THs paper describes the courtship and agonistic behavior of the Com- mon Grackle. The probable derivation, biological significance, and, in some cases, motivation and function of displays are discussed. Basic information on life history is available in Petersen and Young (1950) and Bent (1958). The Common Grackle has a complex display repertory. This, when fully understood, will greatly contribute to our knowledge of social be- havior in general. These observations can also be compared with similar studies of other icterids, thus furthering our understanding of evolution in this group. This study was conducted from August 1957 to May 1960. Most ob- servations were made at two colonies of about 15 and 20 pairs, at Ithaca, New York. In addition, birds were observed in captivity. The Common Grackles of the Ithaca area are intergrades between Quiscalus quiscula versicolor and Q. q. quiscula (see Huntington, 1957). DEFINITIONS Display.--As used herein, refers to "... those peculiarly standardized and often exaggerated performances, including all vocalizations and many movements and pos- tures, which have become specialized and modified as social signals or releasers . . ." (Moynihan, 1955). Ritualization.--This is the process of display evolution. Intention Movement.--This term includes various incomplete and low-intensity movements, which, as explained by Heinroth, reveal to the observer what the animal is "intending to do" (Marlet, 1956). Tendency.--Is used in the sense of Hinde (1955-1956) as given by Mar!er (1956): "... the readiness to show a particular type of behaviour as observed under natu- ral conditions." Intensity.--Here means the degree of exaggeration in form of a movement. Agonistic.--Refers to behavior concerned with escape and attack. UNRITUALIZED AGONISTIC AND SEXUAL BEHAVIOR Attack. In the Common Grackle this consists of biting, pecking, and scratching another bird. Intention movements for these activities and attendant locomotion are included in the definition. A peck may be given with the bill either open or shut. Pecks seem weak and apparently never harm the recipient. In fact, some pecks in- volve no contact. A running attack usually consists of holding the body in a horizontal position while running at the opponent. Gaping and/or wing- and tail- spreading may also occur. Flying at another bird is a frequent form of attack. Flying at or run- ning at usually result in the opponent leaving. A bird may fly up from the ground more or less vertically, extending its legs upward while orient- ing toward its opponent. In fighting, the birds appear to bite and scratch each other repeatedly, and feathers are sometimes dislodged. When the birds are fighting in the air, they always fall downward rather than rising. Escape. This is any avoidance movement stimulated by the object avoided. Copulation. This refers to all cloacal contacts between male and fe- male (see Figure 3). Mounting, a more inclusive term, refers to the male standing on the back of the female, whether or not copulation occurs. During copulation the male's feet are placed on the female's upper back, and his tarsi are flexed strongly. The male's bill is pointed down- ward and his tail lowered to one side of hers, facilitating cloacal contact. At this point he flutters his wings rapidly, copulation taking but a sec- ond or so. The female holds the bill pointed upward with head withdrawn be- tween the shoulders. Her breast is lowered, tarsi flexed, and tail held upward and twisted to one side. DISPLAYS OF THE COMMON GRACKLE AND THEIR PROBABLE DERIVATIONS Displays are identified on the basis of their standardized and exag- gerated form. Their communicatory function has been shown in several species, e.g., Hylocichla spp. (Dilger, 1956) and Larus spp. (Tinbergen, 1959). Head Held Up.--This always involves pointing the bill up at any angle above the horizontal and is given only in the presence of another bird. Other components occur in a wide variety of combinations. The bill may be either open or closed and may sometimes be tipped up and down slightly. The nictitans is often blinked arhythmically, and the eyelids are sometimes partially drawn over the eye. The neck is held in any position from withdrawn to fully extended and the body anywhere from horizontal to nearly vertical. The breast plumage is usually sleeked but is sometimes relaxed or even fluffed while the head feathers are almost always sleeked, sometimes extremely so (see Morris, 1956, for feather posture terminology). The wings are usually folded, but their tips often droop below the level of the under tail coverts. The birds walk in vari- ous speeds and directions or remain stationary. Their orientation varies anywhere from directly toward to directly away from one another. Shifts II 21 31 41 49 61 71 in position of body parts, orientation, and speed and direction of move- ment may occur frequently and rapidly. The Head Held Up is probably derived from an upward flight inten- tion movement. Originally gaping was probably an intention movement to bite (see Marler, 1956). The Head Held Up serves as a distance increasing display (see Tin- bergen, 1959). Both nictitans blinking and eye closure hide the conspic- uous yellow eye of the bird from the opponent. Since the eye is usually wide open in attack, these are probably "reverse movements" (see Mar- let, 1956). Arhythmical Blinking.