172 Marine Investigations in South Africa. Vol. V. 



III.— PTYCHODERA CAPENSIS, sp. n. 



I have from time to time, in shore collectin_<7 found specimens of 

 this Enteropneust chiefly at or near low wai;er, and generally in 

 coarse sand or gravel. It appears to belong to a species not 

 hitherto described, and, in recording its occurrence in this region, 

 I add a brief description of its external features, which I hope to 

 supplement at a later date by anatomical details and a figure of the 

 animal. 



The length of the specimens procured varied from about 80 to 

 130 mm. in the normal expanded condition. Individual specimens, 

 of course, varied much at times when disturbed, but when left in 

 fresh sea-water and coarse sand seemed to assume about the same 

 proportion. The relative size of the various legions of the body 

 did not vary much. Even the caudal region, which in other 

 species has been found to vary so greatly, was in these specimens 

 about the same proportionate length, probal)ly owing to the fact 

 that this species does not show the tendency to break up so 

 markedly as in other species. Some imperfect specimens were 

 procured, but they seem to have been injured in the securing of the 

 animals. None of the specimens kept alive or preserved were 

 observed to break up. 



Proboscis. — The proboscis is of the usual size and shape. It was 

 mostly somewhat longer than broad, egg-shaped, slightly tapering to 

 the anterior end. When prying among the sand and shells, it was 

 somewhat longer and often bent in various directions. A slight 

 notch occasionally appeared on its posterior dorsal surface. In some 

 conditions the longitudinal muscles of the proboscis were seen in 

 distinct bundles. 



The colour of the proboscis was pale chrome-yellow, uniform, 

 except when this organ was expanded so that the longitudinal 

 muscles appeared as distinct stripes. 



Collar. — The dimensions of the collar were also normally fairly 

 constant, its length being usually equal to that of the proboscis. 

 Its breadth was slightly less than its length, being nari-ow in the 

 middle. The free anterior margin was always somewhat folded. 

 This region, which occupied nearly a half of the total length of the 

 collar, was of a palei' yellow than' the succeeding part, which was in 

 the form of a narrow ring. This was followed by a thin band of 

 white pigment in the form of a circular streak, a little further back 

 another thin band or yellow streak, and somewhat behind this, where 

 the collar joined the body, another. 



The collar often showed longitudinal corrugations thicker in thq 



