534 Dr. F. A. Bather — Notes on Yunnan Cystidea. 



posterior is always the longest: in I, 2 it is 8 mm. long; next conies 

 the left anterior, 5-4 mm". ; then the right anterior, 4*5 mm. ; and 

 lastly the riglit posterior, 3-3 ram. The length of the peristome 

 between the forks is 7 mm. The conclusion to which tliese facts 

 lead is that the four-rayed and approximately quadrangular 

 subvective system of Sinocystis is really a modification of the three- 

 rayed system, which I have previously held to be the primitive 

 arrangement in Pelmatozoa (1900, Treatise on Zoology, p. 11, and 

 elsewhere). On this view, the true or primitive oral centre lies at 

 the axil of the left fork ; the anterior of the two branches on the 

 left is the true anterior ray, and it is noteworthy that, in I, 2 at any 

 rate, its line if produced almost coincides with the line joining 

 hydropore and anus; this, then, marks the true or primitive sagittal 

 plane, coincident, as it should be, with the M plane (Text-figs. 2, 4) ; 



hydropore 





"^ i, W 2. pf^r-iprocc 2. 



Fig. 8. — Sinocystis loczyi : diagrams, taken as accurately as possible from 

 specimens I, 1, 2, and 3, to show the varying relations of the subvective 

 system to the thecal openings. All are oriented with the anal plane 

 running N. and S., the oral pole being taken as midway between the 

 forks. If the oral pole were at the origin of the left-hand fork, then a line 

 joining it with the anal pole would pass through the hydropore, and (in 

 2 and 3 at any rate) would be continuous with the anterior branch of the 

 fork. This line would then represent the primitive sagittal plane, and the 

 branch would be the anterior of the primitive three rays (of. fig. 4, antea, 

 also Treatise on Zoology, 1900, p. 11, fig. IX). Nat. size. 



the posterior of the two left-hand branches is the primitive left 

 posterior; and the line of the peristome marks the primitive right 

 posterior branch. This last branch (one supposes) after a time bent 

 slightly towards the anus, and gave off a brauch, which in Sinocystis 

 is the right anterior. This is precisely the same change as took place 

 in the evolution of any normal fi.Te-rayed pelmatozoon, but there the 

 left posterior branch also forked in the same way, thus completing 

 the quintet. 



The Brachiole-facets, which, owing to the biserial structure of the 

 cystid brachiole in general, are composed of two halves, are seen in 

 I, 1, 2, and 4, but by no means clearly (Heed, pi. I, fig. 2a, left 

 anterior branch, is the cleaiest representation). In some cases there 

 is a suggestion of more than one facet at the end of a main branch 

 (see Reed, pi. I, fig. 4, right anterior branch); if there were actually 

 two facets, it would imply a forking of the branch, in which there 

 is nothing impossible. 



The relation of the adjacent thecal plates to the Subvective 



