Experimental tests of predation and food hypotheses for population cycles of voles

Pronounced population cycles are characteristic of many herbivorous small mammals in northern latitudes. Although delayed density-dependent effects of predation and food shortage are often proposed as factors driving population cycles, firm evidence for causality is rare because sufficiently replicated, large-scale field experiments are lacking. We conducted two experiments on Microtus voles in four large predator-proof enclosures and four unfenced control areas in western Finland. Predator exclusion induced rapid population growth and increased the peak abundance of voles over 20-fold until the enclosed populations crashed during the second winter due to food shortage. Thereafter, voles introduced to enclosures which had suffered heavy grazing increased to higher densities than voles in previously ungrazed control areas which were exposed to predators. We concluded that predation inhibits an increase in vole populations until predation pressure declines, thus maintaining the low phase of the cycle, but also that population cycles in voles are not primarily driven by plant-herbivore interactions.     

Landscape effects on temporal and spatial properties of vole population fluctuations

Populations of northern small rodents have previously been observed to fluctuate in spatial synchrony over distances ranging from tens to hundreds of kilometers between sites. It has been suggested that this phenomenon is caused by common environmental perturbations, mobile predators or dispersal movements. However, very little focus has been given to how the physical properties of the geographic area over which synchrony occurs, such as landscape composition and climate, affect spatial population dynamics. This study reports on the spatial and temporal properties of vole population fluctuations in two areas of western Finland: one composed of large interconnected areas of agricultural farmland interspersed by forests and the other highly dominated by forest areas, containing more isolated patches of agricultural land. Furthermore, the more agricultural area exhibits somewhat milder winters with less snow than the forested area. We found the amplitude of vole cycles to be essentially the same in the two areas, suggesting that the relative amount of predation on small rodents by generalist versus specialist predators is similar in both areas. No seasonal differences in the timing of synchronization were observable for Microtus voles, whereas bank vole populations in field habitats appeared to become synchronized primarily during winter. Microtus populations in field habitats exhibited smaller spatial variation and a higher degree of synchrony in the more continuous agricultural landscape than in the forest-dominated landscape. We suggest that this inter-areal difference is due to differences in the degree of inter-patch connectivity, with predators and dispersal acting as the primary synchronizing agents. Bank vole populations in field habitats were more synchronized within the forest-dominated landscape, most likely reflecting the suitability of the inter-patch matrix and the possibility of dispersal. Our study clearly indicates that landscape composition needs to be taken into account when describing the spatial properties of small rodent population dynamics.     

Spatio‐temporal covariation in abundance between the cyclic common vole Microtus arvalis and other small mammal prey species

Populations of the common vole Microtus arvalis in mid‐western France show cyclic dynamics with a three‐year period. Studies of cyclic vole populations in Fennoscandia have often found inter‐specific synchrony between the voles and other small mammals which share the voles' predators. Although predators are central to the favoured mechanism to explain Fennoscandian vole cycles and the spatial variation of small mammal populations, their role in vole cycles elsewhere, including France, is less clear. Establishing whether alternative prey species in France cycle in parallel with voles as they do in Fennoscandia is thus an important step towards understanding the generality of predators' influence on cyclic vole populations. We applied spatial and temporal autocorrelation and cross‐correlation methods to French populations of M. arvalis and two sympatric non‐cyclic small mammal species, Apodemus sylvaticus and Crocidura russula. Patterns of time‐lagged cross‐correlation between the abundance of M. arvalis and the other two species suggested synchrony in their dynamics beyond that expected of stochastic environmental variation, and indicated a weak three‐year cycle in A. sylvaticus and C. russula that was in phase with that of M. arvalis. We interpret the synchrony between these species as the effect of shared predators and environmental stochasticity. Abundance within species showed weak spatial autocorrelation in June at scales consistent with dispersal being the mechanism responsible, but a more general lack of spatial structure within and between species was consistent with the strong spatial synchrony at regional scales often found in fluctuations of small mammal abundance.     

