Rapoport's rule — the tendency for species' ranges to broaden with distance from benign environmental conditions — has been tested in the ocean almost exclusively along the latitudinal axis, with mixed results. Reygondeau and Egorova (2025, bioRxiv) re-examined the question in three dimensions for 21,958 marine species, combining modelled latitudinal distributions (AquaMaps 2.0) with independently compiled depth limits (FishBase/SeaLifeBase). They found that Rapoport's rule is weakly and inconsistently supported along latitude (Pearson r ≈ -0.02 between latitudinal range and absolute mid-latitude) but strongly and pervasively supported along depth (vertical range vs mid-depth, r ≈ 0.88 across habitats and major taxa), with the two range dimensions only moderately coupled (r ≈ 0.24) and a sparse subset of '3-D generalists' combining kilometre-scale depth ranges with basin-scale latitudinal extents. The authors explicitly leave open: (1) whether these patterns are genuine ecological signals rather than artifacts of sampling bias, binning, or heterogeneous data streams; (2) which ecological or evolutionary mechanisms link vertical and horizontal range sizes and produce the 3-D generalists; and (3) the relative contributions of ecological tolerance and dispersal versus evolutionary history to range expansion. This problem asks for a critical, mechanism-level account of why Rapoport's rule holds along the depth axis but only weakly along the latitudinal axis of the ocean, and of what couples the two axes.

Background

Rapoport's rule, formalized by George Stevens (1989), posits that species at higher latitudes have broader geographic ranges than tropical species; Stevens attributed this to greater climatic variability at high latitudes selecting for broad environmental tolerance, which in turn permits large ranges. Stevens (1992) extended the rule to elevation on mountains. The rule has had a contentious history: Gaston and Chown (1999) argued it does not generalize across taxa, regions, or analytical methods, and it remains an active debate (Radomski, 2025). In the marine realm, tests have focused almost entirely on the latitudinal axis, treating the ocean as a one-dimensional horizontal system: Stevens (1996) extended the rule to Pacific marine fishes and proposed a 'bathymetric rule' in which bathymetric range increases with latitude; Fortes and Absalão (2004, 2010) tested both latitudinal and bathymetric (depth-range-vs-latitude) versions globally for marine molluscs and fishes, with regionally mixed but broadly supportive results for the latitude-based bathymetric version; Smith and Gaines (2003) showed that the apparent link between the bathymetric rule and the latitudinal diversity gradient in Northeast Pacific fishes and Northwest Atlantic gastropods is an artifact of the loss of shallow-restricted tropical species at high latitudes; and Kendall and Haedrich (2006) assessed Atlantic deep-sea fish richness against the mid-domain effect and Rapoport's rule, without finding the bathymetric Rapoport prediction supported (the mid-domain effect is the expectation that ranges and diversity peak toward the middle of a bounded geographic domain purely from geometric constraint, since ranges are truncated at domain edges). A separate deep-sea literature asked whether vertical (depth) ranges themselves broaden with depth: Pineda (1993) documented that deep-sea benthic species in shelf and slope regions have smaller vertical ranges than species of the continental rise and abyss, and modelled vertical ranges under hard boundaries at the sea surface and the deepest sea floor; Pineda and Caswell (1998) showed that random rearrangements of ranges under such boundary constraints cannot explain the bathymetric diversity patterns of deep-sea gastropods and polychaetes; and Moreno et al. (2008) found that Southeast