Evidence from community patterns niche differentiation

The various types of niche differentiation in animals and plants were outlined in Chapter 8. On the one hand, resources may be utilized differentially. This may express itself directly within a single habitat, or as a difference in microhabitat, geographic distribution or temporal appearance if the resources are themselves separated spatially or temporally. Alternatively, species and their competitive abilities may differ in their responses to environmental conditions. This too can express itself as either microhabitat, geographic or temporal differentiation, depending on the manner in which the conditions themselves vary. Niche complementarity

In one study of niche differentiation and coexistence, a number of species of anenome fish were examined near Madang in Papua New Guinea (Elliott & Mariscal, 2001). This region has the highest reported species richness of both anenome fishes (nine) and their host anenomes (10). Each individual anenome is typically occupied by individuals of just one species of anenome fish because the residents aggressively exclude intruders (although aggressive interactions are less frequently observed between anenome fish of very different sizes). Anenomes seem to be a limiting resource for the fishes because almost all the anenomes were occupied, and when some were transplanted to new sites they were quickly colonized and the abundance of adult fish increased. Surveys at three replicate reef sites in four zones (nearshore, mid-lagoon, outer barrier reef and offshore: Figure 19.3a) showed that each anenome fish was primarily associated with a particular species of anenome and each showed a characteristic preference for a particular zone (Figure 19.3b). Different anenome fish that lived with the same anenome were typically associated with different zones. Thus, Amphiprion percula occupied the anenome Heteractis magnifica in nearshore zones, while A. perideraion occupied H. magnifica in offshore zones. Elliott and Mariscal concluded that coexistence of the nine anenome fishes on the limited anenome resource was possible because of the differentiation of their niches, together with the ability of small anenome fish species (A. sandaracinos and A.

the ghost of competition past expectations from competition theory evidence from community patterns... ... in anenome fishes in Papua New Guinea,...

1 km

Bismark Sea

1 km

Bismark Sea

Fish species

A. percula

E A. perideraion


■ A. clarkii E A. chrysopterus I A. leucokranos ■ A. melanopus


■ A. clarkii K A. chrysopterus □ A. sandaracinos M A. leucokranos

5o c 25

Heteractis magnifica

Heteractis magnifica

n = Bo n = 28 Heteractis crispa n = Bo n = Bo n = 28 Heteractis crispa n= 1o2 n=B

Stichodactyla mertensii n = Bo

Stichodactyla mertensii

n = 4 n = 17 n = 54 n = 25 Nearshore Mid-lagoon Outer barrier Offshore

Figure 19.3 (a) Map showing the location of three replicate study sites in each of four zones within and outside Madang Lagoon (N, nearshore; M, mid-lagoon; O, outer barrier reef; OS, offshore reef). The white areas indicate water, heavy stippling represents coral reef and light stippling represents land. (b) The percentage of three common species of anenome (Heteractis magnifica, H. crispa and Stichodactyla mertensii) occupied by different anemone fish species (Amphiprion spp., in key on left) in each of four zones. The number of anenomes censused in each zone is shown by n. (After Elliott & Mariscal, 2001.)

leucokranos) to cohabit the same anenome with larger species. The pattern is consistent with what would be expected of communities molded by competition (specifically predictions 1 and 3 above).

Two fUrther points, illustrated by the anenome fish, are worth highlighting. First, they can be considered to be a guild, in that they are a group of species that exploit the same class of environmental resources in a similar way (Root, 1967). If interspecific competition is to occur at all, or if it has occurred in the past, then it will be most likely to occur, or to have occurred, within guilds. But this does not mean that guild members necessarily compete or have necessarily competed: the onus is on ecologists to demonstrate that this is the case.

The second point about the anenome fish is that they demonstrate niche complementarity. That is, within the guild as a whole, niche differentiation involves several niche dimensions, and species that occupy a similar position along one dimension (anenome species used) tend to differ along another dimension (zone occupied). Complementary differentiation along several dimensions has also been reported for guilds as diverse as lizards (Schoener, 1974), bumblebees (Pyke, 1982), bats (McKenzie & Rolfe, 1986), rainforest carnivores (Ray & Sunquist, 2001) and tropical trees (Davies et al., 1998), as described next. Niche differentiation in space

Trees vary in their capacity to use resources such as light, water and nutrients. A study in Borneo of 11 tree species in the genus Macaranga showed marked differentiation in light requirements, from extremely high light-demanding species such as M. gigantea to shade-tolerant species such as M. kingii (Figure 19.4a). The average light levels intercepted by the crowns of these trees tended to increase as they grew larger, but the ranking of the species did not change. The shade-tolerant species were smaller (Figure 19.4b) and persisted in the understory, rarely establishing in disturbed microsites (e.g. M. kingii), in contrast to some of the larger, high-light species that are pioneers of large forest gaps (e.g. M. gigantea). Others were associated with intermediate light levels and can be considered small gap specialists (e.g. M. trachyphylla). The Macaranga species were also differentiated along a second niche gradient, with some species being more common on clay-rich soils and others on sand-rich soils (Figure 19.4b). This differentiation may be based on nutrient availability (generally higher in clay soils) and/or soil moisture availability (possibly lower in the clay soils because of thinner root mats and humus layers). As with the anenome fish, there is evidence of niche complementarity among the Macaranga species. Thus, species with similar light requirements differed in terms of preferred soil textures, especially in the case of the shade-tolerant species.