Spatial dynamics of Microtus vole populations in continuous and fragmented agricultural landscapes

Small mammal populations often exhibit large-scale spatial synchrony, which is purportedly caused by stochastic weather-related environmental perturbations, predation or dispersal. To elucidate the relative synchronizing effects of environmental perturbations from those of dispersal movements of small mammalian prey or their predators, we investigated the spatial dynamics of Microtus vole populations in two differently structured landscapes which experience similar patterns of weather and climatic conditions. Vole and predator abundances were monitored for three years on 28 agricultural field sites arranged into two 120-km-long transect lines in western Finland. Sites on one transect were interconnected by continuous agricultural farmland (continuous landscape), while sites on the other were isolated from one another to a varying degree by mainly forests (fragmented landscape). Vole populations exhibited large-scale (>120 km) spatial synchrony in fluctuations, which did not differ in degree between the landscapes or decline with increasing distance between trapping sites. However, spatial variation in vole population growth rates was higher in the fragmented than in the continuous landscape. Although vole-eating predators were more numerous in the continuous agricultural landscape than in the fragmented, our results suggest that predators do not exert a great influence on the degree of spatial synchrony of vole population fluctuations, but they may contribute to bringing out-of-phase prey patches towards a regional density level. The spatial dynamics of vole populations were similar in both fragmented and continuous landscapes despite inter-landscape differences in both predator abundance and possibilities of vole dispersal. This implies that the primary source of synchronization lies in a common weather-related environment.     

Multiple predators induce risk reduction in coexisting vole species

Large predators may affect the hunting efficiency of smaller ones directly by decreasing their numbers, or indirectly by altering their behaviour. Either way this may have positive effects on the density of shared prey. Using large outdoor enclosures, we experimentally studied whether the presence of the Tengmalm's owl Aegolius funereus affects the hunting efficiency of the smallest member of the vole-eating predator guild, the least weasel Mustela nivalis, as measured by population responses of coexisting prey species, the field vole Microtus agrestis and the sibling vole M. levis . We compared the density and survival probability of vole populations exposed to no predation, weasel predation or combined predation by a weasel and an owl. The combined predation of both owl and weasel did not result in obvious changes in the density of sibling and field vole populations compared to the control populations without predators, while predation by least weasel alone decreased the densities of sibling voles and induced a similar trend in field vole densities. Survival of field voles was not affected by predator treatment while sibling vole survival was lower in predator treated populations than in control populations. Our results suggest that weasels are intimidated by avian predators, but without changing the effects of predators on competitive situations between the two vole species. Non-lethal effects of intraguild predation therefore will not necessarily change competitive interactions between shared prey species.     

Predation rate, prey preference and predator switching: experiments on voles and weasels

We studied the predation rate and prey selection of the least weasel ( Mustela nivalis nivalis ) on its two most common prey species in boreal environments, the bank vole ( Clethrionomys glareolus ) and the field vole ( Microtus agrestis ), in large outdoor enclosures. We also studied the response of weasels to odours of the two species in the laboratory. The enclosure experiment was conducted using constant vole densities (16 voles/ha) but with varying relative abundance of the two species. Weasels showed higher predation rates on bank voles, and males had higher predation rate than females. Females killed disproportionately more of the more abundant prey species, but they preferred bank voles to field voles when both were equally available. Overall, the predation rate also increased with increasing abundance of bank voles. Therefore our results are in agreement with earlier laboratory results showing preference for bank voles, even if no intrinsic preference for odours of either species was observed in our laboratory study. We suggest that the least weasel hunts according to prey availability, prey aggregation and suitability of hunting habitat, and that this causes the observed dependence of least weasels on field voles and emphasises the role of the field vole in the vole-weasel interaction in cyclic vole populations. Furthermore, our results suggest that predation by weasels may facilitate the coexistence of the two vole species via predator switching, and that it may cause the observed synchrony in dynamics between vole species.     

Species‐specific limitation of vole population growth by least weasel predation: facilitation of coexistence?

Interspecific competition is usually understood as different species competing directly with each other for limited resources. However, predators can alter such competitive interactions substantially. Predation can promote the coexistence of species in a situation where it would otherwise be impossible, for example if a tradeoff between the competitive abilities and predation resistance of the prey species exists. The field vole Microtus agrestis and the sibling vole M. rossiaemeridionalis are sympatric grassland species, which compete for the same resources. At the population level sibling voles are suggested to be superior competitors to field voles, yet more vulnerable to predation. We tested the effects of predation on the two species in 0.5 ha outdoor enclosures by exposing vole populations to radio-collared freely-hunting least weasels Mustela nivalis nivalis for three weeks. Lethal and non-lethal impacts of predation limited population densities of both species during and after the experimental period, but the effect was more pronounced in sibling voles in which population densities decreased markedly during the treatment period and even after that. Field vole population densities remained stable under weasel predation, while densities increased in controls. Survival in both species was lower in treatment populations compared to controls, but the effect tended to be more pronounced in sibling voles and in females of both species. The average mass of adults in both species declined in the treatment populations. These results suggest that predation by least weasels can limit vole populations locally, even during favourable summer conditions, and have extended negative effects on the dynamics of vole populations. In addition, predation alleviated interspecific competition between the vole species and is, therefore, a potential factor enabling the coexistence of them.     

