92 GRIESBACH : GEOLOGY OF THE CENTRAL HIMALAYAS. 



As already alluded to in the general chapter on the metamorphic 

 area, and in a former paper 1 the great fold of the central mass of the 

 Nanda Devi has its belt of younger stratified crystalline rocks on 

 both flanks of the flexure, the top of the arch being denuded, as can 

 be well observed in the Milam sections. In the Niti sections I could 

 observe this belt only along the south-western flank of the main 

 flexure ; when entering the great gneissic gorge of the Dhauli Ganga 

 between Joshimath and Niti the younger crystalline beds, the vaikri- 

 tas, seem to dip below the gneiss of the southern range of the Central 

 Himalayas. The great fold which forms this range is formed (as 

 indeed all along the line of the Central Himalayas) of an unsymmetrical 

 and inverted flexure the shorter limb being the southern one, dipping 

 to north-east; hence the apparent underlie of the younger crystallines 

 below the gneiss (see fig. i, Records XIII, p. 84). 



The section north-east of the main axis between Malari and Niti 

 Unconformity shows the lower palaeozoic group as resting im- 



between gneiss of main j* ± i i_ •*.« • r «i ,i 



range and lowest pa- mediately on porphyrinic gneiss of the southern 

 laeozoics. range of the Central Himalayas which is shown 



in the sections of* plate 3. The contact between the gneiss and the 

 semi-altered beds of the lower palaeozoic seems perfectly natural. But 

 it is extremely probable that the vaikritas have been cut out in these 

 sections by a reversed fold-fault along the longer limb of the great 

 central flexure of the main range. The younger crystallines, the 

 vaikritas, pass into the overlying haimantas in all the other areas of 

 the Central Himalayas, and form also a part almost of the great por- 

 phyritic gneiss (central gneiss of Stoliczka) elsewhere, that the in- 

 ference seems natural that the vaikritas are only wanting in the Niti 

 sections because they happen to be cut out by a fold-fault of con- 

 siderable length. 



1 Records Vol. XIII, p. 83. 



( 92 ) 



