REPRODUCTIVE BEHAVIOUR of the EARED GREBE, Podiceps Caspicus Nigricollis by NANCY MAHONEY Mcallister B.A., Oberlin College, 1954
REPRODUCTIVE BEHAVIOUR OF THE EARED GREBE,
Podiceps caspicus nigricollis
by
NANCY MAHONEY McALLISTER
B.A., Oberlin College, 1954
A THESIS SUBMITTED IN PARTIAL FULFILMENT OP
THE REQUIREMENTS POR THE DEGREE OP
MASTER OP ARTS
in the Department
of
Zoology
We accept this thesis as conforming to the required standard
THE UNIVERSITY OP BRITISH COLUMBIA
April, 1955 In presenting this thesis in partial fulfilment of the requirements for an advanced degree at the University of British Columbia, I agree that the Library shall make it freely available for reference and study. I further agree that permission for extensive copying of this thesis for scholarly purposes may be granted by the Head of my Department or by his representative. It is under• stood that copying or publication of this thesis for financial gain shall not be allowed without my written permission.
Department of The University of British Columbia, Vancouver 3, Canada. ii
ABSTRACT
The present study describes and analyses the elements involved in the reproductive behaviour of the Eared
Grebe and the relationships between these elements. Two summers of observation and comparison with the published work on the Great Crested Grebe give some insight into these ele• ments, their evolution, and their stimuli.
Threat and escape behaviour have been seen in the
courtship of many birds, and the threat-escape theory of
courtship in general has been derived from these cases. In
the two grebes described threat plays a much less important role than the theory prescribes, and may not even be important
at all. In the two patterns where the evolutionary relation•
ships are clear, the elements are those of comfort preening.
In the three patterns in which there is still some doubt, all
the elements can be related to general non-reproductive be• haviour or to nesting patterns.
When courtship tapers off, nesting behaviour takes its place. The timing of this change and of the nesting be• haviour patterns is caused by a general climatic stimulus fol• lowed by a more particular behavioural one. Nest-building and egg-laying are the results of the gradual building up of the sexual condition during courtship. Egg laying, however, takes place only in warm, dry weather. If the weather changes for
the worse during laying, the birds stop and wait. The effect of this double stimulus is to put the birds on the borderline between determinate and indeterminate layers. V
ACKNOWLEDGMENT
Br. I. McT. Cowan deserves my greatest thanks for his valuable suggestions and enthusiastic support. My thanks are also due to Dr. M. D. P. Udvardy for his help with the literature and the manuscript, to Dr. W. S. Hoar, Dr. J. P.
Bendell, M. T. Myres and Donald McAllister for reading the manuscript, and to Esther Marion and Mary Jackson for assist• ance in the field work.
Financial support came for the National Research
Council of Canada. Ill
Table of Contents
Page
Introduction 1 Methods 2 Organization 4 General Behaviour Patterns 6 Plight 6 Swimming 7 Resting 7 Diving 8 Bathing and preening 9 Voice 11 Comparisons of the general behaviour ...... 11 Threat and Escape Patterns 12 Occurrence of threat and escape 13 Threat attitudes 14 Alarm 16 Escape 17 Courtship Displays 18 Advertising behaviour 19 Courtship preening 21 Head shaking 22 Penguin dance 24 Cat attitude 27 Origins of the Courtship Patterns 28 Advertising behaviour 29 Courtship preening and head shaking 29 Penguin dance 31 Cat attitude 32 Summary 33 Nesting Behaviour 34 Soliciting 37 Development of platform building 38 Nest building patterns 41 Copulation 43 Abandoned eggs -. 45 Clutch size 46 Interval between eggs 51 Incubation 52 Care and Behaviour of Young 55 Predation and Mortality 57 Outlook 58 Literature Cited 59 LIST OP TABLES AND FIGURES
Page
Table I Water temperatures July 13, 1956 3
Table II Estimated Breeding Birds on Westwick and
Sorensen Lakes 4
Table III Courtship Patterns Compared 30
Table IV Clutch Size 48
Table V Clutch size in Experimental Nests 49
Pigure 1 Map of Sorensen Lake 60
Figure 2 Map of Westwick Lake 61
Pigure 3 Sorensen Lake from the south end looking north and showing in the foreground a Scirpus marsh 62
Pigure 4 Sorensen Lake from the east side looking west toward the mountains at the edge of the Praser River 62
Pigure 5 Westwick Lake from the east side near the north end, looking south toward the narrow part of the lake 63
Pigure 6 Westwick Lake from the middle of the east side looking south at the large part of the lake 63
Pigure 7 Duration of Behaviour Patterns 64
Pigure 8 An early courting platform of Scirpus and pond weeds 65
Pigure 9 An area of sparse reeds showing five courting platforms 65
Pigure 10 A later and more substantial courting
platform of reeds and decaying algae 66
Pigure 11 An abandoned egg on a courting platform 66
Pigure 12 An abandoned egg which has been pecked by the grebes 67 Pigure 13 Drawings of behaviour attitudes ..68 Introduction
The comparative behaviour of birds has been studied
only in the last forty years $ and attention has been limited
to the more advanced groups of birds. The general rules de• rived from these studies may, then, be more restricted in
their application than was at first supposed. Studies of more primitive birds, such as this of grebes, might well add to the understanding of the evolutionary development of innate be• haviour. Special consideration is here given to the develop• ment of territorial and colonial behaviour and the theory of
courtship.
The reproductive behaviour of grebes is known only
through scattered observations over a few days or weeks, and
the greatest confusion has resulted. Many individual behav• iour patterns have been described, but their places in the
species behaviour are not known. The one exception to this is the British work on the Great Crested Grebe, Podiceps
cristatus, from Huxley (5) to Simmons (10). The work is
carefully and exhaustively done, and I have drawn on it
heavily for comparative purposes. My observations are not
influenced by the British work, however; I did not read the
Great Crested Grebe material until my field work was complete.
This project on the Bared Grebe, Podiceps caspicus nigricollis.
is the beginning of a comparative study of North American grebes. Methods
The study area is near Williams Lake in the Cariboo district of British Columbia, 52.00° North latitude, 122.00°
West longitude, and 1500 to 2500 feet above sea level. The
Cariboo Parklands biotic area or biome as described by Munro and Cowan (9) is flat to rolling grassland broken by stands
of lodgepole pine (Pinus contorta), Douglas fir (Pseudotsuga menziesii), and aspen (Populus tremuloides). Frequently the
bottoms of depressions in the grassland hold shallow lakes.
The lakes vary in size from a few acres to twenty square miles. The Eared Grebes choose lakes about a mile by half a mile. The two lakes intensively studied are typical of these
last.
The study lakes, Westwick and Sorensen, were once
one but are now artificially divided by an unpaved road. The
sizes and proportions of the lakes are given in Figures 1 and
2. Their general setting is shown in Figures 3 to 6. The
greatest depth is twelve feet, but more than half the lake
area is less than five feet deep. These highly eutrophic
lakes have at least a foot of thick black mud on the bottom.
The clear water is filled with numerous small invertebrates,
but no fish. The concentration of invertebrates is such that
food does not appear to limit the populations of any of the
birds using the lakes. The bottom flora reaches, by the middle
of July, to within two feet of the surface and extends over
ninety percent of the bottom area. In addition dense mats of
floating algae cover one fourth of the surface by the end of 1
July. Mary Jaekson (pers. com.) identifies the bottom plants
as Scirpus, Ruppia, Ceratophyllum, Myrophillum, and Potamogeton.
and the floating forms as principally Cladophora and Aphanizo-
menon. Temperatures measured with thermocouples at noon on the
thirteenth of July, 1956, are shown in Table I. Table II gives
a list of the breeding waterbirds of the two lakes and the ap•
proximate numbers of breeding pairs per year. I am much endebted
to Mary Jackson for her estimates of breeding ducks.
Table I. Water Temperatures July 13, 1956
Depth in Peet Center of Lake 60 Peet from Shore
0 22.8°C. 24.5°C. 1 22.8 24.5 2 22.5 24.0 3 22.0 4 21.8 5 21.5 Total Depth 7 feet 4 feet
Westwick Lake has a colony of two hundred and ten pairs of grebes, and Sorensen a colony of forty pairs. During
the early part of the season observations were made from shore with a thirty power telescope and 7x35 binoculars. Later a
canoe in the reeds served as an adequate blind. During the nesting season the birds were too shy to be approached by canoe, and blinds on stilts were built in the water at the edges of
the nesting sites. Nests were checked by wading, and tagged with one inch square white cardboard tags numbered with India ink. 4
Table II
Estimated Breeding Birds on Westwick and Sorensen Lakes
The two lakes are combined and numbers are given as lower and upper extremes.
Species Latin Name Breeding Pairs /year Pintail Anas acuta 1-3 Green-winged Teal Anas crecca 5-10 Common Mallard Anas platyrhynchos 7-10 Baldpate Anas americana 10-15 Blue-winged Teal Anas discors 3-8 Cinnamon Teal Anas cyanoptera 0-1 Shoveller Anas clypeata 3-6 Canvasback Aythya valisneria 3-8 Redhead Aythya americana 3-8 Lesser Scaup Aythya affinis 15-25 Barrow's Goldeneye Bucephala islandica 15-30 Buffiehead Bucephala albeola 0-8 Ruddy Buck Oxyiira jamaioensis 15-30 Coot Pulica americana 75-100 Eared Grebe Podiceps caspicus 200-300 Pied-billed Grebe Podilymbus podiceps 0-2 Horned Grebe Colymbus auritus 0-1 Sora Rail Porzana Carolina 1-4
Organization
The purpose of this section is to orient the reader.
