emphasize the need of more extensive explorations, and that it may add its mite toward a better understanding of the fishery problems of the Pacific Ocean. KEYS FOR THE IDENTIFICATION OF THE CHAETOGNATHA. Ritter-Zahony (1911) has been the last investigator to thoroughly revise the chaetognatha. He recognizes six genera Sagitta, Ptero- sagitta, Spadella, Eukrohnia, HeterokroTinia, and KroTinitta. Sub- sequently Germain and Joubin (1912) added two more Pseudo- sagitta and EroTinitella. All are probably valid with the possible exception of Spadella and Pseudosagitta, the status of which is baf- fling. Most of the differences given by Ritter-Zahony (1911) between Pterosagitta draco and Spadella cepJialoptera are certainly no greater than that between those species of Sagitta in which the skeletal part of the vestibular ridge is present and those in which it is absent, and this difference is clearly of subgeneric rather than generic value. On the other hand, Conant's (1895) description of Spadella schizop- tera, although fragmentary and wholly unsatisfactory, reveals a close affinity between that species and S. cepJialoptera and at the same time makes the genus to which it belongs unmistakably distinct from Pterosagitta. Furthermore I have seen specimens of neither S. cepJialoptera nor S. schizoptera, and it seems best, therefore, to tentatively recognize Spadella as valid in spite of the fact that the characteristics by which its one well known species, S. cepJialoptera, differs from P. draco seem of subgeneric value. The validity of Pseudosagitta is ably discussed by Baldasseroni (1915, p. 101), who holds its single new species P. grimaldi to be synonymous with Sagitta lyra. The differential characters described by Germain and Joubin (1912, p. 6) are certainly such as to suggest this synonymy and I find myself in agreement with Baldasseroni. In the following keys seven genera are therefore recognized, of which Sagitta is represented by 23 species, Eukrohnia by three, Spadella by two, and each of the others by one: Pterosagitta draco (see p. 264), HeteroJcroJinia mirabilis (Ritter-Zahony, 1911, p. 42), 238 BULLETIN 100, UNITED STATES NATIONAL MUSEUM. KroJinitta subtilis (see p. 269), and KroJinitella loureei (Germain and Joubin, 1912, p. 133). KEY TO GENERA. 1. Two pairs of rows of teeth 2 1. Teeth entirely absent, or only one pair of rows present 4 2. Two pairs of lateral fins, the posterior pair being partly on body and partly on tail. Fins completely or incompletely rayed; anterior and posterior pairs sometimes connected by narrow membrane Sagitta. 2. One pair of lateral fins, or two pairs of which the posterior one is entirely confined to the tail-segment. Fins completely rayed 3 3. One pair of lateral fins confined entirely to tail-segment. Collarette massive, extending to tail-septum and spreading out over fins. Greatest width slightly anterior to tail-septum, and exceeding half that of the body. Ventral transverse muscles absent. Anterior and posterior teeth both exceed six in number Pterosagitta. 3. One pair of lateral fins confined entirely to tail-segment, or two pairs the posterior one of which is confined to tail-segment. Collarette present but not massive. Greatest width slightly behind head, less than half that of body. Ventral transverse muscles present in body-segment only. Neither anterior or posterior teeth exceed five in number Spadella. 3. One pair of lateral fins partly on body and tail. Collarette absent. Ventral transverse muscles present in both body and tail Heterokrohnia. 4. Lateral fin begins at or in front of ventral ganglion and extends onto tail but never to seminal vesicles. Ventral transverse muscles present in anterior third of body. One pair of rows of teeth EukroJinia. 4. Lateral fin begins about half-way between ventral ganglion and tail-septum, and extends fully to seminal vesicles. Ventral transverse muscles absent. . 5 5. Head small but wider than body. Less than 50 per cent of fin in front of tail-septum. Width of body less than 8 per cent of total length. One pair of rows of teeth. Seizing jaws delicate, but not filliform Krohnitta 5. Head small, narrower than body. More than 60 per cent of fin in front of tail-septum. Width of body exceeds 9 per cent of total length. Teeth absent. Seizing jaws filliform Krohnittella. KEY TO SPECIES OF SAGITTA. 1 . Collarette absent 2 1. Collarette present 13 2. Shaft of seizing jaw serrated S. serratodentata. 2. Shaft of seizing jaw not serrated 3 3. Anterior and posterior fins confluent 4 3. Anterior and posterior fins separated 5 4. Both pairs of fins entirely ray less throughout at least their anterior thirds; tail usually exceeds 15 per cent of total length S. lyra. 4. Fins only rayless adjacent to body but not along outer margins; tail usually less than 15 per cent of total length S. gazelle. 5. Anterior fins longer than posterior fins S. philippini. 5. Posterior fins longer than anterior fins 6 6. Anterior fins entirely rayless; rays of posterior fin perpendicular to body. S. minima. 6. Anterior fins not entirely rayless; rays of posterior fin directed obliquely to body 7 7. Anterior fins extend nearly if not quite to ventral ganglion 8 7. Anterior fins never extend within half their length of ventral ganglion 9 8. Both pairs of fins with rays throughout; mature ovary short and thick, not reaching anterior limit of posterior fins S. setosa. CHAETOGNATHA COLLECTED BY STEAMER ALBATROSS. 239 8. Anterior extremities of both pairs of fins rayless; anterior fin also rayless throughout a narrow strip adjacent to body; mature ovary long and narrow, extending nearly if not quite to ventral ganglion S. serratodentota 9. Tail 28-40 per cent of total length; posterior teeth 20-38, rarely as few as 17. S. macrocephala. 9. Tail less than 28 per cent of total length; posterior teeth rarely exceed 16. . 10 10. Posterior fins extend nearly if not quite to seminal vesicles; both pairs of fins rayed throughout 11 10. Posterior fins never extend more than distance from tail-septum to seminal vesicles; both pairs of fins rayless throughout a narrow strip adjacent to body 12 11. Body very transparent; interval between anterior and posterior fins always less than half the length of posterior fin S. setosa. 11. Body semi-translucent but never transparent; interval between anterior and posterior fins usually greater than half the length of posterior fins S. elegans 12. Vestibular ridge composed entirely of papillae; anterior teeth 0-4, rarely 5; posterior teeth 0-6 S. hexaptera, 12. Vestibular ridge provided with usual skeletal parts; anterior teeth 5-12, rarely less than 6; posterior teeth 7-18, rarely less than 10 S. enflata. 13. Collarette short, not extending over half-way from neck to ventral ganglion. 14 13. Collarette long, extending more than distance from neck to ventral ganglion 22 14. Anterior fins longer than posterior fins 15 14. Posterior fins longer than anterior fins 18 15. Posterior fins never extend nearly to seminal vesicles S. decipiens, 15. Posterior fins extend nearly if not quite to seminal vesicles 16 16. Body transparent; anterior fins exceed 30 per cent of total length of animal. S. pulchra. 16. Body opaque or semi-translucent, but never transparent; anterior fine less than 30 per cent of total length of animal 17 17. Both pairs of fins rayed throughout S. neglecta. 17. Anterior extremities of both pairs of fins rayless S. bedoti. 18. Interval between anterior fins and ventral ganglion exceeds one-fifth length of fins S. bipunctata. 18. Anterior fins extend nearly if not quite to ventral ganglion 19 19. Collarette inconspicuous, extending less than one-fourth distance from neck to ventral ganglion 20 19. Collarette well developed, extending between one-fourth and one-half distance from neck to ventral ganglion 21 20. Pronounced constriction at tail-septum; sexually mature at a length of 5-6 rnm ." S. tennis 20. No constriction at tail-septum; never sexually mature under 9-10 mm.-S'./ncfenci. 21. Anterior teeth 10-18; exceeding number of posterior teeth S. helenae. 21. Anterior teeth 4-9; less than number of posterior teeth S. hispida. 22. Collarette never extending to ventral ganglion S. neglecta. 22. Collarette extending from neck to seminal vesicles S. caUfornica. 22. Collarette extending fully to ventral ganglion, frequently onto anterior fins, but never beyond anterior quarter 23 23. Posterior fins longer than anterior fins S. regularis. 23. Anterior fins longer than posterior fins 24 24. Less than 50 per cent of posterior fin in front of tail-septum S. ferox. 24. More than 50 per cent of posterior fin in front of tail -septum S. planktonis. KEY TO SPECIES OF SPADELLA. 1. One pair of lateral fins entirely on tail-segment. Ventral transverse muscles present throughout entire body-segment S. cephaloptera. 240 BULLETIN 100, UNITED STATES NATIONAL MUSEUM. 1. Two pairs of lateral fins, the posterior pair much larger and confined entirely to tail-segment, beginning directly behind seminal receptacle. Anterior pair extends throughout posterior third of body, beginning directly in front of seminal receptacles. Ventral transverse muscles present only in anterior half of body S. schizoptera . KEY TO SPECIES OF EUKROHNIA. 1. Anterior two-thirds of lateral fins rayless. Posterior extremity of fins less than half-way from tail-septum to seminal vesicles. Width of body less than 8 per cent of total length of animal. Seizing jaws delicate; some- times serrated 2 1. Lateral fins delicately rayed throughout. Posterior extremity of fins at least three-fourths way from tail-septum to seminal vesicles. Width of body exceeds 12 per cent of total length of animal. Seizing jaws massive; not serrated - ^ E. richardi. 2. Eyes without pigment. Seizing jaws 8-10, gently curved throughout. Points sickle-shaped E. hamata. 2 Eye with pigment. Seizing jaws 11-13, sharply curved in anterior quarter. Point slightly curved toward edge of jaw, but not sickle-shaped E.fowleri. SPECIES OBTAINED DURING THE PHILIPPINE EXPEDITION. Genus SAGITTA Quoy and Gaimard. SAGITTA PHILIPPINI, new species. Plate 34, figs. 1-4. General appearance. To the naked eye S. pJiilippini, when placed in formalin upon a white background, appears white in color, scarcely distinguishable from the background. Its head and tail, and in less degree its ovaries, assume a brownish-yellow color in marked con- trast to the body proper. On a black background the head, ventral ganglion, ovaries, tail, seminal vesicles, and to a less extent the intestine appear much more opaque than the body, which resembles ground glass. The lateral fins and tail fin are so transparent as to be invisible to the naked eye. In degree of opacity S. philippini resembles 8. decipiens more than any other species, although it is perhaps less transparent. Characters. Collarette absent. Neck conspicuous. Lateral fields prominent. Body flabby, not retaining its form well; widest behind center, tapering gradually forward toward head and backward toward tail. No constriction at tail-septum. Ovaries, even when immature (pi. 34, fig. 1), extend beyond posterior end of anterior fin. Corona ciliata not observed. Anterior fins (pi. 34, fig. 1) rayless throughout anterior half of fin. They are longer and narrower than posterior fins, and extend ante- riorly beyond posterior end of ventral ganglion. Form triangular, the position of greatest width being in the caudal quarter of fin. Interval from anterior to posterior fins slightly greater than maximum width of body. Posterior fins do not extend caudally to seminal vesicles. More than 50 per cent of fin in front of tail-septum. Form triangular, the position of greatest width being at or just behind tail-septum. CHAETOGlSrATHA COLLECTED BY STEAMER ALBATROSS. 