In a recent comment in this journal, Stuessy and colleagues pointed out a new view on the often detected low genetic variability in island populations (Stuessy et al., 2012). Most biogeographers have explained this genetic feature as a consequence of past founder effects and bottlenecks in the wake of the colonization of the island because only a fraction of the genetic diversity present in the population of origin is introduced to the new habitat (immigration and founder effect, Fig. 1) (Frankham, 1997; Whittaker & Fernández-Palacios, 2007). Subsequent accumulation of genetic diversity due to mutation and recombination may lead to anagenesis and/or cladogenesis and eventually to speciation (speciation box, Fig. 1). In their comment, Stuessy et al. (2012) suggest a new interpretation of low genetic diversity in island populations, arguing that past founder effects are no longer visible. They hold that founder effects and bottlenecks occurred millions of generations ago in most island taxa, which is why their genetic signature is no longer detectable today as more recent processes have veiled past signals. According to their view, the presently low genetic diversity is mostly a result of recent environmental (oceanic island erosion) or anthropogenic habitat deterioration (termed ‘island subsidence, erosion and human impact’ in Fig. 1). While this is a fruitful contribution to island biogeography, the stated scenario should be completed by two additional points.
First, not all island populations/species are old enough to have overcome the initial reduction of genetic diversity in the course of colonization. The authors note that newly established species may still reflect a founder effect. This particularly holds true if generation times are long. While, admittedly, large mammals do only rarely colonize remote islands, large birds of prey with very long generation times are capable of natural colonization of oceanic islands. This is exemplified by, among other species, Galápagos hawks (Buteo galapagoensis), Madagascar fish eagles (Haliaeetus vociferoides) and Egyptian vultures (Neophron percnopterus) on the Canary Islands. The latter colonized the Canaries only about 2500 years ago, and the Galápagos hawks are possibly the latest arrival on the archipelago (probably less than 300,000 years ago) (Bollmer et al., 2006; Agudo et al., 2010). The Cuban oak (Quercus sagraeana), a long-lived plant species with an average generation time of about 100 years, also shows evidence of a mid-Pleistocene island founder effect (Gugger & Cavender-Bares, 2011). All these species harbour less variability than their continental conspecifics or sister taxa (Kretzmann et al., 2003; Bollmer et al., 2006; Johnson et al., 2009), and recent anthropogenic declines may have contributed to the loss of genetic diversity, but it is not at all clear that, particularly in the Egyptian vultures and Galápagos hawks, variability values do not in fact reflect a founder effect. Whether low variability in island taxa is a recent phenomenon or a continuous legacy from the more distant past thus has to be decided based on case-by-case studies, and general conclusions should be drawn only with great caution.
A second point that deserves attention is that the existence (and persistence) in isolation often goes in line with a much stronger impact of environmental stochasticity and, consequently, severe population fluctuations (Melbourne & Hastings, 2008). This greatly reduces long-term effective population sizes (which are particularly sensitive to population fluctuations) and may prevent the accumulation of high genetic diversity through increased inbreeding and drift. It may thus be that island populations, contrary to Stuessy et al. (2012), do not frequently overcome their initially low diversity at all. This is the more likely, the larger the species and the smaller the island. Purging of the genetic load during low population numbers may contribute to even small populations thriving over long periods of time (Frankham et al., 2001), even more so if only a small fraction of deleterious alleles was present in the founding gene pool (Frankham, 1997) (purging in the wake of population fluctuations, Fig. 1). Stuessy et al. (2012) also mention the possibility that founding populations may stay small and genetically poor, and further argue that in such cases no cladogenetic or anagenetic speciation would occur. However, the paradigm of adaptive landscapes holds that particularly small populations overcome the valleys between adaptive peaks and conquer new niches, which is usually accompanied by speciation.
We agree with Stuessy et al. (2012) that the present genetic signature of island species does not necessarily reflect the original founding event, although this cannot be ruled out in many cases (see above). Besides, long-term low effective population sizes as a consequence of demographic fluctuations may be just as important as recent human impacts.