The boletes of Michigan, by Alexander H. Smith and Harry D. Thiers.

18

THE BOLETES OF MICHIGAN

If we consider the other genera in the order in which they appear in the
text, Suillus should be considered first.
If any genus is to be judged on the basis of its combination of
characters, Suillus is, next to Boletus itself, the most distinctive genus in
the family. Perhaps it is pertinent to list some guidelines as to what
should be required in the way of characters or combinations of them to
justify erecting a genus, or recognizing one previously described. This
approach, however, is rather mechanical. Our aim should be to arrive at
a classification which is "natural" in the sense that the most recently
evolved (related by descent from common ancestors) species should be
placed together in a group. We have a number of categories in our
system of classification: Order, family, genus, species, subspecies, variety, and form. Each of these above the rank of species may be divided
into subgroups just as is indicated for species: Subgenus, section, subsection series, and stirps, for instance, under the heading of genus. The
smallest group which we recognize above the rank of species is the stirps
(pl. stirpes). In it we place a central species along with those obviously
very similar to it. The stirps bears the name of the central species. It is
understood, when the term stirps is used, that the author or authors
regard the group as having common parentage. We often refer to a stirps
as a central species with its satellites. The stirpes are then grouped into
subsections, sections; and subgenera, depending on the degree of differences shown by the members of each group. Such a classification is by
its plan of organization a natural one, at least in the eyes of those who
propose it. But if we regard a classification from this point of view, it is
at once evident that what is really important in establishing a genus is
not "a bundle of characters," but rather what value the characters have
in actually indicating the relationships of the fungi involved. This necessitates assigning different values to various characters since some are
better indicators of relationship than others. One might define a genus,
then, as a major segment of evolution (or "line") as this is measured by
the populations that we find in nature and that exhibit the combinations of index characters we believe important in assessing degrees of
relationship. This is why, generally, genera are so different as regards
number of species: Cortinarius with nearly 800 species in North
America, Pulveroboletus with 1-and it is at this point that sharp differences of opinion as to what constitutes a genus are encountered.
We have tried to arrive at a reasonably well-balanced approach to
this problem, which means that we recognize as genera the more distinctive groups of obviously related species in the Boletaceae according to
an evaluation of characters that can be applied on a broad base throughout the fleshy fungi.