UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

An introduction to the behaviour of the goldeneyes : Bucephala islandica and B.clangula (class aves,… Myres, Miles Timothy 1957

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
831-UBC_1957_A8 M9 I6.pdf [ 10.88MB ]
Metadata
JSON: 831-1.0106250.json
JSON-LD: 831-1.0106250-ld.json
RDF/XML (Pretty): 831-1.0106250-rdf.xml
RDF/JSON: 831-1.0106250-rdf.json
Turtle: 831-1.0106250-turtle.txt
N-Triples: 831-1.0106250-rdf-ntriples.txt
Original Record: 831-1.0106250-source.json
Full Text
831-1.0106250-fulltext.txt
Citation
831-1.0106250.ris

Full Text

AN INTRODUCTION TO THE BEHAVIOUR OF THE GOLDENEYES : BUCEPHALA ISLANDICA AND B.CLANGULA (CLASS AVES, FAMILY ANATIDAE) by MILES TIMOTHY MYRES B.A., University of Cambridge, 1953 A THESIS SUBMITTED IN PARTIAL FULFILMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF ARTS in the Department of Zoology We accept this thesis as conforming to the required standard THE UNIVERSITY OF BRITISH COLUMBIA April, 1957 i ABSTRACT In the summers of 1955 and 1956 a field study was undertaken on the behaviour (locomotory, comfort, agonistic, courting, coition and brood) of the Barrow's Goldeneye (Bucephala islandica) in the Cariboo District of British Columbia. This is the fi r s t stage of a five year study of the behaviour of Sea Ducks of the Tribe Mergini. An attempt has been made to summarise the behaviour of B.islandica and also of the closely-related species B.clangula, that is described in the literature. Generic similarities, and specific differences, in the courtship displays of these two species have been described. An attempt has been made to discover the likely sources of these displays from among the locomotory, comfort ar agonistic behaviour patterns. Illustrations of the postures and movements described in the text have been made from films by the author, and two other sources. An attempt has been made to link items of behaviour to the ecological, biological and systematic aspects of duck biology. Thus both the behaviour itself comes to have a biological setting and raison d'etre, and behavioural aspects of breeding and population biology, and population management and manipulation,may be better understood. Among Comfort Movements Drinking appears to have given rise to a courtship display found chiefly in the pre-coition sequence. Wing- and Leg-Stretching is also a comfort movement found i n display (only in the pre-coition behaviour). Preening generally (especially Splash-Bathing) occurs frequently under conditions of stress e.g. after a territorial encounter and after coition. The Upwards-Stretch, Wing-Flap, Tail Wag complex of i i movements is also frequent under conditions of stress, and occurs especially as a "signing-off" (? appeasement) display at the end of an encounter. Agonistic behaviour i s interesting, i n that the diving ducks tend (goldeneye particularly) to use underwater diving as their major aggressive tactic. Threat may derive from intention diving. The "alarmed" position is hard to derive from other postures. Inter-specific aggression i s fre-quent and indicates that the Aythyini are more of an irritant to goldeneye than are the Anatini. Case-histories of inter-specific aggression helped to throw light on the nature of territory in goldeneye, and on fatigue and refractory periods in the attack motivation. Goldeneye have a wide variety of courtship displays. The Head-Up and Head-Raised positions (with the Neck-Withdrawing movement) are rather alike in the two species. The Head-Throw of B.clangula is much more ex-traae in form than the equivalent movement in B.islandica. In B.islandica the Pseudo-Kick and Kick are much less differentiated, than the Head-Throw and Head-Throw-Kick of B. clan aula. An analysis is made of the two forms of pumping motions, to show their basic similarity, but species differentiation. The Coition Sequence is very similar in the two species. There is a marked Post-Coition display. The downies leave the nest by scrambling up the inside of the nest cavity, and tumbling from the nest-hole a l l together. The female has a special ca l l which is used to force a l l the downies off at one time. On the lake the same call gathers; the young together around the female. The i i i importance of this means of ensuring that the young fledge together is discussed. The actions of the female at fledging are described. Females are poor guardians of broods, but they show considerable hostility towards each other during the brood period. Downies greet each other and the female by a movement which at first resembles the Rotary Pumping of the adult drake, and the female. Rotary Pumping thus i s a "greeting" movement. RESERVATIONS ON USE 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 i t freely available for reference and study. I stipulate, however, that owing to the fact that some of the raw data of this thesis was loaned to me by other students of goldeneye behaviour for use in this thesis only, restrictions must be placed on its use hereafter. Publication of the text i s reserved to the author. Use of figures, or quotation, from the text^is only permitted i f both I, and the other parties concerned, have given leave in writing. Mr. Charles Walcott and Mr. Benjamin Dane may be reached through the Dept. of Zoology, Cornell University, Ithaca, New York, U.S.A. Dr. Frank McKinney may be located at the Delta Waterfowl Research Station, Delta, Manitoba, Canada. The author may be reached through the Dept. of Zoology at the University of British Columbia, Vancouver, Canada. The Librarian in making this thesis available for study shall obtain written assurance from the loanee that these conditions are understood, and that the loanee will abide by them. Copying or publication of this thesis for financial gain will, likewise,not be allowed, without the written permission of the persons mentioned above. Department of Zoology The University of British Columbia, Vancouver 8 , Canada A p r i l , 1957. ix AC KNOIvLEDGEffi NTS Dr. I. McT. Cowan, Head of the Department of Zoology, University of B. C, invited me to Canada and suggested subjects for study. I am grate-ful for a l l that he has done to encourage and facilitate this work. With-out his interest and enthusiasm i t would never have been a success. I am indebted to my mother, Mrs. J.M.L.Myres, for financial aid that enabled me to travel to British Columbia, and to live through the first year. During the years 1955-57, I was supported by a Wildlife Conservation Fellowship under the auspices of Canadian Industries Limited. I was also an Overseas Travelling Scholar of the Worshipful Company of Goldsmiths' of London. To the interest in pure learning, and in education, exhibited by these bodies I am especially grateful. In the summers of 1955 and 1956 equipment and supplies were made available under a Grant for Research from the National Research Council of Canada. I am glad to acknowledge the loan of a movie-camera in 1955 from Mr. R.J.Pop, of Vancouver. This thesis would be considerably the poorer, were i t not for the pictorial material, the greater proportion of which has been made avail-able to me by other ethologists. I am especially grateful to Mr. Charles Walcott and Mr. Benjamin Dane, formerly of Harvard College, for allowing me to work over their film, while their own paper was yet unpublished. This was an immense privilege. I cannot express my gratitude strongly enough. Dr. Frank McKinney of the Delta Waterfowl Research Station, Mani-toba, also lent me some film, and has been a constant source of moral support. X Mr. Theed Pearse of Comox, B.C., kindly allowed me to look through his diaries for information on goldeneye displays. Mr. G. F. van Tets and Miss Mary F. Jackson made available translations of foreign papers. Messrs. George Tucker, Clark Tucker and Ward Herrick of Williams Lake, B. C., allowed me to camp on their property, and were kind at a l l times. Mrs. Nancy MacAllister often helped me in the field. I am grateful to my friends Mary F. Jackson, Dr. B. Baggerman, and R.B.Weeden for reading parts of the manuscript (often more than once). Mary Jackson was a constant source of reference for biological facts about Barrow's Goldeneye. Dr. Baggerman helped me erradicate some of the worst errors in regard to ethological theory. Dr. M.D.F.Udvardy supervised the study, and was a continual source of ideas. I am grateful to Dr. W.Drury of Harvard for putting me in contact with Charles Walcott and Benjamin Dane. iv . TABLE OF CONTENTS INTRODUCTION 1 Ethological Studies and Phylogenetic Relationships 2 The Derivation of Sexual Displays 5 Current Studies and the Literature 11 The Pacific Coast 13 Equipment 14 Organisation 16 THE GOLDENEYES (BUCEPHALA ISLANDICA and B.CLANGULA) 17 THE DISTRIBUTION AND ANNUAL CYCLE OF BARROW'S GOLDENEYE 18 Geographical Distribution 18 Moult Migration as an Outbreeding Mechanism within the species 19 Matrilocality (with Coastal Pairing) as an Isolating Mechanism within the species 20 Hybrids 21 Winter Distribution and Winter Behaviour 24 Spring Migration 27 Territory (Mating and Brood) and Nesting 28 Deficiences in Brood Guardianship 30 Age at first breeding 31 Post-breeding Dispersal 31 Summary 32 NON-SOCIAL BEHAVIOUR 34 Locomotory Movements 35 Flight 35 Diving 39 Swimming 42 Walking 44 Comfort Movements 47 Feeding 49 Drinking 51 Tail-Wag 52 Head-Shake Complex 53 Water-Dip Complex 55 Upwards-Stretch 56 Wing-Flap 57 Wing- and Leg-Stretch 58 Yawning 59 Ear-Scratch 59 V Forward-Preening 60 Sideways-Preening 60 Back-Preening and.HeadrEolliBg 60 Splash-Bathing 6l Thigflotaxis 62 SOCIAL BEHAVIOUR , • 73 Agonistic Behaviour 74 Intra-specific Agonistic Behaviour between males -74 Crouched Posture • 79 Laying Neck on Water (Threat Posture) 81 The Attack 84 Surreptitious Encroachment 39 Expectancy Look ("Alarmed" position) 91 Inter-specific Aggressive Behaviour (inc. Table I) 96 Courting Displays 108 General Account •• 108 Female 117 Jiving 120 Deep Jiving 124 Male 128 Head-Up Neck-Withdrawn 128 Head-Raised Neck-Withdrawn 132 Pseudo-Kick and Kick 141 Head-Throw 148 Head-Throw-Kick 158 Rotary Pumping 179 Bowsprit Pumping and Bowsprit Position 187 Masthead • -201 Backwards Swimming Display 204 The Coition Sequence 209 General Account • .• -209 Invitation to Coitus • 211 vi Water-Flip 212 Wing- and Leg-Stretch *... 213 Jabbingj Wing-Preen, Pre-Coition Steaming and Mounting, and Post-Coitfpoion Rotations 215 Post-Coition Steaming and Splash-Bathing 218 The Brood • 234 Leaving the Nest 234 Hostility in the Female 242 Searching for and leading young 243 'Greeting Behaviour in Females and Downies 244 Summary of Similarities and Differences between B.islandica and B.clangula 247 APPENDIX 1 249 LITERATURE CITED 250 V l l Page Barrow's Goldeneye (B.islandica). Adult male, Westwick Lake, B.C. 1955 ~" Ahead of Page 1 A. Comfort Movements; THREE STUDIES OF DOWNIES (B.islandica) Figure 1 46 DRINKING (B.islandica) Figure 2 65 UPWARDS-STRETCH (B.clangula) Figure 3 67 UPWARDS-STRETCH & WING-FLAPPING IN THE DOWNY (B.islandica) Figure 4 69 SPLASH-BATHING (B.clangula) Figures 5 & 6 . . . 71 & 72 B. Aggressive Behaviour; LAYING NECK ON WATER ("THREAT" POSTURE) (B.islandica) Figure 7 107 C. Courting Displays; DEEP JIVING (B.clangula) Figure 8 127 HEAD-UP WITH NECK-WITHDRAWN (B.islandica) ... ' . Side View 7 Figure 9 136 Front View with Head-Turning Figure 10 137 HEAD-RAISED WITH NECK-WITHDRAVJN (B.clangula). Side View Figure 11 139 Rear View with Head-Turning Figure 12 140 PSEUDO-KICK (B.islandica) -Side View Figure 13 167 Rear View Figure 15 170 Front View Figure 16 171 KICK (B.islandica) "Side View Figure 14 168 Front View Figure 36 171 HEAD-THROW (B.clangula) Side View Figure 17 173 Front view. Figure 18 174 Rear View Figure 29 208 With Forward Bowsprit Position ....... Figures 19 & 29 . .175 & 208 HEAD-THROW-KICK (B.clangula) Figures 20 & 2 1 . . 177 & 178 V l l l Page ROTARY PUMPING (B.islandica) Simpie Form Figure 22 194 With Laying Neck on Water Figure 23 195 BOWSPRIT PUMPING & BOWSPRIT POSITION (B.clangula) ~~ With Bowsprit Pumping Figure 24 197 Bowsprit Position static from Forward Scoop Figure 25 198 Bowsprit Position static from Head-^ Throw Figures 19 & 29.. 175 & 208 ANALYTICAL COMPARISON OF ROTARY & BOWSPRIT ? PUMPING Figure 26 200 MASTHEAD (B.clangula) Side View Figure 27 206 Extreme Form Figure 28 207 Front View Figure 29 208 D. The Coition Sequence; B.islandica; WING- AND LEG-STRETCH Figure 30 222 POST-COITION STEAMING - Figure 36 233 B.clangula WING- AND LEG-STRETCH Figure 30 222 WATER-FLIP Figure 31 • 223 PREENING, AND PRE-COITION STEAMING • Figure 32 226 MOUNTING AND COITION Figure 33 228 POST-COITION.-ROTATIONS • Figure 34 230 POST-COITION STEAMING - Figure 35 ..• 232 From film by M.T.Myres (B.islandica) Figures 2,3,4,7,22,23 and 36 From film by Charles Walcott & Benjamin Dane (B.clangula) Figures 1,5,6,8,11,12,17,18, 19,20,21,24,25,27,28,29. From film by Dr. Frank McKinney Figures 9,10,13,14,15,16,30, (B.islandica and B.clangula) 31,32,33,34,35. 1 INTRODUCTION 2 INTRODUCTION Ethological Studies and Phylogenetic Relationships Delacour and Mayr (1945) pointed out, in their much needed revision of the classification of the Family Anatidae, that the use of behavioural characteristics as taxonomic indicators is a respectable activity, because many innate behaviour patterns are as ancient, or more ancient, than the morphological characters which they emphasize. Four years earlier, in a paper that Delacour and Mayr had not seen when they wrote their own, Lorenz (1941) had produced the most elaborate analysis, so far, of the courtship displays of ducks. He described 14 or more elements in the behaviour of 15 species of the surface-feeding ducks (Tribe Anatini). Thereby he estab-lished a precedent for the serious ethological comparison of different groups. Lorenz1 paper (I have referred only to the recent translation in The Avicultural Magazine, 1951-52) was almost entirely descriptive, though in i t he made a number of statements as to the possible origin of certain displays. These may well have to be revised as further work on motivation of activity, and the ontogeny of movements and postures, is done. Primarily Lorenz was interested to use the presence or absence of the various dis-plays, alone or in combination with each other, as a means of determining phylogenetie relationships, and he illustrated this by drawing up a symb-olic "shaving brush" of the species with cross-connections from one "hair" (or species) to another to demonstrate common possession of a cer-tain behavioural element. The number of discrete movements a species 3 shares with another species is an indication of how closely related they are. It was on these lines that the present 5-year survey of the Sea Ducks, the Tribe Mergini, was initiated, of which this report on the goldeneyes (Bucephala islandica and B.clangula) i s the first result. It was considered that i f innate behaviour patterns were "persona" grata with the systematists, and could be treated equally with morphological characters, the time was ripe to try and review the Mergini, which are in a number of ways a very unusual group. Recently ethologists have ceased the use of the word "innate" for patterns which have not been shown (by isolation experiments with young birds from before hatching) to be truly determined by genetic inheritance alone, with no learning involved. There are, however, display patterns which we can be sure will prove eventually not to have been learned, and so, although the word "innate" has been oraraitted from a l l discussion hereafter, the stereotyped nature of the displays to be described will be regarded, in the absence of better evidence, as being sufficient indication of their inborn character. The Mergini merit study, both to elucidate their evolutionary position among the ducks as a whole, and because of some unusual features they possess. The group consists of seven genera, one of which (a single species) is recently extinct, while three others of the remaining six are also mono-typic (Scott, 1 9 5 l ) « Unless the Mergini are very ancient, and these are relict species from a much larger speciation history, there is good reason to look again at these genera to learn more of their relationships. At the same time i t is necessary to study the fossil and zoogeographic histories 4 of the genera i n the group. Delaeour and Mayr (op.cit.) were very clear about what the relation-ships of the various genera were to each other. Yet the evidence i s not, i n my view, entirely complete or beyond re-evaluation. They wrote: Delaeour (1936:376) , as well as Heinroth and other authors, has pointed out the obvious relationship of the merganser (Mergus) with the golden-eyes (Bucephala); and i n spite of the wide difference between the extreme forms of the tribe (Mergus and Somateria), the sea ducks form one of the most closely knit subdivisions of the anatine subfamily. The seven genera are connected with one another by intermediate species. The Hooded iyierganser (Mergus cucullatus), for example, connects the larger mergansers, through the Smew ("Mergellus" albellus) and the Buffle-head (Bucephala albeola) to the golden-eyes. The Harlequin (Histrionicus) i s a link between the Old-squaw (Clangula) and the scoters (Melanitta), as i s the Labrador Duck (Camptorhynchus) between the Old-squaw and the eiders (Somateria). On the other hand, the golden-eyes, the Old-squaw, and the Harlequin are undoubtedly related, as i s proved by the same bold pattern of dark gray and white of a l l their downy young. However, except for the mergansers (Mergus), of which there are seven, no genus contains more than four species. Each genus i s rather uniform i n i t s species, but rather different from every other genus, though both eiders (Somateria sp.) and scoters (Melanitta'sp.) are very heavily-built for ducks. Certainly the Mergini as a group must be deemed less uniform i n type than either the Anatini or the Aythyini. Delaeour (1956) has recently separated the eiders as a distinct tribe, the Somateriini, and he t e l l s me (in l i t t . March 29, 1957) that he has done so on the basis of the plumage patterns of the females and the downies, and also on behavioural characters "as pointed out by Humphrey". We await further details. It is of interest that (as i n the above quotation) the characteristics of the downy young are considered to indicate the common ancestry of modem genera. However i n the matter of displays Delaeour and Mayr (op.cit.) describe the "Head-Throw" 5 display in both mergansers, and, quoting Hochbaum (1944), in the Canvasback (Aythya valTi sneria) • They do not apparently realise that this display is almost certainly homologous in the two tribes, and from des-criptions of eider display (Somateria mollissima) quoted from Hoogerheide (1950) by McKinney (unpubl.M.S.) I believe i t will be shown that the eiders show close relationship both to the scoters and to the scaups jAythya sp.). Delacour and Mayr (op.cit.) and Delacour (in l i t t . March 29, 1957) are convinced however that the Hergini are most closely related to the Perching Ducks (Cairinini). This thesis is confined to an account of the behaviour of the two true goldeneyes (I use this term here instead of Bucephala because the displays of B.albeola are completely unlike those of B.islandica and B.clangula, and B.albeola appears to be quite unrelated to them). The intention, eventually, is to describe the postures and movements employed in courtship throughout the Mergini in the manner employed by Lorenz (1941)« B.islandica and B.clangula have a rather large number of displays having sexual motivation, and in this respect are a useful basis upon which to draw comparisons from other, less effusive, species. They f i l l the same important position as the Mallard (Anas platyrhynchos) in the paper by Lorenz. The derivation of sexual displays. Lorenz (op.cit.) did not go at a l l deeply into the form and variety of the various non-sexual movements, such as those involved in preening, bathing or stretching. These are now classed together as "comfort move-6 ments". They are concerned with the maintenance of healthy body plumage and with satisfactory muscle-tonus. Nor did Lorenz consider in detail the maintenance activities of feeding and drinking. Fortunately McKinney has now performed such an analysis for the Anatini (unpubl. Ph.D thesis University of Bristol 1953)* He has demonstrated the absolute necessity of studying a l l the motor patterns, of a visible nature, that occur in the repertoire of a duck, i f a true understanding, or an exact analysis, of the behaviour of the whole animal is to be made. The second major function, therefore, of an ethologLcal study, is the establishment of the source, or origin, of courtship displays from the motor patterns having non-sexual internal motivation. In many recent studies, notably in highly territorial vertebrates such as birds or fish (Tinbergen, 1 9 5 1 j l 9 5 3 , et al.), i t has been found that sexual displays can generally be explained as biproducts of (or as accid-ental] y derivations from) such conflict situations as arise from opposing motivations e.g. the tendency to aggression being equally strong as the tendency to escape. Furthermore the displays often resemble attitudes as-sumed during aggression or flight. Daanje (1950) has discussed the origin of displays from the intention movements of locomotory activities, such as swimming, diving, and flying. It has become customary, therefore, to suggest that sexual displays generally derive from such aggressive or fleeing attitudes in birds and fish. This judgement is unlikely to be substantiated, as a general law, when rather more species have been studied. For a number of cases are now known where 7 sexual displays are derived by ritualisation, not of aggressive or fleeing attitudes, but of other activities such as comfort movements. This appears to be the case in goldeneyes. It needs to be questioned, furthermore, whether the conflict situation which gives rise to displacement activities, and hence ritualisation of aggressive, fleeing, or irrelevant preening etc., into displays, must always be one involving aggressive motivation and fleeing motivation. E&y there not be a direct tension between the sexual drive per se and the escape drive? Since fleeing by the drake is not an activity which a sexually-motivated female can do much to correct, any irrelevant performance by the drake might rapidly and easily reduce the tendency of the male to flee from the female. It might do this either by becoming symbolic to the female of a tension within the sexual activity of the male, or by means of a feed-back mechanism which would tend further to arouse the sexual moti-vation of the males so that fleeing does not take place. This latter is possibly the case in goldeneyes, for the female may remain motionless in the posture inviting coitus for as long as 20 minutes while the male swims slowly around her "Wing-and Leg-Stretching","Water-Flipping" and performing other sexual displays, for a considerable time, before he is seemingly worked into the mood to mount the female. Since sexual motivation towards copulation must be assumed to be as old, historically, as the activities of mutual shedding of sex products, or copulation itself, there is every reason to look for the origin of sexual displays as much in conflicts between the sexual drive with the ten-dency to escape, as in conflicts of aggressive tendencies with those of 8 escape. Especially is this so since "individual distance" towards the mate is very small indeed, or non-existent, in a solidly paired male and female. The stimulating effect of display upon the female such that she "Jives" or invites coitus may therefore help to correct the actions of the two individuals in favour of success in pairing or mating. Feed-back . as a mechanism reducing the escape tendency, may also be important. It is important here to distinguish the "individual distance" wLthin which another bird i s attacked, from that presence of a bird close-by which induces immediate retreat. Whatever the truth of the matter;the unfulfilled intentions of aggre-ssion or escape, as well as displacement activities in other fields, may acquire sexual signal functions of their own. Lorenz (1941) himself sug-gested some originslof displays from comfort or other activities, though he produced l i t t l e analytical proof of the correctness of his statements. He pointed out, for example, the close resemblance between the display of the Mandarin Duck (Aix galericulata) in which the male dips the beak behind a greatly enlarged, and highly coloured, secondary feather which projects above the back and true wing preening from which i t undoubtedly derives. And this example also illustrates the important principle that a movement precedes, in such cases, the development of morphological charac-teristics, which are now associated with the movement and emphasise its visual effects. In the last 15 years a great many examples of the deriva-tion of displays from comfort movements, or from locomotory movements (or rather the intention movements of locomotory activities) have been described. Movements are regarded as being older than the bright features they display. The colour patterns of the head of the Goosander (Mergus merganser) 9 are different from those in a goldeneye. Yet each has the Head-Throw. Any correlation between colour patterns of the head and this basic court-ship display are therefore secondary. Although other displays may exist (different in the two species) which are emphasised by the differing colour patterns, I am rather inclined to believe, now, that the shape of the patterns is not completely specific to the displays. The effect of a displaying sea duck must always be striking. They are a most highly coloured group. "Threat" in goldeneye is a grotesque spectacle (Fig.7). But i t may be rather in the overall effects as a stimulus that colour patterns are important. The actual colour patterns are so distinct from -species to species that there can be no difficulty whatever in recognising the male. We are left to assume either that the evolution of colour pat-terns is not an isolating mechanism on a distinctly specific level, or that Mergini are a relict group in which the intermediate species have been lost, or that sexual selection is operating to produce fantastically plumaged males. No work has been done on this selectivity of females for males of varying plumages. While there may be a slight surplus of males in the population, plumages (except for juvenal birds in their second sum-mer) do not appear to differ greatly in quality. In the light of the principles enunciated earlier i t has become -increasingly clear that the "whole animal" has to be included in any study of sexual displays, i f any "meaning" is to be found in the various display activities. In order to discover any phylogenetic significance (or any origins, 10 for that matter) of sexual displays i t is necessary to study more than one species, otherwise there is no means of comparison. McKinney was kind enough to warn me (in litt», 1954) that to avoid the unintentional suggestion of homologies with surface-feeding ducks, i t would be necessary to invent quite separate descriptive names for the display movements seen in each species, and only to equate these with each other when one had observed the whole range of movements in each species and had evidence that the displays were homologous. A separate name has therefore been given to each movement in each species, even when they are believed to be the same. The same principle applies when describing the signal meaning, or function of a display. The "Laying Neck on Water" posture of the golden-eyes without any doubt has a "threat" function, but i t is necessary to give i t a descriptive name first and only a functional name (or synonym) afterwards. It would be understood that only when discussing goldeneye would this functional name necessarily apply to the "Laying Neck on Water" posture. A l l the headings in the Table of Contents in this thesis are descriptive names. The meaning or function of a display is sometimes discussed in the text. It may be argued, by the purist, that Lorenz (or his translator) was wrong to call the "Jiving" movement an Inciting movement, for this is completely anthropomorphic. We do not yet know whether this movement has the effect exclusively of stimulating the male, or whether i t also indicates to one male that he i s the chosen mate, and to other males nearby that they are not welcome. In the eiders the movement is not a single 11 whole, but is in two separate parts, with different meanings, i f the homologies that have been suggested are correct. Certainly the movement should, have a descriptive terminology, for the movement occurs in a number of different contexts (e.g. as between courting parties'" and territories), and may have different meanings to different birds i n the group. Current Studies and the Literature The literature of the courtship displays of the ^ergini is scant in the extreme. When I had already done my field work on B.islandica for two summers I discovered that Sawyer (1928) had in fact described, and i l l u s t -rated (from field sketches), a l l the sexual displays that I had seen, including coition. This was a blessing in disguise for my field observa-tions were made without any knowledge of his paper, and our descriptions prove to be extremely alike. We thus confirm each others work. Boa-row's Goldeneye becomes, most probably, the best known of the sea ducks in this respect. For B.clangula the written records are less useful. Bruggemann, as early as 1876, described coition and this description has not been bettered. Brewster (1911) wrote a paper which is the best description of the general courting displays so far. There are a few other literature descriptions of B.clangula from both Europe and America. It must be remarked that ethological observation requires considerable training of eye and mind, and i t is very easy to see and describe a move-ment or sequence of movements erroniously. Most observations recorded in the literature are written by ornithologists unskilled at describing this kind of event accurately, and usually from single occasions only. Thus they are mere estimates of the essential character of the courtship and 12 are comparatively l i t t l e use for a study of this kind. Since most of c the descriptions refer to the "Head-Throw" which i s the most frequent display i n courting parties of B.clangula but i s found also throughout the Mergini and also i n the Aythyini i t i s clear that but few of the distinguishing displays have been adequately described i n any of the genera involved. Coition has been described i n almost none of them. It i s very important that coition should be described i n other species of Mergini i n the near future. It may be easiest to do i n a waterfowl collection (unnatural surroundings, hence other sexual displays may per-haps be li a b l e to errors of frequency and occurrence). P i c t o r i a l illustrations have so f a r only been able to i l l u s t r a t e static poses. These may be assumed during more than one type of display e.g. as i n the Bowsprit pose of B.clangula and are therefore unsatisfact-ory. They are not precise and do not fi n a l i s e any description. It became necessary to i l l u s t r a t e movie-film sequences on a f u l l page plate, by drawing selected frames from the sequence. This technique has been ap-plied i n this thesis. I have made drawings of certain displays from my own films of B.islandica. Dr. Frank McKinney, of the Delta Waterfowl Research Station at Delta, Manitoba, has been kind enough to allow me to make drawings of both B.islandica and B.clangula from a short length of film taken i n England. About half the plates included here are ones made up from a film of B.clangula taken i n 1956 by Charles Walcott and Benjamin Dane (both now of Cornell University) while they were at Harvard College. The film was taken at Newburyport, Massachusetts. It was a great privilege and pleasure 13 to be able to work over this f i l m and use i l l u s t r a t i v e material f o r the purpose of comparison with B.islandica. At the present time the only known studies on the behaviour of the Mergini are those being made currently by McKinney at Delta on B.clangula and on the Aythyini. Dane at Gornell University i s working on B.albeola and mergansers. McKinney, P.S.Humphrey at the University of Michigan, and Gudmundson (Scott, 1953) have unpublished material on the Common Eider (Somateria pfollissima). : s The Pacific Coast Probably the most rewarding area i n the whole free world for the study of the Mergini i n the f i e l d , l i e s along the Pacific Coast of North America from Puget Sound to Point Barrow, Alaska. Eiders, Old-Squaw (Clangula hyemalis), and scoters breed along the shores of the Bering Sea and the arctic coasts of Canada. Two species of Bucephala, the Harle-quin (Histrionicus histrionicus), and even one scoter and three mergansers could be studied i n the summer i n central British Columbia. However much the most rewarding time appears to be i n the winter,when, with the excep-tion of the eiders (all- four of which could however be studied i n the Aleutians) and three species of merganser which have r e l i c t populations i n the southern hemisphere and eastern Asia, a l l but one (the eurasiatic Smew, Mergus albellus) of the remaining world species of Mergini can be studied on the South-East coast of Vancouver Island, i n the Gulf Islands of the Strait of Georgia, and i n the v i c i n i t y of Vancouver, B.C. For this reason the University of British Columbia i s ideally situated for a study of the behaviour of the Mergini, and there i s a great deal to be 13 to be able to work over this film and use i l l u s t r a t i v e material for the purpose of comparison with B.islandica. At the present time the only known studies on the behaviour of the Mergini are those being made currently by KcKLnmey at Delta on B.clangula and on the Aythyini; Dane at Cornell University i s working on B.albeola and mergansers. McKinney, P.S.Humphrey at the University of Michigan, and Gudimindson (Scott, 1953) have unpublished material on the Common Eider (Somateria ntfollissima)^ The Pacific Coast Probably the most rewarding area i n the whole free world for the study of the Mergini i n the f i e l d , l i e s along the Pacific Coast of North America from Paget Sound to Point Barrow, Alaska. Eiders, Old-Squaw (Clangula hyemalis), and scoters breed along the shores of the Bering Sea and the arctic coasts of Canada.. Two species of Bucephala, the Harle-quin (Histrlonicus histrionicus), and even one scoter and three mergansers could be studied i n the summer i n central British Columbia. However much the most rewarding time appears; to be i n the winter.when, with the excep-tion of the eiders ( a l l four of which could however be studied i n the Aleutians) and three species of merganser which have r e l i c t populations i n the southern hemisphere and eastern Asia, a l l but one (the eurasiatic Smew, Mergus albellus) of the remaining world species of Mergini can be studied on the South-East coast of Vancouver Island, i n the Gulf Islands of the Strait of Georgia, and i n the v i c i n i t y of Vancouver, B.C. For this reason the University of British Columbia i s ideally situated for a study of the behaviour of the Mergini, and there i s a great deal to be 14 said in favour of establishing a waterfowl research station in the Fraser Delta area or in the Gulf Islands. The only other Canadian research station of this type i s the Delta Waterfowl Research Station at Delta, Manitoba. There is room for a similar station on the Strait of Georgia or Puget Sound region of the Pacific Coast. Equipment Two technical aids proved essential. In the summer of 1955 a 16 mm. Cine-Pathe model movie-camera was used to take about 1000 feet of Kodach-rome film of B.islandica. Using an Animated Viewer (Editing Apparatus) i t is possible to select individual frames showing the extreme of a move-ment, and to make camera-lucida-type drawings of selected frames, such as illustrate this thesis. Placing such drawings side by side on a single sheet of paper i t is possible to demonstrate in detail the form, extreme positions, and relative speed of a "posture-movement" element the whole "going-into and coming-out-of a posture", in a sequence that may last as short a time as l/lOth of a second, or as long as 2 seconds. The Head-Flick that ends many movements e.g. the "Head-Throw", was not seen in the field, and occupied only 2 frames on film being taken at 24 frames/ second. For the f i r s t time, therefore, i t is possible without actually showing a film to an audience, to illustrate in print the form of a move-ment. Some of the drawings in the literature, because they are not the extremes of a movement, give but a grotesque idea of the movement that actually occurs. The second technical aid (used for the first time in 1956) was a 15 "Midgetape" tape-recorder (made by Mohawk Business Machines, Inc.). This light instrument uses hour-long tapes, broken into two g-hour lengths which are rapidly changed around. It is possible to carry any number of tapes i n the f i e l d . This invaluable machine allows the observer (while holding field-glasses) to dictate an exact record of a l l that happens in front of him, as i t occurs. This rules out the necessity of a prodigious (and accurate) memory, or a stenographer as field-companion. Ducks move rapidly and i t i s not possible without a tape-recorder to record accur-ately the order of movements in a sequence, or to distinguish the individ-uals performing them. In really fast sequences film i s the only completely accurate means of recording a sequence. Film i s also the only way of re -cording the exact "anatomy" of a movement. But film i s too expensive to use a l l the time, while tapes may be used over and over again. Tape-recordings have the power to demonstrate the stereotyped nature of some sequences. For example the goldeneyes have a great many different pos-tures, but very l i t t l e i n the way of an organised use of these i n a stereotyped order, or orientation. On the other hand the Bufflehead (B.albeola), which appears to have very few sexual displays, was clearly shown (by tape-recordings) to employ i t s normal powers of f l i g h t and swimming (with the Nodding display) into an elaborate series of orienta-tion sequences — or "dances". Owing to the speed of movement and i t s repetitive nature, these can best be illustrated by means of tape-recordings of the human voice describing the orientation of the birds, with some film sequences of the individual "sets". 16 Organisation This thesis w i l l be i n three sections: — (1) Distribution and Annual Cycle of Barrow's Goldeneye. It i s v i t a l to place the behaviour ( i t s form, time of occurrence, function and deriva-tion) i n the context of the annual cycle (pairing, migration, nesting, -post-breeding distribution and winter associations). The descriptions of the behavioural elements i s thus preceded by this account of the an-nual cycle and the part played by the sexual activity and i t s form i n zoogeographical distribution, migration etc. One needs to know why pair-ing occurs as, and when, i t does, and what i t s effects are, i n terms of the whole biology of the animal. (2) Non-Social Behaviour. The Locomotory arid Comfort Movements w i l l be described with no particular distinctions drawn between the two species of goldeneye, as i t i s not known that they d i f f e r i n any substantial ways in these a c t i v i t i e s . The descriptions are however necessary so that i t may be seen that certain social displays have similarities with and origins from these more primary act i v i t i e s . (3) Social Behaviour. This includes a number of spheres of activity. In the description of courtship displays in particular the descriptions of displays i n the two species w i l l be separated. This i s often necessary for the further reason that we cannot be certain of the homologies between them. Under each movement there w i l l be separate headings for Field, Film and Literature Descriptions. This i s not an ideal situation, but i s nec-essitated by the nature of my sources, and the necessity at this stage of distinguishing them. 17 THE GOLDENEYES (BUCEPHALA ISIANDICA & B.CLANGULA) THE DISTRIBUTION AND ANNUAL CYCLE OF BARROW'S GOLDENEYE. 18 THE DISTRIBUTION AND ANNUAL CYCLE OF BARROW'S GOLDENEYE Geographical Distribution The Common Goldeneye (B.clangula) breeds i n North America and across Siberia to Europe, but i s not found i n Iceland or Greenland. Barrow's Goldeneye (B.islandica) occupies the remaining suitable areas of the Holarctic. It i s found breeding only i n Iceland, possibly i n Labrador, and westwards of the eastern f o o t h i l l s of the Continental Divide of North America. From these regions B.clangula i s absent. There are only four authoritative records of B.clangula breeding west of the Rocky Mountains i n British Columbia: — twice at Swan Lake i n the Okanagan Valley i n 1930 (Munro, 1935), at Vanderhoof (Munro, 1949), near Enderby i n 1946 (a female taken from a nest,now in the Collection of the Depart-ment of Zoology at the University of British Columbia) and, less reliably, i n 1895 at Sumas (Brooks and Swarth, 1925). Murie (1946) records breeding of B.islandica i n Mount McKinley National Park. A few B.islandica breed i n the Peace River Country, the Yukon and central Alaska, where there is slight overlap with B.clangula (Cowan, 1939; and J.W.Brooks (pers.comm.)). Main breeding ranges are however quite distinct and the examples cited are the only known exceptions. The histories of the distinct pop-ulations of B.islandica i n Iceland and British Columbia are of profound systematic importance. However i t w i l l only be possible here to discuss the part played by courtship displays i n maintaining B.islandica and B.clangula as distinct species, and i n distinct breeding ranges where they nowadays approximate each other geographically i n western North Amer-ic a . 