34 BULLETIN 1059, U. S. DEPARTMENT OF AGRICULTURE. 
applied to all the hours of the day which it represents. This implies, 
of course, that the range of the instrument has been adjusted before 
it is placed in service. 
Summary of Soil Temperatures. 
In addition to the means by decades and months, the "Soil Tem- 
peratures" form may show the number of days, for each depth at 
which readings are taken, with temperatures below 32° F. (frozen) ; 
with temperatures 32.1 to 41.0° (cold) ; with temperatures 41.1 to 
50.0° (cool) ; with temperatures 50.1 to 60.0° (warm) ; and with 
temperatures above 60° (hot). 
Annual Summaries of Soil Temperatures. 
The "Summary' 1 form may be used for a summary of one or several 
soil temperature conditions, such as the mean temperatures by decades, 
months, growing seasons, and years, or the number of days of each 
temperature class in each month. In the case of surface tempera- 
tures, the mean and absolute maxima and the daily ranges are doubt- 
less of great interest. As many forms as necessary may be used. 
It may be found that a given soil temperature sufficiently delimits 
growth so that the occurrence of such a temperature marks the be- 
ginning and end of the growing season. This has been the idea in 
suggesting a division of temperature computations at 41° F. or 5° C, 
such a temperature being approximately the minimum for activity of 
lower forms of plant life, as shown by numerous experiments. 
Apparatus. 
The most simple apparatus for measuring soil temperatures is the 
encased soil thermometer, having a stem of sufficient length so that 
the mercury appears above the surface of the ground when the bulb 
is at the desired depth. As ordinarily made, however, this thermom- 
eter is not only very expensive but is inadequately protected from 
exposure to the elements and to mechanical forces. For this reason 
it is not desirable for permanent stations, but will probably in many 
cases be useful where observations are temporary and light equip- 
ment is desired. 
For permanent stations the most serviceable apparatus that has 
been thoroughly tried is an ordinary thermometer suspended by a 
cord in a 1-inch iron pipe, whose lower end may be sealed either by 
a cap or by welding. The latter is preferable where the pipe must 
be sunk to any great depth, since the welded pipe may be formed 
as a wedge and may be driven into position without seriousty dis- 
turbing the soil. The pipe should, in all cases, be long enough to 
extend well above the ground and above any ordinary snow cover- 
