Small Mammal Assemblages Research Articles

LONG-TERM DYNAMICS OF SMALL-MAMMAL POPULATIONS IN ONTARIO

We analyzed 43 yr of live-trapping data for eight species of small mammals in Algonquin Provincial Park, Ontario. Our primary objective was to test whether complex nonlinear models are necessary to describe dynamics of the Algonquin rodent populations. Variation in abundance among species was related to mean abundance by a simple power function with an exponent of 1.77, implying that population variability did not increase with mean density as much as one might expect for strictly statistical reasons. Time-series analyses of annual population densities indicated no significant autocorrelation functions for five species. Southern red-backed voles, red squirrels, and flying squirrels had significant autocorrelations, but only flying squirrels had an autocorrelation function suggestive of cyclic population dynamics. Per capita rates of population growth were density-dependent in all eight species, although in most cases there was substantial deviation around the fitted regression lines. Response surface models with one- and two-year lags significantly improved the statistical fit to data for three species, but only one of these response surface models was sustainable in the face of realistic stochastic variation in per capita growth. These results suggest that simple logistic models are adequate for predicting the long-term dynamics of the Algonquin small-mammal assemblage. All eight species showed evidence of synchronized population fluctuations over time, suggesting trophic linkages due to shared food resources, shared predators, or both.

Factors influencing amphibian and small mammal assemblages in central Appalachian forests

We studied terrestrial amphibian and small mammal assemblages with drift fences and pitfall traps in five forested stands during 1987–1988 on Shenandoah Mountain in the George Washington National Forest, Virginia, USA. Eleven species of salamanders, five frogs, five shrews, and seven rodents were monitored. Amphibians were significantly more abundant in forest stands consisting of mature hardwoods than in a recent clearcut and a white pine plantation. Although there was considerable variation in abundance among species in the five stands, small mammal abundance was high in all five habitats studied. Amphibian species diversity (Shannon Index) was less than half that for small mammals because red-backed salamanders ( Plethodon cinereus) were dominant in most assemblages. Amphibian and small mammal diversity and total species richness were not related to estimated stand age, total number of canopy trees, tree diversity, or frequency of underground rocks. Maintenance of amphibian biodiversity requires the combination of mature hardwoods and wetland habitats (e.g. wildlife ponds and seepages). Most of the small mammals encountered were habitat generalists. Management focus on mature hardwood forests would maintain populations of small mammals requiring cool, moist situations in upper-elevation habitats in the central Appalachian Mountains.

The effects of spatial patterns in habitat quality on community dynamics within a site

Metapopulation studies of single species have shown that the size and spatial arrangement of patches of assumed uniformly ‘suitable’ habitat can influence their population dynamics and persistence. We investigated whether variation in the spatial arrangement of ‘suitable’ habitat of varied quality within a single site can affect the abundance and persistence of interacting species. We accomplished this by extending a field–based spatial simulation model of four interacting species at two trophic levels (an endangered butterfly, its larval food–plant, and two ants). The habitat on sites with the same average and range of qualities was rearranged to give varying degrees of local spatial heterogeneity or ‘site ruggedness’. We found that the ant species that acts as host to the butterfly caterpillars decreased with site ruggedness. The impact on the butterfly was more substantial: it often failed to persist on very rugged sites. Despite being free–ranging over the whole area, the butterfly9s persistence depends on the arrangement of habitat quality at a finer spatial scale, due to its interactions with species possessing narrower habitat niches and more localized dispersal. Ruggedness also influenced the rate of recovery of the host ant, and hence community structure, for more than a century following the butterfly9s extinction.

Small Mammal Fossil Assemblages as Indicators of Environmental Change in Northern Corsica during the Last 2500 Years

