Describing the Distribution and Abundance of Populations


• A population is a group of individuals of the same species found in the same place at the same time.

• Populations are characterized in terms of their distribution and abundance.

• The distribution of a species can be mapped using historical data, field observations and remote sensing.

• Individuals within a population will not be evenly distributed throughout their range.

• Abundance can be measured as frequency, density, cover or biomass.

• Abundance and distribution do not necessarily reflect a species' ecological impact.


A population is a group of individuals of the same species found in the same place at the same time. Like many ecological terms, this definition is flexible, because it can be used to describe populations at many scales. For example, a population may be the number of individuals contained within a small area (e.g. a field) or it may refer to the local or regional distribution of the species. The first step in understanding any species is to document its distribution and abundance. This gives the researcher an idea of the scope of the potential problem (i.e. weediness). Note that we say potential problem: while distribution and abundance are useful information, more data must be obtained before a decision is made on a species' weediness.

In this chapter, we discuss how to describe a population's distribution and abundance. Distribution is a measure of the geographical range of a species, and is used to answer questions such as: 'Where does the species occur?', 'Where is it likely to occur?' and 'Where is it able to occur?'. Abundance is a measure of the number or frequency of individuals. It is used to answer questions such as: 'Is the number or frequency of individuals increasing or decreasing?'.

© 2003 CAB International. Weed Ecology in Natural and Agricultural Systems (B.D. Booth, S.D. Murphy and C.J. Swanton)

Fig. 2.1. Distribution of Fremont's leather flower (Clematis fremontii var riehlii) in the Ozarks of Missouri. Shown are distributions at the scale of range, region, cluster, glade and aggregate (Erickson, 1945; with permission of the Missouri Botanical Garden).

Population Distribution

A population's distribution (or range) describes where it occurs. In practical terms, it is a description of where the species has been recorded (Gaston, 1991). Mapping a species' distribution can be done on a number of scales depending on how the information is to be used. For example, Erickson (1945) mapped the distribution of the flowering shrub Fremont's leather flower (Clematis fremontii var. riehlii) at several scales. This species was restricted to approximately 1100 km2 in the Missouri Ozarks (Fig. 2.1). Individuals, however, were not distributed throughout the species' range because they live only in sites where the abiotic and biotic conditions are suitable for them. For example, the range of Fremont's leather flower was subdivided into four watershed regions. Within these regions, there were groups of glades (rocky outcrops on south and west facing slopes), and clusters of Fremont's leather flower tended to be located at bases of these glades. Finally, within these clusters, there are loose aggregates of up to 100 individual plants.

Distribution maps have different uses depending on their scale. A researcher wanting to study the pollination of Fremont's leather flower would require a fine-scale distribution map showing the locations of individuals or colonies. Conversely, such a map would not be useful to a researcher interested in the broad-scale environmental controls of the species. They would require a large-scale map of the entire species' distribution.

Distribution change over time

A population's distribution will change over time either naturally or through human influence. Following the retreat of the last North American ice sheet (approximately 10,000 years ago) trees migrated northward, each species at a different rate and following a different route (Davis, 1981). At a smaller scale, a species distribution will change during the process of succession over decades (Chapter 13). Human disturbances, such as changing land-use patterns, will alter the environment such that different species are favoured and therefore population distributions will change. Also, human actions introduce exotic species and this increases their distribution. Thus, a species' distribution is not static; its boundaries are dynamic.

Asking what controls a population's distribution, and whether and why a species' distribution changes over time are fundamental questions of ecology. To better understand a weed species we might want to ask the following questions about its distribution:

• Is the weed at its current limit of distribution?

• Will the weed continue to expand into new locations?

• Is the weed found on specific soil types or land forms?

• Are there likely dispersal routes for this weed?

Distribution boundaries are limited by biot-ic (living, e.g. interactions with other species) and abiotic (non-living, e.g. temperature) factors. The same factor will not necessarily limit all boundaries of the range equally. For example, abiotic factors are more likely to limit distribution at higher latitudes, while biotic factors are more likely to limit distribution at lower latitudes (Brown et al., 1996). Boundaries are rarely sharp, unless the population abuts against a geographic (e.g. river) or human-made feature (e.g. highway). Typically, individuals within the population become less and less frequent toward the limits of their range.

