May 16, 1921 
Heritable Variations in Cotton 
233 
note this evidence of the progress attained. On the other hand, the 
selfed line, the immediate ancestors of which had not been selected with 
reference to this character, had a much lower mean and a higher, though 
perhaps not significantly higher, coefficient of variation for lint index 
than the bulk population of the 5-3 strain from which it was derived. 
In this character, also, five generations of controlled self-fertilization 
apparently had resulted in no further progress toward homozygosity. 
Fuzziness of seeds. —In the earlier years of breeding work with the 
Pima variety no especial attention was given to this character. Conse¬ 
quently it is not surprising that the “bulk” stock, as compared with 
the parent progeny, showed no significant difference in the mean amount 
of fuzz on the seed. On the other hand, the variability had been sig¬ 
nificantly reduced, the standard deviation of the bulk stock being lower 
than that of the parent progeny by an amount equal to 7 times the 
probable error of the difference. The selfed line had a mean fuzziness 
which was significantly greater than that of both the “bulk” stock and 
the parent progeny, the difference in both cases being about five times 
its probable error. The selfed line was significantly more variable in 
seed fuzziness than the bulk stock but was significantly less variable 
than the parent progeny. It is evident that the selfed line had been 
unconsciously selected towards the upper limit of fuzziness represented 
in the immediate progeny of the parent plant of the Pima variety. 
Fuzziness is a marked disadvantage from a practical point of view, the 
separation of the fiber by the type of gin used for this very long staple 
cotton being much more easily effected with smooth than with fuzzy 
seeds. The frequency distribution of the parent progeny indicates that 
selection of a strain having smoother seeds than the present commercial 
stock of Pima is well within the limits of possibility. 1 
EVIDENCE OF THE OCCURRENCE OF HERITABLE VARIATIONS OB¬ 
TAINED IN ROGUING 
The Yuma variety of American Egyptian cotton was grown by Arizona 
farmers previous to the introduction of the Pima variety. In 1914, 
1915, and 1916 fields of this cotton were rogued by agents of the United 
States Department of Agriculture, in cooperation with the Salt River 
Valley Egyptian Cotton Growers’ Association, in order to supply seed to 
be increased for planting the commercial acreage. The total number of 
plants in the fields rogued during the three years was rather closely esti¬ 
mated as 2,290,000, of which (by actual count) 23,537, or 1.03 per cent, 
were removed as being off-type or otherwise undesirable. The work 
involved a rapid examination of every plant in the area rogued and 
thus afforded an excellent opportunity for judging the state of uniformity 
of the variety. 
1 This is true even if we leave out of consideration the single plant (No. 28) which falls into grade 1, this 
having been, in all probability, a first-generation Pima X Yuma hybrid. 
