Climate warming combined with intensive human activities are modifying ecosystems globally, and the Arctic biota is shifting substantially faster than the global average, allowing many new species to expand their range poleward. One such species, is the American beaver (Castor canadensis), a highly specialized rodent capable of greatly modifying ecosystems by altering forest composition through selective foraging and by flooding the landscape through dam and channel building. As rodent cranial morphology is highly related to its functional requirements for foraging and feeding, the beaver provides an opportunity to evaluate the phenotypic response of species to changing environmental conditions. Here, we test the hypothesis that beaver skull morphology is optimized for its local environmental and habitat conditions across Canadian ecosystems. We found that temperature, precipitations, biomass, and local average tree hardness significantly affect the morphology of key masticatory functional traits of the skull, but not its size. Our results suggest that the beaver's phenotype is locally adapted to environmental conditions as a result of its selective foraging behavior. This work provides insight into the adaptive potential of newly established beaver populations in the sub-Arctic to inform management strategies for this keystone species. More generally, our work emphasizes the need to consider traits other than body size in research seeking to better understand the response of species to current global change.

Introduction

The Earth is currently warming at an unprecedented rate, and the Arctic and sub-Arctic are among the most rapidly changing environments, warming 2–3 times faster than the global average (Bush & Lemmen, 2019). Populations may respond to climate change by tracking their preferred habitat via range shift, undergoing rapid evolution, or relying on phenotypic and/or behavioral plasticity to improve their fitness (Baumgartner & Hoffman, 2019; Chevin et al., 2010). The latter two responses will often result in populations being locally adapted in their functional traits and well suited for their new niche requirements (Baumgartner & Hoffman, 2019). Populations may also disperse while becoming locally adapted to a new habitat, gaining a net increase in fitness through both processes (Baumgartner & Hoffman, 2019). Yet, in times of rapid environmental change, it is not clearly established if and how populations evolve their phenotype while also shifting distribution.

Rodents are widely distributed and display a large array of phenotypic adaptations (Baumgartner & Hoffman, 2019; Korth, 1994; Samuels, 2009). Many rodent populations are undergoing rapid morphological change in response to climate change (Baumgartner & Hoffman, 2019; Millien et al., 2017; Pergams & Lawler, 2009; Stumpp et al., 2018; Wolf et al., 2009). One particularly well-suited, yet understudied, species for evaluating the phenotypic response of species to environmental change during range shift is the American beaver (Castor canadensis). Over the last few decades, beaver abundance has drastically increased in sub-Arctic ecosystems across North America (Jarema et al., 2009; Tape et al., 2018), including in Alaska where beaver ponds have doubled in number between 2003 and 2017 (Tape et al., 2022).

Beaver range expansion may have drastic consequences on sub-Arctic ecosystems. Beavers radically impact ecosystems by building dams and networks of channels, consequently altering flow regimes, stream connectivity, and biodiversity (Hood & Larson, 2015; Johnston & Naiman, 1990; Nummi & Holopainen, 2014; Tape et al., 2022). Beavers are also selective herbivores that decrease forest biomass of preferred species, changing the forest composition (Johnston & Naiman, 1990; Mahoney & Stella, 2020). In the sub-Arctic, these impacts may be more pronounced, as water surface area increases caused by beaver activities is known to exacerbate permafrost thaw, tundra soil degradation, and the release of greenhouse gases (Tape et al., 2022). Additionally, beaver dams create warm water refugia for aquatic species, enabling range shifts in other taxa (Tape et al., 2022). Beavers can occupy a given site for many years and their constructions continuing to alter ecosystems even after they have abandoned the site (McMaster & McMaster, 2001; Tape et al., 2022).

The beaver is a prolific agent of ecological change without the need for tools, relying solely on their highly specialized functional traits to enact change. Beavers possess hypsodont cheek teeth adapted for woody and aquatic plant diet (Stefen, 2009), and their ever-growing incisors, specialized for gnawing and anchoring, are used for foraging food and building material (Cox & Baverstock, 2016). As the second largest in size among rodents, with a body mass of up to 35 kg, the beaver is large enough to manipulate significant plant biomass. Further, the beaver's skull configuration is optimized for effective mastication, bite force, and gnawing functions (Cox & Baverstock, 2016), such that their bite force (550–740 N) is greater than expected from body mass and incisor dimensions (Cox & Baverstock, 2016). Beavers have a dorsoventrally deep rostrum and broad zygomatic arch which support important musculature for mastication (Cox & Baverstock, 2016; Korth, 1994; Samuels, 2009). Larger superficial masseter and temporalis muscles are associated with increased mechanical advantage of incisor biting, and are especially important for cutting hard materials such as wood (Samuels, 2009). Therefore, quantifying beaver skull morphology can provide direct insight into their mastication, diet, and foraging functions (Samuels, 2009).

