13 
Effect on soil No. 530. — The table shows that all the salts, with 
the possible exception of calcium phosphate, cause a re^jular decrease 
in rise of moisture with increase in concentration. Calcium phos- 
phate, when added at the rate of 0.66 per cent, increases the 
activity. 
Effect on soil No. J^Jj.8. — Here again the effects of various salts seem 
to be related with few exceptions and the tendency is toward an in- 
crease in capillarity up to a certain concentration beyond which a 
further addition of the salt materially retards the rise of water. 
Exceptions to this rule are sulphate and carbonate of calcium, 
which, being difficultly soluble, w^ould materially change the physical 
nature of the soil, and calcium phosphate, which, being a soluble 
salt, exerts a chemical effect. 
Effect on soil No. ^28. — The effects of salts upon the capillary* 
activity in this soil are similar to those in the former except that the 
variation in moisture rise is greater. The soil, containing a greater 
percentage of sand, shows a higher rise in moisture. 
Effect on silica sand. — Conditions in sand are ideal for measuring 
the effect of salts upon capillary action within certain limits, beyond 
which any further concentration gives misleading results. This is 
due to the filling of the pore spaces and the subsequent drawing up 
of moisture, caused by the salts as they become dissolved by the 
rising moisture, rather than to the action of the salt upon the activity 
of the film surrounding the grains. Taken as a whole the results 
with a concentration of salt below 3.3 per cent are very similar to 
those obtained with the sandy soil, the more marked variation being 
in the action of lime. 
Apparently the most important inference to be drawn from the 
foregoing table is that the action of a high concentration of a salt in 
most instances only magnifies that of a small application. 
With this fact determined it was decided that the salts should be 
used in as small amounts as possible and in amounts proportional to 
their molecular weights, thus more nearly approaching normal con- 
dition. The same molecular proportions as previously mentioned 
were used, w^hile the fertifizing substances, such as mineral phos- 
phates, blood, etc., were added at the rate of 0.1 per cent. The 
results given in Table VI show the relation of various salts to the rise 
of moisture in several of the dominating types of Hawaiian soils. 
