Components of an Igneous Rock — Williams 43 
one nearly uniform in size. This fact cannot but be empha- 
sized by examining a thin section of almost any granular 
rock. 
To carry out this system of measurement Rosiwal rec- 
ommends that the section be very thin, as nearly a plane as 
possible. A number of sections should be examined to ob- 
tain a representative average. It is also desirable to have 
the sections as large as can be conveniently made. In the 
present study, series of fine ink lines were drawn on the 
cover glass to use as indicatrices. The average percentage 
composition calculated from the data obtained in a study of 
three thin sections is included in the summary table. From 
the calculated composition the variety of plagioclase falls in 
the oligoclase group, about Ab^ An^. This identification is 
corroborated by the microscope. The typical specific gravi- 
ties of the minerals are then employed for transferring to 
weight percentages, viz., quartz, 2.65, orthoclase, 2.56, oligo- 
clase, 2.64, and for the iron ores, 5.25, that of magnetite. 
The next method mentioned in the introductory outline 
is the measurement of the areas of the mineral grains in thin 
sections. This may be done in three ways, (i) drawing the 
outlines of the grains with the aid of the camera lucida. 
As used by W. J. Sollas*, this method is the application 
of the Delesse process with the microscope. 
(e). PHOTOGRAPHING AND CUTTING OUT AREAS OF GRAINS. 
(2) Making photomicrographs of several fields in the 
slide, dissecting prints thus obtained, then weighing the 
pieces of paper representing the areas of the constituent 
minerals. Photomicrographs were made of six successive 
fields from one large rock section. Although the magnify- 
ing power of the objective used was but medium so as to 
avoid distortion and also to include as large an area as pos- 
sible, the focusing distance above the eye-piece was adjusted 
so as to get an image approximately 420 times the actual 
size. The Nicols were set at such an angle to each other 
that the contrast in coloration was strong enough to defi- 
nitely outline the grains and yet not enough to seriously 
♦Contribution to a Knowledge of th e Granites of Leinster, Tr.tns 
Royal Irish Acad. Vol. XXIX, pt. XIV, p. 471. 
