An important question, however, remains to be determined; namely, whether the remaining eight granites, amounting to 53 per cent. of the whole, are really composed of these four minerals or not.

It has been demonstrated, mathematically, that they might be formed of four such minerals, having the oxygen-ratios of these minerals as they are found crystallizing out from the mass of the Donegal granites. But is the rock really made up of only four minerals or not? It is the opinion of many skilful mineralogists, that the composition of the crystals which may be picked out of a granite or other crystalline rock may differ essentially from the composition of the same mineral when it enters into the constitution of the general rock-mass; and it is certainly not to be assumed as self-evident that a rock is composed of those minerals, and of those only, which may be seen in separate crystals, and analyzed as such.

To solve this interesting problem in the case of the Donegal granites, we must call to our aid the auxiliary equations not yet used, depending on iron, lime, soda, potash, &c., which have been grouped together hitherto, by us, in the equation containing the oxygen of the protoxides.

In fact, before we can say that the granites are really made of the four minerals whose composition is given in Table II., we must be certain that the equations of condition furnished by each constituent are all fulfilled, as well as those furnished by the oxygenratios. These equations of condition are the necessary supplement of the preceding analysis, which has only proved that certain granites might be made of certain minerals with given mineralogical formula; but which has not proved that they might be made of certain minerals of given chemical compositions.

The method of using these test-equations will be readily understood from the following discussion :

·

Granite No. I.

This granite might be formed of four minerals having the same oxygen-ratios as the minerals of Table II., but it cannot be formed of minerals having the chemical composition of those in Table II., because it fails to satisfy, inter alia, the potash test.

The quantity of potash in this granite, as appears from Tables II. and VI., should be

2.51 x 14.92

+

100

71.03 × 2.07 0.93 × 8.83
+
100
100

= 1.926;

but the quantity actually found by analysis was 4.66 (Table I.); from which it is evident that the calculated and observed amounts of potash cannot be made to agree, and therefore that this granite cannot be composed of the four minerals of Table II.

This granite also fails to satisfy the test-equations derived from iron and lime. It will be found on trial that the calculated quantity of iron, like that of potash, is too small; while that of lime will be found too great.

Granite No. V.

This granite fails to satisfy the test-equation derived from iron. From Tables II. and VI., it appears that the calculated quantity of iron-oxides is

while the quantity actually found by analysis (Table I.) was 3.33. Granite No. V. fails also to satisfy the lime test-equation.

Granite No. VIII.

fails to satisfy the potash test-equation. The calculated quantity of potash is much less than the amount found by analysis.

Granite No. IX.

fails with respect to lime and iron. The calculated quantity of lime is 41.5 x 5.3 2.74 × 0.54 +

25.45 × 2.75

+

100

100

100

= 2.913;

but the quantity of lime found on analysis was only 1.68. It also fails with respect to iron-oxides.

Granite No. X.

fails with respect to potash, soda, lime, and iron.

Granites Nos. XI. and XII.

fail to satisfy the soda, iron, and lime test-equations.

Granite No. XV.

fails with respect to potash and lime.

From the preceding discussion it is plain that not a single granite of those analyzed can be composed of quartz, orthoclase, oligoclase, and black mica, having the precise composition of those minerals given in Table II.

Of the fifteen granites, that which comes nearest to being composed of these minerals is No. IX., from Doocharry Bridge. I here give its observed and calculated composition.

No.

From the entire of the preceding, it may be safe to draw the following conclusions respecting the granites of Donegal:

1. That nearly half of them are certainly not composed altogether of the four minerals, quartz, oligoclase, orthoclase, and black mica, that are found in them in distinct crystals.

2. That the remaining half, even if they be composed of these minerals, must have a paste made of quartz, orthoclase, oligoclase, and black mica, not having precisely the chemical composition of the separate crystals.

§ II. On the Syenites of Donegal.

In many parts of the co. Donegal syenites are found, which appear to form great bedded masses, rudely stratified like the granites; and, like them, also occasionally penetrating the neighbouring rocks in dykes, making various angles with the beds of stratification. The following Table contains the analyses of such as I

have examined.

Locality.

1. Lough Anure

2. Kilrean

3. Doonane

TABLE VII.-Analyses of Donegal Syenites.

4. Precise locality unknown..

49-20 18-32 7-12 1-95 9-72 7-11 192 1·72 1.00 1-20 99-26 | 44-40 25.00 6:45 2-11 10-17 3:51 2:58 2-66 0.84 1-08 98.80 50-08 18:84 7·05 1·0312·37 6·57 2·39 0·57 0-88 0-80100-58

58-05 16:08 8-27 0-45 6:52 294 4-65 2-21| 1·12.

