THE CANADIAN ENTOMOLOGIST. 91 



principal masses of individuals occui near the extremes of the pattern- 

 distance, and the intermediate position is in comparison thinly occupied. 

 A divisive attitude is manifested. 



Additional to tht,- flown males, my series contains 74 males matured 

 in the house from estray pupiB or wild larvae. Of these, 13 are Satdderl ; 

 II are intergrades very near Scudderi ; 15 are somewhat more divergent, 

 yet pretty near to Scudderi. One is a stage agreeing with section 4 of 

 the flown males ; 6 are nearer to Petrosa than to Scudderi, and intergrade 

 closely with fig. 1 1 of the plate and with section 5 of the caught males, 5 

 of them being intermediate between fig. 1 1 and section 5, and the other 

 I between section 5 and fig. 8 of the plate. Twenty-eighty are formal 

 Petrosa. These 74, then, distribute as to pattern development in about 

 the same numerical proportions as the caught males, and with a similarly 

 feeble representation at the centre of the pattern-progress as compared 

 with the extremes. 



The caught females are but 25, all told. Of females matured in the 

 house from wild larvee and estray pupse, there are iii. Total number 

 of females, not including those bred from the egg, 136. Of these, 16 are 

 Scudderi, 20 are intergrades very near Scudderi, 26 are more divergent, 

 yet all pretty near to Scudderi; 9 are of composite pattern, combining in 

 the same individual a considerable degree of Petrosa character, as to 

 some parts of the pattern, with a predominant Scudderi tendency in other 

 pattern elements ; 9 are well-balanced intermediates between Petrosa and 

 Scudderi ; 18 are gradations near to Petrosa; 38 are formal Petrosa. 



The females display a pattern sequence more even and harmonic 

 than that of the males, with less tendency to break, into divisions by lack 

 of intergrades, and more fully sustained in the central part of the chain 

 of variation. At the same time, the females vary as extensively as the 

 males, and they exhibit at least an equal amount of aberration and as 

 great a degiee of inequality or oscillation in the progress of the several 

 pattern constituents. 



All through the range of variation, in both sexes, it is conspicuously 

 evident that Petrosa is exceedingly unstable in regard to the relative 

 development of the various elements of pattern. This is sufticiently 

 illustrated in figures 3 and 5, also by figs. 9 and 11. The oblique bar of 

 primary is the only element which approaches fixedness. The f w. cell- 



