(v) Tests of the theory of island biogeography
The many papers that have been written as tests of island biogeography theory can be described as one of two types:
(1) Observational (or mensural) or (2) Experimental.
1. Observational tests are characterized by survey-type studies where species diversity is estimated by faunal counts along with factors thought to drive island biogeographic processes (island size, isolation, etc) and statistical models are fit to the data to see if the potential factors actually do, statistically, "explain" the variation in species diversity.
A good example is the study by Dennis & Shreeve (1997) on British butterflies. The figure below shows the numerous islands surrounding England/Scotland/Wales and Ireland. Clearly they vary in size (Hebridies/Shetlands are big) and degree of distance from the "continental" source fanuas on the larger islands of Ireland and Great Britain. They counted the numbers of species on each island and measured its size and distance from the source faunas. They also determined the numbers of species at the faunal source closest to each island. They used regression techniques to see if the predictions of island biogeography theory were upheld (i.e. larger, closer islands should have more species, etc).
Results:
Island area had little (non-significant effect)
Isolation only "explained" 20% of the island variation in species diversity (significant)
Faunal source richness explained from 40-60% of the variation in island butterfly diversity (significant)
The small effect of island area was likely due to the small degree of variation in island size (most very pretty small) which could not overcome the "noise" in the data to demonstrate an effect.
They also measured a series of ecological factors that might influence the potential for specific species to colonize certain island and perhaps explain inter-island variation in species diversity: incidence of various species at faunal source, geographical range of each species, dispersal distance within range, hostplant range, duration of flight period before mating, votinism (number of mating periods per season). These variables were combined into a "multivariate ecological index" that explained about 76% of the variation in species diversity on islands (P < 0.001).
This example illustrates the observational kind of study and also a potential weakness of the classic theory of island biogeography – many other factors other than immigration and extinction rates per se or their association with just island area and isolation (or when these effects are small) can influence species diversity on islands.
See Manne et al. (1998) for another example of the observation/statistical approach to testing island biogeographic theory (e.g., do empirical observations of immigration rate and extinction rate on English Islands follow the predicted curves, or another pattern such as a linear decline or increase, respecitively?)
2. Experimental tests are really those studies where the islands/faunas are manipulated in some way to test for the hypothesized effect. Observation studies, of course, suffer from the potentially serious limitation that the factors investigated are not under the strict control of the researcher – some unknown and unmeasured factor may be influencing the outcome of an analysis and any effect may not be simply due to the measured variables.
Experimental tests manipulate the hypothesized factor (say island size) and test for an effect in the predicted direction. All other factors remaining equal, any observed effect in the predicted direction clearly supports the role of the hypothesized (and manipulated) factor in generating the effect.
A classic, and perhaps the best, example is the "experimental defaunation" experiment of Simberloff and Wilson (1970). These authors studied small islets of red mangrove in the Florida Keys. The islets (75-250 sq m in area) were surveyed for their terrestrial arthropod communities and then covered in plastic tents and fumigated with a selective poison that removed all the arthropods. Simberloff and Wilson then followed the recolonization of the islets over a two year period to see if the patterns of species diversity followed expections according to island biogeography theory. The islets varied in distance from the mainland source fauna (300 species) from 20 – 1200 m.
Results: the basic results are shown in the figure below.
Recolonization was quite rapid, all but island 1 reached the "pre-defaunation species diversity" within about 150 days.
Islet four (o) was still below pre-defaunation diversity levels 2 yr after the manipulation and it was the most isolated islet. Note it also has the lowest pre-manipulation species diversity.
Many repeated extinctions and recolonizations were recorded after the species diversity had levelled-off. Species were still "turning over" on the islets.
Despite the numbers of species remaining fairly constant and returning to pre-manipulation levels, only an average of about 30% of the original species re-established themselves on the islets – there had been tremendous turnover.
The experiment of Simberloff and Wilson generally supported the predictions of the theory of island biogeography, but the small spatial and statistical scale (few islands surveyed) of the experiment highlight the limited experimental support for the theory.
(vi) Fate of the theory
Despite the intuitive appeal of the theory of island biogeography and the impact it had on encouraging an experimental approach to zoogeography, in its classic form the theory has a number of limitations. Some of these limitations have been accounted for in slight modifications of the theory.
