620 
GENETICS: H. J. MULLER 
the linkage study, and by dissecting, by means of crossing over, each of 
these two kinds of chromosomes separately, that a complete explanation 
of the beaded case was finally obtained. It would be impossible here to 
^ve even an outline of the detailed steps of analysis, but the following 
are the concrete results and theoretical conclusions which have been 
arrived at. 
1. The difficulty which was experienced in getting pure stock was 
due to the fact that the chief factor for beaded — Bd' — is lethal, killing all 
flies homozygous for it. Heterozygous beadeds are not killed but 
usually show the beaded character. Thus Bd^ like yellow in mice, is 
dominant for its visible effect and recessive for a lethal effect. The 
dominance of Bd' is variable, however. 
2. The reason that a race which bred true to beaded was finally se- 
cured is not because a condition of homozygosis was at last estab- 
lished, but, on the contrary, because of the establishment of a state of 
enforced heterozygosis, wherein not only the homozygous beadeds but 
also the homozygous normal winged flies were prevented from hatching. 
This was due to the appearance, by mutation, of another lethal factor, 
liii I J ill that third chromosome of the impure stock which contained the 
normal allelomorph of Bd'. This lethal, like Bd', kills all flies which 
are homozygous for it, although, unlike Bd^ it has no visible effect when 
it is heterozygous. Since, in the flies of this heterozygous race, it arose 
in the chromosome containing the factor for normal wings, no homozy- 
gous normals will be able to appear except those in which Im i has been 
separated by crossing over from the normal allelomorph of Bd'. The 
usual amount of crossing over between these loci of lim and of Bd' is 
10%, as Ijii 1 was found to be 10 units to the left of Bd'. On this basis 
it is to be expected that the selected beaded stock would throw 90% 
beaded flies and 10% normal winged crossovers. 
3. In this selected stock, however, not even 10% of the flies have 
normal wings, because of the existence of another mutant factor, C, 
which almost entirely prevents crossing over in the region of the chromo- 
some in which Inn and Bd' lie. Imi. on this account, always remains 
with the normal allelomorph of Bd' and all the homozygous normal 
winged, as well as the homozygous beaded flies are thus prevented from 
appearing. It should be noted that the factor C, too, always remains 
heterozygous, for it is contained only in the chromosome having Im i and 
the normal allelomorph of Bd', not in the chromosome with Bd' itself. 
This state of heterozygosis for C is also a necessary condition for the 
mechanism whereby only beaded flies are produced, since C produces its 
inhibition of crossing over only when it itself is heterozygous. Aside 
