The aim of this chapter is to help seed collectors understand the reproductive biology of the genus so that the planning, timing and execution of seed collection operations may be undertaken efficiently. With these objectives in mind, details are given on the reproductive processes that occur during seed development such as floral development and sexuality, pollination, breeding system, hybridization, and seed and fruit development. Details are also given on the phenology of flowering and fruiting, seed productivity and predation of the pod and seed.
When conditions are favourable most acacias produce flowers in great profusion. The process of development from the flower to seed maturity is shown in Fig. 7.
The various stages of seed development are outlined as follows:
Detailed studies of floral anatomy and embryology have been made by Newman (1934) for A. baileyana and by Buttrose et al. (1981) for A. pycnantha. Detailed studies of the style and stigma structure and development of several Australian acacias have been made (Kendrick and Knox 1981). By contrast little attention has been given to the total sequence of floral development in acacia. Buttrose et al. (1981) found that in A. pycnantha floral buds were produced continuously on new shoots in every month of the year. Despite prolific production only buds produced between November and May developed through to flowering while those produced in the remaining months (June to October) aborted at an early stage. Each inflorescence of A. pycnantha consists of 40–100 flowers.
There are few reports of the minimum age at which acacias can produce flowers but it appears to be between one and four years. A. mearnsii in plantations in South Africa begins to flower at about age 20 months and, while providing some ripe seed from the third year, it is not until the fifth or sixth year that appreciable quantities of seed are produced (Sherry 1971).
|Fig. 7||Sequence of flowering and fruiting in A. aneura -|
|(A)||Early flower buds||(B)||Flowering stage|
|(C)||Group of pods and phyllodes|
|(D)||Group of pods|
|(E)||Open pod with seeds|
Most acacias have bisexual (male and female parts on the same flower). Many Queensland species however have inflorescences on individual trees consisting of both staminate (male) and bisexual flowers, a phenomenon also noted for A. baileyana by Newman (1934) and for A. nilotica by Sinha (1971). Individual spikes or heads may even consist wholly of male flowers but usually at least a few bisexual flowers occur. Environmental factors during floral development may determine the proportion of male flowers. It is noteworthy that a rudimentary ovary is often found in functionally male flowers and that the ovary does not have a surface covering of hairs, even in species where ovaries normally developed are hairy.
Insects are the major pollinating vectors of acacias, at least of African species (Coetzee 1955). Bees are the main pollinators of A. mearnsii in South Africa with wind possibly playing a very minor role (Sherry 1971). There is also the possibility that birds attracted to the extrafloral nectaries of acacias play a part in pollination (Ford and Forde 1976).
The breeding system of acacias appears to be one of preferential outcrossing. Estimates of natural outcrossing in A. mearnsii range from 67 to 98% (Moffett 1956; Sherry 1971). Moffett's (1956) experiments showed that while seed set per pod was substantially lower after self-pollination the viability of the seed produced was little different to that of outcrossed seed.
The mating of close relatives especially in species that regenerate readily by vegetative means (e.g. A. albida) and the long distances between individuals of some species e.g. in part of the natural range of A. albida and A. tortilis, could contribute to high level of selfing in certain populations.
Most species have a chromosome complement of 2n = 26 (Pedley 1978; Ross 1981) while species belonging to subgenus Acacia have 2n = 52 (Ross 1979). All species studied in subgenus Heterophyllum have 2n = 26 except for six species and of those A. aneura has 2n = 52 and A. sowdenii has 2n = 38 (Pedley 1978). For further details the reader may consult Atchison (1948) and Guinet and Vassal (1978).
Some natural interspecific hybrids are known to occur, but the full extent of hybridization and introgression in acacias is unknown.
The first clue that an individual might be a hybrid is usually the recognition of morphological features which are intermediate in some degree between those of the two species which are supposed to be its parents. Elamin (1976) showed that A. laeta had morphological characters intermediate between A. mellifera and A. senegal and confirmed the likelihood of its hybrid origin. Very vigorous intermediates between A. auriculiformis and A. mangium occur occasionally in plantations of the latter in Sabah, Malaysia.
