ON: A NEW. CHARACTER, IN. THE FRUIT OF QUERCUS. 141 
stated, this. with. his usual candour,*. and. we. may conclude that he 
thinks his: own subdivisions axe not likely to stand. The result of 
my ọwn long study. is, that in the present state of science the sub- 
genus Lepidobalanus cannot be. subdivided. When the male flowers 
of many species are better. known, and. the evolution of the buds has 
been examined, it may. be possible to establish a. truly natural di- 
vision, but at present, with the help of fruit and leaves only, we cannot 
get. beyond artificial sections, which. frequently separate closely allied 
cies 
"The form and direction, of the involucral scales is a character too 
subject to transition to be depended on, besides it sets aside some spe- 
cies like Q. Cerris, while removing many from it in one mass. 
The duration of the leaves is considered by Webb and other authors 
to. be variable in some species (Q. Lwsifaniea, humilis, etc.), and the 
character has the inconvenience of being. ascertained with difficulty, 
both in herbaria and in travelling through a country. Webb considers 
the leaves of Quercus to be‘ deciduous,” ‘subdeciduous,” or ."* per- 
sistent,’ but. this only indicates the inconstaney of the character. In 
many. southern species, especially in the Mexican ones, it appears that 
the leaves fall in their second year shortly after the shooting out of the 
new leaves, and. in this case are scarcely ever found on herbarium spe- 
eimens, which are usually gathered in fruit in autumn. In general, 
Whether leaves are very persistent is easily ascertained, but the distine- 
tion between leaves which fall a little earlier or a little later than the 
next leafing-season, is too liable to transition between species, and too 
transitory to be of practical use. 
Iam therefore obliged to divide the group Lepidobalanus almost 
artificially ; first, from the duration of the fruit and position of the 
ovules, constant characters of some importance; then, from the dura- 
tion. of the leaves, a less determined and constant character, The result 
is as follows :— 
§ 1. Ocula abortiva infera. Maturatio annua. ; 
/ Folia caduca: Q. Robur, Toza, Lusitanica, alba, Prinos, macro- 
carpa, polymorpha, ete. . 
> ** Folia persistentia: "Q. tomentosa, microphylla, virens, Ilew, 
Suber, etc. 
* Ann. des Sc. Nat., ser. iv, vol. vi. P- 238. 
