Breeding of the Greater Kelp Goose 25

Some field notes on the breeding of the Greater Kelp Goose PETER GLADSTONE and CHARLES MARTELL

Introduction Furthermore, the eggs of various species Together we spent from November 1964 of penguin are more easily collected in to March 1965 in the Falkland Islands, large numbers—and fresh. South Atlantic, with the purpose of study­ We had evidence that the breeding ing the Greater or Falkland Kelp Goose population was in many places at or near Chloëphaga hybrida malvinarum. It was saturation with a reserve of non-breeding also possible to make observations on the birds. When one female died egg-bound majority of other species of birds found another arrived in the territory and paired in these islands. with the gander within twelve hours. We arrived separately in the islands on When a pair with young were washed 4th November transported as guests of from their territory by a storm, a new the Royal Navy. Peter Gladstone travelled pair moved in within a few days. Apart in H.M.S. Protector, the Ice Patrol Ves­ from the area immediately around the sel. Charles Martell travelled down the town of Port Stanley we found no suit­ west coast of South America in H.M.S. able territory which was unoccupied. It Tiger and from South America to Port is almost certainly the specialized Stanley in the tanker R.F.A. Wave Chief. demands of the bird’s ecology which keep We arrived at our first camp on 12th the population comparatively low. November, the day the first Kelp goslings of the season left their nests. Habitat Our chief area of study was at Port The Kelp Goose feeds almost exclusively San Carlos on the west side of East on green algae on the shore. Very occa­ Island, particularly on narrow Big Island sionally birds feed on short grass and, ( l i miles long), named on maps as Fan­ probably more regularly, on the berries ning Island, but not so known locally, of ‘ Diddle-dee ’ Empetrum rubrum in near Fanning Head. We also made obser­ the austral autumn, again nearly always vations round the coast of Port San Car­ close to the shore. los, which covers an area about the same The breeding territories are found on as the Isle of Wight, on the north end rocky shores where the fine green algae, of Great Island in the Falkland Sound, like Enteromorpha, grow in the inter­ on West Point Island, Carcass Island and tidal zone, on comparatively calm coasts. New Island to the west. Thus breeding Kelp Geese are found in We were probably lucky in visiting two bays, in sheltered sounds, in areas where areas where the Kelp Goose is particularly there is a wide shallow underwater shelf, abundant, Great Island and Fanning or, frequently, where the shore is pro­ Island. Roddy Napier, a leading ornitho­ tected from the main waves by the blan­ logist who has spent his life in the islands keting effect of the vast kelp seaweed and knows them well, told us he knew Porphyra umbilicalis. This grows to of no other areas with so high a popula­ great length and the fronds often cover tion. the surface of large off-shore areas. The In spite of the fourpenny bounty on birds can even find territories on the the bills of other ‘ geese ’ the Kelp Geese north coast, where kelp abounds, though suffer little from persecution by man. there is no land between there and the The bills are not acceptable for the British Isles and the sea is seldom calm. bounty and are easily recognised, the It has been recorded by Boyson (1924) female’s being pink and the male’s having and by Cobb (1933), and frequently re­ a white spot on the upper mandible—a peated, that the birds feed on this sea­ fact that may well have saved the bird weed. This is certainly not the case. from near extinction. Other factors Their basic requirements appeared to be favouring their survival are that the eggs at least seven yards width of intertidal are laid rather later than those of the zone (tides average seven feet in height) other Chloëphaga species, at a time when with enough algal growth to make it lamb marking and castrating keep the appear greenish, rocks which will not local inhabitants busy; the nests are move in storms and so destroy the algae, usually less accessible to human settle­ some nest cover close to the high tide ments than tnose of the Greater Magel­ mark, and shelter or a blanket of kelp. lan Goose Chloëphaga picta leucoptera During the breeding season territories and the eggs are said to be foul-tasting. are closely defended by both sexes, the 26 Wildfowl male being seen to chase away intruding are a slightly creamy white. The shape males and females, and the female other varies very considerably, some being long females. The size of the territories varies and thin but the majority tending to a considerably with local conditions. A blunter ovoid. single