vol. 2,bt. 2.] A TAXONOMIC STUDY OF THE SALPIDAE METCALF. 43 



for he shows the peripharyngeal bands connecting dorsally with the 

 horns of the dorsal tubercle instead of with the anterior end of the 

 gill. The latter is of course the arrangement not only in all the 

 Salpidae, but in all Tunicata. 



The results of our studies of the aggregated form agree with 

 Ritter's, except for a few minor details. We show differently the 

 visceral muscle prolonged into the base of the post-abdomen from 

 the median line of the united posterior branches of body muscle IV. 

 We show three more atrial sphincters than he. We show one more 

 sphincter muscle in the lower lip and three instead of four sphincters 

 in the upper lip, and slightly different connections for the sphincters 

 of both lips. The zooids Ritter describes are slightly younger than 

 the youngest we studied and much smaller than our three larger 

 specimens. 



Ihle's careful review of this species showed the accuracy of Rit- 

 ter's studies and the confusion in 'the work of Apstein, who refused 

 recognition to Cyclosalpa balceri as a species distinct from C.jloridana. 



CYCLOSALPA VIRGULA (Vogt 1854). 



Salpa virgula Vogt, 1854. 



Salpa dolichosoma Todaro 1883. 



Cyclosalpa dolichosoma-virgula Herdman, 1888. 



In the aggregated form of this species there is greater bilateral 

 asymmetry than in any other species. Our studies of the species 

 are based upon finely preserved material obtained from the Zoolog- 

 ical Station in Naples. There are 11 individuals of the aggregated 

 form and one large solitary individual. The solitary individual, and 

 one of the aggregated zooids are deposited in the United States 

 National Museum: Cat. No. 6465, (aggregated form), from Naples 

 Zoological Station; one specimen. Cat. No. 6475, (solitary form) 

 from Naples Zoological Station; one specimen. 



CYCLOSALPA VIRGULA, solitary form. 



Plate 11, figs. 26 and 27. 



The muscles of this form have been well described by Streiff (1908), 

 and detailed description is hardly needed here. Observation of 

 our figure will show the points of comparative interest. The inter- 

 mediate muscle is continuous across the mid-ventral line, as in the 

 solitary forms of all Cyclosalpas. All the body muscles are inter- 

 rupted dorsally and ventrally. The forward prolongation of the 

 dorsal ends of the sixth body muscles, seen in the solitary C. bakeri, 

 is, in C. mrgula, carried much further, the dorsal ends of the sixth 

 and first muscles being thus united. Similarly the ventral ends of 



