OIKOS 95: 185–197. Copenhagen 2001

Mechanisms for delayed density-dependent reproductive traits infield voles, Microtus agrestis : the importance of inheritedenvironmental effects

Torbjørn Ergon, James L. MacKinnon, Nils Chr. Stenseth, Rudy Boonstra and Xavier Lambin

Ergon, T., MacKinnon, J. L., Stenseth, N. C., Boonstra, R. and Lambin, X. 2001.Mechanisms for delayed density-dependent reproductive traits in field voles, Microtusagrestis : the importance of inherited environmental effects. – Oikos 95: 185–197.

Reproductive traits of voles vary with the phases of the population density fluctua-tions. We sought to determine whether the source of this variation resides in theindividuals or in their environment. Overwintering field voles (Microtus agrestis) fromtwo cyclic out-of-phase populations (increase and peak phases) were sampled in earlyspring and bred in the laboratory for two generations under standardised conditionswith ambient light and temperature. Monitoring of the source populations bycapture-mark-recapture showed large differences in reproductive performance. In theincrease area, reproduction started six weeks earlier, the probability of maturation ofyoung-of-the-year was more than ten times higher during mid-summer, and reproduc-tion continued nearly two months later in the autumn than in the peak area. Thesedifferences were not found to be associated with a difference in age structure ofoverwintered animals between the two areas (assessed by the distribution of eye lensmasses from autopsy samples). Although the population differences in reproductivetraits were to some degree also present among the overwintered animals in thelaboratory, we found no difference in reproductive traits in the laboratory-borngenerations. There was a strongly declining seasonal trend in probability of sexualmaturation both in the field and in the laboratory under ambient light conditions.However, in the field there were large population differences in the steepness of theseasonal decline that were not seen under the standardised laboratory conditions. Weconclude that seasonal decline in maturation rates is governed by change in photope-riod, but that the population level variation in the shape of the decline is caused bya direct response to the environment and not due to variation in any intrinsic stateof the individuals reflecting the environment experienced by the previous generation(s).

T. Ergon and N. C. Stenseth (correspondence), Di�. of Zoology, Dept of Biology, Uni�.of Oslo, P.O. Box 1050, Blindern, N-0316 Oslo, Norway (n.c.stenseth@bio.uio.no). –J. L. MacKinnon and X. Lambin, Dept of Zoology, Uni�. of Aberdeen, TillydroneA�enue, Aberdeen, UK AB24 2TZ. – R. Boonstra, Di�. of Life Sciences, Uni�. ofToronto, 1265 Military Trail, Scarborough, ON, Canada M1C 1A4.

Since the first scientific description of small rodentpopulation cycles by Elton (1924), much variation hasbeen documented in the density fluctuations of differentpopulations. Through time-series analysis, this varia-tion has been described in terms of differences in thestrength of direct and delayed density dependence onpopulation growth rate (Bjørnstad et al. 1995, Turchin1995, Stenseth et al. 1996, Stenseth 1999). However, less

is known about the demographic mechanisms of theregulation. Indeed, the ecological mechanisms of thelarge variation in life histories of individuals withinmany animal populations are poorly understood (Mc-Namara and Houston 1996).

In fluctuating small rodent populations, there is pro-found between-year variation in body size, timing ofmaturation and reproductive performance of individu-

Accepted 16 May 2001

Copyright © OIKOS 2001ISSN 0030-1299Printed in Ireland – all rights reserved

OIKOS 95:2 (2001) 185

als. In years with increasing population densities, over-wintering animals generally start to breed earlier in thespring and more animals mature in their year of birththan in other years (Krebs and Myers 1974, Hanssonand Henttonen 1985, Bernshtein et al. 1989, Gilbertand Krebs 1991, Boonstra 1994). This general life-his-tory pattern seems to be a universal characteristic ofmicrotine population fluctuations and is also seen insome non-cyclic but multi-annually fluctuating popula-tions (Agrell et al. 1992).

If the properties of individuals vary in relation toprevious densities, there must be a ‘memory’ within thesystem. This memory may reside in the environment(interactions with predators, parasites or food re-sources) and/or within the population itself. The popu-lation may ‘remember’ past conditions in two ways.First, environmental conditions changing over timemay affect demographic processes that alter the agestructure of the population, and this may in turn affectfuture demographic rates. Boonstra (1994) and Tkadlecand Zejda (1998) found that the demographic processescommonly observed in fluctuating small rodent popula-tions cause a shift in age structure towards older ani-mals after peak population densities, and they arguedthat declines become inevitable because of senescence(i.e., the ‘senescence hypothesis’). Second, environmen-tal conditions may cause a variation in the internalstates of the individuals, through genetic selection and/or through persistent changes in the individuals’ physi-ological state. Assumptions of hypotheses forpopulation regulation involving fluctuating genetic se-lection that generates delayed density dependence inreproduction, like Chitty’s (1960, 1967) polymorphicgenetic-behavioural hypothesis and its variants (Krebs1978), have not been supported empirically, and in-traspecific variation in life-history traits of small ro-dents does not seem to have an important genetic basis(Boonstra and Boag 1987, Boonstra and Hochachka1997). However, hypotheses involving time lags main-tained by phenotypic changes and maternal effects havebeen little explored until recently. The environment inearly life, including maternal effects operating duringgestation and lactation, is often an important determi-nant of life histories in mammals (Bernardo 1996,Rossiter 1996, Inchausti and Ginzburg 1998), includingmicrotines (Boonstra and Boag 1987, Boonstra andHochachka 1997, Hansen and Boonstra 2000).

In this paper we examine the proximate causes ofvariation in reproductive performance of overwinteringanimals and their offspring in cyclic populations of fieldvoles (Microtus agrestis). We tested whether variationin reproductive traits seen in the field can be explainedby mechanisms whereby the memory of past conditionsresides in the individual voles. Overwintering volesfrom two areas that differed in previous densities werebred under standardised conditions in the laboratoryalongside monitoring the source populations in the

field. A similar experiment was undertaken by Mihokand Boonstra (1992) on a fluctuating population of M.pennsyl�anicus. However, whereas Mihok and Boonstra(1992) sampled voles from two different years (declineand increase) in the same area, we sampled voles simul-taneously from two cyclic out-of-phase populations.First, we tested if variation in breeding performanceseen in the field was maintained under standardisedlaboratory conditions. Secondly, we bred the voles fortwo generations to test whether the population differ-ences would be reinforced by genetic and/or maternaleffects. By comparing the age distributions through thedistribution of eye lens masses in autopsy samples (cf.Hagen et al. 1980), we sought to assess the potential forsenescence in causing the population differences.

Material and methods

Study system

Field voles (Microtus agrestis) were sampled from theKielder and Kershope forests on the border betweenEngland and Scotland (55°13� N, 2°33� W). These man-made conifer forests have been planted over the last 70yr and now have a mosaic of clear-cuts that are sepa-rated by dense spruce forest. Voles only inhabit thegrassland clear-cuts, which are connected by roadverges and fire breaks throughout the forest. Themesotrophic vegetation in these areas is dominated byDeschampsia caespitosa, Holcus lanatus, Agrostis spp.and Juncus effusus.

