both by native chiefs of the Batclapin (or Koranna) tribes, and by the Orange River Free States (lying south of the river), and even by the Transvaal Republic, whose main territory is higher up to the East on the north side of the Vaal. The rights of ownership and the demarcation of boundaries are to be settled by Commissioners or otherwise; and some kind of regular government has to be organised, in spite of the rowdyism of the gem-seeking cosmopolites and the disputes of the conterminous states and tribes, if the diamond-yielding character of the region be persistent, or if the labour and capital of the present settlers succeed, as is probable, in fixing civilised homes thus far in the Interior, in spite of the sandy and stony soil, and of the floods of one season of the year and the scorching heats of another.

Of the Klip-drift Diamond-field a surveyed plan has been sent to England by Mr. E. T. Cooper, one of the Government land-surveyors of the Cape Colony. It was published in the "Mining Journal" of March 4, 1871, and gives a good notion of the existing topography and drainage, and of their relation to the probable conditions under which the diamonds were deposited there.

At and near Klip-drift the river has an extremely winding course among somewhat flat-topped hills, a mile or so in their greater diameters, and varying from 300 to 450 feet in height, with gullies or creeks running down between them to the river. The tops and slopes of these hills ("Kopjes" they are locally termed) have been the chief sources of diamonds to the diggers. According to Mr. E. T. Cooper, writing in October 1870, Hond's Kopje, 400 feet high, has yielded possibly 15,000l. worth of diamonds; Rosa's Kopje, 400 feet, about 100,000l. worth (including a diamond of fifty-six carats); and Original Kopje, 300 feet, upwards of 100,000l. worth.* The yield of the hills on either side of the river at the drift or ford itself (on the mission ground 450 feet, and opposite about 50 feet high) is not specified.

The flats by the waterside do not appear to have been here successfully worked-only the sides and summits of the hills; and here the diamonds are found in a ferruginous gravelly alluvium,

These estimated values for Klip-drift alone far surpass the declared value of the diamonds shipped to England, according to the statement in the "Times" for February 7, 1871. Klip-drift, rich as it is, has been deserted by the digger for Hebron and Cawood's Hope. The latter place is opposite Gong-gong, and the diggings are in recent alluvium, accumulated by the river between an island and the bank. This we learn, whilst this paper is in the press, from Mr. Barry's Lecture of January 27, reported in the "Grahamstown Journal," and reprinted in the "Colonial News" of March 17.

consisting mainly of lydite, jasper, agate, and quartz, with ochre and rotten felspar, mingled with large and small blocks of the felspathic amygdaloidal trap-rocks that constitute the bulk of each hill. Indeed this basement-rock is undergoing decomposition, some of it breaking up into rotten ochreous felspar, with frequent concretions of chalcedony (agate and carnelian), whilst other portions, less decomposable, remain in angular and exfoliating pieces.

At different spots, however, the gravel varies in its constitution. Rock-crystal is plentiful in some places, and amethyst. also occurs. Garnet, peridot, mesotype, natrolite, and calcite abound here and there; but the exact conditions under which they are met with have not been noticed as yet. Brown mundic (hepatic pyrites), ilmenite, specular iron-ore, diopside, tourmaline, and diamond are rarer minerals in the gravel. Of nearly all the above-mentioned minerals there are water-worn, freshly broken, and perfect crystals. The chalcedony, also, and jaspers occur in both the worn and the unworn state.

It has been suggested by the late R. N. Rubidge and others that large areas of this part of South Africa have been at a long distant period (and yet recently as regards geological time) covered with alluvia, derived from the operations of water and weather on the vast region drained by the head-waters of the Orange and Vaal, and now represented in part by the great Quathlamba or Draakenberg. This mighty range and its southern spurs supply by far the majority of the sources of the present Orange River system, and still yield to the upper valleys agate gravel in abundance, from their amygdaloidal volcanic rocks. To the north, however, in the Transvaal, some of the head-waters of the Vaal (Ky Gariep) rise in the Gats Rand and other mountains, which consist of a different rock-system. To these allusion will again be made when we consider the probable origin of the diamonds. (See my paper on the Geology of the Diamond-fields of South Africa-"Geological Magazine," February 1871-for technical details and. full references to published papers and other materials.)

The ancient alluvial plains, above alluded to, were probably terraced, according to our authors, by the successive subsidences of lake and river; and here and there they were at times coated with beds of calcareous tufa, derived from aggregations of fresh-water and land shells; and this still lies thick on many parts, and serves as a source of lime to the Colonists of the Interior. The work of natural denudation, or the remodelling of the surface by rain and rivers, progressed, whether aided or not by ice-action (as suggested of late by Mr. G. W. Stow for the surface-modification of the Stormberg and Queenstown country further south), and ultimately the levels of the present

drainage-system were arrived at ; portions of the old high-level alluvium being left undisturbed on those stable portions of the valley-floor that resisted the wear and tear of denudation. this be so-and it is likely enough-the precious gems brought to their places by the old system of drainage of the former high-level flats still lie amongst the remnants of local débris and ancient drift on the Kopjes, which remain like the navvy's "dead-men" of excavated plains, and in the fallen talus and running sand and gravel of their slopes, whilst few turn up at lower levels.

Nevertheless, there must be places further down the valley,* probably as "pockets," or patches of small extent, in or near the present river channel, into which the hard and heavy gems have been sorted (like gold-dust or tin-stone) from amongst the miscellaneous sand and gravel. Whether or no it will be worth while to cut off some bends in the river by short cuts, and work over the drained river-bed, energy and experience will probably some day prove.

But whence came the diamonds at first? And, if their origin can be traced, will it be profitable to look for them in their native matrix ?

