244 



Journal of Agricultural Research 



Vol. VI, No. 6 



Table V. — Correlation between the number of sterile spikelets and the total number of 

 spikelets in beardless and bearded varieties of wheat 



BEARDLESS VARIETIES » 



BEARDED VARIETIES 2 



lr=o.S43±o.os4- 



' r=o.598±o.04i. 



CORRELATION BETWEEN THE) PERCENTAGE OF STERILE SPIKELETS AND 

 OTHER CHARACTERS OF THE WHEAT PLANT 



For the purpose of studying the relationship between the percentage 

 of sterile spikelets per plant and other characters, 300 plants of the 

 variety Velvet Chaff were pulled, dried, and later carefully measured. 

 The plants had been grown by the centgener method, 6 inches apart 

 each way. The percentage of sterile spikelets was used rather than the 

 actual number, for the reason that the length of spikes, which deter- 

 mines the number of spikelets, varies so greatly. The measurements of 

 length were taken in -centimeters and those of weight in milligrams. 

 Biometrical data were secured for the statistical relationship between 

 the percentage of sterile spikelets per plant and (i) the number of culms 

 per plant; (2) the yield of grain per plant; (3) the yield of grain per 

 spike; (4) the length of the culm; (5) the length of the spike; (6) the 

 average weight of the kernel; and (7) the number of spikelets. In the 

 above determinations the plant was used as a unit, the value for each 

 character being determined by taking the average of the respective 

 readings. 



CORRELATION BETWEEN THE PERCENTAGE OF STERILE SPIKELETS PER 

 PLANT AND THE NUMBER OP CULMS PER PLANT 



An inspection of Table VI shows only a slight degree of correlation 

 between the percentage of sterile spikelets and the number of culms, 

 which is negative. The coefficient of correlation is — 0.076 ± 0.039. Evi- 

 dently there exists no appreciable relationship between the percentage of 



