1912| HASSELBRING—CUBAN TOBACCO 117 
In harvesting and separating the seeds from the capsules, of 
course every precaution was used to avoid mixing the different 
types. Each bag was taken separately and the seed shelled out in 
a large porcelain evaporating dish in the laboratory. The different 
lots were handled in such a way that there was no possibility of 
a stray seed being blown or scattered among those of another lot. 
Similar precautions were used in sowing the seeds the following 
autumn at the beginning of the next tobacco season. The seeds 
were sown in flats, in soil taken from a nursery where no tobacco 
had been grown, and distant from any tobacco field. The soil was 
sterilized with hot water. After the seeds were sown, the flats 
were covered with burlap frames, and were protected from ants, which 
carry off the seeds, by a ridge of powdered naphthalene around 
the edge of each flat. The posts supporting the benches on which 
the flats stood were kept wrapped in cloth soaked in crude oil. 
The benches had been previously freed from ants by boiling water. 
With these precautions, no trouble was experienced from the insects. 
The flats were kept in an open shed. 
The seeds were sown September 16 and 18, 1908. When the 
seedlings were large enough, they were pricked out in open frames 
kept covered for a time with canvas. The fosturas were planted 
in the field at various times from November 12 to December 9g, 
400-500 plants of each type being set out. 
The results of the cultures were so striking and uniform that 
they can be stated in a few words. Even in the open frames the 
various groups of plants showed differences which made them 
stand out from each other, but the differences were more evident 
when the plants were mature. The descendants of each plant 
were entirely uniform and like the parent plant from which they 
were derived. Even minute and unimportant characters were 
transmitted with surprising definiteness. 
While the different types were indiscriminately intermingled 
in the field, the contrast even among extreme types was obscured 
on account of the many apparently intergrading variations, but in 
the cultures where large numbers of each type were grouped 
together the differences were unmistakable. Thus, for instance, 
the different groups as a whole showed marked differences in height, 
