22
Chapter 2: Biogeographic Partition of the Ocean
about its distribution, it would be necessary carefully to consult the synonymies proposed
by Alvarino. Unfortunately, comprehensive and scholarly reviews like these (S. J. Gould
called them “works of genius”) cover but a small fraction of the total marine biota. The
present, start-up phase of electronic taxonomic data archives seems not to understand that
unless critical taxonomy of this kind has been done, the simple listing of species that have
been described in a genus is at best useless. There appears to be inadequate recognition
of how widely the views of taxonomists may vary: Boltovskoy long ago discussed a case
in which identical sets of specimens of benthic foraminifera, each apparently containing
200 species, were submitted to four taxonomists who were specialists in that group. The
resulting species lists shared only 10 generic names, and but one species. Only those who
have hands-on experience of trying to establish regional species lists can appreciate how
general is the problem of the unconsolidated nomenclature of marine organisms.
But everything concerning the species concept according to Huxley and Mayr has
now potentially been turned on its head. The molecular characterization of individual populations by sequence analysis of subunits of mitochondrial RNA and DNA
has become routine and is widely used in some fields. This technique, supported by
statistical techniques for analyzing relationships between clades, results in neighborjoining or maximum-parsimony trees and other presentations that give taxonomists a
new ability to quantify relationships between populations. This school cleaves to a very
restrictive species concept, suggesting that “the smallest diagnosable cluster of individual
organisms within which there is a parental pattern of ancestry and descent” represents
a “phylogenetic species” (Cracraft, 1983). For Nixon and Wheeler (1990), these are “the
smallest aggregation of populations (sexual) or lineages (asexual) diagnosable by a unique
combination of character states in comparable individuals.” Phylogenetic species are thus
quite distinct from the morphological or nominate species with which we have been
familiar in the past.
A typical example of the kind of problem that phylogenetic taxonomy will induce in
our taxonomic understanding is given by a recent ribosomal RNA analysis of 41 specimens
of the small mesopelagic fish Cyclothone alba, which is ubiquitous and abundant in
all subtropical and tropical oceans (Miya and Nishida, 1997). This analysis identified
five monophyletic populations with low levels of mutual gene flow under conditions in
which there appears to be no discernible barriers to prevent complete dispersion and
intermingling of stocks. The central North Pacific population is genetically closer to
those of the North Atlantic than to the three populations of the Indo-Pacific Ocean.
I note, without comment, that the Web-based Oceanic Biogeographic Information System
(OBIS) associated with the multinational Group on Earth Observations (GEO) offers a
map to represent the global distribution of Cyclothone alba: this shows only 25 North
Atlantic locations, being based on the data and specimen holdings in national and
institutional data banks—there is, apparently, no reference to the scientific literature that,
of course, comprises the majority of our knowledge about this fish.
Consider also the problems that would arise following the application of the phylogenetic species concept to the genus Calanus, whose taxonomy—as I suggested earlier—we
thought was beginning to settle down. Recently, molecular systematics using variations in
the DNA base sequence of the mitochondrial 16S rRNA gene (Bucklin et al., 1995) have
been used to explore the systematics of Calanus, as of other organisms. This approach has
confirmed the reality of the relationship Calanus finmarchicus + glacialis + marshallae
and also of the more genetically diverse C. helgolandicus group, though genetic information on C. orientalis was lacking. Further, the same technique distinguishes between
populations of Calanus finmarchicus from different regions of the North Atlantic: fifteen
samples of C. finmarchicus were thus grouped into four populations, one each in the
Norwegian Sea, in the Gulf of Maine, on Georges Bank, and in the Gulf of St. Lawrence.
Given the oceanic distribution of Calanus, of which these samples were marginal, one may
Chapter 2: Biogeographic Partition of the Ocean
about its distribution, it would be necessary carefully to consult the synonymies proposed
by Alvarino. Unfortunately, comprehensive and scholarly reviews like these (S. J. Gould
called them “works of genius”) cover but a small fraction of the total marine biota. The
present, start-up phase of electronic taxonomic data archives seems not to understand that
unless critical taxonomy of this kind has been done, the simple listing of species that have
been described in a genus is at best useless. There appears to be inadequate recognition
of how widely the views of taxonomists may vary: Boltovskoy long ago discussed a case
in which identical sets of specimens of benthic foraminifera, each apparently containing
200 species, were submitted to four taxonomists who were specialists in that group. The
resulting species lists shared only 10 generic names, and but one species. Only those who
have hands-on experience of trying to establish regional species lists can appreciate how
general is the problem of the unconsolidated nomenclature of marine organisms.
But everything concerning the species concept according to Huxley and Mayr has
now potentially been turned on its head. The molecular characterization of individual populations by sequence analysis of subunits of mitochondrial RNA and DNA
has become routine and is widely used in some fields. This technique, supported by
statistical techniques for analyzing relationships between clades, results in neighborjoining or maximum-parsimony trees and other presentations that give taxonomists a
new ability to quantify relationships between populations. This school cleaves to a very
restrictive species concept, suggesting that “the smallest diagnosable cluster of individual
organisms within which there is a parental pattern of ancestry and descent” represents
a “phylogenetic species” (Cracraft, 1983). For Nixon and Wheeler (1990), these are “the
smallest aggregation of populations (sexual) or lineages (asexual) diagnosable by a unique
combination of character states in comparable individuals.” Phylogenetic species are thus
quite distinct from the morphological or nominate species with which we have been
familiar in the past.
A typical example of the kind of problem that phylogenetic taxonomy will induce in
our taxonomic understanding is given by a recent ribosomal RNA analysis of 41 specimens
of the small mesopelagic fish Cyclothone alba, which is ubiquitous and abundant in
all subtropical and tropical oceans (Miya and Nishida, 1997). This analysis identified
five monophyletic populations with low levels of mutual gene flow under conditions in
which there appears to be no discernible barriers to prevent complete dispersion and
intermingling of stocks. The central North Pacific population is genetically closer to
those of the North Atlantic than to the three populations of the Indo-Pacific Ocean.
I note, without comment, that the Web-based Oceanic Biogeographic Information System
(OBIS) associated with the multinational Group on Earth Observations (GEO) offers a
map to represent the global distribution of Cyclothone alba: this shows only 25 North
Atlantic locations, being based on the data and specimen holdings in national and
institutional data banks—there is, apparently, no reference to the scientific literature that,
of course, comprises the majority of our knowledge about this fish.
Consider also the problems that would arise following the application of the phylogenetic species concept to the genus Calanus, whose taxonomy—as I suggested earlier—we
thought was beginning to settle down. Recently, molecular systematics using variations in
the DNA base sequence of the mitochondrial 16S rRNA gene (Bucklin et al., 1995) have
been used to explore the systematics of Calanus, as of other organisms. This approach has
confirmed the reality of the relationship Calanus finmarchicus + glacialis + marshallae
and also of the more genetically diverse C. helgolandicus group, though genetic information on C. orientalis was lacking. Further, the same technique distinguishes between
populations of Calanus finmarchicus from different regions of the North Atlantic: fifteen
samples of C. finmarchicus were thus grouped into four populations, one each in the
Norwegian Sea, in the Gulf of Maine, on Georges Bank, and in the Gulf of St. Lawrence.
Given the oceanic distribution of Calanus, of which these samples were marginal, one may