187© 2010 ISZS, Blackwell Publishing and IOZ/CAS

Integrative Zoology 2010; 5: 187-197 doi: 10.1111/j.1749-4877.2010.00204.x

REVIEW

Species from Darwin onward

Anouk BARBEROUSSE1 and Sarah SAMADI2

1Institute of History and Phylosophy of Sciences and Technics UMR 8590, National Center for Scientific Research, University

Paris I, ENS and 2Systematic, Adaptation and Evolution, UMR 7138 CNRS-IRD-UMPC-MNHN, France

Abstract

Controversy regarding the species problem has been going on for many decades and no consensus has ever been

reached about what a “species” really is and how best to define the concept. De Queiroz (1998) introduced a

distinction between two aspects of this problem: on the one hand, the definition proper, and on the other, the criteria

allowing biologists to recognize species in practice. This distinction is a first step on the way toward a solution of

the problem. In the present paper, we show that de Queiroz’s distinction is made possible by the radical theoretical

change introduced by Darwin. We emphasize that the species problem did not appear in the 20th century, but long

before, and that Darwin addresses it indirectly in the Origin of Species. It might seem paradoxical to refer to

Darwin’s views about species, because they are usually considered as unclear. However, we propose that an

analysis of these views in the context of Darwin’s own theory of evolution might reveal how a definition of the

concept of species is made possible by being anchored to the very theory of evolution. To this aim, we present a

plausible reconstruction of Darwin’s implicit conception of species and show how this conception fits with the

debates on species that took place in the 18th and 19th centuries. We then turn to today’s biology and show what

changes Darwin’s implicit conception of species has brought about relative to the species concept and species

delimitation.

Key words: Buffon, Darwin, nomenclature, species concepts, taxonomy.

Correspondence: Sarah Samadi, Muséum National d’Histoire

Naturelle (USM 603–UMR7138), Département Systématique et

Evolution, Case postale 26, 57 Rue Cuvier, F-75231 PARIS

CEDEX 05, France.

Email: sarah@mnhn.fr

INTRODUCTION

It is often claimed that, despite its title, the Origin of

Species is not a book about species. Several passages

create the impression that Darwin’s opinion about spe-

cies was not univocal. The book’s subtitle reveals that

Darwin’s aim in the Origin of Species by Means of Natu-

ral Selection, or Preservation of Favoured Races in the

Struggle for Life was to propose a mechanism explaining

(the origin of) biological diversity that human beings com-

monly analyze as being composed of discrete units or

species. Darwin was not so much interested in the de-

tailed narrative of the history of biodiversity or in how to

make an inventory of species and varieties as in the gen-

eral explanation of the evolution of biodiversity. He mainly

focused on the processes that are causally responsible

for evolution. This is one of the reasons why he did not

directly enter the debate over the species concept that

was flourishing at his time:

Nor shall I here discuss the various definitions which

have been given of the term species. No one definition

has as yet satisfied all naturalists; yet every naturalist

knows vaguely what he means when he speaks of a

188 © 2010 ISZS, Blackwell Publishing and IOZ/CAS

species. Generally the term includes the unknown ele-

ment of a distinct act of creation (Darwin 1859, p. 44).

As a matter of fact, Darwin appears less interested in

how naturalists recognize species in practice and define

them, than in why there are species at all. This was not a

core question in the 18th and 19th century debates, which

focused mainly on criteria of species recognition, and not

on the origin of species. Species were most of the time

considered as independent groups, which had appeared

as a result of “special acts of creation”.

In these debates over species, definitions of the

concept, stricto sensu, were often mixed up with lists of

recognition criteria, as is the case today. Whereas Darwin

drastically changed the theoretical framework in which the

question of the definition of the species concept could be

raised and showed that a whole set of former possibilities

had to be eliminated unless one rejected his new theoreti-

cal framework, the basic structure of the debate lived on.

As Darwin never gave the word “species” a new, explicit

meaning, the “species problem” continued to thrive in

biology: indeed, many definitions competed against one

another during the second half of the 20th century (for

examples of the innumerable papers and books on the “spe-

cies problem” (see Otte & Endler 1989; Ereshefsky 1992;

Kimbel & Martin 1993; Claridge et al. 1997; Mayden 1997;

Howard & Berlocher 1998; Wilson 1999; Winston 1999;

Wheeler & Meier 2000; Hey 2001a,b; Mallet 2001a; Wu

2001 and replies; Noor 2002; volume 16(7) of the journal

Trends in Ecology and Evolution; Agapow et al. 2004).

Since the late 1990s, however, a fragile consensus has

emerged, at least when it comes to the necessary separa-

tion between the theoretical definition of the species con-

cept and the listing of practical criteria of species recogni-

tion (Frost & Kluge 1994; De Queiroz 1998, 1999, 2005a,b,

2007). We emphasize that this consensus was made pos-

sible by the theoretical change Darwin introduced: his

theory, as it is developed today, allows for a definition of

the species concept that is theoretically well-grounded.

Such a definition was beyond the domain of scientific

possibility before his work. Several recent articles present

the species problem as solved or “dissolved” (Pigluicci

2003) on the basis of the distinction introduced by De

Queiroz (e.g. Sites & Marshall 2004; Hey 2006; Samadi &

Barberouse 2006; Knowles & Carstens 2007; Rissler &

Apodaca 2007; Giraud et al. 2008). However, the intellec-

tual process through which this important distinction can

be taken into account within the reflection on the species

concept is a rather slow one: for example, Hey (2001b)

focuses on: “our own role, as conflicted investigators, in

causing the problem,” whereas Hey (2006) emphasizes the

common evolutionary idea underlying the multiple spe-

cies “concepts,” which he analyzes as recognition criteria.

To show how a definition of the species concept can be

grounded in the theory of evolution, we first present a

plausible reconstruction of Darwin’s implicit conception

of species. This is best achieved by expounding the 18th

and 19th nineteenth century debates over species, because

Darwin’s theory aimed at solving the problems raised

within these debates. After having made Darwin’s own

conception explicit, we then turn to today’s biology and

show what changes it has brought about relative to the

species concept and species delimitation.

