1g12] BRIGGS & SHANTZ—WILTING COEFFICIENT 33 
and the residuals or the difference between the observed and the 
computed values. 
TABLE V 
COMPARISON OF THE OBSERVED WILTING COEFFICIENT WITH THAT FOUND BY 
COMPUTATION FROM THE MECHANICAL ANALYSIS 
a ae WILTING 
Bin 2 8 ° COEFFICIENT Ratio 
aug ate es te iat bcoere 
He. SOn, TPE wiki ase we 8& | Com- | Ob- |smuars| To 
ac g poe 4o8 208 _— served COMP. 
S58 |) ase Bae és centage | centage 
ee Coarse sand 60.4 } 37.2 }.-0.8) 2.6 | 3.8 |-0.9 |-+0.9 | 0.50 
aoe Fine sand 25.2) CATAL as Fut ee | Bet 2.6 |+0.5 | 0.84 
Se4 Fine sand 86.41 $50 eG St 19-0 | 9.3 10.3 | 0-92 
O.: Fine sand 90.0 |. £0.7:1-6.01 5.311 - 8.84 53.6:}-+0.2 | 0.905 
ee Sandy loam 93:1 -80.0:1- BH 9.8 1. 4-9 1 4.8 [40-* | 0.08 
455 Fine sandy loam| 2.8 | 59.8 | 30.2 | 6.9 | 10.3] 9.7 |+0.7 | 9.94 
P2c4 Loam 3.4] §§.5 [21-8 | 19.4:| 9.8 1 70:9 |—-0.8.) 1.08 
A.....| Sandy loam $3641 28.8 [26.9.4 288 | 6.9 1° $6.97 6.0] 3.60 
| Pca ine sandy loam| 15.8 | 42.4 | 28.7 | 12.9 | 10.7 | 10.8 |—o.1 | I.o1 
Sb ees Fine sandy loam| 19.2 | 35.6 | 30.6 | 14.7 | 11.4 | 11.6 |—0.2 | 1.02 
5... 65) Loam 2:6 PAB-6 377 1. 12.30): 13.5 113-0 [0.4 | 1503 
D Loam 3.6 | 35.2 | 41.4 | 14-4 | 14.6 | 15.2 [0.6 | 1.04 
14 Clay loam Cb ay Oe 36.2 1 94.5 | 34.5 | 40.2: 1-<2.7:} 4.32 
Bas suis Clay loam 2.2) 40.9 [GSE 1 27.2. | 16.0 | .16,5 [0.8 | 2.08 
6 Clay loam 4.4 | 20.5 | 52.6 | 22.0 | 16.6 | 16.3 |+0.3 | 0.98 
The ratio of the observed to the computed value of the wilting 
coefficient is also given in the last column of the table in order to 
provide a basis of comparison as regards accuracy with the other 
physical measurements. The mean ratio® is 1.00, with a probable 
error of +0.025. This large probable error is due mainly to soil 
no. 1, which has a departure no greater than some of the other 
soils, but which on account of its very small wilting coefficient 
gives a ratio which is widely divergent from the rest of the series. 
The formula for computing the wilting coefficient, when affected 
with its probable error, then becomes: 
o.o1 sands+o.12 silt+o0.57 clay 
I=0.025 
= wilting coefficient. 
23 In determining the values of the sand, silt, and clay coefficients so as to give 
a mean ratio equal to unity, soil no. 1 was disregarded, since a better general agreement 
was obtained in this way. This is virtually what would have happened if the method 
of least squares had been applied to the experimental data. In all calculations of 
probable error, however, this soil has been included with the rest. 
