A number of books have been published recently on the subject of germination physiology of seeds. They often have a chapter or two about seed dormancy, either to demonstrate the diversity of mechanisms among seed plants, or to try and simplify the complexity of dormancy mechanisms by establishing general models. A somewhat different approach is used here. Firstly, the subject is confined to seed dormancy in grasses. Secondly, experimental evidence is considered in depth for a single species, the wild oat (Avena fatua L.), probably the most widely studied species for understanding seed dormancy in the plant kingdom. The evidence for this member of the family Gramineae is compared with other examples among the Gramineae to reach some general conclusions about the nature of seed dormancy in grasses.

There are several reasons for confining the book to grasses. The grass family is one of the largest (25 tribes and 600 genera) and most diverse in the plant kingdom. From a human nutrition perspective it is the most important family. Grasses are the principal plant life form covering more than 70% of the land surface of the globe and they are of critical importance to the stability of the fragile arid and semi–arid zones. While seed dormancy is of great adaptive significance for survival in nearly all plant species with seeds, it is also the main reason why grass species cause the most serious weed problems in cultivated crops around the globe.

The several objectives of this book can be stated in the form of simple questions. What can we conclude about the nature of grass seed dormancy in a single, well–studied, species? Are there commonalities between seed dormancy in this single species and other grass species? Can new conclusions be reached about the nature of the physiological and environmental conditions that establish the state of dormancy in grass seeds? Is it possible to describe new models for seed dormancy that simplify our understanding of grass seed dormancy?

The nature of the first question is in part a semantic problem related to correctly matching the etymological meaning of a word to the reality it attempts to describe. The English word ‘dormancy’, derived from the French dormir (to sleep), itself derived from the Latin dormire (to sleep), is defined in the Concise Oxford Dictionary as ‘lying inactive in sleep’. However, biologists have found that this definition does not encompass observed seed behaviour. Hence the many attempts to divide dormancy into sub–categories that cover the situations in nature where some seeds fail to germinate, whereas others can, in a specific environment (Amen, 1968; Bewley & Black, 1982; Lang et al., 1987).

An adequate description of dormancy must involve at least three major components viz. the seed, the environment and a time element describing changes in state of both the seed and environment. In addition, it is useful to have some measure of incidence of dormancy related to genetic variation in a plant population.

Gould (1968) has divided the family of grasses (Gramineae) into 25 tribes and 177 important genera. Seed dormancy occurs within 18 of the tribes and has been recorded in one or more species in at least 78 of the genera (Table 1.1). Seed dormancy is one of a number of adaptive traits, such as seed size and shape, that are polymorphic in character. Together these characteristics provide diversity and fitness for both opportunistic settlement and enduring occupation of temporally and spatially diverse habitats (Jain & Marshall, 1967). Within each grass species there can be considerable variation in the degree of polymorphism associated with the expression of seed dormancy. The degree of polymorphism is a function of the form of reproduction (self– or cross– pollination) in each species. Some species rely on genetic diversity and others on phenotypic plasticity for adapting to varying environments through the trait of seed dormancy.

Water – liquid and vapour phases

Water has a primary role in sustaining tissue activity and viability of both the parent plant and the developing caryopsis. The natural process of seed maturation involves a dehydration of the caryopsis during the final stages that culminates in abscission. Water stress on the intact plant can interact with temperature and influence both the rate of development and level of dormancy attained by the caryopsis. In its new environment, after abscission, the dormant seed may be exposed to a range of varying environmental factors, particularly moisture and temperature. For example, if the seed remains on the soil surface it will be exposed to diurnal cycles of desiccation and hydration according to the level of radiation from the sun and increased relative humidity at night. Alternatively seeds buried in soil can be exposed for long periods of time to excessive moisture or dryness. Variations in available moisture influence the persistence of dormancy. Excess of water can induce secondary dormancy in many grass species. Some optimum amount of water and suitable temperature will ultimately determine the time for germination. Seed coat structures can modify any of the above moisture variations by limiting ingress of water to particular rates or amounts (Chapter 2).

Interactions of moisture with temperature, controlling dormancy and germination, can occur in several ways. Temperature affects the relative humidity of the atmosphere.

Semantic considerations

The discussion of seed structure and environment in relation to dormancy (Chapters 2 and 3) has shown that genetic variation in dormancy is expressed as a distribution of germinability over time. Seeds of nondormant genotypes germinate at maturity, or even before maturity, in a wide range of environmental conditions. Alternatively, at the time of maturity and abscission from the parent plant, seeds of dormant genotypes cannot be germinated within a wide range of temperatures in the presence of water, oxygen and light. With the passage of time the dormant seed becomes sensitive first to a narrow range of environmental conditions that promote germination. Later the range of each environmental factor, within which germination can occur, broadens until germination is limited only by water and extremes of temperature.

The term ‘after–ripening’ has often been used, loosely, to categorize the collective changes that seeds undergo with time as dormancy is lost. Afterripening has been used as a descriptor for loss of dormancy in a dry, stable, storage environment (Crocker & Barton, 1957), in a variable natural environment such as the soil (Baskin & Baskin, 1981), under conditions suited to optimal germination of non–dormant seeds (Simpson, 1966a), and ‘by undefined biochemical changes occurring in seeds’ (Baskin & Baskin, 1985).

Position of the caryopsis on the parent plant

In the grasses the term ‘seed’ is commonly used to describe the dispersal unit. However in some cases the dispersal unit may be a spikelet, a floret, or a naked caryopsis. In addition the term ‘grain’ is frequently substituted for the word caryopsis, particularly among the cereal grasses. In the context of this chapter on structure the term grain will be used as the general descriptor for the fruit. To some extent the terms seed and grain will be used interchangeably, except where clarity in the separation of anatomical structures is of significance.

