CRITICAL
EXAMINATION OF VARIOUS CONCEPTS
RELATED TO SOCIAL SPACE IN ANIMALS.
Jacques P. BEAUGRAND
E-mail: beaugrand.jacques@uqam.ca
ABSTRACT
INTRODUCTION
Science takes
space and time for granted, so much that space and time coordinates are used as
independent variables. We can admit that space and time are not self-existing
objects but a network of relations among factual items, things and their
changes: space is an order of possible co‑existents and time an order of
successive (Leibniz, 1956). In other words, the thing space is nothing but the
collection of spaced things, or the set of things related by their mutual
separations. So, strictly speaking, one speaks of space in a conventional
manner and, a fortiori there is no such thing as a social space.
Applied and
basic ethology are not preoccupied by space in itself, but by separation or
contact between individuals or objects as the result of animal activity and
behaviour. The concept of social space which has recently appeared in the
ethological vocabulary has to be understood as referring to a very general
spatial framework in which separation, distance, proximity or even contacts
between individuals, or between them and valued objects, are the result of
control behaviour aiming at searching or restricting proximity and contact.
Although distances are truly physical distances, measurable in terms of
physical units, they are qualified of social, because they are maintained and
brought up by social behaviour.
Social space
can be used as a very broad concept referring to any area delimited by familiar
objects or defined in reference to the concerned subject(s), the intrusion of
which area may or may not be tolerated according to the social context or may
even be searched in certain circumstances. In the first case, excessive
approach within a minimum distance that an animal attempts to keep between
itself and others, or between some valuable resource and other, congeners can
trigger behaviour having for consequence to put more distance between them. In
this sense, the concept of social space covers more traditional concepts used
by ethologists to qualify and describe patterns of space use in animals,
including the already well encrusted but nevertheless problematic concepts of
home range, territory and territoriality. Though these concepts are of a
very limited application in studies carried out in battery cages and other
industrial settings, they nevertheless still contribute to the background of
such studies. The concept of social space also covers the less known
concepts of personal sphere, space, distance and field. These concepts
concern the portable space of interacting individuals and are en vogue among
ethologists because their illusory relevance to applied and restricted settings
when compared to concepts such as home range and territory which have evolved
from ethological studies carried out on free ranging animals.
In any case,
in order to be valid, scientific concepts such as the ones at hand must have a
definite connotation, and their extensional vagueness must also be kept
minimal.
Ethology, which is an empirical science, is particularly interested in concepts referring to concrete (though perhaps hypothetical) existents. Not all constructs occurring in factual science have factual reference. Logical concepts such as "not" and "set" have no such reference. But concepts which are meant by scientists to refer to concrete things (i.e., which are not solely formal objects) must at least have the possibility to exist on their own, i.e., independently of their being perceived or conceived by the researcher. Otherwise, they contribute to untestable theories about ghosts.
The present
paper is mainly concerned with a critical examination of the concepts just
mentioned which are related to the use of space in animals. After having
very briefly reviewed the essentials of each concept, we will examine their
logical coherence, scientific validity and usefulness in applied
settings. The study of spacing in animals can also be undertaken by
adopting a sociosystemic perspective: active monitoring by individuals of
proximity or distance between them and other congeners becomes understandable
in terms of cooperation and competition which contribute to the formation of
more or less connected social structures. Adopting such an approach can de‑emphasize
the importance of territoriality in accounting for spatial patterns. Such
an approach is presented in the appendix. The measurement of concrete
distances between individuals and of differential occupation or frequentation
of sites by individuals remains the basic scientific operations from which all
concepts related to the use of space gain their support. Some quantitative
instruments, mainly due to Waser and Wiley (1979), for the specification of
obtained patterns of spacing among individuals and between groups are for that
purpose introduced in the second part of this paper.
REVIEW OF THE MAIN CONCEPTS AND CRITICAL EXAMINATION
Portable individual space and related concepts
The tendency
of animals to space relative to congeners is an extremely general feature of
social behaviour. Individuals keep their distance from each other as if there
existed a critical distance, a sphere or bubble surrounding each individual and
moving with him. The intrusion or penetration of the restricted area can
trigger behaviour, sometimes of the aggressive type, having for consequence to
restore a more respectable inter-individual distance. The regular spacing of
birds perched on a roof or wire illustrates the existence of a minimal distance
that can be tolerated between two individuals of the same species, but at the
same time a search for proximity. That stimuli releasing aggression are
never simply those from congeners, but always include an inter-individual
distance component reinforces this idea of a restricted personal space.
Alternately, some animals may be free to enter such areas under some occasions
but not on others. Individuals of one species that keep their distance during
the day may huddle at night, or on especially cold days. Males may tolerate the
proximity of females and juveniles but be intolerant to the approach from other
males in the group. Nevertheless, sociability in animals is evidenced by
individuals not distributing evenly when allowed to disperse over an area of
uniform quality, and by having a definite tendency to aggregate and to
synchronize their activities.
Several concepts were proposed by Hediger (1941, 1955) and McBride (1971) to take into account these regularities. McBride (1971) has proposed that each solitary individual or isolated affiliated group on its home range maintained and defended a personal sphere by preventing others using the same area from approaching within a certain distance in any direction. This personal sphere would be portable with the individual or affiliated group. For McBride (1971), "this is the home range system, where an undefended range is used, but only the portable personal sphere is defended against intrusion". When the species concerned lives in groups, or is a gregarious one, spacing is still present within the group, with individual maintaining personal fields (McBride, 1971) and avoiding entering the fields of neighbours. The personal field was originally called a social force field by McBride (1964). According to him, these personal fields do not necessarily have an equal radius in each direction as do the personal spheres, but are, for poultry at least, well developed directly in front of the face. This was demonstrated in a flock of domestic hens, where most of the birds' movements were concerned with avoiding the personal fields of dominant neighbours (McBride et al., 1963). The spacing of the heads was more regular than it would have been under random spacing and birds seemed to avoid each other's facial aspect. To "simplify" terminology, the concept of personal area was proposed by McBride (1971) to refer to either the personal spheres or fields.
The concept
of personal distance was also introduced by McBride (1971) and it was meant to
refer to the distance from an animal to the limit of its field or sphere.
Another concept, individual distance can be found in Hediger (1941, 1955),
McBride (1971) and Conder (1949), to refer to the minimum distance that an
animal routinely keeps between itself and other members of the same species.
Each species would have a characteristic minimum distance that can be measured
when animals are not on their territories. The animal would enforce the spacing
by either retreating from the encroaching neighbour or by threatening it away.
Individual distance is not to be confused with another concept, the flight
distance, defined as the minimum distance an animal will allow a predator to
approach before moving away (Hediger, 1950).
