ANALYSIS O F PROCESSES IN CONTROL O F INSECTS
31
tory factor. (There may be more than one of each, but for verbal
convenience I write here as if there were not.) A natural early step is
to try to identify these factors by looking for their effects on abundance.
To discern these effects by direct inspection we must look not only at
the adult or any other single stage, but at the whole life-cycle. When
we know at what stage the variation is imposed or the regulation
exerted, it may be possible to identify the factors responsible.
The appropriate data for this approach are tables of the successive
mortalities through the life-cycle. For this purpose, what we need are
not average or standard life-tables, but a series of tables, each of them
based on censuses in a different generation or year. They should preferably be life- and fertility-tables, showing the variation in reproduction as well as in survival.
To demonstrate these points, I shall examine the tables given by
Richards and Waloff (1961) for a population of the broom beetle
Phytodecta olivacea-in the grounds of the Imperial College Field Station
in Berkshire. The species disperses relatively little in this area, and the
population is considered to be nearly self-contained. It is associated
with a plantation of about two acres of broom, Xarothamnus scoparius.
The beetles emerge from the soil from late April to June and lay their
eggs on the plants. The eggs and the four larval instars are found on
the broom simultaneously for most of the time between May and the
end of August. The fully fed larvae descend and pupate in litter and
grass roots under the bushes. The adults emerge in autumn and return
to the plants to feed, but again descend to hibernate, remaining
immature until the spring. A fifth to a third of them survive another
year and hibernate a second time. The thoroughness of the sampling
and the cross-checking of estimates by different methods were impressive, and the paper gives a full account which repays study, but these
matters can be passed over for the present purpose.
Table I shows the annual changes in population density in five
successive phases of the life-cycle. During the &year period the amount
of broom, estimated in “armfuls”, was halved in the winter of 1955-6,
and almost halved again in the winter of 1957-8. Because of these
changes, it is particularly important to deal here with density rather
than with total numbers. I have therefore converted all the population
estimates to numbers per armful, even for insects on the ground, where
2-70 m2 corresponds on average to one armful of foliage above. In the
right-hand column of Table I, I have calculated the coefficient of
variation (100 x S.D./arith. mean) of the five annual values for each
phase of the life-cycle. Comparisons of these expressions of the degrees
of fluctuation suggest the following conclusions about the population
of Phytodecta over the period concerned,
31
tory factor. (There may be more than one of each, but for verbal
convenience I write here as if there were not.) A natural early step is
to try to identify these factors by looking for their effects on abundance.
To discern these effects by direct inspection we must look not only at
the adult or any other single stage, but at the whole life-cycle. When
we know at what stage the variation is imposed or the regulation
exerted, it may be possible to identify the factors responsible.
The appropriate data for this approach are tables of the successive
mortalities through the life-cycle. For this purpose, what we need are
not average or standard life-tables, but a series of tables, each of them
based on censuses in a different generation or year. They should preferably be life- and fertility-tables, showing the variation in reproduction as well as in survival.
To demonstrate these points, I shall examine the tables given by
Richards and Waloff (1961) for a population of the broom beetle
Phytodecta olivacea-in the grounds of the Imperial College Field Station
in Berkshire. The species disperses relatively little in this area, and the
population is considered to be nearly self-contained. It is associated
with a plantation of about two acres of broom, Xarothamnus scoparius.
The beetles emerge from the soil from late April to June and lay their
eggs on the plants. The eggs and the four larval instars are found on
the broom simultaneously for most of the time between May and the
end of August. The fully fed larvae descend and pupate in litter and
grass roots under the bushes. The adults emerge in autumn and return
to the plants to feed, but again descend to hibernate, remaining
immature until the spring. A fifth to a third of them survive another
year and hibernate a second time. The thoroughness of the sampling
and the cross-checking of estimates by different methods were impressive, and the paper gives a full account which repays study, but these
matters can be passed over for the present purpose.
Table I shows the annual changes in population density in five
successive phases of the life-cycle. During the &year period the amount
of broom, estimated in “armfuls”, was halved in the winter of 1955-6,
and almost halved again in the winter of 1957-8. Because of these
changes, it is particularly important to deal here with density rather
than with total numbers. I have therefore converted all the population
estimates to numbers per armful, even for insects on the ground, where
2-70 m2 corresponds on average to one armful of foliage above. In the
right-hand column of Table I, I have calculated the coefficient of
variation (100 x S.D./arith. mean) of the five annual values for each
phase of the life-cycle. Comparisons of these expressions of the degrees
of fluctuation suggest the following conclusions about the population
of Phytodecta over the period concerned,