236 STUDIES IN EVOLUTION 



in the two valves of any genus yet discussed. In some 

 allied genera, as Discina (type D. striata) and Schizotreta, 

 where the pedicle is small and the lower valve rises above 

 the object of support, a similar form in both valves is again 

 produced by the conical growth of the lower valve. 



More primitive types, as Acrotreta and Acrothele, having 

 the plane of the dorsal valve at right angles to the direction 

 of the pedicle, retain a marginal upper beak, while the lower 

 is elevated, sub-central, and perforate. These features in 

 Acrotreta and Discina resemble, in a measure, those in the 

 Rudistes. In Acrotreta, as in Caprotina, the upper valve 

 shows its normal affinities, while the other has become 

 highly modified and dissimilar. But in Discina and Hippu- 

 rites the hinge-line is lost, and the apex of the upper valve 

 is sub-central. This conical habit of growth in erect attached 

 organisms has been explained as the physiological reaction 

 from equal radial exposure to the environment. It consti- 

 tutes the law of radial symmetry, ably discussed by Haeckel, 

 Jackson, Korshelt, and Heider. Its application to the 

 Brachiopoda can be made mainly in forms having the pedicle 

 perforation sub-centrally located in the lower A'alve. 



In Thecidium and Crania the calcareous union of the lower 

 valve to the object of support represents the extreme of 

 unlike conditioning, and such forms exhibit the greatest 

 difference in the features of the opposite valves. Crania, 

 being probably derived from discinoid stock, is without proper 

 hinge. In the history of its development, so far as known, 

 it does not show beyond the protegulum an early hinged con- 

 dition. Hence there is no indication of direct derivation 

 from hinged forms. A false hinge is sometimes present, but 

 it clearly shows a secondary mechanical adaptation, and not 

 a phylogenetic character. On the other hand, true hinged 

 attached genera, such as Thecidium {Lacazella'), Davidsonia, 

 and Strophalosia, possess this feature as a later ancestral 

 character, and in their chronological history tend to shorten 

 and gradually eliminate it. An illustration of this is seen 

 in the succession of the species in Strophalosia, or in the 