19 MoultFMLgration as an Outbreeding Mechanism within the species. For ethological purposes the annual cycle may be said to start at the time that the two sexes meet on salt water, in late October or early November. The adult drakes reaching the coast a week or two ahead of the females and young. This meeting is the climax to a very remarkable phen-omenon occurring in most diving ducks, namely the almost complete disap-pearance of the males from the breeding grounds during June and July. Where the drakes of B.islandica retire is as yet unknown. Swarth (1926) records seeing males flying south in flocks in mid-June at Atlin on the Yukon border. In Iceland males of this species gather in large moulting flocks on the breeding lakes themselves or on the sea nearby (Phillips, 1925 quoting Hantzsch). "Phillips suggests a northward flight and Hochbaum (1955) gives evidence for this in other species. Individual drakes during eclipse would not be quickly identified by any but a competant observer. But assuming that they are gathered together in a very few bands, as in the Canvasback (Hochbaum, 1955)> i t is s t i l l surprising that large groups of moulting goldeneye have not yet been re-ported in British Columbia. It seems most likely that they retreat to a few lakes in the Coast Range, or to one or two of the hundreds of deep fjords of the sea which run up into the coastal mountains of British Col-umbia and the Alaska Panhandle. Both biologically and ethologically this late summer period i s important. While the drakes move out a large proportion of the females remain on or about the breeding ground. It must therefore be certain that, when the sexes meet again on the coast during the winter, there is but a 20 slight possibility of a female goldeneye again pairing with a drake with whom she has previously bred. Thus, within the breeding range, there will be almost complete panmixia. As Mayr (1942, p.241) says after describing how a Maine-bred mallard drake may follow a North Dakota female back to that state from the Gulf coastJ "The result is almost complete panmixia and much-reduced subspeciation". Matrilocality (with Coastal Pairing) as an Isolating Mechanism between species It appears that most female goldeneye return to the neighbourhood of their birthplace during their second summer in order to breed. Many non-breeding yearlings also return to the region of their natal lake. Females thereafter will breed, i f physically possible, in the same stump that they chose for a nest, when they first bred, at two years of age. There is thus a strong tendency in the females to matrilocality. The outbreeding mechanism inherent in the summer and autumn seclusion of the adult males means that when a pair start inland in March or April to the breeding grounds i t must literally be true that the male has no "idea" whither he is being led by the female. She i s homing, by contrast, to a known lake and even a particular nesting cavity. This means that should a mixed pair occasionally be formed on the coast, the drake of the pair will be led by the female (if he goes with her) right outside the breeding range of his species: i f he is B.clangula, short of the Rocky. Mountain chain — i f he is B.islandica, over the Rockies into Alberta, the Peace River Country, the Yukon, or into Alaska. Plumage characters and ethological mechanisms result in accurate 21 species recognition, and so mixed pairings are very unusual. Since the drakes, i f they-pair on the coast, follow the female to the breeding grounds, matrilocality allows neither the females nor the males to pass outside the -. traditional range to breed. It appears therefore that, except for narrow bands of range overlap, competition effectively prevents the two species, B.islandica and B.clangula, from spreading into each other's breeding range. The present dividing line between the two ranges lies just east of the Rocky Mountains, on the edge of the plains. Why this should be so is not certain, but i t seems probable that B.islandica was contained in mountain or inter-montane regions to the west of the Continental Divide (and to mountain regions in the eastern United States) during glacial times and spread north,west of the cordillera,into British Columbia as the ice retreated. Only recently have the glaciers of the high chains of the Rocky Mountains become free of ice and accessible to free passage of range-spreading individual goldeneye. It is likely therefore that B.islandica is spreading east of the mountains at the present time, but only slowly since B.islandica has no reason to fly very far east of the Rocky Mount-ains at any time of year. Competition must be preventing the colonisation westwards of B.clangula for B.clangula traverses the intermontane region twice each year and British Columbia i s good "goldeneye breeding range". Hybrids Around Fairbanks, Alaska, goldeneye of both species are found together during the summer (J.W.Brooks, pers.comm.). It would presumably be pos-sible therefore to find situations, in central Alaska, where both species 22 are present on the same lake and where courting parties and territories of the two species-are in close proximity, unless as in the Peace River district the two species tend to breed on different lakes (Cowan, 1939) . If the two species do sometimes breed on the same lakes, sexual assualts may occasionally occur and mixed matings arise. Hybrids have not been identified commonly so far. The only published reference to a hybrid B. islandica x B.clangula is Snyder (1953)• This is an adult drake taken in New Brunswick and now in the Royal Ontario Museum. Unfortunately we do not know i f the small numbers of B.islandica found on the east coast of Canada and New England in winter come from Labrador, Greenland or Iceland. For the west, Miss Mary F. Jackson has taken a mature hybrid drake on Westwick Lake, Cariboo District, British Columbia on May 13 , 1954, now in the collection of the Department of Zoology at the University of British Columbia. It would be interesting to know whether hybrids are more common in central Alaska, or the Peace River Country, than elsewhere, and i f they are fertile. Maybe the ethological mechanisms are-too strong to permit mixed matings even i f the two species do breed on the same lakes, but whether such competition for space occurs we do not really know yet. Certainly males are too intolerant of each other for small sloughs to be occupied by more than one pair of B.islandica (in British Columbia at least — Icelandic conditions around Myvatn may be different), and ter-ritories may be larger than they are on larger lakes where tolerance appears to be rather greater. Unfortunately the finding of hybrids will not be easy, even in adult 23 males; nevertheless i f they can be found among the breeding drakes in a higher frequency in one overall region, as compared with another, then (assuming a f i f t y : f i f t y sex ratio in hybrid downies) there must be a proportion :of female hybrids in the region of their origin, and in parti-cular wherever females of both species are homing to the same locality, as from Alaskan coastal waters to the breeding grounds inland. It is apparent that female goldeneye do recognise the drakes of their own species on the wintering grounds, and generally pair only with them. Plumage and display differences appear to be successful in preventing mixed pairings from occurring during the winter period, when both species are found together on the salt water. To a trained observer, the males are readily distinguished in nuptial plumage, at a distance, by the amount of white on the wings and flanks. Closer to, the shape of the head, shape of the white patch between the eye and the beak, the colour of the head feathers, and the wing "windows" are distinctively different. No normal female will ever mistake a male. The females may be distinguished by the shape of the head and b i l l , and from October through until lake May, by the amount of yellow on the b i l l . In June when mating for the year is completed, the yellow b i l l darkens over in the female. This is an in-dication of its character as a distinguishing mark. Juvenal males may not be distinguished from females until late winter or their first spring. Yearling males are distinguishable either by the presence of white feathers coming into the brown feathers of the head, where the face patch will ap-pear, or i f signs of this are s t i l l lacking, by theridarker sepia tone of the head colouring, by head-shape and size. Juvenal females of B.islandica 24 cannot be distinguished i n the f i e l d from those of B.clangula. In addition to a l l these plaumage characters, well known to an ob-server familiar with both species, and presumably also easily recognised by the birds themselves, there are the courtship displays. These too, are very different i n form between the two species (though not i n general construction or origin). They may be easily distinguished by a human observer, who knows the body-shape they produce, and the speed of per-formance i n each species. Winter Distribution and Winter Behaviour. Around the Stgdts of Georgia, southwest British Columbia, there i s a d i f f e r e n t i a l distribution of the wintering goldeneye. Around Vancouver (Howe Sound, Point Grey, Stanley Park and Indian Arm, but possibly not the more open coast at Point Roberts), B.islandica is more frequently seen, while along the east coast of Vancouver Island, at least from Comox to Victoria, B.clangula i s more generally found. There is a sim-i l a r d i f f e r e n t i a l distribution i n other Mergini, especially scoters. Display may be seen from al most the moment the-sexes meet again i n •a the f a l l . Relative frequencies of the various displays during the winter (when t e r r i t o r i a l defence i s not needed except around the mate, and mating i s rare) are rather different from those found on the breeding lakes (where mating-territory exists and must be defended, and where mating i s necessary). Nevertheless mating may sometimes be seen on the coast. Charles Walcott and Benjamin Dane have recorded one such coition sequence on film (B.clangula) at Newburyport, Mass., i n April 1956. 25 It is worth mentioning here that sea-ice i s not often found south of the Aleutian chain, whereas on the east coast of Canada and New England i t i s a regular feature of the early months of the year. But this does not reduce display activity. Walcott and Dane's film shows B.clangula courting even among ice-floes i n February, at Newburyport. During the winter goldeneye are seldom found alone, but occur more often i n loose association with other individuals. Quite frequently a number of feeding birds, i n a loose flock, win come together every few minutes and sexual displays, mixed with and perhaps initi a t e d by, brief bursts of threat and counterthreat, w i l l be seen. As quickly as i t began, this activity relapses back into feeding or preening. Preening i s very persistent, and often appears to occur as a community activity. It seems certain that i t has a function i n maintaining the flock as a whole, or i n building up sexual motivation. On the Straits of Georgia, which are a l i t t l e south of the centre of bird's winter range along the Pacific coast, there is a preponderance of males. This may possibly be due to the greater hunting pressure, on females, which occurs i n the interior i n the f a l l (since the adult males are not there during the hunting season). But this preponderance of males is even more marked in B.clangula than i n B.islandica. Both i n B.clangula and i n B.albeola a di f f e r e n t i a l influx of females from the more southern parts of the winter range certainly occurs i n the spring. Courting parties around Vancouver generally consist of at least three males to one female i n the f a l l . Birds, which have the appearance of being paired, may be found at any time during the winter. Towards spring the 26 ratio of females may increase, and isolated pairs, male and female feeding together, become more frequent. Whether this indicates an influx of female during February and March i s unknown. It could do so, but Mary F. Jackson (pers. comm.) finds that coastal banding recoveries of B.islandica show no differential distribution of the sexes. In other diving ducks however i t i s the rule, as with B,clangula. and B.albeola that the drakes predominate i n the more northerly parts of the winter range (Phillips, 1925-26). Whether i n B.islandica there i s an overall preponderance of males or not, there certainly appear to be a sufficency on the breeding grounds and the evidence i s that there i s a higher adult mortality i n females, due to unknown biological causes. The uneven sex ratio in courting parties during the winter i n great part explains their great activity. Clearly a group of birds,in which the males predominate, and are liable to threaten each other, or are a l l performing the "Pseudo-Kick" or "Rotary Pumping" displays again and again around just one or two females w i l l be a more volatile group than would one i n which there i s only one or two males, with a larger number of female Such a situation i s indeed sometimes seen i n the Interior i n July; a lone drake with a drove of inactive post-breeding or immature females. Sexual activity i s then low, but somehow the male always looks as though he did not "belong" (possibly this i s a general phenomenon i n animals i n which the male is- the more active partner in courtship). It would be inter esting to have descriptions of the activity of courting parties i n the more southern limits of the winter range of diving ducks. The implication would seem to be that i n diving ducks generally those females who move furthest 27 south may only pair late i n the winter. The behaviour of a pair i s different from that of the unpaired birds of a courting party. Premium now i s on maintaining the pair bond, rather than on forming i t . Paired birds are noticeably less active, unless, as often happened, the males become mixed up i n a courting party. Possibly the hormonal state of the paired birds differs from that of single birds. If this i s so, the -winter population may not be an homogenious whole, even among mature individuals. In any case i t appears that an unbalanced sex ratio prevents even the farthest-travelling female remaining unpaired by the time she has been on the breeding grounds for a few days. There i s no evidence that juvenal or yearling females go farther south i n win-ter than the adult females, though i t is known that the juvenals wander over considerable distances i n the f a l l . Spring Migration The goldeneye move into the interior of British Columbia during March and Ap r i l , many B.islandica appearing on Okanagan Lake i n March, while a very few winter there (Munro, 1918) . Rather more B.clangula winter there than do B.islandica. We do not know to what extent the migration takes place as pairs, pairs with unmated birds attached, or as groups of unmated males following an unmated female. It is generally assumed that pairing occurs on the coast almost universally. But i t may also presumably happen that an un-paired drake follows another pair. Alternatively we may suppose that matrilocal tendencies might also operate, so that i n males also (when 28 unpaired), wanderings into or settling down i n the range of the other species would be unlikely. Such birds might tend to return to the place of their birth and pair up with an unpaired female there (this i s the converse of the more usual'situation i n which a male ends up almost any-where, but at his birthplace). We do not know the truth about this, and w i l l not do so un t i l as intensive a study has been carried out on the drake of the species as has been done, by Mary F. Jackson, on the female. As the ice melts (often before) the goldeneye move to the lakes and sloughs that they prefer for breeding. The ice may melt any time from mid-March to mid-May depending on latitude, the hardness of the winter, altitude, and the weather conditions at the time of break-up. In 1955 i n the Cariboo District the ice melted i n the last day or two of April. In 1956 a fine weather period melted the ice during the last ten days of April. In 1957 the break-up has probably taken place earlier. Territory (Mating and Brood) and Nesting. Territories are established by the drakes as soon as the ice melts, and are defended even before the ice has departed completely, ^ t i s not long before the old females start laying the f i r s t eggs i n the old nest-sites. Younger birds who have to find and establish a right to a nest-cavity may be slower to begin their clutch. The number of eggs varies from 6 up to 14 and a female may take 10 days or as long as three weeks, from the time she lays her f i r s t egg, before she starts to incubate. Smaller clutches are not l a i d by the younger birds alone; indeed f i e l d data show that each bird lays a more or less fixed clutch, large or small, irrespective of age. Even a young bird can presumably lay a large clutch, 29 and the factor determining clutch-size appears to be genetical and pheno-logical rather than an expression of the age of the bird (Mary F. Jackson, M.S.). Taking a l l years together early nesters may be incubating by May 1st in early years, but late individuals may not start to incubate until the second week in June. By the second week in June, also, the drakes have nearly a l l disappeared. Their departure may be gradual or sudden. In the later season, 1955, the Cariboo suffered from arctic conditions from May 26th to June 1st, but the first week of June, by contrast, was very hot. The males did not leave suddenly during the cold spell as might be expected, but gradually during the middle of June. In the earlier season, 1956, a rainy spell in the f i r s t week of June appears to have caused a rapid exodus of males then, but this is less certain — there were fewer of them on the lakes that season. After the males leave the females do not necessarily frequent the old mating territory, and do not defend anything but the nest until the young reach the water (Mary F. Jackson pers.comm.). The incubation period lasts 32-35 days, so that one-twelfth of the whole year is spent sitting in a tree cavity incubating eggs, with short breaks to feed, drink and preen (at that time about equally important). In any one season hatching takes place over a four week period. The eggs (except the last one or two in some cases) a l l hatch within a few hours, and the young may stay in the nest drying off, and preparing to leave the nest, for a further 24-36 hours. 30 The young are called off the nest by the female, and a l l tumble out of the nest together in the space of a minute or two (in the present study young were seen to leave the nest hole only under experimental conditions -a nest box removed from its tree, and placed on the ground in an enclosure at the edge of the lake). The-young may have to drop as many as 35 to 40 feet, but this does not seem to harm them. They may f a l l directly into the water i f the tree is one killed by flooding, or one standing on the edge of a lake, with the mating territory immediately in front of i t . Mary F. Jackson (pers.comm.) found that the brood territory is often some distance from the nest tree and i s not necessarily coincident with the mating territory which existed a month or three weeks earlier. On the other hand the downies may have to follow the female through the bush or across the range for even as much as three miles (Mary F. Jackson pers. comm.). A brood was watched by the writer, just after leaving the nest tree at the edge of the bush, crossing the range towards the lake. They w«r-e watched during their first hour on the water, and their activity described. Deficiencies i n Brood Guardianship The seeming efficiency of the reproductive process in goldeneye is reduced by the post-fledging behaviour of the females. Most females suc-ceed in hatching their clutches (predation at this stage is slight), but some of them are poor brood "coaches" or "nursery school attendants". Some forsake their broods within a few days. Downies that survive this lack of protection may remain unattended or join other broods, so that some females may be attended by as many as twenty or thirty downies, of 31 whom but a fraction, i f any, are her own offspring. Fighting occur between females for possession of downies. Unattended downies are not usually attacked, but are easily adopted, by other females. Female ducks do not feed their young, which fend for themselves, but they do brood them at night in the early days at least, and this provides protection, but more particularly protects them from unsupportable heat loss on c h i l l nights. The desertion of the young by some females, presumably therefore has a detrimental effect on productivity. - • Age at first breeding Another important fact relating to productivity in goldeneyes, is that the females do not breed until their second year. This habit is found throughout the Mergini, and i s an ancient character. It may be associated with low nest predation both in the arctic ground nesters and in the tree-nesting species found below the tree line. Post-breeding Dispersal. The drakes disappear in June. From June through August, small bands of yearlings and post-breeding females may be seen on most breeding lakes. As far a s is known they do not join the males, but stay in the same general district that they spent the whole summer. The juvenals are s t i l l present during the hunting season (opens September 15th, in interior British Columbia). Mary Jackson's analysis of the banding returns for B.islandica shows that: — l ) . Some juvenal goldeneye are shot on the lakes on which they were banded as flightless downies. 32 2) . that there is a marked tendency among flying juvenals to wander i n any direction—in;.some cases for many hundreds of miles i n the Intermontane area, and even east into Alberta, during the period prior to the true coastal migration. The significance of this phenomenon has been discussed by Hochbaum (1955) , who says that i t is found i n most ducks and allows the juvenals to acquire a geographical knowledge enabling them to home to the breeding area at a later date. 3) . that adult females are also s t i l l present on the breeding lakes, but that they separate from the young before becoming flightless, and usually by the time the young are 6-7 weeks old ( f i e l d observations). Summary Since the ice-age a population of B.islandica has again come i n con-tact with a population of B.clangula i n western North America. On the east coast certain numbers of the former winter with the la t t e r . Pairing i s generally believed to occur on the wintering grounds. Since the males are separated from the females from June to October panmixia i s believed to be almost complete. Since the females home to their natal lakes, a male of a pair breeds i n a different l o c a l i t y from that i n which he was raised. Unmated males may either travel with another pair, or i n a small party, or may even home to their natal lakes. What proportion of the males perform the spring migration unmated i s not known, nor i s i t known which of the three po s s i b i l i t i e s they most usually follow. 33 Species discrimination is evidently good, and displays are recognis-ably different in form (though not in origin). Together these two factors prevent mis^enation and hybridisation and maintain:: the pair bond as far as fertilisation. But displays do not particularly promote reproductive success itself, once the eggs are laid. This distinction deserves to be emphasised. In theory the matrilocality of the female would result in the drake of a mixed pair passing outside the breeding range of his species. Mat-rilocality of the females, combined with pairing on the wintering grounds, and ecological competition of an exclusive "type would prevent either species of goldeneye from enlarging its breeding range at any considerable rate, or from spreading into the area occupied by the other species. These behavioural mechanisms, and the specificity of the plumage characters and displays has prevented hybridisation between B.islandica and B.clangula though they are evidently capable of i t (as are; manykducks; the infertility of hybrids has hardly been studied), and has evidently contained the two species within the breeding ranges occupied when they recently came to-gether again after the tongue of ice along the Rocky Mountains melted. The line of demarcation between the breeding ranges of the two species is parallel to the Rocky Mountains, except in a few areas, although B.clangula regularly flies across the mountains twice each year. As far as i s known the nesting and other requirements of the two species are very similar. 34 THE GOLDENEYES NON-SOCIAL BEHAVIOUR 1 . Locomotory Movements 2. Comfort Movements 35 LOCOMOTORY MOVEMENTS These are a group of muscular activities without which l i f e would be impossible. They are important since they provide the means whereby orientations and hostilities are expressed, and feeding and. other vital biological activities performed. Thus the following of the female i closely by a male during the migration, rising from the water-and its intention movements, the flights preliminary to going to the nest,landing in or leaving the nest tree and landing on the water are points of inter-est i n the Flight of Ducks. The form of diving, especially in downies is also of interest, as is the mode of feeding i n shallow water and in deeper water. Swimming and Walking are of only minor interest. Locomotion movements also have the potential to produce courtship displays, or have produced these from them, by ritualisation of intention movements. Thus the Head-Up position of B.islandica and its counterpart in B.clangula may well be derived from the "alarm" or "anxiety" pose of goldeneye, which in its turn may be derived from the intention movement of flight. Since activities differ considerably between adults and downies they have been described separately. Flight I have not described the intention movement. Millais (1913) however describes i t . He writes: "If observant of danger and about to rise, the neck is straightened and the head held high, with the crest somewhat raised." Only once has an upward li f t i n g of the wings been seen in B.islandica, though a similar flicking of the wings is a ritualised part 36 of the Head-Bobbing display of Bucephala albeola at high intensities, and i s seen in Anatini on occasions. A party of post-breeding female goldeneye were seen at Boitano Lake in June, 1956 and they were sitting on a raft. On seeing me four of them stood up, and raised their heads so better to watch-me. One flew to the water from the raft, the rest remaining standing. One of these suddenly lifted the wings off the back with the carpal joint forwards and raised, the primaries rising no higher than the joint, but the leading primary forming a horizontal line with i t . However this individual did not f l y immediately. In B.albeola, furthermore, landing on the water has become regular display activity. In Anatini actual flying itself has developed display features (e.g. Anas americana) though in other species these so-called "courtship-flights" do not have sexual, so much as territorial, significance, for the female in them is usually paired (Hochbaum, 1944)• Goldeneyes, like other Mergini go a considerable distance over the water before becoming fully airborne. Brewster (1900) describes i n detail the manner in which females (B.clangula) search for a nest, and attempt to alight at a potential s i t e : — Four female Whistlers flying together in a peculiar manner over the trees, now rising high in the air, next descending and dashing among the trunks and branches, vibrating their wings rapidly and continuously as in ordinary flight, but describing circles about a remarkably t a l l stub with a shattered top. Around this they would pass a dozen times or more, gradually drawing nearer until one bird leaving the rest and pitching first down-ward, then sharply upward would fly directly towards the stub and try to alight on i t s jagged top. The attempt usually failed, when the bird, continuing its flight, would disappear among the trees, presently returning to begin circling again; but twice i t gained a foothold and remained perched for several seconds, 37 although i t had to keep i t s wings i n constant motion to maintain i t s balance. Sometimes i t s fl i g h t was directed to a point near the top of the tree where there was a round, neat-looking hole, no doubt the entrance to a nest, for we afterward saw two whistlers emerge from i t i n quick succession. We thought that a l l four birds tried i n turn either to alight on the stub or to enter the hole...no two made the attempt at the same time (how-ever). They were silent for the most part, but occasionally one of them would utter a sound not unlike the quack of a Black Duck but shorter and f l a t t e r and repeated very rapidly six or eight times. I have seen similar flights i n the Cariboo District of B.C. by B.islandica i n which four, or even more, females may be seen flying around together very fast, along the edge of the bush or over the forest, coming back to the v i c i n i t y of the lake over and over again. Brewster's observations sound a l i t t l e l i k e prospecting, but the date was May 30th, which, i s rather late. Mary F. Jackson (pers.comm.) believes that such communal flights by females take place even when each one i s herself incubating a clutch i n a separate tree, and that similar flights may be performed by ganged-up yearlings and adults. Generally the return to the nest i s not a communal a f f a i r . The female takes off from the territory and the male follows after her (Hochbaum, 1955, states that the male follows closely behind the female during migra-tion). The pair circle around the lake and then, i n ever increasing circles, expand beyond the lake as their altitude increases. If the nest i s some considerable distance from the edge of the lake, the circling f l i g h t may continue for some minutes before the male returns. On one occasion I watched the pair circle along the edge of the forest, a number of times, before the female suddenly cut i n opposite the spot where I knew the nest to be, and the male returned to the water. On another oc-38 casion I watched a pair circle round a lake, a number of times, before 1 saw the female cut in sharply towards the nest tree at the water's edge. Brewster (1900) saw a female B.clangula return to the nest on June 2. She first alighted on the water near the tree and for fifteen or twenty minutes swam or drifted idly about preening her feathers. Then she flew out over a space of open water and turned back toward the tree, describing a great loop and rising gradually until she had attained an elevation of about twenty feet when she made directly for the entrance to the nest, which was about thirty feet above the water. On nearing i t she pitched up sharply for the remaining ten feet, keeping her wings in rapid motion up to the last moment but checking her speed very considerably before she reached the hole. and again (Brewster, 1900):— without any preliminary circling dropped into the water within a few yards of the nest tree. After floating motion-less for about two minutes with head and neck erect, evidently watching and listening intently, she flew directly to the hole and alighting on its edge, perched there for an instant, flapping her wings a l i t t l e to maintain her balance. She then popped in, throwing up her spread t a i l just as her body dis-appeared. Other observers (e.g. Harper in Phillips, 1925) have noted how a female plunges, with no pause whatever, directly into the hole and dis-appears. It does seem likely that the habit of flighting around at very high speed, often at a considerable height, before plummeting into the nest hole has protective value. When undisturbed the female may spend some time, in the entrance of the nest, looking around before leaving but, i f disturbed, she emerges from the hole at high speed and is away through the trees, often before she can be recognised. Brewster wrote of this: "The blow (from a paddle on the tree) was Immediately followed by a scratching sound, and the next instant the whistler shot out over our heads." 39 Bernhardt (1940) describes how the female f l i e s around a great deal i n search f o r a nesting site and makes a "slide" over a tree having a suitable hole. He seldom saw a female settle i n the tree and look at a hole. Rather she f l i e s directly into i t , unless disturbed,as she f l i e s into the tree, i n which case she may perch i n the tree. In searching for holes the drake follows the duck, "twice I saw one si t t i n g waiting on the tree, into which the female had flown". During laying and incubation, however, the male waits on the water. Millais (1913) describes B.clangula crash-diving as scoters and eiders may do, when frightened. A Peregrine Falcon (Falco peregrinus) was f o l -lowing a group i n f l i g h t : "I should say 80 yards in the air , and closed their wings as they heard or saw the Peregrine coming, and dropped as i f shot to the surface of the water. On striking the water there was no pause, they just passed out of sight". It i s possible that elaborate "landing" displays exist i n goldeneyes as they do i n B.albeola. Diving Adults Allan Brooks (1920) wrote an important comparative paper on diving habits i n diving ducks with illustrations of B.clangula, Melanitta fusca and M .perspicillata. He says that i n B.clangula the "Wings are never used under water but are held tight to the sides beneath the flank feathers. The t a i l i s expanded to i t s f u l l extent and appears under water to be about as broad as the body." Pearse (M.S.,1923) and Brewster (1911) agree with Brooks that the wings are not used under water (contrast scoters). Brewster 40 (1911) agrees with Brooks that the t a i l i s f u l l y expanded: "to the ut-most possible width just as they disappeared". Pearse (M.S., 1928) remarks the power of swimming under water against a strong current, and he saw single dives for even more than 30 feet. Millais (1913) remarks "the powerful strokes of the legs of these ducks, which seem to beat with great rapidity under water and much power. The strcnke i s more or less para l l e l to the wings, the head i s held out straight i n front". Brooks says that under-water actions are exactly similar i n B.islandica. Brewster (1911) says that the downward plunge i s begun by the bird "simply immersing i t s head and then vanishing with surprising i f not mysterious quickness". He notes that the dive may sometimes appear more as a jump up and a plunge forwards i n the manner of a grebe or merganser, so that the lower contour of the body becomes visible above the water; "also the whole of the legs and feet, just before re-entering the water". Phillips (1925) writes that feeding birds disappear with a push of the feet and "go down at a steep angle or i n circles, with the t a i l spread and the wings closed or nearly closed. Progress i s by the feet alone, these being kicked out at a wide angle from the body; almost i t would seem at a right angle." Feeding birds are readily seen to bob to the surface and, often get l i f t e d a good way out of the water by virtue of their uncontrolled buoyancy. Phillips (1925) writes that "In rising to the surface the body and feet seem to be held r i g i d and the bird bobs up with great speed." Brooks (1920) writes that "Upon emerging, the t a i l i s held f l a t on the water or slightly under the surface; at rest, the t a i l i s elevated or even cocked 41 up at an angle when the bird i s asleep." Alford (1920) writes "In rising to the surface they seem to depend entirely on their own buoyancy and when ascending remain absolutely motionless." It i s apparent, of course, that the adult has some, control (which very young downies may lack) of controlling the way they surface, i f they so desire, for an aggressive bird surfaces very smoothly, and i n a crouched attitude breaks surface at an angle. Often when watching for an aggressive bird to rise one f a i l s to observe his reappearance. Downies Diving i n downy ducklings takes two forms True Diving and Flap Paddle  Diving. Downy goldeneye s i t very l i g h t l y on the water (Figure l a ) . Thus diving requires a sudden sharp effort. Flap Paddle Diving i s the common-est type of underwater behaviour on the f i r s t day that downies are on the lake. The head i s lowered below the surface, and the legs splash out to the side and rear of the body (rather as Figure l c ) . Great jets of water are thrown out at the back, and the young emerge from the water very soon. In this half-submerged position the body i s held at about 45 degrees downwards. The legs are placed l a t e r a l to the body (horizontally) and do not project downwards. Using this technique the downy can become completely submerged, but they do not stay under at a l l long, but bob up to the surface very suddenly, like balloons. The a c t i v i t y is sometimes followed by Upwards-Stretch and Wing-Flap. This activity enables the young bird to gather food just below the 42 surface. The pose greatly resembles that assumed by adults either when feeding i n water too shallow for diving, or prior to diving i n deeper water, the Head-below-Water, or Looking-for-Food posture. This i s some-times observed i n adults even i n deep water, and acts as an appetitive orientation of feeding. The pose also has an intermediate pose such as one might expect of an intention movement of diving, for the head i s not l i f t e d again before a dive takes place. True Diving i s similar to the diving of adults with a forward plunge and smooth breaking of the water (sometimes l i f t i n g out of the water f i r s t ) . It replaced Flap Paddle Diving as early as the f i r s t day, though the latter may continue for some time. First-day downies could dive during their f i r s t hour i n the water. Dives lasted 1-5 seconds. They often preceded this type of diving by holding the face half-under water—as described above for adults (Looking-fopFood pose), though diving did not always follow this pose (rear view seen i n Figure l b ) . On one occasion a downy was seen to run over the water almost treading i t before diving. Swimming Adults Phillips (1925) writes the B.clangula swims with head stretched forward as i f looking down into the water. When feeding the t a i l i s sub-merged or floating on the water (Brooks, 1920; Phillips, 1925). This attitude of the t a i l i s also apparently found when the bird i s alarmed, but Brooks says that "at rest, the t a i l i s elevated or even cocked up at an angle when the bird i s asleep". Other aspects of swimming, i n moving 43 away from danger, in the threat posture^ or a fight or in coition will become apparent when these are discussed. Downies In swimming the feet appear to be paddling much faster than in adults, though at times the downy merely sits idle, or preening. However being very active l i t t l e birds they are seldom s t i l l , but are darting back and forth after insects or other food on the surface, when not diving or preening. Because of their smaller size they can change their orientation much more rapidly than can adults. Their overall rate of swimming i s how-ever low. When a brood is being chased by boat, or the birds are surprised on the shore, the young may skitter, as a group, over the water. In this way they can make good speed and can keep up with their mother. They maintain almost an upright position during skittering and only a small part of them is below the water. I am not sure i f they use their wings. Downies are very curious about objects just above their heads. An object part way up a reed stem would presumably attract their attention, for weed on the netting of an experimental enclosure often attracted atten-tion. To reach i t they would either rise up almost vertically towards i t , or would even (paddling with their feet) jump up in the air in an attempt to snatch i t . As Phillips (1925) writes "they can leap from the ground or water to a height of eight to twelve inches by the power of their legs alone." They are very persistent on such occasions, and failing to get i t will subside onto the water and Tail Wag. 44 Walking Adults Adult diving ducks seldom have occasion to walk on dry land. A female may lead her brood overland in which case she has a very waddling gait with the head drawn quite far back and down. In winter, pairs (especially drakes) may haul out on log booms, or on rocks, but groups of birds do this less often. As the ice melts off a lake, drakes haul out on the ice for short periods between periods of defence of the expanding area of water that they have staked as territory. Both leaving the water and re-entering i t are slow, leisured movements appearing to require very l i t -tle effort. These actions are almost invariably followed by the Tail Wag. Downies The first act of the downy is to climb up the side of the nest cavity (an action for which i t is well fitted as i t has sharp claws). They tumble from the next, as described later, and very often have to walk a considerable distance over dry land to a lake or slough. They are however very strong on their feet when they leave the nest and walk in a rather erect manner, following the female (who is calling). 45 FIGURE 1 B.islandica (B.C. population) THREE STUDIES OF DOWNIES From a film by M.T.Myres A. Rear View. Floating. Shows the puffed out shape of the white cheeks, and the two round white "guiding marks" on either side of the rump. Further forward there are two smaller white spots on the wings. These marks are con-spicuous and also lined up on each other as seen from behind. They are probably visible even on dark nights. B. Rear View about to Dive. .Head down (below the water, in the Looking for Food pose) at the beginning of a dive. Back and rump humped up, and "guiding marks" visible. C. Side-Scratching. The head is being pushed back and forth along the side of the body. The right leg is braced and above the surface of the water, in a similar position to that in the Flap Paddle, which this view of the legs closely resembles. 