Example of Avena fatua:

The inflorescence in Avena fatua is a determinate panicle that matures basipetally but the spikelets mature acropetally (Green & Helgeson, 1957; Raju & Ramaswamy, 1983; Raju, Jones & Ledingham, 1985) (Fig. 2.1). Self–pollination is the rule and the florets are chasmogamous (Raju et al, 1985). Some outcrossing has been observed ranging from one to twelve per cent (Imam & Allard, 1965). Dormancy is present in the grains at an early stage: excised caryopses can germinate as early as three days after anthesis (Morrow & Gealy, 1983) yet by 15 days, dormancy is present and increases up to seed maturation (Thurston, 1957a). The degree of dormancy can vary with position of the spikelet on the panicle (Anghel & Raianu, 1959); grains at the bottom of the panicle are more dormant than those at the top (Schwendiman & Shands, 1943).

The pigeonpea can justifiably be regarded as an under-exploited legume; the reasons why this should be so are not entirely clear. The most likely explanation is that there is an acceptability problem in some parts of the world. Curiously, in the United Kingdom the pigeon pea has been consumed in small quantity, but consistently, as yellow split peas for making pease-pudding. The problem of limited acceptance is one which could at the present time be overcome by judicious stimulation of demand. Present dietary recommendations favour increased consumption of pulses by way of an alternative to red meat as a protein source (avoiding excessive intake of saturated fat) and as a source of dietary fibre. There is also a possible role as a protein source for the manufacture of textured vegetable protein for use in meat substitutes, meat extenders and the like.

The major centre of world production is undoubtedly India, where it is the second most important pulse crop. Production is about 2 million tonnes annually world wide, of which a little less than 85% is produced in India. Potential yield levels in excess of 2 t ha–1 are indicated from trials at ICRISAT (Annual Report, 1982) while in Queensland some breeding lines have indicated yields of over 41 ha–1.

Classification and biosystematics

Taxonomy

The pigeonpea is a member of the sub-tribe Cajaninae of the Phaseoleae and it is the only member of its sub-tribe to have been domesticated. The taxonomy of Cajanus DC.

This chapter concerns legumes of minor current economic significance which have not been covered previously. These include those species which have some food use but which may have other more important uses. In practice it is difficult to draw the line as grain legumes between a crop such as the winged bean, which produces edible pods, seeds and tubers, and one like the yam bean (Pachyrrhizus erosus (L.) Urban), which produces edible pods and tubers but whose mature seeds are toxic (Purseglove, 1974). Guar (Cyamopsis tetragonoloba (L.) Taub.) is another case in point (Hymowitz, 1972). This produces a mucilage-rich seed useful not only for paper-making and textiles but also in food products. The velvet beans (Mucuna spp.) merit at least a brief mention. They are capable of producing prodigious yields of pods, seeds and forage. Their seed can be used as food in times of scarcity if sufficient care is taken during preparation and cooking to eliminate two toxic amino acids (stizolobic and stizolobinic acids) which are present in the seeds.

The crops of major concern in this chapter are the hyacinth bean, the horse gram (formerly included in the genus Dolichos), the Hausa (or Kersting's) groundnut (formerly Kerstingiella) and the sword and jack beans (Canavalia spp.).

The hyacinth bean (Lablab purpureus (L.) Sweet)

Introduction

The hyacinth bean can justifiably be regarded as an under-exploited legume. Herklots (1972) observed that it is useful for soil improvement, as a cover crop and for forage.

Biosystematics is at the present time an unfashionable biological discipline, except perhaps insofar as it now utilises modern computer technology. The once predominant phylogenetic approach is also under attack due to an excessive generation of highly speculative evolutionary hypotheses in the past. This has tended to divert the major thrust of taxonomic research from areas of the greatest practical application, that is those of a phylogenetic and evolutionary nature, to more abstract consideration of taxonomic principles. In spite of this, there has been a welcome kindling of interest in the taxonomic problems of plant groups of economic importance. The applied biologist in the past had probable justification in his suspicion that taxonomists shied away on principle from the study of groups which are of economic importance. The practical value of taxonomy in legume research has however been handsomely, if belatedly, recognised by the publication of work such as that of Polhill and Raven (1981) which provides a good working synthesis of modern ideas on legume taxonomy. The activities of taxonomists can be of particular benefit to plant breeders and those concerned with the collection, conservation and evaluation of germplasm resources. Conversely, the taxonomist benefits from the studies of plant hybridists, cytogeneticists and biochemists in providing additional information for incorporation into the body of taxonomic knowledge. These studies also give taxonomists the opportunity of devising more comprehensive and sophisticated approaches to classification. The proper integration of highly disparate types of information certainly poses problems and challenges.

It is probably true to state that the evolution of cultigens within the genus Phaseolus is as well understood as that of any grain legume species. This is because a great diversity of pertinent evidence has come to light from a range of disciplines in addition to the biological, principally from archaeology and chemistry. At the present time Phaseolus beans are widespread in use both as pulses and green vegetables, particularly the common bean P. vulgaris, and there is considerable interest in their improvement emanating from a wide range of interests. They are of very considerable importance as a subsistence crop in Central and South America as well as in parts of Africa. They are also of commercial interest to the canning and frozen food industries in addition to pulse merchants. Quite obviously the requirements of all these markets are very different. There is therefore the broadest possible interest in the range and extent of the germplasm resources existing in the genus, which can be mobilised to meet the enormous range of actual and potential breeding objectives. The incentives for extensive collection, efficient conservation and evaluation are therefore very considerable and there is ample economic justification for investment in these activities.

One of the most remarkable features of the cultigens of this genus is that they combine very similar basic patterns of morphological divergence from ancestral forms with very marked agro-ecological differentiation.