The current distribution of the American beaver ranges from ~25° N latitude in the South, up to ~69° N latitude in the North (Fig. 1; Cassola, 2016), including a vast array of distinct environments and habitats. Yet, little research has investigated the relation between morphological variation in the beaver skull and environmental conditions across its distribution range. Here, we aim to address this gap by testing the hypothesis that beaver skull morphology is optimized for their local environmental and habitat conditions. We found that the beaver skull morphology was related to variation in environmental conditions and vegetation type across the northern portion of its distribution range in North America. This work provides insight into the adaptive potential of newly established beavers in the sub-Arctic to inform management strategies for this keystone species.

Details are in the caption following the image — **Figure 1**
Open in figure viewer PowerPoint

Sample sizes and localities of the study specimens of *Castor canadensis*. The shaded area indicates current known distribution range of *C. canadensis* retrieved from the IUCN (Cassola, 2016), using the sf package in R (Pebesma, 2018).

Materials and methods

Specimens

We sampled 116 beaver skulls, collected in Canada (Fig. 1) between 1885 and 2021 (Appendix S1) and spanning a significant portion of the north-eastern and some western portion of the beaver's distribution (Cassola, 2016). A majority of the museum specimens were collected between 1945 and 1980 (Appendix S1), and only 7 specimens were lacking information on the year of collection. Sampling locations covered the most north-eastern part of the distribution of the beaver in Canada, between 42.60° N and 57.96° N latitude and from 119.73° W to 57.62° W longitude (Fig. 1).

We considered only adult specimens for this study. As the age of each specimen was almost never available, we developed a custom key based on wear patterns of the molars and developmental features of the skull designed to distinguish adult specimens from juveniles and categorizing skulls into five age stages (i.e., 1-juvenile, 2-sub adult, 3-young adult, 4-mid-life adult, and 5-old adult) without the need for destructive sampling (Appendix S2). Skulls of both sexes were pooled, as beavers are not known to exhibit sexual dimorphism in their cranial morphology (Bond, 1956; Goldberg et al., 2011; Jenkins & Busher, 1979).

Morphometric data

Photographs of the ventral view of beaver skulls were taken using a Nikon D3100 camera with a Micro NIKKOR 85 mm lens by the same investigator (JD). We digitized 24 homologous landmarks on the ventral view of the left half of the skull (Appendix S3) using the StereoMorph package (Olsen & Westneat, 2015) in R (R Core Team, 2022).

We estimated measurement error following Claude (2008) by randomly selecting and landmarking three times nine skulls. We ran a Procrustes ANOVA (1000 permutations) using the procD.lm function in the Geomorph package (Adams et al., 2022) to determine the significance of variance among repeats compared with among specimens, and observed no significant effect of the repeats (P > 0.05).

We performed a Generalized Procrustes Analysis (GPA) to superimpose and align the landmark data, visually inspect skull shape variation across our sample and calculate centroid size, the square root of the sum of squared distances of each 24 landmarks from the centroid of the skull, as a proxy of skull size. We opted to use a size proxy calculated from a bony structure, as it has been shown that external measurements traditionally associated with museum specimens, such as head and body length or body mass are more prone to measurement error, and less reliable to study ecogeographical rules in mammals (Crandall et al., 2023). We then ran a Procrustes ANOVA to test for the effects of size and age on skull shape. The relationship between shape and size was significant in our sample of adult beaver skulls (P < 0.001), while age (categories 2–5) was not (P = 0.71). Therefore, we used the residuals of a linear model with skull shape as a response variable and centroid size as an independent variable in subsequent analyses, but did not include age as a covariate. We performed a PCA on the residuals shape coordinates, using the function gm.prcomp in Geomorph and used the first two axes in shape analyses.

Environmental data

We retrieved climatic normals from the AdaptWest project at a 1.0 km² resolution for two variables (AdaptWest Project, 2022), the mean annual temperature (°C) and mean annual precipitation (mm). The AdaptWest project compiles gridded downscaled data from the WorldClim and PRISM data (AdaptWest project, 2022). We retrieved climatic data from AdaptWest for four time periods (1961–1990, 1971–2000, 1981–2010 and 1991–2020) and calculated the average annual temperature and precipitation across these four datasets. We also retrieved three forest attribute variables from the Canada's National Forest Inventory (NFI) for the year 2001 aggregated at a 250 m × 250 m resolution (Beaudoin et al., 2018): treed land cover, total above-ground biomass, and stem wood biomass. The relative abundance of all tree species present within a 1 km² area surrounding skull locations was estimated from NFI's forest composition information, which was then used to calculate weighted average tree hardness for each location using species level hardness data obtained from Forest Products Laboratory (2010; Appendix S4).

We then used the function extract in the R package raster (Hijmans et al., 2023) to extract the mean value of each environmental variable in a 1.0 km² buffer around the sampling location of each skull specimen. As the buffer size was comparable to the home range of the beaver (0.7–1.2 km², Havens et al., 2013), the mean values extracted within the buffer area served as a good indicator of the climatic and environmental conditions experienced by a given individual. Lastly, we calculated an index of tree hardness for each sampling location as the green side hardness (N) of each 22 tree species weighted by their proportion within a 1.0 km² buffer around this location, using data from Forest Products Laboratory (2010) to obtain the green side hardness of the wood for each tree species (Appendix S5). We then performed a Principal Component Analysis (PCA) on the nine climatic and land-cover variables using the prcomp function in the stats package in R.

