The Geologist/Volume 5/Notes on the Metamorphosis of Rocks in South Africa
NOTES ON THE METAMORPHOSIS OF ROCKS IN SOUTH AFRICA.
By Dr. R. N. Rubidge, of Port Elizabeth.
It is near eleven years since that in travelling through Howison's Poort,[1] one of the most picturesque of the many fine mountain passes through the quartzite ranges of the eastern province of the Cape Colony, my attention was drawn to a geological fact to which observation in other parts of the Colony has since led me to attach no little importance. In the construction of the main road from Port Elizabeth to Graham's Town, many deep cuttings have been made in the solid quartzite rock. In many instances the rock seen in these works lost its crystalline character gradually, and assumed that of a hard blue sandstone, and at length nearly resembled the blue fossiliferous shales and sandstones of the Ecca.
These quartzite rocks have been referred to the age of the Carboniferous formation of Europe by Mr. Bain (Geol. Trans. vol. vii. 2nd series, pp. 54 and 183), and both he and Dr. Atherstone ('Eastern Province Magazine,' vol. i. p. 588) describe them as conformable with the slaty rocks of the district. I have no doubt whatever that they generally are so. They pass gradually into each other, and, as I have described, the quartzite traced downward loses much of its siliceous character, and gradually assumes that of the slate and of the Ecca rock. This last is by Mr. Bain dissociated from the Carboniferous formation, and made the lowermost of the Lacustrine or Karoo series, but I have the following reasons for differing with him:—
1. At the western entrance of Howison's Poort are some beds of rock, intermediate in lithological character between the quartzite and the Ecca beds. These contain vegetable stems which have been recognized by many as identical with those of the Ecca. At Forester's Farm, east of Graham's Town, is a blue rock, just like that of the Ecca, containing the same fossils, which passes gradually into the gneiss. The sandstone on the one side is in relation on the other with the claystone-porphyry of Bain, as is the rock at the Ecca. Near Salem, in the heart of the Carboniferous system of Bain, are similar rocks with like fossils, conformable with the quartzite.
2. The strike of the inclined rocks, quartzites, slates, and Ecca rocks is throughout the province north 60° west nearly. If we draw a line of eight miles through Graham's Town and near Salem, at right- angles to this direction, it will pass through little but quartzite. If we draw a line of the same length through the Commadagga beyond the Zeurberg, it will pass through nothing but slates, Ecca rock, and clay stone porphyry.
3. On the road' to Graaff Reinet is a place called Wolve Krool. It is a plain, bounded by quartzite hills. Its section is this:—
Fig. 1.
Here the Ecca rock contains its characteristic fossils, is conformable with the quartzite, and is separated from the Dicynodon rocks by a highish mountain of quartzite and many miles of slate, porphyry, etc. I could add many other reasons for this belief, but I think these will be sufficient. What is then the true relation of the quartzite to the Ecca rock and the slates? and how is it that at one part of a line of strike the rock will be all of a blue slaty fossiliferous character, and at another all crystalline quartzite, destitute, or nearly so, of fossils? How is it that in deep sections, natural as well as artificial, such as are made by cutting roads or by deep gullies, the slaty rocks are found below gradually passing upwards into quartzite? Of this I could give scores of instances, but I will select only one natural one. The range of quartzite on the left-hand of the section is crossed by a bye-road. This road passes for a mile, or more over well-marked Ecca rock, with the high quartzite hills on either hand. The quartzite on the right-hand dies out, and the road to Graaff Reinet passes over Ecca rock in the corresponding part of the section.
I found what I believe to be the key to the explanation of these facts in Namaqualand. In passing through Ezel's Poort, between Springvontein and Spectakel, I was shown a section which had been noted by Dr. Atherstone as remarkable. The gneiss hills were covered by horizontal layers of quartzose sandstone, and these were continuous on the western side of the hill with like quartzose sandstone dipping at a high angle westward, conformably with the gneiss. It was clear that this change of dip was not due to any upheaval, for the horizontal sandstones were found undisturbed a few yards distant. I soon learnt to regard this juxtaposition of horizontal and inclined beds, this continuity of quartzite conformable and unconformable with its subjacent rock, as a normal state of things in Namaqualand. When I saw high mountains with like structure, I was at first a little staggered, but soon felt convinced that even on this scale the phenomenon was due to the assimilation to each other by a process, common to both, of rocks of widely different ages.
