61 



very numerous county correspondents of the Statistical Division of the 

 Department of Agriculture, aud also of the Weather Service, in addi- 

 tion to the regular correspondents of the Division of Entomology, a 

 much more careful and thorough canvass was possible than had ever 

 before been made. The result has been most satisfactory, the range of 

 these two broods being much more accurately defined than ever before. 



Several thousand replies were received in response to circulars sent 

 out, many of which were negative — the investigation being extended 

 throughout all States in which there was any likelihood of the appear- 

 ance of the Cicada, and necessarily covering many counties and dis- 

 tricts where the Cicada was not expected. The results of this canvass 

 up to June 20 were recorded in Bulletin Xo. 14, new series, of the Divi- 

 sion of Entomology. A large number of replies were received subse- 

 quently to that date, and the corrected list of localities is appended, 

 together with a list of the persons reporting and a brief indication of 

 tbe nature of the record. 



With the exception of the southeastern and northwestern range of 

 Brood YI, most of the records for this brood were of scattering indi- 

 viduals, in many localities only a few specimens being observed. It is 

 quite possible also that the records for Ohio, West Virginia, and Virginia 

 in some cases are based on stragglers from Brood V, which occurred in 

 1897. Dense swarms of Brood VI were, however, reported from the 

 mountain counties of Xorth Carolina, South Carolina, and Georgia, and 

 the limits of the brood, in this portion of its range, are now determined 

 with fair accuracy for the first time. The reports from the mountain 

 counties of Tennessee and Kentucky belong undoubtedly also to 

 Brood VI. A number of strong swarms of this brood are reported in 

 Wisconsin and several in Illinois. Some of the latter assigned to 

 Brood VI may, however, belong to Brood XXIII. The reports if they 

 may be relied upon from northern Michigan (Chippewa and Houghton 

 counties) and from northern Wisconsin (Burnett, Sawyer, and Wash- 

 iugton counties) carry the range of the Cicada farther north than any 

 of the old records. 



The reports of Brood XXIII nearly all indicate the occurrence of the 

 insect in enormous numbers. Unfortunately, however, there enters 

 again with this brood some doubt as to the correct reference of some of 

 the localities in Illinois, Indiana, and perhaps northern Missouri, or, in 

 other words, where the territory occupied by the two races overlaps. 

 In most of the records assigned to this brood, however, in the States 

 mentioned the evidence points pretty strongly to the accuracy of the 

 reference. Where there is uncertainty a query follows the county. 



The records assigned to Brood VI, in North Carolina, South Carolina, 

 and Georgia, and in western Kentucky and Tennessee, can not be 

 questioned. The counties represented are in the main in an elevated 

 mountainous district, and the fact that the Cicada is of the 17-year 

 race is established by the elevation or by the earlier records. 



