Definitions of species

The question of how best to define "species" is one that has occupied biologists for centuries, and the debate itself has become known as the species problem. Darwin wrote in chapter II of On the Origin of Species:

No one definition has satisfied all naturalists; yet every naturalist knows vaguely what he means when he speaks of a species. Generally the term includes the unknown element of a distinct act of creation.^[7]

But later, in The Descent of Man, when addressing "The question whether mankind consists of one or several species", Darwin revised his opinion to say:

it is a hopeless endeavour to decide this point on sound grounds, until some definition of the term "species" is generally accepted; and the definition must not include an element which cannot possibly be ascertained, such as an act of creation.^[8]

The modern theory of evolution depends on a fundamental redefinition of "species". Prior to Darwin, naturalists viewed species as ideal or general types, which could be exemplified by an ideal specimen bearing all the traits general to the species. Darwin's theories shifted attention from uniformity to variation and from the general to the particular. According to intellectual historian Louis Menand,

Once our attention is redirected to the individual, we need another way of making generalizations. We are no longer interested in the conformity of an individual to an ideal type; we are now interested in the relation of an individual to the other individuals with which it interacts. To generalize about groups of interacting individuals, we need to drop the language of types and essences, which is prescriptive (telling us what finches should be), and adopt the language of statistics and probability, which is predictive (telling us what the average finch, under specified conditions, is likely to do). Relations will be more important than categories; functions, which are variable, will be more important than purposes; transitions will be more important than boundaries; sequences will be more important than hierarchies.

This shift results in a new approach to "species"; Darwin

concluded that species are what they appear to be: ideas, which are provisionally useful for naming groups of interacting individuals. "I look at the term species", he wrote, "as one arbitrarily given for the sake of convenience to a set of individuals closely resembling each other ... It does not essentially differ from the word variety, which is given to less distinct and more fluctuating forms. The term variety, again, in comparison with mere individual differences, is also applied arbitrarily, and for convenience sake." ^[9]

Practically, biologists define species as populations of organisms that have a high level of genetic similarity. This may reflect an adaptation to the same niche, and the transfer of genetic material from one individual to others, through a variety of possible means. The exact level of similarity used in such a definition is arbitrary, but this is the most common definition used for organisms that reproduce asexually (asexual reproduction), such as some plants and microorganisms.

This lack of any clear species concept in microbiology has led to some authors arguing that the term "species" is not useful when studying bacterial evolution. Instead they see genes as moving freely between even distantly-related bacteria, with the entire bacterial domain being a single gene pool. Nevertheless, a kind of rule of thumb has been established, saying that species of Bacteria or Archaea with 16S rRNA gene sequences more similar than 97% to each other need to be checked by DNA-DNA Hybridization if they belong to the same species or not.^[10] This concept has been updated recently, saying that the border of 97% was too low and can be raised to 98.7%.^[11]

In the study of sexually reproducing organisms, where genetic material is shared through the process of reproduction, the ability of two organisms to interbreed and produce fertile offspring of both genders is generally accepted as a simple indicator that the organisms share enough genes to be considered members of the same species. Thus a "species" is a group of interbreeding organisms.

This definition can be extended to say that a species is a group of organisms that could potentially interbreed – fish could still be classed as the same species even if they live in different lakes, as long as they could still interbreed were they ever to come into contact with each other. On the other hand, there are many examples of series of three or more distinct populations, where individuals of the population in the middle can interbreed with the populations to either side, but individuals of the populations on either side cannot interbreed. Thus, one could argue that these populations constitute a single species, or two distinct species. This is not a paradox; it is evidence that species are defined by gene frequencies, and thus have fuzzy boundaries.

Consequently, any single, universal definition of "species" is necessarily arbitrary. Instead, biologists have proposed a range of definitions; which definition a biologists uses is a pragmatic choice, depending on the particularities of that biologist's research.

