52 BULLETIN 934, U. S. DEPARTMENT OF AGRICULTURE. 
received equal quantities of .seed. Three strongly parasitic strains 
of Pythium were used, and a total of 12 units of jack pine and an 
equal number of western yellow pine was inoculated with 12 inter- 
spersed units of each species as controls. The mean results are as 
follows : 
Pinus banksiana. — Inoculated plats : Emerged, 64.2±4.9 ; died during the next 
17 days. 25 per cent. Control plats: Emerged, 85.5±3.6 ; died during 
the next 17 days, 13 per cent. 
Pinus ponderosa. — Inoculated plats : Emerged, 34.6±1.8 ; died during the next 
9 days, 39 per cent. Control plats : Emerged, 45.4±1.3 ; died during the 
next 9 days, 25 per cent. 
The difference in emergence apparently due to the inoculation is 
for the first species three and one-half and for the second nearly five 
times its probable error. While, of course, 12-unit means are in- 
sufficient to allow the calculation of entirely reliable probable errors, 
they give some idea of the amount of variability of the results and 
the confidence which can be given them. It is impossible to give any 
such expression applying directly to the damping-off percentages and 
their differences, for the reason that averages for this item have 
been made in the writer's work not by averaging the percentages for 
the individual units but by totaling all the seedlings and the dead 
seedlings on the plats to be averaged and recalculating the percent- 
age from these figures. This seems the only safe method, as other- 
wise units in which germination is low by accident or by the action 
of parasites will be given an influence on the resultant mean entirely 
disproportionate to the number of seedlings which they contain. 
Average values for damping-off obtained by this method and by the 
method of averaging the percentages of the individual plats or pots 
are often very different; it not uncommonly happens that the units 
in which germination is lower than the average also have especially 
high damping-off percentages, both phenomena being caused by an 
unusual activity of parasites. In such case to average the percentages 
themselves usually gives a higher damping-off figure than to total 
the seedlings for the different units and redetermine the percentage, 
and the latter practice is considered the better. In the present case 
the differences in the damping-off percentages obtained by the two 
methods are not great. The figures obtained by averaging the per- 
centages of the ultimate units are as follows : 
Pinus banksiana. — Inoculated, loss 30.9±5.0 per cent; controls, loss 13.2±2.8 
per cent. 
Pinus ponderosa. — Inoculated, loss 40.0±5.1 per cent ; controls, loss 24.1 ±3.3 
per cent. 
The differences between the inoculated and control plats in damp- 
ing-off percentage were for the first species a little over and for the 
second a little under three times their indicated pro'bable errors. 
