In the present paper storm time variations and 27-day geomagnetic periodicity have been analysed to estimate the depth of the substitute conductor, assuming an infinitely (super) conducting core model of the earth. The advantage of using data from a restricted longitude range is that the uncertainties arising from lateral contrasts in the upper mantle and contributions from Sq current systems are considerably reduced. The result of the present analysis, which has been done in the time domain, gives a value of 522 km for the depth of the substitute conductor in case of storm time variations which rises to 870 km for 27-day recurrent storms. A higher value of the depth for 27-day variations indicate that the rise in conductivity inside the earth is not like a step function rather is a gradual one. The value of 522 km for storm time variations for the Indian region is smaller than the global average. This is natural to expect because the Indian sub-continent is known to be a tectonically active region.

The magnetic measurements of declination (D), horizontal (H) and vertical (Z) components of earth’s magnetic field, collected from ground surveys between 1962 and 1966, are used to develop an analytical model of geomagnetic field variations over Indian region for the epoch 1965. In order to reflect spatial features with wavelengths of approximately 1000 km, sixth degree polynomial as a function of differential latitude and longitude is calculated by the method of least squares. The root mean square fit of the model to the input data is better than that accounted by the International Geomagnetic Reference Field for 1965.0. Isomagnetic charts drawn forD, H, Z and total force (F) reflect more details than that shown on world magnetic charts. Further, the values of the field at common repeat stations recorded between 1962 and 1974, after eliminating the field values for the epoch 1965.0, are used to get the secular variation as well as its spatial dependence again by means of polynomial which now includes coefficients which are functions of time and of geographical locations. The accuracy of coefficients is tested against the behaviour of secular variation at permanent magnetic observatories. The merits and limitations of the model are discussed.

This paper describes a comprehensive geochemical study of basalt–andesite–dacite–rhyolite volcanic association in the Khardung volcanic suite along the northern margin of the Ladakh magmatic arc. This volcanic association is outcropped mainly in the segment of the further north of the Khardung village to Khalsar delineating from the Ladakh magmatic arc by $\sim$2 km thick porphyritic sill. The closed association of basalt–andesite–dacite–rhyolite volcanic within a volcanic suite suggests that these rocks may be genetically inter-related and might have derived from the same parental magma source. Felsic lavas (dacite–rhyolite) show $\rm{SiO}_{2}$ range from 64.75 to 79.11 wt.%, while intermediate lavas (basaltic andesite–andesite) ranges from 50.80 to 51.81 wt.% with mafic lavas (basalt) span from 53.39 to 62.05 wt.%. These volcanic suites show enrichment in LIL elements (Rb, Ba, Th, U, and Pb) and depletion in Nb, P, and Ti, which can be evident in spider diagrams with pronounced to mild Eu negative anomalies in REE patterns. Previous reports on zircon U–Pb ages of the Khardung volcanics range between 60 and 69.7 Ma confirm an upper bound eruption age of this volcanic suite as pre-collision continental arc magmas. Hence, the results of geochemical modelling suggest that the Khardung mafic–intermediate-felsic lavas were generated from the melting of 1–4% spinel and garnet-bearing lherzolite sources. The generated parental magmas were modified by crustal materials during the magma ascent along with fractional crystallization and were metasomatized by slab-derived fluids released from the subducting Neo-Tethyan oceanic crust during the Late Cretaceous to Palaeogene in the northern margin of the Ladakh magmatic arc.