Notices respecting New Books. 65 
allied in habit. The characters employed by Mr. Brown are 
said to be easy of detection as soon as the skin of the seed is 
removed, there being no separate albumen ; and these afford 
the most natural, and indeed absolute, primary characters of 
these plants. ‘‘ They serve,” says our author, “ to divide the 
whole into great natural sections, liable, as far as I can find, 
to no exception; the genera under each section being easily 
characterized, and proving much more natural, in habit and 
fructification, than those found by Linnzeus.” 
Whatever objection may, at first sight, appear against the 
‘use of these characters in the cotyledons, as furnishing little 
artificial assistance to the tyro, they are invaluable in the ab- 
sence of more obvious marks, and confirm the empirical 
knowledge of habit and look, which we pointed out at the 
commencement of this paper as so much needed when we can 
no longer detect characters which can be described by bota- 
nical terms. 
Maithiola incana is admitted here; but surely it is an es- 
cape from the gardens. At Hastings even double flowers may 
be observed. The Malva pusilla of Engl. Bot. is here reduced 
to a variety of rotundifolia. ‘The Orobus tenuifolius of Roth, 
which Mr. D. Don had found in Scotland, Mr. Peete in 
Kent, and to these may be added, by ourselves in Glamor- 
ganshire, is regarded (and we think rightly) only as a variety 
of tuberosus. Vicia angustifolia of Sibthorp and others is in- 
troduced, and is no doubt a well-marked species. Lotus de- 
cumbens, an addition of Mr. 'T. F. Forster’s in his Flora 
Tonbridgensis, is also new; while LZ. diffusus turns out to be 
angustifolius of Linneeus. Medicago maculata, muricata, and 
minima, first noticed by our author in the Cyclopzedia, were 
before included in M. polymorpha. 
In Syngenesia all the old Hedypnoides are placed under the 
genus Apargia.. The Linnean and Jussieuian genus Cnicus 
embraces many of our Cardui: but the author thinks the se- 
paration of these two genera justifiable only on the ground of 
convenience, and that they are not naturally separate. Cnicus 
Forsteri, (another discovery of our late estimable friend T. F. 
Forster, Esq., whose inquisitive eye seldom suffered a good 
plant to escape him,) if not absolutely distinct, is a sin- 
gular hybrid, perhaps between palustris and pratensis. San- 
tolina is now Diotis, upon the authority of Desfontaines and 
De Candolle. Under Doronicum Pardalianches our author 
does not quote the figure in the new series of the Flora Lon- 
dinensis ; ond in this we think he is right, the plant there re- 
presented being plantagineum, which is, with the other, an oc- 
casional escape from gardens, as we have evidence from the 
Vol. 67. No. 333. Jan. 1826, I very 