PRE-DARWINIAN CONCEPTIONS OF

SPECIES AND THE CONTEMPORARY

“SPECIES PROBLEM”

In this section, we present some of the themes that

were debated in Darwin’s time, in order to better grasp the

originality of his own conception. We first focus on fixism

about species and associated doctrines to make explicit

the variety of doctrines that could be opposed to his

transformism. We then set out the different kinds of

transformism that have been suggested.

Varieties of fixist conceptions of species

It is a locus communis of the usual statements of

Darwin’s transformism that it opposes fixism; namely, the

doctrine that the most important properties of the mem-

bers of any species, those properties allowing for their

classification as members of this species, cannot vary

beyond definite limits. Transformism and fixism are con-

tradictory positions. However, focusing on fixism might

lead one to miss other important associated doctrines

against which Darwin fought.

Fixism about species was usually related to creation-

ism (the doctrine according to which species are the prod-

ucts of an external act of creation) or essentialism (the

doctrine according to which species are defined by forms,

types or essences) in a complex web of often implicit

assumptions. However, fixism does not strictly imply cre-

ationism because it is possible to believe that fixed spe-

cies appeared by spontaneous generation. However, cre-

ationism does not imply fixism either: for instance, Linné

(1744) imagined a narrative of a historical creation of the

present world and its inhabitants by descent from a few

original forms that had been created by divine action,

where species arise by hybridization of the original forms.

Darwin’s explicit opponent relative to the conception

A. Barberousse and S. samadi

189© 2010 ISZS, Blackwell Publishing and IOZ/CAS

of species was what he calls “special creationism,” the

doctrine that each species arose due to a special act of

creation. In the Origin of Species, he addresses directly

the issue of whether it is necessary to invoke a Creator to

explain the immediate origin of species. He concludes that

such explanations are vacuous: “It is so easy to hide our

ignorance under such expressions as ‘plan of creation,’

‘unity of design,’ &c., and to think we give an explanation

when we only restate a fact” (Darwin 1859, p. 482). The

doctrine of special creationism was largely accepted among

his contemporaries, who did not hesitate to blend scien-

tific and theological considerations. Most of them were

convinced of the truth of Humboldt’s claim that: “these

are among the mysteries which natural science cannot

reach” (quoted in Lyell 1830, ch. 11). However, some with-

stood this attitude and looked for a genuine scientific ap-

proach to investigating the phenomena of life in general

and the origin of species in particular, as we shall see

below.

Even though creationism does not strictly imply fixism,

they are most often associated. Fixism came under two

different varieties. According to the first, the Linnaean va-

riety of fixism, species (namely, definite taxa) are defined

by sets of properties, so that Canis lupus, for instance, is

a universal or a class concept, constituted by individuals

on the basis of possession of defining properties. The

possession of these defining properties is the rational as

well as empirical basis of classification, as developed in

Linné’s system.

The second variety of fixism may also be called

“essentialism”: it is the doctrine that species possess an

eternal, abstract essence that is also referred to as “form,

” “archetype,” “formal principle” and “internal mold.” This

postulated essence or type is claimed to be responsible

for likeness among members of the same species and for

the possibility of interbreeding within a species. The sec-

ond variety of fixism has been vigorously criticized by

Mayr (1964) under the term “typological thinking,” to which

opposes the “populational thinking” introduced by

Darwin. According to “typological thinking,” the type (or

form) comes first, and the properties of individual organ-

isms are dependent upon it. According to populational

thinking, by contrast, the individual phenotypes come first.

Whereas Linné is the best known representative of the

first variety of fixism, Buffon developed a sophisticated

version of the second variety. His conception of species

exerted a profound influence on his posterity, but was

unfortunately often misunderstood because of its subtlety.

Buffon claims that the so-called “species” that are de-

fined and identified according to Linné’s definition are

“artificial.” “Real” (or “natural”) species are defined by

the “internal mold” that is passed on from parent to

offspring. According to Buffon, a “real” species is the

historical succession of ancestor and descendant linked

by material connection through generation. This material

connection causally guarantees the unity of the species

because it contains an immanent formal principle (the “in-

ternal mold”), so that what counts from an explanatory

point of view is not actually the materiality of the

connection, but the form that is materialized in the gem-

mules that are transmitted through reproduction.

We should regard two animals as belonging to the same

species if, by means of copulation, they can perpetuate

themselves and preserve the likeness of the species, and

we should regard them as belonging to different species

if they are incapable of producing progeny by the same

means (Buffon 1749, p. 10).

As is explicit in this quote, for Buffon the empirical sign

of the unity of species over time is fertile interbreeding.

This criterion takes precedence over similarity of anatomy

or habits of life; namely, over the properties defining spe-

cies according to Linné, and often used to delimit species

taxa in practice.

Buffon’s subtle conception of species is a nice illustra-

tion of two often-made confusions. First, Buffon clearly

distinguished between fixism of the Linnean variety and

essentialism; second, he was aware that defining the spe-

cies concept and attributing species membership in prac-

tice are two conceptually different tasks. He acknowledged

that the morphological properties put forward by Linné

may help one in a classification task, even though he con-

sidered interfecundity as a more reliable sign or criterion.

Buffon even elaborated a well-thought justification of the

use of his privileged criterion for species recognition. For

example, according to Buffon, interfecundity is the best

criterion because the unity (or likeness) of a species is

causally produced by the genealogical link. As we shall

see, Buffon’s awareness of the conceptual difference

between the definition of the species concept and the

listing of practical recognition criteria has not been faith-

fully transmitted to his successors. Even nowadays, the

crucial importance of this difference is not always fully

acknowledged.

