The Concept of Environmental Grain

The most historically prominent effort to portray surroundings from the organism’s standpoint is the notion of ‘grain’ (MacArthur and Levins 1964). This attempts to relate environmental variation to an individual’s size, longevity, and activity, especially as regards foraging. In other words, the idea involves more than simply a matter of relative scale of the organism. Not only is the environment per se important but also how the organism experiences or samples it. This is influenced in turn by various attributes such as the kind of organism, its size (state of maturity), and growth form (which includes mobility or sessility) (Chap. 5). Depending on how resource units or ‘grains’ of varying size are presented to an individual or species, environments can be classified as being either fine- or coarse-grained in space or time. The idea is summarized as follows:

- Fine-grained environment
Space: Mobile individual moves among many small patches in its lifetime, consumes resources in proportion to which they occur.
Time: Since the environment undergoes change often in small increments, the effective environment is the average of the units.
Examples: Differences among fruits on forest floor for Drosophila adults (mobile); mobile larvae of barnacles; a bird foraging among many tree types; a bacterial or fungal clone feeding as a generalist.

- Coarse-grained environment
Space: Nonmobile organism spends most or all of life in large patches relative to size of organism.
Time: Environment varies over long periods relative to life span of organism; seasonal food.
Examples: Differences among fruits on forest floor for Drosophila larvae (relatively immobile, cf. adults above); individual bacterial or yeast cell on a leaf; sessile adults of barnacles.

Fine-grained patches in space are small in a relative sense, and the organism does not really distinguish among them: they are used in the proportion in which they occur. The environment is fine-grained in time if it is experienced in many small doses, or if larger fluctuations are encountered by a long-lived organism over many years. Conversely, coarsegrained environments are sufficiently large so that the organism ‘chooses’ among them (space) or spends its life in a single environment (time). An oak-hickory forest appears fine-grained to a scarlet tanager, which forages in both oak and hickory trees, but as coarsegrained to a defoliating insect, which, as a specialist, attacks only the oaks (Chap. 1 in Mac- Arthur and Connell 1966). There are instances for sessile or sedentary organisms (e.g., trees; MacArthur and Levins 1964) where an environment appears coarse-grained to individuals but fine-grained to the widely distributed species. In general, microscale heterogeneity increases as organism size decreases; small creatures, especially if they are relatively sedentary, experience the environment in coarse-grained fashion, living out their lives on a leaf or under a rock.

How organisms might experience the grain in their environments relates to competition among related species, the evolution of specialization in resource use, and how environmental instability affects the degree of specialization. MacArthur and Piankas (1966) theoretical analysis predicts that specialist feeders are favored over generalists in fine-grained systems and the converse in coarse-grained systems, where the generalist can compensate for its less efficient feeding with lower hunting time (see also Chap. 3).

Mathematical models based on the grain concept have been developed to describe selection in heterogeneous environments (see especially Levins 1968) and some experimental tests have been reported. Baythavong (2011) used the logic that plant lineages in fine-grained environments (spatial scale of environmental variation less than dispersal distance of a species) experience habitats different from their maternal parents and selection should favor phenotypic plasticity (discussed later). In coarse-grained environments (spatial variation on scales exceeding dispersal) the dispersing progeny likely experience environments similar to their parents; selection should favor genetic differentiation more so than plasticity (see Sect. 7.5).

Data on attributes of the annual plant Erodium cicutarium (redstem filaree, Geraniaceae) growing in closely adjacent serpentine and nonserpentine soil patches in California s upported the hypothesis. Such soils vary considerably in edaphic and biotic properties within spatial scales of 0-10 m.

Notwithstanding the intuitive appeal of Levins’ model and others like it, there are some caveats worth mentioning. First, it is unclear what is being maximized here by natural selection (Hamilton 1970). For example, at the population level, this could be the mean of individual fitness (as usually stated) over all environments. Selection in this case eliminates extreme individuals, preserving those near the population mean, and is said to be ‘stabilizing. Alternatively, it could be acting directionally to elevate the minimum of individual fitness in the population over all environments.

This is the so-called ‘Maximin strategy’ (Templeton and Rothman 1974). To quote the authors in part (p. 425), “under this strategy the optimum population is that population which maximizes its minimum fitness over all environments instead of maximizing its average fitness. By adopting this strategy, the population further insures its survival by letting the worst conditions it experiences dominate in importance.” Levins (1968) evidently assumes the former mode of natural selection with respect to grain and his predictions have been challenged (Hamilton 1970; Strobeck 1975; Templeton and Rothman 1974).

Second, the grain concept is somewhat abstract and would seem useful mainly conceptually. The level at which a resource or any other environmental attribute appears fine- or coarse-grained is arbitrary; also, whether a species is specialized relative to another in a meaningful way can be in the eye of the beholder (Futuyma and Moreno 1988). Levins’ model was designed to isolate and address specifically the resource component of the environment. Given the simplifying assumptions in a modeling context, this can be a powerful investigative tool. Yet, in reality, multiple life history facets interact among themselves, with the resource, and with the forager, in complex fashion.

Furthermore, as is apparent from the examples summarized above, how any particular organism views the surroundings will vary with its life cycle stage. The sluggish caterpillar sees life in coarse-grained fashion, while the butterfly it will become flits about a fine-grained environment. Finally, different parts of the same modular organism (modules or in some cases ramets; Chap. 5) are exposed to potentially quite different environments. The mathematical models of Levins (1968) and others as applied to hypothetical situations are, of necessity, simplified abstractions. To reemphasize the introductory remarks to this chapter: One should not overlook the fact that we do not know, and can never really know very accurately, how any other organism experiences the world.