ii4 



NEW YORK STATE MUSEUM 



strengthen this point by the " diplograptid " mode of growth, the coales- 

 cence of the proximal parts of the branches and the " web," for it can not 

 be doubted that the upward growth of the primary thecae in the more 

 advanced Dicellograpti and in all Dicranograpti and the coalescence of the 

 branches tend all to the same end, namely that of producing a long and 

 firm connection at the most exposed extremity of the two branches. This 

 tendency has, in my opinion, been the cause of the development of the 

 peculiar Dicranograptus structure, with its biserial proximal portion, from 

 Dicellograptus. 



Freeh has sought the explanation of the structure of Dicranograptus 

 in the assumption that the biserial portion served like the "fins" of the 

 Diplograpti and Monograpti to permit a steering in vertical direction and 

 suggested that the two branches were united by a membrane which was not 

 preservable in fossil state. While this view has the advantage of being 

 part of a general proposition, viz, that the expansions of the sicular ends of 

 all Axonolipa served as rudders, it seems in this special case not so well 

 supported or even contradicted by the observations stated before, for the 

 inference of the growth of the branches in elastic spirals is difficult of recon- 

 ciliation with that of a terminal rudder which requires a rigid rudder pole. 

 Nor is there any tendency to a broadening of the biserial section noticeable 

 in any Dicranograpti, while the origin of the biserial portion finds a plausible 

 explanation as a corollary of the spiral growth of the branches. 1 



The further assumption of the presence of an extensive membrane 

 stretched between the branches for purposes of steering, would also be 

 invalidated by the inference of the spiral growth of the branches. 



The nature of the antisicular or inner end of the branches is still a 

 mystery. Freeh, who first noticed that the branches of large specimens of 

 Dicranograptidae which at first diverge, converge again later on, has based 

 a reconstruction of Dicranograptus and Dicellograptus on this observation, 

 and introduced a semidiagrammatic figure [1897, p.619] in which the 



1 If one, with Freeh derives Dicellograptus through Dicranograptus from ("limaco- 

 graptus, the biserial portion is anyway but a disappearing and transitional structure. 