Varieties of transformism

As is well-known, Darwin defended transformism, the

view that species change over time. We shall discuss be-

low what exactly he opposed to the two varieties of fixism

we identified in the preceding section. Before doing so,

Species from Darwin onward

190 © 2010 ISZS, Blackwell Publishing and IOZ/CAS

we have to recall that even though transformism was rather

unpopular before 1859 (and even shortly after), a few other

naturalists were opposing fixism; namely, Wallace, Cham-

bers (author of the Vestiges of the Natural History of Cre-

ation 1844, published anonymously) and Lamarck. Accord-

ing to Lamarck, the origin of living beings is initially

through spontaneous generation, which is confined to

the origin of the most structurally simple forms of life

(infusoria). All subsequent forms have developed in time

from these elementary beginnings. The primary cause of

transformation Lamarck sees in the dynamical (vitalist)

properties of matter, especially of caloric and electricity.

Inheritance of acquired characters appears in Lamarck

(1809) as a subordinate diversifying process, even though

major transformations between species may occur through

the action of use and disuse of structures.

The kind of transformism advocated by Darwin is radi-

cally different from Lamarck’s transformism, even if it is

gradual. The main opposition between Darwin and

Lamarck is that the former denies any appeal to spontane-

ous generation because no convincing empirical argument

has ever been proposed in its favor. For Darwin, as far as

scientific acceptability is concerned, spontaneous gen-

eration is exactly on the same footing as special

creationism. Darwin’s arguments against Lamarck’s con-

ception of the origin of species are, therefore, identical as

the ones he gave against special creationism.

The genuine theoretical revolution Darwin introduced

in relation to species is a direct consequence of his rejec-

tion of spontaneous generation (as causing the appari-

tion of new species) and special creationism: it consists in

species transmutationism, the doctrine that new species

originate in older ones. This kind of radical transformism

was difficult to accept for the majority of his contemporaries.

The idea that species had no independent origin was con-

sidered to be tantamount to a plain denial of the “reality”

of species. According to Lyell, for instance, the “reality”

of species is connected to their immutability: a “real” spe-

cies cannot be modified with regard to its essential

characteristics. In brief, Lyell was an adherent of typologi-

cal thinking, even though he convinced Darwin that many

new species have come about and many others became

extinct during the long history of the Earth. The most ex-

plicit formulation of anti-transmutationism can be found

in Hopkins:

… every natural species must by definition have had a

separate and independent origin, so that all theories –

like those of Lamarck and M. Darwin – which assert the

derivation of all classes of animals from one origin, do, in

fact, deny the existence of natural species at all (1860, p.

747).

For Hopkins, as for many naturalists of the time, the

very definition of species requires their immutability. He

was not aware that changing the definition of the species

concept might lead to different claims about their existence.

His conceptual error, which is rather common, might ex-

plain why Darwin’s own conception of species was, and

still is, so difficult to grasp. Every naturalist has an intui-

tive knowledge of the meaning of the word “species,” in

as much he/she can identify species taxa. It is a common

mental procedure to move from this intuitive knowledge

to uttering general claims about species without being

aware that their truth crucially depends on the correct-

ness of the intuitive conception of species one has. If this

intuitive conception is useful for recognizing species in

practice but inadequate as a definition of the species

concept, it cannot guarantee the generalizations about

species one is tempted to champion; for instance, that

species cannot originate new species. As soon as one

realizes that the truth of such claims depends on what is

precisely meant by “species,” one can go beyond a

straight denial of Darwin’s theory. Hopkins does not go

that far in his conceptual examination of Darwin’s theory.

After this brief review of some of the conceptions of

and debates about species, we can identify two structural

components of the “species problem” as it manifested it-

self in 18th and 19th centuries. The first of these is the pre-

dominancy of “typological thinking,” as already pointed

out by Mayr (1942). Typological thinking was so over-

whelming that naturalists were unaware that they could

think about living beings otherwise, which was a main

obstacle against grasping Darwin’s conception. Second,

superimposed on typological thinking was the confusion

between what could be accepted as a definition of the

species concept and what could be used as practical rec-

ognition criteria. Buffon had been aware of this distinction,

but his successors did not understand it. Given that Dar-

win did not propose any explicit definition of the species

concept, it is understandable that the “species problem”

did not disappear after 1859, although it did take a slightly

different form.

DARWIN’S CONCEPTION OF SPECIES

One may wonder why Darwin himself did not offer a

clear definition of “species” because this concept appears

to be so central in his novel theory, as shown by the very

title of his book. One should understand, however, that

Darwin’s first objective was to convince his readers of the

plausibility of an entirely new theory of the living world.

A. Barberousse and S. samadi

191© 2010 ISZS, Blackwell Publishing and IOZ/CAS

What he meant by species was thoroughly dependent

upon the details of this theory, as we shall show in this

section.

The species concept and Darwin’s relation to it

According to a broad, but useful, distinction, two kinds

of concepts are used in science: “observational,” or de-

rived from observations, and “theoretical,” or defined rela-

tive to the principles of the theory they appear in. “Mass”

is typically a theoretical concept. Depending on the theory

within which it occurs, it can take different meanings. In

Newtonian mechanics, mass is an unalterable magnitude

that explains how material bodies resist acceleration,

whereas in relativity theory, mass varies depending on

the amount of energy. Therefore, meaning of the concept

of mass depends on the theory it appears in; mass does

not have any independent meaning.

The species concept is a theoretical concept as well.

Its meaning depends on the theory it appears in. Before

Darwin, there was no sufficiently sophisticated, well-ar-

ticulated and scientifically well-founded theory of the bio-

logical world. Lamarck had such a theory in mind, but, as

pointed out above, he resorted to spontaneous generation,

which was, even at the time, scientifically precarious. The

theories upon which the meaning of the species concept

was dependent were not fully scientific theories, but rather

commonsensical theories, most of which were closely as-

sociated with the reality of species to their immutability,

as emphasized above. This theoretical framework was not

recognized as such, however: many naturalists did not

scrutinize the origin of their conception of species and

were not aware that it depended on a set of mingled

assumptions, most of which are associated with

creationism, fixism and essentialism, as emphasized above.