The two forests are in adjacent water sheds approxi-mately 20 km apart within the larger (620 km2) forestedarea. Long-term monitoring of field voles in Kielder hasshown cyclic population fluctuations with a 3–4-yrperiod (Petty 1992, Lambin et al. 2000). Althoughpopulations on nearby clear-cuts fluctuate in syn-chrony, out-of-phase populations are found within thelarger forest region. Kershope forest has been one yearahead of Kielder since 1993 (Petty and Fawkes 1997,Lambin et al. 1998). The voles at Kershope reachedhigh population densities in the autumn of 1996 andmaintained high densities in 1997, while the populationdensities in Kielder were lower in 1996 but increased in1997 (Fig. 1).

Sampling of parental animals for the laboratory

Voles for breeding in the laboratory and for autopsywere sampled from nine clear-cuts within a 50-km2 areain Kielder forest (‘increase area’) and three clear-cuts1–2 km apart in Kershope forest (‘peak area’) in twodifferent periods, 12–24 March and 13–15 April 1997.Voles were caught in Ugglan Special multiple capturetraps placed in active runways. Trapping continued on

186 OIKOS 95:2 (2001)

the sites for 2–3 d to avoid sampling only the mosttrappable animals. On capture, animals were individu-ally marked with ear-tags, weighed and their reproduc-tive status was noted. Pregnancy of captured femaleswas determined by swollen abdomen, parturition in thelaboratory, or by autopsy.

A random subset of the March sample was autop-sied. For determination of relative age (see Hagen et al.1980), eyes from autopsied voles (freshly sacrificed)were fixed in 4% formaldehyde for one week and thenstored in 70% ethanol until being dissected and driedfor one week at 70°C. Eye-lenses were allowed to coolto room temperature in a desiccator where they werekept until being weighed as pairs on a Mettler™ M3balance (precision 0.1 mg).

Laboratory procedures

While field sampling was taking place, animals werekept in single-sex, multi-animal cages (up to 10 individ-uals per cage) after capture until being transported torearing facilities in Aberdeen, Scotland. In the labora-tory, animals were caged in pairs in cages with sawdustand placed in sheds with windows and no heating.

Hence, photoperiod and temperature varied with theambient conditions. Placement of the cages within thelaboratory was randomised.

Animals were fed on fortified cereal pellets (‘Rat andMouse No. 1’ from Special Diets Services; 14.7%protein) and provided with hay for bedding and extrafood. Parental animals were also given apples andcarrots for the first two weeks until they learned todrink from the water bottles. Each cage was suppliedwith a wooden nest box and two water bottles. Thevoles regularly chewed on the nest boxes, hence reducedteeth wear was not a problem. After one of the authors(TE) became infected with leptospirosis (Leptospirasaxkoebing), all animals in the colony were treated withantibiotics (Tetramycine) in the first week of July.

Adults were weighed and checked for reproductivestatus every two weeks. Cages with pregnant femaleswere checked every two to three days in order todetermine the exact date of parturition. Females wereclassified as breeders if they gave birth in the laboratoryor were shown to be pregnant by autopsy. The maleswere usually left in the cages with their mate duringpregnancy and after parturition, but when there wasshortage of sires, males were sometimes taken away tobe paired with another female.

Juveniles were separated from their mother at 18–20d of age, and one female offspring from each litter wasselected at random for breeding in the next generation.These females were paired with non-paternal, knownbreeder, adult males from the parental generations. Insome cases, when excess males were available, twofemales from the same litter were paired but only oneof them (drawn at random) entered the analysis if bothsurvived to parturition.

Population monitoring in the field

Eight sites in the increase-phase area and two sites inthe peak-phase area were monitored before and aftersampling to provide estimates of density trajectoriesand data on breeding and maturation in the field (theeight increase sites were part of another study, Mac-Kinnon 1998). At each site, a 0.3-ha live-trapping gridwas established, consisting of 100 Ugglan SpecialMousetraps in a 10×10 configuration with 5-m spac-ing. The trapping regime followed the ‘robust design’(Pollock 1982) with primary sessions at 27–31-d inter-vals consisting of six secondary sessions over 2 d. Thefirst trapping session was at the end of March in theincrease area and in mid-April in the peak area. Trapswere pre-baited for 3 d, set between 06:00 and 08:00and checked three times per day at 4-h intervals. Thetraps were not set overnight. Individuals were markedwith a pair of uniquely numbered Hauptner™ ear-tags.The first time an animal was caught in each primarysession it was weighed and checked for reproductivestate.

Fig. 1. Density trajectories at the sampling sites before (1996)and after (1997) sampling voles to the laboratory. Densitiesare plotted on a logarithmic scale. Filled symbols are averagedensity estimates of eight sites in Kielder forest (increase area),and the open symbols are average estimates of two sites inKershope forest (peak area). The 1996 estimates from the peakarea are estimated from ‘Vole-sign indexes’ (Lambin et al.2000) and the other estimates are estimated with closed cap-ture-mark-recapture models by MacKinnon (1998). Error barsindicate plus/minus standard error of the averages. Inset showslong-term fluctuations in Kielder forest (from Lambin et al.2000).

OIKOS 95:2 (2001) 187

Estimating timing of reproduction and maturationrates in the field data

In order to compare the pattern of maturation in thelaboratory with maturation rates in the field, we em-ployed multi-state capture-mark-recapture models(Hestbeck et al. 1991, Brownie et al. 1993) in theprogram MARK (White and Burnham 1999, White2001). Males were classified as reproductive if they hadscrotal testes. Females were classified as reproductive ifthey were lactating, if their pubic symphysis or vaginalopening indicated that they had recently given birth, orif their nipples indicated that they had recently weanedyoung. We primarily wanted to estimate maturationrate, and not the rate at which reproducing animalscease reproducing for the season. Hence, animals wereclassified as reproductive if they had been captured inreproductive condition in an earlier session, and theparameter in the likelihood function representing thetransition probability from reproducing to non-repro-ducing state was fixed to zero.

Data from each trapping site were fitted with aseparate model, and the most parsimonious modelswere selected based on Akaike’s Information Criterionadjusted for sample size, AICc (see Burnham and An-derson 1998). The seasonal variation in maturationprobability of young-of-the-year females was modelledas a logit-function of capture date, assuming amonotonic decline in maturation probability over theseason. This assumption was assessed graphically, anda complete time specific structure never increased theparsimony (i.e., never lowered the AICc) of the selectedmodels. To simplify the model selection procedure, westarted with a model structure for apparent survival (�)and recapture probability (p) that was found to be mostparsimonious in standard Cormack-Jolly-Seber modelswithout multiple states. A general model for maturationprobability (different slope and intercept for the twosexes) was then used to determine whether additive orinteracting effects of reproductive state on � and pwould increase the parsimony of the models. With thenew model for � and p, we then searched for the mostparsimonious model for the maturation probability.Alternative models for � and p were then again testedto ensure that the most parsimonious model had beenfound.

Since almost all overwintered animals matured dur-ing the first one-month interval between trapping ses-sions (April–May), we could not estimate the timing ofonset of spring reproduction from the capture-mark-re-capture data. To compare the onset of spring reproduc-tion between the areas we relied on data on thefrequency distributions of reproductive states of cap-tured females. We also compared the seasonal patternsof reproduction by studying the recruitment of juvenilesto the trapping sites.

