74 
Journal of Agricultural Research 
Vol. VIII, No. 3 
DISCUSSION OF RESULTS 
In considering the results of the petrographic examination it should 
be observed that they are such as have been obtained by a search of a 
reasonable number of subsamples and the expenditure of a reasonable 
time. Experience in this work supports the conclusion that by ex¬ 
tending the search indefinitely all the minerals reported as not found 
would have been found in traces. 
Any attempt to correlate the petrographic results with those calcu¬ 
lated from the analytical data must necessarily be rather unsatisfactory, 
due largely to the fact already mentioned, that the petrographic method 
throws no light on the composition of material finer than silt. This 
gives rise to apparent discrepancies. For instance, samples 58 and 59 
showed but a trace of calcite under the microscope, while from the 
analytical data 2.50 and 2.72 per cent of calcium carbonate was calcu¬ 
lated. 
In samples 23 and 24 no calcium minerals were found, while the total 
CaO content was found to be 0.59 and 0.46 per cent. 
In general the prevalence of a number of calcium minerals goes with 
high calcium content as determined by analysis. In samples 3 and 4 
five calcium minerals were found, and the total CaO content was 2.70 
and 2.77 per cent; in Samples 62 and 63 one and two calcium minerals 
were found in traces only, and the total CaO content was 0.27 and 0.37 
per cent. 
Hornblende was the most common calcium-containing mineral found, 
followed in order by plagioclases and epidote. Gypsum, dolomite, and 
apatite were found but once, and minerals Occurring more than once, but 
of rare occurrence, were usually found in soils from the same location. 
For instance, augite was found only in samples 57, 58, and 59, garnet 
in samples 3, 4, 14, and 15, and titanite in samples 3, 4, 15, and 16. 
Calcite inclosed in quartz, a form of inclusion very common in some 
soil types, was found but once, in sample 6. This inclusion of calcite in 
quartz, of course, introduces an error in the results based on analysis, 
calcium in this form appearing with the calcium in difficultly decomposable 
silicates. 
In this series of soil samples the total CaO content varies from 0.27 to 
6.58 per cent, the CaO as calcium carbonate from a trace to 5.67 per cent. 
Calcium sulphate, while usually present, does not exceed 0.02*per cent, 
and CaO in easily decomposable silicates varies from 0.01 to 0.54 per cent. 
Except where large quantities of carbonate are present, CaO as diffi¬ 
cultly decomposable silicates is the predominating class of calcium 
compound, varying from 0.10 to 4.34 per cent. Calcium combined with 
humus compounds varies from 0.01 to 0.58 per cent and is absent frojn 
46 per cent of the samples. This finding is contrary to the common 
belief that humus compounds in soils are usually combined with cal- 
