276 NATURAL SCIENCE. April. 



stem of Antedon is the equivalent of the larval stem of Antedon's 

 ancestors, would be at least a possible explanation, were it not for 

 one fact : the stem-ossicles of the Antedon larva are not at all of a 

 simple type, they have a very peculiar specialised structure, and, 

 broadly speaking, resemble the stem-ossicles of the Bourgueticrinidae 

 so common in the Upper Chalk. It is a mistake to call this larva 

 " pentacrinoid " ; if any genus is known to us with which it may be 

 compared, that genus is Thiollievicyinus, of the Jurassic rocks, many 

 species of which show a gradual loss of the stem and development of 

 cirri around the base of the cup. What the ancestors of Thiollieri- 

 crimis may have been we cannot say with certainty ; this only seems 

 clear, that the structure of the stem is specialised. A somewhat 

 similar structure was developed in the Carboniferous Platycrinidae, 

 from which, however, it is not likely that Thiolliericrinns was derived. 

 Such stems do not occur lower than the Carboniferous, and we must 

 infer that the ancestors of both Platycrinidae and Bourgueticrinidae 

 had stems of the primitive Palaeozoic type, with round ossicles 

 radiately ridged. From such ossicles all other types may easil}- 

 have been derived. Now, to accord with Mr. Hurst's explanation 

 of the Antedon stem, it must be supposed, either that the larval stem 

 of all crinoids has always been of this specialised type, which is 

 contrary to all a\"ailable evidence, or that its structure is a special 

 larval development. Surely it is hardly probable that an organ 

 which is only in use for a few days of the creature's life should have 

 become specialised in just the same way as the stem of the adult 

 ancestors. It is certainly easier for the ordinary mind to imagine 

 that, by acceleration of development (a principle ignored by Mr. 

 Hurst), the structure of the adult ancestor has been pushed back to 

 the larval stages of the existing Antedon. 



But the stem is by no means the only structure in the larval 

 Antedon thdit heiLts on the Recapitulation Theory. " Each transient 

 stage in the development of an individual," says Mr. Hurst, " is a 

 modification of the corresponding stage of development of its 

 ancestors." Let us see how this is borne out by the development 

 of the cup-plates. 



The most important peculiarity of the larva as distinguished 

 from the adult Antedon is the presence of a plate in the anal interradius. 

 At an early period this plate appears between two of the radials, and 

 on the same level with them ; subsequently it is lifted out from 

 between the radials, which close in under it as it passes upwards, so 

 that at last the radial circlet is quite continuous, and the anal plate 

 lies above it. At about this stage the crown is separated from the 

 stem ; shortl}' afterwards, the orals disappear, and a little later, the anal 

 plate is itself absorbed and vanishes. Now, on Mr, Hurst's theory, 

 each of these stages should be represented in the early development 

 of the ancestors of Antedon ; but this would raise far greater difficulties, 

 for the oldest crinoids do not possess this anal plate at all, at all 



