In the last 20 years intensive studies on the ecology and mating system of the Great Reed Warbler Acrocephalus arundinaceus have been conducted at five different localities in Japan and Europe. Although all the populations showed similar territorial polygyny, the studies showed partly different results and conclusions concerning the evolution of polygyny in this species. Here, we compare ecology and behaviour of the warblers from different populations based on published information, focusing on aspects of territorial polygyny inherent in the polygyny-threshold-model. Breeding results in one of the European populations were unique in that secondary females were far less successful than monogamous females compared to almost equivalent success in other four populations. Although causes of offspring mortality were not the same between populations, the European populations commonly showed higher rates of chick starvation in secondary nests than in monogamous nests, but this was not the case for the Japanese populations. Benefits which compensate for these costs of polygyny were found in three of the five populations. Despite variations in breeding results, all studies showed the existence of a potential cost of polygyny: reduced paternal care in secondary nests leading to a realized cost depending on climatic and ecological conditions. Evidences of female choice based on behavioural observations were available at some populations. In one of the populations the benefit of polygyny was linked to variance in territory quality caused by habitat heterogeneity and habitat partitioning pattern between males. But information on these aspects of ecology has not been collected enough for the other populations. Features of territory differed greatly between populations and it suggests the possibility that different mechanisms work to cause polygyny in different populations. Despite these regional differences, however, the polygyny-threshold-model is generally applicable to this species.

View full abstract

A comparison between various populations of the Great Reed Warbler Acrocephalus arundinaceus was made using published and unpublished data. The comparison showed that most differences are consistent with a division into subspecies. The majority of birds from the nominate subspecies go through a complete moult in late autumn in north Africa, and then continue migration to winter quarters south of the equator. The majority of the eastern subspecies moult completely just after the breeding season. In contrast the orientalis birds spent extremely short periods on migration and long periods in the winter quarters. Orientalis also show more habitat tolerance (breeding in bushes, small trees, and no water), higher breeding density, less nestling starvation, and more Cuckoo Cuculus canorus brood-parasitism. However, some of these features may be due to the fact that some of orientalis populations live in extremely changed (man-made) habitats. On the other hand, clutch-size, total nest-losses, eggs hatchability, fledglings production, and nestling diet composition showed interpopulation variation independent of the subspecies division.

View full abstract

An isolated population of Swedish Great Reed Warblers was studied between 1984 and 1993 when almost every breeding adult and nestling were ringed. I examined between year variation in several variables that may affect fintness. In particular I looked at the reproductive success of females that can expect no or little male assistance (secondary females) in relation to females that are assisted by their males (monogamous and primary females). Return rates of females, the frequency of total losses and number of fledlings per female showed significant differences between years. The yearly mean number of fledglings per female was associated with the mean annual temperature during breeding. When an exceptionally warm year was excluded this was significant for both primary and secondary females. The success of secondary females relative to the success of primary females increased with increasing temperature implying that the cost of polygyny is higher in cold than in warm summers. The variation in clutch size and relative reproductive success of secondary and primary females increased over the ten year study period suggesting either that the study period was too short for capturing the variation of the variables or that these were subject to a systematic change over the study period. There were indications that the condition for breeding has deteriorated in the last years; clutch size decreased significantly over the study period and particularly low values of fledging mass and relative reproductive success were found in some of the most recent years. This might reflect that population density had reached a level when effects of intraspecific competition influences breeding results. Alternatively, five extremely mild winters may be responsible for a drastic increase of the lake's population of Crucian carps. Whether a large fish population could have caused a decline in invertebrate food for Great Reed Warblers in recent years is discussed.

