1. Introduction
The three universal ecological responses to global warming are proposed and tested: shifts of species distributions [
1], changes of phenology in plants and animals [
2], and reduction in body size [
3]. The distribution shifts have been quantified: the median rate of the shift to higher latitudes is 16.9 km per decade [
1]. However, the details of the changes in body size are not yet known [
4]. The magnitude of body size shrinkage differs across species [
5]; therefore, heterogeneity of responses might have unpredictable consequences for ecosystems [
4].
Mammals exhibit a wide range of body size responses to climate changes, yet the underlying mechanisms remain poorly understood. Changes in the principal size components can be related to both body length and body mass, as evidenced by Li et al. [
6]. Consequently, comparisons between different species and populations are often challenging. In our understanding, the body condition index based on these two size measures [
7] can reflect both changes equally [
8].
In contrast, some mammals are capable of increasing their body size [
9], a phenomenon that is contingent upon a number of factors, including the availability of resources, the passage of time, and the geographical context, which encompasses factors such as latitude, the presence of an island habitat, and other environmental variables [
10]. Temporal changes in body size can differ due to life history traits, such as reproductive rate, and habitat. However, investigations of these factors have been limited [
11].
In the context of global climate change, the evolution of wildlife is an unavoidable phenomenon [
12]. To survive, species can either shift their ranges or adapt. Adaptations can be observed at the level of animal ecology, behavior, and physiology, all of which are related to body size [
13]. The body size of a species tends to increase in a latitudinal and elevational gradient, a phenomenon known as Bergmann’s rule [
14]. This phenomenon is more pronounced in small mammals, such as the herb field mouse (
Apodemus uralensis) [
15], than in shrews [
16].
The impact of disturbances, such as urbanization, on animal body size and fitness is less understood. However, it is likely that these disturbances will affect animal fitness due to increased temperatures, habitat changes and resource availability [
17]. For rodents, it has been demonstrated that climate change and human-caused disturbance have a simultaneous influence on population decline in various ecosystems [
18].
An understanding of the fitness of animals provides a more comprehensive knowledge about the ecosystems in which they live. Body condition is related to reproduction [
19]. However, intensive reproduction has benefits for population growth but also imposes costs on the individual [
8]. Furthermore, better body condition leads to better survival, enabling animals to overcome weather constraints [
20]. However, lower body mass, which is reflected in a lower body condition index, results in better overwintering in several vole species, as evidenced by studies by Ergon et al. [
21], Balčiauskienė et al. [
22], and Zub et al. [
23]. Fitness can be related to the survival of species [
24], which has implications for species conservation and habitat management [
25]. Besides, body condition of small mammals can be used as an indicator of environmental pollution [
26,
27,
28].
Among the various methods employed to assess the fitness of animals, there are those based on:
Morphological measurements, such as body mass, body length, and indices based on various other traits;
Physiological indicators, such as hormone levels or blood parameters;
Reproductive success, including litter size or breeding frequency;
Survival, including mortality and longevity;
Immunology, including the measurement of immune responses or pathogen loads.
These methods are species-specific and are described in greater detail in [
25].
The assessment of fitness in small mammals is a topic of ongoing debate, with various body condition indices (BCIs) being employed to this end. The optimal BCI remains a matter of contention [
29]. The use of body mass as a proxy for fitness in wild animals is limited due to the incomplete understanding of the relationship between the two variables [
30]. Various indices calculated as a ratio of body mass to body size have been found to function better, but they are still not without critics [
31]. Additionally, estimation of fat amount—either visually, through dissection, or using ultrasound and other techniques—is employed [
24], as is the estimation of lean body mass [
32]. While we agree with the aforementioned authors, the proposed measures have two shortcomings. Firstly, some of them necessitate the use of specific laboratory equipment [
33], and secondly, there is a lack of retrospective data for their comparisons.
Given the natural variability of size in mammals [
34], it is to be expected that extreme values of body mass/BCI will be observed in small mammals. While the presence of overweight individuals is a component of the Chitty effect [
35,
36], the low body mass of various small mammal species remains outside the scope of this analysis, with the exception of the analysis conducted by R. Boonstra and C.J. Krebs [
37]. The researchers observed differences in survival between large and small individuals of four species of voles. However, the range of body mass classified as “small” was quite broad.
