208 
The leaf does not differ from that of the type of A. cam 
panulatus. 
The flower differs in colour and shape. In the typical 
.4. cnm panulatus (as figured in "Bot. Mag.," t. 2812, and in 
Blumes "Rumphia," T. t- 32 and 33) the spathe is broader 
than high, while in the Northern Territory specimen the 
i<pathe is hir/her than hroad. The Northern Territory plant 
is, further, darker inside and more distinctly spotted outside 
than the type, and the sterile expansion on the top of the 
spadix is considerably less wrinkled. 
I do not think these differences are sufficient to warrant 
its description as a new species, and, in view of the amount 
of variation known to exist in the species, I am not altogether 
free from doubt as to the expediency of giving it a varietal 
name. But it may be a convenience to distinguish the Nor- 
thern Territory form, and therefore T propose the name A. 
campaniih(tiis, Blume, var. aiiatralamca, for it. 
Several new species of Amor phoph alius have been de- 
scribed since 1879 (the date of Engler's Monograph), but none 
of the species recorded in the Supplement to the "Index 
Kewensis" come near the Northern Territory plant, so that I 
have no doubt the plant is unrecorded for Australia- 
Mueller mentions A. variahilis, Blume, 3& the only Aus- 
tralian species, and Bailey adds two more species, A. galhra 
and A. angustiloha, but all these three species belong to a dif- 
ferent section of the genus, and are very different from the 
plant under consideration. 
The geographical range of .4. campanvlatKS is from 
Madagascar to the Malayan Archipelago and the Melanesian 
and Polynesian Islands, so that its occurrence in Australia 
is only what could have been expected. 
