214 Cytologia 14 Chromosome Numbers in Pandanus, Sparganium and Typhal By Ititaro Harada Botanical Institute, Faculty of Science, University of Tokyo Received March 10, 7947 In the present paper the chromosome numbers of some species of Pandanus (Pandanaceae), Sparganium (Sparganiaceae) and Typha (Typhaceae) will be reported. These monocotyledonous plants employed in the present work were identified taxonomically by Dr. S. Miki and Dr. H. Hara. Root tips were fixed with either Navashin's solution or Bonn modification of Flemming's solution and sections made by the usual paraffin method were stained with gentian violet after Newton's method. Flower buds were fixed with either Carnoy's fluid or Bouin-Allen's solution and then preserved in 75% alcohol, after being washed in 95% alcohol. From the preserved buds the anther sacs were taken out and smeared in aceto carmine. The writer is much indebted to Dr. T. Hiraoka, who kindly fixed flower sipkes of Typha angustata with Carnoy's fluid. Descriptions of the Results Pandanus (Pandanaceae) P boninensis Warburg: 2n=ca. 64 chromosomes were counted in root tip cells (Fig. 1). All of the chromosomes are short rod or dot shaped, some of them being more or less different in size. Sparganium (Sparganiaceae) S. glomeratum Laestadius var. angustifolium Graebner: 2n=30 short rod chromosomes, in root tip cells (Fig. 2). S. japonicum Rothert: 2n=30, in root tip cells (Fig. 3). In many of the metaphase plates observed, there can be distinguished some consider ably longer chromosomes from the other short rod-shaped ones, especially one pair of them being extremely long (L). Fifteen gemini, in the first meiotic metaphase of pollen mother cells (PMC-I) (Fig. 4). S. stenophyllum Maximowicz: In some root tips collected from the plants in the botanical garden of Kyoto University, 30 chromosomes were counted (Fig. 5), but in other ones collected from the same place, 45 chromosomes (Fig. 6), both consisting of the short or long rod-shaped 1 Contributions from the Divisions of Plant-Morphology and of Genetics , Botanical Institute, Faculty of Science, University of Tokyo, No. 314. 1947 Chromosome numbers in Pandanus , Sparganium and Typha 215 Fig. 1, Pandanus boninensis. Figs. 2-15, Sparganium; 2, S. glomeratum var. angusti. folium. 3 & 4, S. japonicum. 5-9, S. stenophyllum; 5 & 6, from the botanical garden of Kyoto Univ. 7, from Prov. Kida. 8 & 9, from Inada-noborito. 10 & 11, S. stoloniferum. 12, S. stoloniferum var. coreanum. 13, S. stoloniferum var. macrocarpum. 14 & 15, S yamatense. Figs. 16-22, Typha; 16 & 17, T. angustata. 18 & 19, T. latifolia. 20, T. latifolia (variegated form). 21 & 22, T. orientalis. (ca. •~3000). chromosomes. As the root tips of several individual plants were fixed together, it became impossible to decide whether the root tips with the two numbers, 30 and 45, were derived from the same individual or not. Meiosis could not be studied in those plants in Kyoto, because the buds contained no suitable anthers for observation. In material obtained from Province Kida of Sikoku 15 gemini are counted in, late diakinesis of PMC (Fig. 7). Moreover, each individual of eleven from Inada-noborito of Province Musasi shows 2n=30 in root tip cells (Fig. 8), and in two in dividuals, from which favourable anthers have been obtained, 15 gemini Cytologia 14, 1949 15 216 I. HARADA Cytologia 14 were observed in PMC-I (Fig. 9). S. stoloniferum Buchanan-Hamilton: 2n=30, in root tip cells (Fig. 10). Among the short rod- and dot-shaped chromosomes of complements, one pair of distinctly long rod-shaped can be distinguished (L). 15 gemini, in PMC-I (Fig. 11). S. stoloniferum Buch.-Ham. var. coreanum Hara: 2n=30, in root tip cells (Fig. 12). Short rod- or dot-shaped, with some fairly long ones. S. stoloniferum Buch.-Ham. var. macrocarpum Hara: 2n=30, in root tip cells (Fig. 13). Chromosome shape is similar to those of S. stol. var. coreanum. S. yamatense Makino: 2n=30, in root tip cells (Fig. 14). Short rod or dot-shaped. 15 gemini, in PMC-I (Fig. 15). Typha (Typhaceae) T. angustata Bory et Chaubard: 2n=30, in root tip cells (Fig. 16). Short rod- or dot-shaped. 15 gemini, in the late diakinesis of PMC (Fig. 17). T. latifolia L.: 2n=30, in root tip cells (Fig. 18). Short rod- or dot-shaped. 15 gemini, in PMC-I (Fig. 19). T. latifolia L. (variegated form): 2n=30, in root tip cells (Fig. 20). Short rod- or dot-shaped. T. orientalis Presl: 2n=30, in root tip cells (Fig. 21). Among short rod- or dot-shaped chromosomes, one pair of somewhat larger chromosomes (L) is distinguished. 