L. F. WARD — PLANTS OF THE AMERICAN TRIAS. 



27 



fact that twenty species are common to the Richmond coal field and that of North 

 Carolina argues very strongly for the near parallelism of these deposits. The princi- 

 pal problem, then, is whether the Connecticut valley basin and the New Jersey area 

 are really of the same or nearly the same age as the coal-beds of Virginia and North 

 Carolina. The five species common to the Connecticut valley and Virginia, and 

 the six species common to the Connecticut valley and North Carolina weigh for all 

 they are worth directly upon this problem. With regard to the New Mexican beds, 

 we find that out of the thirteen species there found, only two occur also in the east. 

 These are the wide-spread forms Cheirolepis munsteri and Palissya braunii, which 

 have been found in both the northern and southern basins. 



Notwithstanding the thoroughness of this analysis, it nevei'theless leaves the 

 mind in a somewhat unsettled condition with regard to the main question as to 

 whether the data sustain the view that these different deposits are really shown by 

 the fossil plants to occupy about the same horizon or to have been laid down at 

 about the same epoch. This is chiefly due to the great difference in the extent to 

 which the different basins are represented by the fossil plants, especially to the 

 relative meagerness of the flora of the Connecticut valley and New Jersey as well 

 as that of the west as compared with the abundant flora of the Virginia and 

 North Carolina basins. The problem is, therefore, to eliminate this element of 

 obscurity and to reduce all the basins to some common basis of comparison. This 

 can only be done by the use of percentages. For example, it will be instructive 

 and will be the best that we can do to show what per centage of each florule— that 

 is, of the plants of each distinct basin — is also found in any of the other basins. 

 For this purpose we may take the gross number of species or forms that occur in 

 each basin regardless of overlapping. From this gross number we may deduct all 

 those that are confined to each basin, the remainder being common to it and some 

 other basin. Then calculating the percentage of these common forms to the total 

 number occurring in each 1 >asin, we shall not "only have a clear idea of the relation of 

 each florule to the American Trias taken together, but also of the relative homo- 

 geneity of all the florules. 



The following table will show this : 



From this table it appears thai none of the basins excepl thai of the weal con- 

 tains Less than 39 per cenl of common species, and thai one of the basins, viz, that 