241 Vestibular ridge (pi. 34, fig. 3) well developed with large papillae. Wing of ridge covers all but the first two or three teeth, the notch extending to the fourth or fifth. External process long and blunt. Anterior teeth (pi. 34, fig. 2), nine in number. They are short, broad, closely set, and diverge but little distally. Posterior teeth (pi. 34, fig. 3), 20 in number. They are long, narrow, closely set, and diverge even less than do the anterior teeth. Seizing jaws (pi. 34, fig. 4), six in number. Point with an oval base imbedded between 20 and 25 per cent of its height into shaft. Top of shaft and base of point converge toward edge of jaw. Edge of shaft provided with narrow crest. Pulp canal central and slightly swollen at base of point. Pulp evenly distributed. Only a single specimen was obtained. Its measurements follow: Length 13 mm. Width 5. 5 per cent of length. Length of tail 23 per cent of length. Tip of tail to posterior end of ventral ganglion 70. 5 per cent of length. Length of ventral ganglion 6. 5 per cent of length. Length of posterior fin 22 per cent of length. Width of posterior fin 4. 5 per cent of length. Interval from anterior to posterior fins 6 per cent of length. Per cent of posterior fin in front of tail-septum 60 per cent. Length of anterior fin 30 per cent of length. Width of anterior fin 2.5 per cent of length. Anterior fin extends beyond posterior end of ventral ganglion by 2 per cent of length. Ovary extends beyond posterior end of posterior fin by 14 per cent of length. Anterior teeth 9-9 Posterior teeth 20-(?) Seizing jaws 6-6 The single specimen (Cat. No. 17801, U.S.N.M.) was taken from the surface May 14, 1908, off Uanivan Island, at station D 5240, latitude 6 49.5' north and longitude 126 15' east. The same haul also yielded 130 S. enflata, 105. ferox, 6 S. Tiexaptera, 1 P. draco, and 75 S. decipiens. 8. pJiilippini bears a strong resemblance to the latter species, but differs from it in several important details: In the first place, it has no trace of a collarette, although this structure, while not pronounced in S. decipiens^ is conspicuous. Again, the ovaries of S, philippini, though not fully mature, extend nearly to the middle of the anterior fin, while in 8. decipiens they do not, when mature, extend beyond the anterior limit of the posterior fin. Further, the posterior fin of S. philippini is rayed throughout, while in 8. decipiens (pi. 35, fig. 8) the anterior fourth or fifth of the fin is rayless. Lastly, the seizing jaws of the two species are quite different (pi. 34, fig. 4, and pi. 37, fig. 22), the jaw in S. pMlippini being provided with a narrow but conspicuous crest, which is missing in S. decipiens. 242 BULLETIN 100, UNITED STATES NATIONAL MUSEUM. Altogether, these differences would seem to justify the description of a new species, even though it may later prove to be synonymous with S. decipiens. Had more than one individual been obtained, I should feel certain of the validity of S. philippini, but as the matter stands this single specimen might with as much justification be regarded as an abnormal S. decipiens. SAGITTA ENFLATA Grassi. Plate 38, fig. 28. Sagitta enflata GRASSI (1883), p. 13. FOWLER (1906), p. 8. RITTER-ZAHONY (1911), p. 13. MICHAEL (1911), p. 28. This species is represented in the Philippine collection by approxi- mately 2,800 individuals. They usually exceed 20 mm. in length, and the largest taken measures 31.5 mm. In the San Diego region, on the other hand, the specimens rarely exceed 18 mm. in length, the largest recorded (Michael (1911, p. 29)) measuring only 21 mm. Again, the anterior teeth number 6 to 11, typically 7 or 8, in the Philippine specimens, while they number 4 to 8, typically 6 or 7, in San Diego specimens. Similarly, the posterior teeth number 9 to 15 in the Philippine specimens, the usual number being 14, while they number 6 to 12 hi San Diego specimens, the usual number being 10 or 11. In all other respects, however, specimens from the two localities are in agreement, and the Philippine specimens agree in size and number of teeth with specimens described by Fowler (1906 p. 8) from the Siboga region. One puzzling fact is revealed by the Philippine collection. The ovaries in most of the larger specimens are barely approaching maturity, only one case of complete maturity having been discovered in individuals exceeding 20 mm. in length; but among individuals under 16 mm. in length many have mature ovaries (pi. 38, fig. 28). In my San Diego report (1911, p. 56) a table is given of specimens of S. enflata arranged in three groups according as their ovaries were mature, approaching maturity, or remote from maturity. In the first group the specimens varied in length between 12.5 and 19.5 mm., in the second between 15 and 17.5 mm., and in the third between 8 and 15.5 mm. Obviously, these facts are open to two interpretations: First, the ovaries hi San Diego specimens attain maturity only once and that after a length of 12 mm. is reached; and, second, the ovaries in the same individual become mature periodically, first when the individual is not less than 12 mm. in length, and subsequently after it has grown larger. If the second interpretation is eliminated, how is the relation between length of individual and stage of maturity of the ovary in the Philippine specimens to be accounted for ? It could, of course, be readily explained on the assumption that two species had been confused, but I am unable to discover any other differences even remotely indicative of more than one species. In Table 1, for CHAETOGNATHA COLLECTED BY STEAMER ALBATROSS. 243 example, No. 9 alone had mature ovaries, but its measurements agree with the other larger and immature specimens. TABLE 1. Measurements of Sagitta enflata. 1 2.1.2 23. 7 23. 6 23.2 21.1 20.6 KS IS. 4 15.8 is. a 10.2 10.7 10.7 8.7 10.4 9. 9 12.4 12.8 11.8 12.5 9.7 7.7 74.5 74 69.8 73 74.8 71.6 70. 4 73. C 71 71.7 72.3 Posterior fin. 17.T) 10.9 17.3 8.3 4.8 66.7 62.5 80.3 fi2 64.1 64.8 65 63.4 S3. 70 59.9 Anterior fin. 11.8 12.2 14.9 13.4 15. S 13.6 15.6 17.7 11.6 16.2 10.7 15.4 16.9 18.2 10-7 8-7 9-11 7- 7 8- 9 14-14 15-15 14-15 13-13 8-8 8-8 9-9 8-8 8-8 8-7 9-8 9-9 8-9 7-8 1 All measurements made in per cent of total length of animal. ' Per cent of posterior fin in front of tail-septum. Distribution. S. enflata was collected from 39 stations, or 85 per cent of the 46 stations at which chaetognaths were taken, a total of approximately 2,800 specimens having been obtained. Of the 39 stations 8, or 21 per cent, were mesoplanktonic, while 31, or 79 per cent, were epiplanktonic, and 22, or 56 per cent, were surface stations. From the 8 mesoplanktonic stations a total of 247 specimens was obtained, or an average of 31 per station, while from the 31 epiplank- tonic stations a total of 2,559 specimens, or an average of 83 per station, was obtained, and from the 22 surface stations a total of 1,476 specimens, or an average of 67 per station, was obtained. These data, together with the fact that all subsurface hauls were made with open nets, make it clear that S. enflata occurs typically in the upper epiplankton of the Philippine region. The northernmost record of its capture during the Philippine expe- dition is 14 21 '.5 north and 120 23 '.3 east in the China Sea near southern Luzon. The southernmost record is 5 36M south and 127 T. 6 east in Buton Strait. The easternmost and westernmost records are 127 44'.0 east and 1 3'.0 south, south of Patiente Strait, and 117 53' east and 21 31' north in the China Sea off Hongkong. The largest number (950) was taken February 7, 1908, at 8.05 in the morning from 25 fathoms by an open 0000 grit-gauze net towed horizontally 9 fathoms above the bottom of the Sulu Archipelago, near Basilan Island, at 6 44'.2 north and 121 47' east. Other species taken in the same haul are: 318 S. bedoti, 217 8 ferox, 116 S. pulcJira, 2 S. Jiexaptera, and 19 P. draco. The complete records of its capture are given in table 2 : 244 BULLETIN 100, UNITED STATES NATIONAL MUSEUM. -7-7- 111 3 S' - * te-k ^* * k * * I^ t- C* O X^ t-^H^W MM t- <- t- - 3 D^^ 3 S g 5 2 D p 3 3 3 3 OQ C/S QQ CQ 00 CO Ou QQ QQ QC OQ OQ-QO aaaaaaaaasaaaaaaaaaefiseaaaaaaaaasaaasaa joVujvrcj NCN llll;; ill! "..* P i T> '"& S+j feT3 fi^^3 CSI IN ^N M ^ COCOMCQOQM PPP'PPPPPP'P'PPPP " al "So ft! |i ii CHAETOGNATHA COLLECTED BY STEAMEE ALBATROSS. 245 SAGITTA HEXAPTERA d'Orbigny. Sagitta hexaptera D'ORBIGNY (1843), p. 140. FOWLER (1906), p. 11. RITTEE- ZAHONY (1911), p. 7. MICHAEL (1911), p. 30. This species is the second most abundant and frequent obtained during the Philippine expedition, being represented by approximately 700 specimens. Most are large though immature, the largest speci- men measuring 45 mm. in length. Curiously, both anterior and posterior teeth in many specimens are entirely missing. My first impression was that although not seen they must be present, but careful dissection of several heads has made it certain that the teeth are actually missing. This was not noticed in San Diego specimens, two being the smallest number of either anterior or posterior teeth recorded in my (1911) report. Again, Fowler (1906, p. 13) lists the number of teeth in 42 specimens, one being the smallest number of either anterior or posterior teeth recorded. Ritter-Zahony (1908, p. 10), however, while recording three as the smallest number of anterior teeth, gives four instances in which the posterior teeth in individuals 33, 34, 36, and 38 mm. in length were missing. In attempting to account for the peculiar variability in number of teeth, Fowler (1906, p. 14) explains their absence in the Philippine speci- mens. He says: I believe the explanation to lie mainly (perhaps not entirely) in the length and slenderness of the teeth; many of them are probably torn out by the roots; certainly many are broken off short, for their bases may be seen still in place. As an additional weakness, the posterior teeth in older specimens often appear not to be attached to the bony bar with which they are united in other species, but to lie at some distance from it in a superficial plate of chitinous material. TABLE 3. Measurements of Sagitta hexaptera. 1 17.0 65.1 19.1 65.8 19.1 72 10.6 14.5 9.6! 15.51 3.3 11.8! 11.2 2.3 7.2 14.31 2 14.1 1 15.2 2.1 6.9l 13.6! 3 10. o! 2.3 14.9; 39.2 .62-16 14.6 55. 5 1 2.16+ 4.2 11.8 19.7; 1.09-23.2 11.4 24. 5i 1.12; -19.1 -21.6 .46; -32. 3 .46- .711-27 1. 311-12. 4 All measurements made in per cent of total length of animal. Per cent of posterior fin in front of tail-septum. The + signifies extension beyond anterior end of ganglion; the signifies distance from posterior end of ganglion. 246 BULLETIN 100, UNITED STATES NATIONAL MUSEUM. Distribution. S. Jiexaptera was collected from 26, or 57 per cent, of those stations at which chaetognaths were captured. Of these, 8 were mesoplanktonic, while 13, or one-half, were surface stations. From the 8 mesoplanktonic stations a total of only 70, or an average of 9 to each station, was obtained, while the 18 epiplanktonic stations yielded 642, or an average of 36 per station, and the 13 surface sta- tions yielded 491, or an average of 38 per station. It is evident, therefore, that S. Jiexaptera occurs typically in the upper epiplankton of the Philippine region. S. Jiexaptera, often confused with S. elegans and with the large variety (S. maxima) of 8. lyra, is a eurythermal, nearly cosmopolitan species found typically in the lower epiplankton and mesoplankton of the arctic, sub-arctic, north temperate, tropical, and south tropical Atlantic, the south temperate and tropical Indo-Australian, and the north temperate and sub-antarctic Pacific oceans. Its northern and southern limits of distribution are 74 north and 28 south, while the extremes of temperature recorded in connection with its capture are 29 C. and 6 C. A statement frequently encountered in the literature is that surface Chaetognatha of the arctic seas would be found, if at all, in the mesoplankton of temperate and tropical regions, the implication being that temperature plays the all- important part in delimiting the vertical distribution of a species. Obviously, the typical occurrence of S. Jiexaptera in the upper epi- plankton during the Philippine expedition contradicts this statement, which contradiction is further supported by Ritter-Zahony (1911, p. 54), who says: "Es gibt keinen einzigen verbiirgten Fundort der 8. Jiexaptera aus dem Epiplankton der Meere nordlich von 40 N." Rather do the facts point in quite the opposite direction, that is that surface 8. Jiexaptera of tropical and sub-tropical regions are found, if at all, hi the lower epiplankton and mesoplankton of arctic and sub-arctic regions. However, until consistency of identifica- tion of the species obtained during the various expeditions is attained, and until the vertical distribution of the species in diversified regions is critically studied, no conclusion as to the part played by tempera- ture or any other environmental influence, in controlling its distri- bution throughout the world, is justified. The northernmost and westernmost record of its capture during the Philippine expedition is 21 31' north and 117 53' east in the China Sea, off Hongkong. Its southernmost record is 5 36 M south and 122 7'.6 east, in Buton Strait; and its easternmost record is 127 44' east and 1 3' south, south of Patiente Strait. The largest number (153 + ) was taken August 11, 1909, at 7.49 in the evening, from the surface between Siquijor and Bohol Islands, at 9 2 7 '.5 north and 123 38' east. Other species obtained at the same station are: 169 8. enflata, 128 S. pulcJira, 50 S.ferox, and 14 S. bedoti. CHAETOGtfATHA COLLECTED BY STEAMER ALBATROSS. 