26 AUGUSTA ÄRNBÄCK-CHRISTTE-LTNDE, NORTHERN AND ARCTIC INVERTEBRATES. 



the velum is narrow and only two or three rows of tentacles could be observed at 

 its base. Ypsilocarpa thus agrees with Styela as to the arrangement of the atrial 

 tentacles. 



A revision of the Styelae is still wanting; it would be highly desirable. Many 

 a Cnemidocarpa has been described under the name of Styela. Perhaps that will 

 prove to be the case also with the Ypsilocarpa-i orms. Hitherto known Styelids, 

 collected in the Arctic, present no species comparable with the above-mentioned one. 

 However, there are forms dredged in other waters which seem to agree with it as 

 regards the generic characters. In Styela nordenshiöldi Michaelsen, catight in the 

 Magellanic region, the reproductive organs consist of one cylindrical gonad on each 

 side; it is of great length and irregularly curved, the ovary and the testis are not 

 separated. From a re-examination it also appears that the atrial tentacles are 

 scattered över the surface of the velum. 



In Styela curtzei Michaelsen (1900), another Magellanic form, the structure of 

 the gonad seems to be in principal the same as in Styela nordenskioldi, there is only 

 one gonad on each side. The arrangement of the atrial tentacles is also the same: 

 they are scattered över the velum. 



Styela squamosa, dredged to the south of Australia, and described by Herdman 

 (1882), resembles Ypsilocarpa in several respects. 1 In both species a reduction of 

 the branchial folds has taken place; in the former three distinct folds lie on each 

 side, the fourth is in a rudimentary state. In the latter three folds are well-deve- 

 loped on each side, the second is rudimentary, In both, the dorsal lamina is plain 

 and the larger tentacles are short and stout. In both, the surface of the test is 

 peculiarly differentiated. In Styela squamosa it is provided with soft, regular tu- 

 bercles — it is, however, impossible to get any clear idea of the last-mentioned 

 character from the figure illustrating it (1. c. pl. xviii, fig. 1). — In Ypsilocarpa 

 the surface of the test is divided into polygons. Even in Styela squamosa the re- 

 productive organs consist of one hermaphrodite gonad on each side, elongated and 

 curved, the ovary and the testis are not separated. The condition of the specimen 

 did not allow any observation of the atrial tentacles, nor has Herdman given any 

 description of their arrangement. Consequently, no decisive conclusion can, for the 

 present, be drawn from a com parison between the two species. 



If one considers the above-mentioned facts, three groups of Styelids nearly 

 allied to each other may be distinguished. 



1. Cnemidocarpae: gonads from few to many, straight or somewhat tortuous, 

 regularly arranged in a single row on each side. Atrial tentacles sparingly placed 

 in a single row at the margin of the velum. 



2. Polycarpae: gonads numerous short polycarps, diffusely scattered on each 

 side of the mantle. Atrial tentacles in a single row at the base of the velum. 



3. Ypsilocarpae: one gonad on each side, of great length, more or less U- 

 shaped. Atrial tentacles scattered över the surface of the velum. 



' During iny stay in London Uris suminer (1921) I liad opportunity of re-examining Styela squamosa 

 in British Museum, Nat. Hist.; tlie comparison here made is based on tliis re-examination. 



