COMPOSITION OF SOME INDIGENOUS GRASSES. 2'1 i 



To take the case of the chief of the grasses of the chmax 

 veld — Anthistiria, imberbis {TJiemeda triandra) — we have samples 

 taken .on 1/3/20 and 24/2/22, which give practically identical 

 figui'es for protein, but the earlier sample is consistently but 

 slightly inferior to the latter one. The two samples of 4/12/20 

 and 21/12/21 may also be taken as comparable, and here the 

 agreement is remarkably close in all constituents. 



Thus the ordinary veld grass proves to be more constant in 

 composition than the more succulent ruderal grasses. 



As an extreme case of protein variation, Leersia hcxandra is 

 included. This year's sample contains more than double the 

 amount of ^^rotein as compared \A-ith that of two years ago. Both 

 samples came from typical marshy vlei land and from points at 

 no great distance apart. 



Seasonal Variations. 



In connection with the second part of the investigation, the 

 question of seasonal variations, seven of the commonest and most 

 promising species were selected and samples collected as before, 

 as soon as the flowering stage was reached, whilst anthers were 

 showing and before seed had set. Samples of these species were 

 then taken monthly, as regularly as possible consistent with the 

 observation of the weather conditions as already explained. 

 Generally three samples were taken before growth became too far 

 advanced for the present pui-pose. The results of this phase of 

 the investigation are shown in Table IV. They are rather vari- 

 able, but it is too soon tO' draw definite conclusions on the work 

 of a single year. However, certain points of interest are 

 indicated. 



As already noticed, the ordinary climax veld grass, as 

 instanced by Antliistiria imberbis, shows the greatest constancy, 

 its composition varying but little throughout the season. The 

 ruderal species show greater variation, due no doubt to the 

 greater susceptibility of their habitat, and of the species them- 

 selves, to climatic conditions 



Generally, the fibre increases and the protein decreases as the 

 season advances — for example, Digitaria sangiiinalis and Panicutn 

 proliferuni, where the change is progressive. In others the change 

 is irregular — for example, in Eleusine indica and Rottboellia 

 coyyipressa the second sample shows more fibre and less protein 

 than either the sample preceding or following it. In Panicum 

 laevifolium and Sctaria imberbis, on the other hand, crude protein 

 is highest in the intermediate stage, the former also showing here 

 an unusually high maximum of " amides," which normally 

 indicates the less mature herbage. 



In practically every species, however, taking only the first 

 and last samples of the species, the ash and protein diminish, 

 whilst the fibre and soluble carbohydrate increase, this being the 

 general tendency of maturing herbage. In the present instance, 

 although the aim was to secure samples at as even a stage of 



