Morphological, ecological and cytological studies on cultivated and wild butterburs, Petasites japonicus MAXIM. IV.

On the chromosome numbers

Abstract

Somatic chromosome numbers of the cultivated and wild butterburs collected from various localities of Japan were counted in order to elucidate the origin of cultivars, and relationships between polyploidy and their characteristics or geographical distribution were studied.
1.“Aichiwasebuki”and“Mizubuki”, two commonly grown cultivars, were found to be triploid with 2n=87, and the gigantic cultivar called“Aichiwasebuki”, subsp. giganteus KITAM., was found to be diploid with 2n=58. The local strain“Akitaaobuki”was consisted both of triploid and diploid plants. (Table 1, Figs. 1, 6, 7.)
2. In the wild butterburs, 437 diploid plants were collected from 142 different habitats and 191 triploid plants from 62 habitats. Gigantic wild butterbur“Akitabuki”which distributed in Hokkaido and Tohoku districts were found to be diploid with 2n=58 without exception. Diploid and triploid plants showed a marked difference in the geographical distribution. That is, diploids distributed widely in any localities of Japan, while triploids which were not found in Hokkaido district showed a gradual increase toward south. In Kyushu district, southern part of Japan, the number of triploid plants collected was larger than that of diploid ones. (Table 2, Figs. 2-5, 8-10, 11)
3. There is no possibility of the triploid plants being hybrids between diploid and tetraploid plants, since no tetraploid ones have been discovered both on cultivars and wild butterburs. The interspecific hybridization may be also impossible, because only one species of genus Petasites has been found in Japan. Consequently, the triploid butterburs must have been derived form fusion between unreduced and normally reduced gamates of diploiod butterburs, and the triploids are presumed to be autotriploid.
4. The completely sterile male and female plants were found to be triploids and differed morphologically in the individual floret or inflorescence from diploid plants. Namely, the style of triploid male plants was markedly shorter than that of diploid ones and the anther did not open. Of triploid female plants, the ovary and pappus did not develop at all, and the inflorescences or individual floret turned brown in appearance in an early stage of its development. (Figs. 13, 14)
5. The sex expression of butterburs was not influenced by environmental factors and in spontaneous triploid plants sex expression was not related to polyloidy. In the triploid cultivars, however, more female plants were found, which suggested that more female plants has been introduced to cultivation accidentally and propagated vegetatively.
6. Triploid butterburs, both cultivars and wild ones, were distinguished from diploid ones by their earlier sprouting, larger leaves and more vigorous growth habit. (Table3, Fig. 12)
7. The results mentioned above suggested that cultivars of butterbur have been derived from triploid wild plants and not from diploid ones.