192 REVIEWS. 
and what as varieties; in which there is, as he himself observes, nothing' 
new, and they refer to points so obvious, that they seem hardly 
worth the dignified name of principles. One common practice he 
professes not to follow, viz. that of judging of the validity of a 
character in one species by analogy with others; he says, in short, 
given two specimens alike in every respect but one, and granting 
that one to be variable in other Oaks, I would still raise these two 
specimens to the rank of distinct species, till more specimens arrived 
to confirm or confute their claims. Such a practice we venture to 
think neither philosophical nor expedient; it is unphilosophical, 
because it excludes every consideration that experience, study, or 
sagacity suggests, and reduces the systematist to the condition of a 
dividing machine ; and it is inexpedient, because we know that one 
character rarely, if ever, makes a species,—because synthesis is more 
often a safer process than analysis; and, because, if such a doctrine 
were carried out even in part, by one active systematic Botanist, 
synonymy would swallow up system in a very few years. Had such 
been Linnaeus’ practice, with his poor materials, every exotic speci¬ 
men would have been a species; and if it really were M. De Can¬ 
dolle’s, every variety not connected by known intermediates (even 
white and red-flowered races, &c.) must be ranked as distinct species. 
There are no limits to the evil consequences of such an abandonment 
of the reasoning faculties and idolatry of the observing powers. We 
will mention but one, namely, that all M. He Candolle’s excellent 
principles, deduced from a study of Oaks, would be absolutely valueless 
as guides to the study of any other group of plants whatever. If the 
length of the petiole is known as a rule to vary from 1-3, in the indi¬ 
vidual branches of Oaks, surely common sense, no less than expedi¬ 
ency, would indicate that two specimens differing by this character 
alone, should be ranked as varieties till they are proved distinct, rather 
than as species till they are proved the same, which is giving specific 
value to a character of which all we do know is, that it is that of 
many proved varieties, and of no other known species. 
As it is, M. He Candolle regards two-thirds of the 800 species of 
Oaks admitted by him as provisional species , and adds that it is difficult 
to suppose that more than one-third of the published species of 
plants will prove to be fixed. The best known Oaks are those which 
present most varieties and sub-varieties, the maximum being in 
Q . Robur which has 28, Lusitanica, 11, Calliprinos , 10, and coccifera , 
8. Moreover, the best known species are surrounded by provisional 
ones, which may one day be absorbed by them, four species thus 
differing little from Robur , and two being near Calliprinos , which 
touches Q. coccifera. 
After some further comments on the variability of plants in 
general, M. He Candolle observes, “ that the progress of science, no 
less than reflection, leads me to the opinion that the higher the groups 
are, the better they are limited ; in other words, the fewer are the 
doubtful forms that are bandied about from one to the other.” This 
