The Resin-bees 



her means; perhaps the barricade, if hardened 

 into a solid block, would afterwards form an 

 invincible obstacle to the escape of the young- 

 sters; perhaps again the mass of gravel is an 

 accessory rampart, run up roughly as a work 

 of secondary importance. 



Amid these doubtful matters, I see at least 

 that the insect does not look upon its barri- 

 cade as indispensable. It employs it regu- 

 larly in the large shells, whose last whorl, too 

 spacious to be used, forms an unoccupied 

 vestibule; it neglects it in the moderate shells, 

 such as Helix ncmoralis, in which the resin 

 lid is lev^el with the orifice. My excavations 

 in the stone-heaps supply me with an almost 

 equal number of nests with and without de- 

 fensive embankments. Among the Cotton- 

 bees, the Manicate Anthidium is not faithful 

 either to her fort of little sticks and stones; 

 I know some of her nests in which cotton 

 serves every purpose. With both of them, 

 the gravel rampart seems useful only in cert- 

 ain circumstances, which I am unable to 

 specify. 



On the other side of the outworks of the 

 fortification, the lid and barricade, are the 

 cells, set more or less far down in the spiral, 



313 



