Brown Falcons with yellow cere and bare facial parts

To: Paul McDonald <>, Richard Nowotny <>, Jeff Davies <>
Subject: Brown Falcons with yellow cere and bare facial parts
From: David James <>
Date: Thu, 6 Oct 2011 16:05:24 -0700 (PDT)
Part 1.
Chris Watson’s original question about the meaning of yellow ceres in Brown 
Falcons has sparked a lot of responses. Whilst the question has been answered, 
more or less, there have been some conflicting answers, and a lot of peripheral 
misinformation about variation in plumages and bare parts. This stems mostly 
from a study of Brown Falcon by Paul McDonald (2003: Emu 103, 21-28) that, 
despite being a good study, reaches unsupportable conclusions due to 
extrapolation of the findings beyond the capacity of the data set, some 
circular arguments and rejection of prior studies without justification. On 
this latter point, McDonald (2003) rejected virtually every aspect of the Brown 
Falcon ‘Field Identification’ and ‘Plumages and Related Matters’ accounts in 
Marchant & Higgins (1993: HANZAB Vol. 2, pp. 237-8 & 248-51). As the author of 
that plumages section in HANZAB I wish to correct some matters that have been 
dismissed or overlooked in this
 Birding-Aus discussion so far.
I agree with Paul McDonald on the relationship between cere colour and age and 
the lightening of plumage with age (both these processes were identified 
earlier in Weatherly et al. (1985) and HANZAB (1993)). I recognise that Paul’s 
study uncovered a significant component of sexual differences in plumages that 
was not identified in HANZAB or previously. However, I disagree with the 
assertion there are no plumage morphs or “phases” in Brown Falcon and plumage 
variation is all due to age and sex. 
First, let me establish that the Plumages texts in HANZAB were not quick and 
dirty studies. They involved careful and systematic observations of plumage, 
bare parts, measurements, structure and moult of museum specimens from 
throughout the species’ ranges, with large sample sizes when available. 
Variation was systematically assessed to partition the contributions of age, 
sex, season, plumage wear, morphs and geography by sorting the specimens by 
each of these characters to reveal apparent patterns.  
In the bare parts section of Brown Falcon in HANZAB (V.2, 1993 p. 250) I wrote, 
for all morphs, that "Cere and orbital ring usually pale grey to pale bluish 
grey; in a few, yellow, yellowish grey, or greenish yellow; yellowness develops 
with age, may be more common in males, and is associated with pale iris... 
[etc]. This largely answers Chris' question, though Jeff Davies is collecting 
evidence that it is particularly prevalent in rufous morph birds from arid 
regions. Condon (1951: Emu 50, 152-74) wrote about yellow ceres too, but I 
don't recall whether he made the association between bare part colours and age. 
Weatherly et al. (1985: Emu 85, 257-60) looked at changes in captive brown 
morph birds from Tasmania and noted the association between yellower bare part 
colours and age. 
McDonald, in his 2003paper and in recent postings to Birding-Aus, has proposed 
a hypothesis that I will couch as follows <there are no plumage morphs in Brown 
Falcon but rather birds get paler with age, and males are paler than females 
and this gives the illusion of different morphs as erroneously described in 
HANZAB>. This hypothesis contradicts all previous studies of the subject. 
However, it is an extrapolation beyond the applicability of McDonald’s data 
set, because it takes findings from a study at a single small location (Western 
Treatment Plant in Werribee Vic) over a short 3 years and extrapolates them to 
the entire continent of Australia. Considerable evidence is published that 
conflicts with this hypothesis. The descriptions in HANZAB were based on 
examination of approximately 500 specimens (many more than the 160 (or 14, 
actually) birds studied by McDonald) from the entire continent, not just a 
single site, that were accumulated by
 collectors over more than a century. This evidence has been dismissed by 
McDonald as under sampling. Some particular problems with McDonald’s hypothesis 
1) Most juveniles are brown with extensively buff underparts, so if they 
lighten with age how would dark birds described as the dark morph in HANZAB 
come to exist? Reliably aged juvenile skins exist that show entirely dark 
underparts and upperparts consistent with the descriptions of dark morph in 
HANZAB; at this age there can be no lightening of plumage due to age, and both 
males and females occur in this plumage. Therefore, juveniles show different 
2) Juveniles from rufous parents (collected together in central Australia) 
resemble juveniles from brown parents in their underparts, but differ in having 
