MODIFICATION OF COI.OR. 209 



Experiments with increased atmospheric pressure have also been made, but 

 thus far without any decisive results. I conclude, therefore, that as far as 

 variation in atmospheric pressure is concerned, it is a matter of no moment in 

 the color variations of decemlineata. However, altitude in nature is a factor 

 of importance, but whether the modifications produced are the effects of 

 atmospheric pressure or of other factors that are active in alpine regions has 

 not yet been determined. 



From these experiments upon this beetle, extending over a period of ten 

 years, I have reached the following general conclusions : 



(i) Somatic color variations are due to the action of environmental stim- 

 uli which accelerate or retard the physiological processes involved in color- 

 pattern formation, and thus cause the color pattern and its elements to vary 

 in definite directions toward an increase or decrease of pigmentation. 



(2) The environmental factors do not have any specific effect, but each acts 

 as a stimulus, which, when applied to the animals, produces modifications in 

 coloration by accelerating or retarding the physiological processes which are 

 productive of coloration ; hence many different stimuli may produce identical 

 results. 



(3) Temperature and moisture are the two prime factors in the production 

 of color changes, and of these moisture is the more important. 



(4) Soil acts as an important factor indirectly through moisture. 



MODIFICATION OF COI^OR IN I.. SlGNATlCOIvIylS. 



It has already been shown that signaficollis is a stable species, of low varia- 

 bility and of very conservative constitution ; and I have therefore selected it 

 for experimental work as the type of one extreme in certain species of the 

 genus, while decemlineata represents the other extreme. My experiments 

 w4th signaticollis do not extend over as long a period as those with decemline- 

 ata, but the results are quite as conclusive. 



Tkmpkrature Experiments. 



Experiment 28. — To determine the effects of a slight average increase in temperature 

 upon the coloration of L. signaticollis. 



Conditions. — Temperature on the average 6° C. above that in control, with 

 other conditions normal. 



Apparatus. — Glass tank, as in experiment 23. 



In this experiment 800 larvae from three generations were used. The 

 material was brought from Cuernavaca, Morelos, Mexico, and subjected to 

 experimentation in the greenhouse at the University of Chicago. The tem- 

 perature conditions were as shown in table 97. 



