Fraser and Lyons (2017) showed that present-day Western Hemisphere mammal communities carry a strong signature of biotic interchange — dispersal rates into and out of the tropics are the highest of any inter-zone transition, and tropical and Southern Temperate body-size and phylogenetic structure are poorly explained by climate or competition — but their inference rested on extant species only, a single coarse spatial grain, one preferred null model, and the now-criticized DEC+J biogeographic model, and it could not discriminate among candidate mechanisms for high-latitude body-size over-dispersion. The open problem is to attribute the phylogenetic and body-size structure of Western Hemisphere mammal communities to historical processes (interchange-driven dispersal, differential speciation and extinction) versus contemporary ecological interactions (resource competition, environmental filtering) under defensible inference: with extinct taxa incorporated, across multiple spatial grains, robust to null-model choice, under validated biogeographic models, and with explicit discrimination among limiting similarity, character displacement, and flattening of body-size distributions. Substantial partial progress exists (fossil-inclusive diversification analyses of the Great American Biotic Interchange, late-Quaternary North American fossil community analyses, new co-occurrence methods, and a direct critique of DEC+J), but no published work performs this attribution, so the problem remains open.

Background

Community assembly research asks which processes determine which species co-occur locally. Classical verbal models contrasted species-rich tropical communities, presumed to be structured by species-species interactions, with species-poor temperate communities structured by species-environment interactions. Fraser and Lyons (2017) tested these ideas for Western Hemisphere mammals while adding a third, historical axis: intercontinental dispersal. The pivotal historical event is the Great American Biotic Interchange (GABI), the two-way exchange of land mammals between North and South America following closure of the Isthmus of Panama (about 3 million years ago), which ended South America's long isolation, drove the extinction of endemic South American orders (for example the notoungulates), and left a pronounced asymmetry in which lineages of North American origin diversified more successfully in South America than the reverse.

The study computed, for all mammal species with range maps in 100 km×100 km100\ \text{km}\times100\ \text{km} grid cells across the Western Hemisphere: (i) body mass dispersion (BMdistBM_{\text{dist}}), the mean pairwise difference in log body mass among co-occurring species, where high values (over-dispersion) are classically read as competition pushing coexisting species apart in niche space and low values (clustering) as environmental filtering restricting the viable trait range; (ii) the Net Relatedness Index (NRI), the standardized mean pairwise phylogenetic relatedness of co-occurring species, where phylogenetic evenness (distantly related coexistors) suggests competitive sorting and phylogenetic clustering suggests environmental filtering of closely related, similar species; and (iii) checkerboard (C-score) co-occurrence statistics. Deviations from randomness were measured as standardized effect sizes against null communities built with the independent swap algorithm, a randomization that preserves both the number of species per grid cell and each species' number of occupied cells and has well-behaved error rates. Historical dispersal rates among the tropical and temperate zones were inferred by biogeographical stochastic mapping under the DEC+J model in BioGeoBEARS: DEC (dispersal-extinction-cladogenesis) is a likelihood model of geographic-range evolution along a phylogeny, and '+J' adds founder-event speciation, in which a lineage jumps to establish a range separate from its ancestor.

Three findings defined the agenda. Northern Temperate mammal body masses are over-dispersed relative to null communities, increasingly so with latitude, and correlated with mean annual temperature — read as resource competition plus environmental filtering. Tropical and Southern Temperate body masses are clustered or indistinguishable from random, with climate explaining little, and tropical communities show phylogenetic evenness. Dispersal rates into and out of the tropics are the highest of any zone transition, so dispersal and faunal mixing — the legacy of GABI and earlier exchanges — appear to structure tropical and Southern Temperate communities more than local interactions do.