pair may be alone in miles of coast­ The clutch sizes of 44 nests found in line because there is only one small bay which laying was complete and where with the necessary conditions. In other there was nothing to suggest disturbance places nests may be as little as fifty yards or that they were second clutches were apart, though seldom does a territory as follows: cover less than a hundred yards of shore­ Clutch size 6 5 4 3 line. No. of nests 21 17 5 1 Nests We heard that a clutch of eight has We found 56 nests in use and a number been seen but could not get details. which had been in use before we dis­ Clutches of seven eggs do appear occa­ covered them, or were from previous sionally. Mr. Napier found one on West years. Some of the latter were close to Point Island in 1957 but he considered present ones, suggesting traditional attach­ it unusual. Pettingill (1955) records one ment to sites. Indeed in some cases we on Kidney Island. Boyson (1924) recorded found several layers of old down and clutch size as 2 to 4. This is suggested as grass, one on top of the other. the normal size for the Lesser Kelp Goose Nest cover varied from a dense cover Chloëphaga hybrida hybrida on off-shore of ten-foot high tussac grass Poa flabel­ islands in Chile by Johnson (1955). We lata to planks of old sailing ships (which found two newly-hatched broods of littered the coast and provided much of seven young after a storm in territories our fuel for cooking and for heating hot- where neighbouring adults were missing water bottles—for our tame goslings). and there may have been amalgamation, However, with practice, we could usually also two broods of seven well-grown pick out where the nest would be in any young in February on New Island. Mr. given territory. It would be partially hid­ Napier saw a brood of nine on Dunbar den, but allowing the highly cryptic Island in 1957. female to have a view out. Usually the We measured the eggs of 10 clutches, bird could walk at least a couple of yards 48 in all, and found the average to be to the nest, though she might have to 82.5 X 55.6 mm. (ranges 78-89 X 53-58). fly direct to a small hole in the vegeta­ One exceptional egg from an otherwise tion. All nests save one were within ten normal clutch measured 105 X 65 mm. yards of the high tide line, and most were Schcnwetter (1960-1) gave the measure­ closer than that. The exception was 100 ments of 27 eggs as 83.5 X 54.0 mm. yards from the shore of Great Island, (75-91 X 53-58). Boyson (1924) gave against the first bit of cover inland from ranges of 80-84 X 55-58 mm. an otherwise perfect feeding territory. When we arrived on 4th November Where available a ledge amongst stunted most clutches were well incubated and tussac grass four to eight feet up a little we did not get a chance to discover the cliff was preferred. On Great Island, incubation period. (Delacour (1954) gives where sheep have exterminated the tus­ it as 30 days.) From the first sign of sac grass except on such cliffs, ground chipping, individual eggs hatched within nests were beside drift-wood, an out­ 12 to 48 hours and the young normally standing boulder, amongst Empetrum, or left the nest together some hours after in the coarse white grass Cortaderis the last eggs hatched. In the majority of pilosa with which most of the islands cases this was about 48 hours after we are covered. had found the first egg chipping, but The scrape was lined with any avail­ sometimes considerably longer. Cobb able vegetable matter and down was (1937) states ‘ . . . young leave the nest added when the last egg was laid and one by one as they hatch . . . ’ and ‘ . . . incubation started. Where incubation was but the Kelp Ganders just sit about or advanced there was normally a consider­ feed, and do not appear to care if they able quantity of grey down and the eggs lose the lot or not. This is perhaps a were covered with this when the female good thing, or the place would become left to feed. The white male does not smothered in Kelp Geese ’. We saw one take any part in incubation and has no newly-hatched young tended carefully by brood patches. a gander on one occasion whilst the goose sat up a cliff on a nest and Cobb’s Eggs and clutch size description is not of a normal situation. The fresh eggs have a greasy texture and The first young to leave the nest did so Breeding of the Greater Kelp Goose 27 on 12th November and the majority of wild and amongst our tame birds. The nests had hatched by 25th November, actual pre-flight period for any individual though several hatched some days later. bird was not measured; it varies con­ A few still had eggs on 5th December. siderably but is always more than twelve One nest had four eggs which started weeks. The wild goslings fed on fine green chipping on 9th January