174 Rhodora [SEPTEMBER 
When separate, they are more or less crescentic, with the convex side 
facing towards the middle of the stipe. The ad-axial tips of the 
bundles are never inflexed. When the two bundles are united, the 
compound bundle is usually X, Y, or T-shaped. Very rarely, as in 
Asplenium fragrans Sw. (A. mexicanum Mart. & Gal.) and A. biparti- 
tum Bory (A. auritum Watt), the compound is V-shaped, but even 
then it is not entirely like that of Athyrium. Thus we find that in 
Asplenium fragrans, the compound bundle is formed, not by the union 
of the true ab-axial extremities of the component bundles, but by the 
junction of the points of narrowly V-shaped bundles (I and J, fig. 2). 
There are certain minor points in which Asplenium nearly always 
differs from Athyrium and Diplazium. In the first mentioned genus, 
the walls of the sclerenchymatous cells of the stipe have a distinctly 
red color,— it is this which gives the characteristic red-black hue to 
the stipes of many species of Asplenium. In Athyrium and Dilpazium 
on the other hand, the sclerenchymatous cells of the stipe have yellow 
or yellow-brown walls. A similar difference obtains in respect to the 
_ thickened walls of the cells of the scales. 
In Aspleniwm the veins of the ultimate segments tend to be 
repeatedly dichotomous, while in Athyrium and Diplazium the veins 
are pinnate, and the veinlets are either simple or once forked. Very 
rarely indeed in these genera is a vein forked twice. 
The distinctions between Athyrium and the genera Diplazium and 
Dryopteris are based entirely on the character of the sorus. As has 
already been stated, this organ in Athyrium is very variable and in 
many species two, or even three distinct types of sorus occur, even on 
the same frond. The simplest of these types from the descriptive 
stand-point, though almost certainly not the most primitive, is that 
which is characteristic of the genus Asplenium (see fig. 3). Here 
the sorus extends for a greater or less distance along the anterior side 
of a vein,' and is covered by an indusium, which grows out from the 

1 When the subtending vein is forked ith ise b hed, th y sorus occurs on the 
anterior side of the anterior branch of vein. Second ndary 1 sori mae 6 occur on the pretation 
side of the anterior br: anch, oo branch, and in 
positions on the other branches if there are any. It is to es noted that the secondary « sori are 
always on the anterior side of the veinlets, if we consider th 
veinlets, viewed as an ultimate segment of the frond, gato than in respect to the seg- 
ment of the next lower order, which controls the position of the prim . As has been 
pointed out by E. J. Winslow (Double Sori in Athyrium, Am. Hae ha Journ. iii, eth baa dep 
such points of confused and indeterminate orientation that diplazioid and 
cially tend to develop. 


