194 
REVIEWS. 
have been determined with very great difficulty, and only after a com¬ 
parative study of many species of all the contained genera; that the to¬ 
tality of conclusions arrived at are in most cases opposed to the opi¬ 
nions of about half their fellow Botanists; and that difficult as the 
limitation of the Orders has been, it is trifling compared with that of 
the Genera; whilst the grouping of the Orders into cohorts is the most 
difficult, and (as far as regards the means of defining them, however 
natural), the most unsatisfactory of all. Their opinion is most decided 
that, on the whole , the natural grouping of individuals into species, and 
their limitation as such, is far more easy and satisfactory than of 
genera, and of all the other superior groups in the class of Dicotyle¬ 
dons, and this whether in the field or in the herbarium. And further, 
they have no hesitation in affirming, that were half the species of each 
genus, or half the genera of each Order to disappear from the 
earth, and the classification of the remainder to be reconstructed, 
the circumscription of both their Orders and Genera would be extra¬ 
ordinarily modified, but of their contained species not materially. No 
one who has not worked out generically a consecutive series of allied 
Orders, can have any idea of the number of genera whose claims to 
rank in one or the other are determined either by some merely technical 
character, of no apparent functional import, or by some natural cha¬ 
racter, quite undefinable by words. The fact that species do in 
botany stand out as the most prominent term in the series between 
individual and class, is perhaps the most salient obstacle to the re¬ 
ception of the doctrine of the origin of these through variation by 
natural selection. 
Returning to M. De Candolle, this section concludes with some 
excellent remarks on hybridity, and the impossibility of testing, in 
many cases, the validity of species by experiments or observations 
on the permanence of characters during successive generations. 
2 .—Observations and Hypotheses on the History and Origin of the 
Forms of Cupuliferce. 
The discussion to which this section is devoted is one of the most 
interesting in the brochure, and is conducted in M. De Candolle’s 
best method. The following is a specimen. 
Beeches are found in both the Northern and Southern hemispheres; 
but all the other genera of Cupuliferse, including the most numerous 
in species (Oak and Chesnut), are limited to the Northern. The 
exceptions are, a few Oaks which, advancing southwards, reach the 
mountains of New Grenada and the Indian Archipelago. The ma¬ 
jority of Oaks and Chesnuts inhabit Mexico, the United States, the 
Mediterranean basin, and temperate Europe and Asia. Three natural 
groups of Oak are confined to Southern Asia, as are Lithocarpus and 
Castanopsis ; one natural group is confined to California; whilst the 
largest group is European, Asiatic, and American. 
Q. Cerris (the Turkey Oak) appears to be dwindling away, for in 
no other way can he explain the gaps in its distribution. It is found 
