BY R. J. TILLYARD. 



87 



that, by the time I got back, they were all emerging in the 

 box. In any case, the actual internal metamorphoses had 

 proceeded too far fdr an examination of larval structures, 

 even if I had attempted it the moment I took them. 



Returning to the swamp, I set about trying to solve the 

 problem, where and how the larvae lived. My friend. Dr. 

 F. Ris, to whom I had sent specimens of the exuviae, had 

 pointed out to me that, in every case, the anal opening was 

 wide open in the cast-skin. This fact led him to believe 

 that these larvae were not aquatic, but air-breathers. Bear- 

 ing tliis in mind, I selected a large clump of sedge in the 

 middle of the swamp, to which three exuviae were clinging, 

 and carefully examined the clump itself. Close to the base 

 of the stalk on which the first exuviae were found, I dis- 

 covered a neat round liole, about half-an-inch in diameter. 

 On examining the bases of the other two stalks which carried 

 exuviae, I found similar holes. Passing on to other clumps, 

 I found that, in every case where there were exuviffi on the 

 reed-stems, a neat, round hole occurred close by, near the 

 base. Next, with a knife, I cut away the edge of one of the 

 holes, and followed it down. It was very damp, and the 

 sides smooth and plastered with the soft mud of the swamp, 

 enclosing the matted roots of the sedge. At the water-level 

 the hole was, of course, full of water, and thence downwards 

 it got looser, finally becoming indistinguishable in the watery 

 ooze lower down. 



It was now necessary to decide whether these holes wcfc 

 made by the larvae simply for the purpose of emerging, or 

 whether they were more or less permanent, and used as 

 channels of communication, or perhaps for foraging excur- 

 sions, possibly at night-time. As the larvae had nearly all 

 emerged, I had to wait another year before I could carry on 

 this investigation. On November 5th, 1910 (a date chosen 

 as likely to give a prospect of finding the larvae nearly full- 

 fed, but not yet emerged), Mr. C. Gibbons, of Hornsby, and 

 myself went to the swamp at Medlow. We were provided 



