~ 
New YorK AGRICULTURAL EXPERIMENT STATION. 65 
jected to a freezing temperature. The rupturing of_the epider- 
mis and the death of the cells in the brown spots he attributes 
to the direct action of the frost. 
Our theory, which differs somewhat from that of Sorauer, is 
as follows: The leaves were frozen when only partially un- 
folded. The formation of ice crystals between the lower epi- 
dermis and the mesophyll caused these tissues to separate, and 
upon thawing they failed to reunite. Being removed from its 
source of nourishment, the epidermis then ceased growing, 
while the parenchyma cells continued to expand, and not having 
opportunity to spread out laterally, took the form of an arch, 
thus bringing about the wrinkles. In some cases the tension 
produced by the expansion of the growing parenchyma became 
so great that the epidermis was ruptured, and the parenchyma 
cells, being then exposed to the air, died from excessive tran- 
spiration. It is probable, also, that the whipping of the leaves 
in the wind caused the breaking of some of the blisters. 
It was always the lower epidermis and never the upper one 
which was separated from the parenchyma. The explanation 
of this is probably to be found in the fact that the tissue in 
the lower half of the leaf is looser than that in the upper half, 
and consequently the lower epidermis is less firmly attached 
than is the upper one. 
In order to determine definitely whether the rupturing of the 
epidermis may occur as above stated there were selected for 
observation thirteen blistered leaves which showed no rupturing 
of the epidermis. These leaves were selected and marked on 
June 13th. By June 24th four of the leaves showed one rup- 
ture each and one leaf showed two large ruptures. After this 
there was no further change to the end of the season. Al- 
though but few of the blisters broke, enough of them did so to 
prove conclusively that a period of at least one month may 
elapse between the time of the freezing and the rupture of the 
loosened epidermis. It is altogether likely that the breaking 
of the blisters was more common earlier in the season while the 
leaves were expanding more rapidly. The leaves marked for 
observation were nearly full grown at the time they came under 
- observation. 
5 
