234 PROCEEDINGS OF THE ACADEMY OF [1885. 



Erisocrinus the anal plate also is resorbed, and all five radials 

 are perfectly uniform. 



Comparing the gradual reduction of the azygous piece, from a 

 strictly radial non-arm-bearing plate to its ultimate resorption by 

 the light posterior radial, with the modifications which the lower 

 sections of the compound radials undergo among species, it 

 appears to us that the azygous piece may represent the lower 

 segment of the posterior radial. This is further suggested by 

 the genera Anomalocrinus and Heterocrinus, in which the azygous 

 piece, upon its truncate upper side, supports the right posterior 

 radial, which has the form and position of the upper section of 

 the compound radials ; while the az3^gous piece has the form of 

 their lower section. The respective plates in both cases resemble 

 each other so closely, jointly and separately, that they were all 

 described as radials. 



In the Actinocrinidse, Platj'crinidse, Rhodocrinidae, and in all 

 groups in which the general symmetry is not disturbed by the 

 presence of an azygous plate, the radials are more or less equal 

 in size, the only remarkable exceptions being the Catillocrinidse 

 and Calceocrinidae. In Catillocrinus only the two antero-lateral 

 radials are approximately alike. All the others differ widely in 

 shape and size, and while these two plates support from fourteen 

 to thirty arms each, the three others have rarehy more than one. 

 Another peculiarity of this genus is that it has no axillary plates, 

 all the arms being given off directly from the radials without the 

 assistance of brachials. Calceocrinus has but three radials, of 

 which the anterior one is composed of two parts, which, however, 

 are not alwa} r s continuous. 



Our view, that the arms fundamentally commence with the 

 plate above the first radials, whether this is free or incorporated 

 into the calyx, has been fully accepted by P. H. Carpenter, Chall. 

 Rep., p. 48, who further proves it by the developmental history of 

 the plates. The outer radials, he states, " commence as imperfect 

 rings, which soon become filled up with lengthening fasciculated 

 tissue, just as in the case with the stem joints and Inter brachials;" 

 but " the first radials, like the basals and orals, commence as ex- 

 panded cribiform films." He further agrees with us that in 

 practice, for purposes of description, it is more convenient to 

 regard the arms as commencing with the first free plate, provided 

 their real nature is tiol lost sight of. 