and the young algae of the Enteromorpha type, and a left the nest on 14th January but we saw little on similar brown algae. Our birds no evidence of other late broods. We fed on turkey starter crumbs which we suggest that reports of later hatching had imported, short grass, leaves of dan­ may be derived from underestimation of delion Taraxacum officinale, an intro­ the ages of the young due to their ex­ duced plant found round settlements, tremely slow growth rate, particularly porridge oats, flour and anything else during the first few weeks. available. Some even had hard-boiled eggs of Rockhopper Eudyptes crestatus and Goslings Gentoo Penguins Pygoscelis papua. As We got to know the goslings intimately many of the local inhabitants feed their as the two of us lived in a tent 4 ft. 4 in. chickens and tame ducks and geese on by 6 ft. 6 in. with all our equipment carcasses of cattle and sheep we reckon and reared 25 goslings in it. This was our birds did well! We lost a few by not always an easy task as there were chilling in spite of taking them into our frequent gale - force winds often with sleeping bags, one was lost down a near-horizontal hailstorms. We also had to covered hole (together with our only guard the birds against the attentions of young Flightless Steamer Duck Tachyeres Cassin’s Falcon Falco peregrinus cassini brachypterus) and one was killed by and the local Buzzards Buteo polysoma or Martha. Buteo erythronotus—we confirmed what The slow growth rate was worrying to had been suspected before that the Blue both of us as we had had experience of Buzzard is the male and the Red-backed hand-rearing other waterfowl. The white Buzzard the female of the same species, as down soon becomes greyish and the first we found two nests and saw several other feathers showed through after 17 to 25 pairs. Johnson (1965) mentions no sexual days. This period varied with the weather dimorphism for the Red-backed Buzzard and hence the available feeding time, both B. polyosoma in Chile. The Falkland bird for our birds and those living under may prove to be a distinct species. If so natural conditions. The first feathers of it is one of the rarest birds in the world. the females appear before and are darker We also had to protect our ‘ family ’ than those of the males. It is easy to dif­ from the attacks of the Greater Magellan ferentiate between the sexes some days Geese which held the territory in which before the feathers appear as the females we placed our tent. ‘ Martha ’, the female start to get a croaking edge to their voices, of this pair, came each morning to attack the first sign of the musical ‘ qwa ’ of the the goslings round the tent. She also adult females. The normal note of the stole our porridge whilst we were pro­ young is a double ‘ cheep ’, and as the tecting the goslings, being completely female ‘ qwa ’ appears the males develop fearless of human intervention. Picking a triple ‘ cheep ’ which is later further up Martha and enclosing her in our little multiplied and develops into the adult pen whilst the goslings grazed round the male’s whistling ‘ s i’. tent solved this problem at times—the In getting the birds home we had great Lorenz Inverse Cage Law in reverse! A help from Ian Strange, the crew of few near-hits with a red plastic bucket R.M.S. Darwin and the firm of Maclean made the buzzards wary of it even when and Stapleton of Montevideo. We made it was slung flapping on a drift-wood pole cases of weldmesh lined with sacking, and near the tent. having long since run out of turkey The newly-hatched goslings are de­ starter we fed the birds on dandelion, lightful little white birds with pitch black cabbage and porridge. From Port San feet and bills. The egg-tooth is white. Carlos to Port Stanley we travelled on They have sharper claws than any other the Darwin, the local steamer, on the after Anatidae we know and stumble over and deck. This was an easy trip although it round the rocks following their parents. was necessary to struggle through sheep They are only able to feed at compara­ to reach the birds. From Stanley to Mon­ tively low tide for a few hours each day, tevideo we had five days of hell, the only the actual period depending on the available space being on the forward deck weather and the direction of the wind. and the sea being rough. Tending the This may explain why they grow so goslings, together with the other ducks, slowly, a feature we noticed both in the geese and penguins which we brought 28 Wildfowl back, was a task not lightly to be under­ Causes of losses taken again! Before the trip we had tried It was little encouragement to two men to inject the birds against aspergillosis in a tent to be told that they had ex­ but unfortunately the drug proved to be perienced the worst summer in living