278 THE BIOLOGY OF INSECTS 



how these grubs feed by burying the head and thorax in the 

 stems or root-stocks of water-hlies and other aquatic 

 vegetation. They may also be seen with tail-region fixed 

 to water-lily stems or the roots of water Ranunculus, and 

 in this position they are breathing. Attachment of the 

 grub to the plant-shoot is made by a pair of strong, slightly 

 curved, pointed structures which project backwards from the 

 dorsal aspect of the eighth abdominal segment over the 

 tail-end of the larva. These spines serve to pierce the sub- 

 merged plant-tissues and penetrate to the internal air- 

 spaces. Each spine is ovoid in cross-section, has ventrally 

 a thick chitinous wall, and is hollow, containing two large 

 lateral canals separated by a median partition, and also 

 smaller dorsal canals. The arrangement (Fig. 69,C,D,E) has 

 been studied by MacGillivray, in greater detail by Boving. 

 The lateral canals pass into a relatively spacious chamber 

 (atrium) at the base of the spines close to the front edge 

 of the eighth abdominal segment ; into this chamber open 

 the dorsally situated spiracles and the tracheal trunks. 

 The hollow spines penetrating the air-spaces in the plants, 

 open a way by which air can pass into the grub's tracheal 

 system, and the spiracular passage can be opened or closed 

 by suitably arranged muscles. A curiously normal feature 

 of this grub, submerged through its whole hfe, is that it 

 possesses the pair of thoracic and the seven other pairs of 

 abdominal spiracles usually found in insect larvae that Hve 

 in the upper air. It appears that these spiracles, exposed 

 on the surface of the cuticle, are indeed functional, but 

 only for the expiratory phase of the creature's breathing. 

 Thus it takes in air from the plant tissues pierced by its 

 spiracular spines and breathes out through the series of 

 small paired spiracles along the main length of its body 

 which is always surrounded by water. *' That the Donacia 

 larva," as Miall has well remarked, '* should have found out 

 the air- reservoirs of submerged roots, and possess special 

 organs for tapping them, is surely one of the curiosities of 

 adaptation." 



The pupae of the Donaciine beetles are found in closely 



