The moulting seasons of some finches and buntings

Transcription

1 Bird Study ISSN: (Print) (Online) Journal homepage: The moulting seasons of some finches and buntings I. Newton To cite this article: I. Newton (1968) The moulting seasons of some finches and buntings, Bird Study, 15:2, 84-92, DOI: / To link to this article: Published online: 19 Jun Submit your article to this journal Article views: 634 View related articles Citing articles: 33 View citing articles Full Terms & Conditions of access and use can be found at Download by: [ ] Date: 30 November 2017, At: 06:53

2 The moulting seasons of some finches and buntings by I. Newton THE MOULTING SEASONS of most bird populations are only imprecisely known and the duration of moult in individual birds has been determined for only a few species (Stresemann & Stresemann 1966, Snow 1967). Details of moult are now recorded as standard practice by many ringers in Britain under the B.T.O. Moult Enquiry, and enough records for finches and buntings have accumulated to warrant a preliminary analysis. This was undertaken primarily to obtain a better indication of the timing and duration of moult in the various species than is currently available in the literature (chiefly Witherby et al. 1938), so that moult could be related more closely to other events in the annual cycle. It was also hoped to reveal gaps in the data which might be filled in future years. The species involved comprise seven cardueline finches, the Greenfinch Carduelis chloris, Goldfinch C. carduelis, Sisldn C. spinus, Linnet C. cannabina, Twite C. flavirostris, Lesser Redpoll C. flammea cabaret and Bullfinch Pyrrhula pyrrhula; one fringilline finch, the Chaffinch Fringilla coelebs; and two emberizids, the Yellowhammer Emberiza citrinella and Corn Bunting E. calandra. Most records were obtained between 1960 and 1966, and for most species only those records from within Britain south of Lancashire and Yorkshire (i.e. between latitudes 50 and 53 50' N) are considered here. However, a good series of records for the Twite on Fair Isle (lat ' N), collected mainly by Kenneth Williamson, has also been included. At present, there are large gaps in the records for most species in northern Britain, but if future work is aimed towards filling these, a comparative analysis should soon be possible. THE SEQUENCE OF MOULT All the species considered here have nine primaries, six secondaries and three tertialsl in each wing, and twelve rectrices in the tail. All these feathers are moulted in the sequence typical for passerines; viz. the primaries from the innermost outwards (numbered 1-9), the secondaries from the outermost inwards (numbered 1-6) and the rectrices from the central pair outwards (numbered as pairs 1-6); among the tertials, the central one is shed first, foiiowed by the inner and then the outer. In all species, the moult begins with the shedding of the first primary and (with the exception of the Corn Bunting see below) ends with the completion of growth of either the last primary or the last secondary; all other feathers are replaced between these two events. Usually, all the secondaries are shed while primaries 5-9 are in moult (the last 2-3 secondaries being lit is convenient to distinguish the secondaries and tertials in finches, for they moult as independent units, even though anatomically they form part of the same feather series. 84