consistently broader and more rufous fringes to the brown feathers of the 
upperparts at the same age. Therefore juveniles show different morphs before 
any lightening with age occurs (and this holds throughout their lives).
3) There are no rufous birds (rufous morph) or dark birds (dark morph) in 
Tasmania; this means there is by definition a geographical component to the 
variation across the country as a whole, not just age and sex components. 
4) Brown birds are more common in humid coastal regions, rufous birds more 
common in arid interior regions, and dark birds only prevalent in tropical 
northern areas. Once again this is a geographical component to the plumage 
variation. However, it is not a simple case of all rufous birds inland and all 
brown birds on the coast, all three forms occur widely, so these are not 
geographically isolated forms or genetically separated populations (i.e. they 
are not subspecies). Nor is it a cline of gradual transition from brown to 
rufous moving inland. That leaves only one parsimonious alternative: varying 
geographical ratios in the expression of different phenotypes across the 
continent, presumably as a result of differing evolutionary drivers across a 
meta-population in a heterogenous environment - or more simply put, different 
McDonald studied birds in a single location to generate his hypothesis. He 
ignored the lack of variation in Tasmania (from where he had no data), implying 
that it must be present but so far overlooked. Evidence indicates otherwise. 
Similarly the contradictory evidence for geographical variation (higher ratio 
of rufous birds in the interior) was dismissed by the contrived suggestion that 
Brown Falcons live longer in the inland than they do on the coast so more birds 
live to become pale. Where is the data to support that speculation? Where is a 
precedent? There are countless precedents for the alternative explanation that 
there is geographical variation in the expression of colour morphs in 
Australian birds. 
It is still the case that the morphs are not distinct but integrade to some 
extent, and birds do change (mostly get paler) with age. However, the sample 
sizes in McDonald’s study were rather small. Of 14 individuals recaptured after 
moulting to a subsequent plumage, 8 got paler but 6 did not. That is not a big 
sample, not a uniform trend, not statistically significant, and not 
geographically representative. HANZAB had already noted that birds get paler 
with age over several moults, without over generalising. 
HANZAB did not recognise the differences in sexes at the same age (males 
lighter than females) reported by McDonald. This seems to be an important 
advancement from McDonald’s study, and an oversight in HANZAB. McDonald’s birds 
were sexed by methods that have not been totally declared (described merely as 
“behavioural observations and a variety of morphological characters”) and the 
HANZAB method of simultaneously comparing a large sample of adult males with a 
large sample of adult females (sexed by dissection) did not detect this 
pattern. It is possible that I overlooked it. I suggest that the sexual 
difference is likely to be true, but technically it needs further evidence 
before it can accepted as proven. 
Weatherly et al. (1985) was a longitudinal study of a small number of subjects 
over a long period of time. HANZAB was a cross-sectional study of many 
subjects, each preserved at a single point in time (though also incorporating 
the results of Weatherly et al. and others). McDonald (2003) had a bit of both 
aspects with mostly one-off captures of many birds and some recaptures of a few 
banded wild birds, but it lacked the continent-wide coverage and large sample 
size of HANZAB. An advantage with skins over wild birds is you can compare 
large numbers simultaneously for extended periods, and even revisit them later. 
All studies have their advantages and limitations, so combining the insights 
from multiple lines of evidence is the most effective path. My thoughts are 
that McDonald does not replace all previous studies, but adds to HANZAB in 
demonstrating that plumage changes with age and sex are even more marked than 
previously recognised. 
(more to come)

David James, 



To unsubscribe from this mailing list,
send the message:
(in the body of the message, with no Subject line)
<Prev in Thread] Current Thread [Next in Thread>

The University of NSW School of Computer and Engineering takes no responsibility for the contents of this archive. It is purely a compilation of material sent by many people to the birding-aus mailing list. It has not been checked for accuracy nor its content verified in any way. If you wish to get material removed from the archive or have other queries about the archive e-mail Andrew Taylor at this address: andrewt@cse.unsw.EDU.AU