474 
are derived from a common source, while the whole range of 
possibilities has been exhausted in ascribing to them either an 
ectodermal, an endodermal or a mesodermal origin. 
Although it is certainly true that the derivation of these and 
other structures is by no means the same in all Ascidian buds, there 
yet remain many conflicting observations, and until more light has 
been thrown on the subject, much must be left in doubt. Further 
investigation is, therefore, of value, and it is especially desirable 
to study the process in species in which it has not already been 
described, as those whose bud-development has been examined, are 
comparatively few in number. 
However much opinion may differ in regard to individual questions, 
all recent work goes to show that embryonic- and bud-development do 
not proceed along parallel lines, and that the two cannot be brought 
into harmony with each other. The Ascidian bud must be looked 
upon as a “new- formation”, acting under its own laws. The facts of 
the bud-development cannot be harmonised with the germ-layer theory. 
This conclusions is further emphasised hy results which I have 
recently obtained from a study of the process in one of the 
Clavelinidae, namely Ecteinascidia, a genus founded by 
HERDMAN in 1880 for some specimens of the Challenger Expedition. 
The material which I have had at my disposal was collected in 
Jamaica and belongs to HERDMAN’S species E. turbinata. (For a 
description of the species see Challenger Report, Zool., Vol. VI, 
p. 243—-245). The family of the Clavelinidae contains the well 
known genera Clavelina and Perophora. The bud-development 
of the former is familiar from the classic work of VAN BENEDEN and 
JULIN!), and that of the latter has recently been described by RITTER ?) 
for a new species, P. annectens, from the California coast, and 
briefly by myself?) for the eastern species, P. viridis. These are 
the only genera of the family whose bud-development has hitherto 
been studied. 
Although in external appearanee the zodids of E. turbinata 
resemble somewhat those of Clavelina, as they are quite elongated 
and the two siphons are placed anteriorly, the species shows a closer 
similarity to Perophora, both in the structure of the adults and 
the mode of development of the buds. It differs from Clavelina 
1) Archives de biologie, T. VI, 1887. 
2) Journal of Morphology, Vol. XII, No. 1, 1896. 
3) Johns Hopkins University Circulars, No. 119, June, 1895. 