247 SSS33S38 : Ml "11 3 3 I Sg ^ rH " s s s s a a a a a a a a a a a a a a a s a a a e a a 5 >-' >-^ >-' u >> >> >> >' > >> si M"^ bj) 1 ^ i^Ilgllslll j| i MilMMM in .!& IJ : :a : :g :p^|!a||!ll! a : 3lt 10 OOO010U3 OS CM 00 00 U5 M r-(lO ^C:-rt-i-':O^:^fOpXC--Ol^t 248 BULLETIN 100, UNITED STATES NATIONAL MUSEUM. SAGITTA MINIMA Grass!. Plate 37, figs. 16-17; plate 38, fig. 29. Sayitta minima GRASSI (1881), p. 213; (1883), p. 15. KRUMBACH (1903), p. 637.- RlTTER-ZAHONT (1911), p. 25. This species is represented by only two specimens (Cat. No. 17925, U.S.N.M.), both of which are apparently sexually mature. Both were obtained on April 3, 1908, from the surface at station D. 5195, off northern Cebu Island, 10 47' north and 124 6'.5 east. Except for the mature ovaries they are almost inseparable to the naked eye from small S. enfiata. Microscopic examination, however, reveals several marked differences. The species is redescribed on the basis of these two specimens. Collarette absent. Body flabby and widest on a level with anterior end of posterior fins, tapering gradually toward head and tail. Con- striction at tail-septum slight or absent. Ovary (pi. 38, fig. 29) short, not extending to anterior end of posterior fins. Ova large and arranged in a single row within the ovary. Corona ciliata not observed. Anterior fins shorter and narrower than posterior fins and entirely rayless. They fall short of reaching the posterior end of ventral ganglion by nearly two-thirds the length of fins. Interval from anterior to posterior fins approximately equal to half the length of anterior fins. Posterior fins (pi. 38, fig. 29) with rays arranged perpendicular to the body. Interval from fins to seminal vesicles 3 to 5 per cent of total length. More than 50 per cent of fins in front of tail-septum. Position of greatest width behind tail-septum. Vestilular ridge (pi. 37, fig. 16) provided with low, regular papillae, one for each tooth. Wing covers all except the first tooth, the second being just barely covered. Notch extends to fourth tooth. Ex- ternal process apparently missing. Anterior teeth, 4 to 5 in number (2 to 5 according to Ritter-Zahony, 1911, p. 26). They are very closely set and not diverging much distally. Posterior teeth (pi. 37, fig. 16), 10 or 11 in number (6 to 14 according to Ritter-Zahony). They are not so closely set as the anterior teeth, but are more divergent distally. Seizing jaws (pi. 37, fig. 17), 8 or 9 in number (7 to 8 according to Ritter-Zahony). Point with an oval base, inserted into shaft by less than one-fifth its height. Tip of point curved toward its edge. Base of point and top of shaft parallel. Pulp-canal central, with a swollen place below base of point. Pulp evenly distributed through- out canal. Aside from the number of teeth and seizing jaws and length of tail, only one of the two specimens is well enough preserved to permit accurate measurements. CHAETOGNATHA COLLECTED BY STEAMER ALBATROSS. 249 Length in mm 9. 5. Tail: Length 20. 8 per cent of length . To ventral ganglion 68. 5 per cent of length. Posterior fin : Length 17. 6 per cent of length . To seminal vesicles 3. 2 per cent of length. To anterior fin 9. 1 per cent of length. Proportion in front of tail-septum 59. 1 per cent. Anterior fin: Length 16. 5 per cent of length. To ventral ganglion 11. 7 per cent of length. Number of anterior teeth. . . : 5-4. Number of posterior teeth ll-(?) Number of seizing jaws 8-9. The other specimen measured 8.9 mm.; its tail measured 20.7 per cent; the number of anterior teeth are 4-5; the number of posterior teeth, 10-10; and the number of seizing jaws, 9-8. SAGITTA SERRATODENTATA Krohn. Sagitta serratodentata KROHN (1853), p. 272. FOWLER (1905), p. 58. HITTER ZAHONY (1911), p. 22. MICHAEL (1911), p. 39. Approximately 100 specimens were obtained, none of which is sexually mature. The number of anterior and posterior teeth is greater than recorded by Fowler (1905, p. 58) in specimens taken from the Bay of Biscay, or by myself (1911, p. 40) in specimens taken from the San Diego region. In specimens between 7 and 11 mm. in length Fowler records 3 to 6 anterior and 2 to 10 posterior teeth, while in Philippine specimens between the same lengths, the anterior teeth number 8 to 11 and the posterior teeth 13 to 24. The San Diego specimens are considerably larger, those recorded varying in length between 10 and 17 mm.; but the number of teeth is inter- mediate, the anterior teeth numbering 6 to 9 and the posterior teeth 13 to 19. The species appears to be unusually variable. TABLE 5. Measurements of Sagitta serratodentata. 1 10. 5 10.3 10.2 20.0 24 SB 90 24.5 9B 2:.. 5 27 25.5 Posterior fin. Anterior fin. 20. 28 27 27.5 SI 28 28. fl 51.5 55 52 52 50.5 50 49.5 52 55 IS. 5 22.5 21 22 21.5 10-10 23-22 11-11 24-23 11-10 20-18 8 -9 20-? 9 -8 9 -9 9 -8 15-14 9 -9 18-19 9 -? 15-? 9 -9 14-13 17-16 . o> 6-5 6-6 6-6 6-6 6-6 7-7 0-6 6-7 6-1 7-7 1 All measurements made in per cent of total length of animal. i Per cent of posterior fin in front of tail-septum. 59318 19 Bull. 100 2 250 BULLETIN 100, UNITED STATES NATIONAL MUSEUM. < i-l 0* Oi iO r-teO a a a a a a a a BB5IBS Distribution. S. serratodentata was collected from only eight stations. All except three were mesoplanktonic, but the largest number of specimens (73) were taken from the surface. The rec- ords are given in Table 6 : SAGITTA MACROCEPHALA Fowler. Sagitta macrocephala FOWLER (1905), p. 65. RITTER-ZAHONY (1911), p. 31. A single distorted specimen was ob- tained. Its measurements follow: Length in mm 7.8 mm. Width in per cent of length 14.4 per cent. Tail in per cent of length 37. 4 per cent. Length of posterior fin 24. 8 per cent. Per cent of fin in frontof tail- septum 45.5 percent. Width of posterior fin 7 . 7 per cent. Interval from anterior to posterior fin 5.5 per cent. Length of anterior fin 15.0 per cent. Width of anterior fin 2. 7 per cent. Number of anterior teeth 7-7 Number of posterior teeth 25-26 Number of seizing jaws 11-11 Ventral ganglion, corona ciliata, and ovaries not observed. The specimen differs conspicuously in width from those drawn by Fowler (1905, pi. 5, fig. 16) and Ritter-Zahony (1911, fig. 37). In width the Philip- pine specimen measures 14.4 per cent of the length, while Fowler draws it 7.8 per cent, and Ritter-Zahony 5 per cent. However, the Philippine specimen is clearly immature, neither ovaries nor seminal vesicles being visible. More- over, it is poorly preserved, some portions of the body being distorted and others torn away. These facts are probably responsible for the ex- cessive width. Unfortunately the points of all' seizing jaws were broken off, so that their structure could not be determined. CHAETOGNATHA COLLECTED BY STEAMER, ALBATROSS. 251 The single specimen (Cat. No. 17926, U.S.N.M.) was obtained November 6, 1908, in the China Sea, in the vicinity of Formosa, at station D. 5320, 20 58' north and 120 3' east by an open 0000 grit gauze net towed at 3.18 in the afternoon in 500 fathoms for twenty minutes. SAGITTA PULCHRA Doncaster. Plate 35, fig. 5; plate 37, figs. 19, 23. Sagitta pulchra DONCASTER (1902), p. 213. FOWLER (1906), p. 17. RITTER- ZAHONY (1911), p. 21. Approximately 500 individuals were obtained, and few, if any, are sexually mature. In body length and number of teeth they agree remarkably well with specimens described by Fowler (1906) from the Siboga region. He records 5 to 10 anterior and 9 to 15 posterior teeth in specimens between 9 and 22 mm. in length, and in Philippine specimens between 9 and 30 mm. in length, the anterior teeth number 5 to 9, and the posterior teeth 10 to 13. The Philippine specimens are, on the whole, so well preserved that the species is redescribed. Collarette (pi. 35, fig. 5) conspicuous but short, varying in length from one-twentieth to one-tenth the length of the animal. Its length is less than twice the body width and it extends between one- fourth and one-half the distance from neck to ventral ganglion Neck pronounced but rendered inconspicuous by the collarette. Muscles thin but strong. Lateral fields large. More transparent than any other species having a collarette, and similar in transparency to S. enflata. Its body, however, is firmer than that of 8. enflata and is approximately half as wide. Width greatest between one-half and three-quarters the distance from head to tail-septum, tapering gradually forward and more rapidly backward. Slight constriction at tail-septum. Tail 18 to 25 per cent of total length of animal. Corona ciliata not observed. Anterior fins (pi. 35, fig. 5) longer and narrower than posterior fins extending anteriorly beyond posterior end of ventral ganglion, frequently beyond its middle, and rarely beyond its anterior end. No rays except in posterior quarter of fin. Interval from anterior to posterior fins usually less than two-thirds width of body, varying from slightly less than one-half to slightly more than the width. Posterior fins (pi. 35, fig. 5) rayless anteriorly. They extend posteriorly nearly if not quite to seminal vesicles, the interval never exceeding 2.5 per cent of total length of animal. More than 50 per cent of fin in front of tail-septum, varying from 50.5 to 64 per cent. Broadly triangular in form, and widest at or slightly behind tail-septum. Vestibular ridge (pi. 37, fig. 19) provided with large regular papillae, the apices of which usually terminate in two minute spines. Wing of ridge covers all except first two or three teeth. Notch extends 252 BULLETIN 100, UNITED STATES NATIONAL MUSEUM. to fourth or fifth tooth. External process one-third to one-half length of ridge and approximately four times longer than broad. Anterior teeth 5 to 9, closely set and diverging distally. Posterior teeth (pi. 37, fig. 19) 10 to 13, not so closely set nor so divergent dis- tally as anterior teeth. Seizing jaws (pi. 37, fig. 23) 5 to 6 in number. Point with oval base inserted little more than one-tenth its height into shaft. Base of point and top of shaft parallel. Edge of shaft provided with broad thin crest. Pulp-canal central, with pulp evenly distributed throughout. TABLE 7. Measurements of Sagitta pukhra. 1 19. r, 5.5 18 7 17.5 6. j 19.5 5.5! 19.5! 6 I 20 5.5| 21 6 I 19.5 5.5! 21 20 70 72 71 70.5 71) 72 71 70. 5 67 eg li Posterior flu. Anterior fin. 2.5 64 37.5 3.5 56 ! 37.5 2.5! 63.5 36 2 i 58. 5' 36 3 I 59.5 32 1.5: 56.5 35 4.5 59.5i 33.5 7 54 | 32.5 3 58. 5 ! 36 4.5 55.51 33 3.5 60 I 32 ' 56 32 62 i 32 51.1 31.5 1 I 50.5f 32 +4.5 2.5 +1 2.5j +1.5 3 +1.5 2 + .5 3.5 +2 3.51 +1.5 2.5 + .5 3 | +1.5 +1.5 +2.5 Collar- ette. +3.5! +2 +2 10-10 11-12 12-13 0-7 Mi M 7-6 10-11 7-8! 