Age determination and first breeding of British Puffins M. P. Harris At fledging, a young Puffin Fratercula arctica has a small, dark coloured beak (plate 144); several years later when it first breeds, the adult Puffin is much larger and has a big, brightly coloured beak (plate 145), This paper documents the growth and development of the immature Puffin. 144. Head of fledgling Puffin Fratercula arctica showing small beak and lack of eye ornaments; position of cere already visible at base of upper mandible. Isle of May, July 1978 (M. P. Harris) Methods During 1973-79, 5,827 and 2,658 nestling Puffins were ringed on St Kilda, Outer Hebrides, and the Isle of May, Fife, respectively. Most were also given a year-specific colour-ring so that they could be identified in the field. Between 1974 and 198, many of these birds were re-examined by mistnetting them at the colonies (on the Isle of May), collecting corpses killed by Great Black-backed Gulls Lams marinus (on St Kilda) and catching them in burrows (both areas). 246 [Brit. Birds 74: 246-256,June 1981]
Age determination of Puffins 247 145. Head of breeding Puffin Fralercula arctica with three bill grooves. Note eye ornaments and light coloured ridge running diagonally across upper beak. St Kilda, May 1978 (M. P. Harris) On the Isle of May, each Puffin caught was weighed using a Pesola balance accurate to ± 1 g, and a wing measured to the nearest 1 mm using a stopped rule, with the primaries straightened and flattened to give the maximum chord. Three measurements of the bill were made (fig. 1): (a) the upper bill from the tip to the anterior edge of the keratinous cere, (b) the lower (cutting) edge of the upper mandible from the bill tip to the anterior edge of the cere ( = straight bill length), and (c) the maximum bill depth at the base, again excluding the cere. Most measurements were made by S. Murray or by myself and a series of tests showed that our measurements were not significantly different. The cere was excluded from the measurements because experience showed that measurements including it were less repeatable. Puffins killed by gulls had usually been mutilated and were not measured. The number of bill grooves on the outer part of the upper mandible (outside the whitish ridge separating the inner dark area from the reddish Fig. 1. Diagram to show the measurements made of the bill of a Puffin Fralercula arctica
248 Age determination of Puffins distal part) were scored as follows: trace (virtually no sign of a groove), < 1 (an ill-defined groove), 1 (a single distinct groove), 1+ (one distinct, plus the suggestion of a second), IV2 (one, plus a half or more of a second groove), 2 (two distinct), 2+ (two, plus a trace of a third), 2Vi (two, plus a half or more of a third groove), 3 (three complete grooves) or >3 (more than three grooves). The third groove was often ill-defined and difficult to distinguish from the half of a 2V2. Some examples of these categories are shown in plates 145-149. 146. Head of ringed two-year-old Puffin Fratercula arctka with just trace of bill groove. Isle of May, August 1976 (M. P. Harris) 147. Ringed two-yearold Puffin Fratercula arctka with one bill groove. Note irregular outline of top of bill typical of many immatures. Isle of May, August 1976 (M. P. Harris) During 1974-77, the heads of known-age Puffins were photographed using a polaroid camera fitted with a close-up lens, thus enabling checks to be made on the classifications given to the few examined by other observers. For convenience, Puffins were aged by calendar years (i.e. a bird hatched in June 1973 and caught again in 1976 was called a three-year-old, regardless of whether it was caught in April or August). Breeding adults were caught incubating eggs, brooding young or carrying fish. Since Puffins from the Isle of May are significantly larger and heavier than those from St Kilda (Harris 1977), the results from the two areas are analysed separately. Results Measurements in relation to age A total of 192 Puffins of known age was measured on the Isle of May (table 1). The mean wing length, upper beak length and beak depth increased with age until the fifth year of life. Even then, the wing length was shorter than that of breeding adults (most of which would have been at least five years old, see later), though the difference was just not significant (P =.1). The straight beak length did not increase with age after the first year, so the increase in beak depth was alone responsible for the increase in the upper beak length. The beak of an immature Puffin was obviously more triangular
Age determination of Puffins 249 148. Breeding Puffin Fratercula arctica with 2Vi bill grooves. St Kilda, May 1978 (M. P. Harris) Table 1. Measurements (mm) and weights (g) of known-age and breeding Puffins Fratercula arclka from the Isle of May, Fife S.E. = standard error WING LENGTH No. Mean S.E. UPPER BILL LENGTH No. Mean S.E. STRAIGHT BILL LENGTH No. Mean S.E. BEAK DEPTH No. Mean S.E. WEIGHT IN JULY/AUGUST No. Mean S.E. At fledging 1 AGE (YEARS) 2 3 4 5 Breeding in outline than that of a breeding adult (plates 15 & 151), but I did not measure this change. Also, in most immatures,the curve of the upper edge of the bill was not smooth as in most adults, there being either a kink or a steep rise where the culmen met the cere (plates 147 & 149). Some adults also had this rise, but it was generally far less abrupt. There was a gradual increase in body weight with age, but the differences between years were not significant. As a group, however, immatures were significantly lighter (P <.1) than adults feeding young. Number of bill grooves The number of grooves in the bill increased with age in Puffins of both populations up to at least the fifth year, after which the sample size is small.