3 MOULT OF FINCHES AND BUNTINGS shed after the last primary has begun growth), but in all species there is considerable individual variation. The new rectrices begin growth between the shedding of primaries 3 and 9 (usually 5 and 9), but again there is some variation between the individuals of all species. The rectrices of most species are completely replaced within the period of the primary moult, but in the Corn Bunting up to three pairs of these feathers may be shed after the ninth primary, and some may still be growing when the moult of the primaries ends. In most species, the post-juvenile moult is restricted to the body feathers and the small feathers of the wings and tail; the large flight and tail feathers, the alula, the primary coverts and a variable number of secondary coverts are retained for another year. In the Corn Bunting, on the other hand, the post-juvenile moult is apparently complete. Surprisingly, the same is true of the Goldfinches in Sicily (records collected by L. Cornwallis, B. D. H. Dickinson and M. P. L. Fogden), and in southern Spain (E. J. Mackvill and S. L. Pimm), though in Britain juvenile Goldfinches have only a partial moult.2 In all the species considered here, the juveniles moult within the moulting season of the adults. It has not been possible to estimate from the moult cards the duration of the post-juvenile moult in individual birds; but in the Bullfinch I was able to estimate, from recaptures of ringed birds, that the post-juvenile moult is shorter than the adult moult, taking only 7-9 weeks, as compared with in the adults (Newton 1966). In this species, the spread in date of onset of moult in the juveniles is similar to that in adults. Also, the juveniles from the first nests of the year start to moult earlier in the season, but at a greater age, than late-hatched young. The juveniles moult their body feathers in the same order as the adults. ANALYSIS OF MOULT CARDS Since in most species the replacement of the flight feathers spans the entire moult period, the overall stage of moult in a bird can be recorded by reference to these feathers alone. For each bird caught, each flight feather was allotted a score, from 0 (old) to 5 (new), according to its stage of growth. The sum of the scores of individual feathers (the moult score) then provides an index of the overall state of moult in a bird. In calculating the moult score of each species, the score of the primaries alone has been used, for reasons given elsewhere (Newton 1967). Since there are nine primaries per wing, a fully moulted bird would score 45. For each species, the primary scores of all individuals were plotted against date, as in Fig. 1, and from this the mean rate and duration of moult for individuals in the population was judged by eye. A date of start of moult was calculated for each bird by working back from its score at capture; the results for each species are shown in Fig. 2. 2A 1967 moult card from Portland B.O. shows a juvenile Goldfinch (age known from previous trapping) undergoing a complete moult in August. ED. 85

4 BIRD STUDY U zu 10 I UIY 20 io AUGUST 20 SEPTEMBER io OCT. FIGURE 1. Primary moult score in relation to date in the Twite on Fair Isle. Apparently those ringers who record moult data are much more active in August and September than in the following two months, for, although many birds are still in moult then, comparatively few are recorded. This means that the later in the season a bird begins moulting, the less likely it is to be noted. This should not have affected the estimates of the mean rate and duration of moult in the different species, but it will have resulted in a smaller proportion of late, than of early, moulters being recorded. For this reason, I have calculated the median date of onset of moult in each species, rather than the mean (Table I). While the data are too few for a more detailed analysis, they are sufficient to show that the spread in date of onset of moult in any one year or area is for all species almost as great as that of the entire data. Any annual or regional differences in the moulting season of the various species must therefore be slight. Extremely few birds were caught more than once during their moult (less than four for each species), so that it is not possible to compare estimates of the mean rate of moult in the population with retrap data for individuals. These deficiencies in the B.T.O. data can, of course, be overcome by workers operating in one area throughout the season, but, among the finches, such studies have so far been made only on the Bullfinch (Newton 1966) and the Lesser Redpoll (Evans 1966). RESULTS The full data on the moults of the different species are set out in Table I, and for completeness, some figures for the Lesser Redpoll have been added from Evans (1966) and for the Siskin in captivity from Newton (in prep.). As may be seen, the mean duration of moult varies considerably between species. On average, the Redpoll takes 50 days, the Linnet, Twite, Siskin and Chaffinch 70 days, the 86

5 MOULT OF FINCHES AND BUNTINGS GREENFINCH 20 GOLDFINCH 20 LINNET t9 z i= 30 V C l7 Z TWITS BULLFINCH CHAFFINCH AUGUST SEPTEMBER FIGURE 2. The date of onset of moult in various finches. The columns show the percentage starting in successive 5-day periods. Goldfinch 80 days and the Bullfinch and Greenfinch 85 days. However, it is worth noting that at least in two species there is considerable variation between individuals. From retraps, I estimated that individual Bullfinches took days to moult (cf. mean of 85 days given above), while Evans (1966) found that the mean duration of moult in the Lesser Redpoll varied both from year to year and between the sexes, the extremes being 43 and 56 days (cf. mean of 50 days). Despite this variation, the estimates given in Table I should suffice for comparative purposes. The median onset of moult also varies considerably between species, as does the extent of individual variation (Table I and Fig. 2); it is possible that the spread will be increased for certain species (Chaffinch, Goldfinch, Twite) when more records are available. Unfortunately, there are insufficient records for the Yellowhammer in October and November to provide reliable estimates, but the moult probably takes over 80 days in individual birds and 87

