420 GENETICS OF SOMATIC CELLS 



enough to permit a sufficient number of mutations to occur, so as to be detectable in 

 animal groups of suitable size. On the other hand, the number of precursor cells 

 must not be too large, since the progeny of the mutated cells will then become so small 

 that the spot resulting in the adult fur may be too minute to be perceptible. Irradiation 

 on the morning of the tenth day following the observation of the vaginal plug turned 

 out to be practical. This stage was also very sensitive with regard to radiation-induced 

 neonatal death. A dose of 1 00 r was found to be low enough to reduce this complication 

 and still give satisfactory mutation frequencies and was therefore chosen for most of 

 the experiments. The animals were examined in detail for mosaic patches at 32-58 days 

 of age. Nonmosaic animals were discarded at that time, while mosaics were saved and 

 again observed 4-10 weeks later to insure that the spots were not transient. The mosaic 

 area was measured with calipers in fully mature mice, between 8 and 2 1 weeks of age. 

 The total coat area to which the mosaic spots were related was not determined for 

 each individual mouse but was taken to correspond to 45 cm. 2 on an average. 



In offspring of the C57BL x NB cross that had been irradiated with 100 r as 10^-day 

 embryos, the incidence of mosaic animals was about 1 1 per cent, the unirradiated 

 control figure being 0.5 per cent. This figure was corrected for the spontaneous and 

 radiation-induced frequency of animals with coat-color changes that could be attributed 

 to developmental causes or to coat-color dominants, of somatic origin, as indicated by 

 the frequency of spots with altered coat color in the homozygous C57BL x C57BL cross, 

 untreated and irradiated, respectively. Another deduction was made for the spon- 

 taneous frequency of animals with somatic expression of the recessives at the b, c, p, 

 and d loci, as indicated by the frequency of mosaic animals in the unirradiated 

 C57BL x NB F x hybrid group. The corrected figure was still of the order of 10 per 

 cent. Statistical examination of the data indicated that the mosaic animals were 

 distributed randomly among the litters. The appearance of the mosaic spots suggested 

 that in each mosaic animal the closely adjacent spots observed represented not more than 

 a single event of change. The areas, if there were more than one, were of the same color 

 and the gap never exceeded 1 cm. From the colors recorded, it was probable that 

 the spots represented mutations at two loci at least, b and c. The p locus was found 

 to be involved by microscopic examination in at least one case. 



From the size of the coat occupied by the mosaic spot the approximate number 

 of prospective pigmented cells in the irradiated embryo could be calculated. From this 

 figure, it was possible to estimate the rate of mutation — and any other change leading 

 to expression of the four coat-color recessives for which the animals were heterozygous — 

 as being 7.0 x 10 ~ 7 per locus per r. Of course, it could not be definitely decided 

 whether the changes observed were really due to somatic mutations or to other causes. 

 There was one particularly interesting case, however, that made somatic mutation 

 more probable than some of its alternatives. In this case the p locus, which is on the 

 same chromosome as one of the other markers, c, was involved, the crossover distance 

 being 16 in females and 12 in males. Here the expression oip was changed, without a 

 concomitant change in the expression of c that would have been expected if a whole 



