Recovery and genetics of Mexican wolves: a reply to Hedrick et al.

Abstract

Mexican wolves (Canis lupus baileyi) are a gray wolf subspecies found in northern Mexico and part of the southwestern United States. The subspecies was once extirpated in the wild and the current population is highly inbred, having descended from just 7 founders (U.S. Fish and Wildlife Service [USFWS] 2017). Accordingly, the subspecies is federally listed as endangered, and genetic monitoring and management are key components of the current recovery strategy and downlisting recovery criteria (USFWS 2017). In support of this recovery strategy, we recently evaluated inbreeding depression in wild Mexican wolves in Arizona and New Mexico, USA, from 1998 to 2022 (Clement et al. 2024). Our analysis did not detect any statistical association between inbreeding coefficients in a pack, as estimated from the Mexican wolf pedigree, and the number of pups surviving to 9 months (hereafter, recruitment). Hedrick et al. (2025) provided comments on our work, concluding that Mexican wolves “have shown inbreeding depression” and recommending that Mexican wolves hybridize with northern gray wolves (C. l. occidentalis) to induce genetic rescue. Here, we provide our perspective on inbreeding depression, genetic rescue, and other issues raised by Hedrick et al. (2025).

Hedrick et al. (2025) write rather definitively that Mexican wolves “have shown inbreeding depression.” This statement seems to be based on a combination of a prior expectation that low genetic diversity invariably leads to inbreeding depression and some older studies of Mexican wolves, but the current evidence does not support their claim with regard to recruitment in wild Mexican wolves.

Mexican wolves are descended from 7 founders, so they are highly inbred, which increases the risk of genetic problems. However, many endangered species have experienced genetic bottlenecks, yet some species, including some wolves, have not experienced inbreeding depression (Crnokrak and Roff 1999, Kalinowski et al. 1999, Brzeski et al. 2014, Gooley et al. 2017, Powell et al. 2023) and some have experienced robust post-bottleneck population growth (Harding et al. 2016). Because inbreeding coefficients do not consider the relative occurrence of deleterious alleles in the founders or standing population, inbreeding alone does not establish inbreeding depression. Hedrick et al. (2025) also point to syndactyly as evidence of inbreeding depression; however, they did not support this assertion with any evidence or citations.

While some early studies reported inbreeding depression in Mexican wolves, our recent paper represents the most current and best available science for wild wolves. For example, Fredrickson and Hedrick (2002) reported inbreeding depression in the body sizes of captive wolves, but this is a relatively indirect measure of fitness, compared to our analysis of recruitment in wild wolves. In addition to analyses of captive wolves, Fredrickson et al. (2007) reported inbreeding depression in the maximum number of wild Mexican wolf pups observed during summer. However, our recent analysis benefited from a much larger sample size (352 pack-years vs. 39 pack-years), more timely litter size counts (within 6 weeks vs. within several months), and a more comprehensive fitness metric (combining litter production, litter size, and survival vs. summer pup counts). This is not a criticism of Fredrickson et al. (2007) because they used the best data available at that time, but our recent analysis produced more reliable inferences for recovery actions because of the advantages listed above.

Hedrick et al. (2025) recommend hybridizing with northern gray wolves as a remedy for inbreeding in Mexican wolves. However, hybridization would threaten the integrity of Mexican wolves, bringing clear risks and uncertain benefits, all at a time when the available evidence indicates genetic rescue is not necessary. In recommending hybridization, Hedrick et al. (2025), do not address several important considerations, including the legal status of Mexican wolves as a distinct taxonomic entity (USFWS 2015), the risk of genetic swamping (Edmands 1999, Gippoliti et al. 2018, Odell et al. 2018), and recent progress towards recovery goals (USFWS 2024).