15 gemini, in PMC-I (Fig. 22). Considerations The present work has been carried out to contribute to the karyotype investigation of Monocotyledons that is the cooperative work of our labo ratory. It is almost impossible, however, to deal with a detailed karyotypic analysis on the concerned plant groups, Pandanus, Sparganium and Typha, since the chromosomes are very small in size (ca. 2ƒÊ-0.7ƒÊ) and theirr shapes are scarcely distinguishable from each other, being short rod- or dot-shaped. But in some cases, as in Sparganium japonicum, S. stoloni ferum and Typha orientalis, specially long or minute chromosomes can be recognized, although they represent no characteristics enough to make the karyotypic comparison among species. As to the basic number of chromosomes there is nothing to refer in Pandanus, while that of both Sparganium and Typha is b=15, being in accordance with the results of Wulff ('38), Scheerer ('40) and Hagerup ('41) on Sparganium and Roscoe ('27) on Typha. All of Sparganium species observed by the writer are diploid plants (2n=2b=30), except the case where occurs a triploid chromosome set (2n=3b=45) in some root tip cells of S. stenophyllum; the diploid number 1947 Chromosome numbers in Pandanus , Sparganium and Typha 217 of 30 chromosomes was reported in S. minimum Fr. and S. ramosum Huds . by Wulff and in S. simplex Huds. by Scheerer and the 15 bivalents were counted in PMC-I of S. simplex Huds. by Hagerup. In Typha species all the present observations indicate the normal diploid nature (2n=30), showing regular bivalent formation (15II) in the plate of PMC-I. However, Roscoe found somewhat irregular behaviours in the chromosome pairing in some T ypha species: although T. latifolia, growing in the place far apart from that of T. angustifolia, shows thorough ly normal pairing in each 15 bivalents (therefore he regards this type of T. latifolia as the cytological "pure" species of Typha), "where T. latifolia ranges with T. angustifolie, forms appear which show slightly irregular chromosome action" (i.e. failure of pairing). In T. angusti f olia and T. angustifolia hybrid ("forms lying systematically somewhere between T. latifolia and T. angustifolia in taxonomic characters") he observed many figures which, besides bivalents, show univalent chromosomes occur ring in various extent. In T. angusti f olia var. muelleri he counted 30 bivalents; accordingly this variety is a tetraploid plant. In the plants used in the present work, however, there appeared no such irregularities as found by Roscoe who attributed it to hybridity. Most systematists, e.g. A. Engler & L. Prantl (1889, Die natiirlichen Pilanzenfamilien, II. Teil), O. Warburg (1900, Das Pflanzenreich, 3 Heft, IV. 9) and P. Graebner (1900, Das Pflanzenreich, 2 Heft, IV. 8, 10) included Pandanaceae, Sparganiaceae and Typhaceae in an order Panda nales, regarding that these three families are closely related; besides Engler (s. Engler & Prantl 1889) considered that Sparganiaceae stands in a closer relationship with Pandanaceae than with Typhaceae. In this respect we find a sharp discrepancy in opinions between J. Hutchinson (1934, The families of flowering plants. II) and Engler (1889); the former insists that Sparganiaceae and Typhaceae have a remarkably close relationship, forming Typhales, "as a reduced and very advanced group derived from the Liliaceous stock" and Pandanaceae, forming Pandanales only with one family, has entirely no relationship with Typhales and is closer to Palmae and Cyclanthaceae. Moreover, S. Miki (1937, Bot. Mag. Tokyo, 51: 472-480) also recognized a closer relationship between Panda naceae and Cyclanthaceae. Though the cytological results obtained in the present work can not give any clear evidence for these systematical and phylogenical relationships mentioned above, it may leave something for consideration as following: from the facts that one species of Pandanus has 2n=ca. 64 and both Sparganium and Typha have each a basic number 15, the writer is inclined to regard Hutchinson's opinion as a more plausible. Since both Sparganiaceae and Typhaceae represent a considerably similar karyotype and have the same basic number, it may be reasonable to insist 15* 218 I, HARADA Cytologia 14 the closer relationship between them. In regard to the relationship of Pandanaceae with Palmae or Cyclanthaceae nothing can be said at present, for there are lacking cytological studies of other Pandanaceous genera, Freycinetia and Sararanga. Summary Chromosome numbers of some species in Pandanus, Sparganium and Typha are reported, as given in the following table. Basic number is 15 in both Sparganium and Typha and that of Pandanus remains un decided. Somatic chromosomes are generally of short rod or dot-shaped in all species studied. Literature Hagerup, O. 1941. Nordiske kromosom-tal. I. Sot. Tid. Dansk Bot. For. 45: 385-395. Roscoe, M. 1927. Cytological studies in the genus Typha. Bot. Gaz. 84: 392-406. Scheerer, H. 1940. Chromosomenzahlen aus der Schleswig-Holsteinischen Flora II. Planta 30: 716-725. Wulff, H. 1938. Chromosomenstudien an der schleswig-holsteinischen Angiospermen-Flora II. Ber. d. deut. hot. Ges. 56: 247-254. ADDENDUM Love & Love ('48), in their voluminous list of chromosome numbers reported up to 1947 of the northern European plant species, described their own results on the chromosome numbers of some Sparganrum species, i.e.