288 The Scottish Naturalist. 



The galls made by the Gall-midges differ much from those made 

 by Gall-flies (Cynipidae), and by Sawflies (Tenthredinidae), inas- 

 much as the former galls do not begin to appear until the larva? 

 have emerged from the eggs ; and are due to some effect pro- 

 duced by the larvae, while the latter are often fully formed while 

 the eggs are still unhatched in their interior, and must therefore 

 be entirely independent of any action of the larvae in their forma- 

 tion and growth. 



The galls of Cecidomyidae show a considerable diversity as 

 regards the parts of plants attacked, and in the effects produced 

 by them on the plants, as well as in the internal structure of the 

 galls. 



We find among them every transition from the scarcely altered 

 tissues in the flowers of grasses tenanted by midge larvse, through 

 slightly deformed and thickened leaf-margins, or slightly swollen 

 flowers, to galls of a much specialised kind, such as those of Lasio}^- 

 tera Ruhi on Raspberry, or of Cecidomyia Galii on Lady's Bed- 

 straw. In one respect, however, even the most specialised Midge- 

 gall falls short of the galls of the Cynipidac^ viz., in complexity of 

 structure. In none of the Midge-galls can we recognise clearly the 

 different layers of tissue so characteristic of the true Gall-flies' 

 work. Yet they differ among themselves very greatly in respect 

 of complexity of structure. Most of them belong to the class of 

 open galls, in which a small opening communicates between the 

 cavity of the galls and the outer air. Some of the galls of this 

 kind are sunk in the tissues of the plants ; but most are entirely 

 external, and consist of certain parts of the plants rendered fleshy, 

 and rolled round to form a cavity for the protection of the larva-. 

 In the most perfect galls of this family of Diptera, the larval 

 chamber is entirely closed. 



