Soft tissue morphology of the vomeronasal organ in Lontra canadensis and its osteological correlate: Implications for the evolution of the caniform accessory olfactory system

vomeronasal; secondarily aquatic; osteological correlate; lutrinae 22 23 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted January 22, 2026. ; https://doi.org/10.64898/2026.01.19.700381doi: bioRxiv preprint AUTHOR CONTRIBUTIONS 24 Stephanie Palmer: Conceptualization; investigation; writing - original draft; writing – review 25 and editing; methodology; validation; software; visualization; data curation; resources; project 26 administration. William Foster: Conceptualization; writing - review and editing; visualization. 27 Grace Capshaw: writing - review and editing; data curation; resources; visualization. Margot 28 Michaud: writing - review and editing; data curation; resources; visualization. Siobhán Cooke; 29 Supervision; resources; writing - review and editing; project administration. 30 31

We thank Drs. Gabriel Bever, Alistair Evans, and James Rule for 32 reviewing and discussing early editions of this manuscript as part of S. M. Palmer’s dissertation 33 completed at Johns Hopkins University. We also acknowledge the Johns Hopkins University 34

Characterization and Processing (MCP) facility for access to equipment and technical 35 support. We thank both Arianna Harrington (Southern Utah University) and Sandrine Ladevèze 36 (MNHN Paris) for scanning and contributing specimens used in this study, as well as the 37 numerous contributors to MorphoSource for access to other specimens (see Supplemental Table 38 1 for detailed MorphoSource acknowledgements). 39 40 Data availability statement: Contact the corresponding author for availability of data generated 41 that support the findings of this study. 42 43

The olfactory system plays a critical role in mammalian environmental perception, 44 with some clades relying on an expanded accessory olfactory (vomeronasal) system (VNS) to 45 detect larger, non-volatile odorants. Mammals make extensive use of this system for social 46 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted January 22, 2026. ; https://doi.org/10.64898/2026.01.19.700381doi: bioRxiv preprint communication between conspecifics. Recent studies have begun to investigate how the VNS 47 changes in response to or as part of ecological transitions. Several studies have identified trends 48 of VNS-associated gene loss or regression in secondarily aquatic mammals. However, 49 continuing discussion on genotype-phenotype correlation within the VNS means that greater 50 effort should be made to investigate the morphology of the VNS in species where it remains 51 poorly understood. Here, we use skeletal and soft-tissue data to demonstrate that the 52 vomeronasal groove, an established osteological correlate for the VNO in bats and primates, is 53 also a valid indicator for its presence in Caniformia. Additionally, we confirm the presence of the 54 VNO in the secondarily aquatic North American river otter (Lontra canadensis) and compare its 55 morphology with that of two close-related species, the semi-aquatic American mink (Neogale 56 vison) and the terrestrial long-tailed weasel (Neogale frenata). This study expands the valid 57 taxonomic scope of the vomeronasal groove’s proxy as an osteological correlate, confirms the 58 presence of the VNO in the previously undescribed system of the North American river otter, 59 and highlights the complexity of the mammalian accessory olfactory system. 60 61

– Background 62 Vertebrate sensation is supported by a suite of specialized organs that allow organisms to 63 perceive and process environmental stimuli. A key component of this sensory repertoire is the 64 olfactory system, which detects odor molecules in the environment and communicates 65 information to the forebrain for processing. Olfaction is functionally relevant to numerous 66 fundamental vertebrate behaviors, including navigation (Poo et al., 2022), prey tracking (Hughes 67 et al., 2010), reproduction (Aron, 1979; Baum & Kelliher, 2009; Baum & Cherry, 2015). The 68 mammalian olfactory system is highly variable across species, reflecting anatomical and 69 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted January 22, 2026. ; https://doi.org/10.64898/2026.01.19.700381doi: bioRxiv preprint functional constraints (e.g., limitations related to skull morphology and/or respiratory function) 70 as well as adaptation to diverse ecological niches. Understanding the evolutionary response (or 71 contribution) of crucial sensory systems during major ecological transitions is a major goal of 72 comparative biology and demands thorough sampling of phylogenetically relevant phenotypes, 73 especially in poorly known taxa. 74 In mammals (and tetrapods generally), olfaction is functionally and anatomically 75 segregated into the main olfactory system (MOS) and the vomeronasal system (VNS), also 76 referred to as the accessory olfactory system (AOS) (Figure 1) (but see Baum, 2012 and Baum & 77 Kelliher, 2009 for potential for functional overlap between the two systems). The special sensory 78 function of the MOS is supported by the olfactory epithelia of the nasal cavity, which contain 79 olfactory sensory neurons whose axons form the first cranial nerve (CN I) and are connected 80 directly to the olfactory lobes of the brain. The anatomy of the VNS, in contrast, is much more 81 complex. 82 The VNS is supported by specialized epithelia that are organized into a pair of 83 parasagittal organs located lateral to the base of the nasal septum on the maxilla: the 84 vomeronasal organs (or Jacobson’s organs) (VNO) (Salazar et al., 2003; Garrett, 2015; Smith et 85 al., 2024). The VNO is supported by a cartilaginous, sometimes ossified, capsule, referred to as 86 the vomeronasal cartilage (VNC), which protects and provides structural support to the organ 87 (Salazar et al., 1995). The VNO capsule communicates anteriorly with the incisive duct (ID), 88 which travels through the incisive canal and opens into the oral cavity (Figure 1; Ortiz-Leal et 89 al., 2024). The VNO is innervated by vomeronasal nerves (VNN) that originate from 90 functionally distinct receptor cells from the MOS. The VNN connects the VNO to a specialized 91 region of the olfactory lobes, the accessory olfactory bulbs (AOB). Whilst the MOS is 92 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted January 22, 2026. ; https://doi.org/10.64898/2026.01.19.700381doi: bioRxiv preprint responsible for processing small, airborne, and molecularly simple odorants, the VNS has been 93 linked to the processing of larger, non-volatile odorants, such as social chemosignals 94 (“pheromones”) (Garrett, 2015), and is thus hypothesized to be critical for communication 95 between conspecifics, particularly concerning individual recognition and sexual reproduction 96 (Zufall et al., 2002; Keverne, 2004; Tirindelli, 2021). 97 Evidence for reductions of the MOS and AOS-related homologues in secondarily aquatic 98 lineages often support the hypothesis that land-to-water transitions universally result in reduced 99 olfactory acuity (Yu et al., 2010; Berta et al., 2014; Lu et al., 2016; Paulina-Carabajal et al., 100 2015; Zhang & Nikaido, 2020) and may be compensated for by an expansion of other sensory 101 systems, e.g., the “visual-priority hypothesis” (Debey & Pyenson, 2013; Garrett, 2015; Hadden 102 & Zhang, 2023). One of the most studied aspects of the AOS is the underlying genetic 103 components, in which a reduction in AOS-related genes, including VNO receptor genes 104 (ancV1R; TRPC2), has been identified across multiple lineages of secondarily aquatic mammals 105 (e.g., Lutrinae, Pinnipedia, Sirenia, Cetacea; Yu et al., 2010; Hecker et al., 2019; Liu et al., 2019; 106 Zhang & Nikaido, 2020) and bats (Zhao et al., 2011). 107 Other anatomical and genetic transitions in the MOS and VNS in secondarily aquatic 108 mammals have also been identified (Mackey-Sim et al., 1985; Pihlström et al., 2008; Berta et al., 109 2014; De Vreese, 2023, Farnkopt et al., 2025), including several within secondarily aquatic 110 Carnivora (Switzer et al., 1980; Zhang & Nikaido, 2020 Kondoh et al., 2024; 2025). Pinnipeds 111 (seals, sea lions and fur seals, and walruses) have derived morphologies in the VNS (Switzer et 112 al., 1980; Kondoh et al., 2024; 2025), mirroring a reduction in VNO receptor genes (Yu et al., 113 2010; Zhang & Nikaido, 2020). Studies of VNS anatomy in secondarily aquatic carnivorans 114 (Table 1) have revealed the absence of an AOB in the true seal species Phoca vitulina (Switzer et 115 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted January 22, 2026. ; https://doi.org/10.64898/2026.01.19.700381doi: bioRxiv preprint al., 1980) along with the modification of its VNO from a sensory to a secretory structure 116 (Kondoh et al., 2024; Figure 2B). Meanwhile, members of Otariidae, including Callorhinus 117 ursinus, Eumatropias jubatus, and Zalophus californianus, retain an AOB (Switzer et al., 1980), 118 with Steller sea lions (E. jubatus) retaining a functional VNO (Kondoh et al., 2025; Figure 2C). 119 The Steller sea lion VNO is derived from the ancestral condition through its migration from the 120 ventral surface of the nasal cavity to within the incisive canal (Kondoh et al., 2025). 