--This movement of the nictitans occasionally occurs by itself in agonistic encounters and is also frequently a compo- nent of other displays. The Arhythmical Blinking of the nictitans may be a ritualized form of rhythmic blinking. Rapid but rhythmic blinking was noted in several situations. During single one-minute observations of 12 hand-held birds, each blinked its nictitans from 3 to 63 times (i 35.4). An incubating female gave 70 blinks in one minute when I ap- proached her closely. Three caged birds aware of my presence gave 25, 60, and 68 blinks in one minute. Although the rate of nictitans blinking in undisturbed, lone birds was not measured accurately, they almost al- ways blinked more slowly than those described above. This movement, at least at higher rates, is associated with thwarted fleeing. In the Pied Flycatcher, Muscicapa hypoleuca, there was an increase in the rate of beating of the nictitating membrane when the bird "froze" in reaction to a hunting hawk (Curio, 1959). Another possible derivation of Arhythmical Blinking is from the strong closure that occurs when the eye is touched or the bird "anticipates" its being touched, as when feeding young. Such protective closure might also occur during fighting. Waa.--This vocalization of females occurs commonly in a variety of agonistic encounters. Figure 1. Sequence from 16 mm motion picture film showing Head Held Up and Ruff-out Squeak displays in a male. Ten frames equal 0.33 seconds. Frame 1, male directing the Head Held Up toward a male behind him. Frames 11 and 21, beginning of ruffling and spreading components of the Ruff-out Squeak. Frames 31 and 41, male pivoting as the squeak component begins. Frame 49, the complete form of the RuII-out-Squeak. The male has risen on his legs and is beginning to step forward. Frame 61, diminishing of the ruffling and spreading components as another Head Held Up begins. Frame 71, Head Held Up being given while bird is stationary. The opponent is to the right of the picture. Note the partial eye closure and nictitans blinking in frames 11, 21, 31, and 61. Also note the shifting of the iridescent high- lights throughout. TABLE 1 VARIATION IN INTENSIT3/4 NICTITANS BLINKING AND ORIENTATION OF THE MALE RUFF-OUT SQUEAK DISPLAY IN THREE CONTEXTS Context Intensity Blinking o! Orienting nictitans toward other Low High Absent Present Absent Present Male alone Number 18 0 13 2 Per cent 100 0 87 13 Not applicable With female Number 27 5 23 4 22 7 not soliciting Per cent 84 16 85 15 78 22 With male Number 2 48 0 36 8 41 Per cent 4 96 0 100 16 84 Snarl.--A male call similar to the Waa but typically higher pitched and clearer. It occurs in many situations: in hand-held birds, lone, undisturbed birds, during mobbing, in bouts of "song," but was never noted during intraspecific agonistic encounters. Ruff-out Squeak.--This begins with a simultaneous spreading of the wings and tail and ruffling of the contour feathers (Figure 1). As spread- ing and ruffling reach their maximum, the bird rises up on its legs, utters the Squeak, flashes the nictitans arhythmically, and may take a step or two forward. (The entire display lasts two to four seconds.) The Squeak component is the familiar "rusty gate" sound usually written as "readle- eek," "re-lick," or "scudle-eek." The sound of the Squeak seems consis- tent for each male. The Ruf]-out Squeak varies in several ways. A male once gave a out Squeak with no apparent vocal component, and on several other occasions it was barely audible, sounding like "screech," "eek," and "peeku." Table 1 summarizes the form of Ru]f-out Squeaks given toward fe- males that were not Soliciting (see p. 59), toward other males, and by lone males. A comparison of the intensity (degree of ruffling and spread- ing), the presence or absence of nictitans blinking, and the orientation revealed some marked differences. Lone males always performed low- intensity ruffles and spreads and rarely flashed the nictitating mem- brane. Those given toward nonsoliciting females were roughly the same as those given by lone males except that they were sometimes of higher intensity, and the orientation component was occasionally present. Those performed toward other males almost always involved a greater ruffling and spreading, always included the nictitans blinking, and were usually oriented at the other bird. The RuJJ-out Squeak is often associated with other displays. It some- times alternates with Chacks, occasionally with Snarls, and rarely with Peeps. One individual gave RuJj-out Squeaks followed immediately by Snarls. Once, another male performed the Ru]]-out three times but in- stead of Squeaking, gave the Snarl. The Head Held Up often occurs before and/or after the Ru]J-out Squeak (Figure 1). Ru]J-out Squeaks develop slowly in the young. Juveniles in early summer, when among other grackles at food piles, perform an activity that is probably an ontogenetic precursor of the Ru]]-out Squeak. This incorporates ruffling of the contour feathers and a strong blinking of the nictitans but no vocal component. The Squeak shows similarities to "song" of other passerines. An im- mature male gave gurgling noises (between Ru]]-out Squeaks) that may be similar in form to the "warbling" during early "song" of the Song Sparrow, Melospiza melodia (Nice, 1943). Juveniles in winter also have short intervals between utterances, another characteristic of early "song" of the Song Sparrow (Nice, 1943). Another well-known characteristic of passerine "song" is that it occurs in lone birds as does the RuJ]-out Squeak. The Ru]]-out Squeak, at least at close ranges, is probably a threat display since it frequently alternates with the more aggressive Head Held Up threat and during the last part of the nesting season may pre- cede or even accompany attack. The Ru]J-out Squeak seems to have been derived from motor patterns characteristic of attack, combined with some patterns associated with thwarted fleeing, and a vocalization of undetermined history. The spread- ing of the wings and tail resembles the spreading sometimes accompany- ing a running attack. Rising up on the legs is probably derived from springing up to fly. Steps and pointing the bill toward the opponent are seemingly little ritualized components. Ruffling the contour feathers is more difficult to interpret, but is probably derived from the fluffing often seen if fleeing is thwarted in other situations. Ru]]-out Chuga. This display is typical of females. The visual com- ponent resembles that of the male Ru]]-out Squeak, although it is typ- ically less exaggerated. This display includes an unmusical vocalization sounding like chuga or, rarely, like chuga-squeak. One female gave RuJ]- out Chugas followed immediately by Waa. Occasionally, females give Ru]]-out Squeaks, but the female type was observed in only one male. Chack.