Typological species

A group of organisms in which individuals are members of the species if they sufficiently conform to certain fixed properties. The clusters of variations or phenotypes within specimens (i.e. longer and shorter tails) would differentiate the species. This method was used as a "classical" method of determining species, such as with Linnaeus early in evolutionary theory. However, we now know that different phenotypes do not always constitute different species (e.g.: a 4-winged Drosophila born to a 2-winged mother is not a different species). Species named in this manner are called morphospecies^[12]

Morphological species

A population or group of populations that differs morphologically from other populations. For example, we can distinguish between a chicken and a duck because they have different shaped bills and the duck has webbed feet. Species have been defined in this way since well before the beginning of recorded history. This species concept is much criticised because more recent genetic data reveal that genetically distinct populations may look very similar and, contrarily, large morphological differences sometimes exist between very closely-related populations. Nonetheless, most species known have been described solely from morphology.

Biological / Isolation species

A set of actually or potentially interbreeding populations. This is generally a useful formulation for scientists working with living examples of the higher taxa like mammals, fish, and birds, but more problematic for organisms that do not reproduce sexually. The results of breeding experiments done in artificial conditions may or may not reflect what would happen if the same organisms encountered each other in the wild, making it difficult to gauge whether or not the results of such experiments are meaningful in reference to natural populations.

Biological / reproductive species

Two organisms that are able to reproduce naturally to produce fertile offspring of both genders. Organisms that can reproduce but almost always make infertile hybrids of at least one gender, such as a mule, hinny or F1 male cattalo are not considered to be the same species.

Recognition species

based on shared reproductive systems, including mating behavior. The Recognition concept of species has been introduced by Hugh E. H. Paterson.

Mate-recognition species

A group of organisms that are known to recognize one another as potential mates. Like the isolation species concept above, it applies only to organisms that reproduce sexually. Unlike the isolation species concept, it focuses specifically on pre-mating reproductive isolation.

Evolutionary / Darwinian species

A group of organisms that shares an ancestor; a lineage that maintains its integrity with respect to other lineages through both time and space. At some point in the progress of such a group, some members may diverge from the main population and evolve into a subspecies, a process that eventually will lead to the formation of a new full species if isolation (geographical or ecological) is maintained.

Phylogenetic (Cladistic)^{[verification needed]}

A group of organisms that shares an ancestor; a lineage that maintains its integrity with respect to other lineages through both time and space. At some point in the progress of such a group, members may diverge from one another: when such a divergence becomes sufficiently clear, the two populations are regarded as separate species. This differs from evolutionary species in that the parent species goes extinct taxonomically when a new species evolve, the mother and daughter populations now forming two new species. Subspecies as such are not recognized under this approach; either a population is a phylogenetic species or it is not taxonomically distinguishable.

Ecological species

A set of organisms adapted to a particular set of resources, called a niche, in the environment. According to this concept, populations form the discrete phenetic clusters that we recognize as species because the ecological and evolutionary processes controlling how resources are divided up tend to produce those clusters.

Genetic species

based on similarity of DNA of individuals or populations. Techniques to compare similarity of DNA include DNA-DNA hybridization, and genetic fingerprinting (or DNA barcoding).

Phenetic species

based on phenotypes.^{[verification needed]}

Microspecies

Species that reproduce without meiosis or fertilization so that each generation is genetically identical to the previous generation. See also apomixis.

Cohesion species

Most inclusive population of individuals having the potential for phenotypic cohesion through intrinsic cohesion mechanisms. This is an expansion of the mate-recognition species concept to allow for post-mating isolation mechanisms; no matter whether populations can hybridize successfully, they are still distinct cohesion species if the amount of hybridization is insufficient to completely mix their respective gene pools.

Evolutionarily Significant Unit (ESU)

An evolutionarily significant unit is a population of organisms that is considered distinct for purposes of conservation. Often referred to as a species or a wildlife species, an ESU also has several possible definitions, which coincide with definitions of species.

In practice, these definitions often coincide, and the differences between them are more a matter of emphasis than of outright contradiction. Nevertheless, no species concept yet proposed is entirely objective, or can be applied in all cases without resorting to judgment. Given the complexity of life, some have argued that such an objective definition is in all likelihood impossible, and biologists should settle for the most practical definition.

For most vertebrates, this is the biological species concept (BSC), and to a lesser extent (or for different purposes) the phylogenetic species concept (PSC). Many BSC subspecies are considered species under the PSC; the difference between the BSC and the PSC can be summed up insofar as that the BSC defines a species as a consequence of manifest evolutionary history, while the PSC defines a species as a consequence of manifest evolutionary potential. Thus, a PSC species is "made" as soon as an evolutionary lineage has started to separate, while a BSC species starts to exist only when the lineage separation is complete. Accordingly, there can be considerable conflict between alternative classifications based upon the PSC versus BSC, as they differ completely in their treatment of taxa that would be considered subspecies under the latter model

Difficulty of defining "species" and identifying particular species

The Greenish Warbler demonstrates the concept of a ring species.