121 The other secondarily aquatic caniforms, otters (Lutrinae), possess reduced VNO receptor 122 gene repertoires similar to the condition observed in phocid pinnipeds - specifically sea otters 123 (Enhydra lutris), giant river otters (Pteronura brasiliensis), and the Eurasian otter (Lutra lutra) 124 (Yu et al., 2010; Hecker et al., 2019; Zhang & Nikaido, 2020). Some authors hypothesize that it 125 is likely that lutrines do not possess a VNS or VNO, similar to phocid pinnipeds (Zhang & 126 Nikaido, 2020); however, no information has been presented on whether their anatomical 127 components converge on a similar derived morphological condition (Yu et al., 2010; Hecker et 128 al., 2019; Zhang & Nikaido, 2020; Zellmer et al., 2021). While some studies have found a strong 129 correlation between genetics and morphology within the MOS and VNS (Garrett & Steiper, 130 2014; Bird et al., 2018; Farnkopt et al., 2025), other studies show that phenotypes exhibit 131 stability despite reductions in receptor gene repertoires (Niimura & Nei, 2007; Yohe et al., 2020; 132 Yohe & Krell, 2023). As a result, conclusions on the strength and nature of correlations between 133 olfactory genes and phenotype vary across studies (Niimura & Nei, 2007; Salazar & Sánchez-134 Quinteiro, 2009; Yohe et al., 2018; Yohe et al., 2020; Yohe & Krell, 2023), and the nature of 135 phenotype-genotype correlation is often system- and scale-dependent (Garrett & Steiper, 2014; 136 Yohe et al., 2020). The emerging complexity of the correlation between genotype and phenotype 137 in this system is not only intellectually motivating in its own right but also touches on broader 138 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted January 22, 2026. ; https://doi.org/10.64898/2026.01.19.700381doi: bioRxiv preprint themes of phenogenetic drift (Weiss & Fullerton, 2000) and mosaicism (Caianiello, 2024). 139 Ultimately, studying VNS morphology in species for which knowledge gaps exist is critical to 140 understanding the relationship between phenotype and genotype in the mammalian olfactory 141 system. 142 To further evaluate the phenotype of the Lutrine VNS, we utilize both soft-tissue and 143 osteological data. To examine soft-tissue, we used diffusible iodine contrast-enhanced micro-144 computed tomography (diceCT) scans. DiceCT is a method that uses contrast agents (Lugol's 145 iodine and/or Strong solution) to visualize soft tissues in situ, with adjacent and underlying hard 146 tissues (e.g., bone), in CT data (Gignac et al., 2016). Its application in visualizing the soft tissues 147 of the VNO has been validated in recent publications, with authors noting its potential for 148 identifying small structures such as the VNO (Yohe et al., 2018; Smith et al., 2021; Smith et al., 149 2024; Collin et al., 2024; Leng & Shi, 2025). Our diceCT dataset capturing soft-tissue variation 150 of the VNO around Lutrinae consisted of three mustelid species: American mink (Neogale 151 vison), least-tailed weasel (Neogale frenata), and North American river otter (Lontra 152 canadensis). To contextualize the morphology observed, we also consulted the literature to 153 define common features of the VNO across caniforms (Table 1). While it is important to mention 154 that genes associated with the VNS have not yet been explicitly characterized in the North 155 American river otter, phylogenetic bracketing (Witmer & Thomason, 1995) supports the 156 inference of similar genetic reductions based on its relationship to the studied taxa Enhydra 157 lutris, Pteronura brasiliensis, and Lutra lutra (Waku et al., 2016; de Ferran, 2022). 158 In addition to soft-tissue morphology from diceCT, other methods assess the presence 159 and morphology of the VNO by investigating its osteological correlate, the vomeronasal groove 160 (VNG). Morphologically, the VNG is described as a bilateral, rostral-caudally elongate, and 161 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted January 22, 2026. ; https://doi.org/10.64898/2026.01.19.700381doi: bioRxiv preprint medial-laterally narrow trough or indentation located on the nasal septum, with a raised lateral 162 edge and an impressed center (Smith et al., 2011; Garrett et al., 2013; Garrett, 2015; Smith et al., 163 2024). Notably, previous work has established the presence of such structures in both extant and 164 fossil mammals (Hillenius, 2000; Crompton et al., 2017; Bendel et al., 2018), as well as in 165 detailed comparative studies establishing it as a formal correlate in primates (Smith et al., 2011; 166 Garrett et al., 2013; Garrett, 2015) and bats (Smith et al., 2024). Both the presence and 167 dimensions of the VNG have been shown to correlate with the presence or size of the VNO 168 (Garrett, 2015), revealing evolutionary patterns of changes to the AOS, including diverse 169 trajectories of degeneration seen in bats (Smith et al., 2024). 170 To date, no study has used VNG as an indicator of VNO presence in carnivorans, 171 particularly in the application to understudied secondarily aquatic taxa such as Lutrinae. In 172 reviewing existing histological and anatomical studies on the VNO morphology in canids (e.g., 173 Vulpes vulpes, Canis lupus, and [mesocephalic] Canis familiaris), ursids (Ursus arctos), and 174 mustelids (Neogale vison, Mustela furo) [Salazar et al., 1995; Weiler et al., 1999; Kelliher et al., 175 2001; Tomiyasu et al., 2017; Mahdy & Mohamed, 2019; Ortiz-Leal et al., 2020; Dzięcioł et al., 176 2020; Ortiz-Leal et al., 2022; Ortiz-Leal et al., 2024; Sanmartín-Vázquez et al., 2024; Table 1]), 177 a groove-like indentation in the ventral nasal cavity, on either side of the nasal septum or vomer, 178 was present and in close association with the VNO and VNC, suggesting it is homologous to the 179 VNG described in primates and bats (Smith et al., 2011; Garrett et al., 2013; Garrett, 2015; 180 Smith et al., 2024). This evidence from the literature suggests that the VNG is a strong candidate 181 as an osteological correlate for the functional VNO in caniforms, expanding its utility for 182 estimation of VNO presence in other otter taxa (e.g., Lutra, Aonyx, Lutragale, Enhydra) not 183 represented by diceCT data. 184 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted January 22, 2026. ; https://doi.org/10.64898/2026.01.19.700381doi: bioRxiv preprint 185 186 Figure 1. Drawings of the gross anatomy in parasagittal view (A), sub-gross anterolateral view 187 (B), and coronal cross-sectional view (C) of the histo-morphological anatomy of the main 188 olfactory system (MOS), accessory olfactory system (AOS), and vomeronasal organ (VNO). 189 A: the main olfactory bulb and olfactory epithelia represent the MOS (brown-orange), while the 190 vomeronasal organ (VNO) and accessory olfactory bulb represent the AOS (blue-green). B: 191 relationship of the VNO, vomeronasal cartilage (VNC), and incisive (nasopalatine) duct in a sub-192 gross anterior-lateral orientation, with directional axes indicated (dorsal–ventral, rostral–caudal). 193 C: coronal cross-section of the VNO, highlighting the vomeronasal sensory epithelium, 194 respiratory epithelium, vomeronasal nerves, and vomeronasal venous sinus with directional axes 195 indicated (dorsal-ventral). Modified after Salazar and Sánchez-Quinteiro (2009), Evans (1993), 196 and Estes (1978). 197 198 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted January 22, 2026. ; https://doi.org/10.64898/2026.01.19.700381doi: bioRxiv preprint Table 1. Major relevant comparative features of AOS structures across Caniformia, including 199 pinnipeds. VNC = vomeronasal cartilage; VNO = vomeronasal organ; VND = vomeronasal duct; 200 ID = incisive duct; AOB = accessory olfactory bulb. 201 Ecology Taxa Features Source(s) Terrestrial (non- aquatic) Red fox (Vulpes vulpes) VNC is J-shaped “Kidney-like” shaped lumen Numerous and dense vascular network VNO consisting of VND and “vomeronasal capsule” ID inferior and lateral to VNO and VND Ortiz‐Leal, I., Torres, M. V., Villamayor, P. R., López‐Beceiro, A., & Sanchez‐Quinteiro, P. (2020). The vomeronasal organ of wild canids: the fox ( Vulpes vulpes) as a model. Journal of Anatomy, 237(5), 890-906. Wolf (Canis lupus) VNC is U-shaped, and J-shaped in some areas VNC is more open dorsally in the anterior third of VNO/VND Vessels are prominent in the dorsal and lateral regions Ortiz‐Leal, I., Torres, M. V., Barreiro‐Vázquez, J. D., López‐Beceiro, A., Fidalgo, L., Shin, T., & Sanchez‐Quinteiro, P. (2024). The vomeronasal system of the wolf (Canis lupus signatus): The singularities of a wild canid. Journal of Anatomy, 245(1), 109-136. Mesocephalic/mesaticephalic domestic dogs (Canis familiaris) VNC is associated with ID anteriorly AOB is only identifiable by microdissection Spatial relationship between ID and VND/VNC complex VNC described as both U-shaped and J- shaped ID & VNO join immediately behind the upper incisors VNO vascular is more numerous laterally Ortiz‐Leal, I., Torres, M. V., Villamayor, P. R., López‐Beceiro, A., & Sanchez‐Quinteiro, P. (2020). The vomeronasal organ of wild canids: the fox ( Vulpes vulpes) as a model. Journal of Anatomy, 237(5), 890-906. Dzięcioł, M., Podgórski, P., Stańczyk, E., Szumny, A., Woszczyło, M., Pieczewska, B., ... & Wrzosek, M. A. (2020). MRI features of the vomeronasal organ in dogs (Canis familiaris). Frontiers in veterinary science, 7, 483162. Allouch, G. M., & Alshanbari, F. A. (2024). Comparative Anatomy of the Vomeronasal Organ (VNO) in Sheep (Ovis aries) and Dogs (Canis familiaris) with Simple Reference to its Histological Structure and Vasculature Supply. International Journal of Morphology, 42(2). Salazar, I., Cifuentes, J. M., & Sánchez‐Quinteiro, P. (2013). Morphological and immunohistochemical features of the vomeronasal system in dogs. The Anatomical Record: Advances in Integrative Anatomy and Evolutionary Biology , 296(1), 146-155. (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted January 22, 2026. ; https://doi.org/10.64898/2026.01.19.700381doi: bioRxiv preprint Ferret (Mustela furo) More ‘tubular’ lumen rather than ‘crescent-shaped’ Thin epithelia Little to no vasculature in VNO sensory epithelia Weiler, E., Apfelbach, R., & Farbman, A. I. (1999). The vomeronasal organ of the male ferret. Chemical senses, 24(2), 127-136. Kelliher, K. R., Baum, M. J., & Meredith, M. (2001). The ferret's vomeronasal organ and accessory olfactory bulb: effect of hormone manipulation in adult males and females. The Anatomical Record: An Official Publication of the American Association of Anatomists, 263(3), 280-288. Brown bear (Ursus arctos) Lumen differently shaped throughout the VNO J-shaped VNC Veins visible in histological section Tomiyasu, J., Matsumoto, N., Katsushima, H., Nishijima, A., Hagino, K., Sakamoto, H., & Yanagawa, Y. (2025). Association