46 47 COMPORT MOVEMENTS It i s not intended to give more than a brief outline of those body movements which are concerned with the maintenance of a satisfactory rest-ing state i n ducks. Feeding and Drinking are included here, on the grounds that t h i r s t and hunger are to be considered unsatisfactory and uncomfortable states of body condition. True Comfort Movements may be divided into (l) Muscle-relieving (or Stretching and Shaking Movements), and (2) Scratching and Preening Movements. A number of aspects of comfort movements are worth consideration at this stage. The f i r s t i s that while certain of them have developed a second, ritualised form, others appear to occur as displacement ac t i v i t i e s , and yet others are performed most frequently and most actively on occasions when groups of birds (migrating or winter flocks, or courting parties) are as a whole i n a highly excited state. Ritualised comfort movements in golden-eye are the Water Flip (Drinking), Head-Flick, Upwards-Stretch + Wing-Flap •f T a i l Wag at the end of a sequence, and the Wing- and Leg-Stretch. The Water-Flick may be a displacement activity. Splash-Bathing i s an ac t i v i t y which i s expressive of high social stimulation i n courting parties, ^t i s an activity which regularly follows coition, i n the female immediately, i n the male at the end of Post-coition Steaming. It i s also performed (with other types of preening) after a long or a fierce aggressive encounter on territory. The second aspect of comfort movements i s the localisation of preening. The whole body i s , of course, a surface continuum, and since almost a l l 48 preening i s performed by the b i l l there i s but one centre for the ac t i v i t y to originate, or i n other words the various regions of the body bear a certain spatial relationship to the head region, and to the accessibility of theia to the b i l l . We may think of the surface of the body as consisting of a series of overlapping fields -of stronger preening orientation and of the bird of having a general idea of where i t desires to preen. The physical difference between contact of the b i l l with one part of the body, and with another l i e s i n the differences i n tonus i n the various muscles that are involved i n bending and twisting the head and neck so that the b i l l comes i n contact with one region rather than the other. Thus preening of the lower breast forwards di f f e r s from preening of the upper breast and neck only i n the curvature of the breast (and amount of leg paddling), the degree of arching of the neck and articulation of the head forwards, and in the amplitude of the movement of the head up or down, or from side to side, on the spinal column. Thus many of the various types of preening to which names have been given are mere abstractions which have been used to indicate which part of the body is receiving attention, and the descriptions are really only of the contortions the bird goes through to reach them. Again ear-scratching and scratching of the neck are distinguished merely by the relative bowing of the head to the side and the amount of l i f t of the foot. When a comfort movement acquires epigamic signal value, however, the movement becomes stereotyped both i n form and i n timing. There i s a movement in the display of the scoters which greatly resembles the scooping motion that occurs when the upper part of the breast i s being preened. But 49 nibbling as a preening movement requires very rapid and continuous movement. There would seem to be no way i n which i t could become recognisably d i f f e r -ent (as a display) without i t s losing i t s functional effectiveness. On the other hand, the -Water-Flip of goldeneye i s very similar to the Drinking movement, so alike indeed that I have confused them i n the f i e l d . But the movement i s a r i g i d one, with only slight variations i n intensity, and the distinguishing marks are to be found i n the more rapid l i f t of the head, and i n the tendency to frequent repetition, which there i s no reason to suppose i s necessary i n true drinking. I noted i n the f i e l d one time that "The Elaborate Water Flip and the Stretching of the Wing and Leg i n courtship have fixed lines of orientation and muscle tonus. They may occur i n weak form, but the ratio of muscle tonus i n one muscle to that i n another should be the same provided the articulations are not themselves altered by the weakness of the movement...movements not yet f u l l y r i t u a l -ised would show a lack of 'place-to-preen' -localisation, or directed orientation of the movement". The Wing-and Leg-Stretch i n the Pre-Coition Sequence appears to be a slower and more deliberate movement than i n normal r e l i e f of wing tension. Feeding Diving ducks feed on the bottom of the lake, river or sea on which they are swimming, and thus d i f f e r basically from the feeding habits em-ployed by the surface-feeding Anatini. However when goldeneye are swimming in very shallow water as they sometimes do along the tide-line, or around the margins of a slough they may feed merely by dipping their heads below 50 the water and stretching them down to the bottom (Looking for Food posture). But they do not up-end, and paddle with their feet violently, as the pudd-lers always do. The method of diving has already been discussed earlier, but Millais (1913) describes feeding on the bottom i n B.clangula as follows:— On reaching the bottom, i t at once commenced to turn the stones over with the b i l l , and from under these, various water insects were found or caught as they attempted to escape. Some-times i t would find a small batch of young freshwater mussels, and these i t would devour very quickly one after the other... a l l food was swallowed where i t was found, and small pebbles and f a i r l y large stones were pushed over i n the search. Several times I saw the bird just move a f l a t stone. It would go a l l round i t and try i t from every point...In a lake the Golden-Eye w i l l dive i n perpendicular position, but i n flowing water i t dives i n a slant against the stream or tideway...In s t i l l water the Golden-Eye often dives i n circles to get to the bottom... bounce up to the surface l i k e a cork immediately they cease to push-downwards with their feet. Phillips (1925) believed that "nearly a l l food i s swallowed before rising -but no doubt an occasional tough bit i s brought to the surface (as i t often i s with Scaup), readjusted i n the b i l l , and rapidly bolted. A.Chapman (1889) has watched them 'chewing' small fish before swallowing." This refers to B.clangula but Phillips said that i n B.islandica "crayfish are brought to the surface to be swallowed, but this i s not so with the small shellfish" and Munro (1918) wrote that abrasion of the fore-head against stones while foraging for crayfish wore the feathers badly during the winter. Phillips (1925) described the Looking for Food posture as follows: "swim along with i t s head immersed looking beneath the surface preparatory to diving, just as we sometimes see grebes, loons and mergansers do." Downies of B.islandica nearly always bring the food to the surface and generally have a considerable tussle with i t before swallowing i t . They frequently give the food a violent shaking sideways, resembling the Head-51 Flick. The attempt to swallow often involved a continual jabbing at the food, resembling the Jabbing seen in B.clangula (at least) just prior to coition. Drinking This i s an activity whose importance for a water bird i t is not easy to guage. Figure 2 illustrates Drinking in B.islandica. On this occasion a drake had been sitting out on a log. ^ e disturbed from this position and swam slowly away in shallow water. As he swam he made two Drinking movements (as illustrated) and then began feeding by lowering the head below the water in the Looking for Food position. The movement consists of a dip to the water with the b i l l , then raising the head until the beak points vertically upwards, and then lowering the head to the resting position again. Generally only one or two Drinking movements will take place at any one time, i.e. drinking for a water bird can be spaced out very evenly. Exceptions to this will be during the migration, when a flock comes in to a lake to recuperate, and when the female comes off the nest, during the incubation period. One morning, when the temperature in the shade was nearing 80 degrees F., a female was observed to leave the nest. On her alighting on the water I began recording her activity on tape. The tape began "Water-Flip; Water-Flip (really drinking I think); Drinking" and I recorded six more Drinking movements in the next minute or two. Clearly the temperature inside the nesting box had been very high and the bird had become thirsty. In 1955, an experimental group of downies of B.islandica 52 were put on the lake, for the f i r s t time i n their lives, after being held captive over-night: "When f i r s t put into the water a l l f i r s t indulged i n Drinking. This i s a dab to the water followed by a rapid jerk back and up of the head into a position immediately over the base of the neck, but the beak just above the horizontal. In the movement the body i s f i r s t tipped forward slightly, as the neck is dipped, and then raised so that i t bobs as i n the adult. At the f u l l e r extension of the head the head may be dropped a fraction." It w i l l be noticed how great is the resemblance of Drinking to the Water-F-lip (illustrated for B.clangula i n Figure 31) , which i s an important element of the pre-coition behaviour. It should be noticed how much slower true Drinking i s than the courtship display (44 frames instead of 29) and that i n Drinking the neck i s not elevated so far. It i s as much i n the occurrence of Drinking, as i n i t s form and rapidity of occurrence, that the movement i s distinguished from the court-ship display that clearly derives from i t , and so very much resembles i t . Drinking seldom occurs i n bursts of more than one or two movements. The courtship display either takes on a very elaborate form, described later, or else i s distinguished by the mere fact of rapid repetition. Tail-Wag The T a i l Wag i s a slow turning of the whole t a i l area from side to side, such as i s seen i n many Anatidae. It i s not clear what tensions i t relieves but i t occurs i n a great many contexts. The Upwards-Stretch (with or without the Wing-Flap) almost invariably ends with a Tail-Wag. It occurs by i t s e l f , on some occasions, and gives the impression, to an observer, 53 that the bird feels content, and i s i n no-way fearful of anything i n i t s present environment. I believe I have seen i t performed spontaneously when goldeneye were loafing, neither feeding, preening or engrossed i n any other action. It occurs at the end of many preening bouts. It also occurs, i n both downies and adults, when both entering or leaving the water. What the Tail-Wag really signifies i s uncertain. It i s a cosmopolitan movement which often comes at the end of a movement sequence, associated most commonly with the Upwards-Stretch and Wing-Flap. The t a i l may be spread during preening, and spread and Oscillated sideways during coition, but i n the Tail-Wag movement does not take place with the t a i l spread. Head-Shake Complex In the f i e l d I distinguished the Head-Shake movement from two others which might be included with i t . Here I shall discuss a l l three together: Head-Shake (a true comfort movement), Water-Flick (partially ritualised?), and Head-Flick (f u l l y ritualised). Whether they are i n essence the same movement I am not at present sure. (l ) The Head-Shake was used to describe the original comfort movement, as i t occurs i n the Upwards-Stretch movement (with or without the Wing-Flap), and at other times i n the resting position on the water. The comfort movement i s a twist of the head from side to side i n the a i r with beak pointing up or horizontal or down. It i s discussed in the section on the Upwards-Stretch movement. It appears to start as a rolling shake from the shoulder region, which passes up the neck outwards to the base of the skull, u n t i l the skull i s being rotated sideways on the neck. 54 The Head-Shake evidently has the function of shaking off some of the free water from the head, particularly from around the eye and from the nostrils, after diving. When an adult i s feeding the occur- • rence of Head-Shakes between dives is enormously reduced. Sometimes the beak i s dipped^ih/the water so that water is splashed sideways. This quite frequently occurs as a displacement activity, the Water-Flick , during the Laying Neck on Water (Threat) posture. The Water-Flick implies a flick of water (beak down) sideways rather than an upright or horizontal position of the beak. In the Water-Flick the beak starts almost entirely below water level, and with a violent sideways twist of the head on the neck, water i s sprayed side-ways in large drops. Water-Flicks seem to indicate internal tensions as when they occur during the Threat attitude. In the downy the movement is seen most frequently when the l i t t l e bird brings weed or an insect to the surface (as they frequently do). While worrying the food objects forwards (like Jabbing) the object i s often seized and roughly shaken from side to side a few times, between bouts of nibbling. A brood of six downies were watched in their firs t hour on the waters of the lake. "As soon as they reached open water the young began flicking their heads from right to left and vice-versa, but after about five minutes this became converted into true water flick-ing with much water flying sideways." The Head-Flick is a single ritualised flick of the head once to each side occuring at the end of certain courtship displays: Head-Throw and 55 Head-Throw-Kick of B.clangula; Pseudo-Kick and Kick, and at intervals i n Rotary Pumping between periods of Laying the Neck on the Water i n B.islandica. Otherwise than this ritualised form i t does not seem to diff e r from the other movements, though the beak does not contact the water. The Water-Flick (sideways) should not be confused with the Water-Fli p display (upwards), i n which only the t i p , of the whole beak, i s dipped i n the water, but no sideways turning of the head occurs. Water-Dip Complex (1) There i s a frequent movement, the Water-Dip, which does not appear to have any display significance, but equally does not seem to have any comfort movement value. The bird merely dips the t i p of the beak into the water, holds i t there a short while, and l i f t s i t dripping from the water. (2) What the relationship of the Water-Dip to the Water-Flick (see under Head-Shake) i s not known, unless i t be a Water-Flick without the actual f l i c k . (3) The Looking for Food posture, likewise, i s a dipping of the beak, and usually the whole beak into the water. Is there any relation with the weaker, and more stereotyped Water-Dip.? (4) The last movement with similarities to the Water-Dip i s the Jabbing which i s found i n B.clangula i n the pre-coition sequence, and which downies employ to get food into a suitable position for swallowing. 56 Upwards-Stretch This i s the most frequent of a l l the stretching movements. t i s a movement i n which the breast and upper parts of the belly are raised clear of the water and the neck elongated upwards and forwards at an angle of around 7Q degrees. In Figure 3 i t i s illustrated i n B.clangula that the elongation of the neck i s accompanied by a shaking movement which begins at the base of the neck, i n the shoulder region (Frames 16-20), and travels, l i k e a wave, di s t a l l y with increasing amplitude, reaching the base of the crest i n Frame 20. Water i s seen to f l y off the neck during the shaking. As soon as the shaking reaches the head the f i r s t , and weakest, turn of the head takes place. The head had been directed horizontally, but the beak ti p i s now raised and there are (in this sequence) four Head-Shaking move-ments, each of which only lasts 2 frames (l/l2th second). The feet are paddled i n order to enable the bird to retain i t s upright position. In the early part of the movement there i s a resemblance to the i n i t i a l stages of the Head-Throw or Head-Throw-Kick, but as the beak i s held hori-zontal there i s no resemblance to the Masthead posture. While the head i s flicked during the Upwards-Shake, there i s not a Head-Flick at the end of the movement as there i s i n the Head-Throw. Maybe, i f there i s any rela-tionship at a l l , the Head-Flicks have been reduced and deferred to the end of the whole movement i n the Head-Throw. The Upwards-Stretch may occasionally occur without any Head-Shake but this was not certain. 57 Wing-Flap This frequent comfort movement, only takes place when:* the body i s l i f t e d i n the Upwards-Stretch position, but does not always accompany the preceding movement. An interesting analysis remains to be done on the dynamics of the movement, particularly i t s closing portions. Wing-Flapping i n the downy of B.islandica i s illustrated i n Figure l+-As the chest i s raised off the water the wings are l i f t e d back. The f i r s t stroke of the wings takes 3 frames (as opposed to 1 frame i n later wing strokes). The effect of the f i r s t wing stroke-and subsequent l i f t back of the wing i s to raise the body even higher, and the lower part of the belly i s l i f t e d off the water. During the second wing-stroke the head drops forwards again, rather as i t appears to do after the f i n a l beat i n adults, of other species than goldeneyes. After the third (and final) stroke of the wings the belly collapses onto the water as the wings are l i f t e d back again. The wing i s l i f t e d especially high and forwards towards the head before folding. In downies the head was drawn back (and ? up), wings l i f t e d , and the feet paddling. Possibly the feet stop paddling, when the body i s raised, for they appear to start paddling again as the body subsides onto the water. Within a few minutes of entering the water, for the f i r s t time, downies were accompanying the Upwards-Stretch by flapping their tiny wings. The movement was not seen u n t i l the young had Splash-Bathed or done Flap Paddle Diving, and i t followed closely upon their surfacing from these act i v i t i e s . Probably the direct presence of water on the down, or on feathers, i s a stimulus to this shaking activity. However, not every dive was followed by 58 the Upwards-Stretch and Wing-Flap. One individual came up quite entangled i n green alga and had a struggle to free himself. The Upwards-Stretch, Head-Shake and Wing-Flap are one of the common-est forms of comfort movement. To a great extent they occur i n situations of his social stimulation, and, with other preening, are found after border encounters. The signal function may be said to be one of appeasement, for the "end of sequence" aspect of their use together i s very noticeable. It i s most usually the loser of an aggressive encounter i n a courting party or on the boundary of a territory that performs the Upwards-Stretch • Head-Shake 4 Wing-Flap f i r s t of a l l . Almost immediately afterwards the winning bird does them, and the two birds then appear to ignore each other once more. Upwards-Stretch + Wing-Flap occurs at the end of Splash-Bathing. Bernhardt (1940) discusses the wing flapping which i s so common during the early spring and summer: — especially after diving, as i f they were shaking the water out of their feathers, an act which they, as diving ducks, have no need of, and which also they do not usually perform. The drake practices this wing-flapping more often than the duck. Whether he always turns his back, i n order to show off his con-trasty upper surface, I can not say. I have not interpreted this rising on the water as impressive behaviour, and i t i s not directly connected with mating, but seems to be more an expres-sion of general excitement and "nervousness", a state exper-ienced by the ducks at this time. Bright colours would have l i t t l e meaning i n the female. Wing- and Leg-Stretch This i s occasionally seen, but (like false drinking = Water-Flip) only becomes noticeable when i n the ritualised display form (Figure 30). The bird l i e s on i t s side and stretches the wing out behind i t , l i f t i n g the leg out of the water at the same time. It is possible that this muscle-relieving 59 movement may be seen most often in females during intervals on the lake during the incubation period. Downies were seen lying partially side-ways in the water stretching their legs out behind them—with the toes expanded (about 13 days old).-Yawning This is the last of the muscle-relieving comfort movements. I have not recorded i t in adults though I believe I may have seen i t even in them. In downies i t is not infrequent, as they sit out on a log, or idle on the water. I have described i t in Stage 2 downies. The head is held into the shoulders, beak at 45 degrees and the beak is opened and closed, as in nibbling. Possibly i t has threat signal function in downies, as i t appears to have (a similar motion at least, with beak open and head held back at an angle, neck into shoulders) in the "Gesture of Repulsion" of the female Mallard (Anas platyrhynchos). Ear Scratching This, as discussed earlier, is a name for any scratching movement by the leg which is aimed at the head region, whether i t actually relieves the neck, ear region, or face region. The body is turned slightly over and one leg is raised above water level to scratch the side of the face. The head is slightly turned over towards that side. It is a fairly frequent movement on a l l occasions, but does not appear to be commoner under conditions of excitement. The early stage downy frequently rolls right over sideways during Ear Scratching — being too light of body to have complete balance control until i t has grown a l i t t l e more. 60 Forward Preening In this form of preening the head i s depressed forwards and the beak slides sideways, from one side to the other. As the feathers point down-wards and cannot be parted, as they can on the back by manipulation, the beak performs violent nibbling movements, and the contact of the mandibles can be heard as a slight rattling noise on some occasions. The front of the body i s often l i f t e d off the water by means of paddling with the feet as i n the Upwards-Stretch ( t a i l often raised?). In the downy the movement seemed to take the form of 2-3 runs of the beak ver t i c a l l y downwards through the plumage before the body subsides into the water again. Sideways Preening This may develop out of preening or nibbling on the shoulder region or on the flanks lateral to the wing. The bird shifts i t s weight over and l i e s on i t s side, the head bowed forwards and ventrally towards the belly. Often the t a i l i s spread, and bent forwards ventrally. It may be preened. The main preening activity i n this sideways position (which also occurs i n the moments preceding the Wing-and Leg-Stretch) i s that found i n Forward Preening, namely jabs along the line of the feathers and nibbling. It i s often interspersed with Water-Flicks or Water-Dips. In the downy (see Figure 1G) the leg i s braced and flapped above the water, on the side opposite that being preened Back-Preening and Head-Rolling These are performed while the bird i s i n the resting position. It includes preening, or nibbling, on the shoulder regions and preening behind, of under (in front of), the wing when raised to enable this. It i s a very 61 common activity. The theory of "fields" has been discussed i n the infcro^ duction to this section, and applies to preening on the back, around the wings and on the shoulder regions. Head-Rolling, i.e. a dip to the water (Water-Dip — the static pose which i s often seen and resembles the Water-Flick without the f l i c k ) , followed by r o l l i n g the head on the wings, or rump i s probably part of the same motor pattern. This also includes dipp-ing to the o i l gland. Head-Rollings i s seen i n Figure £. Nothing more w i l l be said about this form of preening un t i l film analyses have been made of the actual movements, as has been partly done for Splash-Bathing. Boase thought Head-Rolling had display value. Re wrote: The Goldeneye also uses what may be a form of preening pretence, where both male and female may rub the sides of the head on the plumage of the back, behaving e x c i t e d l y — i n one instance the male began and the female immediately copied. Splash-Bathing This i s an important activity, both as a comfort movement, and also as expressing social excitement. I have seen a flock of eighty B.islandica (nearly a l l drakes) on Lost Lagoon i n Stanley Park, Vancouver i n early f a l l perform t h i s a c t i v i t y (with preening) i n unison for some minutes, then turn over to feeding, and then a l l turn back to Splash-Bathing. Splash-Bathing also occurs (often as the f i r s t activity) after an aggressive encounter. It occurs invariably following post-coition Steaming i n the male, and after coition i n the female. The front of the body and the head are alternately dipped and raised throwing water along the back and causing an undulating movement of the body. A Wing-Flap follows the activity. 62 A film sequence of Splash-Bathing i n B.clangula i s seen i n Figures 5 and 6. In Figure 5 (see explanations) the movement starts as a Water-Dip, or weak Water-Plunge, at the end of which, while the head is s t i l l under water, the head i s flicked (Frames 14-16) from side to side. Head-Rolling follows. The head i s turned over sideways and then drawn back-wards. The crest i s then l i f t e d o f f the back and the beak swuSg forwards (lateral motion) and then lowered and drawn back again. In Figure 6 the Head-Rolling i s not repeated, but the Water-Plunges become increasingly violent. Each s t i l l ends with a Head-Flick (Frames 58-60). The t a i l i s raised, and i n Frames 67-68 the body is thoroughly arched, as the bird were going to make a deep dive. By Frames 73-79, the third part of this action Wing-Shimmering has begun, and i s continued a considerable time before next plunge begins. While Wing-Shimmering the head and t a i l are held up and only the wings are moving. The whole sequence may be repeated over and over again. Thigmotacds Other than Sleep, only one other Comfort movement remains. This is the tendency for downies to proximate themselves. In adults the only signs of this a c t i v i t y would be i n brooding the downies, and i n gregar-iousness i n f a l l flocks. Just after birth when the downy has dried off there i s a nuzzling into the crevices of the human hand, but this i s lost some hours (?) after hatching. Certainly for the f i r s t few nights after they reach the water the female appea/s |broodr~ the downies (Mary F. Jackson, pers.comm.), and this sort of contact would seem to be sought after. 63 An interesting observation was made on some downies 17 days old. The female and 7 of 9 young left a log on which they had been resting and entered the water. The two left on the log had formerly been separated by one duckling. One of the two remaining got up, reversed and sat down again a l i t t l e closer to the other but not in contact. Shortly afterwards i t again got up, reversed again and sat down in contact with i t s neighbour and now facing in the same direction. 64 FUURE 2 . B.islandica (B.C. population) DRINKING From a film by M.T.Myres Frames 1-16 Frames 17-26 Frames 27-33 Frames 34-37 Frame:,- 44 Dip to the Water Raising of the head to a vertical position— neck not elongated. Head passed forwards and beak depressed Head drawn back into shoulder Resting position 24 frames/second. Shutter speed l / 4 8 t h second. 66 FIGURE 3-B.clangula (American population) UPWARDS-STRETCH From a film by Charles Walcott and Benjamin Dane It can be seen that the turning of the head begins in the shoulder region (Frames 16-20) and travels distally. It appears as a swelling at the shoulder in Frame 17 which reaches the crest feathers in Frame 20. The fi r s t , and weakest, turn of the head begins in Frame 21 (to the left). Its compliment to the right is in Frame 22. Turn 2 Lt—Frame 23; Rt—Frame 24 Turn 3 Lt—Frame 25 J Rt—Frame 26 Turn 4 Lt—Frame 27; Rt—Frame 28 The early part of the movement (Frames 7-15) bears some resemblance to the i n i t i a l stages of the Head-Throw-Kick, but d6jfSnot resemble the Masthead as the beak is held horizontal. There is no Head-Flick at the end of the movement. The Paddling which occurs with the feet, to support the upright posture, is not visibly in the drawings. 24 frames/second. Shutter speed l/48th second. < 68 FIGURE 4. B.islandica (B.C. population) UPWARDS^ STRETCH AND WING-FLAPPING IN THE DOWNY From a film by M.T.Myres Three beats of the Wings. Frames 1-5 As chest is lifted wings are lifted back, and then— Frames 6-8 Wing Stroke 1. At the end of the first beat and the l i f t for the second the lower part of the belly comes away from the water. Frame 9 Wings lifted back. Frame 10 Wing Stroke 2. A hard down stroke. Head drops forward again a Little. Frames 11-12 Wings lifted back. Frames 11-13 Head goes back again until almost vertically standing. Frame 13 Wing Stroke 3« Frame 14 As wings are lifted again the bird's belly drops sud-denly as i t collapses onto the water belly f i r s t . Frames 14-28 Collapse onto the water. Wings shuffled. N.B. Note how wing is lifted sharply in Frame 14 as in Bufflehead (B.albeola) courtship, as i t has to be for the base of the humerus to be in the folding position. Shutter speed ? 70 FIGURES 5 & 6. B.clangula (American population) SPLASH-BATHING From a film by Charles Walcott and Benjamin Dane This comfort movement often has social signal effect. It may be broken into three stages (of which the last i s not il l u s t r a t e d ) : — FIGURE 5. Stage 1 . Dip to the Water and Head-Rolling on Shoulder. FIGURE 6. Stage 2. Water-Plunging. Stage 3. Wing-Shimmering (not illustrated.) These three stages occur as one, and Stage 3 is followed again by Stage 1, after a brief pause i n the Normal resting position. In any one instance of Splash-Bathing the three stages may be repeated as many as two dozen times over. Figure 5« Frames 1-13 The dip of the head to the water Frames 14-16 The head i s turned sideways while s t i l l under water and is then directed forwards again. Frames 20-37 Head-Rolling. The head is turned over sideways and then drawn backwards. The crest i s then l i f t e d off the back and the beak swung forwards (lateral motion) and then lowered and drawn back again. During the lateral motion forwards i t may be l a i d down f l a t again (as i n Frame 34). Figure 6. Frames 41-60 First Water-Plunge (with Head-Flick i n Frames 58-60). Frames 62-77 Second Water-Plunge (possibly with Wing-Shimmering beginning i n Frames 72-77)• Frame 78 + Third Water-flunge (start of). Wing-Shimmering thus does not develop u n t i l there have been a number of Water-Plunges (without Head-Rolling, which precedes the deeper plunges). Wing-Shummering occurs i n a pause between plunges. The body i s sunken, with head and t a i l raised, and the back and wings lowered (as i n Frames 77 and 78). The wings are then shimmered i n and out sideways i n the water a number of times. Head-Rolling precedes Plunging and i s i t s e l f preceded by a weak Plunge or Water-Dip (the rear of the body i s humped up—Frame 11—as i n Plunging). The f i r s t plunges have less amplitude than the more rapidly per-formed later ones). ? 24 frames/second and Shutter Speed of l/48th second. 73 THE GOLDENEYES SOCIAL BEHAVIOUR 1. Agonistic Behaviour 2. Courting Displays 3. The Coition Sequence 4. The Brood 5. Interspecific Behaviour 6. Summary of Similarities B.clangula. between B.islandica and 74 AGONISTIC BEHAVIOUR I have more notes on conflict situations between males, than on courtship. This i s because the repertoire of movements i s limited i n true goldeneyes, making the actual locomotory performance easy to describe. In courtship the opposite i s true, the variety of movements i s great dhd the spatial movements circumscribed. This section i s divided into two parts: Intra-specific and Inter-specific agonistic behaviour. The latter i s included here, not because the movements and postures are different, but for the light that some of the Inter-specific aggressive activities throw on the nature of territory i n goldeneyes. INTRA-SPECIFIC AGONISTIC BEHAVIOUR BETWEEN MALES The term "agonistic" i s a ;useful one since i t covers postures and movements which derive from both attack and fleeing drives, and a l l the i n -tergrades between them. It i s possible that the currently accepted notion that a l l actions expressing incompatible social relations are the motor response of one of other of two opposing "drives", may be wrong, and that there are other motivations responsible for some hostile or avoidance be-haviour. Certainly, without proper experimental work, i t would be rash to label a l l such behaviour as due to attack and escape tendencies only. Scott and Fredericson (1951) wrote: When fighting behavior i s analyzed, i t i s found to be one of several patterns of behavior which may provide some measure of adjustment when two organisms come into conflict. Other common and closely related alternate patterns are es-cape behavior, defensive behavior, and passivity. It i s 75 difficult to consider any of these without the others, and i t i s with this general group of behavioral adjustments, which may be given the name "agonistic behavior", that this paper is concerned. Passivity in particular is not readily labelled. The motivation of a passive animal certainly need be neither aggressive nor the desire to escape. Yet appeasement postures (of which the resting position by itself could be one) do tend to prevent an attack from being made without any loss of position being necessary. Indeed the second party to a hostile encounter may avert the threatening actions of the first party in two hostile ways: assumption of the threat attitude also, or an appeasement action. The assumption of an appeasement posture may quite well be a fixed motor response to the sign stimulus of the threat posture, ^t would be a rash man who would say that any particular appeasement posture is necessarily motivated by the escape drive, or is a displacement activity from another field which has or has not yet been ritualised in a conflict situation, rather than that i t is motivated by aggressive tendencies. Diving ducks are particularly interesting, for their agonistic behav-iour is rather different from that of other animals. This is due to the peculiar nature of their territory, and to their powers of diving. In cases of terrestrial territories an attacker can be observed a l l the time as he approaches the victim (unless he has some means of hiding his approach by choosing his route to effect invisibility). In the goldeneyes (which have territories covering a water area) attacks, from any distance, are invar-iably made from below the water surface, although occasionally a bird on the water is attacked from the air. Underwater attacks can only be avoided therefore by observing the behaviour of the attacker, while he i s s t i l l 76 some distance from the victim. Furthermore i f the victim does not Fee the attacker at a distance, he i s not aware of the attack u n t i l the moment of actual body contact. I have seen a Ruddy Duck (Oxyura .jamaicensis) almost turned over on her back by an attacking drake of B.islandica. Thus i n aggressive actions the diving ducks d i f f e r fundamentally from the surface-feeding Anatini. Aggressive actions on the breeding grounds take a number of t a c t i c a l forms: — (1) the male maintains the Threat posture at some distance from the boundary of the territory. Often the male assumes the aggressive posture even when no individual intruders are actually visible, (2) one male has an i n i t i a l advantage from a distance e.g. over an intruder who has flown i n , or other wise come some way inside a t e r r i t o r y — i n which case signs of impending attack, by the territory owner, result i n the intruder flying out of the area, (3) the owner-male again has the i n i t i a l advantage and the intruder retreats by swimming, or more rarely diving, (4) there are times when a male i n the Threat posture i s making dives which look aggressive near a pair of birds on the borders on his territory, but i s content not to direct these dives i n fact at them (although i t looks as though he may be directing an attack at them each time). The male, in these cases, always surfaces about as far away, from them, as he was when he submerged. He i s merely c i r c l i n g about them. The pair, surprisingly, often seem to ignore this behaviour, and he does not actually attack them, 77 (5) both individuals are close to a t e r r i t o r i a l boundary and both i n the Threat posture, and both evidently certain of their claims. Immediate retreat i s rare, and a fight generally occurs. In what follows there w i l l be a discussion of the intention postures of aggression (Crouch and Laying Neck on Water), the mode of attack and fighting, a surreptitious posture of encroachment, and a position i n d i -cating a state of increased awareness, or "alarm" (Expectancy Look). The postures indicating aggressive motivation appear to derive from intention movements of diving. They are therefore fundamentally different from the aggressive acts of the Anatini. 1 Apparently the latter more resemble the Anserinae i n their mode of combat, but this might he due to convergence and lack of the diving habit. The posture indicating alarm may be derived from the intention posture of fli g h t (the opposite i n form i!o intention diving). It resembles the neck-stretched, head-raised position assumed by a l l ducks and geese when "alarmed", and trying to get a better view of the environment, so that they may detect danger i n time to make an es-cape by flying. On the other hand I have not analysed the actual take-off into the a i r from <j„:.. • ;-.• r r ;, u, f i l m to discover the pose of the head at rtv« beginning of flying. It might almost be suggested that since the "alarm" attitude can be maintained for a considerable time (both i n goldeneye and i n the "sentries" among a flock of geese) that this position i s not directly related to the activity of f l y i n g , but i s rather only the appetitive action of escape motivation, or even that i t i s merely an •external"'comfort movement. Whether the "alarmed" position (expectancy Look) i s , i n i t s modern form at least, an intention posture of flying i s 78 at least doubtful, for once body contact of the two contestants takes place escape by flying is unusual in the two goldeneyes (? in B.albeola), 79 Crouched Posture The weakest form of aggressive motivation is expressed by the Crouch attitude. In B.islandica this resembles the Normal resting posi-tion, but the neck i s not drawn into the shoulders so far and the head i s lower, beak perhaps a l i t t l e further forward. It resembles the "threat" posture of Aythya vallisneria (Hochbaum, 1944) . From this posture the Laying Neck on Water (or f u l l Threat posture) i s often assumed as the bird has i t s aggressive motivation increased. A position resembling i t is assumed between dives during feeding (a resting position between dives), when i t i s presumably occurring i n i t s natural setting without any aggressive intent. Brewster (1911) described this posture i n B.clangula under the name tjCrouching Posture". He also noticed that i t was assumed prior to the "Wounded Duck Posture" as he called the Threat posture. He says"the head being thrust forward well above the surface, the neck deeply curved, back somewhat humped". I have noted that i n B.islandica the t a i l i s depressed, that there i s a sharp angle down at the shoulders and that the head i s held forxvard as though on a stick. This illustrates the distinc-tion between the Crouch and the- resting pose between feeding dives. When an intruder appears, a territory owning male drops into this position or directly into the Laying Neck on Water position. The Crouched posture seems to be assumed quite often, even when no l i k e l y victims are within view, and the bird appears restless, turning around, eyeing the environment. The posture may be assumed in the presence of other species 80-e.g. Coots (Fulica araericana) or scaup. If the male goldeneye swims to-wards them i n this posture they tend to d r i f t away from him. It seems l i k e l y that the Crouch i s the earliest intention posture of diving, which i n A.vallisneria now has threat function, while i n the golden-eyes the more extreme intention posture of diving has acquired the specific signal function of indicating the possibility of an underwater-attack. Towards the end of the time that drakes are on territory there i s a marked lowering of their aggressive tendencies. This i s indicated by an observation made on June 5th, 1955. A male B.islandica did not attack a pair of B.albeola which were feeding there. He only swam after them i n a hunched position. He similarly evicted a pair of Eared Grebes (Colymbus  n i g r i c o l l i s ) , simply by following them. Later the Buffleheads returned and tended to keep away from his approach. These birds turned their heads from side to side, and kept swimming ahead of the goldeneye, u n t i l he stopped and turned at what was evidently the boundary of his territory. 