10-11 6-6 8-8 13-13 12-12 12-13 11-12 10- 9 i All measurements made in \ * Per cent of posterior fin in front of tail-septum. Distribution. S. pulchra was collected from 23 stations, or from exactly 50 per cent of those at which chaetognaths were taken. Of these only 5 were mesoplanktonic stations, and 14 of the remaining 18 were surface stations. There can be no question, therefore, that the species is typical of the upper epiplankton in the Philippine region. Its northernmost record of capture during the Philippine expedition is in the China Sea, near Hongkong, 20 58' north and 120 3' east; its southernmost record is in Buton Strait, 5 36.1' south and 122 7.6' east; its easternmost record is in the Gulf of Tomini, Celebes, 125 17.1' east and 1 13.2' north; and its westernmost record is in Macassar Strait, 118 50' east and 2 19.5' south. The largest number (128 + ) was taken August 11, 1909, at 7.49 in the after- noon by a 0000 grit-gauze net towed on the surface between Siquijor and Bohol Islands, at 9 27.5' north and 123 38' east. Other species taken at the same station are S. enflata (169), S. lyra (85), S.ferox (50), and S. bedoti (14). CHAETOGXATHA COLLECTED BY STEAMER ALBATROSS. 253 t L ~ ss* s* ^i ~~ 333333^* 3 g^<N 3 3 3 3 t~ 3 3 030202020202 0202 02020202 O2 02 a a a a a a a a a a a a a a a s a a a s a a a ]HuSWHo2K SS0SSggS I 51! ! sss" 254 BULLETIN 100, UNITED STATES NATIONAL MUSEUM. SAGITTA DECIPIENS Fowler. Plate 35, fig. 8; plate 37, figs. 18,22. Sagitta dedpiens FOWLER (1905), p. 70. RITTER-ZAHONY (1911), p. 27. Sagitta sibogae FOWLER (1906), p. 21. RITTER-ZAHONY (1909a), p. 5. MICHAEL (1911), p. 74. According to Ritter-Zahony (1911, p. 29), there is "keine spezi- fischen Unterschiede in Fowler's Diagnosen und Abbildungen der beiden Arten [8. dedpiens und S. sibogae] und S. sibogae weist danach als alteres Stadium ! gegeniiber S. dedpiens eigentlich nur bedeutendre Dimensionen und hohere Zahlen fiir die Voider- und Hinterzahne auf." Although the species is represented in the Philippine collection by more than 100 specimens, few are well enough preserved to permit accurate measurements and their identification is therefore not certain. They have more anterior and posterior teeth than recorded by Fowler (1905, p. 70) in his original description, 8 to 11 anterior and 19 to 22 posterior teeth against his records of 5 to 10 anterior and 12 to 18 posterior teeth. They agree, however, with his (19,06, p. 21) records for S. sibogae, in which the anterior teeth number 7 to 10 and the posterior teeth 13 to 23. They also agree, not so well perhaps, with Ritter-Zahony's (1911, p. 28) records. He gives the number of anterior teeth as 7 to 9 and the number of posterior teeth as 12 to 20. The species is redescribed on the basis of the Philippine material. Collarette (pi. 35, fig. 8) inconspicuous, varying in length from slightly less than one-quarter to slightly more than half the body width. Body flabby, seldom retaining its form well, and widest on level with posterior end of anterior fins) tapering gradually toward head and tail. No constriction at tail-septum. Ovary short, not extending beyond anterior limit of posterior fins. Corona ciliata not observed. Anterior fins (pi. 35, fig. 8) rayless throughout anterior half, longer and narrower than posterior fins, and extending slightly beyond posterior end of ventral ganglion. Interval from anterior to posterior fins about equal to maximum width of body. Posterior fins (pi. 35, fig. 8) rayless in anterior extremity. They never extend posteriorly to seminal vesicles, the interval varying in length from about 25 to 110 per cent of the maximum body width. More than 50 per cent of fin in front of tail-septum, varying from 54 to 65 per cent. Form irregular, the position of greatest width being at or just behind tail-septum. Vestibular ridge (pi. 37, fig. 18) concealed by a thick cuticle. It is characterized by large fairly regular papillae extending internally beyond the teeth and terminating near the mouth. Whig covers all except the first one or two teeth, the notch extending to the third or fourth. External process not observed. According to CHAETOGNATHA COLLECTED BY STEAMER ALBATEOSS. 255 Fowler (1905, p. 70) it is "a very strong process; sometimes forked at the external edge." Anterior teeth 8 to 11, short, broad, and diverging distally. Pos- terior teeth 19 to 22, longer, and more closely set than anterior teeth. Seizing jaws (pi. 37, fig. 22), 5 to 7. Point inserted slightly less than one-third its height into shaft, with an irregular triangular projection at the middle of its base. Base of point and top of shaft converge toward edge of jaw. Edge of point and edge of shaft on a line with each other, but back of point and back of shaft intersect each other, forming an obtuse angle. Tip of point slightly bent toward edge of jaw. Pulp-canal displaced slightly toward back of shaft, with a swollen place below base of point. Pulp evenly dis- tributed throughout canal. TABLE 9. Measurements of Sagitta decipiens. 1 1 I Posterior fin. Anterior fin. Collarette. 1 .c j ! & 1 4 | 1 fe I UIUIUl 1 "3 1 1 s 1 "a | a <C3 I 1 1 1 J "5 5 *o _2 S A 3 tl 5 fl ** I S, .g g S3 ji A j s fc 1 3 ^ 1 I s 1 2 1 g S 2 ^ g I g i, fc | 1 2 12.4 12.2 5.0| 23.5 5.5 24 70.5 7.0: 20.0 67 1 9.5 20.5 4.0 5 5.5 4.5 7.5 6 64.5 fi5 27.0 98 1.5 2.5 +1.0 + 1.5 3.0 3 15.5 15.5 9-8 9-9 19-20 21-20 6- 6-6 3 11 5.5! 25 72.5 8.5 21.5 4.5 1 5 4 5S 5 W 2 (+1.5 1.5 13 10-11 22-22 5-5 4 9.1 (?) ?5. 5 70 i 8.5 24 5 2.5 4.5 55 27 2.5 +1.5 1.5 14 11-1C 19-21 7-7 5 8.8 5 24 69.5 8 22.5 (?) 2.5 3 58 28 2 +2 1.5 15.5 9-8 20-21 6-6 1 All measurements made in per cent of total length of animal. * Per cent of posterior fin in front of tail-septum. Distribution. 8. decipiens was collected from only six stations, four of which were mesoplanktonic and two surface stations. According to the literature the species is typically mesoplanktonic, only the very young having been taken above 100 fathoms. Its records of occurrence during the Philippine expedition are given in Table 10. SAGITTA BEDOTI Bgraneck. Plate 35, fig. 6; plate 37, figs. 20, 24; plate 38, fig. 30. Sagitta bedoti BERANECK (1895), p. 137. FOWLER (1906), pp. 6-8. RITTER- ZAHONY (1911), p. 20. MICHAEL (1911), p. 75. According to Fowler (1906, p. 6) 8. bedoti has "a very slight thickening of the epidermis at the neck, but no real collarette." In my 1911 report those species not taken from the San Diego region were briefly described and, not having seen S. bedoti, I as- sumed Fowler's statement to be correct and placed this species among those in which the collarette was absent. Subsequently, however, Kitter-Zahony's (1911) report appeared in which he (p. 20) 256 BULLETIN 100, UNITED STATES NATIONAL MUSEUM. II So 1 aasa a B a a a a 'ZKttK'Z, says: "Collerette relativ kurz, nur bis etwa zur halben Corona reichend." The speci- mens collected during the Philippine expedi- tion agree with Ritter-Zahony's statement. Every specimen has a collarette which, while narrow and short, is broader and longer than that of S. bipunctata. Owing to this confusion in the literature, S. ledoti is redescribed from the Philippine speci- mens: Collarette (pi. 38, fig. 30) conspicuous but short, extending caudally a distance nearly equal to greatest width of body. Head small. Lateral fields large. Muscles strong, but narrow. Body firm, retaining its form well, widest slightly behind its middle, and tapering gradually toward head and more rapidly toward tail. No constriction at tail-septum. Tail 20 to 30 per cent of total length of animal. Corona ciliata not ob- served. Anterior fins (pi. 35, fig. 6) longer and narrower than posterior fins, without rays in the anterior half or two- thirds. Fins extend nearly if not quite to posterior end of ventral ganglion, the exact limit being difficult to determine owing to absence of rays. In some individuals the fins may extend beyond posterior end of ganglion, but never to its anterior end. Form acutely triangular, the position of greatest width being in posterior quarter of fins. Posterior fins (pi. 35, fig. 6) extend caudally to seminal vesicles. Rays absent in anterior extremity. Usually, but not always, less than 50 per cent of fin in front of tail-septum, the extremes being 40 and 52 per cent. Triangular in form, the position of greatest width behi'j; behind tail-septum. Interval from anterior to posterior fins varies from hah' to twice the maximum width of body. Vestibular ridge (pi. 37, fig. 2o)- promi- nent and provided with regular and un- usually acute papillae. Number of papillae CHAETOGNATHA COLLECTED BY STEAMER ALBATROSS. 257 less than number of teeth. Wing of ridge covers all except the first and occasionally the second tooth. Notch extends to the fourth or fifth tooth. External process short and blunt, not, as described by Fowler (1906, p. 6), "terminating externally in a well- marked, rather sharp process." Its length is about two-sevenths that of entire ridge. Anterior teeth 8 to 10 and posterior teeth (pi. 37, fig. 20) 20 to 28 in number in individuals 10 to 15 mm. long. Anterior teeth are closely set and diverging distally, while the posterior teeth are more closely set and only slightly divergent distally. Seizing jaws (pi. 37, fig. 24) 5 to 7 in number. Points with curved tip, oval bases, and embedded about 25 per cent of then- heights into shaft. Base of point and top of shaft parallel. Pulp-canal central, extending into point about 75 per cent of its height and converging markedly toward edge of point. Pulp evenly distributed throughout canal. TABLE 11. Measurements of Sagitta bedoti. 1 i S I 29 25 25 25. r. 25 20 25. eg 24.7 4.0 24 4.5 27 3 23.5 4 26 14.4 16 14.5 , 15 14 16 15.5 16.5 16.5 15.5 9-10 9-9 9-10 S-9 24-22 24-23 27-28 22-21J 24-25! 25-24 22-21 24-25 23-22i 20-21; 1 All measurements made in per cent of total length of animal. 2 Ter cent of posterior fin in front of tail-septum. Distribution. S. ~bedoti was collected from only four stations, 352 specimens having been obtained. As shown by the following table all specimens were taken from the upper epiplankton. This indica- tion that the species typically occurs near the surface is supported by other expeditions and collections. It has been taken near the surface in the region of Port Natal by the Gauss expedition; in the Maldive and Laccadive Archipelagoes by Doncaster (1902) under the name S. polyodon; in the Malay Archipelago by Beraneck (1895) ; in the Siboga region by Fowler (1906) ; in Misaki Harbor, Japan, by Aida (1897) under the name S. lipunctata; and in Sharks Bay, Australia, by Ritter-Zahony (1910). Altogether, the evidence war- rants concluding that S. 'bedoti is characteristic of the upper epi- plankton of the tropical Indo-Pacific region. 258 BULLETIN 100, UNITED STATES NATIONAL MUSEUM. ' s. SAGITTA NEGLECTA Aida. Plate 35, fig. 9. Sagitta neglecta AIDA (1897), p. 16. FOWLER (1906), p. 15. RITTER-ZAHONY (1911), p. 23. MICHAEL (1911), p. 46. This species is represented by approximately 425. specimens (Cat. No. 17927, U.S.N.M.), many of which are mature. Curiously enough, all were taken by a single surface haul on November 22, 1909, in Molucca Passage at station D. 5615, 32.5' south and 126 31.5' east. Certain specimens obtained from five or six other stations were at first thought to be 8. neglecta, but closer examinations proved them to be either young 8. serratodentata or 8. ledoti. As in so many other cases, the Philippine specimens have more anterior and posterior teeth than those described from the San Diego region. In my former (1911, p. 48) report, the number of anterior teeth is given as 3 to 5 and the posterior teeth as 8 to 1 1 in in- dividuals between 8 and 13 mm. in length. The Philippine specimens, however, are smaller, ranging as a rule between 6.5 and 8 mm. and the number of anterior arid posterior teeth are 4 to 8 and 12 to 16, respectively. This agrees better with Fowler's (1906, p. 16) records for specimens from the Siboga region. He records 3 to 7 anterior and 7 to 15 posterior teeth in individuals between 5 and 10 mm. in length. Similarly, Ritter-Zahony (1911, p. 24) records 5 to 7 anterior and 11 to 18 posterior teeth in individuals from Port Natal between 6 and 7.5 mm. in length. TABLE 13. Measurements of Sagitta neglecta. 