418 F. P. Mennell — The Northern Margin of Dartmoor. 



the development of minute biotite flakes. The eruptive materials- 

 were clearly of classes quite unlike those represented by the lavas 

 now seen in situ at Was Tor and near Brent Tor on the western side 

 of the Moor. The porphyritic augite-andesite of Doddiscombleigh is 

 also quite different from anything to be seen here. Some of the 

 largest fragments in the tuff, which are upwards of 2 feet in length, 

 are of a light-coloured, inconspicuously banded, sedimentary rock of 

 unknown provenance. 



Blackish flinty rocks outcrop in the river bed above the tuff, and 

 a specimen sliced was found to contain numerous and quite well 

 preserved radiolarian remains. It is, in fact, a typical radiolarian 

 chert. Two specimens from the sedimentary bands interbedded with 

 the tuff also proved to contain radiolaria. The occurrence of these 

 radiolarian rocks is of considerable interest, as although I have not 

 detected radiolaria elsewhere in the contact zone, they are no doubt 

 to be found by careful search and would serve as useful indices of 

 this horizon. Outside the metamorphic aureole, on what is probably 

 the western extension of these very beds, a band of radiolarian rock 

 occurs along the edge of a patch of woodland, on the main road 

 between Bridestowe and the railway station of the same name. 



The dark-coloured beds grade off into the banded types of hornfels 

 so conspicuous in the Okement section. They dip much more 

 steeply, and this is perhaps chiefly responsible for their comparatively 

 narrow outcrop. There is no chiastolite band under the striped rocks 

 corresponding to that of the Okement Valley, and it may be surmised 

 that it is represented by the dark beds, including those with 

 radiolaria. Curiously enough, a well-marked chiastolitic band occurs 

 directly above the lowest epidiorite seen here. This last is 

 a beautiful porphyritic type, containing much secondary biotite, and 

 is well exposed in the river bed under the private footbridge leading 

 into Skaigh Warren. This does not correspond in its position with 

 either of the intrusions seen in the Okement section, though what is 

 evidently the lower of these is exposed close, by at Brinemoor just 

 outside Belstone, with numerous fragments indicating banded 

 hornfels overlying it and chiastolite hornfels below. There is some 

 dislocation, however, at Belstone, as chiastolite hornfels occurs near 

 the village on the high ground on the left bank of the river. This 

 must have reached its present position by faulting, as it does not 

 occur on the opposite side, which has perfectly regular bands of" 

 altered limestone in its proper position between andalusite hornfels 

 and tuff. The chiastolite-bearing rock, exposed at Skaigh, is also, 

 well seen in the bed of the river, which runs along its strike from 

 Skaigh Warren to near Sticklepath. It has some beds above it, very 

 thin at Skaigh, but much thicker at Sticklepath, where they are 

 worked for road-metal, which are often conspicuously spotted, and' 

 appear to correspond to the beds numbered 6 in the Okement 

 section. Above them comes the cordierite hornfels, of which very 

 fresh specimens can be obtained alongside the road just below Skaigh. 

 At this point a narrow epidiorite intervenes between it and the 

 underlying strata. The intrusion is quite different in structure from 

 that exposed in the river bed, and occasionally shows a little 