81 Laying Neck on Water (Threat Posture) In this position the neck i s thrust out ahead of the body and i s lying on the water. The whole body is flattened and rather submerged. Brewster (1911) called this the "Wounded Duck Posture" i n B.clangula• Munro (1939) also described i t i n B.clangula and noted that -the b i l l points slightly upward so that the chin i s raised above the surface. This i s my experience i n B.islandica, but evidently on occasions the lower mandible i s lying on the surface (beak open), and the upper s t i l l pointing upwards. This i s the situation when there i s a ticking noise from the Laying Neck on Water forward position during Rotary Pumping i n B.islandica. I am not sure whether this aspect of the pose is found exclusively then, or whether i t also occurs during true Threat motivation. Bruggemann (1876) gives the earliest, and almost the fu l l e s t , account of this position i n B.clangula. He refers to "territory", nearly 50 years before E l l i o t Howard (1920), though Altum (1868) had already discussed territory. Bruggmann wrote "Jeder Enterich dieser Species hat sein besonderes, gehau begrenztes Gebiet auf dem Gewa'sser. Kein fremdes Mannchen wird innerhalb dieses Bezirkes geduldet sondern bei seinem Erscheinen sofort verfolgt" (my i t a l i c s ) . He says (in transl.) "It goes i n i t i a l l y towards i t menacingly, i t s neck stretched out horizontally l i k e an angry gander, and dives suddenly under the surface." But he believed that boundaries (Grenze) are never crossed and that the owner turns about while under water, so as to get back into his own area. Mean-while the invader has swum back into his own area, and may assume the threat posture. He did not think actual conflicts were common. Brooks 82 (1920) refers to "Chasing" and describes territorial boundary conflicts and attacks in B.islandica. The posture is illustrated in Figure 7- When the motivation i s really high the forepart of the body is placed so low that the beak is touching water, and only black shows along the neck and fore-part of the breast. In the less .extreme form the black triangle or V on the shoulder comes down between the white flanks and the white feathering of the neck, either pinching them apart or completely dividing them. In Figure 1 there are also two impressions of the front view as seen by an intended victim. In ttie;rupper picture the aggressor is seen from some distance. The feathering of the back is s t i l l visible and the white win-dows form a prominent pattern on each side behind the head, a line of contrast leading away on each side. The first window is very prominent between the face plate and the white of the flanks. In the lower picture, seen also from water level, but very close-to, the angle of vision of the intruder which is taken up by the head of the aggressor male is much greater. The head appears enormous, the flanks and wings less striking, and the couatour of the head breaks the contour of the back and rise* above i t . The beak will be more prominent than a black and white sketch can indicate. The white crescents on the face, lining the rear edge of the upper mandible (which reaches high up on the bridge of the skull be-tween the eyes) are very prominent. The yellow i r i s , of a comparatively large eye, stares right out of the black sides of the head behind the white crescents, and the whole appearance is very "mask-like", and presumably intimidating. Though i t has not been noticed in B.islandica Hochbaum (1944) 83 records that in Aythya vallisneria the pupil is contracted to a pin-point, during the Head-Throw of that species, so that the i r i s blazes out a rich verraillion colour. Possibly this same enlargement of colour may occur in the Threat of the goldeneye. The eye will in any case provide contrast, since the feathering of the head (the slope of the face) bears away at an angle of about 45 degrees. Possibly too the feathers behind the eye are raised providing a back-drop to the yellow eye. The eye may even be rotated forwards to increase the angle of binocular vision. A territory owner may initiate an attack as far as 300 yards away from the victim. He makes a number of dives in the general direction of the intruder. The dives are not always in a straight line towards the in-truder, but may shear off at an angle. Possibly differing views of the shore-line, behind the victim, may give the attacking bird an idea of the distance away that the intruder lies. It is noticeable that while in the Threat position, the attacker may swing his body one way and then the other, so that he is seeing the intruder in the monocular field. The final sub-mergence, may be esfcLm£d, as beginning generally from 5-20 feet away, but on occasions from as far away as 30-40 feet. Often two drakes will do much hesitant diving, parallel to each other, without ever getting any closer to each other, or attacking each other. 84 The Attack Briiggemann (1876) i s quoted by Phillips (1925) as saying that actual battles do not occur. But Bruggemann did not actually write this. Hein-roth (1911) remarked on the peculiar form of the attack i n B.clangula saying (in transl.) "No other German duck has a similar method of attack. As a result i t s neighbours usually dash away terrorised, when they see the goldeneye attack them from below". Bernhardt, as late as 1940, believed that'other species were not attacked by B.clangula except i n captivity, but i n light of experience with B.islandica t h i s i s very doubtful. L i l f o r d (1895) i s also reported as having described under-water attacks i n golden-eye. Johnsgard (1955) describes how males attack one another from below both i n B.clangula and B.albeola. Bernhardt (1940) wrote of B.clangula; — If another drake goldeneye approached, he flew over to the latt e r while he was s t i l l 60 m. away, swam up to him, dived, and attacked under water. Two goldeneye females, without males, that were i n the vicinity, were driven away in a similar manner, when they came near the duck. In my experience with B.islandica most diving attacks would appear to begin between 5 f t . and 20 f t . away from the intended victim. Closer than five feet the attacker w i l l rush at the victim over the surface of the water as i n a fight. The submergence for an attack i s smooth and without splash. A moment lat e r the attacker may come up head f i r s t under i t s victim, which w i l l flap over the water, or may turn and dive, or aim head-first at the attacker. I once saw a Ruddy Duck (Oxyura .jamaicensis) almost turned right over, on i t s back, by an attack. There are occasions when the victim changes his position (by swimming), between the time when 85 the aggressor dives and re-surfaces, but this does not necessarily prevent the aggressor from coming up right at the belly of the victim. If the victim i s outside his territory, and has not already flown or swum fastly away when the attacker surfaces he may, when contacted, flap over the water. Sometimes this takes him back onto his own territory, but at other times he traverses a smaller or larger arc, so that the two birds often arrive back on the spot where they started from. If the victim does not actually escape, or i f he does and gets back onto territory he usually defends, a sudden reversal of roles may be seen, as the-fleeing bird i n a Flapping Chase (neck forwards at 45 degrees, and beating the wings against the water, as i n the display of Oxyura jamaicensis) turns and goes beak f i r s t at i t s tormentor. These Flapping Chases are usual i n an attack unless the bird f l i e s away before being struck. The pursuing i n -dividual w i l l continually alternate, as position on the territory changes, or as motivation changes of as one or other bird t i r e s . The losing bird may get airborne, but does not often f l y far, as one might expect. He i s promptly attacked again. If both birds get airborne and one lands again almost at once, i t dives and the other plunges into the water, from the air , after i t . Sawyer (1928) illustrated aggressive encounters i n B. islandica." It i s sometimes the case that the victim anticipates the attack to the extent that just as the aggressor i s about to surface the victim turns and dives towards him. The two birds may meet head-on midway be-tween the former positions on the water. In one case where this happened, the aggressor reappeared (no longer i n aggressive pose) a moment after 86 diving, farther back than where he had submerged. Another time the male of a pair (which were being attacked) reappeared and swam back to his mate. It seems that the two birds met underwater or that the aggressor saw the intended victim coming for him under water and took evasive action. As an attacker surfaced under him, a victim suddenly jumped right over the attacker, with a single flap of his wings. On another occasion a fight lasted nearly a minute. Both birds emerged from the splashing of water facing each other. The head of each was half raised (neither the normal low threat, or the fully- head-raised position). Then one lunged at the other's head, and seemed to get a grip as the victim turned i t s head away. In these fights I do not remember seeing a bird take f l i g h t , once a Flapping Chase had started. Clearly to take flight is;.the easiest form of escape, i f the victim i s an accidental intruder, and the Flapping Chase must be very energy-consuming. I think i t probable that the Flapping Chase over the water requires a faster wing beat than that employed i n the normal take-off. Hence the bird i s not creating enough l i f t and cannot get off the water. Fights end either by the birds diving apart, or by one bird ceasing to chase the other. When this happens the loser d r i f t s away a few feet, and quickly does the Upwards-Stretch • Wing-Flap A Head-Shake. T a i l Wag sequence. This i s almost at once mimicked by the winner of the bout. This seems to have an appeasement signal value. It i s a sign that the incident i s closed. Munro (1939) writes that "usuallyafter such a performance the male which had been the pursuer would stand upright and shake his wings, curving them forward i n front of his body". He i s 87 probably wrong iniimplying that the winner generally initiates this activity. Sawyer (1928) described a fight in B.islandica: — "a pitched-battle, breast to breast (two males); very loud splashing the only demonstration apparent, the water flying and both birds rising about clear of the surface two or three times; i t (the contest) lasted about twenty seconds, then each (contestant) joined a waiting female (some yards away) and a l l four indulged in much head bobbing." Munro (1939) wrote of an encounter in the same species "swimming rapidly towards each other and then meeting standing upright and striking with their wings." I once recorded that an adult male charged one of the more persistent of three juvenile males (of B.islandica) which were doing the Kick a number of times""at his mate whom he was pumping. The female was Jiving. The male, when he charged them, had the head held higher above the water than in the normal threat position and only opened the beak at the last moment. On another occasion a cou^.ng pair were interrupted by an intruding male who chased the female (flapping) over the water. The female darted back and forth doing rapid submergences, and coming up facing another way, an interesting form of excape that I have not seen in males. Only after she had passed her "mate" about five times did he make a jab at the pur-suing male, with his beak. Finally the female flew off with a third male. On another occasion mised up motivations were likewise observed. A male was courting a female (Rotary Pumping + Threat), but then dove at her, and both flew into the middle of the lake. The male then returned to his territory. Most probably he already had a mate and was courting promis-cuously, or mistook her identity. 88 If a female is nearby a fight ends in a Triumph Ceremony. If ter-ritory neighbours have been fighting, i t is possible i t may not matter i f the male won or lost the encounter. A male was once seen to do Rotary Pumping after a fight although no female was in sight. 89 Surreptitious Encroachment In OxyurvcL .jamaicensis (and ? in Anatini and Bucephala) I have seen a pose which resembles the aggressive pose, but which has a defensive motivation apparently, in part at least. It is the opposite extreme of the "alarmed" position, but might represent a state of intention excape by diving, rather than flight. This aspect of motivation remains to be anal-ysed further. Heinroth (1911) evidently saw this position in B.clangula for he wrote (in transl.) that " i f anxious to escape observation without resorting to flight, they swim away rapidly with the head held forward, almost on a levle with the water, and the body 'sunk1 so that the t a i l is out of sxght". Since he also described the "alarm" position and Threat,, i t is unlikely that he was confusing the above posture with the latter. Thus i t seems that escape motivation, or intention, can be expressed i n two ways, the one used when the bird is well-aware that i t is under observation, the second used when the bird believes itself to be camouflaged, or otherwise unobserved. In this crouched, or flattened posture, the bird is evidently attempt-ing to make itself invisible. It assumed the posture when i t appears that the bird knows that the area, in which i t i s , is in the territory of an-other bird. It evidently desires to stay there, but is aware that, i f detected, i t will be; attacked. It is a conscious intruder, on the defensive, but with no intention of fleeing (at that moment). This is a very anthro-pomorphic concept, but human beiags are often aware of behaving and thinking in this wayj i t might be regarded as a surreptitious encroachment, with the 90 intention of gain. Chances are strongly against success, and the individ-uals are aware that an attack, i f provoked, will result in their being defeated. In goldeneye this situation of being forced to enter the territory of another bird does sometimes happen. In 1955 the first brood onto Sorensen Lake was defended by a particularly aggressive female. Two females were s t i l l incubating, however, in trees one on either side of the lake, within the brood territory of the aggressive bird. The two in-cubating females liked to land on the water in front of their nests, after feeding elsewhere, before flying up to fly to the nest itself. But ithey were always attacked. Unfortunately I did not record their behaviour, but a situation where surreptitious encroachment has advantages is illustrated by these incidents. The form of motivation involved in this behaviour, has not so far as I know, been recognised in animals, though i t would appear to occur in primates , and in the play of a number of mammals i t forms a frequent element. There is no valid reason why birds should not be capable of this subtle type of offensive behaviour as well. Perhaps the dominant aggressive component is expressed by the flattened attitude, but the restlessness of the bird, turning from side to side, looking a l l around itself, and often taking shel-ter behind a snag or other part of the environment, suggests that the bird i s aware that an attack is likely. If the attack comes, the bird will, I think, fly away. When a duck is aware that an attack is immanent, and is more fright-ened than other wise, i t will go into the "alarmed" position with raised head. 91 Expectancy Look ("Alarmed"; •position) The difference i n circumstances between this pose and that assumed i n the fore-going section probably l i e s i n the "alarmed" bird, either being uncertain of the whereabouts of the aggressive bird (but knowing there is one), or being uncertain i f an aggressive bird, which i t can see, i s going to press home an attack. In the case of the surreptitious encroachment the intruding bird does know where the territory owner i s , but also knows that the territory owner i s not yet showing signs of aggressive intent. When "alarmed" goldeneye swim with the head held high, as when they are curious about man. The Expectancy Look i s however seldom, i f ever, seen i n a border conflict between equals. Millais (1913) wrote: "If observant of danger at a distance and about to rise, the neck i s straightened and the head held up high, with the crest somewhat raised." Phillips (1925) wrote "alarmed and swimming away from a danger-point the body i s deep i n the water, the t a i l dragging or submerged and the bird ready for an instantan-eous dive or jump. The "alarmed" head-raised position may not i t s e l f , be an intention movement of excape, since when actually struck from below, the victim seldom actually f l i e s . Attacks are followed by Flapping over the water. Attacks are instantaneous events. The victim may suddenly find him-self being l i f t e d out of the water by the beak of the attacker planted firmly i n the middle of his undersides. Only the behaviour of the aggres-sive bird, at a distance, indicates his attentions. Sometimes the sight of him alone w i l l make a potential victim f l y away, sometimes his diving w i l l be sufficient stimulus, sometimes the victim is sufficiently uncertain 92 that he w i l l not f l y u n t i l actually struck from below. In a l l three cases the bird may be i n the head-raised position xvhich i s an indication of alarm or uncertainty but i s possibly not, of i t s e l f , an intention of flying. One moment the bird i s sitting on the water, next i t i s Flapping over the water (unless i t decides to fly) before the attacker presses the attack home. The only true intention movement (such as one might find i n a t e r r e s t r i a l bird or animal-' i s that of moving slowly (swimming) i n a direction away from the aggressor, ^his indeed occurs when the victim is alarmed. It i s a frequent accompaniment of the Expectancy Look. Here a true gradation from no movement through various speeds to fastest swimming rate possible i s found. But there can be no intermediate between flying away and sitting on the water, for the doing of the one i s a direct viola-tion of the other. This distinction between intention flying and intention swimming or walking i s important, for i n the latter the intentions move-ments are weak forms of the movement i t s e l f . The Expectancy Look i s also a sign of curiosity, an increased awareness or sensitivity to the environment, and i t i s used sometimes when facing towards a disturbance, and even when swimming towards an object that i s interesting the bird. A special case, that i s of interest in the study of the motivation behind, the Expectancy Look, i s when the female has been temporarily parted from her brood. She assumes the Expectancy Look i n searching for them, and here the movement expresses a desire to find some-thing (the brood), and of going towards i t , rather than away from i t . The Expectancy Look i s not, I assume, motivated by escape tendencies of a tendency to f l y under these circumstances and this i s why I prefer the 93 term, Expectancy Look, to intention escape or intention i^4$g». The Expectancy Look i s assumed under conditions of alarm or curiosity. It enables a wider view of the environment to be obtained. It i s natural that i t should occur when a bird expects an attack upon i t from below. Whether or not i t i s the intention movement of flying, as well, i s not known, nor whether i t can be regarded as an expression of escape tend-encies. The pose i s resembled, i n ritualised form by the Head-Up and Head-Raised courtship postures of the two species of goldeneye. 96 INTER-SPECIFIC AGGRESSIVE BEHAVIOUR Attacks by B.islandica on other species, as noted during my study i n the Cariboo District of British Columbia are interesting for two reasons: (1) Phylogenetic relationships of the -victims, to B.islandica (2) Circumstances of the attacks, and the light such attacks may throw on territory i n goldeneyes. At this stage a l l that can be noted on the f i r s t score is that Mergini are frequently attacked, Aythyini are attacked f a i r l y frequently, while Anatini have not.been observed under attack, although six species breed on or around the lakes. The following table i s a summary of the species attacked: — TABLE 1 Tribe Mergini Aythyini Oxyurini Species No.of species breeding on lakes Bucephala clangula (migrants) Bucephala albeola  Melanitta perspicillata (migrants) Aythya.americana  Aythya a f f i n i s Oxyura .jamaicensis Anatini n i l Miscel-laneous Attacker Attacked (Sex) (Sex, or young) M+- F M M M + F M M F M +- F M . M M + F F + y Fulica americana M +- F 97 Some extremely vehement attacks were made by drake goldeneye on other species while the mating territory was i n existence. This was especially the case on a small slough where there were no other B.islandica territories. B.albeola and Oxyura jamaicensis (female) were seen to suffer violent and persistent attacks there. Except for the incident between B.islandica (female) and B.clangula which was only a case of momentary Threat as the ice was melting i n 1955, the bird involved i n a l l the other attacks by female B.islandica was the f i r s t one to bring off her brood on either lake i n 1955 ( f i r s t seen at the north end of Sorensen Lake on June 16). Her brood territory was a narrow strip of water at the t i p of the lake, and included the water immediately i n front of a natural goldeneye nest, and also i n front of a nesting box (occupied by incubating females) on opposite shores of this part of the lake. She spent much time harrying the two females of her own species from these two nests, whenever they came off to feed, so that they had to f l y half way down the lake to open water. She made i t very d i f f i c u l t for them to return to their nests, and they were continually being chased i n f l i g h t around the shores of the lake. Below are more detailed accounts of the various interspecific encount-ers, by species: — American Goldeneye (B.glangula) This was only seen on migration i n May, i n small parties. Does not breed i n the Cariboo District. The encounter was on May 4th, 1955 on the lagoon of Westwick Lake, the day the ice was melting. About half a dozen pairs of B.islandica were sittin g around on the ice or swimming at the edge of i t . A l l were 98 highly i r r i t a b l e and as the stray migrating drake B.clangula landed on the water a drake B.islandica landed almost on top of him and both submerged. On re-emergence both flew. The drake B.islandica was seen i n Threat pos-ture a number of times more, but the B.clangula did not f l y , or look apprehensive, u n t i l the B.islandica drake came up under him. The migrant merely fed during the brief intervals when he was undisturbed. A short while later the B.islandica drake deserted him to attack a pair of his own species. Clearly this period of the year i s the. most c r i t i c a l i n the establishment of territory, as the water operis^iip and i s occupied. Bufflehead (B.albeola) A few pairs breed on these lakes, and stay throughout the summer, but i n May a larger number are around, apparently on territories which they later desert, and i n courtingparties. In 1955 courting parties were fre-quently seen but only one brood was brought to the lake. In 1956 two pairs remained on Sorensen Lake for almost a week and I thought they were going to establish themselves, but they disappeared. On a small slough occupied by only one pair of goldeneye I was told that i n 1954 a p&ir of Buffleheads had nested in a tree only a few yards from the nest of the goldeneye at the edge of the water (Mary F. Jackson, pers.comm. ). T e r r i f i c fights were seen between the drakes.of the two species. In 1955 only /the goldeneye nested there. However on May 11th I observed a female Ruddy Duck being violently attacked. A pair of Buffle-head and a single Bufflehead drake were also present on this slough. The single drake Bufflehead became the object of aggression by the drake of B.islandica. When the goldeneye was some ten yards away the Bufflehead 99 would fly, and either escaped or landed somewhere else, but in one case when they landed the goldeneye was only five yards away. The latter then dove and the Bufflehead flew and again landed. Then the goldeneye flew and the Bufflehead swerved away across the slough—the pursuer landing near his female for awhile before.,renewing his attack. On another occasion, two days later, on Sorensen Lake, a goldeneye attacked a Bufflahead under water; the latter did not fly as the bird came up (as a goldeneye would usually do), but was actually levered partly out of the water. The attacker then flapped his wings, and the Bufflehead about five-feet away also did so. Then the goldeneye assumed the Head-Up position opening and clesing the beak. The Bufflehead bathed and wing-flapped and the goldeneye dipped the head below the water and flicked i t . On May 17th, 1956 a drake goldeneye in f u l l Threat dove and a juvenile male and a female Bufflehead flew. The attacker came up only a short distance towards them, however, immediately after they flew. It looked very much as though he saw them fly, from below, or that he was not intend-ing an attack. Surf Scoter (Melanitta perspicillata) Seen only in small groups on migration at the end of May. These parties may stay, usually in the middle of the lake, for some days. Two attacks on these scoters were seen. The first was a fight on the water followed by a dive, and the scoter flew (paddled over the water) on May 20th, 1955• The second occasion a drake goldeneye cameup right under-neath a drake scoter, which jumped away. The goldeneye remained on the 100 surface i n a threatening posture only a foot or two away from i t and another pair of scoters. It then swam slowly away and did the Upwards-Stretch movement without the Ta i l Wag. Redhead (Aythya americana) A breeding species, nesting i n the marshes of Scirpus at the ends of, more sparsely along the sides of, the lakes. Apparently only weakly ter-r i t o r i a l , themselves, for groups of drakes and females are seen i n the middle of the lake, a l l through the breeding season. The female B.islandica on Sorenson Lake flew at a pair of Redhead on one occasion,, dove and forced both Redheads to f l y down the lake. Lesser Scaup (Aythya af f i n i s ) A very common breeding species i n the area. Nests are found out i n the grasslands. like the Redhead appears weakly t e r r i t o r i a l . A drake B.islandica started 100 yards away from a pair of scaup on June 11th, and making three dives towards them, came up under them three times i n succession after he reached them. On the f i r s t two occasions the scaup-merely moved a yard or two away over the water, but upon the attack being repeated a third time they flew away. After the f i r s t submarine attack the drake scaup stretched his neck up i n an "expectancy look" pose. It i s worth noting that at a distance a Lesser Scaup has the general form of B.islandica dark purple head and white flashy flanks. Ruddy Duck (Oxyura .jamaicensis) This i s one of the commonest ducks nesting at Westwick and Sorensen Lakes. It nests i n the marshes, which are localised and only occur i n any 101 depth near the in l e t or outlet at the end of each lake. These marshy-fringes are not the favoured stretches of water for goldeneye te r r i t o r i e s , which are most numerous along the parts of the lake fringed by grass-range. Nevertheless the Ruddy Duck frequently gets i n the way of a goldeneye, and i s i n no way immune to attack from i t . Vicious attacks were made on a female on the small slough mentioned earlier (by the drake goldeneye mentioned i n the section on B.albeola). A female Ruddy Duck (none nest on the slough) was forced to seek refuge among fallen logs just below the nest occupied by the female goldeneye. The male did not intensify his attacks, when she was this close to his nest, imply-ing that i t was the territory alone that counted and the nest had no extra valence for him. Later on the Ruddy Duck was seen out i n the middle of the lake clearly i n a very defensive posture, i t s body very compressed when on the surface, swinging round and keeping a sharp watch on the drake goldeneye while diving for food. In one part of a fight between them the Ruddy Duck was almost turned over backwards by the goldeneye. In chases the drake goldeneye flew after the Ruddy Duck and landed where she dived, sometimes pivoting around as he landed, as though he knew which way she had travelled under the water. On another occasion the female goldeneye at the NW end of Sorensen Lake did five dives or so from a considerable distance, with increasingly threatening crouches, towards a drake Ruddy Duck at the t i p of, the lake. At last he noticed, swam i n towards the reeds and then dove. The female goldeneye remained nearby for a minute or two. Her young were scattered over the middle of this part of the lake. 102 On July 3rd a remarkable case of intolerance was watched from the female goldeneye having her territory at the north end of Sorensen Lake. This bird was also very intolerant of other duck as well as of other ' • female goldeneye. Following an attack on a young Coot (see under that species) she was seen flapping over the water, flying and diving after a drake Ruddy Duck, with at least half a dozen encounters witnessed. The Ruddy Duck apparently got tired, as he did not f l y after the f i r s t two or three encounters and flapped over the water with his wings. Later s t i l l a pair of Ruddy Ducks appeared from the reeds i n attendance on one downy (apparently newly hatched). A l l three ducks swam along the edge of the reeds towards the north west end of the lake, accompanied by two or three other drakes who courted vigorously with the frothing-up-of-the-water move-ment. One drake performed the paddling-over-the-water act which often precedes of follows the "frothing" of the water. On this occasion the "paddling" occurred either before the "frothing" or quite separately, but not after "frothing". An hour later the female goldeneye attacked the pair of Ruddy Ducks and routed them, "and then preceded to keep the young Ruddy duckling under water ( i t dived i n defence) for 1-2 minutes and seized i t by i t s nape when ever she could. I did not see i t again a and i t did not join i t s parents who f i n a l l y fled to the reeds. Occasionally when the goldeneye had the duckling by i t s neck the female Ruddy rushed hunchbacked across the water at her, but seldom would have got that near, and the goldeneye would turn and jab at her and she and the male (who did l i t t l e but accompany the female and do courting movements ("frothing") most of the time would turn and flee from the goldeneye. " Doubtless had there been more than one downy i n the Ruddy Duck brood, no death might have occurred, but as i t was the attack of the goldeneye was 103 concentrated on one individual whichwas, be i t noted, the downy not the adults. Coot (Fulica americana) On May 4 , 1955, during one of i t s long dives towards a B.clangula a drake B.islandica was seen to come up under a Coot. On July 3rd, 1955 just before the incident with the Ruddy duckling mentioned above, the female nearly k i l l e d a two to three week old Coot. The duck was, I think, holding the young Coot below the water or else diving after i t , for the duck was often almost, i f not wholly submerged. The Coot then came to the surface and the goldeneye swam hard after i t pecking at i t of holding i t . Her Stage 2 ducklings gathered round at this and did the nod greeting. The female duck l e f t the Coot immobile on the surface after two or three minutes of harrying. The Coot was immobile for a couple ,of minutes-as the ducks drifted away, and then raising i t s head from the crouched position swam rapidly into the reeds." 104 Discussion The case of the goldeneye on the small slough attacking a female Ruddy Duck was of parti cular interest for periodically i t would break off the fight for about five minutes, and then take up an aggressive pose again. This indicated not only that the whole slough was i t s territory, and that i t tolerated no other diving duck in the area, but also that at highest intensity of stimulation (forced proximity to another duck even of another totally unlike species) an exhaustion of i t s reaction specific energy occurred, and i t s aggressive motivation was quickly exhausted every ten minutes of so, taking some five minutes to build up again to a point where an unlike species within i t s territory was an i r r i t a n t again. This i s quite unlike the behaviour of a goldeneye on a larger lake, where the attacked bird can escape to another part of the lake. In this case the stimulus i s quickly removed and the aggressive motivation quickly declines. On a slough this i s not the case and i t appears that the con-tinued presence of the stimulus results i n a much higher level of latent -no aggressive motivation, which apparently can/longer be expressed by the motor pathways after a certain time, resulting i n a refractory period of some kind. For when a short rest has been had the aggressor once more rushes into the attack. Meanwhile the i r r i t a n t has s t i l l been present and presumably acting upon the nervous system. One gets the impression that the motor path i s quite easily exhausted. Th'e.;nearest analogy, which i s rather a good one perhaps, i s of bouts i n boxing. The fight i s s t i l l on, but the desire to punch the other guy has to be controlled during the intermissions, and because of the frequent rest periods the fight can go 105 on much longer. Without themothe contestants would become exhausted much sooner. In Territorial birds of course there is no great advantage in attacking every bird in sight, and indeed though aggressive encounters on the borders of territories are frequent, each one seldom lasts any time as the loser swims or flies away. Thus a sudden and fierce exhibition of Threat or actual aggressionsis effective. Evidently on small sloughs, however, where another duck is unable to escape to open water at a l l (and so requires considerable beating-up to be persuaded to fly off to another slough or a larger lake} there is a greater resistance to be overcome. When a goldeneye drake attempts in such cases to expel an intruder from a one-territory slough, the fairly rapid fatigue of the efferent path in the central nervous system results in sharp but brief bouts of activity followed by a refractory period. The greater stimu-lation that evidently occurs and greater intolerance towards other birds that is shown has survival value i f territorialism is useful, and is to be maintained, by drakes having territories with long shorelines as in small sloughs, or narrow water boundaries, as in the arms of a lake. 106 FIGURE 7. B.islandica (B. C. population) LAYING NECK ON WATER ("THREAT" POSTURE) Artists drawings by M.T.My res, from f i e l d sketches. Water-level views of an aggressive drake. A. Distant View. The head i s lowered to the water and the back of the b i r d i s seen rising above i t . The white windows are prominent i n series.in the black back and wings. B. Close-up View of Aggressive Drake. The top of the head now rises above the contour of the back, the wing windows are less prominent, the head being outlined rather less against the black dorsal regions as against the prominent white flanks. The white face crescents and golden i r i s are prominent, and "mask"-like i n appearance. C & D. Side Views. In C. the bird i s less highly motivated than i n D. The white of the neck s t i l l shows, and i s divided from the white of the flanks by a black shoulder flash. In D. the neck i s lowered below the water, the beak perhaps t i l t e d a l i t t l e more up. It may (?) be opened with lower mandible on water (or i n courtship only ?). There i s one white flank patch. 107 B. , 108 COURTING DISPLAYS GENERAL ACCOUNT The only detailed or accurate account df display in B.islandica is that by Sawyer (1928). As explained earlier I did not read this paper until after I had completed my field observations. It will become ap-parent, in what follows, that our separate accounts corroborate each other very well. Although Sawyer did not specifically name any displays he did separate them in his own mind and illustrated them in two pages of fig-ures. Unfortunately text and figures f a i l to agree owing to a number of typographical errors. These are detailed in Appendix 1, for the sake of completeness. It needs only to be stated here that Sawyer's illustrations are excellent. I have as- yet, seen very l i t t l e of the display of B.clangula in the field. But there is obviously much to be derived from comparing the displays of B.clangula with those of B.islandica since they are closely related species and occupy mutually exclusive breeding ranges. The l i t -erature of display in B.clangula i s quite large, but i t covers compar-atively few of the discrete display movements and is the work of those untrained in ethological observation or theory. However i t has been felt that a complete review of this literature on ^.clangula would be worth while i f only to clear the background for further study. It has been necessary sometimes to read between the lines or to reinterpret observa-tions in light of recent knowledge, but on the whole the literature descriptions can be broken down into discrete displays. 109 The Literature of B.clangula The reason that displays of B.clangula have been described more frequently i n the Literature than those of B.islandica i s that B.islandica i s confined to British Columbia and adjacent regions and Iceland, while B.clangula winters throughout the popular ornithological hunting grounds of western Europe and the Atlantic seaboard of the united States. The earliest accounts of the display are those of Bruggemann (1876) from Germany (cited i n P h i l l i p s , 1925) and of Gilpin (1880) from Nova Scotia. The latter wrote "the male throws his purple head far backwards t i l l i t rests upon his back". Heinroth (1911) published an important paperron the ethology of the Anatidae, but he had devoted but l i t t l e time to B.clangula and merely wrote (in transl.) "the drake with a jerk throws the head far back onto the back". But most accounts merely<describe one of the many courtship displays of this active species. Curiously the most extensive descriptions come from the period just before the First World War. Townsend (1910) stimul-ated William Brewster to write up his observations on wintering goldeneye near Boston, Mass. He described a number of discrete displays, which he observed one day i n February 1909, i n a paper which he read to the A.O.U. Congress i n Washington i n 1910 (published 1911). This i s the only paper to date i n which an attempt has been made to name discrete displays and i s much the most detailed i n i t s descriptions of the displays. Millais- (1913) added l i t t l e that was new to our knowledge of golden-eye display when he published his great work on the British Diving Ducks, except an apparently exaggerated account of the Post-coition display. n o Nor did Phillips (1925) add anything that was original i n his Natural History of the Ducks (except an account of the seldom seen, "Backwards Swimming Display"). Bent (1925) and Kortright (1942) merely make a few quotations from the literature. Isakov i n Dementiev et a l (1952.) evid-ently takes his account from west European authors, and his illustrations are copies from Bernhardt- (1940)• Papers which have something to add to our knowledge of the display of B.clangula are those of Boase (1924) , Bernhardt (1940) , and Johnsgard (1955) . Munro (1939) i n n:5-s important paper on the American Goldeneye i n B r i t i s h Columbia had nothing to add to the scant account i n Munro and Clemens (1931) . By far the most comprehensive summary of the displays i s that i n The Handbook of Br i t i s h Birds (Witherby et a l , 1939) . This account,compiled from the literature by F.C.R.Jourdain and B.W.Tucker goes a considerable way towards isolating the distinct postures and movements, but does not attempt to name these movements, which would give them the authority of distinctive nomenclature. Nomenclature Comfort Movements, like flying, swimming, eating and drinking are act i v i t i e s which every bird has to perform, and ones which are found i n a l l species with l i t t l e variation i n form. Because of their biological necessity they are deeply inherent i n bird behaviour patterns, and for reasons of functional efficiency they have a similar appearance from species to species. The taxonomic value of sexual displays, however, l i e s i n the 111 fact that they have no adaptive function, and are developed to express specific differences. However, closely related species w i l l tend to have displays which though recognisably different i n form, were at one time (when the species had not yet differentiated from one stock) similar or, rather, the same. Much interest l i e s i n identifying homologous displays. Identification necessitates analysis not only of form but of motivation and function '.'meaning", and occasion of occurrence (Baggerman et a l . , 1956) . If i t can be shown that these latter are similar then the variations i n form can be explained as secondary, and result from the same forces that separated the species. On the other hand, similarity i n form of two movements, i n different species, may not be an indicator of a common origin. Convergence may have occurred, and the way to show whether this has happened or not i s to analyse the situations i n which the two dis-plays occur, and their function or signal meaning. Unfortunately this stage of goldeneye behaviour studies has not yet been reached. Until i t can be shown, by analysis, that movements are homologous, and should have the same functional nomenclature, i t i s necessary that distinct descriptive names should be used, for similar displays i n different spec-ies. This important principle was pointed out to me by Dr. Frank McKinney (in l i t t . 