Nov., 1923] 
HITCHCOCK-TYPE SPECIES 
511 
The remaining 72 genera, containing originally more than one species, 
must be typified by selecting one of the species as the type. It is this 
retroactive application of the type concept that has aroused the most 
opposition from the adherents of the International Rules because of the 
fear that well-established names would be needlessly displaced. I think 
I can demonstrate that the results are not revolutionary and that in the 
main the fears are groundless. A uniform application of any set of rules will 
bring to light certain anomalous cases which in some way should be con¬ 
sidered and either corrected or validated. 
One of the first things to do in selecting a type is to exclude from con¬ 
sideration those species that definitely disagree with the generic description, 
because certainly an author would not illustrate or typify his own genus 
with an anomalous species. As there are no generic descriptions in the 
Species Plantarum, the descriptions in the fifth edition of the Genera 
Plantarum (1754) are used to ascertain Linnaeus’ concept of the genera 
included in the Species Plantarum. The typification on the basis of the 
generic description should be done only by those familiar with the taxonomy 
of the groups. I have not attempted to do this except with the grass 
genera. As an illustration, we have the genus Holcus which does not occur 
in the first 100 genera now under consideration. The generic description 
in the Genera Plantarum certainly applies only to three of the seven species 
included in the Species Plantarum, the three species related to Holcus 
Sorghum which were later segregated as the genus Sorghum. The applica¬ 
tion of the type method here goes contrary to historical development sub¬ 
sequent to Linnaeus and to the general usage of those who recognize the 
group as distinct from Andropogon. In current usage the genus is repre¬ 
sented by Holcus lanatus. 
It is rather exceptional among Linnaean genera to find species definitely 
excluded in this way from consideration in selecting the type. It is possible 
that on taxonomic grounds there may be a few changes in the tentative list 
of type species here presented. 
The next point to ascertain is which species the author of the genus 
appeared to have chiefly in mind, in so far as one species can be singled out. 
We may assume, unless there is evidence to the contrary, that the representa¬ 
tive species to the author would be the one best known to him. This may 
be shown in four ways: First, one to which he has applied a specific name 
like officinalis , communis , vulgaris , or sativus; second, a well-known economic 
species; third, a common species of the native flora or one grown by him in 
a garden; fourth, through a citation in the Genera Plantarum. Another 
method, the selection on the basis of figures accompanying the original 
description, can not be used here because there are no plates in the Species 
Plantarum. These four methods are used coordinately. Sometimes one 
can be applied, sometimes another. Often two or more methods lead in 
the same direction, as in Hordeum, of which H . vulgare is selected as the 
