Northern Prairie Wildlife Research Center
How is the particular habitat distribution of a bird species achieved? Obviously, a bird species will not distribute itself over more than one habitat if it does not have a choice of more than one. Therefore, at least some of the birds must have exposure to several habitats which can be selected, and the distribution is achieved by habitat selection. The choice of habitat will depend on the relative suitabilities of the available habitats. Suitability refers simply to the "goodness" of a habitat in terms of its potential for permitting high reproductive success. So we see that habitat suitability determines habitat selection which in turn determines the habitat distribution.
Fretwell and Lucas (1969) have defined two types of theoretical habitat distributions into which we may classify species of birds. These two distribution types apply most directly to those species which exhibit territorial behavior during their breeding seasons.
The ideal free distribution is a type which carries several assumptions about the birds involved. The most important of these assumptions is that all individual birds will settle in the habitat that is most suitable to them. That is to say, the birds are ideally adapted to their environment, and when presented with a choice of habitats will always choose to settle in the one which offers the best chance for survival and reproductive success. The ideal free distribution also carries the assumption that habitat suitability always decreases with an increase in population density of the bird species in question, so that maximum suitability occurs at a density near zero. Still another assumption (which may not be sound) is that all individuals in a habitat have identical expected success rates. This means that the birds are free to enter any habitat on an equal basis with those birds already residing there. Thus, all individuals in the ideal free distribution have a free choice of habitats and will settle in the one of highest suitability, according to the above assumptions.
Now consider a bird species which exhibits territorial behavior in a manner such that the residents of a given habitat will restrict non-residents from settling in the same habitat. Such territorial behavior would impose a limit on the density in that habitat and would supposedly maintain the suitability of the habitat at a level corresponding to this maximum density so that residents could expect a high degree of breeding success. A bird species of this type would not be distributing itself in an ideal free manner since not all birds would be free to enter any habiat on an equal basis with those already there. The distribution which eventually results from such behavior has been termed by Fretwell and Lucas (1969) as the ideal dominance distribution. This name results from the fact that the free assumption of the ideal free distribution does not hold and that all individuals of this distribution are not equally aggressive, so that "social dominance hierarchies" are extablished. Now that the two types of habitat distributions have been defined, a closer look must be taken at how they are achieved.
Suppose there are three habitats available to a bird species entering an area and that the maximum suitabilities of these habitats are not the same. If this bird species is an ideal free distribution - type, those birds which enter the area first will settle in the habitat with highest suitability. As more birds occupy this habitat, the suitability will go down. At some density, the suitability of this habitat will be lowered to the level of the maximum suitability of the second most desirable habitat. Now any incoming birds can settle in either of these two habitats, and, being ideal, they will settle in these habitats in such proportions that the two habitat suitabilities remain equal. But since density in each habitat increases, the suitability of each goes down. At some density the suitabilities may be lowered to the maximum level of the least desirable habitat. Now incoming birds may occupy any one of the three habitats and expect equal chances of success in any of them. Thus we see that in any two or three habitats occupied by an ideal free species, the density of birds in the respective habitats may differ greatly, yet the respective habitat suitabilities will be equal. In other words, one would not observe a positive correlation between density and suitability among the occupied habitats.
Again consider more than one habitat, each one with a different maximum suitability. An ideal dominance distribution - type species entering the area will first occupy the habitat of highest suitability. Again, suitability decreases with increasing density. Now incoming birds may be restricted from settling by the aggressive territorial behavior of the residents. Because of such behavior, the apparent suitability of the habitat from the point of view of the incoming bird will be lower than the actual suitability of that habitat. Thus, a second, unoccupied habitat of lower actual suitability may appear just as good as the habitat of higher actual suitability and incoming birds will begin to occupy the second habitat. This is consistent with the assumption that the birds are ideal since they are settling in habitats of equal apparent suitability. Apparent suitability of a habitat will always be lower than actual suitability as long as there are resident birds in the habitat. So apparent suitability will always decrease with increased density as does the actual suitability. The habitat distribution will be achieved through settling of poorer habitats as the apparent suitability of high actual - suitability habitats is lowered to a level equal to that of the actual suitability in the poor habitats. In this kind of a situation we would expect to see a positive correlation between density and suitability among the occupied habitats. The habitats could differ greatly in density and the suitability of the habitats would also differ, the highest - density habitat having the highest actual suitability.
