We understand now that this early visiting of the breeding ground is common to several colonial nesting birds; and not only to seabirds, but to some species of land-birds. In Britain the rook is a familiar example. The fulmar petrel, so rapidly increasing on our coasts, returns to the breeding ledges even in November, but only at those cliffs and islands where it is a numerous breeder; to its small new colonies it returns much later, well after the New Year, and where it is attempting a new colonisation it may not appear until March or April. This early return of colonial nesting seabirds has been instanced in the description of the Manx shearwater; it occurs in the gannet, cormorant, guillemot and razorbill. Gannets return to Grassholm's 7,000 nests in February, cormorants to large colonies in January and February (we have found eggs laid early in March, but small colonies may be two months later), guillemots in December, and razorbills early in February. It is a general rule that the larger the community of one species the earlier the individuals return, suggesting that competition for the best nest sites, and the desire to stake a claim and retain it at the breeding ground, induces this early arrival. But, probably because food is not plentiful in the waters inshore, the ledges are not continuously occupied during those early months. There are days, even weeks, when no visit is made, when scarcely a single guillemot or razorbill can be observed in the neighbourhood of the islands. These disappearances are not regularly correlated with definite phases of the weather - although very cold winds from north and east do sometimes occur at the times when the settlers make one of their temporary disappearances. Thus, from April 5th to 7th inclusive in 1946, there were fresh winds alternating with calms at Skomer, and no guillemots or puffins appeared and hardly any razorbills were seen; on the 8th none was seen on the land and only a few on the sea. A like scarcity occurred on April 18th (fresh north wind) and 19th (south breeze or calm); after that the colonies of these seabirds fluctuated numerically, but by early May some birds were continuously present on the island, and the ledges and holes remained occupied until the August leave-taking. We can only attribute the disappearances in the early part of the season to the already-mentioned scarcity of food in the waters immediately around the island: a conjecture which is strengthened by the fact that none of these birds is seen to be feeding when assembled in rafts on the sea at that time.
They are occupied rather with the affairs of courtship, display and mating. Each species adopts its special pattern. The razorbills seem to enjoy forming a kind of water-dance together: a dozen or so will swim in single file for a very short distance, then quietly turn and face together with beaks and tails elevated, sometimes pairing off and swimming away in different directions, only to alter course and form a rough - but sometimes very straight - line ahead again. The whole manoeuvre is pretty to watch, and is varied considerably.
Guillemots are more individual in their behaviour on the water. They do not dance distinct chain dances, although like the razorbills they often perform a concerted panic dive, or rush over the surface together, half-flying, half- swimming, with breasts raised and feet threshing the top of the water. These infectious mass 'alarms' are sometimes due to the appearance of a boat or a predatory gull, but they also occur without these causes and seem to be, like the 'dread' mass flights of terns, due to the general nervous stimulation of the breeding season, and are set in motion by the least untoward movement of an excited individual.
Guillemot and razorbill mate on the ledges of their cliff and boulder homes. But the puffin, perhaps because the interior space of its under ground home is so limited, mates freely on the water. We watched individual males on several occasions, and found that the cock (recognisable both by his larger head and his behaviour) would pursue one female after another until at last he found one who neither swam away nor dived to escape him, but submitted. In coition the male held position by rapidly vibrating his wings but without holding the female's nape in his bill. This promiscuous behaviour is quite at variance with the behaviour of the male on land. At the burrow (as we found by marking several pairs with rings) the male remained faithful to one mate, one nest, egg and chick, and appeared to be then a model husband.
Skomer's cliffs were exciting and beautiful for the bird- watcher in May, June and July. The three auks filled their ecological niches with a pleasing nicety. The groaning gesticulating guillemots occupied the narrow ledges in open situations on the sheer cliff face, the more silent reserved razorbills mingled with them on the broader rock shelves but preferred the shelter of fissures and broken rock masses, even at times using the entrance to a rabbit or puffin hole in the cliff top. In this way the razorbill formed a link between the extremes of habitat: the open dangerous ledge of the guillemot and the safe dark burrow occupied by the puffin. It was not unusual to find all three birds living very close together. We recollect one such spot in Skomer's cliffs where guillemots occupied the narrow front 'veranda' upon which puffins and razorbills alighted on their way to their respective homes, the razorbills under slabs of basalt, and the puffins deep in crevices behind the same slabs. Below all, on the narrowest finger-holds of rock close to the sea, the kittiwakes had built their nests.
