358 APPENDIX B 



when dominant unit met recessive unit, the result was an impure 

 dominant plant that looked like a pure dominant, but produced 

 all three kinds of offspring. His hypothesis may be illustrated 

 by taking black and white marbles, a hundred of each, to repre- 

 sent respectively the dominant and recessive units, and mixing 

 them in a bag. Then if pairs of marbles be withdrawn at 

 haphazard, it is evident that on the average half of them will 

 have partners of their own colour, while the other half will be 

 united to partners of the opposite colour. The twenty-five black 

 pairs will represent the pure dominants and the twenty-five white 

 pairs the pure recessives, while the fifty mixed pairs will represent 

 the impure dominants. If each marble be imagined as a germinal 

 unit capable of multiplication in the embryo, then it is evident 

 that each black pair will for ever give rise to only dominant 

 plants, each white pair only to recessives, while the black-and- 

 white pairs will produce only impure dominants, from which will 

 spring in each generation pure dominants, pure recessives, and 

 impure dominants in the old proportion. The points to be 

 especially noted are (1) that each germ-cell (gamete) contains a 

 single unit for each Mendelian character, (2) that each fertilized 

 ovule (zygote) contains two, one from the ovule and one from the 

 pollen-grain, (3) that these do not blend, but, while one becomes 

 active, the other remains latent, and (4) that both kinds multiply 

 during the formation of the embryo and then separate, so that one 

 unit, and one only, is present in each of the germinal descendants 

 of the fertilised ovule, one-half of the sperm and ovules therefore 

 containing dominant and the other half recessive units. 



In addition Mendel found that several other characters in 

 peas were capable of similar segregation ; for example, colour of 

 seed-skins, colour of cotyledons, shape of seeds, and colour of 

 flowers. In such cases one character was always dominant over 

 its opposite number. Having ascertained the dominant character, 

 he could foretell with fair accuracy the proportion of descendants 

 that would display it. He found, moreover, that, if a given 

 character were dominant, it did not necessarily follow that the 

 other characters associated with it in the original parent were 

 also dominant. In other words, a variety, dominant over 

 another as regards one character, was perhaps recessive as 

 regards another; whence it followed that characters of both 

 varieties might be visible in the first cross-bred generation. 

 Lastly, he found that characters were inherited independently of 

 one another ; thus, when varieties were crossed, descendants might 

 resemble the pollen-parent in one or more particulars, and the 

 pistil-parent in the others ; that is, the original sets of characters 

 tended to break up and form new combinations ; so that, for 

 example, a plant with yellow cotyledons might have a seed- 

 coat or flowers of a colour different to that which had previously 

 been associated with the yellow albumen. 



