76 BOTANICAL GAZETTE. 
AO Mosca AE, a 
was unchanged. Both reagents when applied to cotton fibers 
quickly destroyed them. : 
igs. 1 and 2, Plate I, represent transverse sections, and fig. 
the cork cells is remarkably abrupt (fig 1). On the peripheral 
side of the cork there is no such abruptness, the cork cells, ¢, ¢, 
¢, fig. 2, shading imperceptibly into the wood cells, a, 2. ‘ 
n all the sections the cork cells are seen to be quite irregu- 
lar, much more than this tissue normally is. Some sections from 
other parts of the stem showed a more regular tissue ; in several 
‘he wood cells last formed in autumn are quite different from 
the earlier ones. In fig. 1 the last .12™ of wood cells (from 4 
to 6) are thicker walled and have less angular cavities than the 
older portion. In fig. 3 the difference is still more marked, the 
00 ato 6 consisting of greatly elongated fibers with 
whereas the majority ot the woo 
et of shorter, smooth-walled fibers with oblique ends (@, 
ag 
Tam unable to present any satisfactory explanation of the 
formation of cork in the positions described. Numerous theories 
have suggested themselves, but all are open to too grave objec- 
tions to be worth presenting | 
EXPLANATION OF PLATE.—Figs. 1 and 2. Transverse sections of a portion of 
arder, 
u ‘ 
itted ni Ss; G central side of the section; m, cell of medullary ray, with 
pitted walls; P large pitted vessel ; P, peripheral side of the section ; x, smooth 
walled wood bers ; z', younger wood fibers with irregularly thickened walls. 
All the figures were drawn with a camera, under Beck’s } objective and A ‘ 
eyepiece. es 
ose; ligule very short, truncate: panicle 1} to 3 inches long 
erect, rather lax, its base sheathed by the upper leaf; branches of the panicle 