The complex scene was at its best in the late afternoon when the puffins were gathered outside their burrows for the flighting or general assembly which took place regularly at that time of day. Even during the height of incubation both the hen and her mate would leave the burrow and the egg for an hour or so in order to take part in this communal gathering. What useful biological purpose this assembly served in the life of the puffin was, at first, not easy to see we ourselves were convinced that, as the birds did no work (that is, did not dig, feed or mate) but simply sat about or indulged in short circling flights together, they were merely enjoying each other's company, and in an idle manner which was divertingly if unconsciously human. Strolling visits by individuals across the carpet of the sea-pinks to burrows other than their own were a feature of this assembly. As the puffin is almost a silent bird communication by touch and look may be important; certainly there were frequent bill-to-bill rubbings and shakings, and occasional but not serious scuffles and scrimmages. Bits of grass and seaweed would be carried about, sometimes taken into a burrow, but often dropped haphazardly. A large feather seemed to be a special toy, to be carried about with an air of pride, to be stolen from a neighbour, to be used as a line is used in a tug-o'-war. The look of apparent serious purpose in all these seemingly aimless employments was, to the human watcher, extremely comical. It appeared that the puffins were really enjoying this period of relaxation in spite of their preoccupied appearance, which has a certain resemblance to the mien of a pompous alderman at a civic function.
Yet as a result of the wanderings on foot over several square yards of ground the puffin becomes familiar with the terrain in the vicinity of the burrow; and this familiarity must in some degree always be useful to the adult bird. Moreover, the assembly has another function: it attracts to it later in the summer younger birds which reach the island then, birds which were born in the previous summer but which will not breed until the following year. These sexually immature puffins are in adult plumage and are not easy to detect except by their more timid behaviour, which is not like that of the old confident adult in possession of a mate and a burrow; at present, therefore, the proportion of visiting immatures is not known. Occasionally (as in the razorbill and guillemot too) an immature puffin turns up in summer still wearing the first winter plumage, and with the dark face and very pale lemon-coloured legs of winter. We caught one such on July 3rd, 1946; it had no brood spots, but had the usual yellow false 'lips' (the rictal rosette) and blue-grey orbital decorations.
The puffins were the most amusing as well as by their vast numbers the most spectacular of the auks which thronged Skomer's magnificent cliffs. There too we observed the more sober behaviour of the razorbill and guillemot which, talkative perhaps but not willing to wander, each stayed guarding its single egg or chick until relieved by its mate. This watchfulness in such an exposed site was, of course, essential for the survival of the species. The predatory great blackbacked and herring gulls were constantly patrolling the face of the cliff, waiting the opportunity to snatch an unguarded egg or young chick. Certain 'rogue' herring gulls seemed to live by egg-snatching in the season; they became expert at teasing the sitting guillemot, by lunging at it with yellow scarlet-marked bill until the egg was sufficiently exposed (by the sitting bird's defensive movements) to be seized, either whole or, perhaps more often, by a vigorous blow which cracked it and so gave the robber an edge by which it could be gripped and carried off. The great black- backed gull was more dangerous to the young chick, which it snatched up and often swallowed whole, sometimes at the ledge, sometimes gulping it down in the air, and sometimes taking it to the cliff-top or to the sea to devour. These great gulls, too, were the severest enemies to the puffin, which they watched for and seized as it left its burrow.
