Effects of a Bush Fire on the Vegetation of Signal Hill. 227 
very local in their distribution and formed more or less isolated patches 
all over the area. It was obvious that these seedlings were all growing 
close to the spot that had borne the parent plant (or plants), (fig. 5). 
Another example of this localised distribution of seedlings was found in 
Hakea suavolens. In this case a single, large shrub near the top of the area 
had been burnt, but owing to its very woody nature it could be recognised. 
All round the burnt plant was a host of seedlings, and it called to mind 
a note made by Dr. Harvey in Australia (2) that certain of the Proteaceae 
native to that country only shed their seeds after prolonged basking in 
the sun, or after bush fires. Hakea is one of the genera mentioned in this 
connection. The seeds of members of the Compositae on the area (with 
the exception of Osteospermum moniliferum) are all wind scattered, and 
consequently their distribution was much more general. In point of numbers 
the plants of Elytropappus Rhinocerotis were far in advance of any others, 
and, with the exception of the northern slopes of the valleys, they were evenly 
distributed all over the area. This suggests that, as in the case of the 
Leguminosae, the seeds had withstood the fire and had not been introduced 
from plants growing in the adjoining unburnt part. Had this been the case, 
one would have expected to find the numbers of plants at the margin in 
excess of those near the centre. 
The northern and southern slopes of the valleys showed a somewhat 
striking difference in the bushes which grew on them. The southern slopes 
were the more sparsely populated, and the chief bushes were Rhus lucida, 
Elytropappus Rhinocerotis, Athanasia trifurcata, and Borbonia cordata. 
The northern slopes differed from the southern in the diminution in numbers 
of Elytropappus Rhinocerotis and Athanasia trifurcata., and the presence of 
Myrsine africana, Anthospernmm ciliare, Rhus angustifolia, Stoebe alopecu- 
roides, and Royena glabra. 
Asparagus capensis, the young shoots of which in the previous year 
had been such a feature of the landscape, was most inconspicuous. After 
a careful survey of the area it was clear that the number of plants was about 
the same as in 1919, but this year no young shoots had appeared before the 
winter rains, and the old dull-green shoots were easily overlooked. Another 
noteworthy feature was the absence of flowers in Haemanthus coccineus. 
On the other hand, Andropogon hirtus and Hibiscus aethiopicus were in full 
flower, as in the previous year. A plant that had either been absent or 
overlooked in 1919 was Linum thesioides, which was rather sparsely distri- 
buted on the southern slopes of valley A. 
The area was visited once more in June, but there was little to record. 
The early months of 1920 were remarkably dry, and the whole vegetative 
period was somewhat later than usual over the whole Peninsula. The 
same plants were found as in 1919 (with the exception of the 
