162 



RUMPHIUS'S HERBARIUM AMBOINENSE 



apparently the correct disposition of it and one that has been 

 accepted by all authors. 



KAEM PFERI A PAN DU RATA Roxb. in As. Research. 11 (1810) 328, t. 2. 

 Gastrochilus panduratum Ridl. in Journ. As. Soc. Str. Branch 32 

 (1899) 114. 



Zerumbed claviculatum Rumph. Herb. Amb. 5: 172, t. 69, f. 1. 



Linnaeus originally reduced Zerumbed claviculatum Rumph. 

 to Kaempferia rotunda Linn., in Stickman Herb. Amb. (1754) 

 20, Amoen. Acad. 4 (1759) 129, which is manifestly an er- 

 roneous disposition of it. Roxburgh, Fl. Ind. 1 (1820) 18, 

 placed it under Kaempferia pandurata Roxb., which was de- 

 scribed from Sumatran specimens. This reduction has been 

 accepted by Roemer and Schultes, Blume, Dietrich, Miquel, and 

 Horaninow and is apparently the correct disposition of it. 



GLOBBA Linnaeus 



GLOBBA MARANTINA Linn. Mant. 2 (1771) 170. 



Lampujum silvestre minus Rumph. Herb. Amb. 5: 150, t. 6U, f. 2. 

 Amboina, Soja, Robinson PL Rumph. Amb. 145, August, 1913, along 

 roadsides and on banks, altitude 5 to 400 meters, locally known as kuning 

 utan. 



Rumphius's figure is a very poor one, but it is certainly refer- 

 able to Globba marantina Linn. This reduction seems first to 

 have been made by Roemer and Schultes, Syst. Mant. 1 (1822) 

 46. 



CURCUMA Willdenow 



Rumphius described, in all, nine forms of Curcuma, of which 

 he declares that the first three, considered in chapter sixteen, 

 belong to the true Curcuma. In the following chapter he deals 

 with a group of species, which he considered under the "generic" 

 name Zerumbed, on account of the assertions of the natives, who 

 deliberately distinguish these forms from the true Curcuma, or 

 koenjit, uniting them under a proper group name, tommon. 

 Rumphius confessed, however, that he himself was not able to 

 find characters by which the two groups could be distinguished. 

 However, if one compares the native names cited by Rumphius 

 for the six species mentioned under Zerumbed, with those now 

 used in Java, one is impressed by the fact that, with a single 

 exception, all of them are applied to species belonging in the 

 section Exantha; while the three species placed by Rumphius 

 under the name Curcuma, all belong in the section Mesantha. 

 The clue to Rumphius's enigma apparently lies in the fact that 



