THE INFLUENCE OF HARVESTER ANT MESSOR-CAPENSIS NEST-MOUNDS ON THE PRODUCTIVITY AND DISTRIBUTION OF SOME PLANT-SPECIES IN THE SOUTHERN KAROO, SOUTH-AFRICA
Wrj. Dean et Ri. Yeaton, THE INFLUENCE OF HARVESTER ANT MESSOR-CAPENSIS NEST-MOUNDS ON THE PRODUCTIVITY AND DISTRIBUTION OF SOME PLANT-SPECIES IN THE SOUTHERN KAROO, SOUTH-AFRICA, Vegetatio, 106(1), 1993, pp. 21-35
Nest-mounds of the harvester ant Messor capensis occur on and around n
utrient-rich patches, along minor drainage lines in nutrient-rich soil
s, and on the plains, generally in nutrient-poor soils. Nest-site sele
ction is related to the presence of suitable deep soils, the presence
of stones and the distance from the nearest neighbouring nest. Two pla
nt species, Galenia fruticosa and Pteronia pallens, were significantly
associated with Messor capensis nest-mounds, both in numbers of mound
s occupied and in numbers of individuals. A third species, Drosanthemu
m montaguense, was also more common in numbers of individuals, while a
fourth species, Rhinephyllum macradenium was negatively associated wi
th these mounds. The analysis of species guilds by soil type shows tha
t significantly more species of nutrient-rich soils are present on M.
capensis nest-mounds. Also, significantly more taller, woody species o
ccurred on nest-mounds than in inter-mound spaces. Two species, Pteron
ia pallens and Osteospermum sinuatum, growing on ant nest-mounds had s
ignificantly longer inter-nodes than the same species growing off moun
ds. However, two other species, Pteronia cf. empetrifolia and Galenia
fruticosa showed no difference in inter-node lengths between plants gr
owing on and off mounds. Six of the nine species of plants sampled on
ant nest-mounds had significantly higher seed production than plants o
f the same species growing in inter-mound spaces. The other three spec
ies showed a tendency towards more seeds per plant on ant nest-mounds.
The proportions of live and dead plants on mounds differed between sp
ecies. Only Ruschia spinosa showed a significant difference between th
e numbers of dead plants in the population on and off mounds, with mor
e dead plants occurring on mounds. Significantly more seeds set on ind
ividuals of Pteronia pallens growing on ant nest-mounds than those gro
wing off nest-mounds, but no such difference occurred in P. cf. empetr
ifolia. There was no significant difference in the proportion of seeds
parasitized by the tephritid fly Desmella anceps for individuals of P
. pallens and P. cf. empetrifolia growing on and off mounds.