1 18 28 37.5 47 57 67 77 87 5.f>:',2 Posterior fin. Anterior fin 25 23.52 22.52.5 2 40 10 37.520.52 24 24 2.5 All measurements made in per cent of total length of animal. 1 Per cent of posterior fin in front of tail-septum. CHAETOGNATHA COLLECTED BY STEAMEE ALBATEOSS. 259 SAGITTA FEROX Doncaster. Plate 35, fig. 7; plate 37, figs. 21, 25. Sagitta ferox DONCASTER (1902), p. 212. FOWLER (1906), p. 10. MICHAEL (1911), p. 74. Sagitta robusta (part) RITTER-ZAHONY (1909a), p. 49; (1911), p. 16. Kitter-Zahony (1909a, 1911) synonymises this species to S. ro- busta Doncaster and says (1909a, p. 49): "Ich glaube daher nicht fehlzugehen, wenn ich mich fur die schon von Doncaster (1902, p. 212) als moglich hingestellte Identitat dieser beiden Arten auspreche und S. ferox nur eine altere S. robusta auffasse." Yet, although the two species closely resemble each other, the Philippine specimens do not indicate the slightest convergence with age in three important differential characters (see Table 14): 1 . The collarette is much wider in 8. ferox and nearly always extends beyond the anterior end of the ventral ganglion, while in S. robusta it never extends much over halfway from neck to ventral ganglion. 2. Anterior fins always extend beyond posterior end of ventral ganglion in S. ferox, while in S. robusta there is an interval between the fins and ganglion. 3. Anterior fins are longer than posterior fins in S. ferox, while the posterior fins are the longer in S. robusta. These three persistent differences justify considering S. ferox valid until critical study of variations in these particulars can be made. The species is therefore redescribed from the Philippine specimens in hopes that this description may aid in establishing its valid or synonymical position: Collarette (pi. 35, fig. 7) long and broad, extending past anterior end of ventral ganglion onto anterior fins. Head large. Lateral fields small. Body firm, opaque, and nearly of uniform width from in front of ventral ganglion to tail-septum. Tail 25 to 30 per cent of total length of animal. Corona ciliata not observed. Anterior fins (pi. 35, fig. 7) longer and slightly narrower than posterior fins, always extending anteriorly beyond posterior end of ventral ganglion, frequently past its middle and occasionally to its anterior end. Form acutely triangular, the position of greatest width being in posterior quarter of fin. Posterior fins (pi. 35, fig. 7) extend caudally to seminal vesicles. Interval from anterior to posterior fins 3 to 7, usually about 5 per cent of total length of animal. Less than 50 per cent (41 to 44) of fin in front of tail-septum. Triangular in form, the position of greatest width being about midway between tail-septum and seminal vesicles. Vestibular ridge (pi. 37, fig. 21) strongly mamillated, the number of papillae corresponding to the number of teeth. Wing of ridge covers all except the first two teeth, the notch extending beyond the fourth or fifth. External process short, broad, and blunt. 260 BULLETIN 100, UNITED STATES NATIONAL MUSEUM. Anterior teeth 5 to 9 in number. They are closely set, provided with broad bases, and diverging distally. Posterior teeth (pi. 37, fig. 21) 10 to 14 in number. They are long, broad, closely set, and diverging distally. Seizing jaws (pi. 37, fig. 25) 4 to 6 in number. Point with an oval base inserted into shaft between 15 and 20 per cent of its height. Base of point and top of shaft converge toward back of shaft. Edge of shaft provided with narrow crest. Pulp-canal central, and extend- ing into point about 80 per cent of its height. Pulp evenly dis- tributed throughout canal. TABLE 14. Measurements of Sagittaferox. 1 Posterior fin. Anterior fin. +3.5 +4 +4 +5.5 Collarette. +3 +2 +1 +2 +3.5 +4 +3 13-13 12-11 13-13 6-5 13-12 5-5 13-14 5-4 15-12 I 5-6 12-12 5-4 12-12 ; 5-5 12-12 5-1 12-12 ' 5-5 12-? 5-5 10-12 5-5 i All measurements made in per cent of total length of animal. 1 Per cent of posterior fin in front of tail-septum. The + indicates that the fin extends beyond posterior end of ganglion. The + indicates that the collarette extends beyond anterior end of ganglion. Distribution. 8. ferox was collected from 22 of the 46 stations from which chaetognaths were taken, 598 specimens being obtained. It is difficult to decide whether the species is typically epiplanktonic or mesoplanktonic in the Philippine region. At 15 stations it was taken from above 100 fathoms. Five, or 33 per cent of the epiplank- tonic hauls, and two, or 29 per cent of the mesoplanktonic hauls, obtained more than the average number of specimens (27). Again, the median number of those taken above 100 fathoms is nine, but nine is also the median number of those taken below 100 fathoms. Finally, the greatest number was taken from 25 fathoms, the second, fhird, and fourth greatest from the surface, and the fifth greatest trom 500 fathoms. All subsurface hauls, however, were made by various types of open nets. It is well to remember that such hauls, whether horizontal or vertical, afford no certain evidence of the depth from which specimens CHAETOGNATHA COLLECTED BY STEAMER ALBATBOSS. 261 were collected. This is true even when every haul is made with the same net, and when various types of nets are used the data are worse than useless for this purpose. For these reasons it may well be that, as indicated by other reports, S.ferox is typically epipianktonic in the Philippine region. It was taken in abundance during the Siboga expedition from the surface, but only rarely from the mesoplankton. Indeed the species appears to be restricted to the epiplankton of the Indo-Australian region, although, owing to its questionable synonymy with S. robusta Doncaster, this statement is made with some reservations. But, even assuming the two species to be synony- mical, it is still restricted in distribution to the surface and upper epiplankton of tropical and subtropical regions. Thus, according to Ritter-Zahony (1911, p. 58), it is found "im Atlantischen Ozean zwischen and 20 N., im Indischen zwischen 20 and 30 S. in Vertikalfangen und an der Oberflasche. ..." He continues: "Auf der Westseite des Atlantischen Ozeans wird sie zwar durch den Floridastrom wohl bis in die Gegend des 40 N. gebracht, auf der Ostseite gelangt sie jedoch kaum bis zum 35. Breitegrad, da sie ja schon im Mittelmeer fehlt. Im Siiden diirfte sie gerade noch um die Siidspitze Afrikas herumkommen. Im Stillen Ozean ist sie bisher nur funf Fundorten, die samtlich auf seiner Westseite liegen, bekannt geworden. . . . Ich glaube jedoch, dass die Ver- breitung der S. robusta im Stillen Ozean der im Atlantischen vollig analog ist, d. h. das tropisch-subtropische Gebeit umfasst und nur auf der Westseite etwas weiter nach Norden reicht, wobei der Kuro- Siwo die Rolle des Floridastroms uberniinmt." Clearly S. ferox is a warm water species, but is its absence hi the eastern Pacific not more likely attributable to the abnormally cold water there due to upwelling? (see p. 271). The northernmost and westernmost record of its capture in the Philippine region is 21 31' north and 117 53' east, or in the south China Sea, approximately halfway between the city of Hongkong and the island of Formosa. The southernmost record is 5 36' south at 122 7.6' east off the south end of the island of Celebes in Buton Strait. The easternmost record is 32.5' south and 126 31.5' east, or, less accurately, at the southern end of Molucca Passage east of Tomini Bay. The largest number of specimens (217) was taken February 7, 1908, at 8.05 in the morning, from 25 fathoms by an open 0000 grit-gauze net towed horizontally 9 fathoms above the bottom of the Sulu Archipelago, near Basilan Island, at 6 44.2' north and 121 47' east. The other records are given in the table following. 262 BULLETIN 100, UNITED STATES NATIONAL MUSEUM. iS o S3 c* i- F-i 00 t^ O ssss M ^>S a^o^ -s-sS&Bsi <N M <N S - <N ^gaccoccoooM-o CKAETOGJSTATHA COLLECTED BY STEAMER ALBATEOSS. 263 SAGITTA PLANKTONIS Steinhaus. Sagitta planktonis STEINHAUS (1896), p. 39. RITTER-ZAHONY (1911), p. 29. MICHAEL, (1911), p. 44. Sagitta zetesios FOWLER (1905), p. 67; (1906), p. 22. Eighty-seven specimens were obtained, but unfortunately all except one were preserved in alcohol, with the result that they are so badly distorted as to prevent certain identification. Measurements of the single well-preserved specimen are: Length 27 mm. Width 8. 5 per cent of length. Tail: Length 24. 5 per cent of length. To central ganglion 68. per cent of length. Ventral ganglion (length) 4. per cent of length. Posterior fin: Length 20. 5 per cent of length . Width 4. 5 per cent of length. To anterior fin 4. 5 per cent of length. Proportion in front of tail-septum 77. per cent. Anterior fin : Length '. 23. per cent of length. Width 2. 5 per cent of length. To ventral ganglion 0. per cent of length. Collarette: Length 25. per cent of length. To ventral ganglion ' 0. per cent of length. Number of anterior teeth 8-7 Number of posterior teeth 17-18 Number of seizing jaws 8-8 The length, number of anterior and posterior teeth, and number of seizing jaws of a few other individuals are: Length in mm. Anterior teeth. Posterior teeth. Seizing jaws. 11.5 13 16.5 17 21 25 7-6 9-9 9-10 10-9 8-8 9-8 13-12 16-? 16-15 17-18 16-17 18-17 9-8 7-8 9-9 8-8 8-8 8-9 The number of teeth is greater than in specimens from the San Diego region (Michael, 1911, p. 45). San Diego specimens, ranging in length between 15 and 26 mm., have 4 to 7 anterior, and 11 to 15 posterior teeth. In the Biscayan report, however, Fowler (1905, p. 68) records a variation in number of anterior teeth from 5 to 9 and of posterior teeth from 11 to 19 in individuals between 11 and 21 mm. in length. Distribution. 8. planktonis was collected from only seven stations. Of the 84 specimens obtained, 53 were taken from below 100 fathoms by open nets towed horizontally at four stations for approximately 264 BULLETIN 100, UNITED STATES NATIONAL MUSEUM. 1 a s a a a a a a a a d P. a cs os d 20 minutes at each. The 35 remaining specimens were taken from above 15 fathoms at the three remaining stations by 20-minute tows. These facts in- dicate that the species is typically mesoplanktonicin the Philippine region, which indication is supported by the results of many other expeditions. In the regions covered by the Biscayan, Siboga, and Plankton expeditions, as well as off the California coast, the species occurs abundantly below 100 fathoms, but is rarely found above that depth. It is common between 500 and 1,000 fathoms, and, as Ritter-Zahony (1911, p. 62) says: "S. planktonis ist unter alien Arten der Tiefsee am hau- figste/i in der Literatur erwahnt." Genus PTEROSAGITTA Costa. Syn. Spadella LANGERHANS (part 1). PTEROSAGITTA DRACO (Krohn). Plate 36, figs. 11, 12, 13. Sagitta draco KROHN (1853), p. 273. Pterosagitta mediterranea COSTA (1869), p. 55. Spadella draco FOWLER (1906), p. 25. MICHAEL (1911), p. 54. Pterosagitta draco RITTER-ZAHONY (1911), p. 33. Thirty-two specimens were obtained, of which only one has mature ovaries. These completely fill the body cavity (pi. 36, fig. 13), extending from tail- septum to neck. All other specimens are clearly immature, and in nearly one-half there is no trace of ovaries, and the seminal vesicles are barely visible (pi. 36, fig. 12). Fowler (1906, p. 26) records the tail as 41 to 57 per cent of total length in specimens be- tween 6 and 9 mm. The Philippine specimens, however, vary only between 39.5 and 44.3 per cent. Otherwise Fowler's records agree exceptionally well with the Philippine material. He records 7 to 9 seizing jaws, 7 to CHAETOGKATHA COLLECTED BY STEAMER ALBATROSS. 