Between Back Scent Gland Development and Reproductive Status in Male Brown Bears (Ursus arctos). Journal of Experimental Zoology Part A: Ecological and Integrative Physiology, 343(5), 629-635. Aquatic Harbor seal (Phoca vitulina) No evidence of lumen/luminal structures Present, but degenerated, VNC Patent communication between oral cavity and nasal cavity No AOB Kondoh, D., Tonomori, W., Iwasaki, R., Tomiyasu, J., Kaneoya, Y., Kawai, Y. K., ... & Kobayashi, M. (2024). The vomeronasal organ and incisive duct of harbor seals are modified to secrete acidic mucus into the nasal cavity. Scientific Reports, 14(1), 11779. Switzer, III, R. C., Johnson, J. I., & Kirsch, J. A. (1980). Phylogeny through brain traits: relation of lateral olfactory tract fibers to the accessory olfactory formation as a palimpsest of mammalian descent. Brain, Behavior and Evolution, 17(5), 339-363. Fur seals & Sea lions (Callorhinus ursinus, Eumetopias jubatus, and Zalophus californianus) Present AOB Present VNO (only established in Eumatopias jubatus) VNO located within incisive canal, not nasal cavity (only established in Eumatopias jubatus) Switzer, III, R. C., Johnson, J. I., & Kirsch, J. A. (1980). Phylogeny through brain traits: relation of lateral olfactory tract fibers to the accessory olfactory formation as a palimpsest of mammalian descent. Brain, Behavior and Evolution, 17(5), 339-363. Kondoh, D., Tonomori, W., Iwasaki, R., Tomiyasu, J., Kaneoya, Y., Li, H., ... & Kobayashi, M. (2025). The vomeronasal system of the Steller sea lion. Journal of Anatomy. 202 203 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted January 22, 2026. ; https://doi.org/10.64898/2026.01.19.700381doi: bioRxiv preprint 204 Figure 2. Comparative parasagittal illustrations showing shows the spatial relationships between 205 the incisive duct or canal (orange), vomeronasal organ (cyan), vomeronasal cartilage (blue), and 206 vomeronasal glandular tissue (magenta) within the nasal and oral cavities. Cranial bones colored 207 in yellow. Panel A: a terrestrial carnivoran (Canis familiaris), Panel B: a phocid (Harbor seal, 208 Phoca vitulina), and Panel C: an otariid (Steller sea lion, Eumatopias jubatus). Directional axes 209 (rostral-caudal and dorsal-ventral) are indicated in panel A. Adapted from Kondoh et al., 2024 210 and 2025. Silhouettes from PhyloPic (all dedicated to the public domain): Phoca vitulina, 211 Eumatopias jabatus by Andy Wilson; Canis familiaris by Tracy Heath. 212

213

Dry Skulls 214 This study is based on a comparative sample across Caniformia, selected to capture 215 taxonomic and ecological diversity within the clade, as well as size variation. CT scans were 216 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted January 22, 2026. ; https://doi.org/10.64898/2026.01.19.700381doi: bioRxiv preprint sourced from MorphoSource and from scans taken by Arianna Harrington and Matthew Colbert 217 using CT scanners at the Duke Shared Materials Instrumentation Facility (SMiF) (Duke 218 University) and the University of Texas High-Resolution X-ray Computed Tomography Facility. 219 Specimens were sourced from the National Museum of Natural History (MNHN), Smithsonian 220 Museum of Natural History (USNM), Laboratory of Santiago Palazón (LSP), Yale Peabody 221 Museum (YPM), Illinois State Museum (ISM), University of Florida Museum of Natural History 222 (UF), Duke Evolutionary Anthropology Department (DU), Royal Belgian Institute of Natural 223 Sciences (RBINS), The Pennsylvania State University Department of Anthropology (PSU), 224 North Carolina Museum of Natural Sciences (NCSM), University of Alaska Museum (UAM), 225 and the American Museum of Natural History (AMNH) (Supplemental Tables 1 and 2). This 226 dataset includes specimens representing 45 species across fissiped caniforms. Most species were 227 represented by a single specimen. The nasal and oral cavities of these CT-scanned specimens 228 were examined using 3DSlicer (Federov et al., 2012) to identify evidence of VNGs in coronal 229 and 3D-rendered views. CT scans were loaded into 3DSlicer using the ImageStacks module of 230 the SlicerMorph package (Rolfe et al., 2021). Some scans were imported at reduced resolution or 231 with the "skip slices" (1-2 slices) option to enhance loading performance. Figures were generated 232 using the Capture feature in 3DSlicer, Microsoft PowerPoint, and Adobe Photoshop. Standard 233 adjustments to contrast and brightness were made in these applications to enhance visibility for 234 the identification of structures during analysis and figure preparation. 235

Heads & Soft Tissue 236 All procedures for acquiring soft-tissue specimens and associated data adhered to ethical 237 standards. The river otter (Lontra canadensis) and least weasel (Neogale frenata) diceCT scans 238 were downloaded from MorphoSource (Supplemental Information Table 2). The American mink 239 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted January 22, 2026. ; https://doi.org/10.64898/2026.01.19.700381doi: bioRxiv preprint (Neogale vison) was sourced from Carolina Biological Supply as a whole-body, skinned 240 specimen preserved in Carolina Perfect solution. Decapitation was achieved via atlantooccipital 241 dislocation prior to CT scanning. Scans were generated using a microCT RX Solutions EasyTom 242 scanner at the Johns Hopkins University Materials Characterization and Processing Center. The 243 specimen was scanned at 48.5 µm voxel resolution before staining and re-scanned at 16.5 µm 244 voxel resolution after 41 days in 3.75% buffered Lugol’s iodine solution (Gignac et al., 2016; 245 Dawood et al., 2021). Scans were viewed in 3DSlicer, where structures were identified in 246 coronal, sagittal, axial, and 3-dimensional views. 247 248

249 Osteological Anatomy of the VNG - CT and Gross Morphology (Dry Skulls) 250 All species or specimens sampled exhibited evidence of a VNG. These structures were 251 bilateral and consisted of rostro-caudally elongated and medio-laterally narrow grooves with 252 indented centers and raised edges. VNGs were best visualized in coronal CT slices and 3D 253 renderings of segmentations (Figures 3 and 4). The visibility of the VNG in 3D renderings varied 254 depending on its position relative to surrounding anatomical structures, including the vomer 255 (Figure 3). In some cases, the VNG could be visualized from a dorsal-anterior angle into the 256 nasal cavity, although the vomer sometimes obstructed a direct view. When the VNG was not 257 externally visible from dorsal-anterior views into the ventral nasal cavity floor, the nasal bones, 258 frontal, or vomer were digitally cropped in an axial plane above the ventral floor of the nasal 259 cavity, and/or the specimen was laterally rotated to visualize it. In addition to aligning with 260 descriptions of the VNG in other mammals, remnants of VNC were visible in one specimen 261 (Mustela nivalis, NCSM 8102; Figure 5). 262 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted January 22, 2026. ; https://doi.org/10.64898/2026.01.19.700381doi: bioRxiv preprint The morphology of the grooves was variable and diverse across species, with most 263 variation occurring in the groove’s cranial-caudal length, where some were longer and narrower 264 (Figure 3G, E) and others were shorter and wider (Figure 3D, I). Longer VNGs tended to be 265 mediolaterally narrower, while shorter or more abbreviated VNGs tended to be proportionally 266 wider (Figure 3). The caudal end of the groove steadily (or in some cases, abruptly) narrowed 267 towards the medial direction until its termination. In most species, the rostral termination of the 268 VNG reached the incisive foramen, except for some Lutrine species (Lutragale perspicillata, 269 Figure 3). 270 271 272 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted January 22, 2026. ; https://doi.org/10.64898/2026.01.19.700381doi: bioRxiv preprint 273 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted January 22, 2026. ; https://doi.org/10.64898/2026.01.19.700381doi: bioRxiv preprint Figure 3. Dorsal and oblique views of the ventral surface of the nasal cavities from 3D 274 renderings illustrating diversity in vomeronasal groove (VNG) shape; VNGs are traced in dashed 275 lines on the right side in dorsal views, and oblique views show arrows pointing towards the 276 lateral margin of the VNG. Directional axes (rostral [R]–caudal [C]) are indicated in each image. 277 278 279 Figure 4. Coronal views of microCT scans showing the vomeronasal grooves across a diversity 280 of caniform taxa. Color-coded silhouettes indicate ecological groups (green = terrestrial/arboreal 281 taxa; blue = semi-aquatic taxa). Yellow arrowheads indicate the vomeronasal grooves (VNG). 282 A. Gulo gulo (AMNH 37433); B. Eira barbara (MNHN-ZM-MO 2000-646); C. Martes foina 283 (USNM 173295); D. Ailuropoda melanoleuca (AMNH 89030); E. Helarctos malayanus (AMNH 284 16580); F. Vulpes vulpes (UCLA 13112); G. Lutra lutra (USNM 259466); H. Pteronura 285 brasiliensis (USNM 304663); I. Lontra canadensis (NCSM 19662); J. Enhydra lutris (MNHN 286 ZM MO 1-35-124); K. Enhydra lutris (MNHN ZM MO 1962-1647). 287 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted January 22, 2026. ; https://doi.org/10.64898/2026.01.19.700381doi: bioRxiv preprint 288 Figure 5. Three coronal CT slices of the nasal cavity of Mustela nivalis (NCSM 8102) showing 289 remnant vomeronasal cartilage (VNC) in the dry skull. Slices progress from rostral to caudal 290 from top to bottom. VNC is colored purple in the zoomed-in section. 