ERIC MJOBERG, ISOPTERA. 5 



from the other genera of the Protermitidce, that Holmgren 

 (1910) seems to be well justified in creating a new sub-family 

 Stolotermitidce. The discovery of two quite new species from 

 the rain-forests or jungles of Queensland is of greatest interest, 

 showing that the genus in earlier days apparently had a 

 more northern distribution, and perhaps is to be regarded as 

 a true Australian element with single offshoots to the 

 two big southern islands. Further investigations will doubt- 

 less bring to light additional species in Queensland, and per- 

 haps also in New Guinea, 



The two new species both differ from the two old spe- 

 cies by their flattened heads and a larger number of joints 

 in the antennae. 



Key 



to the tivo known imagines of the genus Stolotermes. 



I. Antennse 17-jointed. 1. St. qiieenslandicus Mjöb. n. sp. 



II. Antennse 16-jointed. 2. St. hrunneicornis Häg. 



Distribution : Tasm a n i a . 



Key 



to the soldiers of the Stolotermes-species. 



I. Antennae 18-jointed. 1. St. queenslandicus Mjöb. 



(Text Fig. la.) Distribution: Queensland. 



II. Antennse 16-jointed. 2. St. australicus Mjöb. n. sp. 



(Text Fig. 1 c.) Distribution: Queensland. 



III. Antennae 15-jointed. 3. St. ruficeps Br. 



Distribution: New Zealand. 



The hitherto unknown worker will be described in the 

 following.^ 



' The genus Porotermes (Hag.) has a very peculiar distribution. One 

 of the five known species (P. quadricollis Raub) lives in Chili, one (P. 

 planiceps Sjöst.) in South Africa, one (P. Adamsoni Frogg.) in New 

 South Wales, a fourth (P. Frogatti Holmgr.) in Tasmania, and a fifth 

 (P. grandis Holmgr.) in Victoria. They represent apparently the last sur- 

 viving remnants of an old group, the centre of development of which is 

 still unknown. Only the discovery of further primitive forms can throw 

 light upon the true home of the Poro^ermcÄ-termites. It must always be 

 borne in mind that the present distribution of genera and species cannot 

 be used as a basis of inference for the centre of development of a group. 



