Gold prospectors recommend that the selection of prospecting samples must be done considering some permanent object, such as a cross-cut, winze, or survey station. The intervals between samples should be measured along the center of the drift, as they otherwise will differ widely according as they are measured on one side of the drift or the other, and will therefore fail to plot correctly on the map. It is poor judgment to mark the points at which the samples are to be taken in advance of the actual sampling; this amounts to an advertisement that a sample is to be taken along a certain line, and permits the evilly disposed to assist nature in the distribution of gold bearing minerals.
It is important to mention that a face that is to be sampled should be thoroughly cleaned. If the ground is soft, a strip a few centimeters wider than the sample cut must be cleaned off with a pick; if the ore is hard, a brush or broom should be used either dry or with water. In driving any working, the fine material, often the richest, is powdered and thrown against the roof and walls, where a portion of it adheres; it is, therefore, of the greatest importance that the face to be sampled should be thoroughly cleaned. Irregular projections and loose pieces should be knocked off, to give, in so far as possible, a flat surface from which to cut the sample. The face to be sampled should be examined carefully for soluble salts. In some gold-copper mines in dry climates an efflorescence of chalcanthite and other salts is usual, and the sampling of old workings is attended with considerable risk of salting from this cause.
In some deposits, the samples that contain this efflorescence, even after boiling in water, show blue crystals under the microscope. These efflorescences are due to the evaporation of migrating solutions on the walls of the workings, and represent an enrichment not found throughout the mass of rock. Where it is necessary to take samples from the floors of drifts it is best to cut large samples and to wash from them and disregard all fine material; fine particles of gold minerals work into cracks in the floor and give deceptively high results. This method may give results somewhat too low, but is not so liable to serious error as would result from the inclusion of the fine material.
Experienced gold prospectors say that samples should always be taken as nearly as possible at right angles to the lines of distribution of the minerals through the ore. The interval between samples depends upon the regularity with which the gold bearing minerals are distributed. One extreme might be considered an absolute uniform mass, of which one sample would suffice, and the other extreme, a segregation of all the valuable minerals into a single mass; it is therefore apparent that the right interval between samples will differ with each exposure sampled. In general, a 6 m. interval will suffice for a large ore-shoot of uniform grade, a 3 m interval in average cases where many samples are to be taken from the same ore-shoot, and 1.5 m. or lesser intervals where the ore is spotty. It is advisable to start with 6 m a property where no data are available, and resample at 3 m intervals, and perhaps again at 1.5 m. intervals, where the results from the first series indicate that such a course is advisable.
In sampling a wide gold vein or deposit it is best to divide the width into sections and to sample them separately, in order to determine the distribution of the values. These widths may be taken over even multiples of the total width if the deposit presents a uniform appearance. If the vein or deposit presents a variegated or banded appearance, however, the several bands or zones should be sampled separately. Where a section of ore is irregularly exposed, as is commonly the case with a vein in the roof of a drift, the sample must be cut deeper over the part that is at right angles to the vein than where the face is slanting, in order not to get an undue proportion from the slanting exposure.