40
M. E. SOLOMON
animals produce large numbers of young most of which die before
maturity. I n many such cases, the greater part of this mortality seems to
be density-independent ; a t least it is difficult to prove otherwise. Suppose in an insect population of 10,000 adults, half of them females, spread
Over a particular habitat, each female produces on average 200 young,
and that density-independent mortality always kills at least goyo, and
sometimes as many as 98%. If there were no subsequent mortality,
the increase would be tenfold at the lower level of mortality and still
twofold a t the higher level. Suppose however there is a sensitively and
promptly density-dependent factor regulating the survivors of this
mortality to approximately the same density as the parent generation.
This regulating factor kills a relatively small percentage of the original
number of young, yet it is entirely responsible for regulation. This is
one way of looking a t the matter. But it may well be that this regulating factor can maintain control only over a limited range of abundance, so that the stability of the system is dependent on the heavy
preliminary mortality as well as on the “finishing touches” by the
regulating factor.
Of course, these “finishing touches” seem to be small because we
have assessed mortality in terms of the original numbers of young. This
is a somewhat unrealistic way of looking at the action of a densitydependent factor, even in an over-tidy example of the sort we are
considering. For its action (when not of the lagging type) is dependent
on the density of population at the time, not on the initial density.
If there has been a preliminary mortality of go%, the regulatory factor
will operate on a population of 100,000, which for perfect stability, and
assuming no other mortalities in the life-cycle, it will have to reduce to
10,000 -a reduction of 90%. If, on the other hand, the preliminary
mortality is as much as 98%, the regulatory factor will have to reduce
20,000 individuals to 10,000 by a 50% mortality.
I n brief, if a regulatory factor operates by “finishing off” after most
of the young have already been killed, it will kill only a small percentage
of the original numbers, but this may be a high percentage of the
numbers present a t the time when it acts. Certainly it must have the
capacity to kill a high proportion of those present when the residual
density is higher than usual, if it is to maintain effective regulation.
A real example of these relationships can be selected from the data
of Richards and Waloff (1961) on the broom beetle (cf. Table I). The
original high numbers of eggs were reduced to low numbers of adults
in the autumn. These were reduced to even lower numbers by the
following spring, as also shown in Table I. The halving of the value of
the coefficient of variation from autumn to spring strongly suggests
the operation of a regulatory process, and this is supported by Fig. 15.
M. E. SOLOMON
animals produce large numbers of young most of which die before
maturity. I n many such cases, the greater part of this mortality seems to
be density-independent ; a t least it is difficult to prove otherwise. Suppose in an insect population of 10,000 adults, half of them females, spread
Over a particular habitat, each female produces on average 200 young,
and that density-independent mortality always kills at least goyo, and
sometimes as many as 98%. If there were no subsequent mortality,
the increase would be tenfold at the lower level of mortality and still
twofold a t the higher level. Suppose however there is a sensitively and
promptly density-dependent factor regulating the survivors of this
mortality to approximately the same density as the parent generation.
This regulating factor kills a relatively small percentage of the original
number of young, yet it is entirely responsible for regulation. This is
one way of looking a t the matter. But it may well be that this regulating factor can maintain control only over a limited range of abundance, so that the stability of the system is dependent on the heavy
preliminary mortality as well as on the “finishing touches” by the
regulating factor.
Of course, these “finishing touches” seem to be small because we
have assessed mortality in terms of the original numbers of young. This
is a somewhat unrealistic way of looking at the action of a densitydependent factor, even in an over-tidy example of the sort we are
considering. For its action (when not of the lagging type) is dependent
on the density of population at the time, not on the initial density.
If there has been a preliminary mortality of go%, the regulatory factor
will operate on a population of 100,000, which for perfect stability, and
assuming no other mortalities in the life-cycle, it will have to reduce to
10,000 -a reduction of 90%. If, on the other hand, the preliminary
mortality is as much as 98%, the regulatory factor will have to reduce
20,000 individuals to 10,000 by a 50% mortality.
I n brief, if a regulatory factor operates by “finishing off” after most
of the young have already been killed, it will kill only a small percentage
of the original numbers, but this may be a high percentage of the
numbers present a t the time when it acts. Certainly it must have the
capacity to kill a high proportion of those present when the residual
density is higher than usual, if it is to maintain effective regulation.
A real example of these relationships can be selected from the data
of Richards and Waloff (1961) on the broom beetle (cf. Table I). The
original high numbers of eggs were reduced to low numbers of adults
in the autumn. These were reduced to even lower numbers by the
following spring, as also shown in Table I. The halving of the value of
the coefficient of variation from autumn to spring strongly suggests
the operation of a regulatory process, and this is supported by Fig. 15.