POPULATION CYCLES IN SMALL MAMMALS
293
Finally, several authors have suggested that litter size is reduced in
peak populations but essentially the same during the other phases of
the cycle. Hoffmann (1958) reported a 1 6 2 5 % drop in litter size during
the peak summer for a Microtus montanus population. Tanaka (1964)
shows a 17% drop in litter size during the peak summer for Clethrionomys smithi in Japan. Patric (1962) claims to have shown a 1 6 3 7 %
drop in litter size for C. gapperi populations; his estimates, however,
assume no significant effects of season, weight of female, or parity on
litter size. Zejda (1964) found a slight depression in litter size of C.
glareolus in the peak summer. Keller and Krebs (1970) found that litter
size was depressed 25% during the peak summer in Microtus ochrogaster.
Koshkina (1966) states that Clethrionomys rufocunw has minimal litter
size in the peak year, compared with the increase phase, but also
minimal in the decline and low phases. Unfortunately the data presented
are not sufficient to verify these claims because only mean values for
the whole year are given. Reichstein (1964) shows that litter sizes of
Microtus arvalis are reduced by 17-22% in peak years.
If the changes in litter size are to be an important driving force in the
population cycle, litter size should be depressed in the decline phase and
enhanced in the increase phase. We can find little evidence of these
trends in the populations studied to date, and we conclude that this
component of reproduction is not a critical link in the demographic
machinery.
2. Pregnancy rate
We next consider the percentage of mature females which are
pregnant during the breeding season. Note that we are not concerned
here with the age of sexual maturity or the length of the breeding
season.
Most workers seem to agree that the pregnancy rate does not vary in
relation to the population cycle. Figure 15 illustrates the similar
proportions of pregnant females in three years for Clethrionomys
rufocanus in Finland. Mullen (1965) found the same result in Lemmw
trimucronatus a t Barrow, Alaska over four years. Krebs (19644 found
no cyclic variation in the percentage of females pregnant for either
Dicrostonyx groenlandicus or Lemmus trimucronatw in northern
Canada. Keller and Krebs (1970) found no significant differences in
percentage of mature females pregnant during the summer months for
Microtus pennsylvanicus and M . ochrogaster.
A few authors have suggested that the pregnancy rate goes up in the
phase of increase. Hamilton (1937) measured the fraction of female M .
pennsylvanicus in New York that were both pregnant and lactating and
293
Finally, several authors have suggested that litter size is reduced in
peak populations but essentially the same during the other phases of
the cycle. Hoffmann (1958) reported a 1 6 2 5 % drop in litter size during
the peak summer for a Microtus montanus population. Tanaka (1964)
shows a 17% drop in litter size during the peak summer for Clethrionomys smithi in Japan. Patric (1962) claims to have shown a 1 6 3 7 %
drop in litter size for C. gapperi populations; his estimates, however,
assume no significant effects of season, weight of female, or parity on
litter size. Zejda (1964) found a slight depression in litter size of C.
glareolus in the peak summer. Keller and Krebs (1970) found that litter
size was depressed 25% during the peak summer in Microtus ochrogaster.
Koshkina (1966) states that Clethrionomys rufocunw has minimal litter
size in the peak year, compared with the increase phase, but also
minimal in the decline and low phases. Unfortunately the data presented
are not sufficient to verify these claims because only mean values for
the whole year are given. Reichstein (1964) shows that litter sizes of
Microtus arvalis are reduced by 17-22% in peak years.
If the changes in litter size are to be an important driving force in the
population cycle, litter size should be depressed in the decline phase and
enhanced in the increase phase. We can find little evidence of these
trends in the populations studied to date, and we conclude that this
component of reproduction is not a critical link in the demographic
machinery.
2. Pregnancy rate
We next consider the percentage of mature females which are
pregnant during the breeding season. Note that we are not concerned
here with the age of sexual maturity or the length of the breeding
season.
Most workers seem to agree that the pregnancy rate does not vary in
relation to the population cycle. Figure 15 illustrates the similar
proportions of pregnant females in three years for Clethrionomys
rufocanus in Finland. Mullen (1965) found the same result in Lemmw
trimucronatus a t Barrow, Alaska over four years. Krebs (19644 found
no cyclic variation in the percentage of females pregnant for either
Dicrostonyx groenlandicus or Lemmus trimucronatw in northern
Canada. Keller and Krebs (1970) found no significant differences in
percentage of mature females pregnant during the summer months for
Microtus pennsylvanicus and M . ochrogaster.
A few authors have suggested that the pregnancy rate goes up in the
phase of increase. Hamilton (1937) measured the fraction of female M .
pennsylvanicus in New York that were both pregnant and lactating and