25 Age determination of Puffins 149. Immature Puffin Fratercula arctica with 1+ bill grooves. Fame Islands, July 1976 (M. P. Harris) 15. Three-year-old Puffin Fratercula arctica with 1+ bill grooves and obviously triangular-shaped beak. Isle of May, June 1977 (A/. P.Harris) 151. Freshly moulted, breeding Puffin Fratercula arctica with three bill grooves. Isle of May, April 1976 (M. P. Harris)
Age determination of Puffins 251 Table 2. Number of grooves on bills of known-age or breeding Puffins Fralercula arclka The breeding Puffins with less than two grooves all had full-sized adult-shaped beaks NO. OF GROOVES Age 3 or Locality (years) Trace < 1 1 1+ 1% 2 2+ 2Vi more There was, however, great individual variation in how quickly individuals attained these grooves (table 2). The three one-year-olds caught had only a trace of a groove (plate 153), but another studied from a distance of 2 m in the field had at least one groove and was indistinguishable from some threeyear-olds. These yearlings had grey-white patches on the sides of the head, similar to, but slightly greyer than, those of older Puffins. Those with bills lacking any distinct grooves, and with poorly developed eye-ornaments, rather drab-coloured eye-rings and the anterior part of the face patch dusky or speckled with dark feathers (plates 154 & 155) were presumably also first-years. Relatively few such Puffins actually land at the colonies, but numbers of them are sometimes seen in the rafts of Puffins swimming below the colonies in July and August. 152 8c 153. Left, head of ringed one-year-old Puffin Fralercula arclka with no bill groove and pale grey face-patch; right, head of probably one-year-old Puffin Fralercula arclka with dusky face-patch. Isle of May, July 1978 (M. P. Harris) Two-year-olds mostly had one or less than one groove. Most three-yearolds had 1 Vi grooves, and some noted as having two grooves were classified as such early in the study, before the importance of noting the completeness of the second groove was realised. Most four-year-olds, and half of the
252 Age determination of Puffins five-year-olds, had two grooves and a trace of a third, while 83% of those breeding had more than two grooves. The ages of the breeding birds were unknown, but most would have been at least five years old. The development of grooves is usually a gradual process, but one individual had half a groove when two years old, and two grooves with a trace of a third when four years old (table 3). Probably most eventually had more than two grooves, as 89 out of 119 birds with two grooves in one year had more grooves the next year. A few, however, may never get more than two grooves, as single 12- and 15-year-olds caught had only two grooves, and Ashcroft (1976) found some Puffins still with only two grooves three years after they had been ringed when showing two grooves. The few breeding birds with less than two grooves all had fully-developed beaks, but lacked grooves. In some cases, the bill was deformed. Only six Puffins out of over 15, handled had four grooves. The individual variation in the number of bill grooves makes it impossible to age any individual Puffin with a high degree of confidence, but the more grooves a Puffin has, the older it is likely to be. Table 3. Changes in number of grooves on bills of 21 known-age Puffins Fratercula arctica caught in two different years on Isle of May, Fife Individual 2 3 AGE (YEARS) 4 3 6 7 8 a b c d 1 1 >/2 1 *\Vi HV 2 *2+ - >2+ >2Y2 g h i J k 1 1V 2 - IV2- IV2- IV2- Wz >2 >2 >2V2 >3 >2»2Y2 >3 n o P q r s t u l'/2 >2 2 S-2V2 1 9-2 1 9-2 2 >2V2 >2+ 2 -^» 2 1%- >2+ 2 3.2Y2 >3 _ Most adult Puffins have a complete moult of the body and wing feathers just before coming to land at the start of the breeding season (Harris & Yule 1977). When they finish breeding, in July and August, their plumage is very bleached and abraded and the bill sheath and eye ornaments are already starting to flake off. Immatures moult much later and usually still have fresh-looking plumage in July (plates 154 & 155), making them easily distinguishable from breeders by plumage. Some adult-billed birds with fresh plumage at this time have almost certainly not bred. Five-year-old non-breeders moult at the same time as adults and usually start to shed the bill sheath before leaving the colonies in August.
