500 
Journal of Agricultural Research 
Vol. XXV, No. I* 
on the dorsal side, the maturity of the leaf, and the amount of tomentum 
covering the stomata. Leaves such as would come under groups i to 3 
above are certainly most easily infected, and in the order given. Owing 
to the fact that it takes teleutosori a much longer time to mature on 
leaves of normal new canes, one is apt to be misled as to the relative 
susceptibility of the very youngest leaves. Very few teleutospores were 
ever developed on leaves which were still folded when the sowings of 
aecidiospores were made, even though these leaves were systemically 
infected, and therefore devoid of tomentum. The same line of infection 
experiments has also been carried on with blackberries and dewberries, 
with the same results. Teleutospores appear first and most abundantly 
if aecidiospores are sowed on leaves of the old canes that are systemi¬ 
cally infected. The comparatively short time (four or five weeks) required 
for the development of teleutosori on leaves already harboring the orange- 
rust stage is accounted for by the rapidity with which such leaves reach 
maturity and fall off. The ease with which they may be infected is at 
least in part due to accessory dorsal stomata and freedom from tomentum. 
Blodgett 3 reported that he had observed that blackberries bearing 
orange rust wilted sooner than uninfected plants, and that this was due 
to excessive transpiration. 
Reed and Crabill 4 think that the greater loss of water by the rusted 
plant reported by Blodgett was due to the rupture of extensive areas of 
the ventral epidermis, which would facilitate evaporation. “Possibly 
other factors connected with the diseased condition may also operate 
to cause increased transpiration.” These authors found that leaves 
of apples affected with Gymnosporangium transpire about the same 
whether they are in daylight or darkness. Healthy leaves transpire 
much more rapidly in daylight. The average for daylight and darkness, 
however, is practically the same for diseased and for healthy leaves. 
They think that the parasite must in some way affect the operation of 
the stomata. The substomatal cavities of rusted apple leaves are 
obliterated. Their figure 13 shows stomata only in the ventral epi¬ 
dermis of an area bearing spermogonia and aecidia. 
In an isolated plant infected with Gymnoconia the chances for the 
production of telia are certainly vastly increased should the leaves which 
will some day have their ventral surface covered with blister-like aecidia 
be provided with stomata on the other side. Just this occurs as, we 
might say, a matter of safety, if not always of necessity, when leaves 
are systemically infected. Aecidia frequently fail to develop in leaves 
covered with spermogonia. In such a case there is no apparent reason 
why stomata on the ventral surface should not function properly. The 
accessory dorsal stomata are formed as the result of the stimulus of the 
gametophytic mycelium present just beneath the epidermis. The 
failure to produce aecidia following spermogonia is undoubtedly due 
to the condition of the host leaf, but it is not a condition under the con¬ 
trol of the host. A great many investigations of the development and 
distribution of stomata have been made in the past, but the writer has 
been unable to find in the literature an account of another such curious 
interaction between host and parasite. There are several systemic 
rusts known which attack herbaceous plants. It would be interesting 
to know whether additional dorsal stomata are also developed by these 
hosts. 
* Blodgett, Frederick H. transpiration of rust-infested rubus. In Torreya, v. i, p. 34-35. 1901. 
4 Reed, Howard S., Crabill, C. H. the cedar rust of apples caused by gymnosporangium 
juniperi-virginianae schw. Va. Agr. Exp. Sta. Tech. Bui. 9,106p., 23 fig. 1915. Bibliography, p. 104-106. 