--This call is characteristic of the grackle in many situations, particularly during flocking, mobbing of predators, and in lone birds. It has several inflections. Tail Flicking.--This is a preflight movement that occurs as a reaction Figure 2. A "leader flight" showing three males behind the female. The lower four birds are viewed from below. Note the V-tails in the males and the aerial Ruff-out Squeak by the center male. to various disturbances. It is an up-down movement (see Andrew, 1956). V-tail.--This display is typical of males throughout the breeding sea- son. The bird depresses its central tail feathers to form a V (Figure 2). The degree of depression varies from barely perceptible to deep. The V- tail is associated with flying, walking, and perching males, and on four occasions was noted in flying females. Head Down.---This display always precedes copulation. In the highest intensity the bill is pointed downward and held in this position through- out, while the wings are lifted to the horizontal and extended slightly and the tail is spread, with some V-shape perceptible. The contour feathers are ruffled. From this position the display takes place in two phases, the first involving a Squeak vocalization and an orientation to- ward the female, and the second, bowing downward and then turning to one side accompanied by a Peep note, wing fluttering, and tail raising. The two phases are completed every two seconds in rhythmic alternation. The higher intensities of the display have many variations. The Squeak phase sometimes differs from the above by: (1) increased ruffling of the body plumage and spreading of the wings with each Squeak; (2) rising up on the legs with each Squeak; (3) Squeaking very weakly (on one occasion a male gave the Chuga). During the Peep phase the plum- age is sometimes sleeked and the number of Peeps may vary from one to eight. Occasionally, a bird raises and extends its wings vertically and then flutters them. Nictitans blinking and eye closure may occur. Bow- ing between phases was only noted in birds that were displaying in places where the movements of their feet were restricted, e.g., on limbs or roof-peaks. The Head Down has a "typical intensity" (Morris, 1957) in which the male holds his bill down and ruffles and spreads (Figure 3) while giving a rhythmically alternated series of Squeaks and Peeps. A frequent lower intensity of the Head Down only omits Peeps. At other times the male advances toward the female in a posture resembling the commonest Head Down, but without vocalization. An unusual, but simplest, Head Down consists of only pointing the bill downward. RuJJ-out Squeaks may grade into a typical Head Down display. Aerial RuJJ-out Squeaks, although seen in lone males as well as those near females, are apparently linked with precopulatory display. These are almost always combined with a deep V-tail. Up to four may occur in rapid succession, sometimes alternating with Snarls or, rarely, a Peep. The Head Down combines components of the Solicit, the RuJf-out Squeak, and the V-tail, with the addition of bill-lowering. Its derivation is obviously very complex. The pointing down of the bill is most prob- ably derived from orienting toward the female during copulation. The occasional flapping of the raised and outstretched wings may also be derived from the act of copulation, i.e., the balancing movements of the mounted male, but on the other hand may be an intention movement of fleeing. The Peep phase of the Head Down is very probably derived from the Solicit and may contain the directly functional components of breast-lowering and tail-raising, as well as the more ritualized compo- nents of Peeping and wing-fluttering. Solicit. This display resembles the position of the female during copulation. The Solicit may vary in several ways. Head-raising, wing- raising, and tail-raising may vary in degree, but are usually given to- gether. Peeping, wing-fluttering, breast-lowering, and tarsal flexion are also often present in various degrees, and Peeping is sometimes uttered while in the "normal" resting posture. As Marler (1956) has pointed out, the components of the female so- liciting posture are mostly functional, aiding copulation by giving the male a place to stand, shifting the female's center of gravity, and expos- ing her cloaca. Wing-fluttering and Peeping, both ritualized components, II 41 51 II 41 51 resemble the behavior of young begging for food. A lone male gave a display on two occasions consisting of raising the tail to about 20 degrees, fluttering the wings, and uttering Peep notes. This display is similar to the female Solicit. PAIR FORMATION Pair formation of the Common Grackle begins in colonies as soon as the first females arrive. Two characteristic behavior patterns, flights and mutual displays, occur between each female and a group of males. Pair-formation flights may be roughly divided into three types: leader flights, chases, and together flights. Each type may grade into the other. In leader flights (Figure 2), the males are usually in a compact group, occasionally straggling, behind the female, and the flight speed is mod- erate to slow. In chases, the males are behind the female, the flight speed is fast, and the female sometimes dodges. In together flights, the speed is also moderate to slow, and one or more males are above, below, at the side of, or even ahead of the female. Pair-formation flights usually begin when a female flies from a tree with males following. Such flights are strong stimuli to males since they rarely remain when the female flies. Once in the air, males usually fol- low a female until she lands, although occasionally males land in one tree and the female in