It is surprisingly difficult to define the word "species" in a way that applies to all naturally occurring organisms, and the debate among biologists about how to define "species" and how to identify actual species is called the species problem.

Most textbooks follow Ernst Mayr's definition of a species as "groups of actually or potentially interbreeding natural populations, which are reproductively isolated from other such groups".^[5]

Various parts of this definition serve to exclude some unusual or artificial matings:

• Those which occur only in captivity (when the animal's normal mating partners may not be available) or as a result of deliberate human action.
• Animals which may be physically and physiologically capable of mating but do not normally do so in the wild, for various reasons.
• Animals whose offspring are normally sterile.

The typical textbook definition above works well for most multi-celled organisms, but there are several types of situations in which it breaks down:

• By definition it applies only to organisms that reproduce sexually. So it does not work for asexually reproducing single-celled organisms and for the relatively few parthenogenetic multi-celled organisms. The term "phylotype" is often applied to such organisms.
• Biologists frequently do not know whether two morphologically similar groups of organisms are "potentially" capable of interbreeding.
• There is considerable variation in the degree to which hybridization may succeed under natural conditions, or even in the degree to which some organisms use sexual reproduction between individuals to breed.
• In ring species, members of adjacent populations interbreed successfully but members of some non-adjacent populations do not.
• In a few cases it may be physically impossible for animals that are members of the same species to mate. However, these are cases in which human intervention has caused gross morphological changes, and are therefore excluded by the biological species concept.

Horizontal gene transfer makes it even more difficult to define the word "species". There is strong evidence of horizontal gene transfer between very dissimilar groups of prokaryotes, and at least occasionally between dissimilar groups of eukaryotes; and Williamson^[6] argues that there is evidence for it in some crustaceans and echinoderms. All definitions of the word "species" assume that an organism gets all its genes from one or two parents which are very like that organism, but horizontal gene transfer makes that assumption false.

Biologists' working definition

A usable definition of the word "species" and reliable methods of identifying particular species is essential for stating and testing biological theories and for measuring biodiversity. Traditionally, multiple examples of a proposed species must be studied for unifying characters before it can be regarded as a species. It is generally difficult to give precise taxonomic rankings to extinct species known only from fossils.

Some biologists may see species as statistical phenomena, as opposed to the traditional idea, with a species seen as a class of organisms. In that case, a species is defined as a separately evolving lineage that forms a single gene pool. Although properties such as DNA-sequences and morphology are used to help separate closely-related lineages, this definition has fuzzy boundaries.^[1] However, the exact definition of the term "species" is still controversial, particularly in prokaryotes,^[2] and this is called the species problem.^[3] Biologists have proposed a range of more precise definitions, but the definition used is a pragmatic choice that depends on the particularities of the species concerned.^[3]

Common names and species

The commonly used names for plant and animal taxa sometimes correspond to species: for example, "lion", "walrus", and "Camphor tree" – each refers to a species. In other cases common names do not: for example, "deer" refers to a family of 34 species, including Eld's Deer, Red Deer and Elk (Wapiti). The last two species were once considered a single species, illustrating how species boundaries may change with increased scientific knowledge.

Because of the difficulties with both defining and tallying the total numbers of different species in the world, it is estimated that there are anywhere between 2 and 100 million different species.^[4]

Placement within genera

Ideally, a species is given a formal, scientific name, although in practice there are very many unnamed species (which have only been described, not named). When a species is named, it is placed within a genus. From a scientific point of view this can be regarded as a hypothesis that the species is more closely related to other species within its genus (if any) than to species of other genera. A genus is commonly included in a hierarchy, with as the best-known taxonomic ranks: life, domain, kingdom, phylum, class, order, family, genus, and species. This assignment to a genus is not immutable; later a different (or the same) taxonomist may assign it to a different genus, in which case the name will also change.

