Transpiration as a Factor in Crop Production. 
27 
data as related to the yield of agricultural crops, and since nearly- 
all such data of importance have been secured by this method 
in the past, it becomes desirable to consider the special sources 
of error entering into it. Such a knowledge is necessary to evaluate 
correctly the results of an investigation. The following outline 
is suggestive of these sources of error. 
OUTLINE OF FACTORS INFLUENCING TRANSPIRATION DETERMINATIONS WITH 
ROOTED PLANTS IN POTS AS SOURCES OF EXPERIMENTAL ERROR. 
1. Character of potometer and contents. 
a. Limitation of amount of soil — 
1. Thru size of potometer. 
2. Thru number of plants grown in potometer. 
b. Limitation of fertility of soil. 
c. Improper distribution of soil moisture. 
d. Evaporation from surface of soil. 
e. Entrance of rain water. 
f. Exposure of potometer and consequent effect on soil 
temperature. 
g. Unintentional lack of uniformity in soil. 
2. Environment. 
a. Testing under unnatural habitat. 
3. The plant. 
a. Plant individuality. 
1. Insufficient number of replications. 
2. Disease and injury. 
b. Stage of maturity. 
1. Insufficient development. 
4. Errors due to methods of computation. 
5. The personal element in drawing conclusions. 
DISCUSSION OF SOURCES OF ERROR IN POTOMETER TESTS. 
CHARACTER OF POTOMETER AND CONTENTS. 
One of the most common sources of error in determining the 
amount of water transpired in crop production is the uninten- 
tional undernourishment of the plants, either thru the use of too 
small a potometer or thru growing too many plants in a potometer. 
Hellriegal (1883), Leather (1911), and Briggs and Shantz (1913a 
and 1914) conclude that the amount of water transpired per unit 
of dry matter produced is increased by a deficiency in the supply 
of food materials, which may become a source of experimental 
error. The same conclusion may be drawn from experiments 
reported in this bulletin. Below a certain range of optimum 
