330 DR. W. A. HERDMAN ON INDIVIDUAL VARIATION 



case of marked and permanent varieties, and not merely of indi- 

 vidual variation. The irregular portion of a branchial sac of 

 Ascidia scabra, which I recently figured in the Journal of this 

 Society (anted, p. 284, PL XVII. fig. 3), I regard, on the other 

 hand, as merely a local abnormality which I think may be accounted 

 for by the splitting or branching of several neighbouring transverse 

 vessels. I have since met with an exactly similar case of great irre- 

 gularity in a branchial sac of Ascidia virginea, 0. F. Miiller. 



In order to try and determine the extent of individual varia- 

 tion in the branchial sac, and so satisfy myself as to what charac- 

 ters are most constant and may be relied upon in the determina- 

 tion of species, I lately examined minutely the branchial sacs of 

 several of our commonest species of simple Ascidians, of which I 

 had a sufficient number of specimens at my disposal. The con- 

 clusion I have come to is interesting, but rather unsatisfactory — 

 namelv, that certain characters, such as the relative sizes and 

 arrangement of the transverse vessels, the number and position of 

 the internal longitudinal bars, the shape of the meshes, and the 

 number of stigmata they contain, are highly characteristic of 

 some species, and not at all so of others # . 



A marked example of the latter class is Styela grossularia, 

 v. Beneden, which has such a variable branchial sac that it is almost 

 impossible to obtain specific characters from its details. This spe- 

 cies belongs to the Cynthiidae, one of the characters of which is that 

 the branchial sac is longitudinally folded. The normal number of 

 folds in the genus Styela is eight, four on each side ; but in this spe- 

 cies the folds are almost obsolete, being entirely wanting on the left 

 side, and reduced to one on the right. Even that one can hardly be 

 called a fold ; it is merely a slight bulging inwards, or projection 

 of the branchial sac, on which there are a number of closely approxi- 

 mated internal longitudinal bars. This single representative of 

 the euormous folds found in most Cynthiidse and Molgulidse is 

 situated in the dorsal part of the right side, and is separated 

 from the dorsal lamina by a broad space free from internal longi- 

 tudinal bars. A similar space is present on the left side of the 

 dorsal lamina and two others at the ventral edge of the sac, one 

 on each side of the endostyle. These spaces are always larger 

 than the ordinary meshes ; but they vary greatly in size in dif- 



* I hare lately arrived at similar results after an examination of the so-called 

 " olfactory tubercle " as a specific character. (See ' Proc. Roy. Pbys. Soc. Edin.' 

 vol. vi.) 