Ali and Faruqi (1969) and Ali and Qaiser (1980) attributed the variability in the A. nilotica complex in Pakistan to hybridization between A. nilotica subsp. indica and A. nilotica subsp. hemispherica. These workers found that the hybrid populations, promoted by man's influence in distributing seeds and creating disturbed habitats, may backcross with A. nilotica subsp. indica and A. nilotica subsp. hemispherica, producing plants similar to A. nilotica subsp. adstringens and A. nilotica subsp. subalata. The probability of hybridization between A. nilotica subsp. indica and A. nilotica subsp. cupressiformis was also suggested.
After pollination, the ovary matures within the flower to form the pod characteristic of the legumes. Considering the number of flowers produced there are few pods per inflorescence (2 or 3 per 1000). In acacias the ovules are fixed in shallow sockets along the centre of the growing pod and connected to the ovary by the funicle. The pod reaches full size before the ovules enlarge into seeds. The pods are at first green and fleshy but with time the fibres lignify and harden. At maturity pods usually dehisce longitudinally along both margins. The seed may not all be shed immediately. Some may remain attached by the funicle until the pod itself is shed.
Acacia seeds are extremely variable in size, shape and weight and, even within a species, marked differences can occur (Cavanagh 1980a). The seedcoat of acacia is relatively thick in comparison with that of many of the small-seeded legumes; in one study it accounted for 33–43% of the total seed mass (Murray et al. 1978). The anatomy of the seedcoat of acacias has been reviewed by Cavanagh (1980a) and a more general account of the features of legume seeds has been written by Gunn (1981).
Pedley (1978) states that the majority of Queensland species of acacia flower at the same time each year (often for less than six weeks) in the driest part of the year, June to September - regardless of weather conditions prior to flowering. This is possibly a photo-periodic response. In many arid-zone species, however, it seems that flowering depends on moisture supply, and may occur in any season (Mott 1979).
Davies (1976) reported long-term phenological observations on perennial shrubs including several acacias in an arid area of Australia. Davies found that although A. aneura can flower after rainfall in any season, only those flowers resulting from summer-rainfall produced fruit, and that a significant quantity of mature fruit resulted only when good rains occurred in the following winter. Preece (1971a) tested the effects of irrigation on individuals in a population of A. aneura and, although the results were inconclusive, they were consistent with those of Davies. Using the criterion that summer rainfall followed by winter falls was necessary for successful reproduction, Preece analysed approximately 80 years of rainfall data for several sites and found that only in 10–15% of these years were climatic conditions favourable for seeding to take place. A summary of the flowering and fruiting times of A. aneura and some seed details are given in Table 1. The phenology of flowering and fruiting of A. albida, A. caven, A. nilotica and A. tortilis is variable depending on species. Reported times by species and country with some details of seed are given in Table 1.
There are numerous ways of expressing seed production in acacias, e.g. seeds per pod, seeds per tree, or seeds produced per unit of land area. Table 1 provides some details for the number of seeds per pod for several species e.g. A. albida in South Africa produces 14–21 seeds per pod, while A. senegal in Pakistan produces 2–5 seeds per pod.
There are few references to the number of seeds produced per plant and at particular plant ages. Monk et al. (1981) for A. pulchella (a small Australian shrub) found that seed production commenced at year 2, reached a maximum of 12 000 seeds per plant per year at year 3 or 4 and declined to 2000 seeds per plant in the 13th year. A mature tree of A. albida may yield about half a million seeds in Sudan and in South Africa a large tree in a good season could produce several million seeds (Wickens 1969).
The number of seeds produced per hectare depends on the density, age and size of trees. Yields of 2–12 kg (approx. 150 000 to 850 000 seeds) of seed per hectare have been reported for A. aneura (see Burrows 1973).
