34 
BULLETIN 1248, U. S. DEPARTMENT OF AGRICULTURE. 
In the case of halved seed the greatest frequency was at two 
stems for 2-ounce. 3-ounce. and 4-ounce sets and at three stems for 
5-ounce and 6-ounce sets. 
For quartered sets the greatest frequency was at one stem for 
3-ounee and 4-ounce sets and two stems for 5-ounce and 6-ounce sets. 
A similar study of the behavior of Irish Cobbler potatoes grown in 
Maine indicates a general tendency of comparable sets of diflfe 
/20 
//O 
4Z ^ 6 V 3 £ 
// /a 
Fig. 7. — Irish Cobbkr stem-frequency correlations from whole sets grown at the Virginia Truck Eiperi- 
nient Station. 1915 to 191S, inclusive. 
weights to show like correlations of stem frequency. For example, 
the data for whole tubers as given in Table IS show that the greatest 
frequency in 2-ounce whole seed occurred at three stems, in 3-ounce 
>iems, in 4-ounce and 5-ounce sets at six stems, and in 
6-oui - i stems. 
In the case of halved sets, the 2-oun< - a showed the great - 
uency at two stems, the 3-ounce at three stems, and the 4-ouneo. 
5-ounce >unce halved sets at four stems. 