Statistical analysis

As animals taken to the laboratory were sampled froma small number of sites within only one year andgeographical region, we stress that the sampled individ-uals do not represent a random sample from particular‘phases’ of the cycles as such, but rather from two areashaving experienced contrasting densities in the previousyear. Phase dependence is a likely cause of the popula-tion (area) differences, but we cannot make statisticalinferences about the area differences. Our intention isto test whether population differences in the field weremaintained under standardised conditions in the labo-ratory. Hence, in the analysis of the laboratory data weconsidered all sampled animals within each of the pop-ulations and sample periods as independent, and com-pare the population differences in the field with thepopulation differences in the laboratory at the level ofthe individuals (not sampling sites).

Each response variable was investigated by two dif-ferent models. First, to assess differences in populationmeans, models with only population and sampling timewere used. Second, models including intrinsic state vari-ables (body mass and reproductive state, i.e., pubicsymphysis, perforate/non-perforate vagina and preg-nancy status) were used to search for underlying mech-anisms behind the differences.

There was high mortality among the parental genera-tion in the laboratory. A total of 69.5% (n=141)‘increase-area females’ and 41.6% (n=113) ‘peak-areafemales’ died before they were paired or within 12 dafter pairing. As the animals were transported and keptin multi-animal cages prior to pairing, and a possibledisease may have been transmitted between individuals,these differences in mortality in the laboratory mayhave little relevance to the wild populations becauseindividuals were not treated independently. However,females of the two populations were very different intheir initial states, so care must be taken to avoiderroneous inferences due to experiment-induced biascaused by selective mortality. In order to reduce possi-ble biases when testing for population differences in thelaboratory, each observation was weighted with theinverse of the estimated probability of survival in theestimation (see Littell et al. 1996). Hence, the sum ofweightings of animals of a particular characteristic di-vided by the total sum of weightings for all animals willbe the same among the survivors as in the originalsample (where all have equal weightings). Since thegroups differed greatly in intrinsic state variables, sur-vival probabilities used to calculate weightings wereestimated with separate logistic models for each popula-tion and sample period, and the best models wereselected by the lowest AIC. These models are given inTable 1.

As we did not have any prior knowledge of theimportance of maternal effects, we considered models

188 OIKOS 95:2 (2001)

Table 1. Survival models (logit[probability of survival], binomial error distribution) used to calculate the weightings used to testfor differences between samples of females. 95% confidence intervals of the parameters are given in brackets. Models wereselected according to the lowest AIC. Weightings for each observation were calculated as the inverse of the predicted survivalprobability (see Material and methods), and the ratios between maximum and minimum weightings in each group are given inthe righthand column.

Sample Intercept Pregnant= ‘no’ Capture body mass Ratio max/min% survived oftotal(g) weights

Peak area – March 0.41 [−0.02, 0.85] 1.060% of 85Increase area – March 2.36 [−1.63, 6.74] 24% of 91−1.76 3.7−0.08 [−0.22, 0.03]

[−3.42, −0.30]Peak area – April 0.14 [−0.60,0.90] 1.053% of 28

6.4Increase area – April −6.30 [−11.32, −2.62] +0.21 [0.08, 0.39] 42% of 50

both with and without weighted observations in thelaboratory-born generations and report the most con-servative result. For logistic and log-linear models, weapplied quasi-likelihood techniques implemented inthe GLIMMIX-macro (version 30 April 1998) in SAS(Littell et al. 1996). When using weighted observa-tions in GLIMMIX, the result is independent of thescale of the weights.

Terms were included in the statistical models ifthey reduced the AIC value (Akaike 1985). Tests ofeffects when all other selected terms are included inthe model (type-III tests) are presented. All date vari-ables were centralised by subtracting the mean value.Litter size had generally larger variance with increas-ing expected value. Hence, litter size was modelledwith log-linear models (Poisson distributed error andlog-link), which proved to give reasonable fits (Pear-son residuals).

Results

Reproductive traits and demography in the field

The voles at Kershope reached high population densi-ties in the autumn of 1996 and maintained high den-sities in 1997, while the population densities inKielder were lower in 1996 but increased in 1997(Fig. 1). To increase readability we will in the follow-ing refer to Kershope as the ‘peak area’ and Kielderas the ‘increase area’. The first spring born juvenilesappeared more than six weeks earlier in the increasearea than in the peak area (Fig. 2). The large differ-ence in the onset of reproduction is also evident fromthe distribution of body mass and reproductive stateof the overwintered animals taken to the laboratoryin both March and April (see below). Young-of-the-year in the increase area continued to mature later inthe season (Fig. 3), and reproduction continued fornearly two months later in the autumn than in thepeak area (Fig. 2).

Fig. 2. Cumulative average of new litters per trapping site inthe increase-area sites (solid line) and the peak-area sites(stippled line). Date of birth of captured juveniles was esti-mated from their body mass using estimated growth curves ofjuveniles in the laboratory. Captured juveniles were groupedinto presumptive litters based on their body mass and locationat capture.

Fig. 3. Estimated probability that an immature female atvarious dates will reach maturity during the next 30 d. Thefigure shows the arithmetic mean�SE of estimates from theeight increase-area trapping sites (filled squares) and the twopeak-area sites (open circles). The estimates were obtained bymulti-state capture-mark-recapture models (see Material andmethods).

OIKOS 95:2 (2001) 189

Fig. 4. Absolute frequency distributions of eye lens masses(mg per lens) of autopsy sample from March. Hatched barsare females, open bars stacked on top are males. There are nosignificant population or sex differences.

Only the largest peak-area females had an estimatedbreeding probability equal to the increase-area females(Fig. 7).

Increase-area females were still heavier after parturi-tion than peak-area females (weighted linear model,population effect: F=9.05; 1, 82 df; p=0.004; 95% c.i.,increase area: 38.3�1.2 g, peak area: 35.5�1.4 g).Thirty of the breeding peak-area females and 27 of theincrease-area females had a mate at parturition of thefirst litter, and hence the opportunity to re-conceive. Allof these females became pregnant except for four in-crease-area females that only had a mate for 4–10 dafter giving birth to their first litter. Time between firstand second parturition ranged from 19–42 d (mean=21.2 d) but there was no significant pattern in thisvariation.

Litter sizeIncrease-area females had on average larger litters (leastsquare mean=4.5; 95% c.i. [4.2, 4.9]) than did peak-area females (least square mean=4.0; 95% c.i. [3.6,4.4]; weighted log-linear model, population effect: �2=4.04, 1 df, p=0.044; sampling month: �2=2.83, 1 df,p=0.093). Animals with closed pubic symphyses whencaptured had smaller litters (mean=3.9; 95% c.i. [3.6,4.2]) than those with open pubic symphyses when cap-tured (mean=5.2; 95% c.i. [4.6, 6.0]; log-linear model:�2=14.4, 1 df, p=0.0002). When state of pubic sym-physis was included in the model, the population effectwas no longer significant (�2=0.99, 1 df, p=0.32;interaction effect: �2=2.75, 1 df, p=0.10). The size oflitters conceived in the laboratory was also related to

Initial characteristics of the laboratory sample

Distributions of eye lens mass (an index of age) fromthe autopsy (Fig. 4) were not significantly differentbetween the two populations (two-sample Kolmogorov-Smirnov tests, p�0.4). However, increase-area animalsof both sexes had generally larger body mass (Fig. 5)and many females had already started to reproduce inthe increase-area sample (Table 2).

