BY VERA IBVVIN-SMITH. 253 



of living larvae were found, full pupal cases were very scarce, and the majority 

 of those found contained dead and decayed pupae. 



Under these circumstances it was impossible to carry out breeding experi- 

 ments with flies obtained in the adult stage, and it became necessary to rely 

 on those bred out in the laboratory. Unfortunately, in most cases they emerged 

 singly, and at such long intervals that flies of both sexes were rarely alive at 

 the same time. However, on the morning of 11th November, two flies, a male 

 and a female, emerged in the one jar, within a short time of one another. They 

 immediately crawled up the grass stems, and remained there, quite motionless, 

 for the first two days. On the third morning, 13th November, the intense light 

 of a hot day seemed to have roused them to activity, for they were found during 

 the morning, engaged in a "courting" dance in the sunshine, following one 

 another in rapid flight up and down the grass stems, and sides of the vessel. 

 Copulation was not observed, but at seven o'clock on the following morning, 

 14th November, a small white cluster of eggs was found deposited in a single 

 irregular clump, against the side of the jar, on a level with the surface of the 

 soil. Next day the male died, and the female lived only one day longer. 



In view of the previous failures with eggs which had been handled, it was 

 judged safer to leave these eggs entirely undisturbed, and trust to the very 

 imperfect observation afforded through the glass of the jar, to determine 

 whether any changes occurred in them. The lid was kept on the jar, and a 

 slight amount of moisture retained in the soil. 



At the end of the first week, the eggs appeared to darken in colour, and 

 by 27th November, the fourteenth day after deposition, they had become amber 

 coloured. No further observation was made until 30th November, when the 

 egg clump was not visible at all. The soil, therefore, was carefully turned 

 out, in small quantities, on to glass Petri dishes, and examined under the micro- 

 scope. The search was rewarded by the finding of newly hatched, living larvae, 

 no bigger than the particles of soil among which they were crawling, and or 

 eggs which contained unhatched, but well-developed larvae. The latter were 

 picked out with a camels-hair brush, and kept, for further observation, on a 

 small quantity of damp soil in a small crystal dish. However, either they were 

 already dead, or the change of environment proved fatal, for no further develop- 

 ment took place in them, and after a few days they were all decayed and 

 covered with fungus. 



The quantity of soil in which it was possible to pick out the almost micro- 

 scopic larvae, was obviously insufficient for the growth of grass for their nour- 

 ishment. Accordingly, it was decided that the best chance of carrying out the 

 chief aim of the experiment, the rearing of the larvae to maturity, was to place 

 them in a large jar full of soil, containing a good growth of grass, and refrain 

 from any further attempt to handle them until they had had time to increase 

 greatly in size. 



Unfortunately, this object was not attained. The jar containing the larvae 

 was left undisturbed until the end of February, a period of three months; but 

 when the soil was turned out, no trace of the larvae was found, under the most 

 searching microscopical examination. Difficulty had been experienced in keep- 

 ing the grass alive, and a constant degree of moisture in the soil, during the 

 hottest summer weather, especially during a period when, through absence from 

 Sydney, I was unable to attend to it' myself. The larvae had evidently died 

 under unfavourable conditions, or been lost during the removal of dead 

 grass. Previous experiments in the rearing of larvae collected in the field had 