Statistical analyses

We first tested for the effect of latitude and collection year on the morphology of the skull. We fit a linear model with the log-transformed centroid size as a response variable, and latitude and collection year as independent variables. We then fit a Procrustes linear model with the residual shape coordinates as a response variable and latitude and collection year as independent variables. In both models, we also included an interaction term between latitude and year to evaluate if the pattern of spatial variation in skull morphology had changed over time.

We then ran a series of models to test the effect of the environmental factors on beaver skull morphology. We first tested for the presence of spatial autocorrelation in the skull size and shape variables running a redundancy analysis on the morphological (size and shape) data and a matrix of geographical distances between each sample skull using the function rda in the package vegan (Oksanen et al., 2022). We then performed a multiscale ordination on the skull morphological and geographic distances data using the function mso in the spdep package (Bivand & Wong, 2018) to determine the threshold distance above which no significant spatial autocorrelation was present. We then ran spatial autoregressive linear models using the function errorsarlm in the spatialreg package (Bivand et al., 2013) to account for spatial autocorrelation in the data. These models were weighted by a neighborhood distance, using the functions dnearneigh and nb2listwdist in the spdep package.

Results

Environmental variables

Lower latitude sampling sites were warmer, experienced higher precipitation, and harbored a larger proportion of hard wood tree species. Tree cover, above-ground and wood-stem biomasses did not vary in a linear fashion with latitude, peaking at intermediate latitudes (Appendix S4). Sites with the lowest weighted average tree hardness (green side hardness <1500 N) included locations dominated by white spruce (Picea glauca; 1200 N), balsam fir (Abies balsamea; 1300), trembling aspen (Populus tremuloides; 1300), and/or black spruce (Picea mariana; 1500), whereas sites with the highest weighted average hardness (>4000 N) were all dominated by sugar maple (Acer saccharum; 4300) (Appendix S5).

We retained the first three PCA components with an eigenvalue larger than 1, and that cumulatively explained 83.05% of the variance in the environmental data. The first PCA axis decreased, while the second axis increased with latitude. When considering variable loadings ≥0.30 in absolute value, the first PCA axis (35.69% of the variance) was positively associated with temperature and precipitations and increased with total above-ground and stem-wood biomasses. The second PCA axis (30.48%) decreased with tree hardness, tree cover and mean annual precipitations, and was positively associated with the total above-ground and stem-wood biomasses. Finally, the third axis (16.88%) was negatively associated with tree cover (Fig. 2).

Skull size and shape variation

The centroid size of the skull ranged from 13.21 to 24.03 mm, with a mean value of 18.84 mm. The first four axes of the PCA performed on the allometry-free shape variables explained 53.2% of the total variance. The shape variation along PC1 (22.47%) revealed a shortening of the rostrum, a posterior displacement of the two anchoring points of the zygomatic arch, and an anterior displacement of the tooth row along increasing PC1 values. PC2 (11.89%) was associated with a markedly broader and shorter rostrum as well as an enlargement of the zygomatic arch. PC3 (11.39%) was associated with the positioning of the maximum reach and a lateral constriction of the zygomatic arch and a lateral displacement of the tooth row. Along PC4, which explained 7.42% of the variance, changes in the shape of the zygomatic arch, both in its breadth and positioning of the attachment to the skull, were most notable (Fig. 2).

Drivers of variation in cranial morphology

There was no significant effect of latitude or collection year on the centroid size of the skull while there was a significant effect of latitude on the shape of the skull (P < 0.001), as well as a significant effect of the year of collection, when in interaction with latitude (P < 0.001).

We detected a small but significant spatial autocorrelation in our combined size (centroid size) and shape (the first seven PCA axes) dataset (P < 0.047), and a variogram analysis indicated significant spatial autocorrelation up to a distance of 15 km. This threshold value was used to calculate the spatial weights we included in the spatial autoregressive models. The model with the three PCA environmental axes as independent variables did not explain any significant portion of the variance in the beaver skull size. However, across all models, the first two environmental PCA axes had a significant effect on the shape of the skull (Appendix S6).

Overall, the beaver skull presented a longer, slender snout at low latitudes, in warmer and more humid climates. At higher latitude, beaver skulls tended to be more robust, with a broader rostrum and zygomatic arches expanding further laterally and shifted anteriorly. Skull robustness was associated with decreasing temperature and precipitation, and decreasing total and stem wood biomasses. Most notably, the skull of beavers foraging in areas with harder wood were shorter with zygomatic arches expanded laterally and a displaced cheek teeth towards the jugal side of the skull, presented a distinctly wider rostrum at its posterior end, and a marked anterior shift of the upper incisors (Figs 2 and 3).

Discussion

Here, we quantified beaver skull morphology across Canadian ecosystems and tested the hypothesis that beaver skull morphology is optimized for their local environmental conditions. The most significant variation in skull shape occurred along key functional traits, such as the rostrum, tooth row, zygomatic arch, and anchoring point of the upper incisor, suggesting functional adaptations for local environmental conditions.

