1867.] TIMIJSrS MALYEEN HILLS. 369 



according to their age. Many of the '' diorites," which are the 

 oldest, are more basic than some of the volcanic beds, which are the 

 most recent, some of the Lower Cambrian than others of the Upper 

 Cambrian period. There is as much variety in the atomic amount 

 of the silica in rocks of the same age and place, as in rocks of dif- 

 ferent ages or from distant localities, and as much resemblance 

 between some of the trap-dykes and the products of ancient and 

 modern eruptions of Hecla and Vesuvius as that which they bear to 

 contemporaneous dykes in their vicinity. 



4. The atomic proportion of the silica to the bases appears to 

 be highest in the largest masses of trap, such as those of the jS'orth 

 HiU (XCI., XCII., LXXXIX., XC), the Worcestershire Beacon 

 (LXXXIT., LXXXIV.), the large irregular mass in the east buttress 

 of the Herefordshire Beacon (LXXIY.), and the great dyke of Mid- 

 summer Hill (LXY.), — and lowest in the smallest masses, as shown 

 in some of the dykes of the Herefordshire Beacon (LXXIII., LXXYI., 

 LXIX.) and in the dyke traversing Swinyard Hill (LXYI.). But this 

 law, though very general, is not invariable ; for in the trap of the 

 middle terminal Hill (LXXXYIII.), which is rather a large mass, 

 the oxygen quotient is about 0'7, and the specific gravity 0*955. 



5. In the same masses of trap there is an appreciable increase in 

 the silica towards their centres, as may be seen by comparing the 

 central portion of the trap in the large dyke over HoUymount (XCI.), 

 which is nearly a bisilicate, with the outer portion (XCIL), and 

 similarly, in the North HiU, LXXXIX. with XC, and, in the Wor- 

 cestershire Beacon, LXXXII. with LXXXIY. It must, however, 

 be observed that at the extreme border of a dyke, if there be not 

 sufficient pressure exerted by the contiguous rocks to make it dense 

 and impervious to water, soluble bases have been removed, and the 

 relative amount of silica thereby increased (LXXXY., LXXXYIL). 



The increase of the silica towards the central parts is not the 

 effect of a liquation in the fluid mass ; for its atomic proportion 

 would then be as much below the average in the exterior portions as 

 it is above it in the interior ; whereas no parts of the larger masses 

 contain more base than is usually found throughout the trap-dykes 

 of the chain. 



From these facts I infer that the primary source of all the trap- 

 rocks in the Malvern Hills was nearly a bisilicate, which, during 

 the various processes by which it has been brought to the surface, 

 has become united more or less with other substances, assimilating 

 metallic oxides, lime, magnesia, or alkalies, according as one or others 

 might be locally prevalent, just as, in modern times, the lava of 

 Yesuvius takes up soda, and that of Etna lime. When these basic ele- 

 ments have been deposited in the rock as free oxides, or have become 

 united with other acids than the silicic, they may be separated more 

 or less completely from the silicate, as shown in Analyses XIY., XY., 

 LXXX.,XCIII. When they have become united in various proportions 

 with the silica, they have caused a diminution in the atomic amount 

 of the silicate of the erupted rock. In none of the traps which I 

 have examined has all the base been saturated by the silica ; but 