The small Monte di Tuda cave is filled with a slightly disturbed, 2 m thick deposit. Very rich small vertebrate bone assemblages from barn owl pellets have been sampled extensively using special techniques to avoid contamination. Stratigraphical analysis and accelerator radiocarbon dating on the bones themselves both point to a rapid sedimentation rate during the last 2500 years, with a break in the building up of sediment between 1960 and 610 bp. Multivariate analyses of the frequencies of the small mammal species show that modern species had approximately the same ecology throughout the time sequence, and give, for the first time, an idea of the ecological preferences of the extinct endemic species of Corsica. The general development of the faunal spectra shows a general increase in human impact (clearance of the vegetation), through the sediment sequence. Other multivariate analyses indicate a sequence of at least three agricultural cycles: the first one during the Early Roman period, the second one between the Late Roman period (2 –5th centuries ad) and the Pisa Peace (11 –13th centuries ad), and the last one during the 14 –19th centuries ad. The two earlier ones were mainly concerned with cereal cultivation, the later with tree cultivation. Whilst most of these cycles have only a local significance, two phenomena have a more extensive significance: (1) the immigration of Rattus rattusto Corsica and probably to the whole North-Western Mediterranean Basin, dated between the 4th –2nd centuries cal- bc; and (2) the mass extinction of the small endemic species, dating either to the Late Roman period or to the 11 –13th centuries adand probably resulting from large scale agricultural deforestation. In addition, the small mammal assemblages recovered appear to reflect a wetter climate at both the beginning of the sediment sequence (the “small alpine glacial age ”) and at the end (14 –18th centuries: “ Petit Âge Glaciaire” of Le Roy Ladurie). This paper gives an example of the importance of small mammal assemblages for reconstructing the environment in the historical period.

Role of Biotic Interactions in a Small Mammal Assemblage in Semiarid Chile

The role of biotic interactions including vertebrate predation and interspecific competition was studied among members of a semiarid small mammal assemblage in a long—term field manipulation in north central Chile. Fences and netting selectively excluded mammalian and avian predators, and a putative rodent competitor, the degu (Octodon degus), from large, replicated grids. Small mammal populations were censused monthly over >4 yr, and effects of predator and degu exclusions on numbers and survival of small mammals were assessed with repeated measures ANOVAs. Marginally significant predator exclusion effects were detected on O. degus, and highly significant year X predation interactions in both wet and dry seasons; degu survival (= persistence) was also significantly longer on exclusion grids. Predation had significant effects on numbers of Phyllotis darwini, and a significant month X predator interaction existed in the wet season; survival was also significantly higher on predator exclusion grids. There was a marginally significant year X competitor interaction in the dry season for this species. Akodon olivaceus had significant month X predator and month X competitor interactions in the wet season only; survival was significantly longer on degu exclusion grids. An irruptive species, Oligoryzomys longicaudatus, showed no significant predator or competitor effects in 1991—1992; however, survival was significantly longer on predator + competitor exclusion (vs. control) grids. An uncommon species, Abrothrix longipilis, had marginally nonsignificant time X competitor interactions but no treatment differences in survival. Finally, a second uncommon rodent, Abrocoma bennetti (not analyzed numerically) had significantly longer survival on predator exclusion grids with degus. Degus and P. darwini were significantly overrepresented (relative to availability) in diets of culpeo foxes (Pseudalopex culpaeus) and three nocturnal owls, respectively, whereas A. bennetti was significantly overrepresented in predator diets generally. Vertebrate predation has pervasive effects on some species that, however, may be related to their intrinsic demographic characteristics. Competition may be mediated by subtle behavioral interactions among assemblage members. Long—term studies are necessary to elucidate the role of biotic and abiotic factors in semiarid environments

Small mammal taphonomy and the middle Pleistocene environments of Dolina, northern Spain

Abstract The study of past ecosystems can be complex due to taphonomic processes. These processes may alter the spatial and temporal location of fossils during site formation. The Dolina site (Burgos, Spain) provides an example of the contribution of taphonomy to the interpretation of the past environments. This site belongs to the Atapuerca-Ibeas cave complex with extensive evidence of human activity throughout the Pleistocene. The small mammal assemblages from Dolina show evidence that predation has biased species composition to some degree. Especially in two levels (TD4 and TD11), predator activity appears to have interfered with palaeoecological interpretations and exaggerated inferred climatic changes during the Middle Pleistocene. Once possible bias due to taphonomic agents is recognized, a more realistic and precise interpretation of the palaeoecology of this area is possible, as well as a history of the cave infilling as a background to human use of the cave.

Fluctuations of rodent populations in response to rainfall and fire in a central Australian hummock grassland dominated by Plectrachne schinzii

