ZOOL.— Vol. III.] BANCROFT— COMPOUND ASCIDIANS. 163 



larvae were not extruded, but metamorphosed within the 

 colony and added to the number of its zooids. Later, 

 Delia Valle (1881, p. 476) discovered the budding in the 

 adult zooids, and gave other evidence against the fusion of 

 the embryozooids with the colony which produced them. 

 Lahille (1890, pp. 122, 134) supports Gegenbaur, but gives 

 no conclusive evidence. 



Seeliger (1898, p. 175) states for the Clavelinidge that 

 larvce may settle on the mother colony, and fuse with it; 

 but does not give any evidence to support this view. 



In families like the Diplosomidge and Clavelinidae, where 

 individual differences are not large, it is ordinarily impos- 

 sible to detect whether any colony is composed of several 

 united colonies or not. Among the Botryllidse, on the 

 other hand, where the individual variation is very great, it 

 might be expected that differences in color would enable 

 one to determine at a glance whether a fusion of two colonies 

 had occurred. But for the Botryllidae almost all the evi- 

 dence heretofore given is also unsatisfactory. Thus Giard 

 (1872, p. 590) stated that adjacent colonies fused; but 

 offered no evidence to substantiate his statement. Still 

 his belief in the process was so strong that he used the 

 absence of fusion in adjacent colonies as a criterion of the 

 specific distinctness of the two forms. Pizon (1892-1893, 

 pp. 196-198; 200-202) says that fusion between adjacent 

 embryozooids frequently occurs, the test first fusing and 

 then the cloacae approaching each other to form the common 

 cloacal orifice of the first system. Later he stated that on 

 five or six occasions he has found agglomerations of young 

 embryozooids of Botrylloides, which were partially fused 

 together. Sections through the mass failed in many cases 

 to show any line of separation between the test of adjacent 

 zooids, though in other cases they did. He believes that 

 from these masses single colonies are formed; but offers no 

 further evidence. 



In his last paper, however, Pizon (1900, pp. 15-16, 

 51-52, PI. IV) describes and figures an undoubted case of 

 fusion in two embryozooids of Botrylloides ruhrum. The 



