298 Sluill. 



both possess in coiuiiioii is ncarlN ail ol' tliat wliicli tbey each i)Ossess, 

 and that the additional clcnieiit or cliMiionts wliich are required for 

 the actual realization of the one w the other sex, are coiiceivalil.v of 

 relativeh' minor value. 



Suiiiinar.v. 



The discovery by lUUK of a iiairow-leafed uiiitanl in l.ijcluiis 

 dioica has led to the denionstratinn tiiat the relatively bioad leaves of 

 • the normal or "typica" foiiii is a sex-limited character. 



This narrow- leafed form is here described and illustrated under 

 the name Lyclinia diuicit angusfifofia. It diffeis from the noi-mal type 

 not only in its narrow leaves, but also in having- narrower petals and 

 long'ei' rudimentary ovaries. 



In crosses with the normal type (oujusti folia is a recessive which 

 reappears in the F- only in half of the males. A complete analysis of 

 the Fi family showed that all of the bi-oad-leafcd males were lietere- 

 zyj^ous for the broad-leaf factor 1!, ami that half of the females were 

 honiozygfous, and half heterozyg-ous. for the same factor. This confirms 

 the conclusion liaiued from the lu'eedinf^- of hermaphrodites, that the female 

 of Li/rhnis (linica is a sex-lioiiio/,ygote and the male a sex-heterozygote. 



There was a remai'kable absence of females in families produced 

 by matinjis between the F^ narrow-leafed males and nnmerous lemales: 

 the total prog-enies from such nuitings included 2741 uuiles and only 

 14 females, and 8 of the latter occurred in one small family of (i(i indi- 

 viduals. The broad-leafed males in the same Fj family pindin'ed males 

 and b-males in ai)i)roximately e(|ual numbers. 



The F-j families resulting!: from crosses between heterozyfi'ous females 

 and narrow-leafed nuiles were exi)e(;ted to produce about 25 per cent 

 narrow-leafed b'liiales. Only 2 b'liiales ajiiieared in such families amouii' 

 a total prof^-eiiy of loltT and these 2 hap|iened to be broad-leafed. 



One narrow-leafed female did appear, but in a family in which it 

 was not expected, haviiit;- been produceil by a cross betwecsn a hetero- 

 zygous female and a broad-leab^l iiiah'. This female was a very small 

 dwarf; its geuotypic relationship to tlie angitatifolia birm is unknown. 



Evidence is presented indicatiui;- that the nmles (d normal strains 

 not directly related to the oi'iginal aiuindifoUa mutant, ai-e heterozygous 

 for the bioad-leaf gene B, but further investigation will be necessary 

 to determiiu' whether this is (jciicndli/ the cas(>. 



The hermaphroditic plants in families derived from my oiiginal 

 herma|)hro(litic mutants are also heterozygous for the gene />'. which 