The concept of social distance was defined by Hediger (1963) as the maximum distance an animal will move away from its group of belonging. Gregarious animals normally live in a living space situated between the personal fields of neighbours and within social distance. Basically, it is then the observance of these two distances which gives, according to McBride (1971), the characteristic spatial architecture to a group of animals in any period of their life. Wilson (1975) and Zayan et al. (1983) do not respect the original meaning given by Hediger (1963) to social distance and they use it in the sense of (inter‑)individual distance, a concept that was previously defined.
What is the
scientific status of these concepts? Concepts such as personal field,
personal sphere, personal area and distance, as used by McBride (1971) and
Hediger (1963), do not refer to concrete entities. They are concepts conceived
to schematize recurrent patterns implying that individuals seek or avoid
contact with each other, or control proximity. But, evidently, there is no
such physical or biological thing as a personal space, field or sphere moving
around with the individual. These are metaphors. Even their authors were
perfectly aware that they were proposing analogies. But users tend to
ignore this and use these concepts as if they referred to real things. For
McBride (McBride et al., 1963) it seemed clear from the kind of evidence
presented above from poultry that the presence of another individual
constitutes a social force on its neighbours. But the word force was used
by McBride to mean that there was a measurable effect on behaviour of
neighbours, that it diminished with distance and was maximum directly in front
of the head. A possible analogy proposed by McBride was then to compare
this gradient with the fields of magnetic force associated with a single pole
of a magnet. But the use of the term field borrowed from physics had the
implication that there exist things (and what things?) like physical fields of
force; but it did not give any indication concerning the nature, structure, and
the measurement of the field itself. The analogy is not fruitful for
ethology because it does not cover the original concept, does not suggest
either fruitful new problems, except the futile challenge of finding ways to
test for the existence of ghosts and to measure them, and cannot be assimilated
by a scientific theory of any sort. These concepts give the appearance of a
scientific approach, but their virtue is, alas, to cover our conceptual
indigence.
The use of
superficial analogies is also dangerous. It is that under apparently
innocuous schemata are left buried deeper explanations which are now absolutely
essential for understanding animal spacing and to solve husbandry and welfare
problems. For example, it hinders the recognition that distances
maintained between individuals, which can be measured concretely, vary greatly
according to the recent histories of the encountering individuals, as well as
to the context in which the encounter occurs.
Enough
metaphorical comments have been made in the literature concerning the so‑called
causes of personal distance, field or sphere, namely, perception of a
conspecific as being too close or too far, behaviour towards a conspecific
where proximity is tolerated or refused (Zayan et al., 1983). These
mechanisms have to be studied and clearly identified when possible. If they
remain unobservable, a theory about them and capable of accounting for these
observed regularities has to be worked out and tested in the search for deeper
regularities.
Simply sticking names on superficial regularities may be satisfactory for descriptive and taxonomic purposes, which was the principal programme of early ethologists. But, modern ethological science aims at understanding spacing mechanisms, their ontogeny and evolutionary history and this knowledge is of primordial importance for sound behavioural intervention and rationale engineering of rearing conditions in zoos and industrial settings. Deeper regularities are to be searched for.
The concept
of individual distance as defined by Hediger and McBride, suffers from
comparable shortcomings. As used by these authors, the concept of
individual distance implies for them a mechanism of segregation explaining the
minimum distance that is actively kept between two individuals. It supposes the
existence of a critical distance, a minimum distance that an animal attempts to
keep between itself and other congeners, the trespassing of which can trigger
aggression. These assumptions could be part of a whole theory accounting
for regularities observed in the spacing out and selective occupation of
area. However, the present author is not aware that such a theory was ever
articulated. Moreover, the concept of individual distance cannot be used
simply as a key‑word referring to a whole series of mechanisms, since it could
only refer to the results of these mechanisms. It could at most be used as
a purely theoretical concept, in a manner similar to the concepts of mean and
variance in the statistical language, to characterize patterns of recurrence
within the spacing of individuals. The concept of individual distance, as
Hediger and McBride accept it, is again a pure metaphor, not pointing to a
concrete entity, and not used as theoretical schemata. It has no descriptive
value, let alone an explanatory one. Such typical and regular distances
effectively maintained between congeners, are the things to be explained, not
explanations in themselves.
Moreover,
individual distance, if used in this last acceptation, is simply a misnomer
(Zayan et al., 1983): distance is, by definition, a relation between two points
or individuals and not an intrinsic
property or state. Being a relational variable of different entities holding
different positions, distance is inter-individual. The concept of (inter‑)
individual distance could be more useful if it simply referred to the
measurable distance between two individuals, without any reference to specific
typical value for each subject.
The concepts of home range, territory, and territoriality
The home range
Nearly every animal spends its life in a circumscribed area, called its home range, except while dispersing or migrating. In field studies, home ranges are mapped by following an animal or group of animals about and plotting their movements on a map or by plotting the various locations where they have been sighted. The outermost points on such a map are then connected to form the smallest possible convex polygon, which is operationally considered the home range. Home ranges seem to have internal structure. Some parts fulfil habitat requirements better than others, and animal movements reflect this heterogeneity. Areas of intensive use, or core area, are centred around sleeping sites, high quality food patches, and water points, and nesting sites, which are the principal resources required by the animal. The home range of many animals, particularly mammals, consist of relatively few core areas interconnected by a network of narrow pathways.
One metrical
problem with home range is that when polygonal maps of such home ranges are
drawn from field observations, the ranges of neighbour animals or groups often
appear to overlap, except for the core‑area. Home range maps based on polygonal
or elliptical methods do not convey a true picture of how space is being used.
Some authors have tried to circumvent the problem by obtaining convex polygons
for different criteria of frequency‑use (Dunn and Gipson, 1977; Michener, 1979,
1981; Anderson, 1982; Bowen, 1982; Bekoff and Wells, 1982). For example, Bekoff
and Wells (1982) calculated the geometric centre for each coyote of their study
and obtained different polygons representing the area encompassed (outward from
the geometric centre) by 25%, 50%, 75%, 90%, and 95% of the total locations as
well as the area defined by all (100%) locations at which individuals had been
sighted. Another way to circumvent such problems is to conceptualize home range
simply as an objective activity field (Waser and Wiley, 1979) or as a space use
map or plot. These concepts are to be covered in a forthcoming section of
this paper.
The concept
of home range has been applied to free ranging animals such as the sheep
(Hunter and Davies, 1963) and the feral domestic fowl (McBride et al., 1969).
But it is evidently not very relevant for welfare studies carried out in
restricted or industrial environments.
The territory and territoriality concepts
In the
earliest accounts, a territory was recognized as the locus of positions in
which a resident animal won agonistic encounters. When a territorial individual
met an opponent beyond this locus, it either had to retrieve against another
territorial individual itself in a situation of residence, or the issue was
left unpredictable (Howard, 1920; Nice, 1941). This was most obvious when there
was aggression manifested against intruders. However, in many instances it is
believed that animals maintain exclusive areas by less overt behaviour. Many
carnivorous mammals avoid each other by detecting the presence of scents
deposited by resident individuals and may very rarely or even never meet.
