Variability of Flowering Sex and Its Effect on Agronomic Trait Expression in White Guinea Yam

Abstract

White Guinea yam (Dioscorea rotundata) is mainly a dioecious tuberous crop that produces flowers of varying sex phenotypes. Agronomic traits in Guinea yam differ according to the sex phenotype, but the precise interaction between the traits and sex phenotype is not clearly understood. This might be due to the high heterozygosity of yam where cultivars with different flowering sex have different genetic backgrounds, which mask the sole effect of sex phenotype on the agronomic traits. This study used F₁-derived clonal progenies from a bi-parental cross to minimize the impact of different genetic backgrounds among the plants with different sex phenotypes. The impact of plant sex on agronomic traits, specifically tuber yield, was evaluated through field trials conducted for four years. The results showed that only plants with a female genotype exhibited varying sex phenotypes even within the clones of same accession grown in the same experimental field. The significant effects of sex genotype and phenotype on agronomic traits were detected. Our results revealed that the flowering date was delayed in the plants with female genotypes compared to male genotypes, even when compared only among the plants with male phenotypes. The flowering date is the most important reason for the sexual differences in tuber yield. A high tuber yield was obtained when plants with the female phenotype flowered before tuber enlargement. This result can be attributed to the fact that the low flowering intensity in female plants increases the availability of carbon resources for leaf development. Female plants also showed a large negative effect of late flowering on tuber yield owing to resource competition between flowering and tuber enlargement. These findings demonstrate the feasibility of yield improvement by controlling the flowering time, with a higher effectiveness achieved in female than in male plants.

Keywords: Dioscorea rotundata; dioecious species; flowering; sex phenotype; tuber yield; white guinea yam.

FIGURE 1

Meteorological conditions during the experimental period. The precipitation, duration of light, and air temperatures are separately shown for all the years. Arrows indicate the plant growth periods in each year. Numbers with the arrows represent total precipitation (mm), total duration of light (hours), and average maximum/minimum temperatures (°C) during the plant growth periods. Vertical dashed lines represent the planting date of each year.

FIGURE 2

Sex phenotypes in Dioscorea rotundata. (A) Photos of male (left), female (right), and monoecious (center) inflorescences taken at the experimental field in IITA were shown. Male and female inflorescences consist only of male and female flowers, respectively. A plant of the first type monoecious has inflorescences with both male and female flowers. (B) Macro photos of the male (left) and female (right) flowers.

FIGURE 3

Distribution of sex phenotype in the parental plants and the cross population. The number of plants in each of the sex phenotype is separately shown for sex genotypes. (A) Sex phenotype distribution in the parents. A total of 15 samples were taken for each parent. (B) Sex phenotype distribution in the F₁ population. The total numbers of plants of male and female genotypes were 363 (121 accessions × 3 replications) and 234 (78 accessions × 3 replications), respectively.

FIGURE 4

Distribution of agronomic traits in each of the sex phenotypes. (A) Dormancy. (B) Flowering date. (C) Plant senescence. (D) Tuber number. (E) Tuber size. (F) Tuber yield. Trait values in the F₁ population are shown in the boxplot. The horizontal lines in the boxes are the median values. The height of the box is equal to the interquartile distance, indicating the distribution for 50% of the data. Approximately 99% of the data fall between the top and bottom of the lines extending from the box. Bars with different letters show significant difference at the P < 0.05. The monoecious phenotype includes both the first and second types.

FIGURE 5

Trait correlations among the F₁ population. Pearson’s correlation coefficient with statistical significance at P < 0.05 level is separately shown for sex phenotypes and years. “Mono” and “NF” denote the monoecious and non-flowering plants, respectively. The monoecious phenotype includes both the first and second types. The empty boxes indicate that the relationship is not significant. The sex phenotypes and years are indicated in rows and columns, respectively. The red and blue colors represent the positive and negative relationships; the deeper the color, the stronger the correlation. The number of plants analyzed in each box corresponds to Figure 3B.

FIGURE 6

Correlation between flowering date and tuber yield. (A) The relationship is separately shown with a regression line for each of the sex phenotypes and years. The open circles with dashed regression line indicate that the correlation was not statistically significant at P < 0.05. The correlation coefficient and significance are the same as indicated in Figure 5. (B) The regression line of different sex phenotypes is compared.