291 292 The VNG in all lutrines, except for the sea otter Enhydra lutris, largely resembled that of 293 other mustelids but were notably shorter in the rostral-caudal dimension and oftentimes wider 294 medio-laterally (Figure 3, Figure 4). The VNG of the semi-aquatic American mink (Neogale 295 vison) was also rostral-caudally abbreviated (Figure 3). In one freshwater otter (Lutragale 296 perspicillata), the VNG terminated early and did not reach the opening of the ID (Figure 3). In 297 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted January 22, 2026. ; https://doi.org/10.64898/2026.01.19.700381doi: bioRxiv preprint contrast, the VNG of the E. lutris exhibited a derived morphology relative to other species 298 (Figure 3, Figure 4J and K). Specifically, it was dorsoventrally deep, mediolaterally narrow, and 299 cranial-caudally shortened. The lateral margin was elevated, forming a distinct bony wall that 300 bordered the groove. Unlike other caniforms, in which the lateral edge of the VNG is lower and 301 merges gradually with the lateral portion of the palate or ventral nasal floor, the VNG in E. lutris 302 was not continuous with these surrounding structures. Instead, the groove appeared more deeply 303 inset, with a steeper lateral wall. 304 305 diceCT Soft Tissue VNO Morphology - Mink, Long-tailed weasel, and Otter. 306 In the diceCT scan of the mink (Neogale vison), the VNO is present and associated with a 307 VNG structure (Figure 6). The VNO had a visible lumen that was dorsal-ventrally elongated and 308 ovoid in cross-sectional shape. A patent incisive duct provided a connection between the oral and 309 nasal cavities. Within the incisive duct, there was evidence of cartilaginous or mucosal tissue 310 present at the periphery on the bony surface (a canal with a radiolucent center with radiodense 311 tissue of differing densities along the walls) (Figure 6). In coronal, axial, and sagittal views, 312 small circular radiolucent regions surrounding the VNO, distinct from the central vomeronasal 313 duct (VND) lumen, were observed (Figure 6Cii; Figure 6Cv; white arrows). These structures 314 were most numerous along the lateral and dorsal aspects of the VNO. They may represent either 315 vascular or neural tissue, but are more likely vascular, as neural tissue typically appears 316 radiodense in diceCT (Camilieri-Asch et al., 2020; Gignac et al., 2021). These features were 317 present bilaterally and resemble the vascular VNO of Vulpes vulpes (Ortiz Leal et al., 2020). In 318 the coronal view, the ID was ventrolateral to the VNO and VND. The lumen of the VND was 319 generally ovoid across the entirety of its length, but some aspects of shape varied across the 320 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted January 22, 2026. ; https://doi.org/10.64898/2026.01.19.700381doi: bioRxiv preprint organ. Most variation in lumen shape was due to differences in the curvature of its superior and 321 inferior margins, ranging from rounded to more angular forms. In the sagittal view, both the ID 322 and the VNO were distinguishable from one another (Figure 6Cii). 323 Findings in the long-tailed weasel (Neogale frenata) were overall similar, with the 324 presence of both a VNO, VNG, and ID (Figure 7). Differences in the N. frenata compared to N. 325 vison include a larger, more dilated VND, VNO, and ID lumen, more distinct separation of the 326 VND and ID lumen, and no visual evidence of VNO vasculature. The ID progresses from 327 anterior to ventral-lateral relative to the VND and VNO, similar to N. vison (Figure 7). 328 329 Figure 6. Views of diceCT of Neogale vison, magnified to see structures of the anterior nasal 330 region and the vomeronasal organ. A. Full-head volume rendering. B. Volume rendering cropped 331 to highlight the anterior nasal region in coronal and parasagittal view; the right vomeronasal 332 organ is also cropped sagitally, indicated by the dashed line. C. Representative diceCT slices 333 shown in i, ii. coronal, iii. parasagittal, and iv, v. axial planes. Directional axes are indicated in 334 panels C iii and C iv. ID = Incisive duct; VNO = vomeronasal organ; VNC = vomeronasal 335 cartilage. 336 337 338 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted January 22, 2026. ; https://doi.org/10.64898/2026.01.19.700381doi: bioRxiv preprint 339 Figure 7. Views of diceCT of Neogale frenata (YPM:VZ:YPM MAM 014733), magnified to see 340 structures of the anterior nasal region and the vomeronasal organ. A. Full-head volume 341 rendering. B i. Three-dimensional volume rendering cropped in coronal section to highlight the 342 anterior nasal region, and B ii. further magnified view showing additional detail. C. 343 Representative diceCT slices shown in i. coronal, ii. axial and iii. parasagittal planes. Directional 344 axes are indicated in panels C ii and C iii. ID = Incisive duct; VNO = vomeronasal organ; VNC 345 = vomeronasal cartilage. 346 347 A VNO was present in the North American river otter (Lontra canadensis) (Figure 8). As 348 in N. frenata, there was no visual evidence of VNO vasculature. An incisive duct (ID) was 349 discernible from the VNO in multiple views and was generally similar to those of N. vison or N. 350 frenata (Figure 8). In L. canadensis, the VND lumen was subtler, exhibiting lower contrast 351 relative to surrounding tissue and being smaller in size compared to those of N. vison and N. 352 frenata. Additionally, the VNO in L. canadensis was surrounded by more open space between it, 353 the surrounding capsule, and the VNG than in either N. vison or N. frenata. 354 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted January 22, 2026. ; https://doi.org/10.64898/2026.01.19.700381doi: bioRxiv preprint 355 Figure 8. Series of coronal slices of diceCT of Lontra canadensis (l-cet:001), magnified to see 356 structures of the anterior nasal region and the VNO. A. Full-head volume rendering. B i. Three-357 dimensional volume rendering cropped in coronal section to highlight the anterior nasal region, 358 and B ii. further magnified view showing additional detail with VNO traced in white dashed 359 lines bilaterally. C. Representative diceCT slices in i. coronal, ii. parasagittal, and iii. axial 360 planes. Directional axes are indicated in panels C ii and C iii. ID = Incisive duct; VNO = 361 vomeronasal organ; VNC = vomeronasal cartilage. 362 363

364 Comparative VNO morphology in lutrines and other mustelids. 365 Overall, the evidence presented here suggests that the VNO is present in freshwater otters 366 but lack some of the derived features that diagnose the ‘rudimentary’ VNO in harbor seals (Table 367 1). It differs from the VNO of the American mink in having a smaller lumen relative to the body 368 of the organ, no discernible neurovasculature surrounding the VND lumen, and increased empty 369 space surrounding the VNO within the cartilaginous capsule. The functional implications of the 370 variation in VND lumen morphology and the expanded capsular space between the two 371 mustelids and the river otter described here are not yet understood. Previous studies have noted 372 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted January 22, 2026. ; https://doi.org/10.64898/2026.01.19.700381doi: bioRxiv preprint that a relatively dilated VNO lumen may indicate a thinner epithelial layer and, therefore, 373 reduced function (Smith et al., 2024). Based on the comparative diceCTs of N. vison and L. 374 canadensis, the lumen of L. canadensis appears to be less dilated than that of N. vison. This may 375 reflect the observation of the more ‘tubular’ or circular-shaped lumen, also seen in the ferret M. 376 furo, rather than an actual difference in volume due to dilation (Weiler et al., 1999). The shape of 377 the VNO lumen in the North American river otter L. canadensis was only discernible when 378 image contrast was increased and was smaller than in the other two species. Its VNO lumen 379 and/or actual VNO may be reduced in size or compressed, creating the empty space observed 380 between the outside of the VNO and the VNG or VNC. This suggests that the VNO is reduced in 381 the North American river otter but does not affect the surrounding cartilage capsule that contains 382 it. While vasculature was, presumably, observable in the N. vison diceCT, no vasculature was 383 identifiable in the L. canadensis or the N. frenata diceCT VNO. 384 Lutrines appeared to have shorter and broader VNG compared to other caniforms. The 385 reduction in VNG length may be produced by covariation or constraint with certain aspects of 386 skull morphology (e.g., relative snout length, extent of nasal turbinates). Kondoh et al. (2025) 387 suggest that such structural constraints may have driven the shift in VNO position to be housed 388 entirely within the incisive canal in Steller sea lions. However, the description of a defective 389 septum resulting in unaffected VNO morphology in domestic dogs suggests that the organ may 390 not strongly covary with specific regions or features within the nasal cavity (Dzięcioł et al., 391 2020). 392 Structure-function and phenotype-genotype decoupling in the accessory olfactory system. 393 VNG length variation may produce differences in relative VNO morphology and 394 therefore, in its function. Shorter VNOs may have reduced surface area for sensory epithelia than 395 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted January 22, 2026. ; https://doi.org/10.64898/2026.01.19.700381doi: bioRxiv preprint relatively longer VNOs. However, this assumes a one-to-one relationship between morphology 396 and function, which is often complicated by kinematics or behavior. For example, the flehmen 397 response, wherein the lips are raised to bring air over the VNO, theoretically supports the 398 function of the organ, but the presence of this behavior is not detectable by the morphology of 399 the VNS overall. In caniforms, this behavior is largely absent, with the exception of the honey 400 badger (Mellivora capensis), a fossorial mustelid (Begg et al., 2003) and reports in the polar bear 401 (Stirling, 2016) (but see the pseudo-flehmen response observed in juvenile E. lutris, Island et al. 402 2017). It is also possible that, even with a thorough understanding of both VNS genotype and 403 phenotype, neither may accurately predict the range of detectable odorants in mammal species 404 (Chengetanai et al., 2020). Ultimately, several interacting mechanisms supporting chemosensory 405 function (morphology, genes, behavior) make interpretations about relative function more 406 complex. 