Frontal shield

Abstract: The study of individual birds in the wild has been made possible by capture and marking techniques followed by further catching or by close observation, as well as the normal flow of information through recoveries. Observation of breeding birds has been the easiest because of their strong attachment to a limited area. Some work has been done outside the breeding season using conspicuous marks in the form of harness or collar attachments. Individual variation in birds of one species has been recognised for many years but it has been exploited for research purposes only in a limited wayfor instance in observing family behaviour in the White-fronted Goose, and in the recording of occurrences of the Lesser White-fronted Goose, at Slimbridge, both concerned mainly with the variation in the black belly bars, and in differences in the shape and extent of the white frontal shield.

Abstract: Displays of the pukeko, Porphyrio porphyrio melanotus Temminck (Gruiformes: Rallidae) are described and analysed quantitatively. Aggressive displays centre around the beak, which is presented to the opponent in a position from which a peck can be given. Submissive or escape displays centre around the white undertail coverts, which appear to have evolved as a distinct submissive signal since they are in sharp contrast to the aggressive releaser, both in location and colour. Raising the wings serves to expose the undertail and to provide a contrasting frame. The development of the white undertail as a submissive signal may have paralleled the development of the large frontal shield, which serves to maximise the aggressive releaser. Most displays are ambivalent, signalling both aggression and escape. With increasing escape tendency the bill and frontal shield are so placed that pecking is decreasingly likely, and the wings are increasingly raised. In postures with high escape tendency the beak is placed on t...

Abstract: The 'status-signalling' hypothesis states that individuals might avoid unprofitable fights by signalling their competitive ability and assessing the signals of others. The frontal shields of rails (Aves : Rallidae) are fleshy structures that are displayed prominently during agonistic interactions. Shield maintenance probably requires elevated levels of testosterone and deposition of carotenoid pigments, which implies that shield size and colour could be honest signals of the ability or willingness to fight. We used field observations and a model experiment to study the signalling function of the red shield of the dusky moorhen, Gallinula tenebrosa. Birds with relatively larger shields were more likely to win aggressive encounters, and the outcome was apparently unaffected, or less strongly affected, by body size, weight, condition and sex. Shield size increased with body size and weight, but neither explained more than 26% of the variation in shield size, even after controlling for age and date. Shield size was not affected by body condition, as estimated from residuals of a regression of weight on body size. Moorhens confronted with a model fitted with a large shield retreated more quickly than they did when presented with the same model bearing a small shield, supporting the results from natural interactions. We conclude that the frontal shield functions as a signal during agonistic interactions, and possibly signals the ability or willingness to fight.

Abstract: The Dusky Moorhen (Gallinula tenebrosa), which is common in swamps and waterways, feeds on seeds and the tips of grasses and shrubs (Marchant and Higgins 1993). During the breeding season, adult male and female Dusky Moorhens are colorful, with bright orange frontal shields. The color of the shield of females and younger males fades to black or greenish brown in autumn and winter (Eskell and Garnett 1979). Juveniles have a black shield and beak with a greenish-yellow tip and olive-green legs. Both adult and juvenile Dusky Moorhens possess white undertail coverts. Like many other members of the Rallidae, Dusky Moorhens foraging on the ground flick their tails vertically, revealing their conspicuous undertail coverts (e.g. Garnett 1978, 1980, Woodland et al. 1980; but see Ridpath 1972). The function of this behavior is poorly understood (see Craig 1982, Alvarez 1993). There are several possible explanations why Dusky Moorhens tail-flick, none of which are mutually exclusive. First, tail-flicking may signal to conspecifics that a predator is present, or it may signal a degree of alertness to a potential predator, thereby discouraging pursuit (see Woodland et al. 1980, Craig 1982, Caro 1986). These ideas can only be tested by examining the behavior of birds in the presence and absence of predators. Second, tail-flicking may be a conspecific signal of an individual's state of vigilance or perception of danger (see Caro 1986). This idea predicts that birds that are more vigilant should also tailflick more frequently. Finally, tail-flicking may signal the social status of the individual (Craig 1982), in which case tail-flicking rates should covary with the social status of the individual. We investigated whether tail-flicking in Dusky Moorhen represents a conspecific signal indicating an individual's perception of danger and/or its social status. We assume that scanning behavior is associated with predator detection (see Lima 1990) and that individuals adjust their scanning rates according to their perceived risk of predation (Lima and Dill 1990). Numerous studies show that scanning rates change with group size and the location of individuals within the group (see Elgar 1989), which may reflect differences in the risk of predation (see Lima 1995). Thus, scanning and tail-flicking rates should vary similarly for individuals in both different-sized groups and different locations within a group. Additionally, if tailflicking signals social status, then individuals of different status should differ in tail-flicking rates. Methods.-Dusky Moorhens were observed between 1300 and 1700 EST during March 1994 at the Prince's Lawn in the Royal Botanical Gardens, Melbourne, Victoria (37?50'S, 145?E). The lawn is an open, grassy area of about 3 ha, flanked by bushes on three sides and inclining to an ornamental lake towards the eastern edge. At least 25 Dusky Moorhens frequented the general vicinity of the lawn at any time. Pacific Black Ducks (Anas superciliosa) also were seen occasionally on the lawn, foraging among groups of Dusky Moorhens. The behavior of Dusky Moorhens foraging on the grass was recorded from a garden bench that afforded a view of most of the study site. The presence of observers on the bench apparently did not influence the behavior of the birds. Individual Dusky Moorhens were selected arbitrarily from throughout the study area. We noted the color of the frontal shield of the focal bird and categorized the bird as either alone or in a group. A group was defined as four or more birds within 2 m of each other. Solitary birds were greater than 2 m from any conspecific. We also noted whether individuals were at the edge or center of the group. A bird was defined as being at the edge of a group if there were no other birds within 1800 of the perimeter of the group. We did not collect data for birds in pairs or trios because it was not possible to discern whether individuals were at the center or edge of the group. It is probable that some birds were observed more than once; while this is unlikely to introduce any systematic bias into the analysis, it may reduce the statistical power. The numbers of tail-flicks and proportion of time spent scanning were recorded over a 2-min period, after which another bird was selected. A tail-flick was defined as a single rapid movement of the tail in an upward direction. Birds were recorded as scanning when they lifted up their heads and turned them on a horizontal plane. Observation periods were terminated early if the composition of the group changed, a disturbance caused the birds to run or fly, or the birds were fed by visitors to the gardens; data from terminated observations were discarded. Data were analyzed using procedures in the package SYSTAT 5.2 for the Apple Macintosh computer (Wilkinson 1992). Tail-flick rates were log(x + 1) transformed in order to approximate a normal distribution. Results and discussion.-A two-way ANOVA, with