too granular for normal syringes. Ham­ memory in the Falkland Islands. The pered by this and the prevailing weather weather in these islands has not a pleasant conditions, which were such that even reputation and during November and the Darwin’s sailing was delayed two days, early December the temperature was we had to abandon the attempt after constantly low with a little frost some succeeding with three birds. At Monte­ nights. The wind was often gale force and video we had fifty-six hours in a heat­ a spade was as useful as armour plating wave feeding, cleaning out, repairing against hailstones as for its accustomed crates and keeping the birds cool with purpose during the calls of nature. How­ an ever-running fire hose. ever there were glorious short periods By plane from Montevideo was an easy when the wind dropped and we did wit­ 17 hours though we were worried by the ness better weather later in our stay. rise in temperature of the hold during a We saw no predation of Kelp Geese delay in Rio de Janiero. We then syringed in the wild, but major gosling mortality water forcibly into the birds using a long­ that season was easily ascribable to necked garden syringe. This had proved storms. These were so violent at one invaluable on many occasions, particularly stage when the birds were hatching that when it was too rough to leave water in we were unable to move round the south the crates. We landed at Gatwick in side of Fanning Island even at low tide. snow and soon had the birds on the Land These conditions lasted several days and Rover, to arrive at Slimbridge at dawn to we then found one colour-ringed pair a a right royal welcome. We released the mile across rough sea from their terri­ birds looking as if they had never been tory, with two dying young. crated. It is sad to relate that in spite of Of 166 eggs we found on Fanning Island devoted care and the use of drugs every or the opposite shore, about eight did bird had died of aspergillosis within a not hatch. We removed 25 newly-hatched few months. The January-hatched birds, goslings from this area leaving about 129 which we had collected, survived best and in the natural state. There were probably were easiest to handle on the trip although two broods (about 11 young hatched) they had only just started to feather when from nests we did not find. Of these 140 we left. young less than 20 were alive at the end The goslings became well imprinted on of February. This 14% survival rate may us which made caring for them much be exceptionally low, but it would prob­ easier than it otherwise might have been. ably have been but little higher if we They took the vicinity of the tent as had left all the goslings hatched. Few their territory and seldom wandered more pairs had complete broods left by Febru­ than twenty yards. In the wild they are ary when we removed two or three (contrary to Cobb (1933)) attended by others. In a storm the parents do not both parents and clearly imprinted on appear able to protect stragglers, and both (Plate 1b, facing p. 28). It is note­ always stay with one or more young, so worthy that they are able to imprint on all our captives might have been surplus. both parents despite such dissimilar In some other areas the survival rate had plumages and voices. There are few other obviously been higher as we saw well- Anatidae where the parents show sexual grown large broods, but many had dimorphism and both look after their apparently been heavily reduced to one young and there is no other species in survivor and many adults had lost all which the sexual dimorphism is more their young. The survival rate of our strongly marked. The female normally tame birds was about 76% from hatching guards the young and when she is present to Slimbridge. they follow her, keeping close behind on We heard reliable accounts of goslings the side opposite danger. If the female is being attacked by Dominican Gulls Larus removed from the brood the young follow dominicus, but did not see this. Great the male, creeping away along the shore, Skuas Catheracla skua antarctica and as closely as they normally follow the mackormici are numerous in some areas female, but when the female is released and probably attack goslings. We found they again switch to her. The female two Kelp nests with eggs apparently eaten usually broods the young, although if by Brown Rats Rattus norwegicus. The the male is sitting down they sometimes parent birds were not alarmed by the snuggle under him. large Carancho Caracara caracara though φ <

