52
Basic biogeography: estimating biodiversity and mapping nature
discoveries and standardize nomenclature (Table 4.1 ).
One example is the Partnership for Enhancing Expertise
in Taxonomy (PEET) ( www.nsf.gov/funding/pgm_
summ.jsp?pims_id = 5451 & org = BIO ), which aims to
target poorly known organisms through supporting
research projects, thereby providing opportunities for
a new generation of taxonomists to be trained in ‘ problematic ’ taxa. A further aim of PEET is to translate
current taxonomic expertise into electronic databases
and other formats that allow broader accessibility.
This fi nal point is exceedingly important. Taxonomic
information is only useful for conservation if it is in a
form that can be easily retrieved and processed.
Technological advances and efforts from numerous
institutions around the world are making such
‘ user - friendly ’ open access databases a reality and are
providing conservationists with powerful new tools
for mapping and understanding nature. Probably
the most ambitious bioinformatics project is the
Encyclopedia of Life ( www.eol.org ), a project inspired
by E.O. Wilson, the aim of which is to ‘ make available
via the Internet virtually all information about life
present on Earth ’ (Wilson, 2003 ). The encyclopaedia
works through a series of linked websites, one of which
is planned for every species that has been formally
described. Each species ’ website will be fl exible and
constantly evolving so that it can easily incorporate
new information on ecology, genetics and conservation as it is generated. By 2014, the project hopes to
have generated a million species pages – a rich resource
for conservation biogeography if it can improve access
to knowledge and improve the quality, accuracy and
speed of data collection. However, some scientists have
questioned the feasibility of such goals in the light of
declines in the number of trained taxonomists and in
resources for both taxonomy and the curation of collections (Gropp, 2003 ).
Serious attempts are also under way to develop computer programs that can identify species from digital
images through the use of a new generation of evolutionary algorithms (Gaston & O ’ Neill, 2004 ). In principle it may be possible to ‘ train ’ identifi cation software
to recognize species from images and, by extension,
identify possible new species for which no records exist.
As we continue to add molecular approaches to
more traditional morphological ones (and behavioural,
physiological, etc.) of recognizing signifi cant breaks
and relationships between ‘ natural groups ’ , we remain
uncertain about how to apply either molecular, morphological, or a combination of both kinds of information in deciding how to diagnose different species of
common pattern, with larger, more conspicuous
species typically being scientifi cally discovered relatively early. Moreover, new species discoveries are often
now restricted to a few geographically localized areas
of the globe where taxonomic capacity has been low,
or where local conditions have made collecting diffi cult
(e.g. remote rain forest, mountains, swamplands, etc.).
A large source of uncertainty in global species richness estimates derives from taxonomic error in species
already described. One common problem is synonymy
– multiple naming of the same species – which is
known to be rife in many groups. For example, due to
taxonomic problems, estimates of the number of
known mollusc species vary from 45,000 to 150,000,
with 70,000 a compromise fi gure (Jeffries, 1997 ). As
new forms of taxonomic analysis become available and
more molecular phylogenies are produced, and as specialist monographers revise their groups, improved
precision in diversity estimates is possible.
As hinted in the Canary Islands ’ beetle example
above, taxonomic adjustments can, of course, go in
both directions. In other words, revisions can result in
multiple - named species being lumped together, but
may also sometimes result in what was thought to be
one species being split into separate species. It can even
sometimes happen that species go in and then out of
synonymy again.
In his analysis of neotropical mammals, Patterson
(2000) noted that between 1982 and 1992, three
times as many species in this group came out of synonymy as became synonyms, and these revalidated
names also outnumbered species being described for
the fi rst time by the same ratio. While some of the
newly described species were essentially newly trapped,
for every such species, Patterson notes that three more
were ‘ found ’ in drawers in museums or identifi ed via
molecular biological analysis. The numbers involved
are surprisingly large for a group of vertebrates generally considered to be well described. Patterson notes
that in the seven years prior to his study, a total of 57
species of neotropical mammals were described, ten
more than the number of bird species described globally in the decade from 1981 to 1990. This indicates
that our knowledge of mammal diversity (and distribution) is probably less satisfactory than for birds
(Patterson 2000, 2001 ).
The resources and expertise needed to resolve all of
these uncertain taxonomies are considerable and there
appears to be limited political will to provide the
funding needed. Nonetheless, there are a number of
initiatives under way that may speed up the rate of new
Basic biogeography: estimating biodiversity and mapping nature
discoveries and standardize nomenclature (Table 4.1 ).
