HBW 4 - Foreword on species concepts and species limits in ornithology by Jürgen H. Haffer

View foreword in PDF format: 

The complexities of the “species problem” in biology are due to the gradual nature of the evolutionary process and to the geographical mode of speciation in animals and plants. Under any species concept proposed it is difficult in many cases to draw a definite line in the evolutionary continuum delimiting species taxa, except in the local, “ non-dimensional” situation. Biologists have discussed these complexities for over 100 years and continue to discuss them intensively at the present time. Several alternative notions on the general nature of species conceive these units as created entities, as groups of non-intergrading populations at any particular time level, or as phyletic lineages through time. These theoretical species concepts need to be distinguished from the different (wide to narrow) taxonomic delimitation of individual species taxa.

Species concepts and species limits

Under each of the theoretical species concepts zoologists delimited and are delimiting “narrow” or “wide” species taxa (splitters and lumpers, respectively), depending on whether these systematists place the taxonomic species limit at relatively “low” or “ high” levels of microtaxonomic differentiation, respectively. A species limit at a fairly high level of differentiation results in relatively few species taxa, each species comprising wide arrays of variously differentiated geographical representatives. On the other hand, a taxonomic species limit at a low level of differentiation results in more numerous, internally rather uniform and narrowly defined species taxa. Zoologists advocating different theoretical species concepts may in practice delimit species taxa in a similar manner. On the other hand, zoologists adhering to the same theoretical species concept may delimit species taxa quite differently.

In a recent survey, Mayden (1997) discussed 22 concepts of species that are in use today. I consider only those few concepts which have been or are being discussed by ornithologists. Most biologists of the nineteenth century applied a typological species concept, labelling as species any sample of organisms that was morphologically “different”. Morphology was used to recognize species taxa that correspond to the typological species concept. During this present century, the majority of biologists have favoured the biological species concept and a placement of species limits at intermediate levels of microtaxonomic differentiation. However, this consensus has been challenged by other proposals during the last 15 years.

1)   Typological Species Concept

Organic diversity is assumed to reflect the expression of underlying “types”, and the observed variation is seen as the result of different manifestations of these “types”. This concept is based on typological essentialism. This was the theoretical species concept of most pre-Darwinian zoologists and of some ornithologists during the late nineteenth and early twentieth centuries. They delimited species taxa following widely varying standards, applying broad to narrow species limits. This concept is basically non-evolutionary, as “type” are believed possible, but no branching evolution leading to the origin of new species. Creationists assume, or assumed, that many or most species (their basic types or their “primordial germs”) have been created independently.

2)   Biological Species Concept

After Darwin’s Origin of Species (1859), the species concept was part of evolutionary theory. This biological species concept (BSC) as elaborated on by Mayr (1942, 1963, 1997) applies only to sexually reproducing organisms and it is truly valid only in local (“non-dimensional”) situations, where species are sympatric. Such species are genetically, reproductively and ecologically separated units, each occupying a species-specific niche in nature (Mayr 1969, 1982; Bock 1986, 1995). Their genetic-reproductive isolation prevents effective interbreeding and merging.

It is the multidimensional species in taxonomy, with its extensions over space and time, which applies to most real units observed in nature, the species taxa (Fig. 1), and these are subject to all the difficulties of any pragmatic application of theoretical concepts (Mayr 1963, 1982; Bock 1979, 1986). The distinctiveness of species becomes increasingly vague as one progresses geographically and chronologically further and further away from a single point where two species occur in sympatry or parapatry, since: “ Species are groups of actually (or potentially) interbreeding populations that are reproductively isolated in nature from other such groups” (Mayr 1942, 1963, 1969). Reproductive isolation is understood to mean genetic isolation, and as Mayr (1968: 164) stated: the “possession of a shared genetic program is the common tie uniting individuals derived from the same gene pool of a given species.” Bock (1986, 1992) made this explicit by emending the definition of biospecies to read: “ a species is a group of actually or potentially interbreeding populations which are genetically isolated in nature from other such groups.” However, reproductive isolation between many biological species is not complete and, therefore, does not prevent hybridization at a low level in these cases (Mayr 1970, 1982; Grant & Grant 1992; Grant 1993). The view of species as “genetic clusters” (Mallet 1995) also represents the transcribed definition of biological species.

In Table 1, I schematically subdivided the process of microtaxonomic differentiation into 6 stages that are here listed from bottom to top in a presumed temporal sequence of gradually increasing differentiation (see also Fig. 1). However, not every differentiating taxon necessarily passes through each of these stages. In this table, aspects of behavioural differentiation between closely related forms are subsumed under “genetic isolation” (e.g. differing calls or songs) and/or “ecological separation” (e.g. different feeding behaviour). Table 1 is an attempt at visualizing the process of microtaxonomic differentiation through a schematic grid of increasing levels of morphological, genetic-reproductive and ecological differentiation. This grid and, in particular, the sharp boundaries of the various stages (microtaxonomic categories) are rather crude means of illustrating schematically the results of the differentiation process. Nature is not necessarily orderly, and extant faunas provide many examples of taxa at transitional stages between the categories distinguished here or of taxa which combine aspects of two categories in different areas of contact; for instance, two particular taxa may hybridize with each other in one of their areas of contact or overlap but not in another. “Ring species” represent other borderline cases, where two taxa that do not hybridize in a particular region are geographically interconnected by a continuous “ring” of intergrading subspecies. Morphological differences may or may not render a group of populations diagnosable taxonomically at an early stage of genetic differentiation (subspecies). However, stages in continuous character clines are not diagnosable and should not be designated as subspecies; only the end points of such clines qualify as such.