Before reproductive behaviour can be discussed at all, the general or non-reproductive behaviour must be understood.
The general behaviour includes flight, resting, swimming, preening, diving, and voice. These are basic patterns seen throughout the year. Then threat and escape behaviour must be dealt with. Often threat has developed into part of territory- holding or courtship. Its elements must be seen clearly sepa• rate from courtship. General behaviour and threat, then, form 5
the first two major sections of this paper.
The next two sections deal with courtship, first descriptively, and second from the point of view of evolution or origin. The courtship period on the breeding lakes is the first three weeks of May. During this time the birds pair and come into breeding condition. The five patterns peculiar to this period are: 1. a position or advertising posture in which a grebe actively searches for a mate, 2. a greatly ex• aggerated form of preening known as courtship preening, 3. a head turning ceremony between a pair called head shaking,
4. the strange penguin dance in which the pair stand upright with only feet and tails in the water, and 5. the cat attitude, a position with head on the shoulders and wings raised.
The courtship stops rather abruptly and nesting behaviour begins. Nesting covers the whole of the time be• tween the building of the first platforms for copulation to the hatching of the eggs. Its diversity prompts chronological treatment. The first pattern to be seen, and described, is soliciting for copulation on the water. Shortly after this active nest building and copulation on unfinished nests or platforms begins. Many eggs are laid and abandoned. Then there is often a break of several weeks where no reproductive behaviour is seen. Following this final nests are built, final clutches laid, and incubation begins.
Care and behaviour of the young are discussed in a separate section, followed by the ecological problems of mortality and clutch size. b
Figure 7 shows the time relationships of the repro• ductive behaviour patterns.
General Behaviour Patterns
The general behaviour patterns are those patterns which are seen throughout the year in the non-reproductive life of the birds. It is necessary to describe them for several reasons. First, they are commonly seen and must be distinguished from patterns having reproductive functions.
Secondly they or their derivatives form parts of the repro• ductive behaviour. Thirdly, they are of survival value and therefore of interest to students of evolution.
Flight
The Eared Grebe rarely flies except during migration, but its straight neck, trailing legs, and the fast beating of the short wings give it a singularly loon-like appearance in the air. From a distance it resembles a very black scaup.
The migration flights are at night with stop-overs on large lakes during the day. The first grebes follow directly behind the receding ice, and reach the breeding lakes the day after the ice goes out. Whether this involves reconnaissance flights over frozen lakes is not known, but grebes do not normally land on ice. Flocks continue to arrive for several weeks after the breeding lakes become open. 7
There is no evidence of movement of individual grebes from lake to lake after they reach their breeding lakes.
The road between my two lakes seemed ah impenetrable barrier to the birds; their numbers remained constant on each lake with the exception of arriving flocks.
On the breeding grounds the Eared Grebe flies only when badly frightened by man. From the escape attitude it suddenly throws its head forward, raises its wings, and flaps across the water a few feet before becoming airborne. It cir• cles low over the water and skids down on the water with a loud splash farther off. There are no intention movements of flight.
Swimming
The Eared Grebe swims with its body high on the water but showing none of the white underparts. The neck is held straight up but relaxed or not unduly extended, and the bill points forward parallel to the water. The feet give strong alternate strokes, but the movement as seen from above water is absolutely smooth. Ordinarily the body feathers are slightly fluffed and the crest relaxed. The various attitudes which modify this basic posture are described separately.
Resting
There is a single resting position in the Eared as well as in the Great Crested Grebe. Simmons (10) describes this position for the Great Crest, "with its head settled on 8
its back and bill tucked away at the side of the neck, a rest• ing grebe transforms itself into a compact bird quite unlike an active one with long, hosepipe neck." Lacking the very long neck, the Eared Grebe puts its head down on its breast and tucks its bill into its median ventral neck feathers. The elements of the pattern are body feathers slightly raised, neck and head down on back, and bill tucked away in the most conven• ient forward position. It is clear then that the attitudes of the two birds are only different where their different propor• tions make this necessary.
On cold mornings the grebes assume the resting at• titude but with the feathers of the crest and back raised and the wing tips lifted. They sit with tails oriented to the rising sun and sun-bathe. Wetmore (12) describes them as look• ing as large as mallards or gadwalls in this position.
Diving
Diving like swimming is a simple functional movement.
The head is drawn back and then arched forward into the water, the body following with a little spring. The legs do not move until the bird is under water. In the usual feeding dive the feathers of the back, neck, and crest are laid down as the head is brought back.
Lawrence (7) cites three variations on the diving pattern for the Western Grebe. These are the same as those in the Great Crested Grebe (10) and in the Eared Grebe. They all have the same pattern as above and depend for their differ- 3
ent appearance on various starting positions. In the "spring dive", used in rough water, the body is higher than usual on
the water and spring action of the movement is accentuated,
the bird landing with a splash a full body length from the start. In the "surface dive" the grebe starts from a half-
submerged position on the water and does not raise its body out of water. This gives a curiously porpoise-like dive. In
the "alarm dive" or "crash dive" the bird lifts its wings to gain impetus and kicks back with its feet. This may be done from a swimming position or after flapping across the water with
the aid of the wings. The wing flapping is part of the escape- over- the- water movements carried over into diving.
The intention movements of diving are the movement of
the head back and then forwards without the accompanying body spring of the real dive. In the lowest intensity form, feathers of neck, back, and crest are relaxed, but at higher intensities they are laid down in preparation for the dive. Eared Grebes are curious and tame in the early phases of their courtship, and often the whole colony approached within twenty feet of my wading figure, all nodding solemnly in this intention attitude.
'When I suddenly moved, the intention movement progressed into real diving and the whole group simply disappeared. Intention diving was seen only in a curiosity situation and probably does not serve as a social releaser.
Bathing and preening
Preening fills every break in grebe courtship activity, 10
is often seen during incubation, and is the most conspicuous activity during the winter. Individual feathers are preened by drawing them through the bill, and whole tracts are straight• ened by smoothing them over with the closed bill. The back is done with long strokes of the bill over the wings and under them. The neck and breast feathers are preened by bringing the bill down through the feathers and then forward and up in a circular movement. The underparts are cleaned with the same motion as the back except that the bird rolls over on its side to expose the feathers being worked on.
When a session of preening is finished the birds
"fluff up". That is, they draw the bill down and raise all the feathers of the back, crest, and neck, at the same time lunging slightly forward and up with a single leg stroke.
The raised feathers are snapped back into place as the bird settles on the water. In the Eared Grebe this is most often done from a stationary position in the water. The Great
Crested Grebe (10) stops swimming, raises its feathers, swims forward, and lunges. It may be that the larger bird needs more momentum to raise its body clear of the water.
Occasionally when swimming along, the grebe will stretch one wing and the foot on the same side, and then snap them back to position. The head and neck are in the resting position. Or sometimes a grebe may be seen in the resting attitude with one leg up under its wing, the foot dangling out behind.
Occasionally too, wing-flapping is seen. The head 11
is nearly in the resting position with crest raised. The wings are raised and flapped clear of the water. The head and neck are not moved, but the flapping raises the body nearly out of the water. Then the wings are closed and the bird settles tail first on the water.
Voice
The Eared Grebe has four calls, all of which have courtship functions. When feeding or resting and in all their non-reproductive behaviour, the birds are silent. There is no alarm call. There is a call for the advertising behaviour, a threat call, a penguin dance call, and a copulation call. It is probable that all grebe species are limited to this vocabu• lary, but comparisons will be made in the courtship sections of this paper.
Comparisons of the general behaviour
In comparing this brief account with the more elab• orate one of Simmons (10) for the Great Crested Grebe, it will be seen that the general behaviour patterns for the two species are identical except for a few details. These details are directly related to the difference in size and proportions of the two birds. Certainly the origins and functions of the patterns are the same for the two species. They are all simple functional patterns which must have been necessary from before the point of divergence of the two species. It is to be ex• pected that further studies v/ill show the same general patterns for the whole genus and probably for the whole family. Threat and Escape Patterns
Classical behaviour theory sees courtship displays as the balancing of threat and escape tendencies over a sub• strate of sexual motivation. Mutually antagonistic birds al• ternately threaten and flee until the sexual tendencies rise and the antagonism is submerged. The pair bond is then formed by mutual tolerance. Courtship patterns are portions of these threat and escape displays ritualized to serve the function of forming pairs without the damage which could result from whole• sale fighting. Following this general train of thought Simmons
(10) says, "Courtship in grebes is caused by a mixture of sexual, aggressive and fleeing impulses, and functions, not only to maintain and strengthen the pair-bond as Huxley sug• gests, but also to inhibit and absorb aggression. There is much evidence to support this theory, not least of all being the fact that the display movements themselves seem to be very highly ritualized, or symbolic fighting ones." (The italics are Simmons1.)