In the Western Province I made, in a rapid journey from Cape Town to Ceres, a selection from the clay-slate to the Upper Silurian of Bain. I had reason, as far as I was able, to confirm the truth of Mr. Bain's section, while differing from him in the inference I drew from it. I believe his wide dislocation of the Ceres beds from the clay-slate to be an error into which he has been led by a state of things like that of Ezel's Poort. I have never been able to get direct proof that this is the case here, although I have elsewhere, as shall presently appear.
On my return to the Eastern Province, I thought I saw evidence of the siliceous change of rocks on a much greater scale than I had observed them in Namaqualand. T wrote a Paper on the subject, and published it in the local magazine I have quoted above ('Eastern Province Magazine,' vol. ii. p. 187). I hoped it would have led my friends here, from whose sections mine differed considerably, to re- examine their data. A little after, I sent home a Paper which was read at the Geological Society of London (see an abstract of it in the 'Geological Society's Journal,' vol. xv. p. 195), in which I explained these views, and predicted that the clay-slate of the west would hereafter be found identical with the Upper Silurian of Bain, and the Carboniferous rocks of the east identical with both, the quartzite being changed rock, sometimes slate itself, sometimes a newer unconformable rock. Of this identity I was enabled to send home strong presumptive proof in the shape of fossils identical with the Upper Silurian of Bain, from the clay-slate on the western shores of Francis Bay. More recently I have obtained the same fossils (pronounced Devonian at home) from various points between the Kromme and Kabeljouw rivers, St. Francis Bay, in the clay-slate, and from Chatty, near Port Elizabeth; from Naroo, near Uitenhage; from Blauw Krants, on the Bezuidenhouts river, on the road to Graaff Reinet; and from the northern base of the Coxcomb in the Winterhoek range in the Carboniferous. Still, it might be objected that there may really be a difference between the clay-slate and the Devonian, though Mr. Bain may have mistaken the line of division. If reference be made to the Admiralty chart of St. Francis Bay, it will be seen that the low shores of the bay are crossed by a range of mountains of considerable elevation. These mountains, which are quartzite, cross the strike at a considerable angle, nearly, in fact, for some distance at a right angle; so that on the beach and the low hills you may cross near ten miles of slate, perhaps five miles of strike, while six or eight miles inland, on the heights, the corresponding part of the section is all quartzite. The quartzite must, consequently, cross unconformably the slates, and therefore be newer than they. The reasons why they cannot be older, I need not give here, as I have given many of them above. These same quartzite hills are continuous with others of the same lithological character, which are decidedly conformable with the Devonian rocks, though they too cross the strike at an angle of 30°. I have not had opportunities for such an examination of the country between this and Cape Town, as to enable me to say positively that there are no beds older than the Devonian; but I think I have shown satisfactorily that the evidence on which the clay-slate is referred to a much higher antiquity is fallacious. I can safely assert that the Devonian beds of this country are crossed by lofty ranges of quartzite, often unconformable with them, which quartzite is continuous with like quartzite conformable with the Devonian beds. Whence I infer that the rocks of a tract of country may he so altered by molecular changes common to all (probably in the instance of our rocks the infiltration of silica), that beds of widely different ages may present the same lithological character, and that when horizontal quartzose (or calcareous[2] or felspathic?) rocks are continuous with inclined rocks of the same kind it is not always safe to infer that beds resting conformably on the latter are much newer formation than those on which the sandstones rest unconformably, that the beds a, b, are very far older than c, d, for instance.
Fig. 2.
It is my conviction then (though I admit that my evidence is not quite conclusive) that the inclined slaty rocks of this Colony, west as well as east, all belong one formation, which geologists at home have, on the evidence of fossils, pronounced to be Devonian; and that the quartzite is a rock which has undergone a superficial change, and may therefore be called metamorphic. This siliceous metamorphosis is associated with other changes. The clay-slaty beds are often converted into ochry, micaceous, and chloritic schists.