Fluctuations in the composition and abundance of a small-mammal assemblage were studied in a hummock grassland dominated by Plectrachne schinzii at Watarrka National Park from 1988 to 1993. During this period an experiment was conducted to examine the short-term effects of fire on the rodents. We caught three species of rodent (Pseudomys hermannsburgensis, Notomys alexis and Mus domesticus). All species reached their greatest density in spring 1989 during an exceptionally wet period that extended from mid- 1988 to 1990. P. hermannsburgensis was the most abundant species and showed a 10-fold fluctuation in numbers over the sample period; N. alexis was the next most abundant species and showed a 5-fold increase but the population took longer to decline. M. domesticus was recorded only during the period of high rainfall. The number of M. domesticus was significantly less on the burnt plots than on the unburnt plots. Neither P. hermannsburgensis nor N. alexis showed significant differences between burnt and unburnt plots. This study illustrates the impact of rainfall events on the composition and density of small-mammal populations in spinifex grasslands in central Australia. Our results lead to the prediction that rodent populations will achieve densities in the order of 10 individuals ha-' or more in regions that experience three consecutive 6-month periods each with rainfall at 150% of the long-term average. This sequence apparently needs to follow a dry period where rainfall is no more than 85% of the long-term annual average for two consecutive 12-month periods.

Age structure of two marsupial species in a Brazilian restinga

This note is part of a population study of a small mammal assemblage undertaken at Restinga de Barra de Marica, Rio de Janeiro State, Brazil. Here we characterize the age structure and reproductive patterns of two didelphid marsupials: Philander opossum (Linnaeus) and Didelphis aurita (Wied). A capturemark-recapture study was conducted on Restinga Forest, the most complex vegetational type in Barra de Marica (see Araujo & Henriques (1984) and Cerqueira et al. (1990) for a complete description of the area). We established a 4 ha grid with 100 trap stations from July 1988 to July 1990, and this was sampled quarterly, each session consisting of seven trapping-nights. We used Young live traps baited with peanut butter, banana, bacon, rolled oat and manioc slices. Individuals were marked by tail tattooing. Ages were estimated by the dental formulas developed by D'Andrea et al. (1994) and Motta (1988) for Philander opossum and Didelphis aurita, respectively. For the former species the experimental standard error of age estimation is 9.2 days and for D. aurita it is 7.9 days. Age classes were divided according to biological parameters. In P. opossum: class I before weaning, up to 80 days old; class II immature young, 81-175 days old, with no reproductive activity; class III mature young, 176323 days old; class IV adults, individuals of 324 days or more with complete dental development. In D. aurita: class I up to 100 days old; class II 101 to 170 days old; class III 171 to 336 days old; class IV individuals of 337 days or more. The minimum estimated age of a weaned individual of P. opossum was 88 days and for a lactating female 176 days; this agrees with data from captive

Patterns of faunal turnover and diversity in the Neogene Siwaliks of Northern Pakistan

The fluvial Neogene Siwalik formations of northern Pakistan contain a long and richly fossiliferous sequence of terrestrial vertebrate faunas in which patterns of faunal turnover and changes in diversity can be documented and analyzed for intervals having durations of 0.5 m.y. The complete sequence extends from circa 18.5 to 1 Ma, but the part between 18.5 and 5.5 Ma is best sampled, and most intervals within it are well represented. Thirteen orders of Siwalik mammals have been identified, with well-sampled intervals having 50 or more species. Most Siwalik mammals, however, are either rodents or artiodactyls. Bovids are the most common and most speciose of the larger mammals, while murid and “cricetid” rodents dominate the small mammal assemblages. Between 18.5 and 5.5 Ma species diversity varied considerably. Among artiodactyls and rodents the number of species first increased between 15 and 13 Ma and then fell. Data on stratigraphic ranges of rodents and artiodactyls show that faunal change in the Siwaliks was episodic, occurring during short intervals with high turnover followed by longer periods with considerably less change. Maxima of first appearances occurred at approximately 13.5 and 8.5 Ma, while maxima of last occurrences were at 12.5 and 8.0 Ma. Some of the observed faunal events can be correlated to climatic and environmental changes. The Middle Miocene diversification occurred during a period of global cooling, while the latest Miocene decline in diversity and increased turnover accompanied oxygen and carbon isotopic changes that correlate to globally increasing seasonality and aridity. Other correlations are ambiguous. The marked decrease in diversity and the major turnover events between 13 and 8 Ma do not correspond to known local or global events. The Neogene Siwaliks and Paleogene Bighorn-Crazy Mountains sequence in Wyoming and Montana share many similarities. They have equivalent levels of temporal resolution and similar levels of completeness of their fossil records. Siwalik ordinal abundance and diversity patterns differ markedly from those of the Paleogene, but generic, and probably species, diversity was approximately the same, although the Siwalik faunas may have been slightly less diverse. Over time, changes in diversity were of comparable magnitude, with monotonic trends persisting for more than 5 million years. The magnitude of faunal turnover was also similar, ranging from less than half to 3.5 times that expected. In both sequences faunal change appears to have been episodic, with strong pulses between intervals of low turnover. The Siwaliks, in contrast to the Paleogene sequence, may have had more distinct pulses and longer intervals between pulses. Neither sequence has peaks of first occurrences coinciding with peaks of last occurrences.

