58 BULLETIN - 1134, U. S. DEPARTMENT OP AGRICULTURE. 



It was observed that in three of the five flowers of the inbred 

 population a few of the pollen grains were very small and did not 

 eject their contents during the period of observation, which was 

 about two and one-half minutes. Only one of the five flowers of 

 the open-pollinated stock had an appreciable number of abnormally 

 small grains, although in this case the proportion of such grains was 

 fully as great as in any of the flowers of the inbred family. Further 

 observations were therefore made upon the occurrence of sterile pol- 

 len grains in the inbred and in the continuously open-pollinated 

 population. Examination of a number of flowers of the continuously 

 open-pollinated stock showed that in most cases from 2 to 5 per cent 

 of the pollen grains were very small and presumably sterile, 20 per 

 cent of the pollen in one flower having been of this character. 

 Counts made on pollen from five plants of each population, with 

 the results given in Table 36, afforded no evidence of an increase in 

 the proportion of sterile pollen grains having resulted from con- 

 tinuous self-fertilization. 



Table 36. — Floivers of a family of Pima cotton inbred during five generations 

 compared with those of a continuously open-pollinated stock of this variety, 

 showing the percentage of abnormally small and presumably sterile pollen 

 grains. 



Inbred family. Open-pollinated stock. 



Plant o n H virrcror N umber of pollen _ Number of pollen ^ 



Plant and i lower. grains.* Percent- grains.* Percent- 



age of age of 



, 1 sterile sterile 



Total. Sterile. grains - Total. Sterile. ' ^^ 



No. 1 ■ I 26 1 3.8 



No. 2 56 1 1.8 ? 



No. 3 ? 30 2 6.7 



No. 4 14 2: 14.3 36 1 2.8 



No.5 ? 25 3 12.0 



1 The total number of grains in the field of the microscope was not determined if no abnormally small 

 grains vrere present. 



It would be expected that if continued inbreeding had impaired 

 the viability of the pollen, fertilization in the inbred strain would be 

 more nearly complete when the flowers are cross-pollinated than when 

 they are self-pollinated. The data given in Table 20 show, however, 

 thai in a family which had been inbred during seven successive gen- 

 erations there was not a significant difference in the mean number of 

 seeds per boll from the cross-pollinated and from the self-pollinated 

 flowers and that a somewhat higher percentage of bolls matured 

 from the self -pollinated than from the cross-pollinated flowers. 



NUMBER OF OVULES IN AN INBRED POPULATION. 



Counts were made in 1919 of the ovules in 20 ovaries of a family 

 which had been strictly inbred by controlled self-fertilization during 

 five generations and in 30 ovaries of a continuously open-pollinated 

 stock, each ovary having been taken from a different plant. The 

 counts were made upon 3-celled ovaries and yielded the mean and 

 extreme numbers given in Table 37. It is obvious that controlled 



