462 W. A. E. TJSSHER ON THE CHRONOLOGICAL VALT7E OF THE 



fragments occur were subject to sufficient heat to produce the 

 change in colour, yet such a change might have been produced 

 by water percolating through ferruginous strata, or by similar chem- 

 ical agencies. Though the occurrence of the fragments in the Trias 

 within a certain distance of Dartmoor would favour this view, it is 

 very problematical that the Dartmoor granite ever formed cliffs in 

 the early Triassic sea ; so I am inclined to refer the fragments to the 

 sources mentioned in the proposition, but more especially to the last 

 named, as a continuation of the Guernsey and Cape la Hogue granites 

 may have furnished the material. 



(a) If the granitic origin of the fragments be disputed, is it pro- 

 bable that the exposed portions of a granitic protrusion would ex- 

 hibit the characters of a quartz porphyry? In that case, as no 

 evidence of such offshoots from the Dartmoor granites is forthcoming 

 in districts uncovered by Triassic rocks, I must agree with De la 

 Beche* that these porphyritic fragments " may readily have formed 

 portions of igneous masses covered up by the red-sandstone series ; 



" That the granite of Dartmoor may not have been covered by 

 water when the red-sandstone series of the district was formed and 

 the igneous rocks associated with it ejected"f . 



The breccias frequently contain igneous fragments distinctly re- 

 ferable to the destruction of such igneous patches as those of Wash- 

 field, Killerton, Silverton, Spencecombe (north of Yeoford), &c. The 

 outflow of these lavas seems generally to have accompanied or 

 heralded the earliest deposition of Triassic sediments in the districts 

 in which they occur ; nor does it appear impossible that the eruption 

 of quartz porphyries may have been in some way connected with 

 their appearance. 



(6) The porphyritic and Murchisonite fragments are confined to 

 the South-Devon breccia, not extending further north than the 

 northern boundary of the Crediton valley, and being limited to the 

 eastward by a line drawn from Heavitree (Wonford quarries) to 

 Exmouth. 



(c) The general rounding of the porphyritic fragments, and the 

 occasional occurrence of quartzites (which seem to me to resemble 

 the Budleigh pebbles) in the breccia near the Labrador Inn, south 

 of the Ness, Teignmouth, give colour to the hypothesis that these 

 earlier Triassic sediments contained materials foreign to the present 

 extent of the county, but perhaps represented beneath the bed of 

 the English Channel. 



Sixth proposition. — That on a comparison of the sequence of de- 

 posits in the midland counties with those of Devon and Somerset, 

 from the evident continuity of the upper division, only, between 

 the two areas, and from the absence of unconformity in the West- 

 Somerset and Devon area, it is manifest — 1st, that the Upper 

 Marls, the Upper Sandstones, and probably the conglomerate and 

 pebble-bed subdivision of Devon and Somerset are equivalent in 

 time to the Keuper series of the midland counties; 2nd, that depo- 



* « Eeport on the Geology of Cornwall, Devon,' &c, p. 217. 

 t Ibid. p. 218. r 