In the same way as we commonly assume that the mass

(or weight) of material bodies is unproblematic and can be

easily determined using a weighing machine, naturalists

from the past thought that species were simply groups of

organisms that shared some important properties and

whose unity depended on the transmission of a formal

principle through reproduction.

The distinction between observational and theoretical

concepts has been criticized (Quine 1953). It has been

emphasized that most scientific concepts have a mixed

origin, both empirical and theoretical, and that no concept

is purely observational or purely theoretical. The species

concept is no exception. By emphasizing its theoretical

aspect, we wish to point out that one should not neglect

the importance of the dependence of its meaning on the

theory it appears in.

It was probably clear for Darwin that he could not pro-

pose any explicit definition of the species concept at the

same time as he put forward an entirely new theory of the

living world. His theory had to be understood and ac-

cepted before he could do so. Because he had to struggle

his life long for the full acceptance of his theory, he never

came back to the definition of the species concept.

However, his implicit conception can be reconstructed

from the Origin of Species.

Before moving to such a reconstruction, let us make

clear why Darwin stuck to a seemingly nominalist use of

the term “species” throughout the Origin of Species. At

the beginning of the book, he specifies that what he is

going to call “species:” the groups of organisms identi-

fied as species by competent naturalists. He does the same

for the term “variety.” This is how he uses these two

words, so that he can be understood by his readers while

explicitly rejecting all current definitions of the species

category. In other words, Darwin did not wish to adopt

any current propositions about the intension (or definition)

of the species concept, but was content with referring to

the commonly recognized extensions of species names;

that is, to the groups of organisms they designate in the

world.

Darwin’s claims about species

Darwin was well-aware of the diversity of the current

definitions of the species concepts, which were reviewed

and discussed in Berhardi (1834) and in Candolle (1855).

To show that none of the definitions were scientifically

tenable, he emphasized the importance of individual varia-

tion to break down the distinction between species and

varieties as these concepts were used in the practical taxo-

nomic literature, thus switching to “populational thinking,”

as Mayr (1942) points out. This allowed him to emphasize

the arbitrariness of the distinction, particularly in plants

and invertebrates, and to undermine the currently accepted

fixity of species.

Darwin’s main claims about species are that:

There is not fixed set of properties identifying any spe-

cies

Every species is subject to gradual change

There is not limit to the scope of change

Existing species are descended from former species

Most varieties go extinct.

The first 3 claims are components of Darwin’s

transformism and the fourth claim is the key to

transmutationism. All five claims are represented in

Darwin’s diagram in chapter 4 of the Origin of Species.

Species from Darwin onward

192 © 2010 ISZS, Blackwell Publishing and IOZ/CAS

These claims and the diagram are consequences of

Darwin’s new theory about the living world. To make their

meaning clearer, we recall his goals and how he achieved

them. The explanatory task he gave himself was to ac-

count for a large amount of independent observations,

including the geographic distribution of species and the

systematic patterns in the fossil records. He conceived

these observational facts as the results of long-term dy-

namics governing the organic history of the earth. For

him, natural history was not descriptive but thoroughly

explanatory. The dynamic component he inherited from

Lyell was the most important aspect of his world view. It

underlies Darwin’s conviction that the best explanation

of the observations is that species gradually change and

give rise to new species. His research program was thus

to explain “the origination of fresh species” (Lyell’s

expression) as a natural, rather than miraculous process.

He wanted to look for its cause in the ordinary course of

nature, to account for the “successive creation of spe-

cies” as “a regular part of the economy of nature.”

A major consequence of Darwin’s dynamic world view

is that he conceived of the set of observations that had

been gathered by himself and his predecessors and con-

temporaries as a sort of snapshot in a long history, not as

the signs of an immutable reality. He was convinced that

before (and after) the time this partial, synchronic snap-

shot was taken, variation is at work, both explaining what

he could observe and that this state of the living world

was transient. Phenotypic variation was, therefore, the

main explanatory factor of what can be observed.

As noticed by Darwin’s contemporaries, there is an

apparent contradiction between his insistence on varia-

tion and transformation on the one hand, and, on the other,

the common experience of naturalists, who observe dis-

crete groups yet not a continuum of forms. Such natural-

ists always wondered how Darwin might explain the fact

that they could not observe forms that were intermediate

ones between existing species. As put forward by Lewens

(2007, 2009), Darwin’s theory about common ancestry re-

quired that the small variations forming species are them-

selves stable. Consequently, the reason why there are

gaps between existing organic forms cannot be that these

forms are unstable. Darwin (1859, p. 171).was aware of this

problem:

… why, if species are descended from other species by

insensibly fine gradations, do we not everywhere see

innumerable transitional forms? Why is not all nature in

confusion instead of the species being, as we see them,

well defined?

Whereas his fixist contemporaries would look for ex-

planation in archetypes, Darwin turned to common de-

scent and natural selection. Both heredity and the shift-

ing competitive demands, that is, the local environments,

might explain the relative permanence of species. This is

one of the major reversals of Darwin’s approach relative

to his predecessors: the coherence of species is achieved

despite differences constantly being introduced among

individuals, not because of a property shared by all indi-

viduals (Lewens 2009).

Darwin’s theory thus manages to explain both the pri-

mary fact of evolution and the secondary fact (according

to his view) of the relative permanence of species. Were

Darwin unable to guarantee the transient (when geologi-

cal times are considered) permanence of species, he would

not be able to use the word “species” in the Origin of

Species as he does, namely as referring to what competent

naturalists recognize as species. The identification of spe-

cies and varieties has to be relativized to a given period of

time, however. Successful varieties being “incipient

species,” the classification of present living forms is itself

evolving, even though this is hard to realize on a human

time scale. This is not to say that Darwin has been a nomi-

nalist about species, as Beatty (1985) and Stamos (2007)

point out.

