IS85.1 ^Od [Lewis. 



numerous calamites (C. arenaceous), overlies the breccia ; (G) The lo-n-er 

 pebbly couglomerate of the Trias follows as a ■well-marked and definite 

 horizon, w^ich may be traced continuously from the Delaware at Centre 

 bridge, past Spring valley, and south of Doylestown to the fault at Chalfont. 

 It forms a ridge of pebbles and is readily recognized. The same horizon 

 occurs at the southern edge of the Triassic area, extending from Norris- 

 town to Trenton ; (7) The long succession of red shales and sandstones 

 which form the bulk of the Triassic strata follow on top of the con- 

 glomerate. 



Had our section been made at Bushington, the Potsdam sandstone 

 would have been absent, and Hudson River slates would have appeared 

 conformably overlying the limestone, and unconformably overlaid by the 

 Triassic conglomerate. These Hudsou River slates are believed by the 

 writer to correspond with the South Valley Hill slates and hydromica 

 schists of Chester county, which latter have there been more highly meta- 

 morphosed. There is a striking similarity in their general appearance. 

 From the observed dips of the limestone at the south- western end of 

 Buckingham valley, there appears to exist here an anticlinal, by which 

 the Potsdam sandstone is covered over, and the Hudson River slates swing 

 around the end of the valley until they meet the fault. The fault is clearly 

 a large and important one, of several thousand feet upthrow. 



Returning now to the long trap dyke which we have followed from the 

 south side of the Bucks county fault to the Stale of Mar} land, it is interest- 

 ing to find that at a point on the north side of the fault, nearly five miles 

 west ol the abrupt termination of the dyke near Bridge Point, another 

 dyke of identical composition and structure as suddenly appears and 

 extends for twenty miles in an arch, until it is again cut off by the fault. 

 There is every indication that it is of the same age and origin as the long 

 dyke previously described, and tliat it should be considered as part of it. 

 The trap has the same characteristic weathered surfaces, the same metal- 

 lic ring when struck, it lorms the same narrow and often inconspicuous 

 dyke, and hand specimens from the two dykes cannot be distinguished 

 from one another, macroscopicaliy or microscopically. 



Starting north of Chalfont, between Pine run and tlie North branch of 

 Neshamiuy creek, in New Britain township, it curves north-eastward so 

 as just to touch the corner of Doylestown and New Britain townships. It 

 passes through the village of Iron Hill, so named on account of the numer- 

 ous boulders of ringing trap which cover the hill. The dyke is here wider 

 and more prominent than at any other place along this section. It passes 

 thence about half a mile south-east of Gurden Glen Mills, whence, keep- 

 ing immediately south of the North branch of the Neshaminy, it passes, 

 one-third of a mile north of Fountainville, into Plumstead township. It 

 is followed past Danborough, where it appears at the fork- roads north of 

 the village ; past Gardenville, crossing the road a quarter of a mile north 

 of the village ; whence, curving towards the east, it enters Solebury town" 

 ship north of Carversville. The next village through which it passes is 