In many bird populations, genetic isolation may be completed before ecological segregation from the nearest relative is reached. This situation leads to geographical replacement (parapatry) of these forms (biospecies) with no or only limited hybridization along their contact zone (Fig. 1, Table 1). The frequent occurrence of superspecies in the avifaunas of the world (Bock & Farrand 1980; Sibley & Monroe 1990) indicates that ecological competition often prevents sympatry of geographical representatives long after speciation is complete (Lack 1944; Mayr 1963). Many species probably perfected ecological segregation and certain aspects of reproductive isolation in neosympatry. However, genetic isolation must evolve fully in the initial allopatric period of speciation (Bock 1979, 1986; Grant 1986). The process of speciation has terminated only after the differentiating taxa have attained genetic-reproductive isolation and ecological separation, leading to sympatry of the species.

Under the biological species concept, most authors currently place the species limit at “intermediate” level III (Table 1), as discussed by Short (1969, 1972), who also indicated the amount of hybridization acceptable between biological species (see also Amadon & Short 1992). Other authors place the species limit at slightly higher or lower levels, thus delimiting species taxa more widely or more narrowly, respectively. For example, adherents to Paterson’s (1985) recognition concept of species place the species limit at level IV. Representative taxa, especially of insects, that hybridize freely along a contact zone (because of the lack of pre-mating isolating mechanisms), but in which cases all or almost all hybrids are infertile (because of fully developed postmating isolating mechanisms) are considered as subspecies under the recognition concept, but are species under the biospecies concept. Some bird species which meet along “zones of overlap and hybridization” (Short 1969) may also represent taxa which are genetically isolated but not fully reproductively isolated in a strict sense.

On continents, the intergradation of contiguous populations or their geographical exclusion with no (or only restricted) hybridization along the contact zone determine their rank as subspecies or species, respectively. The amount of hybridization determining the “exact” limit of the species and subspecies categories is arbitrary. Geographically separated (allopatric) taxa on islands in the ocean or on ecological “islands” on continents are assigned subspecies or species status on the basis of inference, for instance through a comparison with the degree of difference between related sympatric species, the degree of difference between intergrading subspecies within widespread species, or the degree of difference between hybridizing populations in related species (Mayr 1969: 197, Mayr & Ashlock 1991: 104-105). The delimitation of polytypic species taxa uses morphological, geographical, ecological, behavioural and molecular information to infer the rank of isolated populations (Mayr 1996). This procedure presents no conceptual problems, because speciation requires geographical isolation of populations for the differentiation process to proceed. Therefore difficulties of judging the status of allopatric populations as subspecies or species and the existence of intermediate stages between subspecies and species (a “grey zone”) are to be expected. Admittedly, some authors (splitters) emphasize the differences among allopatric forms at intermediate stages of differentiation and treat them as species, whereas ­others (lumpers) tend to emphasize the similarities between these same forms and consider them conspecific.

The “horizontal” concept of species taxa refers to genetically isolated communities of a particular time level (comprising a small number of generations), such as the Recent period or any other time level of the geological history of the earth (Peters 1970; Bock 1979, 1986). The vertical extent (“thickness”) of such a geological time level (“slice”), or in other words the “duration” of a species taxon, is a matter of convention and, in most cases, will be determined by the incompleteness of the fossil record.

The distribution patterns of groups of closely related and geographically representative biospecies on continents in many cases resemble large scale mosaics composed of neatly interlocking patches formed by the ranges of the component species (Figs. 1 and 3-4). Two or more such species are combined in a superspecies if they “were once races of a single species but have now achieved species status” (Amadon 1966). In some groups of animals geographical exclusion (parapatry) probably persists long after the representative taxa have attained genetic isolation, and not only one but two or more periods of successive speciation events have occurred (Haffer 1986). The members of a superspecies are in most cases each others’ closest relatives, because of a basically consistent association between morphological character evolution, genetic-reproductive isolation and ecological differentiation. However, detailed cladistic analyses may reveal that this is not true in some cases when one of the members of a superspecies is the sister taxon of another widely sympatric species. It remains to be determined how frequent such situations actually are.

Several entomologists (Jordan, Poulton) and ornithologists (e.g. Seebohm, Coues) developed the ideas of biological species during the late nineteenth century, although their origin is much older. These ideas were elaborated on and increasingly applied by ornithologists during the present century in Europe (Hartert, Stresemann, Rensch) and in North America (Chapman, Grinnell, Miller and especially Mayr).

3)   Evolutionary and Phylogenetic “Species” Concepts (The Phyletic Lineage)

Paleontologists, beginning with Simpson (1951, 1961), defined species as follows: “ An evolutionary species is a lineage (an ancestral-descendent sequence of populations) evolving separately from others and with its own unitary evolutionary role and tendencies.” Wiley (1981) emended the wording of this definition somewhat. Under this concept, species limits may or may not coincide with speciation events (i.e. branching of lineages), and every geographical isolate has to be treated as a species. Other difficulties refer to the definitions of “ evolutionary tendency” and “historical fate”, as well as to the determination of species limits along evolving phyletic lineages (Mayr 1996). Most phylogenetic systematists (cladists) consider a species as a phyletic lineage between two successive speciation (branching) events or until the lineage terminates. Character change may or may not occur in the two daughter ­species.