We can conclude this section on Darwin’s conception

of species by summing up his claims about the origin of

species. The explanation he offers bears on the origin of

the species he could observe, and, by translating it back-

ward in time, of the ancient species one can only observe

traces of. He looks less for the ultimate origin of life on

Earth than for a causal process explaining the diversity of

observable species, a process that operated in ancient

times and that is still operating. This process he identifies

as natural selection acting on slight phenotypic variation.

The evolving diversity of species is, therefore, condi-

tioned by the occurrence of variation. This might lead one

to think that the primary motor of evolution has to be

looked for in the mechanism of heredity. What is striking

from our retrospective point of view is that Darwin knew

nothing about the mechanism of heredity: he only knew

that it produces variations, some of which are inheritable.

However, he managed to construct a theory that still looks

very modern in many aspects, particularly in relation to

species. The key point in Darwin’s book seems to us that

the (unknown to him) mechanism of heredity explains both

species permanence and phenotypic variation, which is

responsible for species transformation and for the occur-

rence of new species. It is because Darwin was agnostic

about the mechanism of heredity that he could contem-

A. Barberousse and S. samadi

193© 2010 ISZS, Blackwell Publishing and IOZ/CAS

plate the possibility of this double causal pathway that

was a priori dismissed by typologists. Not only was Dar-

win agnostic, but he was willing to make as few assump-

tions as possible about what is going on in the reproduc-

tion process: he only assumes that something happens

that can both explain permanence and variation. Remem-

bering Darwin’s epistemological situation is a good way

not to get confused about the species problem now that

we know a lot about the mechanism of heredity.

SPECIES IN PRESENT-DAY BIOLOGY

Definition versus practical recognition

criteria: A lingering confusion

Biology has changed a lot since the Origin of Species

appeared. Darwinism itself has changed several times

(Depew & Weber 1995; Gayon 2003, 2009). However, the

“species problem” is still hindering a large part of research

in biology (see references above). This situation is para-

doxical for two reasons. First, the species concept is cen-

tral in biology, so that it is surprising that there is no

consensus about its definition. Second, the persistence

of the “species problem” nowadays is paradoxical because

De Queiroz (1998, and other references indicated above)

has proposed a convincing diagnosis of why there is a

problem, which is the first step in solving it. This diagno-

sis is the one we alluded to above: the problem is a result

of an enduring, but only recently recognized, confusion

between the definition of the species concept and the list-

ing of practical criteria for species recognition. Once the

distinction between these two aspects is made, many of

the so-called “definitions” of species no longer appear as

definitions but as practical recognition criteria. This is the

case for the so-called “definitions” based on morphologi-

cal traits, interfecundity and monophyly (or other phylo-

genetic criteria), as we shall see below. Wiley’s (1981) and

Simpson’s (1951) definitions of the species concepts are

exceptions, because they do not fall under the confusion

between a genuine definition and the listing of practical

recognition criteria. However, neither Wiley nor Simpson

analyze the distinction and interplay between definition

and recognition criteria.

The discussions about species during the 20th century

have brought about plenty of “definitions” of species.

Most were meant to be more compatible with the theory of

evolution than former conceptions, whereas others were

deliberately conceived as independent of any theory. Let

us take Mayr’s “biological definition” as an example. As

is well-known, Mayr defines species as “groups of actu-

ally or potentially interbreeding natural populations, which

are reproductively isolated from other such groups” (1942,

p. 120). As is clear from the first section of the present

paper, the criterion of interfecundity was commonly used

long before Darwin. Mayr’s proposal is not a new defini-

tion of the species concept, but a new interpretation of

this criterion in the light of the theory of evolution. Repro-

ductive isolation is one of the main mechanisms of

evolution; this is why interfecundity is viewed, within the

theory of evolution, as a major property of species (at

least, as far as sexually reproducing organisms are

concerned) that can be used in practice, as done in popu-

lation genetics.

By contrast, the motivation behind the phenetic “defi-

nitions” (see Sokal & Crovello 1970), based on resemblance

between characters, is to be independent of any theory in

order to avoid all kind of bias. When viewed as a recogni-

tion criterion, resemblance is fully justified by the prin-

ciple of common descent, which also allows one to under-

stand the limits of its applicability. Consequently, it is

much more interesting and useful to consider resemblance

as a practical criterion grounded in the theory of evolu-

tion rather than as an a-theoretical definition of the spe-

cies concept.

Once De Queiroz’s diagnosis of the “species problem”

is accepted, it is easy to see that the so-called “definitions

of species” or “species concepts” roughly correspond to

the different disciplines dealing with species delimitation:

alpha-taxonomy, population biology and phylogenetics.

These disciplines respectively echo the “phenetic,” “bio-

logical” and “phylogenetic” species concepts, among

others, which are connected with methods of formulating

hypotheses about species delimitation. The current meth-

ods of species recognition and delimitation rely on dis-

tinctive sets of properties: morphological and molecular

traits, interfecundity and phylogenetic relations.

What is most striking in the present-day “species prob-

lem” is how much it resembles the 18th and 19th century

“species problem,” even though the theoretical framework

of the discussion has changed drastically. We see this

surprising continuation through theoretical change as an

impulse to make the present-day theoretical framework as

clear as possible, a seldom undertaken enterprise, though

it may be revealing. This is what we do hereafter, empha-

sizing that as far as the species concept is concerned, not

so much has changed since Darwin, whereas the practice

of species delimitation has undergone invaluable progress.

Species from Darwin onward

194 © 2010 ISZS, Blackwell Publishing and IOZ/CAS

What is the theoretical framework upon which

the definition of the species concept should

rely?

There seems to be a large consensus among evolution-

ary biologists regarding the basic principles of the theory

of evolution as it is used today. The first of these prin-

ciples is common descent, associated with the idea that

the origin of diversity (or variation) among organisms is

mutation, as far as this term is taken in a broad sense,

namely as referring to any modification of an organism’s

characters that is transmitted to its offspring through

reproduction. Typical examples of mutation are the sub-

stitution of one nucleotide for another in a DNA molecule

or chromosomal rearrangement. The next two principles

set out how offspring-leaving organisms are sampled by

selection and drift. According to the principle of natural

selection, certain organisms are more efficient than others

at reproducing. According to the principle of drift, a ran-

dom sampling of offspring-leaving organisms occurs at

each generation. The fourth and last principle emphasizes

the role of context, understood as the geographical loca-

tion of organisms, the associated ecological conditions

(biotic and abiotic) and the evolutionary history of their

ancestors. The effects of selection and drift, as sampling

processes, are heavily context-dependent.

