UTAH ACADEMY OF SCIENCES 97 
With each row of tumblers was one which contained the 
same soil as the others but no salt added. The dry 
weight of plants produced in this was taken as 100 per 
cent and the yield of the other tumblers computed on 
this basis. To express the relative deviation of the 
results the probable error in per cent of the mean was 
used. This figure is the percent of the average result 
which marks the limits above and below the mean within 
which half of the individual results will fall. It is 
worked out by the formula 
>d° 
P. E. (% of mean) = + e745 n 
M 
‘where ‘‘Sd?” is the sum of the squares of the differences 
between each individual result and the mean, ‘‘n’’ is the 
number of samples, and ‘‘M’”’ is the mean or arithmetic 
average. “n” is at least 150 determinations for each 
figure. 
The results are summarized in Tables I, II, and III. 
TABLE I. 
Crop production for soils containing different 
amounts of sodium chloride compared with different 
tests for the salt. 
Ly A IS +P | 39 & Sai 
s ~ is) & r) or Ss 
Zo] == len |/ee=| or /28ela~/ cee] ax 
SH st | pe ea aah BL = yok aie eg | 9 
ao 22 (SF .o2 | eh ies) SP saa| sh 
o ae. Te Bea] 6 BZ {se xe4| 8 
= a Is cee eS Ss | § 
| | ‘cee ore 
100 || o| 51]) 196{ 91 || 918] 72 || .41 | 55 
114 | 200] 49 || 387] 66 || 716 | 61 || .49| 52 
104 || 400] 39 492 | 41 || 963 | 50 55 | 42 
113 ||} 600] 42 656 | 44 || 1155 | 53 || .61 | 44 
104 || 800] 52 858 | 53 || 1484] 62 70 | 53 
98 || 1000] 45 || 1082] 47 || 1760 | 51 | 83 | 48 
81 || 1500 | 59 || 1544] 60 || 2403 | 63 |] .98| 60 
65 || 2000] 54 || 2054] 55 || 3062 | 57 | 1.23 | 55 
50 1 2500 | 63 || 2608 | 64 || 3645 | 68 || 1.44 | 65 
39 || 3000] 70 || 2994| 70 || 4227] 72 || 1.66] 71 
23 3500 | 92 || 3399] 92 | 4704 | 93 1.85 | 93 
16 4000 | 115 | 3812 | 116 || 5016 | 117 } 1.93 | 116 
