DYNAMICS O F FIELD POPULATION O F P . N E LOOPER
289
8, giving rise to the irregular density fluctuations presented in Fig. 27.
That this is not so is demonstrated by another method, which we have
to consider next.
In general, regulation might be expected to be in operation when the
population declines after having reached a high level, or increases
sharply after a very low level. The years in which this occurred are
shown in Fig. 27, and Fig. 31 demonstrates that the density induced
variability of fecundity in these cases is far from being a predominating
factor in the complex of agents involved in the density changes. This
strongly suggests that the population is under the control of additional
regulating factors, and the proof of this will now be given.
It was pointed out on pp. 287-288, that the coefficient of regression
of log Et+l on log Et (Fig. 33-A) tends to be unit,y when numbers are
wholly determined by density independent ageds. This coefficient
proved to be 0.13 only, indicating strong density related action (see
p. 288). In Table XXV it is shown that the gradual addition of the ageinterval mortalities gives a remarkable improvement to the regression
coefficient, whereas adding the reduced fecundity increases the coefficient with a final 4% only. This proves that most ofthe regulation must
be concealed in the mortality factors.
Taking all the evidence together, it demonstrates that density dependent fecundity is of minor importance in the population dynamics
of the pine looper. However, in a later section (p. 297) evidence is presented suggesting that the immediate cause of the fecundity reduction,
namely the increase in the contact interference between larvae at the
higher densities, has additional effects on the viallility of the insects,
and these may well appear to be of primary significance in regulation.
I
c. D E N S I T Y D E P E N D E N T MORTALITY
The regulating agents said to be concealed in the mortality factors
may be density dependent or delayed density dependent. We are concerned here with the analysis of the former category. In Section VII-D
attention is particularly directed to the delayed processes.
1. Mortality in the Egg Stage
Comparing Fig. 33-B with 33-A shows that there is a rise of the
regression coefficient of 0.13 to 0.38 (Table XXV), if the effect of egg
mortality is added to egg density. This suggests that egg mortality
includes a source of compensatory reaction. To check whether this reaction is of the direct density dependent type we plotted k, over the
logarithm of egg density, and it was found that k, increases slowly with
289
8, giving rise to the irregular density fluctuations presented in Fig. 27.
That this is not so is demonstrated by another method, which we have
to consider next.
In general, regulation might be expected to be in operation when the
population declines after having reached a high level, or increases
sharply after a very low level. The years in which this occurred are
shown in Fig. 27, and Fig. 31 demonstrates that the density induced
variability of fecundity in these cases is far from being a predominating
factor in the complex of agents involved in the density changes. This
strongly suggests that the population is under the control of additional
regulating factors, and the proof of this will now be given.
It was pointed out on pp. 287-288, that the coefficient of regression
of log Et+l on log Et (Fig. 33-A) tends to be unit,y when numbers are
wholly determined by density independent ageds. This coefficient
proved to be 0.13 only, indicating strong density related action (see
p. 288). In Table XXV it is shown that the gradual addition of the ageinterval mortalities gives a remarkable improvement to the regression
coefficient, whereas adding the reduced fecundity increases the coefficient with a final 4% only. This proves that most ofthe regulation must
be concealed in the mortality factors.
Taking all the evidence together, it demonstrates that density dependent fecundity is of minor importance in the population dynamics
of the pine looper. However, in a later section (p. 297) evidence is presented suggesting that the immediate cause of the fecundity reduction,
namely the increase in the contact interference between larvae at the
higher densities, has additional effects on the viallility of the insects,
and these may well appear to be of primary significance in regulation.
I
c. D E N S I T Y D E P E N D E N T MORTALITY
The regulating agents said to be concealed in the mortality factors
may be density dependent or delayed density dependent. We are concerned here with the analysis of the former category. In Section VII-D
attention is particularly directed to the delayed processes.
1. Mortality in the Egg Stage
Comparing Fig. 33-B with 33-A shows that there is a rise of the
regression coefficient of 0.13 to 0.38 (Table XXV), if the effect of egg
mortality is added to egg density. This suggests that egg mortality
includes a source of compensatory reaction. To check whether this reaction is of the direct density dependent type we plotted k, over the
logarithm of egg density, and it was found that k, increases slowly with