On the basis of my study of the Eared Grebe and com• parison with the accounts of the Great Crested Grebe, I submit a counter hypothesis. I believe that the courtship patterns of the grebes are motivated by sexual tendencies alone, and are derived by evolution from.nesting patterns and comfort movements. In the highly antagonistic and territorial species fighting behaviour is mixed with courtship and has been con• fused with it. Pighting in the Eared Grebe is seldom seen before courtship is finished, and the patterns can be studied 13
separately and compared. I will follow Simmons' (10) outline and describe first the threat and then the courtship patterns finishing with discussions.
Occurrence of threat and escape
Threat behaviour is connected only with the establish• ment of pairs and the choice and holding of a nest site. There is no territorial defence to confuse the issue. Low and middle intensity threat are seen when an unpaired bird in advertising posture approaches a pair. One or both of the pair may assume the threat attitude. Whether the male always threatens an advertising male, and the female the advertising female, I cannot tell because of the difficulty of telling the sex of the advertising bird. The threat pattern is seldom seen in the first weeks and then increases as the season goes on and pairs become definitive. High intensity threat is seen only when the colony is fighting over nest sites. When nests are established, the threat in low or medium intensity is limited to occasions when an intruding grebe is close to the nest.
The alarm posture is seen during courtship when the birds are disturbed by man. No other animal or natural phe• nomenon produces it. During nesting the alarm attitude is the usual one for birds on open water. They are shy and con• stantly on the alert during this time. When the eggs are hatched, the grebes return to the former behaviour.
Escape behaviour may follow the alarm attitude or it may follow directly from ordinary swimming if the bird is attacked. As a direct result of the increase in threat be• haviour during the period when nests are being chosen, its intensity is increased.
Threat attitudes
At low intensity the grebe assumes only the basic threat attitude. The neck is forty-five degrees forward of the vertical, the bill straight forward and open. The feathers of the neck are laid down and those of the crest raised. A bird in this attitude may swim toward the threatened grebe and brush against its shoulder. The threatened bird usually just swims away. In the middle intensity form the threat attitude is taken first. Then the wings are raised and bird flaps over the water toward the other bird. The threatened bird escapes by diving and is sometimes followed a short distance.
The high intensity form of threat is seen only in the establishment of nest sites. The elements are the same as those in the middle intensity form, but the rush is sudden and the wing flaps much more agitated. The opponent is here always struck or bitten before it has a chance to get away by diving. The biting motion involves drawing the head back and then thrusting it forward. The movement is identical to that performed in diving, and the dive is sometimes actually com• pleted at the end of the rush. The threatened bird always turns away as the rush begins and takes the blow, or most fre• quently the bite, on the side or wing. Threat is always successful in driving the intruder off, though it may come 15
back again and frequently does. This gives moving attack- escape sequences instead of the stationary fighting and up• right threat of the Great Crested Grebe (10).
The threat call is always associated with the middle and high intensity threat, rarely with the low. The call is a loud chitter, the two notes a musical fifth apart and alter• nated rapidly. In spite of the speed, each syllable is indi• vidually accented.
The forward threat and attack of the Eared Grebe are the same as those used by the Great Crested Grebe (10), and the Great Crest bark call is certainly homologous with the chitter of the Eared. It is interesting that the Great
Crested Grebe has so many variations to its threat call.
This may be in connection with its greater use of threat be• haviour. There is no upright threat posture in the Eared
Grebe, because there are never two Eared Grebes evenly matched.
When two Great Crested Grebes are evenly matched at a terri• tory boundary, they both threaten. They raise their heads as they come closer until they sit face to face with necks straight up. The Eared Grebe has no territory boundary since the only territory defended is the nest itself. There is, therefore, no upright threat in this species.
In the fighting of the Great Crested Grebe, Simmons says, "both grebes leap at each other and clash vertically breast to breast." It is not clear whether this is the up• right swimming posture of the upright threat or the standing posture of the penguin dance. This standing position, if it xo
is such, was never seen in the Bared Grebe, and I find no other reference to it in the literature. Dubois (2) describes the Horned Grebe, Podiceps auritus, as having the same lunging attack I saw in the Eared Grebe. Wetmore (12) speaks of a female Pied-billed Grebe, Podilymbus podiceps, with young as
"rising threateningly on the water, made a great boiling noise by treading rapidly with her feet." As the presence or absence of the standing fighting or threat behaviour has great import• ance in tracing the evolution of some of the courtship patterns, it should be more thoroughly investigated.
Alarm
The alarm attitude of the Eared Grebe is called the
"pencil-necked posture" by Munro (8). The feathers of the body, neck, and crest are all laid down. The neck is extended vertically, and the bill is straight forward. The wings are tightly closed. The grebe is ready to escape by diving, sub• marine diving, or flapping across the water. The alarm posture is the same as that of the Great Crested Grebe (10), but is not described for any other species.
There is no call involved in alarm or escape behav• iour. Each bird must be alert to its own safety, and makes its escape absolutely silently. This is particularly notice• able in the nesting colony, where even with the most silent approach one seldom sees and never hears a grebe; only the empty nests are found.
Simmons (10) does not mention whether or not the 17
Great Crested Grebe has an alarm call, but it probably does not. Gross (3) describes the Antillean Grebe, Poliocephalus dominicus, male as having an excited call, yeep-yeep-yeep, followed by a rattled ye-ye-ye-ye-e-e-e-e-e-e. The female and young all dove "in a wild splash at this signal". Deusing
(1) says that the Pied-billed Grebe has a hu.hu,hu,hu call con• tinued for several seconds which he says may be an alarm call.
Suffern (11) writes:
"The Handbook (of British Birds) says the Little Grebe's (Poliocephalus ruficollis) alarm call is a soft, somewhat whistling whit, whit, which exactly describes what I heard; but I have never heard an alarmed Little Grebe give such a note. In my experience, when a Little Grebe is surprised it flies along the top of the water with much splashing and then dives without uttering a cry. The alarm note referred to in the Handbook is not used by birds which are merely startled, but by agitated breeding birds. But it, or a note closely similar, also seems to be used as a call and perhaps the term alarm note would be better not used or at least needs qualifying."
The call he refers to is probably the advertising call to be discussed later. These are all isolated observations and certainly need checking.
Escape
When a person or boat or other strange object ap• pears close to a grebe, it first assumes the alarm posture, and then escapes. "When another grebe attacks it, however, it escapes without showing alarm. Escape may be made by pattering over the surface, by diving, or by the submarine dive. The first two have been discussed in the general habits section since they are so frequently seen. The crash dive is only seen 18
in extreme circumstances, and is a combination of diving and pattering over the water. The wings are raised as if for flight, but the bird dives and uses its wings a short distance under water. The submarine dive is almost as common as the ordinary escape dive and pattering. Unlike these, it is never seen except in an alarm situation. In this pattern the grebe sinks out of sight from the alarm attitude, body first, then neck, then head. The resemblance to a small submarine is un• mistakable. The mechanism of this dive must be loss of buoy• ancy through control of air sacs or feathers, but it has never been studied in the grebes.
Courtship Displays
The courtship displays are all done on open water in the centre of the breeding lakes. There is no territory of any kind involved, as each bird can and does make use of the whole lake area. The Great Crested Grebe (10) courts on territory and uses the defense of territory boundaries as part of its courtship.
Courtship display in Eared Grebe is seen as soon as the birds arrive at the breeding lakes and may even be ob• served on migration stops. Display tapers off abruptly as nesting behaviour begins. Overnight all the birds seem tol ibe paired and the whole character of the lake changes. The colony, stimulated by courtship, acts as a unit in this total change in behaviour. Simmons (10) describes a similar abrupt 19
change of behaviour in the territorial Great Crested Grebe, but each pair has its own change-over time. The increased use of advertising behaviour serves to synchronize courtship in the Eared Grebe and creates a true colonial situation.
Lawrence (7) says that the running penguin dance of the West•
ern Grebe is seen in reduced intensity all summer. Nothing is known of timing in other species.
The courtship is mutual; all patterns may be per• formed by either sex. Each pattern will be described, its frequency indicated, and comparative data given. In the next
section the possible origins of the patterns will be discussed.
Advertising behaviour
Advertising behaviour is seen where an unpaired bird is searching for a mate, or where a paired bird has temporarily
lost touch with its mate. The advertising bird swims up and
down approaching other grebes with a characteristic attitude
and call. The feathers of the whole body are fluffed out making the posturing bird seem larger than its fellows. Both
crest and neck feathers are raised. The neck is erect, and
the bill directed straight forward and closed during the call.
The call is a three noted poo-eee-chk, but the ending
chk can only be heard at very close range. It may be, as
Simmons (10) suggests, produced by the effort of the production
of the first two notes. The first two notes are plaintive and
slurred, a musical fifth apart. They sound much as if they were played on a flute. The third note is the same as the first 20
but much quieter and very guttural.
Eared Grebes give the appearance of being paired on arrival in the spring. More than half were swimming in pairs on the day of their arrival at Westwick Lake. When a single bird approached a pair, situations such as the following occurred
"One of the pair swam toward the advertising intruder, and it turned away and left them." "Intruder appeared and swam between the paired birds and around them. The male of the pair ignored the intruder for a while and then rushed at him from a distance of four feet in a threat attitude. Then the pair moved off while the intruder moved into the reeds." Such incidents as these led Munro (8) to say that the birds were paired on their arrival, and my judgment originally followed his. Closer study of individual pairs, however, showed that none of the Eared
Grebes are paired definitively at this stage. Partnerships last from three hours to five minutes. An advertising single bird of either sex swims up to the pair and takes one of the pair away. The divorced bird then goes off and advertises.