Philippa Scott Plate I. (a) A pair of Greater Kelp Geese Chloëphaga hybrida malvinarum photo­ graphed in the Falkland Islands during the first British tourist excursion to Antarctica. (b) Young Greater Kelp Geese following the adult male, the female was bringing up the rear with two other young. (See p. 28) Philippa Scott Philippa Scott Plate II. (a) A pair of Falkland Island Flightless Steamer Ducks Tachyeres brachyp- terus photographed in the wild. (b) The first Magellanic Flightless Steamer Duck Tachyeres pteneres to be bred in captivity, at 14 days old. (See p. 32) Zoological Society of Philadelphia Breeding of the Greater Kelp Goose 29

these were constantly on and around some when they are flightless. We did not, un­ of the territories. Sea Lions Otaria byro- fortunately, see their reaction when sur­ nia and Leopard Seals Hydruga leptonyx prised with young by the local sheep would certainly take any young or adults dogs. When a dog is near them they they could catch, but the latter is now simply made for the water. uncommon round the islands and the Kelp Geese fed quite close to parties birds never move voluntarily into water of young Sea Lions and Elephant Seals in which seals could swim. Few birds Mirounga leonina. On Carcass Island a except Scoresby’s Gulls Gabianus scores- brood fed between the water and moult­ byi, Sheathbills Chionis alba, Turkey Vul­ ing bull Elephant Seals of several tons tures Cathartes aura and some passerines weight even when these monsters were are seen around Sea Lion colonies. The wallowing their way ashore a few yards gulls and vultures feed on dead seal pups away. A pair was seen feeding close and afterbirths and the Sheathbills, which around a dead Sperm Whale Physeter look like angelic fantail pigeons in flight, catodon, 51 feet long, and they showed feed mainly on the seal faeces, starting no reaction to White Dolphins just off before the evacuation is complete. Obvi­ the rocks. ously the presence of a seal colony elimi­ It is interesting to note that none of nates the possibility of a Kelp Goose terri­ the geese showed alarm at the helicopter tory. The buzzards appear to feed largely (in which we were privileged to make on passerines and we did not see them several excursions). This contrasts mark­ taking goslings though they constantly edly with the reactions of other species hovered over our tent whilst our tame of geese, in Europe for instance. birds were small. No gosling remains were found at their nests. Cassin’s Falcon Behaviour is reported as taking young chickens from When a trespassing bird invaded a terri­ setdements but at Cape Dolphin two tory, the territory-holding gander flew adults and a fledged young shared a terri­ low over the ground towards it and tory with a brood of well-grown Kelp landed facing it a few feet away. He goslings. Here the falcons were feeding landed with his chest raised and puffed on Belcher’s Prion Pachyptila belcheri out and his head tucked in above it, with which were often in large flocks within the tail lifted to make an angle of about sight of land. We saw a male Kelp drive 130° to the ground. This is very similar off a Turkey Vulture which landed in its to the aggressive display of the male territory. Andean Goose Chloëphaga melanoptera. The mysterious Falkland Island Fox On landing he maintained this posture which early explorers including Darwin whilst running direcdy towards the in­ report as having been abundant in the truder with his wings slightly open, pos­ islands may have been a danger in the sibly to display the calouses, possibly to past. Patagonian Foxes have been intro­ add to his apparent size, but certainly to duced to a few of the western islands of aid his balance over die rough rocks. the group to kill penguins and geese with Throughout the aggressive run the attack­ the faulty idea of aiding the sheep. How­ ing male repeats a 4 si-si-si ’ whistie, ever they turned on the sheep and have frequendy composed of three, seven or cleared them from some islands. There eight syllables although any number up are no other wild land carnivores except to and including nine was heard, at con­ the feral cats which we never saw on the stant pitch and at a rate of about five shore-line though they inhabit the wildest notes per second. In many cases that we areas. We found one adult goose that saw, the intruder invariably retreated, died egg-bound. Two dead ganders were running if not near water, then flying found eaten by scavengers near a deserted low, nearly always over water a few yards nest on Great Island. It is possible that off-shore. When moving the birds prefer these had been killed by a Magellan to fly over water, but seldom many yards gander in whose territory the nest was from the shore. Kelp Geese are short­ situated. winged and not strong fliers. We never The Kelp Geese in many areas show saw them fly more than forty feet above no fear of man and it is possible to catch the ground or water. They very seldom them with a fish landing-net, but when even flew over Fanning Island, 30 feet