One example is the Partnership for Enhancing Expertise
in Taxonomy (PEET) ( www.nsf.gov/funding/pgm_
summ.jsp?pims_id = 5451 & org = BIO ), which aims to
target poorly known organisms through supporting
research projects, thereby providing opportunities for
a new generation of taxonomists to be trained in ‘ problematic ’ taxa. A further aim of PEET is to translate
current taxonomic expertise into electronic databases
and other formats that allow broader accessibility.
This fi nal point is exceedingly important. Taxonomic
information is only useful for conservation if it is in a
form that can be easily retrieved and processed.
Technological advances and efforts from numerous
institutions around the world are making such
‘ user - friendly ’ open access databases a reality and are
providing conservationists with powerful new tools
for mapping and understanding nature. Probably
the most ambitious bioinformatics project is the
Encyclopedia of Life ( www.eol.org ), a project inspired
by E.O. Wilson, the aim of which is to ‘ make available
via the Internet virtually all information about life
present on Earth ’ (Wilson, 2003 ). The encyclopaedia
works through a series of linked websites, one of which
is planned for every species that has been formally
described. Each species ’ website will be fl exible and
constantly evolving so that it can easily incorporate
new information on ecology, genetics and conservation as it is generated. By 2014, the project hopes to
have generated a million species pages – a rich resource
for conservation biogeography if it can improve access
to knowledge and improve the quality, accuracy and
speed of data collection. However, some scientists have
questioned the feasibility of such goals in the light of
declines in the number of trained taxonomists and in
resources for both taxonomy and the curation of collections (Gropp, 2003 ).
Serious attempts are also under way to develop computer programs that can identify species from digital
images through the use of a new generation of evolutionary algorithms (Gaston & O ’ Neill, 2004 ). In principle it may be possible to ‘ train ’ identifi cation software
to recognize species from images and, by extension,
identify possible new species for which no records exist.
As we continue to add molecular approaches to
more traditional morphological ones (and behavioural,
physiological, etc.) of recognizing signifi cant breaks
and relationships between ‘ natural groups ’ , we remain
uncertain about how to apply either molecular, morphological, or a combination of both kinds of information in deciding how to diagnose different species of
common pattern, with larger, more conspicuous
species typically being scientifi cally discovered relatively early. Moreover, new species discoveries are often
now restricted to a few geographically localized areas
of the globe where taxonomic capacity has been low,
or where local conditions have made collecting diffi cult
(e.g. remote rain forest, mountains, swamplands, etc.).
A large source of uncertainty in global species richness estimates derives from taxonomic error in species
already described. One common problem is synonymy
– multiple naming of the same species – which is
known to be rife in many groups. For example, due to
taxonomic problems, estimates of the number of
known mollusc species vary from 45,000 to 150,000,
with 70,000 a compromise fi gure (Jeffries, 1997 ). As
new forms of taxonomic analysis become available and
more molecular phylogenies are produced, and as specialist monographers revise their groups, improved
precision in diversity estimates is possible.
As hinted in the Canary Islands ’ beetle example
above, taxonomic adjustments can, of course, go in
both directions. In other words, revisions can result in
multiple - named species being lumped together, but
may also sometimes result in what was thought to be
one species being split into separate species. It can even
sometimes happen that species go in and then out of
synonymy again.
In his analysis of neotropical mammals, Patterson
(2000) noted that between 1982 and 1992, three
times as many species in this group came out of synonymy as became synonyms, and these revalidated
names also outnumbered species being described for
the fi rst time by the same ratio. While some of the
newly described species were essentially newly trapped,
for every such species, Patterson notes that three more
were ‘ found ’ in drawers in museums or identifi ed via
molecular biological analysis. The numbers involved
are surprisingly large for a group of vertebrates generally considered to be well described. Patterson notes
that in the seven years prior to his study, a total of 57
species of neotropical mammals were described, ten
more than the number of bird species described globally in the decade from 1981 to 1990. This indicates
that our knowledge of mammal diversity (and distribution) is probably less satisfactory than for birds
(Patterson 2000, 2001 ).
The resources and expertise needed to resolve all of
these uncertain taxonomies are considerable and there
appears to be limited political will to provide the
funding needed. Nonetheless, there are a number of
initiatives under way that may speed up the rate of new