The number of changes of partners seen drops off rapidly until pairs are definitive a few days before nesting behaviour is first seen.
It is not known whether the pair bond lasts for more than one year or not. It may be that the courtship serves to re-establish old bonds. At any rate birds pairing for the first time cannot be distinguished from older grebes. The process of re-forming a pair, if there is such a thing, seems to be the same as that involved in forming a pair. Simmons (10) ZJL
says that due to lack of a marked population, he has not been able to observe the formation of a new pair. He describes no aberrent courtships which might be young birds, so I think his birds present the same problem as mine. When we get marked populations, the formation of new pairs should be easily followed.
The advertising position of the Great Crested Grebe is the same as that of the Eared except that the crest of The
Great Crest is depressed. Simmons (10) describes the call as
"a loud, drawn-out 1grr-owp' with a metallic tone, somewhat of a cross between the cawing of a Carrion Crow and the crowing of a farmyard rooster. This is followed by a secondary note, a moaning purr, 'row-ah', not audible for more than a few yards and probably a by-product of the effort of producing the main note." These elements are obviously the same as the elements of the Eared Grebe call. The Great Crested Grebe remains stationary while advertising. This is related to the territor• ial nature of its courtship. The Eared Grebe on the other hand may be either stationary or swimming. It may even swim up to other-grebes, but it is never aggressive.
Courtship preening
Grebes, as I have said, preen a very great deal.
The amount of preening must be even greater during courtship as the grebes are more active then. Courting grebes float along in pairs and preen vigorously. This cannot be called courtship preening as it does not differ from ordinary preening and cannot be proved to be a courtship activity. However, between these bouts of comfort preening, a different kind of preening may be seen. The two birds drift within a foot of
each other and preen together. Either,bird may lead the ac•
tivity, but the other repeats the movements exactly. The
timing is so close that it is often difficult to tell which bird is leading. There is no definite sequence of movements,
but both birds preen the same feathers at the same time. The
movements are a little quicker and not quite so thorough as
those of comfort preening. The participants are most frequent•
ly side by side but may end up facing each other. This pattern
is not seen except during courtship, and must have a courtship
function.
Preening is by far the most common courtship pattern
in the Eared Grebe. It also occurs in the Great Crested Grebe, but in that species head shaking is the more common courtship movement.
Head Shaking
Head shaking is seen both as a separate ceremony and
as part of the penguin dance ceremony. As a separate ceremony
the two birds swim along rapidly one in front and slightly to
the side of the other. , Their body feathers are raised, and
their position is very high on the water. Their necks are
stretched up higher than in the advertising behaviour with
crest raised and neck feathers depressed. The general effect
is taller, higher on the water, and thinner than in the adver- 2t5'
tising attitude. The head is turned smartly left and right, all in the horizontal plane. The tempo is military, precise but not hurried. Six to twelve turns constitute a ceremony;
then both birds usually dive. The timing of both the turns and dive is generally quite precise as the following bird apparent• ly copies the leader. Either sex may lead. Once the head shaking was seen done by an unpaired bird. After advertising to a pair, the odd bird did several head shakes, and the female swam off with him.
The head shaking ceremony is as. common as the cat attitude but not as common as the penguin dance. About half the time it is directed toward other grebes, paired or single birds, as a threat. Once the shaking pair were riding so high on the water that white breast feathers were visible. The effect was reminiscent of the Western Grebe courtship dance.
For the Great Crested Grebe (10) the head shaking is the most common courtship pattern. It almost replaces the penguin dance and courtship preening. In addition to courtship, it is used to cement the pair bond during and after territory boundary disputes. The use of shaking as threat toward an in• truder is the only Eared Grebe counterpart of this. The atti• tude of the Great Crested Grebe is the same as that of the
Eared Grebe, but the birds usually face each other and are always stationary. Often, too, their bills are below horizontal.
Easter turns are alternated irregularly with slower ones in the
Great Crest, while the tempo is even in the Eared Grebe.
Huxley (5) describes the Great Crested Grebe as head shaking while hack to back, but Simmons (10) has never seen this variation. Wetmore (12) cites the same for the Eared
Grebe, but I have never seen it. It must certainly be quite rare.
Kilham (6) describes one of a pair of Pied-billed
Grebes as head shaking while they swam near each other, but nothing else of comparative value is known of head shaking.
Penguin dance
The penguin dance is more common than the cat atti• tude or head shaking, but less common than advertising behav• iour and courtship preening. Both birds tread water raising their silvery bellies out of the water and seemingly stand on their tails. Their bellies are only a few inches apart, and they paddle their feet alternately and so rapidly that they sometimes splash. As they rise they utter a strange shrill chattering and head shake violently. There are from six to twenty head turns, but the pace is so fast that the birds are usually out of phase. Dropping to the water, they slow head movements to the pace of the head shaking ceremony, and eight to twenty turns are made from the swimming position facing each other. Only once was the Eared Grebe seen shaking face to face without first going through the dance. Sometimes the birds turn during this last head shaking and continue as in a normal shaking ceremony swimming together down the lake. Some• times they simply stop head shaking, drift apart, and comfort preen. Courtship preening is a part of the post dance ceremony about a third of the time. The beak is dipped eight to six• teen times to the breast feathers or primaries either before or during the head turns. The preening movements are here even more ritualized than is seen in the usual head shaking.
The feathers are touched but not stroked, and the rhythm of the ceremony is not broken.
Very occasionally the birds turn i^hile still stand• ing and run a short distance on the water before settling down to head shake. Usually the birds stand up facing the same direction as in the Western Grebe, but turn while standing to face each other. This turning is the same as that seen in the
Great Crested Grebe discovery ceremony. Here one bird ap• proaches by diving and rises in the ghostly penguin attitude.
This attitude is described by Huxley (5), "He seemed to grow out of the water. First his head, the ruff nearly circular, the bill pointing down along the neck in a stiff and peculiar manner; then the neck, quite straight and vertical; then the body, straight and vertical too; until finally the whole bird, save for a few inches, was standing erect in the water, and reminding me of nothing so much as the hypnotized phantom of a rather slender penguin." In diving it most often overshoots its mate, comes up facing away from it, and turns in this ghostly manner to face it. The mate is in a cat attitude, and the ghostly bird drops to the water and head shakes with it.
A penguin dance does not follow, but in the Eared Grebe whether the mate is in a cat attitude or not, a penguin dance always follows. The Eared Grebe dive is not very ghostly and the much ZD"
smaller bird comes up like a cork bobbing to the surface.
Also it makes no dramatic pause, but immediately begins the head shaking of the dance. I have not considered the ghostly penguin under a separate heading, because I think that it is an incomplete form of the penguin dance. The incomplete form, without the dance itself, is of course the common form for the
Great Crested Grebe, and the complete form is rarely seen.
When it does occur, the Great Crest pair dives and each bird retrieves a bit of pond weed, swims toward its mate and rises in the penguin dance. The Eared Grebe may begin its dance from a dive or from the surface; but the use of weed has never been
seen.
The Great Crested Grebes lean against each other
during the penguin dance. The Eared Grebes, probably because of their smaller size and shorter necks, stand upright without leaning. The awkwardness of the larger bird might be considered
the reason for its less frequent use of the penguin dance. If
this is true, it is strange that the even larger and more awk• ward Western Grebe uses the dance a great deal. Perhaps the forward motion supports it.
The chittering call of the Eared Grebe penguin dance
consists of the two notes of the melancholy poo-eee-chk uttered more quickly and in wild excitement. In construction it is the
same as the threat call, but the quality is more musical.
Simmons does not describe any call connected with the penguin
dance in the Great Crested Grebe. I assume that since head shaking is so closely related to the dance that the head Z'(
shaking call is used. There is no head shaking call in the
Eared Grebe.
Occasionally in the Eared Grebe an unraated bird will flap in threat at a pair doing the penguin dance. It may come from quite a distance at the sound of the dance. This is more likely to occur later in the season when there is less switch• ing partners, and the odd bird has been advertising vainly for a long time.
Cat attitude
In the cat attitude the head is drawn down on the breast with the crest and neck feathers raised. The raised body feathers make the bird look much larger than usual, and the wings are away from the body but bent. The carpal joint is tipped forward and down to almost touch the water. The effect is similar to the bluffing attitude of the frightened young owl. The posturing bird exactly faces the bird elicit• ing the display as if making the most of the orange circle of plumes in the center of the larger black circle of the wings and body. The attitude of the Great Crested Grebe is the same.
The circular cinnamon ruff is displayed in the same way, and in addition the white leading edge of the wing is shown.
The attitude is quite rare compared to the other courtship patterns, although the Great Crested Grebe uses it more than the Eared Grebe. Both grebes use the attitude when an unmated bird has been advertising. Another grebe swims to• ward the advertising bird and dives. The advertising grebe 28
goes into a cat attitude, and the approaching bird rises in
the ghostly penguin attitude. Paired Eared Grebes use the
same ceremony without previous advertising. Mated Great
Crested Grebes use it as a discovery ceremony when one bird
has been away. In the Eared Grebe the penguin dance always
follows; in the Great Crested Grebe head shaking follows.
