20 DE. J. F. GEMMILL ON THE DEVELOPMENT OF 



neck that traces of the deeper pigment still exist, and even these are lost when at a 

 later stage the notch becomes filled up. 



We see accordingly that on the right side a forward sliding movement of the wall 

 of the posterior part of the bilateral larva has taken place, and that this movement is 

 the counterpart of the migration backwards and to the left undergone by the preoral 

 lobe. 



These facts, while they are important in the ontogeny of Solaster (see under Nervous 

 System, p. 46), may also be of great importance from their bearing on general questions 

 of asteroid phylogeny. 



4. The HydroGcele Pouches. — The radial pouches of the hydroccele appear externally 

 as an incomplete circle of lobular swellings on the oral (left larval) surface around the 

 central area where the preoral lobe is becoming incorporated with the body-wall. 

 The gap left in the circle is opposite to the larval neck. Four of the pouches appear 

 practically simultaneously and are succeeded almost at once by a fifth. These form 

 a series the members of which, for reasons that will appear later, are to be numbered 

 in the dextral or watch-hand direction, as seen from the oral side, commencing with 

 pouch No. I immediately to the dextral [i. e., the larval dorsal) side of the preoral lobe. 

 In the fully-formed Solaster the normal number of rays is nine, and accordingly four 

 new pouches remain to be formed in addition to the five that have already appeared. 

 These are added one by one, extending the series onwards from pouch No. V, still in 

 the dextral direction as one looks from the oral side. There is a short pause 

 till VI appears. VII follows after an interval of similar length. VIII is delayed 

 relatively longer, and IX longer still. Indeed, IX is not added until the preoral lobe 

 has almost completely disappeared, the sucker has begun to atrophy, and the larva is 

 attaching itself by means of the first-formed tube-feet. Pouch IX at the end of the 

 series is now adjacent to pouch I at the beginning, and nothing stands in the way of 

 the completion between these pouches of the ring-canal of the hydroccele. 



Soon after they have made their appearance externally, the radial pouches of the 

 hydroccele become three-lobed in shape. This is due to the outgrowth at their sides 

 of the first tube-feet, and at their extremities of the terminal tentacle. Later, a second 

 pair of tube-feet is formed on each radial pouch between the first pair and the terminal 

 tentacle. 



The sequence in the formation of the tube-feet on the radial pouches corresponds 

 with that in which the pouches themselves were developed. For example, pouches I 

 to V are provided with two pairs of feet at a period when VIII is still in the tri-lobed 

 condition and IX has not yet made its appearance. So long as the preoral lobe is 

 large, pouch I of the hydroccele, being somewhat compressed in the angle between 

 that lobe and the disc, is not free to appear externally with as great prominence as 

 the next three pouches in the series ; but after the preoral lobe disappears, the pouch 

 in question, being no longer hidden or compressed, shows in all respects quite as 



