2 14 NEW YORK STATE MUSEUM 



to their possible composition of several tubes. All one can make out in 

 this regard is a depressed central line in the thicker branches which as a 

 rule, ends in the circular pits seen just below the bases of the secondary 

 branches, giving one the impression of a closer structural connection 

 between the central line and the pits and branches [see pi. 12, fig. 1]. 



In M . simplex [see pi. 12, fig. 5] these relations are a little clearer 

 and it can be observed there that each component tube extends from the 

 base of one branch to the second following, passing then out into the 

 branch. It is hence possible that the thin, smooth filiform branches were 

 not as simple in structure as they appear externally. Moreover, the type 

 of M . arundinaceus (Hall) shows indications [see pi. 12, fig. 6 at b] 

 that the pits were not arranged on one side (the " celluliferous side") only 

 but were alternating on both sides. This is also suggested by the relation 

 of the pits to the branches in M . t e n u i r a m osus and M . si m p lex; 

 on some branches the pits are found closely below the bases of the branches 

 [see pi. 12, fig. 5], in others [pi. 12, fig. 4] in the middle of the internodes. 

 This observation, if correct, may also indicate some complication in the 

 structure of the branches. The supposition that the branches of M . 

 tenui ram osus were of composite thecal structure is, of course, antago- 

 nistic to the hypothesis that the appendages were the thecae of nourishing 

 individuals, for assuming the latter case we can not well escape the further 

 conclusion that the branches, and the whole rhabdosome for that matter, as 

 the bearers of the thecae represented nothing but a large continuous 

 " funicle " corresponding to the hydrocaulus of the hydrozoans and Masti- 

 gograptus acquires then a structure and position entirely apart from the 

 other Dendroidea. 



The extremely slender and thin form of the branches which is so 

 little suggestive of their composition of thecae requires the assumption of 

 exceptionally thin and long thecal tubes. But this can not be considered as 

 fatal to the conclusion of the composite character of the branches, since we 

 have here shown that another genus, Thamnograptus, with similar thin 

 whiplike branches is distinctly composed of long, tubular thecae. 