Man has ceased (on Skomer) to prey directly upon the auks. No longer do the fishermen set their nets at night to catch them for bait for their lines and lobster-pots or for human food; and their eggs are no longer collected for any purpose - they were once used for human and pig consumption. In spite of this there is no evidence of increase in any of the three species. We have shown that the predatory gulls have increased, and what harm they do. There are no comparative figures to show that the vast numbers of the puffins have changed within the last century. But we ourselves noticed a very definite decrease in the numbers of guillemots inhabiting the main ledges of the Wick in 1946. When we had last seen these ledges in the summer of 1939 there had been no gaps in that long line of breeding guillemots; but in 1946 there were three distinct gaps of unused ledge, marked plainly by a green growth of some sort, probably the nitrogen-loving Stellaria media. For want of a better explanation we were inclined to blame these reductions in the ranks of the guillemots to losses at sea through waste bilge oil from oil-burning and oil-carrying ships. The beaches and rocks along the whole coast of south-west Wales were in I946 (and each year after, to the date of this book going to press) smirched with this tarry residue; one could not circumnavigate Skomer in a small boat without seeing numbers of seabirds (but auks especially) dead, dying, or badly contaminated by this horrible filth. The guillemot appeared to be the most frequent victim of the three auks.
It was estimated that there were in 1946 5,000 guillemots on Skomer cliffs at one time, or approximately that number of pairs. Of 47346 guillemots examined (through binoculars) at the ledges, 14 were seen to carry the white bridle or spectacle on the face. This gives a percentage of .32, confirming the figure (0.3) given for Pembrokeshire in the Handbook of British Birds (V, page 156); the lowest for the whole of the British Isles. (The percentage rises in colonies north of Pembrokeshire, reaching over 25 per cent. in the Shetland Isles.)
June was the month of general hatching of auk chicks. In spite of being laid a few weeks earlier than the eggs of the guillemot and razorbill, the egg of the puffin does not hatch very much earlier. It is incubated for six weeks, principally by the female, and the chick grows slowly compared with the chicks of the other two auks, which we found were ready to go to sea after two weeks on the cliffs. The young puffin, however, remains in the burrow for seven weeks.
There were differences in feeding methods too. Although all three auks fed their young on raw fish, the puffin always carried into the depths of its burrow a considerable bundle of small fish crosswise in the bill, but the razorbill more often carried home a single rather larger fish lengthwise, or two or three smaller fish crosswise, in the bill, and the guillemot usually (but not always) brought to the ledge a single largish fish lengthwise in the bill. These fish were caught by pursuit under water, the auk swimming by a series of jerks with wings partly extended, but the primaries were closed and used as paddles, with the webbed feet spread to assist the tail as a rudder in turning and twisting movements. These actions of diving auks could be watched through the clear water by the observer on the rocks above the sea. Small and large sand-eels and the young 'fry' of herring, mackerel and other fish which swam near the surface of the sea in summer were the principal food of the auks. Sometimes a puffin would be surprised into dropping its catch outside the burrow, and we could count as many as 20 or more small fish in one bundle (actual counts of 23, 28 and 29 recorded), far more than had been evident to the observer trying to estimate the number of fish held in one bird's bill. It might be added here that puffins certainly do not hold fish exactly, with tails all to starboard (or to port) of the bill, as some writers on natural history have recorded from loose observation or imagination. We have never recorded such precision; on all occasions when we have tried to count a beakload of fish some tails were showing each side of the puffin's bill. What is, however, remarkable is the ability of the puffin to catch and hold in its bill under the water such a number of fish at one time. It is obvious that its bill is adapted for this purpose. Large, powerful and slightly hooked at the tip, it is a natural gripping tool; there are slight serrations within the upper mandible which must assist it to hold the fish when captured. The throat is comparatively small and narrow, not adapted for the swallowing of bulky food which may account for the fact that no large fish are fed to the chick. The guillemot, with its long pointed bill and more elastic gullet, is able to swallow quite large fish. The razorbill has a bill which might be said to be in size and shape a cross between the pointed bill of the guillemot and the parrot bill of the puffin, hence its ability to capture and hold small fish across, and larger fish along, its bill. These adaptations probably enable the three auks to share out the fish supply without undue competition, but in considering the ecological niches occupied in so vast and rich a feeding area as the sea there are many other and complex factors to be reckoned with, such as distances which the auks fly from land, their respective diving limits when fishing, seasonal and daily movements of fish, and the competition of other species (including fish). The examination of these factors, many unknown or scarcely studied, however, leads to speculation on the dynamics of marine biology. We prefer to return to the factual record.
