Document 156179

Journal of Plankton Research Vol 13 no 5 pp. 1093-1099, 1991
Marine phytoplankton: how many species in the world ocean?
A.Sournia, M.-J. Chretiennot-Dinet1 and M.Ricard2
Observatoire Odanologique de Roscoff, do Laboratoire de giologie, Museum
national d'histoire naturelle, 43, rue de Buff on, 75005 Paris, Observatoire
Odanologique de Banyuls, Laboratoire Arago, 66650 Banyulus-sur-mer and
Laboratoire de cryptogamie, Musi-urn national d'histoire naturelle, 12, rue de
Buffon, 75005 Paris, France
Do we classify what we see or see what we classify?
Griffiths (1989)
The number of extant species and genera of the marine phytoplankton has long
been unknown, despite its obvious (not only academic) interest in ecology,
phycology and evolution studies. The reason is that no synopsis of the plankton
flora existed at the world scale. The two more extensive surveys available in the
taxonomical literature thus far were not entitled to fill this need: (i) the botanical
volume of the Nordisches Plankton edited at the beginning of this century by
Brandt and Apstein (1980) with contributions by C.Apstein, H.H.Gran,
C.Lemmermann, O.Paulsen and N.Wille dealt mostly with North Atlantic and
adjacent seas; and (ii) the field covered by the Rabenhorst's Kryptogamen-Flora
von Deutschland, Osterreich und der Schweiz, which came later (Hustedt, 19271966; Gemeinhardt, 1930; Schiller, 1930, 1931-1937; Geitler, 1932), is much
larger than its modest title would imply but, in any case, this flora is ~60 years
old. The other taxonomical works are limited both in the phycological dimension
(i.e. they deal with only one or a few algal groups) and in the geographical
dimension (i.e. they deal with a given region or ocean). Since the Rabenhorst's
series, none of the algal classes (or whatever rank is attributed to the diatoms,
dinoflagellates, and so on) has benefited a treatment at the world scale as far as
marine phytoplankton is concerned, if one expects an exhaustive but noncritical, illustrated catalogue of the living and fossil silicoflagellates (Loeblich et
al., 1968). Note that the freshwater phytoplankton is more favoured on this
respect (Huber-Pestalozzi, 1938-1972; Bourrelly, 1973-1985, 1988; Ettl et al.,
1978-. . .).
There do exist some mentions of total species numbers of marine phytoplankton in the recent literature but these seem to rely on approximate estimates or
© Oxford University Press
Downloaded from by guest on September 11, 2014
Abstract. Towards the end of the 1980s, living plankton flora of the world ocean amounted to 474504 genera and 3444—4375 species if one neglects the increase rate of taxa during the latest years In
the above figures, the lower estimate is that of the 'reliable' taxa (of practical value for identification
tasks), whereas the higher estimate includes the insufficiently known or doubtful organisms, and
synonyms are excluded. The frequency distribution of the numbers of species per genus confirms the
general hyperbolic law, which implies that a relatively large number of genera are uni- or
A.Soumla, M.-J.Cbrltfcnnot-DiiKt and M.Rtcard
unpublished census. For instance, the number of living species of dinoflagellates
has been evaluated as 1000 (Tappan, 1980), 1000-1500 (Steidinger, 1983) or 2000
(Taylor, 1987; Margulis et al., 1990).
A synopsis of the genera of the world ocean has been recently published which
indicates the number of species of each genus (Sournia, 1986; Ricard, 1987;
Chr6tiennot-Dinet, 1990). An answer may thus be given and some comments
The data
The census is summarized on Table I. Some notes must be added for certain
(i) The relatively wide range of species numbers in most of the nanoplanktonic
classes arises mostly from the fact that electron microscopy is now routinely
used for these groups but the relevant information is lacking for many ancient
Downloaded from by guest on September 11, 2014
The Atlas du Phytoplancton Marin considers 15 classes in the marine
phytoplankton, all of which are treated in the same way (although three authors
are involved); these are the Cyanophyceae, Dictyochophyceae, Dinophyceae
and Raphidophyceae (Sournia, 1986), the Bacillariophyceae (Riocard, 1987)
and 11 other, mostly nanoplanktonic classes among them being the Prymnesiophyceae (Chr6tiennot-Dinet, 1990). This classification has been followed here
for obvious reasons but it does not pretend to be sounder than any of the recent
classifications such as those of Levine et al. (1980), Parker (1982), Lee et al.
(1985), Puytorac et al. (1987) or Margulis et al. (1990).
