64 REPORT—18638. 
Taste VII.— Mineralogical Composition of Donegal Granites. 
Orthoclase. | Oligoclase. Mica. Quartz 
2°51 71:03 0:93 25°39 
122-87 —1415 —8:33 —0:51 
—12-45 75°95 8:13 28-74 
38°28 71°75 —2°01 —8:17 
17-98 55°58 2°69 23°91 
22°41 75°60 —10°28 12°76 
36:13 52°38 —3°92 15°85 
5°83 5402 8°70 31:33 
25°45 41°50 2°74 30°30 
16°48 46°31 2°05 35°36 
29°12 37:02 8:77 25°08 
29°45 48:07 1:69 20°78 
158-22 12°83 — 29°48 —42°41 
—541 90°69 —15'02 29°46 
33°58 32°25 2°63 31°65 
125°94, 47-96 —43°79 — 29°14 
39°71 86°50 —19°87 —5'08 
According to this Table, it appears that eight of the granites (those marked 
with an asterisk in the Table) give positive values for all the unknown quan- 
tities, while the remaining nine have one or more of them negative. Hence 
it follows that more than one-half of the granites in question are certainly 
not composed of the four minerals which have been assumed to be in them. 
In the case of eight of the granites it has been proved that they might be com- 
posed of four minerals, having the oxygen-ratios of those which have been 
analyzed ; but it still remains to be proved whether they are so composed or 
not. 
It is well known to be the opinion of many petrologists that it is unsafe to 
draw conclusions as to the constitution of a crystalline rock, like granite, from 
the analyses of crystals picked out from those portions of the rock which are 
coarse-grained enough to allow of such a process of extraction of minerals. 
Here we may be allowed to remark that if there be this doubt as to the 
validity of reasoning based on the analysis of minerals picked out of the very 
rock to which that reasoning is applied, it is @ fortiori much more rash to call 
to our aid analyses of minerals from other localities, which have never been 
proved to exist in the district under examination, and the evidence of whose 
existence depends at best on the results of a microscopic examination. 
In order to ascertain how far this suspicion is founded on truth in the case 
of the granites of Donegal, we must call to our aid the auxiliary equations 
which have not yet been employed. These depend on the iron, lime, soda, 
potash, &c., and have hitherto been grouped together in the equation con- 
taining the oxygen-ratio of the protoxides. 
It is evident that, before we can say that a granite is really composed of 
the minerals whose composition has been given in the preceding part of the 
Report, we must be certain that all the equations of condition furnished by 
each;constituent are fulfilled, as well as those depending on the oxygen-ratios 
of the protoxides, peroxides, and silica. 
On applying these test equations, it is found that not a single one of the 
eight granites which have satisfied the first test fulfils these conditions accu- 
rately, and therefore that not a single granite of those which have been 
examined can be composed of quartz, orthoclase, oligoclase, and black mica 
haying the precise composition which has been assumed for them, 
