128 



RUMPHIUS'S HERBARIUM AMBOINENSE 



Blume, the plate presenting the vegetative parts of Tacca, its 

 infructescence, and detached fruits, /. 1, a, b; and /. 2 the in- 

 florescence of Amorphophallus campanulatus Blume, described 

 by Rumphius as Taccae fungus. The latter figure is cited by 

 Roxburgh in the original description of Arum campanulatum, 

 and is certainly correctly placed. 



HOMALOMENA Schott 



HOMALOMENA COR DATA (Houtt.) Schott Melet. 1 (1832) 20. 



Dracontium cordatum Houtt. Handleid. 11 (1774-83) 200, t. 71, f. 2. 

 Dracunculus amboinicus niger Rumph. Herb. Amb. 5: 322, t. Ill, 

 f. 2. 



Amboina, Batoe merah, Robinson PI. Rumph. Amb. Ill, August 24, 1913, 

 on shaded banks, altitude about 15 meters. 



Dracunculus amboinicus, as figured by Rumphius, was erro- 

 neously reduced by Linnaeus to Arum divaricatum Linn., in 

 Stickman Herb. Amb. (1754) 22, Amoen. Acad. 4 (1759) 131. 

 By Loureiro, Fl. Cochinch. (1790) 532, it was placed with 

 doubt under Calla occulta Lour., which may be the same as the 

 closely allied Homalomena aromatica Schott. By other authors 

 it has been reduced to Homalomena rubescens Kunth, H. rubra 

 Kunth, and other allied forms. The Amboina specimen cited 

 above agrees with the figure given by Rumphius, agrees with 

 the description of Dracunculus amboinicus niger, and agrees 

 also with Homalomena cordata Schott as recently described by 

 Engler, Pflanzenreich 55 (1912) 57. Rumphius does not indicate 

 which of the four forms described under Dracunculus amboinicus 

 he intended the figure to represent, but I agree with Engler in 

 referring it to the one first described, namely, Dracunculus 

 amboinicus niger. 



The form described by Rumphius as Dracunculus amboinicus II 

 albus, is apparently very closely allied to, if not identical with, 

 Homalomena cordata (Houtt.) Schott, and may be the same as 

 Homalomena aromatica (Roxb.) Schott, which is reported from 

 Amboina by Engler ; it can scarcely be Homalomena alba Hassk., 

 to which Hasskarl reduced it, Neue Schlussel (1866) 128, because 

 as far as is known this species is confined to Java. Dracunculus 

 amboinicus III ruber should be compared with both Homalomena 

 cordata Schott and H. aromatica Schott. Hasskarl, following 

 Kunth, reduced it to Homalomena rubescens Kunth, which is 

 manifestly an error, as this species is not known from the 

 Malay Archipelago. More comprehensive collections are neces- 



