OF THE HEKBAlliaM AMBOINENSE. 333 
CAP. LIV. 
SoNGiuM, p. 140. t. 45. 
In a Commentary on the Syalita of Rheede (Hort. 
Mai. iii. 339), I have said most of what occurred to me on 
the subject, as I consider the Songium to be the same 
tree. M. Decandolle does not quote the Songium for 
either Dillenia speciosa or elliptica, but it evidently agrees 
better with his specific character of the former, than with 
that of the latter, the only difference between them being in 
the words foliis elhptico-oblongis appHed to the D. speciosa, 
and elliptico-ovatus applied to the D. elliptica, differences, 
in my opinion, too trifling to deserve notice. Rumphius 
mentions three varieties of this tree, which, being a good 
deal planted near villages, is liable to assume various forms ; 
one has sweet, another sour fruit, and another is sterile, 
producing no fruit. I have never, however, noticed any 
tree, that produced male flowers alone, although such may 
very readily exist. In this case the plant belongs to the 
Polygamia dioecia. 
Sangius mas et foemina, p. 142. t. 46. 
Rumphius erroneously referred this to the Syalita of 
Rheede, from which, however, it is suflSciently distin- 
guished by its peduncules having three yellow flowers. 
That the Songium or Syalita is di(Decious, I have already 
stated, is probable ; but that the Sangius is so, from the 
description given by Rumphius, is certain, and nothing 
seems to be known of the plant, except what he says ; yet 
Thunberg (Lin. Trans, i. SOl)^ by whom it was intro- 
duced into the modern system by the name of Dillenia, 
serrata, takes no notice of this circumstance; and 
Poiret (Enc. Meth. vii. 151), and Decandolle (Prodr. i 
76), are equally silent. 
