THE CLASSIFICATION OF FERNS. 
The characters by which ferns were originally brought 
into groups such as those now called genera, were 
derived from the shape and division of the frond. This 
principle of association, however, as the knowledge of 
species became extended, was found to produce vague and 
unnatural results, and generic or family characters came 
to be sought in the organs of reproduction, the shape of 
the clusters of spore-cases being taken as their most" 
obvious feature. This, too, in its turn, though affording 
better discriminative marks than the former, proved to be 
insufficient. Then it was that the presence or absence of 
an indusium, or cover to the sori, and its form and 
attachment, came to be included among the characteristic 
marks of genera ; and this combination was found to 
mark out much more satisfactory and natural-looking 
groups, than the characters which had been previously 
employed. But a still better and more discriminating 
mode was wanting, and this was at length found to exist 
in the peculiarities of structure of the* venation or 
vascular system of the frond, and in the connection of 
these veins with the sori. Among the earliest proposers 
of these features as characteristic of generic differences, 
the names of Robert Brown in connection with living 
ferns, and Adolphe Brogniart with fossil remains, stand 
prominent; and the subsequent labours of various 
botanists, especially of Presl in Germany, and J. Smith 
in this country, have led to its very wide adoption. In- 
deed, it is made the basis of most modern systems of 
classifying ferns; and, taken in conjunction with the 
peculiarities of the fructification, renders little further 
change necessary or to be desired. 
