Rheindt et al. • LEAPFROG PATTERN IN PTILINOPUS FRUIT DOVES 
431 
FIG. I. Distribution of the three Maroon-chinned Fruit Dove (Ptilinopus subgularis) taxa: epia on Sulawesi, subgularis 
on the Banggai Archipelago (including Peleng Island and Banggai Island), and mangoliensis on the Sula Islands (including 
Mangole Island and Taliabu Island); the two stars indicate the two separate vocal sampling localities for epia in North and 
Central Sulawesi, respectively. 
explicitly noting that it was closely related to 
Columba gularis (Quoy and Gaimard 1830) from 
Sulawesi, but differed in important aspects. 
Similarly. Rothschild (1808) described the Sula 
population as a full species under the name 
Ptilinopus mangoliensis despite being aware that 
it closely resembled the two taxa from Sulawesi 
and Banggai. Pelers (1937), in his compilation of 
the world’s birds, included all three taxa under 
one species and applied the name Ptilinopus 
subgularis (Meyer and Wiglesworth 1896), which 
had priority. Peters's (1937) lumping was con¬ 
ducted without an accompanying rationale and is 
one of many instances in which he united similar 
but allopatric taxa into one species. All subse¬ 
quent authors (e.g., Baptista et al. 1997. Coates 
and Bishop 1997, Gibbs et ai. 2001) followed 
Peters's (1937) treatment, and the question of 
species status for all three taxa has not been 
revisited. 
The objectives of our work are to: (1) document 
the unknown vocalizations of subgularis and 
mangoliensis and the vocal leapfrog pattern that 
characterizes the bioacoustic relationship among 
all three taxa. and (2) combine the new insights on 
the distribution of vocal trait variation in the 
Maroon-chinned Fruit Dove with Pleistocene 
earth-historic information to revise species bound¬ 
aries in the complex. 
METHODS 
Vocal Sampling. —The vocalizations of the two 
undocumented taxa (subgularis and mangoliensis ) 
were recorded during recent visits to the islands of 
Peleng (e.g., Rheindt et al. 2010) and Taliabu (e.g., 
Rheindt 2010). We also obtained recordings of epia 
from our own sound collections and those of 
colleagues. Deluils of all 35 recordings, including 
localities, dates, and names of the recordists, are 
provided (Appendix), Both analog and digital 
recordings were converted into WAV format if 
they were not created in that format. Recordists 
used different equipment but we consider equip¬ 
ment bias to be negligible. For example, a close 
visual examination of sonograms showed the level 
of variability in background noise and slight 
differences in note shape among recordings from 
the same recordist are most often equivalent to, and 
at times even larger than, the variability among 
recordings from different recordists. This suggests 
differences in recording quality are more important 
than equipment differences. We analyzed multiple 
recordings from several different recordists for 
each taxon, which should remove anv such bias. 
