Arber . — On the Past History of the Ferns. 217 
The Leptosporangiatae. 
When we seek to recognize the dominant phase in the life-history of 
the Leptosporangiatae, we have fortunately little difficulty. The evidence 
that, by Rhaetic times, this group had already attained to the status of 
a dominant factor in the flora is very clear. This position was also 
maintained in the Jurassic and Lower Cretaceous vegetations. 
Of the Tertiary floras our knowledge unfortunately is very vague at 
present, but the fact that this line of descent is still one of the characteristic 
features of the vegetation of the world at the present day is proof that the 
broad phase of the life-line has continued more or less constant through 
a large number of geological periods. In other words, of the more ancient 
Mesozoic types, the Leptosporangiate Ferns, like the Coniferales, have on 
the whole been remarkably successful in the struggle for existence. Even 
at the present day they still maintain their position as an important 
element in the vegetation of the world side by side with other races, which 
attained to a position of dominance at a much later geological epoch. 
In contrasting the Fern-like plants of the Palaeozoic and Mesozoic 
periods, one finds that, so far as their external morphology is concerned, 
those of the Mesozoic rocks seem to recall the recent Ferns even more 
closely than one might have expected. It may be admitted that, as regards 
the general configuration of the fronds, and certain other characters such as 
the circinate vernation, both the Palaeozoic and Mesozoic representatives 
agree in being exceedingly Fern-like. But such features are for the most 
part of little value phylogenetically. Young fronds coiled in a crozier-like 
manner, often referred to the frond-genus Spiropteris, Schimper 1 , where 
their real generic nature cannot be ascertained, are known from the Rhaetic 
as well as the Carboniferous rocks. In the latter case some have proved 
to belong to Neuropteris 2 , others to Pecopteris 3 . Among these, the 
Neuropterids at least were not Ferns, but Pteridosperms. Thus this type 
of vernation was common to more than one group in past times, just as 
to-day it is common to both Cycads and Ferns. 
There are, however, some features of the Mesozoic Ferns which are 
sharply contrasted with those of the Fern-like plants of the Palaeozoic. In 
the earlier Mesozoic Floras, the pedate habit was undoubtedly a charac- 
teristic of the fronds of a number of true Ferns. Some of these families, 
the Matonineae and Dipteridinae, have still representatives living to-day, as 
Mr. Seward 4 has shown. The type of sympodial branching exhibited by 
such a Fern-frond as that of the recent Matonia or the Jurassic Matonidium 
was not, however, the high-water mark in this direction. In Rhaetic times, 
as Nathorst 5 has pointed out, we have still more complicated fronds. The 
1 Schimper (’69), vol. i, p. 688, pi. 49, fig. 4. 2 Bunbury (’58), p. 243, text-figure. 
3 Kidston (’01), pi. 26, fig. r ; Potonie (’03), PI. VIII. 
4 Seward (’99), Seward and Dale (’01). 5 Nathorst (’92), p. 169. 
Q ^ 