another. On a few occasions males flying by in another direction were "captured" by a nearby pair-formation flight and then followed along with the original males. Rarely a male took off from a tree first and a female followed and was followed in turn by another male or males. These flights typically involve horizontal arcs and vary from about 15 to 500 meters or more in one direction. In this way the birds fly a long time and never leave the area of the future colony. However, rarely, flights take a straight course, and the birds go out of the colony area. Figure 3. A sequence of precopulatory behavior, ending in copulation, drawn from 16 mm motion picture film. Ten frames equal 0.63 seconds. Numbers at lower left of each picture refer to frames. Frame 1, male performing Head Down display (including Squeaks and Peeps). Frame 11, male moving to the left but continuing to orient toward the female as she begins to flutter her wings. Frame 21, male in pre-mounting position with right eye closed. Female here beginning to adopt the Solicit by lifting tail and lowering breast. Frame 31, male approaching female more closely while still orienting to her head. She is now in the full Solicit. Frame 41, male mounting. Frame 51 shows copulation. The left column shows top views extrapolated from the side views on the right. Usually the female Solicits before the male begins the Head Down display (see text). Flights may last 20 minutes or more but are usually less than a minute. They often take place at treetop level, but are sometimes higher. Vocalizations given during pair-formation flights are also commonly heard in other contexts. Males often give Ru]]-out Squeaks in flight, usually accompanied by a modified, slightly arhythmic wing beat or a glide. Rarely do they Snarl. The females frequently Waa in flight, often as a reaction to apparent thwarting situations, such as when leaving the tree, when a male glides toward her giving the Ru]]-out Squeak, and when males catch up to her after takeoff. Females occasionally Peep, Chack, Chuga, and Chuga Waa. Many flights, however, are without vocalizations. The V-tail is typical of pair-formation flights, occurring in exaggerated form during 37 per cent of 63 flights, and to some degree in an additional 31 per cent. Females show an active interest in males. They occasionally follow males when they take off, but the males usually end following the female. Also, many flights have a male or males in front with the female follow- ing. On one occasion, a leader-type flight was interrupted when the two males engaged in a brief midair fight whereupon the female flew down toward them, and the flight was then resumed. Males were seen fighting other males during flights on two other occasions and sometimes cut one another off. The proportion of males to females during 196 flights was 2:1 (129), 3:1 (47), 4:1 (17), 5:1 (1),and 6:1 (2). Flights containing more than one female were seen on eight occasions, and most of these were together flights. During these flights the propor- tion of males to females was 5:2 (1),4:2 (1), 3:2 (1), 3:3 (1), and 2: 2 (4). In each case where landing was observed, the groups broke up into single females with a male or males. Pair-formation flights are probably ritualized chases (leader flights almost certainly so) with fighting between the males reduced to a mini- mum, and fighting with the female now absent. Ritualized aspects are the frequent slow to moderate flight speeds, the V-tail, females following males, and the fact that males never catch the females, although they obviously could. A pair-formation flight ends when the birds land. At this time, the birds engage in mutual display, which typically consists of males giving Ru]]-out Squeaks and a female giving Ru]]-out Chugas. The Ru]]-out Chuga is the commonest female response, but they also give Ru]]-out Squeaks, Waas, Ru]]-out Chuga Peeps, and various forms of the Solicit. Mutual display is not regular, but tends to occur in bursts with short pauses. One male often Ru]]-out Squeaks more often than do others, and different males may initiate bursts of this display. During mutual display, males sometimes stimulate other males to leave by supplanting or fighting with them, or by advancing in the Head Held Up. Thus, there is active and apparently successful displaying and at- tacking in competition for females. In fact, males were never seen actu- ally to fight in the absence of a female, although they often perform Ruf]-out Squeaks and Head Held Ups at one another. Sometimes they also supplant and chase one another. DEVESE OV TaX NEST-SXTE AREA Nest-site selection by the female begins after she is paired. During this period females walk about the colony area examining old nests. Males sometimes exhibit similar behavior. As soon as nest-site selection begins, the pair defends the site against other pairs. Displays occurring during nest-site defense are essentially the same as those during pair formation; the Head Held Up, the Ruff-out Squeak, and the Ruff-out Chuga are often directed at other pairs. The Waa is the most frequent female call, and females sometimes Waa continuously for minutes. Supplanting and fighting encounters were recorded during a two-week period from the beginning of nest-site selection to the completion of most nests. In 31 heterosexual encounters the male attacked the female in 27 cases and the female attacked the male in only 4. Thus, it seems that males are much more prone to attack females than vice versa. However, females attacked females (22) almost as often as males attacked males (29). In 43 of the male attacks, an attempt was made to observe the coincident behavior of nearby females. During 17 male attacks, the females were either not visible or sitting quietly. On the other hand, 26 of the male attacks took place when there was an interaction (displaying or attacking) between his female and another male or female. These observations indicate that males not only defend an