There is not in the Eastern Province much evidence of ordinary metamorphic action, except in the claystone-porphyry of Bain, which I regard as a product of metamorphic action, as I shall more fully explain hereafter. At the Matland mines, about twenty miles west of Port Elizabeth, are slates like those which at Chatty contain Devonian fossils. Some of these have been converted into chloritic, hornblendic, and micaceous schists, without any evidence of the proximity of eruptive rocks. In the planes of bedding of these schists are veins of quartz, and occasionally carbonate of lime, not very rich in copper-pyrites. I regard the hard blue crystalline limestone of the same locality, in which lead and zinc ore occurs, as partially, at least, metamorphic. At George and other places intermediate between Cape Town and here, granite occurs, but as I have had no opportunity of examining it, I shall trace the evidences of metamorphic action from Cape Town northward.
At Cape Town I found granite-veins varying from one to three feet to as many lines diameter running parallel with the strike of the clay-slate rocks without displacing them, showing, I think, that they had been changed in situ. Other veins crossed the strike. Again, isolated masses of slate preserved their dip unaltered in the midst of granite which appeared to have a dip in the same direction. Passing north-westward towards Namaqualand, I saw the slate near Heer- lozement so little altered and so like some of the fossiliferous rocks of the Eastern Province that I much regretted that my engagements did not permit of a closer examination of it. At Olifant's river the rocks, Still with the same strike as in Cape Town, viz. nearly magnetic north (north 30° west), had assumed a micaceous and talcose character, and on the northern bank of the river were much impregnated with iron. Four or five miles beyond Kokonap I saw the slate for the last time till I met it at the Orange river, and here it abounded in a peculiar form of cyanite which I afterwards found in great abundance in the gneiss and mica-schist of De Kiet, near Hondeklip Bay. Some grassy country intervened between this spot and the next where rocks were visible. These were felspathic in great variety. I could not get a satisfactory observation of their dip for some days' journey, perhaps owing to the little experience I had then of rocks of this class. There are few things I have more to regret in the way of lost opportunities than the want of a careful examination in detail of the country within ten miles' radius of the lowermost ford of the Olifant's river. It would include a section from the clay-slate to the Upper Silurian of Bain which are found in the Cederberg as well as the passage of the former into the felspathic rocks of Namaqualand. Bain has no hesitation in affirming this change, and I have every reason to think that he is correct; but believing as I do in the identity of his clay-slate and the Upper Silurian, I cannot but regret that I was unable to make a thorough examination of the country. I believe Bain's separation of the clay-slate from the Upper Silurian (Devonian) are drawn here as elsewhere from the position of the quartzite crossing the slate and underlying the Devonian. Is not this evidence identical with that on which metamorphic formations are assigned to widely distant epochs in Europe?
In addition to the want of time and of experience referred to, I have to regret the loss of a note-book in which my observations on the rocks in the earlier as well as later part of my journey in Namaqualand were inserted. I cannot therefore tell from my own observations how the strike of the rocks which was north 30° west at Olifant's river, assumes a nearly east and west strike at Springbok Vontein. As we pass northwards it takes a more northerly direction, and at Oograbis it is north 60° west, and at Annies, on the Orange river, it resumes its north 30° west strike with its slaty character. I have no hesitation in affirming the passage of the slate into felspathic rock here.
Assuming, then, the metamorphosis of palæozoic rocks into gneiss, mica-schist, etc., I will merely reiterate my firm belief that those of Namaqualand are the changed condition of the great mass of slaty beds which extend from the mouth of the Fish river in the east to Cape Town, and thence to Olifant's river, and at various points contain fossils which have been referred to the Devonian epoch by geologists of Europe. I again admit that the evidence by which I have attempted to establish this is somewhat defective, but I have shown clearly that that on which it is denied is valueless.