265 10 anterior teeth, and 11 to 16 posterior teeth; the Philippine speci- mens show 7 to 9 seizing jaws, 6 to 10 anterior teeth, and 10 to 16 posterior teeth. Nearly half the Philippine specimens are devoid of the collarette, except for a narrow strip behind the head. This caused consider- able trouble in identification until others were discovered in which the structure was partly missing. These (pi. 36, figs. 12 and 13) indicate that for some unknown reason the collarette has been torn off. Immature specimens in this condition bear a striking super- ficial resemblance to young S. jerox. TABLE 17. Measurements of Pterosagitta draco. 1 Number. Length in mm. Width with colla- rette. Width without collarette. Length of tail. s l I Length of ventral ganglion. Lateral fin. Ovary. Number of ante- rior teeth. | ol ji Number of seizing jaws. a i "6.T 5.7 6.8 5 "i'e" 6.3 5.7 5.3 4.8 6.1 5.5 4.8 5.6 6.5 5.8 5.3 4.8 It 43 & 1 5 1 2 3 4 7 8 9 10 11 12 13 14 15 16 17 18 19 20 7.7 7.6 7.4 7.2 7 6.9 6.9 6.7 6.7 6.6 6.6 6.3 5.8 5.8 5.6 5.4 5.4 5.3 5.1 4.9 7.8 6.9 7.1 10.7 8 7.6 6.6 6.8 7.8 8.5 7.4 7.3 7.9 7.9 S:i 7.8 6.7 6.2 7.1 43.3 44 42.8 39.5 44 40.3 43.1 41.4 40.5 40.9 42 41.3 40 42.3 41.6 42.5 40.8 41.3 41.4 44.3 66.6 66.5 65.6 59.6 66.9 65.2 66.5 65 64 62.7 65 64.8 63.6 66.7 66.5 65.5 64.2 64.5 64.9 68.5 9.1 8.7 9.1 7.8 8.5 9.2 8.1 10 9.9 10.6 9.6 9.5 9.7 6.7 8.7 9.8 9.7 10.7 11 10.7 22.8 23.4 22.4 20 23 23 22.3 23.1 21.3 21.8 19.1 22.9 21.8 18.8 16.8 20.2 20.8 22.7 20.7 21.4 o 24.6 24.8 25.7 47.8 21 10.7 20.8 9.5 2.3 2.7 1.9 5.9 4.5 .5 1.5 1 9-8 9-8 9-8 5-6 8-7 7-6 8-8 8-7 10-9 7-7 6-6 9-8 6-7 9-8 8-7 8-7 7-7 7-7 6-6 7-6 13-13 (?)-14 15-16 11-12 14-13 14-(?) 13-13 13-14 15-(?) 14-15 11-12 14-(?) 11-11 14-14 12-11 14-15 (?)-H 10-11 11-12 13-13 8-7 8-8 8-9 8-8 8-9 8-9 8-8 8-8 8-8 8-8 9-9 8-8 9-9 8-9 9-8 8-9 7-7 8-8 (?)-8 7-7 16.6 15.2 "ie.Y 13 13.3 "isT ...5... ........ 10.4 6.8 1.1 "is"" "5"" 1 All measurements made in per cent of total length of animal. Distribution. P. draco was obtained from only five stations, all of which were subsurface ones. Two were mesoplanktonic, but they only yielded four specimens. The remaining 28 were obtained from between the surface and 25 fathoms, indicating that hi so far as the species occurs in the Philippine region it is typical of the upper epi- plankton. This accords with what is known of its distribution gen- erally. Although nowhere abundant it is apparently restricted to the upper epiplankton of tropical and subtropical regions. Says Ritter-Zahony (1911, p. 63): "Ihre Verbreitung diirfte sich mit der von S. enflata und 'bipunctata decken, doch ist die Art in den gemassig- ten Zonen, soweit sie iiberhaupt noch darin vorkommt, schon seltener als jene beiden. . . ." Its northernmost and southernmost records of occurrence are 42 north, 56 west; and 42 south, 36 east. Its records of capture during the Philippine expedition are given in the table following: 59318 19 Bull. 100 3 266 BULLETIN 100, UNITED STATES NATIONAL MUSEUM. * fc ' op os^rto se- f.!l iilll fifiQOQ Genus EUKROHNIA Ritter-Zfihony. Syn. Krohnia LANGERHANS (part). EUKROHNIA HAMATA (MSbius). Plate 37, figs. 14,27. Sagitta hamata MOBIUS (1875), p. 158. Krohnia hamata KRUMBACH (1903), p. 639. FOW- LER (1905), p. 74; (1906), p. 23. Eukrohnia hamata RITTER-ZAHONY (1911), p. 39. MICHAEL (1911), p. 51. This species is represented by six poorly preserved and badly damaged specimens (Cat. No. 17928, U.S.N.M.). The heads of all but two are missing and in one of those it is torn away from the body. It is impossible, therefore, to accurately determine the length or to count the teeth and seizing jaws in four of the six specimens. The other two are 16 and 16.3 mm. in length; their tails are 28.3 and 30.2 per cent of their lengths; the number of teeth are 21-21 and 24-25, and the number of seizing jaws are 9-10 and 9-9. The number of teeth (pi. 37, fig. 14) greatly exceed that recorded by Ritter-Zahony (1911, p. 39) for specimens taken from the Antarctic Ocean, as well as that recorded by me (1911, p. 52) for specimens from the San Diego region. Although Ritter-Zahony records a variation in number of teeth from 4 to 23, he gives 12 as the upper limit for specimens under 18 mm. in length. Simi- larly, in specimens from the San Diego region between 13 and 17.5 mm. in length, I have never found more than 13 teeth. Fowler (1906, p. 23), however, records 22 in speci- mens from the Siboga area that are only 13 mm. hi length. It is evident, therefore, that the number of teeth vary according to the region in which the species occurs. The pouits of the seizing jaws (pi. 37, fig. 27) are curved to an unusual extent, and the jaws are not serrated as described by Krumbach (1903). Otherwise, however, the structure of the jaws agree with his de- scription. CHAETOGNATHA COLLECTED BY STEAMER ALBATEOSS. 267 The six specimens were all obtained November 6, 1908, in the China Sea in the vicinity of Formosa, at station D. 5320, 20 58' north and 120 3' east by an open 0000 grit-gauze net towed at 3.18 p. m. in 500 fathoms for 20 minutes. EUKROHNIA RICHARDI Germain and Joubin. Plate 36, fig. 10; plate 37, figs. 15, 26. Eukrohnia richardi GERMAIN and JOUBIN (1912), p. 2. This species (pi. 36, fig. 10) is represented by five specimens, only two of which are well enough preserved to permit certain identifi- cation. Measurements of these two (a and 6), and, for comparison, those taken from Germain and Joubin's drawing (c), are as follows: a. b. C. Length in mm ... 27.7 27 8 27 8 Width in per cent of length 12.9 13.3 12.2 Tail: Length in per cent of total length 27.2 26.2 25.2 To ventral ganglion in per cent of total length 70 7 74 2 67 6 Length of ventral ganglion in per cent of total length 6.6 5.2 2.9 Lateral fin: 73 1 69 3 68 4 Width in per cent of total length ... .... 4.9 6 5 6 5 Extends bevond anterior end of ventral ganglion, in per cent of total length Proportion in front of tail-septum (per cent) 10.4 82.7 6.5 86.4 10.8 74.4 Ovary: 4 g 6 5 10 8 Width in per cent of total length 1.26 2.64 2 88 22 21 25-24 24 Number of seizing jaws 10-11 10-10 8 It is questionable whether this species is valid or not. It closely resembles E. hamata, but according to Germain and Joubin (1912, p. 5), "il s'en distingue facilement, en dehors de sa coloration verte caracte"ristique et jusqu'a present absolument unique chez tous les Chetognathes, par la forme tres diff e"rente de sa tfite, beaucoup plus nettement triangulaire allong6e, par ses crochets plus e'troitement allonges et par ses dents, au nombre de 24, alors qu'on en compte seulement de 20 a 22 chez I' Eukrohnia liamata" Of these distinc- tive features it is evident that differences in the number of teeth and shape of head are of no specific value, and it seems probable that the color, which is described (p. 2) as "d'un vert d'eau plus fonce a la region ante"rieure et s'attenuant re'gulierement vers la queue," is also highly variable. In the Philippine material at least, specimens of E. hamata agree quite as well with this description as those of E. richardi. Both, although considerably faded by action of the f ormalin, are dark green and more so anteriorly than posteriorly. Again, the various dimensions of the body recorded by Germain and Joubin (p. 4) do not agree with those taken from the drawing. The drawing measures 139 mm. in length. Assuming the recorded length of 27 mm. to be accurate, the magnification of the drawing is 5.15. This makes the length of tail 6.8 mm., or 25.2 per cent of 268 BULLETIN 100, UNITED STATES NATIONAL MUSEUM. total length, but it is given as 6.5 mm., or 24 per cent of the total length. It is stated that the ovaries are 2.25 mm. in length, or 8.3 per cent of the length of animal, but in the drawing they measure 10.8 per cent, which is equivalent to 2.92 mm. The maximum width of body ("y compris la largeur des nageoires") is given as 2.5 mm., or 9.3 per cent of the length of animal, but in the drawing it measures 11.5 per cent, which is equivalent to 3.1 mm. Finally, it is stated that the lateral fin occupies "17/27 [63 per cent] de la longeur totale de 1' animal," that it extends 5.5 mm. anterior to the ventral ganglion or "a environ 3 millimetres de Textre'mite' du corps," but in the drawing it measures 68.4 per cent of the total length of animal and extends only 10.8 per cent, or 2.92 mm., anterior to the ventral ganglion. Although these discrepancies are, for the most part, small, they make it impossible to depend with certainty upon either the descrip- tions or the drawing, and whether E. richardi is or is not synonymous with fi. Jiamata must remain undecided until the type is more accu- rately described. In spite of this uncertainty there are several points that indicate its validity: 1. The seizing jaws (pi. 37, fig. 26) are more massive than those of E. Jiamata and their points are quite dissimilar. In E. Jiamata the points are always sickle-shaped (pi. 37, fig. 27), although they are not usually curved so much as in the Philippine material; the top of shaft and base of point converge upon approaching the edge of the jaw; the point has an oval base; and the pulp-canal is central and sparsely filled with pulp. In E. richardi, on the other hand, the points are not sickle-shaped; the top of shaft and base of point are parallel; the point has an irregular rather than an oval base; the pulp-canal is irregular in outline and is displaced toward the back of shaft; and the pulp is evenly distributed. 2. The body of E. richardi is between two and three times wider in proportion to the length of animal than is the case with E. Jiamata. 3. Lateral fins extend to, but rarely beyond, anterior end of ventral ganglion in E. Jiamata, while in E. richardi they extend beyond the anterior end of ganglion by 6 to 11 per cent of the length of animal. 4. Lateral fins in E. Tiamata with "fin-rays extending about as far in front of the septum as the fin does behind it, but the fin continued forwards as an expansion of the epidermis up to or to the middle of the ventral ganglion." [Fowler (1905, p. 74).] In E. richardi the fin, according to Germain and Joubin, is delicately rayed throughout, the rays making an acute angle with the body as illustrated by their drawing. This seems to be true of the Philippine specimens (pi. 36, fig. 10), although the rays are fewer and much finer toward the anterior end. CHAETOGNATHA COLLECTED BY STEAMER ALBATROSS. 269 5. Lateral fins never extend more than halfway from tail-septum to seminal vesicles in E. Jiamata, while they extend three-quarters of the distance, according to Germain and Joubin's drawing, and quite to the anterior end of the vesicles in the Philippine specimens. . The Philippine specimens apparently differ from the type in two points : (1) Germain and Joubin state that the ventral neck muscles are composed of longitudinal fibers, together with very fine transverse ones. The transverse fibers are absent in the Philippine specimens. (2) The seminal vesicles are figured and described as very small disks lying within the anterior end of the caudal fin. In the Phil- ippine specimens this is not the case, although the vesicles in all specimens were very immature. Their appearance, however, sug- gests that when mature their posterior extremities may touch the caudal fin. The five specimens (Cat. No. 17929, U.S.N.M.) were all obtained at 12.07 in the afternoon, May 8, 1909, station D. 5437, off the west coast of Luzon, 15 45/9 north and 119 42/8 east, by an open 0000 grit-gauze net towed for 27 minutes in 450 fathoms. Genus KROHNITTA Ritter-ZShony. Syn. Krohnia LANGERHANS (part). Spadella GRASSI (part). Krohnia STRODTMAN (part). KROHNITTA SUBTILIS (Grassi). Spadella subtilis GRASSI (1883), p. 23. Krohnia subtilis STRODTMAN (1892), p. 22. FOWLER (1905), p. 78; (1906), p. 25. Krohnia padfica FOWLER (1906), p. 24. Eukrohnia subtilis MICHAEL (1911), p. 52. Krohnitta subtilis RITTER-ZAHONY (1911), p. 44. Only three specimens were obtained, all of which are immature. Only one is sufficiently well preserved to permit accurate measure- ments : Length in mm 9.8. Length of tail 35.9 per cent of total length. Tail to ventral ganglion 63.5 per cent of total length. Length of ventral ganglion 7.5 per cent of total length. Length of fin . . 