OF THE HERBARIUM AMBOINENSE. 317 
Humphius, it will be impossible to refer his kinds to Decan- 
dolle's species, as he does not mention the number of sta- 
mina. In fact, this division into five species is an attempt 
to revive the division by C. Bauhin, being the same with his 
Mains medicaj Mains Lhnonia acida, Malus aurantia 
major, and Malus aurantia cortice eduli, to which is added 
the Limo dulcis, commonly held to be a variety of the se- 
cond kind. This subdivision may be very well suited to 
include the varieties commonly cultivated in Europe, but 
will be found inapplicable to those produced by long cul- 
ture in Asia. 
Provided it be thought proper to distinguish the Citrus 
decumana from the others, it may be done as follows : — . 
1. Citrus medica (et Aurantium) pulpse in fructu vesiculis concretis. 
2. Citrus decumana pulpae in fructu vesiculis discretis. 
This character, however, although always more or less 
distinguishable, is not so remarkable in the finer varieties of 
the C, decumana, such as those which grow at Canton and 
Batavia ; and I doubt the propriety of separating into dif- 
ferent species trees having such a strong resemblance to each 
other, on account of various forms, sizes, and qualities of a 
fruit, that has been long cultivated with the greatest care ; 
nor does the fruit of any of the varieties of these three Lin- 
nsean species vary more from each other, than do different 
kinds of Apple (Malus). The great size of the Pompelmoes 
is no sort of proof of a distinct species ; as both the Santra 
and J amir of Gangetic India, which have an acid fruit, 
consisting of vesicles closely adhering to each other, which 
generally have a round petiolus, have a fruit fully as large 
as the Pompelmoes. 
CAP. XXXV. 
Malum citrium silvestre, p. 99, t. 25. 
Malum citrium domesticum, p. 99. 
Malum citrium silvestre alterum, p. 100. 