It has been previously reported that temperature, precipitation or food availability may impact rodent body size (e.g. Alhajeri & Steppan, 2016; Souto-Lima & Millien, 2014). Here however, we did not detect any relation between the size of the beaver skull and climatic or vegetation variables. However, we found significant variation in the beaver cranial morphology which entailed a shorter and wider rostrum at higher latitudes, in colder and drier environments. Climate variables have been reported to be good predictors of rodent morphology (e.g. Kang et al., 2020; Martínez et al., 2014; McGuire, 2010; Millien et al., 2017). Here, mean annual temperature was associated with an elongation of the rostrum, as previously documented in Millien et al. (2017). Rodent thermoregulation is partly achieved through their nasal turbinates, therefore, increased surface area of the rostrum, which may be accompanied by a higher complexity of the structure of the turbinates, improves that process (Costa et al., 2013; Stumpp et al., 2018). Both precipitations and temperature may also have indirect effects on beaver skull morphology. Colder regions have shorter and sparce vegetation, longer winters, and less productivity (Mekonnen et al., 2021; Rossi et al., 2015) and drier habitats are less productive.

Others have suggested a similar relationship between climate variables, productivity, and morphology (Millien et al., 2006; Wolf et al., 2009). In fact, Millien et al. (2006), argued that climate variables covaried significantly with vegetation making it difficult to discern their relative effects on rodent morphology. Here, in addition to climatic conditions, the skull shape was associated with the local vegetation conditions. In areas dominated by hard wood trees, beaver skulls possessed a larger surface area for muscle attachment on the zygomatic arch and a longer rostrum. Functionally, the rostrum shape and associated masticatory musculature are directly related to feeding strategy and biting efficiency (Cox & Baverstock, 2016). Further, the anterior contact point of the zygomatic arch and the skull is shifted posteriorly at low latitudes, changing the loading angle of the masseter muscles (Cox & Baverstock, 2016; Samuels, 2009). We also found significant variation across beaver functional features related to mastication, such as the placement of the molar tooth row. These results suggest that beaver craniology in low-latitude environments is more adapted for larger bite force, together with further specialization of their masticatory apparatus. This configuration, leading to increased bite force (Maestri et al., 2016), is likely associated with foraging of hard woody trees. Bite force in rodents is related to body mass or skull size (Freeman & Lemen, 2008; Ginot et al., 2018) and has been shown to be influenced by diet (Maestri et al., 2016) or soil hardness in a burrowing species (Kubiak et al., 2018). In line with these findings, our results suggest that beavers exhibit adaptations to increase their bite force when food sources are physically harder. Furthermore, given their avoidance of conifers, beavers in the boreal forest are likely less optimized for tree cutting (Maestri et al., 2016). In regions characterized by deciduous shrubs of intermediate stems and absence of large trees, such as the taiga shield, taiga plain, and the Hudson plain, the beaver is also less likely to forage on woody vegetation. Hence, the beaver may forage woody trees less frequently and consume softer tree species and softer tissued aquatic plants instead (Milligan & Humphries, 2010; Salandre et al., 2017). In these regions, beaver skulls had larger check teeth, typical of chewing herbivores and generalists (Maestri et al., 2016; Samuels, 2009).

Finally, we did not sample the most western regions of Yukon or Alaska of the beaver range at its most northern edge. It has been shown that the beaver is able to forage in tundra ecosystems, below the tree line, with a net increase of evidence of beaver activity and modification of the ecosystems in Alaska (Tape et al., 2022). An examination of the skull morphology in these recently established beaver populations may provide further evidence for the adaptive potential of this species during range expansion.

Conclusion

We captured geographical gradients in both the environment and the cranial morphology in the beaver suggesting it is functionally adapted to its local environment. However, the skull specimens we analyzed here were collected between 1885 and 2021, thereby not overlapping entirely with the climate data (normal 1961–2020) and the forest inventory data (2001) we used to characterize the local environmental conditions experienced by individual beavers, which may limit our ability to capture fully the signature of adaptation in the beaver. Yet, we showed that the morphological variance in the skull of the beaver was well predicted by environmental and climate factors across space, but did not detect any temporal change in the beaver skull morphology. This result is expected, given that the gradients in climate and habitat are much stronger across space than over time, thereby making it challenging to detect significant temporal trends in size or morphology over historical times, as hypothesized in Teplitsky and Millien (2013).

We found significant variation in functional traits associated with foraging that are closely related to forest cover composition. Our findings may shed some light onto how adaptations may relate to local diet (Mahoney & Stella, 2020), including reliance on alternative food sources at the range periphery. The tundra is composed of deciduous shrubs of preferred stem size, and due to longer winters, are consumed for longer periods in their food cache compared to other habitats (Milligan & Humphries, 2010; Tape et al., 2018). Similarly, tundra aquatic vegetation has a shorter growing period, making woody shrubs more important year-round. Evidently, this is dependent on the type of water body a colony is established in, where lotic environments encourage a higher macrophyte diet (Milligan & Humphries, 2010). Based on our findings, beavers in the sub-Arctic appear to be functionally optimized to these environments, with a cranial masticatory morphology reflective of a more generalized diet. Beavers are not just visiting tundra environments; they are functionally well-adapted agents of change in this dynamic and rapidly evolving ecosystem.