In biological nomenclature, the name for a species is a two-part name (a binomial name), treated as Latin, although roots from any language can be used as well as names of locales or individuals. The generic name is listed first (with its leading letter capitalized), followed by a second term, the specific name (or specific epithet). For example, the species commonly known as the Longleaf Pine is Pinus palustris; gray wolves belong to the species Canis lupus, coyotes to Canis latrans, golden jackals to Canis aureus, etc., and all of those belong to the genus Canis (which also contains many other species). The name of the species is the whole binomial, not just the second term (which may be called the specific name for animals).

This binomial naming convention, later formalized in the biological codes of nomenclature, was first used by Leonhart Fuchs and introduced as the standard by Carolus Linnaeus in his 1753, Species Plantarum (followed by his, 1758 Systema Naturae, 10th edition). At that time, the chief biological theory was that species represented independent acts of creation by God and were therefore considered objectively real and immutable, so the hypothesis of common descent did not apply.

Abbreviated names

Books and articles sometimes intentionally do not identify species fully and use the abbreviation "sp." in the singular or "spp." in the plural in place of the specific epithet: for example, Canis sp. This commonly occurs in the following types of situations:

• The authors are confident that some individuals belong to a particular genus but are not sure to which exact species they belong. This is particularly common in paleontology.

• The authors use "spp." as a short way of saying that something applies to many species within a genus, but do not wish to say that it applies to all species within that genus. If scientists mean that something applies to all species within a genus, they use the genus name without the specific epithet.

In books and articles, genus and species names are usually printed in italics. If using "sp." and "spp.", these should not be italicized.

Historical development of the species concept

The following text needs to be harmonized with text in Species problem.

This section does not cite any references or sources. Please help improve this article by adding citations to reliable sources. Unsourced material may be challenged and removed. (September 2009)

Linnaeus believed in the fixity of species.

In the earliest works of science, a species was simply an individual organism that represented a group of similar or nearly identical organisms. No other relationships beyond that group were implied. Aristotle used the words genus and species to mean generic and specific categories. Aristotle and other pre-Darwinian scientists took the species to be distinct and unchanging, with an "essence", like the chemical elements. When early observers began to develop systems of organization for living things, they began to place formerly isolated species into a context. Many of these early delineation schemes would now be considered whimsical and these included consanguinity based on color (all plants with yellow flowers) or behavior (snakes, scorpions and certain biting ants).

In the 18th century Carolus Linnaeus classified organisms according to differences in the form of reproductive apparatus. Although his system of classification sorts organisms according to degrees of similarity, it made no claims about the relationship between similar species. At that time, it was still widely believed that there was no organic connection between species, no matter how similar they appeared. This approach also suggested a type of idealism: the notion that each species existed as an "ideal form". Although there are always differences (although sometimes minute) between individual organisms, Linnaeus considered such variation problematic. He strove to identify individual organisms that were exemplary of the species, and considered other non-exemplary organisms to be deviant and imperfect.

By the 19th century most naturalists understood that species could change form over time, and that the history of the planet provided enough time for major changes. Jean-Baptiste Lamarck, in his 1809 Zoological Philosophy, offered one of the first logical arguments against creationism. The new emphasis was on determining how a species could change over time. Lamarck suggested that an organism could pass on an acquired trait to its offspring, i.e., the giraffe's long neck was attributed to generations of giraffes stretching to reach the leaves of higher treetops (this well-known and simplistic example, however, does not do justice to the breadth and subtlety of Lamarck's ideas). With the acceptance of the natural selection idea of Charles Darwin in the 1860s, however, Lamarck's view of goal-oriented evolution, also known as a teleological process, was eclipsed. Recent interest in inheritance of acquired characteristics centers around epigenetic processes, e.g. methylation, that do not affect DNA sequences, but instead alter expression in an inheritable manner. Thus, neo-lamarckism, as it is sometimes termed, is not a challenge to the theory of evolution by natural selection.

Charles Darwin and Alfred Wallace provided what scientists now consider as the most powerful and compelling theory of evolution. Darwin argued that it was populations that evolved, not individuals. His argument relied on a radical shift in perspective from that of Linnaeus: rather than defining species in ideal terms (and searching for an ideal representative and rejecting deviations), Darwin considered variation among individuals to be natural. He further argued that variation, far from being problematic, actually provides the explanation for the existence of distinct species.

