1913] Goodspeed: Nicotiana Hyhrids 185 



interpretation of which according to the ]\Ienclelian notation is 

 only possible after a certain number of assumptions have been 

 indulged in. My point is simply that if in one case it is neces- 

 sary to assume an interaction of units within a certain "unit- 

 character" then within all "unit-characters" we must recognize 

 similar fundamental units whether the mode of inheritance of 

 the character in question calls for such an assumption or not. On 

 this assumption that all characters are molecular in structure, 

 representing aggregates of units each of which is capable of modi- 

 iymg the character being dealt with, w^e have a basis for the 

 explanation of all such considerations as those presented by 

 Castle (1912). On this basis there is nothing more inconstant 

 than the individual characters peculiar to an organism except the 

 organism itself which represents the sum of these individual char- 

 acters. On this basis again the range of results capable of inter- 

 pretation according to the Mendelian notation is limitless and 

 there can be no question of stretching the Mendelian notation 

 to cover any particular case for, by reason of its assumptions, it 

 has become immeasureably extensible. 



The question then arises to what extent are these primary 

 assumptions valid or in other words does the end justify the 

 means. The end is nothing more important — with reference to 

 the limits of our present knowledge of the phenomena involved — 

 than the natural desire and valuable effort to interpret all ex- 

 perimental results in accordance with a notation which has been 

 found to apply to a relatively small group of characters — prob- 

 ably less than three hundred characters peculiar to domesticated 

 animals and plants. The means employed to accomplish this end 

 are beginning to involve the complication of the mathematical 

 expressions used to such an extent that nothing can actually be 

 gained for the practical breeder and the necessity of raising 

 and examining the extremely large F^ and F., generations de- 

 manded will restrict independent investigation immensely. In- 

 deed it seems hardly in accord with the "progressive spirit of 

 the times" and the growing recognition of the immensity of the 

 problem involved, to demand a strict conformity of effort and 

 interest. The value of the ]\Iendeli;in notation as a (jencraliza- 

 iion certainly fails if it is true that not more than a few hundred 



