THE IXSTIXCTITE BEHAVIOR OF ANTS. 525 



The males and females do not attempt the nuptial flight till they are 

 thoroughly mature, and the queen does not begin to start a colony till 

 her wings have been removed. The worker will not forage or take 

 part in excavating or guarding the nest till its integument has har- 

 dened and taken on the adult coloration. As a callow, it remains in 

 the nest and functions only as a nurse. The tendency of certain old 

 workers to become sexually mature and to act as gynsecoids may also 

 be regarded as a deferred instinct, depending on unusual powers of 

 assimilating food and the ripening of eggs in the ovaries. To the same 

 category we may also assign the behavior of old queens that shun the 

 light and merely assimilate food and lay eggs, without paying any 

 attention to the brood or to the adult workers. 



As additional evidence of the intimate correlation between instinct 

 and structure we may point to the vestigial, decadent and deceased 

 instincts and the analogy between form regulation and instinct regula- 

 tion. Ants furnish abundant examples of all of these, especially of 

 activities that must once have been of the greatest importance to the 

 species, but have since fallen into desuetude and been overlaid or all 

 but completely replaced by more recently acquired tendencies. Cases 

 of this description are most obvious in the parasitic species, or in those 

 that have changed their nesting habits within comparatively recent 

 times. Forel (1900^, 19047) has called attention to vestigial slave-mak- 

 ing instincts in Strongylognathus huberi and rchbindcri, ants now liv- 

 ing as permanent parasites in the nests of Tetramprium cespitum (see 

 Chapter XXVII), but in all probability descended from slave-makers 

 like Polycrgiis nifcsccus, which they still resemble in the peculiar fal- 

 cate structure of their jaws. P. rufcscens, too, has its vestigial instincts. 

 As we have seen, the workers of this species are no longer able to 

 take food except from the tongues of their slaves, and perish when 

 these attendants are removed, but the queens have retained to a very 

 slight degree the ability to feed independently. This case and many 

 others that might be cited, are interesting as proving that the castes 

 may show different stages in the decay of the same instinct. In other 

 words, we not only find the ants exhibiting vestigial instincts as species, 

 but a certain caste within the species may show vestiges of instincts 

 whose full exercise is the normal prerogative of a different caste. 

 Thus, under extraordinary circumstances, the usually sterile worker 

 may lay eggs, like the female, or the female may forage like the workers 

 or accompany them on slave-making expeditions. In Leptothora.v 

 emersoni I have observed a very striking example of an obsolescent 

 feeding habit (see Chapter XXIII). This ant when living with its host, 

 Myrmica canadcnsis and in a state of nature it is always found in this 