These concepts view species as phyletic lineages through time. However, a “vertical” lineage represents an evolutionary phenomenon quite different from the notion of the “horizontal” biological species discussed above. The phyletic lineage is the continuum of a species as its members reproduce generation after generation through time. The phenotypic characteristics of the members of a phylogenetic lineage, and hence the underlying genetic bases, may remain the same over long geological periods (stasis) or change more or less gradually through time (phyletic evolution). As Bock (1986: 38) and Szalay & Bock (1991: 15) have stated, “a cross-section of a phyletic lineage at any point in time is a species (theoretical, non-dimensional). However, different time slices through the same phyletic lineage are not different species, nor are they the same species. They are simply different cross-sections of the lineage at different times, with the earlier one being ancestral to the later one. Each time slice is a species, but it makes no sense to ask whether they are the same or different species; the question lies outside the theoretical, non-dimensional species concept and hence, from a theoretical perspective, is a non-question.”

In this sense a species has no origin, lifespan or age. Of course, all phyletic lineages need to be studied in detail as they are important entities of the evolutionary history of a group of animals but, in contrast to species, they are not involved in the processes of evolution (i.e. phyletic evolution and speciation), which take place in living populations.

Species taxa under Phylogenetic species concepts (PSC’s) include several basal taxonomic units or only one basal unit. Accordingly, species limits range from (a) fairly wide to (b) intermediate and (c) narrow.

(a) Hennig (1966) and many other cladists (e.g. Willmann 1985) delimit extant species rather broadly, like multidimensional species taxa under the biological species concept (Hennigian species concept of Mayden 1997). They consider it inappropriate to enquire whether species are monophyletic, paraphyletic or polyphyletic, stating that these terms apply only to groups of species. Under this concept, the principles of cladistic analysis apply only when reticulation within a species gives way to the splitting of one species lineage into two new species (phylogenetic relations).

(b) Other cladists disagree and apply cladistic methods to historical units also at lower (i.e. intraspecific) levels of differentiation, as far as this is possible in view of reticulating genealogical relationships between interbreeding individuals. These cladists combine hybridizing forms as subspecies in one species taxon if and only if the resulting unit is monophyletic (monophyly sensu Hennig); paraphyletic entities are not accepted as species (monophyly version, PSC 2, of the phylogenetic species concept; Mayden 1997). Under this concept, organisms are first analysed cladistically, and isolated basal taxa and monophyletic groups of basal taxa are subsequently ranked as monotypic species or polytypic (monophyletic) species, respectively (Mishler & Brandon 1987). This concept leads to intermediate species limits.

(c) Cladists applying narrow species limits assign species status to any population that is morphologically diagnosable (levels I or II in Table 1). Under this diagnosable version, PSC 1, of the phylogenetic species concept (Mayden 1997) many subspecies are ranked as “species”, unless they are merely stages in a continuous cline; and biospecies Homo sapiens is presumably again split into several phylogenetic “species”, as was done 200 years ago. Of the previously named subspecies of animals, only those that are “diagnosably” distinct are considered phylogenetic species. However, the limits of diagnosability, and therefore the decision as to which of the subspecies are accepted as “phylospecies”, are determined subjectively, as Snow (1997) discussed for the Eurasian Blackbird (Turdus merula group) and the Coal Tit (Parus ater) of ­Eurasia.

The concept of narrow phylogenetic species (Cracraft 1983; Zink & McKitrick 1995; Zink 1996, 1997) was proposed under a general cladistic framework; however, the definition does not specifically refer to “species” as lineages. Under this concept, no attempt is made to express taxonomically the hierarchy of increasingly differentiated geographical representatives (Fig. 1), which are all designated uniformly as “species” regardless of whether one is dealing with rather weakly differentiated taxa connected by zones of hybridization (subspecies under the BSC), with non-hybridizing geographical representatives (paraspecies or allospecies), or with independent taxa with no close relatives (isospecies; Amadon & Short 1992). The term “phylogenetic species” subsumes taxa of conspicuously varying biological differentiation from those at early stages of speciation to taxa that have reached phylogenetic independence. Another problem under PSC 1 is that the number of species taxa recognized is a matter of the resolving power of the analytical tools available (Avise & Ball 1990); therefore species limits are highly subjective. Numerous small populations or even groups of individuals may be “ diagnosable” with improved laboratory techniques and would thus qualify as “species”. Moreover, PSC 1 can divide up organisms into overlapping and incompatible species, depending on which characters are picked to be diagnostic (Hull 1997). The difficulties in assessing many subspecies of polytypic biospecies are well known, but this is not very important, because it refers to intraspecific units. On the other hand, this problem is acute under PSC 1, where the decision affects the recognition or non-recognition of species. This difficulty “will make application of PSC 1 hard to achieve with any hope of a consensus or of stability” (Snow 1997: 113). Such difficulties are particularly apparent in many groups of tropical birds in which distribution and geographical variation are incompletely known.

Cladists applying concepts PSC 1 and PSC 2 interpret the capability to hybridize as a plesiomorphic (primitive) character and therefore phylogenetically uninformative. They are concerned that paraphyletic and polyphyletic groups may be ranked as species, if wide species limits are applied without prior cladistic analyses having been carried out. It is not possible to judge how serious this problem is as long as the numbers of paraphyletic and monophyletic biospecies of birds remain unknown. Snow (1997: 116) gained the strong impression that the great majority of polytypic biospecies in Eurasia would be shown by detailed cladistic analyses to be monophyletic. However, there are certainly also numerous biological species taxa in this and other regions of the world that are paraphyletic, as discussed below.

