322 ARTHUR WILLEY. 



and we have before us the stage with four primary stigmata 

 on each side, which have hitherto been regarded as arising by 

 four independent perforations. The budding oS, so to speak, 

 of the two intermediate slits (ii and iii) from the two first- 

 formed slits (i and iv) is subject to a good deal of variation. 

 The constriction appears to take place first from iv in most 

 cases, but occasionally it proceeds in the first case from i 

 (fig. 8). The slits ii and iii are, however, not distinct produc- 

 tions of I and IV respectively, but are essentially ajoint pro- 

 duction of the two latter slits. This is shown clearly in the 

 figures. The ends of the two first-formed slits come into in- 

 timate contact — in fact, fuse with each other prior to and during 

 the formation of the intermediate slits. 



A very interesting variation, demonstrating the common 

 origin and single nature of the two intermediate slits, is shown 

 in PI. XXXI, fig. 20, where the two slits in question are repre- 

 sented by one elongated dumb-bell shaped aperture, which 

 would later divide as indicated by the constriction in the 

 middle. 



The fact that the slits i and iv join together in giving rise 

 to II and III is alone sufficient to render it extremely probable 

 that I and iv, the two simultaneously first-formed slits, are 

 themselves derived from one original slit. This view is 

 further brought to the highest pitch of probability by the 

 occurrence of such a variation as that represented in fig. 16, 

 where the two slits in question have fused together to form one 

 large incurved orifice. In the individual from which this 

 drawing was made the two corresponding slits of the other 

 (right) side were independent. 



In the actual ontogeny, however, these two slits arise by 

 independent perforations, but in such a way as to leave no 

 doubt that they represent one slit divided into two halves by a 

 tongue-bar, the latter forming at an early stage before the 

 actual perforation of the slit, so that the two halves of the slit 

 break through separately. In the case of the secondary slits 

 of Amphioxus — that is, the slits of the primitive right side 

 which are late in appearing (see author, 33) — shortly after their 