Age determination of Puffins 253 154. Probable one-year-old Puffin Fralercula arctica with small eye ornaments and black freckles in front of eye. Isle of May, July 1978 (A/. P. Harris) 155. Two-year-old Puffin Fralercula arctica on Isle of May, July 1978. This bird had been ringed as chick at colony on nearby Craigleith in 1976. In their second and third years, many Puffins visit colonies other than the one where they were reared; some remain to breed, others return to their natal colony. Note that primaries are immaculate even though this is at end of breeding season. Immatures moult in late spring, immediately prior to visiting land; adults moult in winter (M. P. Harris) Age atfirstbreeding Chick survival is high; for example, I recaught 67 out of 187 (35.8%) 1973-young back on the Isle of May in later years. Despite the annual handling of about 25% of the breeding birds and innumerable hours of watching colour-ringed individuals, relatively few known-age Puffins were found breeding (table 4). The youngest was a three-year-old caught in a burrow with a chick, and there was another which probably had a young. Most Puffins, however, did not start breeding until they were five or six years old. Observations of the several hundred colour-ringed known-age birds alive on the Isle of May showed that four- and five-year-old birds
254 Locality Age determination of Puffins Table 4. Ages of breeding Puffins Fratercula arctica YEARS 3 4 5 6 7 8 or older Isle of May St Kilda Fame Islands Skomer Island Westmann Islands 1 5 3 1 7 7 3 4 5 15 1 5 3 12 2 8 14 26 6 returned in late March with the breeding adults. Many of these young birds possessed burrows, but these were usually short or very shallow and unsuitable for breeding. Several four-year-olds with good burrows were soon dispossessed by more experienced birds. Although one three-year-old was caught on 4th April and another on 9th April, most did not return until late April or early May when most breeders had laid. Some other threeyear-olds were seen to enter burrows, but they were very timid and did not defend these holes against other Puffins. Two-year-olds were seen only after the first young had hatched; two of this age caught in burrows had probably been scared by my approach and dived down the nearest holes. This escape behaviour might explain the existence of the skin of an obvious immature (just one ill-defined bill groove) collected by C. B. Ticehurst on Bill's Rock, Co. Clare, and labelled 'caught on the nest'. It is clear that very few Isle of May Puffins bred before they were five years old. Those on St Kilda follow the same pattern of month of return and age of first breeding, but there is no record of a three- or four-year-old Puffin breeding on the Fame Islands, despite 42 known-age birds being caught with an egg or young. Discussion Puffins continue to grow and become heavier until at least their fifth year of life, when many may be old enough to breed (Petersen 1976b; Ashcroft 1976). During this period, the bill does not increase in straight length, but becomes deeper. The bill sheath is shed each winter (Bureau 1877) and renewed just before the start of each breeding season (Harris & Yule 1977). During the first few years of life, each new sheath has more grooves, and a relationship between the number of grooves and age appears to have been first suggested by H. B. Knox in 1891-93 on labels he placed on museum specimens he collected at sea off Co. Mayo. Bird-fowlers, however, in various countries, probably also knew of this (Williamson 197). Salomonsen (1944) came to the same conclusion independently and used the information to separate adults from immatures in his survey of the taxonomy of the Puffin, but it was left to Petersen (1976b) to prove by the use of ringed birds collected at or near colonies that the development of the bill and the number of grooves took several years. He showed that Puffins from the Westmann Islands, Iceland, developed grooves as follows (the number of specimens examined is given in brackets): two-year-olds (11) had one very shallow broad groove, three-year-olds (27) had one deep plus one shallow, broad groove, four-year-olds had two (3), 2V2 (27) or three (6) grooves, five-year-olds had two (2), 2V2 (1) or three (5) grooves. Older birds
Age determination of Puffins 255 156. Left, adult Puffin Fralercula arctica with winter face and beak. Note that, even though bill did not take up breeding form, sheath is starting to flake oil. Isle of May, August 1977 (A/. P. Hams) 157. Right, adult Puffin Fralercula arctica in winter plumage. Note dusky face, lack of eye ornaments, pale cere and pale ridge separating inner and outer parts of bill. Northumberland, October 1974 (.4. M. Taylor) had usually three deep, narrow grooves. He also noted that the third groove in birds five or more years old had a white, chalk-like substance present. Few British birds have a deep, well defined third groove, though many do have the chalky substance (plate 145). Petersen concluded that it should be possible to age two- and three-year-old Puffins with 1% certainty, and that all four-year-old Puffins should be recognisable from older birds by the amount of the chalk in the last groove. Ashcroft (1976) found a less