area around the nest site but also their females against other males and females. Males defend their females away from the colony as well. Observations of birds in or very near the colony compared with those of birds farther away reveal interesting differences. Often from two to four pairs silently approach the nesting area together. The females begin giving Waas when they are close, and typical nest-site defense behavior ensues among them when they land. Similarly, pairs that leave the nest-site area after being attacked are sometimes joined shortly there- after by the same pair that had attacked them in the colony, but now perch peacefully nearby. On a few other occasions, one bird of a pair attacked or displayed at one of another pair, and then both pairs silently left the colony together. When more than one pair forages near the col- ony, they often do not interact at all, or at most give Ruff-out Squeaks and Head Held Ups, unless a male displays at, or comes very close to, another's female. Only twice during the two-week sample period did a female attack another female more than three meters from the nest-site area. Thus, as in most birds, their location has important influence on their agonistic behavior. However, nest-site defense behavior may con- tinue briefly outside the colony. Females that have been attacked may leave the area still giving Waa calls, and if birds have begun a fight in the nest site area, they sometimes continue down to the ground. Such fights are brief, however, and the birds then return to their nest sites. Although agonistic behavior is quite pronounced in the nest-site area, the periods during which birds are attacking other birds are infrequent compared with the periods of relative quiet. Sometimes such quiet peri- ods are broken by encounters involving up to four pairs producing a confusing sequence of flying about in front of the nest-site areas with some supplanting and much Waaing by the females. Females sometimes display in unusual ways during nest-site encoun- ters. Various forms of the Solicit often occur, apparently stimulated by nearby agonistic encounters. Females also give a complex of displays that normally occur singly, such as a Ruff-out Chuga Waa Peep, a Ru-out Chuga Waa Peep Waa, and a Solicit with Waaing in place of Peeping. Similar complexes of displays occur in females leaving the nest during incubation (see below), and in females a short distance from displaying males at a food source. INTERACTIONS OF THE PAIR Interactions of the pair were studied during nest-site selection and nest-building. The basic pattern of behavior is again very similar to that shown during pair formation; i.e., the male usually follows the female when she flies to and from the nest site, and, when perched, they often engage in mutual display. In typical paired flights, the female leads the male at moderate speeds. He often gives Ruff-out Squeaks and she often gives Waas. The male V-tail is very frequent when he is following his female (especially deeply in 73 per cent of 101 flights and to some extent in 19 per cent more), but was seen in the female only once in this context. Paired flights differ somewhat from pair-formation flights in that the former are typically shorter in duration, do not involve much "arcing," and are below tree- top level. There were several uncommon variations of the typical pattern. These included the female following the male, rapid wing-beating (moth flight) by the female, chasing by the male, and diving of the male at the female with Ruff-out Squeaks. Vocalizations occasionally heard were Chugas and Peeps uttered by the female and Chacks given by both birds. Although males often perch in or near the nest-site area, they some- times perch as far as 25 meters away. When females leave the colony, their mates usually follow. Males probably recognize their mates by visual clues such as physical appearance and location. On the return flight, females are usually followed by their mates. Mutual display between the pair, the male typically performing a RuJj-out Squeak answered by a RuJj-out Chuga from the female, occurs throughout nest-site selection and nest-building. This sequence may be performed several times in succession. On three occasions, when a male foraged on the lawn and seemed hidden from the female's view, she answered his Squeak immediately with a RuJJ-out Chuga. Since the Squeaks of males perched in and around the colony seemed to be dis- tinctive for each individual, the female probably is able to recognize and respond to the Squeak of her mate. Females rarely give other responses, such as the Waa, the RuJJ-out Squeak, and Peeps, to the male's Squeak, and sometimes do not even respond. Mutual display is occasionally begun by the female. Thus, each male usually keeps in close contact with his mate by perch- ing near her, by following her, and by engaging in mutual display. Such behavior permits the male to be near his female when she solicits copu- lation, and to be frequently away from other males that often attack or threaten when he attempts copulation. It also enables him to defend her against the copulatory attempts of other males. The probable stim- ulation afforded by mutual display and following may also help to bring both sexes into synchronous reproductive condition, and following may also prevent polygamy. SEXUAL BEHAVIOR The precopulatory display of females (Solicit) and males (Head Down) has been described, as have mounting and copulation. Sequences involving the Head Down in and around the colonies were recorded in as much detail as possible every time they were observed. These obser- vations were analyzed in terms of their location, form, temporal varia- tion, function, and probable motivation. The Solicit sometimes appears "spontaneously," i.e., in the absence of a male. It also differs from the male Head Down in that it is often stimulated by, rather than suppressed by, the agonistic displays of nearby males. TABLE 2 FEMALE