The apparent niche partitioning by Macaranga species was partly related to horizontal heterogeneity in resources (light levels in rela tion to gap size, distribution of soil types) and partly to vertical heterogeneity (height achieved, depth of root mat).

Ectomycorrhizal fungi have also been shown to exploit resources differentially in the vertical plane in the floor of a pine (Pinus resinosa) forest. Until recently, it was not possible to study the in situ distribution of ectomycorrhizal hyphae, but now DNA analyses allow the identification of putative species (even in the absence of species names) and permit their distributions to be compared. The forest soil had a well-developed litter layer above a fermentation layer (the F layer) and a thin humified layer (the H layer), with mineral soil beneath (the B horizon). Of the 26 species separated by the DNA analysis, some were very largely restricted to the litter layer (group A in Figure 19.5), others to the F layer (group D), the H layer (group E) or the B horizon (group F). The remaining species were more general in their distributions (groups B and C). Niche differentiation in time

Intense competition may, in theory, be avoided by partitioning resources in horizontal or vertical space, as in the examples above, or in time (Kronfeld-Schor & Dayan, 2003), for example, by a staggering of life cycles through the year. It is notable that two species of mantids, which feature as predators in many parts of the world, commonly coexist both in Asia and North America. Tenodera sinensis and Mantis religiosa have life cycles that are 2-3 weeks out of phase. To test the hypothesis that this asynchrony serves to reduce interspecific competition, the timing of their egg hatch was experimentally synchronized in replicated field enclosures (Hurd & Eisenberg, 1990). T. sinensis, which normally hatches earlier, was unaffected by M. religiosa. In contrast, the survival and body size of M. religiosa declined in the presence of T. sinensis. Because these mantids are both competitors for shared resources and predators of each other, the outcome of this experiment probably reflects a complex interaction between the two processes.

In plants too, resources may be partitioned in time. Thus, tundra plants growing in nitrogen-limited conditions in Alaska were differentiated in their timing of nitrogen uptake, as well as the soil depth from which it was extracted and the chemical form of nitrogen used. To trace how tundra species differed in their uptake of different nitrogen sources, McKane et al. (2002) injected three chemical forms labeled with the rare isotope 15N (inorganic ammonium, nitrate and organic glycine) at two soil depths (3 and 8 cm) on two occasions (June 24 and August 7) in a 3 X 2 X 2 factorial design. Concentration of the 15N tracer was measured in each of five common tundra plants in 3-6 replicates of each treatment 7 days after application. The five plants proved to be well differentiated in

... and in tundra plants in Alaska their use of nitrogen sources (Figure 19.6). Cottongrass (Eriophorum vaginatum) and the cranberry bush (Vaccinium vitis-idaea) both relied on a combination of glycine and ammonium, but cranberry obtained more of these forms early in the growing season and at a shallower depth than cottongrass. The evergreen shrub Ledum palustre and the dwarf birch (Betula nana) used mainly ammonium but L. palustre obtained more of this form early in the season while the birch exploited it later. Finally, the grass Carex bigelowii was the only species to use mainly nitrate. Here, niche complementarity can be seen along three niche dimensions and differences in timing of use may help explain the coexistence of these species on a limited resource. Niche differentiation - apparent or real? Null models the aim of demonstrating that patterns are not generated merely by chance

Many cases of apparent resource partitioning have been reported. It is likely, however, that studies failing to detect such differentiation have tended to go unpublished. It is always possible, of course, that these 'unsuccessful' studies are flawed and incomplete, and that they have failed to deal with the relevant niche dimensions; but a number have been sufficiently beyond reproach to raise the possibility that in certain groups resource partitioning is not an important feature. Strong (1982) studied a group of hispine beetles (Chrysomelidae) that commonly coexist as adults in the rolled leaves of Heliconia plants. These long-lived tropical beetles are closely related, eat the same food and occupy the same habitat. They would appear to be good candidates for demonstrating resource partitioning. Yet, Strong could find no evidence of segregation, except in the case of just one of the 13 species studied, which was segregated weakly from a number of others. The beetles lack any aggressive behavior, either within or between species; their host specificity does not change as a function of co-occupancy of leaves with other species that might be competitors; and the levels of food and habitat are commonly not limiting for these beetles, which suffer heavily from parasitism and predation. In these species, resource partitioning associated with interspecific competition does not appear to structure the community. As we have seen, this may well be true of many phytophagous insect communities. Plant

