2 THE LARVA. 



problematical, but Adler's suggestion that it is 

 respiratory appears to be founded on some solid 

 grounds : e. g. those species which have very long 

 peduncles are those which place the eggs in posi- 

 tions where they cannot receive oxygen from the 

 plant — the spring generations, which have their eggs 

 placed in the leaves where they can get oxygen from 

 them, have it short ; while the winter generations, as 

 a rule, have it long. The peduncle is always placed 

 uppermost. 



The number of eggs laid is frequently very great. 

 It is certain that the ovaries of some species can con- 

 tain about 600 to 700 eggs ; but as many as 1400 flies 

 have been reared from a gall of Andricus radicis. In 

 this case, however, it is certain that this gall was the 

 conjoint work of two or more individuals. The labour 

 of oviposition with the sexual broods which lay in 

 leaves is comparatively easy ; but in other cases the 

 work is long and laborious. Thus Biorhiza aptera 

 takes over three days to complete its egg-laying. It 

 has first to pierce the bud with its ovipositor to make 

 tunnels in which to introduce the eggs to the centre of 

 the bud, where they form a solid mass. 



A noteworthy feature in the eggs of the Gynipidse 

 is their swelling, which takes place after being intro- 

 duced into the plant. 



The Larva. 



The larva is footless, white and fleshy, has thirteen 

 segments, and possesses little power of locomotion, as 

 might be expected, except with some species — e. g. 

 Andricus saliens, which can make the galls jump. As 

 may be surmised, the duration of the larval life varies 

 with the spring and winter broods ; while a month 

 is sufficient for Spathegaster baccarum, two or three 

 years are necessary for others. In other words, while 

 one year is sufficient to complete the cycle of life 

 for some species, Andricus radicis requires two, this 



