21 6 
Arber. — On the Past History of the Ferns. 
manner, and try first to recognize the particular geological period at which 
one or other of the main divisions of the Ferns attained to its maximum 
diversity and differentiation; and then to trace it backwards and forwards 
in geological time, so far as our present knowledge will permit. As the 
result of comparatively recent research it is now possible to adopt this 
mode of attack in respect to the more important elements of those floras 
with which we are best acquainted, i. e. those of the Palaeozoic (Devonian, 
Lower Carboniferous, Upper Carboniferous, and Permian) and the typical 
Mesozoic floras of the Rhaetic, Jurassic, and Wealden. If, during any 
one or more of these periods, one line of descent was particularly suc- 
cessful in the struggle for existence, there is ample evidence of the fact to 
be found in the rocks of that age in some geographic region or other. 
The imperfection of the Record, largely exaggerated in the past, can be 
wholly neglected where we are considering the larger divisions of the 
Vegetable Kingdom, such as phyla, classes, or groups of plants. For 
instance, the Cycadean line of descent was obviously at its maximum in 
Mesozoic times. This is shown by the abundance of Cycadean remains in 
the rocks of the Rhaetic, Jurassic, and Wealden, in many parts of the world, 
as compared with the rarity of such fossils in the Palaeozoic and Tertiary 
(including Recent) floras (see text-figure i, p. 229). 
A study of the broader botanical features of the Palaeozoic and 
Mesozoic vegetations shows that in the case of a phylum or group which 
has, at one time or another, been successful, we can often recognize 
three phases in the line of descent. One such represents the period of the 
maximum diversity and differentiation, when the race had attained to 
the status of being one of the dominant factors in the vegetation. The 
other two, on the contrary, correspond to periods when it was a subsidiary 
type, and are the stages of adolescence and senility. We can represent 
these three phases diagrammatically if, instead of indicating by a thin line 
the record of a phylum or class of plants extending through a number of 
geological periods, we use a more or less lenticular figure (see text-figure 1, 
p. 229) ; the breadth of the figure being proportionate to the degree of 
differentiation of the group. The broad phase of the life-line indicates 
the periods at which the race was a dominant factor in the then existing 
vegetation, the thinner lines, those when it was subsidiary to other groups. 
No doubt the exact form of the life-figure varies in the case of different 
races ; as will be seen from a comparison of that of the Cycadophyta and 
the Leptosporangiatae indicated in the text-figure on p. 229. 
I propose to attempt to recognize here the various phases in the life- 
lines of the Modern Ferns. In discussing this problem I have found it 
helpful, if indeed it is not absolutely essential, to consider separately the 
lines of descent of the Leptosporangiatae, Eusporangiatae, and Hydro- 
pterideae. 
