MONOGRAPH OF THE EXISTING CRINOIDS. 



121 



most forms are definitely and rather narrowly limited. From the 30-armed 

 species it is but a step to those with 40 arms, and so on up to the highest limit, 

 nearly 200. 



The smaller arm numbers, up to 12 or 14, have long: been conceded to be the 

 result of fracture and subsequent reparation. Since there is a continuous line 

 of species possessing all the intermediate arm numbers from 10 to about 200, the 

 conclusion, in the absence of evidence to the contrary, is logical that all the arm 

 niunbers above 10 result from fracture and subsequent repair by division series 

 and paired or further multiplied arms, instead of by single arms resembling the 

 one lost. 



I have examined a series of specimens of species of Crinometra including 

 practically every stage from the 10-armed young to the (usually) 30-armed adult, 

 and all of them were increasing the number of their arms by fracture and subse- 

 quent reparation. Similar complete series have been studied in Capillaster niulti- 

 radzata, and in Comanthus parvicirra, and very numerous instances of intermediate 

 stages have come to my notice in other species distributed in almost all of the 

 multibrachiate genera, so that I have no hesitation whatever in making the 

 unqualified statement that any increase in the number of arms over 10 is exclusively 

 brought about by the casting off of one or more of the original 10 arms of the 

 young and the regeneration in its place of an axillary bearing two arms or further 

 division series and arms. 



In many species, including all of those which never develop division series 

 further than the second, the original arm, which is usually lost at the brachial 

 synarthrj;-, more rarely at the first brachial syzygy, is replaced by an axillary 

 bearing two arms only. This is the most primitive type of interpolated multi- 

 brachiate condition and is only a slight step in advance of the more or less acci- 

 dental assumption of additional arms, as seen in Antedon hifda — a rare latent 

 tendency become common and of regular occurrence. 



In the species which have more than 20 arms after the casting off of each 

 of the original 10 arms an axillary is formed on the stump from which imme- 

 diately grow, by so-called multiplicative regeneration, additional division series 

 to the full number, followed by free undivided arms to a maximum number of 

 nearly 20. Thus, no matter how large the munber of arms in the adult may be, 

 ' autotomy never normally occurs but once on each of the derivatives from the 

 IBr series. 



The ossicle formed on the stump of a lost arm is always an axillary, and 

 it is joined to the stump by the type of articulation which originally united the 

 stump to the succeeding ossicle. Thus if the original arm was discarded at the 

 synarthry between the first and second brachials, as it most commonly is. the 

 resulting division series will be composed of two elements united by synarthry ; but 

 if it be discarded at the syzygy between the third and fourtli brachials, the result- 

 ing division series will be of four elements, of which the two outer are united 

 by syzygy— that is, 4(3+4). 



