198 



GENEKAL BOTANY 



grow under like conditions. It seems probable, therefore, that 

 the running out of selected races of corn, beans, and other 

 cultivated plants mentioned in the previous pages is due to 

 the selection, on the part of cultivators and breeders, of plants 

 with these fluctuating variations. Fluctuating variations can be 

 accumulated to a certain extent, and a better race may thus be 



produced ; but such 

 a race will always re- 

 main inconstant and 

 limited in range or in 

 the degree to which 

 it can be improved. 

 Thus, in the tobacco 

 plant (Fig. 107) great 

 variation has been in- 

 duced in certain varie- 

 ties by planting in the , 

 soil and climate of 

 Connecticut and other 

 Northern states seed 

 obtained in Sumatra 

 and Florida. Some of 

 these variations are of 

 the fluctuating variety, 



Showing uniformity of type in tobacco due to selec- 

 tion of seed. Photograph by the United States Depart- 

 ment of Agriculture 



FIG. 108. Uncle Sam Sumatra tobacco, selected 



and so will not form 

 the basis for a stable 

 improved race pro- 

 duced by selection. In other instances the variations induced in 

 tobacco by the change in climate and soil are apparently stable 

 from the outset, and form the basis for new constant races. 

 The Uncle Sam Sumatra tobacco (Fig. 108) is supposed to have 

 had such an origin by mutation. 



The Florida-Sumatra tobacco was grown from seed secured from 

 the island of Sumatra. When the Florida-grown seed was taken to 

 Connecticut, the plants grown from it varied in a marked degree, and 

 several new types developed that did not exist in Florida and, accord- 

 ing to the best information obtainable, did not exist in Sumatra. 