The
definition of the territory seems to pose a real problem to the ethologist. A
variety of definitions of the territory and of territoriality have been
proposed. Definitions of either fall in two broad categories, whether they
refer or not to the behavioural mechanisms by which exclusiveness or at least
non‑random distribution of space is obtained.
Here are some examples: "A territory is any defended area" (Noble, 1939); a territory is an "exclusive area" (Schoener, 1968) or a "fixed, exclusive area with the presence of defence that keeps out rivals" (Brown and Orians, 1970); territorialism is "defence of a resource by fighting or displays" (Krebs and Davies, 1981).
Some
definitions do not refer to behavioural mechanisms; they are of the ecological
type. For instance, Pitelka (1959) emphasized that territory is primarily an
ecological phenomenon, and defines it as "an exclusive area, not merely a
defended one...". Pitelka was primarily concerned with the economic
functions of territory, namely by the exclusive use of resources contained by
the territory, and the dispersion of competitors; he dismissed as irrelevant
the mechanisms by which exclusiveness is maintained. A similar definition was
recently proposed by Davies (1978) who recognizes the existence of a territory
"whenever individual animals or groups are spaced out more than would be
expected from a random occupation of suitable habitats".
To these
complementary points of view adds a third one concerning the function of
territories. For example, despite the apparent diversity of territorial
behaviour, Wilson (1975) maintains that its function is simply to defend a
particular resource. The adoption of this functionalist approach allowed him to
recognize five major types of territory.
Type
A. The "all purpose" territory. This is a large, defended area
within which sheltering, courtship, mating, nesting, and most food gathering
activities occur (e.g., benthic fishes, arboreal lizards, insectivorous birds,
some small mammals).
Type B. The
breeding territory. It is a large defended area within which all breeding
activities take place but which is not the site for most food gathering (e.g.,
night-jars, reed warblers).
Type C. Nest
defence territory. A small defended area around the nest of colonial birds
(e.g., ibises, herons, gannets).
Type D.
Mating territories and leks. Pairing and/or mating territories having a
seasonal character (e.g., dragonflies, lek birds, Uganda kob).
Type E. "Home" territory. These territories are centred around roosting and shelter locations (e.g., bird dormitories, roosting place in the domestic pigeon).
Such a
classification fails to grasp the full range of variability that exists. Some
of the dimensions along which territorial forms of organization can be said to
vary are the following: individual territory versus pair or group territory;
defended versus non‑defended territory; exclusive versus non‑exclusive
occupancy; feeding versus non‑feeding territory; mating versus non‑mating
territory; rearing young versus non‑rearing; fixed versus floating territories.
For several
authors, the home range includes the territory which, in this case, corresponds
to the core area of the home range. For others, the basic distinction relies in
the fact that the home range is not defended when intruded, while the territory
is. Finally, for McBride et al. (1969) the territory and home range of feral
fowl can perfectly overlap in some cases, depending on the phase or
reproductive state in which the animals are at a given period of the year.
CRITICAL EXAMINATION
The concepts of home range and territory refer to concrete entities but, these well encrusted concepts are not used without causing major conceptual problems which boil down to their lack of sharpness. In addition, the concept of territoriality shows much ambiguity.
As a first
consequence, there is practical confusion when it comes to report factual
observations concerning the home ranges and the territories of individual
animals or groups of animals.
Although most authors agree with a distinction of principle between the home range and the territory, when it comes to distinguish between them in the field, it is failure. The home range and the territory are one and the same, the terms being used in an interchangeable manner. When one of these spatial system is reported there is a possibility that the use of the term home range or territory is simply arbitrary. Moreover, investigators have a clear tendency to be inclusive and do not hesitate to reduce all activities to territorial ones, as if pruning down to the home range at an early stage of a research project could lead to the loss of some important knowledge for posterity.
A distinction
of principle can easily be established between these two concepts. Both
have concrete connotations and are factually referential. The home range and
territory are composed of sets of physical or biological objects in physical
relation with each others, delimiting a "space" which can be three‑dimensional.
For the ethologist, both the home range and the territory correspond to some
piece of land, area over water, volume in the forest or in the sea, in which
the individual animal, pair or group of animals spend most of their time at a
given period of their life. But, specifying the home range does not
require that the area be defended and whether or not it is of exclusive use. It
is simply a concept accounting for special (e.g. frequentation) relations
that an individual (or group) entertains with a given area.
The
home range can be represented by a triplet accounting for 1) a set of physical
objects in spatial relation with each other and defining a spatial reference
frame (i.e, the "space"); 2) a subject animal, pair or group of
animals frequenting the home range; 3) a set of relations bounding the
individual (or pair, or group) to some physical objects situated within the
spatial reference frame. A comparative criterion of some kind can be used
to decide whether one locus is part of the home range or not.
Specification
of the territory is more complex. It requires, in addition to the elements
already required to specify the home range, an element of comparison between
individuals of the same species. In order word, the territory is an area
with which an individual (or group) enjoys privileged connections when compared
to other areas as well as other individuals. Usually, we say that the area
or space is defended or else used in an exclusive fashion by its
owner(s). Defence and/or exclusive use, which are special kinds of social
relations, can only be defined with reference with other individuals of the
same species (if, evidently, intraspecific territoriality is concerned).
So, in order to make a scheme of the territory, two terms must be added to the triplet modelling the home range: a fourth term must be used to represent other individuals in competition for the same space, and a fifth term to represent social (and anti‑social) bounding relations existing between the territorial individual (pair or group) and other competing ones.
The terms
home range and territory be used in a more rigorous fashion. The term home
range should be restricted to the designation of an area whose map corresponds
to some criterion of frequentation obtained by valid sampling of the positions
of the individual, or by duration of frequentation of the various space units
in which the whole area of study has been partitioned. But, since by
essence home range is a range, information about the structure of space
utilization cannot be conveyed fully. Even the application of successive
criteria for area encompassing various percentages of the locations frequented
by the animal cannot deliver rich information about how the animal uses the
total available area. In addition, since the concept of home range is of
totally irrelevant use in welfare studies carried out in restricted and
industrial environments, and even of poor application to free ranging domestic
animals, the present author suggests to abandon the home range concept in
favour of Waser and Wiley's (1979) activity field which will be covered in the
second section of this paper.
To
this difficulty of distinguishing between the home range and the territory
coincides another problem concerning the vagueness and ambiguity of the
concepts of territory and of territoriality, which justifies the next section.
Attempt to elucidate the concept of territoriality
Brown and
Orians (1970) have suggested that Noble's definition of territory as showed
much flexibility by its simplicity, and most ethologists adhere to this simple
definition of the territory as any defended area.
