A TAXONOMIC STUDY ON PYROSOMA — METCALF AND HOPKINS. 249' 



species P. atlanticum. The dorsal masses of leucocytes are reduced 

 to small size. Dimensions of the zooids: average length 4.6 mm. 

 (4.2 mm.-5 mm.); average height 2.5 mm. (2.2 mm -2.8 mm.). The 

 oral and cloacal chambers are both short. Cloacal muscles rather 

 narrow and short. 



Gonads. The testis is a compact body, consisting of about 14 

 lobes. In form it is narrowly hemispherical to globose. It was 

 relatively small in the specimen studied, which was perhaps not 

 mature. In the colon}^ no eggs were found which had progressed 

 beyond the one-cell stage. Since these were all very small and the 

 testes relatively large, it seems that the zooids are protandrous. On 

 dissection, mature spermatozoa were found in the testis of one indi- 

 vidual. The ovum in this same zooid was nearly mature, but did 

 not appear to have been fertilized. In the flasklike receptacle con- 

 nected with it sperms were found (fig. 5, p. 208). 



Occurrence and distribution. The only known specimen of this 

 form was found in the Pacific Ocean, being collected northwest of the 

 Hawaiian Islands, during the Albatross Philippine Expedition, 

 1907-1910: latitude 25° 10' N.; longitude 166° 20' W., Nov. 3, 1907; 

 one specimen, Cat. No. 6409 (Type) U.S.N.M. 



PYROSOMA ATLANTICUM, form DIPLEUROSOMA Brooks, 1906. 



Plate 32, figs. 39-41; plate 34, fig. 47. 



This Pyrosoma, which we here treat as a form of P. atlanticum, 

 has already been described by Brooks under the generic name 

 Dipleurosoma. That it is not so distinct as to be made the type of 

 a separate genus is very evident to one who has had the privilege of 

 observing large collections. In fact, except for the one feature so 

 prominent, namely the flattening of the colony, it would be rather 

 impracticable to draw any lines of distinction between this form and 

 the typical P. atlanticum atlanticum, which also sometimes shows 

 more or less flattening of its colonies, especially near the open end. 

 It may well be that we do not have a distinct race represented here, 

 but merely an interesting phase, exhibiting itself in various indi- 

 viduals of the P. atlanticum group, both in the Atlantic and in the 

 Pacific Oceans. This is further indicated by the fact that there are 

 many intergrading forms which are flattened only slightly (see 

 description of P. atlanticum atlanticum). The form dipleurosoma 

 appears to be not uncommon. 



The characteristic shape of the colony is represented in figures 

 39 to 41, which are copied from Brooks's drawings of a Dipleurosoma 

 taken in the Gulf Stream. We have material from the vicinity of 

 the Philippine Islands in which the form of the colony corresponds 

 almost exactly with that shown in the above figures. The test wall 

 forms a permanently flattened tube, and, as was shown for P. ellip- 



