11 



These results arc abundanth' corroborated b}' Professor Sanderson^ 

 who states: 



We have made vsimilar experiments with over 500 weevils and were unable to find 

 over 30 per cent dead at the end of 4 or 5 days, while an average of about 9 per cent 

 were found alive at the end of 5 days. The balance of over 60 per cent were missing. 



About 10 per cent of the marked weevils recorded as missing- upon 

 examinations of treated plants were subsequenth' found alive upon 

 other plants. A careful series of tests has shown that weevils are 

 incapable of flight after having eaten Paris green. It has also been 

 found that on the average one ten-thousandth of a grain constitutes a 

 fatal dose. It would appear fairly certain, therefore, that nearly all of 

 the missing weevils really escaped alive. If so, then the number of 

 weevils actually found dead approaches ver}' closely to the total num- 

 ber killed, while the number found alive is really less than one-half of 

 those that actually escaped. The period of the emergence of the 

 weevils from hibernating quarters is known to extend over from G 

 to 10 weeks, so that at least four applications of poison would be nec- 

 essary to keep the cotton poisoned thoroughly enough to kill as large 

 a percentage of weevils as was killed in the foregoing tests. 



Summary. — Summarizing all of the tests with dr^- Paris green on small 

 areas gives a total of 732 weevils used. The average time between treat- 

 ment and last examination was 51 hours. An average of 27.12 per cent 

 of the weevils were found alive, 32.05 percent dead, and 10. 89 percent 

 missing. The lightest application of poison was 20 pounds per acre, 

 and four applications at this rate could not possibly be profitable, 

 especialh' since the last applications would be after the cotton had 

 squares, and the number of weevils the poix^on would kill would be 

 greatl}'' reduced. 



No tests were made with very light applications, but the following 

 experiments made on the Government farms and by individuals show 

 the inefficiency of Paris green where the i^oison was applied at a rate 

 of li to 3 pounds per acre for each application and used throughout 

 the season. 



TREATMENT OF LARGE AREAS DURING SEASON OF 1904. 



Three experiments were conducted by this Bureau at different 

 places, using Paris green on one plat and making frequent applica- 

 tions throughout the season, keeping an adjoining plat as a check, both 

 plats being planted at the same time with the same kind of seed and 

 both cultivated exacth' alike. 



Field experiments at San Antonio. — At San Antonio a field of 1.70 

 acres planted with King cotton was chosen. This field was compara- 

 tively isolated, there being no nearby cotton except on one side and 

 this was separated from the test fields b}^ a dense strip of sorghum 124 



211 