RESPONSES TO THE MALE HEAD-DOWN DISPLAY FROM NEST-SITE SELECTION (EARr,3/4 APRre) TO EO0 LA3/4INO (LAT APmZ), EXPRESSEr n PERCENTAOES Or Ta TOTAL (N) OBSERVATIONS OR TARES PERIOrS Allows Attacks Flies or No overt Some Ru]/-out Waa Date N mounting male waIks responses so!icit Chuga call away April 3-10 49 0 47 27 12 2 6 6 11-18 72 19 35 21 8 3 3 11 19-25 41 44 20 27 5 2 0 2 The Head Down occurs in many places, particularly in trees and on the ground and buildings in the vicinity of the colony, as well as occa- sionally in the nest-site area. This display is always directed at a female. It is rarely seen if a second male is near, in which case both males usually then give agonistic displays to one another. Sequences were recorded from the beginning of the male Head Down toward the female until its termination. Such sequences were always dis- crete units, interrupted by a longer period of other activities. In a typical sequence ending in mounting, the female first Solicits; the male then walks rapidly toward her giving a Head Down of increasingly exagger- ated form until he reaches her and mounts. However, more frequently males give the Head Down to females that are not Soliciting. Such se- quences rarely end in mounting. A sequence, which is unusual because the female Solicited after the male displayed, is shown in Figure 3. When the male gives the Head Down toward a female that is not Soliciting, the female typically attacks, walks or flies away, or gives no apparent response. Attacks by the female on the displaying male are always direct running or flying attacks, often ending in a peck at him. As many as 13 successive attacks by the female were seen. When the female flies or walks away from a displaying male, she usually does not vocalize or move rapidly, indicating more that she is "not interested," than that she is fleeing from the displaying male. The first response of the displaying male to the attacks of the female is to fly off a short distance, turn away, or, most frequently, to continue to display. He is never seen to threaten or attack his mate before or after the Head Down despite her repeated attacks. Sufficient sequences were observed during one spring to determine roughly the changes that took place as the breeding season progressed from nest-site selection until a few females were incubating (Table 2). During this three-week period the most marked change occurred in the frequency of sequences that ended in mounting, from none in the first week to 44 per cent in the last. Attacks by the female, on the other hand, decreased during the same period. The tendencies of the sexes during the sequences seem to be different. The male is always ready to mount if the female permits, but shows only a moderate tendency to escape from her, and no evidence of a tendency to attack her. The female, on the other hand, shows a strong but decreas- ing tendency to attack the male, little tendency to flee from him, and an increasing tendency to allow copulation. During the nonbreeding season the male is dominant over the female. The fusion of motor patterns characteristic of at least four separate displays in the Head Down display suggests very strong selective pres- sures. Apparently it signals the sex of the male, since it is never given by females. It also indicates the male's readiness to copulate, since it always precedes copulation. It may signal lack of aggression because it incorporates the Head Down component (the "reverse movement" of the ttead Held Up threat) and the female Soliciting call, and never causes the female to flee. Since the Head Down rarely leads to Soliciting and subsequent mount- ing, its function may be primarily one of long-term, rather than short- term, stimulation of the female. However, the Head Down always pre- cedes mounting and may stimulate the Soliciting female to maintain her posture until the male mounts. When the female is not Soliciting and the male gives the Head Down near her, it often stimulates her to attack. Since the female was never observed to attack her mate in any other situation, it seems that the Head Down component of his precopulatory display is the factor that leads to her attack. The function of Soliciting is obvious. It stimulates the Head Down and mounting. When the female becomes receptive, she merely has to Solicit in sight of the male; the likelihood of copulation is then very great. The attacks of unreceptive females on displaying males, on the other hand, might function in the long-term stimulation of both members of the pair. Both sexes have typical behavior patterns following mounting. The male often resumes the Head Down while the female often gives whirring movements of the wings that are seemingly identical with those normally occurring when the plumage is wet. Such wing-whirring occasionally follows Soliciting when the male does not mount. Occasionally, whirring of the wings does not follow mounting directly, but occurs after the fe- male has first either attacked the male, flown away, or Solicited again. Males never whir their wings except when they are wet, but occasionally give two other comfort movements (bill-wiping and body-shaking) im- mediately after ceasing the Head Down. BEHAVIOR OF THE FEMALE DURING INCUBATION During the incubation period females give various displays upon leav- ing the nest. The female returning to the nest only gives the Waa, which is probably a reaction to passing near others' nest sites. The frequency of vocal and visual displays in a sample (127) of 12 lone females leaving nests was as follows: Silent (10), Waa (51), Chack (12), Peep (2), and various combinations of Waas, Chugas, Squeaks, Peeps, fluffing, moth flight, glides, and V-tail (52). The stimuli that probably act on the female while she incubates are the constant Ruff-out Squeaking of males and the vocalizations of other females leaving the nest. Apparently the tendencies aroused by these stimuli cannot be expressed because of the overriding tendency