ANOMALIES IX GEOGKAl'HK'AL DISTRIBUTION. 1533 



sudden outbursts of ahundance, and it is probable that its occasional 

 occurrence in regions vvliere it iius not previously been known is due to the 

 rapid distribution of such sudden outbursts ; thus, while it is rarely found in 

 any numbers north of West Virginia, it has apjieared in several instances 

 at some distance to the north, though these distances cannot be compared 

 to the extraordinary gaps which intervene in the case of other species. 

 One of these is Erora laeta, concerning whose distribution we have clearly 

 nuich to learn ; it had been taken only on one or two occasions in single 

 specimens, at somewhat widely distant localities, usually in elevated 

 regions in the north-eastern United States, when suddenly it was discov- 

 ered in great abundance in southern Arizona, thousands of miles away. 

 It is difficult to discover how these widely separated regions may be con- 

 nected and at the same time to explain why so striking and lovely a butter- 

 fly has escaped the notice of collectors in the intervening regions ; this is 

 indeed one of the most extraordinary instances we have, and may perhaps 

 be partly explained by the fact of the small size of the insect, since we 

 have several instances in the same tribe of butterflies of almost ecjual 

 moment ; thus Incisalia irus, as far as we at present know, is an inhabitant 

 almost exclusively of the eastern lialf of the continent and of the middle 

 section of the United States excluding its northern and southern portions ; 

 but apparently the same species has now been discovered in tolerable abun- 

 dance on Vancouver's Island. Or consider its ally, Incisalia niphon, known 

 only from New England and the regions immediately about it, with an 

 extension down the Alleghanies, but which has also been found in central 

 Texas and central Colorado, regions where collectors are of course much 

 more scarce than in the intervening districts. 



A partial explanation may perhaps be ofl^ered by the distribution of 

 Incisalia augustus, which, occupying in the east a region very similar to 

 that of Incisalia niphon, has also been found in Colorado and the region 

 to the west of it, but which we also know extending in a narrow belt 

 across the country north of our boundaries ; it is in the highest degree 

 probable that in the same belt the other species may be found when these 

 have been properly searched at the appropriate season. A similar expla- 

 nation may perhaps be given for the distribution of Thecla calanus and 

 Thecia acadica as plotted upon our maps, which are known both from the 

 extreme east and the extreme west but have not been taken in suffi- 

 cient number in localities in the Rocky ^Mountain and elevated plateau 

 regions properly to connect these two apparent colonies. Or for that of 

 Thecla edwardsii, which occurs in a naiTOw belt from the Atlantic to the 

 Rocky Mountains along the 40th parallel or thereabout and has been 

 found in a single instance ten degrees farther north above the westernmost 

 known area of its distribution ; and it explains also, perchance, why we 

 know so little of the distribution of Thecla Ontario, which is so exceedingly 



