would imply stretching and consequent lowering of temperature, a circumstance favourable to local solidification. Who shall say that in the later and feebler struggles of this kind, as secular cooling went on and the magma approached nearer and nearer to the conditions required for consolidation, some of these tidal waves may not have become in situ sufficiently rigid to outline some of the earliest lines of elevation? *

The term 'Augengneiss' has a definite use in German petrography, and is properly used to denote those gneisses "in which single large felspar-masses (orthoclase), in form ranging from the blunt lenticular to the approximately spherical, occur, segregated from the matrix of the rock which possesses a schistose (schieferig) or coarse grained (flaserig) structure, and around which the mica-laminæ adhere." (Credner, op. cit., 6th ed., p. 101.) Such a structure could easily be developed in parts of a slowly-cooling siliceous magma moving differentially under pressure (shearing); and the difficulty of conceiving the same segregation-structure as developed by the deformation of a solid rock-mass under pressure seems to leave us very good grounds, so far as this point goes, for preferring a primitive and general plutonic origin for such a structure to dynamic action resulting from pressure applied on a ' regional' scale. Of course in either case dynamical action occurs: the crux of the question is as to the origin of such dynamical action and the conditions under which it acted. The introduction of the qualifying word 'dynamic' into the discussion of the structures of the (so called) metamorphic rocks, as between the 'mechanical' school on the one side and the plutonists on the other, leads to nothing but ambiguity.

The habit of speaking of " eyes" in a crystalline rock has however with some writers in this country transgressed considerably the limits of the abovequoted definition of Augen-gneiss, till at last it has become difficult to say that with them even a sheared portion of an intrusive granite, with its included and modified fragments from the adjacent rocks, may not be comprehended under the term.

Orthoclase is probably the embryonic silicate of the terrestrial lithosphere; and it is conceivable that in parts of the primordial magma which were under little or no strain from differential movement of any kind, such felspathic nuclei as are seen in Augen-gneiss proper may have formed by simple segregation, that segregation being arrested when differential movements, such as those which a slow-moving tidal wave would require, began to develope that metataxic process, which, with the different rates of solidifying of the minerals of the residual magma, resulted in the structure which we call 'foliation.' Again, if we extend the term 'Augen-gneiss' so as to include those gneisses in which lenticular masses of a more basic rock (e.g. amphibolite) occur,† and admit (as we must on general grounds) the possibility of local and accidental concentration of the heavy bases in the primitive magma, and bear in mind the greater rapidity with which highly basic slags are known to solidify, we can explain the more pronounced foliation of the acid matrix of the gneiss, by the theory here put forward, in such a way as to involve no mystery, and to conflict with no known laws of nature or observed phenomena.

* Ridges thus formed would have been planed off by subsequent oceanic tidal action. May we not however here find a clue to the rhyolitic-structure frequently observed in the included felsitic fragments of the Cambrian conglomerates? (See App. ii, Note O.)

+ e.g. in the grey gneiss of Elterlein. Credner, op. cit., (6th ed.) Fig. 133,

In such a case as that here cited the feeble foliation of the amphibolite tells us that the mass was undergoing slow differential movement when the individualization of its minerals commenced.*

I would suggest one point more, and that is the possibility that we may come ultimately to associate the feeble foliation of the fundamental gneiss, where it has not been interfered with by mountain-building processes, with the earliest solar tidal waves, and the more pronounced foliation of the archæan schists with the subsequent lunar tidal waves of the magma. Even those appearances which simulate ‘false bedding' in them, on which several recent writers have laid considerable stress, and the not infrequent interstratification of gneiss with schists, may perhaps be partly accounted for in this way. The subject is worthy at least, I venture to think, of some attention. In the developement of foliation and its allied structural characters in the way here indicated in the fundamental gneisses and schists, the chief difference between them and similar rocks, in which in some cases perhaps the general direction of the foliation enables us to connect it causally with lateral pressure operating in the mountainbuilding process, would be that in the former case we should expect to trace effects due to the pulling-out, and in the latter effects due to the squeezing-out of the mass. our present point of view petrology would seem to be relieved of the incubus which has been imposed upon it in some quarters, of having to furnish a theory to account for deformation on a 'regional' scale, quâ production of foliation by pressure acting upon previously solidified rocks.†

From

*With a still more liberal use of the term 'Augen-gneiss' we might include in it cases where, as the outer zone of the lithosphere solidified through the dissipation of energy by radiation, the tidal movements still continuing, huge masses, (even 'regional' masses) of the solidified outer 'crust' would slide over the viscous magma beneath, as the strain produced by tidal movements in the latter caused huge fractures here and there in the crust. As the magma grew more and more viscous, it is certain that under such circumstances large fragments of the thin crust would be torn away, included in the magma, and transferred in some cases to considerable distances. This seems the most natural explanation of the striking facts observed by Lawson in the Laurentian-gneiss, and described by him in his Essay published by the International Geological Congress, (1888). Facts however such as those described by Lawson are not new to science. Facts of a similar nature were described by Macfarlane more than twenty years ago; see his paper On the Geological Formations of Lake Superior,' (Canadian Naturalist, May, 1867).

Then as now the tidal action would vary, (within much wider limits however) with the relative positions of the Sun, the Earth, and the Moon, the maximum effect being produced when the Moon was 'in meridian.' Pfaff (Allgem. Geol. als. ex. Wiss., p. 188, et seq.) has discussed the action of tidal

In Sec. ii we have regarded the earth from the point of view of thermal chemistry and physics as passing through a stage in the history of the evolution of its kooμos out of its original non-differentiated xaos, when such physical conditions (temperature and pressure) prevailed universally at its surface in relation to the fusibility of the different minerals then being deposited to form a 'crust' as now prevail in the glacier-zone of a mountain-range relatively to the liquefaction of ice.