Does Predation Risk Constrain Maturation in Cyclic Vole Populations?

Animals may delay reproduction under temporarily high predation risk. If temporarily high predation risk has a suppressive effect on prey reproduction on a large scale, there should be a negative association between the predation risk level and the proportion of breeding individuals in prey populations. We used long-term trapping data on two vole species and data on predator densities to find out whether there was a relationship between the predation risk level and the proportion of breeding individuals in a small mammal assemblage with cyclic fluctuations. In spring and, in the field vole (Microtus agrestis), in autumn, current predation risk from small mustelids could explain observed variation in the proportion of mature voles better than their own density, current or previous. The association between the proportion of breeding field voles and the current predation risk from small mustelids was negative in autumn, but not in summer when the current reproductive value is high relative to the residual value. For bank voles (Clethrionomys glareolus), which have a lower predation risk than field voles, there was a negative relationship between predation risk from mustelids and proportion of breeding individuals in early summer, but not in late summer or autumn. In late summer, the proportion of breeding year-born bank voles was negatively related to the vole density in spring. Our results indicate that the indirect effects of predators on vole populations may be of great importance on a large scale, and that these effects should be taken into consideration in studies on factors regulating vole populations

Body Size, Insectivory and Abundance in Assemblages of Small Mammals

Data from twenty five assemblages of small mammals are studied. There is no tendency for the species in the assemblages to show any predictable correlation between body mass and abundance. The same is true within the insectivore guild of each assemblage. However, assemblages nearer the equator have insectivore guilds with predictably more positive mass-abundance patterns than those at high latitudes. A survey of twenty years of the published literature reveals that in general, niche overlap in co-occurring small insectivorous mammals is lowest at high latitudes. If high niche overlap is associated with strong competition, this result may explain the patterns in abundance; positive mass-abundance correlations may be found where interspecific competition is most intense.

A five-year population study of an assemblage of small mammals in Southeastern Brazil

Une etude par capture-marquage et recapture de petits mammiferes a ete realisee pendant cinq and dans une vegetation arbustive cotiere bresilienne (restinga). Cinq especes de petits mammiferes, rongeurs et marsupiaux, ont ete capturees et caracterisees par rapport a leur dynamique de population. Akodon cursor et Philander opossum ont presente les plus grandes densites de population, qui sont correlees avec le recrutement. Ils ont aussi presente un recrutement irregulier, des fluctuations de populations frequentes avec des coefficients de variation eleves, des taux de survie bas la plupart du temps et des durees de residence courtes

Mamíferos del yacimiento del Pleistoceno Superior de Cueva Millán (Burgos, España)

The faunal analysis of the mammalian association retrieved in the mousterian cave-entrance site of Cueva Millan (Burgos) uncovers a rather varied assemblage of different cultural and bioecological connotations. On one side, the big-mammal fauna, by far the biggest fraction of the sample, seems to be the likely product of human hunting. As such, most bones are literally smashed and extremely eroded. This, of course, renders identification and interpretation more dificulto The small-mammal assemblage, looks more like the product of a primary accumulation and reflects a late Pleistocene fauna. The whole fauna indicates a rather temperate event with forest and open land criss-corssed by water courses. Particulary interesting is the presence of the beaver, not only a good paleoecological indicator but also arare finding on the Iberian Peninsula.

An Assembly Rule for Functional Groups Applied to North American Soricid Communities

In this paper we demonstrate that an assembly rule for functional groups of Australian small mammals also accounts for the structure of soricid assemblages in mesic forests of northeastern North America. In this insectivore guild, functional groups are defined by body size. A null hypothesis of random assembly was rejected, requiring the acceptance of a deterministic alternative hypothesis, such as this assembly rule. The rule was derived empirically for a small mammal community in Australian heathland and was demonstrated to account for the structure of assemblages of small mammals from many heathland and temperate eucalypt forest sites over a range of 16° of latitude in Australia. The Australian assemblages included representatives of three trophic guilds (insectivore, herbivore, granivore-omnivore), representing three separate evolutionary lines. By using functional groups, rather than individual species, the assembly rule has broader application than previous species-specific rules. The rule reflects the use of available resources and demonstrates a high probability that each successive species in an assemblage will be drawn from a different functional group until each group is represented. Where sufficient resources are available the rule repeats. We conclude that assemblages of small mammals examined in Australia and North America follow the same assembly rule as demonstrated by their highly significant departures from randomly-generated assemblages. Additional evidence from Australia, North America, and England indicates that interspecific competition is the most likely mechanism responsible for the observed non-random interactive patterns. The process producing these patterns appears to act through body size in soricid communities, with larger body size conveying a competitive advantage by improving access to higher-quality foraging micro-habitats.

