150 WALTER GARSTANG, M.A. 



the other hand, Hancock's A. rubicunda, Herdman's A. lata (from 

 Loch Long), and my specimens agree in being merely minutely 

 denticulated along the edge of the lamina. 



It is true, therefore, that we have at last arrived at a point 

 wherein some of the north Atlantic forms agree to differ from their 

 Mediterranean relatives ; but he would be rash who would distinguish 

 the species upon this ground alone, in view of the numerous cross- 

 resemblances in other respects. 



The prsebranchial zone is minutely tuberculated in my specimens 

 just as in Traustedt's. 



Altogether, therefore, there appears to be no sound reason why 

 the numerous mentuloid forms mentioned in this paper should not 

 be grouped together into one species and entitled Ascidia mentula. 

 Some other "species" might even be added to the list. Heller's 

 A. rubescens has rightly been included by Roule as a young individual 

 of the species, and it is just possible that Herdman's A. fusiformis 

 (If inches long ; three specimens) is merely a young condition also. 



It is difficult to form an opinion upon Hancock's A. plana and 

 A. alderi; but they appear to belong to this species also. 



I cannot hope to have altogether avoided error in the course of 

 this paper, but I have certainly endeavoured to do so ; and I trust 

 that, as an attempt to throw a little light upon some of our British 

 Tunicates, my essay will not be without useful results. 



Moreover, it would seem to be serviceable if a word or two should 

 be said upon the desirability of keeping in mind the facts of variation, 

 and of adopting some method by which the broad phenomena of 

 variability within the limits of a species can be properly and 

 systematically recorded. 



It is now a truism that variation does not only consist in the 

 manifestation of irregular abnormalities. The commonest Anemone 

 of our sea coasts, Actinia equina, Linn., sufficiently testifies to the 

 existence of a fixity and a stability even in variation. Yet it would be 

 a strange misconception of the species-idea that would lead anyone to 

 specifically separate the more constantly varieties of Actinia equina or 

 of Gylista undata from one another simply on the ground of that 

 constancy. 



