SHULL—OXYGEN MINIMUM AND GERMINATION 471 
Discussion 
The methods employed in much of the recent work on delayed 
germination of seeds has not been as exact as is desirable, or even 
necessary, for the solution of the problems involved in this phenome- 
non. ‘The factors which are influential in the germination behavior 
have been investigated qualitatively only, without any attempt to 
measure them accurately and determine their relations. Moreover, 
the seed has been regarded too often as an embryo only, which may 
be affected profoundly by chemical and ethereal stimuli, the testa 
being considered as negligible on account of its thinness or on 
other insufficient grounds. In K1nzev’s latest paper (16) he claims 
to have met all the serious criticisms made against his work, but 
in none of his experiments has he eliminated the testa as a factor. 
Recently LEHMANN (17) has shown that other stimuli than light 
can be substituted in its place and produce approximately the same 
effect. For instance, he found that the effect of 1 per cent Knop’s 
solution in the germination of the seeds of Ranunculus sceleratus 
was much the same, inducing practically the same percentage of 
germination. If such substitutions of stimuli are possible, what 
can be stated with certainty as to the cause of delay in these cases ? 
Light and the chemicals undoubtedly affect something, but it may 
as easily be the testa as the embryo that is changed. Even if the 
effect is actually produced in the embryo in these instances, the 
phenomena are not explained by saying that the protoplasm has 
been rendered “lichthart”’ or “‘lichtmiide,” as if it were a sort of 
“weariness of the flesh!’ If there is a change in the permeability 
of the embryo, or other physical or chemical alterations in it which 
leads to activity or cessation of activity, demonstration of such 
changes would throw some light on the phenomena. 
In these Xanthium experiments the determinations have been 
made with as great exactness as possible, the accuracy lying well 
within the variability of the seeds themselves, and the work has 
been almost entirely quantitative. These methods demonstrated 
clearly the physiological difference between the embryos of the 
upper and lower seeds, although the difference is so slight as not 
to affect visibly the germination of the uppers at atmospheric pres- 
sure when the testa has been removed. This fact shows the value 
