THE NATURE OF POLYPLOIDY IN LILIUM TIGRINUM

Abstract

in this species has made it an interesting sub.iect for cytological observations, especially since it is only polyploid found in this genus except -for tetraploids produced by colchicine which have been recently reported by Emsweller (1940). chromosome number of L. tigrinum was reported as 2n 24 by several investigators in early literature (Schaffner, 1906; Hsu-Siang, 1932). In 1930 Takenaka and Nagamatsu first reported a triploid L. tigrinum, and since time counts of thirty-six chromosomes in it and several of its varieties have been repeatedly confirmed (Chandler, Porterfield, and Stout, 1937; Mather, 1934, 1935; Sansome and LaCour, 1934; SatO, 1932, 1937; Stout, 1933; Takenaka, 1933; and Westfall, 1940). Preston (1932) reported a L. tigrinum v. Fortunei which set seed on selfing. Stout (1933) later found this was a diploid. Investigators of meiosis and mitosis in several forms of L. tigrinum have given evidence as to nature of polyploidy present. All have interpreted evidence as indicative of autopolyploidy (Chandler et al., 1937; Mather, 1934, 1935; Sansome and LaCour, 1934; Sato, 1932; Takenaka, 1933; Westfall, 1940). Chandler et al. (1937) reported twelve sets of three morphologically homologous chromosomes. A high number of trivalents were found at meiosis and there were no bridges, but chromosomes lagged at anaphase. Their conclusion was plants were autotriploids since chromosome trios were structurally homologous. Westf all (1940) also described L. tigrinum as an autotriploid species. He reported chromosome sets were similar and there was a high frequency of trivalents at meiosis. He showed bridges and fragments in his figures and stated only one chromosome was heterozygous for an inversion. To explain fragments of several size classes associated with different bridges he stated, The acentric fragment may vary in length, depending on distance from end of chromosome at which chiasma in inverted portion forms. only evidence of non-homology of chromosomes reported is failure of correspondence in all chromomeres of a set of three homologous chromosomes, at meiotic prophase (Chandler et al. 1937). Westfall (1940) reported same nonhomology at prophase. Sato (1937) counted chromosomes in F1 progeny of selfs and intercrosses of triploid L. tigrinum. He found all numbers from 24 to 39 with exception of 31, with 1 Received for publication August 1, 1942. peaks at 26-28 and at 36. Although he referred to the autotriploid mother plant he interpreted chromosome counts in F1 to mean that tiger lilies with 36 chromosomes had a certain number of chromosomes with changed structural portions besides having homologous and non-homologous chromosomes. Preliminary observations of chromosomes in our collections of L. tigrinum indicated they differed in several respects from those already reported in literature. Therefore, it seemed worth while to reconsider some of previous conclusions regarding type of polyploidy present. MATERIALS AND METHODS.-ROOt tips were obtained from three long established clones of L. tigrinum in Maryland and Pennsylvania, labelled 2-42, 3-42, and 4-42, as well as L. ltigrinum, 1-42, and L. tigrinum v. Fortunei, Malmo, Splendens, and Flore-pleno from commercial sources. root tips were killed and fixed in 3:1 absolute alcohol-glacial acetic acid for twenty-four hours and then stored in 80 per cent alcohol. Some tips were treated in a 0.2 per cent colchicine solution one and one-half hours before killing to assure a large number of metaphase figures. Smears were made by usual propionocarmine method within two weeks of collection. Buds were collected and smeared by same method from five of eight clones. No colchicine treatment was used on pollen mother cells. In analysis of somatic chromosome morphology, only very flat anaphase and metaphase figures in which chromosomes showed no sign of breaking or stretching were used. Table level camera lucida drawings were made of each chromosome in ten metaphase and anaphase cells of each variety, and these drawings measured. ratio of each segment of chromosomes, except for short arm