BRYOZOA. 329 



most conservative of all palaeontological specialists, yet in the 

 la.st five years of his life he established more new genera than 

 during the preceding thirty years. With such illustrious pre- 

 cedents, I feel justified in presenting the following classification 

 of Palaeozoic Bryozoa, subject to the decision of future workers 

 in the field. 



In making my divisions I have sought to follow nature as 

 nearly as possible, starting with the propositions, (1) that the 

 classificatory value of a character is determined by its con- 

 stancy, (2) that in the aggregate of characters is found the true 

 test of relationship. From the first proposition it follows that 

 the constant characters, whether zoarial or zooecial are the 

 significant ones, that their relative values are according to the 

 degree of constancy, the degree being determined only by a 

 comparative study of numerous individuals, species or genera, 

 as the case may be. By experience we learn that modifications 

 of certain features are generally more significant than of 

 others. Thus among th CHEILOSTOMATA the variations of the 

 apertures, the degree in which the primitive opening is pre- 

 served or obliterated, and the modifications in the form of the 

 orifice are very important. In the CYCLOSTOMATA, however, the 

 zooecium is extremely simple and uniform, so that here the 

 :natist must depend almost entirely upon variations in the 

 mode of combination of the cells. When we come to the TREPO- 

 STOMATA. both the zocecium and zoarium supply us with impor- 

 tant characters. The same may be said of the CRYPTOSTOMATA, 

 but among these, perhaps as much as with the CYCLOSTOMATA, 

 must we take zoarial variations into consideration in making 

 generic and family divisions. In a general sense this sub-order 

 may be regarded as representing the Escharina group of the 

 CHEILOSTOMATA. but the characters which serve so well in class- 

 ifying the recent forms are almost always obliterated by subse- 

 quent deposits of calcareous material. Thin sections, of course, 



41 



