KIMJJKKIJTK FROM BKLGIAX CONGO. 



181 



instances the inner fibrous zone is very narrow and discon- 

 tinuous. In other instances, it itself exhibits a zonal struc- 

 ture, being- composed of an inner zone of pale brownish-red 

 colour and an outer zone of brownish-yellow colour. 



The n-round-niass of the rock consists of pale greenisli- 

 yellow ser])entiiie crowded with small crystals and anhedrons 

 of perowskile, ilmenite and magnetite, and enclosing- large and 

 small patches of calcite communicating' with stringers of that 

 mineral. Small crystals and anhedrons of apatite are also 

 present. Perowskite is very abundant. It occurs in cubical 

 crystals and rounded grains of purplish-brown colour, 

 exhibiting anomalous birefringence. These often form wreaths 

 about the idiomorphic ground-mass olivines. The mineral is 

 unusually abundant in the neighbourhood of some of the 

 larger phenocrysts of ilmenite surrounded by reaction rims 

 of perowskite that has clearly developed at the expense of 

 the ilmenite. This sug-gests that the crystals and grains of 

 perowskite occurring in the ground-mass are also of secondary 

 origin. The apatite is partly in the form of hexagonal prisms 

 and partly in the form of peculiar ragged grains identical 

 with those found in some varieties of kimberlite of the 

 Kimberley mines. The serpentine, of which the bulk of the 

 ground-mass is composed, again consists of minute radial 

 aggregates of chrysotile fibres. As to the original nature of 

 the ground-mass, there is nothing to go upon. 



The kimberlite. both in its megascopic and microscopic 

 character is, as i)reviously indicated, almost identical with that 

 of the Kimberley Mine, situated some 1,200 miles to the south 

 — proof, if any were needed, of the correctness of Carvill 

 Lewis's contention* that kimberlite is a distinctive rock type, 

 and entitled as such to a sjiecific designation. 



CONSTITUENTS OF THE FLORA OF SOFTHEEN 



RHODESIA. 



Bv F. Eyles. 



Read July 15, 1920. 



No attempt, so far as I am aware, has yet been made to 

 correlate the flora of Southern Rhodesia with that of the rest 

 of Africa south of the equator or with the northern tropical 

 region. Various botanists have from time to time published 

 papers dealing with the local flora. These incdude Marloth, 

 Swynnerton, Gibbs, Rand, Engler, Sim, Burtt-Davy and 

 Monro. 



* Geol. Map. 1887, 3. iv, pp. 22-24; aiso " Papers and Notes on the 

 Genesis and Matrix of tlie Diamond." T>ond,)n. 1897, p. 50. 



