Growth aud variation in maize. jg3 



Summiug up the evidence from these mean (luintile standard 

 deviations it may be said that the standard deviations of all the classes 

 are much smaller than would be expected if the observations were 

 distributed according to pure chance. Further the standard deviations 

 of the extremes, i. e. the very large and the very small plants, are 

 relatively much smaller than the deviation of the middle classes. The 

 standard deviation of the middle class is relatively much greater, i. e., 

 it approaches nearer to the theoretical standard deviation than any of 

 the classes on either side of it. 



It has already been ])ointed out that these results are what would 

 be expected if the growth was determined by Mendelian factors. In 

 the following discussion the facts brought out in this paper will be 

 summarized and an attempt made to show their relation to other facts 

 of growth and genetics. 



Discussion and Results. 



The question which next concerns us is whether it is possible to 

 present a plausible theory to account for these observed facts. Let us 

 keep clearly in mind the main fact that is to be accounted for, namely, 

 that in the frequency distribution of the average relative size for the 

 whole season, the number of plants is practically equal in each class. 

 Further there is a strong tendency for individual plants, especially 

 those in the extreme classes, to maintain the same relative size through- 

 out the season. Two general lines of explanation suggest themselves. 

 First, the differences in individual plants may be due to the action of 

 environmental influences aud, second, they may be due to the action 

 of internal factors. These two possible explanations have already been 

 suggested but they will be discussed more in detail here. 



At first sight it appears quite probable that these differences in 

 individual plants are due to environmental influences. Thus plants 

 which are in better soil or which are less crowded will grow better 

 than plants under less favorable .conditions. These conditions would 

 on the whole remain nearly constant throughout the season and hence 

 might produce the effects noted. It has already been pointed out that 

 the field in which these plants grew was extremely uniform. So far 

 as general observation went there were no places where the soil ap- 

 peared to be better or worse than at any other place. The manure 

 was evenly spread over the gi'ound and then plowed under. The fer- 

 tilizer was distributed by a machine which secured a uniform distribu- 



11* 