to incu- bate. As soon as the female leaves the nest, however, the other tenden- cies are manifested. An alternative explanation is that these are so-called "vacuum activities." REACTIONS TO PREDATORS AND SUDDEN NOISE When the colony was disturbed by sudden, loud noises, such as slam- ming doors, the birds often flew away together as a flock. Such flocking was not fortuitous, since the direction of flight of the flock differed in almost every case. Mobbing was observed as a reaction toward a perched Cooper's Hawk, Accipiter cooperii, a flying Red-shouldered Hawk, Buteo lineatus, a perched Barn Owl, Tyto alba, a stuffed Great Horned Owl, Bubo virgin- ianus, several flying Common Crows, Corvus brachyrhynchos, and man. Almost all of these predators induced flocking and abundant Chack calls, and occasionally Tail Flicks, Snarls, and Waas. The birds chased or dived at the crows and dived at the horned owl. My approach to the nest during construction induced flocking and Chacking, but after the young were fledged, strong Tail Flicking occurred also. Bent (1958) actually had his hat knocked off by a grackle when at the nest examin- ing young. The Red-shouldered Hawk caused a large group to fly out of the colony together down into a pine tree where they perched silently and motionless for over a minute. They sometimes chased and dived at gray squirrels (Sciurus carolinensis), but never flocked in reaction to them. Mobbing always involves attack and escape as well as flocking tenden- cies, which appear to vary seasonally and with the species, location, and activity of the predator. The mobbing reactions are often extremely rapid and contagious, par- ticularly flocking and Chacking. However, the means by which an entire colony becomes alerted and often manages to fly from the colony site up to the same tree within a few seconds are unknown. DISCUSSION The general pattern of reproductive behavior in the Common Grackle may be an adaptation to sociability and colonial nesting, which results in a great reduction in territory size. Thus, pair-formation flights and mutual display are probably adaptations in lieu of individual mate terri- tories into which females could enter and pair undisturbed by other males. The RuJj-out Squeak in which "song" is combined with visual display would be an advantage where close contact between birds is very frequent. Such displays occur also in other, closely related icterids that associate together during the breeding season, such as Euphagus cyano- cephalus (Williams, 1952), Cassidix mexicanus, and C. major (Selander and Giller, 1961) but are apparently absent in Euphagus carolinus (per- sonal observation), which nests singly. The Head Held Up threat, with its many variable components, presumably signals rather precisely the varying degrees of readiness to attack or escape. Such precision of expression would serve to reduce fighting, stress, and time wastage in a species where there is much close contact among individuals. The im- portant role of the female in defending the nest-site area is an obvious result of the reduction of territory size. Since the female is not restricted to a territory, the mate must follow and defend her against other males that may attempt coputation. By following his female, the mate also greatly increases his chances of successful coputation without interfer- ence from other males. Mutual display between members of a pair and the tong-distance recognition of the male's Squeak by his female may also be adaptions to colonial nesting. In addition, many of the above activities may serve to advertise the colony to new birds and probably help stimulate reproductive physiological readiness in the pairs. Fighting in which feathers are dislodged occurs only in males that are in competition for females and in females in competition for nest sites, i.e., in those situations where the survival value of success would pre- sumably be greatest. Such fights are apparently almost harmless to the combatants as they are in many species (cf. Lorenz, 1952; Ditger, 1960), which could easily kilt or seriously injure one another. There are several types of sexual dimorphism in the species. The V- tail is typical of males and is probably a tong-range sex-recognition char- acter. The distinct vocalizations of each sex accompanying the Rujj-out are another form of sexual advertisement. The Soliciting is typical of females and the Head Down of males. All of the displays of the male are given by the female (except Head Down), while the male gives all the displays of the female, even the Ruff-out Chuga. Thus, at least some individuals of either sex are capable of displaying in the manner of the other, but the threshold of these per- formances is apparently higher in the opposite sex. Such presumably sex- linked threshold differences (rather than absolute differences) are to be expected because of the ease with which they could be inherited, and are consistent with information available for other vertebrates. In addition, threshold differences in any behavior allow the species to retain the potential for rapid shifts of behavior in response to changing selective pressures, rather than losing such behavior patterns completely in re- sponse to their uselessness in any particular situation. Threshold differences explain how the potential for the performance of certain behavior patterns may persist for a long time in the species. Although the evolution of displays should proceed in a manner similar to morphological features, displays differ from most morphological fea- tures in that under certain conditions they may never appear during the life of the organism. Thus certain displays can be rare in that they occur in all individuals, but only under unusual conditions, or rare in that they may be performed by some individuals and not by others. The actual extinction of all potential to give a display is probably extremely gradual. For instance Dilger (1960) has found in the genus