H 



SCIENCE 



of emboitement might have proved quite ac- 

 ceptable to Leibnitz, Haller, or Bonnet, but, 

 unless I misunderstand him, Professor Bate- 

 son has presented this view more as an illus- 

 tration of the bankruptcy of present evolu- 

 tionary theory than as a matter of serious 

 conviction of his own. I will not discuss this 

 interesting speculation further than to observe 

 that any interpretation of variation which 

 logically leads to such a standpoint naturally 

 incurs a very justifiable suspicion of unsound- 

 ness. It may be that in the case of any par- 

 ticular variation we are unable to positively 

 declare that it is not due to loss, but on the 

 other hand we are unable to positively declare 

 that most variations are due to loss. I think 

 I am not going too far in stating that a ger- 

 minal variation due to loss has not been 

 proved to occur in any single case. If it is 

 legitimate to explain the appearance of new 

 characters as due to the removal of inhibitors, 

 we may also explain the apparent loss of a 

 character as due to the advent of inhibitors. 

 It is surely justifiable to assume that inhibi- 

 tors can come into an organism somehow if we 

 t are permitted to make such frequent use of 

 their disappearance in accounting for the' 

 origin of new variations. The .plain fact is 

 that we know practically nothing of the 

 i changes in the germ plasm which we postu- 

 \ late as the causes of variability. It is easy to 

 assume the existence of an inhibitor to bring 

 any particular variation into line with one's 

 general theory, but such explanations are 

 purely formal and therefore of little scientific 

 value. 



While few would be inclined to follow Bate- 

 son in his rather paradoxical interpretation of 

 dominance, the doctrine that recessiveness is 

 due to loss is coming to be quite prevalent 

 among workers in genetics. One of the chief 

 reasons for regarding so much of the varia- 

 tion that has arisen among domestic animals 

 as caused by the loss of factors is the fact that 

 the crossing of different varieties often pro- 

 duces a reversion toward the ancestral type. 

 If we regard the ancestor of our races of do- 

 mestic mice for instance as possessing a full 

 complement of factors, and assume that the 

 different varieties have arisen by the dropping 



out of one or more factors in this variety, and 

 one or more other factors in that variety, then 

 when these varieties are crossed the hybrid 

 may possess all the factors of the original an- 

 cestor and hence show a reversion to type. 

 On the basis of this assumption one can make 

 out gametic formulas for the different vari- 

 eties of a species, test them by breeding ex- 

 periments, and thus verify their correctness. 

 Gametic fomulas obtained in this way doubt- 

 less symbolize a truth in regard to the germi- 

 nal constitution of the organisms in question. 

 The value of such formulas is no longer a 

 matter of doubt, and is quite independent of 

 the various interpretations that can be made 

 concerning the nature of the symbolism, just 

 as chemical formula? are of value quite irre- 

 spective of the various theories of the consti- 

 tution of atoms. 



Consider the origin of a black mouse ac- 

 cording to the presence-absence hypothesis. 

 We may explain the origin of a black mouse 

 by saying that it is caused by the absence of 

 the agouti or ticking factor that breaks up 

 the color of the hair into bars. Gray is there- 

 fore black plus an agouti factor. But does it 

 ' J 'follow that because we can interpret the facts 

 in this way, and interpret them consistently so 

 far as breeding experiments are concerned, 

 the change that has taken place in the germ 

 plasm that produced** a black mouse was 

 really a loss? Such a change is frequently 

 assumed to be the result of an actual loss of 

 a little discrete unit of some sort in the germ 

 plasm. De Vries has interpreted recessiveness 

 as due to the latency or loss of potency of pan- 

 gens, but we may also assume that the germi- 

 nal basis of the character in question has 

 undergone a change of such a character that 

 without becoming inactive it ceases to func- 

 tion in its usual way. The agouti factor 

 (commonly designated by G) may be regarded 

 as dependent on a part of the germ plasm, a 

 section of a chromosome possibly, which when 

 present causes the barring of color in the hair. 

 When a black mouse arises we may suppose 

 that something takes place in G. It is not 

 necessarily a change in the direction of either 

 chemical or organic simplicity any more than 

 it is necessarily a loss of substance. The fact 



