154 Transactions of the South African Philosophical Society. 
Taking into consideration the composition of the fine earth, in 
which determinations of lime, potash, and phosphoric oxide were 
made, it may be said at once that in every case the soils taken from 
the low-lying ground were exceedingly poor in lime, and herein lies 
the great difference between the hillock soils and those below ; even 
samples B and C are very deficient in this respect, although con- 
siderably superior to samples A’, B’, and C’. A contains lime in 
fair amount. The potash present in A, B, and GC, is fair in quantity, 
but B’ is rather poor. A’ and C’ likewise show a moderate percent- 
age of potash, though in each case poorer than the corresponding 
samples from the hillocks. As far as the phosphates are concerned, 
there is a fair percentage in the hillock soils, and also, though to a 
_less extent, in sample C’, but A’ and B’ are decidedly poor. Clearly 
the chemical analysis tends to confirm the popular idea; and yet the 
difference all round is not as wide as some of the statements made 
might possibly have led one to expect. To this observation there is 
just the exception already noted—that of the lime. 
Physically, as well as chemically, the hillock soils are slightly 
superior. In water retentive power their average stands higher than 
that of the soils around. To this fact I have already alluded. The 
organic matter in the hillock soils is also better than in the others, 
and the former are likewise the richer in nitrogen. The inferiority 
of the low-lying soils also comes out in the amounts of common salt 
they contain, as indicated by the percentage of chlorine ; thus every 
single one of the seven sets of determinations made shows the soils 
from the hillocks to be better adapted for agriculture, they contain 
more of the six useful constituents and less of the one that is 
harmful. 
Some soils were also collected in the Wellington District, where a 
somewhat similar configuration prevails, and on the farm ‘‘ Groen- 
berg’’ a sample of soil E’ from level ground was taken, as well as 
one Ei from a hillock. Analysis seemed here also to confirm the 
popular theory, as may be seen from the following table :— 
No. | Water. | Qreamiec | chlorine. Potash. | Phgspaone 
Matter. Nitrogen.| Lime. 
E ‘76 5°07 "0255 ‘006 064 ae ‘O74 
ay ‘76 511 ‘006 003 ‘037 ‘084 ‘O77 
Of course the characteristics of the Wellington District are not 
strictly comparable with those of Malmesbury, but it is noteworthy 
that there is a marked difference in both the lime and potash of these 
