Content uploaded by Barbara Helm
Author content
All content in this area was uploaded by Barbara Helm on Jul 22, 2019
Content may be subject to copyright.
ORIGINAL ARTICLE
Barbara Helm ÆEberhard Gwinner
Carry-over effects of day length during spring migration
Received: 15 March 2005 / Revised: 23 July 2005 / Accepted: 24 July 2005 / Published online: 7 September 2005
Dt. Ornithologen-Gesellschaft e.V. 2005
Abstract The day lengths to which migratory birds are
exposed depend on the timing and course of their jour-
ney. While winter day length is known to influence
vernal events, it is not clear if birds also use day length
during the spring migration as a temporal cue. We ad-
dressed this question by exposing captive stonechats
(Saxicola torquata) to two different photoperiodic sim-
ulations of spring migration routes, following common
winter conditions. One group experienced day lengths of
the regular (‘‘fast’’) migration, and the other group, a
‘‘slow’’, or more southerly originating, route. The
resulting small, temporary differences in day length had
lasting effects on the birds. The groups differed in
migratory restlessness during and following exposure to
different day lengths. ‘‘Slow’’ migrants continued noc-
turnal activity longer than ‘‘fast’’ migrants. Further-
more, all activities of the ensuing breeding season were
delayed in the ‘‘slow’’ migrants, indicating a phase shift
in their underlying annual rhythm. ‘‘Slow’’ migrants
delayed terminating their reproductive stage by
regressing testes and the cloacal protuberance later than
the ‘‘fast’’ migrants. Molt started and ended later in
‘‘slow’’ migrants, but the duration of the molt was
unaffected by spring day length. Finally, ‘‘fast’’ migrants
resumed nightly restlessness earlier than ‘‘slow’’ mi-
grants in late summer. These results demonstrate that
Zugunruhe (migratory restlessness) and reproductive
windows are not set exclusively during winter but can be
modified by day length cues during the spring migration.
Because migration modifies the day length exposure of
birds, migration routes can have carry-over effects on
the timing of breeding season events, including the
completion of molt and initiation of autumnal nocturnal
activity.
Keywords Migration ÆPhotoperiod ÆMolt ÆBreeding Æ
Stonechat ÆCarry-over
Introduction
Photoperiod, i.e., the seasonally changing light fraction
of the day, is thought to provide a reliable calendar for
birds to anticipate upcoming seasonal change and pre-
pare for future activities on time. Since the seminal work
of Rowan (1926), the effects of increasing late-winter
day length on spring activities have been extensively
studied. Photostimulation (exposure to longer days) has
been found to induce migratory restlessness
(Zugunruhe) and preparations for breeding in many
species (Farner and Gwinner 1980; Gwinner 1988; Hahn
et al. 1997; Dawson et al. 2001). In addition, vernal
photostimulation has been suggested to function inde-
pendently as a ‘‘remote timer’’ for the initiation of au-
tumn events (King 1963; Farner et al. 1980; Moore et al.
1982; Berthold 2002).