It is very deceiving to realize that some concepts in ethology get their strengths from fuzziness, ambiguity and over applicability. As is the case with the definition of dominance, the simplest definition in terms of aggressive interactions may not be the most useful one for a becoming science.
Clarification
and sharpening of the concept of territoriality or territorial behaviour and
territory are badly required. It seems logical to the present author that
clarification of the concept of territoriality has to be completed before that
of the territory. After all, the map of the territory, although very rarely
drawn, can only be obtained after having recognized a series of loci where some
state of affair, duly qualifying as territoriality, has been obtained.
Elucidation
of the concept of territoriality can be done by interpretation: what does
territoriality mean in a given context, what does it stand for, i.e. what is
its designatum? The intention of the
concept of territoriality is the set of properties characterizing living beings
showing (in their behaviour) territoriality, e.g. site tenacity,
advertisement, overt defence, successful exclusion of intruders and exclusive
use. The extension of territoriality is the set of territorial living
beings.
But, while
trying to elucidate the concept, one realizes that territoriality is not always
the mere conclusion that an animal shows precise properties. Although it should
be that kind of conclusion, territoriality is also a concept related to the
motivation of behaviour and thus of the level of the explanation, rather than
of a purely descriptive one. The concept seems to have acquired an explanatory
value in itself. In this case, a theory about preprogrammed intolerance
and mutual exclusion of congeners, basic individual and reproductive space
needs, and the like, is mistaken for a concept summarizing the (hypothesized)
behavioural manifestations of these, namely territoriality. As for many
other concepts in ethology (e.g. at the taxonomical level), the concept of
territoriality is supported by an implicit theory about its teleonomical
function or selective value of being territorial.
To elucidate or to sharpen the meaning of territoriality one can increase its earmarks. Ambiguity and vagueness can be reduced if the terms are further defined. But definitions cannot, of course, eliminate whatever ambiguity and vagueness there is in the primitive symbols. For instance, one has to choose the primitives.
Recent
contributions from sociobiology and eco‑ethology have shed new lights on the
evolutionary origins of several forms of social organizations, including
territorial ones. However, by proposing different primitive concepts for
explaining sociobiological regularities they have also contributed to create in
the animal behavioural sciences a momentary state of ambiguity and confusion,
because primitive concepts and mechanisms belonging to different levels of
reality were alternately and sometimes simultaneously proposed to explain a
given regularity. A choice must be made between the primitive concepts
corresponding to different levels of reality.
The intuitive
idea about levels is simple: the things at any given level are composed of
things belonging to the preceding levels. Thus ecosystems are composed of
populations, which are composed of organisms, which are composed of organs,
which are composed of cells, which are composed of organelles, which are
composed of molecules, which are composed of atoms, which are composed of so‑called
elementary particles. Each level has its own laws, rooted into laws of the
preceding ones. For example, the behavioural level, behaviour being the action
of individual organisms participating to sociosystems, constitutes a level of
reality ontologically anterior to the ecological level, in the sense that
behavioural or ethological laws support ecological regularities or laws. Levels
cannot act upon another. In particular, the higher levels cannot command or
even obey the lower ones. So, all talk of inter‑level action is elliptical or
metaphorical, not literal.
To be consistent with such an ontological principle, one has to select a given level of discourse. With reference to territoriality, basic ethology has mainly relied on behavioural propositions, not demographic or ecological ones, and the regularities they study are special productions of the CNS, namely behaviours. This is not to say that individual organisms are not influenced by fluctuating elements of their environment. But the laws producing such demographic and ecological fluctuations are surely not the ones that could also cause changes in the behaviour of the organisms under study.
So, at the
ecological level, spacing out patterns of populations not conforming to a
random occupation of suitable habitats may or may not be called territoriality
by ecologists; it is a question of ecological terminology. While some cases of
non‑random occupation can be explained by behavioural territoriality, other
cases of non‑random distribution can surely also be explained by some alternate
behavioural mechanism, e.g. social attraction, pace Davies (1979).
True territoriality, when defined in terms of behavioural events, requires that stringent conditions be satisfied in order to eliminate rival explanations. That one individual successfully chases an intruder from an area can be explained by differences in familiarity with the given area, prior residence, differences in size and strength, the expression of an already established dominance relation between the pair members, individual recognition, &c, and is thus not a sufficient condition to declare that territoriality is at work.
Five
behavioural criteria must be realized conjointly in order to declare that true
territoriality is present: First, individuals must show site tenacity to a
given area. This is to distinguish territoriality from mere personal defence
when approached, or from the defence of a valuable mobile resource such as a
mate, a young, a food item that happens to be in the area. Second, there must
be defence of the site (Noble, 1939). The proposition to defend a site is more
easily said than shown. It is in itself non‑observational and has to be
constructed, by the interpretation of the raw empirical data, into a
theoretical proposition. When we say that an individual animal defends a given
space we implicitly declare that the individual residing within a given space manifests
aggressive behaviour toward a category of congeners considered as intruders,
and that it has for result (at least in a majority or modal number of cases)
that intruders leave the defended space while the resident stays. It is not at
all desirable to add any speculation about the "intentions" of the
defender, which are not verifiable publicly. These conditions have to be
strictly realized in the observations to be in a position to declare that one
individual defends a given area. The question, however, remains whether a fixed
space was defended, or only an area set in relation to a more or less mobile
defender or defended valued object (e.g. mate, nest, young). A third condition
for the existence of true territoriality is that the individual's presence in
the area be advertised. Overt defence (attacking, chasing intruders) or display
identifies the site holder and makes him conspicuous to rivals (Brown and
Orians, 1970). A fourth condition for the existence of territoriality is that
the territorial animal putatively excludes potential competitors from the
defended area. The holder (an individual, a pair, a family or a whole group of
animals) must have exclusive use of the defended area (Emlen, 1957). Finally,
as fifth condition for true territoriality, putative reversibility of exclusion
must occur when the resident becomes an intruder for an adjacent site‑holding
conspecific (Tinbergen, 1953; DeBoer and Heuts, 1973; Zayan, 1974). In the fish
Hemichromis bimaculatus, territorial defence could be induced at will by
alternatively provoking intrusion into each of two fish's residence area
(DeBoer and Heuts, 1973). This specific‑area‑linked dominance has to be
realized in order to declare the existence of true territoriality in a given
situation. In several species of fish the present author as tested, including Xiphophorus
helleri, Salvelinus fontinalis, Trichogaster trichopterus,
reversibility cannot be obtained. Zayan has also failed to show reversibility
in the domestic fowl in experimental conditions.
The application of such strict criteria for true territoriality sharpens the concept but has for inevitable consequence to greatly reduce its scope. True territoriality, as defined from these earmarks, becomes restricted to extreme situations, and probably very rare ones.