Darwin's work drew on Thomas Malthus' insight that the rate of growth of a biological population will always outpace the rate of growth of the resources in the environment, such as the food supply. As a result, Darwin argued, not all the members of a population will be able to survive and reproduce. Those that did will, on average, be the ones possessing variations—however slight—that make them slightly better adapted to the environment. If these variable traits are heritable, then the offspring of the survivors will also possess them. Thus, over many generations, adaptive variations will accumulate in the population, while counter-adaptive traits will tend to be eliminated.

It should be emphasized that whether a variation is adaptive or non-adaptive depends on the environment: different environments favor different traits. Since the environment effectively selects which organisms live to reproduce, it is the environment (the "fight for existence") that selects the traits to be passed on. This is the theory of evolution by natural selection. In this model, the length of a giraffe's neck would be explained by positing that proto-giraffes with longer necks would have had a significant reproductive advantage to those with shorter necks. Over many generations, the entire population would be a species of long-necked animals.

In 1859, when Darwin published his theory of natural selection, the mechanism behind the inheritance of individual traits was unknown. Although Darwin made some speculations on how traits are inherited (pangenesis), his theory relies only on the fact that inheritable traits exist, and are variable (which makes his accomplishment even more remarkable.) Although Gregor Mendel's paper on genetics was published in 1866, its significance was not recognized. It was not until 1900 that his work was rediscovered by Hugo de Vries, Carl Correns and Erich von Tschermak, who realised that the "inheritable traits" in Darwin's theory are genes.

The theory of the evolution of species through natural selection has two important implications for discussions of species—consequences that fundamentally challenge the assumptions behind Linnaeus' taxonomy. First, it suggests that species are not just similar, they may actually be related. Some students of Darwin argue that all species are descended from a common ancestor. Second, it supposes that "species" are not homogeneous, fixed, permanent things; members of a species are all different, and over time species change. This suggests that species do not have any clear boundaries but are rather momentary statistical effects of constantly changing gene-frequencies. One may still use Linnaeus' taxonomy to identify individual plants and animals, but one can no longer think of species as independent and immutable.

The rise of a new species from a parental line is called speciation. There is no clear line demarcating the ancestral species from the descendant species.

Although the current scientific understanding of species suggests that there is no rigorous and comprehensive way to distinguish between different species in all cases, biologists continue to seek concrete ways to operationalize the idea. One of the most popular biological definitions of species is in terms of reproductive isolation; if two creatures cannot reproduce to produce fertile offspring of both genders, then they are in different species. This definition captures a number of intuitive species boundaries, but it remains imperfect. It has nothing to say about species that reproduce asexually, for example, and it is very difficult to apply to extinct species. Moreover, boundaries between species are often fuzzy: there are examples where members of one population can produce fertile offspring of both genders with a second population, and members of the second population can produce fertile offspring of both genders with members of a third population, but members of the first and third population cannot produce fertile offspring, or can only produce fertile offspring of the homozygous gender. Consequently, some people reject this definition of a species.

Richard Dawkins defines two organisms as conspecific if and only if they have the same number of chromosomes and, for each chromosome, both organisms have the same number of nucleotides (The Blind Watchmaker, p. 118). However, most if not all taxonomists would strongly disagree^{[citation needed]}. For example, in many amphibians, most notably in New Zealand's Leiopelma frogs, the genome consists of "core" chromosomes which are mostly invariable and accessory chromosomes, of which exist a number of possible combinations. Even though the chromosome numbers are highly variable between populations, these can interbreed successfully and form a single evolutionary unit. In plants, polyploidy is extremely commonplace with few restrictions on interbreeding; as individuals with an odd number of chromosome sets are usually sterile, depending on the actual number of chromosome sets present, this results in the odd situation where some individuals of the same evolutionary unit can interbreed with certain others and some cannot, with all populations being eventually linked as to form a common gene pool.