Breeding performance of the overwinteredgeneration

Frequency and timing of breeding in the labIn the laboratory as in the field, the overwinteredanimals from the two populations showed differentpatterns of reproduction. A higher proportion of theincrease-area females conceived and the increase-areafemales had shorter average time between pairing andparturition (26 vs 32 d) than the peak-area females(Fig. 6a; weighted Cox proportional hazard regression,population effect: Z=2.44, p=0.015). Excluding ninefemales that died within 12 d after pairing, 95% of theincrease-area females vs only 77% of the peak-phasefemales either gave birth or showed signs of pregnancywhen autopsied (weighted logistic model, populationeffect: �2=6.40, 1 df, p=0.013). Non-breeding femaleslived up to 205 d (1st to 3rd quartile: 32 to 111)together with their mate. Breeding probability of peak-area females increased significantly with body mass atcapture (logistic regression: �2=5.15, 1 df, p=0.023).

Fig. 5. Body mass distributions of sampled animals groupedby sex, month sampled and population. Both data from au-topsy sample and animals used in the laboratory experimentare used in the March samples. Box-plots show median andinter-quartile distance. Whiskers show 1.5 times inter-quartiledistance (approx. 95% of data) and outliers are plotted ashorizontal lines. There are highly significant differences indistributions between the populations in all four groups (Kol-mogorov-Smirnov two-sample tests, p�0.0001), also whenexcluding all pregnant females (March: p�0.0001, April: p=0.011).

190 OIKOS 95:2 (2001)

Table 2. Reproductive state of sampled females. The March sample includes animals sampled for autopsy as well as thosebrought into the laboratory. The April sample is only animals sampled for the laboratory. p-values were calculated usingFisher’s exact test.

March sample April sample

p-valuePeak area Increase area p-value Peak area Increase arean=123 n=118 n=29 n=55

Pregnant 2.4% 28.0% 0.008�0.0001 6.9% 32.7%Nipples Small 100.0% 88.5% 92.6% 58.3%

Lactating 0.0050.0% 0.03% 0.001 0.0% 29.2%Post-lact.1 0.0% 8.2% 7.4% 12.5%

Perforate vagina 7.8% 52.7% 0.059�0.0001 37.0% 60.4%Pubic Closed 89.5% 53.8% 88.9% 68.1%symphysis �2 mm 10.5% 38.7% �0.0001 11.1% 21.3% 0.078

�2 mm 0.0% 7.5% 0.0% 10.6%Mature uterus2 60.5% 81.5% 0.10(�0.5 mm diameter) (n=38) (n=27)

1 Reproduced earlier in life.2 Data on uterus are only from the autopsy sample.

pregnancy status at capture in the increase-area sample(there were very few pregnant females in the peak-areasample, see Table 2); mean of 5.2 (95% c.i. [4.4, 6.1])among those pregnant at capture versus 4.1 (95% c.i.[3.6, 4.7]) for those not pregnant at capture. Litter sizeof those females that re-conceived in the laboratoryincreased significantly from first to second parturition(average increase: 1.1 pups; Wilcoxon matched pairstest: Z=4.30, p�0.0001). Together, these results sug-gest that the population difference in litter size may bedue to a parity effect, as a larger proportion of theincrease-area females had previously bred in the field.

Breeding performance of the laboratory-borngenerations

As there was large variation in date of birth of thelaboratory-born females, and since date of birth dif-fered between the two populations, date of birth wasused as a covariate in all further analysis.

Body mass at weaningBody mass at weaning (18–22 d old, age included as acovariate in the model) of the first laboratory-borngeneration (F1s) generally decreased with increasingnumber of littermates (95% c.i. −0.73�0.69 g peradditional littermate) and increased with date of birth(95% c.i. 0.09�0.05 g d−1). However, there was nodifference between the two populations (model includ-ing littermates and date of birth, population effect:F=1.16; 1, 82 df; p=0.28, alone: F=1.68; 1, 84 df;p=0.20).

In the second lab-born generation (F2s), body massat weaning decreased with increasing date of birth (95%c.i. −0.16�0.08 g d−1), and estimated weaning masswas 2.47 g (�2.12 g; 95% c.i.) higher for the peak-area

females than for the increase-area females born on thesame date.

Frequency and timing of breeding – F1sFour of the 49 breeding peak-area parental females didnot wean any daughters, and a further two of thepeak-area F1-females died shortly after pairing. Of theremaining F1 females, 49% (n=41) of the increase-areafemales and 55% (n=43) of the peak-area femalesconceived (revealed by parturition or autopsy) in thelaboratory (Fisher’s exact: p=0.7). There was no sig-nificant population effect on breeding pattern (Fig. 6b,Cox proportional hazard regression, population effect:Z=0.12, p=0.9). However, date of birth under theambient photoperiod conditions had a highly signifi-cant effect on breeding probability (logistic regressionmodel: �2=16.05, 1 df, p�0.0001). Females born earlyin the season had a much higher estimated probabilityof becoming a breeder than those born late in theseason (Fig. 8a).

Frequency and timing of breeding – F2sIn the F2 generation, four of the 21 (19%) increase-areafemales versus seven of the 24 (29%) peak-area femalesconceived (Fisher’s exact test: p=0.5). As in the F1generation, there was a decreasing probability of breed-ing for F2 females born later in the season (Fig. 8b;logistic regression model: �2=4.77, 1 df, p=0.029),but there was no significant population effect on thebreeding pattern (Fig. 6c; Cox proportional hazardregression, population effect: Z=1.03, p=0.3). How-ever, a different pattern in timing of breeding wasgenerally seen in the F2 generation than in the F1generation. Only one F2 female raised pups beforemid-August, even though they were paired 50–95 dearlier. This female had only one pup in mid-July,whereas all other females had three pups. The five

OIKOS 95:2 (2001) 191

earliest born F2 females had their first litter killed byone of the parents, but all except one of these pairsraised young later in the season. Infanticide of ownlitters occurred only twice among the 90 breedingparental females and never among the 44 breeding F1females.

Non-breeding females remained small throughout thesummer, and by 1 August they were significantlysmaller (mean=22.8 g, SD=3.1) than breeding fe-

Fig. 7. Estimated probability that females bred in the labora-tory given that they survived. The broken line and opensymbols are peak-area females, while the solid line and sym-bols are increase-area females. Observations are symbolisedwith circles above (breeders) and below (non-breeders).

Fig. 6. Maturation in a) the parental generation, b) the F1-generation, and c) the F2-generation. Figures show the cumu-lative daily probabilities (Kaplan-Meier estimates) of givingbirth for peak-area (stippled lines) and increase-area (solidlines) females. Crosses indicate censored animals; i.e, animalsthat either died, lost their mate before maternity, or had stillnot reproduced by the end of the study. There is a significantdifference between the curves only in the overwinteredparental generation (see text).

males (mean=28.0, SD=2.4, one pregnant femaleomitted; two-tailed t-test: p�0.0001).

