68
T.J. Givnish
1 Introduction
Adaptive radiation - the rise of a diversity of ecological roles and attendant adaptations within a lineage, as exemplified by Darwin's finches (Grant 1986), the African rift lake cichlids (Stiassny and Meyer 1999), and the Hawaiian silversword
alliance (Baldwin 1997) - is one of the most important processes bridging ecology
and evolution, with profound implications for the origin of adaptations and for the
genesis and maintenance of biological diversity (Givnish 1997). Ecological divergence among the members of a radiation is often thought - and in three recent cases
(Schluter 1994; Turner et al. 1996; Rainey and Travisano 1998) has been directly
shown - to reflect selection to avoid competition with close relatives. Adaptive
radiation is often apparent in organisms that have invaded oceanic islands or similarly isolated lakes or mountaintops (e.g., Losos et al. 1998; Albertson et al. 1999).
Few groups are able to colonize such remote areas, eliminating competition from
many lineages, while increasing the importance of competition with close relatives.
Although less easy to recognize, adaptive radiation can also occur on continents
and in oceans - often after a group invades a previously unoccupied adaptive zone,
through the origin of a key innovation or extinction of competing groups (Simpson
1953; Clark and Johnston 1996) - as part of the grand diversification of life on
earth at all taxonomic levels (Foote 1996; Goldblatt et al. 1995; Jernvall et al.
1996; Givnish et al. 1997, 1999; Hapeman and Inoue 1997; Kirsch and Lapointe
1997; Smith and Littlewood 1997; Springer et al. 1997; Bond and Opell 1998;
Johnson et al. 1998; Price et al. 1998; Chiba 1999; Sato et al. 1999).
In this paper, I examine the importance of adaptive radiation and other processes in promoting the remarkable diversification of the Hawaiian lobeliads. Molecular systematics provides an essential tool for evolutionary studies of this and
similar groups: plants that have undergone extensive adaptive radiation often have
diverged so dramatically from putative ancestral groups that it can be difficult to
ascertain relationships based on morphology alone (Givnish et al. 1994, 1995, 1997;
Sang et al. 1994; Baldwin and Robichaux 1995; Bohle et al. 1996; Francisco-Ortega
et al. 1997; Givnish 1998). In the lobeliads, limited dispersal and adaptive radiation appear to have interacted synergistically to generate more species than might
be expected from either process alone, and has helped generate a number of convergent radiations on different islands. Paradoxically, some of the processes that may
accelerate speciation in these tropical plants also may increase their likelihood of
extinction. The roles of limited dispersal and sexual selection in promoting speciation are also discussed in the context of other groups, with a focus on how these
processes might intersect with adaptive radiation. Finally, I consider how interactions with natural enemies and mutualists - long considered to be crucial factors
favoring high plant diversity in tropical forests (Janzen 1970; Connell 1971) - may
also help generate strong ecological gradients in plant species diversity within the
tropics.
T.J. Givnish
1 Introduction
Adaptive radiation - the rise of a diversity of ecological roles and attendant adaptations within a lineage, as exemplified by Darwin's finches (Grant 1986), the African rift lake cichlids (Stiassny and Meyer 1999), and the Hawaiian silversword
alliance (Baldwin 1997) - is one of the most important processes bridging ecology
and evolution, with profound implications for the origin of adaptations and for the
genesis and maintenance of biological diversity (Givnish 1997). Ecological divergence among the members of a radiation is often thought - and in three recent cases
(Schluter 1994; Turner et al. 1996; Rainey and Travisano 1998) has been directly
shown - to reflect selection to avoid competition with close relatives. Adaptive
radiation is often apparent in organisms that have invaded oceanic islands or similarly isolated lakes or mountaintops (e.g., Losos et al. 1998; Albertson et al. 1999).
Few groups are able to colonize such remote areas, eliminating competition from
many lineages, while increasing the importance of competition with close relatives.
Although less easy to recognize, adaptive radiation can also occur on continents
and in oceans - often after a group invades a previously unoccupied adaptive zone,
through the origin of a key innovation or extinction of competing groups (Simpson
1953; Clark and Johnston 1996) - as part of the grand diversification of life on
earth at all taxonomic levels (Foote 1996; Goldblatt et al. 1995; Jernvall et al.
1996; Givnish et al. 1997, 1999; Hapeman and Inoue 1997; Kirsch and Lapointe
1997; Smith and Littlewood 1997; Springer et al. 1997; Bond and Opell 1998;
Johnson et al. 1998; Price et al. 1998; Chiba 1999; Sato et al. 1999).
In this paper, I examine the importance of adaptive radiation and other processes in promoting the remarkable diversification of the Hawaiian lobeliads. Molecular systematics provides an essential tool for evolutionary studies of this and
similar groups: plants that have undergone extensive adaptive radiation often have
diverged so dramatically from putative ancestral groups that it can be difficult to
ascertain relationships based on morphology alone (Givnish et al. 1994, 1995, 1997;
Sang et al. 1994; Baldwin and Robichaux 1995; Bohle et al. 1996; Francisco-Ortega
et al. 1997; Givnish 1998). In the lobeliads, limited dispersal and adaptive radiation appear to have interacted synergistically to generate more species than might
be expected from either process alone, and has helped generate a number of convergent radiations on different islands. Paradoxically, some of the processes that may
accelerate speciation in these tropical plants also may increase their likelihood of
extinction. The roles of limited dispersal and sexual selection in promoting speciation are also discussed in the context of other groups, with a focus on how these
processes might intersect with adaptive radiation. Finally, I consider how interactions with natural enemies and mutualists - long considered to be crucial factors
favoring high plant diversity in tropical forests (Janzen 1970; Connell 1971) - may
also help generate strong ecological gradients in plant species diversity within the
tropics.