A biological species taxon becomes paraphyletic when a daughter species is originated through “budding” (Fig. 2); for example, a derivative population of a widespread mainland species may have reached species status on a nearby island. However, this speciation event had no effect on the parental biospecies (no. 3, Fig. 2) on the mainland from which neospecies 4 has budded off. The mainland species (no. 3) is real in the sense that it represents a biological unit characterized by close genetic-reproductive and ecological relations among its component subspecies taxa, even though it represents a “non-historical” (paraphyletic) group from a cladistic point of view. Traditionally, such biological clusters have been designated as “species”. Cladists who favour the narrow version (PSC 1) of the phylogenetic species concept would consider each of the 9 lineages illustrated in Fig. 2 as “species”, regardless of their forming 4 separate clusters through genetic cohesion and intergradation. Cladists applying the monophyly version (PSC 2) of the phylogenetic species concept would recognize monophyletic species 1 and 2 but would dissolve paraphyletic species 3 into three separate species. I emphasize that the cladistic analyses schematically illustrated in Fig. 2 (if feasible at that infraspecific level) yield relevant phylogenetic (“vertical”) and biogeographical data on the origin and differentiation of the various groups and their component taxa. Classification of the taxa as subspecies or species is a separate matter.

It would be an important task to analyse the bird species of the world from a cladistic point of view and to determine approximately what percentage of bird species are monophyletic entities and how many biospecies are paraphyletic or polyphyletic taxa. It is true that some authors went too far when, early during this century, they combined morphospecies as subspecies into broadly delimited biological (multidimensional) species taxa. This is now being remedied by detailed analyses of such broadly delimited biospecies. However, I would discourage the procedure followed by some modern authors (e.g. Hazevoet 1995), who, on the basis of PSC 1, fall to the opposite extreme and elevate many subspecies on islands and on continents to species status without any cladistic or evolutionary analysis and discussion of grouping and ranking criteria for the isolated basal taxa and monophyletic groups. Moreover, the decision as to which subspecies are elevated to species status is determined subjectively, as explained above. This would be a return to the situation over 100 years ago when many ornithologists applied species names to any population that was morphologically “different”.

It is obvious from this discussion that “phylogenetic species” as lineages have been delimited and are being delimited taxonomically following standards that are as different as (or even more different than) under the other main species concepts. Not one but several theoretical notions are subsumed under the name of the “phylogenetic species concept”.

Three examples of tropical birds

Within each of the three groups of birds discussed below, ornithologists recognize a different number of species depending on the particular species concept they apply. The birds in each group remain the same but their classification differs according to the theoretical ideas of ornithologists as to what constitutes species and subspecies. I have chosen for discussion groups of birds whose distribution and geographical variation are fairly well known and that are composed of rather well marked forms, in order to illustrate clearly the differences in interpretation under different species concepts. Application of the narrow PSC 1 to many poorly known groups of tropical birds would be much more difficult than in the examples chosen.

1)   Ramphastos toucans in Amazonia (Fig. 3).

The plumage of these toucans is mainly black, while the uppertail-coverts are red, yellow or white; other differences among the various forms refer to the colour of the throat and breast, the tail, the iris and the bill. Two distinct and widely sympatric species groups form mosaic distribution patterns over almost the entire Neotropical Region. These assemblages are composed of: (a) large, smooth-billed species with yelping calls (R. tucanus group); and (b) medium-sized channel- or keel-billed species which have croaking vocalizations (R. vitellinus group). The larger species inhabit the canopy level of rain forest, whereas the smaller toucans live mostly at middle levels in the forest. The latter are also more insectivorous than their larger relatives, and are seen occasionally at ant raids near the forest floor. Some of the geographical representatives of each group hybridize along their zones of contact, whereas others have reached the species level replacing each other geographically without hybridization. The latter species are probably still too similar ecologically to be able to co-exist in the same forest. Each of the two groups of Ramphastos toucans is represented in upper Amazonia by one form (cuvieri and culminatus, respectively), the smaller toucans by two in lower Amazonia (vitellinus and ariel) and the larger ones by one form only (tucanus). Similarly, in the rain forests west of the Andes there are more forms of the smaller group (three) than of the larger toucans (two). Ramphastos toco, a large species with a keeled bill that inhabits mosaics of forest and savanna, is not included in this discussion.

The phylogenetic relations of the taxa included in the R. vitellinus group indicate that an early west-east separation of upper and lower Amazonian forms preceded the later differentiation of extant forms citreolaemus/culminatus in the west and vitellinus/ariel in lower Amazonia to the east. R. dicolorus and the two trans-Andean species originated from stocks that separated prior to the differentiation of the Amazonian forms. The upper and lower Amazonian forms of the large smooth-billed species R. tucanus also document a separation of a western from an eastern population in Amazonia that later re-established contact forming an extensive hybrid belt similar to that found in the smaller channel-billed toucans. The eastern Amazonian forms vitellinus and ariel are separated from each other by the broad lower Amazon River. They are sister taxa and would certainly hybridize with each other if they could meet without an intervening barrier zone (both hybridize with the western Amazonian culminatus). All the Amazonian forms represent historical entities that probably originated in formerly isolated “refuge” areas during periods of altered distribution of forest and non-forest vegetation, subsequently establishing zones of secondary contact during expansive phases (Haffer 1974).