Photostimulatory effects are reportedly tailored to
the particular living conditions of species and even
populations. For example, the photoperiodic thresholds
timing the reproduction of resident great and blue tits
(Parus major,P. caeruleus) show local differentiation,
even over relatively short distances (Silverin et al. 1993;
Lambrechts et al. 1996). Silverin et al. (1993) observed
that the threshold day length for rapid gonadal growth
preceded egg laying by approximately 6 weeks in great
tits. In contrast to resident populations, migratory birds
cannot directly act on local photoperiodic thresholds
prior to breeding. They must anticipate the upcoming
reproductive season before the return migration, while
still located in winter quarters with often very different
photoperiodic conditions (Gwinner 1987,1988,1990;
Coppack et al. 2003; Gwinner and Helm 2003). Captive
Communicated by F. Bairlein
B. Helm (&)ÆE. Gwinner
Max-Planck Institute for Ornithology,
von-der-Tann-Str. 7, 82346 Andechs, Germany
E-mail: helm@orn.mpg.de
Tel.: +49-8152-373114
Fax: +49-8152-373133
J Ornithol (2005) 146: 348–354
DOI 10.1007/s10336-005-0009-5
migrants use a combination of circannual clocks and
winter photoperiod to time vernal events. When exposed
to different day lengths, migratory birds showed mean-
ingful timing responses, at least to a typical range of
winter conditions. Simulated photoperiods affected the
subsequent timing of Zugunruhe, prenuptial molt, and
testicular development (Gwinner 1987,1988; Coppack
et al. 2003). These studies mimicked photoperiod during
the fall, winter, and spring, as encountered by migrant
birds wintering at different latitudes. However, the re-
sults of these experiments do not clarify whether the
birds used the timing cues exclusively during winter or
also during the spring migration. If day length en route
further modifies the timing of future activities, then
migratory decisions, including the speed of migration,
could directly affect the ‘‘window’’ for breeding (i.e., the
time during which reproduction is physiologically pos-
sible; Gwinner 1999). This intriguing possibility can be
tested in captive birds, if, after exposure to common
winter photoperiods, spring day length is modified dif-
ferently in different groups of birds.
We thus exposed European stonechats (Saxicola
torquata rubicola) to a photoperiod of a typical winter
latitude of the species (Fig. 1a). During early vernal
migration we divided the birds into two groups. One
group was exposed to the rapidly increasing spring day
length that would be encountered by stonechats
migrating directly to their breeding areas. The other
group experienced an initially slower increase in vernal
day length that would be typically experienced by
migrants wintering further south and migrating later
or advancing more slowly. The birds of this latter
group were exposed to slightly shorter day lengths
during the spring period, i.e., March–May. To test the
hypothesis of a ‘‘remote timer’’, we monitored sea-
sonal activities for carry-over effects until the birds
completed molt.
Methods
Birds
From March through September 2004 we monitored
activity, molt, and reproductive cycles of 16 male ston-
echats. All birds had been bred in our institute and were
derived from an Austrian population (4814¢N, 1622¢E;
Gwinner et al. 1987). The birds were kept in individual
registration cages and maintained on a standard insec-
tivorous diet. Room temperature was kept constant at
21C, and light intensities at 300 lx (daytime) and 0.01 lx
(at night). Over winter all birds were kept under the day
length experienced in the winter quarters of European
stonechats at 40N. When spring Zugunruhe started, the
birds were divided into two groups (see below). We
carefully minimized additional variation: in each group,
one-half of the birds were first-year males, the other half
were 2- or 3-year-old males. Of the eight males in each
group, seven had a full sibling in the other group. Thus,
it is safe to assume that group differences in seasonal
behavior were largely a consequence of the differential
experimental treatment.
Day length regimes
All stonechats were wintered at a simulated latitude of
40N. Following initiation of migratory restlessness, the
experimental groups were exposed to different day
lengths from the 5th of March (Fig. 1a) until they were
simulated to reach their breeding range at 47.5N.
Migratory latitude was reconstructed from passage
times according to literature and ringing recoveries
(Glutz von Blotzheim and Bauer 1988; Helm 2003).
The corresponding experimental day length was derived
from the period between the onset and end of civil
twilight at the respective latitude. One group of birds
was exposed to day lengths of the natural migration
route rapidly followed by arrival at the breeding
grounds on March 12 [‘‘fast’’ (40N) route]. The other
group was exposed to day lengths of a ‘‘slow’’, or more
southerly originating, route, as experienced by ston-
echats departing from 25N (Helm and Gwinner 2001).