Figure 19.4 (right) (a) Percentage of individuals in each of five crown illumination classes for 11 Macaranga species (sample sizes in parentheses). (b) Three-dimensional distribution of the 11 species with respect to maximum height, the proportion of stems in high light levels (class 5 in (a)) and proportion of stems in sand-rich soils. Each species of Macaranga is denoted by a single letter: G, gigantean; W, winkleri; H, hosei; Y, hypoleuca; T, triloba; B, beccariana; A, trachyphylla; K, kingii; U, hullettii; V, havilandii; L, lamellata. (After Davies et al., 1998.)



1 1

1 1




1 1

1 1


1 II 1

2.8 ZZL




1 1 III

Crown illumination index

V (n = 103)


U (n = 229)

1 1

1 li i

Crown illumination index

Figure 19.5 The vertical distribution of 26 ectomycorrhizal fungal species in the floor of a pine forest determined by DNA analysis. Most have not been named formally but are shown as a code (TRFLP, terminal restriction fragment length polymorphism). The vertical distribution histograms show the percentage of occurrences of each species in litter, the F layer, the H layer and the B horizon. (After Dickie et al., 2002.)


Vertical distribution


Unknown TRFLP 009 Unknown TRFLP 010 Ramaria concolor Unknown TRFLP 007 Tylopilus felleus Unknown TRFLP 008 Unknown TRFLP 006 Lactarius sp. Unknow_nJRFLP_005_ _ Trichoderma sp. Unknown TRFLP 001 Unknown TRFLP 002 Scleroderma citrinum Russula_ s_p_(white 1_)_ _ Unknown TRFLP 003 Clavulina cristata

Cenococcum geophjlum

Unknown TRFLP 004 Unknown TRFLP 014 Suillus intermedius

Unknown TRFLP 013 Russula sp. (white 2)

Aman[ta_ rubescens___

Unknown TRFLP 015 Amanita vaginata w////m I

20 40 60 80 100 Percentage occurrences


H layer


F layer

B horizon

studies involving taxa as diverse as phytoplankton (see Figure 19.1) and trees (Brokaw & Busing, 2000) have similarly failed to provide evidence consistent with a strong role for niche partitioning in promoting coexistence and species diversity. Whilst patterns consistent with a niche differentiation hypothesis are reasonably widespread, they are by no means universal.

Moreover, a number of workers, notably Simberloff and Strong, have criticized what they see as a tendency to interpret 'mere differences' as confirming the importance of interspecific competition. Such reports beg the question of whether the differences are large enough or regular enough to be different from what might be found at random among a set of species. This problem led to an approach known as null model analysis (Gotelli, 2001). Null models are models of actual communities that retain certain of the characteristics of their real counterparts, but reassemble the components at random, specifically excluding the consequences of biological interactions. In fact, such analyses are attempts to follow a much more general approach to scientific investigation, namely the construction and testing of null hypotheses. The idea (familiar to most readers in a statistical context) is that the data are rearranged into a form (the null model) representing what the data would look like in the absence of the phenomenon under investigation (in this case species interactions, particularly interspecific competition). Then, if the actual data show a significant statistical difference from the null hypothesis, the null hypothesis is rejected and the action of the phenomenon under investigation is strongly inferred. Rejecting (or falsifying) the absence of an effect is reckoned to be better than confirming its presence, because there are well-established statistical methods for testing whether things are significantly different (allowing falsification) but none for testing whether things are 'significantly similar'.

Lawlor (1980) looked at 10 North American lizard communities, consisting of 4-9 species, for which he had estimates of the amounts of each of null hypotheses are intended to ensure statistical rigor a null model of food resource use in lizard communities...

Figure 19.6 Mean uptake of available soil nitrogen (± SE) in terms of (a) chemical form, (b) timing of uptake and (c) depth of uptake by the five most common species in tussock tundra in Alaska. Data are expressed as the percentage of each species' total uptake (left panels) or as the percentage of the total pool of nitrogen available in the soil (right panels). (After McKane et al., 2002.)

Was this article helpful?

0 0
Lawn Care

Lawn Care

The Secret of A Great Lawn Without Needing a Professional You Can Do It And I Can Show You How! A Great Looking Lawn Doesnt Have To Cost Hundreds Of Dollars Or Require The Use Of A Professional Lawn Care Service. All You Need Is This Incredible Book!

Get My Free Ebook

Post a comment