By following changes in a species' distribution over time, it is possible to tell whether a population is expanding or contracting. In the case of weeds, this may warn us where problems are likely to occur, or alternatively where control measures have been effective. We can also gain information on species' characteristics such as dispersal mechanisms or habitat preferences. Forcella and Harvey (1988) analysed how the distribution patterns of 85 agricultural weeds introduced into the northwestern USA changed between 1881 and 1980. They found that species' migration patterns were dependent on the species' point of entry and on the types of agriculture (e.g. grain, cattle) with which the weed was associated. Furthermore, migration patterns tended to follow land transportation routes. Similarly, Thompson et al. (1987) mapped the expansion of purple loosestrife (Lythrum salicaria) from 1880 to 1985 along canals, waterways and later along roads (Fig. 2.2). These examples give insight into how future introductions of new plant species might spread depending on their point of origin.

Fig. 2.2. Distribution of purple loosestrife (Lythrum salicaria) in North America in 1880, 1900, 1940 and 1985 (from Thompson et al., 1987).

Estimating and mapping distribution

The traditional method for collecting data on the actual distribution of a species is to consult public records such as government documents, herbaria, field notes or academic journals. This type of data allows for the construction of historical distributions as was done by Thompson et al. (1987) and Forcella and Harvey (1988) (Fig. 2.2). These give a clear view of a species' regional distribution and change with time. Such records, however, are dependent on the accuracy and precision of the data collected, and this may be difficult to judge. Also, all sites and species will not be sampled equally and therefore, areas with less-intense sampling will be under-represented on maps (Schwartz, 1997). There will also be a sampling bias towards large or more obvious species. For example, purple loosestrife has large purple inflorescences and is more likely to be observed and recorded than a co-occurring weed, Japanese knotweed (Polygonum cuspidatum).

Field sampling and herbaria records give us information about the current or recent past distribution of species because

Fig. 2.3. Pollen diagrams of sediment taken from Crawford Lake, Canada. Shown is the per cent of pollen for each species. Note the increase in maize (Zea mays), purslane (Portulaca oleraceae) and grass (Gramineae) pollen during the Iroquoian period from 1360 to 1660, and the increase in ragweed (Ambrosia), dock (Rumex), and plantain (Plantago) pollen following land clearing by European settlers in 1820. (Adapted from McAndrews and Boyko-Diakonow 1989; with permission of the authors and the Minister of Public Work and Government Services Canada, 2002 and Courtesy of Natural Resources Canada, Geological Survey of Canada.)

Fig. 2.3. Pollen diagrams of sediment taken from Crawford Lake, Canada. Shown is the per cent of pollen for each species. Note the increase in maize (Zea mays), purslane (Portulaca oleraceae) and grass (Gramineae) pollen during the Iroquoian period from 1360 to 1660, and the increase in ragweed (Ambrosia), dock (Rumex), and plantain (Plantago) pollen following land clearing by European settlers in 1820. (Adapted from McAndrews and Boyko-Diakonow 1989; with permission of the authors and the Minister of Public Work and Government Services Canada, 2002 and Courtesy of Natural Resources Canada, Geological Survey of Canada.)

these records may only go back a few hundred years. Thus, the initial invasions of some species cannot be tracked in this way. One possible method for tracing early introductions of species and their distribution changes is to use palaeoecological records. Microfossils, such as pollen grains and other plant parts, are preserved occasionally in peat or lakebed sediments. These can be retrieved and then identified (often to species level) to obtain a record of past vegetation. These records can be dated because the sediment is laid down in yearly layers, which can be radiocarbon dated. Changes in species composition over time can be traced by identifying pollen grains in successive layers of the sediment and constructing diagrams that show changes over time (Fig. 2.3). Using this method, extended time series can be constructed. Interestingly, this method has proven that some species previously thought to have been introduced to North America are actually indigenous. The pollen diagrams of Crawford Lake, Ontario, Canada, for example, show that purslane (Portulaca oleracea) was not an invasive weed from Europe as previously thought; in fact, it existed in the area from at least c. ad 1350 when the Iroquois began cultivating maize (Jackson, 1997).

