Effect of laying date on chick production in Oyster catcher s and Herring Gulls M. P. Harris INTRODUCTION It has been widely believed that birds timed their breeding seasons so that the young were raised when food was most abundant or easily obtained, but Perrins (1965, 1966) showed for two such different species as the Great Tit Parus major and the Manx Shearwater Puffinus puffinus that the pairs breeding earliest in a season left more progeny than those breeding late. In the cases of some species it may perhaps be that external factors, such as a shortage of food at the laying period, force many pairs to raise their young after the time of maximum availability of food. In the Great Tit late nestlings tend to be lighter in weight at fledging than early ones, and such light chicks survive their first three months less well than heavier ones (Perrins 1963). In the Manx Shearwater nesting success is similar among late and early nesters and the mortality comes after fledging. Young Manx Shearwaters migrate soon after fledging to wintering grounds off South America, and the early ones may either find feeding conditions easier for this migration or have more time to learn to feed efficiently before being forced to migrate (Harris 1966). The aim of this paper is to give some pertinent information for two other species, the Oystercatcher Haematopus ostrakgus and the Herring Gull harm argentatm, studied on Skokholm and Skomer Islands, Pembrokeshire. OYSTERCATCHER The Skokholm population of about 50 pairs of Oystercatchers raises an average of 0.7-0.9 young per pair each season, but breeding success is not uniform throughout a season, being highest for early nesters at 1.7 young fledged per pair and lowest for late nesters at 0.2 young fledged per pair (fig. 1). This decline is due to a smaller average clutchsize and lower hatching and fledging success (table 1). Safriel (1967) has shown that the higher success of early nesters is due to their young being fed on terrestrial food, such as moth caterpillars and crane-fly larvae (Tipulidae) when these prey items are largest. Later young are fed either on similar prey which, although still present, are at a lower density or on limpets Patella spp. which are brought to them from the shore. On Skokholm limpets are apparently a less satisfactory food source. 70
Effect of laying date on chick production 71 Table 1. Nesting success and post-fledging survival of Oystercatchers Haematopus ostralegus in relation to breeding date on Skokholm, Pembrokeshire Data mainly from Harris (1967), except that fledging success relates to 1963-64 and 1967-68, and the number of young surviving after fledging is brought up to November 1968 Before 20th May 2ist-3oth May HATCHING DATE 31st May 9th June ioth-i9th June 20th-29th June After 30th June Clutches which could have hatched Average clutch size Hatching success Percentage of hatched chicks which fledged 14 3.0 88% 67% 40 3.2 7 % 59% in 3-i 67% 5i% 95 2.8 58% 44% 29 2.4 60% 25% 7 1-9 38% 33% Total young ringed Calculated number fledged Percentage surviving more than one month after fledging 35 *3 17% 128 75 19% 343 175 15% 242 106 14% "3 28 17% 69 23 13% Data for the post-fledging survival of the young, as shown by retrapping and recoveries after fledging, are also given in table 1. The hatching dates of these young are not known, but, since most were ringed soon after hatching, the dates of ringing have been used and, generally speaking, the sequence of the groups shows the sequence of hatching. If the number surviving more than a month after fledging is considered in relation to the number fledging (calculated from data in table 1), it appears that the date of fledging does not have any further influence on survival. Once young leave the island they can presumably feed with post-breeders and non-breeders on the beds of Mussels Mytilus edulis and Cockles Cardium edule where food is abundant and in good condition. Thus early and late young are experiencing the same good feeding conditions, and might be expected to be equally efficient by the winter when presumably this food supply is depleted and there is competition for food. HERRING GULL During 1962 a study was made of 400 pairs of Herring Gulls nesting on the cliffs of Skomer (for details see Harris 1964). It must be stressed that fledging success was not known for all the young which hatched, but there is no reason to suspect any bias in the sample. Clutch-size remained constant, except in a few, probably repeat, nests, but both hatching and fledging success increased, so that late pairs produced about three times as many fledglings as early pairs (fig. 1, table 2). There were too few recoveries after fledging to determine any pattern in post-fledging survival. Broods of two (170 young hatched, 32 known to have fledged)
7 2. Effect of laying date on chick production YOUNG Der PAIR Fig. i. Decrease in production of young per pair of Oystercatchers Haematopus ostrakgus in relation to hatching date, contrasted with increase in nesting success of Herring Gulls Larus argentatus, Skokholm and Skomer Islands, Pembrokeshire. The data for Oystercatchers are analysed by ten-day periods and those for Herring Gulls by five-day periods (see tables i and 2) Table 2. Nesting success of Herring Gulls Larus argentatus in relation to laying and hatching dates on Skomer, Pembrokeshire, in 1962 The hatching success refers to all eggs laid, whereas the young known to have fledged is only a sample LAYING DATE Before ist-5th 6th-ioth iith-i5th After April 30th May May May i6thmay Eggs laid Percentage hatched 219 56% 150 57% 148 72 72% 68% 17 7i% Before May 31st HATCHING DATE ist-5th ( Sth-ioth nth-i5th i6th-20th June June June June After 21 st June Young hatched Percentage known to have fledged 14 7% 102 "% 79 i 7 93 13% 17% 14% 90 26%
