PAPERS ON BIOLOGY AND AGRICULTURE 115 



data given consists of averages, taken in most cases from 

 eight seedlings. In a few cases, where germination was 

 poor, or seedlings were discarded because of abnormality 

 or accident, the averages include a smaller number. Meas- 

 urements of seedlings were taken to tenths of centimeters, 

 but the calculation of averages was made to the third deci- 

 mal place and are recorded to the second. 



III. DISCUSSION 



Daily observation of the seedlings made evident a strik- 

 ing correlation between the amount and rate of growth and 

 the specific gravity and size of the seed. So marked and 

 regular is the correlation that it was possible, as a rule, to 

 select the seedlings from seeds of certain densities and sizes 

 by their general appearance. This was especially true for 

 the seedlings from seeds of Densities 2 and 3, for these ap- 

 peared more uniform in size and consequently in the rate 

 of growth. They were also more sturdy and of a deeper 

 yellow color than those from seeds of the lower densities. 

 Now and then a group from Density 4 would be mistaken for 

 those of the higher densities. This is apparently in agree- 

 ment with the results that Degrully (3) obtained in his 

 work with wheat. He found that the plants from the den- 

 ser grains were greener, more vigorous, and, during their 

 early growth, showed a great superiority over those from 

 the less dense grains. Because of heavy rains, his plants of 

 both groups suffered greatly from rust and he was unable 

 to make comparisons of the final growth. A study of the 

 data as recorded in the tables shows that the differences 

 noted in the seedlings are not differences of appearance 

 only. 



Because seedlings from seeds of all sizes and densities 

 were grown at 25=^ C the *Tables 1 to 17 and 21 to 36 have 

 the data given for individual seedlings. A comparison of 

 individuals is not undertaken because that would lead to 

 a study of individual variation. In order to show the su- 

 periority of some groups over others, the groups will be 

 compared in respect to their average, maximum and mini- 

 mum values. Because of differences between the seedlings 

 grown in water and those grown in soil each culture will be 

 studied separately. 