Ecological correlates of small mammal community structure in the semi‐arid Karoo, South Africa

The understanding of the determinants of small mammal community structure in arid and semiarid ecosystems is of importance, both in the light of the role that small mammals play, and the impact of livestock grazing on the flora of these systems. In a study aimed at identifying these determinants, small mammal assemblages and environmental features were quantified at six localities (a gradient of floristic structure, with constant annual rainfall) across the southern Karoo, South Africa. Stepwise variable regression indicated that small mammal diversity was correlated with plant and rock cover, as well as plant cover and horizontal foliage diversity at intermediate heights (40–60 cm). Initially, small mammal diversity increased with increasing plant cover, but decreased at cover levels greater than 30%. This relationship is similar to that found in other desert systems, although the peak in diversity found here is at higher levels of plant cover than found previously. I suggest that this higher peak may be owing to the lack of reliance on granivory by these animals, which are relatively omnivorous. This model may therefore explain the conflicting reports on the impact of livestock grazing on desert small mammals, with small mammal diversity decreasing with grazing below the peak, and increasing with grazing above the peak.

Patterns of faunal turnover and diversity in the Siwalik Neogene record in relation to regional and global events

The fluvial Neogene Siwalik formations of northern Pakistan contain one of the longest and richest sequences of terrestrial vertebrate faunas known. The complete sequence extends from ca. 18 Ma to 1 Ma, with the interval between 18 and 7 Ma being best sampled. Throughout this best known interval vertebrate remains are frequently abundant in channel fills and less common in large channel sands, levees, and paleosols. Although the abundance and quality of fossil preservation varies, all stratigraphic levels have some fossils and the record of most subintervals is good to excellent. As a consequence the patterns of faunal turnover and changes in diversity can be documented and analyzed for 0.5 my long subintervals.Thirteen orders of Siwalik mammals have been identified, with well sampled subintervals typically having 50 or more species. Despite the ordinal diversity, however, most Siwalik mammal species belong to just three orders: rodents, artiodactyls, and perissodactyls. Among the larger mammals, the bovids and equids are the most common and have the most species, while the murid and cricetid rodents dominate the small mammal assemblages. These Siwalik abundance and diversity patterns differ markedly from those of the Paleogene and are a result of Neogene radiations in these four families and extinction of Paleogene groups.Between 18 and 7 Ma species diversity varies considerably. Among artiodactyls and rodents the number of species first increases between 15 and 13 Ma and then falls after 12 Ma. Significant changes in relative abundance are also known, including an increase in the abundance of bovids between 16.5 and 15 Ma and a very abrupt increase of murids at 12 Ma.Data on stratigraphic ranges of rodents and artiodactyls show that faunal change in the Siwaliks was episodic, occurring as short intervals with high turnover, followed by longer periods with considerably less change. Maxima of first appearances occur at approximately 13.5 and 8.5 Ma, while maxima of last appearances come at 12.0, 9.5, and 8.0 Ma. It is thus apparent that in the Siwaliks increased extinction did not accompany or closely follow maxima of first appearances.Correlations of these faunal events to global climatic trends are ambiguous. However, it is apparent that the middle Miocene diversification of Siwalik faunas occurred during a period of global cooling, while the late Miocene decline in diversity preceded a second episode of cooling and increasing aridity.