they have young they often take to the high and 100 yards wide, after a visit to water between and round the rocks on the 4 mainland ’. Instead they would make the approach of humans. (The birds nor­ a detour right round the island, each mally dislike getting their feet wet.) It paired gander displaying as they passed. is difficult to get within ten yards of some The female will also chase straying non-breeders on a beach and of all adults birds in the same posture as the male, 30 Wildfowl with her white patch on the wing-joint lished when we arrived. There was a little showing (as it does every time she lands). squabbling in non-breeding parties. She resembles an old woman with an When the female left the nest either to enormous swaying front angrily chasing feed or when disturbed she flew at once chickens, as she stumbles over the rough to the male who was standing guard some ground, with flank-feathers flapping like distance away, landing sideways on, an apron in the wind. On one occasion almost touching him. She immediately an incubating female was seen to leave took up a breast-down position with the the nest and drive off a stray female to tail right up, similar to that seen in the whom her gander was making advances. aggressive run, with the white wing-joint When a paired female was picked up well forward and very distinct. Mean­ it was usually easy to catch the male as while she gave a repeated musical ‘ qwa- well as he either attacked us or displayed qwa-qwa ’ (frequendy three notes, vary­ violently within a few feet. On some ing to seven), and the male would almost occasions we picked up the male whilst fall over backwards with an upright pos­ handling the female and threw him into ture, his neck pumping up and down and the air several times to be rid of him, his chest puffed up, uttering a ‘ si-si-si ’ but he kept coming back to display at us whistle, seven or eight syllables being the and to attack. If there were young present most usual, at constant pitch but sound­ the male usually slunk off along the shore ing more sing-song than when he is instead, with the brood following. When aggressive. This whistie of the male is less we caught a male the female stood around fluting than that of a Magellan gander. looking bewildered, though again if there The birds would move and repeat the was a family she would make off along performance several times until the female the shore with it. When we caught the either flew to fresh water to drink or young the parent birds showed no sign started to feed. On Fanning Island there of aggression but rather appeared to try was no fresh water except temporary rock to call the young from us, keeping mean­ puddles and some birds went over to while within a few feet. The female drink at rivulets on the coast of East would give a constantly repeated single Island. Birds also appeared to drink more ‘ qwa ’ note, the same as when leading the at dawn than at other times of day, young. It may be that the female did not though there were many birds that we recognise humans as dangerous animals never saw take water. The male seldom but rather reacted as if the young had ate when the female was on the nest. got trapped behind a rock. On these Some pairs flew to a neighbouring non­ occasions the male’s call seldom had more territorial beach to feed when the female than three notes and sometimes only one came off. or two. We did record up to eight notes We could see no significant difference a few times. between the greeting display and the pre­ Occasionally a brood of Greater Magel­ sumed triumph display which occurred lan Geese would wander on to the shore regularly when the male returned to the in a Kelp’s territory. On these occasions female after a chase. each gander would display at the other, Although we witnessed and filmed dis­ but neither responded to the other’s traction display in the Greater Magellan aggressiveness by flight and there was Goose, we did not see it in Kelp Geese. some fighting, the caloused wings being Several times we saw display with cal­ brought into use. Once a Kelp female was ling as the male flew close behind the seen to drive off a Magellan female. Kelp female low over the water, especially Geese took little notice of Ruddy-headed early in the morning as the pair flew Geese Chloëphaga rubidiceps, Flighdess back from drinking. Here the neck pump­ Steamer Ducks T achy eres brachypterus, or ing was semi-horizontal although there Crested Ducks Lophonetta specularoides was considerable vertical movement, with specularoides (which are possibly distinct the chest again puffed forward, making from the South American birds). These flight appear difficult. The male called might well all share the territory together (usually seven syllables) and the female with a pair of