I tliink therefore that Wallace's '"Polynesian" Subregion shouht 

 be rejected. 



Celebes is usually included in the Papuan, but is ])crhaps, at 

 present, best regarded as an unattached subregion. 



The Fijian Isles seem to have a well marked affinity with New 

 Guinea and the neighboring groups, or with the first of the fol- 

 lowing Australian. 



The Australian Continent seems to passess four ])rincipal 

 faunas : 



1. The tropical forest, an impoverished extension of the Aus- 

 tromalayan, little known, and producing such forms as A'eomcli- 

 charia, Ancipo and some fine Derbidae. 



2. Eucalyptus-Forest, characterized by Eiiryiiicla, Eiiriiiop- 

 syche and Platyhrachys. This is typical of the Euronotian. 



3. Western Australia, which stands apart from the others. 



4. Grass-lands and coast lands, probably typical in another 

 way of the Euronotian, and characterized by common-place 

 Deltocephalus and Asiracidac. 



In the habitats, "K" stands for Prof. Koebele, "P" for Dr. 

 Perkins, and "M" for Mr. Muir. 



Mr. Muir's researches in Fiji were confined to Viti Levu. 

 Suva is on Suva Bay in the Southeastern part ; Naviia on the 

 river of the same name in the South-central ; Rcwa, on the Rewa 

 river, a little east from Suva, and Ba (=:Mbaj on the Ba River 

 in the northwest. 



An expression of ap[)reciation is due to Mr. W. E. Chambers, 

 who has enriched this memoir with more than three hundred 

 figures. No work on Hemiptera has previously appeared, in 

 which such careful and detailed drawings of genitalia are given, 

 affording characters which are the final arbiters of "species" in 

 certain groups. 



i 



