6 G. E. J. NIXON 



THE SUBFAMILY MICROGASTERI N AE 



Telenga (1955), the most recent worker on the subfamily, divides it into three 

 tribes : — Microgasterini, Cardiochilini and Acoeliini. 



The Microgasterini, limited by Telenga to Apanteles, Microgaster, Microplitis and 

 Hygroplitis is almost exactly coextensive with the tribe as I have interpreted it in 

 this work. 



Telenga's Cardiochilini contains the two genera, Cardiochiles Nees and Asio- 

 cardiochiles Telenga. The subfamily Cardiochilinae is reduced to the status of a 

 tribe but its content, though a little smaller, remains unchanged. 



Szepligeti (1904) included in the Cardiochilinae only three genera : Cardiochiles 

 Nees, Toxoneuron Say and Psilophthalmus Szepligeti. Toxoneuron, along with several 

 other genera either unknown to Szepligeti or described after the appearance of the 

 Genera Insectorum, has been sunk as a synonym of Cardiochiles by Muesebeck and 

 Walkley (195 1). Psilophthalmus, which I know from a single specimen (unnamed) 

 from Peru and identified by Mrs. J. Clark in the British Museum (Natural History), 

 is certainly close to Cardiochiles, though showing several uncharacteristic features. 



I think Telenga is right in stressing the affinity of Cardiochiles s.l. with the Micro- 

 gasterini by means of the tribal concept. 



Telenga's Acoelinii is the least natural of the three tribes and is made up of the four 

 genera, Mir ax Haliday, Myriola Shestakov, Dirrhope Forster and Adelius Haliday. 



All these genera are discussed below. 



APANTELES Forster 



Muesebeck (1920 : 485) expressed the belief that Apanteles was a homogeneous 

 group and not susceptible of division into smaller units. He therefore placed in 

 synonymy all the genera into which Ashmead and Viereck had attempted to break 

 up the genus. These genera have remained in disuse ever since. Nor has Muesebeck 

 in any of his more recent papers on Apanteles modified his earlier opinion. 



Wilkinson (1932) divided Apanteles into five groups that he arbitrarily designated 

 with the letters, A, F, S, U and M. The first four of these, as he himself admits, were 

 essentially extensions of the four sections proposed by Marshall in 1885 for the 

 British species. Later, Wilkinson (1934 : 155) added a sixth group to which he 

 gave the letter G. 



Wilkinson's subdivisions were used by de Saeger (1944) and Granger (1949) but not 

 by Telenga who, nevertheless, could not avoid using the fundamental characters on 

 which the subdivisions are based. 



Of Wilkinson's groups, A and F represent for the most part natural aggregates. 

 The others are too artificial to have taxonomic value. Group S, which Wilkinson 

 reserved for all species having a propodeal areola, can be split into two main divisions 

 that are convergent in that they both possess this areola or at least a vestige of it. In 

 other respects they are widely dissimilar on a combination of characters, of which one, 

 the shape of the vannal lobe of the hind wing, is striking and probably of philogenetic 

 significance. This character, incidentally, is perhaps the most useful I have been 



