Emilio asks about subspecies. Okay, here is the technical aspect.
 
The International Code of Zoological Nomenclature governs only three levels
of taxa: species-level taxa, genus-level taxa, and family-level taxa.
Within each level, names "compete for priority", that is, the oldest valid
name is the correct one, regardless of rank within the level. A subspecies
name can compete successfully with a species name.
 
Species-level taxa include subspecies, species, and superspecies.
Genus-level taxa include subgenera, genera, and supergenera. Family-level
taxa... well, you get the idea. There are also some extra terms like
"tribe" that can be tucked into the system when people are dealing with a
very finely divided group like the ants. The higher taxa (orders, classes,
phyla, kingdoms) are not governed by the Code, but are sort of a
free-for-all that is governed by good sense and good taste. It's all
voluntary, but it's like obeying the rules of a game: No one will want to
play with you if you don't follow the rules.
 
By naming a species, an author automatically names the corresponding
subspecies and superspecies. For example, if Lynn Scheu named a new species
of flying cowry, Coca cola Scheu, 1997, she would automatically be
considered the author of Coca cola cola Scheu, 1997, even if the subspecies
name were never printed or used. The subspecies has the same type specimen
as the species of the same name. This becomes interesting information only
when someone, let's say Jerry Brown, later names another subspecies, C.
cola californiana Brown, 1998. There are similar rules for genus-level and
family-level taxa.
 
In the usual course of events, the most conspicuous or most typical
subspecies of C. cola would be named first, leading to the situation that
Emilio noted when he wrote about out-of-the-way subspecies being of
"inferior" rank. But this isn't an ideal world, and sometimes the most
common and "central" subspecies is named second. This is too bad, but
there's nothing that can be done to correct the situation as long as we
obey the rule of priority. The earliest valid name has priority over all
junior synonyms and homonyms.
 
As to varieties and forms, they are not recognized by the Code. That is, if
you name a new form or variety in print today, no one will pay much
attention to it. Of course, people were naming forms and varieties long
before the Code was written, so forms and varieties named before the cutoff
date of January 1, 1961 are automatically considered to be subspecies.
"Forms" and "varieties" are almost interchangeable, but not quite. In a few
cases, an author named forms within varieties within species, and this
poses a problem. The Code solves this problem by saying that forms named
within varieties are below the subspecies rank and are therefore not
recognized by the Code. You can name them if you wish, and some people do,
but they do not "compete for priority" or for homonymy and are not
protected by the Code. Most people ignore them.
 
Why do we bother with subspecies? The simple answer is to say that it
allows a taxonomic rank below the species, and that some people find it
useful occasionally to split a species up. It's enough to say that having
an extra rank adds flexibility to the system. The complex answer involves
significant, consistent differences between populations that don't
ordinarily interbreed (but could if they were put together), geographic
separation, ecologic separation, and/or temporal separation, and a lot of
arguments over what to use in what group and when, and frankly I'd rather
leave it to someone else. I'm going to dance instead.
 
Andrew K. Rindsberg
Geological Survey of La Bamba