What happens
with other cases in which one or more of the above-mentioned conditions were
not satisfied? Are these also manifestations of territoriality? The
answer is no. The concept of territoriality refer to a global property
that is qualitative, i.e. the concept refers to a property variable taking
a value of 0 when not territorial or 1 when territorial. Intuitively and in the
everyday use, the concept of territoriality is used to cover various patterns
of social distancing, territorial defence, isolation (on territory) and
appropriation or defence of resources momentarily fixed both in space and time.
As rigorously defined above, the qualitative concept of territoriality is
surely inappropriate to account for such a variability.
A possible
avenue to explore as a solution would be to consider the properties used to
define the concept of territoriality as variables. These "qualities",
the concept is supposed to cover in the everyday ethological language, can be
found realized to various degrees in nature. They exist in determinate amounts
and in space‑time as characteristics of the relations animals have with each
others and with objects in their environment. True territoriality, as
defined above, would correspond to the conjoint and full covering of these
variables but in their extreme value. But, situations where
only partial and incomplete covering of the criteria is obtained should
not qualify as territoriality and should simply be referred to as social
spacing or preferably social distancing, which are more generic concepts
referring to a continuum of realities of which territoriality would be a
special case.
Applied
ethologists are much concerned by space requirements in terms of properties of
habitats and proximity of congeners. The concept of territoriality is
simply not relevant for intensive situations in which ethologists try to solve
welfare and productions problems. True, territoriality in the weak sense
of a defended area has been reported in older bulls (Kilgour and Campin, 1973)
and Soay sheep (Grubb, 1974) but these are the very rare cases reported. Even
so, overt aggression observed in these cases and interpreted in terms of
defence of the area could well be explained by other factors.
It is obvious that an investigation of
spacing obtained by social behaviour can no longer be reduced to the
naturalistic tenet that individual areas are basically the result of
territorial control (Zayan et al., 1983).
Spacing systems are part of social systems. As stressed by Zayan et al. (1983), spatial relations are full components of social systems and should be discussed in the light of cooperation and competition which are the basic forces fashioning social systems, a view already announced in McBride (1971), and that we have presented in Appendix A.
Conclusion
The
conclusion is methodological. Ethologists should measure more
often. Precise measures of distance, especially between individuals, are
badly needed. In the field of animal welfare, specific evaluations of physical
space can be made, e.g. cage dimensions and actual distances between cage mates.
Needless to say, a full account of social space requires that both the
behavioural and cognitive processes of spacing be also investigated. Concepts
such as personal space, personal sphere and fields have to be discarded as
potentially useful for the science of behaviour. The concept of home‑range can
be used to refer to a map obtained from the transposition over a geographical
representation of an area, of objective and specific measures of frequentation
of the various loci (or quadrats) that are part of the area. However, the
territory is more difficult to delineate or to characterize. In principle, the
concept of territory can be used to refer to a map obtained in a similar
fashion for points or loci neatly satisfying the criteria of territoriality.
However, we have seen that the concept of territoriality is a fuzzy concept,
and as used in the ethological literature, difficult to apply. These concepts
are not rich enough to cover most cases of space use and their too more or less
rigorous application in research cultivate the danger of concealing, under
apparent conceptual conformity, mechanisms that have highly different
evolutionary origins. On the other hand, those concepts related to the mobile
individual space are analogies and furnish only a very superficial
schematization. They should be replaced by more profound explanations, based on
the cognition and experience of the individual animal and grounded by concrete
the measurements of distances between individuals among themselves, distances
between individuals and valued resources, and of use of specific sites.
Waser and Wiley (1979) have proposed to abandon the search for a unitary
definition of territoriality, and instead, to address the variation among
species in relationships of aggression, isolation, and activity fields, which
are measures of space utilization, showing more flexibility, coming in degrees
and thus more readily quantifiable. These concepts, which are of the
metrical level, can be used to substantiate theoretical concepts, serve as
indices for them or, as more usually said, serve to render them operational. A
concept of spacing, defined in terms of site frequentation, exclusive access,
site defence and inter-individual distance, would have the definite advantage
of being in a position to refer to graded properties of social systems and thus
to be represented by magnitudes or quantities.
APPENDIX A
QUANTIFICATORS
Activity field
Isolation field
The relative
exclusiveness of an individual's use of space as a function of location further
defines an isolation field (Waser and Wiley, 1979). The value of an
individual's isolation field at any location, is the ratio between the time
spent by the subject at that location and the time spent by all individuals
including the subject at the same location. If instantaneous sampling was used,
frequencies of occurrence at a given location are substituted to total periods
of time. This ration varies from 1, when the subject has exclusive use of the
location, to zero, when it never uses the location but others do. Isolation
fields are obtained for each individual and thus describe their pattern of
exclusive use, and in combination with activity fields and information on
resource distribution within the studied area, can give good indications on the
degree to which each individual monopolizes access to resources. Individual
isolation fields are thus obtained by comparing the activity field of one
individual to the total activity field of the group or at least focussed
individuals.
Aggression
field
Wiley (1973)
proposed the measurement of a third field, the aggression field. As we know, an
individual's reaction to an approaching conspecific depends on several factors
such as the state of the reacting individual, the identity and apparent state
of the approaching individual, the distance separating the two and the absolute
location of the encounter. The concept of an aggressive field can help to
understand the influence of several of these determinants if they are known to
the researcher. The value of an individual's aggression field at any location
is defined by the proportions found there of attacks or retreats initiated by
the concerned individual. By noting in which quadrats are situated both
aggressive interactors, it becomes possible to obtain the aggression field for
a focal animal; also, information about the absolute location of the encounter,
the proximity of actors from each other or from a fixed resources, as well as
the identity of the target individual can be obtained by partitioning the
original observations. I am not aware of any work published having measured in
detail the aggression field as defined by Wiley (1973). Perhaps van Iersel
(1958) came closest by determining the decline of aggressive behaviour of male
sticklebacks toward a standard opponent as distance from the subject's nest
increased. In some way, the aggression field of Wiley (1973) is an operational
definition of McBride's personal field. The measurement of individual
aggression fields is of particular relevance to researchers interested in
social and space requirements in industrial cages as well as in the disposition
of water nipples and feeding‑trough. One can also imagine several other
specific behavioural fields based on the ratio of presumed opposed tendencies,
such as cooperation versus rivalry.
Inter‑individual spatial relations.
How are the individual's activity, isolation and aggression fields
related to those of its neighbours? How can we say that the concerned field is
not random? There are several ways to answer these questions. Some are
presented by Waser and Wiley (1979). The only thing I can do here is to explain
the essential of some of these techniques.
Methods for measuring the instantaneous dispersion of individuals, particularly those based on nearest‑neighbour distances are well known and were imported from plant ecology. By comparing the distribution of nearest‑neighbour distances with that expected from a set of randomly positioned points, groups or populations can be classified as over dispersed (when close spacing is more frequent than expected by chance alone), random or aggregated (when close spacing is less frequent than predicted by a uniform distribution).
