DYNAMICS OF FIELD POPULATION O F PI-SE LOOPER
27 1
By chance the density estimate of a particular stage may turn out to be higher
than that in an earlier stage. Examples of this have already been given in Tables
IV and VII, and are more frequent in the life tables. Thus in 1953-54 nymphal
density was estimated to be 2.5, whereas one month' later pupal density appeared
to be 3.0. Such chance effects may even occur in sitiiation!3 where by the application of direct measurements mortality factors could be shc wn to operate. Thus in
1951-52 pupal density in April amounted to 3-6/mL, in D'xember to 3.5. Nevertheless 2.2% of the pupae collected in April had been killed, in the main by
predators during the previous winter. Similar!y, in 1957-58 the small difference between the density measurements of pupae in April and of moths in June suggests
a relatively low mortality rate in that interval of time, but the rearing of pupae
proved that more than 5094 of them were parasitized.
Another source of inaccuracy is caused by rounding off. Thus, in December and
April 1959-60, pupal density was estimated to be 1-19 amd 1-13, respectively.
The size of the confidence intervals shows (Table VITI) that t,liere is no sense in
presenting the estimations with an accuracy of two decimril places, talius resulting
in 1.2 and 1.1 specimens/iii', respectively. The wint,er inortality of pupae as
measured in April can only account for a difference of 0.06, which might be
rounded off to 0.1 specimen/m'.
Finally, larval density was not measured in Oc!t,obcr 19C.O and 1961 (Table IV),
and nymphal density not measured in 1950, 1951, and 1952 (Table V). Pupal
density in December was measured directly only to 1!h53 inclusive (p. 223).
Fecundity was not measured in 1961 and 1964, but determined by interpolation
in Fig. 25. Larval parasitism due to Apanteles was not rtssessed in 1953, 1957,
1958, and 1959 (p. 237 and Table XIV).
One point which is cleady brought to light in bhe tables is the incompleteness of the analysis, as shown by the recurrent use of the word 1
miscellaneous. It indicates mostly that the mortality is due to unknown
causes, and in addition that several non-separa ble factors operate
simultaneously. All cases in which the term is used will be considered
briefly.
( a ) Egg Mortality. The miscellaneous mortality lisited in the life tables
is specified in Table XI11 and discussed on p. 234.
( b ) First Instar Larval Mortality. This mortality rate is computed as a
difference between the July and August densitieri. The former value
results from the product of the egg density and the riurvival rate of eggs;
the latter is a direct estimate made by sampling trees in the study area.
The mortality of the tiny larvae is nearly alwa,ys very high, but is
difficult to measure directly because dead specimens shrivel up, leaving
extremely small head capsules, which are hardly detectable amidst all
other debris. The causes of their deaths can be manifold, but are probably mainly abiotic. A possible biotic factor is a spider, Xysticus audax.
This thomisid searches for prey on the needles and it readily accepts
first instar caterpillars as food.
27 1
By chance the density estimate of a particular stage may turn out to be higher
than that in an earlier stage. Examples of this have already been given in Tables
IV and VII, and are more frequent in the life tables. Thus in 1953-54 nymphal
density was estimated to be 2.5, whereas one month' later pupal density appeared
to be 3.0. Such chance effects may even occur in sitiiation!3 where by the application of direct measurements mortality factors could be shc wn to operate. Thus in
1951-52 pupal density in April amounted to 3-6/mL, in D'xember to 3.5. Nevertheless 2.2% of the pupae collected in April had been killed, in the main by
predators during the previous winter. Similar!y, in 1957-58 the small difference between the density measurements of pupae in April and of moths in June suggests
a relatively low mortality rate in that interval of time, but the rearing of pupae
proved that more than 5094 of them were parasitized.
Another source of inaccuracy is caused by rounding off. Thus, in December and
April 1959-60, pupal density was estimated to be 1-19 amd 1-13, respectively.
The size of the confidence intervals shows (Table VITI) that t,liere is no sense in
presenting the estimations with an accuracy of two decimril places, talius resulting
in 1.2 and 1.1 specimens/iii', respectively. The wint,er inortality of pupae as
measured in April can only account for a difference of 0.06, which might be
rounded off to 0.1 specimen/m'.
Finally, larval density was not measured in Oc!t,obcr 19C.O and 1961 (Table IV),
and nymphal density not measured in 1950, 1951, and 1952 (Table V). Pupal
density in December was measured directly only to 1!h53 inclusive (p. 223).
Fecundity was not measured in 1961 and 1964, but determined by interpolation
in Fig. 25. Larval parasitism due to Apanteles was not rtssessed in 1953, 1957,
1958, and 1959 (p. 237 and Table XIV).
One point which is cleady brought to light in bhe tables is the incompleteness of the analysis, as shown by the recurrent use of the word 1
miscellaneous. It indicates mostly that the mortality is due to unknown
causes, and in addition that several non-separa ble factors operate
simultaneously. All cases in which the term is used will be considered
briefly.
( a ) Egg Mortality. The miscellaneous mortality lisited in the life tables
is specified in Table XI11 and discussed on p. 234.
( b ) First Instar Larval Mortality. This mortality rate is computed as a
difference between the July and August densitieri. The former value
results from the product of the egg density and the riurvival rate of eggs;
the latter is a direct estimate made by sampling trees in the study area.
The mortality of the tiny larvae is nearly alwa,ys very high, but is
difficult to measure directly because dead specimens shrivel up, leaving
extremely small head capsules, which are hardly detectable amidst all
other debris. The causes of their deaths can be manifold, but are probably mainly abiotic. A possible biotic factor is a spider, Xysticus audax.
This thomisid searches for prey on the needles and it readily accepts
first instar caterpillars as food.