Effect of laying date on chick production 73 FISH LANDED (tons) Fig. 2. The tonnage of fish landed at Milford Haven, Pembrokeshire, in each month in 1962 (circles and solid line) compared with the average for 1962-68 (squares and broken line). The 1962 figures are above the average because of a general reduction in landings in more recent years were more successful than broods of one (82 hatched, 11 fledged), suggesting that food was not short. The slightly lower success (108 hatched, 12 fledged) of broods of three may have been due as much to the inability of adults to protect their young from bad weather as to their inability to feed them (Harris and Plumb 1965). A large proportion of Skomer Herring Gulls obtain their food from the fish docks ten miles away at Milford Haven (Harris 1965) and an examination of the fish catches landed there indicated nothing unusual during the 1962 breeding season which might have explained these results (fig. 2), for landings are normally lower than average in June. It is difficult to see what prevents later breeding (if the 1962 results were typical) or earlier breeding (if atypical) in the resident Skokholm population of Herring Gulls. Since much of the food, including that fed to small young, comes from man, there is no reason to suppose it is in short supply either just before or just after the breeding season. Indeed, as the species is now increasing, food may well be superabundant. Possibly the breeding season evolved in response to a food supply which had remained unchanged until very recently and the species is still in the process of readjustment. DISCUSSION Among sea- and shore-birds, a decline in nesting success with the date of laying has been observed in the Madeiran Petrel Oceanodroma castro (Harris in press), Kittiwake Rissa tridactyla (Coulson and White 1958), Lesser Black-backed Gull L.fuscus (Brown 1967) and American Herring Gull L. a. smithsonianus (Paynter 1949), but not in the Gannet
74 Effect of laying date on chick production Sula bassana (Nelson 1966), Swallow-tailed Gull Creagrus furcatus (personal observations) or British Herring Gull (this study, although Brown 1967 noted a slight fall off in the very latest nests). Owing to less predation, Black-headed Gulls L. ridibundus laying within the peak period have a higher breeding success than those laying early or late in the season (Patterson 1965), but I could find no such correlation in the Herring Gull. Post-fledging survival appears to be affected by the date of laying in the Manx Shearwater (Perrins 1966) and possibly in the Gannet (Nelson 1966), but not in the Oystercatcher. Given that there is some advantage in early nesting, why do birds not nest even earlier? Perrins (1965, 1966) suggested that the females may be unable to obtain enough food to form the egg or eggs. Most Oystercatchers feed on an apparently rich food supply on the shore before breeding (Safriel 1967), but the frequent rough seas around Skokholm during March and April may prevent their utilising this to the full. In the month before laying, female Oystercatchers lose up to 16% of their peak pre-breeding weight (Mercer 1968), probably due to the demands of territorial defence and egg formation. These factors could well put a limit on how early a female can lay. The peak of laying is normally the end of April and early May. One pair which laid exceptionally early in mid-march 1966 lost their young during a period of adverse weather in April (C. K. Britton verbally). This pair has remained unchanged for at least six seasons and is always among the first to lay. In the five seasons for which I have data, six or seven young were reared successfully, much more than average. Possibly this experienced pair is efficient enough to get into breeding condition early, though a genetic effect cannot be ruled out. The few dates I have for individuals breeding for the first time suggest that these are usually late nesters, possibly due to inexperience in protecting a territory and finding food. If food shortage before laying was preventing earlier breeding in some species, it might be that females nesting very early laid smaller eggs than those laying when food was more abundant later in the season. To test this I measured eggs of known date of laying in the Herring Gull (too clutches), Lesser Black-backed Gull (11 clutches), Great Black-backed Gull JL, marinus (27 clutches), Manx Shearwater (97 eggs) and, in Galapagos, Audubon's Shearwater P. Iherminieri (125 eggs), Madeiran Petrel (194 eggs) and Swallow-tailed Gull (427 eggs). In none of these species was there a change in egg or clutch volume with date of laying. Therefore birds appear either to lay an egg of normal size or not to lay at all. This does not prove that food was not limiting earlier breeding, but indicates that egg size is a compromise between any advantage to be gained by nesting early and the need to have a large egg to give a chick which can better survive any adverse conditions at hatching (Lack 1968).