Litter size in the laboratory generationsThere was no significant difference in average litter sizeof F1s between the two populations (increase area 3.9vs 4.1 for peak area; �2=0.15, 1 df, p=0.7). Allfemales in the F2 generation had three offspring exceptone increase-area female having only one pup.

Cross-generational correlations

Despite large variation in reproductive traits withinboth the parental generation and the first laboratory-born generation, there was no indication of any strongeffects carrying over between the generations (Table 3).The weak correlation between time before parturitionin the parental generation and the pre-weaning growthrate of their daughters (Table 3) may be a date (pho-toperiod) effect. Heavier animals at capture also tendedto be heavier after giving birth, and laboratory-bornanimals with higher pre-weaning growth rate alsotended to be heavier after parturition.

Discussion

We have demonstrated large variation in the seasonaltrends in reproductive traits between female field volesfrom two nearby cyclic populations that had experi-enced contrasting densities in the previous year. How-ever, except for the parental laboratory generation,reproductive traits varied only seasonally under stan-

192 OIKOS 95:2 (2001)

dardised laboratory conditions. Hence, the large popu-lation differences in maturation rates that we found inthe field are unlikely to have been caused by maternalor genetic effects, as assumed by the maternal effectshypothesis for population cycles (Boonstra and Boag1987, Boonstra and Hochachka 1997) and Chitty’sfluctuating selection hypothesis (Chitty 1960, 1967).

Although we cannot make statistical inferences aboutthe phase of the cycles from this study, the reproductivetraits that we have been investigating (onset of springreproduction and maturation of young-of-the-year) areknown from many studies to vary consistently with thephase of small rodent cycles (see review in Krebs andMyers 1974, Hansson and Henttonen 1985, Bernshteinet al. 1989, Gilbert and Krebs 1991, Boonstra 1994).

Our field observations are in agreement with thesegeneral findings; later onset of spring reproduction and‘poorer’ reproduction when the populations have gonethrough high densities.

Overwintered animals – does the memory fordelayed density dependence reside in the individualvoles?

Onset of spring reproduction in the field was more thana month later in the peak area than in the increase area.The large differences between the sites are shown bybody mass distributions, frequency of reproductivestates among overwintered animals when sampled andthe time that the first juveniles in the spring appeared.Such differences in onset of reproduction were alsofound under the standardised laboratory conditions.Increase-area animals had also a larger proportion ofbreeders as well as larger litters in the laboratory. Ourevidence indicates that the latter may be due to the factthat more animals in the increase area had previouslyreproduced in their lives.

Although the population differences in onset of re-production were much smaller in the laboratory than inthe field, the clear differences between the populationsseen in the laboratory must represent different states ofthe animals when they were sampled. These initial statedifferences may have resulted in larger differences inreproductive performance in a natural but shared envi-ronment (or with a different experimental protocol).However, it is not clear from our experiment at whatstage the individuals’ internal state diverged betweenthe populations. On the one hand, the state differencesmay have been shaped early in the life of the individu-als or resulted from genetic differences, in which casethere would be a time-lag residing in the individuals.Hansson (1989) reported that laboratory-born Microtusagrestis started to breed two months earlier in thespring and had a higher frequency of winter breedingunder laboratory conditions with ambient light thanwild-born females living under the same conditions.This suggests that the environment in early life may bean important determinant of onset of spring reproduc-tion. On the other hand, since some of the voles hadalready started to reproduce in the increase area whenthey were sampled in the field, it may also be that theincrease-area animals had reacted to some cue in theenvironment just before sampling, and that the peak-area animals had not yet received this stimulus. Thefact that the smallest peak-area animals failed to breedin the laboratory supports the latter suggestion – thatthe smallest animals may not yet have received anenvironmental stimulus to initiate spring growth andreproduction.

Boonstra (1994) and Tkadlec and Zejda (1998) foundthat low juvenile recruitment during the peak phase of

Fig. 8. Estimated breeding probability (solid line) as a func-tion of date of birth of laboratory-born females; a) F1s, and b)F2s. Broken lines show the 95% confidence limits of theestimate. Observations are symbolised with circles above(breeders) and below (non-breeders); open circles representpeak-area females, while solid circles are increase-area females.All breeding F1 females bred before the summer, while all butone of the breeding F2 females postponed reproduction untilafter the summer (see text for details).

OIKOS 95:2 (2001) 193

Table 3. Spearman rank correlations between reproductive traits of females within and across the parental (mothers) and thefirst lab-born (daughters) generation. The top number is the correlation coefficient for both populations pooled, the middle isfor the peak-area population, and the lower number is for the increase-area population. Asterisks indicate significance levels;*: p�0.05, **: p�0.01, ***: p�0.001.

Pooled peak Parental generation F1 generationincrease

LitterTime before Pre-weaningLitter Age at firstMother body Mother bodymass aftermass atsizeparturition growth rate sizeparturition

capture parturition

Parental Litter size −0.42***generation −0.38**

−0.36*Mother body −0.23* 0.30*mass at capture −0.19 0.34*

0.05 0.02Mother body −0.01 0.26* 0.37***mass after 0.02 0.31* 0.19parturition 0.11 0.15 0.39*

F1 generation Pre-weaning 0.25* −0.35** 0.04 0.07growth rate 0.13 −0.30 0.07 0.08

0.34 −0.33 0.11 0.13Age at first 0.20 −0.17 0.18 0.11 −0.12parturition 0.22 −0.08 0.21 0.24 −0.09

0.30 −0.26 0.10 0.00 −0.14Litter size −0.25 0.01 −0.08 0.06 0.07 −0.12

−0.29 0.15 −0.21 −0.05 −0.04 −0.24−0.24 −0.16 0.02 0.07 0.44 0.05

Mother body 0.19 −0.01 0.43*0.21 0.38* 0.51* −0.09mass after −0.13 0.51*0.53* 0.38 0.34 0.25 0.05parturition 0.36 −0.42 0.40 0.43 0.380.84*** −0.19

the cycles causes a shift in age structure toward olderanimals, and they argued that poor reproduction ofoverwintered animals in the decline phase is due tosenescence. We assessed differences in age structurebetween the two populations in our study by comparingthe distributions of eye lens masses (Hagen et al. 1980),but found no evidence for a difference in age structure.Hence, there was no support for the idea that senes-cence was responsible for the large differences in onsetof reproduction and reproductive performance in thisstudy. However, since the relationship between eye lensmass and age flattens out with increasing age, detectingdifferences in the prevalence of older age classes may bedifficult.

Laboratory-born animals – there is no supportfor maternal or genetic effects

Although the parental generation showed clear differ-ences in reproductive performance in the laboratory,these differences did not carry over to the laboratory-born generations. In fact, despite large variation inreproductive traits in all generations, there were noconvincingly strong cross-generational correlations forany trait. Thus, maternal and genetic effects are notmajor sources of variation in reproductive traits ofthese cohorts under the laboratory conditions. Hence,the notion that maternal effects are important determi-nants of demographic traits in small rodent populations

in general (Boonstra and Boag 1987, Boonstra andHochachka 1997) is not supported by our study. Boon-stra and Hochachka (1997) found strong maternal in-fluence on growth rate and age at maturity of collaredlemmings (Dicrostonyx groenlandicus) under laboratoryconditions. However, such effects remain to be demon-strated under natural conditions.

