376 L, Lesquereux on the Coal Formations of North America. 
be comprised in the three sections formerly examined: the 
Neuropteridece, the Pecopteridece and the Sphenopteridee. 
2d. That, from the scarcity of fructified specimens of fossil 
ferns in the Coal Measures, it would be supposed that most of 
the species were without fruit. If not, how can we account for 
the total destruction of the sporanges, either borne on peculiar 
stems, or attached to the lower surface of the leaves, as we fin 
them in the species of our time : 
3d. That the scarcity of large stems that have been or might 
be referred to ferns would lead us to suppose that, during the 
formation of the coal, the fern trees were of rare occurrence, at 
least when compared with the great number of ferns, which, i 
— to arborescent species, can be called herbaceous oF 
shrubby. 
These three questions must be considered separately. | 
Ist. If it is certain that characters taken from the form of the 
leaves and from their nervation are sufficient for a kind of gen- 
eral classification, applicable to the stratification of the © 
easures, it is true also that this classification fails to give us & 
clear insight into the true relation and the affinity of our f 
species. ‘T'o be exact and scientific, an analysis of the ferns must 
take into account the form and the position of the fructifications; 
and when these are absent or undiscernable, as is generally 
the case with the specimens found in the Coal Measures, we ate 
not authorized to believe that all the species, referable by their 
nervation and the form of the leaves to a commo 
esil ferns 
r- 
tive effort, the little we know of these fructifications shows 4 pe : 
greater diversity of typical and generic forms than are in : 
by the leaves and their nervation. The fruiting leaves 0 
original shape. American specimens of this species pe! of it 
agree with the beautiful figures that Mr. Goeppert has aha : 
tlet and rather resemble a raceme of 
1 bearing buds of flowers, of a dicotyl 
