I'd like to see Stephen Ambrose's evidence for the notion that some birds of
the arid and semi-arid zones of Australia moult their primaries twice a
year. It comes as surprising news to me, as it is inconsistent with two
fairly clear trends from the admittedly scanty literature on moult in arid
(1) In arid and other regions the timing of moult is fairly regular every
year, while the timing of breeding is not. This trend was first reported
some time ago (e.g. Snow 1966, Keast 1968, Payne 1972) and I haven't seen
anything published since then to suggest it is seriously wrong.
(2) In arid regions, wing moult usually takes longer than in other habitats
(e.g. Keast 1968, Wyndham et al. 1983).
I'd like to expand a bit on these two trends:
REGULAR MOULTS AND IRREGULAR BREEDING
David James hit the nail on the head when he pointed out that birds do not
moult because they breed, but primarily because their feathers wear out and
thus have to be replaced. There are lots of birds in Australia known to be
multi-brooded (e.g. Ian Rowley and Elanor Russell demonstrate this nicely in
their recent book on Fairy-wrens). Such birds, which nest or attempt to nest
several times in each season, do not moult after each attempt at nesting -
just once, at the end of the breeding season. Many migrant birds from the
Holarctic routinely breed, and then perform a long southwards migration
before they moult their wing feathers - so in these breeding does not
automatically lead to primaries which are so worn that they need instant
replacement. Of course, this is not quite the same thing as the scenario
Stephen had in mind: "If an Australian arid-zone bird breeds in early spring
(as a seasonal breeder), moults in late spring/early summer at the
conclusion of breeding, and then summer cyclonic rains falling in late
summer stimulates further breeding, then it's likely that these birds will
moult again in autumn/early winter. Thus wing moult would occur more than
once in a 12-month period."
It might be worth mentioning here some of the results of a banding study of
Yellow-tufted and Fuscous Honeyeaters that we have been running on the
box-ironbark forests of central Victoria for the last 16 years. Analysis is
far from complete, but here are the basics:
1. Adults usually moult wing-feathers once a year, between about December
2. In the spring and early summer of the 1982/83 drought, these honeyeaters
failed to breed; some might have attempted to do so but none got far enough
along the track to develop brood patches or cloacal protrubrance.
3. In the 1982/83 spring & summer, primary moult occurred far earlier than
it has in any of the subsquent years of the study. There was some scatter,
but in general birds started wing-moult a couple of months early and
completed it around January.
4. When the drought broke with heavy rains in April 1983, many of the
surviving adult tufties and Fuscous Honeyeaters bred. Having started this
breeding attempt just after they had completed moult, when they completed
breeding their primaries were still pretty fresh.
5. After this breeding attempt, the honeyeaters did not perform an extra
primary moult. Their next moult occurred in the following summer/autumn
(roughly January to March 1984), the timing being consistent with that in
all subsequent years.
No suggestion here that opportunistic breeding made the primaries of these
honeyeaters remarkably tatty, or that it forced birds to squeeze more than
one primary moult into a year. However it does suggest that they have pretty
sophisticated mechanisms for controlling their timing of moult and breeding
so they can get through droughts and exploit unpredictable rainfall.
SLOW MOULTS IN ARID PLACES
Slowish wing-moult in arid places was first reported in Australia by Keast
(1968). His estimate of duration (usually about 4.5 months in inland NSW)
may need some revision in view of fancier analytical methods now available,
but I don't think the basic trend is in doubt. Slow moults might reduce the
moult-induced stress that a bird undergoes at any particular time but they
aren't easy to perform twice a year. Indeed, some of the estimates of the
duration of primary-moult of birds in the arid zone are for longer than half
a year, so such birds couldn't perform two wing-moults a year without
substantial and senseless overlap between two moult cycles. An extreme
reported for quite a few arid-zone birds (e.g. Zebra Finch, Zann 1984;
Budgerigar, Wyndham et al. 1983) is for wing moult to be so slow that is
more or less continuous except when interrupted for breeding. Slow wing
moults of this kind can be suspended at times, leading to unexpected
patterns of wear in the wing that are easily misinterpreted as evidence of
partial moults or of some feathers being moulted more than once a year. This
doesn't mean that such processes cannot occur, but fairly careful studies
are often needed to work out what is really going on.
Stephen's Spinifexbirds do not seem to fit in with this trend; if they
dumped all of their secondaries at once, they were moulting fast. I wonder
if this might be because they live in such low and prickly and dense habitat
that they can get away with moulting at such a rate that their flight is
impaired? In this respect they might be similar to rails (which become
flightless while moulting) or Arctic Warblers (which moult in dense
willow-scrub and lead a very skulking existence while moulting their
primaries so fast that they are close to being flightless, e.g Haukioja
1971; Underhill et al. 1992). As for how often Spinifexbirds moult their
primaries a year, I couldn't say - if they do it twice, that might not be an
arid zone thing but an effect of the abrasive habitat they live in (i.e.
analagous to the double wing-moult of Cisticolas and Prinias, which are also
Sylviidae living in abrasive, grassy places). In Victoria, most of our
goldfields Honeyeaters have started breeding in August, and are in still in
their once-a-year moult in February. It doesn't sound too unlike the timing
Stephen described for Spinifexbirds.
Haukioja, E. 1971. Ornis Fennica 48: 101-116.
Keast, A. 1968. J. Zool., Lond. 155: 185-200.
Payne, R.B. 1972. Mechanisms and control of Moult. Chapter 3 in Farner et
al. Avian Biology, volume 1.
Snow, D.W. 1966. J. Ornithol. 107: 283-291.
Underhill et al 1992. Ibis 134: 286-297.
Wyndham, E. et al 1983. Emu 83: 242-246.
Zann 1985. Slow continuous moult of Zebra Finches.... Ibis, vol. 127.