All Amazonian representatives of both groups would be ranked as species under the diagnosable version (PSC 1) of the phylogenetic species concept, whereas they are combined as subspecies of one species of yelping toucans and one species of grunting toucans under the biological species concept, because they hybridize extensively or would do so if in direct contact: thus cuvieri and tucanus are combined as biospecies R. tucanus; similarly the forms vitellinus, ariel and culminatus, plus the trans-Andean citreolaemus, form biospecies R. vitellinus. The cladogram of R. vitellinus (Fig. 3) indicates that this biospecies is monophyletic (Haffer 1985, Prum 1988). The trans-Andean forms sulfuratus and brevis do not hybridize where they meet in north-western Colombia, and they probably began to differentiate somewhat earlier than the Amazonian subspecies of R. vitellinus. This last species is also parapatric with R. sulfuratus. The latter species and brevis are considered as specifically distinct under both versions of the phylogenetic species concept as well as under the biological species concept and the Hennigian concept. The taxonomic classification of basal taxa under the BSC emphasizes the distinction between hybridizing (or potentially hybridizing) taxa (= subspecies) and non-hybridizing sympatric or parapatric taxa (= species). The classification of the Amazonian forms under the monophyly version of the PSC would coincide with that under the BSC, because biospecies R. vitellinus and R. tucanus are monophyletic. On the other hand, the distinction between hybridizing and non-hybridizing forms is dismissed under PSC 1, and all forms are ranked as species. Under this concept, broadly hybridizing forms (e.g. culminatus/vitellinus and cuvieri/tucanus) that treat each other in life as biologically “the same unit” are separated as “species” like two widely sympatric and unrelated toucans. Moreover, the large Ramphastos populations forming the vast hybrid belts in Amazonia can not be assigned to particular species taxa under the PSC 1, whereas under the BSC and PSC 2 they belong to R. tucanus and R. vitellinus, respectively. The ‘hybrid’ populations consists exclusively of intermediate and highly variable (hybrid) individuals; parental phenotypes are lacking in the broad central portions of these hybrid zones.

It is true that, under the BSC, the often conspicuous geographical and historical variation at the subspecific level is “concealed” at the level of polytypic species if only the species names are considered (e.g. R. vitellinus). However, any zoogeographical analysis would be incomplete without an analysis also of well characterized historical units at the subspecies level. The ongoing process of speciation predicts that there are species with inconspicuous internal differentiation (monotypic) and others that show increasingly conspicuous intraspecific differentiation (polytypic).

 2)  Amazonian manakins of the Pipra serena group (Fig. 4).

This group of small forest manakins comprises several sharply differentiated representatives which, in many areas, are among the commonest forest birds (Haffer 1970). The males of serena and coronota north of the Amazon River (and west of the Andes) are mainly black, while those of the southern Amazonian forms are extensively green with a yellow belly; the cap of the males is, respectively, blue or white. The females of the various forms are generally green and very similar to each other. P. serena inhabits the lowlands to the north of the lower Amazon and the hilly interior of the Guianas. In southern Venezuela, P. (serena) suavissima is found in montane forests above 500 m in some regions which lie above lowland forests that are inhabited by the western Amazonian P. coronata, with its blue cap in males. At some stage, this latter form crossed the upper Amazon River southward and here hybridized extensively with the P. c. exquisita group of south-western Amazonia.

Geographically representative species of the serena group in south-eastern Amazonia are P. nattereri, with snow white cap and rump in males, and P. iris, with a glistening opalescent cap and green rump. P. nattereri occurs between the lower Rio Madeira and Rio Tapajós and is widespread in south central Amazonia. It has crossed the narrow upper Rio Tapajós and Xingú in an eastward direction establishing contact with P. iris in this general region. The latter species inhabits the area east of the lower Tapajós and most of the Rio Xingú valley east to the mouth of the Amazon River. The male plumage colour of the exquisita group in south-western Amazonia, with a blue cap (like coronata) and a green body with yellow belly (like nattereri-iris), is somewhat intermediate between coronata and the south-eastern Amazonian species. Because it hybridizes extensively with coronata, the exquisita group is included as a subspecies group in P. coronata under the BSC. The Rio Madeira separates the exquisita group from P. nattereri, and it remains unknown whether they would also hybridize. Should future genetic analyses reveal closer phylogenetic relationships between exquisita/nattereri than between exquisita/coronata, this would prove biospecies P. coronata (including the exquisita group) to be paraphyletic, unless this species is broadened to include nattereri and iris too. It is also still unknown whether serena and suavissima hybridize in the area of western Surinam where they may meet.

All representative forms of the Pipra serena group are historical (phylogenetic) entities and represent phylospecies under the diagnosable version of the PSC. Unless biospecies P. coronata turns out to be paraphyletic, application of the monophyly version of the PSC would result in similar species limits as those discussed above under the BSC.

3)  Kingfishers of the Tanysiptera galatea group in the New Guinea region (Fig. 5).

Paradise kingfishers are blue above and white below with long, streamered central tail feathers. The Common Paradise Kingfisher (T. galatea) inhabits lowland forests of mainland New Guinea in three barely distinguishable subspecies which may actually be in contact and may broadly intergrade with each other. On the other hand, the island populations off the coast of New Guinea are quite distinct. T. hydrocharis on the Aru Islands and in southern New Guinea is certainly differentiated as a species, because in the latter area it lives side by side with T. galatea minor (3). Some of the other island forms have probably also reached species level (populations 4-7); but only T. carolinae on Numfor Island (8) is treated as a species by all recent authors. Mainland T. galatea budded off these various daughter species. However, the daughter populations on the islands are not derived from the mainland species as a whole, but it is always a single local population of the widespread species on the mainland that produced a given founder population (Mayr 1987: 312). A detailed cladistic study relating each island population to a particular subspecies on the mainland is not yet available. However, it is clear that polytypic T. galatea on mainland New Guinea is paraphyletic (Fig. 2 B) and, according to cladistic principles, would have to be split up into a number of separate “species” that form the “stem species” of the five island populations, if the latter are indeed considered as species. Yet in reality T. galatea as a whole represents one widespread biological entity (biospecies) on mainland New Guinea and shows only weak geographical variation.

