COSSING 95 



curred. In autogamous plants it is hard enough, but in animals 

 and allogamous plants the proof of a real mutation can be given 

 only under very favourable circumstances, which make all the 

 necessary test-matings possible. It is quite clear that Morgan 

 actually distinguishes between Mutation in the sense of de 

 Vries and in the sense of Burbank emd Davenport. But it is 

 evident that some of the other workers with Drosophila are not 

 aware of the distinction. From the Weismannian standpoint, 

 and looking upon genes as determinants for characters, and 

 upon "unit-characters" as determined by one or a few genes, it 

 is evidfent that the spontaneous production of an individual 

 which is shown to have a clear-cut dominant and new unit-char- 

 acter proves a dominant mutation in de Vries' sense. And 

 such a mutation would imply either the creation of a new gene 

 or the change of a "locus" into something different. We have 

 seen, that Davenport does not hesitate to assume this sort of a 

 progressive mutation to account for dominant "mutations" in 

 poultry. 



Crossing in the widest sense, mating between individuals not 

 identical genotypically, will produce those new dominant char- 

 acters, these new mutations in the sense of Davenport. And if 

 we find out the facts in cases of the production of a new domin- 

 ant "unit-character" in Drosophila, we will always find that 

 such dominant "mutants" happened to originate in cultures 

 which were not inbred for a long series of generations. The fact 

 that the dominance and distinctness and novelty of a "mu- 

 tant" character are looked upon as sufficient proofs for a muta- 

 tion, is evident from the very tests to which animals with the 

 new character are submitted. As an example we want to quote 

 a few sentences from the first lines of a recent paper by Otto 

 Mohr. 



"A female, just hatched, having wings serrated or not- 

 "ched at the end, was found in the purple stock-bottle, 

 "October 7 1918. In order to test whether this character 

 "was due to a new mutation, the female was mated to 

 "males from an unrelated stock. The majority of her F^ 



