DOROCIDARIS PAPILLATA. 257 



serrated spines, and specimens possessing thicker smooth spines, fine sec- 

 ondary spines seemed to leave no differences of any value. The abactinal 

 system identical in the two forms. These differences in the spines are 

 probably the result of wear, for the young of both the varieties are 

 identical. In one case the serration becomes worn, stimulating the growtb 

 of the spines, as is the case also in Heterocentrotus, where broken or 

 bruised spines frequently exceed in size the spines which have grown 

 normally. As far as I have observed, in the genus Cidaris the spines 

 seem to vary but little in one and the same species, excepting those 

 modifications which are due to the very variable size of the spines upon 

 different parts of the test; but they are all after one pattern; the spines 

 round the actinostome being the only ones which, owing to the constant 

 wear, are liable to be worn out of their regular pattern. Such is at 

 least the case in Cidaris tribuloides, metularia ; P. imperialis, baculosa, gi- 

 gantea; Goniocidaris geranioides, S. bispinosa. 



The tubercles are not perforated in small specimens, and it is only long 

 after all the specific characters (PL IP. f. S) are well developed, that the 

 perforation appears, showing that too great value ought not to be attached 

 to this character. The function of this perforation will only be satisfactorily 

 settled from an examination of living specimens. 



The mode of growth of the buccal plates of the Cidaridae is totally un- 

 like that of all the other Echinidae, with the exception, probably, of Asthe- 

 nosoma. These plates perform the part of ambulacral and interambulacral 

 plates, and appear near the test at first, forming in full-grown specimens 

 rows made up of more than two plates, as in the Palaechinidae, suggesting 

 that the test of Palaechinidae must have been made up of plates homologous 

 to the buccal plates of Cidaris. The test, of course, would then have been 

 capable of considerable compression and change of outline, as is the case in 

 Astrop3'ga and Asthenosoma. This similarity is very striking in young 

 Cidaridae, where the number of coronal plates is small, and when the young- 

 Sea-urchin seems to consist almost entirely of an abactinal and an actinal 

 system, separated by a narrow band of coronal plates. Let this narrow 

 band of coronal plates disappear entirely, and the buccal plates take a corre- 

 spondingly great development, and we have a Palaechinus made up of small 

 ambulacral and interambulacral plates consisting of several rows, and con- 

 tinuous from the teeth to the abactinal system, similar to that discovered by 

 Meek and Worthen, the whole test surmounted by short spines, articulating 



