382 
THE GEOLOGIST. 
nada. The report did not contain any statements relative to the age of 
the rocks of Donegal, but it expressed a hope that the labours of the 
Geological Surrey of Ireland in Connaught and of the Survey of Scotland 
would shortly afford data on which sound reasoning as to the age of the 
Donegal rocks might be based. The chemical constitution of the granites, 
syenites, and the included minerals, was then discussed. Most of the 
anaij'ses have already been brought before the Geological Society of Lon- 
don, and published in volume xviii. of the Quarterly Journal. In addi- 
tion to these, two analyses of Scotch granites were given, one from Stron- 
tian and the other from Tobermory, which had been selected owing to their 
similarity to some of the Donegal granites. The report concluded with 
an important investigation into the mineralogical constitution of the gra- 
nites, which was furnished by Professor Haughton. Only the results of 
the calculation were laid before the Association. 
It wiU be seen from the paper which has been already printed (Quart. 
Journ. vol. xviii. p. 403), that there are four equations to determine four 
unknown quantities, namely, the quartz, orthoclase, oligoclase, and black < 
mica, which are assumed to compose the granites in which they are ac- 
tually present, and whose composition is taken exclusively from the ana- 
lyses of specimens obtained in the district under examination. From the 
coeflScients of the four equations, which may be at once reduced to three 
by elimmation of the quartz, ten constants are obtained, by the use of 
wliich the percentages are at once found. The application of this method 
of calculation to the seventeen analyses of granite and granite rocks, 
which are given in the report, leads to results which are unexpected. 
JXine of the granites give negative values to some one or more of the un- , 
known quantities, and therefore cannot consist of the four minerals above r 
mentioned. The remaining eight give positive values, and therefore may 
consist of these minerals. On applying to these eight granites further 
tests furnished by equations relating to the different protoxide bases, it is 
found that not a single one satisfies all the conditions exactly ; however, ' 
the dogree of approximation between the calculated and observed per- 
centages of the constituents is very close, as was shown by an example. 
From this result it follows that not a single granite of those which were 
examined can be represented by four minerals having the precise compo- 
sition given in the report, although nearly one-half of them might be re- ' 
presented by minerals having the same oxygen ratios as those assumed. > 
Hence the authors would conclude that, as has been suggested by many 
petrologists, it is unsafe to draw conclusions as to the mineralogical com- ' 
position of a rock like granite from the analysis of minerals picked out of 
veins and other coarse-grained portions of the rock, and that it is a fortiori 
more imsafe to apply to the case of any granite analysis of foreign 
specimens of minerals, which have never been proved to exist in the dis- 
trict in which the granite occurs. The actual specimens which have beeni 
analj^sed are deposited in the museum of Trinity CoUege, Dublin, and a' 
duplicate series in that of the Eoyal Dublin Society. A catalogue of the i 
mineral localities of Donegal was appended to the report. Upwards of | 
sixty species have been observed. 
liEPOET OF THE SHETLAND DrEDGING CoMMITTEE IN ITS GeOLOGI- ' 
CAL Bearings. By Mr. J. Gwyn Jeffreys. — The paper had been pre- 
viously read in the Zoological Section. 
On the Origin . of the Jointed Prismatic STErcTUEE in Basalts 
AND OTHER Igneous Eocks. By Profcssor James Thompson. — The Profes- 1 
sor's belief was that shrinkage was the cause of fracture and columnar struc- i 
ture, and that a peculiar tension starting from a central line was the cause 
