376 BULLETIN 82, UNITED STATES NATIONAL MUSEUM. 



proximal side of the pinnule attachment. This is as fertile as the rest of the cord 

 within the pinnule, so that the whole structure appears to be attached to the genital 

 cord at some little distance from its end. It descends into the arm almost to the 

 brachial genital cord, the lateral branches of which are quite short. Thus in most 

 sections of the arms an ovary is to be seen on either side of the central genital cord. 

 P. H. Carpenter, quoting Semper's observation, notes that the same condition 

 occurs in Comanthina sehlegelii, and from an examination of the ventral perisome 

 of the arms suspects that it is tolerably common in the larger tropical comatulids. 

 In one specimen of Antedon lAiida Carpenter found a small but well-developed 

 ovary occupying the position of the genital cord beneath the left posterior am- 

 bulacrum of the disk. The first traces of it appeared in the sections which pass 

 through the hinder part of the spongy organ, and it extends outward to the point 

 where the primary ambulacral groove divides into the two which proceed to the 

 arms. It contained the nuclei of half a dozen ova in various stages of develop- 

 ment, some with a germinal spot and some without. 



Carpenter found a still larger and more fully developed ovary in the disk of 

 one of the three specimens of Neocomatella pulchella which he sectioned. It com- 

 menced close to the peristome and extended outward beneath the left anterior 

 ambulacrum nearly to its bifurcation, lying close down upon the upper surface of 

 the digestive tube and molded to the plications of its wall. 



In Tropiometra pictn Carpenter met with detached portions of ovaries con- 

 taining more or less fully developed ova in various parts of the body cavity, in 

 the spaces of the connective tissue network forming the lip, in the intervisceral 

 portion of the body cavity between the two parts of the coiled digestive tube, and 

 in the subtentacular canals between the genital plexus and the water vessels. 



In some sections through the arm of an unknown species belonging to the 

 Pentacrinitidse, which liad been cut for Sir Wyville Thomson, Carpenter found 

 an ovary occupying the position where the imdeveloped genital cord is normally 

 found. He himself cut sections of the arms of all the more common species of 

 that family, but in none of them found any trace of such a condition. 



Sperry found in sections through the division series of Metacnnus rotundus 

 a well-developed functional ovary in the IIIBr series, which occupied a large part 

 of the ventral side of the arm. The genital cord remained distinct nearly all the 

 way, merging in but one spot into the ovary and soon reappearing to give off 

 a branch to the succeeding pinnule. The branches of the genital cord running into 

 the pinnules were much larger than the main strand. 



In sections through the calyx of another specimen of Metacrinus rotundus 

 Sperry found a small ovary lying in the body cavity surrounded by loose connective 

 tissue. 



In Holopus Carpenter found ovaries in the lower part of the arms. 

 In the sexually mature comatulid the primitive germ cells of the genital 

 cord in the pinnules have by their development expanded the genital tube to such 

 an extraordinary extent that it usually entirely fills the genital canal. On the 

 periphery of the nearly solid mass filling the genital tube lie the developing e<rgs, 

 while the ripe eggs lie in the center. At a later stage the contents of the genital 



