GRAIN-SORGHUM EXPERIMENTS IN THE PANHANDLE OF TEXAS. 59 
Six heads were saved from this head row of selection No. 30. The 
seed from these heads was sown in 1908 in six head rows, numbered 
30-1 to 30-6, inclusive. Later in the year Cereal Investigations No. 
340 was given to the selection No. 30. Two of these six rows, Nos. 
and 2, proved uniformly dwarf and early, and heads selected in row 
, have ieee the progenitors of the Dawn (dwarf) kafir hereafter con- 
sidered. Rows 3 to 6 were all early, but heterozygous for height. 
Row 4 contained both tall and dwarf plants, but was most uniform 
in other respects. Heads were selected from the tall plants. Rows 
3, 5, and 6 were discarded. 
In 1909 five head rows were grown (C. I. No. 340-4-1 to 5). These 
all bred fairly true to the tall early type, though containing some . 
dwarf stalks. Heads were selected from both tall stalks and dwarf 
stalks. Their records in the plats begin with this year. 
In 1910, seed from heads taken from tall plants was sown in 10 
head rows (C. I. No. 340-4Bt-1 to 10). The resulting plants were 
nearly all tall, though a few dwarfs appeared again. The average 
height was 5.3 feet. The tall and early selection being now fairly 
well fixed, it was given the distinct number C. I. No. 472, and later 
named Sunrise on account of its earliness. Ten head rows (C. I. No. 
340-4Bd-1 to 10) also were sown with seéd from the dwarf plants in 
No. 340-4 of 1909. The plants in these rows were all homozygous for 
dwarf stature and differed not at all from those oe from the seed 
of No. 340-1 or Dawn (dwarf) kafir. 
Like all the other kafirs, this selection did not produce heads in 
1913. In the other seven years the average duration of the vegetative 
period has been 88 days, of the fruiting period 36 days, and of the whole 
growing period 124days. Blackhull kafir did not mature in 1916, and 
the record of Sunrise is not complete in 1909, so that comparison can 
be made only for the five years, 1910 to 1912, 1914, and 1915. In these 
five years the average duration of the three periods in Sunrise kafir 
was 85.6 days, 38.6 days, and 124.2 days, respectively. In Black- 
hull kafir the average duration of these periods in all selections was 
88.8 days, 42.2 days, and 131.0 days, respectively. This shows an 
average shortening of the growing period by 7.5 days in favor of the 
Sunrise kafir, which is an item of much importance in districts where 
the season is fairly short and summer cmon is of Hieergpetenie occur- 
rence. 
The complete experimental data are found in Table XXI, a study 
of which shows that fairly good germination and good stands were 
obtained in all eight years. The closest spacing was one plant to 
each 4.7 inches, the widest was one plant to each 17 inches, and the 
average was one plant to each 9.2 inches in the 8-year mariog. In 
the very dry spring of 1910, when poor germination was recorded 
from nearly all varieties, the plant space in Sunrise kafir was 17 inches. 
