Dr. F. A. Bather — Studies in Edr toaster oidea. 121 



tracts of the rays, which do not stretch half-way across the inter- 

 anihulacrum. Owing to lateral compression, the curves are not very 

 clearly seen. The proximal curve is solar in tlie right anterior, 

 contrasolar in the right posterior ray, as in A. The distal bend is 

 clearly contrasolar in tlie left anterior ray, and appears to have been 

 the same in the others, except in the right posterior, where it is solar 

 but cannot be traced far. 



The rays in E 16054 are not only relatively shorter, but taper 

 more rapidly than in older individuals. Thus the left anterior 

 ray is 4"1 mm. wide in the extreme proximal region; 3"1 mm. 

 wide at about half-way to the periphery; 24 mm. at the periphery, 

 whence it rather quickly rounds off. 



Specimens E 15900 and E 16173 [E. levis) differ conspicuously 

 from the others in that the main distal curve is solar in all rays, 

 and that the right posterior ray, instead of being recurved, passes 

 along the periphery of the posterior interradius and bends round 

 like the other rays to end on tiie adapical face. In E 15900 the 

 proximal curve is solar in all rays except the left anterior (II) raj*, 

 where it is contrasolar (Plate XII). In E 16173 the proximal tract 

 is not clearly seen in ray II ; in all others the curve is solar. 



Summing up the evidence before me as to the distal curvature of 

 the rays, it appears that in all individuals that of the right posterior 

 ray, when known, is solar, that in six individuals (A, B, C, E 15930, 

 E 16054, and E 16172) that of all the other known rays is 

 contrasolar, and that in two individuals (E 15900 and E 16173) 

 it is solar. Billings (1858), pi. viii, fig. 2, also represents two rays 

 with a conti'asolar bend. Although Billings (1858, p. 83) says that 

 the rays curve "towards the right in some specimens, and towards 

 the left in others", I know of no evidence for a contrasolar bend 

 of the right posterior ray, and seeing that it is solar in all of our 

 six specimens in which the other rays are contrasolar, it is hardly 

 likely to have been contrasolar in any other case. Neither do I find 

 evidence for any individual ray or rays (the right posterior always 

 excepted) bending in a direction contrary to the majority, as does 

 the right anterior in the holotype of Agelacrinus liamiltonensis. 



In his original account (1854, p. 272) E. Billings said that he 

 had found seven specimens, and that in one only did "the rays 

 turn to the right [solar] instead of the left [contrasolar]". It is 

 therefore curious that in his "partly restored figure" lu (p. 271) 

 he should have represented all the rays as eq\iab]y solar. This 

 must have been a pure mistake, for Mr. Walter Billings, who kindly 

 looked up this point for me in April, 1900, said that all the 

 ten imperfect specimens then in the museum at Ottawa showed 

 a contrasolar flexure. The one exception mentioned by E. Billings, 

 further distinguished as showing the adapical face, could not then 

 be found. Subsequently specimen C. with the adapical face exposed, 

 turned up in tliat museum and was sent to me. In it the rays, 

 thougli contrasolar, appear solar because seen only from the under 

 face. This may have been the specimen refern d to by E. Billings, 

 iu which case none of the Ottawa specimens would have the rays 

 solar. 