To make clear what the consequences of these prin-

ciples are, another important element has to be mentioned;

namely, what the domain of the theory is exactly. We claim

that the domain of the theory of evolution as it is used

today is the set of all organisms on Earth, past, present

and future. As we shall see below, we consider that adopt-

ing the “point of view of the genes” is not the appropriate

way to solve the species problem, even though it is the

most common point of view since the Modern Synthesis

to conceive of evolution as change in gene flow.

According to the principle of common descent, the set

of all organisms in the past, present and future history of

the Earth constitutes a huge, unique genealogical

network, because every organism is related with each other

by (usually indirect) genealogical relationships. The other

principles of the theory of evolution imply that groups of

organisms diverge from one another. This in turn can be

represented by the genealogical network acquiring a tree-

like topological structure. This would not be the case were

selection and drift not acting; that is, were no evolution

occurring.

In Samadi and Barberousse (2006), we show how it is

possible to define a rigorous partition of this genealogical

network through an equivalence relation that can be inter-

preted as defining species. We thus define species as in-

ternodal segments of the genealogical network of

organisms, following Simpson (1951, 1961), Hennig (1966),

Wiley (1981), Kornet et al. (1995) and de Queiroz (1998).

We emphasize that the most important aspect of the ge-

nealogical relationship is the transmission of a piece of

matter from a parent organism to its descendant during

reproduction. It happens that this piece of matter con-

tains DNA; had it been another molecule, nothing essen-

tial would have been changed in the global, tree-like struc-

ture of the genealogical network.

The contribution of present-day biology to the

methodology of species delimitation

The definition provided in Samadi and Barberousse

(2006) gives taxonomy a unified theoretical basis as far as

it allows one to account for the efficiency of the practical

criteria for species recognition and delimitation that are

currently used. In this theoretical framework, the criteria

are easily conceived of as complementary rather than

competing, as is the case when they are wrongly consid-

ered as definitions of the species concept. It has to be

forcefully reminded that these criteria allow biologists to

formulate hypotheses, not facts, about the grouping of

organisms into species. As emphasized by Sites and

Crandall (1997), described species are not facts: they are

hypothesized when the available criteria are fulfilled. These

hypotheses may be supported or rejected when further

investigation brings out new data or when new criteria are

applied.

Alpha-taxonomy, to begin with, describes species on

the basis of morphological, behavioral or ecological

characters, and, as a result, emphasizes primary hypoth-

eses about the structure of the global genealogical net-

work of organisms. These primary hypotheses may then

be tested against more or less indirect clues indicating

reproductive isolation, or against signs of apomorphy

acquisition. Apomorphy acquisition may be further

investigated, in some cases, by coalescence theory. What

is important here is the interplay of various techniques:

for instance, the grouping of organisms on the basis of

morphological, behavioral or ecological resemblance, or

the phylogenetic reconstruction that aims at revealing the

topological structure of limited parts of the global genea-

logical network.

The vast amount of knowledge we have acquired since

Darwin’s times about the mechanisms of heredity is of

insuperable help in achieving taxonomic tasks. It allows

us to use molecular characters as well as morphological,

behavioral and ecological ones, and gives us access to

A. Barberousse and S. samadi

195© 2010 ISZS, Blackwell Publishing and IOZ/CAS

unprecedented knowledge of the past. Molecular charac-

ters allow biologists not only to draw phylogenetic trees

of genes, but also to infer the demographic history of

populations, including the detection of barriers to gene

flows. Some of the most recent methodological develop-

ments have stemmed from the coalescence theory. New

methods that allow refined inferences about groups of

isolated organisms have been proposed and generate, in

turn, richer hypotheses of species delimitation (see for

example in Knowles and Carstens 2007).

In this context, “integrative taxonomy,” as advocated

by several authors (e.g. Sites & Marshall 2003; Dayrat

2005; DeSalle 2006; Vogler & Monaghan 2007), may be

understood as an incentive to support hypotheses of spe-

cies delimitation by all possible sources of evidence. One

of the motivations of integrative taxonomy is probably to

invite biologists to propose, insofar as it is possible, spe-

cies names that are linked to long-lasting hypotheses of

species delimitation. However, new technical developments

will inevitably bring out new data and methods that will

question the hypotheses that were earlier supposed to be

robust.

The process of validation or modification of taxonomic

hypotheses is achieved through so-called “taxonomic

revisions.” The revisions consist of evaluating anew ear-

lier hypotheses of species delimitation in the light of new

data, which may be new specimens, new characters, or the

results of new methods of analysis. Then, using the rules

of a nomenclature code, the attribution of species names

may be modified accordingly. The application of nomen-

clature rules, however, is not an easy task, and is often

not carried out completely. For instance, in Smith et al.

(2007) some scientific names are given between quotation

marks because it was not possible to rigorously link the

new hypotheses of species delimitation, drawn from mo-

lecular data, to available species names.

