1922] 
and the KENNEDY (2) modifica- 
tion of the SWEENEY method 
was used for crude fiber. 
Results 
The data obtained in these 
analyses are shown graphically in 
figs.1 and 2. All values are cal- 
culated to a moisture free basis. 
In fig. 1 the sound and rotted por- 
tions of each sample are placed 
side by side, so as to show the 
change in composition during 
rotting. It will be seen that in 
almost all cases the rotted portion 
shows a higher content of all con- 
stituents except crude fiber; the 
latter shows no general tendency 
in either direction, sometimes 
being greater and sometimes less 
in the rotted than in the sound 
material. The increase in these 
constituents is probably due to 
loss of dry matter through res- 
piration. Previous work has 
shown that respiration is higher 
in infected than in sound plums, 
and this results in a relative in- 
crease in the substances deter- 
mined. The figures for ether 
extract and for crude fiber are 
erratic in their changes, which is 
in keeping with the empirical 
nature of the methods of analysis. 
In fig. 2 the same data are 
used, but they are arranged so as 
to give more direct comparison of 
varietal characteristics. The first 
WILLAMAN & DAVISON—PLANT DISEASES 
107 
ASH 
%o 
so} 
40 
Tl | 
oe NITROGEN 
20 : 
7 i 
ef 
ull 
“ov O 
go} : 
es 
ie 
2 | i 
, ts 
La] 
ss 
.; 
be ETHER EXTRACT 
6.0} 
40} 
H 
20} ae pes 
22 5s 
aes 
ee 
RUDE FIBER 
265 205 
/50} 
HOr 
; 
70} .s 
tt 
LEE 
30 c dabig WOK) IW CS 
yy. IT He 
fstashee of sound and of 
rotted pls grouped to bring out varietal 
: solid line, sound sa 
dotted ‘line, rotted samples; first two vari- 
rot; second two are susceptible; s 
for stages of growth and names ie varias. 
