508 
Philippine Journal of Science 
1920 
solution conditions. For the same reason, the detailed values 
need not be considered, attention being confined to the group 
designations only, for both dry yield and green weight. Table 
12 shows the group designations (high, medium, or low) for 
dry yield and green weight of tops, for each set of salt pro- 
portions and for each of the four different total concentrations. 
In this table each series has a double column, the first for the 
dry yield and the second for the green weights. Where the 
group designations are different in the two cases the letter X 
is placed in the space between the two values; this brings out 
the frequency of discrepancies and shows for what sets of salt 
proportions and total concentrations they occur. 
Inspection of Table 12 leads to the impression that dry yield 
and green weight are generally in agreement as far as placing 
the various culture solutions in the low, medium, or high group 
is concerned. Out of one hundred forty comparisons between 
the indications given by these two criteria only thirty-one, or 
22 per cent of the total number, fail to show agreement. This 
means that the two criteria agree in 78 per cent of the cases; 
and, since these cases of agreement represent the whole range 
of environmental complexes here studied, it may be concluded 
that the two criteria are generally both influenced in about the 
same way, by total concentration, salt proportions, and non- 
solution conditions. It should be emphasized that this con- 
clusion is to be regarded as holding only for the plants and 
environments dealt with in these experiments; such a general- 
ization is not warranted for the whole plant kingdom or for 
other solution types, etc. It may or may not hold for these 
other conditional complexes, for other plants (or for lowland 
rice plants in later developmental stages), etc. 
With this warning as to the danger of attempting to generalize 
beyond the limits of the field actually investigated (a danger 
that seems not always to be appreciated fully in biological 
writing) , attention may be turned more specifically to the cases 
that show disagreement in the indications of the two criteria 
here considered. In the first place, it may be inquired whether 
these discrepancies are any more or any less frequent with one 
total concentration than with another. Table 12 shows nine dis- 
agreements for the lowest total concentration, eight for the low- 
medium, ten for the high-medium and only four for the highest 
total concentration ; it therefore appears that, with higher total 
concentrations than are well suited to the development of this 
plant with this type of solution, the agreement between dry yield 
