ANALYSIS O F PROCESSES I N CONTROL O F INSECTS
51
fecundity at the higher host densities would inevitably lead to greater
increases of the host (of. Tinbergen and Klomp, 1960).
B. DO PARASITE-HOST OSCILLATIONS TEND TO INCREASE
I N AMPLITUDE?
Experimental studies have provided a little support for the conclusion of Nicholson and Bailey that parasite-host interactions tend
towards increasingly violent oscillations. In the early experiments of
Gause (1934) with the predatory mite Cheyletus eruditus (Schr.) and an
Acarid mite as prey, the predators killed all the prey and then died out.
This has happened in more recent laboratory or greenhouse experiments with mite predators and prey (e.g. Bravenboer and Dosse, 1962,
with mites on peach in a greenhouse).
Burnett (1958a), in experiments with Trileurodes and Encarsia
in a small cage, produced two cycles with’an increase in amplitude in
the second. However, this experiment was so highly manipulated
(somewhat in the style of the DeBach and Smith experiments), that it
seems closer to the theoretical model than to any set of natural conditions. In a brief account of experiments with the same animals on
tomato plants in a greenhouse, Burnett (1959) reported that “in general,
the growth forms of the host and parasite populations exhibited
fluctuations of increasing amplitude”. However, the experiments
stopped after three oscillations, and these did not always increase in
amplitude. At least one experimenter (Utida, 1953, 1955a,b,c) has
reported parasite-host oscillations that did not increase in size. He used
pulse beetles Callosobruchus spp. and Hymenopterous parasites in small
dishes of bean seeds, in which the host increase was restricted by
crowding; Burnett (1959) suggested there must also have been some
form of protection of the hosts.
Burnett (1959) has.expressed the opinion that in some of the parasitehost or predator-prey experiments in the laboratory too little space
has been allowed for the populations to become properly established.
Nicholson (1954b) expressed a similar opinion about the well-known
experiments of Gause (1934) with Paramecium and Didinium. Varley
and Edwards (1957) raised a series of objections to the procedure used
in the experiments of DeBach and Smith (1941a,b, 1947) on the housefly and Mormoniella. There is no doubt that the conditions imposed by
experimental design need scrutinizing as closely as the assumptions of a,
theoretical model.
C. HOW MIGHT EXPANDING OSCILLATIONS B E DAMPED?
Another approach to the Nicholson-Bailey model and its prediction
of increasingly violent oscillations has been to assume that it may apply
51
fecundity at the higher host densities would inevitably lead to greater
increases of the host (of. Tinbergen and Klomp, 1960).
B. DO PARASITE-HOST OSCILLATIONS TEND TO INCREASE
I N AMPLITUDE?
Experimental studies have provided a little support for the conclusion of Nicholson and Bailey that parasite-host interactions tend
towards increasingly violent oscillations. In the early experiments of
Gause (1934) with the predatory mite Cheyletus eruditus (Schr.) and an
Acarid mite as prey, the predators killed all the prey and then died out.
This has happened in more recent laboratory or greenhouse experiments with mite predators and prey (e.g. Bravenboer and Dosse, 1962,
with mites on peach in a greenhouse).
Burnett (1958a), in experiments with Trileurodes and Encarsia
in a small cage, produced two cycles with’an increase in amplitude in
the second. However, this experiment was so highly manipulated
(somewhat in the style of the DeBach and Smith experiments), that it
seems closer to the theoretical model than to any set of natural conditions. In a brief account of experiments with the same animals on
tomato plants in a greenhouse, Burnett (1959) reported that “in general,
the growth forms of the host and parasite populations exhibited
fluctuations of increasing amplitude”. However, the experiments
stopped after three oscillations, and these did not always increase in
amplitude. At least one experimenter (Utida, 1953, 1955a,b,c) has
reported parasite-host oscillations that did not increase in size. He used
pulse beetles Callosobruchus spp. and Hymenopterous parasites in small
dishes of bean seeds, in which the host increase was restricted by
crowding; Burnett (1959) suggested there must also have been some
form of protection of the hosts.
Burnett (1959) has.expressed the opinion that in some of the parasitehost or predator-prey experiments in the laboratory too little space
has been allowed for the populations to become properly established.
Nicholson (1954b) expressed a similar opinion about the well-known
experiments of Gause (1934) with Paramecium and Didinium. Varley
and Edwards (1957) raised a series of objections to the procedure used
in the experiments of DeBach and Smith (1941a,b, 1947) on the housefly and Mormoniella. There is no doubt that the conditions imposed by
experimental design need scrutinizing as closely as the assumptions of a,
theoretical model.
C. HOW MIGHT EXPANDING OSCILLATIONS B E DAMPED?
Another approach to the Nicholson-Bailey model and its prediction
of increasingly violent oscillations has been to assume that it may apply