The predation risks of interspecific eavesdropping: weasel–vole interactions

Competing species benefit from eavesdropping on each other's signals by learning about shared resources or predators. But conspicuous signals are also open to exploitation by eavesdropping predators and should also pose a threat to other sympatric prey species. In western Finland, sibling voles Microtus rossiameridionalis and field voles M. agrestis compete for food and space, and both species rely upon scent marks for intraspecific communication. Both vole species are prey to a range of terrestrial scent hunting predators such as least weasels, however, the competitively superior sibling voles are taken preferentially. We tested in large out‐door enclosures whether field voles eavesdrop on the signals of its competitor, and whether they behave as though this eavesdropping carries a risk of predation. We presented field voles with scent marks from unknown conspecifics and sibling voles and measured their visitation, activity and scent marking behaviours at these scents under high (weasel present) and low (weasel absent) predation risk. Field voles readily visited both field and sibling vole scents under both high and low predation risk; however their activity at sibling vole scent marks declined significantly under increased predation risk. In contrast, predation risk did not affect field voles’ activity at conspecific scents. Thus, field voles were compelled to maintain eavesdropping on heterospecific scents under an increased risk of predation, however they compensated for this additional risk by reducing their activity at these risky scents. Scent marking rates declined significantly under high predation risk. Our results therefore reveal a hidden complexity in the use of social signals within multi‐species assemblages that is clearly sensitive to the potential for increased predation risk. The predation risks of interspecific eavesdropping demonstrated here represents a significant generalisation of the concept of associational susceptibility.     

Competition, predation and interspecific synchrony in cyclic small mammal communities

Although competition and predation are considered to be among the most important biotic processes influencing the distribution and abundance of species in space and time, the relative and interactive roles of these processes in communities comprised of cyclically fluctuating populations of small mammals are not well known. We examined these processes in and among populations of field voles, sibling voles, bank voles and common shrews in western Finland, using spatially replicated trapping data collected four times a year during two vole cycles (1987–1990 and 1997–1999). Populations of the four species exhibited relatively strong interspecific temporal synchrony in their multiannual fluctuations. During peak phases, we observed slight deviations from close temporal synchrony: field vole densities peaked at least two months earlier than those of either sibling voles or bank voles, while densities of common shrews peaked even earlier. The growth rates of all four coexisting small mammal species were best explained by their own current densities. The growth rate of bank vole populations was negatively related to increasing densities of field voles in the increase phase of the vole cycle. Apart from this, no negative effects of interspecific density, direct or delayed, were observed among the vole species. The growth rates of common shrew populations were negatively related to increasing total rodent (including water voles and harvest mice) densities in the peak phase of the vole cycle. Sibling voles appeared not to be competitively superior to field voles on a population level, as neither of these Microtus voles increased disproportionately in abundance as total rodent density increased. We suggest that interspecific competition among the vole species may occur, but only briefly, during the autumn of peak years, when the total available amount of rodent habitat becomes markedly reduced following agricultural practices. Our results nonetheless indicate that interspecific competition is not a strong determinant of the structure of communities comprised of species exhibiting cyclic dynamics. We suggest that external factors, namely predation and shortage of food, limit densities of vole populations below levels where interspecific competition occurs. Common shrews, however, appear to suffer from asymmetric space competition with rodents at peak densities of voles; this may be viewed as a synchronizing effect.     

Vole population cycles in northern and southern Europe: is there a need for different explanations for single pattern?