Pacific polychaetes at abyssal depths have extremely wide bathymetric ranges, a pattern robust to mid-domain-effect simulation and consistent with Rapoport's effect together with source–sink dynamics (in which sink populations at depth are maintained by dispersal from shallower source populations). These regional, taxon-narrow studies anticipated the vertical range–mid-depth relationship but did not assemble a global, cross-taxon, three-dimensional dataset, and none related vertical to latitudinal range structure. For the horizontal axis, Pie et al. (2021) showed with phylogenetic comparative methods (statistical analyses that account for species' shared evolutionary history) that latitudinal range size in reef-associated fishes has strong phylogenetic signal and follows an 'inverse' Rapoport pattern (smaller ranges at higher latitudes) — addressing the evolutionary question for one clade but only in one dimension. On the theory side, Saupe et al. (2019) showed by simulation that latitudinal gradients in realized niche breadth (the breadth of environmental conditions a species actually occupies) and range size can emerge from the spatial dynamics of climate alone, without intrinsic differences in fundamental tolerance — a direct caution for tolerance-based interpretations of any Rapoport pattern. Into this setting, Reygondeau and Egorova (2025) combined AquaMaps 2.0 modelled distributions (AquaMaps is a global species-distribution-modelling framework that maps suitability of ~0.5° ocean cells from occurrence records and environmental layers; its next-generation revision, AquaX, is described in Reygondeau et al., 2026; a species distribution model, or SDM, correlates occurrences with environmental variables to predict habitat) with depth limits from FishBase and SeaLifeBase (global species-level databases for fishes and other marine organisms) for 21,958 species spanning pelagic and benthic habitats and major animal phyla. Latitudinal range was related to absolute mid-latitude and vertical range to mid-depth. Three findings define the open problem. First, latitudinal Rapoport correlations are near zero or slightly negative (all species: Pearson r = -0.02, Spearman ρ = -0.04), even though the scatter decomposes into two regimes — a tropical-dominated regime with a steep range–latitude slope (1.6° of range per 1° of latitude, r = 0.72) and an extra-tropical regime with a shallower slope (0.4, r = 0.50) — whose superposition cancels the global signal. Second, vertical Rapoport correlations are very strong (r = 0.88, ρ = 0.94; Pearson r = 0.87–0.89 for both pelagic and benthic/demersal species, and r > 0.8 in every major taxon with sufficient sample size), with the widest vertical ranges occurring at intermediate-to-bathyal mid-depths (the bathyal zone, roughly the continental slope and rise down to ~3,500–4,000 m). Third, latitudinal and vertical range sizes are positively but only moderately coupled (r = 0.24 overall; pelagic r = 0.27 vs benthic/demersal r = 0.21; echinoderms r = 0.42, marine mammals r = 0.36, down to arthropods and annelids r = 0.21), with a sparse tail of '3-D generalists' — species combining kilometre-scale depth ranges with basin-scale latitudinal extents — while most species remain restricted in at least one dimension. The authors attribute the vertical pattern tentatively to depth-structured environmental tolerance (depth integrates monotonic gradients in light, temperature, pressure, oxygen and food supply), while noting that latitudinal gradients are fragmented by currents, basin geometry, and the patchiness of continental margins. Companion work emphasising the three-dimensional treatment of marine biodiversity (Owens and Rahbek, 2025) and global data on marine dispersal potential (Álvarez-Noriega et al., 2020) supply context and candidate data layers but no mechanism test.

