io8 On the Formation of 



great natural difficulties the fly has to overcome in 

 laying the ^^^^ with this degree of accuracy, it is not 

 surprising if many eggs come to lie amiss. It is not 

 possible to believe that the emerging larva can alter its 

 position in any way, because it has not the means of 

 locomotion; and, besides, the ^%% is so closely sur- 

 rounded by the tissues of the bud that any movement of 

 the larva is rendered impossible. 



I certainly believe that failures are most common in 

 those cases where buds are pricked, and where the 

 emerging larva produces a bud-gall ; they occur much 

 less frequently in those species which produce leaf- 

 galls, because in these the fly can utilize a much 

 wider territory, extending over the whole of the rudi- 

 mentary leaves in the bud. But even then it certainly 

 does sometimes occur that eggs are not within reach 

 of the rudimentary leaves, and unavoidably fail. Another 

 confirmation of the view that malposition of the ^^% is 

 the chief reason for failure of gall formation, is found in 

 the fact that failures are not usually observed in those 

 cases where the fly cannot easily miss the cambium-zone. 

 This is well seen in those species that prick the bark 

 and the surface of the leaf; in these, only the outer 

 layer of epidermis requires to be pierced, and the cell 

 region sought for is always situated about one uniform 

 depth ; while in the case of buds of indefinite size and 

 shape, the form of the bud axis, sometimes longer, 

 sometimes shorter, gives a varying measure for the depth 

 to which the egg must be sunk. 



The development of the gall is closely dependent 

 upon the vegetative periods of the oak, and ceases with 