The most prominent patterns seen in the laboratory-born generations were the differences between the gen-erations and the changes over the season. In both of thelaboratory-born generations, the estimated probabilityof breeding declined with date of birth of the female.All F2 females that raised any pups (except for onefemale having a single pup) did not do so until end ofAugust to end of September. The only environmentalchange over time in the laboratory was that of seasonalchange in ambient light conditions and temperature.Change in photoperiod is known to be an importantcue for reproductive regression and inhibition of matu-ration before the winter in several species of voles(reviewed in Bronson and Heideman 1994) includingfield voles (Spears and Clarke 1988). However, regula-tion of reproduction by natural change in photoperiodduring mid-summer has, to our knowledge, not beenpreviously described.

The seasonal decline in maturation seen in the labo-ratory closely resembles the pattern seen in our live-trapping sites, especially in the peak area wherematuration ended early in the season. Spears andClarke (1988) suggested, based on their findings of

194 OIKOS 95:2 (2001)

strong heritability of photo-responsiveness on matura-tion of juvenile field voles, that phase-dependent varia-tion in the seasonal reproductive pattern in cyclic volepopulations may be due to genetic differences (see alsoNelson 1987, Bronson and Heideman 1994). However,in our study, the large season-specific difference inmaturation between the areas was not maintained un-der the standardised laboratory conditions. This sug-gests that phase-dependent variation in juvenilematuration is not genetically based but rather causedby differences in the environment.

Mihok and Boonstra (1992) reported that femalemeadow voles (Microtus pennsyl�anicus) from a lowyear after a severe winter decline had a much poorerbreeding performance under laboratory conditions thanfemales sampled during a strong increase the followingyear. In their study, however, low-year animals weresampled in early June and July (50/50 overwintered/young-of-the-year), whereas the increase-year animalswere sampled at the start of the breeding season in Mayand had all overwintered. In the light of our ownresults, it is possible that the difference in reproductiveperformance observed by Mihok and Boonstra (1992) isdue to seasonal effects rather than to phase-dependentdifferences.

Possible environmental effects on reproductivetraits in voles

Although deterministic seasonal change in photoperiodcan alone cause the seasonal trends in maturation prob-ability, environmental mechanisms must be invoked toexplain the large between-year/site variability in lifehistories of cohorts born in mid-summer cohorts whichwe found in this study.

Social constraints may affect juvenile maturation inthe field (Krebs and Myers 1974, Morris 1989, Puseniusand Viitala 1993, Boonstra 1994). At the beginning ofthe year, there were differences in density between ourtwo study areas, so the differences in maturation ratemay partly have been due to direct density-dependentsocial suppression of maturation (e.g. Boonstra andRodd 1983, Rodd and Boonstra 1988, Boonstra 1989,Gilbert and Krebs 1991). However, the difference inmaturation frequencies was largest late in the yearwhen the densities had more or less converged. Nor cansocial suppression of breeding explain why all overwin-tered animals in the peak area delayed maturation untillate in the spring. The above points to the importanceof extrinsic effects.

Suppressed reproduction and delayed maturation inmicrotines have been suggested to be an adaptive re-sponse to high risk of predation, and several laboratorystudies have found strong effects of mustelid odours onreproduction (Ylonen 1989, Korpimaki et al. 1994,Koskela and Ylonen 1995, Mappes and Ylonen 1997).

However, such an effect may be a laboratory artefactwith little relevance for natural populations (Hansson1995, Lambin et al. 1995, Korpimaki and Krebs 1996).A recent field experiment failed to find any effects ofmustelid odours despite simulating very high predatordensity, which led the authors to conclude that theirprevious findings in the laboratory had little relevance(Mappes et al. 1998).

Feore et al. (1997) found that non-lethal cowpoxinfections in bank voles (Clethrionomys glareolus) andwood mice (Apodemus syl�aticus) increased the timebefore first litter under laboratory conditions. Sucheffects may be profound under natural conditions andmay have interacting effects with, e.g., food availability.Cowpox is highly prevalent in our study system (Ca-vanagh et al. unpubl.).

Food quality and/or availability directly affect thephysiological state of the individuals and may cause adelay in maturation (Berger et al. 1977, 1981, Gallaherand Schneeman 1986, Andreassen and Ims 1990, Ryan1990). Negus et al. (1977) provide evidence that consid-erable variation in onset of reproduction in M. mon-tanus between years may be related to the availability ofgreen plants in the spring, which in turn is related tothe time of snow melt. The phenolic plant compound,6-methoxy-2-benzoxazolinone (6-MBOA), which ispresent in all growing grasses and sedges but has nonutritional value (Moffatt et al. 1991, Nelson 1991), isknown to enhance gonadal development and triggerreproduction in several north-American Microtus spe-cies (Sanders et al. 1981, Korn and Taitt 1987, Gowerand Berger 1990, Nelson and Blom 1993, Meek et al.1995). Korn and Taitt (1987) found that Microtustownsendii in natural populations that were fed withoats coated with 6-MBOA started to breed four weeksearlier than voles in a control site 200 m away whereoats coated with the solvent only were provided. Sev-eral mechanisms may cause a delayed response in foodquality and/or availability to earlier grazing; e.g. quan-titative reduction and slow recovery (Kalela and Ko-ponen 1971, Oksanen and Oksanen 1981, Moen et al.1993), depletion of the plants’ stored resources in theroots or altered allocation of resources to growth anddefence (Herms and Mattson 1992), and induced plantresistance (Karban and Baldwin 1997). It is also wellknown from agricultural science that harvesting orgrazing in a critical period in the autumn can interruptcold acclimation, and thereby strongly reduce tillersurvival and grass production in the following spring(Sheaffer and Marten 1990, Sheaffer et al. 1992, Shi-mada 1994). Bergeron and Jodin (1993) found a 52%reduction of green biomass in the spring after manipu-lating high densities of Microtus pennsyl�anicus in en-closures the previous year.

To understand the mechanisms of population fluctu-ations of microtines, there is clearly a need for bothbetter descriptions of the direct and delayed density-de-

OIKOS 95:2 (2001) 195

pendent structure of variation in life-history traits, aswell as studies on the individual level responses to theenvironment. Diseases and food quality/availabilitymay impose severe constraints on the individuals’ per-formance, and they therefore represent plausible agentsto cause variation in reproductive traits. However, therole of diseases and food quality/availability in thepopulation dynamics of microtines is as yet largelyunexplored.

Acknowledgements – The laboratory part of this work wasfunded by a special grant from the Univ. of Oslo to NCS andby support from the Natural Sciences and Engineering Councilto RB. The field monitoring was supported by a studentship(to JLM) and grants from Natural Environmental ResearchCouncil, UK to XL. The Centre for Advanced Study of theNorwegian Academy of Science and Letters together with theUniv. of Oslo supported RB to visit Oslo and the field sites inScotland. TE has been supported by the Norwegian ScienceFoundation. Rearing facilities were generously provided atCulterty Field Station and we thank Robert Donaldson andSue Way for assistance with maintaining the animals. Wethank George Batzli, Karen E. Hodges, Robert Moss andNigel Yoccoz for constructive comments on the manuscript.