Collecting field data can be a long, expensive process and therefore new methods to map the distribution of weeds are being developed. Such methods use remote sensing with either aircraft or satellite imagery. Photos or videos are taken to record the spectral reflectance of plants and ground terrain. To detect and map a species using this method, it must be possible to distinguish a species' reflectance pattern from the background of surrounding vegetation, ground, roads and other features. To date there has been some success mapping weeds of rangeland and pasture. Lass et al. (1996) were able to map the spatial distribution of common St John's wort (Hypericum perforatum) and yellow star thistle (Centaurea solstitalis) in rangeland using multispectral digital images taken from aircraft. Everitt et al. (1992) were also able to obtain area estimates of falsebroom (Ericameria austrotex-ana), spiny aster (Aster spinosus) and

Chinese tamarisk (Tamarix chinensis) in rangelands and wild land of the southwestern USA. Remote sensing also has the benefit of covering large patches of land, so it can be used to follow the invasion of a species and monitor whether management practices are working. However, before it can be employed, we must have biological information about the species in order to be able to remote sense it properly and interpret the images. For example, a species' spectral reflectance pattern may change over its life cycle: we must know this in order to remote sense at the appropriate time. With advances in the technology, remote sensing may become applicable to more situations in the future.

Potential distribution

A species' ability to grow and reproduce can extend well past the boundaries of its natural or native distribution, because species distributions are not always limited by abiotic conditions. A species' distribution may be limited either by its inability to disperse to other sites or its inability to compete with other species. However, many species thrive after being artificially transplanted into a new habitat for agriculture or forestry. Many weeds were introduced purposefully and, once introduced, were able to rapidly expand their distribution. For example, kudzu (Pueraria montana var. lobata), which was introduced into the USA in 1876 as an ornamental vine and later used as a forage crop and for erosion control, is now considered to be a serious threat in the southeastern USA.

The area in which a species can (in theory) survive is its potential distribution (i.e. physiological distribution or climatic range). The potential distribution is based on the abiotic environment only and does not take into consideration how the species might survive in 'real situations' where, for example, it competes with other species. The potential distribution of a species may be far greater than its native distribution. For example, the natural distribution of Monterey pine (Pinus radiata) is limited to approximately 6500 ha in the coastal fogbelt of California and there have been attempts to place it on the 'threatened species' list in California. While the trees in the native range may be threatened, Monterey pine is also found all over the world and is a weed in some places. How can it be threatened in California, yet be found almost everywhere in the world and even be considered a weed? The answer is that in its native range, Monterey pine has been threatened by development, logging, changing weather patterns and diseases. However, humans have made the tree the most widely planted plantation tree in the world such that it covers over 4,000,000 ha (Clapp, 1995; Lavery and Mead, 1998). It is planted extensively in countries with habitats similar to California, e.g. New Zealand, Australia and Chile, where it is a fast-growing tree that can be harvested in 25-year rotations. Since it does not face the diseases that exist in its native California and is drought tolerant, Monterey pine is a weed in places with a Mediterranean climate and has invaded grasslands and native eucalypt forests (Richardson and Bond, 1991).

By comparing the native and potential distributions of a species, it may be possible to predict where it is likely to spread. The potential distribution of a species can be estimated in several ways. Patterson et al. (1996, 1997) estimated the potential distribution of a number of agricultural weeds using laboratory-based studies to determine the temperature and light conditions required by each species. From these data, they can create a mathematical model to predict where the right combinations of

Fig. 2.4. Observed and predicted distributions of bridal veil (Asparagus declinat) in Australia. Solid dots indicate the predicted distribution while crosses indicate sites unsuitable to this species. Regions of known infestations are around Adelaide, Perth and Bunbery. The inset shows observed and predicted distributions of bridal veil in South Africa. (Pheloung and Scott 1996; with permission of R.G. and F.J. Richardson and P. Pheloung.)

conditions exist for the species to survive and reproduce. For example, after growing tropical soda apple (Solanum viarum) in growth chambers under a variety of day and night temperatures and photoperiods, Patterson et al. (1997) compared their results with climatic conditions in 13 southern states of the USA. They concluded that temperature and photoperiod were not likely to limit the expansion of this species and suggested that measures should be taken immediately to control the expansion of soda apple beyond its current distribution in Florida. This type of approach uses only abiotic factors that can be experimentally controlled, and it does not take into account seasonal temperature extremes or precipitation patterns (Patterson et al., 1997).

