ones. These crystals occur in the grains of sand. Their occurrence is, however, somewhat rare. The common form in which the platinum occurs is round or oval thin plates, or leaves. But the fact that crystals do occur is important, as proving that the iron-sand is their true matrix. Roughly crystalline platinum also occurs, sometimes with gold crystallised on it, as well as iridium, iron, and native copper. The iron-sand in question is not the ordinary titanic iron-sand of our beaches. It is non-magnetic, or but slightly so, heavier, and lumpier. Lumps of comparatively large size often occur in it, which are iron pyrites derived from wood, and still retaining the original mineral elements of timber, and often some carbon. This is proved from the fact that twigs and small branches are often found along with the sand, completely changed into pyrites, though still retaining the grain, the bark, and all the characteristics of timber. Specimens of these undeniable branches occur where the wood structure is perfect in some parts, while in other parts it is broken up into masses resembling duck-shot, partially fused together. This, I apprehend, gives us the key to the origin of the sand, which seems to be nothing else but the pyritized débris of ancient vegetation subjected to special conditions, which we may yet come to understand.

Just as wood is often silicified into stone in large quantities, or carbonized into coal, so it would apppear that it may be metallized into the iron-sand of our goldfields, auriferous, cupiferous, or platiniferous, from either some obscure conditions of process or inherent quality of the original substance. These pyritized twigs are curiously shrunken to a far smaller size than their original, some of them being reduced to the thinness of needles, whilst still showing wood structure. The iron-sand of many of our goldfields seems to be derived from the breaking down of this pyritized wood by mechanical and chemical means, such as water-wearing and rusting. The sulphur of the pryites is gradually replaced by oxygen to form the magnetic oxide, probably determined by the conditions of deposit, temperature, etc. In this way, the magnetic iron-sand of our beaches would be the ultimate product of timber, after being first reduced to wood pyrites, and then broken down by oxidation and the action of mechanical agents, and finally changed into the magnetic oxide, the other metals crystallizing out.

This change can actually be effected experimentally by the artificial oxidation of the non-magnetic lumps of wood pyrites, with the production of magnetite in all respects similar to the titaniferous iron-sand of our shores. Either sulphuric or nitric acid will effect this by long continued action. The grains of sand do not dissolve, but become semiplastic, lose their sulphur, and recrystallize into highly magnetic angular grains of the ordinary magnetic iron-sand.