1905.] IMPROVEMENT OF TOBACCO BY SEED SELECTION. 1 85 



of the seed plants is extremely important, because the char- 

 acter and the comparative value of the tobacco cannot be wholly- 

 determined by the study of the plants in the field. From 

 the fact that it requires some care and continuous attention, 

 in order to keep the individual samples in order and good con- 

 dition, this test of the seed plants is usually neglected. The 

 leaves from the seed plants are not ordinarily of as good 

 quality as the leaves from the plants that have been topped, 

 and this fact must be kept in mind during the comparison in 

 the warehouse. 



In making selections of seed plants it is desirable to run 

 over the entire field and mark the best plants for a second 

 examination. This preliminary selection can be made just 

 before topping, but it should not be done before the plants 

 are pretty well developed, so that the type can be determined. 

 A second, more careful selection should be made, and each one 

 of the selected plants examined with regard to the character 

 and features which are important in the variety of tobacco. 

 The best plan for making such studies is to follow some regu- 

 lar outline of points, such as are used in the plant breeding 

 records of the Department of Agriculture. In this way the 

 definite data secured as to the actual condition of the plants 

 can be kept for use when the leaves are compared in the ware- 

 house. The number of desirable leaves should be noted, and 

 in counting them only the leaves useful for manufacture should 

 be counted. The size of leaf can be pretty accurately deter- 

 mined by measuring the middle leaf of each seed plant. The 

 length should be measured from the tip to the point of at- 

 tachment of the leaf to the stalk, and the width measured at 

 the widest point of the leaf. The shape of the leaf is somewhat 

 difficult to record, and the most effective means of making this 

 note is to describe the shape as briefly as possible. The number 

 of suckers is easily recorded, and the time of ripening of the 

 leaves should be kept in order to throw some light on the time 

 of maturity and the comparative earliness of the plants. In 

 addition to these notes a record should be made of the quality 

 of the cured tobacco when the cured tobacco can be examined. 



The number of seed plants to be saved depends upon the 

 size of the tobacco fields for which they are expected to sup- 

 ply plants. Each seed pod contains from five to seven thou- 

 sand seed, and, as there is from fifty to one hundred and fifty 



