162 
THE WILSON JOURNAL OF ORNITHOLOGY* Vol. 123, No. 1, March 2011 
METHODS 
Waterproof pressure-zone microphones were 
used to concurrently record nocturnal flight calls 
from 2 to 16 October 2008 at three locations 
within 6 km of Gardiner, Maine, USA. Site #1 
was at 44° 13' N, 69° 46' W; Site #2 was at 44° 
13' N, 69 47' W; and Site #3 was at 44° 16' N, 
69 47' W. Two microphones were within 2 km of 
each other and the third was ~6 km distant. Two 
microphones were in suburban neighborhoods with 
low level street-light illumination while the third 
was on the border of an extensive forested area 
with no artificial lighting. All three locations were 
within 1 km of the Kennebec River, a major 
southward flowing river. Each microphone was 
placed and pointed upwards so that it had an 
unobstructed path to the sky. Acoustic XLR cable 
connected the microphone to a Rolls MP13 pre¬ 
amplifier housed in a nearby building. The signal 
was sent from the pre-amplifier into a computer 
which automatically activated two simultaneously 
running bird flight call detection software pro¬ 
grams (Thrush-r.exe and Tseep-r.exe; both distrib¬ 
uted as shareware from www.oldbird.org) at 2000 
his EST and de-activated the programs by 0600 hrs 
EST the next morning (the detector programs at 
one station at times were allowed to continue 
slightly past 0600 hrs EST). Each potential bird 
flight call detected by either program was saved as 
a WAV tile with a file name reflecting the date and 
time it was detected. All sound files collected were 
reviewed aurally and spectrograms inspected 
visually using Glassofire sound analysis and file 
sorting software (distributed as shareware from 
www.oldbird.org). Non-bird sounds were re¬ 
moved and bird sounds were sorted and saved 
by date. More detailed spectrographic analysis 
and measurements were completed using Raven 
sound analysis software (available from Cornell 
Laboratory of Ornithology, Ithaca, NY, USA). 
Flight calls that we considered to be of Pine Siskins 
because they were identical to the well-known 
and described “Kdeew” flight call of the spe¬ 
cies (Sibley 2000), based on our own field 
experience, were saved and sent to flight call 
and bird identification experts William Evans 
Michael O’Brien, and David Sibley for external 
review. 
RESULTS 
We identified 212 of 2,432 flight calls detected 
at all three stations that we considered likely 
characteristic of Pine Siskins. Our three expert 
reviewers independently concurred that 90% of 
the calls were clearly identifiable as Pine Siskin 
flight calls. We were left with 190 calls that were 
confiimed as those of Pine Siskins after removing 
calls for which there was not consensus among 
our experts. All flight calls were archived at 
Macaulay Library of Natural Sounds at Cornell 
Laboratory of Ornithology (catalog numbers 
140388-140396). 
Our stations recorded Pine Siskin flight calls 
from 10 to 16 October 2008 when all three 
stations were shut down for the season. Calls were 
detected as early as 2146 hrs EST and as late as 
0606 hrs EST (Fig. 1), but detections occurred 
throughout the night with calls detected in every 
hour between 2300 and 0600 hrs EST the 
following morning. Approximately 90% of re¬ 
corded flight calls were between 0000 and 0500 
hrs EST, and ~80% of calls were at least 1 hr 
before sunrise (Table 1). One hundred and thirty 
of the 190 recorded calls occurred on 15 October 
2008. Recording stopped on 16 October and the 
full extent of migration dates is unknown. 
DISCUSSION 
Many species of North American finches thought 
to be virtually exclusive diurnal migrants, including 
Pine Siskin, have been recorded producing flight 
calls in the hour before sunrise as they begin their 
diurnal migrations (Evans and O'Brien 2002). 
Flight calls of Pine Siskins were recorded in our 
study in significant numbers throughout the night 
over multiple nights and multiple locations, sug¬ 
gesting these birds were likely undergoing noctur¬ 
nal migration. To our knowledge, this is the first 
documented observation of apparent nocturnal 
migration in this species. 
The migratory irruptive behavior of Pine Siskins 
is apparently induced proximately by a lack of food 
resources, primarily conifer seeds (Dawson 1997). 
Pine Siskins are known to make long-distance 
migratory movements biennially on average (Bock 
and Lepthien 1976, Yunick 1983, Hochachka et al. 
1999), apparently due to broad scale synchronicity 
of conifers in their biennial cycle of cone 
production (Pielou 1988). Birds will not show these 
long-distance migratory movements when cone 
production is high in a particular region, while in 
poor seed production years, large numbers will 
move in search of food (Dawson 1997). 
Nocturnal migration could be a behavioral trait 
that is only expressed by finches under extreme 