The classification of species has been profoundly affected by technological advances that have allowed researchers to determine relatedness based on molecular markers, starting with the comparatively crude blood plasma precipitation assays in the mid-20th century to Charles Sibley's ground-breaking DNA-DNA hybridization studies in the 1970s leading to DNA sequencing techniques. The results of these techniques caused revolutionary changes in the higher taxonomic categories (such as phyla and classes), resulting in the reordering of many branches of the phylogenetic tree (see also: molecular phylogeny). For taxonomic categories below genera, the results have been mixed so far; the pace of evolutionary change on the molecular level is rather slow, yielding clear differences only after considerable periods of reproductive separation. DNA-DNA hybridization results have led to misleading conclusions, the Pomarine Skua – Great Skua phenomenon being a famous example. Turtles have been determined to evolve with just one-eighth of the speed of other reptiles on the molecular level, and the rate of molecular evolution in albatrosses is half of what is found in the rather closely related storm-petrels. The hybridization technique is now obsolete and is replaced by more reliable computational approaches for sequence comparison. Molecular taxonomy is not directly based on the evolutionary processes, but rather on the overall change brought upon by these processes. The processes that lead to the generation and maintenance of variation such as mutation, crossover and selection are not uniform (see also molecular clock). DNA is only extremely rarely a direct target of natural selection rather than changes in the DNA sequence enduring over generations being a result of the latter; for example, silent transition-transversion combinations would alter the melting point of the DNA sequence, but not the sequence of the encoded proteins and thus are a possible example where, for example in microorganisms, a mutation confers a change in fitness all by itself.

The isolation species concept in more detail

This article is written like a personal reflection or essay and may require cleanup. Please help improve it by rewriting it in an encyclopedic style. (December 2007)

It has been suggested that this article or section be merged into Species problem. (Discuss)

A mule is the infertile offspring of a male donkey and a female horse.

In general, for large, complex organisms that reproduce sexually (such as mammals and birds), one of several variations on the isolation or biological species concept is employed. Often, the distinction between different species, even quite closely related ones, is simple. Horses (Equus caballus) and donkeys (Equus asinus) are easily told apart even without study or training, and yet are so closely related that they can interbreed after a fashion. Because the result, a mule or hinny, is not fertile, they are clearly separate species.

But many cases are more difficult to decide. This is where the isolation species concept diverges from the evolutionary species concept. Both agree that a species is a lineage that maintains its integrity over time, that is diagnosably different from other lineages (else we could not recognise it), is reproductively isolated (else the lineage would merge into others, given the chance to do so), and has a working intra-species recognition system (without which it could not continue). In practice, both also agree that a species must have its own independent evolutionary history—otherwise the characteristics just mentioned would not apply. The species concepts differ in that the evolutionary species concept does not make predictions about the future of the population: it simply records that which is already known. In contrast, the isolation species concept refuses to assign the rank of species to populations that, in the best judgement of the researcher, would recombine with other populations if given the chance to do so.

The isolation question

There are, essentially, two questions to resolve. First, is the proposed species consistently and reliably distinguishable from other species? Second, is it likely to remain so in the future? To take the second question first, there are several broad geographic possibilities.

• The proposed species are sympatric—they occupy the same habitat. Observation of many species over the years has failed to establish even a single instance of two diagnostically different populations that exist in sympatry and have then merged to form one united population. Without reproductive isolation, population differences cannot develop, and given reproductive isolation, gene flow between the populations cannot merge the differences. This is not to say that cross breeding does not take place at all, simply that it has become negligible. Generally, the hybrid individuals are less capable of successful breeding than pure-bred individuals of either species.

• The proposed species are allopatric—they occupy different geographical areas. Obviously, it is not possible to observe reproductive isolation in allopatric groups directly. Often it is not possible to achieve certainty by experimental means either: even if the two proposed species interbreed in captivity, this does not demonstrate that they would freely interbreed in the wild, nor does it always provide much information about the evolutionary fitness of hybrid individuals. A certain amount can be inferred from other experimental methods: for example, do the members of population A respond appropriately to playback of the recorded mating calls of population B? Sometimes, experiments can provide firm answers. For example, there are seven pairs of apparently almost identical marine snapping shrimp (Alpheus) populations on either side of the Isthmus of Panama, which did not exist until about 3 million years ago. Until then, it is assumed, they were members of the same seven species. But when males and females from opposite sides of the isthmus are placed together, they fight instead of mating. Even if the isthmus were to sink under the waves again, the populations would remain genetically isolated: therefore they are now different species. In many cases, however, neither observation nor experiment can produce certain answers, and the determination of species rank must be made on a 'best guess' basis from a general knowledge of other related organisms.

