Agelacrinidae and Lepadocystinae 407 
9. Aboral Surface op Lepidodiscus 
The oral part of the specimens being at the center of the 
exposed part of the theca, that part of the theca of the various 
Agelacrinidae which rested upon some shell or other support may 
be known as the aboral surface. It is probable that this aboral 
surface varied considerably in different genera, but nothiug 
is known of it excepting in very few species. Clarke, in New 
Agelacrinites, has figured the aboral surface of two specimens of 
Lepidodiscus alleganius. From these specimens it is evident that 
'this Lepidodiscus was derived from some more or less globular or 
oval ancestor covered by imbricating scaly plates which overlapped 
in a direction from the base toward the upper or oral end. There 
is no indication of the animal having become attached to a support 
even temporarily. (Plate VI, Fig. 2B.) 
10. Mobile Marginal Part of Peripheral Ring 
In all of the Ordovician species referred to Agelacrinus or 
Lepidodiscus, however, the animal evidently was capable of attach- 
ing itself to various objects for support, although this attachment 
was not permanent, and occasional specimens are found unat- 
tached. The main means of attachment evidently was the margin 
of the peripheral ring, since the latter always is composed of small 
plates, closely adjusted to the varying curvature of the underlying 
support. These small marginal plates are merely the protective 
covering of the underlying soft fleshy margin of the individual, 
the close application of which to the underlying Rafinesquina or 
other support permitted the attachment. The slight elevation 
of the central part of the aboral surface evidently caused the 
latter to act like an ordinary sucker. 
The ready mobility of the marginal part of the peripheral ring 
is suggested not only by the close application of the latter to the 
underlying support, but also by the sagging of the specimen, on 
death and probably also, to a certain extent, during the life of the 
animal. To this sagging frequent reference has been made in the 
preceding lines. The fact that some specimens are found unat- 
tached also suggests that, when attachment again was desired, 
the animal was able again to apply this margin of the peripheral 
ring to the varying curvature of the underlying object sufficiently 
