Subsurface Methods as Applied in Mining Geology 983 



judgment of the sampler after the deposit has been studied. Indiscrimin- 

 ate sampling without regard for geologic conditions and the extent and 

 grade of ore should be avoided. Wherever feasible, individual samples 

 should represent material of only one general grade or character. Not 

 only are more accurate determinations of mineral content possible, but 

 the assay maps will show distribution by grade and width, and the con- 

 tent-distribution pattern may indicate otherwise unrecognizable guides 

 and controls. 



Several techniques can be used to obtain good samples. As stated in 

 preceding paragraphs, cuttings and cores from drill holes can accurately 

 represent the rock through which the hole passes. Another common method 

 of sampling is to cut across the body along a predetermined channel, re- 

 moving proportional amounts of the differing material. In addition to 

 sampling in definite zones, grab samples are taken either according to a 

 fixed pattern or more or less at random. Grab sampling is less accurate 

 than channeling but serves a useful purpose for preliminary work or for 

 checking other samples. Grab samples underground are taken from the 

 piles of broken rock, ore chutes, or ore cars. Before taking a sample care 

 must be exercised to see that the walls or workings are clean and repre- 

 sent the rock to be tested. Leached zones and accumulations of soluble 

 salts must be removed along with all dust and mud. 



A sample too large to be taken to an assay office economically or 

 conveniently may be reduced to one-half, one-fourth, or one-eighth. By 

 the use of a mechanical splitter fractions can be obtained that are of uni- 

 form grade. In most prospecting work, however, mechanical splitters are 

 not available, and the sample must be divided by coning and quartering. 

 As the term "coning" implies, the rock is first coned by pouring or shovel- 

 ing all the sample onto one spot on a smooth, clean surface. The cone is 

 uniformly flattened, the resultant mass is divided into quarters, and op- 

 posite quarters are combined to make a sample one-half the volume of 

 the original. If further reduction is necessary, the procedure may be re- 

 peated one or more times. 



Representation and Correlation of Data 



Considerable data may be accumulated on a particular property, but 

 unless the data can be evaluated and correlated they are of little value. 

 One important factor and aid in correlation is proper representation of 

 data.^^ ^^ Standard methods of showing underground geology include 

 plan maps of all levels, cross sections, and longitudinal sections. The 

 level map is most important in that sections can be constructed from a 

 set of level maps. In mines where the geology is complicated, stope maps 

 generally are made. These are of most value if made at regular intervals 

 between levels and to be usable must be related in plan and elevation to 



*^ Forrester, J. D., op. eit., pp. 445-497, 1946. 

 ^McKinstry, H. E., op. cit., pp. 162-198, 1948. 