ReferencesAgrell, J., Erlinge, S., Nelson, J. and Sandell, M. 1992. Body

weight and population dynamics: cyclic demography in anoncyclic population of the field vole (Microtus agrestis). –Can. J. Zool. 70: 494–501.

Akaike, H. 1985. Prediction and entropy. – In: Atkinson, A.C. and Fienberg, S. E. (eds), A celebration of statistics.Springer-Verlag, pp. 1–24.

Andreassen, H. P. and Ims, R. A. 1990. Responses of femalegray-sided voles Clethrionomys rufocanus to malnutrition: acombined laboratory and field experiment. – Oikos 59:107–114.

Berger, P. J., Sanders, E. H., Gardner, P. D. and Negus, N. C.1977. Phenolic plant compounds functioning as reproduc-tive inhibitors in Microtus montanus. – Science 195: 575–577.

Berger, P. J., Negus, N. C., Sanders, E. H. and Gardner, P. D.1981. Chemical triggering of reproduction in Microtusmontanus. – Science 214: 69–70.

Bergeron, J.-M. and Jodin, L. 1993. Intense grazing by voles(Microtus pennsyl�anicus) and its effect on habitat quality.– Can. J. Zool. 71: 1823–1830.

Bernardo, J. 1996. Maternal effects in animal ecology. – Am.Zool. 36: 83–105.

Bernshtein, A. D., Zhigalsky, O. A. and Panina, T. V. 1989.Multi-annual fluctuations in the size of a population of thebank vole in European part of the USSR. – Acta Theriol.34: 409–438.

Bjørnstad, O. N., Falck, W. and Stenseth, N. C. 1995. Ageographic gradient in small rodent density fluctuations: astatistical modelling approach. – Proc. R. Soc. Lond. B.262: 127–133.

Boonstra, R. 1989. Life history variation in maturation influctuating meadow vole populations (Microtus pennsyl-�anicus). – Oikos 54: 265–274.

Boonstra, R. 1994. Population cycles in microtines: the senes-cence hypothesis. – Evol. Ecol. 8: 196–219.

Boonstra, R. and Rodd, F. H. 1983. Regulation of breedingdensity in Microtus pennsyl�anicus. – J. Anim. Ecol. 52:757–780.

Boonstra, R. and Boag, P. T. 1987. A test of the Chittyhypothesis: inheritance of life-history traits in meadowvoles Microtus pennsyl�anicus. – Evolution 41: 929–947.

Boonstra, R. and Hochachka, W. M. 1997. Maternal effects ofadditive genetic inheritance in the collared lemming Di-crostonyx groenlandicus. – Evol. Ecol. 11: 169–182.

Bronson, F. H. and Heideman, P. D. 1994. Seasonal regula-tion of reproduction in mammals. – In: Knobil, E. andNeill, J. D. (eds), The physiology of reproduction. RavenPress, pp. 541–583.

Brownie, C., Hines, J. E., Nichols, J. D. et al. 1993. Capture-recapture studies for multiple strata including non-Marko-vian transitions. – Biometrics 49: 1173–1187.

Burnham, K. P. and Anderson, D. R. 1998. Model selectionand inference: a practical information-theoretic approach.– Springer-Verlag.

Chitty, D. 1960. Population processes in the vole and theirrelevance to general theory. – Can. J. Zool. 38: 99–113.

Chitty, D. 1967. The natural selection of self-regulatory be-haviour in animal populations. – Proc. Ecol. Soc. Aust. 2:51–78.

Elton, C. 1924. Periodic fluctuations in the numbers of ani-mals: their causes and effects. – Br. J. Exp. Biol. 2:119–163.

Feore, S. M., Bennett, M., Chantrey, J. et al. 1997. The effectof cowpox virus infection on fecundity in bank voles andwood mice. – Proc. R. Soc. Lond B 264: 1457–1461.

Gallaher, D. and Schneeman, B. O. 1986. Nutritional andmetabolic response to plant inhibitors of digestive enzymes.– Adv. Exp. Med. Biol. 199: 167–184.

Gilbert, B. S. and Krebs, C. J. 1991. Population dynamics ofClethrionomys and Peromyscus in southwestern Yukon1973–1989. – Holarct. Ecol. 14: 250–259.

Gower, B. A. and Berger, P. J. 1990. Reproductive responsesof male Microtus montanus to photoperiod, melatonin,and 6-MBOA. – J. Pineal Res. 8: 297–312.

Hagen, A., Stenseth, N. C., Østbye, E. and Skar, H. J. 1980.The eye lens as an age indicator in the root vole. – ActaTheoriol. 25: 39–50.

Hansen, T. F. and Boonstra, R. 2000. The best in all possibleworlds? A quantitative genetic study of geographic varia-tion in the meadow vole, Microtus pennsyl�anicys. – Oikos89: 81–94.

Hansson, L. 1989. Tracking or delay effects in microtinereproduction? – Acta Theriol. 34: 125–132.

Hansson, L. 1995. Is the indirect predator effect a special caseof generalized reactions to density-related disturbances incyclic rodent populations? – Ann. Zool. Fenn. 32: 159–162.

Hansson, L. and Henttonen, H. 1985. Regional differences incyclicity and reproduction in Clethrionomys species: arethey related? – Ann. Zool. Fenn. 22: 277–288.

Herms, D. A. and Mattson, W. J. 1992. The dilemma ofplants: to grow or to defend. – Q. Rev. Biol. 63: 283–335.

Hestbeck, J. B., Nichols, J. D. and Malecki, R. A. 1991.Estimates of movement and site fidelity using mark-resightdata of wintering Canada geese. – Ecology 72: 523–533.

Inchausti, P. and Ginzburg, L. R. 1998. Small mammal cyclesin northern Europe: patterns and evidence for maternaleffect hypothesis. – J. Anim. Ecol. 67: 180–194.

Kalela, O. and Koponen, T. 1971. Food consumption andmovements of the Norwegian lemming in areas character-ized by isolated fells. – Ann. Zool. Fenn. 8: 80–84.

Karban, R. and Baldwin, I. T. 1997. Induced responses toherbivory. – Univ. of Chicago Press.

Korn, H. and Taitt, M. J. 1987. Initiation of early breeding ina population of Microtus townsendii (Rodentia) with thesecondary plan compound 6-MBOA. – Oecologia 71: 593–596.

Korpimaki, E. and Krebs, C. J. 1996. Predation and popula-tion cycles of small mammals. – BioScience 46: 754–764.

Korpimaki, E., Norrdahl, K. and Valkama, J. 1994. Repro-ductive investment under fluctuating predation risk: mi-crotine rodents and small mustelids. – Evol. Ecol. 8:357–368.

196 OIKOS 95:2 (2001)

Koskela, E. and Ylonen, H. 1995. Suppressed breeding in thefield vole (Microtus agrestis): an adaptation to cyclicallyfluctuating predation risk. – Behav. Ecol. 6: 311–315.

Krebs, C. J. 1978. A review of the Chitty hypothesis ofpopulation regulation. – Can. J. Zool. 56: 2463–2480.

Krebs, C. J. and Myers, J. H. 1974. Population cycles in smallmammals. – Adv. Ecol. Res. 8: 267–399.