The precision as regards the number of species in a given genus is very
variable as it depends on the proportion of the species which has been poorly
described or which remains 'doubtful' for one reason or another to the eyes of
the specialist. This reservation does neither include the strictly taxonomical
uncertainties nor the failings to the nomenclature rules (Greuter et al., 1988)
because a taxonomical statement is ascribed to each alga in the Atlas, how
appropriate it may prove to be in the future (in other words, a given organism
cannot be counted twice in different parts of the Atlas). Because of the
uncertainty about species numbers, the following census considers either a
number or a range, depending on the genus considered; in the latter case, the
lower number is that of the taxa which look presently reliable and have a
practical value to the taxonomist in charge of identifying a species, while the
higher number includes the insufficiently known or doubtful species, synonyms
The authors are aware of several new entries or taxonomical propositions
which have been published during the last years or months but, for the sake of
clarity, such recent information is not taken into consideration, so that the
reference level remains that of the Atlas as it was published.
How many marine phytopJankton species in the world ocean?
Table I. Numbers of genera and species of the living marine phytoplankton, as compiled from
Sournia (1986), Ricard (1987) and Chretiennot-Dinet (1990)
474-504 = 489 ± 15
3444-4375 = 3910 ± 465
(ii) The mutual delimitations of the Chlorophyceae and Prasinophyceae are
particularly amenable to changes in the near future, with new classes being
possibly added (0.Moestrup, personal communication).
(iii) The Diatomophyceae are subdivided here into their two orders because of
the well-known ecological and evolutionary implications of this taxonomical
delimitation. The particularly wide ranges for the species numbers result from
the presence (or dominance, among the Pennales) of tychoplanktonic diatoms
whose ascription to either plankton or benthos is utterly questionable.
Because of this fundamental ambiguity, the estimation of genera and species is
particularly unprecise in the case of diatoms.
(iv) Among the Dinophyceae, no effort has been made to distinguish the freeliving heterotrophic and mixotrophic organisms from the autotrophic ones
(see Larsen and Sournia, 1991) but the parasitic and symbiotic genera have
been omitted.
The uncertainty about the total numbers of genera and species may reach 3.1
and 11.9% respectively. This indicates that practices in taxonomy are such that
genera are more firmly established and/or that they are felt as more reliable to
taxonomists than species are.
Comparisons can be made with the freshwater phytoplankton on the account
of the census provided by Bourrelly (1985) in the second edition of his Algues
d'Eau Douce, from which the total numbers of genera and species reach 1173
and 14 900 respectively. For many classes, however, biases are introduced by
inclusion of multicellular and/or benthic organisms among the freshwater
Downloaded from by guest on September 11, 2014
Central es
A.Soomla, M.-J.Chrttfcnnot-Dtnet and M.Ricard
Number of genera
200 -i
All classes of marine phytoplankton
150 -
100 -
50 -
Number of species per genus
Fig. 1. Frequency distribution of the numbers of species per genus for the whole plankton flora of the
world ocean (lower and higher estimates as defined in the text).
Downloaded from by guest on September 11, 2014
genera. If the comparison is restricted to classes whose members are essentially
planktonic and which have been given the same taxonomical acceptance, then
the comparative diversity of the marine and freshwater phytoplankton seems to
differ widely depending on classes. The continental plankton would thus contain
many more chrysophytes (159 genera and 860 species) and euglenophytes (55
genera and 1100 species) but many fewer dinoflagellates (32 genera and 240
A picture of the biological diversity can be provided by the frequency
distribution of species numbers per genus. It was soon recognized by Willis
(1922) that this distribution is not linear but rather hyperbolic, which means that
a relatively high number of genera comprise a few or only a single species. Willis
dealt mostly with flowering plants and, briefly though quite demonstratively,
with several groups of animals. His findings have been recently verified by
Griffiths (1989) as regards several groups or protista. The latter work includes a
regional synopsis of freshwater algae but does not consider marine phytoplankton (neither partially nor totally). The present study makes it thus possible to
extend Willis's laws to the marine plankton flora. The frequency distribution of
species among genera, resulting from the present data, is shown in Figure 1;
although Griffiths (1989) uses arithmetic scales, a logarithmic scale was felt
preferable for the abscissae. A 'hollow curve' (Willis, 1922) is found again. At
the height of the curve, 182-195 genera, i.e. more than one-third of the total
flora (38.4-38.7%), are shown to be monotypic (i.e. consisting of a single species
each). A strikingly identical proportion (38.6%) was found by Willis (1922) for
the flowering plants of the world, 4853 of them out of a total of 12 571 being
monotypic (as taken by Willis from the 1919 edition of his Dictionary). The 'tail'
of the curve on Figure 1 shows very few genera having >50 species each. From
How many marine pbytoplankton species in the world ocean?
