The DNA-Content of nuclei in the meristem of onion roots

Summary The techniques of making and evaluating measurements of nuclear DNA-contents by the two wave length method, including wave length determination were worked out. The efficiency of the analysis increased by unbiased elimination of off-values. Correction for non-specific absorption was made but remained, with the present material (sections), necessarily imperfect. This seems to be the factor which limited the, otherwise apparently very high, optical accuracy of which instrument and methods are capable. Errors in the determination of the nuclear dye-content were, even so, small compared with the more extreme cases of proportionality errors in the determination of the DNA-content that resulted from disproportionality of Feulgen dye- and DNA-content.Direct evidence of proportionality errors entering the DNA determination was found when different mitotic stages were compared. The ratio of mean dye-contents of two stages varied in slides which were prepared differently in regard to fixation, hydrolysing agent, or duration of hydrolysis. In seven such samples the mean dye-content of metaphases ranged from 80 to 115% (with mean 100.2%) of the prophase mean, the mean dye-content of mid-anaphases from 92 to 112% (with mean 105.5%). Similarly the mean dye-contents of daughter nuclei in late anaphases and telophases bracketed 50% of the prophase value. It appears virtually certain that this variation is not caused by a corresponding natural DNA variation but is due to proportionality errors.The results strongly support the view that the DNA-content does not change at all during mitosis apart from its halving brought about by the anaphase separation of sister chromatids. The results are further fully compatible with the hypothesis that the DNA-content per complete genome is strictly constant except during the period of DNA doubling. The dye-contents of prophase nuclei (also of metaphases) of the same sample have a coefficient of variation (corrected for non-systematic reading errors) of about 5.5%. This appears to be well within the limits of errors due to imperfectly corrected non-specific absorption and of proportionality errors.The proportionality errors found in measurements of different mitotic stages reveal a pattern similar to that prevailing in published DNA measurements: apparent deviations from DNA constancy are usually small and often negligible, but are occasionally large enough to be misleading. None of the claimed disproofs of DNA constancy is in any way conclusive, as such claims necessarily rest upon the unwarranted assumption that proportionality errors are always very small or absent.The development of the DNA-content during interphase was studied by correlating DNA-content and nuclear volume. Earlier findings on roots of other plants are confirmed. An interphase I with the DNA-content 2C is followed by a period of DNA synthesis (during this period the nuclear volume seems to increase only slowly, if at all) and this by an interphase III with the DNA-content 40. Each phase lasts several hours. Mitosis and DNA synthesis are essentially independent processes.Measurements on nuclei which are technically more favourable than those in sectioned onion roots have yielded a smaller or even disappearing inter-nuclear variation of the Feulgen dye-content. A strict DNA constancy per complete genome appears more and more likely.Contribution from the programme in Cytology, Department of Botany, University of Wisconsin, Madison, supported by grants to the late Dr. C. Leonard Huskins, from the American Cancer Society upon recommendation of the Committee on Growth of the National Research Council, from the Rockefeller Foundation and from the Research Committee of the Graduate School with funds supplied by the Wisconsin Alumni Research Foundation.