318 CALIFORNIA ACADEMY OF SCIENCES. [Proc. 3D Ser. 



appearance of the substance surrounding the cell and 

 that composing the yellow layer shows conclusively the way 

 in which the latter is formed. It does not, however, grow 

 exclusively by incorporating masses that were originally 

 separate, for some cells do not take on this function until 

 they are closely pressed against the outer layer, so that 

 from the first the product of their activity is in contact with 

 this layer. In some cases, especially in the siphonal ridges, 

 a number of cells do not seem to begin forming the yellow 

 substance until they have been nearly or quite surrounded 

 by it. The ultimate fate of the cells in the yellow layer 

 seems to be degeneration, for of those nearest the surface 

 nothing can be seen but a small vesicle containing some 

 refractile material in which usually not even the remains of 

 a nucleus can be made out. As there are extensive regions 

 of the test which contain no blood-vessels at all, the bladder 

 cells must be in great measure physiologically isolated from 

 the rest of the body, for they could receive nutritive fluids 

 only by way of the rather dense cellulose substance. On 

 this account I think it likely that the yellow substance is 

 formed almost entirely by a direct transformation of the 

 cellulose, and accordingly it would be interesting to know 

 the exact chemical relations between the two substances. 



Blood-vessels in the Test. 



These are usually confined to the posterior region near 

 the base of attachment, at times extending half way up 

 the free wall towards the disk; but they have never been 

 found extending into the disk. The proximal part of the 

 system is composed of a large number of pairs of ves- 

 sels running side by side, as in Ascidia and other tunicates, 

 but the terminal part in which these two vessels communi- 

 cate with each other is very greatly developed. It is much 

 more capacious and consists of large single or compound 

 and irregularly branching ampulla?, present in such num- 

 bers that the substance of the test is fairly honeycombed in 