Agapornis that occasional individuals will rarely perform displays that otherwise occur only in related species. This may be the rule rather than the exception. In the Common Grackle the Ruff-out Chuga and the Ruff-out Squeak Snarl were each characteristic of one individual male but involve merely the substitution of the typical female vocalization for the male Squeak in the former case, and the addition of a second call to the Ruff- out Squeak in the latter. Thus, although different from the typical dis- plays of the male, these variants may be threshold differences. The typical form of most of the visual displays of the grackle showed little variation among individuals, as would be expected from the need for accurate intraspecific signalling. The most variable call is the Squeak, which functions in intrapair recognition, and here variability would be an advantage. ACKNOWLEDGMENTS I wish to thank particularly William C. Dilger for his guidance and criticism. William L. Brown, Millicent S. Ficken, Robert B. Klopman, and Olin Sewall Pettin- gill, Jr., offered many useful suggestions for improving the manuscript. Alan Van Sant did the drawings. Arthur A. Allen, David G. Allen, Dr. Thomas Eisner, Paul R. Gladstone, James M. Hartshorne, Ronald J. Hock, Sally F. Hoyt, Morley R. Kare, Miss Judith Lieberman, Malcolm Peckham, and Mr. and Mrs. David Williams gave valuable help in various ways. This study was financed in part by a Chapman Memorial Fund grant from The American Museum of Natural History. The work was conducted while I was a Research Assistant of The Laboratory of Ornithology at Cornell University. SUMMARY Reproductive behavior patterns of the Common Grackle are described, and their probable derivation and biological significance discussed. Functional and causal analyses are presented for a few displays. The most important displays are: (1) the Head Held Up threat; (2) the Ruff-out Squeak of the male, which is probably homologous to "song" but adds visual components, and the Ruff-out Chuga, the female counterpart; (3) the male Head Down, which is the precopulatory display combining four displays; and (4) the female Solicit, which invites copulation. Pair formation consists of flights of two to six males and a female, and of mutual display (Ruff-out Squeak and Ruff-out Chuga) by these birds when perched. Both males and females defend a small area around the nest site in the colony by threatening and attacking. Males also defend their females against other males. In paired birds the male typically follows the female around, and when perched they may engage in mutual display. By staying near his female, the male is able to defend her against the copulatory attempts of the other males and to be often alone near her when she Solicits copulation. Copulation is always preceded by the Head Down of the male and the Solicit of the female. During the period before egg-laying the male is always ready to mount if the female permits, but shows only a moderate tendency to escape from her, and no evidence of a tendency to attack her. The female, on the other hand, shows a strong but decreasing tendency to attack the male, little tendency to flee from him, and an increasing tendency to allow mounting. Incubating females give displays when leaving the nest, which are probably aroused during incubation but are not able to be expressed until the females leave the nest. Mobbing involves the interaction of escape, attack, and flocking tendencies. Many reproductive behavior patterns of the grackle are probably adaptations to sociability and colonial nesting. Serious injury during fighting was never observed in the wild. Fighting was confined to situations where the survival value of success is probably highest. Behavioral differences between male and female are due mainly to threshold differences rather than absolute incapability. Threshold differences also help explain intraspecific variation. They are a convenient way of retaining potentiality for certain behaviors. LITERATURE CITED ANDREW, R.J. 1956. Intention movements of flight in certain passerines, and their use in systematics. Behaviour, ]0: 179-204. BENT, A. C. 1958. Life histories of North American blackbirds, orioles, tanagers, and allies. U.S. Natl. Mus. Bull., 211: 1-549. CURIO, E. 1959. Verhaltensstudien am Trauerschnpper. Beiheft 3 zur Zeits. f. Tierpsychol., 1-118. DILOER, W. C. 1956. Hostile behavior and reproductive isolating mechanisms in the avian genera Catharus and Hylocichla. Auk, 73: 313-353. DILGER, W. C. 1960. The comparative ethology of the African parrot genus Aga- pornis. Zeits. f. Tierpsychol., 17: 649-685. Hm)E, R. A. 1955-56. A comparative study of the courtship of certain finches (Fringillidae). Ibis, 97: 706-745, 98: 1-23. HUNTINGTON, C. E. 1952. Hybridization in the Purple Grackle, Qulscalus quiscula. Syst. Zool., 1: 149-170. LORENZ, K. 1952. King Solomon's ring. New York, Crowell. MARLER, P. 1956. Behaviour of the Chaffinch, Fringilla coelebs. Behaviour Suppl., 5: 1-184. MORRIS, D. 1956. The feather postures of birds and the problem of the origin of social signals. Behaviour, 9: 75-113. MORRIS, D. 1957. "Typical intensity" and its relation to the problem of ritualisa- tion. Behaviour, 11: 1-12. MOYNI(AN, M. 1955. Remarks on the original sources of displays. Auk, 72: 240- 246. NICE, M. M. 1943. Studies in the life history of the Song Sparrow II. Trans. Linn. Soc. New York, 6: 1-328. PETERSEN, A., and H. YouNo. 1950. A nesting study of the Bronzed Grackle. Auk, 67: 466-476. SEZAN)ER, R. K., and D. R. GILnER. 1961. Analysis of sympatry of Great-tailed and Boat-tailed grackles. Condor, 63: 29-86. TINBERGEN, N. 1959. Comparative studies of the behaviour of gulls (Laridae): a progress report. Behaviour, 15: 1-70. Wrc;Ams, L. 1952. Breeding behavior of the Brewer Blackbird. Condor, 54: 3- 47. Laboratory of Ornithology, Cornell University, Ithaca, New York. Present address: Department o! Zoology, University of Maryland, Col- lege Park, Maryland.