The changing numbers of bird species

Because of the different opinions among ornithologists and other biologists as to the taxonomic delimitation of species taxa, a higher or lower number of bird species has been recognized at all times. These different counts refer to the birds of the world as a whole, of a large continental region or an island archipelago, though not to the number of bird species of a small area or at a single locality which, of course, coincides under the various species concepts discussed. Ornithologists delimiting species taxa narrowly arrive at high numbers of species which, during the last century, were rapidly increasing due to the continuous discovery and description of new forms made known through the work of numerous scientific expeditions (Fig. 6). The leading museum ornithologists in Europe applied narrow species limits toward the end of the nineteenth century, resulting in the recognition of high numbers of species taxa, mainly through the influence of the authoritative Catalogue of Birds in the British Museum (27 volumes, 1874-1898). This trend culminated when R. B. Sharpe published his A Hand-list of the Genera and Species of Birds (1899-1909) recognizing 18,939 species, many of which represent allospecies and subspecies.

During the following 20 years, the situation reversed itself entirely. Numerous Linnaean morphospecies were reinterpreted as subspecies and combined in more widely conceived biological species taxa. The result was a precipitous decline in the number of species recognized (Fig. 6). A period of moderate stability regarding species numbers followed during the 1930’s and early 1940’s, when Mayr (1946: 68) estimated the total number of known birds to be 8616 species. A gentle increase of species numbers began during the late 1940’s when many geographically isolated representatives formerly considered subspecies were reinterpreted as species and combined in superspecies. This “quiet revolution” (Mayr 1980) at the microtaxonomic level during the last 30-40 years led to a continuous increase in the number of bird species, only slightly boosted by the discovery of genuine new biospecies (153 species in the period 1938-1985; Vuilleumier & Mayr 1987). Bock & Farrand (1980) counted a world total of 9021 species and Sibley & Monroe (1990) 9672 species. In the latter species list, superspecies are indicated to give a measure of ecological units in the world’s avifauna.

The world total would go back up to c. 20,000 bird species if the narrow diagnosable version (PSC 1) of the phylogenetic species concept were applied to the avifauna of the world, with many subspecies being ranked as “species”. However, in view of the wide variety of species concepts proposed during the last 15 years (Claridge et al. 1997; Wheeler & Meier 1997), it would be premature and inadvisable if the current stability and agreement in delimiting bird species under the biological species concept were upset at this time of rapid development of taxonomic methods and thinking, and a narrow version of the phylogenetic species concept (PSC 1) generally applied. The biological species concept, while not perfect, is useful and meaningful. This is probably the main reason why the authors of almost all recent ornithological textbooks and all regional handbooks have preferred the traditional concept of biological species as the basis for discussing the avifaunal diversity of the world (Sibley & Monroe 1990, del Hoyo et al. 1992 ff.) or of individual continents (Cramp & Simmons 1977 ff.; Brown et al. 1982 ff.; Ridgely & Tudor 1989 ff.; Marchant & Higgins 1990 ff.).

The biological species concept continues to form the basis of global conservation biology (Collar 1996, 1997), the main problem with the narrow version of the phylogenetic species concept being the limit of “diagnosability”. Stotz et al. (1996: 118) also used the biological species concept as the basis of their work on the ecology and conservation of Neotropical birds stating that “we can more effectively discern patterns useful in guiding conservation action by focusing on biological species.” Many conservation biologists analyse “evolutionary significant units” and judge their need for protection regardless of whether these units are considered as species or subspecies under different species concepts. Views in conservation biology as to which taxa need to be protected are independent of the particular species concept applied by systematists. Therefore, taxa do not have to be recognized as species in order to be conserved (Bock MS).

An ideal species concept for all groups of organisms that is at the same time general, applicable and theoretically significant may be unattainable. As Hull (1997) remarked, theoretical significant species concepts tend not to be very operational. Attempts to make them more operational result in their being theoretically less significant. I emphasize that many levels of differentiation at which species limits have been proposed are biologically significant. It will be advisable, therefore, that these stages of increased microtaxonomic differentiation continue to be analysed in depth. In this way, the conceptual relations among the various taxonomic categories and their component taxa may be studied, and the various entities may be used in analyses of the biogeographical and phylogenetic history, as well as the ecological divergence, of genera and families of birds.


Note: In practical terms, the BSC remains the most efficient tool at present in most fields of ornithological research, and in the same way the organization of the species recognized equally requires a good degree of stability, which leads on to another controversial topic of current systematic ornithology, namely the distinction between classifications and sequences (Mayr & Bock 1994; Mayr 1997; Bock 1994, In press). Provisional classifications of genera, families and orders of birds are proposed by specialists on the basis of their analyses of the phylogenetic relationships among such macrosystematic units. Even well established classifications are subject to eventual change. A standard sequence such as Peters’ Check-list of Birds of the World (1931-1987) is derived from widely accepted classifications, and represents a consensus in the linear arrangement of taxa to permit optimal communication among avian biologists. A standard sequence should be followed by authors and editors of ornithological textbooks and faunal lists until enough new data about avian relationships have been gathered to permit agreement on a revised standard sequence. As valuable as Sibley & Ahlquist’s (1990) and Sibley & Monroe’s (1990) work on the phylogeny, classification and taxonomy of birds of the world has unquestionably been, their classifications and sequences are provisional and full of uncertainties. They should not be considered as the basis of a new standard sequence, except with respect to certain well established shifts that might be made now, though even such minor modifications of the standard sequence might not be wise (Mayr & Bock 1994).