As there was some disparity between the records of the fledging periods for the razorbill and guillemot given by various observers and quoted in the Handbook, observations on the fledging periods of these birds were carried out in 1946. Joan Keighley made a daily round of the breeding sites of 24 razorbills and 20 guillemots, as well as of 1O kittiwakes. In 1947 she studied 50 breeding sites of the razorbill on Skokholm.
The razorbill thus takes fifty days from the date of egg- laying to the day the chick leaves the ledge. The guillemot probably does the same; although the incubation period of the guillemot is not yet accurately measured, it is known that this auk lays and hatches its egg at the same time as the razorbill. There is an interesting explanation for the very much longer fledging period of the puffin, although it is not quite so clear why the egg should take an extra week to hatch. We can postulate that the earliest hatched egg and the earliest fledged (i.e. strong, fast-growing) chicks of the cliff- nesting auks are best able to survive the hazards of their exposed situation because they are open to its dangers for a shorter period. Therefore these two auks tend to evolve short incubation and fledging periods. This has been achieved in the so-called 'fledging' period by the device of the young birds leaving the cliff (and its dangers of exposure to predators and to accidental falls) long before they are full-grown and before they are independent of their parents. In order to fit themselves for so early a contact with the sea they grow a thick coat of feathers during the first ten days after hatching, and even at that tender age some of the chicks under observation were seen to make a successful 'flight' to the sea, but the average fledging-period, as the tables show, was just over a fortnight. The young birds become restless for a day or so before leaving, and walk up and down the ledge, calling to, and being called by, an adult swimming on the sea below. At last the young bird jumps into space, and, although its primaries are quite undeveloped, it manages to flutter down with the aid of its secondaries and wing- coverts on a long plane to the sea. It may hit a boulder or projection of rock, but its fluttering descent, and its thick wad of feathers and the down beneath them, check momentum and enable it to bounce off unhurt. Once in the sea, even in a heavy surf, it is able to dive deep when threatened by wave or by gull. It quickly reaches an anxious adult (presumed by the observer to be its legitimate parent) which convoys it to the open sea. The adult - usually one but sometimes the pair - will accompany it, protect it from attack by other birds, feed it and help it to find fish. The chick learns to follow its guardian, and we have often seen one of these young birds, now apparently fullgrown, still following an adult as late as November, and have recognised it by its juvenile plumage and, in the guillemot, by its characteristic loud juvenile call-note which it reiterates with monotonous effect.
Very different is the story of the young puffin. In the safety of the burrow there seems to be no such haste to hatch and grow and be off to sea. The long incubation period is followed by a leisurely fledging period during which the young puffin has time to grow to its full size; and actually it becomes fatter and heavier than the adult before it leaves the burrow. So long is this period in fact that the adults seem gradually to lose interest in the chick which they have fed on small raw fishes with such liberality during the earlier weeks of its life. About the fortieth day of its existence it is finally deserted by the old birds, which are probably troubled by the next important physiological change in their annual cycle- the onset of the moult. In late July and early August the adult puffins gradually abandoned Skomer. The young birds remained fasting in their burrows. At night they came out to exercise their wings, as we discovered on our night patrols then. And it was at night (thus avoiding the diurnally active gulls) that the young birds eventually moved independently to the cliff edge and fluttered down to the sea - to swim alone into the wilderness of ocean, and to live there on the reserves of their 'baby' fat until hunger taught them to fish for themselves. Unguided except by that sure and wonderful sense of geographical position which seems to be inherited in this and other species of seabirds, these young birds would find the wintering grounds of the adults and, as ringing has proved, return with them to the islands in the following spring.
With the disappearance of the auks, when the last puffins left in August, Skomer, losing more than half its number of birds, seemed to lose more than half its life and attractiveness