THE AMERICAN NATURALIST [Vol. XLVII 



more or less ignored. Thus, to take two instances from 

 the fossil crinoids, in the genus Marsupites the only fea- 

 ture which can possibly give a clew to its true affinities is 

 the arm structure, which is that of an ordinary comatu- 

 lid ; and in the allied genus Uintacrinus the arm and pin- 

 nule structure alone are found to be reliable. 



Eldonia— In Eldonia there are only two structures 

 upon which we can hope to base our deductions concern- 

 ing its systematic position: (1) the bell-like general 

 shape, and (2) the coiled digestive tube with two tentacle 

 clusters at the anterior end. 



1. The bell-like shape suggests the ccelenterates, and 

 such forms as Trochosphcera or trochophore larvae. 



The highly specialized digestive tube at once negatives 

 the supposition that Eldonia may be a ccelenterate. 



Trochosphcera has a general form and an internal 

 structure which is certainly suggestive of Eldonia; but 

 there are many reasons why it is not possible to connect 

 the two. In the first place there is the question of animal 

 mechanics ; the size of the members of each group of ani- 

 mals is limited by physical and mechanical considerations 

 due to the requirements of fundamental structure, etc. 

 Thus we do not find butterflies as large as ordinary birds, 

 nor cetaceans so small as the average fish ; their structure 

 is not adaptable to the limitations imposed by such sizes. 

 Trochosphcera is surrounded by a band of cilia, just below 

 which is the mouth, and below that another band of cilia. 

 Ciliated bands do not transform into broad body fringes 

 such as we see in Eldonia; they are more or less uncertain 

 structures, and are present as ciliated bands, or are absent 

 altogether. Trochosphcera has a powerful retractor mus- 

 cle attached to the posterior portion of the alimentary 

 canal ; powerful retractor muscles are a feature of all the 

 rotifers; there is no trace of any retractor muscle in 

 Eldonia. 



In Trochosphcera, and in the so-called trochophore 

 larvae, the anus opens at the pole determined by the cili- 

 ated band as the equator, while the mouth is just below 



