202 Tuttle. — Induced Changes in Reserve 
where lower winter temperatures occurred, showed less starch than that 
from the southern districts. Kiichi Miyake simply recorded the starch 
content as he found it. He did not try to account for its variation. It 
would have been interesting to know whether part of this decrease in 
starch was due to the formation of fats and oils at the beginning of winter, 
but no mention is made of investigation in this direction. 
The present investigation was undertaken, therefore, not so much with 
a view to ascertaining the food reserve in a large number of plants as that 
of a single one, Linnaea borealis , L. var. Americana , Forbes, and to investigate 
the changes which these reserves undergo when subjected to different 
temperature conditions. * 
In view of the fact that temperature is vitally connected with the 
problem, a brief statement may be given of the meteorological conditions 
of the district. Early frosts are characteristic of Western Canada, occurring 
usually from the 6th to the ioth of September — frequently followed by 
warmer periods. The precipitation during September and October is usually 
high, but varies considerably with the different years. The winter seasons 
are quite variable, but the extreme temperatures remain the same from year 
to year, ranging from about — 40° F. ( — 40° C.) to an occasional 50° F. ( 1 o° C.). 
The humidity of the air is moderately low and high winds are rare. The 
depth of snow averages from one to two feet on the level. 
The winter 1916-17, however, has been unusual. Cold periods of five 
to ten days in which the thermometer ranged from o° F. (— 17 0 C.) to 
— 40° F. ( — 40° C.) have alternated quite regularly with slightly longer 
periods of rising temperature, reaching on one occasion 47°F. (8° C.). As 
will be mentioned later, these fluctuations have been useful during the 
progress of the investigation, especially in connexion with reconversion. 
I. Preliminary Tests of Reserves. 
The work was commenced late in October, just about the time of leaf- 
fail of deciduous trees. A number of leaves were examined then for their 
starch and oil content. A solution of iodine and another of chloral hydrate 
and iodine were used in the tests for starch, the former in the preliminary 
tests and the latter in the material during conversion, as the chloral hydrate 
swelled the starch grains and also decolorized the sections, thus making 
the reaction much clearer. For fats and oils, sections of the material were 
tested with a 1 per cent, solution of osmic acid and also with neutral red 
solution, both tests yielding similar results. 
Table I shows the starch and oil content of the majority of local plants 
which still retained their leaves. 
