Element repertoire: change and development with age in Whitethroat Sylvia communis song

Abstract

Song repertoires are often important determining factors in sexual selection. In several species, older males have larger
repertoires than 1-year-old males. The development of large song repertoires by an individual is, however, poorly understood.
We studied song element repertoire changes in five individual male Whitethroats Sylvia communis sampled as 1- and 2-year olds. These males increased the size of their element repertoire between their first and second
year, but song length and number of different elements per song did not change. On average, 44.3% of the song elements in
the first-year repertoire were also found in the second-year repertoire. Elements shared between years were found earlier
in the songs and tended to occur in sequences. Sequences of shared elements also seemed to be conserved between years. The
study suggests that the song element repertoire of the second year is partly based on the first-year repertoire, which may
explain why large song repertoires are mainly expressed by males at least 2 years of age. It would appear, therefore, that
song element repertoire size could be a reliable signal of male age.

KeywordsAge-Delayed maturation-Song development-Song repertoire-
Sylvia communis
-Whitethroat

Full text

Page 1

1
Element repertoire: change and development with age in
Whitethroat Sylvia communis song


Thorsten J. S. Balsby* and Poul Hansen**

*Behaviour group, Department of Biology, University of Copenhagen,
Universitetsparken 15, DK-2100 Copenhagen Ø, Denmark. E-mail:
tjsbalsby@bio.ku.dk. Phone +45 51862515
**Natural History Museum Århus, Wilhelm Meyers Allé 210, Universitetsparken,
DK-8000 Århus C, Denmark

Page 2

2
Abstract
Song repertoires are often important in sexual selection. In several species older males
have larger repertoires than one-year old males. The development of large song
repertoires within an individual is, however, poorly understood. We studied song
element repertoire changes in five individual male whitethroats Sylvia communis
sampled as one and two years old. Males increased the size of their element repertoire
between first and second year. Song length and number of different elements per song
did not change between years. Song element repertoire size, therefore, could be a
reliable signal of male age. On average 44.3 % of the song elements in the first year
repertoire were also found in the second year repertoire. Elements, shared between
years, were found earlier in the songs and tended to occur in sequences. Sequences of
shared elements also seemed to be conserved between years. The study suggests that
the song element repertoire in the second year is partly based on the first year
repertoire, which may explain why large song repertoires mainly are expressed by
males at least two-year old. Song element repertoire can thus be a reliable indication
of male age.

Keywords: song repertoire, song development, age, delayed maturation, whitethroat,
Sylvia communis.

Page 3

3

Introduction
Several oscine passerine species have very elaborate song repertoires, which may
serve both inter- and intrasexual functions (Catchpole and Slater 2008). Development
of large repertoires has been hypothesized to be costly (Catchpole 1996; Buchanan et
al. 2004), but see Gil and Gahr (2002), and large repertoire size has been shown to
reflect aspects of male quality such as age, condition and parasite (Catchpole and
Slater 2008; Gil et al. 2001; Spencer et al. 2005) or nutritional stress early in life
(Nowicki et al. 1998).
Cross-sectional studies have demonstrated that old males have larger song
repertoires in e.g. sedge warbler Acrocephalus schoenobaenus (Birkhead et al. 1997),
pied flycatcher Ficedula hypoleuca (Lampe and Espmark 1994), great tit Parus major
(McGregor et al. 1981), European starling Sturnus vulgaris (Eens et al. 1992), canary
Serinus canaria (Nottebohm et al. 1986), whitethroat Sylvia communis (Balsby 2000a)
and red-winged blackbird Agelaius phoeniceus (Yasukawa et al. 1980). An age
difference in song repertoire size could arise if older males include new elements in
their repertoire or if males with larger repertoires have higher survival rates (Espmark
and Lampe 1993). However, only in great tits a relation between song repertoire size
and survival rate has been documented (Lambrechts & Dhondt 1986). Longitudinal
studies have shown that individual males increase their repertoire size with age,
especially between first and second year, in pied flycatcher (Espmark and Lampe
1993), willow warbler Phylloscopus trochilus (Gil et al. 2001), starling (Mountjoy
and Lemon 1995), sedge warbler (Nicholson et al. 2007) and great reed warbler
Acrocephalus arundinaceus, although the increase in repertoire size only applied to
one of the two populations of great reed warblers (Forstmeier et al. 2006). These

Page 4

4
longitudinal studies indicate that the repertoires only develop fully in older males in a
number of species.
Longitudinal changes are interesting, because they offer insight to the
development of the song repertoire and may explain why mainly older males display
large repertoires. Although a large repertoire would be advantageous in both inter-
and intra-sexual interactions, the full repertoire is often delayed to older age, even
though the songs in the repertoire could have been learned early in life. One
explanation may be that males can not learn a large repertoire in one season, but need
to build on the repertoire from one or more previous years. This could be tested by
comparing the song element repertoire between years. In this case we should expect to
find that many elements and the most frequently used elements are shared between
years. Shared elements should also occur in identical sequences between years. The
alternative possibility is that older males develop their repertoire independently of the
repertoire of previous years. In that case, elements, shared between years, should not
be used more frequently than non-shared elements and sequences of shared elements
should not be conserved between years.
In this paper we present data on the longitudinal change of song element
repertoires and composition in the whitethroat in order to investigate how males
develop their song repertoire from first to second year. Male whitethroats use their
song for mate attraction and territory defence and have a substantially higher song
activity before they attract a female to their territory (Balsby 2000b). Cross-sectional
studies of whitethroats have shown that one-year old males had smaller song
repertoires than males two-years old or older. The song element repertoire seems to be
a sexually selected trait in whitethroats. Females have been found to show a
preference for males with a large song repertoire (Balsby 2000a), while males did not

Page 5

5
respond to differences in song repertoire size in a playback experiment (Balsby and
Dabelsteen 2001). Given the importance of repertoire size in sexual selection, we
wanted to study the development of the repertoire size by comparing the song element
repertoire in individual males as one- and two-year old

Methods
The birds in this study lived in hedge rows between agricultural fields on the south
side of Fussing Lake, Denmark (5628N, 954E) in May and June 1995, 1996 and
1997. To colour band and age each male, we captured the birds in mist nets by means
of playback. Each male was marked with a unique combination of colour rings and
one metal ring for individual identification. One-year old males could be
distinguished from older males on the basis of iris colour and moulting patterns
(Karlson et al. 1985; Svensson 1992). In total we colour banded 9 one-year old males
and recorded their song. The following year we identified 5 of these males by their
colour bands and recorded their song. We recorded and analysed a total of 7021 songs
from the 5 males. We analysed an average of 813 songs (range 196 to 1162 songs)
from one-year old males and an average of 590 songs (range 440 to 729 songs) from
the same males in their second year. We recorded all males with a Sennheiser MKT
816T directional microphone and a Sony 5000 EMV tape recorder. All recordings
took place between 0500 and 1200 h. Males were only recorded prior to attraction of a
female, as song activity post female attraction is very scarce. For the spectrographic
analysis we used Avisoft-SASLab Pro (www.avisoft.de). We sampled sounds at 22.05
kHz. Spectrograms were made with the hamming window, a time resolution of 5.8 ms
and a bandwidth of 111 Hz.