For the ‘‘slow’’ route migrants, day length during the
migration period was therefore shorter than for the
‘‘fast’’ route migrants. The rate of vernal increase in
day length was initially lower for ‘‘slow’’ route migrants
but later increased faster than that for ‘‘fast’’ route
migrants as a result of the combined effects of latitu-
dinal migration and increasingly longer days at north-
ern latitudes after the spring equinox. Photoperiodic
differences between the groups rose from 3 min to a
maximum of 42 min in early April and then decreased
again. Along the ‘‘slow’’ (25N) route, migrants
reached photoperiodic conditions of the breeding
grounds on May 7. Both groups remained thereafter
under day lengths of the normal breeding range of
47.5N.
Nocturnal activity
Activity was recorded using passive infrared sensors
(range: 12 m/40; Intellisense XJ-413T; CK Systems,
Ann Arbor, Mich.). In contrast to microswitches,
infrared detectors record not only perch hopping but all
of the spatio-temporal changes that the birds make. We
recorded the number of movements per 2-min-long
interval and then pooled five consecutive intervals. Be-
cause infrared detectors are highly sensitive, a threshold
was applied to minimize white noise, thereby omitting
activity counts below 20 per 10-min-long interval. We
thus scored a 10-min interval as ‘‘active’’ if 20 or more
movements were registered. When summing up the
number of active 10-min intervals per night, we dis-
carded one transitional 10-min interval in the morning
and evening, respectively; the remaining nocturnal
349
periods of restlessness were used as a measure of
migratory activity. Because nocturnal activity was sup-
pressed during the nights after gonadal measurements,
we replaced the data of the three nights following lap-
arotomy by the means of the 2 days preceding and
2 days following the temporary suppression.
Fig. 1 Effects of day length
during the spring migration
period on subsequent life-cycle
events of stonechats (Saxicola
torquata rubicola).
aPhotoperiodic simulations.
Dotted lines indicate day length
experienced by all birds. After
wintering at 40N, birds were
exposed to the day length of
either the regular ‘‘fast’’ route
(black line) or the ‘‘slow’’
migration route (gray line), as
experienced when departing
from 25N. From May
onwards, the birds were kept on
a breeding day length that
would be encountered at
47.5N. bNocturnal activity
and molt. Nocturnal
restlessness (dots) is given by
group means of the number of
active 10-min intervals per 24 h.
Molt is shown by two pairs of
gray bars in which the upper
pair indicates the primary molt
and the lower pair, the body
molt. Median onset and
completion ± standard error
(SE) are indicated by upper
(primary molt) and lower (body
molt) triangles.cSizes of left
testis and basal width of cloacal
protuberance (inlay). Data are
group medians ± SE. In all
cases, black symbols indicate the
‘‘fast’’ route migration and gray
symbols, the ‘‘slow’’ route
migration
350
Reproductive windows
Birds were laparotomized every 3–5 weeks to measure
the width of the left testis, usually to the nearest 0.1 mm
(Gwinner 1975). Permits for measurements were issued
by the animal welfare protocol of the state of Upper
Bavaria. To evaluate non-invasive alternatives to lapa-
rotomy, we also measured the basal width of the cloacal
protuberance (CP), which enlarges during the breeding
season due to sperm storage in the seminal glomus
(Baumel 1993). This measurement had previously been
found to be significantly correlated with testis diameter
in pilot studies on stonechats (B. Helm and E. Verme-
irssen, unpublished data).
Molt
All birds were checked for molt at weekly intervals.
Flight feather molt was recorded over primaries one to
nine following the protocol of Newton (1966). Its onset
was defined as the checking date if no more than one
feather had been dropped. If molt had proceeded
further, it was defined to have started on the mid-date
between the last check without molt and the current
date. Body molt was monitored in 19 plumage parts
(Helm and Gwinner 1999,2001). It was considered to
have begun on the control date if the birds had started
molt in less than five plumage parts. If molt had oc-
curred in more feather tracts, its onset was defined as the
mid-date between the last check without molt and the
current control date. Molt completion was defined cor-
respondingly.