Acknowledgments

VM is supported by a NSERC Discovery Grant #2017-03839, MMH was supported by funding from Ouranos and the McGill – Institut nordique du Quebec Northern Research Chair. We thank Mikhaela Neelin for collaborations with the sample collection and Norman and Mark Blake for providing beaver heads from their community. We are grateful to all the natural history museum staff who helped us with data collection: Burton Lim (Royal Ontario Museum), Kamal Khidas (Canadian Museum of Nature), and Anthony Howell (Redpath Museum).

Conflict of interest

The authors declare no conflict of interest.

Author contributions

JD, MMH, and VM conceived the ideas and designed methodology; JD collected the data; JD and VM analyzed the data; JD and VM led the writing of the manuscript. All authors contributed critically to the drafts and gave final approval for publication.

Supporting Information

Filename	Description
jzo13138-sup-0001-AppendixS1.xlsxExcel 2007 spreadsheet , 49 KB	Appendix S1. Specimens included in this study. Geographic coordinates are in decimal degrees. CMN, Canadian Museum of Nature (n = 44); RMMA, Redpath Museum McGill University (n = 17); ROM, Royal Ontario Museum (n = 56). Age categories: 2-sub adult, 3-young adult, 4-mid-life adult, and 5-old adult.
jzo13138-sup-0002-AppendixS2.docxWord 2007 document , 331.5 KB	Appendix S2. Beaver aging key developed to distinguish adults from juveniles, and to categorize relative life stage using the ventral views of the beaver skull. The key was based on previous aging methods for both extinct and extant Castor species but avoids the need for destructive methods (Bejenaru et al., 2015; Mayhew, 1978; Yang et al., 2021). While this key makes the aging process more accessible, the key was not designed to age skulls precisely to the year. Rather, the key was well suited for distinguishing adult specimens from juveniles and categorizing skulls into five age stages (i.e., 1-juvenile, 2-sub adult, 3-young adult, 4-mid-life adult, and 5-old adult), based on wear patterns of the molars and developmental features of the skull. Specimens: CMN_Z-422, CMN_75025, RMMA2021.09.07, CMN_25381, ROM_18583.
jzo13138-sup-0003-AppendixS3.TIFFimage/TIFF, 2 MB	Appendix S3. Positions of the 24 landmarks digitized on the ventral view of the skull and their anatomical definition (specimen pictured: RMMA2021.09.03). 1: most anterior point on the buccal edge of the upper left incisor; 2: most anterior point on the lingual edge of the upper left incisor; 3: anterior most point of the incisive foramen at the midline of the skull; 4: posterior most point of the incisive foramen at the midline of the skull; 5: posterior most point of the infra-orbital foramen; 6: lateral most extent of the left suture between the premaxilla and the maxilla; 7: posterior end of the squamosal root of the zygomatic arch; 8: anterior most lateral point on the suture between the jugal and the maxilla; 9: anterior most point of the glenoid fossa; 10: intersection point at the midline on the anterior face of the first premolar; 11: contact point between the first premolar and the first molar; 12: contact point between the first and the second molars; 13: contact point between the second and the third molars; 14 most posterior point on the occlusal surface of the third molar; 15: extent of the palate at the midline of the skull; 16: midpoint on the suture between the basisphenoid and the basioccipital; 17: posterior tip of the pterygoid process; 18: anterior most external border of the left ectotympanic; 19: posterior most point of the occipital condyle; 20: tip of the paraoccipital process; 21: anterior most lateral margin of the occipital condyle; 22: anterior most point on the dorsal margin of the foramen magnum at the midline of the skull; 23: anterior most point on the ventral margin of the foramen magnum at the midline of the skull; 24: lateral most point on the external edge of the zygomatic arch.
jzo13138-sup-0004-AppendixS4.docxWord 2007 document , 97.6 KB	Appendix S4. Variable attributes included in this study and their variation with latitude; climatic data were retrieved from AdaptWest at https://adaptwest.databasin.org/pages/adaptwest-climatena/ (Retrieved 16 Jul. 2023), while forest attributes data were downloaded from the Open-Government Canada and the Canada Geospatial Data Infrastructure at https://open.canada.ca/data/en/dataset/ec9e2659-1c29-4ddb-87a2-6aced147a990 (Retrieved 16 Jan. 2023).
jzo13138-sup-0005-AppendixS5.docxWord 2007 document , 192.9 KB	Appendix S5. Methodology used to estimate an index of tree hardness at each sampling site, using the forest composition at these sites and the wood hardness of each tree included in the data.
jzo13138-sup-0006-AppendixS6.xlsxExcel 2007 spreadsheet , 17.6 KB	Appendix S6. Spatial autoregressive models with weights calculated for a threshold distance of 15 km. In all models, the first three environmental PCA axes (Dim.1, Dim.2, Dim. 3) were independent variables. Individual models were run for the log-transformed centroid size (Size) and each four principal components of the PCA conducted on the allometry-free shape coordinates (PC1–PC4). ⁺P < 0.1, P < 0.05, P < 0.01, **P < 0.001.

