NOTICES OF BOOKS.  , 293 
fate of such books is ultimate neglect, for we hold it as a well-established - 
law, that every one who writes upon a subject is not only bound to take 
every possible pains to acquire a perfect knowledge of all that has been 
written previous to his own attempts in the same line, but must do so 
if he has any regard for his reputation. | 
Another very important point presented by this work is the full and - 
ample commentary which it presents, both in the way of remark and 
ilustration, on the more recent labours of the younger Agardh, than 
whom no one has done better service to this branch of science, which 
has been perplexed more than most others with ill-defined genera, so 
as to make it for a long time the reproach of natural history. The 
figures and dissections will enable many to read the works of Agardh 
with more profit, and may ultimately lead to a more accurate know- - 
ledge of many obscure points, in the hands of persons favourably situ- 
ated for observation. : 
We think it, at the same time, a great merit that Dr. Harvey has not 
attempted to throw out any new system of his own, though he has not 
hesitated, where he saw good cause, either to dissent or amend, and he 
has established several good and interesting genera. 
A single instance of this mode of disposing of needless genera will 
give a good example of the sort of matter to be met with in the more 
general observations. 
“A considerable number of species" —speaking of Nytophyllum, Grev. 
—“ have been described, which Kützing divides into three genera, 4g- 
laiophyllum, Schizoglossum, and Cryptopleura, distinguished by some 
differences in the structure of the frond. Aglaiophyllum is limited to 
those genera which are altogether nerveless, and have a membrane com- 
posed of a single stratum of large cellules; Cryptopleura, to those in 
which nerves are more or less obvious, with a similarly constructed 
membrane ; and Schizoglossum, to such nerveless species as have a mem- - 
brane composed of one stratum of cellules. I do not adopt these. 
genera, because I find the characters untenable. The more or less ner- 
vated frond varies even in the same species; Schizoglossum Gmelini, 
Kiitz., for instance, which ought by generic character to be nerveless, is 
frequently abundantly supplied with nerves, much more abundantly 
than Cryptopleura Bonnemaisoni and C. Crozieri. The character foun- 
ded on compound structure is equally nugatory, for Cryptopleura lace- 
rata, which is moreover the type of the genus so called, has a frond 