35.1 per cent of total length . Width of fin 6.4 per cent of total length. Proportion of fin in front of tail-septum 32.7 per cent. Length of ovary 8.6 per cent of total length- Width of ovary 2.1 per cent of total length. Number of seizing jaws 8-8. The teeth could not be counted, but in the two other specimens, which were about the same length, the number of teeth on the right and left sides are 11-10 in one and 12-13 in the other specimen. Similarly, the number of seizing jaws are 8-7 in the first specimen and (?)-7 in the second. 270 BULLETIN 100, UNITED STATES NATIONAL MUSEUM. Two of the specimens (Cat. No. 17931, U.S.N.M.) were obtained between Panay and Negros at station D. 5185, 10 5/8 north and 122 18/5 east, on March 30, 1908, at 5.26 in the afternoon, by an open 0000 grit-gauze net towed horizontally for 20 minutes in 550 fathoms. The third specimen (Cat. No. 17930, U.S.N.M.) was obtained in the China Sea in the vicinity of Formosa, at station D. 5319, 21 31' north and 117 53' east, on November 5, 1908, at 7.23 in the afternoon, by the same net towed for 27 minutes in 20 fathoms. COMPARISON OF PHILIPPINE CHAETOGNATHA WITH THOSE FROM THE SAN DIEGO REGION. Aida (1897) describes 10 species from Misaki Harbor, Japan, of which all except 8. regularis and 8. Jiispida ( = 8. robusta Doncaster) are represented in the Philippine collection. Again, Doncaster (1902) records 10 species from the Maldive Archipelago, of which all save 8. regularis and 8. Jiispida are present in the Philippine material. Likewise, Fowler (1906) describes 14 species from the "Siboga" area, of which all save 8. regularis and 8. hispida have been taken from the Philippines. Lastly, Bitter-Zahony (1910) lists 10 species from Sharks Bay, Australia, of which all except 8. regularis and 8. bipunctata are represented in the Philippine collection. The only species obtained from the Philippines which are not recorded from any of these regions are 8. pJiilippini and E. richardi, the former a new species represented by a single individual, and the latter a rare species represented by only five specimens. Obviously these facts strongly point toward a uniformity in the chaetognath fauna, espe- cially the epiplankton, over the Indo-Pacific Ocean, notwithstanding the curious absence of 8. regularis and 8. Jiispida in the Philippine collection. The situation is quite otherwise when the Philippine chaetognatha are compared with those from the San Diego region, for those species most characteristic of the Philippines are those that are either absent or least characteristic of the San Diego region, and the op- posite. Thus, 8. enflata, by far the most typical and abundant species about the Philippines, has been obtained in the San Diego region by less than 20 out of nearly 4,000 hauls. Conversely, 8. bipunctata is by far the most typical and abundant species in the San Diego region, but not a single individual was obtained from the Philippines. Again, Sagittaferox, 8. pulcJira, 8. bedoti, 8. decipiens, 8. minima, 8. macrocephala, and Eukrohnia richardi have not been taken from the San Diego region, although the first three are third, fourth, and sixth in order of abundance in the Philippine region. On the other hand, Sagitta lyra and 8. californica, in addition to 8. bipunctata, were not taken from the Philippine region, although the former is CHAETOGNATHA COLLECTED BY STEAMER ALBATROSS. 271 the most characteristic mesoplanktonic species in California waters. Again, of those species common to both regions, Sagitta Jiexaptera and S. serratodentata are second and eighth in order of abundance in the Philippines and tenth and second in the San Diego region. Finally, aside from S. neglecta, which is rare in the San Diego region and ob- tained by only one haul from the Philippines, the order of abundance of the remaining species common to both regions is as follows: S. planktonis, P. draco, E. Tiamata, and K. subtilis in the Philippine region; and E. hamata, K. subtilis, S. planktonis, and P. draco in the San Diego region. Taken in connection with what is known of the distribution of chaetognatha throughout the world, the above comparisons show that the Philippine species are characteristic of the Tropics and warm water, while those of the San Diego region are, on the other hand, more characteristic of the Arctics or sub-Arctics and cold water. As a matter of fact there is less difference between the California chaetognatha and those of the region about Spitzbergen than there is between the California and Philippine faunas. Furthermore, this sub-Arctic nature of the California chaetognatha is not peculiar to that group. Calanus finmarchicus , the commonest copepod of the California coast is, according to Cleve (1900, p. 47), a "characteristic inhabitant of the Arctic regions, along the coast banks of Greenland, Iceland, etc." Similarly, Eucalanus elongatus, the second most typical copepod of California waters, is "noted from 60 north, 7 west in August and the Skagerak in February." [Cleve (1900, p. 63.)] Likewise Acartia clausii, obtained in abundance off the pier of the Scripps Institution, is typical of the North Sea " and follows the coast of Norway to about 70 or 74 north." [Cleve (1900, p. 42).] Again, the most prevalent ctenophore of the California region, PleurolracMa lachei, "is found in vast swarms in the cold water of Maine and Nova Scotia." [Esterly (1914, p. 28).] Lastly, among the diatomaceae, CJiaetoceros criophilum is the commonest diatom in San Diego waters, although Cleve (1900, p. 295) states that it is a "decidedly Arctic, pelagic species." Another common San Diego CJiaetoceros is C. debile, but Cleve (1900, p. 296) says that it is abundant along the south coast of Iceland and at the Faroes." Similarly with Nitzschia seriata, its "principal area of distribution is between Scotland, Iceland, and Greenland" [Cleve (1900, p. 336)], although it is among the common diatoms of the San Diego region. So the list might be continued. To be sure, there are many tropical and semitropical species occurring in California waters, but they are not the characteristic and prevalent ones. These have their nearest allies, not in other parts of the world at corresponding lati- tudes, but in the Arctic and sub-Arctic regions. May this not be attributable in part to the marked upwelling of cold bottom waters 272 BULLETIN 100, UNITED STATES NATIONAL MUSEUM. along the western coast of America ? To establish this would require an extensive series of collections off the coast of Central and South America, and comparisons of the faunas at the same latitudes on the two sides of the Pacific. But, if it is true, it emphasizes the necessity of recognizing this fact in fisheries investigations and demonstrates an essential difference between the problems, economic and otherwise, of the coastal waters of the eastern and western Pacific. That the chaetognath faunas of the two regions are fundamentally different is made more certain by the fact that in four of the five Sagitta l common to both Philippine and San Diego regions, the Philippine specimens have a greater number of both anterior and posterior teeth. The same is true with respect to P. draco, and in E. hamata the Philippine specimens have nearly twice as many teeth (21 to 25) as do the San Diego specimens (10 to 13). These differences are mentioned in the preceding pages in connection with the account of each species, but are better revealed, perhaps, in the following list: Species. Anterior teeth. Posterior teeth. Philippines. San Diego. Philippines. San Diego. 8. enflata 6-11 4-8 8-11 6-10 6-10 4-8 3-5 6-9 4-7 4-4 9-15 12-16 13-24 12-18 10-16 7-12 8-11 13-19 11-15 8-9 S. serratodentata S. planktonis P. draco In the case of P. draco the differences may mean little, owing to the fact that only a single very immature specimen is recorded from the San Diego region. It is interesting, however, to note that it was 7 millimeters in length, whereas 15 of the 20 from the Philippine region recorded in Table 17 are smaller and quite as immature, the smallest being less than 5 millimeters in length, although it has 6 to 7 anterior teeth and 13 posterior teeth. To demonstrate that the differences in number of teeth given for the four species of Sagitta are significant, from 10 to 30 or more individuals of each species were selected at random from the two collections, the number of teeth counted, and the mean number and corresponding probable errors computed. The results are entered in Table 19: TABLE 19. Comparison of mean number of teeth in specimens from the Philippines and from the San Diego region. Species. Anterior teeth. Posterior teeth. Philippines (P). San Diego (S). Difference (P-S). Philippines (P). San Diego (S). Difference (P-S). 8. enflata 8. 29 0.256 5. 61 0.160 9.320.133 8.360.200 6.270.035 3. 800. 243 7. 220. 114 5. 94 0.335 2. 02 0.258 1.81 0.291 2. 100. 175 2.420.390 13.450.241 13. 75 0.184 18. 24 0.536 16. 15 0. 145 10. 18 0.133 8. 96 0.135 15.440.203 12.600.218 3. 27 0.275 4. 79 0.22? 2. 800. 573 3. 55 0. 262 S. ner/lecta S. serratodentata S. planktonis. hexaptera is not considered owing to loss of teeth. See p. 245. CHAETOGNATHA COLLECTED BY STEAMER ALBATBOSS. 273 Table 19 shows (1) that in every case the mean number of both anterior and posterior teeth in Philippine specimens exceeds that in San Diego specimens, and (2) that the magnitude of the excess is between 5 and 20 times the corresponding probable error. That this excess is not merely an expression of the larger size of the Philippine specimens is evident, for the counts were made on specimens of Philippine Sagitta enflata between 10 and 21 mm. in length and on San Diego specimens between 10 and 25 mm.; on Philippine S. neglecta between 6 and 8 mm., and on San Diego specimens between 8 and 13 mm.; on Philippine S. serratodentata between 7 and 11 mm., and on San Diego specimens between 10 and 17 mm.; and on Philip- pine S. planktonis between 13 and 27 mm., and on San Diego speci- mens between 17 and 26 mm. Obviously, some differential influence is at work in the two regions causing an excess of teeth in the Philip- pine fauna, or a deficiency in the San Diego fauna. Unfortunately, a similar comparison of Philippine chaetognatha with those from other regions of the Pacific is impossible, owing to the fact that no one except Fowler (1906) has published a series of tooth counts, and he has not kept the individual counts distinct. The range of variation, however, in the Siboga material is much the same as that in Philippine specimens. This is pointed out in the foregoing pages for every species common to the two regions, but these data are brought together and amplified in Table 20 : TABLE 20. Comparison of number of anterior and posterior teeth in Philippine species and those from the Siboga region. SAGITTA ENFLATA. L^U, specimens In mm. Anterior teeth. Posterior teeth. Philip- pines. Siboga. Philip- pines. Siboga. 10-15 15-20 20-25 25-30 6- 7 7-9 7-11 8 7- 9 8-9 8-10 7-10 9-12 13-15 14-15 14 9-14 12-17 12-16 14-17 SAGITTA NEGLECTA. 6.5 4-6 4- 5 12-14 9-10 7 5-8 4- 6 12-16 9-12 7.5 5- 6 4- 5 14-15 10-13 8 5-6 4- 6 14-15 9-14 SAGITTA SERRATODENTATA. 7- 8 9 5-9 13-15 9-16 8- 9 8- 9 6-9 14-18 13-18 9-10 8-9 6-9 ' 16-19 13-19 10-11 8-11 8 9 18-24 15-19 274 BULLETIN 100, UNITED STATES NATIONAL MUSEUM. TABLE 20. Comparison of number of anterior and posterior teeth in Philippine species and those from the Siboga region Continued. SAGITTA PLANKTONIS. Length of specimens in mm. Anterior teeth. Posterior teeth. Philip- pines. Siboga. Philip- pines. Siboga. 11-13 13-15 15-17 17-19 19-21 21-23 23-25 25-27 7- 9 9 9-10 9-10 8 8 8- 9 7- 9 5- 8 5- 9 8- 9 7- 9 7-11 8-11 8-10 8- 9 12-10 16 15-18 17-18 16-17 16-17 17-18 17-18 13-16 14-18 17-18 16-18 16-20 17-20 18 18-19 SAGITTA PULCHRA. 10-15 15-20 20-25 5- 8 6-9 6- 8 6-10 6-10 7-10 9-12 11-13 10-13 10-14 10-15 12-15 SAGITTA DECIPIENS. 9.0 11 12 12.5 9-11 10-11 9 8-9 8 7 8 8- 9 19-21 22 20-21 19-20 16 16 15-17 15-18 SAGITTA BEDOT1. 10-12 12-14 14-16 8-10 8- 9 8-10 9-11 9-13 9-10 20-23 21-25 22-28 18-27 20-32 21-29 SAGITTA FEROX. 