Ostrichrdi

Normal 0 false false false EN-US X-NONE AR-SA

that for an ostrich feeding alone (Fig. 3a), the head was raised again within 10 s on ahnost twothirds

of the occasions. Insofar as could be seen from the much smaller samples, the distribution

of interval lengths of invividual birds corresponded to the overall pattern of all single

birds in Fig. 3a, and there were no indications of temporal patterning. There is an ahnost

exponential decay in the distribution of interval lengths, suggesting that bout lengths were

randomly determined. The distribution of interval lengths in Fig. 3a is not significantly

different from a negative exponential (Z 2 test), suggesting that the lengths of these intervals

were randomly determined. Thus although the head sometimes stayed down for a long period,

it was impossible to predict when this would be. For birds in groups of two, and in groups of

three and four (Figs. 3b, c), the mean length of time for which any iridividual bird's head was

down was greater, and these distributions depart increasingly from the negative exponential.

However the distribution of lengths of intervals when all the heads in the group were down (Fig. 3d)

was similar to that for a single bird, and likewise indicated randomly determined interval lengths

Discussion

Functions of Raising the Head In the absence of any detectable external

stimuli causing a feeding ostrich to raise its head, it is worth considering the other possible causes,

bearing in mind that there may well be more than one.

(i) The bird might raise its head in order to swallow a food bolus. Ostriches store many

beakfuls of food in the elastic top part of the gullet, and at intervals allow a large bolus of food

to move slowly down the neck (Smit 1963). However, the raising of the head is apparently not in

order to allow the passage of a bolus, for two reasons. First, during the majority of periods

with the head up no bolus is swallowed; a bolus can contain the results of 40-I00 pecks (personal

observation) spread over a considerably longer time than the average interval between head

raises. Second, a bolus will often pass up the neck of a feeding ostrich whose head is down.

(ii) The bird might raise its head in order to look for food nearby. This possibility is rendered

improbable by the observations that in the great majority of cases the ostrich continued to feed

at exactly the same spot, and that the relatively few steps taken while feeding were ahnost all

taken when the bird had its head down. The ostrich certainly does not appear to be searching

the ground adjacent to it when its head is up.

(iii) The bird might raise its head in order to observe the feeding rates of its companions

(Drent & Swierstra 1977), or the food types they were taking (Mutton 1971). 'This explanation is

not consistent with the ostrich data presented, where birds raised their heads most often when

they had no companions to observe.

(iv) The bird might raise its head in order to look for conspecifics, either to attract or approach

them, or to avoid them if they are hostile (Dimond & Lazarus 1974). For such interactions

within the feeding group, this explanation is not consistent with the ostrich data given above,

because both the need and the opportunity to observe companions would be expected to become

greater rather than less as the group size increases. However, it is likely that the detection

or observation of very distant conspecifics is one function of head raising; solitary individuals of a

species which is usually gregarious might be expected to spend a greater proportion of time

looking for other birds (Lazarus & lnglis 1978). Searching for distant conspecifics would be

expected to be a more important function for males than for females for two reasons. First,

males are territorial, and they approach, display to and drive out other males, whereas females

largely ignore other females. Second, males approach and display to distant females, and

copulation often follows such approaches; ostriches are frequent and promiscuous maters

(personal observation). A male probably increases his reproductive output more than a

female does from an extra copulation: he fertilizes more eggs, but she does not lay more eggs.

The data presented in Fig. 1 show that males were indeed more vigilant than females. Their

increased vigilance was produced mainly by their keeping their heads up for longer than females

(Table I) : improved scanning of the horizon for distant, slowly-moving, ostriches is probably

best achieved by increasing the length of each scan rather than by increasing their frequency.

(v) The bird might raise its head in order to look for approaching predators. The decline in

percentage vigilance with group size is consistent with reduced vulnerability to predation (see

below). The increased vigilance of single birds over birds in groups was produced mainly by

their raising their heads more often (Table I). Improved detection of a nearby approaching

predator is probably best achieved by frequent and unpredictable raising of the head (as in the

game of Grandmother's Footsteps), rather than by long scans (during which a predator would

remain motionless). Figure 3 showed that the distribution of feeding bout lengths was such

that any prediction by a predator of when a head would stay down for a long time would be unreliable.

Because of their colour, male ostriches are more conspicuous than females, and their greater

vigilance might be thought to be due simply to a greater vulnerability to predators. This is unlikely

for two reasons. First, their increased vigilance was effected mainly by their putting their heads

up for longer, not more often. Second, at the ranges (0-50 m) at which lions are both a danger

and likely to be detectable, any adult ostrich of either sex is highly conspicuous because of its

size.

