le2ndopterous lance and pup<2. 



175 



In this case the pupa, when first weighed, had under- 

 gone the rapid decrease immediately following pupation, 

 so that the fall in the rate of loss was gradual, as is 

 shown by a comparison of the weighings at regular 

 periods of one hour. There is also irregularity, which 

 may have been due to atmospheric conditions affecting 

 the pupa and the chip-tray. Nevertheless, the fall in 

 the rate of loss is well seen on comparing the first 10 

 with the second 10 of the 20 weighings at regular 

 intervals. In the first there was a loss of 9| milligrams, 

 in the second only 2|- milligrams. Here also subsequent 

 weighings prove that the tray was a source of error. 

 There was a loss of 24 milligrams in the 8 days between 

 Aug. 26th and Sept. 3rd, and a loss of 56 milligrams in 

 the 24|- weeks between Sept. 3rd, 1885, and Feb. 21st, 

 1886, being at about the same rate as in pupa 11. , i. c, 

 rather over 2 milligrams a week, under similar conditions 

 as to dryness, &c. 



Air moist, raining all night. 



