230 BENEDICT: NEW VARIETIES OF NEPHROLEPIS 
relatively high coefficient of variability, as for example, Teddy Jr. 
They may be sought by cytological study of similar variations, as is 
shown by the production of dwarfs in many widely separated lines, 
or by the reversions. The cytological method would only show 
some possibly correlated facts of cell structure. The experimental 
method might show some connection between cultural methods 
and the appearance of new forms. In this connection it may be 
noted that Eimer and Coulter, perhaps the earliest and the 
latest to discuss the possible causes of orthogenetic variation, 
have both professed the opinion that climate is the probable 
causative agency. 
The query has been expressed as to whether a hybrid origin 
might not account for these variations of bostoniensis. With this 
possibility in mind, one species of Nephrolepis, N. biserrata, which 
is co-extensive with N. exaltata in distribution, would seem to be 
worthy of investigation. It is hard to imagine, however, how the 
combination of any two twice-pinnate species of Nephrolepis, or 
indeed of any more dissimilar species, could account for the ap- 
pearance not onlyof the three linesof variation but of the progressive 
intensification of these lines thereafter. The sterility of these forms 
is apparently the only feature these forms have in common with 
most hybrids, and this must presumably be accounted for in some 
other way. 
For the present, all that can be said as to the genesis of these 
forms is they apparently proceed from some internal cause. It 
may be added that similar variations have occurred as a result of 
spore sporting, and also that similar variations are common wild. 
These forms may therefore be considered as indicating how species, 
showing similar differences, have arisen, that is, by what lines of 
variation. The evidence offered along this line by the progressive 
forms alone is most incomplete, however, without an understand- 
ing of the regressive variations as well. 
D. SUMMARY OF FACTS AND CONCLUSIONS 
1. Nephrolepis, a genus of once-pinnate tropical ferns, includes 
several species which are cultivated. Nearly all of these culti- 
vated species show variation by bud sports, but JV. exaltata 
through its variety bostoniensis is most prolific, having given rise 
