THE CANADIAN ENTOMOLOGIST. 141 



This development of macropterism in so many species usually 

 regarded as normally brachypterous is the most interesting feature in the 

 Orthoptera of the Thunder Bay District. We find the same tendency in 

 less degree in Northern Michigan, as given in the papers cited above, but 

 farther south long-winged examples of some of these same species are 

 unknown, or of such rare occurrence that they are usually considered 

 somewhat abnormal and as representing cases of reversion to an ancestral 

 type. Such species are, e. g., Chloealtis conspersa, C. abdominalis, 

 Melanoplns fasciatus and M. extremus, though in the last named species 

 the macropterous form is relatively common. The question arises : Why 

 is the proportion of macropterous to brachypterous individuals so much 

 greater in these northern regions than it is farther south ? 



In, the first place, where dimorphism in wing-length occurs the fully- 

 winged type is of course the more primitive one, the flightless type the 

 more recently evolved. The species is tending to become wingless, and 

 the short-winged individuals are therefore better adapted to their environ- 

 ment, while the long-winged individuals are gradually becoming 

 eliminated. Where this process has been most completely carried out 

 only the flightless type remains, and in such cases the species may be 

 apterous, as in Podisma glacialis, or extremely brachypterous, the wings 

 remaining as mere vestiges, as in Melanoplus islandicus. 



Now, where the environment is least favourable to the needs of the 

 species, or where favourable conditions are localized, the elimination of 

 the unfit will proceed more rapidly than it will under favourable conditions, 

 so that we might, a priori, expect to find that where the trend of evolution 

 is towards brachypterism this condition will become established most 

 rapidly where favourable conditions of environment are localized, e. g., 

 towards the limits of the geographical range of the species. 



It will be noted that all the dimorphic species in question are boreal 

 forms, and are more abundant and generally distributed in the north, 

 where the long-winged forms are plentiful, than towards the southern limits 

 of their geographical range, where this form occurs only sporadically or 

 not at all. Hence it may be concluded that the elimination of the less 

 fit macropterous forms at the south, where the environment is least suited 

 to the species, has been more complete than in the north, where the 

 conditions are more favourable. 



There is possibly another factor entering into this question. It is 

 well known that differences in wing-length are correlated with differences 