Number of genera
Minimum or exact number of species per genus
Fig. 2. Comparative frequency distribution of the numbers of species per genus for the
dinoflagellates (top) and the diatoms (bottom)
Downloaded from by guest on September 11, 2014
the above figures, in so far as this can be estimated from the numbers of species
per genus, biological diversification appears to be very low. The shoulder and
the peak at 5 and 10 species per genus result from the artefact that many genera
of tychoplanktonic diatoms have been ascribed the approximate number of 'say',
5 or 10 planktonic species (Figure 2, bottom). The curve of the minimal or exact
species numbers and that of the maximal numbers do not differ significantly in
shape from each other (Figure 1). Differences among classes are exemplified by
the comparison of the Bacillariophyceae and the Dinophyceae (Figure 2).
The present census of the marine plankton flora is obviously liable to two
(i) It is already underestimated, even before being actually published, due to the
increase rate of taxa in the literature. Defining these rates for the various
classes or for the whole plankton flora would deserve a special study but an
order of magnitude has been provided by VanLandingham (1967): 'New
diatoms [either recent or fossil, planktonic or benthic, and freshwater or
marine] are being described at a rate of approximately one per day or about
A-Sournta, M.-J.Chretieniiot-DiDrt and M.Rkard
The thoughtful comments of two anonymous reviewers are gratefully acknowledged.
Anderson,D.M. (1990) Toxin variability in Alexandnum species In Gran61i,E- (ed.), Toxic Marine
Phytoplankton. Elsevier, New York, pp. 41-51.
Bounelly.P. (1973-1985) Les algues d'eau douce. Initiation a la systtmatique (2eme edition revue et
corrigee). I. Les algues vertex (1973). II. Les algues jaunes et brunes (1981). III. Les algues bleues et
rouges, les Engltniens, Ptridmiens et Cryptomonadtnes (1983). Boubee, Pans.
Bourrelly.P. (1988) Compliments. Les algues d'au douce. Initiation a la systtmatique Tome I. Les
algues vertes. Soc. nouv. Edit Boubee, Paris.
Brandt.K. and Apstein.C. (eds) (1908) Nordisches Plankton Botanischer Ted. XVIII, 1-124; XIX,
1-146; XX, 1-29; XXI, l^M); XXII, 1-5. Upsius & Tischer, Kiel and Lepizig.
Campbell.L., Shapiro.L.P., Haugen.E.M. and Morris.L. (1989) Immunochemical approaches to the
identification of the ultraplankton: assets and limitations. In Cosper.E M. et al. (eds), Novel
Phytoplankton Blooms. Springer, Berlin, pp. 39-56.
Chretiennot-Dinet,M.-J. (1990) Atlas du phytoplancton mann. Vol. 3. Chlorarachnwphyctes,
Chlorophycies, Chrysophydes,
Cryptophydes, Euglinophydes,
Eustigmatophydes, Prasinophycies, Prymntsiophydes,
Rhodophyctes, Tribophycies. Editions du CNRS, Paris.
Ettl.H., Gerloff,J., Heynig.H. and MollenhauerJ). (eds) (1978-. . .) SOsswasserflora von
Mitteleuropa. Gustav Fisher, Stuttgart, Jena.
Geitler.L. (1932) Cyanophyceae von Europa unter Beriicksichtigung der anderen Kontinente. In
Rabenhorst's Kryptogamen-Flora von Deutschland, Oesterreich und der Schweiz, Vol. 14 Akad.
Verlag, Leipzig, pp. 1196.
Gemeinhardt.K. (1930) Silicoflagellatae. In Rabenhorst's Kryptogamen-Flora von Deutschland,
Oesterreisch und der Schweiz, Vol 10 (2), pp. 1-87.
Greuter.W. et al. (eds) (1988) International Code of Botanical Nomenclature Adopted by the
Fourteenth International Botanical Congress, Berlin, July-August 1987. Regnum vegetabile, 118
Koeltz Scientific Books, Koenigstein.
Griffiths.A.J. (1989) The protistan landscape Prog. Protistol., 3, 1-20.
Huber-Pestalozzi.G. (1938-1972) Das Phytoplankton der Susswassers. Systematik und Biologie In
Thienemann.A. (ed.), Die Binnengewasser. Schweizerbart'Verlags., Stuttgart, Vol. 16(1), pp. 1342, pi. 1-66.
Hustedt.F. (1927-1966) Die Kieselalgen Deutschlands, Osterreichs und Schweiz mit Beruksichtigung der iibngen Lander Europas sowie der angrenzenden Meeresgebiete. In Rabenhorst's
Kryptogamen-Flora von Deutschland, Oesterreich und der Schweiz, Vol. 7: 1 [1-5] (1927-1930); 2
[1-6] (1931-1959); 3 [1-4] (1961-1966). Akad. Verlag, Leipzig.
