pp:rophora annectens. 45 



crowded, but not compounded colonies. This is natural 

 size, and the number and arrangement of the zooids are 

 reproduced as faithfully as possible. 



The appearance of the zooids when taken from the 

 test is fairly well represented by fig. 3, pi. i, the outlines 

 of which were drawn by the aid of a camera lucida. As 

 compared with the other known species of the genus, the 

 approximately spherical form and the terminal position of 

 both orifices are noticeable. When the anterior end of 

 the fully expanded living zooid is looked down upon, the 

 outline presented is that of a rather broad ellipse with the 

 two orifices situated at the foci. 



The ease with which the zooids can be removed from 

 the test in the fully compounded colonies is worthy of 

 mention. In a preserved colon}^ that has been cut in 

 pieces, they may be picked from their little cavities in the 

 test, reminding one of the way in which very young frog 

 embryos may be picked from their gelatinous envelop. 

 The only points at which they seem to be adherent are 

 the extreme edges of the orifices, and the point of pas- 

 sage of the blood vessel from the body into the stolon. 

 When we come to examine the minute structure of the 

 test and the body layer in contact with it, we shall see 

 that the union between the two is much more intimate 

 than appears on gross dissection. 



Another point to which attention may well be called 

 in connection with the general appearance is the course 

 of the transverse vessels of the branchial sac. The body 

 is sufliciently transparent to permit these to be seen with 

 considerable distinctness in a good light. As may be 

 seen by fig. 3. the planes of these circular vessels are 

 not situated at a right angle to the antero-posterior axis of 

 the body, but they all converge on the dorsal side of it. 

 This convero-ence is associated with a sliofht dorsal curv- 



