REPORT ON THE ASTEROIDEA. • XXV 



when the sub-class is viewed as a whole, two modes of growth ; in the one the production of 

 parts is accelerated in relation to the growth of the starfish, in the other the production 

 of parts is retarded, or proceeds pari passu with the general development of the skeleton. 



In the first group, the production of the ambulacral elements is so rapid and prolific in 

 relation to the development of the starfish and the length of the ray, that the plate3 

 become crowded and crushed together, and their individual development, though not essen- 

 tially reduced, is greatest in a plane at right angles to the axis of the ray, the plates beino- 

 very short in the longitudinal direction of the ray, and crushed together in consequence of 

 the action of growth-pressure. The segments may become so numerous, and the growth- 

 pressure so great, that the normally single series of ambulacral tube-feet on each side of the 

 median line may be crushed into a zigzag or alternating double series of tube-feet, thus pro- 

 ducing a quadriserial arrangement of the tube-feet in each furrow. This character is secon- 

 dary, however, in my opinion, and results from the degree to which the crushing is 

 carried. I do not therefore consider the quadriserial disposition of the tube-feet to be 

 of sufficient importance to define the primary divisions of a class as previous syste- 

 matists have done. A further result of this growth-movement and crushing of parts 

 in the ambulacral skeleton is found in the form and posture of the adambulacral plates, 

 which assume a more or less crowded character, and their position becomes transverse. 

 A secondary effect of this action on the adambulacral plates is the influence on the 

 form of the mouth-plates, which is especially conspicuous in the actinostomial ring, the 

 ambulacral elements being rendered the most prominent and define the mouth character, 

 whilst the adambulacral elements are retarded, and have a relatively insignificant character. 

 In consequence of the mode of its formation, I consider the form of the mouth a secondary 

 character. I do not consider it to be of such importance as Viguier does, for although it 

 is true that all Asterids in which the ambulacral skeleton is of the retarded type, or with 

 its growth concurrent with the general development, have an adambulacral mouth, the 

 occurrence of the ambulacral mouth is simply dependent on the degree to which the growth- 

 pressure is extended, as well as the influence of other factors which determine the form of 

 the adambulacral plates. 



In the second group, in which the production of parts in the ambulacral skeleton is 

 retarded, or proceeds pari passu with the general development of the skeleton, there is 

 little or no action of growth-pressure, the ambulacral and adambulacral plates are not 

 diminished in the longitudinal direction of the ray by crushing, the ambulacral tube-feet 

 are biserial in arrangement, and the adambulacral plates are most prominent in the actino- 

 stomial ring, and define the character of the mouth. These two groups may be called the 

 Leptostroteria and the Eurystroteria respectively. I consider the last mentioned of these 

 two groups (the Eurystroteria) to be the older. Embryology supports this view, for at an 

 early stage of a Leptostroterate Asterid (in which when adult the ambulacral tube-feet are 

 quadriserial and the mouth ambulacral), the ambulacral skeleton shows no effects of growth- 



(ZOOL. CHALL. EXP. — PART LI. 1888.) d 



