64 



The pharyngeal epithelium of Ceplialodiscus closely resembles in structure the epidermis 

 of the oral side of the operculum, with which it is continuous through the nmuth. It is composed, 

 for the most part, of deeply staining cells, with numerous nuclei in man\- layers. These probably 

 indicate a columnar epithelium composed of very long and narrow cells. Clear evidence of 

 ciliation is noticed in various preparations, and it is probable that the whole of the epithelium 

 which has this character is ciliated. The remainder of the pharyngeal epithelium consists of the 

 dorso-lateral regions which Masterman has termed the "pleurochords", and of the walls of the 

 ti-ill-slits with which these are continuous. The epithelium is here vacuolated, and takes up staining 

 materials with much less readiness than the other part, the nuclei being much less numerous. 



The differentiation of the pleurochords is of course well marked in C. dodecalophus^ and 

 it is also particularly evident in the neuters of C. sibogae. In this case the distinction between 

 the deeply staining epithelium of most of the pharynx and the slightly stained pleurochords is 

 very marked. The species seems to be further remarkable for possessing an unusually distinct 

 hypopharyngeal groove (PI. VIII, figs. 90—92). It is difficult to speak w'ith certainty as to the 

 reality of a distinction, in this respect, between the species, as the condition of the pharynx 

 differs so w'idelv, in different individuals of the same species, in different degrees of contraction. 

 Thus C. dodecalophtis may have a very well marked median ventral groove, while in fig. 157, 

 for instance, nothing of the kind is visible. In C. lcvi)iscni the ]jharynx is usually found flattened 

 in an antero-posterior direction. 



N o t och o r d. 



I use this term as indicating the homology of this organ with the structure in Balano- 

 glossus called by that name by Bai-eson (84, p. 228; 86, pp. 550, 562) and compared by iiim 

 with the notochord of Chordata. Whether the latter homology holds good is an open question, 

 but I think there can be no doui)t that the "notochord" of Cephalodiscus is the homologue of 

 the "notochord" of Balanogflossus. The organ has been termed the "Eicheldarm" bv Spengel 

 (in Balanoglossus), the "stomochord" by Willev (99, 1, p. 234; 99, 2, p. 224) and the "sub- 

 neural gland" by Mastermax (97, 2, p. 351). 



The notochord of Cephalodiscus is a tubular ') structure which ]M-(jbably always opens 

 into the anterior wall of the dorsal diverticulum of ihc pharynx (fig. i on p. 23), although the 

 opening is not as a rule easy to demonstrate. The organ lies in septum ' .,, on the dorsal .side 

 of that ])art of the anterior body-cavil\- which lies in the- pn^boscis-stalk (PI. I\', figs. 33, 34, 

 42; PI. XIII, fig. 181 ; PI. X, figs. 113 — 117) and on the ventral side of the two collar-cavities. 

 The dorsal mesentery of the collar supports the notochord along its whole length ; and, as is 

 well shewn in Masterman's figures (03, PI. XXXII, figs, i, 2), the length of the mesentery 

 becomes less as it passes forwards, so that at its ape.x the notochord almost reaches the central 

 nervous sy.stem. At its anterior end the notochord is intimately connected with the pericardium 

 (PI. IV, fig. 33; PI. XIII, fig. 181 ; PI. XI, fig. 138) to which its relations are the same as 

 those found in Balanoglossus. The resemblance to this animal is further increased by the presence 

 of a glomerulus (see p. 29) on the ventral side of the notochord. 



l) SCHEPOTIEIF (04, p. 15) denies the tubular chaiacler. 