Problem Statement

Why does Rapoport's rule hold strongly along the depth axis of the global ocean but only weakly along the latitudinal axis, and what mechanisms couple marine species' vertical and horizontal range sizes? The question, posed by Reygondeau and Egorova (2025) as the explicit next step of their global analysis of 21,958 marine species, has three linked components that jointly constitute one scientific objective. (1) Standing of the pattern: establish whether the strong vertical Rapoport relationship (vertical range vs mid-depth, Pearson r ≈ 0.88), the weak latitudinal one (r ≈ -0.02), and their moderate coupling (r ≈ 0.24) are genuine ecological signals rather than artifacts — specifically, (a) whether they generalize to under-sampled invertebrate groups (e.g., bryozoans, ctenophores) and to the under-sampled ocean below the bathyal zone; (b) whether they survive the biases of the heterogeneous data streams (species-distribution-model-derived latitudinal extents versus literature-compiled depth limits, including spatial smoothing and environmental extrapolation in the former); and (c) whether they persist under relaxation of the binning and linear-regression assumptions, allowing the non-linear and scale-dependent structure the source's own figures suggest (widest vertical ranges at intermediate-to-bathyal mid-depths). (2) Mechanism of the coupling and of the '3-D generalists': identify and critically test the ecological and evolutionary drivers that link vertical and horizontal range size — the candidates named in the source are broad physiological tolerance, flexible life histories, dispersive pelagic stages, and habitat continuity — specifying which traits or processes distinguish the small subset of species combining kilometre-scale depth ranges with basin-scale latitudinal extents from the bulk of species restricted in at least one dimension — for this deliverable the solver must state and justify an explicit operational membership rule for the '3-D generalist' subset (e.g. numeric thresholds applied consistently to both vertical range and latitudinal extent) rather than relying on these qualitative descriptors alone — and accounting for the documented heterogeneity of the coupling across habitats (pelagic r ≈ 0.27 vs benthic/demersal r ≈ 0.21) and clades (echinoderms r ≈ 0.42 down to arthropods and annelids r ≈ 0.21). (3) Ecological-versus-evolutionary partition: determine the relative contributions of ecological processes (physiological tolerance, dispersal capacity) and evolutionary history (clade-specific constraints, niche conservatism) to range expansion; the source suggests that integrating finer-scale oceanographic structure, phylogenetic information and explicit measures of environmental tolerance could help disentangle these contributions. The problem is complete when all three components are addressed. (1) The standing of each of the vertical (r ≈ 0.88), latitudinal (r ≈ -0.02) and coupling (r ≈ 0.24) patterns — genuine ecological signal versus artifact of sampling bias, heterogeneous data streams, or binning and linearity assumptions — is settled for the global ocean, and the weak, two-regime latitudinal signal is either explained within the same framework or shown, with quantified evidence, to require axis-specific processes. (2) The drivers of the coupling, including the 3-D generalists, are identified with quantified evidence. (3) The ecological-versus-evolutionary partition of range expansion is resolved. All three components are independently required.

References

  1. Reygondeau, G., & Egorova, Y. (2025). Rapoport's rule in the marine realm: wrong axis, right pattern. bioRxiv preprint (2025-12-31). doi:10.64898/2025.12.30.696717. https://www.biorxiv.org/content/10.64898/2025.12.30.696717v1
  2. Stevens, G. C. (1989). The latitudinal gradient in geographical range: how so many species coexist in the tropics. The American Naturalist, 133(2), 240–256. doi:10.1086/284913. https://doi.org/10.1086/284913
  3. Stevens, G. C. (1992). The elevational gradient in altitudinal range: an extension of Rapoport's latitudinal rule to altitude. The American Naturalist, 140(6), 893–911. doi:10.1086/285447. https://doi.org/10.1086/285447
