NEAR NEW GALLOWAY. 579 



contact, have been altered to a rock very like that of the lower 

 cairn of Knocknairling Hill without the garnets ; and on the Bennan 

 Hill to the east, and Laughenghie Hill to the south, there are in 

 purple-brown grits spots, an inch or so long, closely resembling the 

 most altered grits of Knocknairling Hill. 



§ 8. Remarks. 



In connexion with metamorphism the following points seem note- 

 worthy : — 



1. The extreme variation in the amount of alteration undergone 

 at different places at the same distance from the granite. 



a. At the actual contact in different parts of the margin. 



h. In the case of beds close together, as on Knocknairling Hill, 

 where, at a distance of 200 yards, grits are highly altered and shales 

 little affected. 



c. In the case of different layers of the same rock, as on Knock- 

 nairling Hill, where bands of shale, a few inches thick, are much 

 more altered than those with which they alternate ; and even more 

 noticeably in the shales marked A on the map, where, in an exposure 

 of hardened shale about 8 feet high, are bands, an inch or two thick, 

 much altered, with white mica and garnets. 



2. The way in which the extreme alteration of the grits has 

 spread from the margin on Knocknairling Hill, the whole being re- 

 crystallized near the margin, parts of irregular outline similarly 

 affected at a distance of 100 yards, and smaller portions farther off. 



3. That material seems to have travelled through the rock. 



a. The larger lenticles of fig. 3, and the smaller of fig. 2, con- 

 taining quartz, garnet, and sillimanite, as well as the numerous 

 eyes and lenticles of quartz alone, have the appearance of having 

 been filled by material which has been brought from the surrounding 

 rock. 



h. The more altered grits consist largely of clusters of crystals, 

 here of one mineral, and there of another. It would seem as if 

 material must have been conveyed from one part of the rock to 

 another to form such nests of only one mineral. 



4. The apparent order of succession of the minerals. Garnets 

 extremely rarely contain anything but colouring matter and quartz ; 

 chiastolite contains garnets ; bands of mica bend round both. 



5. The microscopical slides give the impression that the minute 

 folding could not all have been prior to the mineral alteration. One 

 would suppose that the black colouring matter of fig. 4, PI. XXIII., 

 must have been heaped up on one side of the waves during the folding, 

 and the form of the knots in the knotted schist connect them with the 

 folding. jN'o inconsiderable part of the rocks consists of twisted bands 

 and eyes of segregated material. The spaces thus filled might have 

 been caused by shrinkage in consequence of crystallization in the 

 more altered rocks, but could not have been so caused in the slightly 

 altered shales. They suggest easing during folding. The little 

 garnet quartz-eyes suggest the same idea. 



