206 ME. J. PARKINSON ON THE GEOLOGY [May I9OI, 



secondary — the former are malacolite, mica, and spinel, the last- 

 named brownish hornblende. This replaces the malacolite com- 

 pletely in many places.^ It is yellowish-brown for vibrations 

 parallel to the prismatic cleavage, and possesses a fairly strong 

 absorption. The mica includes plates of the hornblende, and ex- 

 tends its irregular edges amongst the hornblende-crystals. It is 

 brownish-red for vibrations parallel to the basal plane, and pale 

 straw at right angles to this direction. The mica also encloses small 

 grains of green spinel, which make only the faintest attempt at an 

 idiomorphic outline. The same mineral occurs embedded in the 

 malacolite, and occasionally forms conspicuous aggregates which 

 measure -1 inch across. The malacolite-grains are closely packed 

 together, cracked, and rather stained. Very commonly the horn- 

 blende and mica contain a large number of small rounded greenish 

 inclusions, with a high refractive index. They are so minute that 

 their double refraction cannot be safely estimated, but it seems that 

 a regular gradation can be traced from them to indubitable grains 

 of spinel, and it may be inferred that they are that mineral. 



The laterite-exposures found alternating with those of the 

 limestone must originally have been represented by a crystalline rock, 

 of some such type as that described from the west of Kandy or from 

 the neighbourhood of Bandarawella." Accordingly we find the less 

 difficulty in correlating the small rock-mass termed the ' Band ^ 

 with those between Newara Eliya and Hakgala, which it closely 

 resembles. The presence of crystalline limestone above and below 

 this ' Band ' and the way in which it partially encloses a mass of 

 the former are unaccountable by any explanation other than that 

 of intrusion.^ 



rV. HOENBLENDE AND PyEOXENE-GeANULITES. 



(a) The neighbourhood of Bandarawella. — In the quarry 

 on the hillside above the station we find a garnet-bearing hornblende- 

 felspar-quartz rock with a very granulitic structure (specific gravity 

 = 2' 76). It is often well banded, the more felspathic containing 

 much quartz, but frequently such parts, instead of forming bands, 

 occur as patches coarser than the rest. The felspar is greenish, 

 which gives the rocks a rather dark appearance, and, by the aid of 

 the quartz, a greasy lustre. Occasionally we find large eye-shaped 

 felspars about 1*5 inch in length. A typical specimen is a finely- 

 banded rock, the darker parts being composed of garnet, magnetite, 

 and hornblende. These bands vary in breadth from about •! inch 



^ Some iron must be present in the original mineral. 



^ Decomposition i7i situ seems to be the true explanation of the formation of 

 this type of soil in Ceylon. 



^ I am of opinion that the peculiar mineralogical composition of tlie two 

 rocks described above, namely, the representative of the ' Band' at the junction 

 with the limestone, and the malacolite-mica-spinel rock, may be best accounted 

 for on the hypothesis that local incorporatioii of the limestone accompanied 

 the intrusion. 



'Jil 