Acknowledgement: I thank Dr H. Pieper (Kiel, Germany) and Professor W. Bock (New York) for discussions.

Jürgen Haffer
Research Associate,
Section of Biology and Phylogeny of Tropical Birds,
A. Koenig Zoological Research Institute and Zoological Museum,
Adenauer-Allee 150, D-53113 Bonn,


References Cited  

Amadon, D. (1966). The superspecies concept. Syst. Zool. 15: 245-249.
Amadon, D. & Short, L.L. (1992). Taxonomy of lower categories – suggested guidelines. Pp. 11-38 in: Monk, J.F. ed. (1992). Avian Systematics and Taxonomy. Bulletin of the British Ornithologists’ Club. Centenary Supplement 112A.
Avise, J.C. & Ball, R.M. (1990). Principles of genealogical concordance in species concepts and biological taxonomy. Oxford Surveys in Evol. Biol. 7: 45-67.
Bock, W.J. (1979). The synthetic explanation of macroevolutionary change – a reductionist approach. Bull. Carnegie Mus. Nat. Hist. 13: 20-69.
Bock, W.J. (1986). Species concepts, speciation, and macroevolution. Pp. 31-57 in: Iwatsuki, K., Raven, P.H. & Bock, W.J. eds. (1986). Modern Aspects of Species. University of Tokyo Press, Tokyo.
Bock, W.J. (1992). The species concept in theory and practice. Zool. Science (Zool. Soc. Japan) 9: 697-712.
Bock, W.J. (1994). Foreword. Pp. 13-15 in: del Hoyo, J., Elliott, A. & Sargatal, J. eds. (1994). Handbook of the Birds of the World. Vol. 2. New World Vultures to Guineafowl. Lynx Edicions, Barcelona.
Bock, W.J. (1995). The species concept versus the species taxon: their roles in biodiversity analyses and conservation. Pp. 47-72 in: Arai, R., Kato, M. & Doi, Y. eds. (1995). Biodiversity and Evolution. The National Science Museum Foundation, Tokyo.
Bock, W.J. (In press). Heuristics in systematics. In: Proceedings of the IX Pan-African Ornithological Congress. In press.
Bock, W.J. & Farrand, J. (1980). The number of species and genera of Recent birds. Amer. Mus. Novit. 2703.
Brown, L.H., Urban, E.K. & Newman, K. eds. (1982 ff.). The Birds of Africa. Academic Press, London & New York.
Claridge, M.F., Dawah, H.A. & Wilson, M.R. eds. (1997). Species. The Units of Biodiversity. Chapman & Hall, London.
Collar, N.J. (1996). Species concepts and conservation. Bird Conserv. Int. 6: 197-200.
Collar, N.J. (1997). Taxonomy and conservation: chicken and egg. Bull. Brit. Orn. Club 117: 122-136.
Cracraft, J. (1983). Species concepts and speciation analysis. Curr. Orn. 1: 159-187.
Cramp, S. & Simmons, K.E.L. (1977 ff.) The Birds of the Western Palearctic. Oxford University Press, Oxford.
Fry, C.H., Fry, K.F. & Harris, A. (1992). Kingfishers, Bee-eaters and Rollers. Christopher Helm, London.
Grant, P.R. (1986). Ecology and Evolution of Darwin’s Finches. Princeton University Press, Princeton, New Jersey.
Grant, P.R. (1993). Hybridization of Darwin’s finches on Isla Daphne Major, Galápagos. Phil. Trans. Roy. Soc. London 340: 127-139.
Grant, P.R. & Grant, B.R. (1992). Hybridization of bird species. Science 256: 193-197.
Haffer, J. (1970). Art-Entstehung bei einigen Waldvögeln Amazoniens. J. Orn. 111: 285-331.
Haffer, J. (1974). Avian Speciation in Tropical South America. Publications of the Nuttall Ornithological Club 14, Cambridge, Massachusetts. 390 pp.
Haffer, J. (1985). Avian zoogeography of the neotropical lowlands. Orn. Monogr. 36: 113-146.
Haffer, J. (1986). Superspecies and species limits in vertebrates. Z. Zool. Syst. Evolutionsforsch. 24: 169-190.
Haffer, J. (1992). The history of species concepts and species limits in ornithology. Pp. 107-158 in: Monk, J.F. ed. (1992). Avian Systematics and Taxonomy. Bulletin of the British Ornithologists’ Club. Centenary Supplement 112A.
Hazevoet, C.J. (1995). The Birds of the Cape Verde Islands. An Annotated Cheklist. BOU Check-list 13. British Ornithologists’ Union, Tring, UK.
Hennig, W. (1966). Phylogenetic Systematics. University of Illinois Press, Urbana, Illinois.
del Hoyo, J., Elliott, A. & Sargatal, J. eds. (1992 ff.). Handbook of the Birds of the World. Lynx Edicions, Barcelona.
Hull, D.L. (1997). The ideal species concept – and why we can’t get it. Pp. 357-380 in: Claridge et al. (1997).
Lack, D. (1944). Ecological aspects of species-formation in passerine birds. Ibis 86: 260-286.
Mallet, J. (1995). A species definition for the modern synthesis. Trends Ecol. Evol. 10: 294-299.
Marchant, S. & Higgins, P.J. eds. (1990 ff.). Handbook of Australian, New Zealand & Antarctic Birds. Oxford University Press, Melbourne.