The development of the Barcode of Life project pro-

vides us with increasing more of these situations. This

project is at the crossroad of a shift in the perception of

the extent of biodiversity on the one hand and the devel-

opment of sequencing technologies and diffusion of data

through electronic data bases that are accessible through

the web on the other hand. At the middle of the 20th

century, biodiversity was estimated to a few millions of

species (among which approximately 1.5 million were

described), whereas today’s estimations of biodiversity

range between 10 and 100 millions of species (Blaxter 2004;

Savolainen et al. 2005). Consequently, traditional meth-

ods of taxonomy need to be reconsidered and new ap-

proaches proposed to increase the rate of species descrip-

tion (Agapow et al. 2004; Wiens 2007). Hebert et al. (2003)

launched the Barcode of Life project to face this large

scale variation, taking on the success of molecular ap-

proaches in systematics and the accessibility of data

through electronic databases. The aim of this project is

both to facilitate taxonomic expertise (i.e. attributing speci-

mens to hypotheses of species delimitation using taxo-

nomic names) and to accelerate the rate of discovery of

yet undetected species. This standardized approach has

actually accelerated the growth rate of taxonomic

knowledge. To avoid falling into typological thinking, such

a standardized approach should become fully integrated

in the practice of taxonomy, including the application of

nomenclature rules that allow us to follow the revision

process.

To sum up, our recent knowledge of the transmission

of genes is important from the practical point of view of

species recognition and delimitation, not from the theo-

retical point of view of the definition of the species concept.

Information about how characters are inherited is useful

in the context of the disciplines whose main objects of

investigation are characters, like taxonomy and phyloge-

netic systematics; but as far as the definition of the spe-

cies concept is concerned, it does not add anything to

Darwin’s own conceptions that we have tempted to bring

up to date in the present paper.

CONCLUSION

Even though Darwin’s conception of species is not

explicit in the Origin of Species, we have attempted to

present a plausible reconstruction of his main claims about

species by contrasting them with the opinions of his pre-

decessors and contemporaries. We have also argued that

the “species problem” that could be found in Darwin’s

time, arising from a conflict between theoretical ideas about

species and current practices of recognition and

delimitation, has unfortunately perdured, whereas the con-

ceptions of species have changed thanks to Darwin.

To solve today’s “species problem,” we have sug-

gested that it is necessary to have a clear view of what the

theory of evolution amounts to nowadays. We have pro-

posed a definition of the species concepts that relies on a

consensus version of today’s theory of evolution and

that is compatible with Darwin’s insights.

REFERENCES

Agapow PM, Bininda-Emonds OR, Crandall KA et al.

(2004). The impact of species concept on biodiversity

studies. The Quarterly Review of Biology 79, 161–79.

Species from Darwin onward

196 © 2010 ISZS, Blackwell Publishing and IOZ/CAS

Anon (Chambers R) (1844). Vestiges of the Natural His-

tory of Creation. John Churchill, London.

Beatty J (1985). Speaking of Species: Darwin’s Strategy.

In: Kohn D, ed. The Darwinian Heritage. Princeton Uni-

versity Press, Princeton, pp. 265–81.

Bernhardi J (1834). Uber den Begriff der Pflanzenart und

seine Anwendung. Otto, Erfurt.

Blaxter M (2004). The promise of a DNA taxonomy. Philo-

sophical Transactions of the Royal Society: Biological

Sciences 359, 669–79.

Buffon GL (1749). Histoire naturelle, générale et

particulière, avec la description du cabinet du Roy.

Paris, Tome Second.

Candolle A de (1855). Géographie biologique raisonnée.

Librairie Masson, Paris.

Claridge MF, Dawah HA, Wilson MR, eds (1997). Species:

The Units of Biodiversity. Chapman & Hall, London.

Darwin C (1859). On the Origin of Species by Means of

Natural Selection, or the Preservation of Favoured

Races in the Struggle for Life, 1st edn. John Murray,

London.

Dayrat B (2005). Towards integrative taxonomy. Biologi-

cal Journal of the Linnean Society 85, 407–15.

Depew D, Weber BH (1995). Darwinism Evolving: Systems

Dynamics and the Genealogy of Natural Selection. MIT

Press, Cambridge, MA.

De Queiroz K (1998). The general lineage concept of species,

species criteria, and the process of speciation: A con-

ceptual unification and terminological recommendations.

In: Howard DJ, Berlocher SH, eds. Endless Forms: Spe-

cies and Speciation. Oxford University Press, Oxford,

pp. 57–75.

De Queiroz K (1999). The general lineage concept of spe-

cies and the defining properties of the species category.

In: Wilson RA, ed. Species: New Interdisciplinary

Essays. MIT Press, Cambridge, MA, pp. 49–89.

De Queiroz K (2005a). Different species problems and their

resolution. BioEssays 27, 1263–9.

De Queiroz K (2005b). Ernst Mayr and the modern concept

of species. PNAS 102, 6600–607.

De Queiroz K (2007). Species concepts and species

delimitation. Systematic Biology 56, 879–86.

DeSalle ROB (2006). Species siscovery versus species iden-

tification in DNA barcoding efforts: Response to

Rubinoff. Conservation Biology 20, 1545–7.

Ereshefsky M, ed (1992). The Units of Evolution: Essays

on the Nature of Species. MIT Press, Cambridge, MA.

Frost DR, Kluge AG (1994). A consideration of epistemol-

ogy in systematic biology, with special reference to

species. Cladistics 10, 259–94.

Gayon J (2003). From Darwin to today in evolutionary

biology. In: Hodge JJ, Radick G, eds. The Cambridge

Companion to Darwin. Cambridge University Press,

Cambridge, pp. 240–64.

Gayon J (2009). Death or persistence of Darwinism? A

philosopher’s point of view. Comptes rendus-Palevol

8, 321–40.

Giraud T, Refrégier G, Le Gac M, de Vienne DM, Hood ME

(2008). Speciation in fungi. Fungal Genetics and Biol-

ogy 45, 791–802.

Hebert PD, Cywinska A, Ball SL, deWaard JR (2003). Bio-

logical identifications through DNA barcodes. Proceed-

ings of the Royal Society of London B 270, 313–21.

Hennig W (1966). Phylogenetic Systematics, (translation

from German, Dwight Davis and Rainer Zangerl). Uni-

versity of Illinois Press, Urbana, IL.

Hey J (2001a). The mind of the species problem. Trends in

Ecology & Evolution 16, 326–9.

Hey J (2001b). Genes, Categories, and Species: The Evo-

lutionary Causes of the Species Problem. Oxford Uni-

versity Press, New York.

