14 BULLETIN 700, U. S. DEPARTMENT OF AGRICULTURE. 



may differ appreciably in their crop-producing potentialities. Thor- 

 ough mixing of the soil, therefore, regardless of the care with, which 

 it has been selected, can. not be overemphasized as a means of avoiding 

 outstanding errors from this source in experimentation. 



MEASUREMENT OF PLANTS. 



At each of the type stations measurements of growth were made 

 on the following number of specimens : 20 of wheat, 10 of peas, and 5 

 of native brome grass. The lid of each potometer contained five 

 evenly spaced perforations (fig. 1). In the case of wheat, 10 seed- 

 lings were planted in each pot, 2 in each perforation; while in the 

 case of peas and brome grass, each pot contained only 5 plants. 



Throughout the growing season both leaves and stems of the plants 

 were measured at regular intervals. The object of the measurements 

 was to obtain data as to the relation of the environment to (1) the 

 tendency of the plants to elongate their stems and (2) the tendency 

 of the plants to expand their leaves. Measurements of the stems 

 furnished direct data as to the rate of elongation of the plants. In 

 obtaining leaf expansion, however, indices had to be used instead 

 of actual figures on leaf area. Hence in the case of the grasses the 

 leaf expansion was obtained by recording the length of the leaves; 

 for, as will be shown in the calculation of the experimental error, 

 leaf length is proportional to leaf area. In the case of the peas, an 

 index of leaf area was obtained b}' recording the number of leaflets, 

 as they were found to be of rather constant average size and were 

 considered as units of area. 



Since the seedlings were all of uniform size and inconsiderable in 

 comparison to the subsequent growth of the plant, the measurements 

 Were considered as beginning at zero. During the first half of the 

 growing season all the plants were measured at 10-day intervals. 

 Owing to the number of plants grown and their luxuriant develop- 

 ment, it became impossible in the first week in August to remeasure 

 all of the plants at lO-clay intervals; so from then on the measure- 

 ments were made once a month. 



At the end of the growing season in each type the plants were cut 

 at the junction of the stem and the lid of the potometer, and the 

 measurement again recorded. In addition, the dry weight and the ash 

 content were determined. In the case of plants grown in the aspen- 

 fir association, the soil was washed away from the roots and the dry 

 weight of the latter obtained. 



EXPERIMENTAL ERROR. 



In determining the rates of growth and other physiological activi- 

 ties for a given species some variations are sure to be found in indi- 



