POTATO BREEDING AND SELECTION. 19 
in May, it is desirable to sow the seed in the greenhouse about March 
20 to 25. (PL V.) The plants are usually spaced in the field in rows 
3 feet apart and are set 2$ feet apart in the row. (PL VI, tig. 1.) In 
tjie ensuing year the tubers are spaced 18 inches apart in the row. 
(PL VI, fig. 2, and PL VII, tig. 1.) 
The method which has been pursued in determining the merits of 
the seedlings which have been grown by the Department of Agri- 
culture since 1910, some 60,000 in all, has been rather more compre- 
hensive than is considered advisable from a commercial standpoint. 
Out of some 28,000 seedlings grown in 1910, nearly 19,000 developed 
tubers; most of the remaining 9,000 either failed to grow after being 
transferred from the greenhouse to the open ground or else they 
failed to produce tubers. All of those which developed tubers 
were saved, described, and grown in 1911. At harvest time all were 
again saved for further study and description. This entailed a large 
amount of work and the recording of many data. Some idea of the 
way in which these studies were performed may be secured from 
Plate VII, figure 2. The object in taking so many data and in grow- 
ing a large number of seedlings which would ordinarily be discarded 
was to note whether any change occurred in the seedling in the second 
and subsequent generations. In other words, it was thought desirable 
to determine whether one might safely discard all unpromising 
looking seedlings the first season. The results secured indicate that 
there is little likelihood that a first-year seedling producing pronged, 
irregular-shaped tubers similar to those in Plate VIII, figures 1 and 2, 
will ever develop into a smooth-tubered variety. It is also equally 
apparent that a deep red or blue skinned seedling is never likely to 
become a desirable commercial variety. On the other hand, the 
hybridist is not always justified in discarding a seedling which has 
produced only two or three small tubers, weighing in the aggregate 
possibly not over 1 ounce, provided the tubers are smooth, shapely, 
and white skinned. (PL VIII, tig. 3) . It is an easy matter, however, 
to decide on the advisability of retaining such seedlings as those shown 
in Plate IX. Occasionally a seedling is found that is unusually 
prolific in both tubers and seed balls; such a potato is shown in 
Plate X. 
The data secured from the 1910 and 1911 seedlings have served to 
make it possible to discard first-year seedlings rather freely, with a 
fair degree of reliance, it is believed, both with respect to those dis- 
carded and those retained. 
These seedlings were tested in 1913 and 1914 at three rather widely 
separated points — Houlton, Me., 1913; Caribou, Me., 1914; Honeoye 
Falls, N. Y., and Jerome, Idaho. Of the 35,000 seedling plants grown 
in 1910 and 1911, there now remain less than 150 numbers at the first 
two points and about 120 at the last point. Since 1911 the practice 