Please note: The publisher is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.

References

Adams, D. C., Collyer, M. L., Kaliontzopoulou, A., & Baken, E. K. (2022). Geomorph: Software for geometric morphometric analyses. R package version 4.0.4.
Google Scholar
AdaptWest Project. (2022). Gridded current and projected climate data for North America at 1km resolution, generated using theClimateNA v7.30 software (T. Wang et al., 2022). adaptwest.databasin.org
Google Scholar
Alhajeri, B. H., & Steppan, S. J. (2016). Association between climate and body size in rodents: A phylogenetic test of Bergmann's rule. Mammalian Biology, 81(2), 219–225.
10.1016/j.mambio.2015.12.001
Web of Science®Google Scholar
Baumgartner, J. M., & Hoffman, S. M. G. (2019). Comparison of the responses of two Great Lakes lineages of Peromyscus leucopus to climate change. Journal of Mammalogy, 100(2), 354–364.
10.1093/jmammal/gyz063
Web of Science®Google Scholar
Beaudoin, A., Bernier, P. Y., Villemaire, P., Guindon, L., & Guo, X. J. (2018). Tracking forest attributes across Canada between 2001 and 2011 using a k nearest neighbors mapping approach applied to MODIS imagery. Canadian Journal of Forest Research, 48(1), 85–93.
10.1139/cjfr-2017-0184
Google Scholar
Bivand, R., Hauke, J., & Kossowski, T. (2013). Computing the Jacobian in Gaussian spatial autoregressive models: An illustrated comparison of available methods. Geographical Analysis, 45(2), 150–179.
10.1111/gean.12008
Web of Science®Google Scholar
Bivand, R. S., & Wong, D. W. S. (2018). Comparing implementations of global and local indicators of spatial association. TEST, 27(3), 716–748.
10.1007/s11749-018-0599-x
Web of Science®Google Scholar
Bond, C. F. (1956). Correlations between reproductive condition and skull characteristics of beaver. Journal of Mammalogy, 37(4), 506.
10.2307/1376644
Google Scholar
Bush, E., & Lemmen, D. S. (2019). Canada's changing climate report. Government of Canada.
10.4095/314614
Google Scholar
Cassola, F. (2016). Castor canadensis. The IUCN red list of threatened species 2016: E.T4003A22187946. https://doi.org/10.2305/IUCN.UK.2016-3.RLTS.T4003A22187946.en
10.2305/IUCN.UK.2016?3.RLTS.T4003A22187946.en
Google Scholar
Chevin, L.-M., Lande, R., & Mace, G. M. (2010). Adaptation, plasticity, and extinction in a changing environment: Towards a predictive theory. PLoS Biology, 8(4), e1000357.
10.1371/journal.pbio.1000357
CASPubMedWeb of Science®Google Scholar
Claude, J. (2008). Morphometrics with R. Springer Science & Business Media.
Google Scholar
Costa, D. P., Houser, D. S., & Crocker, D. E. (2013). Fundamentals of water relations and thermoregulation in animals. eLS. John Wiley & Sons, Ltd.
10.1002/9780470015902.a0003216.pub2
Google Scholar
Cox, P. G., & Baverstock, H. (2016). Masticatory muscle anatomy and feeding efficiency of the American beaver, Castor canadensis (Rodentia, Castoridae). Journal of Mammalian Evolution, 23(2), 191–200.
10.1007/s10914-015-9306-9
Web of Science®Google Scholar
Crandall, K., Olson, L. E., & Millien, V. (2023). Rocky rules: The idiosyncrasy of spatial and temporal size variation in mammals. Journal of Mammalogy, 104, 1216–1229.
10.1093/jmammal/gyad083
Web of Science®Google Scholar
Forest Products Laboratory. (2010). Wood handbook – wood as an engineering material. General technical report FPL-GTR-190. U.S. Department of Agriculture, Forest Service, Forest Products Laboratory.
Google Scholar
Freeman, P. W., & Lemen, C. A. (2008). A simple morphological predictor of bite force in rodents. Journal of Zoology, 275(4), 418–422.
10.1111/j.1469-7998.2008.00459.x
Web of Science®Google Scholar
Ginot, S., Herrel, A., Claude, J., & Hautier, L. (2018). Skull size and biomechanics are good estimators of in vivo bite force in murid rodents. The Anatomical Record, 301(2), 256–266.
10.1002/ar.23711
Google Scholar
Goldberg, C. S., Woodruff, K., Toldness, R., & Waits, L. P. (2011). Robust molecular sex identification of beaver (Castor canadensis) from non-destructive samples. Conservation Genetics Resources, 3, 729–731.
10.1007/s12686-011-9443-2
Web of Science®Google Scholar
Havens, R. P., Crawford, J. C., & Nelson, T. A. (2013). Survival, home range, and colony reproduction of beavers in east-central Illinois, an agricultural landscape. The American Midland Naturalist, 169(1), 17–29.
