SEA-URCinxs. 465 



has been made already (p. 455) to the truncated form of many 

 of the Pjf(jast^,vs, and of (Jalcropifijus, and the comparatively 

 elongated outline of some Ilolectijpi and of Coaidus. The 

 truncation of the posterior interaml)ulacrum is associated with 

 the presence of a more or le^s defined posterior sulcus ; and, in 

 fact, whenever the periproct is on the adapical surface, this 

 sulcus is developed. The functional value of a sulcus below the 

 anus is obvious, as it would tend to restrict the passage of ftecal 

 matter to a definite channel, and so to prevent it from coming 

 in contact with the podia. But the developmental meaning 

 of the structuie would seem to concern the interference in the 

 growth of the iiitei-ambulacral plates by the periproct. These 

 plates, formed at the apex and forced downwai-ds towards the 

 ambitus, have to separate along their median sutures to pass 

 i-ound the periproct, and subsequently have to close together 

 below it. The irregularity thus caused results in a retardation 

 of their downward movement (and a consequent shortening of 

 the distance from the apex to the posterior margin of the test), 

 and in a sagging inwards of the reconstructed portions of the 

 plates to form a groove. The probability of this explanation of 

 the structure so characteristic of Pygaster, Galeroj^ygiis, and the 

 Nucleolitidfe, becomes increased when the opposite conditions 

 are considered. When the periproct is marginal or inframarginal 

 in position, the interambulacral plateg can pass freely over the 

 adapical surface until the edge of the periproct is reached. Here, 

 in consequence of the lessened width of the divided halves of the 

 area at the sides of the periproct, a delay in the progress of the 

 plates occurs. As a, result, the oncoming plate.s lecome heaped 

 up against one another above the perijn'oct, and give I'ise to the 

 elongated, carinate posterior shape which characterizes //o/ee^T/^j'jfS 

 sens, str., Comdas, and also the fSpatangoida. 



Although it must be admitted that the two opposite conditions, 

 sulcate and carinate, of the posterior interra^lius have, from a 

 teleological standpoint, an obvious and similar functional value, 

 the explanation given above seems natural in view of their 

 regular association with the position of the periproct. In the 

 case of the Spatangoida., there is the complication of a "posterior 

 siu'face " to the test, at the upper part of which the peripi'oct is 

 situated. If this surface were curved in conformity with the 

 rest of the test, the periproct would open at a point about 

 midway between the apex and the ambitus, as in PUeus or 

 Anortkopi/giis. It seems to me to be a very striking fact that, 

 in the last-named genus, no trace of a posterior sulcus is developed, 

 but that the declivity of the test is appreciably incieased as a 

 lohole in the region behind the periprcct. The Spatangoid 

 posterior surface might, then, be regarded as the product of a 

 retarded growth of the entii-e postei'ior interambulacrum owing 

 to the interfeience of the periprcct, while the carina above it 

 would be cau?jed by the same agent in its opposite infiuence. 



In the few cases where absolute circularity of outline is 

 Proc. Zool. Soc— 1912. No. XXX. ' 30 



