130 DR. A. L. Dl' TOIT OX THE [vol. l.KXV, 



from the granite veins by from a few up to fully 12 inches of 

 marble rich in diopside, scapolite, or forsterite, and grading on the 

 other side into clear calcite, generally rather abruptly. Some- 

 times the mineral is scattered evenly through several inches of 

 rock ; at other times it is concentrated into narrow bands, along 

 with either phlogopite. or spinel, or both. 



At Swakopmund, in the South- West African Protectorate, are 

 certain belts of crystalline marble evenly flanked by intrusive 

 gneiss, showing an association of minerals identical with that at the 

 Marble Delta. Layers rich in one or more of the minerals diopside, 

 serpentine, phlogopite, and spinel, which (even though narrow) can 

 be followed along the strike for long distances, mark out the original 

 bedding-planes of the marbles. The chondrodite, on the other hand, 

 may be scattered rather sparsely through the white marble, and 

 when followed along the strike, the orange-coloured mineral tends 

 to vanish but to reappear upon an adjoining bedding-plane ; more- 

 over, it never approaches nearer than a couple of yards to the 

 gneiss-contacts. Owing to the absence of transverse dykes in that 

 part of the belt which I studied, the behaviour of the mineral 

 towards such intrusions could not be determined. 



In the account given by Kemp & Hollick of the crystalline 

 limestones of Warwick (X.Y.), a similar relationship holds, the 

 chondrodite zone — with phlogopite, spinel, and sometimes fluorite, 

 and with the calcite beyond — being separated from the granite by 

 a diopside-scapolite zone. 



Though the yellow mineral is referred to comprehensively as 

 chondrodite. clinohumite is also included ; but the specific distinc- 

 tion is often difficult to make, owing to the feeble development of 

 the basal cleavage. Moreover, the two species are not unusually 

 intergrown, in certain cases having the principal plane in common ; 

 humite is probably absent. While the yellow colour and the pleo- 

 chroism will serve to distinguish the chondrodite from forsterite 

 in thin section, there are cases in which only a limited part of a 

 crystal is yellow, the remainder being colourless : but in such cases 

 the whole still extinguishes simultaneously between crossed nicols. 



The varieties that are without colour in thin section are, whether 

 fresh or partly serpentinized, all but indistinguishable from forsterite, 

 and resort must be had to the hand-specinien, in which the latter 

 mineral possesses a greenish tint. In one instance, cited below, 

 there Avas an intergrowth of the two minerals, but almost invariably 

 and fortunately the one occurs to the exclusion of the other, a fact 

 of prime significance. 



In the instance cited, a band 5 cm. wide in the calcite is par- 

 ticularly rich in chondrodite and phlogopite, the former appearing 

 in the thin section (3108) as large strongly-pleochroic grains of 

 irregular outline, including mica and occasionally small octahedra 

 of pale pink spinel ; sometimes the mineral is in aggregates, with 

 the prisms arranged in a more or less parallel fashion. 



An interesting point is that certain of the grains, which 

 are from 1 to 3 mm. in diameter, consist of a shell of yellow 



