46 The Botanical Gazette. [Februa 
rie, though I had made out these facts for myself before 
knowing what their statements were. Marie also refersto 
the two forms of structure as found in roots like those of Act@a 
alba. . 
IV. Miscellaneous considerations. 
Treub, among others, proposed to use anatomical charat 
ters to aid in the classification of plants, using root characters 
as well as those of the stem and leaf. Erickson in discuss 
this idea said that he did not think this could be accomplished, 
for his study showed him that very nearly related plani | 
showed very different histological structures, and that in many 
roots which he studied he could not tell the one from the other ~ 
From my study of the roots of the Ranunculacee 1 should say 
that anatomical characters, especially root characters, could — 
only be used in a very general way, for I think these valy_ 
more according to environment than according to specific 1 
lations, and much change of structure is seen in the roots of . 
the same species, especially between younger and older col 
ditions of it. With a few orders and genera, and perl 
species, it is possible that some anatomical character ml 
be found which would be an aid in classification, but I 
certain that among the Ranunculacee 1 could not tell th 
ference between the roots of many of the genera; and as 
species I should be entirely at a loss to characterize them 
root structure. I emphasize the fact that environment inl 
ences structure more than specific relations. For exam 
all the plants of Ranunculacee inhabiting wet places, no me 
ter of what species, presented in transection rounded col 
cells, and these loosely packed, the spaces between the 
being numerous and varying in size and number in al 
direct proportion to the aquatic or terrestrial habit of 
plant. On the other hand, plants of very dry ground § 
erally showed angular cortex cells and almost invariabl 
and small spaces between them. Another difference 
was that the roots of water Ranunculacee present a po? 
velopment of the vascular system, its office probably 
filled by the more abundant spaces between the cells, 
cially in certain roots in which many of the spaces are 
tinuous cavities surrounded by cells. These differences 
well illustrated in plate 11, figs. 1 and 3, and plate 11, 18 
Figs. 1 and 3 represent the roots of Ranunculus circinatus 
R. aquatilis var. trichophylius. Both of these grow in Wé 
asl 
cS 
f x 
ch 