Effect of laying date on chick production 7 5 ACKNOWLEDGEMENTS I am grateful to Dr David Lack, Dr C. M. Perrins and P. J. Jones for helpful criticism of this paper. C. K. Britton kindly supplied some details of the Skokholm Oystercatchers, and the Manager of the Milford Haven Dock Company allowed me access to details of fish landings. SUMMARY Early nesting Oystercatchers Haematopus ostralegus raise, on average, about eight times as many young as late ones do. Once the young are fledged, all stand an equal chance of surviving. In this species even earlier breeding may be prevented by the females' difficulty in obtaining food and defending a territory. The situation is confused in the Herring Gull Larus argmtatus as the species is rapidly increasing in numbers. There is an increase in nesting success with laying date, but this may indicate that the species is still in the process of adjustment to a recently altered food supply. Egg size in sea- and shore-birds does not vary with date of laying. REFERENCES BROWN, R. G. B. 1967. 'Breeding success and population growth in a colony of Herring and Lesser Black-backed Gulls Larus argmtatus and L. fuscus'. Ibis, 109: 502-515. COULSON, J. C, and WHITE, E. 1958. 'Observations on the breeding of the Kittiwake'. Bird Study, 5: 74-83. HARRIS, M. P. 1964. 'Aspects of the breeding biology of the gulls Larus argmtatus, L. fuscus and L. marinus'. Ibis, 106: 432-456. 1965. 'The food of some Larus gulls'. Ibis, 107: 43-5 3. 1966. 'Breeding biology of the Manx Shearwater Puffinus puffinus'. Ibis, 108: 17-33. 1967- 'The biology of Oystercatchers Haematopus ostralegus on Skokholm Island, S. Wales'. Ibis, 109: 180-193. in press. "The biology of storm-petrels in the Galapagos Islands'. Proc. Calif. Acad. Sci. and PLUMB, W. J. 1965. 'Experiments on the ability of Herring Gulls Larus argmtatus and Lesser Black-backed Gulls L. fuscus to raise larger than normal broods'. Ibis, 107: 256-257. LACK, D. 1968. Ecological Adaptations for Breeding in Birds. London, pp. 293-294. MERCER, A. J. 1968. 'Individual weight change in breeding Oystercatchers'. Bird Study, 15: 93-98. NELSON, J. B. 1966. "The breeding biology of the Gannet Sula bassana on the Bass Rock, Scotland'. Ibis, 108: 584-626. PATTERSON, I. J. 1965. 'Timing and spacing of broods in the Black-headed Gull Larusridibundus'. Ibis, 107: 433-459. PAYNTER, R. A., Jr. 1949. 'Clutch-size and the egg and chick mortality of Kent Island Herring Gulls'. Ecology, 30: 146-166. PERRINS, C. M. 1963: 'Survival in the Great Tit Parus major'. Int. Orn. Congr., 13: 717-728. 1965. 'Population fluctuations and clutch-size in the Great Tit, Parus major L.'. /. Anim. Ecol., 34: 601-647. 1966. 'Survival of young Manx Shearwaters Puffinus puffinus in relation to their presumed date of hatching'. Ibis, 108: 132-135. SAPRIEL, U. 1967. 'Population and food study of the Oystercatcher'. D.Phil, thesis, Oxford. Edward Grey Institute of Field Ornithology, Botanic Garden, High Street, Oxford