Independence of movements
In some cases
it is relevant to know whether two individuals or two groups of individuals are
attracted to each other or avoid systematically. If inter‑individual or
intergroup distances are sampled at regular time intervals, the relative
probabilities of approach or withdrawal can be calculated as a function of
intergroup or inter‑individual distance (Waser, 1976). These data can be
contrasted to specific hypotheses predicting frequencies of encounters to a
specified separation by using, for example, a statistical model of a perfect
gas having known density and velocity (refer to Waser and Wiley, 1979 for full
details). Other statistical techniques are also available. Essentially,
comparison of observed and expected frequencies of approach at given distances
can determine whether or not avoidance occurs, as well as the radius of
avoidance, if such a radius exists (Waser, 1975b, 1976).
Spatial
overlap
The simple
measure of the spatial relationships of activity fields is the percentage of
individual home range overlap. Indices of overlap that takes intensity of
use into account should include measures of overlap in the use of quadrats of
the activity fields concerned. A numerically simple index was suggested by
Holmes and Pitelka (1968) based on differences in proportions of use of each
quadrat. Another is Pianka's measure of
overlap (Pianka, 1975). But, the only index of overlap so far used for intensities
of quadrat frequentation is the Pearson product‑moment coefficient of
correlation. Similar indexes can be
calculated for individual isolation fields.
Group configuration
The concepts
of activity, isolation, and aggression fields can at most serve to process and
to condense original observations. They have more in common with matrices and
summary tables than with concepts which are part of theories. In other words,
they are instrumental concepts referring to ways of condensing observations;
they do not even point to forms of regularities and, still less, they do not
have any explanatory pretension as the concept of territoriality did, at least
implicitly. This might be a reason why these metrological concepts cannot
replace the concept of territoriality. Only refined data are subjected to
interpretation and only such refined information can be fed into a theory in
order to test it or to derive further data. There are several other instruments
which can be very useful to visualize global tendencies when it comes to the
comparison of several "fields" from different subjects.
Multidimensional scaling is one of them. Multidimensional scaling, or MDS for
short, is a set of mathematical techniques that enable the researcher to
uncover the "hidden structure" in data. It is simple matter to take a
map and to fill an entry in a table of distances between two cities. One just
applies a rule on the map and multiplies by the scale of the map. Now, suppose
you only have the distances between cities, as is the case with research
results from which the researcher wants to obtain a spatial configuration
taking into accounts proximities between individuals. The map can be easily obtained
with MDS, even though measures contain considerable noise due, for example, to
repeated measurement under non optimal conditions. In addition, a map can be
obtained from proximities or distances for which it is not known in advance
whether a two‑ or three‑dimensional representation will be adequate. A solution
using a least‑square monotonic regression is most often used in order to obtain
the best configuration that fits the original data. Results obtained from
recent experimental work on fish can serve to illustrate the application of MDS
to spacing patterns. In a recent research (Beaugrand et al., 1985), 16
populations, each of 4 male and 4 female green swordtail fish were observed in
54 litre tanks separated into two unequal volumes by a partition allowing swimming
from one volume to another only at the surface. Each population was observed on
2‑4 occasions, making a combined total of 50 observation periods of 2 hours
each. Each aquarium was gridded into 12 sectors and 8 samples of instantaneous
positions were taken each day at 15 minutes intervals. The tridimensional
coordinate of the position of each individual were simply noted. A distance
matrix was obtained for each sample by taking the differences between the
positions of each pair of individuals in the tank and by transforming these
differences into distances by the application of the theorem of Pythagoras.
These upper right half matrices of inter-individual distances, one for each
sample, 8 per period of observation, for 50 periods, had to be summarized in some
way to reveal the general (spatial) configuration of the group. MDS was applied
to these matrices and a one‑dimensional representation realized a satisfactory
stress of 0.10. A three‑dimensional representation is presented in the next Figure for
the sake of visual clarity (final stress of 0.002).
The obtained configuration
supports the following social and spatial regularity that was, anyhow, confirmed
by more conventional manners: the alpha male and the four females form a first
group in the large compartment of the tank, while the three subordinate males
form a second group in the small compartment. This spacing‑out pattern emerged
with neat regularity from the data and was apparently caused and maintained by
aggressive behaviour, especially by charges, initiated by the alpha male toward
male rivals. However, it is clear that the behaviour of the alpha male should
not qualify as territorial defence, since specific‑area‑linked dominance was
never realized in this study. Moreover, spatial proximities between the various
males and females in this study correspond perfectly with epigamic proximities
between the males and the females, and with agonistic distances among the
various males. Epigamic proximities were obtained by weighing frequencies of
sexual displays initiated by the males toward the females; epigamic behaviours
such as copulation attempts were given more weight than others such as sexual
pursuits at a distance. It was found that the alpha male was responsible to 80%
of all sexual activity and had privilege to behaviour leading to insemination
with a high probability in 85% of the cases. In that sense, the alpha male was
much more in proximity of the various females than the other males. When a
similar weighing system was applied to agonistic behaviour among the males,
agonistic distances obtained between males corresponded to the distances that
had been obtained from repeated sampling for spatial positions. The same
experiment was carried out with two compartments having the same volume and
essentially the same regularities were obtained with the following exception:
the spatial association of the alpha male with the females was found to be site
independent.
APPENDIX
2
Sociosystem
A set of
socially linked or connected animals can be considered as a sociosystem. Social
links, bonds and connections are special cases of social relations, but unlike
a mere relation, a connection makes some difference to the thing to which it is
related. If a link holds between two individuals, then at least one of them
will behave differently from the way he/she would behave if not so coupled.
When a connection affects two group members, it implies that one of them, or
both (reciprocally), act upon the other and can, potentially at least, modify
the latter's behavioural trajectory. This is the basic notion of behavioural
influence. A sociosystem also includes connections with the environment; its
immediate environment or milieu (i.e., the composition of the next supersystem)
must be included in the description of a sociosystem because the behaviour of
the members of the micro group depends critically on the nature of the immediate
milieu. In addition, members of the group use, exploit and transform elements
of their immediate environment. So, it should be clear that social structure
(internal, between members) and ecosociological structure (external, between
members and elements of the environment) are inter‑dependent.
What elicits
the formation of a sociosystem and keeps it together despite somewhat
(apparent, see the sociobiologists) divergent interests of its members ? The
social structure can be considered as the result of a balance of forces that
act in opposed directions: cooperation (mutuality) between group‑members tends
to social cohesion and competition between them, to social dispersion.
