ANT AND TEBMITE GUESTS 199 



hosts. Two others, on the other hand Dinar da Pygmcea 

 and Hagensi which live with Formica rufibarbis and F. 

 exsecta are only on the way to develop : they appear, 

 for instance, only in a limited portion of the domain 

 occupied by their hosts and in their typical form, 

 outside of which they are absent or are replaced by 

 transitional forms which have arisen from Dinarda 

 dentata ; they stand thus as outposts at different stages 

 of species formation, always according to the various 

 points of their extensive domain. In this way we 

 conclude that there is being perfected before our eyes 

 a so-called process of species formation within the 

 genus of Dinarda. 1 



The same causes of adaptation which at present 

 still determine the process of differentiation between 

 our northern bicoloured Dinarda forms, suffice, how- 

 ever, perfectly to explain the systematic differences 

 which exist between the genus Dinarda and the closely 

 related therewith Mediterranean genus Chitosa. The 

 host ant of the latter is namely Aphcenogaster testaceo- 

 pilosa, thus belonging to quite another sub-family 

 of the Ant stock (Myrmicini) belonging to the Formica 

 (Camponotoni). The adaptation of an Aleocharina 

 of the offensive type such as are Dinarda and Chitosa 

 to Formica on the one hand and to Aphsenogaster 

 on the other demands, however, much greater morpho- 

 logical differences than the adaptation to different 



i The objections raised by H. Muckermann in Natur und 

 Offenbarung, 1909, No. 1, have already been contested by me therein 

 in No. 6. 



