THE CANADIAN ENTOMOLOGIST. 229 



Delaware and Morrow counties to a point in Richland county about a 

 mile west of the village of Ontario, when it changes again to the north- 

 west, crossing the north-east corner of Crawford and the south-east 

 corner of Seneca, then a little east of north to a point near Lake Erie, a 

 mile and a half south-west of the city of Huron, Erie county. This 

 area in Erie county is, however, but little more than a peninsula-like 

 extension, and will probably not appear again. Near the south line of 

 Erie county the line of demarcation makes a broad sweep to the south- 

 east, thus leaving both the north-west and north-east corners of Huron 

 county unoccupied, as well as all of Lorain county, except the southern 

 portion and south-eastern border, and the western end of Cuyahoga 

 county. Just west of Cleveland another peninsular extension occurs 

 lakeward, where the Cicada appeared for a few days at first, but sud- 

 denly disappeared before the brood reached its maximum in numbers in 

 the adjacent counties. This also will hardly appear again. From this 

 point to near the eastern end of Lake county the insect keeps well back 

 from the lake, though it formerly occupied ground much nearer to the 

 shore. The eastern terminus also comprises but little more than a 

 promontory, as the course here changes broadly to the south-west and 

 then to the south east, leaving a considerable portion of eastern Geauga 

 county and the north-east corner of Portage county unoccupied. The 

 dividing line here only touches the south-west corner of Trumbull county 

 and includes the western end of Mahoning and Columbiana counties 

 and the southern border of the latter, the line passing into West Vir- 

 ginia or Pennsylvania, near East Liverpool, Ohio, where this year a very 

 few Cicadas appeared, and where brood XV. overlaps brood XX. 



I shall be obliged to confess that when I began to map out the 

 area covered by brood XV. it was with more enthusiasm than I could 

 command when I finished the survey. The map indicates, with a good 

 degree of accuracy, the area over which the brood occurred in 1897, but 

 that it will as accurately show the area covered by the brood in 191 4, I 

 have no expectations. The continued destruction of forests and the 

 inroads made upon the brood by its natural enemies will result in great 

 changes, not only in the outline of the area of habitation, but this will 

 be composed of more and more isolated and continually decreasing 

 " Cicada Islands," as I might term them, until the well-known notes of 

 the male will have ceased forever, and the voiceless female will have 

 followed her spouse into the shades of oblivion, 



