284 CONTRIBUTIONS FROM THE NATIONAL HERBARIUM. 
for there was a still older name Callicarpa that had been applied by 
Linnzus himself to another genus of flowering plants from the same 
regions as the sapote. It is true that the forms of the name used by 
Linnzus and Pierre, ‘“Callicarpa” and ‘“Calocarpum,”’ are capable of 
bibliographic discrimination, but essentially they are merely variants 
of the same word. Such names do not differ as words, but merely 
as combinations of letters. Instead of aiding in the recognition of 
plants these ambiguous designations serve rather to confuse them. 
Some writers have proposed to admit variations of spelling, or even 
typographical errors, as constituting distinct names. The danger of 
this tendency finds a striking illustration in the present instance. 
There would be no need to stop with a few variations like Callicarpa 
and Calocarpum, for Kallikarpon, Kalokarpus, and many others are 
possible. Indeed, this name is capable of no less than 64 variations 
of spelling, to say nothing of the possibilities of developing some 
genuine typographical errors. Each family of plants might have its 
Calikarpum or Kalocarpon, or a whole family might be provided 
with generic names based on the same flexible combination. 
A NEW GENERIC NAME FOR THE SAPOTE. 
Hence, it appears that no satisfactory generic designation is avail- 
able in literature for the sapote, notwithstanding the many names 
that have been applied to it during the past two centuries. As no 
combinations of the word Achras are known to have been used hitherto, 
the name ‘‘Achradelpha,” recently proposed,! may avoid the danger 
of homonymy. Allusion to the sapote as the ‘‘sister of Achras”’ is 
warranted by the fact of similarity which has been the occasion of so 
much confusion. 
The type species of Achradelpha is Achradelpha mammosa, based 
on Achrasmammosa L. Some might hold that the specific name pre- 
viously used by Jacquin in the binomial Siderorylum sapota should 
be revived under the new genus, instead of adopting the Linnean 
name mammosa. This is one of the cases where the rule ‘‘Once a 
homonym, always a synonym” would find a useful application. 
Though it may not seem likely in the present case that the sapote 
will ever be referred back to Achras and thus cause a direct conflict 
of homonyms, if the alternative combination were adopted, there 
are other cases of more closely related genera where confusion would 
be created if the names of species were to be altered with each change 
of generic assignment. This course becomes necessary unless names 
that have been subject to rejection as homonyms are permanently 
discarded. Moreover, it is undesirable to have the same specific 
? An advance summary of the results of this study has been published in the Journal 
of the Washington Academy of Sciences, 3: 158, March 19, 1913. 
