SEASONAL PLASTICITY IN SONGBIRDS

Such seasonal plasticity in the adult brain has been highly studied in songbirds. Each year, during early stages of seasonal reproductive development, the volume of certain areas of the brain associated with song learning and production in various songbird increases (sometimes dramatically), along with the number, size and spacing of neurons (see review by Tramontin & Brenowitz, Reference Tramontin and Brenowitz2000 and references there in). Stimulated primarily by increasing levels of testosterone in the blood (possibly through the direct action of one or more of its metabolites; see Tramontin et al., Reference Tramontin, Wingfield and Brenowitz2003), but also affected by lengthening days, thyroid hormones and other factors, this occurs several weeks before the actual onset of breeding, at the same time as (or even a little before) gonadal recrudescence (i.e. the redevelopment of the testes) and the reappearance of secondary sexual characteristics (see Tramontin & Brenowitz, Reference Tramontin and Brenowitz2000; Tramontin et al., Reference Tramontin, Perfito, Wingfield and Brenowitz2001; Brenowitz & Lent, Reference Brenowitz and Lent2002).

These changes do not affect the memory of previously learned songs. Song sparrows (Melospiza melodia), a species whose ability to learn new songs is age-limited (i.e. they do not learn to produce new songs as adults), obviously retain the songs learned when young, despite the seasonal reduction in size of the neurological song control system. Similarly, the same seasonal changes in the song control system do not necessarily affect the ability of song sparrows to make difficult song discriminations—although it may in other species (Reeves et al., Reference Reeves, Beecher and Brenowitz2003). This would allow them to discriminate and respond appropriately to all other con-specific and environmental sounds equally well throughout the year.

SEASONAL PLASTICITY IN HUMPBACK WHALES

We propose that similar changes occur seasonally within the brains of humpback whales: certain parts of the humpback whale brain increase in size (or ‘hypertrophy’) as breeding season approaches, allowing and/or promoting song-related behaviour, before again shrinking in size once breeding is over. Indeed, Piersma & Lindstrom (Reference Piersma and Lindstrom1997) noted that the performance of a seasonally predictable behavioural task is often preceded by hypertrophy of the organs or tissues that are required for that task in many species. Furthermore, there is some evidence that, as in birds, humpback whale testis change in size seasonally, being larger on the breeding grounds (e.g. Chittleborough, Reference Chittleborough1955; Symons & Weston, Reference Symons and Weston1958). This is in line with changes observed in the size of testis of other baleen whales (e.g. grey whale, Eschrichtius robustus, Rice & Wolman, Reference Rice and Wolman1971; minke whale, Balaenoptera sp., Best, Reference Best1982; but see discussion of B. acutorostrata subsp. and B. bonaerensis in Rice, Reference Rice1998; Reilly et al., Reference Reilly, Bannister, Best, Brown, Brownell, Butterworth, Clapham, Cooke, Donovan, Urbán and Zerbini2008a, Reference Reilly, Bannister, Best, Brown, Brownell, Butterworth, Clapham, Cooke, Donovan, Urbán and Zerbinib). It has also been suggested that such seasonal changes in testis size will occur in any baleen whale species that exhibit seasonal reproductive cycles (Brownell & Ralls, Reference Brownell, Ralls and Donovan1986). Additional weight is thus added to the empirical evidence, as humpback whales tend to migrate annually, as briefly mentioned above (see review by Clapham, Reference Clapham1996, for additional details; but also see Mikhalev, Reference Mikhalev1997). We therefore submit that this hypothesis addresses many of the outstanding issues related to humpback whale song behaviour, as we shall now demonstrate.

Firstly, it would be reasonable to assume that, as in birds, the reduction in size of the neurological song control system in humpback whales during the non-breeding season would decrease the frequency of singing, perhaps to the point that no singing occurs at all. The advantage of this to the whale would be a reduction in metabolic costs, as the maintenance of any hypertrophied organ systems or tissues is thought to be energetically expensive, thus probably making such hypertrophy common to many vertebrates that live in variable environments (Jacobs, Reference Jacobs1996; Piersma & Lindstrom, Reference Piersma and Lindstrom1997).

These increased metabolic costs may also add weight to the idea that the song is an honest signal of reproductive fitness (as discussed by Parsons et al., Reference Parsons, Wright and Gore2008). The metabolic costs of rebuilding and maintaining the song control system each year would be in addition to the handicap conferred by the song production itself, indicating fitness to a potential mate or competitor.

Although this hypothesis does not explain why humpback whale song changes continuously throughout the season (or why others adopt the new song) it may explain why it does not change between breeding seasons (e.g. Payne et al., Reference Payne, Tyack, Payne and Payne1983). The ability to learn new song elements would not be required during the non-breeding season and thus the area of the brain responsible for song change could also be included in the song control system reduced during this period.

Similarly, if the neurological song control system in humpback whales increases in size prior to the onset of migration in a similar manner to the neurological changes seen in birds, then singing males could be expected on the feeding grounds just prior to their migration to the breeding grounds, as well as perhaps also immediately upon their return. In fact, there is increasing evidence that this may indeed be the case (e.g. Clark & Clapham, Reference Clark and Clapham2004; Vu et al., Reference Vu, Risch, Clark, Gaylord, Hatch, Merrick, Thompson, Tougaard, Wiley and Van Parijs2009). This may also offer a mechanism for transfer of song components from one breeding population to another, as foraging populations may contain individuals from different breeding populations, and vice versa.

Finally, neurological changes may also explain why songs can be detected, on occasion, out-of-season (e.g. Charif et al., Reference Charif, Clapham and Clark2001). The males of various bird species produce songs more frequently (and with less song-to-song variability) during the breeding season, but do not cease entirely during the non-breeding season (see Tramontin & Brenowitz, Reference Tramontin and Brenowitz2000). It may be that it is normal for certain elements to remain in the humpback repertoire out-of-season in a similar way, although they might be largely indefinable as song components. Alternatively (and perhaps more likely), it might be possible for certain disease-related conditions to circumvent the inhibition of song through the reduction in size of the song control centre. For example, the seasonal development of the song control system in song sparrows appears to occur at sub-maximal levels of circulating testosterone (Tramontin et al., Reference Tramontin, Perfito, Wingfield and Brenowitz2001) and can be artificially induced (to some extent) in Gambel's white-crowned sparrows (Zonotrichia leucophrys gambelii) through the experimental application of testosterone (Brenowitz & Lent, Reference Brenowitz and Lent2002). This suggests that physiological conditions resulting in unusually high levels of testosterone might be able to induce at least some parts of the song control system to hypertrophy, possibly resulting in out-of-season and/or female song production.