2i8 Arber . — On the Past History of the Ferns . 
frond of Camptopteris spiralis , Nath., for instance, exhibits a combination of 
the sympodial with the spiral type of branching in a very remarkable 
manner,, a character which is also found, but to a much less extent, in 
certain leaves of the genus Dictyophyllum . Such features are quite unknown 
among the Palaeozoic Fern-like plants, and although among Mesozoic Ferns 
the frond is by no means always pedate, the frequent occurrence of this 
habit is significant. 
Similarly the Palaeozoic Fern-like plants exhibit certain features which 
are more especially their own. Stress has been laid in more than one 
quarter on the occurrence of the repeatedly dichotomous type of branching 1 
found in many Palaeozoic fronds, and especially on the fact that stipular or 
adventitious outgrowths 2 , known as Aphlebiae, occur on the rachis in many 
cases. I am inclined to agree with a suggestion which was made recently 
by Professor Oliver that these features may probably be regarded as indicative 
of the fronds of members of the Pteridospermae rather than of true Ferns. 
If, however, we turn to the subject of the fertile fronds of Mesozoic 
and Palaeozoic age, we are at once struck by the fact that, while such 
leaves, bearing annulate sporangia often but not always on fronds of 
unreduced lamina, are almost common in the Mesozoic sediments, such are 
comparatively rarely found in the older rocks. For instance, to take an 
example from our best-known British Mesozoic flora, Mr. Seward 3 has 
described from the Lower Oolite of Yorkshire twenty species, regarded as 
probably of the nature of Ferns, and in thirteen cases the sporangia are 
known, which in eight or ten species are borne on fronds practically identical 
with the sterile foliage, and in three to five others on reduced fronds. 
Among the Palaeozoic Fern-like plants, the number of species, on the fronds 
of which annulate sporangia have been discovered, is, by comparison with 
the total number known, almost infinitely small. 
Further it is clear from a study of the fertile fronds of Mesozoic age 
that, even by the Jurassic period, most of the modern families of ferns had 
become differentiated. Mr. Seward 4 has recognized members of the 
Schizaeaceae, Osmundaceae, Cyatheaceae, Polypodiaceae, as well as of the 
Matonineae and Dipteridinae, in the rocks of -that period. 
It would be easy to discuss the Mesozoic Ferns at greater length, but 
enough has perhaps been said to show that, so far back as the Rhaetic 
period, Ferns belonging to modern families of Leptosporangiatae were not 
only in existence, but flourishing in sufficient degree to form one of the 
dominant features in the vegetation of that period. 
The Primofilices. 
When, however, we try to trace the life-line of the Leptosporangiatae 
back to Palaeozoic times, we are at once in difficulties. A race which rose 
1 Potonie (’95). 2 Potonie (’03). 
3 Seward (’00). 4 Seward (’04), p. 843. 
