345 



can I confirm Mr. Masterman's statement (p. 352) that what I have 

 termed the central blood-space opens into the anterior body-cavity. 

 I must however admit that I have not been able to convince myself 

 with certainty that the central blood-space opens into any other vessels, 

 although I have some slight evidence pointing in that direction. 



The small size of the parts in question makes it very easy, to be 

 misled with regard to the way in which one space communicates with 

 another; but while Mr. Masterman is probably right in colouring 

 certain parts of figs. 6 — 9 red, to indicate that they belong to the vas- 

 cular system, I believe that the only part which should have been 

 coloured red in figs. 2 — 4 is the uncoloured space in the interior of 

 the «subneural sinus«. Fig. 1 and probably fig. 5 also show the »sub- 

 neural sinus«. 



If this point has been demonstrated, additional support is given 

 to the view that the median notochord of Cephalodiscus is homologous 

 with that of Balanoglossus. The view is further strengthened if I am 

 right in thinking that there is some indication of the proboscis gland 

 (Spengel's »glomerules«) on the ventral side of the base of the noto- 

 chord, in the position indicated by radiating lines in figs. 8 — 10 of Mr. 

 Masterman's paper. The principal reason given for rejecting the 

 homology discussed above appears to be that the »subneural gland« of 

 Actinotrocha is formed in the immediate neighbourhood of the mouth. 

 Although our knowledge of the development of this structure in. Balano- 

 glossus is by no means complete, what evidence there is indicates that 

 the notochord of the metamorphosing Tornarla^ appears in precisely 

 the same position as the »subneural gland« of Actinotrocha. 



3. What view then are we to take of Mr. Masterman's »lateral 

 notochordsa? It is easy to convince oneself of the existence, in Cepha- 

 lodiscus, of paired dorso-lateral regions of the pharynx, which have 

 the position and the appearance shown by Mr. Masterman in PI. 

 XXIV. fig. 13 and PI. XXV. figs. 16, 17 and 19. It will be observed 

 that these structures are not sac-like diverticula, but are long grooves 

 of the pharynx, extending »throughout its whole length«. Their epi- 

 thelium is further identical in character with the epithelium of the 

 gill-sacs. Mr. Masterman's description and figures (especially figs. 17 

 and 22) seem to lead to the interesting conclusion that the pharynx of 

 Cephalodiscus is divided , like that of some species of Balanoglossus, 

 into a dorsal, branchial and a ventral, alimentary portion. The former 

 has a histological character of its epithelium, not greatly differing from 

 what has been described by Spengel'' for Balanoglossus, which is re- 



5 See Morgan, Journ. of Morphol. V. 1891. p. 423. PI. XXVI. fig. 40. 



6 T. cit. PI. II. fig. 1, PL XX. fig. 11. 