6 BIRD STUDY TABLE I-THE TIMING AND DURATION OF MOULT IN SOME FINCHES AND BUNTINGS Number examined Mean duration of moult (days) Median date of start Median date of finish 95% spread in start Mean no. of primaries in growth at once per wing* Chaffinch June 1 Sept. 54 days 2.7 ( ) Greenfinch Aug. 28 Oct. 81 days 2.5 ( ) Goldfinch July 17 Oct. 54 days 2.6 ( ) Siskin 6 70 Linnet July 2 Oct. 65 days 2.3 ( ) Redpoll 50 6 Aug. 25 Sept. 42 days ( ) Twite July 28 Sept. 40 days 1.8 ( ) Bullfinch Aug. 2 Nov. 64 days 2.9 ( ) Yellowhammer Corn Bunting 2.3 * between the shedding of the 3rd primary and the completion of growth of the 7th primary. NOTE. This table is based on records of birds caught in southern England, except for the Twite (Fair Isle), Redpoll (Northumberland; Evans 1966), and Siskin (in captivity, personal obs.). the onset is spread over at least 60 days. There are few records for the Corn Bunting, but some individuals certainly begin moulting in August and some are completing moult in October and early November. The spread and duration of the moult determine the period each year during which birds may be found in moult. The moulting season of each species is shown in Fig. 3. As may be seen, it varies from three to five and a half months, according to species. Thus, not only does the moult occupy a substantial part of the annual cycle of individual birds, but in all species the moult period of the population as a whole is longer than the breeding season a point apparently not previously appreciated. Among the Greenfinches in southern Britain, the spread in moult is so great, that at no time is the entire adult population in moult at once. The same was true of the Bullfinch in one year near Oxford, when some individuals moulted unusually late (Newton 1966). For all species, the present data extend the moulting seasons given by Witherby et al. (1938) by 6-12 weeks, in different species. It seems that for most species Witherby gave only the main moulting period, when a high proportion of the population could be found in moult. However, he gave the moult periods of the Twite and Lesser Redpoll as September and October, but some 88

7 MOULT OF FINCHES ANT) BUNTINGS J ^ FIM 'AlM.^ J ^ J ^ A ^ S ^ O ^ N ^ D FIGURE 3. The moulting seasons of some finches. The thick line shows the period when the entire adult population is in moult. 95 per cent of the adults moult completely within the period shown by the thin line, the remaining 5 per cent having at least part of their moult outside this period. The figure is based on the data in Table I. Twites begin as early as June and some Redpolls in July. In both species, most individuals have started by early August, and very few are in moult after mid-october. The number of primaries growing concurrently in each wing depends both on the intervals between the shedding of the old feathers and the growth rates of the new ones. There is no consistent relationship between the duration of moult in the different species and the number of primaries growing concurrently (see Table I). This means that both the growth rate of the feathers and the rate at which they are shed varies independently between species. The fact that the Yellowhammer has, on average, more primaries in growth at once (mean 3.8, range 2-8) than any other species and yet has a protracted moult (probably more than 80 days) suggests that its feathers grow very slowly. The Twite, on the other hand, has a comparatively short moult (mean 70 days), yet fewer primaries in growth at once (mean 1.9, range 1-3) than any other species, which indicates that its individual feathers grow very rapidly. Measurements of feather growth have been made on captive Bullfinches and Greenfinches (Newton 1967). Within any one of these species, the rates of growth of particular feathers were remarkably consistent from bird to bird, and variation in the duration of moult was the result of variations in the intervals between the shedding of feathers. Similarly, Evans (1966) concluded that in the Lesser Redpoll differences in the duration of moult from year to year and between the sexes were directly related to the number 89

