Genetics. In the article “Spontaneous mutations recovered as mosaics in the mouse specific-locus test” by L. B. Russell and W. L. Russell, which appeared in number 23, November 12, 1996, of Proc. Natl. Acad. Sci. USA (93, 13072–13077), the authors request that the following correction be noted.

The line below should be added to Table 4:

Table.

Treatment	Allele	Offspring scored	Percentage of germ-line mosaic*
X	ajv	25	16.0

The addition of a seventh cluster requires minor corrections in some of the frequencies appearing elsewhere in the paper. Thus, in Table 5 (last line), the frequency (with 95% confidence limits) of masked mosaics becomes 13.3 (6.2, 26.1) × 10⁻⁶; this, like the original figure, 11.4 (4.9, 24.3) × 10⁻⁶, is not significantly different from the frequency of visible mosaics. As regards the per-generation rate, the combined frequencies of masked mosaics and heterozygotes (G₁ information, see p. 13077, second paragraph) is now 1.6 times the rate derived from singletons—changed from 1.4 and thus even closer to the more reliable multiple of 1.7 calculated from the G₂ information. The fact that the previously omitted cluster is at the a locus increases the significance of the difference between the masked–mosaic locus spectrum and that for singletons (from χ² = 27.53 to χ² = 31.97, with 7 degrees of freedom) (p. 13076, third paragraph), further strengthening the conclusion about the different nature of mutations arising spontaneously during the perigametic interval and those arising in mitotic divisions.