Black Oystercatchers Hae­ answered during this display. matopus ater and a pair of pied Gamot’s Oystercatchers Haematopus leucopodus. Non-breeding movements Magellan Penguins Spheniscus magellan- The non-breeding birds form flocks and icus were constandy on the shore in Kelp move about depending on the conditions territories and the birds took no nonce of the weather and the availability of c sea of each other. Intra-specific fighting was lettuce ’, bright green and similar to luche not observed, one bird always acting re- Ulva, growing below high tide level. cessively, but territories were well estab­ This is normally washed up on shores Breeding of the Greater Kelp Goose 31 not used by breeding birds. It does not with black feathers, one with a thin appear to occur on the most exposed broken line down the back of his neck, beaches where there was a lot of drift­ one with some black tail-feathers and wood. The birds do not often move as a one with a black primary. One female of flock, but rather in small parties or singly. a territory-holding pair had black feathers Several times we watched this food on her rump. There were pure white accumulating on the beach near our tent males in non-breeding groups and these and the number of birds would gradually tended to be the most aggressive. They increase until there were more than 50. may have been adults which were unable On 10th February we saw one flock of to secure a territory. On the whole the legs 316 Kelp, all over a year old and many of the birds which were not breeding flightless, on a beach on New Island. were less bright in colour than those of The first flightless bird we saw was a the breeders, and this may be a method male on 26th November. Throughout of telling fully mature from other post­ males moulted earlier than females though juvenile birds. In the areas we visited there was considerable overlap. there were more birds in flocks than there There is some winter movement round were holding territories. the coasts, but we were unable to ascer­ tain its extent. Apparently some of the Acknowledgements birds remain in their summer haunts. One of us (P.G.) was given leave of They occur regularly in the bay of Port absence from Shrewsbury School and a Stanley and we saw some there in early Travel Award by the Goldsmiths’ Com­ March before we left. pany. We owe a great deal to the Cap­ Although the birds normally keep their tain and crew of H.M.S. Protector·, of broods within their territorial limits until H.M.S. Tiger, of R.F.A. Wave Chief and fully grown, they are occasionally driven R.M.S. Darwin and to Admiral Sir Hugh from it by storms and then some young Martell, to Dr. S. Slessor, Mrs. White, may become separated to join another Ian Strange at Port Stanley, to the Napier family. Though this behaviour could be family of West Point Island, the Bertrand considered accidental and may be un­ family of Carcass Island and the Davies common, we found one pair of adults family of New Island. Every inhabitant feeding on grass beside a sandy beach of Port San Carlos settlement gave us with twenty-five well-grown young of unstinting help, particularly Bill Hogarth varying ages. The area did not seem suit­ who fed the goslings whilst we toured able for a breeding territory and we can other areas. Above all we must thank the only conjecture how they arrived there. Cameron family. No better hosts than Also this beach was sometimes infested Mr. and Mrs. Norman Cameron could with Leopard Seals which eat sea birds be found anywhere; we camped on their freely and have even attacked swimming ground, spent Christmas with them, and dogs, cattle and men. made free use of their home whenever While breeding males were white, nearly we wished. Everywhere we met kindness all non-breeding ones had one or more and help. We also thank Dr. G. V. T . black wing - feathers, usually primary - Matthews for many helpful comments on coverts. We saw three breeding males the manuscript.

References b o y s o n , V. F. 1924. The Falkland Islands. Oxford : Clarendon Press . CAWKELL, e . M. and J. E. HAMILTON. 1961. Falkland Island Birds. Ibis 103a : 1-27. Co b b , A. F. 1933. Birds of the Falkland Islands. London : Witherby. De l a c o u r , J. 1954. The Waterfowl of the World. Vol. I. London : Country Life. J o h n s o n , A. w . 1965. The birds of Chile and adjacent regions of Argentina, Bolivia and Peru. Vol. 1. Buenos Aires : Platt Establecimientos Gráficos S.A. PETTINGILL, o. s. jr. 1965. Kelp Geese and Flightless Steamer Ducks in the Falkland Islands. The Living Bird 4 : 65-78. schönwetter , M. 1960-61. Handbuch der Oologie, ed. W. Meise. Parts 2 and 3. Berlin : Akademie-Verlag.

Peter Gladstone, Shrewsbury School, Shrewsbury, Shropshire. Charles Martell, Long Reach, Bosham Hoe, Chichester, Sussex.