April, '19] FLINT. TURNER AND DAVIS: EXPERIMENTAL METHODS ISl 



percentage of infestation, but this method should not be used where 

 accurate records are desired and especially in experiment plots where 

 it is of much importance to determine the stages of the insect, severity 

 of infestation, etc. Likewise this method cannot be used where the 

 wheat has made a heavy growth, for frequently an infested plant will 

 send up new shoots and it is often impossible to determine whether such 

 tillers are individual plants or simply tillers of an infested plant, 

 thus again showing a possible source of error if infestations are secured 

 by simple examination of the plants as they appear above ground. 



Summarizing we are led to conclude that the above-ground appear- 

 ance of plants should be used only in generalizing the infestation, such 

 as heavy, medium or light; that the picking method should be used 

 only for fall scouting work when estimating Hessian fly infestations; 

 and that for experimental plots where simplicity, accuracy and com- 

 parableness are essentials, the linear foot method should be followed 

 and that at least ten linear feet be taken from each plot where counts 

 are required. 



Method of Taking Yields 



While the problem dealing with the accuracy of obtaining yields 

 from small plots is largely agronomic, it is as important for the ento- 

 mologist as for the agronomist to be familiar with the most accurate 

 and practicable means of obtaining these data. 



Various methods have been practiced. Some agronomists insist 

 that reliable records can be obtained only by harvesting the entire plots. 

 Others believe equally reliable or even more accurate yields can be 

 obtained by harvesting only small areas from each plot, usually about 

 one-thousandth of an acre. In taking such small areas as one-thou- 

 sandth of an acre, some prefer to take so many linear rods of individual 

 wheat rows, while others take so many square yards. 



Another year it is hoped that we may have the use of a portable 

 threshing outfit in order to make a comparison of the different methods, 

 but up to the present time we have not had this opportunity and wish 

 simply to place on record the method which has been practiced in 

 obtaining yields in our Hessian fly sowing experiments and to discuss 

 its possible advantages and disadvantages. 



Briefly, the method is to select five square yards from each plot, 

 this to be bagged, shipped to a central point and there threshed, 

 weighed and graded. Probably the most important point to be con- 

 sidered in securing yields from such small areas is the selection of the 

 square yard. Observations indicate that selecting the areas by the 

 hoop or other similar method which depends on chance is not accurate 

 when as few as five square yards are to be taken. A fairer way is to 



4 



