52 



together and plant food is scarce, the}^ ma}^ develop cannibalistic pro- 

 pensities and begin feeding on adjacent unhatched eggs. Owing to 

 the way the eggs are scattered al)out in nature, such an occurrence 

 must be rather rare under normal conditions. 



PERCENTAGE OF EGGS THAT HATCH. 



In all lots of eggs laid by fertile females, the percentage that fail to 

 hatch is so small that it was disregarded except in a few cases where 

 careful counts were made. These show that of 493 eggs 6 did not 

 develop, giving an average of 1.22 per cent, or about 1 in a hundred. 

 This is doubtless the approximately^ correct percentage for the hatch- 

 ing of eggs laid out of doors. 



LENGTH OF THE EGG STAGE. 



During the summer of 1904 a long series of observations was made 

 in the laboratory b}^ Mr. A. A. Girault to determine the length of the 

 egg stage at different dates during the entire season and at var\dng tem- 

 peratures. In all, the developmental period for over 4,300 eggs was 

 recorded. These results may be tabulated as follows: 



Table XIV. — Length of the egg stage at different dates throughout the season at Paris, 



Tex., 1904. 



Date of hatching. 



Length. 



Date of hatching. 



Length. 



Date of hatching. 



Length. 



April 14 



Days. 

 8 



Ih 

 8 

 5i 

 4i 

 4^ 

 4i 

 3^ 

 4 



f 



Juhe 23 



Days. 



f 



3 



2| 

 2| 

 2i 

 2i 



August 31 



Days. 

 21 

 2h 

 2i 

 3? 



April 15 . 







April 21 



June 26 



September 12 



April 28 



June 27 



September 25 



May 18 



July 10 



September 30 



3' 



May 20 



July 12 





4i 



Mav 23 



July 13 



OctoJjer 1'^ 



4 



May 29 



July 18 



October 14.. 



4i 



May 30 



July 19 



17 





July 25 - . 









Aug. 5 











A clearer conception of the varying length during the season may 

 be had from the following curve (figure 4), which is based on the data 

 given in the table: 



In general it is seen that during the warmer parts of the season the 

 egg period is much shorter than in the spring and fall. Although the 

 embr3^onic period is thus inversel}^ proportionate to the temperature, 

 it is not so in any constant ratio. A number of calculations regarding 

 the sum of effective temperatures^' to which different lots of eggs have 



« Following the theory of Merriam (Nat. Geog. Mag., VI, 229-238, 1894), the sum 

 of mean daily temperatures above 43° to which the eggs had been exposed during 

 development was calculated with the following result: Lot 1, April 14-22, 189°; 

 Lot 2, June 22-25, 111°; July 15-18, 105°; November 1-17, 203°. Assuming 45° to 

 mark the inception of embryonic development, the figures agree somewhat more 

 closely: 173, 105, 100, 169. The summer sums are lower than the spring ones in 

 either case. 