• The proposed species are parapatric—they have breeding ranges that abut but do not overlap. This is fairly rare, particularly in temperate regions. The dividing line is often a sudden change in habitat (an ecotone) like the edge of a forest or the snow line on a mountain, but can sometimes be remarkably trivial. The parapatry itself indicates that the two populations occupy such similar ecological roles that they cannot coexist in the same area. Because they do not crossbreed, it is safe to assume that there is a mechanism, often behavioral, that is preventing gene flow between the populations, and that therefore they should be classified as separate species.

• There is a hybrid zone where the two populations mix. Typically, the hybrid zone will include representatives of one or both of the 'pure' populations, plus first-generation and back-crossing hybrids. The strength of the barrier to genetic transmission between the two pure groups can be assessed by the width of the hybrid zone relative to the typical dispersal distance of the organisms in question. The dispersal distance of oaks, for example, is the distance that a bird or squirrel can be expected to carry an acorn; the dispersal distance of Numbats is about 15 kilometres, as this is as far as young Numbats will normally travel in search of vacant territory to occupy after leaving the nest. The narrower the hybrid zone relative to the dispersal distance, the less gene flow there is between the population groups, and the more likely it is that they will continue on separate evolutionary paths. Nevertheless, it can be very difficult to predict the future course of a hybrid zone; the decision to define the two hybridizing populations as either the same species or as separate species is difficult and potentially controversial.

• The variation in the population is clinal; at either extreme of the population's geographic distribution, typical individuals are clearly different, but the transition between them is seamless and gradual. For example, the Koalas of northern Australia are clearly smaller and lighter in colour than those of the south, but there is no particular dividing line: the further south an individual Koala is found, the larger and darker it is likely to be; Koalas in intermediate regions are intermediate in weight and colour. In contrast, over the same geographic range, black-backed (northern) and white-backed (southern) Australian Magpies do not blend from one type to another: northern populations have black backs, southern populations white backs, and there is an extensive hybrid zone where both 'pure' types are common, as are crossbreeds. The variation in Koalas is clinal (a smooth transition from north to south, with populations in any given small area having a uniform appearance), but the variation in magpies is not clinal. In both cases, there is some uncertainty regarding correct classification, but the consensus view is that species rank is not justified in either. The gene flow between northern and southern magpie populations is judged to be sufficiently restricted to justify terming them subspecies (not full species); but the seamless way that local Koala populations blend one into another shows that there is substantial gene flow between north and south. As a result, experts tend to reject even subspecies rank in this case.

The difference question

Obviously when defining a species, the geographic circumstances become meaningful only if the population groups in question are clearly different: if they are not consistently and reliably distinguishable from one another, then we have no grounds for believing that they might be different species. The key question in this context, is "how different is different?" and the answer is usually "it all depends".

In theory, it would be possible to recognise even the tiniest of differences as sufficient to delineate a separate species, provided only that the difference is clear and consistent (and that other criteria are met). There is no universal rule to state the smallest allowable difference between two species, but in general, very trivial differences are ignored on the twin grounds of simple practicality, and genetic similarity: if two population groups are so close that the distinction between them rests on an obscure and microscopic difference in morphology, or a single base substitution in a DNA sequence, then a demonstration of restricted gene flow between the populations will probably be difficult in any case.

More typically, one or other of the following requirements must be met:

• It is possible to reliably measure a quantitative difference between the two groups that does not overlap. A population has, for example, thicker fur, rougher bark, longer ears, or larger seeds than another population, and although this characteristic may vary within each population, the two do not grade into one another, and given a reasonably large sample size, there is a definite discontinuity between them. Note that this applies to populations, not individual organisms, and that a small number of exceptional individuals within a population may 'break the rule' without invalidating it. The less a quantitative difference varies within a population and the more it varies between populations, the better the case for making a distinction. Nevertheless, borderline situations can only be resolved by making a 'best-guess' judgement.

• It is possible to distinguish a qualitative difference between the populations; a feature that does not vary continuously but is either entirely present or entirely absent. This might be a distinctively shaped seed pod, an extra primary feather, a particular courting behaviour, or a clearly differentDNA sequence.

Sometimes it is not possible to isolate a single difference between species, and several factors must be taken in combination. This is often the case with plants in particular. In eucalypts, for example, Corymbia ficifolia cannot be reliably distinguished from its close relative Corymbia calophylla by any single measure (and sometimes individual trees cannot be definitely assigned to either species), but populations of Corymbia can be clearly told apart by comparing the colour of flowers, bark, and buds, number of flowers for a given size of tree, and the shape of the leaves and fruit.