Mexican wolves evolved a distinct genetic and phenotypic profile in Mexico and adjacent portions of the American Southwest, and they fill a unique ecological niche in their native habitats. Accordingly, the USFWS has designated Mexican wolves as a separately listed endangered subspecies (USFWS 2015). Hybridizing with large, Canadian northern gray wolves, as proposed by Hedrick et al. (2025), would compromise the Mexican wolf as a distinct subspecies. Given the smaller body size, higher inbreeding, smaller pack size, and lower abundance of Mexican wolves, hybridization would create a risk of genetic swamping, in which unique portions of the Mexican wolf genome are lost (Odell et al. 2018). Even if hybridization initially induces heterosis, in the long term, introgression of traits that evolved in Canada could lead to outbreeding depression, reducing the fitness of hybridized Mexican wolves in some southwestern habitats (Edmands 1999, Edmands and Timmerman 2003, Gippoliti et al. 2018).

Hedrick et al. (2025) argue that hybridization is necessary because Mexican wolves “have shown inbreeding depression.” As discussed above, their claim rests on older, smaller studies (Fredrickson and Hedrick 2002, Fredrickson et al. 2007), while dismissing a larger and more current study that did not detect inbreeding depression (Clement et al. 2024). Additionally, they do not address the fact that the wild population is currently thriving, which is not consistent with a population in need of genetic rescue. For example, the wild Mexican wolf population has averaged 14% annual growth since 2010, yielding a population that exceeded interim recovery goals by 67% just 5 years into the current recovery plan (USFWS 2024). In addition, gene retention and mean inbreeding in wild Mexican wolves have been better than interim recovery targets (USFWS 2024), indicating that current management is maintaining genetic diversity without hybridization. Hybridization with northern gray wolves given no demonstrable need would be an irreversible ecological experiment that could harm recovery of Mexican wolves.

Hedrick et al. (2025) critiqued our analysis of artificial food caches (which they call “supplemental feeding”). Food caching is a practice in which some denning wolves are provided food during some or all of the pup-rearing season (April to September). Food caches may be provided to wolves near ranches or residential areas to reduce conflicts with people, or to packs with foster pups with the intent of increasing pup survival. Inbreeding levels are not used to decide which packs receive food caches and packs with foster pups were excluded from our analysis. Hedrick et al. (2025) argue for estimating an interaction between inbreeding and food caches on the grounds that food caches may mask inbreeding depression. We fit a simpler additive model (which indicated that food caches increased recruitment) because estimating an interaction for an attenuating effect (i.e., food caches reduce the impact of inbreeding) requires large sample sizes (Sommet et al. 2023). Even without estimating an interaction, we can assess the argument by Hedrick et al. (2025) logically. The overall association between inbreeding and recruitment was not significantly different from zero (Clement et al. 2024). If, as Hedrick et al. (2025) hypothesize, recruitment in unfed packs was negatively associated with inbreeding, then recruitment in fed packs (which constituted 46% of packs) must have a positive association with inbreeding (to maintain the overall slope close to zero). But a positive association between inbreeding and recruitment is biologically implausible and not consistent with a hypothesis of masking. In addition, since 2019 the percentage of packs receiving food has been substantially reduced from that projected in the 2017 Recovery Plan, yet the population has surpassed interim abundance and genetic targets (USFWS 2024). There is no indication that feeding is masking inbreeding depression.

Hedrick et al. (2025) also critiqued our handling of packs that did not produce litters, writing “Clement et al. (2024) only considered inbreeding in litters when there was one or more offspring.” Their statement is incorrect. We estimated the temporal trend in mean inbreeding coefficient only for successful litters (≥1 pup) because non-existent litters are not part of the population. However, our fitness metric, pup recruitment, included all packs (that were not translocated or given foster pups), including those that failed to produce a litter. For example, we wrote that “we also examined how inbreeding might affect components of recruitment [including] the probability of producing a litter,” and “we estimated the probability of producing a litter using logistic regression.” We also provide estimates in Table 3 that show non-significant associations between inbreeding in pups, dams, and sires and the probability of producing a litter (Clement et al. 2024).

We appreciate the interest of Hedrick et al. (2025) in our paper and the larger issue of Mexican wolf management and recovery. Continued management and monitoring of inbreeding, and its relationship to fitness, remain priorities for Mexican wolf recovery. Any decisions about Mexican wolf recovery and management should be guided by the best available science, which currently includes our analysis of inbreeding depression (Clement et al. 2024).

The authors declare no conflicts of interest.