![Thumbnail: [Volume 37, 1904 page 486]](/journals/rsnz/images/rsnz_37/thumbnails/rsnz_37_00_0513_0486_ac_02.gif)
Art. LV.—On the Cause of Border - segregation in some Igneous Magmas.
[Read before the Otago Institute, 13th September, 1904.]
It has been ascertained by microscopic investigation that nearly all igneous rocks vary in composition in different parts, due to the aggregation of certain minerals in clusters, which vary in size from the smallest dimensions up to large rockmasses. Dr. W. C. Brogger and Professor J. H. L. Vogt were the first to direct attention to the fact that igneous rocks are generally more basic near their borders than in the interior of the mass. It is found that the heavier minerals that first crystallize from the magma—namely, magnetite, olivine, chromite, hornblende, pyroxene, &c.—are principally concentrated in the outer portions of the igneous body.
Many valuable deposits of magnetite and chromite occur as border-segregations, and for this reason much discussion has taken place among mining geologists as to the force or energy concerned in the migration of these minerals from one part of the magma to another. The subject is one having a profound bearing upon magmatic differentiation generally.
For years it was commonly believed that border-aggregation was due to molecular flow, in accordance with Soret's principle that molecular concentration may be caused by difference of temperature—that is, in a homogeneous solution unequally heated in different parts concentration will be greatest in the region of lowest temperature.
A. Harker* and G. F. Becker† have contended that molecular flow in a cooling magma is too slow in its operation to be the cause of magmatic differentiation. And more lately Becker‡ and T. E. Spurr,§ simultaneously and independently, have argued that the migration and concentration of the first crop of heavy minerals near the borders, to a greater or less extent, may be due to convection currents resulting from differences of temperature. But even this hypothesis does not seem quite satisfactory. In the case of an eruptive magma in finite mass confined between walls, the direction of the convection currents in the central portion would be upward
[Footnote] *A. Harker, Quart. Journ. Geol Soc., London, vol. i., 1894, p. 311.
[Footnote] † G. F. Becker, “Some Queries on Rock Differentiation,” Amer. Journ. Sci, Jan 1897, vol. iii., p 21.
[Footnote] ‡ G. F. Becker, “Fractional Crystallization of Rocks,” Amer. Journ. Sci., Oct., 1897, p. 257.
[Footnote] § J. E Spurr, “Igneous Rocks as related to Occurrence of Ore,” Trans. Aust. Inst. Ming. Eng., vol. xxxiii., 1903, p. 288.
![Thumbnail: [Volume 37, 1904 page 487]](/journals/rsnz/images/rsnz_37/thumbnails/rsnz_37_00_0514_0487_ac_02.gif)
and outward, and in the outer portion downward and inward. This cycle of flow would apparently be as likely to form central as border aggregations, or aggregations only in the outer lower limits of the magma.
Convection currents even in a very liquid magma would be very slow, except there was a constant accession of heat from below in amount greatly above that normal to the depth. Of such an accession in the case of a magmatic mass forming either a dyke or laccolite there is no evidence whatever.
I am inclined to ascribe border-segregation mainly to differences of osmotic pressure in the magma, with perhaps convection currents as a contributing cause. Osmotic pressure is a form of energy of great intensity. When precipitation takes place from a homogeneous solution of dissolved salts it instantaneously establishes a condition of equal concentration throughout the whole mass.
It may be urged, and not without reason, that an igneous magma is not a homogeneous body in the same sense that a solution of auro-potassic cyanide is said to be homogeneous. For a magma is composed of watery vapour, various gases, and a solution of solid constituents. It forms, however, a two-phase system the constituents of which are homogeneous in themselves, and consequently in a state of equilibrium at constant temperature and pressure.*
According to Van t'Hoff's law, the osmotic pressure of a substance in solution is the same pressure which that substance would exert were it in gaseous form at the same temperature, and occupying the same volume.
[The section below cannot be correctly rendered as it contains complex formatting. See the image of the page for a more accurate rendering.]
| That is, | p = Rt/V |
| where | p = osmotic pressure in pounds per square inch; |
| R = the gas constant = 1,206 lb. per square inch; | |
| t = absolute temperature, centigrade; | |
| V = volume of the solvent containing one molecular weight of the solute. | |
| And | V = 100 M/r |
| where | M = the sum of the atomic weights of the atoms in a molecule of the dissolved substance; |
| r = the strength per cent. of the solution. | |
| Therefor | p = 1206 (t° + 273)r/100 M |
| or | p = 12r (t° + 273)/M |
[Footnote] *Professor W. Ostwald, “Elements and Compounds,” Faraday Lecture delivered before the Fellows of the Chemical Society in the Theatre of the Royal Institution, London, 19th April, 1904.
![Thumbnail: [Volume 37, 1904 page 488]](/journals/rsnz/images/rsnz_37/thumbnails/rsnz_37_00_0515_0488_ac_02.gif)
For example, the osmotic pressure of a 1-per-cent. solution of auro-potassic cyanide (AuKCy2) at a temperature of 20° C. is as under:—
| Here | r = 1.0 |
| and | M = (197 × 1) + (39 × 1) + (26 × 2) = 288 |
| then | p = (12 × 1) (20 + 273)/288 |
| therefore | p = 12.21 lb. per square inch. |
The osmotic pressure for a 4-per-cent. solution is 48.84 lb. per square inch, and for a 10-per-cent. solution 122.1 lb.
If we regard a molten magma as a mass of rock-material in solid solution, it is manifest that the separation of the first crop of minerals will result in a disturbance of the condition of equilibrium—for we know that the osmotic pressure varies directly as the concentration—and osmotic energy will at once exert itself to again establish a state of equal concentration throughout the magmatic solution.
The temperature at the borders of the magma will be less than that in the interior; and, since osmotic pressure varies directly as the temperature, it follows that the osmotic pressure will be less at the borders than in the interior. But osmotic pressure holds good for the Boyle-Henry laws and the law of Avogadro, and must therefore hold good for the laws of thermo-dynamics.* Hence there will be a transference of osmotic energy from the interior to the borders. The minerals first crystallized will be inert, and, being unable to offer ionic resistance, will be carried towards the cooler parts of the magma—that is, towards the borders.
Manifestly this unequal struggle between different potentials of osmotic energy will continue so long as the difference of temperature exists; but a point will be reached where the pressure will be neutralised by the increasing viscosity of the magma.
The minerals which first crystallize are generally basic, and these being carried outward as they form necessarily leaves the interior more siliceous or acid than the borders.
Further, it seems not improbable that molecular concentration,† in accordance with the Ludwig-Soret principle, may be the result of osmotic pressure due to unequal temperature in a homogeneous solution.
[Footnote] * Lord Rayleigh, “The Theory of Solution,” Scientific Papers, Cambridge Press, vol. iv., 1903.
[Footnote] † J. H. van t'Hoff, “The Role of Osmotic Pressure in the Analogy between Solutions and Gases,” “The Modern Theory of Solution,” New York, 1899, p. 21.