Once a female Eared Grebe did a cat attitude to an advertising male as he passed. Her mate immediately drove the odd bird
away. Rarely in either species one bird may flap over the
water a little way and go into a cat attitude turning to face
its mate. Or it may just patter away and swim back without
using the cat attitude. The penguin dance or head shaking
follows respectively as seen before. In addition the Great
Crested Grebe unmated female may use the attitude when threat•
ened by a mated male. I have included this section under
courtship and made no mention of it under threat as Simmons (10)
does. This is because I believe that it is not related to
threat behaviour at all. I will discuss this further in the
next section.
Origins of the Courtship Patterns
In order to understand the courtship>of birds more
fully, we attempt to determine how the courtship attitudes
developed and how they are related in different species. The
method used is comparison of the elements of courtship patterns
to those of non-reproductive behaviour, comparison between 29
species and the observation of development of the patterns in
the young. The first two form parts of this discussion. Great
care must be taken, since a simple pattern may not be recently
evolved, and two similar patterns may parallel and not be re•
lated. Table III shows a comparison between the courtship patterns of the Eared Grebe and those of the Great Crested
Grebe.
Advertising behaviour
Simmons (10) suggests that the advertising posture is not ritualized, but simply the posture necessary for the- production of the advertising call. I agree with this since
the feathers are only slightly fluffed out and the head slight•
ly raised from the normal swimming position. Why the Great
Crested Grebe lowers its crest and the Eared Grebe raises his is a problem. Both birds raise their crests in all the other
courtship patterns, in threat, and in incipient diving. They
lower them in nesting activities and in alarm.
Courtship preening and head shaking
Courtship preening is certainly ritualized comfort
preening as both contain all the same elements and are con•
nected by an unbroken chain of intermediates. The head shaking
of the Eared Grebe and the slow phase of the Great Crested
Grebe shaking are also connected to preening by a series of
intermediate forms. In this derivation I again agree with
Simmons (10). The fast shaking, lacking in the Eared Grebe, 30:.
Table; in. Courtship Patterns Compared
Advertising Behaviour::; Eared Grebe Great Crested Grebe done by lone bird mated or not X X body feathers fluffed x X body high on water X X neck erect, bill forward X 3iX crest and neck feathers raised X crest down X call: poc—eee-chk X grr-owp row-ah X done while: moving X stationary X Courtship: Preening:: precise, stereotyped movements X X primaries and breast used mostly X X preens whole body X ? seen alone or with head shaking X X Head Shaking:;: neck erect, medium stretched' X X crest upv X X neck feathers down, body feathers medium X X bill: forward X slightly down X facing:,, forward X each other X done; swimming X done: stationary X done between battles X used by pair as threat or courtship- X X Penguin Dance: standing position X X crest up, bill forward X X body and neck fluffed X X neck medium stretched X X head turning X X leaning X not leaning X chitter call X ? call X elicited: by diving for weed X by ghostly penguin X from surface X by cat attitude X turn forward while shaking X stay facing each other X followed by head shaking X X Cat Attitude: head withdrawn, crest raised X X wing out, tipped forward X X body feathers raised X X elicited by ghostly penguin X X faces partner X X followed by: penguin dance X head shaking X elicited by mate approaching on surface X female may do it when threatened X Simmons (10) derives from the movement used to shake water and feathers from the bill. This last must have developed after the point of separation of the two species; the rest probably- developed before.
Simmons (10) suggests that the upright position used in head shaking is related to the upright threat posture, but grebes in their normal swimming, alarm, and advertising use an upright attitude. There is no reason for postulating a special connection here.
Penguin dance
The three major elements to be explained in the pen• guin dance are the head turning, the upright posture, and the call. The head turning is similar only to the head shaking display. The fast phase of the head shaking is identical to that in the dance and continues as a regular head shaking cere• mony after the dance.
The standing posture has been connected by Simmons
(10) to the upright threat posture. He sees the coming to• gether in the penguin dance as a modified fighting situation.
In the upright threat, however, the birds are in swimming position with necks up; here they are actually standing with only their feet and tails in the water. I have discussed before the importance of knowing whether the Great Crested
Grebe in fighting uses a standing posture. If it does, then there is a possible connection. If not (and the Eared Grebe does not), there is no reason for supposing a relationship. The one place where the grebe does stand is on the nest, in copulation and nesting. It seems clearer, then, to postulate a relationship in this direction. The same is true for the incomplete case where the Great Crested Grebe rises in a ghost• ly penguin attitude near its mate. Why this should be a "sym• bolic attack" (10), I don't know. A grebe approaches its nest by diving, and may come up from a feeding dive near its mate in a situation where no threat is intended.
The use of weed by the Great Crested Grebe is taken directly from nest building. Weed is not eaten and is not used in any other connection.
The dance call of the Eared Grebe is similar to the threat call. Whether this means anything or not is not certain.
The call of the Great Crested Grebe when doing the penguin dance has not to my knowledge been described.
Cat, attitude
In the cat attitude we must explain the raised wings, raised crest and body feathers, the withdrawn head, and the orientation. Simmons (10) describes the cat attitude as an - escape reaction usually following the ghostly penguin ritual• ized attack. The head is withdrawn, he says, in the opposite of the stretched neck of threat. This sounds reasonable, but a grebe showing a tendency to escape never withdraws its head.
In fact it assumes the alarm attitude with head up and crest and body feathers down. The only time a grebe ever does with• draw its head is in resting and sun-bathing. When a grebe raises its wings in threat or escape, it holds the first primary feather parallel to the water. In the cat attitude the carpal joint is down with the primaries raised to form a nearly forty-five degree angle with the water.
This angle is seen only in sun-bathing.
The orientation of the cat attitude is toward the bird eliciting it. This is also true of threat, but not of alarm where the bird does not orient at all, or of escape where it orients away. In sun-bathing the orientation is away from the sun.
It seems clear from the previous arguments that the most reasonable theory of the origin of this posture is origin from sun-bathing, but the matter requires further detailed ob• servation and comparison with other species.
All of Simmons' examples of the cat attitude could be explained by sexual motivation rather than escape. All of my observations of the cat attitude were followed directly by the penguin dance. This still does not explain why on some occasions a bird flies away from its mate or another bird and goes into the cat attitude. Perhaps the cat attitude is an invitation to court.
Summary
Two of the five courtship patterns discussed are definitely derived from comfort movements. The other three may tentatively be related to nesting and comfort patterns.
Threat and escape behaviour probably play no part in their derivation. Much more research on the grebes as a group is necessary to work out all the details of courtship derivation.
But still, there is an indication that in some groups of birds at least, threat and escape behaviour may have become necessary in the courtship period only after courtship patterns were al• ready established. The development of territory and fighting over territory boundaries would then cause the intermingling and confusion of courtship and threat.
Nesting Behaviour
During the course of one week, the behaviour of the grebes changes completely. Before the sixteenth of May 1956 none of the elements of the nesting behaviour was seen; after
May twentieth none of those of the courtship period was seen.
The situation was the same and the dates almost identical in
1955. The change is evident even to the casual observer. In the first period feeding and resting are closely intermingled with courtship; there is almost always a pair courting some• where in sight. The birds are scattered all over the open water and they are moving randomly. The number of advertising calls makes the whole lake seem noisy all day, and the birds are very tame when approached.
In contrast, nesting behaviour occurs each day for short periods only. The short periods are only about half an hour, and the whole colony is usually involved. The mating parties are always in the reeds or floating vegetation, and 35
are marked by threat and copulation calls. But between parties the birds spend most of their time resting in large groups near the reed area where they last courted. They scatter to feed periodically but always return to the group.
While on open water they are absolutely silent and increasing• ly shy.
Late migrants arriving during the platform behaviour period remained separate from the original birds and performed courtship instead of nesting behaviour. They stayed separate until their courtship was finished, when nesting behaviour commenced. They were then absorbed into the first group.
Nesting behaviour is that portion of grebe repro• ductive behaviour which is associated with nests or platforms, and may be divided into nest establishment, egg laying, and incubation. Incubation and final egg laying last for a fixed number of days, but the nest establishment varies in length.
In 1955 and 1956 clutches were not begun until June twenty- seventh and June twenty-second respectively, but Munro (8) reports newly hatched young June 14, 1941 on Westwiek Lake and August 10, 1940 at Cummings Lake only ten miles away.
These represent clutches started May twentieth and July six• teenth.
These changes in behaviour, from courtship to nesting and from nest establishment to the laying of final clutches, are the result of both behavioural and climatic factors. The first change from courtship to nesting depends on the date the ice goes out and the temperature in the early part of May. 36
The higher the temperature in May (it never reached excessive heights) the greater the degree of courting activity. Pre• sumably a certain amount of courtship must be done before the birds begin to come into breeding condition. This is shown, by the fact that late flocks continue to court after the early flocks,have already started nesting activities.
The change from nest establishment to final clutches is influenced by climate and behaviour but in a, different way.