407 Further evaluation of whether a shortened VNO observed in Lutrinae correlates with 408 reduced function would require a more diverse sample that preserves the soft tissues of the VNO. 409 However, the reduction of both VNG length and MOB volume in Lutrines (Gittleman, 1991) 410 suggests a synchronous reduction of the two systems in the transition to an aquatic environment 411 (Suárez et al., 2012). This is noteworthy given that the relationship among components of the 412 VNS (i.e. genotype and phenotype, including morphological structures and behavioral outputs of 413 the AOB and VNO) can evolve in a mosaic manner between and within the two systems (this 414 study; Yohe & Kroll, 2018; Yohe et al., 2020; Smith et al., 2024). The observed reduction of 415 both the MOS and AOS in Lutrinae suggests that, at this phylogenetic scale, phenotypic change 416 associated with their ecological transition is sufficiently captured (Garrett & Steiper, 2014; Yohe 417 et al., 2020). 418 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted January 22, 2026. ; https://doi.org/10.64898/2026.01.19.700381doi: bioRxiv preprint The current literature does not present a strong consensus on the correlation between 419 VNS phenotype and genotype: some studies support a strong correspondence (Garrett & Steiper 420 2014; Ibarra-Soria et al. 2014), where others demonstrate that the relationship weakens or 421 becomes more complex at systemic or phylogenetic scales (Yohe et al., 2019; Yohe et al. 2020; 422 Yohe & Krell 2023; Smith et al. 2024). Such correlations touch on the idea of phenogenetic drift 423 (Weiss & Fullerton, 2000), which theorizes that selection differentially acts upon those two 424 components. The concept of phenogenetic drift suggests that different genotypes produce the 425 same phenotype, where phenotypes remain fixed or conserved while the potentially multi-loci 426 genotypes underlying them change with time and space (Weiss & Fullerton, 2000). This 427 typically applies when phenotypes are strongly selected for across different populations. The 428 present study suggests the opposite dynamic in which pinnipeds and lutrines present with similar 429 genotypes (reduced functional VNO genes, Yu et al. 2010; Zhao et al., 2020) and dissimilar 430 phenotypes (Kondoh et al., 2024). Beyond their relevance to studies of the VNS, carnivorans, 431 and secondarily aquatic mammals, these findings contribute to broader biological theory by 432 supporting a mosaic relationship between chemosensory genotype and phenotype. At the 433 taxonomic scale of Lutrinae within Caniformia, our results align with evidence that genotype-434 phenotype correspondence in chemosensory systems weakens at finer phylogenetic scales (Yohe 435 et al., 2020; Yohe & Krell, 2023) and mirrors patterns of component-level VNS mosaicism 436 described in bats (Smith et al., 2024). 437 A potential explanation for the longevity of VNS phenotypes following or during periods 438 of genomic simplification is that they are, at least in part, patterned by surrounding tissues of the 439 palate during embryological development. In this paradigm, information required for patterning 440 the VNS is not only contained within genes expressed in the VNO or vomeronasal epithelia, but 441 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted January 22, 2026. ; https://doi.org/10.64898/2026.01.19.700381doi: bioRxiv preprint by the signaling environment of the developing palate as a whole. A growing body of literature 442 suggests complex signaling dynamics exist between tissues that develop in close spatial 443 proximity (e.g., Fabbri et al., 2017). Foster et al. (2024) identify a correlation wherein reptile 444 lineages that undergo a radical restructuring of the palate also lose many, if not all, aspects of the 445 VNS in adult phenotypes. Interestingly, embryos of said lineages actually exhibit transitory 446 VNOs in early development, implying that the information required for their early patterning is 447 present, but is disrupted by derived palatal morphogenesis. These insights have relevance for the 448 retention of VNS phenotypes in mammalian lineages that have a reduced gene repertoire. 449 Sampling of palatal development in these taxa, particularly those with morphologies related to an 450 aquatic existence, might reveal explanatory factors for the presence or absence of VNS 451 phenotypes. 452 Hypothesized mechanisms for the derived VNS in pinnipeds. 453 Generally, throughout the literature, a trait that appears particularly relevant to the 454 function of the AOS is its connection to the oral cavity via the ID (Sanmartín-Vázquez et al., 455 2024). In dogs (Canis familiaris), the ID communicates between the two cavities even at birth 456 (Sanmartín-Vázquez et al., 2024)., and one of the derived traits of the VNO in harbor seals 457 (Phoca vitulina) is a lack of connection between the VNO and the oral cavity (although the 458 incisive canal is still present; Kondoh et al., 2024). The retention of communication between the 459 oral cavity and the nasal cavity where the VNO is housed in Steller sea lions (Eumetopias 460 jubatus) provides increasing evidence that this feature serves as additional anatomical support for 461 the presence of a patent VNO (Kondoh et al., 2025). 462 The loss of sensory-functional tissues in the derived VNO in harbor seals (Phoca 463 vitulina) may be due to various selective, adaptive, or structural changes. One hypothesis is that 464 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted January 22, 2026. ; https://doi.org/10.64898/2026.01.19.700381doi: bioRxiv preprint the nasal cavity requires increased mucosal secretions to maintain a protective layer in salt water 465 (Kondoh et al., 2024). This may be compounded by the loss of the nasolacrimal duct in 466 pinnipeds (Berta et al., 2018; Colitz, 2022). However, this would not explain why phocids (such 467 as the harbor seals) have derived VNOs while otariids do not, as all pinnipeds have lost the 468 nasolacrimal duct and nearly all inhabit salt water. Another alternative explanation proposed is 469 that otariids are more gregarious than phocids and generally spend more time on land - a 470 combination of increased opportunities for social communication and other behavioral and 471 biological factors creates selective pressure in maintaining a functional VNS (Kondoh et al., 472 2025). 473 Another alternative explanation for the transformation seen in pinnipeds reflects changes 474 in their respiration. In horses, which are obligate nasal breathers, the connection between their 475 oral and nasal cavities (ID) closes, although their VNO remains, and they retain a flehmen 476 response (Mader, 2019). Pinnipeds, along with other marine mammals, highly prioritize nasal 477 breathing, as the nose is superiorly located on the head relative to the oral cavity/mouth requiring 478 them to raise less of their head from the water to inhale (Maust-Mohl et al., 2019; Reidenberg & 479 Laitman, 2024). Kondoh et al. (2025) hypothesize that the derived VNO of the Steller sea lion 480 (Eumetopias jubatus) functions through air intake via the oral cavity rather than relying on a 481 venous pump. However, the extent to which pinnipeds are obligate nasal breathers remains 482 unclear, and the literature lacks comprehensive reports on panting or mouth breathing across 483 species, further complicating interpretations of VNO function in this group. In other aquatic 484 mammals, such as cetaceans, evolutionary pressures have led to a significant reorganization of 485 the facial skeleton, repositioning the nasal opening dorsally, which allows for efficient breathing 486 at the water’s surface without full emergence (Berta et al., 2014; Maust-Mohl et al., 2019). This 487 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted January 22, 2026. ; https://doi.org/10.64898/2026.01.19.700381doi: bioRxiv preprint illustrates how obligate nasal breathing and associated craniofacial adaptations have evolved in 488 response to the demands of an aquatic environment, further underscoring the need to examine 489 whether similar respiratory constraints or strategies exist in pinnipeds. 490 Variation in the VNO may also correlate with changes to the role of respiration in 491 thermoregulation. Pinnipeds appear to lack a panting mechanism for heat dissipation (Khamas et 492 al., 2012), with the exception of the Northern Fur seal (Callorhinus ursinus) (Bartholomew & 493 Wilke, 1956)—although future behavioral studies may find more exceptions to this observation. 494 Some have suggested that control of the arterial supply in the nasal mucosa may be an alternative 495 mechanism for regulating expired air temperature in pinnipeds (Folkow, 1992). Additionally, 496 correlations have been found between the number of respiratory turbinates and climate within 497 pinnipeds (Mason et al., 2020). Environmental temperature also differs across the range of 498 phocids and otariids making an ecological correlation between the presence of a derived (phocid) 499 and non-derived (otariid) VNS plausible. It has been documented that cold, dry air increases 500 nasal epithelial shedding in humans (Cruz et al., 2006). The loss of any mouth-breathing and the 501 potential for increased nasal mucosa to cope with colder temperatures, compounded with the loss 502 of the nasolacrimal duct, may produce the selective pressures that have shaped the derived 503 condition of the Harbor seal VNO. Future studies should also investigate whether AOS structure 504 or function covaries with environmental factors or other aspects of the nasal cavity (e.g., width or 505 turbinate area). 