1. Students of population cycles in small rodents in Fennoscandia have accumulated support for the predation hypothesis, which states that the gradient in cycle length and amplitude running from southern to northern Fennoscandia reflects the relative influence of specialist and generalist predators on vole dynamics, itself modulated by the presence of snow cover. The hypothesized role of snow cover is to isolate linked specialist predators, primarily the least weasel, Mustela n. nivalis L. and their prey, primarily field voles Microtus agrestis L., from the stabilizing influence of generalist predators. 2. The predation hypothesis does not readily account for the high amplitude and regular 3-year cycles of common voles documented in agricultural areas of western, central and eastern Europe. Such cycles are rarely mentioned in the literature pertaining to Fennoscandian cycles. 3. We consider new data on population cycles and demographic patterns of common voles Microtus arvalis Pallas in south-west France. We show that the patterns are wholly consistent with five of six patterns that characterize rodent cycles in Fennoscandia and that are satisfactorily explained by the predation hypothesis. They include the: (a) existence of cycle; (b) the occurrence of long-term changes in relative abundance and type of dynamics; (c) geographical synchrony over large areas; (d) interspecific synchrony; and (e) voles are large in the increase and peak phase and small in decline and low phase, namely. There is a striking similarity between the patterns shown by common vole populations in south-west France and those from Fennoscandian cyclic rodent populations, although the former are not consistent with a geographical extension of the latitudinal gradient south of Fennoscandia. 4. It is possible that the dominant interaction leading to multiannual rodent oscillations is different in different regions. We argue, however, that advocates of the predation hypothesis should embrace the challenge of developing a widely applicable explanation to population cycles, including justifying any limits to its applicability on ecological and not geographical grounds.     

Smaller Microtus vole species competitively superior in the absence of predators

Interspecific competition is assumed to generate negative effects on coexisting species, possibly including slower population growth and lower survival. The field vole ( Microtus agrestis ) and the sibling vole ( M. rossiaemeridionalis ) are sympatric close relatives which compete for similar resources. Previous non-experimental studies suggest that the smaller sibling vole is a superior competitor, yet more vulnerable to predation than the larger field vole. We studied the effects of coexistence on population densities, reproductive parameters, and survival in these two species by means of experimentation in large, predator-free outdoor enclosures. While populations of both species reached higher densities in the absence of the other, field voles appeared to suffer more from interspecific competition than sibling voles. The proportion of young individuals in the population was higher in the sibling vole than in the field vole at the end of the experiment. The presence of a coexisting species reduced the survival of field voles. Sibling voles, on the other hand, appeared to suffer more from intraspecific competition than interspecific competition. On a population level, the sibling vole seems to be a superior competitor in the absence of predators due to better survival and possibly a higher reproductive capacity. However, predation probably has a profound influence on the interspecific dynamics of these two species indicating that in natural surroundings apparent competition (i.e. competition via shared predators) is stronger than direct competition.     

Experimental tests of the role of predation in the population dynamics of voles and lemmings

• 1 Reasons for fluctuating populations of small mammals have been intensively investigated since the early days of modern ecology. Particular interest has been taken in vole populations exhibiting multiannual oscillations. Much empirical and theoretical work has been accomplished to find out the key factor(s) driving these population cycles and many reviews have been written about the results.
• 2 One of the most plausible processes for explaining regular fluctuations in small mammals is predation. Here I review the existing literature on the experimental studies of the role of predation in vole population dynamics in the hope that a critical examination of these studies will help researchers improve the design of future experiments.
• 3 Most predation manipulations have been done in exclosures, but there are also studies that have attempted to reduce or increase predator numbers in non‐fenced areas, islands and enclosures.
• 4 As the number of experimental studies has increased, their quality in terms of replication, use of controls and realistic spatial and temporal scales has also improved.
• 5 Most studies have found population‐level effects of predator manipulations on prey populations. The effects have varied from very weak to very strong, reflecting dissimilar experimental designs and the great variety of predator–prey interactions among different kinds of species in different landscapes. Most of these studies show that predation limits population growth of voles, and in some circumstances even regulate vole population fluctuations, but none of them clearly demonstrates that predation consistently changes fluctuation patterns of voles.
• 6 To be able to assess more reliably the true role of predation on (cyclic) population fluctuations of voles, more competent experiments are still needed not only over the geographical range of cyclic population dynamics, but also in areas of weakly or non‐cyclic populations of voles.

     

Seasonal changes in the numerical responses of predators to cyclic vole populations

