ENVIRONMENTAL AND HISTORICAL FACTORS. 93 
may be a number of days after the occurrence of a heavy shower or series 
of showers, before any effect of these is felt at the depth of 30 cm. This 
consideration emphasizes the inadequacy of mere precipitation data in 
any attempt to determine the moisture conditions under which the plants 
of any region live. The lagging in the opposite direction is also well 
shown by the curves, and is of even greater importance to plant-life in 
the desert. Thus, a dry period of several weeks may ensue without the 
occurrence of any appreciable change in the moisture conditions of the 
deeper soil layers. In general, the heavier the soil, the more important 
is this principle, but other factors must be considered. In the case of 
the hiil soil, the water contained in the deeper portions is lost only through 
plant transpiration and direct evaporation at the upper surface, but in 
that of the soil of the river-plain water is lost from a depth of 30 cm. 
both by an upward and by a downward movement. The underlying rock 
of the hill prevents water-loss downward, while the dry underlying soil- 
mass of the flood-plain absorbs water with great avidity and thus acts 
not unlike a process of evaporation effective from below, as in the case 
of a suspended porous clay pot of soil. 
From very limited data’ on the relation of the activities of desert 
plants to soil-moisture, it appears that the minimum for the activity of 
forms not possessed of storage organs probably lies in the vicinity of 10 
per cent as here calculated. Making such an assumption, it appears 
that the hill soil at a depth of 30 cm. was in good condition for plant 
activity throughout the period of observation, and that the soil of the 
river-plain at the same depth was not in as good condition. The con- 
ditions at a depth of 30 cm. in the wash and at a depth of 20 cm. on the 
Larrea slope can not be regarded as favorable to plant growth, although 
during February and early March considerable moisture might have been 
available here. 
The moisture conditions of the surface layers of the soil must be re- . 
garded as the prime factor in the determination of germination, since 
seeds seldom find their way to very great depths and are unable, in most 
instances at least, to germinate without a relatively great supply of 
oxygen. It appears from the curves that germination was probably 
possible in the surface layers of all the soils throughout February and 
early March. In order that most perennials may succeed in the desert, 
their seeds must be able to germinate in the surface soil when this is 
moist, and the seedlings must send their roots downward rapidly enough 
to reach the deeper soil layers before they are overtaken by the desicca- 
tion of the surface soil. A large proportion of the seeds which germinate 
in the vicinity of the Desert Laboratory meet an untimely end through 
failure to fulfil this condition. 

1For some data on the minimum of soil-moisture with which plants can be active, 
see Publication No. 50, Carnegie Institution of Washington, pp. 66-67. 