  4. Stevens, G. (1996). Extending Rapoport's rule to Pacific marine fishes. Journal of Biogeography, 23(2), 149–154. doi:10.1046/j.1365-2699.1996.00977.x. https://doi.org/10.1046/j.1365-2699.1996.00977.x
  5. Gaston, K. J., & Chown, S. L. (1999). Why Rapoport's rule does not generalise. Oikos, 84(2), 309–312. doi:10.2307/3546727. https://doi.org/10.2307/3546727
  6. Smith, K. F., & Gaines, S. D. (2003). Rapoport's bathymetric rule and the latitudinal species diversity gradient for Northeast Pacific fishes and Northwest Atlantic gastropods: evidence against a causal link. Journal of Biogeography, 30(8), 1153–1159. doi:10.1046/j.1365-2699.2003.00898.x. https://doi.org/10.1046/j.1365-2699.2003.00898.x
  7. Fortes, R. R., & Absalão, R. S. (2004). The applicability of Rapoport's rule to the marine molluscs of the Americas. Journal of Biogeography, 31(12), 1909–1916. doi:10.1111/j.1365-2699.2004.01117.x. https://doi.org/10.1111/j.1365-2699.2004.01117.x
  8. Fortes, R. R., & Absalão, R. S. (2010). The latitudinal and bathymetric ranges of marine fishes: a global analysis to test the application of Rapoport's Rule. Marine Ecology, 31(3), 483–493. doi:10.1111/j.1439-0485.2010.00357.x. https://doi.org/10.1111/j.1439-0485.2010.00357.x
  9. Kendall, V. J., & Haedrich, R. L. (2006). Species richness in Atlantic deep-sea fishes assessed in terms of the mid-domain effect and Rapoport's rule. Deep Sea Research Part I: Oceanographic Research Papers, 53(3), 506–515. doi:10.1016/j.dsr.2005.12.005. https://doi.org/10.1016/j.dsr.2005.12.005
  10. Pineda, J. (1993). Boundary effects on the vertical ranges of deep-sea benthic species. Deep Sea Research Part I: Oceanographic Research Papers, 40(11-12), 2179–2192. doi:10.1016/0967-0637(93)90097-M. https://doi.org/10.1016/0967-0637(93)90097-M
  11. Pineda, J., & Caswell, H. (1998). Bathymetric species-diversity patterns and boundary constraints on vertical range distributions. Deep Sea Research Part II: Topical Studies in Oceanography, 45(1–3), 83–101. doi:10.1016/S0967-0645(97)00090-8. https://doi.org/10.1016/S0967-0645(97)00090-8
  12. Moreno, R. A., Rivadeneira, M. M., Hernández, C. E., Sampértegui, S., & Rozbaczylo, N. (2008). Do Rapoport's rule, the mid-domain effect or the source–sink hypotheses predict bathymetric patterns of polychaete richness on the Pacific coast of South America? Global Ecology and Biogeography, 17(3), 415–423. doi:10.1111/j.1466-8238.2007.00372.x. https://doi.org/10.1111/j.1466-8238.2007.00372.x
  13. Pie, M. R., Divieso, R., Caron, F. S., Siqueira, A. C., Barneche, D. R., & Luiz, O. J. (2021). The evolution of latitudinal ranges in reef-associated fishes: heritability, limits and inverse Rapoport's rule. Journal of Biogeography, 48(9), 2121–2132. doi:10.1111/jbi.14150. https://doi.org/10.1111/jbi.14150
  14. Saupe, E. E., Myers, C. E., Peterson, A. T., Soberón, J., Singarayer, J., Valdes, P., & Qiao, H. (2019). Non-random latitudinal gradients in range size and niche breadth predicted by spatial patterns of climate. Global Ecology and Biogeography, 28(7), 928–942. doi:10.1111/geb.12904. https://doi.org/10.1111/geb.12904
  15. Radomski, T. (2025). Revisiting Rapoport's rule: the history of an idea and alternative explanations for latitudinal variation in geographic range size. The Quarterly Review of Biology, 100(2), 63–81. doi:10.1086/736000. https://doi.org/10.1086/736000
  16. Owens, H. L., & Rahbek, C. (2025). Depth matters for marine biodiversity. Journal of Biogeography, 52(12), e70080. doi:10.1111/jbi.70080. https://doi.org/10.1111/jbi.70080
  17. Álvarez-Noriega, M., Burgess, S. C., Byers, J. E., Pringle, J. M., Wares, J. P., & Marshall, D. J. (2020). Global biogeography of marine dispersal potential. Nature Ecology & Evolution, 4(9), 1196–1203. doi:10.1038/s41559-020-1238-y. https://doi.org/10.1038/s41559-020-1238-y
  18. Reygondeau, G., Egorova, Y., Boerder, K., Tittensor, D. P., Kaschner, K., Kesner-Reyes, K., Bailly, N., & Cheung, W. W. L. (2026). AquaX: an enhanced and revised AquaMaps framework to model marine species distributions and biodiversity. PLOS One, 21(2), e0335823. doi:10.1371/journal.pone.0335823. https://doi.org/10.1371/journal.pone.0335823