Reproductive Biology of Ammophila breviligulata

Abstract

Sexual reproduction of Ammophila breviligulata Fernald was initiated by the emergence of panicles in late June during 1976 and 1977. Only a small proportion (28-48%7) of the florets on a panicle produced mature, well-developed caryopses. The hypothesis of nonviability of pollen was rejected as a reason for poor fertility. Low percent fertility could result from the activity of biotic agents (a beetle, Strigoderma arboricola, aphids and ergot, Claviceps purpurea), self-incompatibility, pollen shortage or adverse environmental conditions during maturation. INTRODUCTION Ammophila breviligulata Fern. is a pioneer, colonizing, perennial grass that occupies the foredune habitat along the Atlantic Coast from Nova Scotia to North Carolina and along the Great Lakes. The species possesses specialized adaptations that allow it to tolerate burial by sand, sand-blasting, salt spray and drought conditions (Woodhouse et al., 1976). The possession of these characters, along with the ease in vegetative propagation and relative freedom from pests and diseases, make A. breviligulata and the introduced A. arenaria the two most frequently used species to stabilize wandering dunes, blowouts and denuded shorelines in North America. Several studies (Harper, 1965; Sharitz and McCormick, 1973; Payne and Maun, 1981) have shown that pioneer successional plant species, especially annuals, are characterized by high reproductive potential. Such species typically produce and disperse large numbers of seeds and suffer high juvenile mortality (Sharitz and McCormick, 1973). Ammophila breviligulata, although a pioneer in succession, expands into a contiguous area primarily by vegetative reproduction. Krajnyk and Maun (1981a) showed that the plants are highly successful in clonal growth during the juvenile phase and rapidly colonize an area by the addition of serial rhizome modules. The plants devote very little energy to sexual reproduction. The ratio of flowering culms to all culms in Ammophila breviligulata populations along Lake Huron was 0-0.33 (Krajnyk, 1979) and along Lake Michigan was ca. 0.13 (Laing, 1958). The density -of flowering culms was only 0-8.5 per m2 along Lake Huron (Krajnyk, 1979) and 1.3 per m2 along Lake Michigan (Laing, 1958). Eldred and Maun (1982) showed a significant positive correlation between sand accretion and proportion of flowering to nonflowering culms in a population. Even in flowering sites, only a small proportion of florets on a panicle contain normal, well-developed caryopses. The reasons for this are obscure. The causes of low percent fertility could be nonviable pollen, insect damage, sensitivity to environmental factors, shortage of pollen and meiotic abnormalities. An investigation was undertaken to examine some critical factors affecting the age-specific survival from flowering to the predispersal phase of Ammophila breviligulata. The objectives of this study were (a) to determine fecundity, percent fertility, seed weight and dimensions, and (b) to study five causes of low fertility: anthesis cycle, nonviability of pollen, self-incompatibility, shortage of pollen and effects of biotic agents. MATERIAL AND METHODS These studies were conducted on Ammophila breviligulata plants growing along the shorelines of Lakes Huron and Erie at Pinery and Rondeau provincial parks, respectively, in Ontario, Canada. For details on locations, see Maun (1981) and Krajnyk and Maun (1981a). Fecundity and percent fertility in natural populations. Panicles of Ammophila breviligulata