• i:;i J W. G. Van Name — Bermuda Ascidians. 



New Genera. 



Diazona picta Verrill requires a new genus, which I have termed 

 Rhodozona. It is in many respects intermediate between Diazona 

 and Glavelina. In the general shape of the colony it is not unlike 

 Stereoclavella australis Herdman (8) from Australia, hut only the 

 anterior portions of the zooids project from the common test, and 

 the colony has the beautifully transparent and gelatinous character 

 of a Diazona. The branchial sac resembles Glavelina in the absence 

 of internal longitudinal bars, but the strong longitudinal muscle 

 bands along each side of the dorsal lamina, and the man}?- transverse 

 muscle bands in the mantle are very different from any species of 

 that weniis with which I am familiar. 



The genus Echinoclinum is also an intermediate form, serving to 

 unite more closely the genera Didemnum and Diplosoma, and fur- 

 nishes an additional argument for uniting the Didemnida? and 

 Diplosomidre in one family. In its zooids and in the gelatinous 

 nature of the colony, it resembles the last named genus or family. 

 The large cavities present in the test in that genus are however 

 wanting, and the large tetrahedral spicules, though peculiar in their 

 shape and arrangement, are more suggestive of the former genus. 



The two new genera of the Polystyelida*, Michaelsenia and Dian- 

 droccapa, appear to be connecting forms linking that family with 

 the Halocynthiidse and Botrylliclse respectively. In the latter genus 

 one or more of the forms included by Michaelsen (12) in his genus 

 Gynandrocarpa may also be placed, though the Bermuda form 

 approaches the genus Botryllus more closely in the general character 

 and pigmentation of the colony than any of the other species. 



Classification. 



The classification employed is based upon that of Herdman, but 

 with a number of modifications, such as separating the Perophoridiv 

 from the Clavelinidre, and uniting the Diplosomida' with the Didem- 

 nida?, as many writers have done. Moreover, I include the Clave- 

 linidse and Perophorida" among the Compound Ascidians, because 

 they are compound. They reproduce by budding and form colonies, 

 and are by no means so closely related to the Simple Ascidians as 

 some of the Polystyelidre are, though Herdman includes these among 

 the Compound Ascidians. 



I have made free use of Herdman's diagnoses of families and 

 genera, as given in his Revised Classification (7), with many changes 

 and omissions, and wish to acknowledge my obligations to that 

 author. 