Very few convincing experimental tests of the theory have been completed and those that have usually do not explore the underlying processes (i.e. they don’t study immigration or extinction explicitly)
It fails to recognize within island divergence as an important factor generating variation in diversity. Isolated and large islands may have high degrees of endemism, speciation on archipelagos like the Galapagos (birds) and Caribbean (reptiles) islands. Note also the difference in species-area relationship between the African (up to 1-2 million years old) Rift Lakes and temperate North American lakes (< 15,000 years old) in section (ii) above.
Immigration and extinction may not be independent as the theory implies. Continued immigration from source faunas may "rescue" taxa from extinction on islands. The rescue effect will clearly be dependent on distance (therefore, isolation influence extinction as well as immigration).
The target effect. Large islands may be bigger "targets" for immigration – therefore, island size not only effects extinction rate, but also immigration.
Historical effects on island diversity. For instance, island age may influence diversity (see Borges 1999). Older islands may have had greater potential for within-island divergences to influence diversity, may be closer to equilibrium. Also, historical interconnectedness may influence diversity patterns. Trinidad was once connected to South America by a land bridge (now submerged). If extinctions are still ongoing the island may not yet have "relaxed" to a species diversity level consistent with its now island status (it has more species than its size or isolation alone would suggest because of its historical connection to a larger landmass). Remember that it is an equilibrium theory, but some habitats may not yet have reached an equilibrium after a historical disturbance (see Torres and Snelling 1997).
In sum, the equilibrium theory of island biogeography is currently regarded as a very useful, if incomplete, paradigm under which to begin investigations of species diversity in island landscapes. It is certainly regarded by most as providing a very useful stimulation for encouraging experimental studies of ecological factors and, more recently, their interaction with historical factors in explaining variation in species diversity. See Powledge (2003) for discussion of the lasting legacy of McArthur and Wilson's fundamental contribution.
References:
Borges, P.A.V. 1999. Effect of island geological age on the arthropod species richness of Azorean pastures. Biol. J. Linn. Soc. 66: 373-410.
Brook, B.W., N.S. Sodhi, and P.K.L. Ng. 2003. Catastrophic extinctions follow deforestation in Singapore. Nature 424: 420-423.
Bush, M.B. & R.J. Whittaker. 1991. Krakatau: colonization patterns and hierarchies. J. Biogeog. 18: 341-356
Dennis, R.L.H. & T.G. Shreeve. 1997. Diversity of butterflies of British islands: ecological influences underlying the roles of area, isolation and the size of the faunal source. Biol. J. Linn. Soc. 60: 257-275.
Gilbert, F.S. 1980. The equilibrium theory of island biogeography: fact or fiction? J. Biogeog. 7: 209-387.
Gurd, D.B. & T.D. Nudds. 1999. Insular biogeography of mammals in Canadian parks: a reanalysis. J. Biogeog. 26: 973-982.
Lomolino, M.V. 2000. A call for a new paradigm of island biogeography. Global Ecology and Biogeography. 9: 1-16.
MacArthur, R.H. & E.O. Wilson. 1963. An equilibrium theory of island biogeography. Evolution 17: 373-387.
MacArthur, R.H & E.O. Wilson. 1963. The theory of island biogeography. Princeton University Press, Princeton, Mass.
Manne, L.L., S.L. Pimm, J.M. Diamond & T.M. Reed. 1998. The form of the curves – a direct evaluation of MacArthur and Wilson’s classic theory. J. Animal Ecol. 67: 784-794.
Powledge, F. 2003. Island biogeography's lasting impact. BioScience 53: 1032-1038.
Quammen, David. 1996. The Song of the Dodo. Island Biogeography in the Age of Extinctions. Scribner, New York. A VERY readable and fascinating discussion of evolution and ecology of island biotas and extinction. Nice chapter on Wallace with refreshing (and candid!) insights into Darwin.
Ricklefs, R.E. & I.J. Lovette. 1999. The roles of island area per se and habitat diversity in the species-area relationships of four Lesser Antillean faunal groups. J. Animal Ecol. 68: 1142-1166.
Simberloff, D.S. & E.O. Wilson. 1970. Experimental colonization of islands: a two year record of colonization. Ecology 51: 934-937.
Simberloff, D.S. 1986. Design of nature reserves. Chap. 14, In: M.B. Usher (ed.). Wildlife conservation evaluation. Chapman and Hall, London. Pp. 315-337.
Torres, J.A. and Snelling, R.R. 1997. Biogeography of Puerto Rican ants: a non-equilibrium case? Biodiversity and Conservation 6: 1103-1121.