BRYOZOA. 331 



distinguish those of one genus from those of another. (Com- 

 pare sections of fenestrate genera figured on pi. 54 and 56). 

 Taking this uniformity into consideration it seems evident that 

 the Silurian species which 1, have united under the genus Pbyl- 

 loporina, are sufficiently distinguished from the FEXESTELLID.E 

 to rank as a separate family. Indeed, the tubular zocecia, which 

 are occasionally intersected by diaphragms, suggest a decided 

 affinity with the TREPOSTOMATA and CYCLOSTOMATA. Granting 

 this relationship, we must not lose sight of the fact that a 

 tubular primitive cell is also characteristic of many of the 

 ARTHROSTYLOXE, and all of the RHABDOMESONTID.E as well as 

 occasionally in other families of the sub-order (e. g. Stictoporella, 

 and Phyllodictya ) . It also appears highly probable that in 

 Fhylloporina we see the ancestors of the more highly developed 

 FEXESTELLID.E. A comparative study of that family brings to 

 light some important facts. Thus the earliest of the species 

 really belonging to the restricted genus Fenestella, differ from 

 the typical Devonian and Carboniferous species in having the 

 thin walled primitive cell more inclined to be tubular; or per- 

 haps it would be better to say, that the anterior and posterior 

 walls are less nearly vertical than we find them in later repre- 

 sentatives of the genus. On plate 54, fig. 2, represents a deep 

 tangential section of a branch of F. bellistrmTa (?) Hall, from 

 the Niagara shales at Waldon, Ind. This should be compared 

 with 16 on same plate, representing a similar section of Pbyllo- 

 porina ? dawsoni, from the Trenton group at Montreal, Canada,* 

 and it will be seen that they have one peculiarity in common 

 which is not possessed by the Devonian and Carboniferous 

 species represented by figures 3, 4, 5. Namely, the partitions 



PHYIXOPOBIXA DAWSONT, n. sp. Branches lax, very slender, 0.3 to OJ> num. wide, 

 irregularly inosculating, often united by non-poriferous dissepiments. Fenestrules 

 acutely elliptical, usually wider than the branches and from 1.5 to 3.0 m.m. long. Reverse 

 convex, smooth or faintly striate. Poriferous side with a sharply angular! mesial ridge 

 and one range of cells on each side. Zocecia sub-tubular, with rather prominent sub- 

 circular apertures. 0.18 num. in diameter, five or six in 2 m.m. Tangential sections show 

 a number of minute accessory pores in the interspaces between the apertures. 



The reverse of this species very much resembles that of P. reticulato. Hall, from 

 the same foi mation, but sections of the two forms cannot be confounded. Hall's species 

 has three ranges of zooecia and generally has smaller fenestrules. 



Quite abundant in the Trenton strata at Montreal, Canada, and Chimney Point, Ver- 

 mont Specimens were received from Sir Win. Dawson and Prof. H. M. Seely. 