clear-cut relationship between age and the number of bill grooves in Puffins on Skomer Island, Dyfed, as follows (numbers of specimens and their grooves in brackets): two-year-old (lxl + ), three-year-old (4X1 +, lx IV2), fouryear-old (3X2, lx2vi), five-year-old (3x2, 4x2+). Petersen (1976b) thought that this might have been because British Puffins had smaller beaks than those from the Westmann Islands. My results (table 2), however, show a great variation in the number of grooves, despite the Isle of May Puffins having bills as large as those on the Westmann Islands (Petersen 1976b; Harris 1977). Thus, it is not possible to age British Puffins with any high degree of certainty using bill characters, though it is possible to separate most immatures from those old enough to be breeding. The significance of these grooves is not known, but, as Puffins do not normally breed until they have at least two grooves, and the bill is used in display, they presumably have some sexual importance. The few which bred when they had less than two grooves had fully-sized bills, suggesting that they were at least four years old; probably, something had gone wrong during the formation of the bill sheath. A Puffin usually retains the same mate from one season to the next, so they may recognise their mates as individuals and the bill grooves would then be of less importance in display than if they were getting a mate for the first time. Adult Puffins sometimes visit the colonies while in winter plumage (plate 157), but I have only one record of such a bird breeding: one caught taking fish to a chick on St Kilda (S. Murray inlitt.). In winter, the bill grooves, though still present, are far less distinct. The ridge separating the outer and inner parts of the bill is less obvious (plate 156) and often dark, and care must be taken not to include the groove posterior to the ridge in the count. Birds in winter plumage can be divided
256 Age determination of Puffins only into juvenile or first-winter (still in juvenile plumage, bill very small), adults (two or more grooves) and immatures (intermediates). Although few Isle of May Puffins bred before they were five or six years old, Ashcroft (1979) recorded three four-year-olds and five five-year-olds breeding on Skomer Island. On the Westmann Islands, Petersen (1976a) examined 14 known-age Puffins caught with young. Of those, five were aged five years, three were six years, the remainder were older. Only two four-year-olds were found in burrows, and neither was breeding. He also examined the gonads of freshly killed females. At least 14 (and possibly all) of 16 four-year-olds had certainly not bred; neither had two of three five-year-olds. Five older females had all laid. He concluded that most Westmann Island Puffins started to breed when five or six years old. The Puffin population on the Isle of May has trebled during the period of this investigation, suggesting that conditions were very favourable. The Fame Islands can probably not support more Puffins, and the increase in the Isle of May colony is partly due to immigration from the Fame Islands. The apparent lower age of first breeding on the Isle of May and Skomer (where there is also plenty of room for Puffins), as compared with the Fame Islands, could well be due to less competition for burrows. Acknowledgments My thanks are due to the many people who helped with the fieldwork but especially to S. Murray who made many of the measurements. Mrs G. Hickling kindly supplied details of Fame Island Puffins. Dr D. Jenkins criticised the manuscript. The study was partly carried out under Nature Conservancy Council Contract No. F3/3/3. Summary Puffins Fratercula arctica continued to grow and become heavier until at least their fifth year of life. The bill became deeper and the number of grooves on it increased. Although it was possible to separate most immature Puffins, which had less than two bill grooves, from those probably capable of breeding, it was not possible to age British Puffins accurately. Although a single Puffin bred in its third year, most Isle of May Puffins do not breed until they are at least five or six years old. References ASHCROFT, R. E. 1976. Breeding biology and survival of Puffins. Unpublished D. Phil, thesis. Oxford. 1979. Survival rates and breeding biology of Puffins on Skomer Island, Wales. Ornis Scand. 1: 1-11. BIRKHEAD, T. R., & ASHCROFT, R. E. 1975. Auk numbers on Skomer Island. Nature in Wales 4: 222-233. BUREAU, L. 1877. De la mue du bee et des ornaments palpebraux du Macareux arctique, Fratercula arctica. Bull. Soc. Zool. France 14: 377-399. HARRIS, M. P. 1977. Puffins on the Isle of May. Scot. Birds 9: 285-29. & YULE, R. F. 1977. The moult of the Puffin Fratercula arctica. Ibis 119: 535-541. PETERSEN, A. 1976a. Age of first breeding of Puffin, Fratercula arctica (L.). Aslarte9: 43-5. 1976b. Size variables in Puffins Fratercula arctica from Iceland, and bill features as criteria of age. Omis Scand. 7: 185-192. SALOMONSEN, F. 1944. The Atlantic Alcidae. Gbteborgs Kungl. Verterhets Samhalles. Hand. 6: 1-138. WILLIAMSON, K. 197. The Atlantic Islands. London. Dr M. P. Harris, Institute of Terrestrial Ecology, Hill of Brathens, Banchory, Kincardineshire, Scotland AB3 4BY