EEPORT ON THE OPHIUROIDEA. 259 



Passing upward through the mouth of a Gorgonocephahis, and getting above the 

 mouth papillas (c?) (PL XLV. fig. 2) and tentacles [r), we come to the usual contractile 

 aperture, which may weU be called the stomach sphincter {du). It is considerably 

 wrinkled or even a little papillose on its border, and opens into a flattened digestive 

 cavity [St). Thus far the structure is normal, but beyond this point all is novel. Instead 

 of remaining simple, the digestive cavity passes outwards and upwards into a number of 

 membranous pouches, which, in profile, present a fluted aspect {St', St"). Their outer 

 ends are attached in three ways ; first (St"), tliey stretch upwards and are strongly fixed 

 to the roof of the disk wall ; secondly, they reach horizontally and grow to the inner 

 points of the egg-bearing lobes (8, S) ; thirdly, they incline downwards, and are power- 

 fully attached at ten points encircling the mouth. Of these points five are brachial {St') 

 (fig. 4) and five interbrachial {St). It is to the outer open angle of the mouth frames 

 that the latter are attached, by a part of the floor of the digestive cavity, which is there 

 much thickened {Sf) (fig. 2). Immediately above this attachment opens out the much 

 folded and fluted interbrachial pouch {St") (fig. 4), which, at its outer end, adheres to the 

 inner points of the corresponding genital lobes ; and, above, grows fast to the roof of the 

 disk. In like manner there is a brachial attachment to the upper side of each arm {St') ; 

 and above it opens a brachial pouch which has a similar shape, and is made fast at 

 corresponding places. From these ten points the attachment of the floor of the digestive 

 cavity is continued outward over radiating lines, respectively across the interbrachial 

 spaces and along the tops of the arms quite to the body wall. This structure would 

 divide the body cavity in ten radiating compartments completely separated from each 

 other, were it not that an open space exists {8f) between the inner point of each attach- 

 ment and the stomach sphincter (fig. 2). This open space corresponds to the ring canal 

 surrounding the entrance to the stomach of Ophiurans (inner perihsemal canal, Ludwig), 

 but difi'ers in being a mere continuation of the body cavity and not a closed anniilar 

 tube. It may be seen in wider section in fig. 4. The main digestive cavity directly 

 above its own centre passes upwards to the roof of the disk as a simple cone, round 

 which appear the folds of the radiating pouches (fig. 2). To give a general notion of 

 this complex organ, we may suppose a large loose bag, having a hole at the bottom 

 (mouth), and whose periphery is gathered in numerous radiating folds, leaving within 

 a central flask-shaped open space communicating dii-ectly with these folds ; and, further, 

 that the folds are divided into ten lobes, and each lobe is attached at the bottom by a 

 radiating adhesion. 



The central portion of the digestive cavity was empty, but its lobes were stufi"ed with 

 a coagulated, yellowish, pasty' substance, which, either simple or with reagents, presented 

 no special structure under the microscope, and which contained no organic remains. 

 It might weU be the decomposition of a thick layer, which had an hepatic character, or 

 was simply epithelium. 



