TRANSACTIONS OF SECTION C. 



6G1 



the followinj^ uiinerals : forsterite, or serpentine after forsterite, tremolite, 

 (liopside, and brucite. The dominant carbonate of tlie marble is calcite, but 

 dolomite occurs in variable cpantity. The amount of dolomite decreases as the 

 total amount of the ma?nesian t^ilicatea and brucite increases. The original 

 dolomite contains a variable amount of silica in the form of chert. 



When the altered rocks are examined under the microscope it is seen that 

 forsterite, serpentine, and tremolite are invariably associated with calcite, but 

 that diopaide is sometimes associated with dolomite. These facts of parag'enesis 

 can be easily accounted for if we assume that the silica of the orio-inal dolomitic 

 rock has combined with the bases of the carbonate, and preferably with the 

 mac^nesia, for diopside is rare. Thus forsterite, a magnesian silicate, cannot have 

 been formed in the dolomite without the liberation of lime, and consequently we 

 find either detached crystals of forsterite surrounded by aureoles of calcite in 

 a matri.x of dolomite, or, when forsterite is abundant, a simple afrgrpg-ate of 

 forsterite and calcite ; the formation of tremolite in wiiich the ratio of CaO : MgO 

 is 1 : o also implies the separation of lime from magnesia; and it is invariably found, 

 like the forstrrite, in direct contact with calcite. But in diopside the ratio of 

 CaO : MgO is the same as in dolomite ; so that in accordance with the principles 

 above explained we should expect to find these two minerals in contact, and this 

 has been observed. 



The above facts clearly point to the conclusion that the cherty dolomites have 

 been dedolomitised by the formation of magnesian silicates. Carbonic acid has 

 been driven off, but the ratio of the bases has not been disturbed. The ratio of 

 CaO : MgO in the altered as in the unaltered rocks is approximately 1 : 1. 



But dedolomitisation has also been produced in another way. C'ertain 

 varieties of the marble are composed of calcite and brucite. The brucite is pro- 

 bably a pseudoniorph after periclase, just as the serpentine is a pseudomorph after 

 forsterite. We are therefore compelled to conclude that, under the conditions 

 which prevailed during the intrusion of the plutonic rocks, the carbonic acid freed 

 itself more readily from the magnesia thnn Irom the lime; thus, in the absence of 

 silica, giving rise to the formation of periclase and converting the original dolomite 

 into an aggregate calcite and periclase, the latter mineral subsequently being 

 clianged to brucite. The resulting rock is identical with the well-known predaz/.ito 

 of the Tyrol, which was probably formed in a similar way. 



9,. Fossil Floras of South Africa. By A. C. Seward, F.R.S. 



1. Vitenhacie Flora. — The plants from the Uitenhage series of Cape Coloi.y 

 include types characteristic of Wealden and others more closely allied to Jurassic 

 species. On the whole there is a balance of evidence in favour of a Wealden 

 horizon. 



Onychiopsis Mantelli (Brongn.) 

 Cladophlebis Brotvninna (Dunk) 

 Cladophlcbis dentkiilata (Brongn.) 



forma Atherstonei. 

 Sphenopteris Fittoni, Sew. 

 Sphetiopteris sp. 

 Treniopteris sp. 

 Zamites recta (Tate). 

 Zamites Morrisii (Tate). 

 Zamites africana (Tate). 

 Znvntcs Rubidgei (Tate). 

 Nilssonia Tatei sp. nov. 



Cycadolepis Jenldnaiana (Tate). 

 Benstedtia sp. (cf. Coni'ferocaulon 



Columhcafonne , Fliche), 

 Carpolithes sp. 

 Araucarites Jtorjersi sp. nov. 

 Ta.vites sp. 

 Brachiphi/llum sp. 

 Conitcs sp. a .1 

 Conites sp. fi .j 

 Coniferous wood. 

 Tlanta incertce sedis. 



2. Stormhcrg Flora. — The plants from the Stormbcrg series point to a flora of 

 Rbsetic age. The Rh?etic vegetation, of which remnants have been recorded from 

 Scania, Franconia, and other parts of Germany, North America, New Me.xico, 



