1903.] 239 



that Y. .^ylvestris, which habitually builds in trees, occasionally makes a ground-nest 

 (Sinitli, '58) ; here apparently we see the beginning of the change in habit which has 

 already been carried far by the rufa-austriaca race. In most of the points wherein 

 V. atistriaca differs from V. rufa, it approaches the tree-wasps {e.g., the hairy shins, 

 the yellow-streaked scape) ; and we see in this further evidence that austriaca is to 

 be regarded as the older form. Moreover, as all the workers of these wasps are 

 clearly referable to V. rufa, it seems that V. austriaca points us back to a time in 

 the history of the race before the worker had become differentiated from the queen. 

 The workers doubtfully referred by Smith to his V. arhorea ('43, p. 171 ; Ormerod, 

 '68, pi. 3) might well be considered varieties of V. rufa in which the reddish abdo- 

 minal markings are wanting, while a yellow streak on the scape of the feeler and 

 additional yellow marks on the scutellum are present. Unfortunately he gives no 

 structural details of these workers. Marchal ('96) has shown that, even among 

 our commonest social wasps, a sharp distinguishing line between the two forms of 

 female — the queen and the worker — cannot always be drawn. 



Some very interesting problems as to the origin of specific distinctions may, 

 perhaps, be elucidated by the relationship between our two wasps. Their structural 

 differences are quite sufBcient to warrant " specific " distinction in the ordinary 

 sense of the term, so that if our view be established, the development of rufa- 

 offspring from aMs^rifflcff-parents would be a very striking instance of " discontinuous 

 variation" (Batcson, '94). It would, indeed, furnish an instance in support of 

 Bateson's theory " that the Discontinuity of Species results from the Discontinuity 

 of Variation." We think that we see hero a new species arise by tlie production, 

 through many generations, of an increasing number of intlividuals (n</i7-forms) 

 among the offspring, that are markedly unlike the parents [austriaca-iorms). We 

 believe that austriaca-^ovms give rise to ri^/a-forms, but we have no evidence of the 

 reverse process ; and it seems that those specimens of V. rifa varying towards V. 

 austriaca must be regarded as examples of reversion towards the ancestral type. 

 The slight but constant difference between the male armature in our two wasps 

 illustrates Jordans' ('96) principle of " mechanical selection," and supports Romanes' 

 view ('97, p. 46) that variations affecting the reproductive organs, and tending to 

 prevent intercrossing between an incipient species and its parent form, arise before 

 any striking differences in the general body-structure become developed. These 

 constant differences between the male armature in the two wasps and the absence 

 of intermediate individuals suggest that interbreeding never takes place, and that 

 V. rufa can never be " swamped " by crossing with its parent form, V. avsiriaca. 

 Indeed, the latter is now so scarce, that all danger of such " swamping " is past. 

 Except in Ireland and possibly in some mountainous continental localities, it has 

 been almost entirely superseded by its more vigorous and robust descendant. 



References. 

 '84. Andee, E. — Species des Hym^nopteresd'Europe et d'Algerie, vol. ii. Beaune, 



1882—6 (p. 597). 

 '01. Babrington, R. M., and Moffatt, C. B. — Wasps in Co. Wicklow. Irish 



Nat., vol. X, 1901, pp. 197—201. 

 '94. Bateson, W. — Materials for the Study of Variation. London, 1894. 



Y 2 



