60 WALTER GAltSTANG, M.A. 



when the group in question is in reality no natural group at all, but 

 an "artificial assemblage" composed of several quite nnrelated 

 phyla. The primary subdivision of the Ascidiacea into these three 

 sub-orders will therefore not be adopted in my Report ; the various 

 genera will be grouped into families upon morphological grounds 

 pure and simple, and will be taken as far as possible in the order of 

 their affinity. From the nature of the case it is impossible to do 

 this with perfect satisfaction, because the families of Tunicates, as 

 of other order of animals, do not form a single series; but upon 

 completing the description of the species I will present a scheme of 

 classification in which the various families will be bound together 

 according to their most probable phylogenetic relationships. 



I desire here to express my warm thanks to Professor Herdman 

 for the assistance which he has liberally given me from time to time ; 

 as regards the present paper, I am particularly indebted to him for 

 his kindness in rendering me various information concerning still 

 unpublished work of his upon members of the C lavelinidcv and upon 

 Tunicate classification* generally. I am equally indebted to Professor 

 Milnes Marshall for the excellent facilities and help which he has 

 afforded me in his Laboratories. 



Order ASCIDIACEA. 



Family 1.— CLAVELINIDiE. 



Body consisting of a thorax and abdomen connected by a slender, 

 more or less elongate, oesophageal region. Stolonial tubes arising 

 from the posterior end of the abdomen, rarely from its lateral walls. 



Test gelatinous or cartilaginous ; forming either distinct sheaths 

 round the stolonial tubes or a common mass investing them ; common 

 test never extending above the abdominal region ; apertures circular, 

 not lobed, placed near together, terminal. 



Musculature consisting almost exclusively of longitudinal bundles ; 

 transverse muscles rare. 



Branchial sac not folded ; horizontal membranes well developed, 



* Professor Herdman' s views upon the classification of the Tunicata will 

 form the subject of a comprehensive memoir in the Transactions of the 

 Linnaean Society, to which Society they were recently commuuicited. 



