296 
Journal of Agricultural Research 
Vol. XXVII, No. 5 
The ratios of the differences to their probable errors (column 9) are 
in all cases larger than, and in some cases several times as large as, 3.5 
times their probable error. There can, therefore, be no reasonable 
question concerning their significance. The results show clearly that the 
tissue fluids of the Egyptian cotton are distinctly more acid than those of 
the Upland cottons. 
In the calculation of the percentage difference between the hydrogen- 
ion concentrations of the tissue fluids of the two species, a procedure 
must be adopted which is somewhat different from that used for the other 
constants. When expressed in terms of P H , both acidity and alkalinity 
are measured from a neutral point determined by the dissociation con¬ 
stant of pure water. We may, therefore, logically take the deviation of 
the P H value of the Upland cotton from neutrality as a base in calculating 
the percentage differences. 
The exact value of P H taken to represent neutrality is of relatively 
little importance. The dissociation constant of water varies with the 
temperature. We may arbitrarily take P H = 6.860 for neutrality at the 
average temperature of our determinations. In determining the per¬ 
centage difference between Egyptian and Upland cotton we therefore 
work from the formula 
100 (E—U) 
6Mo^ IT 
where E and U represent the P H values of Egyptian and Upland cottons, 
respectively. 
The percentage differences (Table XIII, column 10) calculated in this 
way range from 2.75 to 8.67. 
The correlations between the Upland and Egyptian plants of the same 
duplet or triplet (Table XIII, column 7) are of considerable interest. 
The values for the second series are very much larger than those for the 
first. All of the coefficients are positive in sign. These results are of 
importance in that they show that the hydrogen-ion concentrations of 
adjoining plants are correlated. This must be assumed to be due either 
to (a) extremely local differences in the soil of the field, or to (b) the 
influence of variations in the atmospheric conditions or in the time of 
collection of the samples. It is impossible at present to make any 
suggestions concerning the real factor to which these correlations are 
due. Their existence indicates clearly, however, that here is an impor¬ 
tant field for further investigation. 
Turning now to the differences between the first and second collec¬ 
tions (Table XIII, columns 5 and 6), we note that in all cases the value 
of P H is somewhat larger in the second series than in the first. Thus the 
tissue fluids were on the average more nearly neutral at the time of the 
second series of determinations than when the first series was made. 
A comparison of these differences with their probable errors shows 
that in every comparison except that based on plants grown from bulk 
seed of Meade in rows 3 and 7 the differences are at least 6 times as large 
as their probable errors. 
There can, therefore, be no reasonable doubt concerning the validity 
of the conclusion that at the time of the taking of the second series of 
samples the tissue fluids of both Egyptian and Upland cultures were 
more nearly neutral than they were earlier in the season. 