Statistics
Activity data were available on a daily basis and there-
fore suitable for time-series analyses of group differences
(Helm 2003). We calculated the mean difference between
experimental groups for each night of the experiment,
which gave us a series of nightly differences. Group
differences deviated from white noise (Diggle 1990) and
indicated phase-shifting of the birds’ schedules. We
determined the timing of the phase shift by a simple edge
detector filter function (ED; W. Zucchini, personal
communication; Helm 2003). For each nightly activity
value xat time (t
0
), the five preceding and consecutive
values were added with reversed signs:
EDðxðt0ÞÞ ¼ SUM ðxðt05Þxðt01ÞÞ
þSUM xðt0þ1Þxðt0þ5ÞÞ:
The resulting series of computed ED values had a
clearly defined maximum, indicating the phase shift, i.e.,
the onset of the maximal increase in group differences.
Additional statistical techniques included mixed-
model analyses of repeated measures by restricted
maximum likelihood (REML) and non-parametric
testing of non-normally distributed data. To avoid in-
flated testing, we compared activity data by months.
Results
Nocturnal activity
Patterns of nocturnal restlessness are shown in Fig. 1b.
At the beginning of the experiment in early March
stonechats had just started Zugunruhe, and initial
migratory restlessness did not differ between experi-
mental groups (Mann-Whitney U-test of daily means in
March: U=206.5; p=0.408; n=22 days). However,
differences did develop rapidly and subsequently devi-
ated highly significantly from white noise (Box-Ljung
test: r=0.772; Box-Ljung statistics=124.45; p£0.000;
n=206 days). The date identified as the onset of phase
shift between the two groups was March 22, some
2 weeks into the experiment, at a time when the pho-
toperiod experienced by the two groups differed by
21 min (Fig. 1a, b).
Stonechats exposed to day length of their natural
(‘‘fast’’) migration route showed a decreased Zugunruhe
earlier than those experiencing a ‘‘slow’’, or more
southerly, route. ‘‘Fast’’ route stonechats ceased show-
ing this restlessness in late May, whereas ‘‘slow’’ route
birds continued to be active at night until mid-June.
From April to June, ‘‘slow’’ route migrants showed
significantly higher activity than ‘‘fast’’ route migrants
(April: U=78.0; p<0.001; n=30 days; May: U=85.0;
p<0.001; n=31 days; June: U=180.0; p<0.001;
n=30 days). Surprisingly, phase differences between the
groups reappeared when nocturnal restlessness was re-
sumed in the summer and fall. From July until the end
of September, ‘‘fast’’ route migrants showed signifi-
cantly more nightly activity than ‘‘slow’’ route migrants
(July: U=68.5; p<0.001; n=31 days; August: U=0.00;
p<0.001; n=31 days; September: U=0.00; p<0.001;
n=30 days).
Reproductive windows
Figure 1c shows the testicular sizes of birds before and
during the experiment. Testes in both groups had
reached approximately one-half of their reproductive
size in late February when Zugunruhe started. The
groups did not differ in testicular development prior to
the experiment, except during early March when
‘‘fast’’ route migrants had slightly larger testes
(U=0.12.5; p=0.038; n=16). Photoperiodic manipu-
lations began when the birds’ testes were almost
completely developed. Subsequent to this, the size of
the testes was significantly affected by the choice of
simulated migration route, but group differences were
highly dependent on date (mixed-model analysis,
group differences: Wald statistics=7.43 with 1 df;
351
p=0.006; interaction between group and date: Wald
statistics=30.20 with 4 df;p<0.001). Testes were fully
grown in both groups in April as well as early May.