8 BIRD STUDY of primaries growing concurrently and did not result from changes in the growth rate of the feathers. DISCUSSION To what extent can the timing and duration of moult in the various species be related to their ecology and to other events in the annual cycle, such as breeding and migration? All the species considered here start breeding in late April or early May, and rear one or more broods through the summer. In all species, the peak of the moulting season follows the peak of the breeding season. In the Bullfinch and Redpoll, individual adults begin moult either just before or just after their last young of the year reach independence. The same is likely to be true of the remaining species, though detailed studies have not been made. The cardueline finches, which feed their young primarily on seeds, raise up to three broods a year and occupied nests may be found until August or September. The Chaffinch, on the other hand, rears its young primarily on defoliating caterpillars, which are plentiful for only a few weeks in early June, and permit most pairs to raise only one brood a year (Newton 1964); associated with the shorter breeding season, adult Chaffinches start moult, on average, several weeks earlier than do the carduelines. The buntings also raise their young primarily on insects, but chiefly from herbaceous plants, and take large quantities of adult insects as well as larvae. Food suitable for nestling buntings is thus available in quantity for a longer period each year than for nestling Chaffinches and in fact the breeding seasons of the buntings in southern England are about as long as those of the carduelines. As a result, the adults of both groups begin moult at the same time of year. As already mentioned, the spread in date of onset of moult in any one area or year is almost as great as that of the entire data. This spread probably results mainly from some pairs breeding till a later date than others and delaying their moult accordingly. For the populations as a whole, the overlap between the breeding and moulting seasons varies from about six weeks in the Redpoll and Twite to twelve weeks in the Greenfinch. In all the species, the overlap is so great that some individuals might raise one or (in the Greenfinch) even two broods after others have started to moult. Little information is available on the factors terminating the breeding of individual birds. In the Bullfinch, a greater proportion of older birds than of yearlings prolong breeding and delay their moult, though both age-groups are among those moulting latest (Newton 1966). Perhaps also, the breeding success in the early part of the season has some influence, those pairs whose previous attempts have failed being more likely to try again with a late nest. The fact that in the two species studied in detail, individual birds start moult as their own breeding ends, irrespective of the date at which this occurs, suggests (1) that breeding normally inhibits moult in individual adults, and (2) that they can begin moult any time towards the end of the normal breeding season (because 90

9 MOULT OF FINCHES AND BUNTINGS those birds that finish breeding earliest are among the first to moult and do not wait to moult with the latest birds). It is not surprising that individual birds start moult as soon as the cessation of breeding activities permit, for feather growth requires extra food, and as the season advances, days become shorter and colder, and food generally scarcer. Moult and migration also tend to be mutually exclusive, presumably partly because both events need an increase in food-intake. Also, a long flight with defective wings might result in too great an expenditure of energy (Evans, in press), as well as exposing the bird to greater risk from avian predators. There is an apparent relationship between the moult and the migratory habits of most of the species considered here (see also Dolnik 1965). The Bullfinch, Greenfinch and Yellowhammer, which are resident in southern England, take longer to moult and start later than the remaining species. Individuals of at least two of these species may be found in moult for over five months each year (Fig. 3). The remaining cardueline finches are partial migrants; the majority of individuals winter several hundred miles from the breeding grounds, the Goldfinch and Linnet mainly in southern France and Iberia, and the Siskin and Redpoll mainly in southern Britain or the Low Countries. The Goldfinch, Linnet and Twite start earlier and moult more quickly (the Goldfinch only slightly so) than the resident species, with the result that most individuals finish towards the end of September or in October and then migrate. The Lesser Redpoll, which is also a partial migrant, starts comparatively late (in early August), but moults so quickly that it too has finished by the end of September or in early October (Evans 1966). Thus all the migrant species finish their moult, on average, about a month before the resident species and leave towards the end of September or in October. Also, I have noticed that late-moulting individuals of all three migrant species occasionally migrate with the last one or two primaries and secondaries not quite full grown. The Chaffinch does not fit the pattern shown by the buntings and the carduelines, for although resident, it has an early and relatively fast moult. The early moult is explicable in terms of the shorter breeding season, as mentioned above, but it is difficult to see the advantage in its being so rapid. There is no apparent relationship between the pattern of the moult in these different species and their feeding ecology. Thus, considerable differences in moult occur between species despite the fact that they_all live primarily on seeds at this time. Further, the species most similar in feeding habits, such as (on farmland) the Linnet, Chaffinch and Greenfinch, differ considerably in the timing, duration and spread of their moult. Similarly, the species which in some years feed primarily from trees at this time, the Siskin, Redpoll and Bullfinch, also differ from each other in the duration of their moult. It seems, therefore, that the timing and duration of moult in finches is more closely related to their breeding and migratory habits than to their feeding ecology. 91