The Organization of Vigilance It was shown in Fig. 2 that ostriches would be

able to achieve a considerably improved group vigilance if they were 'sequentially organized' in

the sense that each were to raise its head not at random in relation to others but only at times

when the others' heads were down. Some species,such as dwarf mongooses (Helogale undulata)

(Rasa 1977) and jungle babblers (Turdoides striatus) (Gaston 1977), do achieve a high degree

of organization of vigilance. There are several different reasons for the lack of organization in

ostriches; among them are the following:

(1) Any stimulus of common interest will tend to cause a lack of organization because both or

all birds look at it simultaneously.

(2) A bird whose head is down may often be unable to see whether a companion's head is up

or down.

(3) The groupings of birds are temporary ones, and different individuals aggregate in different

combinations at different times and places. The modification of behaviour in relation to that of

frequently changing individuals is probably more difficult to achieve than in relation to the behaviour

of a few fixed companions.

(4) An ostrich cannot predict accurately when another will keep its head down for a long time,

because as was shown in Fig. 3c the length of time a companion's head was down was variable,

although predictability was greater than for a single bird. Note that a predator cannot benefit

at all from this increase in predictability, for as Fig. 3d shows, the distribution of intervals until

the next head in the group was raised was similar to that for a single bird.

Pulliam et al. (unpublished manuscript) discuss the problem of cheating by a bird which

benefits from its companions' vigilance but does not participate in vigilant behaviour itself.

Group Size and Vulnerability to Predators Making certain assumptions about possible

predation on ostriches, it is practicable to some extent to quantify their vulnerability to predators

when feeding alone or in groups. The assumptions are the following:

(l) An ostrich standing with its head up is virtually invulnerable to predation at that

moment. It can run at considerable speed, almost certainly greater than the 50-60 km/h attainable

by lions (Schaller 1972), and can maintain such speeds over far greater distances. Comparative

acceleration rates are not known. Lions usually rush at a prey animal when its head is down

(Schaller 1972).

(2) If in a group there is one (or more) ostrich whose head is up, the whole group is invulnerable,

because sudden movement by one bird would be rapidly detected by others. Therefore,

one head up is as good as two or more.

(3) If a predator makes an attack, its chances of success are not influenced by group size, and

only one ostrich at random in the group is captured.

(4) A small group of ostriches does not attract more predation attempts than a single bird does.

From personal observations of lions hunting, this is a more reasonable assumption than might

appear.

Figure 4 shows the changes in vulnerability with group size. Line 1 shows the mean proportion

of the time that any particular bird has its head down; it is simply 100% minus Line 1 of

Fig. 2. Line 2 shows the group's vulnerability: it is defined as the proportion of time that all the

heads in the group are down. It can be seen that there is little change with group size in the group's

vulnerability. Thus the predator's chances of success are scarcely influenced by group size,

because it is scarcely any more likely to be detected

by its prey. Line 3 shows each bird's individual vulnerability: it is Line 2 divided by

the number of birds in the group. It can be seen that individual vulnerability decreases rapidly

with group size, and that the first companion is by far the most important. Protection is gained

almost entirely by 'diluting' the predator's success (Bertram 1978) rather than by reducing it.

It did not prove practicable to measure experimentally ostriches' actual ability to detect

dangerous stimuli, mainly because of the disadvantage of alarming ostriches which I needed

unafraid for other observations. Nor was it possible to test whether groups of ostriches were

more likely to detect a novel stimulus, as was done by Lazarus (in Dimond & Lazarus 1974)

for red-billed weaver birds (Quelea quelea). Dimond & Lazarus (1974), Drent & Swierstra

(1977), Inglis & Isaacson (1978), and Lazarus & lnglis (1978), observing different species of geese,

have demonstrated a positive relationship between the number of birds in a flock and the

proportion of them who are feeding at any instant. These studies differ from the ostrich

observations in that they deal with flock sizes running into the hundreds, and in that in these

geese the Head Up posture is often important also in observing other geese and grazing conditions

nearby.

Predation is likely to be one important ultimate cause of grouping by ostriches. Bertram

(1978) reviewed the variety of ways by which animals in general may gain protection from

predation by aggregating. The relevance of these to the ostrich case is outlined below.

(1) Grouping may enable them to avoid detection by the predator (Vine 1971, 1973

Treisman 1975; and references given by Lazarus 1972), under certain specialized circumstances

which are unlikely to apply to lions and ostriches.

(2) A group of animals may be able to detect a predator sooner than a single animal can (as was

shown by Powell 1974, and Kenward 1978), and a larger group may detect it sooner than a

smaller group (as was shown by Siegfried & Underhill 1975, and I-Ioogland & Sherman 1976).

The ostrich data refer only to very small groups, and indirectly suggest only a minimal improvement

in predator-detecting ability with group size.

(3) Groups of animals may sometimes be able to deter a predator from attacking, especially if

they co-operate. No defence against lions by ostriches has been observed, and it would

probably be most unwise

www.ostrichrdi.com.