Our interest in the BCI extremes in small mammals was also prompted by the observation that a greater proportion of high-cited investigations relate to the higher latitudes [
38,
39,
40,
41,
42] than to mid-latitudes [
43,
44,
45,
46]. Furthermore, the majority of studies have focused on a limited number of species, namely gray voles and shrews, with minimal sampling of other small mammal species in a given area. Additionally, the Chitty effect has been reported as absent in shrews, as evidenced by Nordahl and Korpimäki [
41], yet it has been observed in Lithuania [
8].
The objective of this study was to examine the distribution of small mammals exhibiting either extremely high (referred to as the Chitty effect) or low body mass with respect to their trophic group, species, sex, age, and reproductive intensity. Additionally, we sought to estimate the relationship between body condition index and the mass of individuals of different species required for the Chitty effect to occur.
3. Results
3.1. Body Condition Index Extremes in Small Mammal Species and Trophic Groups
A total of 361 individual small mammals belonging to 11 species exhibited a BCI below 2.0, representing 1.3% of the total number (
Table 2). The observed proportions of individuals with low BCI were not in accordance with the number of analyzed individuals of different species (
Figure 1a; χ
2 = 181.8,
p < 0.0001). The proportion of individuals with low BCI was found to be higher than expected in
S. minutus (χ
2 = 226.5,
p < 0.0001),
S. araneus (χ
2 = 146.8,
p < 0.0001), and
M. arvalis (χ
2 = 8.3,
p < 0.01). The proportions were found to be lower than expected in
A. flavicollis (χ
2 = 57.4,
p < 0.0001),
A. agrarius (χ
2 = 31.0,
p < 0.0001),
C. glareolus (χ
2 = 9.9,
p < 0.01),
and M. musculus (χ
2 = 4.2,
p = 0.05). In other species, the proportions of individuals with a BCI < 2 were found to be in accordance with their representation in the sample (
Table 2). No underweight individuals were found in
M. minutus.
A total of 1502 individuals exhibited BCI > 4, with observed proportions differing from those expected (
Figure 1a, χ
2 = 194.5,
p < 0.0001). In two species,
M. musculus and
M. minutus, the proportion of overweight individuals was 3–6 times greater than expected (χ
2 = 105.0 and 234.4, respectively,
p < 0.0001). In
S. minutus (χ
2 = 4.3,
p < 0.05),
A. agrarius (χ
2 = 24.1,
p < 0.0001), and
A. flavicollis (χ
2 = 23.3,
p < 0.0001), the proportion of overweight individuals was approximately 1.4 times higher than expected (
Table 2).
The proportion of individuals with a BCI > 4 in voles, C. glareolus, was less than expected (χ2 = 87.5, p < 0.0001), as was the proportion in M. arvalis (χ2 = 7.6, p < 0.01), M. agrestis (χ2 = 6.9, p < 0.01), and A. oeconomus (χ2 = 4.4, p < 0.05).
A total of 140 individuals (0.52%) exhibited BCI values greater than 5, which deviated from the expected proportions (
Figure 1a, χ
2 = 56.0,
p < 0.0001). The highest proportion of overweight individuals was observed in
M. musculus and
M. minutus, with a prevalence that was 5.4–13.6 times greater than expected (χ
2 = 39.3 and 226.7, respectively;
p < 0.0001). The other species with a higher than expected proportion of BCI > 5 were
S. minutus (χ
2 = 12.9,
p < 0.001) and
A. oeconomus (χ
2 = 3.9,
p < 0.05). In contrast, the proportion of overweight individuals in
C. glareolus (χ
2 = 22.9,
p < 0.0001) and
M. arvalis (χ
2 = 5.7,
p < 0.02) was less than expected (
Figure 1a,
Table 2).
The proportion of individuals with low or high BCI did not correspond to the number of individuals analyzed from different trophic groups (
Figure 1b). So, among individuals with a BCI < 2, there were four times more insectivores and 1.4 times more herbivores than expected, but nearly eight times fewer granivores and 1.5 times fewer omnivores (χ
2 = 530.4,
p < 0.0001).
