38 ILLINOIS BIOLOGICAL MONOGRAPHS [432 



the group, in the direction of general concentric extension in which 

 the original type of pattern may be recognized (Figs. 160, 167, 169, and 

 181, PL XII ; 204, 204a, 196, PL XIII ; 378 to 437, Pis. XVII and XVIII) . 

 Thus one who inspects the figures as arranged is impressed with 

 the fact that there are a great many directions in which patterns have 

 been modified and these figures are numerous and intentionally substi- 

 tuted for less satisfactory descriptions. The material afforded by the 

 600 or more species is rich in possibilities and excels in this respect the 

 butterflies of Eimer or pigeons of Whitman. 



EXPERIMENTAL MODIFICATION OP PATTERNS 



To test the laws of modification of the typical patterns of Cicindela 

 larvae of several species — C. tranquebarica, repanda, hirticollis, lim- 

 balis, lepida, and lecontei — were subjected to low temperature, high tem- 

 perature, and moist and dry conditions. The temperature was raised 

 about 10 degrees C. above that encountered in the normal outdoor life 

 history. The experiments were carried on in the apparatus shown in 

 figure 455, plate XXIX, and described in connection therewith. 



The larvae were put into the high-temperature (near 37°, 1906; 

 40° C, 1905) about May 15. They were placed in a lamp chimney con- 

 taining fine sand. The apparatus as arranged gave 2° to 4°C. higher 

 temperature at the top than at the bottom. The average of the two 

 was used in computing the mean. Temperatures were taken twice a day 

 as a rule. The temperature rose each day as the sun shone on the 

 cases so that during the hottest weather daily maxima in soil temper- 

 ature went to 40 to 42 degrees at times. 



The results of the experiments on C. tranquebarica so far as the 

 patterns are concerned are shown in figure 456 a to g, plate XX, and 457 

 ato&,458,459 and 460; these should be compared with control 456 a' toft', 

 w' , 457 a' to e f ', w' and 458 a'b' . A comparison of these experiments 

 with their control and the representative of the forms collected 

 in the field from the same generation shows that in the controls the 

 normal middle band reaches to the margin of the elytron where it is 

 expended in the line of the longitudinal band a A; the longitudinal part 

 is parallel with the anal side of the elytron ; the middle band is hooked 

 at the end or turns into a horizontal position in compliance with the 

 normal direction of the transverse band from which it is derived. The 

 humeral lunule is usually hooked. The angle in the middle band is a 

 right angle and there is a forward extension of the middle band at the 

 angle. 



The patterns which result from the experimental conditions almost 

 without exception differ from the control in the following respects : 