The tempo of the nest establishment activities rises gradually to a culmination in the laying of final clutches. Again a certain amount of the first activity seems to be necessary be• fore the next can be begun. The birds are in breeding con• dition and laying eggs, but the incubation pattern is not yet mature and the eggs are abandoned. Also as before, the amount of behaviour is governed by climate. If June is warm, the grebes incubate the beginning of June, or even the end of May in exceptionally good years, and bring off young the end of
June. But usually June is cold and rainy in the Cariboo, and the birds stop all reproductive activity and rest and feed until good weather in July. When good weather returns, a burst of nest-establishing activity leads to incubation. Here again, the activity of the colony itself is important as there may be a difference of as much as a week in the resumption of nest establishment between Westwick and Sorensen lakes.
All the birds in one colony react to behaviour changes together. It will be interesting to see how the changes are made in territorial species and in species such 37
as the Horned Grebe which may be either extremely territorial or very colonial.
This is the picture after two summers of observation• al work. It is not known which aspects of the climate are act• ing as releasers, or whether they act as specific threshold values or as cumulative factors. I suspect that the climatic factor is temperature slightly above the water surface, and that the mechanism is a cumulative one. Also unknown is the physiology of the behaviour changes, both as to maturing of the gonads and as to hormonal changes. I would postulate, though, that first change is associated with the increase in ... size and condition of the gonads, and the second with increase in the prolactin concentration in the blood.
Soliciting
Soliciting is the first pattern of nest establishment to be seen in the field. The body is low in the water, neck forward, and head and bill laid straight out on the surface of the water. The neck is not unduly extended, and the crest is down. This is the same as the position of the female in copulation except that in copulation her bill is usually point• ed slightly down. This would of course be impossible on the water. There is a single call associated with both patterns.
It is an eerie note somewhat similar in quality to the scream of a Red-shouldered Hawk, Buteo lineatus. It does not however change in pitch. The soliciting bird may be answered by its mate with the same call. Both birds dive and disappear into 38
the reeds. Later in the season both birds solicit copulation on the nest platform.
The Great Crested Grebe (10) rarely solicits on the water, and does not start soliciting until the platform is at least partly built. The position taken is the same as that of the Eared Grebe except that the neck is somewhat stretched.
Simmons (10) does not describe any call for this pattern such as that used by the Eared Grebe. The Great Crested Grebe has another soliciting display where it stands on the nest and arches its head down to peer at its feet. Simmons refers to this as rearing. Hdsking (4) describes a similar pattern in the Horned Grebe. I have not yet seen this in the Eared Grebe, but its presence should not be excluded.
Development of platform building
Platforms are simply unfinished nests on which cop• ulation first takes place. Building begins at a very low in• tensity a few days after the first soliciting is seen. The group begins to drift all together into the reeds. Almost any place may be chosen, but some floating algae or reeds must be present. Isolated tufts and sparse floating pond- weeds totally unsuitable for finished nests may be used for these early platforms. As the group swims slowly around in the chosen area, a single female may pull a broken reed over and lay it across another reed caught between two upright stalks. Then the whole colony swims away and resumes feeding.
With each occasion of group activity the intensity 39
of the behaviour increases and more pairs become involved.
On Sorensen Lake the evening of June 9, 1955, this sequence was seen: At sunset the birds were all feeding on open water.
A pair drifted into the entrance to the marsh, and the birds swam to opposite sides of a reed tuft in the middle of the channel. This tuft was one of a group of three each contain• ing less than a dozen reed stalks. The female pulled at a long floating reed and added some floating algae to that al• ready caught in the reed tuft. Several pairs swam by and were driven off with much chittering, the area of the territory be• ing only a foot or two around the platform. This pair held the tuft until dark at eleven P.M. and copulated on it about a dozen times. Several other pairs could be heard similarly occupied in the marsh, but the majority of the birds had not yet chosen tufts and trouped up and down in a body, heads down in resting posture. These early platforms are shown in figures 8 and 9, and a later one in figure. 10.
As more and more pairs become involved in the party the noise becomes continuous and the action so fast as to be' completely confusing. On May 18, 1955 an active party was seen on Westwick Lake. I was less than fifty feet from the previous year's nesting area, and the grebes completely ignored me. The day was warm and sunny, and the grebes were feeding all over the lake. At four P.M. a bunch of cattle, driven by on the far shore forced the grebes to drift toward me. In half an hour they were all in the reed fringe in front of the canoe.
An eerie trilling began and increased as pair after pair joined 40
in. Six pairs1, solicited and copulated on the old grebe plat• forms. Pairs rushed at each other, wings raised and beaks open in threat. A dozen small battles raged at once and the threat calls of fifty individuals were so continuous that the birds which were calling could not be distinguished. Six single birds tried to attract mates but were dealt with violent• ly by both members of each pair they approached. Coots and ruddy ducks swam unconcernedly through this scene of chaos and were not bothered by the grebes.
The parties start spontaneously and may take place in a different area each time. There is a tendency to use one reed area for several days and then to move to another. The moves took place both on the lake that I was watching from the canoe and on that which I watched only from the distant hill• side. I could not, then, have been disturbing the birds and causing the moves. The moves may have been connected with the increasing maturity of the nesting patterns. The movements are indicated by numbers on the charts of the lakes, figures
1 and 2.
In each of the two summers I watched them the grebes at one time chose an area of reeds far too sparse to support nests. June 21, 1955 the grebes were in an area of new reeds in the centre of Westwick Lake. The tufts each contained less
than a dozen stalks and were several feet apart. I could see in this open area the beginnings of nearly fifty nests. The grebes were alternately soliciting on the new platforms and adding new material to them. Five nests had one egg; one nest 41
had two. A thundershower and strong winds the next night destroyed the nests completely, and activity switched to a new area.
Nest building patterns
The patterns are the same regardless of the intensity of the nest building activity. All the building is done by the female. This is the only pattern besides copulation which is not reciprocal. She dives near the nest and comes up about two feet from it with a large beakful of algae, lays it on the edge of the heap with a quick reaching movement and dives again.
The nest is set on a foundation of several reeds, broken down to form a geometric outline. It may be in the center of a small clump of reeds or at the side of a large one. The heap of debris finally makes a more or less pyramid-shaped pile reaching from a foot or two from the bottom of the lake to slightly above the surface. Three hours is sufficient for the construction of a nest. Additional material may be added to the sides of the nest during the egg laying period. On several occasions I tagged reeds near nests which I thought finished, and found the tagged reeds built in when I returned.
The center of the nest is cupped, but the rim is never more than an inch above the center. The whole nest is soggy, and the center may be slightly below water level even when the adult is not on it. The lower half of the eggs is always wet.
The color of the nest depends on the bottom in the 42
immediate vicinity of the nest, because the female always
takes what is handiest. Nests built in a new reed bed were
bright green, being completely constructed of living algae;
those in old reed beds were nearly black with rotting reeds
and algae. All the nests had the appearance of being modeled
in wet clay, smooth and soft. There were few rough edges of reeds to be seen.
The colonies when complete were compact units in
the densest part of the reed bed and visible neither from the
shore nor the open water. The final nests of other lakes, however, may be built much more in the open. The nests are from one foot to three feet apart and are in water one to three and a half feet deep. Places where there are surface algae are avoided. The grebes do not ordinarily swim through algae;
perhaps it gets their feathers dirty. The first nests in the
final colony are in a group and form a nucleus for the develop• ment of the colony. The colony is usually fairly compact, but
twice it was formed in a circle around a small lagoon in the
reeds.
When nest building, the females assume the shy atti•
tude characteristic of the egg laying and incubation periods.
Their crests and neck feathers are sleeked down making them
look snake-like and furtive. This is a low intensity alarm
attitude used habitually. They are no longer tame and curious
little creatures, and have become shy and restless. The males
cruise around after the females as if supervising them, but
they were never seen to handle nest material. They do not 43
adopt the pencil-necked alarm posture until the middle of egg laying.
Nest material was never seen used ritually in court• ship or as an invitation to nest. This may be related to the fact that.the male does not build. In the Great Crested Grebe where the male as well as the female builds, weed is used in the penguin dance.
Copulation
Copulation occurs frequently from the time the first platforms are made until the end of egg laying. The female hops up on the platform with a thrust of her legs, wings closed.
She stands on it with her legs out like props for a few seconds and then drops down on her belly. She stretches her neck and points her bill down forty-five degrees. Head and neck feathers are flattened. The male swims around behind and hops up on her back without opening his wings. He stands at an angle slightly forward of vertical with crest and neck feathers down. His neck is curved and bill pointed down. His posture is very similar to escape bathing. Both birds are motionless. The male gives the copulation call. The female does not join in the call and may even threaten passing birds. After about half a minute the male slides into the water over the shoulder of the female, turns to the nest, and bows. The bow, sometimes omitted, is done by dipping the bill to the breast feathers and at the same time raising the flattened crest. It is a single courtship preening movement. The female stays on the nest a minute and then slides off and preens. She usually invites copulation by hopping-onto the platform, although the male sometimes solicits on the empty platform and then slides off, turns, and swims to the female..
The Great Crested Grebe (10) female holds her head and beak straight out, probably because of her long neck. The male takes the same position as seen in the male Eared Grebe, but the closed wings are quivered as in the rearing soliciting attitude. The call is a harsh rattle "sounding like a creaking
ee eee" (Simmons' 10 quote of Hanzak). As he slides off the nest, he takes the escape bathing position except with the
crest and head up. He settles and begins head shaking with his back to the female. The female raises her head to head shake with him.
