vou2.pt. 2.] A TAX0N0MIC STUDY OP THE SALPIDAE METCALF. 37 



dorsal sphincters and one instead of our three ventral sphincters. 

 Brooks figures the dorsal and ventral oral retractors in this embryo, 

 but gives them different relations with the oral sphincters. We have 

 not had this embryo for study with Brooks' other specimens. 

 Brooks does not figure the atrial retractor. We find the intermediate 

 muscle and the first two body muscles approximated dorsally, but 

 not actually united as Brooks figures and describes them in the adult. 

 Part of the discrepancies between Brooks' results and ours may be 

 due to the fact that he describes the appearance of the animal, in 

 side view, and only for the embryo, while our studies are of adult 

 individuals. 



As to the structure of the aggregated form also, we find some dis- 

 crepancies between Brooks' results and ours. He shows two fewer 

 ventral oral sphincters than we. He prolongs the intermediate muscle 

 and body muscle II, instead of the intermediate muscle and body 

 muscle I, into the peduncle, carrying the ventral end of body muscle 

 I back to join that of body muscle III. Observation of our figure 

 15 on plate 5 will explain the condition in this region, which might 

 easily mislead if not studied under high magnification. Brooks 

 shows no prolongation of any branches of body muscle III (as we 

 number it) into the visceral protuberance. 



Ihle's description of the solitary form is based upon Brooks' 

 studies. He gives a good description of the aggregated form, and 

 he shows very clearly Apstein's mistake in confusing Cyclosalpa flori- 

 dana (Apstein, 1894), and C. bakeri which Apstein (1906) calls C. 

 floridana. Ihle worked upon much younger zooids of the aggre- 

 gated form than we, and our slight differences from his results may 

 be due to this fact. We show one more dorsal and two more ventral 

 sphincters than he. Also he does not figure the posterior branch of 

 the third body muscle (our number) as continued, after union with 

 its fellow, into the visceral protuberance. It is interesting to note 

 that he figures the large dorsal eye in his young zooid as pointing 

 postero-dorsally, while in our older individuals it is directed antero- 

 dorsally. This indicates a forward rotation of this eye in its later 

 development, similar to that which occurs in the aggregated Cyclo- 

 salpa pinnata. 



CYCLOSALPA BAKERI Ritter. 1905. 



C.jloridana Apstein, 1906, a, not 1894, b. 

 C. bakeri Ihle, 1910. 



This species, confused by Apstein (1906, a) with G. floridana, 

 shows more than superficial resemblance to the latter species in both 

 solitary and chain forms, but is a clearly distinct species, differing 

 in many points of structure and form. Our figures and the descrip- 