Theoretical models predict that a delayed density-dependent mortality factor with a time lag of ca 9 months is able to drive 3–5-yr population cycles of northern voles. We studied numerical responses of predators in western Finland during 1986–92, in an area with 3-yr population cycles of voles. Abundances of small mammals were monitored in several farmland areas (each 3 km²) by snap-trapping in April, June, August, and October (only in 1986–90), and the abundances of avian, mammalian, and reptilian predators by visual censuses during trapping occasions. The 3-yr cycle studied was a cycle of Microtus voles (field vole M. agrestis and sibling vole M. rossiaemeridionalis ) and their small-sized predators (small mustelids and vole-eating birds of prey). The numerical responses of both migratory avian predators and small mustelids to changes in vole densities were more alike than different. In late summer (August), the time lag in the numerical response of all main predators was short (0–4 months), whereas longer time lags prevailed from spring to early summer. The length of the time lag in spring appeared to be related to the length of the winter, which indicates that strong seasonality may create longer time lags to the numerical response of predators at northern latitudes than at more southern latitudes. Our results suggest that, from spring to early summer, predation by migratory avian predators may act in concordance with mustelid predation to produce the long time lag necessary to drive the 3-yr cycle of voles, whereas almost direct density-dependent predation by all major predators in late summer may dampen spatial variation in prey densities.     

Functional response of the least weasel, Mustela nivalis nivalis

We investigated the functional response of the least weasel ( Mustela nivalis nivalis ) in a series of experiments conducted in large outdoor enclosures (0.5 ha). Radiocollared Microtus voles were released in the enclosures at different densities (4, 8, 16 and 100/ha) three days before the release of a radiocollared weasel. During the three-day experiment every vole killed was replaced with another one as soon as possible to retain constant prey density. The results demonstrated type II functional response with the predation rate reaching 50% of the asymptotic rate at a vole density of 15 individuals per ha. More voles were killed at the highest densities than would be expected from the known energy demands of weasels. Female and male voles were killed in proportion to their abundances in the enclosures, and no difference in predation risk was detected between voles released in the enclosure before the weasel (residents) and during the experiment (transients).     

Dynamic impacts of feral mink predation on vole metapopulations in the outer archipelago of the Baltic Sea

Predation impacts by introduced predators are predicted to be most intense in island ecosystems, and also variable depending on environmental conditions, but large-scale experimental field testing is rare. In this study we examine the factors that determine the distribution and abundance of vole metapopulations preyed upon by feral American mink Mustela vison in the outer Finnish archipelago of the Baltic Sea. Specifically, we follow the dynamics of field voles Microtus agrestis and bank voles Clethrionomys glareolus on 40 small islands under variable rainfall as part of a large-scale mink removal experiment. For both vole species occupancy rates were negatively influenced by island isolation, as were extinction events for field voles. High summer rainfall in 1998 corresponded to large vole populations where mink were absent, populations that then crashed in 1999 and 2000 when below average rains fell during the summer breeding season. Where mink were present however, vole abundance remained more constant between years with no boom-bust apparent. We conclude that weather and predation may drive vole abundance whereas habitat patchiness and metapopulation processes more strongly drive vole distributions. There may also be potential for interaction between these factors: because feral mink prevent rapid vole population growth after good summer rains, and vole dispersal is influenced by population size, feral mink may be changing vole dispersal patterns to disrupt the natural metapopulation dynamic. Hence this indirect impact of mink could lead to gradual erosion of vole populations in the outer archipelago by reducing recolonisation processes.     

Dynamic effects of predators on cyclic voles: field experimentation and model extrapolation

Mechanisms generating the well-known 3-5 year cyclic fluctuations in densities of northern small rodents (voles and lemmings) have remained an ecological puzzle for decades. The hypothesis that these fluctuations are caused by delayed density-dependent impacts of predators was tested by replicated field experimentation in western Finland. We reduced densities of all main mammalian and avian predators through a 3 year vole cycle and compared vole abundances between four reduction and four control areas (each 2.5-3 km(2)). The reduction of predator densities increased the autumn density of voles fourfold in the low phase, accelerated the increase twofold, increased the autumn density of voles twofold in the peak phase, and retarded the initiation of decline of the vole cycle. Extrapolating these experimental results to their expected long-term dynamic effects through a demographic model produces changes from regular multiannual cycles to annual fluctuations with declining densities of specialist predators. This supports the findings of the field experiment and is in agreement with the predation hypothesis. We conclude that predators may indeed generate the cyclic population fluctuations of voles observed in northern Europe.     

