WOUNDING OF FOLIAGE LEAVES 
301 
Fig. 54. lyCaves with several slashes parallel to the midrib, and extending through 
the outer end of the leaf. 
gin of blade (Pig. 54 ). — Results here are seemingly unfortunate 
as all except the shade leaves of Lilac (d) show death of parts 
of the strips. The causes of this failure are not far to seek. In 
the first place the leaves were robbed of a contact with a contin- 
uous marginal strip which has been shown in these experiments 
to possess marked efficiency for conduction. Secondly, the di- 
rection of the slashes left the major veins as barriers across the 
strips. Independent of these retardative features to the flow of 
materials, is the major difficulty due to the fact that the strip is 
attached at but one end, while fully exposed to double water loss 
both right and left. This halves the possible sources of supply 
and leaves the central ones interrupted by the larger veins. 
A comparison of parallel strips in a given leaf reveal^ the im- 
portance of favorable basal supply and also the advantage of 
wider over narrower strips. Comparison of shade leaf (d) with 
sun leaves (f, g) of the same species (Lilac) reveals the influence 
of exposure. 
Most of these leaves, and especially the more sensitive, Bass- 
wood (c), and Asclepias (h), show a marked efficiency for modi- 
fied conduction in the submarginal zone. This is in part due 
