CLASSIFICATION OF AMERICAN WHEAT VARIETIES. 13 
emergence, heading, ripening, and height of the many varieties. 
During the summer the writers visited the various points and took 
detailed notes on the characters of the varieties. It was here, in the 
field, that the descriptions of the varieties were written and the keys 
designed and perfected to distinguish the different varieties. The 
descriptions were checked and rechecked at the various points and 
the different descriptive classes were established on a basis broad 
enough to include the varieties wherever they were grown. 
NATURE OF THE MATERIAL. 
The early studies showed the necessity of working with pure lines. 
When bulk seed was used it often consisted of mixed varieties and 
a wrong description might easily become applied to a variety. For 
that reason careful notes were made on the material that was sown in 
each nursery. A typewritten outline was prepared each year which 
showed the classified arrangement of the varieties based on the re- 
sults to date and also the row numbers at each station. The same 
variety often was represented by different lots of seed obtained from 
different sources. These were distinguished by different C. I. numbers, 
which are accession numbers of the Office of Cereal Investigations. 
The varieties, however, have always been distinguished by names 
rather than by numbers. For this reason Cereal Investigations 
numbers are not used in this publication. The nursery outlines also 
contained columns showing the source of the seed sown and the orig- 
inal source of the variety. In addition, they showed whether the 
seed sown was bulk grain or a pure line, and if a pure line, whether 
the same pure line was sown at all stations or whether different pure 
lines were used. In this way it was easily possible to compare field 
notes accurately with those of the previous year or to account for 
differences which existed in the same variety at different stations in 
the same year. This latter condition often occurred when bulk grain 
or different pure lines were used. Natural field hybrids thus were 
easily distinguished from mixtures. 
formerly scientific assistant, in charge of the cereal nursery at the Amarillo Cereal Field 
Station, Amarillo, Tex.; Mr. A. D. Ellison, formerly scientific assistant, Mr. H. P. Ames, 
formerly agent, and Mr. J. W. Taylor, scientific assistant, respectively, in charge of the 
cereal investigations at the Arlington Experimental Farm, near Rosslyn, Va.; Dr. H. H. 
Love, professor of plant breeding, and Mr. W. T. Craig, agent, Cornell University, 
Ithaca, N. Y. 
The writers also acknowledge with gratitude the assistance received from the follow- 
ing officers of the State experiment stations not formally cooperating with the Office of 
Cereal Investigations: Prof. G. R. Hyslop, professor of farm crops, and Prof. C. C. Ruth, 
assistant professor of farm crops, at the Oregon Agricultural Experiment Station, Cor- 
vallis, Oreg.; Prof. E. F. Gaines, assistant professor of farm crops at the Washington 
Agricultural Hxperiment Station, Pullman, Wash.; Prof. P. V. Cardon, agronomist at the 
Montana Agricultural Experiment Station, Bozeman, Mont.; Mr. Breeze Boyack, formerly 
assistant agronomist at the Colorado Agricultural Experiment Station, Fort Collins, 
Colo. ; and Prof. 8. C. Salmon, professor of farm crops at the Kansas Agricultural Experi- 
ment Station, Manhattan, Kans. 
