A biological continuum

George Garrity george_garrity at MERCK.COM
Mon Mar 20 17:02:32 CST 1995

                       Subject:                               Time:11:26 AM
  OFFICE MEMO          A biological continuum                 Date:3/20/95
The current debate on Latin binomials has proven to be most interesting.
Microbiologists must also contend with a similar set of naming conventions and
rules.  I think the problem that biologists in general face is how to adapt an
18th century code of nomenclature to late 20th century classification

There has been much debate about replacing the hierarchical binomial system
with a numbering system.  I agree that a numerical code that consists of
either arbitrary or sequential numbers (regardless of base), cannot and will
not suffice.  However, there are numerical nomenclature systems in existance
that have proven to be quite successful outside of biology.  Some examples
are: the Dewey decimal system, postal codes, telephone numbers with area and
country codes.  These have all stood the test of time as practical and useful
approaches of identifying objects that have been classified.  These systems
are, however, categorical.  They do not necessarily provide an indication of
proximity of one item to the next, which is what our current systems of
biological nomenclature attempt to do.  Chemists also have an interesting
approach (IUPAC naming conventions) that uses a hybrid set of numerical and
character names to describe a potentially infinite array of compounds.  Yet
another approach that everyone is familiar with is the coordinate systems used
in cartography and navigation.  Not only can this approach provide an
indication of proximity, it can also provide information about direction (a
simple map), and elevation (topographic   maps).  By adjusting scale, one can
either look at small or extremely large areas in readily comprehensible terms
that can be easily visualized.

We've been experimenting with the latter approach in our lab.  We are working
with large, multivariate data sets (> 10,000 strains).  Providing that
"benchmarks" can be suitably placed within the "biological terrain", we can
place groups of organisms into a taxonomic map and estimate distance and
species density in a readily comprehensible form.  What we are learning from
this approach is that the species of microorganisms we normally refer to are
merely points along a biological continuum.  I rather suspect that higher
organisms will behave in an analogous manner, provided that a sufficiently
large sampling is done with enough characteristics.  Rather than focus on
names, we can now pinpoint the location of an individual and how it relates to
the larger population of presumably related or unrelated organisms.  We can
refer to it by coordiates, "locale", or neighbors of known identity.  We
needed system like this because we ran out of names a long time ago and we
routinely ecounter new species as a by-product of our work.

While it is uncertain whether this approach is any more legetimate than Latin
binomials it does provide a means of visualizing large amounts of taxonomic
information and "navigating" through uncharted terrain in prospective studies
of microbial populations.  Is it perfect ? No.  Will the map change ? By all
means.  However, by updating the map as soon as new infomation becomes
available we can get an idea of where we are going and where we have been.
It is offered here simply as an alternative to those who would beleive there
are none.

George Garrity
Natural Products Research
Merck Reserach Labs
Rahway, NJ 07065

More information about the Taxacom mailing list