ON THE BRANCHIAL SAC OF SIMPLE ASCIDIANS. 329 



Fig. 12. Attenuate straight-limbed, knobbed, simple rosette, magnified. 



13. A spinulose hexactinellid spicule with knob, magnified. 



14. An hexactinellid with serrate limbs, magnified. 



15. An hexactinellid with prolonged axis, magnified. 



16. A long, attenuate serrate fibre, magnified. 



17. Plain processes of the skeleton, magnified. 



18. Clubbed cylindrical spined processes, magnified. 



19. The lattice-work at the free edge, magnified. 



20. The lattice-work of the base, magnified. 



On Individual Variation in the Branchial Sac of Simple Ascidians. 

 By W. A. Herdman, D.Sc, F.L.S., F.K.S.E. 



[Eead April 21, 1881.] 



The difficulty of determining the value of specific characters in 

 Ascidians is well known to all who have worked at the group. 

 It is now universally admitted that the old method of describing 

 merely the external appearance of the animal is insufficient ; as 

 in many cases it is impossible, from an examination of the external 

 characters alone, to determine the genus, and even in some cases 

 the family, to which the specimen belongs. Consequently, most 

 writers on the Tunicata in recent years have described in more 

 or less detail certain internal characters, including the branchial 

 sac and its related organs, the circlet of tentacles, the dorsal 

 lamina, and the olfactory tubercle. The conditions of these 

 important structures furnish most valuable generic and specific 

 characters, and an account of them should undoubtedly form part 

 of the description of an Ascidian. 



It must not be forgotten, however, that some of these charac- 

 ters, in many species, vary considerably according to the indivi- 

 dual ; or, in other words, not only do varieties exist, but most indi- 

 viduals differ slightly from each other in points which are given as 

 specific characters : this, of course, is only in certain species. 

 Hence when the number of specimens for comparison is small, 

 it is often a delicate matter to determine what is a good species. 

 My attention was first directed to this variation by reading Lacaze- 

 Duthiers's description, in his great work on the Molgubdse *, of 

 three marked varieties of branchial sac in Ctenicella Lanceplaini, 

 L.-Duth. This appears, however, from the account given, to be a 



* H. de Lacaze-Duthiers, " Histoire des Ascidies simples des cotes de France," 

 2 e partie, 'Archives de Zoologie ex peri in en tale et generale,' t. vi. p. 619, pi. xxiii. 

 figs. 9, 10, 11 (1877). 