Downloaded from by guest on September 11, 2014
320 per year. New diatom genera are being described at a rate of about 3 or 4
per year'. Note that a plot of the increase rate through time, starting at the
early days of planktology, may allow a simple extrapolation of the 'true'
number of taxa in the oceans to be made, provided that the resulting trend
would be a rather regular one. This exercise has not been carried out yet but it
can be attempted, in the case of the dinoflagellates at least, as a catalogue of
the new taxonomic entries of the marine planktonic Dinophyceae is currently
up-dated (Sournia, 1991, and references therein).
(ii) The count is worth what the Linnean classification is worth. The delimitation
of many genera and species, and the validity of the taxonomical criteria are
widely questioned. At least two current perspectives of planktology, viz.
picoplankton ecology and the identification of toxic species, tend to
emphasize infraspecific or even clonal variability by means of immunochemical or genetical approaches (e.g. Campbell et al., 1989; Anderson, 1990).
How many marine phytoptankton species in the world ocean?
Received on December 17, 1990; accepted on March 20, 1991
Downloaded from by guest on September 11, 2014
Larsen,J. and Sournia.A (1991) The diversity of heterotrophic dinoflagellates. In Patterson.D.J.
and LarsenJ (eds), The Biology of Heterotrophic Flagellates. Oxford University Press, Oxford, in
Lce,J.J., Hutner.S H. and Bovee.E C. (eds) (1985) An Illustrated Guide to the Protozoa. Soc.
Protozool., Lawrence
Levine.N.D. et al. (1980) A newly revised classification of the protozoa. / ProusuA., 27, 37-58
Loeblich.A.R ,111, Loeblich.L.A , Tappan.H. and LoebIich,A.R.,Jr (1968) Annotated index of
fossil and recent silicoflagellates and ebridians with descriptions and illustrations of validly
proposed taxa. Ceol. Soc. Am. Mem., 106.
Margulis.L., Corliss^I.O., Melkonian.M. and Chapman,D.J. (eds) (1990) Handbook ofProtoctista
Jones & Bartlett, Boston.
Parker,S P. (ed.) (1982) Synopsis and Classification of Living Organisms. McGraw Hill, New York., GrainJ. and Mignot,J.-P. (1987) Precis de Prottstologie. Boubee, Pans.
Ricard.M. (19KT) Atlas duphytoplanctonmarin. Vol. 2 Diatomophydes Editions du CNRS, Paris.
Schiller.J. (1930) Coccolithineae. In Rabenhorst's Kryptogamen-Flora von Deutschland, Oesterreich
und der Schweiz, Vol. 10 (2), Akad. Verlag, Leipzig, pp. 89-273.
Schiller^I. (1931-1937) Dinoflagellatae (Peridineae) in monographischer Behandlung. In Rabenhorst's Kryptogamen-Flora von Deutschland, Oesterreichs und der Schweiz, Vol 10 (3)- Teil 1 [ 1 3] (1931-1933); Teil 2 [1-4] (1935-1937). Akad. Verlag, Leipzig.
Sournia.A. (1986) Atlas du phytoplancton marin. Vol. 1 Introduction, Cyanophydes, Dictyochophycies, Dinophyctes et Raphidophycees. Editions du CNRS, Paris
Sournia.A (1991) Catalogue des especes et taxons lnfraspecifiques de Dinoflagelles marins actuels
publics depuis la revision de J.Schiller, V: Compldment Ada Protozoologica, in press.
Steidinger.K.A. (1983) A re-evaluation of toxic dinoflagellatc biology and ecology Di Round,F.E
and Chapman,D.J. (eds), Progress in Phycological Research Elsevier, Bristol. Vol. 2, pp 147188.
Tappan.H. (1980) The Paleobwlogy of Plant Protists. Freeman, San Francisco
TaylorJF.J.R. (1987) General group characteristics; special features of interest; short history of
dinoflagellate study. In Taylor^F.J R. (ed ), The Biology of Dinoflagellates Blackwell Scientific
Publications, Oxford, pp. 1-23.
VanLandingham.L. (1967) Catalogue of the Fossil and Recent Genera and Species of Diatoms and
their Synonyms. A Revision of F.W.Mills' An Index to the Genera and Species of the Diatomaceae
and their Synonyms. Part I. Acanthoceras through Bacillana. Cramer, Lehre
WillisJ.C. (1922) Age and Area a Study in Geographical Distribution and Origin of Species
Cambridge University Press, Cambridge.