Concurrent with the phase shift in nocturnal activity,
‘‘slow’’ route migrants regressed their testes later than
‘‘fast’’ route migrants (late May: U=16.5; p=0.105;
June: U=7.5; p=0.007; July: U=14.0; p=0.065;
n=16). As a result, the period of enlarged gonads was
lengthened for ‘‘slow’’ route migrants. The width of
the CP showed similar patterns (Fig. 1c, inlay). The
CPs of the two groups of birds were initially identical,
but they later differed, again dependent on the date
(group: Wald statistic=4.48 with 1 df;p=0.034;
interaction group and time: Wald statistic=5.10 with
4df;p<0.001). The CPs of both groups were of full
size, indicating massive sperm storage, until late May.
In June, ‘‘fast’’ route migrants had significantly de-
creased CPs compared to ‘‘slow’’ route migrants
(U=9.0; p=0.015; n=16), while in July differences in
CP were not significant (U=26.5; p=0.374; n=16).
Sizes of CP and testes were highly correlated (Pear-
son’s r=0.75; p£0.000; Spearman’s rho=0.643;
p£0.000; n=104).
Molt
The timing of post-nuptial molt was consistently affected
by simulated spring migration day length (Fig. 1b).
Primary and body molt were initiated later in ‘‘slow’’
route migrants than in ‘‘fast’’ route migrants, thereby
corroborating phase shifting towards later seasonal
activities. The median ± standard error (SE) for the
primary molt for ‘‘fast’’ route migrants was June 7±2.3;
for the ‘‘slow’’ route migrants, it was June 16±3.7
(U=7.5; p=0.007). For the body molt, the median ±
SE was June 24±3.1 for the ‘‘fast route migrants and
July 14±2.5 for the ’’slow‘‘ route migrants (U=0.5;
p<0.001; n=16 for both molts). Both molts were also
completed later, but differences were significant only for
body molt (primary ‘‘fast’’: September 4±6.3; ‘‘slow’’:
September 17±5.4; U=21.5; p=0.279; body molt
‘‘fast’’: September 22±4.6; ‘‘slow’’: October 13±5.9;
U=13.0; p=0.05; n=16). Despite differences in the
timing of the molts, the duration of the primary and
body molts did not differ between the two groups
(p>0.7).
Discussion
Subtle modifications of the day length during vernal
migration had consistent effects on captive stonechats:
simulation of a slower, or more southerly originating,
migration delayed all subsequent life-cycle stages at the
breeding grounds. ‘‘Slow’’ route migrants terminated
spring Zugunruhe at a later date than ‘‘fast’’ route mi-
grants. They also regressed their testes and CP at later
dates, implying a lengthening of the reproductive win-
dow following ‘‘slow’’ route migration.
Furthermore, ‘‘slow’’ route migrants molted later and
resumed nocturnal restlessness later in the summer. The
consistent seasonal delays strongly suggest that the day
lengths experienced by the ‘‘slow’’ route migrants caused
a phase-shift of all breeding-ground activities. These
results have implications for the timing of migration and
for our understanding of avian circannual rhythms.
Both the timing and the course of the vernal migra-
tion, due to changes in latitude, influence the day length
which birds experience in the spring. Our data show
that, in turn, day length during spring migration is
‘‘read’’ by the birds and utilized by them to time ensuing
life-cycle stages, including the termination of Zu-
gunruhe. Day length during the spring migration thus
provides a feedback system in which photoperiodic
exposure is both modified by migration and influences
the migratory program. Via day length, the timing and
course of migration can thus have carry-over effects on
ensuing schedules. By means of this mechanism, late-
arriving birds can extend their reproductive window and
potentially still produce offspring. Furthermore, because
the phase shifts induced in our experimental groups
persisted until the fall, the amount of time at the
breeding grounds (e.g., to complete molt) may be
unaffected by arrival date. This said, a cautionary note
should be added in that the patterns observed on captive
birds could be altered in the field. For example, birds
that fail in their breeding attempt may molt and depart
earlier than otherwise (Jenni and Winkler 1994). Data
from captive birds indicate underlying migratory pro-
grams, but the correct interpretation of Zugunruhe,
especially towards the end of vernal migration, is still
unresolved (Gwinner and Czeschlik 1978; Gwinner
1990). Conversely, our data caution against explaining
the fast progress of the spring migration, as opposed to
the fall migration, mostly by the longer days available
for feeding (Bauchinger and Klaassen 2005). While the
length of the day (light) available for fueling may indeed
contribute considerably to the speed of migration, it also
directly influences migratory programs, as shown here.
