BY R. J. TILLYARD. 391 



Thus it appears that considerable moisture is essential to the 

 survival and development of the embryo. Consecjuently, the 

 oxygen necessary for the metabolism of the growing embryo 

 cannot be obtained directly from the air, but nuist be alisorbed 

 from the moist medium surrounding the i^i^g. Tt appeared to 

 me, therefore, that if I could lessen the necessary oxygen-supply 

 in some manner, I might succeed in so weakening the embryo, 

 that its emergence from the egg \vould take a long time to 

 accomplish, instead of being completed vigorously in about half- 

 a-minute. By this means, it would be possible to observe at 

 leisure those processes which could only be imperfectly noted 

 under normally active conditions. 



Having brought the eggs of Anax pnpuensis safely home in 

 their piece of stem of Water-Milfoil, I placed them in a fairly 

 large Petri dish, with plenty of water. The stem was then 

 broken up inttt small pieces, from which a number were selected 

 for examination. From these pieces, Ucenty-eight eggs were 

 extracted without any damage, and were cleai'ed as far as pos- 

 sible from vegetable tissues clinging around them. Several eggs 

 were inadvertently damaged, and were thrown away. Eight eggs 

 were cut out from a hard part of the stem, and were left sur- 

 rounded by small blocks of vegetable tissue. All the pieces of 

 tissue not containing eggs were returned to the Petri dish, which 

 also contained twenty eggs in situ in the remainder of the 

 Milfoil stem. 



The twenty-eight eggs free of vegetable tissues were then 

 placed all together in a small crystal dish, two-thirds filled with 

 tap-watei', and covered with a glass slide. The eight eggs 

 enclosed in small blocks of vegetable tissue were similarly placed 

 in a second crystal dish. The Petri dish was also kept covered. 

 The three dishes were only occasionally uncovered for purposes of 

 examination. Fresh tap-water was only added sutiiciently to 

 make up for slight losses by evaporation. 



If wi' now denote the twenty-eight eggs as Series A, the eight 

 eggs as Series B, and the twenty normall}' placed eggs as Series 

 C, we can easily state the object of the experiment. Firstly, 



