46 TOBACCO BREEDING. 



In order that the grower or breeder may select seed plants intelli- 

 gently it is necessary for him to form an ideal of the type of plant 

 which it is desirable to grow. Without a clear conception of the type 

 of plant desired any improvement by seed selection will be accidental, 

 and as a rule the efforts in this direction will be unsuccessful. It is 

 also necessary in forming the ideal to keep in mind the purpose for 

 which the tobacco is produced in order to develop a type which will 

 meet the demands of the market. For instance, in the growing of 

 cigar-wrapper tobaccos a broad, round leaf, adapted for the cutting of 

 the largest possible percentage of wrappers, is most desirable, so far 

 as the shape of the leaf is concerned. The production of a high- 

 yielding type must be governed in all cases by the effect of such 

 change in the size and number of leaves upon the quality of the 

 tobacco. The information necessary for the intelligent selection of 

 desirable plants for seed can onh^ be gained by a careful study of the 

 plants, the cured and fermented product, and the market demands. 



The plan of selection of seed plants followed by the writers is to 

 examine with greater or less care, several days before topping, ever}' 

 plant in the field from which selections are to be made. As indi- 

 cated before, it is always advisable to study the plants from the time 

 they are set out, whenever this plan is practicable, with a view to 

 picking out the best plants for seed. Such plants, when found, can 

 be marked with a tag, string, or heavy rag, so that they can be 

 readih' identified when the final selection is made. Some characters, 

 such as time of maturity, are more easily observed in the young 

 plants than later; hence the importance of marking the plants show- 

 ing the characters desired whenever they are found. 



The size of at least three leaves in apparently desirable plants 

 should be measured — one at the bottom, one near the middle, and 

 another at the top of the plants. These measurements can easily be 

 made with an ordinary yardstick, taking the length from the point 

 of attachment of the leaf to the stalk to the tip and the width at the 

 broadest point. The development of the top leaves b}?^ reason of 

 further growth can be taken into consideration, tho in most tobaccos 

 well-developed top leaves are correlated with early-maturing plants 

 and always with uniformity of leaves on the same plant. The 

 shape, size of veins, color, texture, and other characters of the leaves 

 should be taken into consideration. 



The number of leaves should be coimted, also the number of suck- 

 ers, and observations made, and, if possible, recorded of the uni- 

 formity of the shape, size, and other characters of the leaves in dif- 

 ferent jjortions of the plants, the presence of rust or other fungous or 

 bacterial diseases, the height of the plants, and the space between the 

 leaves, or length of internodes. The transmission of these charac- 



96 