506 More broadly speaking, our understanding of some of the more specific functional 507 mechanisms of the VNS is largely limited to studies in rodent models (Salazar & Sánchez 508 Quinteiro, 2009). While these studies provide a great wealth of information regarding the VNO 509 and AOS, it is evident that our limited knowledge affects assessments of the characteristics of the 510 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted January 22, 2026. ; https://doi.org/10.64898/2026.01.19.700381doi: bioRxiv preprint non-rodent mammalian AOS (Salazar & Sánchez Quinteiro, 2009). For example, the ferret VNO, 511 which is smaller than that found in rodents, is described as rudimentary by Weiler and colleagues 512 (1999), despite the lack of research on the phylogenetic structure and diversity of VNS structures 513 in caniforms. These observations ultimately reflect an assumption that any regression of the AOS 514 relative to the rat model may reflect a non-functional system. However, comparison within a 515 phylogenetic context of caniforms highlights that their morphology reflects their own 516 evolutionary specialization relative to rodents, some of which can be described as potential 517 ‘regressions’ (Salazar et al., 2012). Future studies on the influence of ecology and phylogeny on 518 the diversity of AOS structure and function in mammals require a broader integrative approach, 519 examining the greater morphological diversity of this system within its phylogenetic context. 520 This investigation of the VNS in secondarily aquatic caniforms is essential to understanding the 521 response of mammalian sensory ecology and anatomy to ecological change. 522 The definition of the ‘functional’ VNO. 523 Future research is needed to standardize the terminology used to relate VNO structure to 524 function. The literature currently employs a wide range of overlapping and sometimes 525 ambiguous descriptors, including distinctions such as “functional” versus “non-functional” and 526 “vestigial” or “rudimentary” versus “well-developed” or “true” VNOs (Smith et al., 2014). 527 Recent research has attempted to provide more impactful definitions of a functional VNO as “a 528 neuroepithelium is essential to VNO function as a chemosensory organ, and we use the term 529 'neuroepithelial VNO' to refer to those that possess both a neuroepithelium and bundle axons 530 departing the basal aspect of the neuroepithelium. Based on our observations of the morphology, 531 we might refer to bilateral epithelial tubes lacking a neuroepithelium and coextensive with VNCs 532 as “putative rudimentary VNOs” (Smith et al., 2024), in contrast to the “true” VNS defined by 533 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted January 22, 2026. ; https://doi.org/10.64898/2026.01.19.700381doi: bioRxiv preprint Salazar and Sánchez-Quinteiro (2009) as comprising three components (the VNO, accessory 534 olfactory bulb, and vomeronasal amygdala) with corresponding nerves and connections. 535 Patterns observed in bats highlight the complexity of the relationships between AOS 536 anatomical components that may reflect ‘functional’ versus ‘rudimentary’ VNOs (Smith et al., 537 2024). Bats that lack AOBs and functional VNO genes retain neurological and vascular VNO 538 structures that persist into adulthood. Bats without a ‘functional’ VNO (“possess(ing) both a 539 neuroepithelium and bundle axons departing the basal aspect of the neuroepithelium”) often 540 retain multiple structural elements of the VNO (Smith et al., 2024). This would mean that the 541 definition of the ‘true’ VNO outlined by Salazar and Sánchez-Quinteiro (2009) would define 542 those bats as likely having “true” VNOs, while Smith et al. (2024) would describe them as 543 “putative rudimentary VNOs”. The presence of neurovascular tissue as a criterion for a 544 functional VNO/AOS is complicated further by the cases of nerves from the VNO 545 communicating with the MOB instead of the AOB in species that lack an AOB (Salazar & 546 Sánchez-Quinteiro, 2009). This suggests that the presence of neural tissue and vascular structures 547 may be a poor predictor of a “functional” VNO in carnivorans or, more generally, across 548 Mammalia, although further research is needed. In general, future research should explicitly 549 investigate the VNS across its various levels of organization (genotype and multiple aspects of 550 phenotype) before defining groups as having “non-functional” VNS. 551 Other future directions. 552 Evaluating the soft-tissue morphology of the sea otter VNO would greatly contribute to 553 further determining correlations between ecological, genetic, and anatomical transformations in 554 the VNS. Sea otters (Enhydra lutris) are widely regarded as ecologically distinct from freshwater 555 otters (Estes, 1989; Bird et al., 2020), spending significantly less time on land and exhibiting 556 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted January 22, 2026. ; https://doi.org/10.64898/2026.01.19.700381doi: bioRxiv preprint minimal reliance on terrestrial environments for feeding, reproduction, offspring care, or other 557 ecological functions. In birds, gene patterns suggest that varying degrees and types of aquatic 558 lifestyles are associated with corresponding, gradational changes in the olfactory gene repertoire 559 (Lu et al., 2016), a pattern that may likewise be reflected in the AOS structures of aquatic 560 mammals. Based on the derived VNG morphology compared to other otters and their lifestyle, 561 sea otters would be the most likely lutrine to converge on the AOS morphology seen in the 562 Harbor seal (Phoca vitulina). It is worth noting that captive juvenile sea otters have been 563 observed exhibiting a flehmen-like response (Island et al., 2017); however, the relationship 564 between this behavioral observation and the state of the sea otter VNS is not clearly understood. 565 Another future contribution that would significantly contribute to our understanding of 566 the AOS in secondarily aquatic caniforms would be a thorough comparative description or 567 identification of the AOB. While here we identify and describe soft tissue of the VNO in the 568 North American river otter, there are no data in the literature describing their AOB. A 569 description of AOB structures has previously only been achievable through histology or 570 dissection (Ortiz-Leal et al., 2020, 2024). However, recent work (Gignac et al., 2021; Straight et 571 al., 2024) presents the exciting possibility of identifying and describing it in detail using diceCT 572 techniques. 573 In addition to otters, greater comparative sampling across phocid seals would be highly 574 beneficial, as they represent immense ecological and taxonomic diversity that is 575 underrepresented in the descriptive literature on VNS morphology. While describing or 576 investigating these morphologies using diceCT or histology would be valuable, some aspects of 577 the VNS may potentially be represented using osteological correlates. For example, it is 578 hypothesized that the AOB may be observable on the endocast of some otariid pinnipeds (Loza 579 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted January 22, 2026. ; https://doi.org/10.64898/2026.01.19.700381doi: bioRxiv preprint et al., 2023), and the presence of a patent communication between the oral and nasal cavities can 580 be investigated in hard tissues via the incisive canal. 581 While the VNG serves as a strong proxy for dimensions of the VNO in primates (Smith 582 et al., 2011; Garrett et al., 2013; Garrett, 2015), it is possible, considering the condition seen in 583 pinnipeds (Kondoh et al., 2024), that the VNG is simply an osteological correlate for the VNC in 584 carnivorans. While this is still promising, as the VNC is strongly correlated with the presence of 585 a VNO, or at least derived structures homologous to a VNO, it means that the presence of a VNG 586 is not necessarily an osteological correlate of a functional VNO. The harbor seal (Phoca vitulina) 587 retains what is regarded as a rudimentary (or exapted) VNO (secretory tissue) and VNC (Kondoh 588 et al., 2024). Further, the presence of the VNG as evidence for a VNO in pinnipeds is 589 complicated by the fact that the Steller sea lion (Eumetopias jubatus) VNO is present but is 590 constrained to the incisive canal and does not extend into the nasal cavity (Kondoh et al., 2025). 591 While this study confirms the presence of a VNO and establishes that the VNG is a strong 592 correlate for the presence of VNO-derived or homologous tissue, further study is required to 593 better reconstruct histo-morphological features that better capture ‘functional’ aspects of the 594 VNO. These would include examination of the epithelium expressed throughout the VNO lumen 595 and a detailed study of VNO neurovasculature. 596

597 Here, we established the presence of a vomeronasal organ in one species, Lontra 598 canadensis, a member of the family Lutrinae for which very little was known regarding the 599 functional morphology of their accessory olfactory system. The correlation between a VNO and 600 VNG across the sample of Caniformia suggests that the VNG can offer insight into the evolution 601 of the AOS within carnivorans. Our findings suggest that there is phenogenetic drift within the 602 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted January 22, 2026. ; https://doi.org/10.64898/2026.01.19.700381doi: bioRxiv preprint caniform VNS, at least in secondarily aquatic species, highlighting the mosaic, irregular, and 603 complex nature of structural and genetic change in the VNS. Future research should prioritize 604 characterizing the histo-morphological aspects of the VNS in secondarily aquatic carnivorans to 605 elucidate phenogenetic patterns, especially during ecological transitions. Such work would also 606 facilitate comparisons with VNS changes documented in other mammalian clades (Yohe & 607 Krell, 2023) and may help identify generalized characteristics of the system across mammals or 608 even tetrapods. 609 610 611 612 613 Sources cited: 614 Allouch, G. M., & Alshanbari, F. A. (2024). Comparative Anatomy of the Vomeronasal 615 Organ (VNO) in Sheep (Ovis aries) and Dogs (Canis familiaris) with Simple Reference to its 616 Histological Structure and Vasculature Supply. International Journal of Morphology, 42(2). 