498 PROCEEDINGS OF TEE NATIONAL MUSEUM. vol. 39. 



be better described as circuli than as genera. A circulus was one of 

 the small groups of individuals who clustered round speakers in 

 the Roman forum. Most of the individuals in the forum were defi- 

 nitely attached to a particular group; the groups were less crowded 

 around their margins, and between them people were irregularly 

 scattered and crossed from circulus to circulus. They thus pre- 

 vented any rigid division of the crowd into definite groups." 



Wliile Gregory's view of genera among the simple Cyclostomata is 

 a very ingenious one, still it seems to me that no distinction need be 

 made between the conception of a genus of Bryozoa and that of 

 any other group of organisms. It must always be recognized that 

 sharply outlined genera are impossible in any class, and that generic 

 names are instituted for convenience in designating a series of organ- 

 isms having certain characters in common, as well as to show natural 

 affinity. The middle forms of such a series will have the most dis- 

 tinctive characters, while those at either end often show relation- 

 ships to allied groups of species. 



The simplest of the tubuliporoid bryozoans comprise the forms in 

 which the zocecia are adnate and arranged uniserially. To-day 

 most writers restrict the genus Stomatojjora Bronn to such unilinear 

 species. More complex, adnate species have the zooecia arranged 

 in two or more rows and are generally referred to the genus Prohos- 

 cina Audouin, while zoaria of incrusting circular or irregular patches, 

 resulting from the union of numerous contiguous zooecia, belong to 

 Lamouroux's genus Berenicea. The present article deals only with 

 simple, unilinear forms, which, as stated above, have hitherto been 

 classed as Stomatopora. 



My experience with Stomatopora began some years ago with the 

 two very abundant Ordovician forms, S. arachnoidea SindSAnflataHall. 

 These two species are so different that I could never reconcile their 

 recognition as species of the same genus; but it is only recently that 

 sufficient material has been accumulated in the collections of the 

 U. S. National Museum to warrant a close study of the subject. 

 Now all of the known Paleozoic and most of the Mesozoic and later 

 forms of the genus are represented by specimens, and of many of 

 them the material is abundant enough to fairly test their hitherto com- 

 monly supposed great instability in specific characters. The present 

 paper is therefore based upon a study of thousands of specimens from 

 widely separated localities and from many horizons in the geologic 

 column. 



The two Ordovician species mentioned above being the most 

 numerously represented, received a correspondingly greater amount 

 of study. S. arachnoidea is essentially like the Jurassic S. dichotoma, 

 the genotype of Stomatopora, but the zocecia of S. injiata are so dif- 

 ferent and its peculiarities are so constantly reproduced in a long 

 line of descendants, that it seems desirable to establish a new genus 