Among the individuals with high BCI, the differences were less expressed. Among the individuals with a BCI > 4, there were 1.5 times more granivores, 1.5 times fewer omnivores, and 1.4 times fewer herbivores, while the number of insectivores was approximately the same as expected (χ2 = 202.1, p < 0.0001). Among individuals with a BCI > 5, there were 1.5 times more granivores and 1.4 times more insectivores than expected, but the number of overweight omnivores was only half that expected, and the number of herbivores was 1.25 times less (χ2 = 28.7, p < 0.0001).
3.2. Sex- and Age-Related Issues in Body Condition Index Extremes
A consistent pattern of sex-related variation in extremes of body condition index was not identified. Furthermore, the number of significant differences was limited (
Figure 2a). Among individuals with a BCI < 2, the proportion of females (1:1.75) was higher than expected only in
S. araneus (χ
2 = 6.2,
p < 0.05).
Among individuals with a BCI > 4, the proportion of females was higher than expected in A. agrarius (1:1.15, χ2 = 6.2, p < 0.05) and A. oeconomus (1:14.5, χ2 = 13.7, p < 0.0002). Additionally, a trend was observed in C. glareolus (1:1.20, χ2 = 2.9, p < 0.10), although it did not reach statistical significance. In the remaining species, the observed proportions were consistent with the expected sex ratio for the entire species sample.
The observed male-to-female ratio among individuals with a BCI > 5 did not differ from the expected ratio (
Figure 2a).
Inconsistency with expected age proportions was observed in five small mammal species with a BCI < 2 (
Figure 2b). The number of low-fitted juveniles was found to be higher than expected in
S. araneus (χ
2 = 17.3,
p < 0.001),
S. minutus (χ
2 = 5.4,
p < 0.01), and
C. glareolus (χ
2 = 15.1,
p < 0.001). In
A. oeconomus, adults were overrepresented, while subadults and juveniles were underrepresented (χ
2 = 6.5,
p < 0.05). In
M. arvalis, subadults were overrepresented, while adults were underrepresented compared to expected numbers, though the difference was on the trend level only (χ
2 = 4.6,
p = 0.10).
Among individuals with a BCI > 4, there was a notable overrepresentation of adults and an underrepresentation of juveniles in S. araneus (χ2 = 11.7, p < 0.005). Among individuals with a BCI > 5, juveniles of M. musculus and A. agrarius were overrepresented, while the proportion of adults in both species was lower than expected (χ2 = 7.2 and 6.7, respectively, p < 0.05 in both species).
3.3. Reproduction-Related Issues in Body Condition Index Extremes
The occurrence of extreme values of BCI was less prevalent in reproducing individuals, both males and females. The frequency of reproductive males with a BCI < 2 was only 0.3%, while that of females was 1.2% (
Table 3). The frequency of reproductive males with a BCI > 5 was 0.1%, while that of females was 0.3%. These figures are less than those observed in general (see
Table 2). No reliable statistical differences were found depending on the intensity of reproduction.
3.4. Body Mass of Individuals with Highest Body Condition Indices
A total of 123 small mammals of known age were identified in the sample, with a BCI > 5. Of these, 61 were juveniles, 33 were subadults, and 29 were adults (
Table 4). The distribution of age groups across species was similar, with some exceptions.
No juveniles were present among individuals characterized by BCI > 5 in S. araneus, S. minutus, and M. agrestis, no subadults were present in M. arvalis and A. oeconomus, and no adult individuals were present in A. agrarius and M. arvalis. These species are representative of insectivores, granivores, and herbivores.
The range of body masses producing BCI values greater than 5 was likely to be quite broad, with a ratio between the minimum and maximum values being less than 2 in shrews,
M. musculus,
M. minutus, and
C. glareolus (
Table 4). This ratio was generally higher in granivores and herbivores, with a maximum in
A. flavicollis of 2.5 in juveniles and 3.3 in adults, and in
A. oeconomus, 2.6 in juveniles and 2.4 in adults.
3.5. Chitty Effect? Maximum Body Mass and BCI of Adult Individuals
Our findings indicate that the highest BCI values in small mammals were not correlated with body mass. In other words, the heaviest individuals were not necessarily characterized by the highest BCI (
Table 5). The threshold body mass for each species was arbitrarily set. Given the absence of differences in the male-to-female proportion among the heaviest individuals, we did not present the sex ratio.