617 Aron, C. (1979). Mechanisms of control of the reproductive function by olfactory stimuli 618 in female mammals. Physiological Reviews, 59(2), 229-284. 619 Bartholomew, G. A., & Wilke, F. (1956). Body temperature in the northern fur seal, 620 Callorhinus ursinus. Journal of Mammalogy, 37(3), 327–337. 621 Baum, M. J. (2012). Contribution of pheromones processed by the main olfactory system 622 to mate recognition in female mammals. Frontiers in neuroanatomy, 6, 20. 623 Baum, M. J., & Cherry, J. A. (2015). Processing by the main olfactory system of 624 chemosignals that facilitate mammalian reproduction. Hormones and behavior, 68, 53-64. 625 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted January 22, 2026. ; https://doi.org/10.64898/2026.01.19.700381doi: bioRxiv preprint Baum, M. J., & Kelliher, K. R. (2009). Complementary roles of the main and accessory 626 olfactory systems in mammalian mate recognition. Annual Review of Physiology, 71(1), 141-627 160. 628 Begg, C. M., Begg, K. S., Du Toit, J. T., & Mills, M. G. L. (2003). Scent-marking 629 behaviour of the honey badger, Mellivora capensis (Mustelidae), in the southern 630 Kalahari. Animal behaviour, 66(5), 917-929. 631 Bendel, E. M., Kammerer, C. F., Kardjilov, N., Fernandez, V., & Fröbisch, J. (2018). 632 Cranial anatomy of the gorgonopsian Cynariops robustus based on CT-reconstruction. PLoS 633 One, 13(11), e0207367. 634 Berta, A., Churchill, M., & Boessenecker, R. W. (2018). The Origin and Evolutionary 635 Biology of Pinnipeds. Annual Review of Earth and Planetary Sciences, 46, 203–228. 636 Berta, A., Ekdale, E. G., & Cranford, T. W. (2014). Review of the cetacean nose: form, 637 function, and evolution. The Anatomical Record, 297(11), 2205-2215. 638 Berta, A., Sumich, J. L., & Kovacs, K. M. (2014). Marine Mammals: Evolutionary 639 Biology (3rd ed.). Academic Press. 640 Bird, D. J., Hamid, I., Fox‐Rosales, L., & Van Valkenburgh, B. (2020). Olfaction at 641 depth: Cribriform plate size declines with dive depth and duration in aquatic arctoid 642 carnivorans. Ecology and Evolution, 10(14), 6929-6953. 643 Bird, D. J., Murphy, W. J., Fox-Rosales, L., Hamid, I., Eagle, R. A., & Van Valkenburgh, 644 B. (2018). Olfaction written in bone: cribriform plate size parallels olfactory receptor gene 645 repertoires in Mammalia. Proceedings of the Royal Society B: Biological Sciences, 285(1874), 646 20180100. 647 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted January 22, 2026. ; https://doi.org/10.64898/2026.01.19.700381doi: bioRxiv preprint Caianiello, S. (2024). The Strange Story of Mosaic Evolution. In: Delisle, R.G., Esposito, 648 M., Ceccarelli, D. (eds) Unity and Disunity in Evolutionary Biology. Springer, 649 Cham. https://doi.org/10.1007/978-3-031-42629-2_13 650 Camilieri-Asch, V., Shaw, J. A., Mehnert, A., Yopak, K. E., Partridge, J. C., & Collin, S. 651 P. (2020). diceCT: A valuable technique to study the nervous system of fish. Eneuro, 7(4). 652 Chengetanai, S., Bhagwandin, A., Bertelsen, M. F., Hård, T., Hof, P. R., Spocter, M. A., 653 & Manger, P. R. (2020). The brain of the African wild dog. II. The olfactory system. Journal of 654 Comparative Neurology, 528(18), 3285-3304. 655 Colitz, C. (2022). Ophthalmology of pinnipedimorpha: seals, sea lions, and walruses. 656 In Wild and Exotic Animal Ophthalmology: Volume 2: Mammals (pp. 269-309). Cham: Springer 657 International Publishing. 658 Collin, S. P., Yopak, K. E., Crowe‐Riddell, J. M., Camilieri‐Asch, V., Kerr, C. C., 659 Robins, H., ... & Chapuis, L. (2024). Bioimaging of sense organs and the central nervous system 660 in extant fishes and reptiles in situ: A review. The Anatomical Record. 661 Crompton, A. W., Owerkowicz, T., Bhullar, B. A., & Musinsky, C. (2017). Structure of 662 the nasal region of non-mammalian cynodonts and mammaliaforms: speculations on the 663 evolution of mammalian endothermy. Journal of Vertebrate Paleontology, 37(1), e1269116. 664 Cruz, A. A., Naclerio, R. M., Proud, D., & Togias, A. (2006). Epithelial shedding is 665 associated with nasal reactions to cold, dry air. Journal of allergy and clinical 666 immunology, 117(6), 1351-1358. 667 Dawood, Y., Hagoort, J., Siadari, B. A., Ruijter, J. M., Gunst, Q. D., Lobe, N. H. J., ... & 668 Van Den Hoff, M. J. B. (2021). Reducing soft-tissue shrinkage artefacts caused by staining with 669 Lugol’s solution. Scientific reports, 11(1), 19781. 670 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted January 22, 2026. ; https://doi.org/10.64898/2026.01.19.700381doi: bioRxiv preprint de Ferran, V., Figueiro, H. V., de Jesus Trindade, F., Smith, O., Sinding, M. H. S., 671 Trinca, C. S., ... & Eizirik, E. (2022). Phylogenomics of the world’s otters. Current Biology, 672 32(16), 3650-3658. 673 De Vreese, S., Orekhova, K., Morell, M., Gerussi, T., & Graïc, J. M. (2023). 674 Neuroanatomy of the cetacean sensory systems. Animals, 14(1), 66. 675 Debey, L. B., & Pyenson, N. D. (2013). Osteological correlates and phylogenetic analysis 676 of deep diving in living and extinct pinnipeds: what good are big eyes?. Marine Mammal 677 Science, 29(1), 48-83. 678 Dzięcioł, M., Podgórski, P., Stańczyk, E., Szumny, A., Woszczyło, M., Pieczewska, 679 B., ... & Wrzosek, M. A. (2020). MRI features of the vomeronasal organ in dogs (Canis 680 familiaris). Frontiers in Veterinary Science, 7, 159. 681 Estes, J. A. (1989). Adaptations for aquatic living by carnivores. In Carnivore behavior, 682 ecology, and evolution (pp. 242-282). Boston, MA: Springer US. 683 Estes, R. D. (1972). The role of the vomeronasal organ in mammalian reproduction. 684 Evans, H. E. (1993). Miller’s Anatomy of the Dog (3rd ed.). W.B. Saunders. 685 Fabbri, M., Mongiardino Koch, N., Pritchard, A. C., Hanson, M., Hoffman, E., Bever, G. 686 S., Balanoff, A. M., Morris, Z. S., Field, D. J., Camacho, J. and Rowe, T. B. 2017. The skull roof 687 tracks the brain during the evolution and development of reptiles including birds. Nature ecology 688 & evolution, 1(10), 1543-1550. 689 Folkow, L. P. (1992). Adrenergic vasomotor responses in nasal mucosa of hooded 690 seals. American Journal of Physiology-Regulatory, Integrative and Comparative 691 Physiology, 263(6), R1291-R1297. 692 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted January 22, 2026. ; https://doi.org/10.64898/2026.01.19.700381doi: bioRxiv preprint Foster, W., Gensbigler, P., Wilson, J. D., Smith, RMH., Lyson, T. R., & Bever, G. S. 693 2024. Cranial Anatomy of the Triassic Rhynchosaur Mesosuchus browni based on computed 694 tomography, with a discussion of the vomeronasal system and its deep history in Reptilia. 695 Zoological Journal of the Linnean Society, 201(4), zlae097. 696 Garrett, E. C. (2015). Was there a sensory trade-off in primate evolution? The 697 vomeronasal groove as a means of understanding the vomeronasal system in the fossil record. 698 City University of New York. 699 Garrett, E. C., Dennis, J. C., Bhatnagar, K. P., Durham, E. L., Burrows, A. M., Bonar, C. 700 J., ... & Smith, T. D. (2013). The vomeronasal complex of nocturnal strepsirhines and 701 implications for the ancestral condition in primates. The Anatomical Record, 296(12), 1881-702 1894. 703 Gignac, P. M., Kley, N. J., Clarke, J. A., Colbert, M. W., Morhardt, A. C., Cerio, D., ... & 704 Witmer, L. M. (2016). Diffusible iodine‐based contrast‐enhanced computed tomography 705 (diceCT): an emerging tool for rapid, high‐resolution, 3‐D imaging of metazoan soft 706 tissues. Journal of anatomy, 228(6), 889-909. 707 Gignac, P. M., O’Brien, H. D., Sanchez, J., & Vazquez-Sanroman, D. (2021). Multiscale 708 imaging of the rat brain using an integrated diceCT and histology workflow. Brain Structure and 709 Function, 226(7), 2153-2168. 710 Gittleman, J. L. (1991). Carnivore olfactory bulb size: allometry, phylogeny and ecology. 711 Journal of Zoology, 225(2), 253-272. 712 Hadden, P. W., & Zhang, J. (2023). An overview of the penguin visual 713 system. Vision, 7(1), 6. 714 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted January 22, 2026. ; https://doi.org/10.64898/2026.01.19.700381doi: bioRxiv preprint Hecker, N., Lächele, U., Stuckas, H., Giere, P., & Hiller, M. (2019). Convergent 715 vomeronasal system reduction in mammals coincides with convergent losses of calcium 716 signalling and odorant‐degrading genes. Molecular Ecology, 28(16), 3656-3668. 717 Hillenius, W. J. (2000). Septomaxilla of nonmammalian synapsids: soft‐tissue correlates 718 and a new functional interpretation. Journal of morphology, 245(1), 29-50. 719 Hughes, N. K., Price, C. J., & Banks, P. B. (2010). Predators are attracted to the olfactory 720 signals of prey. PLoS One, 5(9), e13114. 721 Island, H. D., Wengeler, J., & Claussenius‐Kalman, H. (2017). The flehmen response and 722 pseudosuckling in a captive, juvenile Southern sea otter (Enhydra lutris nereis). Zoo 723 Biology, 36(1), 30-39. 724 Kelliher, K. R., Baum, M. J., & Meredith, M. (2001). The ferret's vomeronasal organ and 725 accessory olfactory bulb: effect of hormone manipulation in adult males and females. The 726 Anatomical Record: An Official Publication of the American Association of Anatomists, 263(3), 727 280-288. 728 Keverne, E. B. (2004). Importance of olfactory and vomeronasal systems for male sexual 729 function. Physiology & behavior, 83(2), 177-187. 730 Khamas, W. A., Smodlaka, H., Leach-Robinson, J., & Palmer, L. (2012). Skin histology 731 and its role in heat dissipation in three pinniped species. Acta Veterinaria Scandinavica, 54, 1-732 10. 733 Kondoh, D., Tonomori, W., Iwasaki, R., Tomiyasu, J., Kaneoya, Y., Kawai, Y. K., ... & 734 Kobayashi, M. (2024). The vomeronasal organ and incisive duct of harbor seals are modified to 735 secrete acidic mucus into the nasal cavity. Scientific Reports, 14(1), 11779. 736 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted January 22, 2026. ; https://doi.org/10.64898/2026.01.19.700381doi: bioRxiv preprint Kondoh, D., Tonomori, W., Iwasaki, R., Tomiyasu, J., Kaneoya, Y., Li, H., ... & 737 Kobayashi, M. (2025). The vomeronasal system of the Steller sea lion. Journal of Anatomy. 