Impacts and extent of biotic invasions in terrestrial ecosystems

Invasive species have become a priority concern for conservation biologists and land managers. For basic research scientists, they are a tool with which to examine controls over community structure and ecosystem processes. However, the field of invasion biology has been almost as resistant to generalization as the broader field of ecology. As a step toward greater synthesis, the Impacts and Extent of Biotic Invasions in Terrestrial 0169-5347/02/ see front matter 2002 Elsevier Science Ltd. All rights reserved. PII: S0169-5347(02)02454-0 TRENDSin Ecology & Evolution Vol.17 No.5 May 2002 with d2 than with heterozygosity are very limited. So why has fitness been found to be more closely correlated with d2 in so many studies (reviewed in 1,2)? One possible explanation is that important factors other than those considered in Tsitrone et al.s 1 model are involved, although it is unclear what these factors could be. Alternatively, it might be that too much enthusiasm has accompanied the development of d2 and that a more careful examination of the data will lead to a different conclusion. For example, only one or a few components of fitness were correlated with d2 in most studies where fitness was better correlated with d2 than was heterozygosity. Perhaps these few significant results correspond to the 5%false positive expected in statistical testing. Indeed, the significant correlation between d2 and both birth weight and juvenile survival reported previously in the deer population on the isle of Rhum 5 disappeared when a larger number of loci was used (71 instead of nine; 2). Conversely, heterozygosity, which was not significantly correlated with birth weight in the original study, became significantly correlated in the new study. Moreover, Slate and Pemberton 2 show that individual heterozygosity correlates, albeit weakly, across loci, whereas d2 does not, leading them to conclude that heterozygosity is a more robust measure of inbreeding than is d2. The work of Tsitrone et al. 1 is therefore important for several reasons. First, it indicates that one should not use heterozygosity and d2indiscriminately as measures of inbreeding. If inbreeding is expected to have occurred only recently, then heterozygosity should always be preferred to d2. Second, it shows that the conditions under which d2 performs better than does heterozygosity are quite restricted. Third, their work provides hope that it will soon be possible to determine whether inbreeding occurred deep in the pedigree or more recently (close inbreeding). This will probably require developing new measures of inbreeding and modelling the effect of long- and short- term inbreeding on these measures, but the prospect that this will be done is good. Indeed, new measures of inbreeding accounting for the level of variability at each marker locus have been developed recently 6,7 and Tsitrone and colleagues are currently generalizing their modelling approach to multiple loci. References 1 Tsitrone, A. et al. (2001) Heterosis, marker mutational processes and population inbreeding history. Genetics 159, 18451859 2 Slate, J. and Pemberton, J.M. (2002) Comparing molecular measures for detecting inbreeding depression. J. Evol. Biol. 15, 2031 3 Weber, J.L. and Wong, J.C. (1993) Mutation of human short tandem repeats. Hum. Mol. Genet. 2, 11231128 4 Lynch, M. et al. (1995) Mutation accumulation and the extinction of small populations. Am. Nat. 146, 489518 5 Coulson, T.N. et al. (1998) Microsatellites measure inbreeding depression and heterosis in red deer. Proc. R. Soc. London B Biol. Sci. 265, 489495 6 Coltman, D.W. et al. (1999) Parasite-mediated selection against inbred Soay sheep in a free- living, island population. Evolution 53, 12591267 7 Amos, W. et al. (2001) The influence of parental relatedness on reproductive success. Ecosystems workshop was held with a goal to amass and evaluate the current state of data on three topics central to invasion biology.