BY G. D. OSBORNE. 187 



extremity of the andesite, where a slight bend toward the south takes place. The 

 dip-surface is broken by a number of cross-cuts made by small tributaries of 

 Martin's Creek, so that when viewed obliquely to the dominant strike the topo- 

 gTaphy has the appearance of a series of dip-slopes arranged somewhat en 

 echelon. The termination of the andesite on the south is certainly due to a fault, 

 and it is probahle that the north-western extremity is close to a fault-line. 



The andesite at the locality under description is in every case of the variety 

 that comes under the general category of "lithoidal phase," and the most wide- 

 spread variety of that phase is the one described above in tlie stratigraphy sec- 

 tion as Variety I . 



A development of the andesite, almost as important as the Martin's Creek 

 occurrence, is that in Mr. A. J. Vogele's property, portions 113, 104, 28, 21 and 

 99, Parish of Barford. Limited to the west, near Mt. Douglas, by a heavy 

 fault {see map), the rocks trend easterly for some distance and then the strike 

 changes to north-north-west in portion 28. Here the flow is faulted to the east 

 and forms a bare ridge in portion 99, with a steep slope directed to the west. 

 This ridge of andesite does not extend very far to the south, as it soon ends 

 against a heavy fault which has placed higher members of the Volcanic Stage 

 against the andesite. 



The slopes of andesite in portion 113 carry heavy masses of talus, and are 

 cut into by a number of very youthful streams flowing south. The glassy phase 

 of the rock is only scantily developed in portion 28, but a larger amount of the 

 two varieties of the andesitic pitclistone, together with the lithoidal phase makes 

 up the rocks occurring in portion 99. 



It is difficult now to follow the andesite from here to the outcrops near 

 Clarencetown and the Williams River. If one traverses across the strike north- 

 wards towards Welshman's Creek from near portion 99, Parish of Barford, 

 where the hornblende-andesite is terminated by a fault, the majority of the lower 

 beds of the Volcanic Stage are passed over, — some are missing but certain dis- 

 tinct types occur. On account of the strata assuming a flat dip in a general 

 south-south-west direction, a considerable horizontal distance has to be covered 

 before one arrives at the stratigTaphical position where the hornblende-andesite 

 should occur. In portions 66 and 67, Parish of Barford, the biotite-quartz- 

 keratophyre which has been shown to be the first important flow following the 

 hypersthene-andesitic pitehstone is observed, and then the absence of the latter 

 is apparent, and only tuffs and conglomerates of almost undoubted Volcanic 

 Stage identity intervene between the keratophyre and an outcrop, in portion 217, 

 Parish of Wallarobba, of lithoidal hornblende-andesite identical in hand-specimen 

 with that at Martin's Creek. Immediately to the north of this occurrence is a 

 very much disturbed area, comprising outcrops of both Burindi and Kuttung" 

 rocks, but the andesite appears to be without the zone of disturbance and is 

 here viewed rather as the directly-faulted equivalent of that forming the ridge 

 in portion 99, Parish of Barford, than as a lava of Burindi age, which has suf- 

 fered almost complete devitrification, for in all the Burindi hornblende-andesites 

 occurring elsewhere, devitrification has not proceeded to any great extent. 



The outcrop of andesite at Welshman's Creek does not proceed far to the 

 south-east, which is the direction of its strike, and from the point of its dis- 

 appearance in portion 217, Parish of Wallarobba, to a spot near the Maitland 

 Road, three and a half miles from Clarencetown, no evidence of its existence has 

 been found, the stratigTaphy between the two localities mentioned being in fact 

 somewhat obscure. From the Maitland Road just near the northern boundary 