738 Leng, L., & Shi, L. (2025). How Foraging Mode Sculpts Sensory Systems: 739 Morphological Evidence From DiceCT and Histology in Sympatric Lizards. Ecology and 740 Evolution, 15(8), e72042. 741 Liu, A., He, F., Shen, L., Liu, R., Wang, Z., & Zhou, J. (2019). Convergent degeneration 742 of olfactory receptor gene repertoires in marine mammals. BMC genomics, 20, 1-14. 743 Loza, C. M., Sánchez-Villagra, M. R., Scarano, A. C., Romero, M., Barbeito, C. G., & 744 Carlini, A. A. (2023). The brain of fur seals, seals, and walrus (Pinnipedia): A comparative 745 anatomical and phylogenetic study of cranial endocasts of semiaquatic mammals. Journal of 746 Mammalian Evolution, 30(4), 1011-1028. 747 Lu, Q., Wang, K., Lei, F., Yu, D., & Zhao, H. (2016). Penguins reduced olfactory 748 receptor genes common to other waterbirds. Scientific Reports, 6(1), 31671. 749 Mader, B. J. (2019). The narial morphology of Metarhinus and Sphenocoelus 750 (Mammalia, Perissodactyla, Brontotheriidae). 751 Mahdy, E. A., & Mohamed, S. K. A. (2019). Comparative morpho-histological analysis 752 on the vomeronasal organ and the accessory olfactory bulb in Balady dogs (Canis familiaris) and 753 New Zealand rabbits (Oryctolagus cuniculus). Journal of Advanced Veterinary and Animal 754 Research, 6(4), 506. 755 Mason, M. J., Wenger, L. M., Hammer, Ø., & Blix, A. S. (2020). Structure and function 756 of respiratory turbinates in phocid seals. Polar Biology, 43, 157-173. 757 Niimura, Y., & Nei, M. (2007). Extensive gains and losses of olfactory receptor genes in 758 mammalian evolution. PloS one, 2(8), e708. 759 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted January 22, 2026. ; https://doi.org/10.64898/2026.01.19.700381doi: bioRxiv preprint Ortiz‐Leal, I., Torres, M. V., Barreiro‐Vázquez, J. D., López‐Beceiro, A., Fidalgo, L., 760 Shin, T., & Sanchez‐Quinteiro, P. (2024). The vomeronasal system of the wolf (Canis lupus 761 signatus): The singularities of a wild canid. Journal of Anatomy, 245(1), 109-136. 762 Ortiz-Leal, I., Torres, M. V., Villamayor, P. R., Fidalgo, L. E., López-Beceiro, A., & 763 Sanchez-Quinteiro, P. (2022). Can domestication shape Canidae brain morphology? The 764 accessory olfactory bulb of the red fox as a case in point. Annals of Anatomy-Anatomischer 765 Anzeiger, 240, 151881. 766 Ortiz‐Leal, I., Torres, M. V., Villamayor, P. R., López‐Beceiro, A., & Sanchez‐Quinteiro, 767 P. (2020). The vomeronasal organ of wild canids: the fox (Vulpes vulpes) as a model. Journal of 768 Anatomy, 237(5), 890-906. 769 Paulina-Carabajal, A., Acosta-Hospitaleche, C., & Yury-Yáñez, R. E. (2015). 770 Endocranial morphology of Pygoscelis calderensis (Aves, Spheniscidae) from the Neogene of 771 Chile and remarks on brain morphology in modern Pygoscelis. Historical Biology, 27(5), 571-772 582. 773 Pihlström, H., Thewissen, J. G. M., & Nummela, S. (2008). Comparative anatomy and 774 physiology of chemical senses in aquatic mammals. Sensory evolution on the threshold: 775 Adaptations in secondarily aquatic vertebrates, 95-109. 776 Poo, C., Agarwal, G., Bonacchi, N., & Mainen, Z. F. (2022). Spatial maps in piriform 777 cortex during olfactory navigation. Nature, 601(7894), 595-599. 778 Reidenberg, J. S., & Laitman, J. T. (2025). Review of respiratory anatomy adaptations in 779 whales. The Anatomical Record, 308(4), 1179-1213. 780 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted January 22, 2026. ; https://doi.org/10.64898/2026.01.19.700381doi: bioRxiv preprint Rolfe, S., Pieper, S., Porto, A., Diamond, K., Winchester, J., Shan, S., ... & Maga, A. M. 781 (2021). SlicerMorph: An open and extensible platform to retrieve, visualize and analyse 3D 782 morphology. Methods in Ecology and Evolution, 12(10), 1816-1825. 783 Salazar, I., & Sánchez Quinteiro, P. (2009). The risk of extrapolation in neuroanatomy: 784 the case of the mammalian vomeronasal system. Frontiers in neuroanatomy, 3, 982. 785 Salazar, I., Cifuentes, J. M., & Sánchez‐Quinteiro, P. (2013). Morphological and 786 immunohistochemical features of the vomeronasal system in dogs. The Anatomical Record: 787 Advances in Integrative Anatomy and Evolutionary Biology, 296(1), 146-155. 788 Salazar, I., Lombardero, M., Cifuentes, J. M., Quinteiro, P. S., & Alemañ, N. (2003). 789 Morphogenesis and growth of the soft tissue and cartilage of the vomeronasal organ in pigs. 790 Journal of Anatomy, 202(6), 503-514. 791 Salazar, I., Quinteiro, P. S., & Cifuentes, J. M. (1995). Comparative anatomy of the 792 vomeronasal cartilage in mammals: mink, cat, dog, pig, cow and horse. Annals of Anatomy-793 Anatomischer Anzeiger, 177(5), 475-481. 794 Sanmartín-Vázquez, E., Ortiz-Leal, I., Torres, M. V., Kalak, P., Kubiak-Nowak, D., 795 Dzięcioł, M., & Sanchez-Quinteiro, P. (2024). Functional Role of the Incisive Duct in Neonatal 796 Dogs. Cells Tissues Organs, 1-30. 797 Smith, T. D., Corbin, H. M., King, S. E., Bhatnagar, K. P., & DeLeon, V. B. (2021). A 798 comparison of diceCT and histology for determination of nasal epithelial type. PeerJ, 9, e12261. 799 the language of anatomical reduction. The Anatomical Record, 297(11), 2196-2204. 800 Smith, T. D., Downing, S. E., Rosenberger, V. B., Loeffler, J. R., King, N. A., Curtis, A. 801 A., ... & Santana, S. E. (2024). Functional microanatomy of the vomeronasal complex of 802 bats. The Anatomical Record. 803 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted January 22, 2026. ; https://doi.org/10.64898/2026.01.19.700381doi: bioRxiv preprint Smith, T. D., Garrett, E. C., Bhatnagar, K. P., Bonar, C. J., Bruening, A. E., Dennis, J. 804 C., ... & Morrison, E. E. (2011). The vomeronasal organ of New World monkeys 805 (Platyrrhini). The Anatomical Record: Advances in Integrative Anatomy and Evolutionary 806 Biology, 294(12), 2158-2178. 807 Smith, T. D., Laitman, J. T., & Bhatnagar, K. P. (2014). The shrinking anthropoid nose, 808 the human vomeronasal organ, and the language of anatomical reduction. The Anatomical 809 Record, 297(11), 2196-2204. 810 Stirling, I., Spencer, C., & Andriashek, D. (2016). Behavior and activity budgets of wild 811 breeding polar bears (Ursus maritimus). Marine Mammal Science, 32(1), 13-37. 812 Straight, P. J., Gignac, P. M., & Kuenzel, W. J. (2024). A histological and diceCT-813 derived 3D reconstruction of the avian visual thalamofugal pathway. Scientific Reports, 14(1), 814 8447. 815 Suárez, R., García-González, D., & De Castro, F. (2012). Mutual influences between the 816 main olfactory and vomeronasal systems in development and evolution. Frontiers in 817 neuroanatomy, 6, 50. 818 Switzer, III, R. C., Johnson, J. I., & Kirsch, J. A. (1980). Phylogeny through brain traits: 819 relation of lateral olfactory tract fibers to the accessory olfactory formation as a palimpsest of 820 mammalian descent. Brain, Behavior and Evolution, 17(5), 339-363. 821 Tirindelli, R. (2021). Coding of pheromones by vomeronasal receptors. Cell and Tissue 822 Research, 383(1), 367-386. 823 Tomiyasu, J., Kondoh, D., Sakamoto, H., Matsumoto, N., Sasaki, M., Kitamura, N., ... & 824 Matsui, M. (2017). Morphological and histological features of the vomeronasal organ in the 825 brown bear. Journal of anatomy, 231(5), 749-757. 826 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted January 22, 2026. ; https://doi.org/10.64898/2026.01.19.700381doi: bioRxiv preprint Waku, D., Segawa, T., Yonezawa, T., Akiyoshi, A., Ishige, T., Ueda, M., ... & Sasaki, T. 827 (2016). Evaluating the phylogenetic status of the extinct Japanese otter on the basis of 828 mitochondrial genome analysis. PLoS One, 11(3), e0149341. 829 Weiler, E., Apfelbach, R., & Farbman, A. I. (1999). The vomeronasal organ of the male 830 ferret. Chemical senses, 24(2), 127-136. 831 Weiss, K. M., & Fullerton, S. M. (2000). Phenogenetic drift and the evolution of 832 genotype–phenotype relationships. Theoretical population biology, 57(3), 187-195. 833 Witmer, L. M., & Thomason, J. J. (1995). The extant phylogenetic bracket and the 834 importance of reconstructing soft tissues in fossils. Functional morphology in vertebrate 835 paleontology, 1, 19-33. 836 Yohe, L. R., & Krell, N. T. (2023). An updated synthesis of and outstanding questions in 837 the olfactory and vomeronasal systems in bats: Genetics asks questions only anatomy can 838 answer. The Anatomical Record, 306(11), 2765-2780. 839 Yohe, L. R., Fabbri, M., Hanson, M., & Bhullar, B. A. S. (2020). Olfactory receptor gene 840 evolution is unusually rapid across Tetrapoda and outpaces chemosensory phenotypic change. 841 Current Zoology, 66(5), 505-514. 842 Yohe, L. R., Hoffmann, S., & Curtis, A. (2018). Vomeronasal and olfactory structures in 843 bats revealed by DiceCT clarify genetic evidence of function. Frontiers in neuroanatomy, 12, 32. 844 Yu, L., Jin, W., Wang, J. X., Zhang, X., Chen, M. M., Zhu, Z. H., ... & Zhang, Y. P. 845 (2010). Characterization of TRPC2, an essential genetic component of VNS chemoreception, 846 provides insights into the evolution of pheromonal olfaction in secondary-adapted marine 847 mammals. Molecular biology and evolution, 27(7), 1467-1477. 848 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted January 22, 2026. ; https://doi.org/10.64898/2026.01.19.700381doi: bioRxiv preprint Zellmer, N. T., Timm-Davis, L. L., & Davis, R. W. (2021). Sea otter behavior: 849 morphologic, physiologic, and sensory adaptations. In Ethology and behavioral ecology of sea 850 otters and polar bears (pp. 23-55). Cham: Springer International Publishing. 851 Zhang, Z., & Nikaido, M. (2020). Inactivation of ancV1R as a predictive signature for the 852 loss of vomeronasal system in mammals. Genome Biology and Evolution, 12(6), 766-778. 853 Zhao, H., Xu, D., Zhang, S., & Zhang, J. (2011). Widespread losses of vomeronasal 854 signal transduction in bats. Molecular biology and evolution, 28(1), 7-12. 855 Zufall, F., Kelliher, K. R., & Leinders‐Zufall, T. (2002). Pheromone detection by 856 mammalian vomeronasal neurons. Microscopy research and technique, 58(3), 251-260. 857 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted January 22, 2026. ; https://doi.org/10.64898/2026.01.19.700381doi: bioRxiv preprint