Some Recent Investigations in Sugar-Beet Breeding

1. Differences in the size and sugar content of individual beet roots show no evidence of inheritance. They are fluctuations, therefore, and apparently play no part in beet improvement. 2. No correlation was discoverable between percentage or quantity of sugar in sugar-beet roots of ordinary sizes and their yield of seed, nor between their yield of seed and the average percentage of sugar in their progeny. 3. The fluctuations of beet families planted in progeny rows in alternation with check rows exceeded their real differences, but real differences were distinguishable by the use of a large number of replications (cf. fig. 1). 4. Areas of beets in an apparently uniform field of small dimensions showed a difference of 2 per cent sugar. 5. Percentage of sugar and yield of sugar of sugar-beet rows vary independently. Progeny rows should be graded on both percentage and yield of sugar, therefore, or on yield of extractable sugar. 6. The average weight of root per row increases with yield of sugar and decreases with percentage of sugar. 7. The discontinuance of selection for one generation caused no deterioration in percentage of sugar. In fact, there was some apparent gain. 8. No improvement in yield or percentage of sugar was obtained by continuous selection. Both the good and the poor families transmitted average qualities.