4. Comparison of X-ray and Ultra-violet Mutation. Following the experiment on X-ray and ultra-violet mutation of A previously reported (News Letter 1941, page 45-47, 1942, page 24-27), Roman and I set up a somewhat similar experiment with A^b. This was designed to take advantage of the fact that the spontaneous mutations of A^b are to an intermediate allele and are therefore clearly distinguishable from the effects of deficiency. The previous experiment had shown that the apparent mutations induced by X-rays ware in fact minute deficiencies and that the apparent mutations induced by ultra-violet were distinctly different and behaved as if they represented transformation of the gene to a recessive allele. It did not, however, exclude the possibility that the ultra-violet mutations were still more minute deficiencies, or cases of destruction of the single gene. With A^b this distinction could be made, if ultra-violet mutations actually are mutations of the type represented by spontaneous mutation of the same gene.

Extensive pollinations with untreated, UV-treated, and X-rayed pollen of a single A^b plant were made upon ears of a Dt, and numerous deficiencies and mutations were identified in the progeny. But the experiment failed in its main objective, because the natural frequency of mutation of A^b to a^p is so high that no significant increase in a^p mutations was produced by the treatments used.

The results, however, give additional support to the indication that the UV mutations are true gene mutations in two ways.

(1) No apparent mutation of A^b to a was found in the very extensive ultra-violet series.

(2) Among the endosperm mosaics induced by ultra-violet treatment, there were several cases in which a mosaic of clearly pale aleurone tissue showed typical dots of Dt type. Although an endosperm sector does not permit progeny testing, these can only have resulted from mutation of A^b to a^p, induced by the ultra-violet treatment. An endosperm mosaic of pale appearance could result from any one of numerous causes, but it could not provide a background for visible dots of A tissue unless it resulted from a change in A-action, and this background could not be pale if the A loss were due to deficiency.

The effect of ultra-violet treatments upon A^b mutation is sufficiently frequent for detection in the endosperm and not in the embryo because of the much higher frequency of induced alterations in endosperm than in embryo, which has previously been reported as characteristic of ultra-violet treatment.

This heightened frequency of endosperm alterations may be used to simplify various studies involving ultra-violet effects, and to make possible certain studies which otherwise could not be carried out. For example, it would be very desirable to determine the effect of varying ultra-violet wave lengths on the frequency of mutation. The action spectrum for A-losses in endosperm has been determined, but these include both deficiencies and mutations, and presumably consist very largely of deficiencies. It would not be possible to make significant comparisons of wave length effectiveness in inducing mutation if the mutations could be identified only by the growing and testing of progeny plants.

The use of A^b, with recognition of mutants by the a^p phenotype, as described above, is effective for identifying positive cases of mutation in the endosperm, but it is not suited to quantitative work because of frequent failure of a^p sectors to color positively. Laughnan and I have therefore made use of a different method, which permits identification of the alterations in the endosperm but with confirmatory tests on the plant grown from the accompanying embryo.

Pollen of homozygous A A with the recessive markers gl3 and j was used on ears of a-X1/a^p. The x-ray mutants a-X1, a-X2, etc., are inviable when homozygous and in all possible combinations inter se, and sectors homozygous or hemizygous for them are also inviable (News Letter, 1942, page 25). If all X-ray induced A-losses involve the loss of the associated viability factor, X-rayed pollen will never yield a colorless seed or sector; if any apparently colorless or sectorially colorless seed is found, it may be tested by growing the plant to determine whether the female gamete was a-X1 or a^p. A colorless seed yielding a plant not heterozygous for a^p is selfed or tested for the recessive markers to exclude the possibility of pollen contamination.

The A-losses shown by a^p tissue include the deficiencies plus the mutations among the seeds from a^p gametes; those shown by a tissue include the mutations alone among the seeds from a-X1 gametes. Control pollination by a C R on a number of ears of the female stock show that ♀ gametes of a^p and a-X1 functioned in approximately equal numbers.

In the limited populations now completed, X-ray treatment has failed to yield colorless seeds or sectors. Ultra-violet treatment has given 3 proven cases of colorless sectors. The total number of A-losses in endosperm in the ultra-violet population on which the tests have been completed was 92. This indicates a ratio of deficiency to mutation of about 86:3 under ultra-violet treatment for the A stock used in the experiment. This is not greatly different from the proportion found among progeny plants representing A losses in the embryo.

The induced alterations classified as mutations are subject to the same reservations regarding their genetic nature as are the ultra-violet mutations identified in progeny plants following treatment of A. The method permits the determination of relative frequency of mutation (in this sense) with a fraction of the effort required in determining mutation from progeny plants. By this method it is feasible to compare the effect of different wave lengths upon deficiency and mutation simultaneously, and to compare different A alleles in relative frequency of mutation. With slight modifications the method may be used also for the identification of gene mutations of A^b critically distinguishable from the effects of gene-deficiency.

The results of the above experiment have a further interest in connection with the problem of the endosperm-embryo difference in frequency of ultra-violet alterations. The cause of this difference is unknown, and the most plausible guess has been that it is somehow connected with the difference in breakage-fusion phenomena in endosperm and embryo, which might appropriately be termed the McClintock effect. It might be expected that deficiencies, initiated by equal effects of the treatment upon the two sperm nuclei, might differ greatly in frequency of realization under the very different conditions of endosperm and embryo. But this experiment indicates that the heightened frequency of alterations in the endosperm applies to mutations as well as deficiencies.

While the various experiments with induced mutation of A and A^b indicate that ultra-violet treatment produces true gene mutation and that X-ray treatment does not, they are disappointing in their failure to yield induced gene mutations which may be established in stocks subject to critical analysis. This is due to the failure of the A^b experiment described on an earlier page of this report. The advantage of regular spontaneous mutation to an intermediate allele, which makes A^b suitable for this experiment, applies also to R^r, since its spontaneous mutations are regularly to R^g rather than to r^g. In the case of R^r distinct alleles available, including types with varying frequency of spontaneous mutation. Mrs. Elena Perak has undertaken an extensive study of the effects of X-ray and ultra-violet treatment upon mutation of various alleles.