Hey J (2006). On the failure of modern species concepts.

Trends in Ecology & Evolution 21, 447–50.

Hopkins W (1860). Physical theories of the phenomena of

life. Fraser’s Magazine 61, 739–52.

Howard DJ, Berlocher SH, eds (1998). Endless Forms. Spe-

cies and Speciation. Oxford University Press, Oxford.

Kimbel WH, Martin LB, eds (1993). Species, Species

Concepts, and Primate Evolution. Plenum Press, New

York.

Knowles LL, Carstens BC (2007). Delimiting species with-

out monophyletic gene trees. Systematic Biology 56,

887–95.

Kornet DJ, Metz JAJ, Schellinx H (1995). Internodons as

equivalence classes in genealogical networks: building-

blocks for a rigorous species concept. Journal of Math-

ematical Biology 34, 110–22.

Lamarck JB (1809). Philosophie zoologique. Dentu, Paris.

Lewens T (2007). Darwin. Routledge, London.

Lewens T (2009). The origin and philosophy. In: Ruse M,

Richards AJ, eds. The Cambridge Companion to the

“Origin of Species”. Cambridge University Press,

Cambridge, pp. 314–32.

Linné CO (1744). Oratio de telluris habitabilis incremento.

Cornelius Haak, Leiden.

Lyell C (1830). Principles of Geology, Being an Attempt to

A. Barberousse and S. samadi

197© 2010 ISZS, Blackwell Publishing and IOZ/CAS

Explain the Former Changes of the Earth’s Surface, by

Reference to Causes Now in Operation, Vol 1. John

Murray, London.

Mallet J (2001). Species, concepts of. In: SA Levin, ed.

Encyclopedia of Biodiversity, Vol 5. Academic Press,

New York, pp. 427–40.

Mayden RL (1997). A hierarchy of species concepts: The

denouement in the saga of the species problem. In:

Claridge MF, Dawah HA, Wilson MR, eds. Species: The

Units of Biodiversity. Chapman & Hall, London, pp. 381–

424.

Mayr E (1942). Systematics and the Origin of Species. Co-

lumbia University Press, New York.

Mayr E (1964). Introduction to Facsimile Edition of

Darwin’s Origin of Species, of 1859. Harvard Univer-

sity Press, Cambridge MA.

Noor MAF (2002). Is the biological species concept show-

ing its age? Trends in Ecology & Evolution 17, 153–4.

Otte D, Endler JA, eds. (1989). Speciation and Its

Consequences. Sinauer, Sunderland, MA.

Pigliucci M (2003). Species as family resemblance concepts:

The (dis-)solution of the speciesproblem? BioEssays

25, 596–602.

Quine WV (1953). Two dogmas of empiricism. In: From a

Logical Point of View. Harvard University Press,

Cambridge, pp. 20–46.

Rissler LJ, Apodaca JJ (2007). Adding more ecology into

species delimitation: Ecological niche models and

phylogeography help define cryptic species in the Black

Salamander (Aneides flavipunctatus). Systematic Biol-

ogy 56, 924–42.

Samadi S, Barberousse A (2006). The tree, the network, and

the species. Biological Journal of the Linnean Society

89, 509–21.

Savolainen V, Cowan RS, Vogler AP, Roderick GK, Lane R

(2005). Towards writing the encyclopedia of life: An in-

troduction to DNA barcoding. Philosophical Transac-

tions of the Royal Society B 360, 1805–11.

Simpson GG (1951). The species concept. Evolution 5, 285–

98.

Simpson GG (1961). Principles of Animal Taxonomy. Co-

lumbia University Press, New York.

Sites JW, Crandall KA (1997). Testing species boundaries

in biodiversity studies. Conservation Biology 11, 1289–

97.

Sites JW, Marshall JC (2003). Delimiting species: A Renais-

sance issue in systematic biology. Trends in Ecology &

Evolution 18, 462–70.

Sites JW, Marshall JC (2004). Operational criteria for delim-

iting species. Annual Review of Ecology, Evolution,

and Systematics 35, 199–227.

Sloan PR (2008). Evolution. In: Zalta EN, ed. Stanford En-

cyclopedia of Philosophy, 2008 winter edn. [cited May

2 2010.] Available from URL: http://plato.stanford.edu/

archives/win2008/entries/evolution/

Sloan PR (2009). Originating species: Darwin on the spe-

cies problem. In: Ruse M, Richards AJ, eds. The Cam-

bridge Companion to the “Origin of Species.” Cam-

bridge University Press, Cambridge, pp. 67–86.

Smith MA, Wood DM, Janzen DH, Hallwachs W, Hebert

PDN (2007). DNA barcodes affirm that 16 species of

apparently generalist tropical parasitoid flies (Diptera,

Tachinidae) are not all generalists. PNAS 104, 4967–72.

Sokal RR, Crovello TJ (1970). The biological species

concept: A critical evaluation. American Naturalist 104,

107–23.

Stamos DN (2007). Darwin and the Nature of Species SUNY

Series in Philosophy and Biology. State University of

New York Press, New York.

Vogler AP, Monaghan MT (2007). Recent advances in DNA

taxonomy. Journal of Zoological Systematics and Evo-

lutionary Research 45, 1–10.

Wheeler QD, Meier R (2000). Species Concepts and Phy-

logenetic Theory. Columbia University Press, New York.

Wiens JJ (2007). Species delimitation: New approaches for

discovering diversity. Systematic Biology 56, 875–8.

Wiley EO (1981). Phylogenetics: The Theory and Practice

of the Phylogenetic Systematics. Wiley and Sons, New

York.

Wilson RA (1999). Species: New Interdisciplinary Essays.

MIT Press, Cambridge, MA.

Winston JE (1999). Describing Species: Practical Taxo-

nomic Procedure for Biologists. Columbia University

Press, New York.

Wu C-I (2001). The genic view of the process of speciation.

Journal of Evolutionary Biology 14, 851–65.

Species from Darwin onward