314 ROBINSON. 
18. DEBREGEASIA Gauclich. 
r 
Alcoholic material, collected with Phil. PL 808 Merrill, from Mount Tonglon, 
Benguet, appears to show that the pistillate perianth when in the flowering stage 
may be considered very loosely adnate to the ovary. The latter is separated with 
ease, but usually has attached fragments of tissiie from the inner wall of the 
perianth, which in their turn' are readily removable. Still, as the ovary appears 
roiifrhened, the term adnate may be applicable. But its use as a generic character 
is to be avoided wherever passible, as there is much reason to believe that the 
mechanical difficulty of ascertaining the facts has often led botanists working 
upon dried material to wrong conclusions; if this is not'lhe case, it becomes 
certain that the ovarj- and the perianth may or may not be adnate within the 
limits of the same genus.^ 
Trimen, in his key to the Urticaceae,-^ distinguislies Dehregeasia from Ville- 
brunca {Oreoaiide), by the character ''Fem. fls. naked" as contrasted with "Fern, 
fls. surrounded by fleshy bracteoles." This does not hold true for the Philippine 
species of Dehregeasia, as its flowers are surrounded by braeteoles, although these 
are short, and can hardly be called fleshy. 
^ See discussion under Piptunts, above. 
-■•Handb. Fl. Ceylon 4 (1898) 80. 