10 BIRD STUDY ACKNOWLEDGEMENTS My thanks are due primarily to all those members of the B.T.O. who contributed moult records. I am also grateful to Drs. P. R. Evans, D. Lack, D. W. Snow and P. Ward for their comments on the manuscript and to Miss Waina Cheng for drawing the figures. SUMMARY The B.T.O. moult records for some finches and buntings are analysed. It is estimated that, on average, the Lesser Redpoll takes about 50 days to moult (Evans 1966), the Linnet, Twite, Siskin and Chaffinch 70 days, the Goldfinch 80 days and the Bullfinch and Greenfinch 85 days. There is, however, considerable variation between individuals of the same species. There is also variation between species in the timing of moult and in the spread in date of onset. On average, Chaffinches start moulting several weeks earlier than the cardueline finches and the buntings; this may be linked with the fact that Chaffinches finish breeding earlier. Within a species, the spread results from some individuals breeding later in the year than others and delaying their moult accordingly. There is no consistent relationship between the duration of moult in the different species and the number of primaries growing concurrently. This suggests that the growth rate of the primaries and the rate at which they are shed vary independently in the different species. The B.T.O. data extend the moulting seasons given by Witherby et al., by 6-12 weeks in the different species, and modify considerably the information given there for the time of moult in the Lesser Redpoll and Twite. In all species, the period of year when individuals may be found in moult is longer than their breeding seasons. In general, those species which are partial migrants start moulting earlier and/or moult faster than the resident species. As a result the migrants have mostly finished moult by the end of September or early October and are ready to migrate then, while the residents are still in moult. The Chaffinch is an exception in that, although resident, it has an early and relatively fast moult. REFERENCES DOLNIK, V. R Bioenergetics of moults of fringillids as adaptations of migration. Nov. Om: (4th All Union Om. Conference, Alma-Ata). EVANS, P. R Autumn movements, moult and measurements of the Lesser Redpoll Carduelis flammea cabaret. Ibis, 108: EVANS, P. R. (in press). Migration and pre-migratory fat deposition in the Lesser Redpoll Carduelis flammea cabaret. Condor. NEWTON, I The breeding biology of the Chaffinch. Bird Study, 11: NEWTON, I The moult of the Bullfinch Pyrrhula pyrrhula. Ibis, 108: NEWTON, I Feather growth and moult in some captive finches. Bird Study, 14: SNOW, D. W A guide to moult in British birds. British Trust for Ornithology Field Guide no. 11. STRESEMANN, E. & STRESEMANN, v Die Mauser der Vogel. J. Orn., 107. Suppl. WITHERBY, H. F. et al The Handbook of British Birds. Vol. 1. London. Dr. I. Newton, The Nature Conservancy, 12 Hope Terrace, Edinburgh, 9. 92