Lambin, X., Ims, R. A., Steen, H. and Yoccoz, N. G. 1995. Volecycles. – Trends Ecol. Evol. 10: 179–220.

Lambin, X., Elston, D., Petty, S. and MacKinnon, J. 1998.Spatial patterns and periodic travelling waves in cyclic fieldvole, Microtus agrestis, populations. – Proc. R. Soc. Lond.265: 1491–1496.

Lambin, X., Petty, S. J. and MacKinnon, J. L. 2000. Cyclicdynamics in field vole populations and generalist predation.– J. Anim. Ecol. 69: 106–118.

Littell, R. C., Milliken, G. A., Stroup, W. W. and Wolfinger,R. D. 1996. SAS Systems for Mixed Models. – SAS Inst.,Cary, NC.

MacKinnon, J. 1998. Spatial dynamics of cyclic field voleMicrotus agrestis populations. – Univ. of Aberdeen.

Mappes, T. and Ylonen, H. 1997. Reproductive effort of bankvole females in a risky environment. – Evol. Ecol. 11:591–598.

Mappes, T., Koskela, E. and Ylonen, H. 1998. Breedingsuppression in voles under predation risk of small mustelids:laboratory or methodological artifact? – Oikos 82: 365–369.

McNamara, J. M. and Houston, A. I. 1996. State-dependent lifehistories. – Nature 380: 215–221.

Meek, L. R., Lee, T. M. and Gallon, J. F. 1995. Interaction ofmaternal photoperiod history and food type on growth andreproductive development of laboratory meadow voles (Mi-crotus pennsyl�anicus). – Physiol. Behav. 57: 905–911.

Mihok, S. and Boonstra, R. 1992. Breeding performance incaptive meadow-voles (Microtus pennsyl�anicus) from de-cline- and increase-phase populations. – Can. J. Zool. 70:1561–1566.

Moen, J., Lundberg, P. A. and Oksanen, L. 1993. Lemminggrazing on snowbed vegetation during a population peak,northern Norway. – Arct. Alp. Res. 25: 130–135.

Moffatt, C. A., Bennett, S. A. and Nelson, R. J. 1991. Effectsof photoperiod and 6-methoxy-2-benzoxazolinone on male-induced estrus in prairie voles. – Physiol. Behav. 49: 27–32.

Morris, D. W. 1989. Density-dependent habitat selection: test-ing the theory with fitness data. – Evol. Ecol. 3: 80–94.

Negus, N. C., Berger, P. J. and Forslund, L. G. 1977. Repro-ductive strategy of Microtus montanus. – J. Mammal. 58:347–353.

Nelson, R. J. 1987. Photoperiod-nonresponsive morphs: apossible variable in microtine population-density fluctua-tions. – Am. Nat. 130: 350–369.

Nelson, R. L. 1991. Maternal diet influences reproductivedevelopment in male prairie vole offspring. – Physiol.Behav. 50: 1063–1066.

Nelson, R. J. and Blom, J. M. C. 1993. 6-Methoxy-2-benzoxa-zolinone and photoperiod: prenatal and postnatal influenceson reproductive development in prairie voles (Microtusochrogaster). – Can. J. Zool. 71: 776–789.

Oksanen, L. and Oksanen, T. 1981. Lemmings (Lemmus lem-mus) and grey-sided voles (Clethrionomys rufocanus) ininteractions with their resources and predators onFinnmarksvidda, northern Norway. – Rep. Kevo Subarct.Res. Stn. 17: 7–31.

Petty, S. J. 1992. The ecology of the tawny owl, Strix aluco, inthe spruce forests of Northumberland and Agryll. – TheOpen Univ., Milton Keynes, p. 295.

Petty, S. J. and Fawkes, B. L. 1997. Clutch size variation intawny owls (Strix aluco) from adjacent valley systems: canthis be used as surrogate to investigate temporal and spatialvariation in vole density. – In: Duncan, J. R., Johnson, D.H. and Nicholls, J. H. (eds), Biology and conservation ofowls of the northern hemisphere. USDA Forest ServiceNorth-Central Forest Research Station, General TechnicalReport 140–190, pp. 314–324.

Pollock, K. H. 1982. A capture-recapture design robust tounequal probability of capture. – J. Wildl. Manage. 46:752–757.

Pusenius, J. and Viitala, J. 1993. Varying spacing behaviour ofbreeding field voles, Microtus agrestis. – Ann. Zool. Fenn.30: 143–152.

Rodd, F. H. and Boonstra, R. 1988. Effects of adult meadowvoles, Microtus pennsyl�anicus, on young conspecifics in fieldpopulations. – J. Anim. Ecol. 57: 755–770.

Rossiter, M. C. 1996. Incidence and consequences of in-herited environmental effects. – Annu. Rev. Ecol. Syst. 27:451–476.

Ryan, C. A. 1990. Proteinase inhibitors in plants. Genes forimproving defenses against insects and pathogens. – Annu.Rev. Phytopathol. 28: 425–449.

Sanders, E. H., Gardner, P. D., Berger, P. J. and Negus, N. C.1981. 6-Methoxybenzoxazolinone: a plant derivative thatstimulates reproduction in Microtus montanus. – Science214: 67–69.

Sheaffer, C. C. and Marten, G. C. 1990. Alfalfa cuttingfrequency and date of fall cutting. – J. Prod. Agric. 3:486–491.

Sheaffer, C. C., Barnes, D. K., Warnes, D. D. et al. 1992.Seeding-year cutting affects winter survival and its associa-tion with fall growth score in alfalfa. – Crop Sci. 32:225–231.

Shimada, T. 1994. A new concept on the critical harvest periodof forage crops in autumn. Japan-Russia workshop: Lowtemperature physiology and breeding of northern crops.Hokkaido National Agricultural Experiment Station, Sap-poro, pp. 35–41.

Spears, N. and Clarke, J. R. 1988. Selection in field voles(Microtus agrestis) for gonadal growth under short photope-riod. – J. Anim. Ecol. 57: 61–70.

Stenseth, N. C. 1999. Population cycles in voles and lemmings:density dependence and phase dependence in a stochasticworld. – Oikos 87: 427–461.

Stenseth, N. C., Bjørnstad, O. N. and Falck, W. 1996. Is spacingbehaviour coupled with predation causing the microtinedensity cycle? A synthesis of current process-oriented andpattern-oriented studies. – Proc. R. Soc. Lond. B 263:1423–1435.

Tkadlec, E. and Zejda, J. 1998. Small rodent populationfluctuations: the effects of age structure and seasonality. –Evol. Ecol. 12: 191–210.

Turchin, P. 1995. Population regulation: old arguments and anew synthesis. – In: Cappuccino, N. and Price, P. W. (eds),Population dynamics; new approaches and synthesis. Aca-demic Press, pp. 19–40.

White, G. 2001. Program MARK. – Available at �http://www.cnr.colostate.edu/�gwhite/�.

White, G. C. and Burnham, K. P. 1999. Program MARK:survival estimation from populations of marked animals. –Bird Study 46 Suppl.: 120–138.

Ylonen, H. 1989. Weasels Mustela ni�alis suppress reproductionin cyclic bank voles Clethrionomys glareolus. – Oikos 55:138–140.

OIKOS 95:2 (2001) 197