1954) . Thus an attempt has been made here to give the most suitable names to the various movements or postures. The rule of priority should, wher-ever possible, be used. However there i s no virtue i n perpetuating a name which i s no longer descriptive, or which has a functional s i g n i f i -cance which i s s t i l l undemonstrated. Thus the term used i n the trans-112 lation of Lorenz (1951-52) for the females twisting of the head, from side to side, should never have been called "Inciting" i n the f i r s t place. We do not know whether i t has the meaning "I belong to you" directed at the male she i s meeting or following, or "See here, I am mated to this male", directed at other males. It so happens that i n eiders (Somateria) a display occurs i n this context which consists of two quite different parts (McKinney, M.S.), which probably have two rather different motiva-tions. In goldeneye the behaviour of the female takes two forms of which the "Inciting" (here called "Jiving") i s one. Possibly the "Deep Jiving" posture i s from the same stock. Another example of confusion i s that of "Pumping". This i s used by Lorenz (in transl.) to describe the up and down movements of the head by both sexes i n the Prelude to Mating, i n Mallard (Anas platyrhynchos). In Bucephala I f i r s t used this term for the rotary movement of the head of the drake in B.islandica. It i s conceivable that the two forms of Pumping might have the same origin, but unt i l we know one way or the other the movement i n B.islandica must be known as "Rotary Pumping", ^f they prove to be different movements the names may then s t i l l stand, since they are distinguishable, and specific, which "Pumping" by i t s e l f would not be. Another example i s that of the "Masthead" of B.clangula (as here re-^ defined) which resembles the "Neck-Stretch" of Aythya vallisneria (Hochbaum, 1944) , but we do not know that they are homologous. Indeed I am uncertain i n my own mind what the homologous movement i n B.islandica may be (possibly the "Water-Flip"). Their signal functions may not be the same, even i f there i s phylogenetic relationship. By contrast, when the 113 signal functions, and occasions, of the various movements, are known we w i l l be using the same Functional name for quite unlike movements i n many-cases. Thus the crouched "Threat" posture of A.vallisneria (Hochbaum, 1944) has the same function (namely intention of aggression i f certain circumstances arise) as the "Laying Neck on Water" posture of B.islandica. This l a t t e r posture greatly resembles the "Sneak" posture of A.vallisneria which does not apparently have a threat function. By distinguishing descriptive names between species there i s an avoidance of confusion should the function of a named display have to be revised. For example i f the "Laying Neck on Water" posture of B.islandica were to be called the "Sneak", as i t might, i t would be presumed that the "Sneak" also was aggressive in function, which apparently i s not the case. The i l l u s t r a t i v e material and descriptions Certain of the figures are taken from my own film of B.islandica, But as stated earlier i t was preferable to make a comparison with B.clangula i f this were possible. I have had the great privilege of being allowed to make use of p i c t o r i a l i l l u s t r a t i o n from a f i l m of B.clangula taken by Charles Walcott and Benjamin Dane, and of a short film taken by Drl Frank McKinney at the Wildfowl Trust, Slimbridge, Gloucestershire, England. The latter film contained shots of B.islandica (Icelandic stock), and B.clangula (both American and European races). Naturally, while making drawings from these films I have been able to compare B.clangula display sequences with those I know from f i e l d observation and film of B.islandica. It must be stated categorically 114 here, that the opinions expressed i n the following descriptions are mine and mine alone. They do not necessarily represent those of the persons who took the films. It w i l l however be of considerable interest to compare the descriptions made here, with those that w i l l be found i n Walcott and Dane's forthcoming paper (in preparation) on B.clangula. and B.albeola. The Courtship and Coition displays. For the sake of convenience I have separated those displays which are seen in the courting party and between the pair on territory (without mating), from those occurring i n the coition sequence, alone or for the most part. Thus Pre- and Post-Coition Steaming are clearly found only in.the mating sequence. The Wing and Leg Stretch are found almost ex-clusively i n the coition sequence, but occasionally at other times (they are almost impossible to t e l l apart from the comfort movements they so closely resemble either i n a courting party, or even when they occur i n isolation). Rotary Pumping, and to a lesser extent Bowsprit Pumping and the Kick Complex of displays are seen to arcertain extent i n the coition sequence, prior to mating, but are a much commoner feature of courting parties. Indeed B.islandica and B.clangula tend to d i f f e r in the degree to which these movements may be found i n mutual or communal courting situati ons. As w i l l become apparent Rotary Pumping i s the most frequent display element i n B.islandica. But a movement from a different complex, the Head-Throw, is the most important i n B.clangula. This species has a more highly differentiated Kick Complex than has B.islandica, but i t s pumping 115 movement i s apparently less differentiated. The Masthead has not been identified i n B.islandica, unless i t be the same movement as the Water-Fl i p . Thus i n most of the courtship displays the two species d i f f e r quite considerably i n the form, the relative frequency, and the occurrence of each display. It took me some considerable time to realise that apart from the coition sequence (Wing and Leg Stretch, Water-Flip, Jabbing, and the two forms of Steaming), the only sexual display found i n courting parties, which takes exactly the same form i n the two species is the Head-Up or Head-Raised (with Neck-Withdrawn) posture-movement sequence. Whether this i s a basic courtship movement which has not differentiated, but has been replaced (as a criterion for species recognition, by more differentiated displays), or whether i t was the last basic display type developed before the species diverged, or whether i t has a somewhat generalised and unspec-i f i c signal function, i s not yet certain. In both species i t i s more ; closely a l l i e d to the pre-coition displays than to the pairi-~-r-forming and pair-maintaining displays. It i s not as frequently seen as Rotary Pumping or the Head-Throw. I have distinguished the following male courting displays i n the order i n which they are described: — Complex B.islandica B.clangula Head-Up Complex Head-Up Neck-Withdrawn Head-Raised Neck Withdrawn. Alike Kick Complex Pseudo-Kick ) Head-Throw ) ) not unlike ) very distinct Kick ) Head-Throw-Kick ) Pumping Complex Rotary Pumping Bowsprit Pumping Not unlike — — Masthead Backwards Swimming 116 The female's repertoire of displays i s restricted by comparison. She has two regular displays, Jiving and Deep Jiving directed at the opposite sex and an Invitation to coitus, which is a static posture dis-played at the mate. Often the male is not ready for coitus when she as-sumes this posture and the preliminairesto mating may continue for as long as 20 to 30 minutes. She has no post-coition display though she regularly performs Splash Bathing movements, as does the male. 117 THE FEMALE There are three actions which are characteristic of the female alone, Jiving, Deep Jiving and the Invitation to Coitus. The last w i l l be deferred u n t i l the Coition Sequence. However the female may also perform, in weak or infrequent form, some of the courtship displays of the male. Deep Jiving (B.clangula at least) has resemblances (probably non-relational) to the Head-Throw-Kick i n one of i t s forms. Jiving i s only seen i n the female, and i s the main movement with sexual motivation i n female goldeneye, as i n many other ducks (= Inciting display of Lorenz). The circumstances of female use of other displays i s very uncertain. It remains to discuss the literature on them. B.islandica I have observed Rotary Pumping by the female i n the f i e l d . This i s t\oV however very common and i s described i n the sections on inter-female and Brood behaviour. It i s a form of greeting display, and appears to have a similar function in B.clangula. The Water-Flip i s commonly performed i n B.islandica. B.clangula In the literature a number of displays are attributed to the female. It i s of interest that the chief of these i s the Bowsprit position, which i s believed to be the homologue of Rotary Pumping in the other species. It evidently also has a greeting function. The next most commonly de-scribed of the displays, chiefly found i n the male, i s the Masthead. There i s some l i t t l e evidence that the Head-Throw and/or the Head-Throw-Kidk may occasionally occur in weak form, but i t cannot be ruled out that 118 the "females" i n which these displays were supposedly seen were not immature males. The alarm posture resembling the Head-Raised posture i s presumably found i n the female as i t commonly i s i n B.islandica. The Bowsprit performance i n the female has been described as follows (B.clangula): — In one instance the female made display by extending the neck with head in line at an angle of about 45 degrees, (Boase, 1924). She dips her b i l l ; splashes water sideways with i t ; pro-trudes her neck at an angle of 45 degrees (the "bowsprit pose" of Brewster); she stabs the zenith; kicks up water behind her; and as I say, r o l l s and extends her wing. (Gunn, 1939). ....the females. The .latter were also active, jerking with the neck, making drinking movements, lying low on the water, and swimming up to the males. (Bernhardt, 1940). a female, f i r s t nod, next crouch, and then take the bowsprit poseJ This behaviour on her part created intense excitement among the attendant drake who....crowded close about her... (Brewster, 1911). Display has also been used as a welcome to newcomers. In one case, two individuals judged to be females....showed great excitement at the approach on the wing of a male....the two 'females' called quar-quar and held the head and extended neck s t i f f l y upwards at a steep angle, the head feathers raised, and one bobbed the head by repeated retractions of the neck, calling. (Boase, 1924). The Masthead i s described above when Gunn (1939) refers to "stabbing the zenith". In another passage Townsend (1910) describes i t as occurring i n response to the Head-Throw of the male: — ....the female, although usually passive, sometimes responds by protruding her head close to the water i n front, and then - bringing i t up so that i t also points to the zenith. Further than this I have not seen her go. He admits the possibility i n the spring that young males are involved 119 but states that he i s convinced that females do actually perform this act. Boase (1924) writes: — -The female, which had been watching the male with chin on the water, and neck retracted, replied by jerking head and neck i n line upwards at a steep angle, holding the f i n a l extended position for an instant, and returning to normal. These dis-plays were repeated several times. Although Gunn (1939) above says he has seen.lthe female kick up water, he denies that she has the Head-Throw. Boase (1924) says that he did not see the Head-Throw from the female. Phil-lips (1925) remarks that a "curious hollow groaMng note like owhh was heard from a female by Harper on the Athasbasca, which seemed to stimulate the male to throw back his head on to his rump". Brewster -likewise believes the female stimulates the males to display, by some action of hers. He writes: "she might occasionally single out and obviously encourage one of them by approaching him closely and bobbing her head up and down a few times. To this salu-tation he would immediately respond by a corresponding action before begin-ning his more elaborate performances again.", and as remarked earlier (quoting from Brewster, above) and later (under Head-Throw) the Bowsprit posture seems to stimulate the Head-Throw. Mr. Gerald Legge had two pairs and an immature female i n captivity. He i s reported i n Millais (1913) as writing: — At the same time she suddenly throws up her head and neck u n t i l the b i l l i s quite perpendicular, and at the same time she makes a loud c a l l , ('very loud and harsh 1) quite unpronounce-able, and quite distinct from that uttered by the male. She also kicks with both of her feet, but does not kick the water so high as the male....the kick i s done with both feet, one after the other, i n quick succession.... It i s usually after the young bird has chased her elders....the drake approaches his duck and does his f u l l show and kick, and the duck seems to follow i n imitation. I have, however, several times seen the old duck show when she has not been pursued by the young bird. 120 Millais (1913) also says that the female may advance towards her mate with "head and neck outstretched, and somewhat similar i n attitude to that of the male." and that "both sexes make use of a hoarse kraa-aa as a c a l l , the male being somewhat louder than the female....I have heard the female make a low note like wah-wah when talking to her young". Phil l i p s (1925) describes the c a l l of the female as a "loud guttural churr-ing noise not often repeated. I have heard her give vent to a squeaking ach or heck uttered sharply on suddenly being surprised on the water with her young brood. Again the harsh note of alarm may sound like the quaak of a surface feeder, very loud, but coarse and rough. Harper (M.S.) heard a note like cuk-cuk-cuk-cuk, quite different from the ordinary grating grrrk when four or five female were cir c l i n g through the woods hunting for nest-cavities." Jiving This i s the response of the female to courting activity by the males. It occurs on territory when a male rejoins his female, either when he re-turns from a boundary dispute, or when she rejoins him on other occasions. It thus appears to have the meaning "I belong to you". It occurs i n courting parties and i s expressed i n such a way that i t appears the female i s repelling some of the males, with the meaning "See here, I am mated to this male". In such situations Jiving may lead into a short rush with the beak open at another male than the preferred one. It need not be assumed that Jiving i s aggressive in motivation, however, since the female i s capable of aggressive threat i n the form of the Laying 121 Neck on Water posture of the male, and may even be capable of a "Gesture of Repulsion" from this position. Boase (1924) writes of the female making short runs at the male "with head and neck i n line extended horizontally— this i s , of course, an action common to many ducks". Jiving often occurs between a pair when another male i s threatening to attack a pair. In this movement in B.islandica the female lowers the head forwards but the head i s some way off the water. She then swings i t sharply from side to side, f i r s t one way, then another, i n broad sweeps of considerable amplitude, so that she faces right around towards other males i n her wake. This usually occurs when the female i s following a male swimming ahead of her i n a courting party. In a courting party the extreme lateral position of the Jive comes, very often, to be directed at another male than the one she prefers. If a female has no chosen partner she may Jive at a l l follow-ing males. When the female i s behind the male of her choice he is usually i n the Head-Up position, Head-Turning, rather than Rotary Pumping. On territory she may also do i t as she comes to meet him, and then swing into line beside or behind him, while he makes Rotary Pumping movements. If a male approaches too close she may rush at him with neck stretched forward and beak open. In extreme form the rear of her body may swing out of line from the direction of movement, and her body may come side on to the following male. This may, I think, lead into a jab towards him. But this i s exceptional. However i t raises the question of the signal inten-tion of the movement, and i t s orientation. After Jiving the female T a i l -Wags emphatically. Once when a female was disturbed from the shore by the observer she swam out towards her male and Jived as she approached him. 122 t A female who had three times refused coition to a courting male, began Jiving, behind her male, after they had both dived and he had begun Rotary Pumping after surfacing. Shortly afterwards she was driven away by another male. In one very peculiar instance Jabbing was observed performed, by both birds, at the rump of the other. It seems that the birds were not paired and that an element of aggressiveness took charge i n an otherwise heated courtship sequence. It ran as follows: — Both birds dove and the female upon surfacing did the Upwards-Stretch movement and then Jiving (low croaking as she did so). The Jiving became so violent that she came alongside the male from behind and then faced away from him at the extreme of the outward movement. He turned at right angles to follow. Once she jabbed him i n the side of the rump during the Jiving. She as often turned towards him from i n front of him. He would jab then at her rump as she turned (and she once at his). Then he took up the Threat posture and dove twice at her. She had to scoot away. As he came up he once seized her rump feathersaandda flapping chase ensued over the water. He dove again and she flew. He followed and during this flight swerved up at her once. She avoided this and they circled together. He came back to his territory and she landed close inshore 300 yards away. On two occasions the Upward-Stretch with Head-Flick immediately preceded Jiving. This movement is found i n the Anatini, being the only display about which the common homology i n the two groups cannot be doubted. As discussed earlier Lorenz referred to i t as the "Inciting" movement (in.transl., 1951-52). It seems l i k e l y that light on both Jiving and Deep Jiving w i l l come from comparison of goldeneyes and surface-feeding ducks with other groups 123 e.g. scaups and eiders. McKinney (M.S.) cites Hoogerheide (1950) to the effect that i n Somateria mollissima "the two components of inciting are not fused into the one movement as they are i n many Dabbling Ducks. The duck stretches threateningly toward one bird and then c h i n - l i f t s repeatedly toward 'her own1 male." This i s a very important discovery. Sawyer (1928) writes "The extreme example i s a snaky movement of the neck as the head i s swept forwards and backwards, from extreme side to side of the bird, i n a generally horizontal plane". He illu s t r a t e s this i n o Figures 19 and 21-23 on Plate 3 of his paper. Mary Jackson (M.S., 1952) describes Jiving as follows: — The head is held erect and the b i l l describes an arc i n which the following positions may be identified: (l) B i l l horizontal, pointing straight ahead, neck relaxed; (2) B i l l pointed downward toward the water at an angle of 30 degrees, and toward the right, neck outstretched; (3) B i l l horizontal, pointing straight ahead, neck relaxed; (4) B i l l pointing downward toward water at an angle of 30 degrees, and toward the l e f t , neck outstretched; (5) B i l l horizontal, pointing straight ahead, neck relaxed. Thus the b i l l describes a horizontal angle of nearly 180 degrees, and a vertical angle of about 30 degrees. '124 Deep Jiving I have no f i e l d data on this movement in B.islandica though i t probably occurs. However Figure 8 shows this movement i n B.clangula taken from walcott and Dane's film. It appears to resemble the second half of the Head-Throw-Kick Position 2, i n that the neck i s sunk below the water, beak pointed steeply upwards, and there i s a rounding up of the,rear of the body and a considerable paddling (or kick) with the feet. It seems to be directed at one particular male, and appears to be threatening i n function (though females can also assume the Laying Neck on Water posture, especially towards other females when they have broods). Gunn (1939) has described Deep Jiving i n B.clangula as follows: — She has one curious l i t t l e performance which i s never seen i n the male, and which I may describe by an extract from my notes. "A movement by the duck when attracting the attention of the drake; usually associated with periods of 'lying along' the surface. The head i s slowly and completely submerged and then carried for-wards i n a curve having an upward direction, so as to reappear with the b i l l pointing upwards at an angle of 45 degrees. As soon as the eyes are clear of the water the movement i s arrested, and this position of the head i s retained while the whole of the neck i s out of sight. At the same time the stern is well elev-ated i n a humped-up way, the heels being exposed though the t a i l i s not l i f t e d from the water. The bird then i s more or less i n the position taken by a Great Crested Grebe chick which has been bullied by a stronger brother, and, after making his escape by diving, rises and peeps to see i f 'the coast i s clear'. This pose forms part of an active display by the female bird; and, as such, differs from the common attitude of solicitation, when she half sinks her body and l i e s extended f l a t along the surface of the water." Briiggemann (1876) wrote (in transl.) of B.clangula that: — The female bends i t s e l f completely f l a t on the water while i t lays the head down on the same. From time to time i t bends the neck contortion-wise downwards and points the spread out t a i l feathers downwards into the water, while i t makes i t s whining .(Winselnde) or purring (Knurrende) voice, heard i n reply to the c a l l of the male. 125 It seems possible that this is a description of Deep Jiving. Boase (1924) also believed that a call was uttered by B.clangula in this pose !jswam around him with chin on the water and neck almost submerged, the body being sunk low; they appeared to be calling, but wind made the sound in-audible." 126 FIGURE 8 . B.clangula (American population) DEEP JIVING From a fi l m by Charles Walcott and Benjamin Dane. Read from right to l e f t . The neck is sunk below the water, and there are signs of a kick owing to the elevation of the rear of the body. The movement resembles the late phases of one form of the Head-Throw-Kick. 24 frames/second. Shutter Speed l / 4 8 t h second. 12$ THE MALE Head-Up Neck Withdrawn (B.islandica) Field Descriptions The main posture consists of the neck being stretched vertically up-wards and maintained i n this position. The beak is held horizontal (Figures 9 & 10). The male gives a clicking sound from his b i l l which i s opening and closing a l l the time. The noise i s presumably caused by contact of the mandibles, and thereby differs from the noise made i n the Laying Neck on Water pose which occurs during Rotary Pumping, and i n which the b i l l i s kept open. The head is turned slightly from side to side while i t i s held up i n this position (Fig.10): "there i s no movement i n this position except a 'startled' sideways twisting of the head from right to l e f t and vice versa". While i n this position the neck i s frequently drawn back quickly (Fig. 9., Frames 10-11), and shortened, so that i t is brought between the shoulders. It i s held here for a moment, and then the head i s elevated again more slowly (Fig. 9., Frames 18-22) and the neck straightened (Fig. 9). It i s of interest that i n 1955 I did not recognise this as a discrete move-ment, although i t was described i n my f i e l d notes. However i n 1956 I ob-served the withdrawing of the neck (or bending of the neck) i n the Redhead (Aythya americana) and shortly afterwards recognised i t for what i t i s i n B.islandica. The Neck Withdrawn position i s only assumed momentarily i n the course of the Head-Up display. My notes record that : "they would also slowly -stretch and lower their necks as i n Rotary Pumping. There seem to be two components here — a slow forward pump and a vertical courtship pumping 129 proper.". In another case I recorded (tape): "began to pump at the female (high pumps with very l i t t l e amplitude). In this position the high position resembles the Expectancy Look. In the high position the beak i s sometimes opened.. .continually opened and shut...". Sometimes the withdrawing of the neck i s not a fast jerk but a slower movement. There i s an angled aspect to this movement on such occasions— a lowering of the up-stretched head into the shoulders and raising i t again to the head-stretched position. This low motivation movement sequence has considerable resemblance to Rotary Pumping, as well as to the Head-Up Neck-Withdrawn movement. It remains to be discussed what form low motivation Rotary Pumping may take (possibly i n the high-up pose). But i t seems that Rotary Pumping may intergrade into the Head-Up Neck Withdrawn posture-movement display. Juvenal males may perform the Head-Up Neck-Withdrawn position. The resemblance of the Head-Up position to the Expectancy Look pose has been mentioned earlier. It seems l i k e l y that the former has derived from the latter. However i t i s now a highly ritualised movement, frequently seen i n courting parties. Very often more than one bird may be seen doing i t at any one time. The Head-Turning which accompanies the Head-Up posi-tion, and the ticking noise, i s a constant feature and also adds likelihood to the suggestion that the Head-Up Neck-Withdrawn display originated i n the Expectancy Look. On one occasion a male realising that another male was directing a submerged attack upon him, began swimming towards his female with neck raised and (?) courting weakly (Rotary Pumping). A dis-tinct quacking was heard from his open b i l l "while i-n this position of 'alarm'", and i t seems that the position i s essentially one which increases 130 the visual f i e l d (head raised, and the turning of the head) of the alarmed bird. Whether this pose has any resemblance to the beginning of f l i g h t has not yet been analysed, but i t i s a fact that i f the attack i s pressed home, or i f the bird i s alarmed sufficiently the Expectancy Look position (or Head-Up position) i s followed by f l i g h t . We may look therefore for elements of escape motivation when i t i s observed i n a courting party, although mimetic stimulation seems to be one cause for i t s occurrence there. The Head-Up position i s maintained even when the female i s submerged. The movement also has resemblance to the Head-Turning which i s inter-posed between the Head-Throws of a male, when a female i s following a B.clangula drake. There i s a suggestion from one observation that the Head-Up position i s a releaser of Jiving, which commonly accompanies the Head-Turningjby a following female of B.clangula. Furthermore the neck i s withdrawn into the shoulder i n B.islandica no less than i t i s i n the Kick and this backward position of the head i s considered to be equivalent to the extreme head-on-rump position of B.clangula i n the Head-Throw-Kick. There may be a connection here therefore. On the other hand B.clangula has an identical pose to that of the Head-Up (Head-Raised Neck-Withdrawn) so that the Head-Throw cannot be said to be the exact equivalent of the Head-Up position. The Head-Throw i s evidently the equivalent to the Pseudo-Kick. However i n B.islandica the Head-Up (Neck-Withdrawn, the Pseudo-Kick and the Kick, are much more alike than are the Head-Raised (Neck-Withdrawn), Head-Throw and Head-Throw-Kick of B.clangula. Literature Description Sawyer (1928) illu s t r a t e s this pose. He writes "head bobbing and i 3 1 elaborate neck pumping and stretching are the most outstanding actions; these motions are grotesque and spectacular as, for example, i n the sudden upward thrust of the head," (See Figures 10-11 , Plate 3 i n his paper. He also says that "the drake often or usually gives a continued short quack i n throwing the head back and while i n the attitude which ensues", which i s the Head-Up position. Mary F. Jackson described (M.S. 1952), a movement which she called "Head-wagging": — The head i s held level, with the b i l l horizontal, and the neck stretched out (means up^ as far as possible. The head i s then moved backward and forward (means sideways) through an angle of about 135 degrees, with the slightest momentary pause at the end of each "wag", so that the entire pattern appears to be slightly jerky. The body i s carried high on the water throughout. 132 o Head-Raised Neck-Withdrawn (B.clangula) Film Description This movement resembles the Head-Up position of B.islandica almost exactly (compare Figs. 11 & 12 with 9 & 1 0 ) . Curiously enough i t has not been specifically named before, nor has i t been seen very often. The head appears to be held a l i t t l e further backwards i n the Head-Up position than i n B.islandica. The contour of the head and front of the neck and breast from b i l l - t i p to the breast at the water-line traverses a smooth half-S. The bird gives the appearance of leaning back a l i t t l e (like a man looking down his nose) i. e . the b i l l i s t i l t e d a l i t t l e upwards from the horizontal. In this position the beak is opened and closed (Fig. 11,Frames 35-37 , and Fig. 12,Frames 2-4 and 29-32). The position i s assumed from the Normal resting position ( a l i t t l e forward and the neck a l i t t l e more stretched than usual), by the head being withdrawn into the shoulder, and then reflected forwards and upwards off i t , thus raising the angle of the b i l l a l i t t l e and extending the neck. From this extreme position the head i s periodically withdrawn onto the back. The movement i s rapid (Fig. 11,Frames 11-13 , Fig. 12, Frames 13 -16) . The head l i e s much further down the back than i n B.islandica and beak t i l t e d upwards resembling the Head-Throw-Kick (Position 2) instead of horizontal as i n Head-Up position. The raising of the head afterwards appears as though i t may be as rapid. The beak i s opened and closed 2-3 times between each withdrawing of the neck. There i s Head-Turning (Fig.12) as i n B.islandica. It appears from scrutiny of Walcott & Dane's fi l m that i f the female 133 is behind him, the male w i l l swing his body from side to side, or w i l l turn his head from side to side, while calling. In this position i s appears that the neck is less frequently withdrawn—a comparable position to that found i n B.islandica. i n which i t is withdrawn rather less often, and i n which the head may be turned from side to side more frequently. Most usually the movement is made towards a female from behind her, but i f the female turns to go with him, he then turns to lead and i s then i n front of her and facing away from her. Then Head-Turning replaces the Neck-Withdrawing rather more. This could be because the Neck-Withdrawn position i s less readily seen i f done i n front of the female, ^ead-Turning i s then more visible and (?) the white face patch becomes v i s i b l e on each side alternately (this i s not the case however i n Post-Coition Steaming i n which Head-Turning also occurs). When swimming behind the female the male-may be a: l i t t l e at a slant away from her so that the Neck-Withdrawn position i s visible to her. Orientation i s indicated here, to be important i n the genesis,moulding, and selection of movements. If each sex i n each of two species behaves and i s responsive to subtly different orientation situations, and signal-exposure i s dependent on these we can explain how the Neck-Withdrawn position has become more frequent i n B.clangula and Head-Turning more fre-quent i n B.islandica. Further analysis alone can t e l l us whether these f i r s t impressions are i n fact correct. Meantime they are worth recording. Literature Description Both literature descriptions (Boase, 1924 and Bernhardt, 1940) say that the head feathers are puffed out. The best description i s found i n 134 Bernhardt (1940) i n which the orientation i s indicated and the exact f o l -lowing of the female (turn by turn) by one or more males i s emphasised. Also the withdrawing of the neck is indicated by the phrase "continually stretching their necks": — Seven males swam around a female continuously stretching their necks; they swam back and forth with elevated Crests; i f the female changed her direction, the males did so also, like commandos. Each i s determined to outdo the others... T his description describes exactly a similar action observed i n the Lesser Scaup (Aythya a f f i n i s ) by myself, and I think I have seen i t i n Melanitta sp. as'welll. It would be possible to argue that the description was of the Bowsprit pose, but I doubt i f i t were, for Boase (1924) has also apparently described the Head-Raised position: — the head feathers of the male were raised, making the head appear large; the head was held rather higher than i n the normal attitude, and, swimming near the female, the males displayed as follows.... and he goes on to describe the Head-Throw, which may evidently be intro-duced into a sequence of Head-Raised (Neck-Withdrawn) positions. Head-Turning i s not mentioned however. 135 FIGURES 9 & 10. B.islandica (Icelandic population) HEAD-UP WITH NECK-WITHDRAWN From a film taken by Dr. Frank McKinney at Slimbridge, England Figure 9 . Side View Frame 1. Frames 3 & 4 Frames 6-9 Frames 10 & 11 Frames 11-18 Frames 18-22 Frames 22-28 Frame 28 Head-Up position Beak opened Beak closed •Neck being withdrawn Neck Withdrawn position Head being raised to Head-Up position Head-Up position Beak opened Figure 10. Front View with Head-Turning Frame 1 Neck Withdrawn position Frames 5-14 Head being raised to Head-Up position Frame 15 Head-Turn to right Frame 16 Beak opened Frame 19 Beak closing Frame 25 Beak opened when head directed forwards Frame 27 Beak closing Frame 33 Beak opened and Head-Turn to l e f t Frame 35 Beak closing. Head turned f u l l y to l e f t 16 frames/second. Shutter Speed l/32nd second. 138 FIGURES 11 & 12. B.clangula (American population) HEAD-RAISED WITH NECK WITHDRAWN From a film by Charles Walcott and Benjamin Dane Figure 11. Side View Frames 1-10. Frames 11-13 Frames 13-15 Frames 16-18 Frames 19-38 Frames 35-37 Head-Rai sed Withdrawal of Neck Neck-Withdrawn Head-being raised to Head-Raised position Head-Raised position Beak Opened (In Frames 25-38, the body of the bird was hidden by a wave), 24 frames/second. Shutter Speed l/48th second. Figure 12. Rear View with Head-Turning i Frames 2-4 Beak open Frames 14-18 Neck-Withdrawn Frames 29-32 Beak open At the beginning and end of the Neck-Withdrawing movement the head faces away from the observer. The beginning and end of this sequence i s therefore with Head turned to the l e f t . ? 24 frames/second and Shutter Speed of l/48th second. m Pseudo-Kick and Kick (B.islandica) In the f i r s t year of my study I was unable to separate these movements. Both movements are very rapid and hard to describe, and i n my recordings they appear merely by name whenever they occur. I have not personally re-corded either movement on film, but have been able to make drawings of both from a film of Icelandic birds taken by Frank McKinney at Slimbridge. These two movements are not as distinct, or as extreme, as are the Head-Throw and the Head-Throw-Kick of B.clangula which are quite different from each other, and take a much more r i g i d and amplified form than do either of the equivalent movements i n B.islandica. The Pseudo-Kick In the Pseudo-Kick the feet are kicked but do not break surface. In one place I described i t as a " l i f t i n g of the head back and up not quite onto the shoulders and with beak held almost ve r t i c a l l y upwards. It i s a rapid movement both up and down and the upper position i s not held. During the movement the b i r d - c a l l s — l i k e the quack of a duck". The film (Fig. 13) shows that the head is l i f t e d from the resting position upwards, then swung backwards u n t i l i t i s being held s t i f f l y over the back at an angle of 45 degrees, beak being held- not quite vertical. The-beak i s then opened and evidently a stroke of the legs occurs for the breast sinks into the water. A Head-Flick follows the movement as the head moves forward and then back into the resting position. A point of interest brought out by the film i s that i n the Pseudo-Kick (but not i n the Kick proper) the rear (raised) position of the head, the t i p of the back, and the head as a whole, are t i l t e d over sideways towards the female, giving the male a distorted 142 appearance, as seen from behind or i n front (Figs. 15 & 16a). There • are slight indications that the beak i s opened twice, i . e . before the.kick (Fig.. 13, Frame 4) as well as after i t . This may also occur i n B.clangula i n the Head-Throw-Kick (Fig. 20, Frame 26, and Fig. 21, Frame 11). The Pseudo-Kick thus resembles the Meek Stretch of Aythya vallisneria (Hochbaum, 1944), the Masthead of B.clangula, and i n weak form might resemble the Water-Flip of both B.islandica and B.clangula. In B.clangula the Head-Throw (without a splashing of water behind) i s distinct from the Head-Throw-Kick. In B.islandica the Pseudo-Kick seems to be different from the Kick, and i s probably homologous with the Head-Throw of B.clangula. The Kick In the Kick the feet are convulsively thrust backwards, as i n B.clan- gula, i n such a way that a shower of spray i s produced behind the bird. The film (Figs. 14 and 16b) shows that the head i s thrown back and up i n the same way as i n the Pseudo-Kick, but i s dropped onto t-he middle of the back between the wings (as compared with the rump in B.clangula). This takes place as the bird sinks into the water, as the propulsive movement of the foot takes place. I once noted that: "The head is thrown back as though merely by the force of the jerk of the feet", which in B.islandica occurs when the head is s t i l l elevated (Fig. 14, Frames 9-11). In B.clan-gula the head i s not, however, thrown back by the force of the jerk, for the head is l a i d on the rump before the feet are even brought forward (Fig. 21, Frames 6 & 7). The sinking action of B.isla.ndica i s also quite distinctive, for i n B.clangula the rear of the body i s elevated by the kick of the feet and the neck- is thrown forward, often to sink well below 143 the water. The actions d i f f e r i n timing too, for the beak i n B.islandica i s opening at the same time as the propulsive movement i s taking place (Fig. 14, Frames 10-15) while i n B.clangula the kick occurs before the beak opens. The head does not appear to be t i l t e d over sideways i n the Kick as i t i s i n the Pseudo-Kick. The movement ends with the head being thrown forward with a slight f l i c k sideways (Fig. 16b, Frames 13 -15) . Mary F. Jackson (M.S., 1952) managed to compile a f u l l e r description of the Kick i n the f i e l d , than I was able: — The head i s thrown rapidly up and back u n t i l the back of the crown and occiput touch the back between the wings, and the b i l l i s vertical, pointing upwards, with the mandibles slightly parted. The throat at this time i s somewhat i n -flated and swollen...As the head is thrown back, the bird utters a short, soft, somewhat gutteral note...The head i s immediately returned to i t s normal position...At the same time the bird usually kicks back vigorously with both feet, raising a splash of water several inches above the surface. The Kick i s clearly homologous with the Head-Throw-Kick _of B. clangula. Literature Descriptions: Sawyer (1928) wrote that "a frequent act of the drake i s the backward kick which sends a spurt of water backward and upward i n the wake of the swimming bird" (Fig. 2 on P1 .3-in his paper). Munro (1918) mentions the backward kick. A photograph by Phillipa Talbot-Ponsonby i s reproduced i n the 4 th Annual Report of the Severn Wildfowl Trust (1950-51) i n which the opening of the beak (and production of the call) i s seen to be contempor-aneous with the kick of the feet. This i s not the case i n B.clangula, i n which the c a l l occurs when the head i s being raised from the forwards position after the kick. We also see the high position of the head above 144 the back which i s the most posterior position of the Pseudo-Kick i n this specie s• P h illips (1925) points out the difference between the Head-Throw-Kick of B.clangula and the Kick of B.islandica. He wrote: "The principal difference i s that the whole performance i n the Barrow's i s less exagerrated and the male does not throw the head nearly so far back". He then quotes from Brooks (1920) who wrote: "The most common form of display i n the drake i s the 'swallowing1 or 'gulping' action, this may or may not be followed by a kick which throws water up behind...Once i n November I watched a lone drake going through the whole performance by himself—water-kick and a l l . " Perhaps the Pseudo-Kick should, by priority, be called the "Swallow-ing Action", but i t would tend to imply a form of drinking which i t i s not. Munro (1939) described the Pseudo-Kick and Kick as follows: — WJien displaying the male throws back the head with a quick jerk u n t i l the b i l l points backward, or merely stretches the neck upward to i t s greatest extent, or makes a succession of quick bows. Usually the head movements are accompanied by a quick backward thrust of the foot that may send spray flying. These actions very i n intensity and have numerous modifications. So like Brooks, Munro also did not distinguish the Pseudo-Kick (or "Swallow-ing") display from the Kick display. There can be l i t t l e doubt that they are elaborations of a single primitive display, as mentioned under B.clangula. In B.clangula the two movements are rather more distinct than i n B.islandica, though i n both the Kick i s the essential distinguishing feature. Origin and Circumstance The origin of these movements i s completely uncertain i n either species. The Head-Throw of B.clangula i s found i n many other ducks (Aythyini and 145 and Mergus). The opening of the beak appears to be rare, and i s only found in l i k e circumstances i n the Head-Up position and i n Threat. The back-throwing of the head is found i n the Neck-Withdrawn position from a resting or raised -position of the head. In both the Masthead of B.clangula and the Water-Flip of both species the head is l i f t e d back from a low position, with the beak extended forwards over the water. A kick of the feet i s only otherwise found i n the Deep Jiving of the female (Fig. 8). In this posi-tion the neck i s below the water, as i t i s i n one form of the Head-Throw-Kick of B.clangula. The movement i s found i n four differing kinds of situations: — 1). Courting Parties, 2). when another drake f l i e s over one on the water, or lands near him, 3)- i n Isolation, often after landing on the water, 4). i n the Coition Sequence. (1) . In courting-parties on lakes i n the Interior, the Kick i s seen more frequently from f i r s t year, juvenal, males than from any adult males (who may however perform i t , or mimic juvenal birds as they do i t ) . The juvenals are unmated, of course (and i n more female-like than male-like plumage). When they perform the act i t i s often immediately followed by another bird doing i t (often an adult). Some sort of social stimulation appears to be involved, as i t i s with other displays i n Bucephala. A juvenile male following a female did the Kick and then an Adult male behind her did i t as well. Shortly after the juvenal was Rotary Pumping and scooted through the water for a second ("probably an incipient Kick"), and then did another Kick. (2) . When a drake f l i e s over, or past, a drake on his territory the latter sometimes does the Kick. At other times they may assume the Threat 146 posture. It is not known yet i f a threat position i s assumed before the Kick i s performed. On one occasion two juvenals were seen to do the Kick as an adult male flew low over them. Since they were neither mated nor on territory i t seems that.the Kick display does not have threat mot-ivation i n these circumstances. I have a number of records of the Kick being performed either just after a drake lands on the water, either by himself or another drake nearby. Two males and one female landed on the lake, and the second male began Rotary Pumping vigorously. A third male landed, threatened the second male and performed the Kick display at him. Then i t swam away in the Expectancy Look position. Two males were flying together and landed a l i t t l e way apart. One then dove at the other which flew, landed not far away and almost at once did the Water Flip and the Kick. Soon afterwards another male landed near a pair, the male of which dove at him. The intruder flew, landed and performed the Kick. The tape recording continues: "The owner male dove at him again and he fLew—to the far end of the lake and landed 15-20 yards from a pair. The male of the pair went into Threat. The intruder did the Kick again, nothing else and then dove" (Cummings Lake, May 16, 1956) . One time a male did the Kick as an intruding female landed on his territory. Another time when a neighbouring-male flew an adu.lt male nearby did the Kick twice. The stimulus of a bird flying, taking off, or landing i s most clearly illustrated perhaps by an occasion when an adult male after threatening a group of goldeneye did the Kick display just as 147 a Baldpate (Anas americana) which rose from the water nearby began to c a l l , as they do i n f l i g h t . (3) . Once an adult male landed alone on the lake and did three or four Pseudo-Kicks i n a couple of minutes, with some Water-Flips, and preening. (4) . The Kickj or perhaps more li k e l y the Pseudo-Kick with i t s head-orientation towards the female, may be seen i n the preliminaries of the Coition Sequence. Unmated birds perform the Kick most commonly, adults often i n reply to a juvenal. The movement i s performed i n courting parties under these circumstances. It may occur in isolated birds also however, but i s as often seen in birds as they land either alone, or close to other drakes or pairs. It often follows an aggressive incident i n which the performer has beaten a retreat. The indications seem to be less that i t i s a form of aggression, than that the display has a "greeting", "advertising" or "appeasement" significance. 