TAKEDA—ASIATIC POLYPODIUMS. 267 
other, and how these are quite naturally connected with P. 
Griffithianum and P. trifidum through P. hastatum. Yet these 
members of one natural group have been classed in several 
different sections of a certain subgenus, or even into different 
subgenera, on account of different nervations. It is impossible 
to understand why different types of nervation should not 
occur within one group of ferns. It is almost absurd to 
establish genera on a slight difference in nervation alone. The 
case above mentioned is only an instance to show how natural | 
groups of ferns have been unnaturally systematised. It is 
much to be desired that a more natural classification of ferns 
based upon sound judgment and careful observations on 
actual plants in nature should be undertaken. 
As to the distinction of species, the writer is of opinion 
that the ramenta on the rhizome generally afford a good dis- 
tinguishing character, though their colour may vary to a certain 
extent. However, it should be remembered that in some cases, 
such as in Phymatopsis, closely related species possess ramenta of 
very slight difference. The position of sori, as to whether they 
are marginal or intramarginal, etc., seems to be subject to little 
variation, and particularly in Eupleopeltis one can _ usually 
rely upon it. In the pinnate forms, the direction of pinnae, 
especially of the lowermost pair, can be commonly taken into 
consideration. The margin of the frond, whether entire, notched, 
or toothed, etc., is not to be neglected in many cases. Lastly, 
I may mention that the consistency of the frond can also 
frequently be utilised. However, one has to be very careful in 
regard to the age of the frond and also to the circumstances of 
the habitat, whether sunny or shaded; young fronds of thick- 
leaved species have repeatedly been described by herbarium- 
botanists as “‘ membranaceous.”’ 
The present investigation has been carried out in the Kew 
Herbarium, while a few specimens in the British Museum have 
also been examined. The writer wishes to take this opportunity 
of thanking the authorities of both establishments for their 
courtesy in allowing him to use the herbarium and library. 
1. P. lineare, Thunberg. 
O to its wide distribution over various parts of the Old 
World and to its polymorphic nature, this species has several 
varieties, the majority of which have been described as distinct 
species. Although the different varieties are connected with each 
other by intermediate forms, they are fairly well fixed, so that 
