1876.] and Development of Antedon rosaceus. 227^ 



along the outside of the ridges that border the tentacular furrow in the 

 arms and pinnules, and which, by their peculiar property of attracting the 

 colouring-matter given off after death by other parts of the body, become 

 the well-known " red spots," have long been a puzzle to students of 

 Antedon. I have long been myself inclined to regard them as Sensory 

 organs, my surmise being founded on the fact that in the oral pinnules, 

 which have no tentacular apparatus, but which are much more irritable 

 (i. e. more susceptible of impressions which call forth contraction of the 

 arms) than the tentaculiferous pinnules, these sacculi are crowded 

 together so as to form two continuous rows *. 



Development. 



In the earliest stage at which I have been hitherto able to study the 

 Pentacrinoid larva (which may be designated as " Allman's stage f"), the 

 calyx rather resembles an inverted bell than a shallow basin, its lower 

 part being supported by the five basals, whilst its upper is surrounded by 

 the five incipient first radials which alternate with them. Alternating 

 in position with the five first radials, and at this stage resting immedi- 

 ately upon them, are five oral plates, somewhat triangular in form, 

 enclosed in valvular folds of the delicate perisome which forms the 

 general investment of the body. These oral valves, when inclined to- 

 wards each other, form a sort of five-sided pyramid, which completely 

 covers in the oral disk ; but when opened out they surround it like the 

 expanded petals of a flower. Immediately within the oral valves is a 

 thin elevated lip, from which spring the oral tentacles ; and the wide 

 mouth, leading into a funnel-shaped oesophagus, occupies the whole of the 

 area surrounded by this lip. A vertical section of the calyx in this stage 

 shows the gastric sac loosely suspended in the perivisceral cavity or 

 coelom (jpv, fig. 10), with the inner wall of which, however, the outer wall of 

 the gastric sac is connected by scattered threads and lamellss of connective 

 tissue. And it is very distinctly seen that the circular lip is formed by 

 an annular plication of the bounding membrane of the perivisceral cavity, 

 a space being left between the two folds, which constitutes the original 

 tentacular or ring-canal, re. Hence this tentacular canal is clearly 

 a derivative of the coelom, the only separation between them at this stage 

 consisting in a circlet of threads of connective tissue, which passes be- 

 tween the inner and outer folds of the lip. The increased development 

 of this connective tissue subsequently constitutes a partition that cuts 

 off the oral ring from the coelom, and finally breaks up the canal itself 

 into an irregular areolation. 



The gastric sac at this stage is elongated horizontally into a form 



* They have been carefully studied by M. Edmund Perrier (Joe. cit.), who has 

 corrected (as I had long since privately done) the mistake into which Prof. Wyville 

 Thomson fell in regarding the corpuscles they contain as calcareous, but has not 

 thrown any further light on their function. 



t See his Memoir in the Edinb. Trans. Eoy. Soc. vol. xxiii. (1861), p. 241. 



