446 TUNICATA. 



in S. democratica, which form was further peculiar in the structure and origin 

 of its placenta, inasmuch as this organ was said to arise, in part at least, from 

 the ectoderm of the embryo, whereas, in other Salps, the embryo was covered 

 by protective folds and the placenta was purely follicular, i.e., maternal. 

 Recent researches, especially those of Brooks (No. XIII.) and Korotneff (No. 

 XVIII.), tend to show that this distinction was greatly overrated by Salensky, 

 and that in all probability traces of covering folds exist in S. democratica, as 

 Brooks has shown is the case in S. hexagona. Further, when we remember 

 that the folds are extremely transitory even in those forms in which they are 

 best developed, it will be agreed that the division of the Salpidae into two 

 groups on this character is inadvisable, since such a division does not appear 

 to involve any fundamental difference between the embryos. 



In the matter of the placenta also, Salensky appears to have been in error 

 and in all probability that organ is entirely follicular (maternal) in origin in 

 every case (even in .S'. democratica, Korotneff). Neither does there appear 

 to be any essential difference in the structure of this organ in the two groups. 

 Consequently, the more recent writers have ceased entirely from considering 

 the development of the Salpidae as taking place along two distinct lines, 

 regarding the differences rather as slight and unimportant modifications of 

 one and the same type.] 



The changes which occur towards the end of embryonic develop- 

 ment have been well described in S. piniiata. When the embryo is 

 approaching its final form, it passes through the dilated aperture of 

 the enveloping fold into the respiratory cavity of the parent. The 

 enveloping fold then draws back more and more (Fig. 224 A, f), 

 soon leaving the placenta also free. The latter, at this stage (Fig. 

 224 B), becomes sharply marked off from the embryo, being finally 

 connected with it merely by a narrow umbilical cord. This cord, as 

 a rule, is short, but, in S. scutigera, it attains a considerable length. 

 Finally, after the embryo has, for some time, carried on independent 

 respiratory movements, it detaches itself altogether, with the placenta 

 connected with it, from the respiratory wall of the parent and passes 

 out at birth through the atrial aperture of the latter. Vestiges of 

 the placenta and of the covering fold are retained for some time on 

 the wall of the atrial cavity of the parent. The placenta attached to 

 the embryo gradually decreases in size, but, even in somewhat well 

 developed young Salps, can be recognised as a small stalked mass 

 embedded in the mantle-substance and attached to the posterior half 

 of the endostyle. 



It has been pointed out that the development of the Salpidae is 

 direct ; a tailed larval stage such as is characteristic of the Ascidian 

 does not develop. With this is connected the apparently complete 

 absence of the chorda dorsalis. The absence of this provisional 

 structure, although explained by the direct nature of the develop- 



