1 84 



LILLIE AND KNOWLTON. 



[VOL. I. 



It will be noticed that the form of the curve is similar to 

 the temperature regeneration curve of Planaria. The angle in 

 the curve of Amblystoma would probably disappear with a 

 sufficiently large number of observations ; that is, it is prob- 



475 

 45 

 425 

 4QO 



375 

 35 

 3-5 

 300 



275 

 25 



22 5 

 2OO 



175 

 i5o 



125 



100 



75 

 5 



3 6 9 I2 15 l8 2I 2 4 27 30 33 36 39 42 45 



FIG. 2. Curves showing the effect of the normal temperature range on the early development 

 of Amblystoma and the frog constructed from the above tables The broken line is the 

 curve of Amblystoma. The ordinates give the time in hours, 25 hours for each abscissa ; 

 and the abscissae represent degrees Centigrade, 3 for each ordinate. 



ably clue to individual variation in the eggs observed. It is 

 important to notice that the optimum and maximum tempera- 

 tures actually coincide. This is due to the difference in tem- 

 perature range of the dark and light hemispheres ; the cells of 

 the light hemisphere are much more easily affected by low and 

 high temperatures than those of the dark (see section on sub- 

 minimal and supramaximal temperatures). The cells of the 



