288 Journal of Agricultural Research voi. xvi, no. h 



of the seed at planting time and the symptoms displayed during growth. 

 These experiments were carried out in the field under conditions approxi- 

 mating commercial field practice, and no methods of culture or treatment 

 were used at any time after planting which would not have been used 

 by a commercial grower. For the purpose of this experiment it was 

 conceded that parts of the same seed potato, grown under like condi- 

 tions, would follow within reasonable limits pretty nearly the same course 

 of procedure in growth, disease symptoms, and general appearance. A 

 difference in two plants from twin seed pieces must be accredited to 

 different conditions encountered during the growing season after planting. 

 Various lots of seed were assembled in 191 6 to test out this assumption. 

 Among others, they consisted of one lot of certified Wisconsin Pearl, 

 one lot of certified Wisconsin Rural, two lots of Early Ohio from the 

 Red River Valley in Minnesota, one lot of Rural from the Carbondale 

 District of Colorado. 



All tubers were cut from bud to stem end, dividing the tuber into two 

 equal parts. All tubers above 6 ounces were cut into four pieces. These 

 were cut in the field and planted immediately side by side in parallel 

 adjacent rows. They were given as good care as possible during the 

 growing season. The summer was excessively warm until July 30, 

 1 91 6, at which time 3.09 inches of rain fell. The remainder of the sum- 

 mer was comparatively cool. Notes were taken four times during the 

 summer: Once when the plants were about 6 inches high, then when 

 they averaged 12 inches high, again when they were full grown, and 

 finally when no normal change was to be expected. No reference was 

 made to any previous note; other members of the force were asked to 

 assist in the work, and every method employed by w^hich an impartial 

 diagnosis could be made. 



KEY To TABLE I. 



Pbr the purpose of summarizing and presenting the performance of these lots of 

 tubers, a new form of table, known as an aggregation table, has been constructed. 

 This table must not be confused with a correlation table, which it resembles in general 

 appearance, but not in context. In any single row of potatoes six different ultimate 

 types of plant were recognized. These have been designated as "H," "HD," "DH," 

 "D," "A," and "O." The meaning of these symbols are as follows: "H" denotes a 

 plant which appeared healthy throughout the growing season; "HD" denotes a plant 

 whicli was healthy during the first part of the season, but finished by being diseased; 

 "DH" denotes a plant which gave manifestations of disease during the first part of 

 the season, but finished by being healthy; "D" denotes a plant which was diseased 

 throughout the season ; " A " denotes a plant so badly diseased as to be merely existing, 

 with no hope of progeny ; "O" denotes no germination, or a case in which the seed piece 

 suffered the maximum of disease and rotted in the ground. All plants in a row fell 

 into one of these divisions. In comparing the plants from twin seed pieces in the two 

 adjacent rows at the same time it is seen that in classification certain coincidences, or 

 lack of tlicrn are significant. In the case of some plants classed as "H" in one row, 

 the twins in the adjacent row Vv-ere "H" also; in more cases they differed for all the 



