818 
Journal of Agriculutral Research 
Vol. XXVIII, No. 8 
ABNORMALITIES OF THE FLOWERS 
Most of the flowers that were produced on the late top growth of Hartsville 
cotton near Casa Grande were very small and the anthers did not open. Many 
of the anthers were shrunken, as though the pollen had not developed. A recent 
period of cold rainy weather might explain the failure of the anthers to open, 
but the failure of the pollen-grains to develop may be associated with crazy-top. 
A case of complete sterility was observed in 1922 in the Coachella Valley, 
between Indio and Palm Springs, in a field of an okra-leaf strain of Acala cotton 
isolated by several miles from any other cotton, which may represent an analogous 
case of pollen-suppression by a growth disorder. A careful examination of many 
flowers in different parts of the field showed that no normal stamens were being 
developed, and that all of the anthers failed to open, although there were many 
flowers. 
Earlier in the season a few bolls had been set on some of the plants, but widely 
scattered through the field. This might indicate that the disease began at the 
early flowering stage, or that some of the plants were not stricken as soon as 
others, as might be the case if a growth disorder were present. In view of the 
isolation of the field from other cotton any infectious disease must be supposed 
to have come from the native vegetation. A partial deficiency of pollen has 
been observed also in a planting of Acala cotton in another mountain valley near 
Palm Springs, confined to individual plants, which might result from a scattering 
infection with a growth disorder. 
The reduction in the size of the flowers at Casa Grande was not regular, nor 
in definite proportion to other symptoms of the disease. In some cases rather 
large flowers were produced on badly distorted and reduced growth. Even on 
normal plants the size of the flowers may be somewhat reduced toward the end of 
the season. 
DWARFING OF BOLLS 
Though plants that are badly affected with crazy-top became completely 
sterile, under the stress conditions of the summer months, there is a partial 
recovery or mitigation of the symptoms under more moderate late-season con¬ 
ditions. In the crazy-top areas of the Hartsville field at Casa Grande only a 
small proportion of the plants recovered to the extent of retaining and developing 
bolls, and most of the bolls in such cases were very small and misshapen (PI. 15) . 
In Pima cotton the late season return to a fruiting condition is more general 
and the bolls are more normal in shape and size, though fewer seeds are devel¬ 
oped than in the late-season bolls of adjacent plants that are not affected with 
crazy-top. Data recorded by Robert D. Martin on numbers of seeds and 
abortive ovules of bolls from 25 crazy-top plants and 25 normal plants of Pima 
cotton, grown at the Sacaton, Ariz., seed-farm in 1923 are given in Table I. 
Smaller numbers of good seeds were found in the bolls of the late-season 
crazy-top growth than in the earlier or mid-season bolls of the same plants, and 
both classes of bolls of the crazy-top plants had notably fewer seeds than 
the corresponding classes of bolls from adjacent plants not affected by crazy-top. 
The reduced number of ovules in the crazy-top bolls may not be significant, 
except as an indication that the ovules aborted at an earlier stage so that the 
rudiments did not persist in visible form in the mature bolls. Some of the 
“middle crop” bolls, though borne on apparently normal fruiting branches be¬ 
low the crazy-top growth, probably were produced during the crazy-top period, 
and thus may have shared any adverse influence of the disease. 