10.1674/0003-0031-169.1.17
Web of Science®Google Scholar
Hijmans, R. J., Van Etten, J., Cheng, J., Mattiuzzi, M., Sumner, M., Greenberg, J. A., Lamigueiro, O. P., Bevan, A., Racine, E. B., Shortridge, A., & Hijmans, M. R. (2023). Raster: Geographic data analysis and modeling. R package version 3.6–14.
Google Scholar
Hood, G. A., & Larson, D. G. (2015). Ecological engineering and aquatic connectivity: A new perspective from beaver-modified wetlands. Freshwater Biology, 60(1), 198–208.
10.1111/fwb.12487
Web of Science®Google Scholar
Jarema, S. I., Samson, J., McGill, B. J., & Humphries, M. M. (2009). Variation in abundance across a species' range predicts climate change responses in the range interior will exceed those at the edge: A case study with north American beaver. Global Change Biology, 15(2), 508–522.
10.1111/j.1365-2486.2008.01732.x
Web of Science®Google Scholar
Jenkins, S. H., & Busher, P. E. (1979). Castor canadensis. Mammalian species, no. 120. American Society of Mammalogists.
10.2307/3503787
Google Scholar
Johnston, C. A., & Naiman, R. J. (1990). Browse selection by beaver: Effects on riparian forest composition. Canadian Journal of Forest Research, 20(7), 1036–1043.
10.1139/x90-138
Web of Science®Google Scholar
Kang, Y., Su, J., Yao, B., Ji, W., Hegab, I. M., Hanafy, A. M., & Zhang, D. (2020). Geometric morphometric analysis of the plateau zokor (Eospalax baileyi) revealed significant effects of environmental factors on skull variations. Zoology, 140, 125779.
10.1016/j.zool.2020.125779
PubMedWeb of Science®Google Scholar
Korth, W. W. (1994). Specialized cranial and dental anatomy of rodents. In W. W. Korth (Ed.), The tertiary record of rodents in North America (pp. 7–21). New-York: Springer US.
10.1007/978-1-4899-1444-6_2
Google Scholar
Kubiak, B. B., Maestri, R., de Almeida, T. S., Borges, L. R., Galiano, D., Fornel, R., & De Freitas, T. R. (2018). Evolution in action: Soil hardness influences morphology in a subterranean rodent (Rodentia: Ctenomyidae). Biological Journal of the Linnean Society, 125(4), 766–776.
10.1093/biolinnean/bly144
Web of Science®Google Scholar
Maestri, R., Patterson, B. D., Fornel, R., Monteiro, L. R., & de Freitas, T. R. O. (2016). Diet, bite force and skull morphology in the generalist rodent morphotype. Journal of Evolutionary Biology, 29(11), 2191–2204.
10.1111/jeb.12937
CASPubMedWeb of Science®Google Scholar
Mahoney, M. J., & Stella, J. C. (2020). Stem size selectivity is stronger than species preferences for beaver, a central place forager. Forest Ecology and Management, 475, 118331.
10.1016/j.foreco.2020.118331
Web of Science®Google Scholar
Martínez, J. J., Millien, V., Simone, I., & Priotto, J. W. (2014). Ecological preference between generalist and specialist rodents: Spatial and environmental correlates of phenotypic variation. Biological Journal of the Linnean Society, 112(1), 180–203.
10.1111/bij.12268
Web of Science®Google Scholar
McGuire, J. L. (2010). Geometric morphometrics of vole (Microtus californicus) dentition as a new paleoclimate proxy: Shape change along geographic and climatic clines. Quaternary International, 212(2), 198–205.
10.1016/j.quaint.2009.09.004
Web of Science®Google Scholar
McMaster, R. T., & McMaster, N. D. (2001). Composition, structure, and dynamics of vegetation in fifteen beaver-impacted wetlands in Western Massachusetts. Rhodora, 103(915), 293–320.
Web of Science®Google Scholar
Mekonnen, Z. A., Riley, W. J., Berner, L. T., Bouskill, N. J., Torn, M. S., Iwahana, G., Breen, A. L., Myers-Smith, I. H., Criado, M. G., Liu, Y., & Euskirchen, E. S. (2021). Arctic tundra shrubification: A review of mechanisms and impacts on ecosystem carbon balance. Environmental Research Letters, 16(5), 53001.
10.1088/1748-9326/abf28b
CASWeb of Science®Google Scholar
Millien, V., Kathleen Lyons, S., Olson, L., Smith, F. A., Wilson, A. B., & Yom-Tov, Y. (2006). Ecotypic variation in the context of global climate change: Revisiting the rules. Ecology Letters, 9(7), 853–869.
10.1111/j.1461-0248.2006.00928.x
CASPubMedWeb of Science®Google Scholar
Millien, V., Ledevin, R., Boué, C., & Gonzalez, A. (2017). Rapid morphological divergence in two closely related and co-occurring species over the last 50 years. Evolutionary Ecology, 31(6), 847–864.
