ZooL— Vol. III.] BANCROFT— COMPOUND ASCIDIANS. I41 



ascidians that died were thirteen and fifteen respectively, 

 and of these twelve in each lot were found attached to the 

 slides after death. The cause cannot have been crowding, 

 for the embryozooids were usually situated one centimeter 

 apart, or more; therefore, death must have resulted from 

 some disease or a general inability to cope with the environ- 

 ment. One of the main difficulties that the embryozooids 

 have to contend with seems to be the acquisition of a 

 sufficient quantity of food to tide them over the period 

 during which the first blastozooid is formed. The embryo- 

 zooid is not able to go on growing and assimilating 

 indefinitely until it has reached a sufficient size so that 

 the first blastozooid can be safely formed. Its existence 

 is limited to about six days; then it degenerates and the 

 growth of its bud is accelerated. But there is usually a 

 period of about a day during which the siphons of neither 

 embryozooid nor blastozooid are open, and during which 

 no food can be taken in.^ It is only very exceptionally that 

 the young colony is so vigorous that the development of 

 the first blastozooid can be effected without the resorbtion 

 of other parts of the colony besides the embryozooid. The 

 tips of the ampullae are almost invariably retracted away 

 from the edge of the test to which they formerly reached, 

 their volume is considerably diminished; and sometimes 

 the whole common vascular system is resorbed to such 

 an extent that it can hardly be detected at all.^ It is here 

 that we have a source of danger which is probably re- 

 sponsible for a good part of the mortality of these young 

 colonies. It is in the earhest stages, then, that there is 



1 Pizon (1899 and 1900) has studied in detail the length of life of the zooids in Botryllus 

 and Botrylloides ; but, as might have been expected from the fact that his colonies were 

 observed in aquaria, while thoseof the writer were observed in their natural habitat, his 

 figures do not agree very closely with those of the latter. A detailed comparison will 

 not be attempted here, but in general it may be said that while in the colonies studied 

 by the writer the adult life of any generation of zooids was equal to or longer than that 

 of the zooids studied by Pizon, the time from the beginning of the degeneration of 

 the old generation to the adult condition of the next generation (marked by the 

 opening of the siphons to the exterior) is very much shorter. This latter period was 

 sometimes even reduced to less than zero, the siphons of both generations being open 

 at the same time. 



-'For a discussion of the results which Pizon (1900) has arrived at with respect to the 

 decrease in the size of the ampullee, see Appendix. 



