716 KEPOKT— 1895. 



Polycarpa which I have found on our Califoinian coast and Gooihiria dura from 

 the same locality that I am sure no zoolojrist would ever think of recognising 

 more than a specific diflerence between them, did not the one reproduce by budding 

 while the other does not. Two points, however, must be briefly dwelt upon — one 

 of resemblance, the other of difference. 



It is well known that the hj'pophyseal duct in ascidians is usually situated on 

 the ventral side of the ganglion; but it is also well known that Botry llusf oims an 

 exception to this rule, lor in this species the duct is doisal to the (janglion. 1 find 

 that Goodsiria agrees with Botryllm \\\ this peculiarity. I also find that the only 

 species of Polycarpa which I have examined with reference to the point, viz., 

 Polycarpa pomaria, possesses the same unusual character. In some cases, at least, 

 it is well nigh, if not wholly, impossible to ascertain the relation of the duct to 

 the ganglion without the aid of sections. It appears to me, consequently, that the 

 occurrence of this very exceptional condition in both Polycarpa and Goodsiria, 

 when considered together with their many other close resemblances, adds consider- 

 able weight to the belief in their close kinship. 



The difference to which I refer is the presence usually of well-marked folds in 

 the branchial sac of Polycarpa and the rudimentary condition or entire absence 

 ■ of these folds in Goodsiria. In the sfic of Goodsiria dura there is no trace of folds 

 proper, but two of the five internal longitudinal vessels on each side of the sac are 

 distinctly nearer together than are the others. A similar approximation of these 

 vessels where no true folds exist occurs in numerous sjiecies of ascidians, and 

 Herdman has given good reasons for regarding it as evidence of folds that have 

 been lost. The vessels are usually crowded on the folds more than elsewhere ; 

 the folds in some cases disappear, but the crowded vessels, being a deeper 

 morphological character, persist though on a plain surface. Now it can be shown 

 that in genera whore these folds are present as a rule, but where they may be 

 rudimentary or absent, it is, in a gener.rl way, in the larger species that they are 

 best developed, and in the smaller ones that they are rudimentary or absent. 

 From this fact and others I am di''2>osed to look upon Goodsiria as a pigmy 

 Polycarpa. 



Next, concerning the structural likenesses between Ascidia and Perophora, 

 they are close in most points and not remote in any. Perhaps the most important 

 difference is found in the branchial sncs, but this difference is interesting. Ascidia 

 has internal longitudinal vessels which are papillated, while Perophora has no 

 internal longitudinal vessels ; but it does have long interserial papillae, each of 

 w^hich is provided with two processes of variable length, one directed anteriorly, 

 the other posteriorly. Now if these processes on one series of the papillse were to 

 reach across and unite with the corresponding ones of an adjacent series of papillae, 

 internal papillated longitudinal vessels would be produced entirely similar to those 

 existing in Ascidia. Suggestively enough, individuals of at least two species of 

 Perophora have been observed in which just such a union does exist. From this 

 it would appear that the lateral processes of the papillae in Perophora are remnants 

 of internal longitudinal vessels ; and this fact, taken in connection with others, 

 inclines me to regard Perophora as a pigmy Ascidia, just as we have seen that 

 Goodsiria may be regarded as a pigmy Polycarpa. 



In comparing Goodsiria and Perophora on the one hand and Polycarpa and 

 Ascidia on the other, we find a marked contrast in each case in nearly every organ 

 of the body. 



Turning to reproduction by gemmation, the buds of Perojjhora are produced 

 by a proliferous stolon, while tho^e of Goodsiria are I'ormed directly from the body 

 of the parent zooids, the inner layer of the bud being an evagination of the parietal 

 wall of the peribrancbial sac. Are these two methods of origin of buds reducible 

 to a common type ? I may say at once that a conclusive answer to this query is 

 not yet possible, for the reason that we do not know how the first bud, i.e., the bud 

 from the embryozooid, arises in either genus. 



I cannot leave this part of the subject without saying a single word about the 

 epicardium, a structure that is certainly of much importance in connection with 

 thft budding of many compound ascidians. The only structures in Goodsiria that 



