1910.] ON THE FOSSORIAL WASPS OF AUSTRALIA. 253 



The following papers were read : — 



1. Additions to our Knowledge of the Fossorial Wasps of 

 Australia. By Rowland E. Turner, F.Z.S., F.E.S. 



[Received February 16, 1909.] 

 (Plates XXXI. & XXXII.*) 



The material for the following descriptions is nearly all in the 

 British Museum collection, most of the Thynnidse having been 

 received from Mr. H. M. Giles, and the Ceropalida? from the 

 collection of the late Gilbert Turner. We may now assume that 

 our material, though very incomplete, is fairly representative of 

 the Fossorial Hymenoptera of Australia. It is most striking 

 that of the species described up to the present time nearly one- 

 half belong to the family Tbynnidae, which is almost entirely 

 without representatives in any other part of the world excepting 

 South America. 



The wingless condition of the females must operate against a 

 wide distribution, but I think it quite possible that the family as 

 it exists to-day is derived from less specialised and much more 

 widely distributed ancestors of a form more nearly resembling the 

 modern Scoliidee. The mouth-parts of the females in most 

 Australian species are almost rudimentary, while in South- 

 American species they are usually developed, so that the degree 

 of specialisation in the two countries differs considerably. I am 

 inclined to believe that the ancestors of the group have been 

 exterminated in the northern continents and Africa by the keener 

 competition existing on large continental areas, and that their 

 descendants have survived on the smaller and more isolated land- 

 areas of the south. In South America the species are very few 

 in the tropical lowlands, but increase in number in the temperate 

 and mountainous regions. In Australia they are most numerous 

 in the south-west and decrease in number as they come into 

 contact with a fauna of a more Oriental character in Queensland. 

 I do not therefore look on the occurrence of the group in such 

 distant localities as any proof of a land connection in the past, 

 but rather regard the South American and Australian species as 

 isolated survivors. The Mutillida? of Australia mostly seem 

 generically distinct from those of other countries, and the Scoliid 

 genus Anthobosca is elsewhere only represented by a few species 

 in South America, South Africa, and Madagascar, but has been 

 found fossil in Colorado. The Ceropalidte do not show any 

 remarkable peculiarities, being nearly all representatives of wide- 

 ranging genera. It is unfortunate that the old name for this 

 family, Pompiliclse, has to be superseded. 



* For explanation of the Plates sec p. 356. 