The authors' own caveats define the open problem. First, the analysis used only extant species, so the contributions of historical speciation and extinction rates to present community structure cannot be accurately estimated from extant-only phylogenies; extinct taxa must be included. Second, competition is expected to act at small spatial scales, so its invisibility at the coarse analysis grain (10,000 km210{,}000\ \text{km}^2 cells) is not evidence of absence; the scale-dependence of the signal is undetermined (the paper's supplementary finer-grid check of BMdistBM_{\text{dist}} found no pattern differences, but the discussion explicitly refuses to infer absence of competition). Third, the two null models tried disagree on whether tropical body mass dispersion is random or under-dispersed, so conclusions depend on null-model choice. Fourth, the inferred dispersal rates depend on the DEC+J model, on randomly resolved polytomies (100 trees, roughly 15 hours of computation), and on an inference framework whose validation remains incomplete. Fifth, high-latitude over-dispersion is consistent with three mechanisms that the data could not separate: competitive exclusion via limiting similarity (species too similar excluding one another), character displacement (in situ divergence of body size where competitors coexist), and passive flattening of the body-size distribution at high latitudes (more even richness across body-size categories, without any competitive sorting). Subsequent methodological work sharpened the fourth caveat: Ree and Sanmartín (2018) showed that DEC+J is conceptually and statistically flawed — cladogenetic range-inheritance events are not modeled as probabilistic with respect to time, inflating their likelihood contribution — making DEC-versus-DEC+J model comparisons inappropriate and any dispersal rates inferred under DEC+J suspect.

Problem Statement

Determine the relative contributions of historical processes — dispersal associated with biotic interchange (GABI and earlier exchanges), and zone-specific speciation and extinction rates — versus contemporary ecological interactions (resource competition, environmental filtering) to the present-day phylogenetic and body-size structure of Western Hemisphere mammal communities, with a defensible inference chain. Specifically, an answer must: (1) incorporate extinct taxa so that the contribution of historical speciation and extinction rates — which cannot be accurately estimated from extant-only phylogenies — to present-day phylogenetic community structure and body-mass dispersion can be quantified; (2) establish the scale-dependence of competition's signal by testing community assembly at multiple spatial grains, given that competition is expected to operate at scales finer than the original 100 km×100 km100\ \text{km}\times100\ \text{km} cells; (3) establish whether the characterization of tropical body-mass dispersion (random versus under-dispersed) is robust to null-model choice, using formal sensitivity analysis over occupancy- and richness-constrained randomizations; (4) re-estimate dispersal rates among the tropical and temperate zones under a validated biogeographic model (DEC+J being conceptually and statistically flawed), with quantified uncertainty over phylogenetic resolution and model choice; and (5) discriminate among the candidate mechanisms for high-latitude body-size over-dispersion — competitive exclusion via limiting similarity, character displacement, and flattening of body-size distributions — which make overlapping predictions for dispersion metrics taken alone. These components are the joint robustness conditions of a single attribution: each addresses a specific way the original inference could fail, and all five must be addressed for the attribution to be complete; together they constitute the one overarching answer about what structures Western Hemisphere mammal communities.

