570 REPORT— 1889. 



■with consequent reconstitution of the granite in the presence of salt water, with 

 injection into the fissures. 



There has been no eviience of veins ia the trawled Channel rocks. 



The Dartmoor granites abound in A'eins both injected and infiltrated. The 

 felspathic matter in the infiltrated veins occasionally presents indications of the 

 polysynthetic twinning of plagioclase. 



The tourmaline in the crystalline veins is occasionally the straw-coloured 

 idiomorphic variety: that in the aqueous veins is of the radiating dark-green 

 variety (both occur together in Luxulyanite). 



On the southern borders of Dartmoor in the culm- measures, chiastolite ( = anda- 

 lusite) is developed— sil. 36-8, al. 63-2,> sp. gr. 3-16— 3-2.- 



On the northern borders of the Channel, i.e. in the Bolt Head, mica schist, 

 kyanite ( =disthene) is developed — sil. 36-8, al. 63'2,' sp. gr. 3'5 — 3'7.- 



Chiastolite and kyanite having the same chemical composition, the heavier 

 kyanite may be taken to indicate a region of greater compression, where it occurs, 

 viz. the borders of the Channel crystalline rocks. 



The Channel granites and gneisses have been considered by experts archsean. 



The Dartmoor granites have been commonlj' considered post-carboniferous on 

 the evidence of the veins injected into the culm-measures. 



The facts above noted seem to indicate that the culm-measures on the flanks of 

 Dartmoor were deposited on an ocean bed of pre-Devonian granite, and that in 

 post-carboniferous times the Dartmoor area, by a heavy north and south squeeze, 

 was elevated and fractured, whilst still beneath the sea, thus giving access to the 

 superincumbent salt water. The Channel area being depressed would by the same 

 earth movement undergo heavy compression at its surface which would prevent any 

 ingress of salt water. 



On this hypothesis the Channel and Dartmoor granites are both pre-Devonian, 

 and probably archfean. The former having occasionally been altered by compression, 

 the latter by moderate heat in the presence of salt water. . The one chiefly aflPected 

 by compression, the other by solution, but both having imdergone their mineral 

 transformations, whether dynamical or chemical, or both combined, before emerg- 

 ing from the waters of the post-carboniferous sea. 



5. The Island of Paros, in the Cyclades, and its Marhle Quarries. 

 By Robert Swan, F.C.S. 



The island of Paros is II miles long and 8 miles broad at its widest part. There 

 is a broad belt of nearly level land round the coast ; but the interior is moun- 

 tainous, rising to a height of 2,530 feet at Movmt St. Elias (probably the ancient 

 Mount Marpessus). 



The northern and western parts consist of schist and gneiss, granite appearing 

 also in the environs of Parekhia. The southern part of the island consists chiefly 

 of crystalline limestone. There is no evidence here of the age of this limestone ; 

 but that of Attica is now known to be Cretaceous, and probably that of the 

 Cyclades is of the same age. The finest statuary marble, or lychnitis, varies 

 from 5 to 15 feet in thickness at the quarries of St. Minas; it occurs in a bed of 

 coarse-grained white marble, with bluish-black veins. The coarse marble becomes 

 dark in colour near the lychnitis, both above and below it, and thus the layer of 

 statuary marble is distinctly marked oft". The dark colour is due to traces of 

 binoxide of manganese and magnetic oxide of iron. It seems probable that the 

 impurities have been withdrawn from the lychnitis and have become concentrated 

 near the edges of the adjacent seams of limestone. 



The rocks are much disturbed and folded, and often dip at high angles. The 



* Systevi of Mineralogy, Dana, pp. .372, 375. 



^ Microsoojfioal Physiography of the Roch-forming Minerals, Kosenbusch, pp. 194, 



» See B. A. Report 1888, p. 520. 