The relative contributions of photoperiodically set
endogenous programs and environmental factors, such
as food availability, remain to be clarified.
Phase shifting during spring migration has implica-
tions for our understanding of avian annual clocks.
Captive birds can use winter day length to adjust
departure and breeding preparations according to lati-
tudinal distance from the breeding grounds (Gwinner
1987, 1988,1990; Coppack et al. 2003; Gwinner and
Helm 2003). Previous studies combined the photoperi-
odic effects of winter and spring and, consequently, the
question of whether day length during migration modi-
fies summer schedules was left unanswered. Our data
demonstrate that the vernal day length has clear and
lasting effects on seasonal timing. This observation
supports the hypothesis that vernal photostimulation
acts as a ‘‘remote timer’’ for initiating events occurring
in the fall (King 1963; Farner and Gwinner 1980; Moore
et al. 1982; Hahn et al. 1997; Dawson et al. 2001;
352
Berthold 2002). This hypothesis has been based on
experimental data which indicate that subsequent to
photostimulation, birds need a set amount of time to
complete their breeding cycles and resume autumnal
activities (e.g., King 1963; Farner et al. 1980; Moore
et al. 1982; Berthold 2002). However, the evidence from
these experiments was mixed and often derived from
birds transferred to a constant day length (Farner and
Gwinner 1980; Farner et al. 1980) which, therefore, re-
ceived no further cues to readjust their schedules. The
results of our study revive the idea of remote timing
because the effects of the spring photoperiod on the
birds lasted until fall under synchronizing, natural day
length. The persistent phase shift shows that the circ-
annual clock was reset in spring and thereafter not until
late September. Our experiment ended before we could
determine when the clocks were eventually reset. The
estimated timing of the vernal phase-shift, March 22,
coincided almost exactly with the vernal Equinox.
Equinoxes are primary candidates for resetting annual
clocks for several reasons. First, around the Equinoxes,
day length is globally uniform and hence gives date
independent of location. Second, during the Equinoxes
daily changes in day length are maximal so that photo-
periodic information is most pronounced. Third, Euro-
pean stonechats spend the time between Equinoxes at
their breeding grounds. Resetting the circannual clock at
the Equinoxes would allow the most precise ‘‘gauging’’
to minimize photoperiodic misinterpretation related to
migration. Extrapolating from the evidence we have
presented here for the spring migration, we consider it
possible that circannual systems may rely on at least two
periods of clock-setting, one close to the spring and fall
Equinox, respectively. This idea remains to be tested by
long-term studies of phase-shifting effects on birds.
