New York AGRICULTURAL EXPERIMENT STATION. 283 
or not. A laboratory test indicated that potatoes in growing sprouts 
from one to three inches long lose 2 per ct. of their dry matter. 
Size of seed.— Whether it is best and most profitable to use large 
or small tubers for seed is a question that will probably never be 
settled until we can limit the terms “large” and “small” by other 
terms which shall tell whether the small potato is of “ good”’ parent- 
age or the “large” tuber merely an accident in a “ poor” family. 
The early work of the Station nearly all indicated an advantage in 
the use of large tubers as seed ; but a carefully checked experiment*® 
was made in 1906, in which small potatoes were compared with 
equal weights of seed cut from large tubers, gave a different result. 
In this test the advantage was plainly with the small tubers. The 
influence of cutting enters this test, while in most of the early tests 
the factor of unequal weight of seed pieces makes a complication. 
In tests* in 1884 largest and smallest tubers, selected from produc- 
tive and from unproductive hills of nine varieties of the crop of 
1883, were used for seed. In nearly every case the large tubers 
gave greater yields of merchantable potatoes than small tubers 
from the same kind of seed. In case of seed from unproductive 
hills, the large tubers gave more than 50 per ct. increase in mer- 
chantable potatoes. In case of three varieties the influence of size of 
seed appeared greater than that of productivity of the hills, since 
the large tubers from poor hills of these varieties gave more mer- 
chantable potatoes than the small tubers from productive hills. 
In 1887, the computed acre yields of 116 varieties tested on small 
areas showed an average gain of six bushels to the acre from the 
use of large rather than small tubers. 
In 1890 tubers of different sizes but of uniform weight for each 
lot were compared. The weights of the tubers used were two, 
three and four ounces, respectively. Deducting the amounts of seed 
used, there was still a net gain in favor of the successively larger 
seed pieces. Rot, however, lessened the yields fully one-half in 
this experiment. 
Cutting seed—As a result of the first season’s test, it seemed that 
single eye cuttings of very small size were as good for seed as 
halves or quarters of the potatoes, so in succeeding years compari- 
sons along this line formed a prominent part of the Station’s 
potato program. Not only were single eyes tested against larger 
pieces, but various ways of cutting the tubers to secure these eyes 
® Syllabus, Normal Institute Lecture, 1906. 
“Rpt. 3:301-305 (1884) ; 6:78-86 (1887) ; 9:375-379 (1890). 
