April, '12] - HEADLEE: HEtSSIAN FLY 103 



the stations, but since that time, the stations have been secured and 

 managed by the Kansas State Experiment Station, representatives 

 of the Bureau taking such data as they desire. Taking the experi- 

 mentally-determined safe-sowing date of Marysville (latitude 39° 49', 

 altitude 1153 ft.) as October second, the theoretic safe-sowing dates 

 for Manhattan (latitude 39° 11', altitude 1012 ft.), Sedgwick (latitude 

 37° 56', altitude 1375 ft.), and Caldwell (latitude 37° 4', altitude 

 1107 ft.) for which we have the averages of two or more years of 

 sowings, are October 6, October 7, October 13, respectively, while the 

 actual dates determined by the average of two or more years of experi- 

 mental sowing are October 7, October 9, and October 14. Likewise 

 taking the experimentally determined safe-sowing date of Great 

 Bend (latitude 38° 22', altitude 1843 ft.) (in the western series 

 where only two stations have as yet given promising results) as 

 October 5, the theoretic date for Sa^\yer (latitude 37° 29', altitude 

 1913 ft.) would be October 8, while the actual date by one year's test 

 is October 7. 



The correspondence between theoretic and actual date is close 

 enough that Hopkins' law may be said to apply to the eastern series 

 and western series when considered as separate units. Here, as in 

 the case of Ohio and West Virginia, the only factor of climate suffi- 

 ciently variable within the limits of the individual series to produce 

 such variation in time of safe-sowing in relation to latitude and altitude 

 is temperature. 



For the purpose of further testing the universality of this law, let 

 us take the date of safe-sowing at Marysville as October 2. Apply- 

 ing the law, the theoretic date of Great Bend (latitude 38° 22', 

 altitude 1843 ft.) is found to be October 1, whereas two years sowing 

 tests shows the actual date to be October 5. No such discrepancy 

 as this appeared between the actual and theoretic dates in Ohio and 

 West Virginia according to Hopkins. For the purpose of further 

 testing the law, taking the safe-sowing date at Columbus (latitude 40°, 

 altitude 800 ft.) experimentally determined as September 25, we find 

 that the theoretic safe-sowing dates for Marysville and Great Bend are 

 September 23 and September 21 respectively, whereas the actual dates 

 by experimental sowings are October 2 and October 5 respectively. 

 The actual date at Marysville is 9 days and at Great Bend 14 days 

 later than the theoretic. Evidently, this difference is the indication 

 of another powerful factor. 



When we review the factors of environment known to retard Hessian 

 fly, which do not vary enough to produce a material difference in 

 Ohio and West Virginia, or in the individual stations of either the 

 eastern or western series, but which do vary enough to make large 