I have already described the interpolation of masses of granite among the slate of Cape Town without displacement: this phenomenon obtains to a much greater extent in Namaqualand; great masses of granite, with little if any evidence of stratification, pass gradually into gneiss on either side, and, in fact, all round, without change of dip. These are called locally "bosses," and their scaling off is remarkable, giving them the rounded outline, whence their name. The same thing is seen in the change of hornblende-schist into greenstone or syenite, with large crystals of hornblende. Numerous instances of this occur; one of the most striking is between Klein Pella and Oomsdrift.[3]
I have mentioned in a former Paper that the twists of the strata in which the copper-ore is deposited occur in gneiss, and when a section is seen on a hill-side no granite is visible, but when worked to any considerable depth, the rock loses its laminated character and becomes a felspathic granite or greenstone. A remarkable section was observed near Pella: a stream had worked a deep channel in the rocks; the edges of the ravine so formed were of well-marked gneiss, while the water ran ever a bed of granite without trace of lamination, the gneiss preserving the same dip on either side of the ravine. Indeed, it appeared to me as if metamorphosis of the rock into felspathic granite was the normal state below, while the gneissic lamination was a superficial indication of the old stratification-planes. While on this subject I will mention what appears to me to be a singular character of our palæozoic rocks here. The specimens I have sent home will show that all the Devonian fossils here lose every trace of their carbonate of lime. They are preserved, often very perfectly, in oxide of iron, but in my experience they are seen only on the exposed edges of the rocks, be these greatly inclined, as at Chatty and Hermansdorp, or only slightly so, as at Coxcomb and Jeffrey's bay. At Chatty I have seen a mass hollowed out in all directions by the decay of the encrinites on the edges, while tracing the same layer deeper in, it lost all trace of fossils. Frequent repetitions of this seemed to me to establish it as a rule that the fossils in the rock were only exposed by decomposition. Still it may be merely accidental. I should be glad to learn whether it is so or not.
Fig. 3.
I have stated that in the metallic twists, or saddles, I never saw granite in what I could consider the position of an intrusive rock. In one of the accessory twists which meet the metallic saddles at various angles, and which in section on a flat surface have the appearance of a feather, the shaft (a b) of the feather was composed of micaceous schist, with a few rather large crystals of felspar. I have frequently seen irregular-shaped patches of mica-schist following neither strike, nor any law that I could perceive, among the gneiss. Granite occurred in the same way in other spots.
It is well known that prehnite is a common mineral in some parts of this colony. It occurs in the dioritic or syenitic dykes of the Dicynodon strata, and is evidently a product of the re-arrangement of the minerals of these dykes during their decomposition. The prehnite is found in laminæ between the decomposing masses of the dykes and on their faces as digitate concretions. Unlike M. Daubrée's zeolites it has not required heat for its formation, but, like them, it is forming at this hour by aqueous action. I have almost as little doubt that dykes and other masses of granite are forming in the same way from the re-arrangements of the constituent minerals of the gneiss. I find it easier to feel the truth of this on the spot than to convey it to others; still I will endeavour to give some reason for my faith. One mass of gneiss, near Henkrees, in which a vein of granite, terminated by an expansion about a foot square, proceeded along the gneiss. It was quite evident that there was no intrusion, no connection with any mass below. Spindle-shaped masses of granite placed between beds of felspathic, micaceous, and other rocks, were numerous and could often be entirely removed by a hammer, showing they had no connexion with any subterranean mass. Spherical lumps of granite or syenite were frequently quite isolated in the schists of the mines. I have stated before that beds of quartzite appeared to be intercalated by a filtration from above; some of these could be traced down till they became mixed with mica and then with felspar, and not very much deeper assumed the character of the gneiss of the country. Other masses, which seemed to be intermediate between these, toward the spindle-shaped granite lumps above, were composed principally of quartz, with a few grains of mica and felspar, and occasionally garnets: these were surrounded by mica-schist in bent-up strata; yet the whole mass, perhaps twenty pounds or less in weight, could be removed. These circumstances seemed to me to prove clearly that if felspathic rocks of any kind are the products of metamorphic action, then are granite, syenite, etc., traceable to a like origin; that