An integrati e framework for understanding the origin and e olution of a ian migration

In a recent article in the Journal of A ian Biology’s ‘‘Forum’’ section, entitled, ‘‘Towards a framework for understanding the evolution of migration,’’ R. M. Zink (2002) has stimulated discussion about necessary future directions to take in approaching this subject. Zink’s article addresses the pertinence, or lack thereof, of studies other than phylogenetics to the understanding of migrant evolution and origin. He states, ‘‘A phylogenetic tree is required to make inferences about the evolutionary history of migration.’’ (emphasis added). His initiative is timely and potentially promising, and we herein subject the theme to further consideration based on data concerning migration biology. Before doing so, we need to clarify an apparent confusion over the term origin. In the sense in which it is used by migration theorists, Zink included, the word ‘‘origin’’ clearly has two meanings, i.e., 1) the (first) appearance of migration in evolutionary history, and 2) appearance of migration in a particular lineage of migrants. The first meaning involves ‘‘deep history.’’ We consider it now largely consensus that ‘‘original’’ migration in the first sense occurred early in organismic development, which is why it is found in such disparate groups as insects, fish, reptiles, mammals, and most orders of birds (Baker 1978), and that there is extensive evidence to indicate that most, if not all, birds possess the basic mechanistic repertoire required for migration (Dingle 1996, Berthold 1999, Prum 2002, Helbig 2003, Rappole 2003, Rappole and Jones 2003). In contrast, ‘‘origin’’ in the second sense refers to a different time scale. Migration, as we see it, is a flexible system. At least since the Miocene, fluctuations in seasonality and recurrent glacial cycles have shaped migration systems repeatedly, and modifications of migratory behavior continue to occur, even on observable time scales (Sutherland 1998, Fiedler 2003, Helbig 2003, Rappole 2003, Rappole and Jones 2003). Thus, ‘‘origin’’ in the second sense often refers to relatively recent events in a given lineage. Clarification of the different meanings of the term ‘‘origin’’ is important since Zink’s (2002) discomfort with ecological approaches to understanding the origin and evolution of avian migration appears to rest in part on a fundamental misunderstanding of the two senses in which the term ‘‘origin’’ is used. For instance, he criticizes the discussion of evolution of migration in Rappole (1995), as an inadequate and flawed attempt to address the origin of migration in the first sense. However, the stated purpose of Rappole’s discussion was to address the origin of migration in the second sense; specifically to present arguments for reconstructing the evolution of migration in birds that breed in temperate North America and migrate to the Neotropics. Understanding the evolution of migratory behavior in given lineages of birds entails reference ’’to its modification and maintenance,’’ as Zink (2002) suggests in his example of fox sparrows Passerella iliaca. We take this view as a common starting point for explorations into the evolutionary course of migratory behavior, and suggest that cooperative efforts by ornithologists from different fields are not only valuable, but necessary to understand the evolution of complex behaviors such as migration (Alerstam and Hedenström 1998). The framework Zink (2002) suggests involves phylogenetic analyses that consider the various components of migratory behavior separately. There is no doubt that migration, or the ‘‘migratory syndrome’’ (Dingle 1996), consists of multiple components that merit individual consideration. We suggest, however, that there are problems with Zink’s approach. Firstly, extensive evidence indicates that migratory traits are parts of an