Daniel Mosquin posted a link to this document in a Botany Photo of the Day posting, and I thought it was worth saving in this forum. On The Use Of The Terms Subspecies And Variety It's a long read. Daniel also quoted the definition from the USDA’s National Plant Materials Program Manual (4th ed.): The terms “subspecies” and “variety” are used to designate the first and second divisions of a species. A “subspecies” is a grouping within a species used to describe geographically isolated variants, a category above “variety”, and is indicated by the abbreviation “subsp.” in the scientific name. A “variety” consists of more or less recognizable entities within species that are not genetically isolated from each other, below the level of subspecies, and are indicated by the abbreviation “var.” in the scientific name.
Christensen (1987; Nordic J. Botany 7: 383-408; see p.384) published a set of morphology-based guidelines for selection of rank, which I have long found extremely useful and apt:
Worth adding that in the USA in the past, particularly by USDA forestry botanist E. L. Little (and still even now by some authors who treat Little as deified), any use of the rank of subspecies at all was rejected; this has led to many clear subspecies (under all the above criteria) being treated as just varieties. Notable examples include in Douglas-fir Pseudotsuga menziesii (it would be hard to find a better example of a subspecies than P. m. "var." glauca), Lodgepole Pine Pinus contorta, and Ponderosa Pine Pinus ponderosa. The latter two are increasingly being treated with subspecific divisions following Critchfield's monograph on P. contorta and Callaham's recent work on P. ponderosa, but all three are still all too frequently only cited as divided at varietal rank. A similar pair of cases in Abies 'suffering' from the same rejection of the rank of subspecies, is found with A. concolor and A. lasiocarpa. Traditionally they were treated as divided into two varieties each, but in his contribution to the Flora of North America, R. S. Hunt realised they were more than varieties, though was unable or unwilling to consider use of the rank of subspecies, so split them as species (A. concolor / A. lowiana, and A. lasiocarpa / A. bifolia, respectively). A far better treatment is to have them as A. concolor subsp. lowiana, and A. lasiocarpa subsp. bifolia, respectively (and also A. lasiocarpa subsp. arizonica for the 'Corkbark Fir' in Arizona). Conversely, the use of subspecies has often been overdone in Europe, with e.g. Pinus nigra divided by some authors into eight or more subspecific names, when both the genetic and morphological evidence shows a clear divide into two major groups (best treated as subspecies), each subspecies then further divided into varieties (see its treatment at the Gymnosperm Database). I look forward to the day when a global consensus is reached on the usage of subspecies rank, but I fear it will be a long time coming yet.