References

  1. Fraser, D. & Lyons, S. K. (2017). Biotic interchange has structured Western Hemisphere mammal communities. Global Ecology and Biogeography 26(12): 1408–1422. doi:10.1111/geb.12667, https://doi.org/10.1111/geb.12667
  2. Carrillo, J. D., Faurby, S., Silvestro, D., Zizka, A., Jaramillo, C., Bacon, C. D. & Antonelli, A. (2020). Disproportionate extinction of South American mammals drove the asymmetry of the Great American Biotic Interchange. Proceedings of the National Academy of Sciences 117(42): 26281–26287. doi:10.1073/pnas.2009397117, https://doi.org/10.1073/pnas.2009397117
  3. Pineda-Muñoz, S., Jukar, A. M., Tóth, A. B., Fraser, D., Du, A., Barr, W. A., Amatangelo, K. L., Balk, M. A., Behrensmeyer, A. K., Blois, J., Davis, M., Eronen, J. T., Gotelli, N. J., Looy, C., Miller, J. H., Shupinski, A. B., Soul, L. C., Villaseñor, A., Wing, S. & Lyons, S. K. (2021). Body mass-related changes in mammal community assembly patterns during the late Quaternary of North America. Ecography 44(1): 56–66 (published online 2020). doi:10.1111/ecog.05027, https://doi.org/10.1111/ecog.05027
  4. Fraser, D., Villaseñor, A., Tóth, A. B., Balk, M. A., Eronen, J. T., Barr, W. A., Behrensmeyer, A. K., Davis, M., Du, A., Faith, J. T., Graves, G. R., Gotelli, N. J., Jukar, A. M., Looy, C. V., McGill, B. J., Miller, J. H., Pineda-Muñoz, S., Potts, R., Shupinski, A. B., Soul, L. C. & Lyons, S. K. (2022). Late Quaternary biotic homogenization of North American mammalian faunas. Nature Communications 13: 3940. doi:10.1038/s41467-022-31595-8, https://doi.org/10.1038/s41467-022-31595-8
  5. Fraser, D. & Lyons, S. K. (2020). Mammal community structure through the Paleocene-Eocene Thermal Maximum. The American Naturalist 196(3): 271–290. doi:10.1086/709819, https://doi.org/10.1086/709819
  6. Whittingham, M. A. J. B., Korasidis, V. A. & Fraser, D. (2024). Functional stasis and changing habitat preferences among mammalian communities from the PETM of the Bighorn Basin, Wyoming. Cambridge Prisms: Extinction 2: e20. doi:10.1017/ext.2024.25, https://doi.org/10.1017/ext.2024.25
  7. Gearty, W., Uricchio, L. H. & Lyons, S. K. (2024). Investigating the biotic and abiotic drivers of body size disparity in communities of non-volant terrestrial mammals. Global Ecology and Biogeography 33(12): e13913. doi:10.1111/geb.13913, https://doi.org/10.1111/geb.13913
  8. Ree, R. H. & Sanmartín, I. (2018). Conceptual and statistical problems with the DEC+J model of founder-event speciation and its comparison with DEC via model selection. Journal of Biogeography 45(4): 741–749. doi:10.1111/jbi.13173, https://doi.org/10.1111/jbi.13173
  9. Botta-Dukát, Z. (2018). Cautionary note on calculating standardized effect size (SES) in randomization test. Community Ecology 19(1): 77–83. doi:10.1556/168.2018.19.1.8, https://doi.org/10.1556/168.2018.19.1.8
  10. Tóth, A. B., Alroy, J., Lyons, S. K. & Allen, A. P. (2025). Effects of phylogenetic distance, niche overlap and habitat alteration on spatial co-occurrence patterns in Neotropical bats and birds. Proceedings of the Royal Society B: Biological Sciences 292(2051): 20241679. doi:10.1098/rspb.2024.1679, https://doi.org/10.1098/rspb.2024.1679
  11. Novella-Fernandez, R., Chalmandrier, L., Brandl, R., Pinkert, S., Zeuss, D. & Hof, C. (2024). Trait overdispersion in dragonflies reveals the role and drivers of competition in community assembly across space and season. Ecography 2024(4): e06918 (published online 2023). doi:10.1111/ecog.06918, https://doi.org/10.1111/ecog.06918
  12. Rodríguez-Tricot, L. & Arim, M. (2020). From Hutchinsonian ratios to spatial scaling theory: the interplay among limiting similarity, body size and landscape structure. Ecography 43(2): 318–327 (published online 2019). doi:10.1111/ecog.04462, https://doi.org/10.1111/ecog.04462
  13. Swiston, S. K. & Landis, M. J. (2025). Testing relationships between multiple regional features and biogeographic processes of speciation, extinction, and dispersal. Systematic Biology 74(2): 282–300 (published online 2024). doi:10.1093/sysbio/syae062, https://doi.org/10.1093/sysbio/syae062
  14. Freitas-Oliveira, R., Lima-Ribeiro, M. S., Mendoza-Rodriguez, V. H. & Terribile, L. C. (2025). Reviewing the Great American Biotic Interchange: climate change as a trigger for biodiversity dispersal. Proceedings of the Royal Society B: Biological Sciences 292(2058): 20251745. doi:10.1098/rspb.2025.1745, https://doi.org/10.1098/rspb.2025.1745
  15. Croft, D. A. (2026). The Great American Biotic Interchange. Current Biology 36(14): R783–R784. doi:10.1016/j.cub.2026.05.049, https://doi.org/10.1016/j.cub.2026.05.049