410 THE TROPICAL AGRICULTURIST. [Dec. 2, 1901. 
granitic : some of the veins are only 5 inches across, 
and frequently contain a fair propoition of the 
darker minerals. Tlie hornblendic rock itself often 
exhibits much felspar. Sometimes 1 'djc lenticular 
bands of tlie dark rock appear ; at otiiers it is 
represented merely by a few broken minerals. 
A very fine section is exposed near the 54 bench- 
mark. Tills shows in some places the rock 
beautifully and wavily bandeil, the darker parts 
varying somewhat in their proportions of horn- 
blende and mica, and hence in colour. In places a 
light-coloured vein cats across an older banding 
and yet seems to form an integral part of the rock ; 
at ochers, the hands are disjointed and lie in frag- 
ments in the lighter matrix as though broken, 
while occasionally we find strongly-marked puckers. 
The gigantic rock varies in texture, and some- 
times for a foot or so, is almost free from dark 
folia. 
Thin sections show the dark rock to consist of a 
mosaic of green hornblende-crystals, extremely 
irregular in shape, and plagioclase, the former 
predominating in quantity. Occasionally, rounded 
grains of the hornblende are enclosed in the felspar. 
Apatite is very plentiful, and small zircons are not 
uncommon. The section also contains a little 
pleocliroic augite with the characteristic colours 
of hypersthene. The specilic gravity of the rock 
is 2 95. 
The granitic rock, taken at a point where it 
possesses a minimum of hornblende or mica, is of 
medium grain or rather pink in colour. A thin 
section discloses microcline in considerable quan- 
tity, some orthoclase with microperthitic inter- 
growth, and a good deal of plagioclase. Qaartz 
is plentiful. In addition, the rock contains a few 
irregular flakes of biotite, zircon, and irou- 
oxide. 
Specimens fiom the banded part of the same 
mass are characterized by rather small horn- 
blendes and micas, often with a few larger indi- 
viduals of the former scattered through the rock. 
The specific gravity of such a rock is 2-82. 
One possessing well-marked bands has a peculiar 
though quite distinct foliation, making an angle 
of about 45* to the direction «f the bands. Under 
the microscope the rook is seen to consist of pla- 
gioclase, orthoclase, quartz, green hornblende, a 
little brown mica, zircon, and apatite : the darker 
minerals predominating in certain jiarts. _ and, 
together with less quartz than in the remainder 
of the slide, constituting the banding. The pro- 
portion of mica varies and in several instances 
it is associated with hornblende in a way which 
suggests very strongly its formation from that 
mineral. Iron-ores are abundant : for the most 
part they are probably ilmenite, together with 
some pyrites. The rock shows no indication of 
crushing and the foliation referred to above seems 
probably due to a fresh movement preceding final 
consolidation, causing the mica-flakes to take 
up a new position oblique to the plane of 
banding. 
(b) ON THE Ceylon Govrrnmejmt Railway : 
WESTWARD FROM KADUGANNAWA. —Along the 
rail way line good but monotonous sections are ex- 
po.sed, essentially resembling those just described. 
The rock is both' finely and coarsely banded, in 
some places the granitic veins measure 4 inches 
across, and although they frequently remain 
straight for some distance, yet they often vary in 
thickness, thinning out, then swelling. Occasion- 
ally the rock is gnarled, but this is not charac- 
teristic ; while not frequently the veins, by inter- 
secting, produce an appearance of brecciation. This 
seem-; incompatible with any theory which would 
impute the banded structure to crush. A little 
fartlier west tlie lock, banded iu its lower part, 
has a much more brecciated look in its upper. 
The whiter veins seem partly to enclose lenticular 
masses of the darker rock, and, in a few cases, to 
form a rather high angle with the underlying 
bands. Tn one case a long band of the darker 
rock, containing one or two small white bands, has 
been broken in two (lig. 3), so that probablj' the 
granitic rock is not all quite of the same age (lig. 
4). Home coarse felspar-veins are certainly 
younger than the granitic rock which constitutes 
tlie more regular bands. 
The cross cutting granite-vein of fig. 4 is fine- 
grained and white in the hand-specimen, thereby 
differing from the granite described on p. 199. A 
thin slice exhibits large and very irregular grains 
of quartz, orthoclase in considerable quantity, 
microcline rare or absent, plagioclase, and a few 
greenish flakes of mica. The rock has a granulitio 
structure, the quartz usually penetrating tLe fels- 
par, or in some places forming quartz vermicule. 
Its specilic gravity is 2 62. 
A dark band (specific gravity 3'28) in the banded 
rock below Kadugannawa shows in a thin section 
much pale green augite, altering to darker 
green hornblende. A brown mica is pre- 
sent, often presenting the appearance of a 
further product of alteration, but at others seem- 
ingly independent and ia well-formed flakes with 
a distinct orientation. Tire rest of the slide 
consists of pyrites with some magnetite, untwinned 
felspar (not ^ery plentiful), perhaps a little 
dolomite, and apatite. 
A thin section, from a well-banded specimen 
from the same cutting, exhibits botli iu constituent 
minerals and in structure a great resemblance to 
those already described in § (a). In the lighter 
bandstheonly ferro-magneiianmineral isa mica; the 
remainder consists of grains of quartz, plagioclase, 
and some orthoclase. The extinction of the 
plagioclase symmetrically with regard to the trace 
of the composition-plane varies from 7 dg. to 9 dg. 
The darker parts of the slide are distinguished 
by large plates of green hornblende and a quantity 
of brown mica. The two minerals are intimately 
connected the one with the other ; in one instance 
the mica seems to be an alteration-product of the 
hornblende. The constituent grains of the light 
and dark parts interlock, and there is nothing 
appr<iachiiig to a sharply defined line of contact. 
Quartz is less plentiful in the darker parts , a few 
grains of an apatite-like mineral occur, and an 
occasional crystal of zircon. There is very little 
quartz vermicule, but no microperthite. 
In the cutting just below Kadugannawa (that is, 
on the west) the banded rock becomes strikingly 
garnetiferous. The garnets favour the dark parts 
in their distribution, but occur also in the lighter. 
One or two are of the size of a very small pea, 
but this is exceptional. A typical specimen is 
dark and micaceous, markedly foliated, slabby in 
fracture, and crowded with small garnets, about 
•025 inch in diameter. A thin section shows that 
hornblende is entii ely absent, its place being taken 
by the garnets. These are irregular in shape, 
rather cracked, and may be either pyrope or alnian- 
dine.* Occasionally we find mica-flakes embedded 
in the garnet. The plagioclase gives -syminetrical 
* M Lacroix refers the garnet to almandine, Ball, 
Soc, mia. France, vol, xii (1889) pp. 288, 306, etc. 
