No. 638] 



271 



individual pairs always bear tlie same relation to ciii-h otlin-. al- 

 lowing the conclusion that whatever influences the size ot the 

 chromosomes generally affects all similarly. The fi-invs df Kii- 



recognizes the factors playing: roles iii the liehavmi' <>\ iiie chro- 

 mosomes, but does not feel that his results eaii he eiitii-elv ex- 

 plained by them. 



To illustrate what is meant by the above criticism let us con- 

 sider a cross made by Kuwada between sugar corn 22, . and Black 

 Mexican 58(,.), another sugar corn. In both, dividing cells from 

 adventitious roots were uniformly selected. The former has 

 chromosome complexes averaging 149.05 mm. in length while the 

 latter gives a total of 172.17 mm. This to Ku%\ada indicates a 

 real and genetic dilference in chromosome lengtii, althoutrh in 

 the same Blaet Mexican plant 58,,., complexes from side root 

 tips average only 145 mm. in length. This would signify that 

 the length 172.11 was no more fundamental in plant 58,,., than 

 was 145, and lessens the ^\elght of the CMd. lu - th it ih> hi_h. r 

 number betokens genetic chromosome dilfereiie,-^ ■ • : - ti 

 149.05 in plant 22,,.,. When the two plair^ - ■ 



chromosome lengths in the hybrids are almost , \ : i 



the sum of the lengths of these structuns m in ^ r.lT 



1 J.u'iOl + 172 17)'^160 61 ltnM\h, pmn , d . ut h- i tnat 



1/2(172.17 + 145) =158.58. The F, plants tnmi the ah,,ve 



that expected on Kuwada's assumption of 149.05 and 17J.17 as 

 the basic or typical lengths of the parental eliroiiiusonie-. 1 he 

 chiomosomes In the F, plants varied from IV, 75 mm 1(>-9 

 mm and a\eraged 161 86 mm This numl.ei ti . i i ^^ Mi 



sideration of 172 17 mm a^ the n pi , nt it u . ' n_rh to p' oit 