100-28

1. Lough Anure.-A medium-grained syenite, or crystalline greenstone, composed of small plates of black mica, with some hornblende, aggregated together; and of a felspar, which seems to be oligoclase.

2. Kilrean. A coarse-grained syenite, containing

(a.) Long crystals of green hornblende.

(b.) White felspar (?oligoclase).

(c.) Specks of iron pyrites.

In addition to the constituents given in the analysis, this rock contained 1.078 per cent. of sulphur, which, no doubt, was present in the form of the bisulphuret of iron. It is one of the toughest rocks I have ever met with. 3. Doonane Hill, near Donegal.-A crystalline, greasy trap-rock, forming a dyke penetrating the lower arenaceous Carboniferous Limestone, and then expanding into a mass on the summit of the hill. It contains

(a.) Black hornblende.

(b.) Green felspar.

4. This specimen closely resembles those found in the Black Gap, near Pettigo; it forms a connecting link between the true granites and the greenstone-syenites, as it contains

(a.) Quartz.

(b.) Oligoclase, pinkish yellow, in large crystals with brilliant cleavage. (c.) Hornblende, dark glossy black-green.

(d) Occasional facets of sphene in minute crystals.

§ III. Comparison of the Granites of Donegal with those of
Scotland and Sweden.

The granites of Scotland and Sweden resemble the granites of Donegal in being stratified and in containing the same constituent minerals.

I here give the analyses of two Scotch granites, for the opportunity of examining which I am indebted to the courtesy of Sir R. I. Murchison, K.C.B.

TABLE VIII.—Analyses of Scotch Granites.

No.

Locality.

1. Strontian.
2. Tobermurry,
Mull

62.00 17.60 4.78 0.74 4-95 3·17 4.08 3.25 0.40...100.97 70.60 16:40 1·52 0·36 2.47 1.00 4.14 4.29 0.48 ... 100-26

1. Strontian. This granite is somewhat like the Ardara granite in appearance, and also like the black gneissose granite, which found as a variety at Glen, in the co. Donegal. It is medium grained. It

contains

(a.) Quartz.

(b.) Felspar, white (oligoclase), having the characteristic striæ.

(c.) Black mica, abundant.

2. Tobermurry, Mull.-A coarse-grained granite, resembling the coarser varieties of the typical granite of Donegal. It contains

(a.) Quartz, abundant.

(b.) Pink orthoclase, large crystals (in. to

(c.) White oligoclase, striated.

(d.) Black mica, not abundant.

(e.) Sphene-facets, occasional.

in.).

It remains to be considered how far the analyses of the Scotch granites admit of being represented by a mixture of quartz, orthoclase, oligoclase, and black mica. In order to solve this question, it is necessary to calculate their oxygen-ratios, and apply equations (4) to them.

Locality.

TABLE IX.-Oxygen in Scotch Granites.

Strontian

32-192 8-227 1·432 0-253 1·407 1.239 1047 0.551 Tobermurry |36-657 7-6660-455 | 0·186 | 0-702 0-427 | 1062 0-728

From this Table the following is deduced:

TABLE X.-Oxygen contained in the Silica, Peroxides, and Protoxides of Scotch Granites.

Introducing the preceding values of A, B, C into equations (4), we find the following per-centages of quartz, orthoclase, oligoclase, and black mica:

The negative values of z and u, in the foregoing Table, demonstrate that the Scotch granites, like many of those of Donegal, cannot be represented by any combination of quartz, orthoclase, oligoclase, and black mica; and yet they contain all those minerals.

The granites of Norway, Sweden, and Finland, so far as I have had an opportunity of examining them in the field, have the same stratified geological structure as the granites of Donegal, and they are composed of the same constituent materials, namely, 1, Quartz ; 2, Orthoclase; 3, Oligoclase; 4, Black mica; 5, White mica (in veins). I do not remember to have seen hornblende or tourmaline in the Swedish granite and gneiss.

The orthoclase is generally pink, and the oligoclase either white or waxy greenish, very like the oligoclase of the Garvary Wood veins, near Castle Caldwell.

The black mica is much more abundant than the white, which is generally found in veins, as in the quarry of Ytterby and other celebrated mineral localities.

I have made the following analyses of the constituent minerals of the Swedish granite and gneiss.