ICHTHYOCRINIDAE 269 



free as in many other crinoids, or can be distinguished from the ray divisions of which they 

 are mere direct extensions. In other words, we cannot tell from the dorsal side where the 

 calyx ends and the free arms begin. This is a character largely prevalent throughout the 

 group, but it is especially marked in Ichthyocrinus. From this interlocking of the rays, and 

 the manner in which the arms are usually found tightly folded over the tegmen, it might be 

 supposed that the spread of ventral surface for the gathering of food was not very wide. 

 It is evident however, that the arms were much longer than would be judged from the general 

 appearance of the specimens ; they become very slender, and are coiled at least two or three 

 times after disappearing from view on the outside (PI. XXXV, fig. 16). 



In its basal structure Ichthyocrinus exemplifies to the extreme the character upon which 

 the family is now defined. The infrabasals, and to a considerable extent the basals, have 

 ceased to form a part of the calyx wall, and there is a tendency in the former to atrophy ; 

 they are very. small, always completely covered by the top columnal, and in one species, prob- 

 ably two, they have entirely disappeared, leaving in their place a large opening communicat- 

 ing with the chambered organ. In considering this character Angelin's figures must be 

 altogether ignored ; his plate 22, figure 22, of piriformis, shows prominent and projecting 

 infrabasals, whereas there is nothing of the kind in the specimen (PI. XXXII, fig. ga, 

 herein). 



The axial canal entering the stem from the base is obtusely pentagonal in outline, its 

 angles radial in position ; immediately inside the calyx it enlarges and projects above the 

 floor of the cavity in a raised funnel-shaped rim the lobes of which, so far as observed, coin- 

 cide with the basals and are therefore interradial (PL XXXIII, figs, gb, c; text-fig. la). In 

 the specimen illustrated, which is the only one showing the inner floor of the base, the funnel 

 is not perfect all around owing to injury, but it probably surmounted all the basals with five 

 lobes, upon each of which a shallow channel passed down over the inner edge of a basal 

 toward the infrabasals. Comparing this with the similar funnel of Forbesiocrinus saffordi 

 shown in several figures on Plate XXX, it will be noted that there it is exteriorly trilobate, 

 coinciding with the three infrabasals from whose inner edges it rises ; but that the lobes are 

 divided by septa in such a way that the five resulting channels pass downward through the 

 infrabasals interradially. 



In Ichthyocrinus, the infrabasals being too minute to afford space for such structure, it is 

 formed upon the edges of the five basals, and is therefore five-lobed. In this shape at the 

 level of the basals it doubtless lodged the chambered organ, from which the axial cords 

 passed down toward the stem. Being interradial in conformity with the orientation of the 

 basals, they would be expected to continue in that position ; but such is not the case. The 

 channeled lobes of the funnel are bounded by very thin septa, which instead of being straight 

 curve to the right sufficiently to make the channels in the radii of the stem lumen radial in 

 position. In other words, the axial canal in passing from the basals to the stem has under- 

 gone a revolution of 36 degrees in order to become radial ; whereas in Forbesiocrinus saffordi 

 the channels are interradial where the funnel originates on the infrabasals and continue so 

 into the stem. 



When the original figure gc was drawn the exact structure of the funnel was not under- 

 stood, and the extremely thin septa were overlooked. These may now be seen in text-figure 

 10 (P- 35) °f the interior, drawn with sufficient enlargement to show how the septa twist to 

 the right as they pass downward. This evidence of an actual revolution in the position of 

 the axial canal may help to explain the perplexing variations of its orientation in the few 

 genera where it is known. The rarity of specimens in which it can be traced, except in one 

 species, renders generalization upon this character uncertain. 



The basals are never entirely exposed to view ; they are always more or less covered by 

 the column, and therefore their lower angles cannot be seen from the exterior. Hence in 