An alternative way to predict a species' potential distribution is to compare the environmental conditions of the species' native habitat with those of a potential habitat. CLIMEX is one computer model suitable for this (Sutherst and Maywald, 1985). climex considers measures of growth such as temperature, moisture and daylength, and then adjusts this based on stress indicators such as excessive dry, wet, cold and heat, to give an ecoclimate index. Pheloung and Scott (1996) used climex to compare the distribution of bridal creeper (Asparagus asparagoides) and bridal veil (Asparagus declinat) (Fig. 2.4) in their native South Africa to potential habitats in Australia. They concluded that both species had the potential to continue spreading and that measures should be taken to control or eradicate them. Similarly, Holt and Boose (2000) were able to map the potential distribution of velvetleaf (Abutilon theophrasti) in California. They concluded that the distribution of velvetleaf was not likely to increase, because its range was limited by water stress.

Thus, potential distribution gives us an idea of the climatic regions where a species is able to survive the physical environment. This does not mean that the species will live there, because a species' distribution is controlled by non-climatic factors such as lack of dispersal or by interactions with other species.

Population Abundance

While distribution describes the geographical extent of a population, abundance describes a population's success in terms of numbers. Individuals will not be equally dispersed throughout their entire range; there will be areas of high and low density. Abundance can be measured in a number of ways. The type of measure selected will depend on the species in question, the habitat type (e.g. forest, field), the goal of the study and the economic resources.

Measures of abundance

Frequency and density

Frequency is the proportion of sampling units (e.g. quadrats) that contains the target species. It is easy to measure because only a species' presence or absence is noted for each quadrat. Frequency is a fast, nondestructive method and is less prone to incorrect estimates by the researcher. Density measures the number of individuals in a given area (e.g. square metre or hectare). It too is non-destructive, and while it is more complicated to measure, it provides more information than frequency.

While frequency and density are probably the most commonly used measures of abundance, there are some difficulties associated with using them. Density assumes that you are able to separate individuals. This is not a problem in higher animals because they are distinct individuals. In plants, however, many species are capable of reproducing vegetatively and therefore, it is often difficult to distinguish one genetic individual from another (see Chapter 5). Frequency does not have this problem.

A further difficulty in identifying individuals is that individuals of the same species may appear morphologically different depending on their age, stage of growth or environment. Many plants differ in appearance from one life stage to another (i.e. they are phenologically plastic). For example, a tree seedling will look very different from a mature adult. In addition, plants may be morphologically plastic: their appearance may differ depending on their environment. Leaves of aquatic plants often appear different depending on whether they are above or below the water, or leaves of terrestrial plants may differ depending on whether the leaf is produced in the sun or the shade. The variable appearance of a species may make it difficult to count. Therefore, measures of frequency and density might exclude individuals that are morphologically different and result in an underestimation of their abundance.

A final problem in using frequency and density as a measure of a population is that they do not distinguish between the sizes of individuals. Therefore, larger individuals are scored the same as smaller ones, even though they will have different influences on the community. Larger plants will probably have more effect on the physical environment (e.g. through shading) and they tend to produce more seed than smaller ones, thereby having a greater influence on subsequent generations. Therefore, frequency and density are better used when vegetation is of uniform size. Other measures of abundance such as cover and biomass can be used when an indication of size is desired.

Cover and biomass

Cover and biomass are sometimes used in place of frequency and density when an indication of individual size is important. Cover is the proportion of ground covered by a given species when viewed from above. Cover is useful when a non-destructive sampling method is required; however, it is sometimes difficult to quantify. It may be difficult to get an accurate value of cover because it is typically done as a visual estimate, so percentage cover estimation is often broadly categorized (e.g. 0%, 1-5%, 5-10%, 10-25%, 25-50%, 50-75% and 75-100%). Measuring cover is subjective and therefore not precise; however, this method is widely used and considered valuable because it provides useful information with relatively low effort by the researcher.

Biomass is the weight of vegetation per area. Biomass is useful when an accurate indication of plant size is needed. It is sampled usually by collecting the shoots and roots from a given area. When collecting, the plant can also be divided into roots, stems, leaves and reproductive structures to observe how plants allocate biomass to different structures. Collecting actual plant samples to determine biomass is not practical for larger organisms such as trees. Therefore, some mathematical equations have been developed to calculate biomass based on size. For example, we may harvest several plants of varying height to establish if there is a correlation between height and biomass. If there is, then height can be measured instead of harvesting the plant. For trees, stem diameter at breast height (dbh) is often taken as a measure of tree size.