10.1007/s10682-017-9917-0
Web of Science®Google Scholar
Milligan, H. E., & Humphries, M. M. (2010). The importance of aquatic vegetation in beaver diets and the seasonal and habitat specificity of aquatic-terrestrial ecosystem linkages in a subarctic environment. Oikos, 119(12), 1877–1886.
10.1111/j.1600-0706.2010.18160.x
Web of Science®Google Scholar
Nummi, P., & Holopainen, S. (2014). Whole-community facilitation by beaver: Ecosystem engineer increases waterbird diversity. Aquatic Conservation, 24(5), 623–633.
10.1002/aqc.2437
Web of Science®Google Scholar
Oksanen, J., Simpson, G., Blanchet, F., Kindt, R., Legendre, P., Minchin, P., O'Hara, R., Solymos, P., Stevens, M., Szoecs, E., Wagner, H., Barbour, M., Bedward, M., Bolker, B., Borcard, D., Carvalho, G., Chirico, M., De Caceres, M., Durand, S., … Weedon, J. (2022). vegan: Community Ecology Package. R package version 2.6–4.
Google Scholar
Olsen, A. M., & Westneat, M. W. (2015). StereoMorph: An R package for the collection of 3D landmarks and curves using a stereo camera set-up. Methods in Ecology and Evolution, 6(3), 351–356.
10.1111/2041-210X.12326
Web of Science®Google Scholar
Pebesma, E. (2018). Simple features for R: Standardized support for spatial vector data. The R Journal, 10(1), 439–446.
10.32614/RJ-2018-009
Web of Science®Google Scholar
Pergams, O. R. W., & Lawler, J. J. (2009). Recent and widespread rapid morphological change in rodents. PLoS One, 4(7), e6452.
10.1371/journal.pone.0006452
PubMedWeb of Science®Google Scholar
R Core Team. (2022). R: A language and environment for statistical computing. R Foundation for Statistical Computing.
Google Scholar
Rossi, S., Cairo, E., Krause, C., & Deslauriers, A. (2015). Growth and basic wood properties of black spruce along an alti-latitudinal gradient in Quebec, Canada. Annals of Forest Science, 72(1), 77–87.
10.1007/s13595-014-0399-8
Web of Science®Google Scholar
Salandre, J. A., Beil, R., Loehr, J. A., & Sundell, J. (2017). Foraging decisions of north American beaver (Castor canadensis) are shaped by energy constraints and predation risk. Mammal Research, 62, 229–239.
10.1007/s13364-017-0312-6
Web of Science®Google Scholar
Samuels, J. X. (2009). Cranial morphology and dietary habits of rodents. Zoological Journal of the Linnean Society, 156(4), 864–888.
10.1111/j.1096-3642.2009.00502.x
Web of Science®Google Scholar
Souto-Lima, R. B., & Millien, V. (2014). The influence of environmental factors on the morphology of red-backed voles Myodes gapperi (Rodentia, Arvicolinae) in Québec and western Labrador. Biological Journal of the Linnean Society, 112(1), 204–218.
10.1111/bij.12263
Web of Science®Google Scholar
Stefen, C. (2009). Intraspecific variability of beaver teeth (Castoridae: Rodentia). Zoological Journal of the Linnean Society, 155(4), 926–936.
10.1111/j.1096-3642.2008.00467.x
Web of Science®Google Scholar
Stumpp, R., Fuzessy, L., & Paglia, A. P. (2018). Environment drivers acting on rodent rapid morphological change. Journal of Mammalian Evolution, 25(1), 131–140.
10.1007/s10914-016-9369-2
Web of Science®Google Scholar
Tape, K. D., Clark, J. A., Jones, B. M., Kantner, S., Gaglioti, B. V., Grosse, G., & Nitze, I. (2022). Expanding beaver pond distribution in Arctic Alaska, 1949 to 2019. Scientific Reports, 12(1), 7123.
10.1038/s41598-022-09330-6
CASPubMedWeb of Science®Google Scholar
Tape, K. D., Jones, B. M., Arp, C. D., Nitze, I., & Grosse, G. (2018). Tundra be dammed: Beaver colonization of the Arctic. Global Change Biology, 24(10), 4478–4488.
10.1111/gcb.14332
PubMedWeb of Science®Google Scholar
Teplitsky, C., & Millien, V. (2013). Climate warming and Bergmann's rule through time: Is there any evidence? Evolutionary Applications, 7(1), 156–168.
10.1111/eva.12129
PubMedWeb of Science®Google Scholar
Wolf, M., Friggens, M., & Salazar-Bravo, J. (2009). Does weather shape rodents? Climate related changes in morphology of two heteromyid species. Naturwissenschaften, 96(1), 93–101.
10.1007/s00114-008-0456-y
CASPubMedGoogle Scholar

Volume322, Issue3

March 2024

Pages 251-260

Chipping in: functional morphology of the American beaver under range expansion

Abstract

Introduction