News & Views | Published:


Density and diversity

Naturevolume 417pages698699 (2002) | Download Citation


One explanation for the especially rich diversity of trees in the tropics is that a process called 'density-dependent mortality' operates there. It turns out, however, that this process occurs in temperate forests too.

The latitudinal gradient in species diversity is arguably the most universal pattern in global biodiversity: the lower the latitude, the higher the number of species in a given area. This pattern, with biodiversity peaking in the tropics (Fig. 1), is found in most taxonomic groups and may be as old as life itself1,2,3. For plants, one explanation centres on a phenomenon called 'density-dependent mortality', in which the survival rates of species decrease as they become more common, leaving space for rarer species. It has been suggested4,5 that density-dependent mortality is more intense at lower latitudes, so, at least in part, accounting for the gradient in diversity. As they describe on page 732 of this issue, however, Hille Ris Lambers and colleagues6 have tested this hypothesis and found it wanting.

Figure 1
Figure 1


Trees in the tropics — the height of diversity.

It is not surprising that many researchers have sought to find an explanation for the latitudinal pattern in species diversity. Biologists study diversity at different scales and it is becoming clear that scale is an important bridging element between the various theories that have been developed for regional and local levels. As far as trees are concerned, local processes such as the rate and extent at which gaps for seedling colonization occur, as well as density-dependent mortality, may limit the extent to which one species excludes another, and thus promote local diversity. Density-dependent mortality, however, will maintain high diversity only if it is species specific — that is, if it decreases the density of a species as a function of the density of that species alone, rather than of the density of all species.

If species-specific, density-dependent mortality contributes to the latitudinal gradient in species diversity, the effect should be stronger in the tropics than in temperate areas. The rationale for this is that much seed and seedling destruction stems from attack by insects and fungi, which reach higher densities in tropical regions because the tropics do not experience seasonal variations and tend to remain hot and humid4,5.

Hille Ris Lambers et al.6, however, show that density-dependent mortality acts in temperate as well as in tropical forests. Their results come from a forest in North Carolina, where they found that six out of seven tree species experience density-dependent mortality at one or more transitions in their early stages: from seed to seedbank; from seed or seedbank to seedling; and in seedling survival (the seedbank phase is a latent period in which seeds lie dormant in the soil before germination). The results clearly show not just the effect of general density-dependent mortality, but also species-specific, density-dependent mortality. This suggests that rather than the driving factors being resource competition for light or nutrients alone, predators or pathogens are responsible. Hille Ris Lambers et al. also compared previous studies carried out in temperate and tropical areas, in which several species were tested for density-dependent mortality, and conclude that the proportion of species affected is not greater in the tropics.

All in all, this study6 adds support to those who claim that general ecological mechanisms, such as the creation of gaps in vegetation that allow colonization by seed-lings, and density-dependent regulation, operate similarly in the tropics and temperate zones. So theories invoking these processes fail to explain the higher diversity in the tropics compared with temperate zones.

Still, further investigations are required. As Hille Ris Lambers et al. point out, many different protocols have been used in this kind of research, producing varying and sometimes confusing results. By re-analysing their own data with the approaches used in other studies, the authors show that these approaches have invariably underestimated the effects of density dependence. They propose a framework for simultaneously assessing the impact of seedling and adult density on seed and seedling fate. This framework should now be applied in future work at various latitudes.

The authors' main message, then, is that the proportion of species subject to density-dependent regulation is not lower in temperate areas than in the tropics. But the ecologically more significant question may be to what extent species are actually regulated by the process. Are the effects strong or weak? And is there a latitudinal component? Addressing these questions will require identifying and quantifying latitudinal trends in the relevant factors. Earlier work demonstrated that density-dependent patterns of seed loss tend to be associated with predation by insects rather than by vertebrates, and that if spatial and temporal availability of seeds is predictable it can have an effect on seed loss7. So which predators determined the outcomes of the studies by Hille Ris Lambers et al.? And is there a latitudinal difference in the relative prevalence of insect and vertebrate predation, and of fungal attack? There is likely to be a difference in the fruiting time of particular plant communities, and in the occurrence of irregular heavy fruiting that swamps consumers and means a higher proportion of seeds escapes predation.

If indeed the small-scale processes observed in temperate and tropical areas are much the same, how can the latitudinal gradient in biodiversity be explained? Diversity is influenced by regional as well as local processes2,3,8, all operating at specific temporal scales (Fig. 2). At the large spatial scale at which the gradient is evident, the answer must, at least partly, lie in the balance between speciation and extinction. A further question then centres on the relative importance of local processes in this balance. It is obvious that there are several mechanisms that can contribute to the latitudinal gradient1,2,3,4,8. There is unlikely to be a simple answer, and any explanation probably consists of a mix of local, regional, global and historical factors.

Figure 2: Determinants of biodiversity.
Figure 2

Impacts on diversity occur at both regional and local scales8,9: here, green arrows indicate processes that increase diversity (species are added) and red arrows those that decrease it (species are lost). Regionally, diversity is mainly influenced by speciation and extinction and, to a certain extent, by immigration. These processes determine the number of regional species from which communities can 'draw' particular species. Of course not all species can occur everywhere, either because the habitat is unsuitable (environmental filters) or because of constraints on their movement (dispersal limitation). Locally, processes such as interspecies competition, random extinction and predation affect diversity. Predation can have a double effect. It can remove species. But in the case of density-dependent mortality of seeds and seedlings discussed by Hille Ris Lambers et al.6, it can increase diversity if it favours rarer species by selectively reducing the population density of the most common species.


  1. 1

    Gaston, K. J. Nature 504, 220–227 (2000).

  2. 2

    Rosenzweig, M. L. Species Diversity in Space and Time (Cambridge Univ. Press, 1995).

  3. 3

    Huston, M. Biological Diversity: The Coexistence of Species on Changing Landscapes (Cambridge Univ. Press, 1994).

  4. 4

    Givnish, T. J. J. Ecol. 87, 193–210 (1999).

  5. 5

    Harms, K. E. et al. Nature 404, 493–495 (2000).

  6. 6

    Hille Ris Lambers, J., Clark, J. S. & Beckage, B. Nature 417, 732–735 (2002).

  7. 7

    Hammond, D. S. & Brown, V. K. in Dynamics of Tropical Communities (eds Newbery, D. N., Prins, H. H. T. & Brown, V. K.) 51–78 (Blackwell Science, London, 1998).

  8. 8

    Ricklefs, R. E. & Schluter, D. Species Diversity in Ecological Communities (Univ. Chicago Press, 1993).

  9. 9

    Hubbell, S. P. The Unified Theory of Biodiversity and Biogeography (Princeton Univ. Press, 2001).

Download references

Author information


  1. the International Institute for Geoinformation Science and Earth Observation (ITC), PO Box 6, Enschede, NL-7500 AA, The Netherlands

    • Hans ter Steege
  2. the National Herbarium of the Netherlands, Utrecht University Branch, W. C. van Unnik Building, Heidelberglaan 2, Utrecht, 3584 CS, The Netherlands

    • Roderick Zagt


  1. Search for Hans ter Steege in:

  2. Search for Roderick Zagt in:

About this article

Publication history

Issue Date


Further reading


By submitting a comment you agree to abide by our Terms and Community Guidelines. If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate.

Newsletter Get the most important science stories of the day, free in your inbox. Sign up for Nature Briefing