130 
Journal of Agricultural Research 
Vol. XXIV, No. a 
From the data in Table XII we see that the limber pine was little 
stimulated, probably because its transpiration is always moderate. 
The spruce was still less affected, apparently because it is always able 
to satisfy its needs. Lodgepole, Douglas fir, and yellow pine were about 
equally helped and seemed greatly invigorated. The relatively high 
transpiration of the fir on October i and 3 can be accounted for only 
by an error in weighing. 
To a certain extent, these performances may be accounted for by the 
root habits of the trees. It is possible that the finely divided and 
numerous roots of the spruce give it immediate control over so much 
more soil that it exhausts the available water much less quickly than 
those species which usually develop only a few coarse roots. To some 
extent this would also explain the ability of limber pine to obtain its 
water more steadily. On the other hand, either lodgepole or Douglas 
fir ordinarily has much better roots than yellow pine, yet these three 
were about equally stimulated by a heavy addition to the water supply. 
On the whole, this matter is only suggestive and does not, we believe, 
explain the relative behavior of the species. That Engelmann spruce 
possesses a remarkably great ability to supply itself witti all the water 
that is needed under the most trying circumstances, and that this 
ability is exceeded, among the species studied, only, possibly, by that 
of Siberian larch, seems proved beyond a shadow of doubt. This is 
plainly shown in the day-to-day records where, if there is a marked 
contrast in the amount of sunlight on two succeeding days, or in other 
conditions conducive to high transpiration, it is the spruce which is 
invariably able to live up to these conditions most fully. Thus we are 
enabled to say quite co^dently that the relatively high rate of tran¬ 
spiration of spruce on cloudy days, as shown by Table V, truly expresses 
an ability to make use of all available light and does not signify that 
this species is unable to meet the conditions which cause high trans- 
piratipn from all the species. 
SUMMARY 
It has seemed desirable to go into this matter fully on account of the 
complicating features introduced by the radically different results 
secured in 1917 and 1920, and in order that we might not deceive our¬ 
selves as to the true meaning of the results. It has been necessary for 
us to go through with this analysis in order to reach a conclusion, and 
it is hopeless to expect the reader to reach a conclusion by any other 
process. 
It now becomes fairly apparent that transpiration is very much 
dependent on water supply and that the relatively low water use of 
some of the species in 1920, when the water supply was maintained at 
a low level, is not to be considered as a virtue but rather as evidence 
of a lack of ability to supply needs. And, even though in some cases 
growth may not have been seriously impaired by the inability of certain 
species to keep the leaves well stocked with water, yet it is perfectly 
evident that the species which show this inability in the most marked 
diegree would soonest succumb in time of real drought or in the usual 
autumnal drought that occurs where there is strong competition. 
Tha'e are, apparently, two slightly different problems to be considered 
in comparing the species. The one has to do with the relative require¬ 
ments of different tree species of a unit size. The other has to do with 