Social overwintering and food distribution in the bank vole Clethrionomys glareolus

We studied overwintering in the bank vole Clethrionomys glareolus in four 0.5 ha enclosures in an abandoned field in central Finland in the winter 1987/88. In two of the enclosures food was offered evenly distributed over the whole enclosed area (Even Enclosures = EE), in the two others food was offered in one feeding patch with four feeding chambers 2 m apart (Patchy Enclosures = PE). Food was provided in about the same amount in both enclosures. The experiment commenced in early October, with 13 females and 11 males in EEs and 12 + 13 voles in PEs. After two months the voles in the PEs were concentrated around the feeding patches. Territoriality was not observed in EEs, instead the voles formed small exclusive overwintering groups consisting of 2-3 females and at least one male. The size of the home range of the females and males was identical during mid-winter as the voles were non-breeding. By the onset of breeding, range size increased in both sexes, but significantly more in males, however. The survival was about the same in all populations. Every population showed a mid-winter decline suggesting the effect of the mustelid predators observed in and around the enclosures. In the PEs the overwintering aggregations lasted until the maturation of the first litters. Food distribution affected the spatial distribution of the populations. We conclude that the patchiness of the habitat and especially the availability of food are the most important factors determining the social structure of overwintering populations.     

Predator-induced synchrony in population oscillations of coexisting small mammal species

Comprehensive analyses of long-term (1977-2003) small-mammal abundance data from western Finland showed that populations of Microtus voles (field voles M. agrestis and sibling voles M. rossiaemeridionalis) voles, bank (Clethrionomys glareolus) and common shrews (Sorex araneus) fluctuated synchronously in 3 year population cycles. Time-series analyses indicated that interspecific synchrony is influenced strongly by density-dependent processes. Synchrony among Microtus and bank voles appeared additionally to be influenced by density-independent processes. To test whether interspecific synchronization through density-dependent processes is caused by predation, we experimentally reduced the densities of the main predators of small mammals in four large agricultural areas, and compared small mammal abundances in these to those in four control areas (2.5-3 km(2)) through a 3 year small-mammal population cycle. Predator reduction increased densities of the main prey species, Microtus voles, in all phases of the population cycle, while bank voles, the most important alternative prey of predators, responded positively only in the low and the increase phase. Manipulation also increased the autumn densities of water voles (Arvicola terrestris) in the increase phase of the cycle. No treatment effects were detected for common shrews or mice. Our results are in accordance with the alternative prey hypothesis, by which predators successively reduce the densities of both main and alternative prey species after the peak phase of small-mammal population cycles, thus inducing a synchronous low phase.     

Regulation of root vole population dynamics by food supply and predation: a two-factor experiment

This paper reports the effects of food supply, predation and the interaction between them on the population dynamics of root voles, Microtus oeconomus , by adopting factorial experiments in field enclosures. This two-factor experiment proved the general hypothesis that food supply and predation had independent and additive effects on population dynamics of root voles. The experimental results proved the following predictions: (1) predation reduced population density and recruitment significantly; (2) food supply increased population density; (3) predation and food supply influenced spacing behavior of root voles separately and additively: Exposure to predation reduced long movements of root voles between trapping sessions; additional food supply reduced aggression level and home range size of root voles. Less movement of individuals that exposed to predators possibly reduced their opportunity of obtaining food and lessened population survival rate, which led population density to decrease. Smaller home range and lower aggression level could make higher population density tolerable. The interactive effect of predation and food on home range size was highly significant (P=0.0082<0.01). The interactive effect of food and predation on dispersal rate was significant (P<0.01). From the experimental results, we conclude that the external factors (predation, food supply) were more effective than internal factors (spacing behavior) in determining population density of root voles – under the most favorable external conditions (−P, +F treatment), the mean density and mean recruitment of root vole population was the highest; under the most unfavorable external conditions (+P, −F treatment), the mean density and mean recruitment of root vole population was the lowest.     

Rate of population change in voles from different phases of the population cycle

Factors involved in causing cyclic vole populations to decline, and in preventing populations from recovering during the subsequent low density phase have long remained unidentified. The traditional view of self-regulation assumes that an increase in population density is prevented by a change in the quality of individuals within the population itself, but this is still inadequately tested in the field. We compared the population growth of wild field voles ( Microtus agrestis ) from the low phase (conducted in 1998) with that of voles from the increase phase (conducted in 1999) in predator-proof enclosures (each 0.5 ha) in western Finland. Within a few months, enclosed vole populations increased to high density, and the realised per capita rate of change over the breeding season did not differ between the populations from different cycle phases. This implies that the recovery of populations from the low phase was not hindered by an impoverishment in quality of individual voles. Accordingly, we suggest that population intrinsic factors (irrespective of the mechanisms they are based on) are unlikely to play a significant role in the generation of cyclic density fluctuations of voles. Instead, we discovered direct density-dependent regulation in the vole populations. Accurate estimates of population growth and the observed density dependence provide important information for empirically based models on population dynamics of rodents.