if, as I think, I have clear evidence, in the Maitland Mines and other places of this neighbourhood, palæozoic rocks are convertible into micaceous, chloritic, and felspathic schists, without evidence of eruptive agencies, then are the so-called igneous rocks equally so. I think it will scarcely be disputed by any who admit the re-arrangement of felspathic rocks I have contended for, that causes by which such re-arrangement was made, may have effected the original conversion of sedimentary into crystalline rocks. We have in this province a rock which has been pronounced decidedly igneous by the highest European authorities; it is the claystone-porphyry of Bain. Yet its position among the other rocks is, in many respects, just that of the quartzite; and, like the latter rock, its individual beds and its minor masses are conformable with the stratified rocks, while in ranges many miles in length, it crosses the strike of the strata, generally at an angle of about 30°. This fact was first demonstrated to me by Mr. Pinchin, a gentleman who has made many interesting observations on the geology of this country. The sections of the Zeurberg, of Yan Zonder's Plain (given above) on the Graff Reinet road, together with a somewhat imperfect recollection of that of Graham's Town, had led me to believe that the relation of the porphyry to the slate was constant. It is now evident that I was in error here. The fact of the masses of the porphyry crossing the slate without disturbing it seems to me greatly to strengthen my views as to its metamorphic origin by rendering the hypothesis of Messrs. Bain and Wylie untenable. The former gentleman supposed that the porphyry had been poured over the surface of the strata as liquid lava. Mr. Wylie referred its origin to volcanic action, producing ash, which was deposited at the bottom of the ocean, and formed this igneous-like rock with its contained granite pebbles arid fragments of rock. The fact of the direction of its masses being at an angle of 30° with the strike seems to me to be incompatible with either of these hypotheses. The ranges of porphyry, like those of quartzite, die out and reappear. The normal position of the porphyry appears to be as in fig. 4, but I have seen it placed as in figs. 5 and 6:—
Slate. Porphyry. Slate.
Fig. 4. |
Slate. Porphyry. Slate.
Fig. 5. |
Slate. Porphyry. Slate.
Fig. 6. |
The character by which we all agree to recognize this rock is the presence of masses of quartz and granite of various sizes with occasional fragments of slate and other rocks. Sometimes these masses are as much as fifty pounds in weight, at others they are so minute as to be scarcely recognizable by the naked eye. In a recent journey to Paardenpoort I met with a mass of this porphyry which terminated in a vein about a foot thick, with very minute crystals. Now what is the character of the rock among which this porphyry is interposed or interstratified? It is such that no one acquainted with the two would pretend to diagnose them, save by the presence of the crystals above mentioned. Nor would the blowpipe, or even more careful analysis, so far as I am aware, enable him to do so. If then the base of the rocks differ so little, and there is evidence that no displacement has taken place in any known section (see Bain and Wylie), is it not clear that this rock has originated in slow conversion? Yet I believe whatever may be predicated of it may be equally so of granite; for it contains granitic masses in great numbers, and often of large size: besides granite, veins occupy precisely the same position among rocks which I have given reasons for believing to be the same strata in the Western Province.
But it will be seen above that I am not disposed to admit that the evident displacement of strata is at all times due to eruptive agency. I have given instances, on a very small scale, in which I feel sure it is not so. I hope ere long to be able to show that the infiltration of quartz from above has produced this- effect, but my evidence on this subject is still somewhat defective; I will defer what I have to say on this subject till a future period.
I have had but little time or opportunity for the microscopical studies which have done so much for the views on the nature and origin of granite which I am here advocating. I should hardly have ventured indeed to have given observations so crude as my own, but for a conviction that probably no country in the world offers greater facilities for studies of this kind than does this Colony, and more especially the district of Namaqualand, which is probably barer of vegetation and more intersected by gullies than any other country in the world not absolutely uninhabitable.
I will give a brief résumé of the observations which led me first to doubt and at length to abandon the igneous theory of granite, in which I was a firm believer ere I visited the Western Province of the Colony.
1. The undoubted change which rocks have undergone into quartzite and its equally evident origin in superficial and igneous agency. Mr. Darwin admits this origin of the Table Mountain sandstone.