All the world knows that diamonds, whether in India, Borneo, Sumatra, South Australia, the Ural, Algiers, California, the United States, or Brazil, are got from alluvial gravel derived from more or less distant mountains. In Brazil only have these gems been found in their native beds, namely, in a granular quartzose schist (itacolumite), and some other schists (micaceous, chloritic, talcose, hornblendic, and argillaceous) associated therewith. These, together with some accompanying limestone bands, evidently represent, in metamorphosed (highly altered) conditions, some very old sandstones, clays, shell-beds, &c., such as constitute any one formation of marine and fluviomarine deposits. The diamond crystals that occur in these Brazilian schists may also be, and indeed are always regarded as being, the results of some of the changes that have affected the strata in question; and they may represent the carbon of old carbonaceous deposits, separated and purified from hydrogen, clay, and other matters, whether within the original mass of the strata, or sublimed through pores and fissures from still more deeply seated sources of carbon.

It was suggested by the late Dr. Rubidge that the direct heat and pressure of volcanic dykes passing through coal-beds might bring about a change of hydrocarbons, producing pure carbon (= diamond), as they have changed certain coal-seams

* See the foregoing footnote.

in South Africa and elsewhere into coke, anthracite, and graphite, which are almost pure carbon. Although the chain of evidence is here incomplete, this hypothesis has ardent supporters in the Cape Colony, inasmuch as the trap-rocks above referred to as constituting a main portion, if not all, of the Kopjes at Klip-drift are, without doubt, of volcanic origin, whether they be dykes or outspread masses, and have passed through the fissured strata that here and there in the gullies are seen to lie beneath, forming the yet lower foundation of

the district.

The low-lying strata, moreover, are known to be a part of the great stratified formation that crops out along the hill-sides south of the Orange River basin, in the Colesberg, Smithfield, Harrismith, and other districts to the south and east, and which, indeed, constitutes all the Interior of the Colony within the circling ranges of Namaqualand, the Bokkeveld, Zwartebergen, Winterhoek, and Zuurbergen, ending near the mouth of the Great Fish River in Albany, on the south-eastern coast. Within this great area the nearly horizontal and probably lacustrine (Triassic) strata, first examined, defined, mapped, and described by the late A. G. Bain, spread far and wide; and their geological name, "Karoo Formation," is derived from the Great Karoo Desert, which is a characteristic feature of a considerable tract in the Worcester and Beaufort Divisions. They are also known as the Dicynodon strata,' on account of the prevalence of the remains of that remarkable two-toothed reptile in some of the beds. The Karoo Formation consists of an enormous series of shales and sandstones, accompanied by some calcareous bands, and rich at places with the wonderful remains of the above-mentioned reptile and many others; also with fishes of the palæoniscan type, together with seams of lignite and coal, remains of coniferous trees, ferns, and other plants. Throughout their whole extent these Karoo strata are crossed by frequent dykes of doleritic, dioritic, and syenitic trap-rocks, at different angles, and are often overlain by, or intercalated with, similar igneous rock. Here, then, are some of the elements required in Dr. Rubidge's hypothesis of the formation of diamond from coal by volcanic interference; but direct proofs are altogether wanting. We see here also the reason why many of the Colonists, who have read geological books, and observed something of the structure of their country, are so ready to suppose that the diamonds are native to the spots where they are found-converted, they imagine, out of hidden coal-beds and plant-remains by the subterranean heat, of which, truly enough, the volcanic rocks bear witness.

Let us look again at the accompaniments of the diamond crystals at Klip-drift and thereabouts; and although the agate

and carnelian, the peridot, mesotype, natrolite, calcite, and probably the rock-crystal and amethyst, have been derived from the veins, glodes, and kernels of the amygdaloidal and other trappean rocks in place—and we may add the garnet too, for I have a specimen of melaphyre (?) loaded with garnets, labelled "from near the Orange River"-yet the lydite must have come from some old metamorphic rocks; and we may associate with it the hepatic mundic, the ilmenite, specular iron-ore, diopside, tourmaline, and most probably the diamond as well.

Many of these specimens, not being waterworn, could not have travelled far. There are, however, two very probable local sources for them: namely, 1st, Outcrops of old rocks, still lower in the series than the Karoo beds which lie on the floor of the valley, pierced and covered by the greenstone and amygdaloidal lavas; and such outcroppings are indicated here and there in the Orange River Free States and the country to the west. 2nd, The blocks and smaller fragments of old rocks, constituting the materials of some of the Karoo beds themselves.

We must further observe that, on the one hand, the great Draakenberg, based, no doubt, on old metamorphic rocks, such as Sutherland and Griesbach have met with on the Natal side of that range,* has supplied, and may still supply, to some of its rivers the minerals of the older series of rocks, as well as of the Karoo beds and their dykes. On the other hand, the northern head-waters of the Vaal come direct from off old metamorphic rocks, and have some diamonds in their valleys before they reach that part of the Vaal which flows over Karoo strata.

We may add that down all the valleys of the Orange Riversystem ice may have played its part, as among the hills further south, and have carried blocks and crystals for miles, and left them to be detached, unharmed, by subsequent operations of rain and rivers.

Lastly, what are the metamorphic rocks of the Transvaal and the Upper Vaal, on the one hand, and of the base of the Draakenberg, on the other? We know that "Devonian" schists, supporting an old sandstone, are present in the latter, and that similar rocks lie against still older mica-schists, marble, gneiss, and granite in the Gats Rand about the head-waters of the Vaal. And, further, we now know that the Karoo beds of the Stormberg, between Washbank and Queenstown, lie on the palæozoic carboniferous strata containing Lepidodendron, Sigil

With similar rocks on this eastern side of the range, probably diamonds may also be found in the alluvium of the valleys.