Mayden, R.L. (1997). A hierarchy of species concepts: the denouement in the saga of the species problem. Pp. 381-424 in: Claridge et al. (1997).
Mayr, E. (1942). Systematics and the Origin of Species from the Viewpoint of a Zoologist. Columbia University Press, New York.
Mayr, E. (1946). The number of species of birds. Auk 63: 64-69.
Mayr, E. (1963). Animal Species and Evolution. Harvard University Press, Cambridge, Massachusetts.
Mayr, E. (1968). Illiger and the biological species concept. J. Hist. Biol. 1: 163-178.
Mayr, E. (1969). Principles of Systematic Zoology. McGraw-Hill, New York.
Mayr, E. (1970). Populations, Species, and Evolution. Harvard University Press, Cambridge, Massachusetts.
Mayr, E. (1980). Problems of the classification of birds, a progress report. Pp. 95-112 in: Nohring, R. ed. (1980). Acta XVII Congr. Int. Orn. Berlin, 1978. Verlag der Deutschen Ornithologen-gesellschaft, Berlin.
Mayr, E. (1982). The Growth of Biological Thought. Diversity, Evolution, and Inheritance. Harvard University Press, Cambridge, Massachusetts.
Mayr, E. (1987). The species as category, taxon and population. Pp. 303-320 in: Histoire du Concept d’Espèce dans les Sciences de la Vie. Colloque International, Mai 1985, Organisé par la Fondation Singer-Polignac, Paris.
Mayr, E. (1996). What is a species and what is not? Philosophy of Science 63: 261-276.
Mayr, E. (1997). This Is Biology. The Science of the Living World. Harvard University Press, Cambridge, Massachusetts.
Mayr, E. & Ashlock, P.D. (1991). Principles of Systematic Zoology. 2nd edition. McGraw Hill, New York.
Mayr, E. & Bock, W.J. (1994). Provisional classifications versus standard avian sequences: heuristics and communication in ornithology. Ibis 136: 12-18.
Mishler, B.D. & Brandon, R.N. (1987). Individuality, pluralism, and the phylogenetic species concept. Biol. Philos. 2: 397-414.
Nelson, G. & Platnick, N. (1981). Systematics and Biogeography. Cladistics and Vicariance. Columbia University Press, New York.
Paterson, H.E.H. (1985). The recognition concept of species. Pp. 21-29 in: Vrba, E.S. ed. (1985). Species and Speciation. Transvaal Museum Monographs 4, Pretoria.
Peters, D.S. (1970). Über den Zusammenhang von biologischem Artbegriff und phylogenetischer Systematik. Aufsätze und Reden der Senckenberg. Naturf. Gesellschaft 18. 39 pp.
Prum, R.O. (1988). Historical relationships among avian forest areas of endemism in the Neotropics. Pp. 2562-2572 in: Acta XIX Int. Orn. Congr. Ottawa, 1986. Vol. 2.
Ridgely, R.S. & Tudor, G. (1989 ff.). The Birds of South America. University of Texas Press, Austin & Oxford University Press, Oxford.
Short, L.L. (1969). Taxonomic aspects of avian hybridization. Auk 86: 84-105.
Short, L.L. (1972). Hybridization, taxonomy and avian evolution. Ann. Missouri Bot. Garden 59: 447-453.
Sibley, C.G. & Ahlquist, J.E. (1990). Phylogeny and Classification of Birds: a Study in Molecular Evolution. Yale University Press, New Haven & London.
Sibley, C.G. & Monroe, B.L. (1990). Distribution and Taxonomy of Birds of the World. Yale University Press, New Haven & London.
Simpson, G.G. (1951). The species concept. Syst. Zool. 5: 285-298.
Simpson, G.G. (1961). Principles of Animal Taxonomy. Columbia University Press, New York.
Snow, D.W. (1997). Should the biological be supersed by the phylogenetic species concept? Bull. Brit. Orn. Club 117: 110-121.
Stotz, D.F., Fitzpatrick, J.W., Parker, T.A. & Moskovits, D.K. eds. (1996). Neotropical Birds, Ecology and Conservation. University of Chicago Press, Chicago, Illinois.
Szalay, F.S. & Bock, W.J. (1991). Evolutionary theory and systematics: relationships between process and patterns. Z. Zool. Syst. Evolutionsforsch. 29: 1-39.
Vuilleumier, F. & Mayr, E. (1987). New species of birds described from 1976 to 1980. J. Orn. 128: 137-150.
Wheeler, Q.D. & Meier, R. eds. (1997). Species Concepts and Phylogenetic Theory: a Debate. Columbia University Press, New York. In press.
Wiley, E.O. (1981). Phylogenetics, the Theory and Practice of Phylogenetic Systematics. J. Wiley & Sons, New York.
Willmann, R. (1985). Die Art in Raum und Zeit. Parey, Hamburg.
Zink, R.M. (1996). Species concepts, speciation, and sexual selection. J. Avian Biol. 27: 1-6.
Zink, R.M. (1997). Species concepts. Bull. Brit. Orn. Club 117: 97-109.
Zink, R.M. & McKitrick, M.C. (1995). The debate about species concepts and its implications for ornithology. Auk 112: 701-719.