ON THE MIGRATION OF MATERIAL DURING THE METAMORPHISM OF ROCK-MASSES.

THE researches of numerous geologists during the last two decades have placed at our disposal a large amount of information respecting the metamorphism of rocks, and from the facts thus collected we are now in a position to draw conclusions which we may expect to have a wide application. The important changes that affect the character of rock-masses divide roughly into two classes.

First, there are those dependent on meteoric agencies. These changes, though not necessarily superficial in the ordinary sense, are due in the first place to the action of circulating waters in communication with the atmosphere, and as a rule they involve the addition or subtraction of various ingredients or the transference of material from one place to another. The ordinary "weathering" effects illustrate the removal of alkalies and silica, the addition of water, oxygen, carbonic acid, etc. We must also include the processes which have given rise to many crystalline limestones and quartzites, serpentine-rocks, dolomites, ironstones, and jaspers, and even (as appears from Van Hise's researches in the Penokee region) some mica-schists and finegrained gneisses. The characteristic of almost all these transformations is that they are metasomatic as well as metamorphic.

Secondly, we have those transformations more usually understood by the term metamorphism: viz., dynamic metamorphism, due to high pressure operating upon rock-masses, and thermal metamorphism, due to high temperature, whether produced by an intrusion or by the mechanical generation of heat. In these various cases of metamorphism proper, metasomatism is rather the exception than the rule. I shall deal here with thermal metamorphism only, and shall draw my data chiefly from the rocks surrounding the large igneous intrusions of the English Lake District, investigated by Mr. Marr and myself, but the conclusions are confirmed in other areas.

Metasomatic changes are known to take place during thermal metamorphism as regards the volatile constituents of the rocks. affected. A (usually partial) loss of water and the elimination (under proper conditions) of carbonic acid from carbonates are instances of this; a more special case is the accession of boric and hydrofluoric acids near the contact of metamorphosed rocks with certain acid intrusives. Several observers have recorded a transference of other materials (silica and soda) from an invading igneous magma to the neighboring rocks, but such a phenomenon seems to be of uncommon occurrence, and to be confined to the immediate vicinity of the contact. Apart from the exceptions noted, there is every reason to believe that thermal metamorphism involves no alteration in the bulk-analysis of the rocks affected. Whatever part water may play in the various chemical changes that are set up, it does not (as in atmospheric metamorphism) act as a medium to transfer material to or from the rocks in question.

I believe that we can go further, and assert that within the mass of a rock undergoing thermal metamorphism any transference of material (other than volatile substances) is confined to extremely narrow limits, and consequently that, for a given temperature of metamorphism, the mineral formed at any point depends only on the chemical composition of the rock-mass within a certain very small distance around that point. Illustrations of this principle, as stated in the latter form, are familiar to all who have studied cases of "contact metamorphism" they are very striking when some of the constituent substances of the original rock were, by weathering or otherwise, locally aggregated prior to metamorphism. By studying such cases we can not only verify the principle here laid down, but also arrive at an estimate of the actual limits within which interchange of material has taken place.

An excellent test-case is afforded by rocks containing calcite. It is well known that impure calcareous rocks are readily metamorphosed by heat into rocks rich in lime-silicates, with total 'Compare Bull. Geol. Soc. Amer. (1891) vol. iii., pp. 16–22.

elimination of the carbonic acid, while pure limestones or dolomites, under the same conditions, merely recrystallize without chemical change. In other words, the carbonates are decomposed in thermal metamorphism only in the presence of silica in some available form to take the place of the carbonic acid. Interesting illustrations of this are given by some of the rocks which have come under our notice. The Strap granite in Westmoreland metamorphoses certain basic lavas containing amygdules of various dimensions, many of which were occupied, prior to the metamorphism, by calcite. Near the granite the smallest of these calcite-amygdules are converted into various silicates rich in lime, the silica having been derived from decomposition-products lining the original vesicles or from the immediately adjacent portion of the rock. In the larger metamorphosed amygdules, on the other hand, only the outer layers are transformed into lime-silicates, the interior still consisting of calcite; which, however, has recrystallized during the metamorphism, as is proved by its moulding the silicates and being penetrated by needles of actinolite, etc. Analogous appearances characterize veins and lenticles of calcite in shales and the converse case of argillaceous nodules imbedded in pure limestones and dolomites. The conclusion is that carbonic acid is displaced from the calcite only when there is in the immediate neighborhood either free silica or some substance capable of furnishing silica. Where calcite and quartz have recrystallized side by side in a metamorphosed rock, they are always separated by some one or more lime-bearing silicates, but their distance apart may be very small, and we deduce that the migration of silica to take the place of carbonic acid has been restricted to extremely narrow limits. In some highly altered rocks the distance is not more than onetwentieth of an inch.

The limit of migration of material no doubt increases with the temperature of metamorphism. This is well illustrated by some calcareous ashes or tuffs. At a considerable distance—say a thousand yards—from a large granite intrusion, the carbonic See especially Quart. Journ. Geol. Soc. (1893) vol. xlix., pp. 359–371.

acid is entirely expelled only from very fine-grained mixtures of calcareous and ashy materials: approaching the contact, the complete decomposition of the calcite is found to extend to successively coarser-grained rocks. Another line of inquiry is offered by the texture of the metamorphosed rocks themselves, of whatever lithological nature, in a district of metamorphism surrounding a large igneous intrusion. The size of the individual crystals of secondary minerals increases towards the contact with the intrusive rock: this may be taken to indicate that the migration of material within the mass of a rock undergoing metamorphism has more latitude when the temperature is higher. For various reasons, however, it would be unsafe to found numerical results upon such observations. The crystals of certain metamorphic minerals attain to considerable dimensions by virtue of their power of enclosing a large amount of foreign material; others, again, can apparently push aside solid impurities to make room for their own growth. The texture of the metamorphic rocks examined is still, however, in general accord with the conclusions reached by other methods of inquiry.

The question naturally arises whether the limit of migration of material is the same for different substances. On this point we have but little information. Among the various types of "spotted" rocks described in aureoles of metamorphism is one in which the spots are simply spaces free from the secondary brown mica abundant in the general mass of the metamorphosed rock. Since the iron compounds in the rock must originally have had a generally uniform distribution, the phenomena of the spots indicate a movement of ferrous oxide, and the radius of the spots gives a measure of the extreme limit of such movement. In the cases examined this is about one-twentieth of an inch, and we may infer that the greatest distance of migration of ferrous oxide has been about the same as that of silica at a similar tempera

ture.

Not to insist unduly upon precise estimates, these and similar observations certainly tend to show that in thermal metamorphism no interchange of material takes place except between

closely adjacent points. The law that, apart from volatile constituents, the total chemical composition remains unchanged is true not only of the rocks in bulk, but of any individual cubic inch of the rocks. This might be followed out into various corollaries, of which I note only one, viz., that the greatest variety of metamorphic minerals is to be found in rocks which were the most heterogeneous prior to metamorphism. Such rocks are breccias and fault-breccias, etc., and especially basic igneous rocks more or less weathered before being metamorphosed. ALFRED HARKER.

CAMBRIDGE, ENGLAND.