399] EH. E. Free and 8S. F. Trelease 201 
due to the nitrate in the uranium salt. No stimulation was 
observed with any other ofthe elements tested. 
This failure to secure determinable stimulating effects with 
most of the elements is surprising and is contrary to the re- 
sults of many previous investigations. It seems possible that 
it may be due to the fact that the Shive solution, in the con- 
centration and salt proportions employed, is itself slightly 
toxic because of its high content of magnesium. This solu- 
tion, although it gives the best production of dry weight of 
tops, produces plants many of which show the’ morphological 
modifications characteristic of magnesium poisoning.? These 
observations form one of several bits of evidence which sug- 
gest that the best growth of a plant, as measured by produc- 
tion of dry matter, occurs only when the plant is slightly 
poisoned. It may be a general rule that increased growth 
is the first response to agents or circumstances which would 
prove injuriously toxic in greater concentration or on longer 
exposure. 
We have found some confirmation of this suggestion in our 
experiments on the effect of boron on Canada field pea. Using 
the Shive solutions containing salt proportions other than 
the ones above referred to, and adding borax to these solu- 
tions, considerable stimulations were obtained. The experi- 
ments need to be extended and confirmed, but the present 
indication is that borax is stimulating in those nutrient solu- 
tions which contain less magnesium than the one giving great- 
est dry weight of tops. In other words, slight poisoning, 
such as that caused by magnesium or boron, is essential for 
the production of the greatest dry weight of tops. Hither 
magnesium or boron will serve. Probably other poisons 
would be equally efficacious. 
2 Shive, loc. cit. (2), p. 370-374. Tottingham, William. E., “A 
quantitative chemical and physiological study of nutrient solutions 
for plant cultures.” Physiol. Res. 1: 133-245. 1914. 
