304 Ward. — On Relations between Host and Parasite 
When, on August 4, I made the last records of the results 
of infection of the two series named, the first green leaf of 
both the infected and uninfected seedlings from each pot 
was removed, and put into tubes of absolute alcohol and 
of Fleming’s solution. Each tube received also a card- 
board slip with the serial experiment-number, name of the 
species, and date. These tubes were then put aside (the 
specimens in Fleming’s solution having been properly 
hardened, washed and transferred to absolute alcohol) to 
await examination during the autumn and winter. Mean- 
while, notes were accumulated as to the length, breadth, number 
of ribs, thickness, height of ridges and so on of the fresh 
leaves, in order that future comparisons could be made, in 
case any question arose during the examination of the pre- 
served material. 
In due course, the examination was made as follows. The 
upper part — three centimetres or so in length — of each leaf 
was taken, and examined as follows. The uppermost centi- 
metre (apex) was removed and the rest cut into short lengths 
of about three to four millimetres. One of these was mounted 
with its upper face upwards, and its next segment with its 
lower face upwards on the same slide. Various methods — 
balsam, glycerine, glycerine-jelly, warming in eau-de-javelle, 
&c., were tested, but the most satisfactory results were ob- 
tained by transferring through warmed glycerine and alcohol 
to glycerine jelly. 
These translucent preparations were then examined for 
such details as the breadth of the leaf, the number of ribs 
and their distances apart, the number and size, distribution, 
&c., of hairs, stomata, and so forth. 
The remaining segments were employed to obtain trans- 
verse sections, similarly mounted and used for measurements 
of the thickness, breadth, &c., of the leaf, the vascular bundles, 
chlorophyll-tissue, &c., as well as to check the preceding 
measurements. 
The results are set forth in the accompanying Tables. 
I may here state that no measurable differences in the thick- 
