36 A Study of the Vegetation of 
dry weight. The chart shows that at no time did these virgin 
forests of fir-tamarack or cedar exhaust all of the water in the 
surface ten inches of soil. Likewise, samples at a depth of 2 
feet on August I, 1914, gave a water content of 66 per cent. and 
19 per cent. in the cedar and fir-tamarack forest respectively. 
Similar determinations at 2 feet in the rocky pine-covered soils 
gave a water content of about Io per cent., which is very near the 
wilting coefficient. The slow growth and dwarfed appearance 
of these trees reflect the unfavorable habitat conditions. 
These studies on the evaporating power of the air and the 
amount and range of soil moisture in different plant communities 
may be briefly summarized. 
Evaporation at different stations within the same plant com- 
munity exhibits variations similar both in character and degree, 
and the variations in the rate of evaporation gradually become 
less and less as the climax type of vegetation is approached. 
The great amount of evaporation in the earlier stages of succes- 
sion seems a sufficient cause for the xerophytic character of the 
vegetation. 
A study of the differences of the rates of evaporation in the 
various plant communities shows that these differences are suffi- 
cient to be important factors in causing succession, at least 
through the earlier stages, where light does not play an important 
roles 
If sufficient light is available, there is no question but that 
humidity of the air and the soil are the most important factors 
1 While reading proof of this manuscript the recent paper by Gates came 
to the writer’s notice. (Gates, F. C. The relation between evaporation 
and succession in a given area. Am. Journ. Bot., 4: 161-178, 1917.) Asa 
result of an investigation carried on at Douglas Lake, Mich., he concludes 
that a decrease in evaporation is not a prerequisite to succession; lowered 
evaporation being coincident, with and not an antecedent to it. Gates 
worked in a region with summer rainfall, where the average daily rates 
of evaporation were low. The highest average daily summer evaporation 
obtained on open ground was only 14.7 cc. Certainly his conclusions are 
not applicable to a region of practically no summer rainfall and where, in 
areas covered with bunch-grass, the daily evaporation rate for the whole 
growing season averaged 28 to 37 cc., with the average daily rate for cer- 
36 
