2/2 SMITHSONIAN MISCELLANEOUS COLLECTIONS [vOL. 45 



Other species of Rhopalonaria are so different that comparison is 

 unnecessary. 



Occurrence. — Devonian of France. 



RHOPALONARIA KEOKUKENSIS new species 

 (Plate LXVI, ii) 



In this species the branches divide and throw off equal branches 

 at such frequent intervals that the pinnate arrangement of the cells 

 is generally very much obscured. The fusiform cells are narrow 

 and' vary from 0.5 mm. to 0.6 mm. from center to center. They are 

 longer than the connecting threads, the relations of the former to the 

 latter being, respectively, about as 3 is to 2. 



The proportions of the cells and stolons are very similar to those 

 found in the Devonian R. tenuis, still there is a slight difference in 

 the length of their internodes, while the pinnate arrangement of the 

 latter is much less obvious in R. keokukensis than in R. tennis. If 

 both forms had occurred in the same geological formation, we might 

 have considered them as varieties of one species, but since the differ- 

 ences noted are supported by great disparity in their ages, we cannot 

 hesitate to distinguish them specifically. In cases like this we should 

 also remember that the unknown parts of the organisms most prob- 

 ably were more sharply differentiated. 



Occurrence. — Keokuk formation, Keokuk, Iowa. 



Cat. No. 43,122, U. S. N. M. 



Family YINELLID^'E new family 

 In its simplest form, Hcteronema, the creeping base of this family, 

 consists of usually simple, though locally jointed, delicate, partially 

 ramifying, orderless tubular threads. Where pores have been ob- 

 served on these they always occurred in a single row. As a varia- 

 tion from this simplest type, we have Vinella, consisting of similar 

 delicate threads but having them arranged in such manner that they 

 proceed from more or less definite centers. In the third generic type, 

 AUonema, the orderless arrangement of the threads observed in 

 Heteronenia is maintained, but its segmentation has assumed the im- 

 portance of a constant character and instead of a single row of pores 

 the internodes are covered with them. Perhaps, on account of the 

 last feature, AUonema should have been arranged with Ascodictyon 

 rather than Vinella, the punctate internodes being comparable with 

 the similarly punctate vesicles of that genus. Possibly they are 

 homologous, even, but in either case Ascodictyon possesses charac- 



