THE NEW PHYTOLOGIST VOL. XVI., Plate I.

THE NEW PHYTOLOGIST VOL. XVI., Plate I. Flg. 1. Fig. 2. Fig. 3. DAVEY AND GIBSON My/?/a4 GALE

A. J. Davey & C. M. Gibson. 147 NOTE ON THE DISTRIBUTION OF SEXES IN MY RICA GALE. BY A. J. DAVEY, M.SC, AND C. M. GIBSON, B.SC. [WITH PLATE I AND ONE FIGURE IN THE TEXT.] M Y RICA GALE, the common Bog Myrtle, is described as typically dicecious, thus agreeing with the majority of the other members of Myricaceae. The present note calls attention to the characters of various types of monoecious individuals and to variations in sex. The occasional occurrence of some expression of the monoecious condition has heen noted by several authors.' Thus Chevalier' mentions the exceptional presence of monoecious shoots, and of androgynous catkins, and Marshall Ward' states that hermaphrodite flowers may occur. Observations during several successive years on a large area of Myrica Gale in the peat moors of Somerset show that there always exists a small proportion of monoecious plants which present all gradations between the normal staminate and pistillate types. The series includes plants bearing catkins which contain hermaphrodite flowers. This has been found to be true as regards other areas in different parts of the country. Further it has been found that the sex (if it may be so termed) of a bush or shoot may vary from year to year. The variations observed during several years have been almost entirely in the direction of change from the pistillate to the staminate condition; but in the present season (1916) several instances of the reverse change have been noted. Tbe Somerset area includes situations differing considerably in level, and consequently differing as regards moisture conditions. The low levels produced by the removal of a considerable depth of peat are always wet and boggy, and are flooded to various depths during the winter; the high levels consist of dry peat which is never flooded. In all situations, staminate plants appear to be more numerous, but the relative proportion of pistillate plants was found to be greater in the wet than in the dry areas. Thus in a dry area Moss, C. E. The Cambridge British Flora. Vol. II., Camb. Univ. Press, 1914, p. 69. ^ Chevalier, Aug. Monographie des Myricacees. Mem. Soc. Sci. Nat. Cherbourg, XXXII., 1901-2, p. 178.» Marshall Ward, H. " Trees." Vol. III., Camb. Univ. Press,1905, p. 204.

148 A.J. Davey S C. M. Gibson. the proportions were roughly: staminate 75%, pistillate 10%, the remaining!5% being monoecious or hermaphrodite. At the same time, in a very wet area, the proportions were: staminate 56%, pistillate 38%, hermaphrodite or mixed 6%. This represents countings of shoots traceahle to the ground, which cannot with certainty be taken to represent individual plants. Since the staminate plants are always much more vigorous than the pistillate it is prohable that the excess of staminate over pistillate individuals may be more apparent than real, although in the dry levels large areas of plants are met with, in which the pistillate type is almost entirely lacking. As is well known, the catkins are borne towards the ends of special shoots, and up to the time of flowering the staminate catkins are much larger than the pistillate. Tbe bracts of the staminate catkin are broad and spreading, while those of the pistillate catkin are narrow and compactly arranged. In both male and female catkins, the flowers occur singly in the axils of the bracts. The male consists of a group of four (sometimes more) stamens, without scales or bracteoles; the female consists of a bicarpellary ovary flanked by two adnate scales or bracteoles, which enlarge considerably after fertilisation to form the characteristic wings of the fruit. As the pistillate catkin develops its axis lengthens and becomes woody, while its bracts enlarge and become green. The woody axis persists long after the fruit has been shed, thus providing an indication of the amount of fruit that has been set in preious seasons. The monoecious or "mixed" plants may be grouped as follows, but all the types grade into one another. A. Plants or shoots bearing staminate and pistillate catkins of the normal type. These may occur on different shoots, or hoth may be found in the same shoots. Staminate, pistillate and mixed shoots may he found on the same plant. On the mixed shoots staminate catkins are generally found helow, and pistillate ahove, but the reverse order may obtain. Androgynous catkins may be found in the intermediate region. There is much variation in the proportions of the different kinds of shoots present. B. Plants bearing androgynous catkins. These catkins are large and resemble the normal staminate catkins as regards form, and the size and shape of their bracts. Stamens generally occur in the lower part of the catkin and pistils above ; this is the arrangement commonly met with in those species of Myrica

