RICE-PEST OF BRITISH BURMA. 



9 



the detached piece upon the leaf that the upper surface 

 of the former comes into apposition with the same sur- 

 face of the latter, and, by thus turning the two conca- 

 vities towards one another, gains a more spacious shelter. 

 Secure beneath the bit of green leaf, it devours the leaf, 

 to which the former is so loosely applied as to give the 

 freest possible access on all sides to the currents of 

 aerated water with which it is necessary that its tracheal- 

 gills should be constantly bathed. In correlation with 

 this habit, we find that its body is markedly depressed, 

 and that the tufts of tracheal-gills all occupy the sides 

 of the body, where, owing to the interval which exists 

 between the leaf and its bit, these respiratory organs are 

 completely exposed to the aerating medium without risk 

 of injury from the pressure of the bit. In the Burmese 

 insect, on the contrary, the body is scarcely at all de- 

 pressed, and the tufts are attached much higher up on 

 the back ; from which facts itmay with confidence be 

 inferred that the creature constructs no shelter, but 

 crawls about free and uncovered in the midst of the 

 water. 



In the Brazilian species (Cataclysta pyropalis) lately 

 described by Dr. Willi. Miiller-Blumenau, the body is 

 still more depressed, the tufts are quite lateral in position 

 and reduced to simple and unbranched filaments, and 

 the shelter is a much more complex structure than in the 

 European species. 



It is worthy of remark that the aquatic caterpillars 

 which possess no tracheal-gill tufts, and may hence 

 be inferred to have entered so recently on their water 

 life that the changed conditions have not had time to 

 produce the structural rearrangements needed for a 

 truly aquatic life, live in a sort of case from which water 

 is entirely excluded, and are no more aquatic animals in 

 the sense that they are provided with special organs for 