The highest proportions of the heaviest individuals, exceeding 5% of the total number of adult individuals, were observed in
M. minutus and
N. fodiens. In contrast, in
C. glareolus,
A. flavicollis, and
M. arvalis, these proportions were less than 1% (
Table 5).
In only two species, N. fodiens and M. agrestis, the maximum observed BCI was observed in one of the heaviest individuals.
The heaviest individuals exhibited BCI values that were significantly higher than those observed in all adult individuals of C. glareolus (p < 0.01), as well as in S. araneus, S. minutus, M. minutus, and M. arvalis (p < 0.05). In other species, these differences were not found to be statistically significant.
4. Discussion
The results demonstrate that only a small proportion of individuals exhibited extreme body condition indices. Those with a BCI < 2 comprised 1.33%, and those with a BCI > 5 just 0.52% of the analyzed small mammals. Among individuals with a BCI < 2, the proportion of insectivores and herbivores was higher, while that of granivores and omnivores was less than expected. Among individuals with a BCI > 5, the proportion of granivores and insectivores was higher, while that of omnivores and herbivores was less than expected.
The average proportion of extra-large adults was 1.0%, with the highest proportion observed in M. minutus and N. fodiens (6.9% and 6.3%, respectively). The least represented large adults (>1%) were in C. glareolus, A. flavicollis, and M. arvalis. This finding is compared to data from other populations.
The absence of extra-large individuals in
S. araneus was reported for Finland [
41]. Furthermore, the loss of cyclic fluctuations, which has resulted in the disappearance of the Chitty effect in eight shrew species, particularly those with non-cyclic population dynamics, was reported from the Middle Yenisey at high latitude [
58,
59]. Consequently, our report of the maximum body mass observed in three shrew species from Lithuania represents a novel contribution to the existing body of knowledge.
In the California vole (
Microtus californicus), the proportion of extra-large individuals was 12.7% [
44]. In the meadow vole (
Microtus pennsylvanicus), the proportion extra-large individuals during a three-year period was 0.47–0.77% [
60].
A mere 0.06% of
C. glareolus individuals with a body mass of 33.5–35.0 g was observed in the Białowieża National Park, Poland, a region with a similar latitude to Lithuania [
61]. The proportion of overweight
C. glareolus in our sample was significantly higher (χ2 = 14.8,
p < 0.001) at 0.4%, despite the use of a minimum body mass of 35.0 g to select for this category (see
Table 5). In the Middle Ural region,
C. glareolus with body mass over 35.0 g comprised 0.98% of all individuals [
42], which is a significantly (χ
2 = 7.0,
p < 0.01) higher proportion than in our sample, this being 0.38%.
In a study conducted in high latitudes by J. Tast in 1972, no individuals of the species
A. oeconomus were reported to exceed the size of an extra-large individual [
62]. However, in an earlier study by the same author, 6.5–10.0% of males were reported to be very large, exceeding 80 g [
63]. In our study, the maximum body mass of this species was 77.0 g, which is comparable to the body mass reported for Germany by G.H.W. Stein in 1952 [
64].
D. Chitty [
38] reported no individuals of
M. agrestis with a body mass exceeding 40.2 g for males and 36.2 g for females. However, the minimum body mass of adult females (18.0 g) and males (23.0 g) was comparable to that observed in our sample. We found a total 15 out of 301 adult
M. agrestis to be of low weight, with males weighing between 18 and 21.5 g and females weighing between 16.5 and 21.3 g. These figures are somewhat lower than those reported in [
38].
Similarly to our findings, the minimum and maximum body masses of adult
A. agrarius from Slovakia were reported to be 14.5–44.0 for males and 14.0–46.0 for females [
65]. However, the proportion of these large individuals in the sample was not presented.
