MODIFICATION OF COI.OR. 



199 



In experiments 23 to 25, where deviations in temperature and relative hu- 

 midity were both high, the results obtained were very much like those already 

 described for experiments in which temperature and moisture were used 

 alone. Not only are the mortality percentages closely parallel, but the order 

 and degree of color changes are the same. I conclude, therefore, that tem- 

 perature and moisture, when varying together in the sam.e direction, act as a 

 stimulus upon the coloration of decemlhieata, producing, with slight devia- 

 tions, an increase in pigmentation, or melanism, and with larger deviations a 

 decrease, or albinism, the curve of the response being the same as that found 

 in experiments where the varying factors are used alone. 



I have also conducted series of experiments corresponding to 23 to 25, 

 in which temperature and moisure varied together in decreasing average 

 amounts. The results, however, were in every way the same as those obtained 

 from like experiments in which these factors were used alone — that is, 

 the production of melanism by slight deviations, and of albinism by large 

 deviations. 



Experiment 26. 



In this experiment I have brought together the results of twenty separate 

 sets of experiments which were conducted during the years 1895 ^o 1904. 

 The object of these was to determine the effects in combination of deviations 

 of temperature and relative humidity in opposite directions by which cool dry, 

 cool moist, hot dry, and hot moist conditions in different degrees were pro- 

 duced. The conditions of the experiments, stated briefly, are as follows : 



Experiments in which the Temperature and Rei^ative Humidity Varied 

 Together in Opposite Directions Above or Below the Normai,. 



No. 



Temperature. 



Relative hu- 

 midity. 



No. 



Temperature. 



Relative hu- 

 midity. 



Above. 



Below. 



Above 

 aver- 

 a.e. 



Below 

 aver- 

 age. 



Above. 



Below. 



Above 

 aver- 

 age. 



Below 



aver- 

 age. 



26^ 



26b 



2ec... .. 



26d 



26<? 



26/. 



26^ 



26h 



261 



267 



26/^ 



6 

 10 



'I 



10 



13 



6 



10 



13 



65 

 6.5 



Pe> 



. 



I 

 2 



' Ct. 



' 







I 



Per ct. 1 

 10 i 

 10 i 

 10 i 

 21 i 

 21 

 21 ! 



35 

 35 

 35 



1 



26/ 



26m 



26n 



260 



26/ 



26^ 



26r. 



26:? 



26/f 



26U 



° 



C. 



6.5 

 10 

 10 

 10 

 15 

 15 

 15 

 24 

 24 

 24 



Per ct. 



35 

 10 

 21 



39 

 10 

 21 



39 

 10 

 21 

 39 



Pe 



ret. 



In these experiments, in which 10,600 larvae, hatched from about 17,000 

 t.g'g^, were used, 39 per cent died in the larval stage, 31 per cent in the pupal, 



