1892.] 



Development of Ciona intestinalis, fyc. 



517 



Fig. 3. — Primary branchial apertures of right side of Ciona, showing the way in 

 which the stage with four slits on each side becomes established. Drawn from 

 living object. Zeiss 4 B, cam. luc. 



stigmata on each side. The slits which, form after this (5th and 

 6th, &c.) arise by independent perforation. 



The point is now to determine the origin of the first two slits, 

 and this is by no means easy. By the study of great numbers of 

 living specimens, and more especially of horizontal sections, I have 

 convinced myself that the following is what takes place. It has been 

 known for many years that the slits of Amphioxus become each di- 

 vided into two halves by the formation of a tongue-bar. In the 

 larva, however, the unpaired slits are simple, and remain simple 

 till near the end of the period of metamorphosis. During this 

 period the slits of the second row make their appearance, and very 

 soon afterwards, the primary slits being still simple, tongue-bars 

 begin to form in the secondary slits. Thus in this case the tongue- 

 bars are considerably hastened in their development. If, now, they 

 were hastened a little more, what we should see would be that the 

 two halves of the slit would become independently perforated. This 

 is actually what occurs in Ciona. I have very convincing evidence for 

 this point of view, which I hope to produce in detail, accompanied by 

 figures, later. The first four stigmata, therefore, are derived from, 

 and represent, one gill-slit. 



The primary branchial canals of van Beneden and Julin are simply 

 the first pair of gill-slits. The atrial involutions are at first nothing 

 but the ectodermic portions of a pair of gill-slits. They remain per- 

 manently in this condition in Appendicularia ; but in the fixed 

 Ascidians they become secondarily expanded to form a distinct 

 chamber. The view of van Beneden and Julin, that the visceral 

 wall of the atrium is endodermic, while the parietal wall is ecto- 

 dermic, I consider to be unfounded. 



