104 



The following are ruderal species : Digitaria sanguinalis, 

 Panicum crus-galli, Setaria verticillata, Cynodon dactylon, 

 Phalaris minor, Eleusine indica, Bromus commutatus , B. 

 patulus, Brachypodiwrn distachyiim, Lolium multiflorum, 

 Eragrostis major. 



In the mountain ranges of the Karroo region — the Rogge- 

 Teld, Komsberg, Nieuwveld, and Sneeuwberg, most of the 

 species enumerated above occur, bat at altitudes of 4,000- 

 8,000 feet, there are considerable differences in the environ- 

 mental factors, and a number of other interesting species are 

 found. The Roggeveld are said by Thunberg to have been so 

 named from the wild rye grass, Secale africanum, but it does 

 not seem to be common. It is unfortunate that all those great 

 ranges have not been better explored botanically. 



It has already been pointed out that" the South Western 

 species, or nearly allied species, belonging to the same tem- 

 perate tribes, tend to spread eastward along them. Danthonia 

 is represented by D. stricta, D. purpurea, D. cincta, D. 

 disticha. There are two or three species of Achneria (A. seti- 

 folia, A. viicrofhylla, A. hirsutaj. Pentachistis has P. angus- 

 tifolia var. cirrhulosa, P. densifolia, P. aeroides. The genus 

 Melica, of the tribe Festuceae, is peculiarly characteristic of 

 this region. There are half-a-dozen species, M. racemosa, 

 M. holusii, M. ovalis, M. decumbens, M. pumila, M. neesii. 



Other mountain species include the bok-baard, Festuca 

 caprina (a useful pasture plant), F. scahra, Antlioxanthum 

 ecklonii, Tetrachne dre.gei, Fingerli.uthia africana, Microchloa 

 caffra, Harpechloa capensis, Koeleria cristata, Andropogon 

 hirtus, Anthistiria vmberbis. 



THE TRANSITIONAL BELT BETWEEN KARROO AND 

 EASTERN GUASSVELD. 



This belt is of particular interest because it gives us an 

 opportunity of studying tj^^ establishment of definite Grass- 

 veld, of the type which extends over the whole eastern side 

 of South Africa. It is very difficult to draw any definite 

 boundary line between Karroo and Grassveld, since every- 

 where in the transitional belt there are patches of Karroo and 

 Composite Veld interspersed with grassland, or towards the 

 west it is better to say that patches of grassland are inter- 

 spersed with Karroo. Further, owing to the effects of over- 

 stocking, the grasses are often eaten out, the surface layers of 