Zusammenfassung
Einfluss der Tagesla
¨nge wa
¨hrend des Fru
¨hjahrszugs auf
Brutzeitraum, Mauser, und Beginn der Herbstzugunruhe
Zugvo
¨gel beeinflussen durch ihre Zugroute und Zugzeit
die Tagesla
¨nge, in der sie sich befinden. Wa
¨hrend
experimentell belegt ist, dass die Wintertagesla
¨nge das
saisonale Verhalten im Fru
¨hling beeinflusst, ist bisher
unbekannt, ob Vo
¨gel auch wa
¨hrend der Fru
¨hjahrszugzeit
photoperiodische Zeitinformationen nutzen. Um diese
Frage zu beantworten, haben wir Schwarzkehlchen
(Saxicola torquata) unter den Lichtbedingungen von zwei
verschiedenen Zugrouten untersucht. Die Vo
¨gel hatten
unter einheitlichen Bedingungen u
¨berwintert. An-
schließend erlebte eine Gruppe die Lichtbedingungen
ihres regula
¨ren (‘‘schnellen‘‘) Zugweges, die andere die
eines ‘‘langsameren‘‘, oder weiter su
¨dlich beginnenden
Zugweges. Die damit verbundenen geringen Unterschie-
de in der Tagesla
¨nge zeigten nachhaltige Wirkung. Vo
¨gel
der beiden Gruppen unterschieden sich in Zugunruhe
wa
¨hrend und nachdem sie unterschiedliche Tagesla
¨ngen
erlebten. ‘‘Langsame’’ Zieher setzten ihre na
¨chtliche
Aktivita
¨tla
¨nger fort als ‘‘schnelle‘‘ Zieher. Daru
¨ber hin-
aus verzo
¨gerten sich bei den ‘‘langsamen’’ Ziehern alle
Aktivita
¨ten der anschließenden Brutsaison. Dies deutet
auf eine Phasenverschiebung der zugrunde liegenden
Jahresrhythmen hin. ‘‘Langsame‘‘ Zieher verzo
¨gerten
den Abschluss ihrer Brutphase, indem sie Hoden und
Kloakenprotuberanz spa
¨ter zuru
¨ckentwickelten als
‘‘schnelle’’ Zieher. Die Mauser begann und endete spa
¨ter
als bei ‘‘schnellen‘‘ Ziehern, dauerte aber gleich lang an.
Schließlich nahmen ‘‘schnelle’’ Zieher die na
¨chtliche
Unruhe fru
¨her wieder auf als ‘‘langsame‘‘ Zieher. Diese
Ergebnisse zeigen, dass der Brutzeitraum von Zugvo
¨geln
nicht allein im Winter, sondern auch wa
¨hrend des Fru
¨h-
jahrszugs von der Tagesla
¨nge beeinflusst wird. Somit
ko
¨nnen sich Zugzeit und Zugroute auf das anschließende
zeitliche Verhalten bis zu Mauserende und Beginn der
Herbstzugunruhe auswirken.
Acknowledgements We thank our colleagues at the institute, espe-
cially Lisa Trost and Willi Jensen, for their thorough collection of
the data and surveillance of the experimental setup. Walter Zuc-
chini gave valuable advice on time-series analyses, and Bill Brad-
shaw contributed by an inspiring discussion of clock mechanisms.
Ian Newton has been a huge help by providing rich information,
critical reading, thoughtful discussions and, in general, great
supportiveness. The experiment complied with current laws in
Germany.
References
Bauchinger U, Klaassen M (2005) Longer days in spring than in
autumn accelerate migration speed of passerine birds. J Avian
Biol 36:3–5
Baumel J (ed) (1993) Handbook of avian anatomy: nomina anat-
omica avium, 2nd edn. Nuttall Ornithological Club, Cambridge
Berthold P (2002) Control of bird migration, 2nd edn. Chapman &
Hall, London
Coppack T, Pulido F, Czisch M, Auer DP, Berthold P (2003)
Photoperiodic response may facilitate adaptation to climatic
change in long-distance migratory birds. Proc R Soc London
Ser B 270[Suppl 1]:S43–S46
Dawson A, King V, Bentley G, Ball G (2001) Photoperiodic con-
trol of seasonality in birds. J Biol Rhythms 16:365–380
Diggle P (1990) Time series: A biostatistical introduction. Oxford
University Press, New York
Farner D, Gwinner E (1980) Photoperiodicity, circannual and
reproductive cycles. In: Epple A, Stetson M (eds) Avian endo-
crinology. Academic Press, New York, pp 331–366
Farner D, Donham R, Moore M, Lewis R (1980) The temporal
relationship between the cycle of testicular development and
molt in the white-crowned sparrow, Zonotrichia leucophrys
gambelii. Auk 97:63–75
Glutz von Blotzheim UN, Bauer KM (eds) (1988) Handbuch der
Vo
¨gel Mitteleuropas, 11. Aula, Wiesbaden
Gwinner E (1975) Die circannuale Periodik der Fort-
pflanzungsaktivita
¨t beim Star (Sturnus vulgaris) unter dem
Einfluss gleich-und andersgeschlechtiger Artgenossen. Z Tier-
psychol 38:34–43
Gwinner E (1987) Annual rhythms of gonadal size, migratory
disposition, and molt in garden warblers Sylvia borin exposed in
winter to an equatorial or a southern hemisphere photoperiod.