When using a combination of characteristics to distinguish between populations, it is necessary to use a reasonably small number of factors (if more than a handful are needed, the genetic difference between the populations is likely to be insignificant and is unlikely to endure into the future), and to choose factors that are functionally independent (height and weight, for example, should usually be considered as one factor, not two).

Importance in biological classification

The idea of species has a long history. It is one of the most important levels of classification, for several reasons:

• It often corresponds to what lay people treat as the different basic kinds of organism – dogs are one species, cats another.
• It is the standard binomial nomenclature (or trinomial nomenclature) by which scientists typically refer to organisms.
• It is the highest taxonomic level which mostly cannot be made more or less inclusionary.

After years of use, the concept remains central to biology and a host of related fields, and yet also remains at times ill-defined.

Implications of assignment of species status

The naming of a particular species should be regarded as a hypothesis about the evolutionary relationships and distinguishability of that group of organisms. As further information comes to hand, the hypothesis may be confirmed or refuted. Sometimes, especially in the past when communication was more difficult, taxonomists working in isolation have given two distinct names to individual organisms later identified as the same species. When two named species are discovered to be of the same species, the older species name is usually retained, and the newer species name dropped, a process called synonymization, or colloquially, as lumping. Dividing a taxon into multiple, often new, taxons is called splitting. Taxonomists are often referred to as "lumpers" or "splitters" by their colleagues, depending on their personal approach to recognizing differences or commonalities between organisms (see lumpers and splitters).

Traditionally, researchers relied on observations of anatomical differences, and on observations of whether different populations were able to interbreed successfully, to distinguish species; both anatomy and breeding behavior are still important to assigning species status. As a result of the revolutionary (and still ongoing) advance in microbiological research techniques, including DNA analysis, in the last few decades, a great deal of additional knowledge about the differences and similarities between species has become available. Many populations which were formerly regarded as separate species are now considered to be a single taxon, and many formerly grouped populations have been split. Any taxonomic level (species, genus, family, etc.) can be synonymized or split, and at higher taxonomic levels, these revisions have been still more profound.

From a taxonomical point of view, groups within a species can be defined as being of a taxon hierarchically lower than a species. In zoology only the subspecies is used, while in botany the variety, subvariety, and form are used as well. In conservation biology, the concept of evolutionary significant units (ESU) is used, which may be define either species or smaller distinct population segments.

Numbers of species

Undiscovered and discovered species

Bearing in mind the aforementioned problems with categorising species, the following numbers are only a soft guide. They break down as follows:^[13]

Total number of species (estimated): 7–100 millions (identified and unidentified), including:

• 5–10 million bacteria^[14];
• 74,000–120,000 fungi^[15];

Of the identified eukaryote species we have:

• 1.6 million, including:
  • 297,326 plants, including:
    • 15,000 mosses,
    • 13,025 Ferns and horsetails,
    • 980 gymnosperms,
    • 258,650 angiosperms,
      • 199,350 dicotyledons,
      • 59,300 monocotyledons,
    • 9,671 Red and green algae,
  • 28,849 fungi & other non-animals, including:
    • 10,000 lichens,
    • 16,000 mushrooms,
    • 2,849 brown algae,
  • 1,250,000 animals, including:
    • 1,203,375 invertebrates:
      • 950,000 insects,
      • 81,000 mollusks,
      • 40,000 crustaceans,
      • 2,175 corals,
      • 130,200 others;
    • 59,811 vertebrates:
      • 29,300 fish,
      • 6,199 amphibians,
      • 8,240 reptiles,
      • 9,956 birds,
      • 5,416 mammals.

At present, organisations such as the Global Taxonomy Initiative, the European Distributed Institute of Taxonomy and the Census of Marine Life^[16] (the latter only for marine organisms) are trying to improve taxonomy and implement previously undiscovered species to the taxonomy system. Due to the fact that we know but a portion of the organisms in the biosphere, we do not have a complete understanding of the workings of our environment. To make matters worse, despite the discovery of new species, according to professor James Mallet, we are wiping out these species at an unprecedented rate. ^[17]This means that even before a new species has had the chance of being studied and classified, it may already be extinct.