84 EXPERIMENT STATION. [Jan. 



The plants liave l)een grown each year upon a different plot of 

 ground and those have not always been as satisfactory, especially 

 as to uniformity, as might be desired. In 1908 the plot, while 

 fairly uniform, was gravelly, and sutlered somewhat from 

 drought, which modified the character of the plants grown to 

 a considerable degree. In 1900 the soil was heavier, but one 

 end of the plot was inferior as shown by the appearance of the 

 plants grown. In 1910 the soil of the plot seemed fairly uni- 

 form, but the error was made of applying fertilizer in the row. 

 While an effort was made to have this uniform all through the 

 plot, it appeared that it was not fully successful, some portions 

 of the rows receiving more stimulus than others. The plots 

 used in 1911 appeared to be more desirable than those used pre- 

 viously, and were so on the whole, yet they were not all that 

 could be desired, some portions being evidently poorer than the 

 average, as indicated by the slightly less flourishing plants. 

 Perfect uniformity of soil, however desirable in work of this 

 kind, is very difficult and perhaps impossible to attain, and this 

 must be compensated for by duplication of results. It is felt 

 that in this work sufficient dui)lication has been carried out to 

 neutralize this variation in soil conditions, and that the conclu- 

 sions reached are not materially affected thereby. 



The mathematical calculations have been carried out with the 

 aid of millionaire and comptometer calculating machines and 

 fully checked, and it is felt that they are free from errors that 

 could sensibly affect the results. The methods that have been 

 used are for the most part the usual ones and substantially as set 

 forth in " Principles of Breeding," by E. Davenport. 



Heredity. 

 In what degree may individual pea plants be expected to 

 transmit their characters to their descendants ? Table T. sums 

 up the measure of inheritance of vine length and productive- 

 ness of about 10,000 plants in an effort to throw light on 

 this question. Before discussing this table it is necessary to set 

 forth the history and nature of the four groups dealt with. 