Cooperation can be said to occur when members in a group of socially connected individuals share goods for defence against predation, in order to increase their foraging efficiency, or to ease reproduction between mates, for mutual defence against parasites and infestations (Wittenberger, 1981). More schematically, if a and b are animals, then a and b cooperate with one another iff the social behaviour of each is valuable to the other or to a third animal. A thing is said to be valuable when it contributes to the degree of health of the individual. When cooperation concerns things (e.g. goods) of some kind, it is called sharing, and participation when it concerns activities. But cooperation may take the elementary form of being together at the same time and at the same place. Being two may have deterrent effects on predation. But, cooperation is only possible when the gains obtained from simply being together, from mutualism, sharing and social life override their costs (Wilson,1975; Barash, 1982; Wittenberger, 1981; Hinde, 1982; Deag, 1977). On the other hand, competition can be said to occur when one individual in a sociosystem does not share goods and does not participate in social activities because these goods or activities are more valuable consumed or practised individually than shared, or more valued when not obtained or practised at all. Competition may take the mild form of indifference for activities that else would be more adaptive when done cooperatively, of parasitism, of over‑exploitation of fundamental resources at the expense of others. In extreme cases, it takes the form of direct interference with exploitation or practice done by others, with or without the aid of aggressive behaviour. Competition is repulsive and highly dispersive in nature, both socially and spatially and temporarily. Individuals compete for food, shelters, females, sites for reproduction, &c.
The social or
internal structure among group‑members can be considered as the result of
cooperative and competitive forces acting in opposed directions. When the
balance is neatly in favour of cooperation, group structures exist; no
cooperation, no sociosystems. A set of conspecific animals forms a social
system if (and only if) each of them cooperates (on the average) with some
other members of the same set. The supremacy of cooperation over competition
contributes directly to the tightening of links between individuals;
connections between members are attractive and strong and it can be said that
the degree of integration is high. If the links are still positive but weak,
due to the small supremacy of cooperation over competition in a group, the
degree of integration is low. However, if the links are very repulsive, there
is no systemicity or integration at all. The highest level of integration
corresponds to no differentiation: all individuals are equivalent. But this is
an extreme case since there are always elements of competition in a group, and
there are differences between individuals in competitive abilities. The lowest
level of integration corresponds to complete differentiation, for example, in a
territorial situation where individuals do not participate to a society.
However, intermediate levels of integration or differentiation exist, for ex.,
in a true straight‑line hierarchy or even in despotism.
The balance between cooperation and competition has profound effects on the spatial structure or configuration of the sociosystem. By definition, space is an element of any social system: the configuration or spatial structure is a subset of the internal environment relating some of the system's components by contact and distance. The living individuals part of the group stand in definite spatial relations to one another. Moreover, spatial relationships can be used by them to create and reinforce connections among themselves. Hence, as stressed by Zayan et al. (1983), in the context of social activities, participation of individuals in a common activity such as feeding, copulation, rest, mutual preening, flocking, mobbing and group defence against predation, implies mutual tolerance both spatially and chronologically, namely synchronism and sharing of objects that stand in the same immediate space. Cooperation implies proximity. On the contrary, competition implies spatiotemporal incompatibility, particularly in the execution of individually valuable activities. While cooperation and participation imply decrease of inter‑individual distances and cohesion, spacing‑out is the rule when competition overrides cooperation. Individuals do not share the same places, avoid encountering each others, get desynchronized spatiotemporalily. Ultimately, extreme competition can lead to complete differentiation, to complete disconnection from the social structure, as in group expulsion and emigration, and to the formation of a territorial system.
So, social,
spatial and temporal organizations are networks of relations existing between
individuals, objects and resources of the milieu whose function is to optimize
the partitioning of resources and necessities of life and reproduction.
Plasticity of sociosystems
A given
sociosystem is capable of much plasticity in its social organization. Eco‑ethology
has tried, with some success, to explain much of the diversity of animal social
systems as adaptive consequences. But, while this variation in social systems
among species confirms the central predictions of eco‑ethology, it nevertheless
contradicts one of its basic premises according to which the social system of a
given species is a fixed product of natural selection (Lott, 1984). One has to
recognize that, for a given species, social predispositions are selected rather
than social systems per se (Mason, 1978). Instances of intra‑specific
variations in the social systems are rather the rule than the exception and
this plasticity suggests that in many species selection has produced a
predisposition that can have more than one social system outcome, depending on
the individual history and current circumstances. For example, coyotes within a
single population may defend pair territories, live in a nomadic fashion, or
form a well-organized pack (Bekoff and Wells, 1980). Chars in a single pool may
share the defence of a group territory around a reed bed at one side of the
pool, contribute to a hierarchy at the pool entrance, defend individual partial
territories in rotation, or even be transient from pool to pool without making
any social association in each pool visited (Jenkins, 1969). But, genuine
territoriality, as would have been indicated by the presence of territorial
mosaics, was never observed in adults chars. Similarly, the presence of
congeners of the opposite sex may have profound effects on the social structure
of a group of animals. Roosters and hens form hierarchies when placed in
isosexual groups (i.e. with individuals of the same sexual gender). But, when
hens are introduced to roosters already forming a hierarchy, despotism appear
and a harem is formed around the most dominant male. Very similar results are
obtained with green swordtail fishes (Xiphophorus helleri) kept in tanks
separated into two communicating compartments (Beaugrand et al., 1985). Intra‑specific
plasticity in social organization permits animals to efficiently exploit
momentary situations produced by increase or decrease of intra‑specific or
inter‑specific competition, climatic changes and habitat deterioration, prey
increase or decrease, increase or decrease of predation, etc. In environmental
conditions changing rapidly and frequently, intra‑specific plasticity of social
system should be favoured. It has also been recognized by several authors that
the concept of territoriality had much in common with dominance. Both may be
given a behavioural definition in terms of overt aggressive interactions giving
way to a certain form of precedence of one individual over another. In each
form of organization, there exists a mechanism permitting relative stability in
the precedence over resources, without the necessity of repeated overt fighting
whenever a resource might be contested. Individual recognition plays that
role of a convention for sure in a hierarchical system (Beaugrand and Zayan,
1985; Beaugrand et al., 1985) and is also most probably at work when
territories are adjacent, as attested by the "dear enemy phenomenon".
When territories are not adjacent, conflicts can be solved according to other
conventions. In a dominance hierarchy, an individual has privileged access to
resources regardless of their location within a group territory or home‑range.
