112 NEW YORK STATE MUSEUM 



straight ventral walls of the thecae and straight apertures, while the others 

 have strongly convex ventral walls and introverted and introtorted apertural 

 parts — and D. sextans and smithi are again representative of a 

 phylum different from that of D . moffatensis, — it is quite apparent 

 that this evolution has taken place in several phyla at the same time. 



The tendency to a concave curvature of the proximal branches mani- 

 fests itself still in many other species of Dicellograptus, though more faintly 

 and without combination with the " diplograptid " arrangement of the 

 primary thecae. As examples could be mentioned D . i n t o r t u s , 

 divaricatus, p a t u 1 o s u s , m o r r i s i , f o r c h a m m e r i and 

 e 1 e g a n s . 



It is interesting to note just the opposite tendency in Dicranograptus. 

 After the formation of the biserial portion by the adnascence of the sub- 

 parallel proximal parts of the branches, a tendency to a convexity is devel- 

 oped in the bases of the remaining free uniserial branches. This tendency- 

 can be faintly seen even in forms with very straight branches, as D . rectus 

 [see Elles & Wood's excellent drawings, pi. 24, fig. 4], D . c e 1 t i c u s [ibid. 

 fig. 5], D. ramosus [ibid. fig. 6 and pi. 23, fig. 2 of this paper] and I), 

 nicholsoni [ibid, pi. 25, fig. 1 ], but it becomes most strongly developed 

 in those forms, where the biserial portion attains its greatest length, viz, 

 D. ramosus, D. spin if er and D. spinifer var. geniculatus. 

 There it leads in its extreme development to a geniculation. 



Both these tendencies, that to a parallelism or concavity in the proxi- 

 mal parts of the branches of Dicellograptus 1 and that to a convexity in 

 some parts of the branches in Dicranograptus, arc in my opinion, but 

 expressions of a more general tendency of the entire group to a spiral 

 growth of the branches. 



That the branches grew in compact spirals in a few species, namely, 

 Dicranograptus furcatus and I), ziczac, can be readily seen 



1 There is also a form of 1 )icranograptus with subparallel proximal branches, viz, 1 I . 

 cyathiformis Elles & Wood. Tn this the gradual adnascence is still in progress 01 

 just beginning, as the short biserial portion evinces. 



