of Edinburgh, Session 1879-80. 
721 
The first of these characters he had admitted to be insufficient 
alone, and I am unable to recognise the value of the other two. It 
is difficult to understand how the shape of the stigmata can be a 
characteristic of much value, and the statement that oval stigmata 
are characteristic of the Synascidice may be easily refuted, as many 
Simple Ascidians have oval stigmata, while a species of Colella 
(. Aplidium pedunculatum , Q. and G.), an undoubted Synascidian, has 
very long slit-like stigmata with parallel sides. The stage of develop- 
ment in which the embryo is hatched cannot be considered of much 
importance, as it seems to vary in closely-allied forms ; and Giard’s 
generalisation that only in Compound Ascidians does the embryo 
remain in the egg-membrane till far advanced in development, will 
certainly not hold, as Kupffer * describes and figures the embryo of 
Molgula macrosiphonica as being still, when almost completely 
developed, covered by the “ eihaut.” This is also the case in several 
others of the few Simple Ascidians, the development of which has 
been observed. 
In conclusion, it appears that the power of reproducing by gemma- 
tion is the only difference of more than generic rank between Glavelina 
and Ciona , while Edeinascidia , a new genus of the ClaveliniDjE 
might, were it not for the fact that it reproduces by budding, and 
that the individuals are united into colonies, be included in Ciona. 
In Clavelina in an adult colony, the stolons connecting the bases of 
the individuals often atrophy and in places entirely disappear, leaving 
the individuals without any connection. They are now practically 
Simple Ascidians. In Edeinascidia , the same seems to happen ; 
and I believe that if a Ciona intestinalis , a solitary Clavelina lepadi- 
formis, and a solitary Edeinascidia turbinata were submitted to a 
naturalist who did not know the several species, he would declare 
that they were all Simple Ascidians, that the Edeinascidia and the 
Ciona were species of the same genus, and that the Clavelina was a 
nearly allied one. This would be the natural arrangement were 
it not for the budding, which, however, should be considered of 
sufficient importance to characterise a family, and, therefore, I unite 
those Simple Ascidians which reproduce by gemmation and form 
colonies, including the genera Clavelina , Edeinascidia , Perophora 
* ‘ ‘ Entwickelung der einfachen Ascidien,” Archiv fur Microscopische 
Anatomie, 1872. 
