MONOGRAPH OF THE EXISTING CBINOIDS. 417 



There is no perceptible difference in the speed of regeneration between animals 

 in water with an abundant supply of food and animals in water with a minimum 

 of food. 



There is also no difference in the speed of regeneration between animals with 

 fully developed sexual products and those from which the sexual products have 

 been given off. If several weeks before sexual maturity large portions of the 

 arms be cut away regeneration takes place on an injured arm in conjunction with 

 the undisturbed maturation of the sexual products. The pinnules remaining on 

 the proximal portion of an amputated arm develop the sexual products simul- 

 taneously with those on uninjured arms. Thus the maturation of the sexual 

 products is not affected by regenerative processes, and the material intended 

 for the former is not diverted to the latter. 



POLOR OF REGENERATING PARTS. 



In the crinoids as a whole regenerating parts are usually yellow, more rarely 

 whitish or pink, and much lighter than the remaining portions of the animal. 

 This is especially evident in dark-colored types in which the contrast between 

 the original and the new structures is often strongly marked. But in the lighter 

 colored species exceptions are frequent, and in Antedon red regenerating buds 

 on yellow arms or brown on whitish specimens are commonly met with. 



The specimens of Isocrinus and Metacnnus are often quite white in color 

 when alive, and if dark the color quickly changes or entirely disappears in alcohol. 

 In preserved material regenerating parts as a rule are lighter than the rest of the 

 specimen, though often indistinguishable. In the latter case it is quite possible 

 that the color may have been different in life. 



VISCERAL MASS. 



It is well known that in many of the comatulids the visceral mass is more 

 or less readily lost and is soon regenerated without any apparent inconvenience. 



The facility with which this occurs varies greatly in different groups. In the 

 Comasteridse it appears to be very exceptional. Only very few specimens without 

 disks have come under my notice, and in these the disks were evidently lost as 

 a result of very rough methods of capture. No case of a regenerated disk in 

 this family is known. 



In the Zygometridse, on the other hand, the loss of the visceral mass is a 

 very frequent occurrence, both under natural conditions and as a result of capture. 

 Indeed, so easily is the disk lost that it is as yet unknown in several of the species 

 of Catoptometra and Eudiocrinus, and in one of the species of Zygometra. The 

 detached disk of a species of the last-named genus served as the type specimen 

 of Loven's Hyponome sarsii, described as a recent cystidean, while another species, 

 microdiscus, was so named because of its small disk, which in the type specimen 

 is in the early stages of regeneration. 



In the Himerometridse loss of the disk occurs, but is extremely rare. 



In those species of Mariametridse which have narrow and well-separated 

 division series, such as the species of Pontwmetra and TAparometra and a few 



