ersonal
interest to him.

21. I have used the word 'Order' as the name of our widest groups, in
preference to 'Class,' because these widest groups will not always include
flowers like each other in form, or equal to each other in vegetative rank;
{189} but they will be 'Orders,' literally like those of any religious or
chivalric association, having some common link rather intellectual than
national,--the Charites, for instance, linked by their kindness,--the
Oreiades, by their mountain seclusion, as Sisters of Charity or Monks of
the Chartreuse, irrespective of ties of relationship. Then beneath these
orders will come, what may be rightly called, either as above in Greek
derivation, 'Genera,' or in Latin, 'Gentes,' for which, however, I choose
the Latin word, because Genus is disagreeably liable to be confused on the
ear with 'genius'; but Gens, never; and also 'nomen gentile' is a clearer
and better expression than 'nomen generosum,' and I will not coin the
barbarous one, 'genericum.' The name of the Gens, (as 'Lucia,') with an
attached epithet, as 'Verna,' will, in most cases, be enough to
characterize the individual flower; but if farther subdivision be
necessary, the third order will be that of Families, indicated by a 'nomen
familiare' added in the third place of nomenclature, as Lucia
Verna,--Borealis; and no farther subdivision will ever be admitted. I avoid
the word 'species'--originally a bad one, and lately vulgarized beyond
endurance--altogether. And varieties belonging to narrow localities, or
induced by horticulture, may be named as they please by the people living
near the spot, or by the gardener who grows them; but will not be
acknowledged by Proserpina. Nevertheless, the arbitrary reduction under
Ordines, Gentes, and Familiae, {190} is always to be remembered as one of
massive practical convenience only; and the more subtle arborescence of the
infinitely varying structures may be followed, like a human genealogy, as
far as we please, afterwards; when once we have got our common plants
clearly arranged and intelligibly named.

22. But now we find ourselves in the presence of a new difficulty, the
greatest we have to deal with in the whole matter.

One new nomenclature, to be thoroughly good, must be acceptable to scholars
in the five great languages, Greek, Latin, French, Italian, and English;
and it must be acceptable by them in teaching the native children of each
country. I shall not be satisfied, u