*

We can recognise a real distinction between granular, vitreous, and crystalline ice; and the metatropic and metataxic changes by which a glacier is developed through the stages of snow, Firn (névé), and that crystalline-granular form which constitutes glacier-ice' (with its more or less slabby stratification and frequently subordinated banded structure with white air-charged and blue air-free lamina) have been so fully studied during the last quarter of a century by some of our foremost physicists (Helmholtz, Tyndall, Forbes, and others) that we have little difficulty in referring the peculiar structure of glacier-ice to unequal rapidity of motion under pressure of the particles of different parts of the mass of the glacier. To put the matter more definitely: since equal forces must do equal work in equal times,

if w=the work done upon a yielding mass,

t = the tenacity of cohesion of the mass,

and d = the amount of change of form produced by the action of any force p;

then for any value of p we have in unit of time

w=dxt;

that is to say d varies inversely as t. As in the glacier two forces are mechanically at work-the constant action of the earth's gravity resulting in a constant strain of the mass downwards, co-operating with the minor movements produced by alternate expansion (by day) and contraction (by night); so in the case of the semi-fluid magma out of which the earliest solid crust' was formed, we have in this section suggested the action of two mechanical forces-the constant tug of the moon's attraction tending to produce a tidal wave

movements in the magma upon the earliest rind of the Earth, initiating the first permanent inequalities upon its surface. If the archæan schists (taken as a whole) represent this first-formed rind, their materials as they accumulated by precipitation from the heavy atmosphere being bathed through and through with H2O in a highly-superheated condition, we seem to have at once an explanation both of the frequent recurrence of gneiss (in a subordinate degree) among the schists as a result of tidal movements, and of those lithological characters by which they have misled the Neptunists into regarding them as 'sediments.'

*See App. ii, Note F.

after it, with a consequent constant strain in the direction of the motion of that wave, and a series of minor tides due to the attraction of the more distant sun. In the case of the glacier we certainly have to take into account liquefaction under pressure and regelation; but this we have to do by way of explaining the viscous flow' of the mass. The analogy drawn above is between the behaviour of a body such as glacier-ice possessed de facto of the yielding property which enables it to flow,' and that of a viscous mass of rockmaterial. The banded structure of both may be explained by reference to the same physical laws.

For the banded structure of fundamental rocks—whether on a large scale or that smaller scale which we see in 'foliation' (in the strictly-limited sense of the latter term)—we thus come to recognise a complex result arising out of the co-operation of a complex series of causes; the complicated metataxic changes being connected with such conditions as—

(a) inequality of distribution of thermal energy;

(b) variations in the rate of deposition of the original materials;

(c) variations in the fusibility and consequent yielding property, and in the specific gravity of the minerals resulting from primary paramorphic changes;

all these acting concomitantly with a general dissipation of energy tending to make the magma approximate nearer and nearer to the resultant condition of a solid crust. If we regard any one great class of foliated rocks-the gneisses, for example-and the great difference in the degrees to which their foliation is accentuated, from that feeble degree of foliation which makes it difficult to distinguish it in a handspecimen from a normal granite, to such a marked foliation as may be seen in some of the gneiss of the Alps,† we can hardly avoid the conclusion that pressure exerted subsequently by overlying palæozoic strata has contributed its quota in some cases, before the materials of the gneiss were completely solidified.

If we eliminate the large factor of paramorphic change (which is essentially chemical) and confine our attention to

* The behaviour of highly acid slags from the blast-furnace as compared with that of more basic slags from copper and lead furnaces, the former flowing sluggishly and solidifying slowly, while the latter flow quickly and harden suddenly, has a most important bearing upon the question of foliation on a regional scale. This was pointed out more than twenty years ago by Macfarlane. (Canadian Naturalist, loc. cit., 1864).

+ In some of the crystalline rocks of the Engadine (e.g.) composed of felspar, mica, and quartz, I have observed an unusually distinct foliation.

metatropic (physical) and metataxic (mechanical) changes, it is not difficult to draw a parallel between the process by which we have conceived the vast thickness of crystalline rocks of the primordial 'crust' with their coarsely-banded (slabby) and their finely-banded (foliated) structure to have been developed out of the earth's original nebulous mass and the process whereby a crystalline rock-mass of glacier-ice with its banded structure is known to be derived from the water-vapour suspended in the present atmosphere. To the geologist who refuses to look at phenomena of the class generally called 'geological' in the light of physical and chemical ideas all this must read rather like romance than sober science. It is not for such that it has been written.

§ v. HYPERPHORIC CHANGE.

As a typical process under this head let us consider dolomitization. A limestone containing originally but a small percentage of carbonate of magnesia is in course of time so altered in the proportion of its constituents that it becomes more and more dolomitic, and in some instances is at last converted into a true dolomite. This must occur in one of two ways; either more and more carbonate of magnesia is deposited within, or carbonate of lime is removed from, the rock-mass. The former process would be accompanied by an increase, the latter by a decrease, in bulk of the rock-mass. Observations in the field afford abundant evidence of decrease of volume as a concomitant of dolomitization. What has the chemistry of the laboratory to say to this? Its answer is twofold:

(1) Experiment shows that water containing free carbonic acid dissolves a certain amount of carbonate of magnesia, but that this is insignificant as compared with the amount of carbonate of lime which the same quantity of equally carbonated water can dissolve.*

(2) Thermal chemistry leads us to consider the relative stabilities of the two carbonates, and reveals to us the fact that carbonate of lime begins to undergo dissociation with only a moderate elevation of temperature (having the least stability of all the carbonates of the alkaline earths), while carbonate of magnesia is distinguished from most artificially* See Roth Allgem. u. Chem. Geologie, Bd. I, pp. 70 80.