414 Mr. H. St. J. Donisthorpe on the Colonisation of 



observe in the wild state. Exception may perhaps be 

 found in such species as Cetojiia fioricola, and Clytho^a 

 quadripnnctata, which only inhabit the nests in the larval 

 and pupal states. The methods of colonisation are further- 

 more very different according as the different species of 

 beetles belong to the different categories of true guests, 

 hostile persecuted lodgers, or indifferently treated lodgers. 

 The first of these supplying their hosts with a sweet secre- 

 tion and fed by them, naturally come into much closer 

 relationship with the ants than any of the others. In 

 order to give anything like a satisfactory answer to 

 Darwin's question we really require to study the whole 

 life history of ants'-nest beetles and their connection with 

 their hosts. 



When an ants' nest is disturbed a scene of apparent 

 chaos ensues : the ants all rush about, some attacking the 

 intruder, others hurrying off into safety, with their brood, 

 and with the true guests and their larvae. Their other 

 guests likewise endeavour to escape : some " feign death," 

 while others seek the interior of the nest. It is obvious 

 that in such a scene we can gain no insight into the 

 normal procedure of a fresh colonisation, and are therefore 

 forced to rely on evidence obtained from observation nests, 

 and experiments on ants'-nest beetles, and on beetles bred 

 in such nests ; on their pairing habits, and on when and 

 where one finds the beetles, their larvae, etc. It is neces- 

 sary to consider all records of true myrmecophilous beetles 

 found at large, away from ants' nests ; and of all specimens 

 found with ants other than their usual hosts. These 

 latter examples will afiord evidence that the beetles not 

 only have not returned to the nest in which they were 

 bred, but have not even sought the same species of ants. 

 The rarity of such records tends to prove that ants'-nest 

 beetles are as a rule very constant in associating with 

 their regular hosts. I would also remark that it is most 

 important that all records of ants'-nest beetles should be 

 accompanied by full and accurate data, stating the number 

 of specimens taken, the exact date, and the species of ant 

 with which the beetle occurred. Without these facts the 

 subject cannot be investigated with any hope of success. 



Many species no doubt simply fly direct to a new nest, 

 either to deposit their eggs in it, or from the necessity of 

 pairing with other individuals of their own species. Be- 

 sides this they may reach new nests by the following 