Spatial Distribution of Individuals Within a Population

Within a population, individuals are not distributed evenly throughout their range. Individuals can be arranged at random, in clumps or in a regular pattern. These distribution patterns are the result of the abiotic environment, seed dispersal patterns, the species' biology, interactions among species or management practices. When we measure population abundance, it is an estimate of the average value over the entire area. It is important to consider spatial arrangement within a population, especially when determining effects of weeds on crops or on natural communities. Early studies on the effect of weeds on crop yield loss assumed that weeds were randomly distributed; however, it is now clear that this may not be so (Hughes, 1990; Cardina et al, 1997). Crop yield loss due to weeds will be overestimated if weed distribution is not taken into account (Auld and Tisdel, 1988). If weeds are clumped in a few areas of the field, then crop loss estimates for the entire field will be lower than if they were randomly distributed. Another field with the same overall density, but a more random distribution of weeds, will probably have more yield loss.

Problems of Predicting Weediness Based on Distribution and Abundance

Purple loosestrife has been characterized frequently as an invasive species and certainly the distribution and abundance of purple loosestrife has increased dramatically since it was first introduced into the New England states in the mid-1800s (Fig. 2.2) (Thompson et al., 1987). What has not been documented, however, is the effect that this species has on the native vegetation. Just because a species is increasing in distribution and 'appears' to be a dominant species does not mean that it is having a pronounced effect on plant communities. In reality, there is surprisingly little evidence to indicate that purple loosestrife is, in fact, an aggressive weed that has negative effects on other plant populations (Anderson, 1995; Hager and McCoy, 1998). The conspicuous appearance of this plant acts against it, because subjective observation will overestimate its abundance and underestimate the abundance of less conspicuous species. Other species that may disrupt the shoreline component of ecosystems may be more pernicious and problematic but less attention has been given to these species, in favour of the more obvious purple loosestrife (e.g. Japanese knotweed; see Chapter 1).


The first questions to ask when considering a potential weed problem are: 'Where is it?' 'How abundant is it?' and 'How is it spatially distributed?'. The answers to these questions allow us to characterize the distribution and abundance of the weed. These are important first steps towards understanding the ecology of species, but they are not necessarily good indicators of the species' influence on other populations or on the community as a whole. While we gain some information about whether a species is increasing or decreasing from abundance and distribution data, we need to go further to understand fully the dynamics of a weed and whether it will affect other populations. Although the concepts in this chapter are simple, they are important. If incorrectly applied they could lead to the conclusion that a weed is a problem when in fact, it is not. In the next chapter we begin to 'go further' and look at population structure and dynamics. Individuals within populations are not all identical: they differ in age, size, sex and developmental stage. We look at the repercussions of population structure.


1. Using the species you selected in Chapter 1, research its distribution. Map the distribution of your species using the appropriate scale (e.g. field, regional, continental) of map. What resources other than maps are available? Consider the following questions:

• At what scale do we know the species' distribution?

• Can we follow changes in its distribution status over time?

• What types of data were used to construct this map?

2. For each of the following environments, which method of estimating abundance (density, cover, biomass or frequency) would be best and why? (i) A natural forest, (ii) planted woodlot, (iii) a maize field, and (iv) a pasture.

3. Why is it important to consider spatial distribution of a weed within: (i) a field of maize, (ii) a natural forest?

4. By understanding abundance and distribution, how would you determine the ecological impact of a weed?

General References

Brown, J.H., Stevens, G.C. and Kaufman, D.W. (1996) The geographic range: size, shape, boundaries, and internal structure. Annual Review of Ecology and Systematics 27, 597-623. Gaston, K.J. (1991) How large is a species range? Oikos 61, 434-437.

Weber, E. (2001) Current and potential ranges of three exotic goldenrods (Solidago) in Europe. Conservation Biology 15, 122-128.

Literature Cited

Anderson, M.G. (1995) Interactions between Lythrum salicaria and native organisms: a critical review. Environmental Management 19, 225-231.

Auld, B.A. and Tisdel, C.A. (1988) Influence of spatial distribution of weeds on crop yield loss. Plant Protection Quarterly 3, 81.

