SUGGESTIONS TO PROSPECTORS. 1()9 
often be observed first within the peridotite area, and, when this is the 
case, a larger deposit may very reasonably be expected to be found 
near the contact. 
If, when examining a peridotite formation, no corundnm is found 
along the contact of this rock with the gneiss or other country rocks, 
there is little or no need of searching in the midst of the formation, 
for it is only at the contact that any large deposit may be expected to 
])e found. Again, if there are no fragments of corundum to be found 
at the loAver borders of a peridotite formation and none can be 
obtained by panning the gravel of the streams flowing by them, there 
is but little probability that success will attend any further search. 
It should also be borne in mind that at all the peridotite localities 
(>xamined no more than traces of corundum have been found where 
there is a quantity of chromite, and that, on the other hand, where 
there is much corundum there is a scarcity of chromite. 
Where crystals and fragments of corundum are found in gravel 
l)eds or in the soil, they of course indicate the existence of corundum 
(h^posits at some distance from where they are found. On account of 
the stability of this mineral when exposed on the surface, it may have 
l)cen transported long distances down the mountain slopes without 
showing any appreciable alteration and l)ut little abrasion. When, 
liowever, it is associated with certain minerals, as fragments of chrys- 
olite (olivine), chlorite, or talc, it is very probable that they have 
been transported but a short distance, for these minerals would have 
been almost entirely disintegrated and worn away before they could 
have been transported very far or exposed very long to surface 
influences. AVhere such fragments have been found, they should be 
followed up the valley or incline by which they could have most 
naturally reached their position. AYlien following up these gravel 
deposits, whether they are in the actual or in the old beds of the 
streams, they should be tested at intervals by panning, until a point 
is reached beyond which no corundum is foinid in the gravel, and then 
the search should be carried on up the hillsides until the parent rock 
is located. If this rock is a peridotite, its contact with the other coun- 
try rock should be carefully examined. Where there is considerable 
soil and decomposed surface materials, the contact can not always be 
at once located, and then ditches should be cut through the soil at 
right angles to the strike of the rock itself. When the contact has 
been found, the border zones of the minerals should be examined for 
corundum, a guide at this point being the chlorites and vermiculites 
that have always been observed surrounding corundum at the contact 
of a peridotite with a gneiss or other country rock. 
