
          25.  Whatever method you adopt or lean to, or even should you frame a new
one, remember that it will be needful to compare & quote the labours of others,
on families & orders, particularly in all the doubtful, disputed or shifting
families, that are not yet properly fixed.  But do not follow the blunder of
Jussieu, who had Genera of incertadedis, belonging to no families!  There is not
a single Genus, but what finds its place in the Natural Method.!

26.  There are families or Orders so distinct and well characterised, that
they admit of no doubt.  Such are the Graminea, Cyperacea, Filices, Musci
Umbellifera, Crucifera, Malvacea, Orchidea, Labiata etc, which are in fact
Natural Orders containing subordinate Natural families.

27.  But there are other more limited or not so well distinguished, that have
often been rendered more obscure, by adding there to genera of disparate characters.
It is those Nat. [Natural] families that require the utmost attention & investigation.  A
very good plan is to take a well known genus as the type & annex thereto the
akin general having common characters.  Such types are found in Viola, Linum,
Dionea, Menispermum, Podophyllum, etc.

28.  But beware of annexing Genera from mere habit, or without adequate
common characters of an obvious nature: this is yet the stumbling block of
many botanists.  Thus, Consolida does not belong to Delphinium nor the Ranunculacea,
nor Botrophis to Actea nor Cimicifuga! not even congeneric, how can
they be of the same Order?  or since of different family types how can they be congeneric?
If you are in doubt about any Genus true affinities, I will easily
clear it, provided the G. has been well described.

29.  It is not necessary to seek for [crossed out:obvious] [added:recondite] characters in the seeds & embryos
there are always corresponding characters by far more [added:obvious &] evident in the flowers &
inflorescens, habit etc that normally guide but fix the place in the Natural System.

30.  The essential axiom stands good for families as for general.  No genus &
no Species ought to be put in a family or genus of dissimilar habit & character
combined; or even if the character is different, [crossed out:even] should the habit agree

31.  Many Botanists have followed Linneus in making genera from mere
habit, overlooking the characters:  this was an oversight of his, following the
plan of the Old Botanists, he took a Sp for type of a genus, & refered other Sp. to it
by mere habit, without noticing the differences.  Thus we had once 100 Genera
that have been split on review whenever improving Botanists have arisen such as
Geranium, Orchis, Fumaria etc. but we have as many more that need the same
reform, Veronica, Scabiola, Commelina, Aristolochia, Polygala, Euphorbia etc.

32.  A single G. [genus] of Linneus Cances in Zoology is now become a class.  Many
of his general of insects, birds, fishes etc are become Natural Orders.  In Botany
the same with Lycopodium, etc.

33.  It is a wrong principle that numbers ought to guide in forming genera, this
was a mistake of Lamark.  He thought all extensive genera ought to be split upon
trifling characters, & genera of a few species let along even if they had anomalous
species.  It is obvious that such a paradox is destructive of all correct occurences.
Yet it has had much influence in some cases, and may partly guide us, as it really
does among the Umbellifera, Crucifera, Graminea etc which might be deemed
single genera with more propriety than Convallaria, Plygala, Carex etc.

34.  It will be difficult to explain why the Cyperacea & Graminea have been
        