Of course, no bird species can be expected to perfectly demonstrate the ideal free or ideal dominance distribution since no bird is ideal. However, bird species can approximate these distributions in real life situations. With the basis of the two theoretical distributions, it is possible to examine the relationships of territoriality to habitat distribution. The main territorial hypotheses that currently exist can be fitted into the framework of the theoretical distributions just described.
Basically, three hypotheses for the role of territoriality have been advanced. The density assessment or density cue hypothesis was described by Kluyver and Tinbergen (1953) as a result of their work with titmice. They observed great differences in density of titmice in two habitats, one habitat supporting a high density of titmice and the other always a low density. When total population levels were low, most individuals were found in the more attractive habitat. They suggested that territorial behavior exhibited by settled individuals was used as a cue to density by unsettled birds so that these birds could avoid the densely populated habitats where the chances of reproductive success had presumably been lowered as a result of the high density of birds there. No differences in breeding success were observed between the two habitats meaning that suitabilities were equal. Thus, territorial behavior of this type certainly leads to an approximation of the ideal free distribution -- widely different densities yet equal suitabilities. Density assessment cues other than territoriality could conceivably exist that would also allow for achievement of an ideal free distribution, but Kluyver and Tinbergen observed none.
The second main hypothesis for the role of territoriality assigns a density - limiting function to territorial behavior. This was alluded to earlier in the discussion of the ideal dominance distribution. Huxley (1934) was the first to make detailed note of this phenomenon. The gist of this hypothesis is that aggressive territorial behavior by resident birds limits the density in that habitat at a certain level by forcing unsettled birds into suboptimal habitats. As noted earlier, this invalidates the "free" assumption of the ideal free distribution. This type of territorial behavior will lead to the ideal dominance distribution in which habitats have different densities and different suitabilities, the habitat with higher density having higher suitability. But the apparent suitabilities from the point of view of unsettled birds is still assumed to be equal since this type of territorial behavior will make a habitat appear less suitable than it actually is. Thus, the ideal assumption still holds. Zimmerman (1971) described territoriality in the dickcissel as apparently having a density - limiting function. Brown (1969) assigns a density - limiting function to most territorial behavior, but only limiting on a local level and only if a surplus of individuals can be shown to exist. Nice (1941) cites several early authors who imply that territoriality functions to limit density.
Lack (1964) and Johnston (1961) have promoted the hypothesis that territorial behavior serves only as a mechanism of spacing individuals. If territorial behavior in a species only spaces individuals and does not limit density or cue density, then it would have virtually no bearing on where the individual birds settled. This type of territorial behavior could neither affect nor effect the habitat distribution of a species.
Other hypotheses for the role of territorial behavior have been developed but most, if not all of them, it seems to me, could easily be considered as modifications of one of the three hypotheses just described. Verner (1977) ascribes to the territorial behavior of a bird the role of preventing conspecific birds from establishing territories nearby so that that bird's offspring would comprise a larger proportion of the total number of young produced by the population. Brown (1964) also seems to suggest this. But this hypothesis seems very similar to the density - limiting hypothesis and would most likely result in the same type of habitat distribution.
Applying the theory which has thus far been worked out requires that the habitat distribution of the bird species being studied be identified as approximating either the ideal free or ideal dominance distribution or neither. Once the distribution has been identified, it can be taken as evidence in support of the hypothesis for territorial behavior corresponding to that distribution. This is assuming, of course, that if territorial behavior has one of the roles hypothesized above, it will result in a distribution approximating the ideal distribution described for that hypothesis. In arriving at conclusions, one must make sure that the distribution observed is a result of territorial behavior and not some other phenomenon. Also it must be remembered that no bird is really ideal and thus an individual bird's judgment of habitat suitability will not be perfect.
Based on these theoretical developments, Fretwell and Lucas (1969) give the following ways in which the role of territorial behavior in the habitat distribution of a territorial bird can be determined: If high density habitats show consistently higher success rates (ideal dominance distribution) and if no density - limiting mechanism other than territoriality is evident, the role of the territorial behavior is most probably to limit density. If suitabilities are equal, but densities are not, and no other density cue is apparent, then the territorial behavior is presumed to be serving as a density cue mechanism. If neither of the above two conditions are met, the territorial behavior observed must be only functioning to space the individuals.
It was the objective of this research to observe the habitat distributions and territorial behavior of the dickcissel and red-winged blackbird, and from the data thus obtained, to arrive at tentative definitions of the functions of the territorial systems employed by these two species. The preceeding introduction is based in large part on the theoretical work of Fretwell and Lucas (1969).