130 FOSSIL CRINOIDS. 



except the anterior, which uniformly divides much higher up, having 9+, 13, 

 15, 15, and 19 IBr respectively in five specimens; the halves then bifurcate 

 repeatedly at intervals of from 5 to 20 brachials on a fairly regular plan, with 

 minor variations; the outer branches in each dichotom continue in a definite 

 direction, while the others branch from them towards the inner side to the 

 number of four or five bifurcations ; the inner ones mostly branch once or twice, 

 so far as preserved, and probably more. This gives from 20 to 25 ultimate divi- 

 sions to the ray, or 100 to 125 in all, with the arms still strong and but little 

 tapering. It is probable that five bifurcations was the usual limit, as I have a 

 specimen, not figured, with one ray five inches long extending far beyond the 

 fifth and no sign of further branching. There is far more regularity in the arm 

 distribution than in the type species. Radial facets large, occupying the 

 greater part of the distal face of the plates. The calyx plates are rather thin, 

 and are broken and displaced in all the specimens, so their exact form and 

 position cannot be stated, and in all the base is injured beyond recognition. 

 Loose columnals are found, however, in the same beds, showing that the axial 

 canal is quadripartite (PI. I, fig. 3). Anus small, with no plate between the 

 radials above it; its position is shown in Fig. 2, of Plate I. 



This species comes from strata called the Upper Helderberg ( = Corniferous 

 = Onondaga) by the collectors at the type locality, the Falls of the Ohio, at 

 Louisville, Kentucky, where it is found only in the Nudeocrinus bed below the 

 Hydraulic bed. The type specimen was labeled by Dr. Knapp as from the 

 Hamilton, and so stated in the original description; but this is now known to be 

 erroneous, the exact horizon of all the specimens being definitely fixed as above 

 stated. The type and principal specimens used herein are from the above 

 locality, but some specimens were found in the equivalent rocks in Livingston 

 county, New York, which apparently belong to it. I give figures of two of them 

 (PI. I, fig. 4; PI. II, fig. 2), which show the characteristic arm structure, so 

 different from that of A. bulbosus, with which they were associated. Fig. 4 

 shows the anterior ray with about 19 brachials below the bifurcation. 



Types. The magnificent specimen figured on Plate I is the type used for the 

 original description, then in the collection of Dr. Knapp, of Louisville, and given 

 to me some years ago together with the type of A. knappi, by the late Professor 

 W. W. Borden, who acquired the collection. This specimen seems to have six 

 rays, the extra one being at the right posterior; on account of the broken condi- 

 tion of the caljrx we cannot see exactly how they start, but in the space between 

 the right anterior ray and the posterior interradius there must have been two 