148 Head-Throw (B.clangula) Even Brewster (1911) appeared unable or unwilling to distinguish the Head-Throw from the Head-Throw-Kick. They are however distinct stereo-typed movements and probably have different signal functions. Brewster's name for them "Folded Duck-Skin Posen no longer has any valency i n the language, and has been replaced by more universal terms, descriptive of the actions. Field Description I have l i t t l e experience as yet, of this movement. I have watched i t , however, at 73-Mile Lake, Cariboo Highway, during the spring migration of 1956. A courting party was observed containing both males and females. In the Head-Throw the head was thrown back onto the rump, often so that the nape feathers dipped behind the t a i l into the water, the chest being raised and the beak held v e r t i c a l or backwards some 10 degrees. In this extreme position the beak was opened and a c a l l emitted and then, after a pause, the head was rapidly brought forward to the normal position again. The legs were held out sideways. The action was repeated over and over again. ^receding this convulsive movement, or alternating with repetitions of i t , the head was turned very rapidly, from side to side, showing each face patch i n turn. This i s nowhere mentioned i n the literature, ^ t re-sembles the Head-Turning that occurs i n the Head-Up position, and i n the post-coition display. 149 Film Description Walcott & Dane's film shows that the Head-Throw begins with a vert-i c a l l i f t i n g of the breast, and stretching of the neck, so that the head i s elevated (Figure 17). The neck is then kinked, and the head rotated, so that the head and neck are brought back and l i e on the rump. The move-ment resembles Position 1 of the Head-Throw-Kick, but i s quite unlike the more common Position 2 of the Head-Ahrow-Kick. The head i s held for about one second (22 frames) i n this position, and almost as soon as the position i s assumed the beak i s opened and is held open for 5/6th second (20 frames]), closing as the head starts forward again. This holding of this position i s much longer (about 3 times as long) as i t i s i n the Head-Throw-Kick (Posi-tion 1). The Head-Flick occurs before the head reaches the normal resting position. In an extreme form the Head-Throw is continued, and the head i s carried forward into a Forward Position resembling the Bowsprit position (Figure 19) but there i s not usually any Bowsprit Pumping from this position. In one instance on the film of the sequence Head-Throw Forward Position Bowsprit Pumping Head-Throw Head-Flick the ^ ead-Flick was delayed to the end, and did not come i n i t s usual position at the end of f i r s t Head-Throw. This i s evidence that the Head-Flick i s an indicator of the end of this motor element. The Forward Position i s therefore a normal variant at the end of the Head-Throw proper. The Head-Turning between Head-Throws (Figure -18) may be very energetic, and i s especially frequent when the male i s swimming i n front of the female, leading her. The face-patches may perhaps be exposed alternately. As seen 150 from i n front there is a very sudden break between the extreme posture of the Head-Throw (when the head i s on the rump and the t i p of the beak alone shows black over a mountain of white chest), and the clearly exposed side-ways fli c k i n g of the black head over the white breast as the bird returns to the normal resting position (Figure 18). This same Head-Flick would also be very visible from behind. Literature Descriptions This movement i s the one most commonly described i n the literature as "the display" of the goldeneye (with the Head-Throw-Kick as a variant). It has been described by Bruggemann (1876), Gilpin (1880), Townsend (1910), Brewster (1911), Heinroth (1911), Boase (1924), Phillips (1925), Munro & Clemens (1931), Munro (1939), Bernhardt (1940), and Johnsgard (1955). It i s illustrated i n Brewster (1911—-in which the head i s not placed far enough back, and the beak i s turned backwards at too great an angle), i n Scott and Boyd (1954, P-6.,—again not i n the most extreme form), Phillips (1925,—tail shown as spread which i s probably incorrect), Bern-hardt (1940) and Johnsgard (1955). Bruggemann (1876) was the f i r s t to describe the Head-Throw (in transl.): The male ruffles the crest up, snaps i t suddenly onto the back and emits a loud sound, almost lik e quaatsch, while i n the mean time i t makes one energetic thrust movement with the rear end of the body and feet. Alternately i t lays the crest f l a t down and stretches one or the other wing wide out. This play usually lasts about five to ten minutes. Again and again i t is pointed out that the goldeneye has the most spectac-ular of a l l duck displays. Johnsgard (1955) writes that "the goldeneyes have developed the head-throw to i t s extreme", and compares this movement 151 of B.clangula with the similar movement found i n the couirtship of the mergansers and the Aythyini. Boase (1924) wrote "the male threw the head and neck back, the nape touching the t a i l , immediately recovering and jerking the head i n line with the extended neck held s t i f f l y v e r t i c a l " . Gunn (1939) says "his astonishing 'jack-knife 1 movements...when he flashes his fluffed-out head on to his back, holds i t there for a second, and then as quickly snaps i t forward again." Alford (1921) wrote: — The most acrobatic display that I have ever witnessedis that of the male Goldeneye....With a t e r r i f i c j e r k — s u f f i c i e n t , one would suppose, to dislocate i t s neck—the head is thrust upwards to the f u l l extent of the neck, with the b i l l closed, then bent sharply backwards u n t i l the tip-df the mandible touches the t a i l . The next second, and with equal force, head and neck are jerked forward again to the normal position, with b i l l agape, the bird uttering at the same moment a loud squeak, not unlike the sound produced by the creaking of rusty hinges, and splashing the water behind i t by the force of the convulsion. The whole performance occupies but two seconds. Munro aansl Siemens (1931) description gives aaa idea of the marionette-l i k e quality of the movement: "First one male then another would jerk back the neck convulsively as i f pulled by a wire, u n t i l the neck was lying on the back with the b i l l pointing up or ©-en backward. Then the same convulsive jerk would bring head and neck back to normal position." The rearmost position of the Head-Throw i s more extreme i n B.clangula than i n B.islandica and the head is held on the rump stati c a l l y longer in the Head-Throw than i n the Head-Throw-Kick. Townsend (1910) wrote: "Sometimes the head i s held on the rump for several seconds before i t i s snapped into place". Brewster (1911) wrote: "The l i v i n g bird would ordinarily remain i n the attitude just described from half a second to a 152 f u l l second or perhaps two seconds, but rarely longer than that". Heinroth (1911) wrote that during this action the head i s made more noticeable by spreading of the feathers of the head. In the Head-Throw-Kick this has been noticed more often. Brewster's (1911) description i s the most complete: — Back thrust of head and neck ending i n the folded duck- skin posture. Sometimes made from the mast-head posture but far oftener directly from the normal position, by a single uninterrupted upward and backward swing of the head and neck; this movement was so swift and sudden that I often failed to follow it...At i t s termination the neck lay extended along the back and so very f l a t and close that i t was scarce not-iceable. The head showed conspicuously enough, resting apparently on the occiput at a point anywhere between the middle of the back and the rump, with widely-parted mandibles pointing straight upward...At the close of this brief period of inaction the head and neck would swing forward, usually less rapidly and abruptly than when carried backward, some-times pausing for a moment when the mast-head posture might be taken, but, as a rule, continuing to move without decided halt u n t i l the normal position was resumed. Townsend (1910) also apparently believed that the Head-Throw could be assumed from a "Masthead" pose (breast elevated, neck stretched upwards, c a l l emitted). After describing this he continued "the head is then quickly snapped back u n t i l the occiput touches the rump, whence i t i s brought for-ward again with a jerk to the normal position." Evidently Brewster was doubtful that the Masthead was often a part of this display, and since he did not distinguish the Head-Throw completely from the Head-Throw-Kick we may assume that he only saw the "Masthead" pose only on occasions when the kick of the feet was present. It i s true the head, as i t goes back, i s l i f t e d off-the back and then placed down onto the rump, but no c a l l i s emitted until' the head is i n the latter position. 153 Most usually the Head-Throw ends with the head being returned to the normal s i t t i n g position. In some instances the head i s raised off the rump with the neck stretched and s t i f f , so that as i t rises to the vertical the Masthead pose i s assumed. This i s however only to be mistaken for the true Masthead display i f a static view of the most upright position i s seen, for the head i s not poised here, but continues to swing on forwards into a "Bowsprit" pose. But here again, since oscillations of the head do not seem to occur on these occasions, the forward position i s not to be confused with the true Bowsprit display (which i t only resembles i n a static view). Boase (1924) writes of this relaxation from the head on rump position "immediately recovering and jerking the head i n line with the extended head held s t i f f l y v ertical". He talks of a c a l l urr i n this last position, but was probably hearing only the c a l l accompanying the earlier phase of the Head-Throw. In another instance the same author wrote as follows: — The display began as before with the upward extension of the neck and the swing back so that the nape touched the t a i l , but, held there an instant and instead of the up-thrust of the rear of the body (Head-Throw-Kick), the head and neck were jerked up i n line vertically, and a c a l l quee-reek uttered, • then a relapse to normal. Boase writes that this display was repeated several times, but as thepe i s no mention of a kicking motion i t seems evident he i s describing the For-ward Position following the Head-Throw. Johnsgard (1955) is clearest about the Forward Position. "The head is then rapidly brought forward to i t s normal position, or perhaps more com-monly to about a forty-five degree angle. The bird holds this position, with neck extended, for several seconds". 154 "Stretching of the trachea of syrinx" We need to discuss the form of the c a l l , and the supposed origin of the Head-Throw i n the necessity to stretch the trachea or syrinx to produce the c a l l . A c a l l i s emitted from the open b i l l when the head i s l a i d on the ramp and as the head i s moving forward again. Brewster ( I 9 l l ) was unable to decide exactly when i t occurred. Both he and Townsend (1910) claimed that they had seen the movement unaccompanied by any sound but this seems improbable. The c a l l i s variously described. H einroth writes (I911.transl.): "While he returns the head into the normal posture one,hears a f a i r l y loud •creaking note'.i! is Bernhardt,; (1940).Jtalks.-Of a 'creaking note' also (Knirrlaut). Johnsgard (19-55) after describing the submergence of the posterior half of the-body' wrote: "at the same time they utter a double note, zzee-at, which i n quality reminds me of the cry of a Nighthawk (Chordeiles minor)". Townsend (1923) talks of "a harsh and rasping double note, vibrating and searching i n character and expressed by the syllables zzee-at." Heinroth (1911) discussed the contorted position of the Head-Throw in relation to the production of the c a l l that i s associated with i t . He wrote as follows (transl.): — Obviously this peculiar head and neck movement i s necessary for the production of the courtship c a l l , which i s demonstrated, i f by nothing else, by the exceptional structure of the windpipe! It should again be mentioned here that the trachea and the bulla ossea of the drake Clangula is built completely differently from that of the other Fuligulinae; i t reminds one very strongly of the Merglnae. Lorenz (1941, transl. 1952) perpetuates this explanation. After describing the exaggerated 'burping' movement of the Garganey (Anas querquedula)— a 155 form of 'head-throw' he writes: "The similarity with Bucephala certainly rests upon convergence, which has arisen from the necessity, found i n both forms, of stretching the bone-drum to a high degree". But he has not expanded this theme, and there seems to be considerable evidence to show that i t i s not correct. The question i s whether the c a l l developed f i r s t and forced the exaggeration of a less extreme movement, or whether (as seems more likely) tjne movement developed f i r s t and the c a l l was d i f f e r -entiated later. The Head-Throw— o r something resembling i t i n actual body form—is found not only i n B.clangula but also i n Mergus and even i n Aythya vallisneria (Fig.2 ) . Allen's account of the Canvasback (in Bent, 1923) reads as follows: The c a l l consists of three syllables, ick, ick, cooo, with a l i t t l e interval between the second and third. When the f i r s t two syllables are being produced the bird opens his b i l l s lightly and then with considerable force appears to inhale quickly, jerking his b i l l as he does so. ^t appears as though this sudden inhalation abruptly closes the glottis so as to produce the two rather high-pitched, sharp, quick, ick, ick notes.. Accompanying these notes the back of the neck swells and the feathers rise as though a gulp of air were being swallowed. Immediately, however, i t seems as though exhalation occurred with the b i l l closed, accompanied by a low cooo like a muffled baric or distant moo of a cow... .Accompanying this note the chin- swells out for an instant with a curious swelling about the size of an ordinary marble. Very frequently this note was accompanied by the head-throwing performance, already referred to, the ick, ick notes being given when the head was thrown back, and the cooo when the head was brought forward again, the swelling on the chin being noticeable as the head assumed the normal position. Hochbaum (1944) called this movement i n the Canvasback the "Head-Throw". He writes, significantly for our purpose, that the ick ick notes do not carry for any great distance. The cooo note -.hioh i s loudery'an asthmatic groan which, on a s t i l l day, may be heard half a mile or more". when heard 156 close-to i t appears as a s p l i t note, cu-oo. When many birds are court-ing at a distance i t "seems soft and dovelike". He points out further-more, that "This note, i n part, or i n f u l l , i s given frequently without the head-throw. When thus offered i t i s accompanied by a short, forward thrust of the head, while the.neck is arched s l i g h t l y backward. A con-spicuous swelling appears on the chin the moment the sound i s uttered." Thus the' fact that the Head-Throw i s identical i n the Canvasback and B.clangula (and also i n Redheads ^ Aythya americana),and scaups, as well, according to A l l e n — i n Bent, 1923), but the notes are different, indicates that the notes as such have no part i n determining the movement for the notes are different and are determined solely by the form of the trachea and the force of a i r (and stops) behind i t . That the Head-Throw i s not a necessity for the appropriate a i r blast to be created i s ind-icated by Hochbaum's statement that the cooo c a l l (the ick, ick syllables are probably included i n his meaning) may be given i n the absence of the Head-Throw movement. The movement i s not therefore a necessity. Furthermore since the Head-Throw takes the same form i n B. clangula and A.vallisneria, but the calls differ, the movement must ( i f they are related genera) be older than the c a l l (in the same way that a movement i s older than the plumage characters displayed by i t ) . Presumably stretch-ing (or some.other action) of the "bone-drum" (syrinx presumably) i s part of the production of the note, since a swelling of the throat appears i n the extreme-position. But that the movement of the head onto the rump i s necessary to produce the required tensions or muscular actions on 157 the trachea (as suggested by Heinroth and Lorenz) seems improbable. The swelling of the neck i s found also i n Anas querquedula. Lorenz (transl. 1951-52)describes i t as follows: — Then while the head is brought back i n a sweeping curve to the normal position, a loud rattling noise i s belched out, during which the stretched windpipe springs, l i k e the cord of a bow, to the front of the neck, Li f t i n g a high fold of skin. This is invariably followed by drinking. The rejection of any phylogenetic relationship between the "Head-Throw" of the Garganey and Aythya, and Bucephala and Mergus, needs to be re-examined. There are three common features a), the movement, b). a c a l l being emitted c). the swelling of the neck as the windpipe springs forward. If the movement i s not a functional correlate of the c a l l , then the combination of a l l three items being-found i n two groups which are unrelated i s highly improbable. In Aythya the c a l l may be given in.a normal posture, but there are apparently no other elaborations. In the Mergini the Head-Throw-Kick appears to be an elaboration of the simpler Head-Throw display. In A.querquedula drinking follows the "Laying of the Head Backwards" (as Lorenz describes i t ) , but drinking does not follow i n Aythya or Bucephala. To explain such a situation i t i s only necessary to suggest that Aythya represents a central stock from which the Mergini have derived and retained the essential postures, while A.querquedula i s a species linking the Aythyini to the Anatini. In view of other considerations about the Surface-Feeding Ducks (Anatini) such an hypothesis is not impossible. 158 Head-Throw-Kick (B.clangula) -This movement has considerable resemblance to the Head-Throw and presumably has differentiated from i t . It is similar to the Kick of B. islandica but as with the Head-Throw i t is a neater movement than either of the equivalent postures in B.islandica. As noted by Brewster (1911) the head lies on the rump for a considerable portion of the whole move-ment in the Head-Throw, but i n the Head-Throw-Kick i t is held there only for a very short period. It i s this, with the propulsive movement of the feet, and the position of the head in the second half of the movement which (in B.clangula at least) distinguishes the Head-Throw proper from any version of the Head-Throw-Kick. Field Description I have seen this movement only once, during the spring migration of 1955* I described i t as consisting of a convulsive movement of the legs back, and the head up. It was like a sudden spasm, as i ts arresting (end) was as quick as i ts start. It seemed probable that the head was held on the rump for a shorter time than i t was in the Head-Throw. Film Description Walcott and Dane's film illustrates that this movement may take two rather different forms (though perhaps there are intergrades). In both the head is thrown back onto the rump, as in the Head-Throw, but i ts dura-tion there is curtailed. But the two forms start and end in different ways. The common element is the violent propulsive movement which gives the two movements their common name. 159 Position 1. (Figure 20). In this form the breast is elevated, and the head and neck are erect before being swung backwards and placed on the rump. This resembles the Head-Throw but the head remains on the rump only about one-third of the time. The rear of the body is then humped up as the feet are drawn forward ventrally, and there i s a downward pressure of the feet at the start of the kick. The wings show as a projecting spur beyond the rounded angle of the body. At this time the head i s brought forward and l i f t e d off the back at the end of the foot-stroke and the beak, which has been facing 10 degrees backwards when the head i s on the rump, remains i n this position. The head i s flung forward u n t i l the breast and front of the neck form an upright line, but since the beak i s s t i l l being held 10 degrees back, from the vertical, the position i s very striking. The throat appears very swollen (bulging forwards) and the beak is opened. Then the head i s l i f t e d back and up at about 10 degrees back from the vert i c a l . As the beak closes the head i s rotated, so that the beak returns to the horizontal, and the Head-Flick follows. The bird then relaxes into the Normal resting position. The whole sequence only takes 2 seconds. Position 2. (Figure 21). In this form the head i s not elevated, before being thrown back, and does not reach the rump. It is drawn between the shoulders more rapidly than i n Position 1, and remains on the back about half the time that i t does i n Position 1. The rear of the body i s humped and the tips of the wings appear as i n Position 1. The head passes forward u n t i l the neck i t -160 self i s right underwater. The head i s then l i f t e d back, with beak open, as i n Position 1 . Head-Flick and relaxation follow. Owing to the shorter time taken to assume the extreme backwards position.this sequence only took l«l/3 seconds (31 frames). The forward movement of the head, i n both forms, occurs as the body angle i s inclined decidedly forward, and the mass of the body i s behind the neck, i.e. forward propulsion. But the head should really be flung back-wards mechanically by the kick of the feet. It does not do so, and i t s for-ward passage implies that i t i s r i g i d l y s t i f f and that the protracting muscles are contracting strongly. Literature Descriptions It is evident that almost everyone has regarded this movement as an extreme form of the Head-Throw, and undoubtedly they have the same deriva-tion. The head-Throw-Kick now has a quite different form (and timing) from the Head-Throw even i n the portions of the sequence i n which the kick of the feet i s not the dominant motion. It is also evident (as was mentioned for B.islandica) that the behavioural context, in which i t occurs, differs somewhat from that i n which the Head-Throw occurs. The f i r s t part of the Head-Throw-Kick (to the head-on-rump position-^ may resemble the Head-Throw exactly. As the head passes forward the rear part of the body and t a i l are raised, as the feet are brought forward, and then thrust backwards. Spray i s then thrown up behind the bird. The Hand- book of British Birds quotes H.Wormald, and Phillips (1925) , as saying that the kick of the feet may occur as the head i s s t i l l moving backwards. 161 There i s doubt whether one of both feet are kicked and, i f both are i n -volved, whether they occur together or successively. The head is raised, after the kick, to aiforwards position and then l i f t e d up and back. The second half of the head movement (forwards) may however by d i f f e r -ent, the neck being brought forward and actually lowered into the water, b i l l pointing upwards (rather l i k e Deep Jiving i n the female), before being elevated as before. Accounts of this movement begin with Gilpin (1880) who talks of water being "thrown into mimic surf by their play". Millais (1913) wrote: "The third action is to drop the head...suddenly between the shoulders, the b i l l s t i l l pointing heavenwards, and to kick...so as to throw a jet of water high i n the a i r behind the bird." Brewster (1911) describes the move-ment as follows: — Just as the paaap was uttered—or perhaps a fraction of a second l a t e r — a slender shower or spurt of water...might often, be seen rising immediately behind the bird to a haght of one or two feet which, indeed, I saw plainly more than once, jerked out of the water..Owing to i t s force and direction the kick caused the hinder portions -of the body to sink perceptibly i n the water for an instant, after which these parts bobbed s t i l l more ob-viously upward... Phillips (1925) wrote that sometimes when the head was snapped back (he did not distinguish the Head-Throw from the Head-Throw-Kick) the bird "shoots i t s e l f forward by a rapid kick of both i t s bright red leggs, which results i n a spurt of water showing behind the bird for a distance of sev-eral feet". In fact, of course, the kick does not begin u n t i l the head i s already on the rump, but i f the showing of the red feet has signal value this i s a major point of distinction of the Head-Throw-Kick from the Head-162 Throw. It is possible also that the spray, per se, has extra-corporal signal value. Townsend (1910) likewise emphasised the colour of the legs. He wrote "As the head i s returned to i t s place the bird often springs for-ward, kicking the water i n a spurt out behind, and displaying l i k e a flash of flame the orange-coloured legs." Townsend furthermore adds that there are "many variations of this curious action" and goes on to delimit the Head-Throw proper and a movement which reads as though i t were the Masthead, or the elevated pose of the head at the beginning (of Position l ) or the end of both forms of the Head-Throw-Kick. Phillips (quoting Brewster, 1911 and Boase, 1924) also suggests a Masthead-like posture. Boase (1924) explains that the movement starts from the Normal position, with neck retracted, and how the neck i s "suddenly extended upwards, the head remaining normal to the neck, a back-throw of the head and neck so that the nape touched the t a i l , the neck being arched over the back." The Masthead-like pose at the end of the Head-Throw-Kick i s described by Johnsgard (1955) who wrote: "A second typical action i s a partial head-throw, where the head is rapidly jerked upward to a position directly perpendicular to the water,usually accompanied by a frenzied kick that splashes water far behind". This reads like a description of the Kick of B.islandica, but i t i s also a description of the elevated pose assumed at the end of the Head-Throw-Kick. The beak i s not opened as the head i s riased and goes back, but i t i s opened i n the elevated pose at the end of the movement. Indeed Brewster's description of the Masthead pose probably applied to this for he said the beak was open i n i t . As explained later the name has been reapplied to a similar pose (but different display). 163 Boase (1924) describes the Position 2 as follows: — followed by an up-thrust of the t a i l , the bird resting momentarily on the base of the neck i n the water, and a resumption of the normal posture. This i s illustrated by Bernhardt (1940) who states that the t a i l i s fanned. He wrote that the head i s suddenly thrown onto the back where a "creaking note" (Knirrlaut) i s uttered—" and jerks the head and neck well forward, at the same time spreading the t a i l and shooting a stream of water out behind with the feet." I know of no other reference to the spread t a i l , except the i l l u s t r a t i o n by Allan Brooks of the Head-Throw (in Phi l l i p s , 1925). Bernhardt (1940) also i l l u s t r a t e s the position of the feet i n an-other figure. After breaking surface they are seen to be placed quite a bit lat e r a l to the body. Boase records that the whole performance is carried through with a swing and great rapidity. Bernhardt (1940) wrote that the "creaking note" i s uttered "not only when 'throwing out' the throat as I assumed earlier, but also i n the dorsal position of the head, and indeed with opened b i l l " . Boase mentions a "c a l l or grunt". Gilpin (i860) mentions a "short s h r i l l cry comes...from his upturned b i l l " . Millais wrote: "At the moment of throwing back the head the c a l l keek-kee is emitted, but sometimes i t i s given after the kick". Brewster (1911) described i t as a "short vibrant paaap, not unlike that of the Wood-cock but a t r i f l e more prolonged and also less harsh and incisive, i t re-minded me somewhat of the blast of a penny trumpet, less forcibly of the wheezy quack of a drake Black Duck...It was sometimes doubled (paaap-paaap) and occasionally trebled (paap-paa-paa) ...syllables were each shorter than 11 the normal single c a l l and otherwise slightly different . 164 Millais (1913) mentions that the head feathers are puffed out during this movement, even more than i n the Bowsprit. Discussion We do not know i f the c a l l is different from that which accompanies the Head-Throw, or whether no differentiation has yet taken place (as compared with the movement). The occasions on which the movement i s found have been discussed i n the section dealing with B.islandica. Munro and Glemens (1931) wrote: "The males did not face a female whe n displaying thus. The action seemed entirely mechanical and would be i n -dulged i n just as freely by members of a band composed entirely of drakes". This raises the question of whether this movement has primarily a court-ship or "greeting" or other significance. This has already been discussed in the description of the Kick i n B.islandica. There i s some confusion as to whether one or both feet are kicked during this movement. Bernhardt (1940) i n his i l l u s t r a t i o n confirms that both feet perform the kicking movement together. Millais (1913) wrote that they "kick once alternately with both feet". Brewster (1911) wrote: "When, as occasionally happened, the jet was doubled i n volume, and also apparently somewhat divided at the base, I thought that the bird had made simultaneous use of both feet". This implies that Brewster thought that the more usual form was a kick with a single foot. However this would presumably have the effect of turning the bird's orientation through some degrees. There i s no indication that this happens. Rather i t may be suggested that both feet are always kicked, but that occasionally one 165 breaks surface a l i t t l e before the other so that two jets of water are produced, one a l i t t l e ahead of the other, while normally the jets are following equal and ^parallel paths and appear as one from the side. If alternate beats do prove to be the more usual on the other hand this would presumably be a reflection of the alternate foot paddling of the ducks during leisurely swimming. B.clangula differs from B.islandica i n that i n both the displays of the Kick Complex the head is l a i d back on the rump, and i n that the two displays are more clearly differentiated. The kick i n Head-Throw-Kick has the effect of raising the rear of the body and throwing the neck forward and down, while i n B.islandica the whole body sinks and the neck does not got forward into a "Deep Jiving" pose. There must be considerable d i f f e r -ences i n the way the foot-stroke i s made between the two species. There does not appear to be any t i l t i n g of the head sideways i n either display i n B.clangula, as there i s i n the Pseudo-Kick of B.islandica, i n which the head does not come i n contact with the back at a l l . The origins of the displays i s not known, but the Head-Throw i s found i n Mergus and i n Aythya and may be a very old display. The c a l l i s not coincidental with the kick i n B.clangula as i t i s i n B.islandica. The need to stretch the bone drum of the respiratory tract has already been discussed i n the discussion of the Head-Throw. FIGURES 13-14 B.islandica (Icelandic population) PSEUDO-KICK & KICK From a f i l m taken by Dr. Frank McKinney at Slimbridge, England Figure 13. Pseudo-Kick (Side View) Though a kick of the feet occurs i t i s not as violent as i n the Kick, and the bird merely bobs on the water, not sinking as far as i n the Kick. The head.does not reach the back between the shoulders. A Head-Flick ends the sequence. Frame 4 Frames 5 Frame 6 & 6 Frames 7-10 Frames 11-13 Frames 14-15 Beak opened Neck extended up and back to f u l l extent, beak apparently closed Rear of body is humped up as feet are brought forward for the Kick of the feet. Body sinks i n water, beak open. Head-Flick as head passes forwards Head drawn back and down into Normal resting position. Figure 14 Kick (Side View) The head comes to l i e on the middle of the back, not as far however as i n the Head-Throw-Kick of B.clangula. The body sinks much deeper than i n the Pseudo-Kick. Frames 3 & 4 Frames 6 & 7 Frames 8-12 Frames 9 & 10 Frames 11-14 Frames 15-17 Frames 17-19 Head i s drawn sharply into the shoulder Head i s elevated forwards Head is thrown back between the shoulders (and beak opened while the head i s s t i l l passing back) Feet brought forward and rear of body humped. The Kick causes the sinking of the body (as compared with B.clangula, i n which the rear of the body remains elevated) Body rises Head-Flick 16 frames/second. Shutter Speed l / 3 2 n d second 169 FIGURES 15-16 B.islandica.-(Icelandic population) PSEUDO-KICK & KICK From a fil m taken by Dr. Frank McKinney at Slimbridge, England Figure 15. Pseudo-Kick (Rear View) Shows t i l t i n g of the head towards the female to the right of the male. Figure 16. A. Rseudo-Kick (Front View) Shows t i l t i n g of the head toward the female to the le f t of th e male B. Kick (Front View) Shows the Head-Flick at the end of the movement (Frames 13-15)- Any t i l t i n g of the head that may occur, occurs only i n the extreme position (Frames 10-12 only). 16 frames/second. Shutter Speed l/32nd second. 172 FIGURES 17-19 B.clangula (American population) HEAD-THROW From a f i l m by Charles Walcott & Benjamin Dane Figure 17. Side View The head and neck are l i f t e d (Frames 2-5) as i n the First position of the Head-Throw-Kick (Fig.20). The head remains on the rump for 22 frames (Frames 7-28). The beak i s open from Frames 10-29. There is a Head-Flick (Frames 32-35). Figure 18. Front View. Breast appears as a mound of white from which the beak may be seen projecting upwards. Head-Flick at the end of the Head-Throw. 1st Head-Turn to right Centre position held between Head-Turns Head-Turn to l e f t 2nd Head-Turn to right Centre position after 2nd turn to right Movement of head back towards rump during 2nd Head-Throw. Figure 19. With Forward Bowsprit Position The head i s laid on the rump for 22 frames, as i n the normal Head-Throw (compare Fig.17). Instead of being drawn forward into the normal resting position (compare Fig.17), the head i s l i f t e d up and forwards into the Bowsprit pose, and held there for nearly 20 frames. It i s then drawn back into the shoulder with a Head-Flick. That this Bowsprit pose i s part of the Head-^hrow i s evidenced by the delaying of the Head-Flick u n t i l after i t , and to the apparent fact that Bowsprit Pumping does not occur. 24 frames/second. Shutter Speed l/48th second. Frame 1 . Frames 10-12 Frames 28-40 Frame 45 Frame 48 Frame 67 Frame 81 Frames 83-85 13-35. 1* 36 176 FIGURES 20 & 21. B.clangula (American population) HEAD-THROW-KICK From a fi l m by Charles Walcott & Benjamin Dane Both frames start with the same two individual birds (Frame 2, of Figure 20 = Frame 1 of Figure 21). In Figure 20 the far-ther of the two i s followed through Position 1. In Figure 21 the nearer individual i s followed through Position 2 and completes the movement i n about g the time taken by the bird i n Figure 20. Figure 20. Position 1. The head i s l i f t e d up, on extended neck, as i n the Head-Throw. The head i s held on the rump from Frames 22-29, which i s one-third of the time i n the Head-Throw. ' The only sign of the beak being opened i n this position i n either Figure i s i n Frame 26. The rear of the body is drawn i n in,Frani§ 30 (when the f i r s t and main Kick occurs and the head begins to move forward. The rear of the body remains humped un t i l Frame 35 when the wing becomes v i s i b l e as an acute projection beyond the con-tour of the body. A second kick appears to occur i n Frame 3 9 , and the head i s then elevated from the lowdown forward posi-tion which i t has reached. The beak closes. A Head-Flick ends the sequence (Frames 47 -50) . Figure 21. Position 2 (less extreme). The Head i s not elevated at the beginning of the movement, but i s drawn directly into the shoulder, not as f a r back along the back as Position 1. The head i s held on the back for an even shorter time (Frames 5-6) and the beak i s not opened. In Frame 7 the feet have been brought forward and the rear of the body i s humped. In Frame 9 the wing appears as an acute projection as the feet break surface. The equiv-alent position at this time i n Position 1 i s shown by a dotted outline (the head s t i l l passing backwards). The head and neck pass even farther forward than i n Position 1 and the neck i s submerged (Frames 13 & 14 ) . From this position the head is raised (beak open) and a Head-Flick (frame 25) ends the sequence. The bird i n the foreground i n Frames 22 and 25 i s the female. The beak appears to be open i n Frame U , as well as later. 24 frames/second. Shutter Speed l/48th second. 179 Rotary Pumping (B.islandica) This movement ( i t i s hardly a posture, as the movement i s continuous) i s the most frequently performed of a l l displays i n B.islandica, whereas the Bowsprit Pumping of B.clangula i s much less common than other displays, i n particular the Head-Throw. Rotary Pumping varies i n intensity to a con-siderable degree and has at least one variant form. Field Description The common form i s il l u s t r a t e d i n Figure 22, and analysed in Figure 26, i n which the path of the t i p of the beak is traced. The head i s l i f t e d back and up from the resting position, then l i f t e d forward, by rotation of the stiffened neck, u n t i l the neck i s at an angle of about 75 degrees, then un-dergoes a gradual lowering and contraction of the neck into a semi-crouched position before being l i f t e d back and up again. The path traced by the beak i s pear-shaped. The neck is considerably straighter and more upright (neck axis at a steeper angle in the forward position) than i n B.clangula. The speed of the movement varies greatly, as does the number of rotations. It seems that when birds are either on territory, or the sexual motivation is weak, only Rotary Pumping occurs. The male w i l l give but l i t t l e jerks of the head (hardly recognisable as pumping) and this may go on for as long as 45 seconds. The movements may be very slow, or even be stopped temporarily while the male i s almost motionless, with the head held high ("Head-Up" pose), as he watches the female. In the high up position the beak may be opened, and i t i s possible that we have here an intergrada-tion of motivation with that of the Head-Up Neck-Withdrawn position (discussed earl i e r ) . 180 Motivation i s increased, either through some sort of feed-back i n the true courtship sequence or because of an increase i n aggressive motiv-ation e.g. after a t e r r i t o r i a l boundary dispute, when a triumph ceremony occurs. The speed of rotations of the head (Rotary Pumping) i s increased and of fu l l e r amplitude, and there i s a head forward (Laying Neck on Water) posture, which i s frequent at this time (Figure 2 3 ) . In this "Threat" position the male makes a faint ticking or clicking noise. The beak i s held wide open continuously at an angle of about 30 degrees (contrast Head-Up position) and the noise i s produced either by a tapping of the tongue against the mandibles, or by exhalation through the syrinx. The ticking sound occurs i n bursts of 4-5 clicks and greatly re-sembles the noise made by a pair of hand-operated hair clippers. The slight pause between bursts might be due to a pause for breath. Since the sound appears to be a broken series of stops i t seems just possible that i t i s produced by a current of exhaled a i r , rather than by the tongue. This form of Rotary Pumping i s very regular. The number of head rotations before thrusting the head forward i s fixed and few, reduced usually to two between each forward thrust of the head down to the water. We thus have a sequence: Pumping, Pumping, Threat, or PPT, PPT, PPT, PPT. At such times the H e ad-Flick which normally accompanies the Threat a t t i t -ude (occurs during i t at intervals) may occur while the head i s i n the head-raised position of the Rotary Pumping (Figure 23), between periods of Laying the neck along the water. This i s an indication that the laying of the neck along the water has a truly aggressive motivation. 