10-12 12-14 14-16 5- 7 7- 9 5- 9 4- 8 7- 9 6- 9 10-12 12-15 10-14 9-13 10-14 10-14 PTEROSAGITTA DRACO. 6.0 7 7.5 7- 9 5-10 8- 9 8 7-10 8- 9 11-14 11-15 13-16 12 11-15 12-16 Inspection of this table reveals a pronounced similarity in number of both anterior and posterior teeth between Philippine and Siboga speci- mens oiSagitta enflata, S. planktonis, S. pulchra, S. bedoti, S.ferox, and Pterosagitta draco. In S. neglecta, S. serratodentata, and 8. decipiens specimens from the two regions agree in number of anterior teeth, but those from the Philippines have a greater number of posterior teeth. Even in these instances the differences are slight and prob- ably insignificant except for 8. decipiens. In this species there seems to be no doubt that the Philippine specimens have more posterior teeth than Siboga specimens of the same length, although, as pointed out on page 254, where the length of animal was neglected, Fowler's records show a range of 13 to 23 posterior teeth as against that of 19 to 22 for Philippine specimens. CHAETOGNATHA COLLECTED BY STEAMER ALBATROSS. 275 On the whole there is close agreement in number of teeth between specimens of species common to the Philippines and the Siboga region, while the same species are represented in the San Diego region, if at all, by specimens having markedly fewer teeth. In the face of this fact it is evident that, as formerly (1911, p. 68) stated, "variation in number of both anterior and posterioi teeth in many species is not referable to specific differences, but probably to some distribution factor." When it is recalled that a subnormal ocean temperature characterizes the region adjacent to the western coast of America, due to pronounced upwelling of bottom water, and that the chaetognath fauna off southern and Lower California is representative of more northern latitudes, it suggests that one of these distribution factors is temperature. I believe the small num- ber of teeth in San Diego specimens is an expression of the slower rate of metabolism due to a lower ocean temperature. This is not a new suggestion, but it is one that merits thorough investigation. Fowler (1906, p. 29), after stating that specimens of Siboga Sagitta serratodentata have nearly twice as many posterior teeth as speci- mens of the same length from the Bay o^f Biscay, says: "It is possible that this may be correlated with the respective temperatures at which the specimens live, but a long series of similar observations from different latitudes would be necessary before this could be regarded as even probable." LITERATURE CITED. AlDA, T. 1897. The Chaetognatha of Misaki Harbor. Annot. Zool. Japon., vol. 1, pp. 79-81, pi. 4. BALDASSERONI, V. 1915. Chetognati. Raccolte Planktoniche fatte dalla R. N. "Ligura" nel viaggio di circonnavigazione del 1903-05, vol. 2, pp. 85-117, pis. 6-7. BERANECK, E. 1895. Les Che"tognathes de la Bale d'Amboine. Rev. Suiese Zool., vol. 3, pp. 137-159, pi. 4. CLEVE, P. T. ; r 1900. The seasonal distribution of Atlantic plankton organisms. Goteborg. Vetensk. Vitterhets-Samhull. Handl. F., vol. 4, Heft 3, No. 3, 368 pp. CONANT, F. S. 1895. Description of two new chaetognaths. J. Hopkins Univ., Circ., vol. 14, pp. 77-78, pi. 1. COSTA, A. 1869. Di un nuovo genere di Chetognati. Ann. Mus. Zool. Univ. Napoli, vol. 5, pp. 54-57. DONCASTER, L. 1902. Chaetognatha, with a note on the variation and distribution of the group. Fauna and Geogr. Maldive-Laccadive Archip., vol. 1, pp. 209-218, pi. 13, figs. 39-40 in text. ESTERLY, C. O. 1914. A study of the occurrence and manner of distribution of the ctenophora of the San Diego region. Univ. Calif. Publ. Zool., vol. 13, pp. 21-38. 276 BULLETIN 100, UNITED STATES NATIONAL MUSEUM. FOWLER, G. H. 1905. Biscayan plankton collected during a cruise of H. M. S. Research, 1900. Part 3. The Chaetognatha. Trans. Linn. Soc. London, ser. 2, vol. 10, pp. 55-87, pis. 4-7. 1906. The Chaetognatha of the Siboga expedition, with a discussion of the synonymy and distribution of the group. Siboga Exped. Monogr. No. 21, 86 pp., 3 pis., 6 maps. GERMAIN, L., and JOUBIN, L. 1912. Note sur quelques Che'tognathes nouveaux des croisieres de S. A. S. le Prince de Monaco. Bull. Inst. Oc&inogr., No. 228, 14 pp., 15 figs, in text. GRASSI, B. 1881. Intorno ai Chetognati. Atti Inst. Lombard, ser. 2, vol. 14, pp. 193-213. 1883. I Chetognati. Fauna u. Flora Golfe Neapel, Monogr. No. 5, 126 pp., 13 pis. KROHN, A. 1853. Nachtragliche Bemerkungen iiber den Bau der Gattung Sagitta, nebst der Beschreibung einiger neunen Arten. Arch. Naturgesch., vol. 19, No. 1, pp. 266-277. KRUMBACH, T. 1903. Ueber die Greifhaken der Chatognathen. Zool. Jahrb. Abt. System. vol. 18, pp. 579-646, figs. A-U in text. MICHAEL, E. L. 1911. Classification and Vertical Distribution of the Chaetognatha of the San Diego region, including redescriptions of some doubtful species of the group. Univ. Calif. Publ. Zool., vol. 8, pp. 21-186, pis. 1-8. 1913. Sagitta californica, n. sp. from the San Diego region, including remarks on its variation and distribution. Univ. Calif. Publ. Zool., vol. 11, pp. 89-126, pi. 2. 1916. Dependence of Marine Biology upon Hydrography and necessity of quanti- tative biological research. Univ. Calif. Publ. Zool., vol. 15, pp. I-XXIII. MOBIUS, K. 1875. Vennes. Die Expedition zur physicalisch-chemischen und biologischen Untersuchung der Nordsee im Sommer 1872. Wiss. Meeresuntersuch., n. F., Abt. Kiel, vol. 2, pp. 158-159. D'ORBIGNY, A. 1843. Voyage dans 1'Amerique m&idonale. Paris, quarto, vol. 5, pp. 140-144. VON RITTER-ZAHONY, R. 1908. Chatognathen. Ber. Comm. Erforsch. ostl. Mittelmeer, Zool. Ergebn., vol. 14, 18 pp., 1 pi. 1909a. Expeditionen S. M. Schiff Pola in das Rote Meer, nordliche und sudliche Halfte, 1895-96, 1897-98. Chatognathen. Ber. Comm. ozeanogr. Forsch., Zool. Ergebn., vol. 27, pp. 43-54, 4 figs, in text. 19096. Die Chatognathen der Gazelle Expedition. Zool. Anz., vol. 34, pp. 783-793, fig.l. 1910. Die Fauna Siidwest-Australiens. Lief. 3. Chatognathen. Ergebn. Ham- burg, siidw.-austral. Forsch., vol. 3, pp. 125-126. 1911. Revision der Chatognethen. Deutsche siidpolar Exped., vol. 13, Zool. p. 5, 71 pp., 51 figs, in text. STEINHAUS, O. 1896. Die Verbreitung der Chatognathen im sudatlantischen und indischen Ozean. Inaug. Diss. Kiel, vol. 5, No. 13, 49 pp., 1 pi., 2 charts. STRODTMANN, S. 1892. Die Systematik der Chatognathen und die geographische Verbreitung der einzelen Arten im nordatlantischen Ozean. Inaug. Diss. Kiel, vol. 2, No. 10, 47 pp., 2 pis. CHAETOGNATHA COLLECTED BY STEAMER ALBATROSS. 277 EXPLANATION OF PLATES. [All figures drawn with camera luclda.] PLATE 34. FIG. 1. Sagitta philippini, new species. X 18. . 2. Anterior teeth and few of posterior teeth of S. philippini. X 250. 3. Vestibular ridge of S. philippini. X 250. 4. Seizing jaws of S. philippini. X 800. PLATE 35. FIG. 5. Sagitta pukhra Doncaster. X 8. 6. Sagitta bedoti Be>aneck. X 8. 7. Sagitta ferox Doncaster. X 8. 8. Sagitta dccipiens Fowler. X 18. 9. Sagitta neglecta Aida. X 18. PLATE 36. FIG. 10. Eukrohnia richardi Germain and Joubin. X 6. 11. Pterosagitta draco (Krohn). X 25. 12. Pterosagitta draco with collarette partly torn away and with no vestige of ovary or seminal vesicle. X 25. 13. Pterosagitta draco with fully mature ovaries and with collarette almost entirely missing. X 25. PLATE 37. FIG. 14. Teeth of Eukrohnia hamata (Mobius). X 100. 15. Teeth of Eukrohnia richardi Germain and Joubin. X 100. 16. Vestibular ridge of Sagitta minima Grassi. X 400. 17. Seizing jaw of Sagitta minima. X 800. 18. Portion of vestibular ridge and posterior teeth of Sagitta dccipiens Fowler. X 400. 19. Vestibular ridge of Sagitta pukhra Doncaster. X 400. 20. Vestibular ridge of Sagitta bedoti Be"raneck. X 400. 21. Vestibular ridge of Sagitta ferox Doncaster. X 400. 22. Seizing jaw of Sagitta decipiens Fowler. X 800. 23. Seizing jaw of Sagitta pukhra Doncaster. X 800. 24. Seizing jaw of Sagitta bedoti Be>aneck. X 800. 25. Seizing jaw of Sagitta ferox Doncaster. X 800. 26. Seizing jaw of Eukrohnia richardi Germain and Joubin. X 800. 27. Seizing jaw of Eukrohnia hamata (Mobius). X 800. PLATE 38. FIG. 28. Posterior extremity of a small mature Sagitta enflata Grassi. X 25. 29. Posterior extremity of a mature Sagitta minima Grassi. X 25. 30. Ventral view of anterior extremity of Sagitta bedoti Be"raneck. X 50. U. S. NATIONAL MUSEUM BULLETIN ICO PL. 34 CHAETOGNATHA COLLECTED IN PHILIPPINE ISLANDS. FOR EXPLANATION OF PLATE SEE PAQE 277. U. S. NATIONAL MUSEUM BULLETIN 100 PL. 35 CHAETOGNATHA COLLECTED IN PHILIPPINE WATERS. U. S. NATIONAL MUSEUM BULLETIN 100 PL. 36 S CHAETOGNATHA COLLECTED IN PHILIPPINE WATERS. U. S. NATIONAL MUSEUM BULLETIN 100 PL. 37 23 24 25 CHAETOGNATHA COLLECTED IN PHILIPPINE WATERS. U. S. NATIONAL MUSEUM BULLETIN 100 PL. li. CHAETOGNATHA COLLECTED IN PHILIPPINE WATERS. INDEX. Page. bedotil (Sagitta), distribution of 257 measurements of 257 Philippine records of occur- rence 258 redescription of 255 chaetognatha, comparison of Philippine and San Diego faunas 236, 270 distribution relative to fishery problems . 237, 272 distribution relative to up- welling water 271,275 key to genera of 238 significance of variability in number of teeth 272 species collected 235 validity of genera 237 decipiens (Sagitta), distribution of 255 measurements of 255 Philippine records of oc- currence 256 redescription of 254 resemblance to S. phil- ippini 241 variability in number of teeth 254 distribution of E ukrohnia hamata 267 Krohnitta subtilis 270 Pterosagitta draco 265 Sagitta bedoti... 257 S.decipiens 255 ferox. neglecta planktonis pulchra serratodentata draco (Pterosagitta), distribution of measurements of Philippine records of oc- rence variability in number of teeth enflata (Sagitta), distribution of measurements of peculiarities in maturation of Philippine records of occur- rence variability in number of teeth Eukrohnia, key to species of validity of ferox (Sagitta), distribution of 260 measurements of 260 Philippine records of occur- rence 262 redescription of 259 resemblance to S. robusta 259 genera, key to 238 validity of 237 hamata (Eukrohnia), distribution of 267 variability in number of teeth 266 hexaptera (Sagitta), distribution of 246 loss of teeth 245 measurements of 245 Philippine records of oc- currence 247 Heterokrohnia (validity of) 237 Keys, discussion of 237 to genera 238 species of Eukrohnia 240 Sagitta 238 Spadella 239 Krohnitella (validity of) 237 Krohnitta (validity of) 237 macrocephala (Sagitta) 250 measurements of Eukrohnia richardi 267 Krohnitta subtilis 269 Pterosagitta draco 265 Sagitta bedoti 257 S.decipiens 255 enflata 243 ferox 260 hexaptera 245 neglecta 258 pulchra 252 serratodentata 249 minima (Sagitta), redescription of 248 neglecta ( Sagitta), distribution of 258 measurements of 258 variability in number of teeth 258 philippini (Sagitta), description of 240 resemblance to S. decip- iens 241 validity of 242 planktonis (Sagitta), distribution of 263 Philippine records of oc- currence 264 variability in number of teeth 263 Pseudosagitta (validity of) 237 Pterosagitta (validity of) 237 m IV INDEX. Page. pulchra (Sagitta), distribution of 252 Philippine records of oc- currence 264 variability in number of teeth 263 richardi (Eukrohnia), measurements of 267 validityof 267 Sagitta, key to species of 238 species collected 235 validityof 237 serratodentata (Sagitta), distribution of 250 measurements of . . . 249 Philippine records of occurrence 250 variability in num- ber of teeth 249 Page. Spadella, key to species of 239 validityof 237 subtilis ( Krohnitta), distribution of 270 measurements of 269 teeth (variability in number of), Eukrohnia hamata... 266 Pteros agit- ta draco.. 264 Sagitta de- cipiens . . . 254 S. enflata. . . 242 neglecta . 258 plankto- nis 263 serrato- dentata 249 Signifi- cance of 272 o UC SOUTHERN REGIONAL LIBRARY FACILITY 001 298 800 2 3 1205030163362 QL THE LIBRARY UNIVERSITY OF CALIFORNIA 1 Santa Barbara THIS BOOK IS DUE ON THE LAST DATE STAMPED BELOW. U 477 2 * 50m-l,'63(D4743s8)476