Escape bathing Simmons named because he thought it was a real escape pattern. It is a particularly violent version of the bathing behaviour which ends with a peculiar attitude.
The body is held upright but half submerged, that is half the white breast feathers are showing. The neck is arched and the crest down; the head is on the breast. I have seen it only
three times, and Simmons (10) saw it only five times. In all of my records the pattern followed defeat in a nest establishing encounter. Three of Simmons' records followed aggressive en• counters with other grebes; his other two records followed crash dives. The Great Crested Grebes put their heads under water and looked around as if looking for something; the Eared
Grebes did not do this. Simmons felt that the grebes were 45
looking for fish predators and that the whole pattern was one of escape. He says that the assumption of this attitude after copulation is a return to escape behaviour after this close contact with another bird. I do not doubt that the post- copulatory attitude and the escape bathing posture are related.
But the upright posture, the arched neck, and the lowered bill are never seen, even separately, in escape behaviour. I sus• pect that both patterns are the assumption in the water of the male copulation posture, as a result of an excess or carry-over of sexual motivation. This, I think, explains the use of escape bathing as an aftermath to the loss of a battle for a platform.
Certainly the theory needs careful testing by comparison with other species.
Abandoned eggs
After high intensity nest establishment has begun, eggs begin to appear. At first they are not brooded but simply abandoned on flimsy platforms. These are yard eggs in the same sense as those of galliforms and represent readiness of the gonads before maturation of the incubation behaviour. The first few days these eggs are not pecked, but then all eggs found have a hole peeked in the top. Presumably the grebe pair gaining a platform from a pair with an egg peck the egg and push it off the nest. Just before final clutches were started in
Westwick Lake in 1956 there was a broken egg under nearly every platform and two under some. The grebes were never seen to peck an egg, but there were no other animals around at the times that eggs were being pecked. The birds become more and more vicious in guarding their platforms, and they begin to incubate the eggs as they laid them. These incubated eggs are the eggs of the final clutch.
Two clutches of pecked eggs were found in the last week of incubation. One bird of each pair may have died or deserted the nest. At any rate, it seems necessary to protect the clutch from other grebes within the colony. Figures 11 and 12 show abandoned eggs; the egg in figure 12 has been pecked.
If the month of May is warm, abandoned eggs appear before the end of May. Then through June there is no nesting activity and no egg laying. The egg laying starts again the end of June. There must, then, be some behavioural or climatic triggering to ovulation. Perhaps, as in some other species, ovulation is induced by copulation.
Clutch size
Pull clutches vary from one to six eggs, but clutches of one, two, five, and six are relatively rare. The mean clutch size for all nests observed except those robbed experi• mentally is 3.48. The data for these normal clutches is de• tailed in table IV. The difference between the nest totals and the figures given earlier for colony size represents nests broken during laying plus nests not revisited. The nests shown in the table, then, are all those where data is suitable for clutch size analysis.
There is a significant difference (P~.01) between 47
the distribution of clutch sizes for all nests observed in 1955 and those observed in 1956. Sorensen Lake, when the two years are compared, shows a less significant difference (P —.05).
Westwick Marsh 1956 compared with Westwick 1955 shows a sig• nificant difference (Ps.Ol). The Westwick colonies have many more nests than the Sorensen colonies, so the total year figure approaches that for the comparison of the two large Westwick
colonies. At any rate significant differences between the
clutches of one lake from year to year are indicated.
When two colonies are compared in the same year,
there is less difference between them. Westwick compared to
Sorensen in 1955 shows no significance (P—.30). Westwick
Marsh 1956 when compared to Sorensen 1956 and to Westwick
Willows 1956 both show some significance (P=.05). Two col• onies in the same environment may, then, have their clutches
significantly different, but the differences are not as great as those between the years.'
There are three possible factors responsible for differences in clutch size between colonies. These are genetic differences between colonies and between years, environmental
differences, and variability in behaviour. Genetic differences between colonies would be possible if the birds homed to their
colonies year after year and did not mix. The eonsistant size
of each colony suggests that most of the birds do home to their
colonies, but it is probable that enough birds change colonies
to keep gene composition the same. Also Sorensen Lake is higher than Westwick Lake in 1955 and lower in 1956. Genetic CO O O CM CM to CM to CO • « • • • • • to to to to to to to
o to O to CM rH O to
o cr. CM r-i CM O r-i to r-i
in CM f- to O c- o CM VO to to CD N •ri CO. to, c- to O C7\ VD cn VO o r-i VO rH rH O +> r-i O G0 O 00 O O rH rH 6 r-i r-i
r-l O rH O O O O o 3
«H CO +3 O 0) o tO 0 VX> LTi CM to rH O H to tr% 0 rH CM 0$ +> r-i rH rH rH O r-i EH O CD rH £> a EH
ITl^-v LO >» i1n0
rH O rH rH - O HJ CO Hi rH rH differences are certainly not responsible for the large dif• ferences seen. Environmental differences explain the fact that there is more variability between years than between lakes in the same year. But it cannot explain why there is so much differ• ence between colonies in the same year and under the same environmental conditions. Experimental nest robbing brings this out. One fresh egg was taken from each of thirty-two nests within the Westwick Marsh colony in 1956. The environ• mental conditions were exactly the same for robbed and un• disturbed nests. Clutch sizes for the robbed nests are shown in table V. Table V. Induced Clutch Size (by robbing) Westwick Lake 1956 Total of (Marsh colony) Clutches c/l c/2 c/3 c/4 c/5 c/6 c/7 Mean second egg removed 15 0 5 7 2 1 0 0 2.93 third egg removed 14 0 5 7 1 0 1 0 2.92 fourth egg removed 3 0 0 2 1 0 0 0 total (robbed nests) 32 0 10 16 4 1 1 0 2.97 total eggs laid 32 0 0 10 16 4 1 1 3.97 The total number of eggs laid in the experimental clutches is significantly different (P—.01) from the control clutches in the same colony. In the absence of environmental differences the birds responded to the loss of an egg by re- placing it, making up fifty percent of the loss. They are not entirely indeterminate layers, however, as they only made up half the loss they were subjected to. The same percentage of loss was made up whether the third or the second egg was taken. The first egg could not be taken because the nest is then abandoned, and the data for the fourth egg removed are too few for analysis. As a result of this work, I would postulate a dual mechanism for the control of clutch size. First, the birds seem to prefer a certain clutch size. Ovulation is suppressed after this preferred number of eggs is laid, but at any time before the last egg is laid a lost egg is replaced. The pre• ferred number must have some relationship to the fitting of the eggs in the brood pouch, but how this is accomplished is not known. The preferred number must be specific for each bird. Secondly, there are only a limited number of follicles avail• able. If too many eggs are lost in pecking behaviour of nest establishment, loss during egg laying cannot be made up and the preferred clutch size is not reached. The preferred size might not be reached even in the absence of robbing. This explains clutches of one and two. Clutches of five and six represent birds with high preferred clutches and low loss. The preferred clutch size may be an individual character or specific to the age of the bird. At any rate slight differences in the composition of the colonies coupled with equally slight differences in the number of eggs pecked would explain the variability in mean clutch size and distri- bution of clutch sizes. Interval between eggs Clutches are all started very close together in time. In the Westwick Willows colony 1956, the first eggs for all 22 clutches started were laid in four days. (Only 15 of these lend themselves to clutch size analysis.) The larger colonies were more spread out, all first eggs in 18G clutches started in Westwick Marsh 1956 were laid in a thirteen day period. The increased social stimulation of the larger colony is apparently counteracted by decreased social cohesion. The reasons for colony synchronization have been discussed under nest establishment. Data on the interval between eggs in a single clutch are difficult to obtain. In order to get large numbers of records, work was concentrated in the Westwick Marsh colony. Counts took nearly two hours and could not be made more than once a day for fear of disturbing incubation too much. On cold rainy days they had to be omitted altogether because of the danger of the eggs being chilled. Further studies should include more intensive study of fewer nests. In one four day period before their clutches were complete, fourteen birds laid four eggs, fifteen laid three, and three laid five. I have three records of two days skipped between eggs and two of three days between eggs. In one instance I made counts in the evening, and the morning and evening of the next day. The following were noted: 52 1 egg laid during time: no. nests 8:00 P.M. to 10:00 A.M. 15 10:00 A.M. to 6:00 P.M. 13 no eggs whole 22 hours 21 All these birds laid at least one more egg after this period. If the grebes lay at regular intervals, the interval must be more than twenty-four hours and less than thirty-six. This explains a skipped day during laying and laying at any hour of the day. It does not, however, explain clutches where four or five eggs were laid in four days or where two or three days were skipped between eggs. Variability in interval between eggs may be related to the age or gene constitution of the birds. On the other hand, it may be that variation is caused by individual triggering of ovulations. Incubation Incubation begins with the laying of the first egg, and is continued until the last egg hatches. The incubation period is a time of furtive behaviour on open water and strong territorialism in the nesting colony. On open water the adults feed and preen or sleep. They still stay in groups, but there is little contact between individuals. Crests and neck feathers are kept down, and the birds are quiet and shy. At the appear• ance