181 An example of the aggressive intermixture is seen i n the following record: a male flew towards a female, and the mate of the l a t t e r threat-ened him, with a ticking noise, and the head i n the Threat posture. This continued while the mated male did Rotary Pumping towards his female. He was making the ticking noise both while i n the Threat position and also while pumping ( i t i s not known i f the beak was held open the whole time i n this latter position). This aspect of Rotary Pumping, which Saw-yer (1928) illustrated, namely the beak open during the movement, i s not usual during pumping, and acts as a possible link with the Head-Up position, i n cases of extreme stress or excitement. There i s seldom any likelihood of such an individual becoming involved i n a fight, for he i s usually the territory owner. Rotary Pumping i s often performed by an intruding pair, when'; threat-ened by a territory owner, as the threatened pair swim away from the aggres-sor. In these cases the pumping by the threatened bird i s quite violent but the head i s held high and i s not lowered to the threat position. Once when a pair landed on the water the male began Rotary Pumping with head held high. Then into Threat, flew, landed and came up head-on with another male which went into a position resembling the Head-Up position, and flew with a different female. The aggressor did Rotary Pumping and the threat position a number of times, and then performed a considerable amount of preening i.e. even after the victim had flown away the aggressive mot-ivation was s t i l l high, as was the sexual motivation expressed by Rotary Pumping. 182 Threat i n Rotary Pumping " -T h e T r i u m p h G r e e t i n g C e r e m o n y After an encounter the male returns to his female. Rotary Pumping i i s the primary display he exhibits but very often the PPT, PPT, PPT, PPT sequence i s added. - The female responds on-these occasions (and on other oacasions when the male is doing Rotary Pumping towards her) by Jiving, and even occasionally a few pumping motions herself as they f i r s t meet. The return of the male to the female deserves to be regarded as a "Triumph Ceremony". Delaeour and Mayr (1945) say that this mutual display i s confined to geese (where i t consists of similar movements by both male and female), but i n i t s motivation and pair-bonding function the Triumph Ceremony of goldeneye i s exactly similar, even though the movements of the two sexes may d i f f e r (the female can and does do Rotary Pumping, however, on occasions). It occurs after nearly every encounter. Whether the threat attitude i s assumed as the male goes to meet his female or only when they are swimming alongside each other I am not sure. The female shows no sign of being intimidated by the male, however and she Pumps as they go towards each other (= Greeting i n downies) and Jives when she come along-side or behind him. Sometimes the male may direct his Threat outwards towards another male. It i s very rare indeed that the male attacks his own female. There i s a suggestion here that i n the case of the Laying Meek on the Water posture which occurs during Rotary Pumping, on such occasions, there i s no aggressive or fli g h t motivation at a l l i n respect of the female, although the Rotary Pumping i t s e l f is being directed at her. The aggressive 183 motivation present i n the Triumph Greeting Ceremony is not, apparently, directed at the mate, but at third parties, whether present or not. A l l that i s seen i s Rotary Pumping, Threat and occasionally beak-opening i n a high position of Rotary Pumping (?Head up position, with escape motivation). In ethological theory aggressive elements i n courtship behaviour towards the female have been explained as either indicating a past con-f l i c t situation (which has been ritualised i n the intention movement of attack as a courtship display) or as indicating that aggressive motivation i s s t i l l actually present during a courtship sequence. But need this be the only conflict (as discussed in the Introduction), and need the aggres-sive postures be necessarily directed at the mate, rather than at third parties? It i s not hard to think of analogies from human behaviour i n which the presence of the female, and/or an adversary, leads to greater degrees of sexual activity with or without aggressive a c t i v i t y associated with i t , than when the female i s absent. The Triumph Ceremony i s just such a situation i n ducks. There i s no indication that the female accepts the actions of the male as being a threat to herself, although the Pumping which she may mutually perform with him, or the Jiving, may have the fun-ction of appeasement. Circumstances i n which the motivation of another bird (e.g. i n a courting party) arriving on the scene, i s not made clear at once, or i s not known to other birds, requires a greeting ceremony with appeasement function. It i s possible that the Kick display has such a function for i t occurs at times when a bird i s a new a r r i v a l , or i s about to land near some other birds. 184 I n d i v i d u a l D i s t a n c e In any discussion of the aggressive content of sexual displays one must clearly consider the nature of "individual distance", both as regards the origin of the displays, and as regards their present day motivations. Goldeneyes pair on the wintering grounds and are not there being more than slightly aggressive towards each other. They move around alon^, i n pairs, or flocks (of from a few birds to hundreds, or even thousands eiders). Courting ac t i v i t i e s occur i n groups and, at this time, the individual distance o f t h e male i s very small. The birds swim close together and, individual distance being small, most displays are derived from conflict situations which have been expressed as displacement preening, bathing, stretching or drinking. It may have an aggressive element, as i n Rotary Pumping with the Laying Neck on Water attitude, or an escape element (Head-Up position), but these are not the only displays. This i s because the individual distance (which i s defended) i s at that time so small that the ne-cessity to curb and ritualise aggression i n courting parties i s slight. Indeed i t seems that i t i s only a favoured male who seems to go i n for any great aggression against i t s neighbours i n a courting party (at least i n scoters) i.e. the individual distance is suddenly increased as the female pays attention to him. Nevertheless the Threat position i n Rotary Pumping may be an essential element i f individual distance increases on the breeding grounds. When males clash on territory Rotary Pumping towards the female may need to be alternated with Threat towards the intruder to keep him away. Thus Threat may come to reinforce the pair pond, x>rhile s t i l l having valency for an intruder. 185 Unlike gallinaceous -birds that have a •lek' and i n which we might expect the displays to derive from aggressive postures, for the lek i s both courting ground and a t e r r i t o r i a l l y divided area, the ducks as a whole have very few courtship postures which originate from aggressive postures. As discussed i n the introduction the tensions which may have given rise to displacement preening and to comfort movement displays are those between the sex drive i t s e l f and escape. These displays are commonest i n the pre-coition sequence. It is true, nevertheless, that threat appears commonly to be included i n the performance of Rotary Pumping, and that the Head-Up position may be the intention movement of flight. It is of interest therefore that Rotary Pumping i s the most common mutual display, or dis-play by the male towards the female while on territory. At this time the individual distance of the male has increased to several acres. Literature Descriptions Mary F. Jackson (M.S., 1952) described Rotary Pumping as follows: — The- head i s slightly t i l t e d backward u n t i l the b i l l points upward at an angle of 45 degrees to the vertical. Then the head i s pumped up and down four or five times i n rapid succession, by alternately straightening and bending the neck into an S-shape. Sawyer (1928) illustrated the movement (with the beak open) i n Figures 6-9 of -Plate-3 i n his paper. -He includes i n this the Threat posture (beak open). He writes: "head bobbing and elaborate neck pumping and stretch-ing are the most outstanding actions" and then refers to the Head-Up position. He goes on "In many of his gestures the.-neck i s extended to a surprising length. With neck stretched upward the b i l l bmay be opened and shut at one to three second intervals i n repeated low quacks. I 186 think this i s more common during rivalry, but I have noted i t when the drake seemed to be courting a certain female". As explained earlier i t i s not my experience that the beak i s open very commonly during Rotary Pumping. Plate 9 i n Stonor (1940) shows Rotary Pumping and the Laying Neck on Water position which accompanies i t . 187 Bowsprit Pumping and Bowsprit Position (B.clangula) Field Descriptions I saw this movement, on two occasions, i n the Interior during the spring migration (in which B.clangula crosses British Columbia to i t s breeding grounds east of the Rocky Mountains). The movement is quite clearly com-parable to the Rotary Pumping of B.islandica though, from s t i l l photographs of any stage i n this movement, i t would be impossible to compare them. When the movement of the head i s seen i n the f i e l d , however, the oscillation of the head i s seen to be quite similar i n timing. The movement was seen between a pair i n 1955, and was observed to be i n the form of a forward and upward movement along a plane 25 degrees from the horizontal. In the forwardmost position the neck axis (to the water-line) forms an angle of 50 degrees. The effect of this forward, and up-ward, elongation of the neck i s that the beak appears to be longer, and more finely pointed, than i n B.islandica. This effect i s also accentuated by the gentler slope of the front part of the skull i n b.clangula. At 73-Mile Lake on April 30 , 1956, the Bowsprit Pumping was seen to be followed by the Head-Throw a number of times. In the f i e l d the movement gives, at f i r s t , the impression of being backwards and forwards, but i s later seen to be actually following an e l l i p t i c a l path (Figure 2 6 ) . It could be de-rived from exaggerated oscillations of the head during swimming, such as occur when the bird is alarmed and i s trying to hold the head s t i l l for a good view while swimming (as a chicken does when walking). 188 Film Description The Bowsprit Position may be assumed i n three ways: 1) from the Head-^hrow (Figure 19), by the head being passed forwards of the normal resting position, 2) via the Forward Scoop (Figure 25), or 3) from the Normal resting position. Only i n the last case i s the head pumped while i n the Bowsprit posi-tion (Figure 24 ) . In the other cases i t i s held s t a t i c a l l y i n the Bow-sprit Position form some time. It i s never assumed from the Head-Throw-Kick, another-feature that distinguishes that movement from the Head-Throw. The Forward Scoop movement resembles that leading to the Masthead position, but with a different directional form. The head i s lowered for-wards from the Normal position and a scooping movement is made over the water forwards. The head then goes up to a 45 degree angle, or i s raised a l i t t l e higher and then lowered an imperceptible fraction (as i n the Mast-head), and appears by faint forward and backward oscillations to be held in this position under tension. *he Bowsprit Position may be held for about 1 second (20 frames) The head i s then drawn back into the shoulders. A Head-Flick follows. Even when the oscillations are very weak the Head-Flick follows the Forward Scoop and Bowsprit pose, and i s an indication that the movement i s a true display. From the Normal resting position the Bowsprit position i s assumed (apparently) directly by l i f t i n g the head forwards, and upwards, at some 25 degrees. When the pumping i s very pronounced, the neck also bends a l i t t l e and the movement then has a considerable resemblance to that found 189 i n B.islandica. During Bowsprit Pumping the back of the crest i s raised, forming a cobra-like shield, and the contour of the head is sharply angled against the neck. The white face patch i s greatly enlarged. This treat-ment of the head'feathering occurs also when the position i s assumed i n other ways but no oscillations (pumping) occur. It distinguishes the move-ment from the Masthead i n which, i f anything, the head feathers are smoothed. The Bowsprit occurs chiefly, i t seems, when the male i s approaching another individual, and has come quite close to that bird. It appears to have greeting function as i t does i n downies of B.islandica. It may also be associated with Head-Turning i n the Normal position. There i s a sug-gestion from Walcott and Dane's f i l m that the Bowsprit position i s a re-leaser of the Head-Throw i n other males; indeed I saw t h i s myself a number of times on April 30, 1956. The Bowsprit position (static at least) may be assumed by the female, and so the female herself may be able to release the Head-Throw. Analysis of Pumping Figure 26 illustrates the path of the beak t i p as traced from fi l m sequences, and modified (slightly) i n the light of f i e l d observation. The movement takes the beak-tip through a smaller traverse than i n B.islandica and thus gives the impression (generally) of being a slower rotation. The path i s a crude ellipse, as compared with a pear-shape i n B.islandica. The head i s held much farther forward, and lower down, than i n B.islandica, the eye being held at only just over half the height from the water i n the ex-treme forward position, that, i t i s i n B.islandica. In B.clangula the angle of pumping i s a plane of 25 degrees from the horizontal, as compared with 190 a mean plane o$ about 35 degrees i n B.islandica (in which the head passes forwards above the horizontal, but i s then„lowered at an angle of 70-80 degrees, at f i r s t , and drawn back further at an angle of about 40 degrees). The angle of the neck-axis i s inclined at 50 degrees forwards of the water-line, as compared with 70 degrees i n B.islandica. In the rearmost position of the head (well into the shoulders), the neck axis i s much more sinuous than i n B.islandica. In B.islandica the head is held much more upright, and much farther back, hence the neck-axis i s straighter. Literature Descriptions The movement was named by Brewster (1911) who noted the resemblance of the extreme forward position to the prow of a sailing vessel. It has also been described by Townsend (1910), Millais (1913), Boase (1924), Phillips (1925), Bernhardt (1940) and Johnsgard (1955). The Bowsprit Position i s apparently illustrated i n Brewster (1911), Phillips (1925) and Johnsgard (1955), but only Brewster (whose i l l u s t r a t i o n shows too great a bend i n the neck) observed that a similar position was assumed soemtimes following the Head-Throw. His descriptions are as follows (Wounded-Duck posture" resembles the Threat posture): — Forward Thrust of Head and Neck ending i n the Bowsprit  posture. The drakes assumed this attitude by suddenly thrust-ing their heads and necks forward and upward from the normal position (this was evidently the usual way) or by raising them more slowly from the crouching of the wouhded-duck pos-ture.... On each occasion the" bird remained absolutely motion-less for several seconds with i t s neck elongated to the utmost and held perfectly straight and r i g i d at an angle with the water of about 45 degrees suggesting the bowsprit of a vessel of ancient type....During i t s continuance the feathers of the head were sometimes fluffed out, but oftener pressed down so 191 very f l a t that the head looked scarce thicker than the neck which because of i t s unusual elongation, appeared abnormally slender. The b i l l was only slightly opened....when the head was raised to the bow-sprit posture from the crouching or wounded-duck pose the movement was not especially rapid; but when i t was thrust directly fonvard and upward from the normal position the action was so swift and abrupt that my eye could scarce follow i t . . . . sometimes take the bowsprit or topmast posture without becoming rigid i n i t , or maintaining i t for more than a-fraction of a second....on closely approach-ing....they often bobbed their heads up and down two or three times i n quick succession.... As a rule the bird kept silent when in this posi-tion, but twice I saw, as well as heard, i t bleat. In one of these instances i t kicked up water just as i t uttered the paaap; i n the other this action was omitted. Phillips (1925) adds that the b i l l i s dipped into the water at inter-vals from the Bowsprit position. Bernhardt (1940) wrote: "The mating takes place about as follows: the drake swims up to the female, perhaps also swims around her, at the same time stretching the neck obliquely forward. The crest i s erect." He described how this led into the Head-Throw-Kick. At copulation he again says that "the drake swims around her with repeated thrusts of the neck". Townsend (1910) may also have seen the Bowsprit when he writes: — One or more males swim restlessly back and forth and around a female. The feathers of the cheeks and crest of the male are so erected that the head looks large and round, the neck cor-respondingly small. As he swims along the head is thrust out i n front close to the water, occasionally dabbing at i t . . . . He goes on to describe the Head-Throw as though i t followed on the Bowsprit. Perhpas Boase (1924) was also describing the Bowsprit when he says: — He seemed to make an impression, for, swimming away from the group he was followed by the particular female addressed.... (he) then began a new display, bobbing the head i n a curved path by rapid retractions of the neck and was imitated or re-plied to by the female with a similar performance.... Millais (1913) says that the t a i l i s raised, and the head may be raised to 192 I the vertical, but we may doubt either of these features. H e wrote: — The male approaches the female with his head and neck held s t i f f l y up at an angle of from 60 to 7 5 ° , the feathers on the cheeks being much puffed out. He swims, sometimes with raised t a i l , i n a semicircle round her, gradually elevating the b i l l u n t i l i t i s quite perpendicular". It i s perhaps odd that none of these descriptions except that of Brewster (who appears to have seen a l l three origins of the Bowsprit pose) i s a really accurate, account of this common movement. 193 FIGURES 22 & 23 B.islandica (B.C. population) ROTARY PUMPING From a film by M.T.Myres Figure 22. Simple -Form The head is drawn back, then forwards and upwards, and fi n a l l y back and down. Compare with Figure 24 (B.clangula) and Figure 26, where the patbeof the beak t i p i n both species i s traced. Figure 23 . With laying Neck on.Water This f u l l sequence of 43 frames was repeated twice over — making about 85 frames altogether. As the head was passing forwards (and while i n the highest position) there was a Head-Flick on both occasions. This led into the lowering of the head to the water. Because of the lowering of the head and neck to the water a different path i s traced from that found i n normal Rotary Pumping (Fig.26) . When the neck i s l a i d on the water as i n this case, there is greater aggressive motivation than normal, usually towards a third bird. The sequence befean with normal Rotary Pumping (with the female Jiving), then laying Neck on Water raising of the head Head-Flick Laying eck on Water (19 frames), of which the last frame i s Frame 1 of this figure. The head was raised, flicked again and layed back on the water (Frame 4 3 ) . Shortly afterwards the bird dove and may have attacked another bird. The Head-Flick lasted only l / 6 t h of a second and was not visible i n the f i e l d . Frame 1 . Frames 9-10 Frames 12-14 Frames 15-21 Frames 28-30 Frames 39-43 End of Laying Neck on Water position Head starts forward Head-Flick Head passes forward again i n Rotary Pumping. Head i s jerked back into the shoulder before going down and forward to water level. Assumption of the Laying Neck on Water position. 24 frames/second. Shutter speed l / 4 8 t h second. 196 FIGURES 24 & 25 B.clangula (American population) BOWSPRIT PUMPING & BOWSPRIT POSITION From a film by Charles Walcott and Benjamin Dane. Figure 24. With Bowsprit Pumping This is a compounded series of drawings to be read from l e f t to right (no frame numbering). Figure 25 . Bowsprit Position static from Forward Scoop Read from l e f t to right This i s one of two other ways i n which the Bowsprit pose i s reached (the other—see Fig.19) . The beak i s lowered to the water (Frames 1-7) and l i f t e d i n a scooping motion (Frames 7-11 ) . The Bowsprit pose is maintained for as long as 22 frames (Frames 12-34) and i s not pumped. The head is f i n -a l l y drawn back into the shoulder (Frames 34-44) • H CO 199 FIGURE 26 B.islandica and B.clangula ANALYTICAL COMPARISON OF ROTARY AND BOWSPRIT PUMPING Compounded from f i e l d notes of both species by M.T.Myres, and from tracings from films by M.T.Myres (B.islandica) and Charles Walcott and Benjamin Dane (B.clangula). Upper pear-shaped trace (with broken-lined head) i s the path taken by beak of B.islandica during Rotary Pumping. Lower, almost e l l i p t i c a l trace (solid-lined head) i s the path taken by beak of B.clangula during Bowsprit Pumping. The path of the beak t i p i s much wider, and placed much higher i n B.islandica than i n B.clangula. • The base line represents the length of the body at water-level. The solid lines arising from water level represent therneck-axis of the goldeneye at the extreme forward and extreme backward position of Pumping i n each species. The rearmost position of the neck of B.clangula i s most curved, that of B.islandica most elongated. The fore-most position of the neck has a shallower slope i n B.clangula than B.islandica. N.B. the path of the eye could, and should, be traced i n a similar fashion. The traces would be placed more nearly over the front of the body, and the shape would be more e l l i p t i c a l i n B. islandica, as there appears to be considerable l i f t of the beak t i p i n the rearmost position. Thus the NW segment of thebeak-tip trace i n B.islandica would be absent i n an eye-trace. 201 Masthead (B.clangula) There i s on Walcott & Dane's f i l m a display (Figures 27-29) which I have not seen i n B.islandica. It resembles the Water-Flip of both species and also resembles Drinking. But whereas the Water-Flip is chiefly found i n the pre-coitaoh sequence, the Masthead, as I shall c a l l i t , i s found i n courting groups, -^ t i s usually directed at another bird from the Laying Neck on Water posture, and may have certairTaggressive adjuncts. It may vary somewhat in amplitude, depending presumably upon the intensity of motivation. Film Description The neck is lowered to the Crouch or may even be dipped into the water at an angle forwards and downwards, -^ t is held here for a considerable time (Figure 27, Frames 1-23), then the head i s suddenly jerked up from the base of the neck, as though the t i p of the b i l l had a string attached and were being pulled upon by a playful boy. T n e neck appears quite s t i f f and straight. The angle of the beak rises to 60 degrees, or even further and thethead is then dropped a l i t t l e way as suddenly, held at this lower angle for a moment, and then slowly lowered to the water again, as though the string had been allowed to s l i p . The movement i s rapid and jerky and i s s t i f f , almost l i f e - l e s s . The head i s held close to the water for as long a time after the movement as before i t (Figure 27, Frames 38-92), or even longer. The movement looks much less li k e Drinking than display of aggressive motivation, although the beak may be actually dipped into the water before the head i s raised. 202 Literature Description The name derives from that used by Brewster (1911). However i t has become evident that Brewster may really have been using the name for the second part of the Head-Throw-Kick. He was not completely clear, i t seems, about the movement, for although he says that the " b i l l was invariably well opened" he also says in the next sentence "oftener than not I heard no;sound". That he confused i t with the Head-Throw-Kick is evident from his remark that "At the close of t h i s brief period of inaction the bird frequently uttered i t s paaap and kicked up spray". However the beak i s quite definitely not opened during the Masthead display as redefined after Townsend's description (see below). Nor does the Masthead lead into a performance of the Head-Throw-Kick.' It i s not clearly detectable i n B.islandica, although i t seems to resemble the Water-Flip i n some respects. Brewster's original description i s as fellows: — Upward Thrust of head and neck ending in the mast-head  posture. Ordinarily this movement was complete i n i t s e l f and made directly from as well as back to normal position. Occasionally, however, i t closely preceded or immediately followed a s t i l l longer backward swing yet to be described. In the pose to which i t commonly led i.e., the mast-head  pose the Whistler's neck might be elongated and straight-ened as i n the bow-sprit posture, and held s t i f f l y erect, or i t might be so shortened, and curved that the occiput almost touched the back between the shoulders. In either case the b i l l was invariably well opened and pointed skyward for several seconds during which the head and neck were kept perfectly r i g i d . At the close of this brief period of inaction the bird frequently uttered-its paaap and -kicked up spray, but oftener than not I heard no sound and saw no water f l y . A l l that can be said of this i s that he thought he saw these variations, but unlike the rest of his descriptions which are substantially correct, 203 he seems to have been rather innaccurate here. Townsend (1910) describes a display which also indicated that the beak was opened at the start of the Head-Throw-Kick. This has not been confirmed however, as indicated above. But i n another passage Townsend does seem to have seen the Mast-head. He wrote: — I have also seen them thrust out the head i n front i n such a way as apparently to scoop up the water, and then elevate the head, the b i l l pointing straight up, but closed as i f they were drinking the water, although I doubt i f this i s the case. It i s to be noted that he notices that the b i l l was held closed i n this display. Also that the neck was stretched forward along the water, before the head and neck were r i g i d l y swung back into the almost vertical position. 204 The Backward-Swimming Display (B.-clangula) Phillips (1925) gives an account of a display which he saw, which was otherwise unrecorded at that time. On April 11 one year he was watch-ing nineteen goldeneye at Wenham Lake: — A l l of these, except two males and a female, were diving and feeding, but these three were in especially active display. Besides the ordinary actions just portrayed I saw the males with head held s t i f f ly forward (bowsprit pose) for a second or two, suddenly shoot backward for a distance of one or two feet. I cannot imagine how this was done, for there was no way of observing the feet. But this curious action was repeated many times and I feel sure that i t must have had some significance. There is a further mention of this behaviour in Gunn (1939). He states that A. Holte Mcpherson described this same display in a letter to him in 1932. I have recently come across a second, unpublished, record of this backwards movement. Mr. Theed Pearse of Comox, British Columbia has kindly allowed me to look through his f ie ld notes, and under the date February 9, 1941 there is the following entry for B.clangula: — The male does not always open the beak wide to display nor always use the foot for splash effect; sometimes he raises the feathers on the top of the head, other times gives a jerky twist of the head. Another form is to depress the head horizontally to the surface, hunch up the back and move backwards. 265 FIGURES 27-29 B.clangula (American population) MASTHEAD From a film by Charles Walcott & Benjamin Dane Figure 27. Side View From the!Normal resting position the head is lowered towards the water, and held there for as many as 20 frames. It is then raised until the peak points vertically. The beak is not opened. The head is then lowered to the water again, and held in this position for at least 50 frames. Note resemb-lance to Water-Flip (Fig.31) of B.islandica and to Drinking As above, starts and ends from a Crouched Position with beak close to water or dipping in i t . Difference lies in the degree to which the stretched neck is sloped backwards, and in the elevation of the breast from the water. The movement is very rapid (Frames 17-24 up, and 25-29 down.). The beak is not opened. Shows the large dark, spatulate, shape of the head above white neck and breast. B. HEAD-THROW Rear View Shows head in its extreme posterior position over the rump. The b i l l points upwards and the white face patches are only just visible, and may not be visible to a female following along behind i t . The dark back is tilted back at 45 degrees and the shoulder regions form dark outwardly pointing flanges. C. With Forward Bowsprit Position Two views. Frame 1 shows the head passing forward after the Head-Throw. The dark head sits atop a brilliant white neck. Below the square black back lies between white flanks. In Frame 13 the head is static in the Forward position. The length of white neck is greatly reduced over Frame 1 . 24 frames/second. Shutter Speed l/48th second. (Fig.2). Figure 28. Extreme Form Figure 29. A. Front View 209 THE COITION SEQUENCE The displays separated out from the courting displays and placed here, are those which directly lead to, occur during, or follow, the act of f e r t i l i s a t i o n . Copulation may be seen as early as February. The coi-tion sequence i s often i n i t i a t e d by the female (especially on the sea). She does this by assuming the Invitation to Coitus, and may remain i n this position for a considerable period. The male tends to ignore her at f i r s t but on the breeding grounds the position i s generally assumed after a certain amount of courting activity by the male has taken place. The liling- and Leg-Stretch display of the male i s much the most certain indicator of a coming mating. However, the Water-Flip (either alon£, or i n combination with the Wing- and Leg-Stretch) occurs most frequently i n the pre-coition sequence as well. The male assumes the Pre-Coition Steam-ing posture i n B.islandica from a lying sideways pose (e.g. Wing- and Leg-Stretch) or from a normal posture after Water-Flips or twitching of the water sideways. Jabbing of the water violently appears immediately to precede the Steaming position i n B.clangula (in which also there i s a preening movement begind the wing). The Kick -is seldom seen among the displays preceding coition. Rotary Pumping i s the most common element i n the early part of the sequence when the pair meet on territory (in B.islandica), or when the male starts dis-playing i n a pre-coition sequence. When the sequence i s broken off for any reason there i s a reversion from the specific coition displays back to Rotary Pumping. On one occasion a male (? off territory) made three attempts at cop-210 ulating. The female scooted away from him each time, but he could not follow in the Steaming position; i t appeared too tiring. Failure to mount may result in a reversion to Rotary Pumping. On two occasions both members of the pair dove, and on surfacing the male did Rotary Pump-ing and the female Jiving, i.e. there may have been the necessity to dive before Jiving could be performed (?). The Post-Coition Steaming display does not resemble that of Anas  platyrhynchos or Aythya vallisneria. 211 Invitation to Coitus The "solicitation" posture of the female has been described by a num-ber of authors. In the Mallard (Anas platyrhynchos) the female performs the up and down "pumping" movements (Lorenz, 1941) mutually with the male. In the Canvasback (Aythya vallisneria), however, the female stretches her-self low on the water as i n goldeneyes (Hochbaum, 1944). Alford (1921) i n particular was impressed by the invitation posture of the female and remarked how long the posture could be maintained before copulation occurred (as long as 15 minutes): "With neck outstretched and her body quite limp and apparently l i f e l e s s , she allows herself-to d r i f t upon the surface ex-actly after the manner of a dead bird". This observation was made on Vancouver Island i n February (species ?). Bernhardt (1940) wrote of B.clangula that copulation was initiated by the female, who lowers herself f l a t on the water and stretches her neck well out i n front. Sawyer (1928) illustrated the posture i n B.islandica (Figure 20 of Plate 3 i n his paper): "to indicate the fact that the females thus often manifest desire without relation to the mood or proximity of any male". He discussed this further as follows: — "inviting" females to be seen floating like half submerged logs on the pond...For many minutes at a time the bird looks like a rounded piece of driftwood as she l i e s half submerged for her entire length, including head, neck and b i l l . She sometimes emits a low ducking c a l l i n this attitude.. .the females, so far as one might judge by behavior, were decidedly more precocious than the males i n their desire. With a given pair the female's period begins long i n advance of the male's and continues unabated u n t i l the male's period, only two or three minutes i n duration, i s over. It was a common thing to see a half sunken duck float and d r i f t invitingly about a drake for a quarter of an hour or more, while he showed not the slightest knowledge of her exist-ence. 212 Water-Flip This movement i s placed i n the coition sequence because i t i s here that i t is most frequently found, but i t does occur between pairs on other occasions, and i s even performed by unaccompanied birds. Furthermore the female may perform the movement, not i n the coition sequence, but at other times, especially i n response to i t s performance by the male. The movement (Figure 31) i s almost impossible to distinguish from Drinking (Figure 2 ) . In B.islandica the head i s dipped to the water and then rapidly the beak i s raised to an angle of about 45 degrees (or more). There seem to be a number of postures i n which the head dips i n a medial plane to the water: the symbolic dip to the water (Water-Dip) the Looking for Food pose, and Jabbing as well as the Water-Flip. But i t seems clear from i t s similarity to Drinking, that the Water-Flip is derived from the comfort movement of Drinking, and that i t i s more in the timing and repeti-tion of the movement than i n i t s form that i t has acquired sexual signal value. It i s distinguished most clearly during the Wing- and Leg-Stretch movement of the pre-coition sequence. The form i n this case appears to be a l i t t l e different as the neck is thrust forwards down to the water i n a contorted position, and then sharply l i f t e d . In this forward position from the sideways-tilted posture of the body, the body seems to bob i n the water. The mutual performance of the Water-Flip between the pair (at other times than the pre-coition sequence^ i s very striking. I once recorded the following sequences i n B.islandica: "Six times when the male has done 213 an Elaborate Water Fl i p , the female has responded to him by doing the same thing, deriving i n the male from something like the Crouch position. Once he repponded to an Elaborate Water-Flip i n i t i a t e d by her. Another time he did i t though she was dove." Wing- and Leg-Stretch This movement is almost exclusively found i n the coition sequence, just prior to the act of mounting, though i t may be performed for some minutes at a time, then broken off and replaced by other movements, and then resumed. The male l i e s on his side i n t h i s movement. Most usually, but not invariably his belly is towards the female, gleaming white. He stretches the wing and the leg which are then uppermost, so that the orange leg i s displayed against the undersides of the wing, and the snow white breast and underparts aretgleaming (especially when the sun is out, since the breast i s covered i n water-droplets). The wing i s stretched slowly backward and ..then brought forward again, after a pause, for folding. The leg passes back parallel to i t , and appears to support the wing (Figure 30). The head is often turned forwards (? always), and pointed out over the water i n front. A deep and contorted Water-Flip i s often performed while i n this pose, when the wing i s stretched farthest back the end of the Wing- and Leg-Stretch movement and as the wing i s starting forward again. Sawyer (1928) illustrated this movement i n Figures2-3of Plate 2 i n his paper. ^ e wrote (B.islandica): — 214 The f i r s t positive indication of his desire i s apt to be a peculiar and animated twitching of the water with his b i l l ; then he i s apt to stretch, turning over on his side and ex-tending the upper wing and l e g — i n this he i s quite deliberate; pluming of the back feathers follows, and looks like a gesture of ostentation. Bernhardt (1940) remarked that he had seen a. wing-stretching display i n breeding pairs. Gunn (1939) thought he was the f i r s t to notice this dis-play but Briiggemann (18?6) mentions i t i n B.clangula. Gunn wrote that i t consisted i n "the exhibition by the drake of the upper surface of his out-spread wing while i t i s supported by his foot. In order to do this effect-ively, he careens his body and extends the wing that then comes upper-most", ^e went on to discuss the display further, as follows: — "There i s , of course, a great difference i n the adver-tisement value of the two wing-surfaces for the under i s partly "mouse-grey" i n colour and shows l i t t l e of interest while-the upper surface is made up of large black and white areas i n conspicuous contrast, i t follows therefore, that the most complete'method of showing his wing-pattern i s for him to r o l l towards the duck and while his back is towards her to spread his uppermost wing. The object i n sxipporting the wing with the foot i s , presumably, to enable the drake to retain this somewhat awkward pose with the wing at f u l l stretch, longer than he could i n the absence of such a strut; that i t serves i t s purpose efficiently i s certain, for, i n a back view, his orange-coloured toes may sometimes be seen projecting between the q u i l l s . When, as sometimes happens, he-makes the mistake of careening away from the duck instead of towards her the exhibition becomes something of a damp squib, for he then shows only the under surface of the wing; but, as a result of watching him over a good many years, I can state confidently that he i s much more often right than wrong i n his selection of the proper attitude to adopt and wing to extend. I am sure that i n B.islandica i t i s the belly that i s most often exhibited towards the female, though both orientations may be seen. It i s possible that there i s a difference between the two species i n this 215 respect. Gunn (1939) mentions that the female B.clangula also performs this display. During the coition sequence however the female i s posed in the Invitation -to Coitus and so I believe he may merely have been describing the comfort movement. Jabbing, Wing-Preen, Pre-Coition Steaming, •.andbMounting and Post-Coition Rotations Field and Film Descriptions In B.islandica the male assumes the Pre-coition Steaming position direct from the Wing- and leg-Stretch position, or there i s an interval during which he may be performing water-Flicks. The Pre-Coition Steaming posture^resembles the Post-Coxtion Steaming posture. The male often has to swim from a lateral to a posterior position i n relation to the female before actually assuming the posture, but the speed of motion i s consider-able during this preliminary orientation. In the Pre-Coition Steaming position the chest i s raised very high (white) off the water. The head feathers are b r i s t l i n g as the male races towards the female. It i s not easy to see what happens as he mounts, but possibly he brings his feet forwards as he reaches the sloping rear quarters of the female. It is possible that the Water-Flicks are the equivalent of Jabbing i n B.clangula. In B.clan'gula Jabbing intervenes between the Wing- and Leg-Stretch and Mounting, i t consists oT a series of violent jabs or dips to the water, downwards i n front of the male. It resembles the preliminary dips to the Water found i n Splash-Bathing. As the drake B.clangula performs these movements he is moving backwards u n t i l he is behind the female. 