With territorial organization, different individuals have priority of access to
resources depending on their location. Territoriality is thus a form of locus
dependent dominance. Dominance orders and territorial organization can be seen
to grade into each other in some ways rather than to exist as absolute
alternatives. Several species have been observed to adopt a territorial
organization at low population densities, but they shift to a hierarchical
organization when population densities increase. Wilson (1975) has qualified
this shift of "behavioural scaling". It may be more appropriate to
call it intra‑specific variation in social system (Lott, 1984). It may have a
genetic basis and may permit animals to adopt different organizations according
to environmental conditions. Noble (1939) and Greenberg (1947) had several
years ago recognized the interference of dominance hierarchies and the use of
space in fish. They had identified two kinds of territory in fish: the
territorial mosaic and partial territory. The territorial mosaic corresponded
to the situation where each adjacent territory holder was dominant over its own
territory but submitted when on an adjacent territory. This situation of
reversibility satisfies the definition of true territoriality as defined
previously. In the case of partial territory, "nip right" dominance
hierarchy superimposed over territorial organization in such a way that
residents were successful in defending their territory against subordinates but
submitted to a dominant intruder. Both forms of territorial organizations could
be found in the same species, at different phases of their life. In several
species of Salmonidae, pars and juveniles holding adjacent stations can
be said to organize in the form of territorial mosaics. As suggested by Jenkins
(1969), such a neat partitioning of space can be explained by the rather
uniform sizes, and possibly, level of aggression of the fishes, uniform
substratum conditions and minima displacements over the substratum. When
heterogeneity in size, more mobility and other asymmetries appear, minimal
conditions for the formation of dominance orders are present, and partial
territories can be formed. Partial territories seem to persist throughout adult
life in some trout species except during the reproductive seasons where couples
may defend small adjacent sites (territorial mosaics) on river beds in which
eggs are to be deposited. Patterns of social organization are thus extremely
diverse, and may vary within a species according to phases of their life, and
according to sexes. True territoriality, dominance orders, despotism, flocking‑schooling,
coloniality and lekking are alternative adaptive social systems. They are the
most appropriate solutions for a given species in a given situation, for
efficient exploitation of alimentary resources, coping with intra‑ and inter‑specific
competition, defence against predation, mate choice and fecundation, parental
care, defence against infectious diseases, and also protection against the
physical environment (e.g. strong currents). They vary within a single species
and even according to sexes within a species (e.g. females do not lek, males do
not take care of the young). Individuals involved in environmental
circumstances will manifest one type of organization when it is economically
(i.e. energetically) justified and physically possible but shift to other
types when either of those conditions does not hold. To each form of social
organization correspond one or several forms of spacing and distancing. The utilization
of space is, by definition, tied to resources that stand in definite positions
with regard to competitors, and social organization is partitioning of these
according to a balance resulting from cooperative and competitive forces.
REFERENCES
Beaugrand, J.P., Caron, J. and Comeau, L. 1985. Social organization of small heterosexual groups of green swordtails (Xiphophorus helleri, Pisces, Poeciliidae) under conditions of captivity. Behaviour, 91, 24‑60.
Beaugrand, J.P. and Zayan, R. 1985. An experimental model of aggressive dominance in Xiphophorus helleri (Pisces, Poeciliidae). Behavioural Processes, 10, 1‑52.
Black, C.H. and Wiley, R.H. 1977. Spatial variation in behavior in relation to territoriality in Dwarf Cichlids Apistogramma ramirezi. Z. Tierpsychologie, 45, 288‑297.
Brown, J.L. and Orians, G.H. 1970. Spacing patterns in mobile animals. Ann. Rev. Ecol. System., 1, 239‑262.
Bunge, M. 1979. Treatise on Basic Philosophy, Vol. 4. Ontology II: A world of Systems. (D. Reidel., Dordrecht).
Crofton, H.D. 1958. Nematode parasite populations in sheep on lowland farms. VI. Sheep behaviour and nematode infections. Parasitology, 48, 251‑260.
Davies, N.B. 1978. Ecological questions about Territorial Behaviour. In "Behavioural Ecology: an evolutionary approach". (Ed. J.R. Krebs and N.B. Davies). (Sinauer Ass., Sunderland, Mass.). pp. 317‑350.
Krebs, J.R. and Davies, N.B. 1981. An introduction to Behavioural Ecology. (Sinauer Ass., Sunderland, Mass.).
Holmes, R.T. and Pitelka, F.A. 1968. Food overlap among coexisting sandpipers on northern Alaskan tundra. System. Zool., 17, 305‑318.
Hediger, h. 1963. The evolution of territorial behavior. In "The social life of Early Man". (Ed. S.L. Washburn). (Methuen, London). pp. xxx‑xxx.
Howard, E. 1920. Territory in Bird Life. (Collins, London).
Hunter, R.F. and Davies, G.E. 1963. The effect of method of rearing on the social behaviour of Scottish Blackface sheep. Animal Produc., 5, 183‑194.
McBride, G., Parer, I.P. and Foenander, F. 1969. The social organisation and behaviour of the feral domestic fowl. Anim. Behav. Monog., 2, 125‑181.
McBride, G. 1971. Theories of animal spacing: the role of flight, fight and social distance. In "Behaviour and Environment". (Ed. A.H. Esser). (Plenum, New York). pp. 53‑68.
Nice, M.M. 1941. The role of territory in bird life. Am. Midl. Natur., 26, 441‑487.
Noble, G.K. 1939. The role of dominance in the social life of birds. Auk, 56, 263‑273.
Pianka, E.R. 1975. Niche relations of desert lizards. In "Ecology and Evolution of Communities". (Ed. M.L. Cody and J.M. Diamond). (Belknap Press, Cambridge, Mass.). pp. 292‑314.
Schoener, T.W. 1968. Sizes of feeding territories among birds. Ecology, 49, 123‑142.
Waser, P.M. 1976. Cercocebus albigena: Site attachment, avoidance, and intergroup spacing. Am. Natur., 110, 911‑935.
Waser, P.M. 1977. Individual recognition, intragroup cohesion, and intergroup spacing: Evidence from sound playback to forest monkeys. Behaviour, 60, 28‑74.
Waser, P.M. 1975b. Spatial associations and social interactions in a "solitary" ungulate: The bushbuck Tragelaphus scriptus (Pallas). Z. Tierpsychol., 37, 24‑36.
Waser, P.M. and Wiley, R.H. 1979. Mechanisms and evolution of spacing in animals. In "Handbook of Behavioral Neurobiology, Vol. 3: Social Behavior and Communication" (Ed. P. Marler and J.G. Vandenbergh). (Plenum, New York). pp. 159‑223.
Wiley, R.H. 1973. Territoriality and non‑random mating in sage grouse Centrocercus urophasianus. Animal Behaviour Monograph, 6, 85‑169.
Wilson, E.O. 1975. Sociobiology: the new synthesis. (Belknap Press, Cambridge, Mass.).
Wittenberger, J.F. 1981. Animal Social Behavior. (Duxbury Press, Boston).
Zayan, R.C., Doyen, J. and Duncan, I.J.H. 1983. Social and space requirements for hens in battery cages. In "....". (Ed. Baxter and Baxter). (Martinus Nijhoff, The Hague).
Aucun commentaire:
Enregistrer un commentaire