io The Colorado Experiment Station 
erties of the berries. These would appeal to most persons familiar 
with them as quite sufficient to justify the opinion expressed. But, 
whatever the composition and properties of these berries may have 
been, they (Roberts and Freeman) assume that climatic factors 
are, above all others, important in determining these and set them¬ 
selves the task of finding out the critical period in the development 
of the grain, or that period when it is most susceptible to the action 
of those factors determining the yellow-berry. They experimented 
with winter wheat and consequently have two periods of growth, 
the autumn and spring periods. They find that the longer the au¬ 
tumnal period, especially with favorable conditions in the early part 
of the period, the higher the percentage of yellow-berry. In the 
season 1905-06, they found for wheat planted 28 September in 
moist soil 53.80 percent yellow-berry; for wheat planted 16 Octo¬ 
ber in rather dry soil 19.58 percent yellow-berry. They find in 
these results, the two quoted are the extremes of a series, a rela¬ 
tion between the length of the autumn growing period and the 
amount of yellow-berry produced. In 1906-07 they obtain results 
of the same significance; a difference of thirteen days in planting 
is followed by a difference of 12 percent in the yellow-berries. 
This difference was, as in the preceding year, in favor of the late 
planting, this giving the lower percentage of yellow-berry. They 
divided the spring into practically three periods, and studied the 
effect of the mean temperature and rainfall and also considered the 
whole of the spring growing period; so that they treated of the 
whole period from first heading to< maturity, for three weeks be¬ 
fore maturity, and for two weeks before maturity. I can do no 
better than to give their own words to state their results. 
“If we attempt to interpret the effect of these combined influences 
in terms of the rapidity of growth of the plants and of the ripening of the 
grain, and to correlate them with the percentage production of yellow- 
berry, we shall find, as previously stated, that the prevalent idea that 
slow ripening is correlated with high percentages of yellow-berry is ap¬ 
parently justified by the data collected from our wheat cultures for 
1906-07 when the total averages for the two years are compared. When, 
however, the data for the different varieties within a given season are 
considered, this rule does not hold. Now, since it is improbable that in¬ 
fluences that would operate as between different seasons to bring about 
such a result would fail to operate in a given season, there must, there¬ 
fore, be other factors which complicate the results, and which are not 
analyzable without taking into consideration data with reference not 
alone to climatological conditions but also to the hereditary tendencies 
of the varieties concerned.” 
These sentences state as clearly as can be stated the conclusion 
at which Roberts and Freeman arrived, though they subsequently. 