The viability of large-sized Townsend’s vole (
Microtus townsendii), prairie vole (
Microtus ochrogaster),
M. pennsylvanicus, and
M. californicus was demonstrated in [
37], yet the authors do not present the numbers of these individuals. Another paper, which deals with body mass of six small mammal species, does not present data that can be used in our context [
66]. Therefore, we believe that the reference data provided are valuable for comparison with other territories, as well as species that have not been analyzed in this respect. In our dataset, the issue of measurement bias, which was highlighted in [
7,
66], has been effectively addressed.
The proportions of underfit individuals exhibited significant differences from those expected in S. araneus, S. minutus, A. flavicollis, M. arvalis, A. agrarius, C. glareolus, and M. musculus. Additionally, the proportions of overfit individuals in the last four species differed from those expected.
In our sample,
M. minutus was distinguished by the absence of underfit individuals and the highest proportion of overfit individuals. In Austria, W. Haberl and B. Kryštufek reported the maximum body mass of adult females as 12.0 g, and that of adult males as 10.6 g [
67]. In the high-latitude site in Finland, the maximum adult body mass was 13.5 g in females and 9.0 g in males [
68]. In our sample, adult females with a body mass > 12.0 g comprised 10%, while the maximum body mass in males was 13.5 g.
The observed male-to-female ratio among individuals with a BCI > 5 and those with a BCI < 2 did not differ from the expected ratio in our sample; the proportion of females was higher than expected only in
S. araneus. This can be explained by the life history of this species, which involves short age and large litters at considerable energetic costs [
69,
70]. Consequently, the reproductive cost represents a trade-off between fitness and body condition [
19].
Additionally, our findings revealed that the age groups of certain small mammal species with extreme BCI values did not adhere to the expected distribution. In comparison to the anticipated ratio, the number of juveniles was higher than expected in S. araneus, S. minutus, and C. glareolus, while the number of subadults was disproportionately represented in M. arvalis, and the number of adults was overrepresented in A. oeconomus.
With regard to shrews, an explanation may be sought in their metabolic rate, which is high in shrews in general and particularly high in the smallest ones (juveniles). This is coupled with the specificity of their diet, which contains 100% of the high-quality food [
71]. It is also possible that differences in the diet of
C. glareolus juveniles may be expected in connection with their use of lower-quality habitats [
72]. The most promising explanation for the low BCI in subadults is that they experience stress and competition at postnatal dispersion [
73,
74]. Low body mass
A. oeconomus are characterized by better survival in the beginning of winter [
23]; moreover, this species does not tolerate even short-term food deprivation [
75].
Our results indicated that the proportion of juvenile
M. musculus and
A. agrarius among individuals with a BCI > 5 was higher, while the proportion of adults was lower than expected for both species. The high body condition in juveniles may be related to a lower pathogen load, which allows for more energy to be utilized for growth [
76], and possibly to maternal care. It can be postulated that species with high adaptability and potential for spread should exhibit better body condition in juveniles and adults. This is evidenced by the fact that both
M. musculus, which occupies a wide range of habitats [
77], and
A. agrarius, which has recently been documented to have a wide distribution [
78,
79], exhibit this characteristic. However, data on the body condition of these species from different regions within their ranges are currently unavailable.
Consequently, we posit that the comparison of the extremes of body condition for the most abundant small mammal species in a single region, employing a methodology that permits the re-recruitment of retrospective data from other sites, represents a significant advantage of this publication.
It is important to acknowledge the limitations of this study. Firstly, it is not possible to retrospectively assess the nutritional condition of individuals. Our BCI was traditional (based on morphometry); therefore it was not possible to correctly assess the nutritional condition of an individual due to the dynamics of body tissues [
80]. It was not possible to perform retrospective measurements of body composition, as in [
81], and thus we relied on the assumption that higher BCI values were beneficial to the individual. Secondly, we did not relate BCI values to population cycles across species. In Lithuania, population cycles are not a mandatory attribute of small mammal species [
82]. This is also true of other territories outside high latitudes [
83]. Furthermore, in the 21st century, many populations of small mammals may undergo a transformation to a noncyclic dynamic [
59].