284 



The Journal of Heredity 



distinct improvement over them in 

 some respects. Such improvement ap- 

 parently could not be obtained from 

 the standard variety by simple selection 

 alone, as has been shown by such trials 

 carried on simultaneously. 



The chief advantages claimed for 

 Connecticut Havana No. 38 are due to 

 the desirable habit of growth it presents 

 in the field, the leaves being quite erect 

 in position, rendering the necessary 

 field operations more easily accom- 

 plished. The leaves are quite uniform 

 in size from top to bottom of the plant 

 and are somewhat broader in proportion 

 to their length than those of the ordinary 

 Havana seed. It possesses on the 

 average one or two more leaves to the 

 plant than does the ordinary Havana, 

 while, at the same time, it has a greater 

 average size of leaf. The quality is not 

 inferior to that of Connecticut Havana 

 seed, at least under Wisconsin con- 

 ditions. Although no carefully con- 

 trolled experiment on the relative yield 

 of Connecticut Havana No. 38 and of 

 ordinary Havana seed has been made, 

 it appears from the data secured on leaf 

 number and size that on the average it 

 will yield 7 to 10% more than the 

 ordinary Havana seed. 



The data upon which most of these 

 conclusions are based are presented in 

 Table I, covering a period of nine sea- 

 sons. During each year measurements 

 and counts were made on about twenty- 

 five plants of each type and the averages 

 only are given for the four types 26, 27, 

 38, and 33, the latter being a product of 

 a continued selection and imbreeding 

 experiment with the ordinary Connec- 

 ticut Havana type. During two or 

 three years all the data could not be 

 secured, either because of insect injury 

 or, as in 1916, when the data were dis- 

 carded as unreliable because of the 

 effect of Mosaic disease upon the plants. 

 In other years the root-rot disease due 

 to Thielavia was a factor in reducing 

 size, but it was considered that the 

 influence had the same relative effe.ct 

 on each, since in these types there was 

 no evidence of variation in resistance to 

 the disease. It should be said here 

 that Connecticut Havana No. 38 is no 



more resistant to the Thielavia root- 

 rot disease than the ordinary Spanish 

 or Havana types of seed, and the 

 grower is likely to be disappointed in 

 low yields obtained on infested soils. 

 This strain preferably should not be 

 grown on Thielavia infested soils. 

 Where the grower expects to plant 

 tobacco on root-rot infested soils or on 

 old tobacco fields, he should obtain a 

 resistant type of seed. Such types are 

 now being developed at the Wisconsin 

 Station in cooperation with the United 

 States Department of Agriculture and 

 may in the future largely displace the 

 susceptible strains of Havana tobacco, 

 provided an equally high quality can 

 be produced in the resistant types. 



Referring to Table I, it will be noted 

 that numbers have been gradually 

 added to the designation. Each digit 

 following the original number indicates 

 an added generation of inbreeding so 

 that the number of generations a seed- 

 plant has been inbred can be readily 

 determined. The stalk height refers 

 to the height of a "topped" plant. 

 The height of topping was in some 

 measure determined b}^ judgment from 

 a commercial standpoint in earlier 

 years, but in the later years the plants 

 were to]3ped uniformly with reference 

 to the position of the first "bald" 

 sucker. The leaf number is the number 

 of marketable leaves. The length and 

 breadth of the leaves are the maximum 

 measurement of these dimensions. The 

 "top leaf" is the uppermost leaf of 

 the plant after topping, the "middle 

 leaf" the largest leaf selected at random 

 at about the middle position of the 

 plant and the "bottom leaf," usualh- 

 the third from the bottom counting in 

 the "sand" leaves. The average leaf 

 length and width is the average of the 

 top, middle, and bottom leaf, and rep- 

 resents quite satisfactorily the average 

 size of the leaves on the plant. The 

 average breadth-index is secured by 

 dividing the average leaf breadth by the 

 average length and multiplying by 100. 

 This figure is a good relative index to 

 the shape of the leaf, i. e., it indicates 

 the width as compared to the length, 

 the greater the number the more nearh' 