The Atapuerca sites and the ibeas hominids

The Atapuerca railway Trench and Ibeas sites near Burgos, Spain, are cave fillings that include a series of deposits ranging from below the Matuyama/Bruhnes reversal up to the end of Middle Pleistocene. The lowest fossil-bearing bed in the Trench contains an assemblage of large and small Mammals includingMimomys savini, Pitymys gregaloides, Pliomys episcopalis, Crocuta crocuta, Dama sp. and Megacerini; the uppermost assemblage includesCanis lupus, Lynx spelaea, Panthera (Leo) fossilis, Felis sylvestris, Equus caballus steinbeimensis, E.c. germanicus, Pitymys subterraneus, Microtus arvalis agrestis, Pliomys lenki, and alsoPanthera toscana, Dicerorbinus hemitoechus, Bison schoetensacki, which are equally present in the lowest level. The biostratigraphic correlation and dates of the sites are briefly discussed, as are the paleoclimatic interpretation of the Trench sequences. Stone artifacts are found in several layers; the earliest occurrences correspond to the upper beds containingMimomys savini. A set of preserved human occupation floors has been excavated in the top fossil-bearing beds. The stone-tool assemblages of the upper levels are of upper-medial Acheulean to Charentian tradition. The rich bone breccia SH, in the Cueva Mayor-Cueva del Silo, Ibeas de Juarros, is a derived deposit, due to a mud flow that dispersed and carried the skeletons of many carnivores and humans. The taxa represented are:Ursus deningeri (largely dominant),Panthera (Leo) fossilis, Vulpes vulpes, Homo sapiens var. Several traits of both mandibular and cranial remains are summarized. Preliminary attempts at dating suggest that the Ibeas fossil man is older than the Last Interglacial, or oxygen-isotope stage 5.

Distribution and Abundance of Small Mammals along an Elevational Transect in Temperate Rainforests of Chile

Assemblages of small mammals in temperate Andean rainforests of southern South America are diverse but poorly known. Herein, we describe an elevational transect through temperate rainforests in Valle de La Picada, Chile, during February and March, 1984. The transect consisted of standardized removal trapping (165 traps set in lines at ca. 7-m intervals and operated for 6 consecutive days) at each of seven elevations: 425 m, 505 m, 620 m, 715 m, 820 m, 920 m, and 1,135 m. A total of 514 small mammals representing two species of marsupials and seven species of sigmodontine rodents was collected. Trap success averaged 7.4% across traps, days, and elevations. Number of captures by individual traps during the sample period was indistinguishable from a Poisson distribution, both throughout the transect and at individual levels, suggesting independence of captures among traps. Captures were biased significantly by trap type, with museum specials securing significantly more Dromiciops australis and all forms of Akodon than Sherman live traps; the same biases were suggested by captures of Irenomys tarsalis and Auliscomys micropus . As expected, total number of individuals taken at each elevation declined throughout the sampling period while the cumulative number of species rose. Number of species, number of individuals, and species diversity varied inversely with elevation. Number of species and species diversity taken each day reached a maximum after 2–4 days of sampling. Captures of Rhyncholestes raphanurus, Akodon sanborni , putative hybrids of A. sanborni and Akodon longipilis , and I. tarsalis were correlated inversely with elevation. Conversely, D. australis, A. longipilis , and A. micropus were taken more commonly at higher elevations, whereas captures of Oryzomys longicaudatus, Akodon olivaceus , and Geoxus valdivianus appeared unrelated to elevation. Systematic sampling procedures are viewed as essential for describing fundamental patterns of abundance and distribution that can be compared with other studies.

The contribution of raptorial birds to patterning in small mammal assemblages

Seven taxa of raptorial birds were experimentally fed a controlled sample of 50 house mice (Mus musculus). Bones recovered from the pellets were examined for interspecies variability in preservation to assess the potential contribution of specific raptors to patterning in fossil assemblages. Quantitative analyses demonstrate that patterns in bone fragmentation may assist in the identification of particular raptor species as depositional agents in small mammal assemblages.

Species assembly and the evolution of community structure

In this paper I consider the evolutionary and ecological implications of an assembly rule which was derived empirically from studies on a heathland small-mammal community in south-eastern Australia. This rule has been tested successfully against 52 heathland small-mammal assemblages. Here it is shown to hold also for 80 forest assemblages of small mammals spanning a latitudinal range from 27°S to 43°S in south-eastern Australia. The observed forest communities are predicted by the rule and they deviate significantly from random assemblages. I suggest that the unique evolutionary history of the Australian fauna has made these patterns more apparent. The rule is simply stated as: ‘There is a much higher probability that each species entering a community will be drawn from a different functional group (genus or other taxonomically related group of species with similar diets) until each group is represented, before the cycle repeats’. A theoretical basis for the rule is proposed which extends the niche compression hypothesis to cover evolutionary time. Evolutionary constraints on adaptations for diet selection are greater than those operating on habitat selection. Successful tests in North America for the granivorous desert rodent guild and the mixed-forest insectivore guild support a wider application of this rule than the Australian communities from which it was derived. A speculative model is proposed in which the mechanisms involved in the operation of this rule shape the evolution of community structure.