216 When behind the female the male turns the head and preens on the back between the wings or behind the h a l f - l i f t e d wing (Figure 32, Frames 1-9) i n B.clangula. Then he suddenly l i f t s the neck s t i f f l y upwards (Figure 32, Frames 9-12) and moves very fast towards the female i n a pose resembling that following mating. The head may be rotated back a l i t t l e (Figure 32, Frames 12-19) and then swung forward again as the drake begins to cover the female. It i s noticeable (Figure 33) that the male actually covers the female and treads her down (? she actively sinks her-self i n addition), before he manages to acquire a grip on the crest feathers of the head. The female is almost completely submerged. In the coition sequence illustrated, the spreading of the t a i l and wagging of the t a i l ( ? origin of Tail-Wag) by which intromission i s effected, continue for about 7 seconds (103 frames from Frame 35-Frame 13&). When coition i s completed the male slides sideways off the female's back (either side), but continues to hold onto the feathers of the nape. The male begins to paddle with his feet. Since the female i s s t i l l motion-less she acts as a dead weight, and the male's head i s dragged back at a sharp angle (Figure 34, Frames 145-149). In the Figure he appears to be dragging the female around him, since he i s turning to the right, while she i s on his l e f t , but this i s an i l l u s i o n . In fact there i s a combina-tion of the female acting as a dead-weight, and the male propelling him-self away from the female (outwards) to counteract her weight. The female turns around parallel to him, but since she i s almost stationary the male can only turn away from her by traversing a circular path around her, moving backwards. Presumably the leg on the side of the female i s beating 217 more powerfully than the other one. In the sequence illustrated a f u l l l g turns are madebbefore the male releases the head of the female and she emerges completely from her submergence i n the water. In Frame 200 (Figure 34) i t i s noticed that the head of the male is placed as far back as i n Frame 149, but there i s a greater turn of the head, the axis being more upright than i t was earlier. literature Descriptions Bruggemann (1876) who saw coition about twenty times in B.clangula describes Jabbing. His i s the earliestdescription of the coition of B.clangula. He wrote (in trans1.): — "Then the male pushes the beak, sneezes into the water, immediately afterwards sneezes again and thus four or six times after each other with always short intermissions. At the same time i t moves away to the side of the female swim-ming up to about two metres. Then i t turns suddenly around with a tremendous speed at the female." Sawyer (1928) mentions "water-twitching" and described coition i n B.islandica. Saxvyer illustrated the performance in Figures 8-17 of Plate 2 i n his paper, ^e writes of Mounting: "then the upright position i s assumed, which marks the beginning of the spurt to the female; the birds are usually within a yard of each other when the spurt begins", and he mentions that an unusually plumaged male (probably a yearling) when trying to force copulation began his Steaming movement "at a distance of some dozen to twenty-five or thirty feet from the intended mate, instead of the customary yard or less". Nor was his performance preceded by the"dabbling, stretching or preening so characteristic". The Post-coition Rotations are described by Bruggemann as follows: — 218 "After the pairing the drake holds onto the crest of the female for a few moments and pulls her, turning backwards two or three times i n a narrow circl e around him while ^he gives off the previously mentioned c a l l " . Sawyer (1928) also described th*m as follows: — "While f i n a l l y , i n coitu, they begin to swim i n a very small c i r c l e a note i s repeated at regular intervals of about a second; I wrote i t , "Gr-err-er" or "cr-err-er", and i t seemed to come from the female, yet the latt e r point i s i n doubt. The middle syllable, high and accented, seems jerked forth. Another note (I thought from the drake) i s a low cluck; these two notes were timed with each other so that one appeared an echo...there seems a possibility that both calls came from the same individual. Having circled, as mentioned, two or three times around, the pair separates, each bird swimming away instantly from the spot; dabbling and vigorous bathing begins at a distance of some forty feet on the part of the female, rather farther i n the case of the male. Post^Coition Steaming and Splash-Bathing The male, upon dismounting swims directly away from the female, usually at an angle forwards i n B.islandica. He has his head held high and crest f u l l y expanded (Figure 35 and 3 6 ) . He grunts as he turns his head from side to side. Once I saw the movement from i n front. The head turning was a l i t t l e twist of the head to the l e f t and right as he steamed away. The difference between this and the Head-Up (in B.islandica) l i e s i n the curvature of the neck axis (compare Figure 9), the long sloping "set" of the feathers at the back of the head, and the fact that the beak i s pointed downwards at an angle of 45 degrees. The white face-patches flash j e r k i l y and alternately with small amplitude. M, seems probable that seen from behind they would not be visible to the female. Bruggemann (18?6) was the f i r s t to describe this display. He wrote 219 of B.clangula (in transl.) "Then he releases the female and then swims for a few metres i n a straight line with neck stretched out vertically, and with breast out of the water, and only by and by does i t get out of i t s exalted state". Millais (1913) figures this display i n B.clangula but his figures exaggerates the posture enormously. He wrote: — ..after pairing with the female....a mad rush along the surface of the water for 15 or 20 yards. Whilst doing this he adopts the somewhat remarkable attitude....breast is held high out of the water, and head depressed backward with b i l l up i n the a i r . He makes the water f l y in a l l directions and not infrequently makes his rush at any other bird that may happen to be near.' This description compliments the i l l u s t r a t i o n , but i s not accurate. For B.islandica, Sawyer (1928) provided the f i r s t description: — The male's appearance i n this swimming away i s note-worthy. He has an extremely self-conscious bearing; in the li v e bird the effect i s enhanced to a ludicrous extent by the regular ticktock movement of the b i l l from side to side. The set pose, the straight course with uniform speed, the mechanical movement of the head—all give every appearance of an automaton, personifying egotism and wound up to run a set course.. .How.. can we explain the remarkable 'proud' swim, so uniform i n every remarkable detail, of the male; this especially, since i t follows complete consummation of the mating.? Sawyer illustrates the display i n Figure 18 of Plate 2 i n his paper. Sawyer raises the question of the selective value i n a post-coition display. Such displays are, of course, quite frequent in animals. The movement i n this case i s completely stereotyped. Post-coition comfort movements could be understood as an outlet of emotional tensions, and as a normal cleansing activity after the "contamination of bodily con-tact" (as one might c a l l i t ) of coition, just as occurs after a fight. In both cases the feathers need to be readjusted from the ruffled condition 1.220 J that has been brought about. But why a specific ritualised, display should be evolved is not clear to me. After some seconds the male subsides and begins to Splash-Bathe. After an unsuccessful copulation the male did not take on the Steaming posture (nor did the female begin Splash-Bathing as she normally does). Shortly after the unsuccessful attempt a successful copulation occurred and the usual display followed. The female generally pauses a moment after the male lets go of her head and then begins to Splash-Bathe (ahead of him, i n time). Sawyer (1928) remarks "Just after the bathing which follows a mating completely enacted the female i s especially given to repulsing intrusive males and females alike". I have not noticed this. Vigorous preening follows the Splays-Bathing for some minutes i n both sexes. Hochbaum (1944) makesl*o mention of a Steaming display i n Aythya  vallisneria: "After treading the pair part, flap their wings vigorously, preen for a moment and then swim off together." 221 FIGURES 30 & 31 B.islandica (Icelandic population) WING— & LEG-STRETCH From a film taken by Dr. Frank McKinney at Slimbridge, England. Figure 30 Upper Line A. White patches on the upper side of the wing are visible. Sometimes these are exhibited to the female, instead of the white belly and orange legs. B. The tipped-over-sideways position i s seen, and the extended foot i s seen to be using the wing as a backcloth, or i s supporting the wing. B. clan gu la WING- & LEG-STRETCH From a film taken by Dr. Frank McKinney at Slimbridge, England. Figure 30 Lower Line A & B. Equivalent positions to above. WATER-FLIP Figure 31. This individual xvas performing i n a pre-coition sequence. The resemblance to Drinking (Fig. 2) w i l l be noted, and to the Masthead (Fig. 27). 16 frames/second. Shutter Speed l/32nd second. B. B . 224 FIGURES 32-35 B» clangurla THE COITION SEQUENCE 225 FIGURE 32 B.clangula PREENING, AND PRE-COITION STEAMING From a film taken by Dr. Frank McKinney at Slimbridge, England. Read from right to l e f t From the normal position which i s preceded by some form of preening (on the back between the wings, or behind the wings) the neck i s shot upwards with raised beak and the male swims rapidly towards the rear of the female. The head i s f i r s t drawn back steeply (Frame 14) and then forwards into a curved posture (Frame 1 9 ) . 16 frames/second. Shutter Speed l/32nd second. 227 FIGURE 33 B.clangula MOUNTING AND COITION From a film taken by Dr. Frank McKinney at Slimbridge, England Read from right to l e f t . This sequence i s a continuation of the one i n Figure 32. Mounting directly from behind, the male presses the female down into the water, with the weight of his body and, by treading with his, before seizing the feathers of the female's crest. Frame 35 shows the spread t a i l during intromission. 16 frames/second. Shutter Speed l/32nd second. 229 FIGURE 34 B. clangula POST-COITION ROTATIONS From a f i l m taken by Dr. Frank McKinney at Slimbridge, England Read from right to l e f t . This sequence i s a continuation of the ones in Figures 32 and 33 • During coition a break occurs from Frame 35 to Frame 133. The male slips off sideways and begins moving forward so that his hold on the female's head draws his own head backwards (Frame 147). In Frame 149 the paddling of the feet i s seen to produce some spray as the male accelerates. This continues u n t i l Frame 200. Frame 149. At start of Rotation Frame 200. End • of f i r s t turn (male's head is raised more upright, but rotated a greater angle posteriorly than i t was i«\ Frame 149) Frame 214. After Ig turns the male lets go the head of the female, leaving a projecting tuft of feathers (Frame 215). 16 frames/second. Shutter Speed l/32nd second. 231 FIGURE 35 B.clangula POST-COITION STEAMING From a f i l m taken by Dr. Frank McKinney at Slimbridge, England Read from right to l e f t . This sequence i s a continuation of the ones i n Figures 32, 33 and 34 The head i s raised, beak horizontal, and the male swims rapidly away from the female with neck raised. The view of him that i s seen by the female i s that of Frame 226. Although he i s Head-Turning she probably does not see the white face-patches. In the example i l l u s -trated, the male gets smaller and smaller as he swims away on the film. He never did any Splash-Bathing i n this instance. 16 frames/second. Shutter Speed l/32nd second. B.islandica POST-COITION STEAMING From a f i l m by M.T.Myres In the right hand picture the male i s s t i l l holding onto the female. He then splashes as he paddles away. The Post-Coition Steaming pose was held i n this case for 245 frames (speed of shutter ?), before the bird began to Splash-Bathe. It i s not known yet i f the wing i s always held up as shown. Compare with Figure 35• ? frames/second (either 24 or 48) 349 234 THE BROOD Leaving the Nest Field Description of B.islandica At midnight on July 11, 1955 I took a nesting box off the tree to which i t was attached, blocked the entrance and transported i t to camp. The female was inside and the six eggs had hatched in the previous 24 hours. I cut the primaries of one wing of the female and returned her to the box. At 1000 hours on July 12, I placed the box on the ground in an enclosure a foot or two from the edge of the lake and removed the plug from the hole. Nothing happened for the first 2f hours, during which time there was scuffling inside the box and pecking noises as though the young were pecking at the walls. Occasionally there were low grunts from the female. After 2g hours the female appeared. "The female came out first and the f i r s t young only a moment later and the rest in a rapid succession and a l l were out within two minutes, but I had no watch to time i t . " The young of course have sharp toe-nails and these are probably a special adaptation for the tree-nesting habit, since holes from which they must escape may be as much as, or more than,three feet deep. In captivity their climbing ability both against a rough wood vertical surface and against string net fencing is remarkable, and there is evidently great flexor power in the legs which, when roughly horizontal, can yet maintain the body of the duckling in a vertical posture. "I noticed that the young perched on the edge of the hole with their heads held quite high and then gave a goodly-springy-jump (? with both legs together) down at 45 235 degrees to the ground." where a nest hole i s at i t s normal height of 5-40 or more feet above the ground this jump would result i n the downies landing 3-10 feet from the base of the tree. When calling the young off the nest (nest-box in enclosure) the female gave a rapidly repeated, continuous, series of calls. This was a short note cue, repeated over and over again. I heard i t 15 minutes later used to c a l l the young to her. They would rush after herwhen they heard i t , leaving off whatever they were doing, to do so. The cuc-cuc-cuc c a l l sometimes broke into the more normal following grunt. By the next day four of the six downies had developed much independence and did not always re-spond to the cuc-cuc c a l l i f given i n a low tone. When she later called i t louder and more urgently these four joined the two which were following her. The female was s t i l l giving this c a l l , when alarmed, on the second day after the young l e f t the nest box. Literature of B.islandica There are two literature descriptions of this event, both with part-icular interest because of their setting. The f i r s t is found i n Munro (1945) who quotes Mr. Sam Sorensen's description of young leaving the nest, close to my campsite of 1955, at Westwick Lake, B.C. In the 35 years that he lived there (prior to 1945) Mr. Sorensen often saw the young fledge. The nest was 16-20 feet from the ground and about 50 yards from-the edge of the lake. The young ducks "tumble out from the entrance and run to water....While the young were coming out the female kept flying back and forth between the lake and the-tree. She would be seen on the water with one young, then with two and so on u n t i l f i n a l l y a l l the brood was there." 236 Munro suggests that this habit of flying from the nest tree to the water may be the basis of the oft-quoted myth that the female carried her young. Presumably the female does not behave in this way when the nest hole is actually over water. Nor does i t seem to occur when the distance to be travelled by the downies i s some way, for I have seen a female leading her young on foot across the range from a nest 400 yards from the water's edge. However, i t may be the usual, or a common practice, when the young have only a short distance to walk, for the female to keep i n the a i r , and f l y to and fro over the heads of the downies, rather than for her to land on the ground and lead them. Doubtless they are guided by seeing her i n flight and by the direction from which she i s heard calling. One wonders, however, whether the young do not i n addition have some other means of detecting the direction of water. The second description, appropriately comes from Iceland, where the nests are not i n trees. It was made by Millais (1913) himself, at Kyvatn. He describes how he came to the edge of a high bank overlooking the Skalfandi River. On the water below him, and not 15 yards away, there was a female i n a great state of excitement. She kept swimming up and down uttering an oft-repeated kra-kra. As he was about to leave he observed a small black and white spot on the edge of a hole i n the farther bank of the river. It disappeared and was replaced by two similar objects: — A moment afterwards I sa^ v the head and neck of a baby Golden-Eye peering over the edge of the hole. Immediately another chick seemed to leap upon i t s head in less time than i t takes to t e l l i t . Eight l i t t l e Golden-Eyes l i t e r a l l y poured out of the nesting-hole one after the other, and f e l l into the water close to the mother... The distance they had to jump was, i n this case, only 10 feet, but there 236 i s l i t t l e doubt that a considerably greater height would have not been any more hazardous. Literature of B.clangula. Tales of the mother goldeneye carrying the downies from the nesting hole to the ground, either on her back or i n her beak run back far i n the folk literature of Europe. They are apparently s t i l l told by northern countrymen:,in Iceland, Scandinavia and Germany, and are actually reported by such writers as Seebohm, Dresser, and NaUmann (Millais, 1913). Lloyd (1854) reports a description i n the Proceedings of the Academy of Sciences at Stockholm. Whilst a Pastor Bjorkman was out with his servant hunting ducks, they hid i n ambush near a lake. They ~gaw a female goldeneye alight among some willows, and then f l y up again. They noticed a newly-hatched downy there. Shortly afterwards the female rejoined i t "and after deposit-ing a second duckling, flew off hurriedly as before...On her third v i s i t they remarked her head to be inclined i n a very peculiar manner; and on the following they very clearly perceived that i n a sort of Olga, or hollow, formed by the head and the b i l l resting on the breast, she conveyed them under her throat." While i t i s generally accepted that such accounts are fabrications, i t would be unwise to exclude them t o t a l l y . It i s possible that under certain conditions of distance and country an alternative behavioural pattern i s invoked. Phillips (1925) evidently quoting from Brewster (1900) , says that downies "easily climb up any sort of material, smooth or wet wood, or cloth, by a series of jumps." A good account of B.clangula downies leav-the nest is that by R. A . Gilbert (in Brewster, 1900) . He describes how 237 the female spent much time at the nest entrance peering outwards, and f i n a l l y flew down to the water, and swam around the tree stub three times, "clucking and calling". Finally she stopped directly under the hole "and gave a single loud cluck or c a l l , when the ducklings began scrambling up to the entrance and dropping down to the water i n such quick succession t as to f a l l on top of one another..." They l i t e r a l l y poured out of the nest much as shot would f a l l from one's hand. One or tw> hesitated or paused for an instant on reaching the mouth of the hole, but the greater number toppled out over the edge as soon as they appeared.... They did not seem to strike the water with much force. While the downies were emerging the female sat motionless. When a l l were out she swam off ahead of the brood and quickly disappeared. The tree de-scribed i n this account i s illustrated, with other nesting sites, i n Brewster (1900), from the Lake Umbagog, Mainejregion. In a similar account from Quebec, Macartney (1918) describes how he saw the female standing on the ground at the bottom of the tree. "She gave several low quacks or calls, and out of- the hole i n the tree overhead promptly tumbled about a baker's dozen of fledgling ducks"...The old bird gathered them i n a bunch and led them to the water. Phillips (1925) summarises this behaviour by saying that various reliable observers described i t as a veritable "pouring out" or "boiling over" of downies. The l i t t l e birds land eight or ten feet from the base of the tree. H e cites an M.S. of Harper's and i V ierrikallis (quoted i n the Bulletin of the British Ornithologist's Club 40:151-152), as well as Gilbert's account (above). Phillips writes "swims around i t , clucking for a few turns, and then, stopping directly under the hole,gives a single 238 loud chuk which i s inunediately responded to by the appearance of the l i t t l e ones", ^his may simply be an adaptation of Gilbert's account. Saunders (1916) writes of a house-site, such as i s sometimes used by goldeneye: "the duck then si t s on the top of the chimney and waits for the youngsters to climb up and project themselves over the edge, to r o l l down the roof and bang on the ground, which never seems to hurt them at a l l " . Bernhardt (1940) and his helpers saw the downies leave the nest. The young appeared at the entrance to the hole (which was almost 2 m. deep) f e l l to the lawn below, hurried to the fence and tumbled down the 35 m. high wall into the water where the mother awaited them, In another place the nest was 3 m. high "the old bird sat under the tree and called korr  korr u n t i l the last young had jumped out of the nest hole." There are only two more recent accounts. The f i r s t was one by Miss Jessie MacDonald i n 1948 (in Carter, 1952), in New Brunswick. The other is that of Siren (1952). H e ^ Q particularly interested i n the distances travelled by broods of B.clangula after they had l e f t the nest, both over-land (1500 -m. i n 3 hours) and across bodies of water (2I4O m. i n 67 minutes). The female made fl i g h t s , for periods of \ hour each, i n the direction that she subsequently led the young from the nest. Finally she sat at the nest opening (once as long as 51 minutes), and then alighted on the ground or the water beneath the nest. It i s at this moment that "a special sound never registered on other occasions" was heard. When this was uttered the young "jumped one after the other from the nest to the ground or water. In the five cases observed the departure of the young took 40 , 86, 95, 106, and 150 seconds—"• i.e. an average of 1 minute and 35 seconds only. "The 239 female s t i l l waited for some time...45 seconds, to 5 minutes, 25 seconds... then she started leading the brood away". Discussion These accounts a l l agree i n a number of points, F i r s t , that the female i s on the water or the ground when the young emerge. Second, that there i s a special c a l l used at this time, which has an immediate and powerful effect upon the behaviour of the downies inside the nest cavity (very often i n considerable darkness). Third, that the young a l l react quickly, by clambering up the sides of the cavity, and with l i t t l e hesi-tation springing out into space. It seems that i n some accounts the female calls for some short time before the young appear, at others that they appear as soon as she starts to c a l l . Perhaps the early clucking i s her natural expression of nervous-ness, such as she uses.as an alarm signal, apparently, when a brood i s pursued on a lake by a boat, or i s disturbed from the edge of a lake by a pedestrian. It may be that when this period i s over the releasing c a l l i s uttered, but that the human observer does not necessarily distinguish the two c a l l s . It cannot be known un t i l we have recordings of both whether the c a l l i s exactly the same i n the two species. There is a certain agreement that the downies do use their wings during the descent. Gilbert (loc.cit) wrote: " A l l used their tiny wings freely, beating them continuously as they descended.", while Macartney (loc.cit) wrote: "They were unable to f l y , but were sufficiently grown to be able to ease their f a l l to the earth, and not unlike a flock of butterflies, they came down pell-mell, fluttering and tumbling, some of 240 them heels over head, u n t i l they reached the ground, unharmed." Macartney (loc.cit) entitled his paper "Golden-eye Duck carrying young", which might imply a persistence of the old tradition. But what he meant i s indicated i n the remark: "The old duck then sank low i n the water and the ducklings gathered over her back in a compact clump. She took them across the bay..". From this i t i s clear that he thought that the downies were actually being carried on the back of the parent. This i s extremely doubtful, and must be very unusual, i f i t occurs. It must be mentioned however since there are references in the literature to such a carrying-of-the-young by parent mergansers on the water and goldeneyes are thought to have relationships with the mergansers. Thus the main ethological problems needing c l a r i f i c a t i o n are 1). whether only one c a l l (the specific "calling-off" note) i s given at fledging, or whether the anxiety c a l l may also be given (the problem i s whether the "calling-off" c a l l always has an immediate effect or not), and 2). what the physiological changes (end of incubation) and motivational changes are i n the female i n the few hours before fledging. In the "dry-ing-off" period of the hatched doxvnies her incubation rhythm must be reduced, and her exploratory tendencies i n the direction the downies w i l l be led, must be developed. Finally comes the period of hours, or even minutes, when she comes to v i s i t the nest for the last time, "when she i s "satisfied" about the route to be taken, and the preparedness of the young to leave ( i t i s presumably necessary for her to go into the nest so that their dry and restless condition may act as a stimulus to her on this score) she must leave the nest, give the releasing c a l l from below and then wait for the young to come to her. But the timing must f i t not 241 only into her own preparedness but into the stage of growing strength and activity of the young themselves. They must be made to leave the nest, before any individual might do so on his own account, for i f this were to happen the period of time between each downy leaving the nest might (for the early ones at least) be considerable, and the brood would never be united on the lake. There would possibly be total loss of young from nests back i n the forest. On ecological ^grounds alone therefore such a special stimulus i s a necessity for survival of the race, although I have suspicions that downy B.islandica may have a tendency to walk downhill rather than u p h i l l - this presumably leading them to water. Interesting experimental work remains to be done using recordings of the "calling-off" c a l l . For example i f the female could be caught about 24 hours after the f i r s t young hatched (or could be held i n the nest for another 12 hours) the penultimate, and sometimes the last egg l a i d might be enabled to hatch, which i s often not the case i n the wild. The " l i f e " of receptivity of the young to this c a l l could be investigated, and how long they may remain i n the nest i f they are deprived of i t . To what ex-tent i s the response to the c a l l a case of imprinting learning? 242 Hostility i n the Female If disturbed while she i s on the nest the female may f l i n g herself from the hole and f l y away. Very often however she s i t s tight, and does not t ry and escape. One may suppose that predation of the eggs by small mammals may sometimes take place when the female is absent. When the female remains on the nest she w i l l hiss Like a snake i f an arm i s i n -serted into the next cavity. ^ do not know i f this Hissing i s found only i n the hole-nesting species of duck, but coming from the inside of a dark deep caivty, i t may be suggested that i t has survival value. Aggressive a c t i v i t i e s (e.g. defence of the mating territory) are gen-erally the province of the male during the early part of the breeding season, though occasionally the female may threaten an intruder. However at the time of the brood territory the female alone must defend the area involved,and females are very easily stimulated into aggressive gestures when another female approaches too closely to a brood she is guarding. Whether such motivation is reduced after the young have outgrown Stage 1 , I do not know, but since only a proportion of the females with broods are in company with their own offspring by the time the downies are three weeks old (Stage 2 of Aythyini, Gollop and Marshall, 1954) , i t is probable that rights of sole possession are less important then. Multiple broods with a number offemales i n company are a frequent feature of scaup brood aggregations, and this feature i s also found i n goldeneye. Probably the individual distance of the female declines so that only when another (accompanying) female gets very close w i l l she be chased off. There 243 will be a change in behaviour therefore from the Crouch and Laying Neck on Water postures, to a quick assumption of Threat followed by a rush across the water (with the beak open and jabbing at the other bird), such as might lead into a fight in males, but i n females seldom does, for the attacked bird usually escapes. The Threat posture resembles that of the male, but in the female i t can easily be confused with the Invitation to Coitus pose. The female does not appear to open the beak and "klick" in this pose. There may be a signal difference in the time i t takes to assume the Laying Neck on Water and Invitation to Coitus postures, the latter being assumed more grad-ually. Once assumed, there is no movement except slight swimming or re-orientation to keep up with the mate, while the Threat posture is generally being constantly re-oriented in relation to another bird, at some distance in the visual field. It happens on occasions that as two females go towards each other (especially i f one has a brood) both will perform Rotary Pumping. This is often followed by the female with the brood assuming the Crouch or Threat posture. The other female then retreats or is driven off. Searching for, and Leading the Young (B.islandica) The call cue which occurs at the time of fledging has already been discussed. It was noted that i t also occurred when the female wished to call the brood together suddenly while on the water. It seems probable that i t is a derivative of the normal leading call, mwek, but with an altered pitch. 244 On one occasion upon returning to her brood a female twice uttered a nasal grunting c a l l , while on another occasion a female (with 9 downies) was disturbed by another bird, and took her brood slowly across the lake opening and closing her beak. Later I heard the nasal grunt, and l a t e r s t i l l she swam along the lake i n front of me, up and down, uttering the same c a l l as a consequence of my presence. A female searching for her young assumes the Expectancy Look (as described earlier) and utters a nasal gruntingccall. This same c a l l mwek-mwek was also used when I observed a female leading her brood of six young downies across the range to Westwick Lake on June 20th, 1955• She uttered the single note again and again at equal intervals apart as she walked ahead of tham with her head held high and far back, tt "Greeting" behaviour i n females and downies (B.islandica) Females who have brought their broods to the water often lose them to other females, and sometimes very soon after they come to water. Some females take an interest i n broods of downies that they see (particularly early ones). The females direct the Rotary Pumping movement towards other females, and towards their own young or those of other females, at this period. I only saw i t used occasionally by females before the males l e f t the lakes. It i s almost entirely confined (in the female) to use i n the Triumph Greeting Ceremony before the drakes leave. The Rotary Pump-ing movement is also seen being used by Stage 2 downies. They use i t when the female returns to the brood after a hostile encounter, or when 245 two parts of the brood come together after a separation caused by some disturbance. Often approaching females do Rotary Pumping movements and the female with the brood does the same and then goes into the Crouch or Threat position. On one occasion a female which had 12 young a l i t t l e distance away did rapid up and down (pumping) movements followed by the Threat position with neck held forwards making a clucking noise towards a single downy. She then did Jiving movements directed at i t . It dove and the female did so a few minutes later. After an encounter on one occasion the female did 3-4 Jiving movements facing the intruder (but ? directed at the downies). In Stage l a downies the Rotary Greeting movement is a much more up and down bobbing, movement (like the Head-Bobbing of drake B.albeola) than i t i s like Rotary Pumping. The greeting movements are performed when a disturbance has excited the brood, or when the female returns to the brood after driving off an intruder, and herself performs the Rotary Pumping movement towards them (however, i t i s not necessary that she should per-form i t for them to start doing i t ) . I described the act i v i t y as follows on July 7 t h , 1955 when the female returned to her brood, after diving: "The six young (which were seen for the f i r s t time today and were probably experiencing their f i r s t or second day on the lake) a l l or nearly a l l , began to indulge individually i n a fast head up and down movement (pumping) exactly comparable to the court-ship nods of Bu^leheads. This went on for about two minutes, after the adult joined them and as they swam after her across the pond. The nods 246 were interspersed with Wing-Flappings and Upwards-Stretch movements". Eleven days later there were 22 downies (between Stage l c and 2a) with one female at the same place. Four of these were' apart from the rest and I noticed that as they returned they greeted her from some distance "not by nodding but by the male courtship up and down movements with i t s antero-posterior rotary component". Next day the same group were inactive on the water except for three who were rejoining the group "doing the slow court-ing movements, and their brothers and their guard adult also did i t , but the group of eighteen had s p l i t into two parts and i t may have been due i n this case (instead) to the fusion of the two parts". On one occasion Stage 2 downies did the nodding (as of B.albeola) for three or four minutes after (and perhaps during) violent attacks by the parent on a young Coot. On turning to her young she did one Rotary Pump-ing movement. It seems clear that Rotary Pumping is derived from the normal Greet-ing displays used to maintain broken brood cohesion both by downies and their parent. 247 SUMMARY OF SIMILARITIES AND DIFFERENCES BETWEEN B.islandica and B.clangula In their locomotory, comfort and (probably) agonistic behaviour these two related species do not appear to d i f f e r . In their courtship displays there has been considerable specific differentiation from the basic pattern: — (1) Head-Up Complex: •''his display hardly differs in the two species (Figures 9 and 11). Its derivation i s uncertain, though i t greatly re-sembles the "alarmed" pose. Whether this i s intention f l i g h t i s doubtful. The context in which the display occurs is not clearly understood. (2) Kick Complex: These displays (two in each species) are basically sim-i l a r (Figures 13 and 17; 14, 20 and 21), but have become considerably differentiated. They are more clearly distinguishable from each other i n B.clangula i n which the movements are also more extreme (the head for ex-ample i s thrown much farther back, and retained there longer). These displays would indicate that B.islandica was closer to the ancestral stock, but the "pumping" displays seem to indicate otherwise. (3) Pumping Complex: While the Head-Throw (Kick. Complex) i s the most fre-quent display i n courting parties of B.clangula, Rotary Pumping i s the most common display in B.islandica. Thus both form and frequency have been a l -tered as the species have differentiated. Rotary Pumping (B.islandica) i s a more striking movement than Bowsprit Pumping (Figure 26), but i t i s a less differentiated than in .clangula i n which the Bowsprit pose i s assumed i n three possible ways. (4) The Masthead and Backwards Swimming have not been seen i n B.islandica. 248 (5) The displays of the Coition Sequence (especially Water-Flip, WLng-and Leg-Stretch, and the steaming movements) do not appear to diff e r much i n the two species. Two at least of them are derived from comfort movement s. 249 APPENDIX 1 The Illustrations of B.islandica displays i n Sawyer (1928) The captions of Plates 2 and 3 i n Sawyer (1928) should be transposed. Of the displays named i n this thesis the following are illustrated by Sawyer. Plate 3» "Attitudes during r i v a l r y " Fig. 2 Fig. 7 Kick Threat Figs. 10-11 Figs. 12-18 Figs. 19 & Head-Up Neck-Withdrawn Fight Jiving 21-23 Fig. 20 Invitation to Coitus Plate 2. "Attitudes during mating II Fig. 1 Figs. 2-3 Fig. 4 Figs. 5-7 Figs. 8-11 & Water-Flick Wing- and Leg-Stretch Wing Preen Invitation to Coitus Coition 13-16 Figs. 12 & 17 Fig. 18 Post-Coition Rotations Post-Coition Steaming 250 LITERATURE CITED Alford, C.E. Alford, C.E. Altum, J.B.T Baggerman, B., Bent, A.C.. Bent, A.C. Bernhardt, P. Boase, H. Brewster, W. Brewster, W. Brooks, A. Brooks, A. 1920. Some Notes on Diving Ducks. Br i t i s h Birds 14:106-110 1921. Diving Ducks: Some notes on their habits and courtship. British Birds 15:33-38 1868. Der Vogel und sein Leben. Miinster G.P.Baerends, H.S.Heikens and J.H.Mook. 1956. Observations on the behaviour of the Black Tern, Chlidonias n. niger (L.), i n the breeding area. Ardea 1923 Life histories of North American wild!fowl. U.S. National Museum Bull. 126 1925 Life histories of North American wild fowl. U.S. National Museum Bull. 130 1940 Beitrag zur Biologie der Schellente, Bucephala  clangula. Journal fur Ornithologie. 88:488-497• 1924 Courting display of the Golden Eye on Salt water. British Birds 18:69-71 1900 Notes on the breeding habits of American Golden-eyed duck or "Whistler. Auk 17:207-216 1911 Courtship of American Golden-eye or Whistler. Condor 13:22-30 1920 Notes on some American ducks. Auk 37:353-367 1945 The under-water actions of diving ducks. Auk 62:517-523 Brooks, A. and H.S.Swarth. 1925. A distributional l i s t of the birds of British Columbia. Pacific Coast Avifauna 17:1-158 Bruggemann, F. Carter, B.C. 1876 Ueber die Paarung der Schell-Ente. Zool. Garten 17:366-368 1952 The American Goldeneye i n Central New Bruns-wick. M.Sc. University of Maine. 251 Chapman, A. Cowan, I.McT. Daanje, A. Delacour, J. 1889 Bird-Life of the Borders. Records of wild sport and natural history on moorland and sea. London pp x i i . 286 1939 The vertebrate fauna of the Peace River d i s t r i c t of Br i t i s h Columbia. Occasional  Papers B. C. Provincial Museum. No. 1: 1-102. Birds 11-66 1950 On locomotory movements i n birds and the intention movements derived from them. Behaviour 3:48-99. 1938 La systematique des &natides et leur moeurs. Proc. 8th int.Orn.Congr. Oxford 1934 Delacour, J. and E. Mayr. 1945. The jPamily Anatidae. Wilson Bulletin 57:3-55-Delacour, J. 1956 The Waterfowl of the World. Vol.2, p.17-Country Life, London. Dementiev, G.P. and N.A.Gladkov. 1952. The Birds of the Soviet Union. Vol.4 State Publishers "Soviet Science", Moscow. Gilpin, J.B. 1880 On the semi-annual migration of sea fowl i n Nova Scotia. Proc. and Trans. Nova Scotian  Inst. Nat. Sci. 5:138-151 Gollop, J.B. and W.H.Marshall. 1954- A guide for aging duck broods i n the f i e l d . . Mississippi Flyway Council Tech- nical Section, pp. 14 Gunn, D. Heinroth, 0. Hochbaum, H.A. Hochbaum, H.A. Hoogerheide, C. Howard, H.E. Jackson, Mary F. 1939 On the courtship display of the Golden-eye. British Birds 33:48-50 1911 Beitrage zur Biologie, namentlich Ethologie und Psychologie der Anatiden. Verh. V.Tnt. Orn.Kongr. Berlin 1910:589-702 1944 The Canvasback on a Prairie Marsh. The American Wildlife Institute, Washington, D.C. 1955 Travels and Traditions of.Waterfowl. Univer-sity of Minnesota Press, Minneapolis. 1950 De Eidereenden Somateria mollissima L., op Vlieland. Ardea 37(3/4):139-161 1920 Territory i n Bird Life. London 1952 M. S. 252 Jackson, Mary F. Johnsgard, P.A. Kortright, F.H. Lilford, Lord Lloyd, L. Lorenz, K. Macartney, W.N. Mayr, E. McKinney, D.F. Millais, J.G. Munro, J.A. Munro, J.A. Munro, J.A. Munro, J.A. Munro, J.A. 1957 1955 1942 1895 1854 1941 1918 1942 1953 1913 1918 1935 1939 1945 1949 Munro, J.A. and W.A.Clemens Ph.D. thesis (in preparation). University of British Columbia, Vancouver, Canada. Courtship act i v i t i e s of the Anatidae i n eastern Washington. Cpndor 57:19-27 The Ducks, Geese and Swans of North America. The Stackpole Company, Harrisburg, Perm, and Wildlife Management Institute, Washington, D.C. Notes on the birds of Northamptonshire and neighbourhood. London, 2 vols. Scandinavian Adventures. Vol. 2 London. Vergleichende Bewegungsstudien an Anatinen. Journal fur Ornithologie 89:194-294 (Translated by C.H.D.Clarke i n the Avicultural Magazine, 1951-52.) Goldeneye duck carrying young. Bird Lore 20(6):418-419 Systematics and the Origin of Species. Columbia University Press. New York Studies on the behaviour of the Anatidae. Unpublished Ph.D. thesis, University of Bristol British Diving Ducks. 2 Vols. Longmans, London The Barrow Golden-eye i n the Okanagan Valley i n British Columbia. Condor 20:3-5 Recent records from British Columbia. Condor 37:178-179 Studies of Water-Fowl i n British Columbia, No.9 Barrow's Oolden-eye, American Golden-eye. Trans. Royal Canadian Institute Vol.22(Pt.2):259-318 The birds of the Cariboo Parklands, British Columbia. Canadian Journal of Research, D 23:17-103 The Birds and Mammals of the Vanderhoof Region, British Columbia. American Midland Naturalist 41:1-138 . 1931. Waterfowl i n relation to the spawning of herring i n British Columbia. Biol.Bd.of  Canada Bull. 17:1-46 253 Murie, A. Pearse, Theed Phillips, J.C. Saunders, W.E. Sawyer, E.3;. Scott, P. Scott, P. and H. Scott, P. and J. Scott, J.P. and Siren, M. Snyder, L.L. Stonor, CR. Swarth, H.S. Tovjnsend, C.W. Townsend, C.W. 1946 Observations on the birds of Mt. McKinley National Park, Alaska. Condor 48:253-261 Unpublished observations 1925-26 A Natural History of the Ducks. Vol. 3 (1925), Vol. 4 (1926). Houghton Mi f f l i n Co., Cambridge, Mass. 1916 Bird protection i n Canada. C o n s e n a t i o n of f i s h , birds and game. Comm.Cons.Can. The Methodist Book."and,.Publishing,House, Toronto. 1928 The Courtship of the Barrow's Golden-eye. Wilson Bulletin 40:5-17 1951 Key to the Wildfowl of the World. Severn Wild-fowl Trust, Slimbridge, Glos., England. Boyd. Ed. 1954- Seventh Annual Report of the Wildfowl Trust (1953-54). Country Life Ltd., London Fisher 1953 A Thousand Geese. Collins, London E.Fredericson 1951 The causes of fighting i n mice and rats. Physiol. Zool. 24:273-309 1952 Undersb'kningar oVer knipans, Bucephala clangula, fortplantningsbiologi. (Studies on the breeding biology of the Goldeneye, Bucephala clangula) Papers on Game Research No. 8. Finnish Game Foundation, Helsinki. 1953 An apparently hybrid golden-eye. Wilson  Bulletin 65:199 1940 Courtship and display among birds. Country Life Ltd., London (also British Birds 34:202) 1926 Report on a collection of birds and mammals from the Atlin Region, northern British Columbia. Univ. California. Publ. Zool. 30(4):51-162. 1910 The courtship of the Golden-eye and Eider Ducks. Auk 27:177-181 1923 The music of the Golden-eye (GlaUcionetta clan^ula americana). Bull. Essex County (Mass-) Ornith.Club. 1922. 4:6-9 254 Tinbergen, M. 1951 The Study of Instinct. Oxford Tinbergen, N. 1953 The Herring Gull's World. Collins, London Walcott, C. and B. Dane. 1957• (Paper on the behaviour of Bucephala clangula and B.albeola, in preparation). Witherby, H.F. Editor 1943- The Handbook of British Birds. 2nd (Revised) Impression. H.F. & G. Witherby Ltd., London. Vol. 3 

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.831.1-0106250/manifest

Comment

Related Items