Brown, J.H., Stevens, G.C. and Kaufman, D.W. (1996) The geographic range: size, shape, boundaries, and internal structure. Annual Review of Ecology and Systematics 27, 597-623.

Cardina, J., Johnson, G.A. and Sparrow, D.H. (1997) The nature and consequences of weed spatial distribution. Weed Science 45, 364-373.

Clapp, R.A. (1995) The unnatural history of the Monterey pine. Geographical Review 85, 1-19.

Davis, M.B. (1981) Quarternary history and stability of forest communities. In: West, D.C., Shugart, H.H. and Botkin, D.B. (eds) Forest Succession: Concepts and Application. Springer-Verlag, New York, pp. 131-153.

Erickson, R.O. (1945) The Clematis freemonlii var. riehlii populations in the Ozarks. Annals of the Missourri Botanical Garden 32, 413-459.

Everitt, J.H., Escobar, D.E., Alaniz, D.E., Villarreal, R. and Davis, M.D. (1992) Distinguishing brush and weeds on rangelands using video remote sensing. Weed Technology 6, 913-921.

Forcella, F. and Harvey, S.J. (1988) Patterns of weed migration in Northwestern U.S.A. Weed Science 36, 194-201

Gaston, K.J. (1991) How large is a species range? Oikos 61, 434-437.

Hager, H.A. and McCoy, K.D. (1998) The implications of accepting untested hypotheses: a review of the effects of purple loosestrife (Lythrum salicaria) in North America. Biodiversity and Conservation 7, 1069-1079.

Holt, J.S. and Boose, A.B. (2000) Potential for spread of velvetleaf (Abutilon theophrasti) in California. Weed Science 48, 43-52.

Hughes, G. (1990) The problem of weed patchiness. Weed Research 30, 223-224.

Jackson, S.T. (1997) Documenting natural and human caused plant invasions with paleoecological methods. In: Luken, J.O. and Thieret, J.W. (eds) Assessment and Management of Plant Invasions. Springer-Verlag, New York, pp. 37-55.

Lass, L.W., Carson, H.W. and Callihan, R.H. (1996) Detection of yellow thistle (Centaurea solstitalis) and common St John's wort (Hypericum perforatum) with multispectral digital imagery. Weed Science 10, 466-474.

Lavery, P.B. and Mead, D.J. (1998) Pinus radiata: a narrow endemic from North America takes on the world. In: Richardson, D.M. (ed.) Ecology and Biogeography of Pinus. Cambridge University Press, Cambridge, pp. 432-449.

McAndrews, J.H. and Boyko-Diakonow, M. (1989) Pollen analysis of varved sediment at Crawford Lake, Ontario: evidence of Indian and European farming. In: Quaternary Geology of Canada and Greenland. Geology of Canada, No. 1. Geological Survey of Canada, Ottawa, pp. 528-530

Patterson, D.T. (1996) Temperature and photoperiod effects on onionweed (Asphodelus fistulosus) and its potential range in the United States. Weed Technology 10, 684-689.

Patterson, D.T., McGowan, M., Mullahey, J.J. and Westbrooks, R.G. (1997) Effects of temperature and photoperiod on tropical soda apple (Solanum viarum Dunal) and its potential range in the U.S. Weed Science 45, 404-418.

Pheloung, P.C. and Scott, J.K. (1996) Climate-based prediction of Asparagus asparagoides and A. dec-linatus distribution in Western Australia. Plant Protection Quarterly 11, 51-53.

Richardson, D.M. and Bond, W.J. (1991) Determinants of plant distribution: evidence from pine invasions. American Naturalist 137, 639-668.

Schwartz, M.W. (1997) Defining indigenous species: an introduction. In: Luken, J.O. and Thieret, J.W.

(eds) Assessment and Management of Plant Invasions. Springer-Verlag, New York, pp. 7-17. Sutherst, R.W. and Maywald, G.F. (1985) A computerized system of matching climates in ecology.

Agriculture, Ecosystems and Environment 13, 281-299. Thompson, D.Q., Stuckey, R.L. and Thompson, E.B. (1987) Spread, impact, and control of purple loosestrife (Lythrum salicaria) in North American wetlands. Fish and Wildlife Research Report No. 2, US Department of Interior, Fish and Wildlife Service, Washington, DC.

0 0


  • kathryn
    What can be characterize for population abundance?
    4 months ago

Post a comment