58 MISCELLANEOUS PAPERS. 



HOW TO GROW PURE SEED OF GOOD QUALITY. 



It is a well-known fact that varieties of cotton become mixed and 

 impure unless special care is taken to prevent crossing with other vari- 

 eties. If growers receiving seed of any of the varieties sent with this 

 circular desire to grow the same variety another year, precaution 

 should be taken to plant the seed in an isolated patch, situated as far 

 as possible from any other varieties. It should be at least a quarter 

 of a mile from any other cotton and preferably in a field surrounded 

 by a forest, particularly on the side nearest to other cotton fields. 

 Before any seed is gathered for planting all plants which are not true 

 to the tvpe of the variety should be carefully weeded out. 



If it is desired to keep the variety up to its full productiveness and 

 better adapt it to local conditions, the planter may easily accomplish 

 this by following a simple and inexpensive method of selection. Before 

 beginning the picking, go over the patch carefully and select and 

 mark with a white cloth the best plants; that is. those most productive, 

 earliest in ripening, and having the largest, best formed, and most 

 numerous bolls. Care should also be exercised to select plants that 

 are true to the type of the variet}^. Before each picking send a care- 

 ful man over the patch to pick the seed from the selected plants. 

 Preserve such seed separately, gin it separately to avoid mixing, and 

 use this to plant the crop the next year. If this simple method of 

 selection is carried out each year, the yield will doubtless be greatly 

 increased, and as much or more added to the crop than would result 

 from special fertilization or cultivation, though these factors should 

 by no means be neglected. The importance of careful seed selection 

 is seldom fully recognized, and growers are urged to give this factor 

 of cotton culture more careful attention. 



Herbert J. ^'ebber. 

 Physiologist, in < %arge of Laboratory of Plant Breeding. 



Approved: 



A. F. Woods, 



Pathologist and Physiologist. 



REPORT OF RESULTS DESIRED FOR PUBLICATION. 



In order to determine the comparative values of the different varie- 

 ties of cotton in various parts of the United States, the growers 

 receiving this seed are requested to give it a thorough trial in com- 

 parison with the variety or varieties that they generally grow, and be 

 prepared in the fall of 1903 to report the results of the test to the 

 United States Department of Agriculture. A report will then be 

 requested covering the following points: 



(1) Character of the soil. 