2. The existence of beds of granite and other rocks of felspathic bases in association with sedimentary rocks in positions which it is impossible to believe they could have occupied by forcible intrusion from below. Many veins of the claystone-porphyry exceed a thousand yards in width, yet they do not in the slightest degree disturb the strata adjacent to them. At Kleinpoort I measured the slate eighteen inches from its junction with the porphyry. It dipped towards the latter at an angle of 35°, the porphyry itself having a dip in the same direction.
3. The irregular masses of granite taking the place of gneiss and not connected with the granite below.
4. The origin of prehnite and other zeolitic minerals from decom- position of igneous dykes of the Dicynodon-strata. Prehnite, as well as of quartz, is formed thus between the decomposing "boulders" of igneous rocks. Veins of carbonate of lime are often formed in the same way. Nor can I hesitate to refer the felspathic veins and irregular masses in decomposing gneiss in Namaqualand to a like process of re-arrangement. I have there seen carbonate of lime in felspathic rocks; fluor-spar mixed with epidote and felspar; phosphate of lime with felspar and quartz.
5. I have mentioned the igneous dykes of the Dicynodon-strata. They have always been referred to plutonic agency, but it appears to me that there are great difficulties in admitting such origin. They take, I believe, every direction of the compass, vary from eighteen inches to some hundreds, perhaps thousands, of yards in breadth, and some of them are probably fifty or more miles in length; they are numerous, and occur frequently from near Somerset East to the Vool River, but never, in my experience, or that of any one I know, pass the boundary of the Dicynodon-strata, nor do they disturb the rocks through which they cut in the base.
6. I have mentioned the occurrence of granite-veins conformable with the strata among which they lie. The claystone-porphyry of Bain, appears conformable as to individual beds, while in the mass it crosses the section of the country. I have never found igneous rocks in the positions of upheaving rocks. I have repeatedly found them in positions (4, 5) where they could not possibly be so. In Namaqualand the rocks between Springbok and Concordia were perhaps more decidedly gneiss-like than in any other part of the section, except perhaps near Kok Vontein, yet I regard these two spots (the former about a mile north of Springbok Vontein, the latter two miles south-east of Kok Vontein) as the main axial lines of the country. Many facts concur to prove that whatever may have been the cause of the upheaval of strata in this country, igneous rocks have had nothing to do with it. That there are considerable difficulties about the stratification of this neighbourhood, I fully believe. That I have no clue to the satisfactory explanation of those difficulties I am obliged to confess. To mention one or two, I believe that encrinites are generally local in their distribution, that is, individual species are confined to a few beds; and that if the same species of encrinite is found in these spots, the rocks containing them may be safely assigned to the same age or near it. At the northern base of the Coxcomb[4] are some nearly horizontal beds of blue and ferruginous schists containing trilobites, shell-fish, and encrinites, pronounced Devonian on good authority. The strike of these rocks is north 60° west nearly, and this line of strike would pass through Cape Reciffe. The Chatty beds of shale, which are in hills continuous with those of Port Elizabeth, would nearly correspond in strike with these beds; and at Chatty two or three encrinites identical with those of Coxcomb occur. Yet at the former place the rocks dip at an angle of 45°. There do not seem to be any igneous rocks to account for this difference. At Naroos, near Uitenhage, the slaty beds are associated with quartzite, and dip at 60°–70°.
Again, the beds containing spirifers or this encrinite at Kabeljouw river's mouth, Jeffrey's Bay, have but a slight dip on the seashore; a little inland they have a greater dip, but at Hermansdorp, where the same spirifers or this of encrinite occur, they have a dip of 80° close to their junction with the quartzite. I cannot account for these things. I suppose no one in the present day would call quartzite an igneous or upheaving rock. Yet it is certainly my impression that if any rock in this country influences the change of dip in either rocks, quartzite does. Mr. Niven, the gentleman from whom I have the last fact, and who has done so much in throwing new light on the geology of this province, tells me that the quartzite, a hundred and eighty yards from the slate, dips 45°. If compelled to suggest a reason for these things, it would be, that whereas quartzite might be metamorphosed by addition of matter infiltrated, claystone, porphyry, granite, etc., might owe their origin to mere crystalline action under the agency of water, thermo-electric currents, etc. This last is Mr. Sterry Hunt's view, I think.