448 



SCIENCE. 



[N. S. Vol. XVII. No. 429. 



studied in the pea, as well as for the seed- 

 shape. He did not conclude, however, that 

 it holds good for all plants, but left the 

 subject for further investigation. He him- 

 self found different results in Hieracium. 

 It will be seen at once that it will be a 

 very difficult matter to follow this law when 

 many characters are to be contrasted, par- 

 ticularly when the characters are merely 

 qualitative and grade into each other. 

 The dominant characters pertain to either 

 parent: some of them may come from the 

 mother and some from the father. 



When this roundness is dominant from 

 the male parent, it falls under the denom- 

 ination of what we commonly know as 

 xenia, or the immediate effect of pollen; 

 when it is from the female parent, there 

 is no xenia. In the case of the pea, the 

 seed-content is embryo and we are not sur- 

 prised if there is xenia. In those plants 

 in which the embryo is embedded in endo- 

 sperm, however, it would seem to be diffi- 

 cult to account for xenial dominance, unless 

 there is double fecundation, as appears to 

 be the case in Indian corn, as pointed out 

 by De Vries, Webber* and others. It 

 looks as if the question of dominance would 

 introduce a new point of view into the 

 study of xenia. There is now a strong 

 tendency, however, to use the word xenia 

 to designate only those effects occurring 

 outside the embryo. 



Which characters will be dominant in 

 any species we cannot determine until we 

 perform the experiment; that is, there is 

 no mark or attribute which distinguishes 

 to us a~ priori a dominant or a recessive 

 character. However, the mere fact as to 

 whether the one or the other character is 

 dominant is relatively unimportant, for 

 constant dominance is no more a regular 

 behavior than recessiveness is. In various 

 subsequent experiments it has been found 



* ' Bull. 22, Div. of Veg. Pliys. and Path.,' 

 IT. S. Dept. Agric, 1900. 



that even when marked dominance is not 

 shown in the first product, the hybridiza- 

 tion may follow the law in essential numer- 

 ical results. The really important points 

 are two: (1) that the characters typically 

 remain pure or do not bend, (2) and that 

 their reappearance follows a numerical 

 order. 



After finding such surprising results as 

 these, Mendel naturally endeavored to dis- 

 cover the reasons why. The product of 

 his speculations is the theory of gametic 

 purity (to use our present-day terminol- 

 ogy), which is a partial theory of heredity. 

 Every plant is the pi-.oduct of the germ cell 

 fertilized by the sperm cell. When con- 

 stant progeny is produced, it must be be- 

 cause the two cells, or gametes, are of like 

 character. When inconstant progeny is 

 produced, it must be because the sperm cell 

 is of one character and the germ cell of 

 another. When these unlike gametes come 

 together, they will unite according to the 

 law of mathematical probabilities, one 

 fourth of those of each kind coming to- 

 gether and one half of those of both kinds 

 coming together. If A and B represent 

 the contrasting parental characteristics, 

 they would combine as 



A + A=A' 

 A + B=AB 

 B + A=BA 

 B + B = B- 



A^ and B' are equivalent only to A and 

 B. Since both of the opposed or con- 

 trasted characters can not be visible at the 

 same time, we have the following: 



A 

 A' 

 A» 

 B 



in which small b represents the character 

 that for the time being is not able to ex- 

 press itself, or is recessive, and large B 

 represents the same character fully ex- 

 pressed. 



