400 



THE AMERICAN NATURALIST [Vol. LV 



below the normal. The reason for this may lie in the fact 

 that the flower is an independent axis rather than a lat- 

 eral organ like the leaf and may therefore be less affected 

 by the size of the main axis from which it springs. It is 

 well knowTi that flowers are more constant in size, and 

 other characters and less affected by environmental con- 

 ditions than are vegetative organs. 



It should be recognized that in the grasses, where most 

 of the work along this line has been done, conditions 

 are somewhat different from those in dicotyledons such 

 as the bean ; since the plant body, or at least each shoot 

 or culm, is essentially determinate in growth, with a lim- 

 ited number of parts. This may sometimes affect the 

 statistical results obtained, but we believe that conditions 

 are fundamentally the same in the two groups. It is of 

 interest to note the conclusions of Leighty and others, for 

 Fmall grains, viz., that the correlations between organ 

 size and ])lant size are considerably higher in small, 

 ])oorly (loveloi)ed [)lants than in large ones,— a situation 

 ])recis('ly similar to that which we have reported in beans. 

 Tlic whole problem can pci-lmps lu» soIvcmI best, liow- 

 rvi-r. l.> ^iii(!yiiiu- ^]ieli an in.lel enninate lypi- of plant 

 h.xlx 1^ .-liaracl.TiMir <.r llic ..r.liiia r> .li.-nl \ lodoii. 



CONCT.USIONS 



We may conclude, therefore, that the size of the plant 

 l)ody is not tlie direct causative factor in determining the 

 size of tlie knaves, fruits or seeds which it i^roduces, as 

 lia- been suggested or implied by many investigators, but 

 tliat the size of any given organ depends rather upon 

 the size of the growing i)oint out of which it has been 

 de\'el()i)ed. Any factor, be it age, moisture or food sup- 

 ply, which alters the size of the nieristem, will thus alter 

 tlie -ize of the oi'gans pi'odnced by tliis nieristem. There 

 seems to he iiotliiim- in fhe-e liiu-]i<'i- j)lants closely cor- 

 re^pMiMlin- tn liiM .h-liniP- ..r-ani/aliou with which we 

 ai'e familiar in the animal in(li\ idnal, where size of body 



