326 ARTHUR WILLEY. 



tion (PI. XXX, fig. 14, and PI. XXXI, fig. 21), and in this way 

 we arrive at the stage with six primary stigmata on each side, 

 which was also described by van Beneden and Julin for 

 Phallusia scabroides. 



Assuming that the stigmata v and vi have the value of true 

 gill-slits, and are each of them equivalent in morphological 

 value to the first four stigmata taken together, we conclude 

 that in the simple Ascidians under consideration three pairs 

 of primary gill-slits are formed in due order of succession from 

 before backwards, as follows : 



Stigmata 





TTT j- — 1st gill-slit. 



IV. 

 V = 2iid 

 VI = 3rd „ 



The irregularity in the order of appearance of these primary 

 stigmata has not, therefore, in any sense the signification 

 which van Beneden and Julin attached to it, and as soon as 

 the stigmata are properly identified with gill-slits we have 

 seen that there is no irregularity at all, so that this argu- 

 ment cannot be used against homologising the gill-slits of 

 Ascidians with those of Amphioxus. If the above table be 

 compared with that given by van Beneden and Julin, which I 

 have repeated above, it will be seen how great is the difference 

 between them. 



For convenience of description it is necessary to make a 

 mental distinction between the terms '' primary stigmata " 

 and "^ primary gill-slits." The former expression refers to the 

 six simple, undivided apertures which are developed in the 

 way above described, while the latter is a morphological term. 

 Thus while there are six primary stigmata, they only represent 

 three gill-slits. 



As time goes on, the six primary stigmata elongate very 

 considerably in the transverse direction, and eventually be- 

 come divided into two by the formation of small tongue-like 

 projections, which arise indifferently from the anterior or 