of the canoe, even at a distance, they scatter and dive. Accurate counts of the population cannot be made because of the furtiveness of the birds. One gets the impression that 53 three quarters of the population has disappeared. In the nesting colony, however, noise and confusion reigns. On cool days the birds incubate constantly and almost no noise is heard, but on warm days only half the nests are occupied at any one time. Each bird incubates for ten or fifteen minutes, covers the eggs, and drifts out a little way to feed. It returns in less than five minutes and incubates again. Each time it passes another nest, the occupant of the nest raises its crest, thrusts its head out in a threat display, and chitters. The territory defended is only that within reach of the sitting bird. It threatens in an area a foot beyond its reach. Rarely there are a few bites exchanged in the area be• tween the nests. Between threat incidents the birds sit with heads up or down but eyes always open. Every few minutes they stand, pick at a few bits of nest material, walk a few steps in one direction or the other on the rim of the nest and sit down facing this new direction. The eggs are never turned except by accident in pulling a bit of reed. Even this accidental turning is rare. When alarmed by a strange sound or object, each bird slips silently off the nest and disappears. It swims under• water past the observer or under the canoe and only comes to the surface on open water. When°the intruder is heard some distance away, the bird stands up on the nest and covers the eggs with algae from the rim in three or four deft pokes. It then slides off and disappears. No matter how silently I approached by wading or by canoe, I never surprised a grebe on its nest. This is undoubtedly the origin of the superstition that grebe eggs are incubated by the heat of the decaying vege• tation covering them. If I stayed visible in the marsh over ten or fifteen minutes, the birds began to return. The only sign of their presence was an occasional glimpse of a red eye or a tiny splash as a bird dove again on seeing me. They were all there, however, because less than two minutes after I enter• ed the blind, the nests were all occupied. Both birds incubated, but they were harder to tell apart in this stage because of the lowered crest and neck feathers. The division of incubation time and even the length of an individual turn could not be determined without color marking. Often, I suspect, one bird left to feed and the other came back. When they did meet at the nest, the changing of the guard was very simple. The new bird swam up, the sitting bird hopped off and swam away, and the new bird replaced it. The wing flapping to cool the eggs referred to in the Pied-billed Grebe, Deusing (1), was never seen, on hot or any other days. The incubation period was twenty to twenty-one and a half days for each egg. As incubation was begun as soon as the first egg was laid, the eggs hatched in series, about one day apart. My data are insufficient to prove it, but there is an indication that the last eggs in the clutch hatch in less time than the first ones. This could be a result of more continuous incubation on the full clutch than on the first eggs. A full day before the egg is pipped the peeping of the chick can be heard. By listening carefully, one can hear this even without taking the egg out of the nest. When the egg is pipped, the call is as loud as that of a distressed baby chicken. It takes about three hours for the chick to hatch after pipping, and once hatched it crawls out on the top of the nest to dry. The calling of the unhatched chicks seems to keep the parent birds near the nest until all the eggs are hatched. At any rate I did not find a single abandoned egg in an incubated clutch. The newly-hatched young sit on the parent's back under its wings while the remaining eggs are being incubated. The other parent hunts food for the young and feeds them. The little birds poke their heads out of the parent's wing feathers and take the food from the other parent's bill. Whether both birds continue to incubate after the first young hatch is not known. Care and Behaviour of Young At hatching the young grebes are damp and weak; they lie with heads and legs outstretched on the nest. Within an hour their down is fluffy, their heads up, and their feet tucked under them. They are alarmed by nothing and may even threaten the observer with the adult threat attitude. They make no attempt to escape by swimming. When placed in the water they swim vigorously and dive well. While they are quiet on the nest, in the water they peep continuously. They will accept food from a sharp forceps, held with the points up and down. Pood is taken by grabbing it with a sudden thrust of the head and a twist to wrench the food free. They accept food from the parents as they sit on the adult's back, on the nest edge, or in the water. As they rarely swim during the first four days except when dumped off the parent's back by accident, this may be the time required to develop adequate temperature regulation. Both adults feed and carry the young, one adult carrying as many as three little birds. The pair stays close together and one bird usually carries all the young while the other feeds them. I never saw a grebe carrying more than three young. Higher clutches are, then, very quickly reduced. Crus• taceans and insect larvae are brought one at a time from the bottom of the lake, and the young grab for them as the parent holds them or as they are dropped on the water. If the young bird fails to seize the food, the adult puts its bill in the water and shakes the fragile food into bits. The young birds then pick up the bits. Only one young bird was fed at a time, and active feeding was kept up during all the time that the grebes would otherwise be active. Much additional work on the development of the adult behaviour patterns should be done. This would undoubtedly clarify behavioural relationships. Predation and Mortality The Eared Grebe has no predators during the summer. The lakes on which it nests have no large fish to take eggs or young, and I have never seen the Marsh Hawks, Circus cyaneus. hunting over the lake. There are a great many crows, Corvus brachyrhychos, but they seem to prey entirely on land nesting birds. The many muskrats, Ondatra zibethica. do not bother even abandoned eggs. It has been suggested that coots, Pulica americana, may kill young grebes, but I have never seen coots attack grebes, much less overtake and kill them. The loss of eggs is extremely low. In the Westwick Marsh colony 1956, eleven nests were abandoned and twelve single eggs lost. The nests lost represent 6.1 percent of the total. The egg loss includes all eggs which slide off the unsubstantial platforms. In the early incubation stages these eggs float, but even when on the edge "of the nest they are not retrieved by the parents. An extremely high loss of young balances the low egg loss. About a dozen newly-hatched young were seen on the bottom of the lake in the nesting area. Many more must have been further out because the brood size drops markedly during the first two weeks. The reason for this mortality may well be exposure of the young to wetting or chilling. Their temperature regulation is not established, and they probably cannot stand a rain storm. When kept in a cotton-lined box overnight, two day old chicks could not maintain their temperature and were cold and stiff in the morning. They became active, however, when warmed. They died the second night, but whether from cold or insufficient food or both, I am not sure. Munro (8) suggests that the grebes may kill their own young when the chicks are half grown. Presumably the little grebes continue to try to climb onto the backs of the parents after the parental tolerance of such behaviour has waned. I have never seen evidence of death or injury by this means. Outlook Up to this point, our study of the grebes has been a catalogue of behaviour patterns. This must be continued to include other species in the family. The end here is the under• standing of evolution and relationship. But in addition, work should be done to find out how behaviour is related to ecology in these birds. This involves discovering the physiological basis of sudden changes in behaviour and examining the environ• mental causation of the physiological changes. These studies will be much concerned with colonialism in the grebes and what causes it. The conclusions reached will not only add to our knowledge of grebe behaviour. They will fill gaps in the overall theories of behaviour and ecology of animals. 59 Literature Cited 1. Deusing, M. Nesting habits of the Pied-billed Grebe. Auk, 56:367-373. 1943. 2. Dubois, A.D. An experience with Horned Grebes (Colymbus auritus). Auk, 36:170-180. 1918. 3. Gross, A.G. The Antillean Grebe at Central Soledad, Cuba. Auk, 66:42-52. 1949. 4. Hosking, Eric J. Homed Grebe - courtship and display. Canadian Nature, 11:150-153. 1949. 5. Huxley, J.S. The courtship habits of the Great Crested Grebe; with an addition to the theory of natural selection. Proc. Zool. Soc. London, 25:491-562. 1914. 6. Kilham, L. Courtship behaviour of the Pied-billed Grebe. Wilson Bull., 66:65. 1954. 7. Lawrence, G.E. The diving and feeding activity of the Western Grebe on the breeding grounds. Condor, 52:3-16. 1950. 8. Munro, J.A. The grebes. Occas. Papers B.C. Prov. Mus., no. 3. 1941. 9. Munro, J.A. and Cowan, I. McT. A review of the bird fauna of British Columbia. B.C. Prov. Mus. Special Pub., no. 2. 1947. 10. Simmons, K.E.L. Studies on Great Crested Grebes. Aviculture Mag., 61:3-13, 93-102, 131-146, 181-201, 235-253, 294-316. 1955. 11. Suffern, C. Note of Little Grebe. Brit. Birds, 43:90. 1950. 12. Wetmore, A. Observations on the habits of birds at Lake Burford, New Mexico. Auk, 38:221-247. 1920. 60 Figure 1. Map of Sorensen Lake and the northern tip of Westwick Lake. Reed areas are stippled. Numbers show the sequence of areas used for nesting behav• iour in 1955; letters show the same for 1956. 61 1/4 mile Figure 2. Map of Westwick Lake. Reed areas are stippled, and numbers show-the sequence of areas used for nesting behaviour in 1955. Pig. 4. Sorensen Lake from the east side looking west toward the mountains at the edge of the Praser River. Fig. 6 Westwick Lake from the middle of the east side looking south at the large part of the lake. +5 3 o CD +> a aJ o u .rH o o »"3 •H •H cd o •s o c o •ri +o crj JH • C- 0) JH a 6 o rH •rl » a5 •ri rH -P •ri ft O o •ri o rH O CO ft •rl I JQ CO •P JH o o Pig. 8. An early courting platform of Scirpus and pond weeds. Fig. 9. An area of sparse reeds showing five courting platforms. bb Fig. 10. A later and more substantial courting platform of reeds and decaying algae. Fig. 11. An abandoned egg on a courting platform. cat attitude — penguin dance swimming Figure 13. Drawings of behaviour attitudes. CO 49