Figure 1.
Flowers of H. fulva (A), F1 hybrid (B), H. citrina (C) and F2 hybrids (D-F).
Figure 2.
Variation of flower colors in F1 (A) and F2 hybrids (B).
The horizontal axis is the standard color chart score. Larger scores indicate reddish color and smaller scores indicate yellowish color. Color chart scores from 2–13 were classified as the yellow group, and color chart scores from 14–23 were classified as the yellow-orange group. The vertical axis is the number of plants.
Figure 3.
Reflectance spectra of tepals of two Hemerocallis species, F1 hybrid and Standard Color Charts.
(A) Reflectance spectra of the central part of tepals (upper, H. citrina; center F1 hybrid; lower, H. fulva). (B) Reflectance spectra of the peripheral part of tepals (upper, H. citrina; center F1; lower, H. fulva). (C) Reflectance spectra of three representative Standard Color Charts (upper, SCC=3; center SCC=13; lower, SCC=23).
Figure 4.
Typical reflectance spectra of F2 hybrids (above) and the relationship score, reflectance (below).
(A) Reflectance spectra of the central part of tepals. Three representative F2 hybrids, DG11 (SCC=3), BD3 (SCC=13) and BC12 (SCC=21), are showed. (B) Reflectance spectra of the peripheral part of tepals. (C) The relationship between color chart score and relative reflectance at 525 nm of the central part of tepals. (D) The relationship between color chart score and relative reflectance at 360 nm of the peripheral part of tepals.
Figure 5.
Variation of fragrance intensity in F1 (A) and F2 hybrids (B).
The horizontal axis is the intensity of floral scent measured with a handheld odor meter. The odor meter can show relative intensity of scent in an arbitrary scale. All data sets were measured by the same odor meter for reproducibility. The vertical axis is the number of plants.
Figure 6.
Relationship between flower color and fragrance in F2 hybrids.
The horizontal axis is the standard color chart score. The vertical axis is the intensity of floral scent measured with the odor meter. A Pearson's product-moment correlation coefficient is –0.0336 (P=0.684).
Figure 7.
The number of four groups of flower visitors observed in each time zone.
The number of visitors were pooled during the observation period of experiment 1 (A) and experiment 2 (B).
Table 1.
Statistics for four pollinator groups in field experiments.
Figure 8.
Partial regression coefficients of frequency of pollinator visits on floral types, H.fulva or F1 hybrid.
Daily results (gray bars) and total results (black bars) are shown along the horizontal axis in the order of date. The positive regression coefficient means that the pollinators prefer H. fulva. Conversely the negative regression coefficient means that the pollinators prefer F1 hybrids (***, P<0.001; **, P<0.01; *, P<0.05; after Bonferroni correction, (A) swallowtail butterflies, n=324, (B) hawkmoths, n=360).
Table 2.
Plant-to-plant transitions made by pollinators in plots containing H. fulva and F1 hybrids.
Figure 9.
Partial regression coefficients of frequency of pollinator visits on flower color and scent intensity.
Gray bars represent the daily results and black bars represent total results (**, P<0.01; *, P<0.05 after Bonferroni correction. panel A, C: swallowtail butterflies, n=285, panel B, D: hawkmoths, n=286). In flower color (A, B), the positive regression coefficient means that the pollinators prefer reddish flowers to yellowish flowers. Conversely the negative regression coefficient means that the pollinators prefer yellowish flowers to reddish flower. In floral scent (C, D), the positive regression coefficient means that the pollinators prefer the flowers with stronger scent. The negative regression coefficient means that the pollinators prefer the flowers with weaker scent.