Ornis Scan 18:251–256
Gwinner E (1988) Photorefractoriness in equatorial migrants. In:
Ouellet H (ed) Proc 9th Ornithological Congr, pp 626–633
353
Gwinner E (1990) Circannual rhythms in bird migration: con-
trol of temporal patterns and interactions with photoperiod.
In: Gwinner E (ed) Bird migration: physiology and eco-
physiology. Springer, Berlin Heidelberg New York, pp 257–
268
Gwinner E (1999) Rigid and flexible adjustments to a periodic
environment: Role of circadian and circannual programs. In:
Adams, Slotow R (eds) Proc 22nd Int Ornithological Congr.
BirdLife South Africa, Johannesburg, pp 2366–2378
Gwinner E, Czeschlik D (1978) On the significance of spring
migratory restlessness in caged birds. Oikos 30:364–372
Gwinner E, Helm B (2003) Circannual and circadian contributions
to the timing of avian migration. In: Berthold P, Gwinner E,
Sonnenschein E (eds) Avian migration. Springer, Berlin Hei-
delberg New York, pp 81–95
Gwinner E, Neusser V, Engl E, Schmidl E (1987) Haltung, Zucht
und Eiaufzucht afrikanischer und europa
¨ischer Schwarzkehl-
chen Saxicola torquata. Gef Welt 111:118–120, 145–147
Hahn T, Boswell T, Wingfield J, Ball G (1997) Temporal flexibility
in avian reproduction. Patterns and mechanisms. In: Nolan V,
Ketterson E (eds) Current ornithology, vol 14. Plenum, New
York, pp 39–80
Helm B (2003) Seasonal timing in different environments: com-
parative studies in stonechats. PhD thesis, Ludwig-Maximilian
University, Munich
Helm B, Gwinner E (1999) Timing of postjuvenal molt in African
(Saxicola torquata axillaris) and European (Saxicola torquata
rubicola) stonechats: effects of genetic and environmenal
factors. Auk 116:589–603
Helm B, Gwinner E (2001) Nestling growth and post-juvenile molt
under a tight seasonal schedule in stonechats Saxicola torquata
maura from Kazakhstan. Avian Sc 1:31–42
Jenni L, Winkler R (1994). Molt and ageing of European Passe-
rines. Academic Press, London, San Diego, New York
King JR (1963) Autumnal migratory-fat deposition in the White-
crowned sparrow. In: Proc 13th Int Ornithological Congr, pp
940–949
Lambrechts MM, Perret P, Blondel J (1996) Adaptive differences in
the timing of egg laying between different populations of birds
result from variation in photoresponsiveness. Proc R Soc
London Ser B 263:19–22
Moore M, Donham R, Farner D (1982) Physiological preparations
for autumnal migration in white-crowned sparrows. Condor
84:410–419
Newton I (1966) The moult of the Bullfinch Pyrrhula pyrrhula. Ibis
108:41–87
Rowan W (1926) On photoperiodism, reproductive periodicity and
the annual migrations of birds and certain fishes. Proc Boston
Soc Nat Hist 38:147–189
Silverin B, Massa R, Stokkan K (1993) Photoperiodic adaptation
to breeding at different latitudes in great tits. Gen Comp
Endocrinol 90:14–22
354