View full abstract

A reduction of paternal feeding to the secondary brood is regarded as a potential cost of polygynous breeding for female Great Reed Warblers. To examine the effect of a cool climate on breeding activities, I studied a population in Aomori Prefecture, northern Japan, in two seasons. The mean temperature was lower in 1989 than in 1990, and females delayed mating with already mated males in 1989. The food delivery rate of polygynous males to their secondary broods was significantly lower than that of monogamous males, suggesting that nestlings of the secondary brood are fed insufficiently during cool weather when females devote more time for brooding. However, the nestlings of the secondary broods of polygyny did not suffer very cool weather, because they hatched later in the season after the cold spell had ended. As a result, the reduction in food delivery rate by parents in the secondary brood was not significant and no adverse effects were detected either in the incidence of starvation or in growth of the nestlings in the secondary broods of polygyny. Delayed settling of females as the secondary mate of polygyny in the year of cooler weather might have been effective in avoiding nestling starvation.

View full abstract

We examined reproductive success of monogamous, primary and secondary females as well as factors influencing male mating status in a polygynous Great Reed Warbler population by analysing long term data. The analysis is based on 428 nests of which at least one parent was individually colour-ringed. The number of males studied over a 15 year period averaged 30/yr. of which 11.3% were polygynous and 13.9% unmated. Primary broods of polygynists started 10 days earlier than broods of secondary females. Average fledging success of secondary females was lower than that of monogamous females (relative success 0.79) but did not differ significantly from that of simultaneously monogamous females (relative success 0.85). The percentage of fledglings recruited to the breeding population overall was highest in primary broods, but did not differ significantly between monogamous and secondary broods. Mating status of males was influenced by age, with a higher proportion of older males becoming polygynous whereas correlations between age and mating status of females was not significant. We also investigated the relative importance of male and territory quality in female choice by analysing 3 territorial and 11 male characteristics (physical traits, age, repertoire size, aggressiveness, hormones) of 29 males (4 polygynous, 3 unmated, 22 monogamous) using a discriminant analysis. The best predictors for polygyny were a long reed edge against open water, lower aggressiveness, and a higher song repertoire size. We conclude (1) that it is not necessary to invoke female deception, which had been favoured in explaining polygyny in our population (2) that ecological conditions for polygyny are suboptimal in our study area but there seem to exist enough predictable differences in breeding situations to allow female choice and (3) that females base their mate choice on both male characteristics and territory quality, and that these factors are positively correlated.

View full abstract

I analysed demographic parameters and lifetime reproductive success in a Swedish population of Great Reed Warblers. The population was followed on a daily basis in 1985-1993 when all breeding Great Reed Warblers were individually marked and >95% of the young fledged in the study area were ringed. Between-year breeding site fidelity of adults was relatively high both for males (55%) and females (51%), and it seemed not to be affected by previous breeding results. Natal site fidelity was also relatively high as 14% of the fledglings hatched in the study area returned to breed in subsequent years. Each year, however, about 50% of the breeders were hatched outside the study area (i. e. immigrants). The production of lifetime recruits was biased so that 17% of the males and 13% of the females produced more than 50% of the recruits in the population. There was no difference in lifetime reproductive success between native birds and immigrants, suggesting that immigrants contributed substantially to the local gene pool. Despite this, effects of non-incestuous inbreeding, measured as lowered hatchability of eggs among broods of genetically more similar pair members, were found in the study population. This may be a result of the dispersal pattern between local Great Reed Warbler populations in Sweden and also a consequence of the short history of these small populations.

View full abstract

This paper reviews the mating systems of Acrocephalus warblers, and their probable evolution. Polygamy is seen as a gradual departure from the usual avian mating system of monogamy and biparental care. The main constraints are imposed by the environment, but due to high productivity the environmental potential for polygamy occurs at marshland ecotones. In such conditions, polygyny, polyandry and promiscuity have been found to occur in varying degrees. DNA fingerprinting is now clarifying the distinction between social mating systems and realised genetic reproductive success. Morphological adaptations, such as bill size and testes size, can also be correlated with different mating systems. There is also a clear correlation between mating systems, the resulting sexual selection pressure and the structure and function of song. Monogamous species produce long, complex songs for mate attraction, and polygamous species produce shorter, simpler songs for territorial defence. Polygamous species also have separate longer songs for mate attraction, and shorter songs for territorial defence and mate guarding.

View full abstract