The Distribution of Sexes in Myrica Gale. 149 in which androgynous catkins are the rule {e.g. M. conifera, described by Chevalier'). Sometimes the reverse arrangement is found. The middle regions of these catkins may contain hermaphrodite flowers or flowers with rudimentary pistils. C. Plants with shoots the hulk of whose catkins consist of hermaphrodite flowers. The catkins with hermaphrodite flowers are intermediate in size between the normal pistillate and staminate catkins, but may incline towards either kind in form and appearance, according to the nature of the bracts. The larger staminatelike type with broad and wide-spreading bracts is the more common. The pistillate type is smaller and more compact. The hermaphrodite flower contains a centrally placed ovary, similar to that of the normal female flower. Round it there are three or four stamens united to the base of the ovary, and just below these on the very short floral axis are two minute lateral outgrowths, corresponding to the " bracteoles " of the normal female flower. (See Fig. 1.) Both stamen and stigmas are functional, and the flowers are proterogynous. They are capable of producing fruit. PJP 2 Transverse sections of hermaphrodite flower at different levels, showing ovary and four stamens (1, 2, 3, 4, numbered downwards). St style; o, ovary; ov, ovule; /, filament; a, anther; b, bracteole ; JB,'bract. ' Vascular bundles supplying styles are shown blacu. In the upper scales of the catkins various abnormal stages may occur; e.g., a centrally placed pair of stamens, flanked by two 1 Chevalier, loc. cit., p. 229.

150 Distribution of Sexes in Myrica Gale. small green scales corresponding in position to the" bracteoles" of the normal female fiower, was found persisting, while fruits were developing in lower parts of the catkin. It is common to find rudimentary ovaries in which the " bracteoles " are replaced by a pair of stamens. Variations in Sex front year to year. Bushes and shoots of various types have been marked and examined in successive years, their previous state as regards fruit production being recorded. There is considerable difficulty in identifying an individual since the plant possesses underground stems. The changes recorded include only sucb as took place on the branches of single shoots traceable to the ground; in a few cases where a tree or a bush is mentioned, reference is made to isolated tree-like plants possessing only one stem or trunk. In fourteen cases clumps or bushes found to be entirely pistillate in 1913 and in 1914, produced staminate catkins in 1915, in some cases abundantly, and in others sparsely. One plant became almost entirely staminate. In some cases staminate shoots were produced, the upper catkins of which were hermaphrodite ; other branches bore pistillate and staminate shoots in about equal numbers, while a few remained for the most part pistillate, producing only a few staminate shoots. Thus sometimes the change may be generally spread over a tree or clump of bushes, while in others only certain shoots may be affected. In the latter case there may be progressive change extending over several seasons, since in successive years the proportion of staminate catkins increased. Thus certain trees and branches which produced abundant fruit in 1913 developed mixed shoots in 1914, and in 1915 had become almost entirely staminate. In 1916, however, some of these showed an increase in the production of pistillate catkins. It would seem that it should be possible to gain some insight into the causes underlying these variations in sex, and to ascertain whether external conditions play any part in determining tbe proportions of the sexes which appear. The conditions which naturally suggest themselves are moisture, temperature, and light (witb their influence on nutrition), and also the previous state of a plant as regards the production of fruit. Since the staminate flowers are developed early in the season preceding that in which they flower, while the pistillate catkins

H. Stuart Thompson. 151 develop much later, it is possible that conditions accelerating or retarding the development of catkin buds may influence the proportions of the two kinds. Variations similar to those above recorded for Myrica Gale are described by Chevalier as occurring to a marked extent for certain monoecious species of Myrica (notably il/. cali/ornica, M. conifera and M. pubescens), in which a polygamous condition obtains. Chevalier suggests nutrition as the factor which controls the distribution of the sexes on monoecious individuals in any one season, as well as the changes taking place in different seasons. In conclusion, our thanks are due to Dr. E. N. Thomas for encouragement and helpful criticism. BOTANICAL RESEARCH LABORATORY, UNIVERSITY COLLEGE. DESCRIPTION OF PHOTOGRAPHS ON PLATE I. 1. Different kinds of flowering shoots (April, 1913), a, b, c, d, reading from left to right. a. Branch with staminate catkins, showing remains of small amount of fruit of previous years. b. Branch with remains of fruiting catkins (1912) bearing one staminate and two pistillate shoots. c. Branch terminated by fruiting shoot of 1912, below which have arisen shoots bearing staminate and " hermaphrodite ' 'catkins d. Is entirely pistillate, with remains of previous season's fruiting catkins. 2. Fruiting, branch in August, bearing a shoot with young staminate catkins for next season. 3. Branch bearing staminate catkins on 1915 shoots. On the left is a bare persistent staminate shoot of the previous season, and on the right a fruiting shoot showing woody axes of catkins and bracts persisting after the fruit has been shed. RADICAL LEAVES OF PARNASSIA AND VALERIAN A DIOICA. BY H. STUART THOMPSON, F.L.S. [WITH TWO FIGURES IN THE TEXT]. PALUSTRIS CONFUSION may possibly occur between young plants of these two species (which often grow in the same habitat) owing to the superficial similarity of their radical leaves. The writer fell into this mistake two years ago, believing he had gathered the former plant, which had not been recorded from Somerset since