The islands' story: Genomics and adaptive radiation on remote islands

inhospitable matrix3. Typically, we imagine an oceanic island surrounded by a matrix of seawater; however, from the perspective of particular organism, an isolated rock, tree, mountaintop (‘sky‐islands’), cave system, river or lake may represent an island depending on its degree of isolation; which in turn depends upon the ability of the organism to disperse through or across that matrix. Although isolation is fundamental, the history of an island – how and when it came to be – critically deter‐ mines the composition of the biota and the nature of the evolutionary and ecological processes that occur (for a detailed review see reference 3). Truly remote islands, for example volcanic oceanic islands, are typically formed de novo and are initially devoid of life (in contrast with fragment islands that are formed when an area of habitat becomes separated from a previously contiguous habitat and consequently have their ecological niche space already largely filled at the time of isolation). On these Darwinian islands (so called because Darwin promoted this, at that time controver‐ sial, mechanism1), the number of species is initially zero and increases over time. On highly isolated islands, the number of colonizing species that can reach the island is extremely small and their rate of immigration is much lower than the rate at which new species can form. Multi‐ ple neo‐endemics may form through adaptive radiation. The outer limit for the dispersal ability of an organism is referred to as its radiation zone and one result of the differential abilities of species to disperse is that the fauna and flora of islands tends to be disharmonious – statisti‐ cally non‐representative of the mainland or source. This disharmony increases with distance from the source, as fewer and fewer lineages are able to colonize – a proc‐ ess called attenuation. Consequently, many higher taxa tend to be poorly represented on remote islands. The Hawaiian archipelago is extremely isolated (4000 km from the North America and 3200 km from the nearest island group) and exhibits extreme disharmony. Only 50% of insect orders and 15% of the known families are So wrote Charles Darwin in On the Origin of Species (1859)1. The “very close relation of the distinct species” that he observed on numerous occasions, most famously the Galapagos finches, is of course the product of adap‐ tive radiation, a term attributed to the American palae‐ ontologist H.F. Osborn2. An adaptive radiation can be defined as the rapid diversification of a lineage into mul‐ tiple ecologically different species. In Darwin’s spirited observation, he encapsulated the four criteria that define an adaptive radiation: that the species share a common ancestry i. that there is a correlation between species’ phenotype ii. and their environment that the species’ phenotype includes traits that are iii. useful (adaptive) in their environment that speciation has occurred rapidly iv. It is likely that episodes of adaptive radiation have gen‐ erated much of the diversity of life; however, most of the best‐known examples come from remote islands, archipelagos or similar settings. From Darwin and Wal‐ lace on, the concepts and theories of adaptive radiation and much of evolutionary biology have been developed through studies on islands. Indeed, the parlance of pop‐ ulation genetics is pregnant with the reasoning invoked by islands – we talk of ‘island models’ and sample organ‐ isms from ‘populations’ which are then often treated as ‘islands’ and their degree of genetic divergence tested. Both real and imagined islands are amenable to study because of their discreteness. In this short article, we briefly define what we mean by remote islands, why they promote adaptive radiation and illustrate how molecular genetics and genomics are transforming our understand‐ ing of the adaptive process using examples from our own work on spiders and other key examples.