Table 1 Flowering and seed collection times by species and countries for six acacias, together with seed crop and germination information
|Species||Country and reference||Months of flowering||Seed collection months||Duration of seed crop||Periodicity of good crops||No. of seeds/ pod||No. of seeds/ kg||Germination|
|FAO 1980||April–June||Production limited|
|FAO 1974a, 1975||20 400||70–100|
|Maydell 1978||Seed is easily available|
|Giffard 1964||About 2 months after end of rainy season||Pods fall 3 months after flowering||11 500||Seed has waxy impermeable cuticle; seed viability retained for several years|
|Giffard 1971||March–May||11 500|
|Carr 1976||Throughout winter||Good seed in quantity usual||14–21||Up to 95|
|Radwanski and Wickens 1967||Beginning of dry season|
|Wickens 1969||21 (based on 200 pods)||Trees have yielded 135 kg of pods per annum, 95% of pods and seeds damaged by insects. Up to 50% destroyed in pods by bruchid beetles.|
|FAO 1975||40 000||0–30|
|West 1950||Start of dry season (July or earlier)||Aug. onwards|
|FAO 1974b||April (Senegal)||11 000||Seed keeps well if free from insect attack.|
|NAS 1979||125–135 kg of pods on a single tree have been recorded. Yields of 400–600 kg of pods/ha usual in in Sahelian region.|
| CSIRO unpub.|
Hall et al. 1979
|Summer and winter; only summer flowers produce mature fruit; quantity determined by winter rain||Oct.–Nov.||Oct.–March (most shed early)||Variable, depends on rainfall patterns; less than year in 10 reported||50 000– 100 000||Yields of 2–11,45 kg of seed/ha have been recorded|
|Anon. 1973||Aug. onwards|
|H.R. Mangieri pers. comm.||Oct.–Nov.||Feb.–March||Several weeks||Variable, depending on climatic factors|
|FAO 1975||10 000||30–70|
|FAO 1980||Dec. onwards||10 000|
|Instituto Forestal pers. comm.||March||8 530||83|
|Flinta 1960||10 000||Regenerates abundantly by natural seeding|
|Goor and Barney 1976||10 000||Seeds germinate easily|
|FAO 1980||April–June||Abundant||7 000– 11 000|
|Good seed readily obtainable||Up to 15||0–60|
|Palmer and Pitman 1972||Oct.||March–April||10–15|
|FAO 1980||March–June||8 000 (subsp. tomentosa)|
|Khan 1970||July–Jan. (peak Sept. – Oct.)||Jan–May (peak March–April)||Temperature affects flowering and fruiting while seeding is controlled by evaporation|
|8 000||Bruchid beetles can seriously attack seed in pods|
|FAO 1980||Feb.–March||7 000|
|FAO 1975||10 200||70–100|
|FAO 1975||33 000||70–100|
|Cheema and Qadir 1973||Aug.–Dec.||Aug. onwards||2–5||16.7 to 82.5% of seed destroyed by buffalo tree hopper|
|Giffard 1975||Dec.–Feb.||Several weeks||3–8||10 000– 19 500||Dust seed with insecticide immediately|
|Ross 1968||Several, peak spring and early summer||Bruchids destroy vast quantities of seed|
|Carr 1976||Dec.–Jan./April||Oct.||var. rostrata may persist throughout winter||Profuse||1–6||Undamaged seed usually readily available|
|Sief el Din and Obeid 1971||Every 2–3 yr||84% of seed destroyed on ground before germination|
|Abundant every year||7000– 8000|
|FAO 1975||16 100||30–100|
|Pathak et al. 1980||45–49||Seed polymorphism in treatment requirement noted|
|FAO 1980||June–July||Seeds copiously||50 000||40% of seed affected by bruchid larvae. Fumigation prior to storage essential; use ‘Phostoxin’|
|Karschon 1961||June–Jan.||July||Moisture content of fresh seed is 11–15%. Immature seed from green pods germinates without pre-treatment|
|Carr 1976||Jan.–Feb.||Up to 14||Viability varies|
|FAO 1980||March (subsp. raddiana)|
April–July (subsp. spirocarpa and tortilis)
|NAS 1980||Seed production is often severely reduced by insects|
Despite the fact that only a very small fraction of flowers develop into fruit the potential seed production of the many profusely-flowering acacias is enormous. Conversely the potential for the destruction of seed is also very high. Ross (1979) found that insects, failure of fertilization and other factors accounted for 50% (43 415) of the potential crop of a specimen of A. caffra in South Africa. Some of the many insect species whose larvae feed upon the buds, flowers, pods and seed of African acacias are listed by Palmer and Pitman (1972).
Bruchids (Bruchidae, ‘seed weevils’) cause very considerable damage to seed of African and Central American acacias. Reports in the literature of the extent of seed death attributed to bruchids range between 20 and 100% (some figures are given in Table 1). Ross (1979) described the sequence of attack and the end result -typically the adult bruchid lays eggs in or on the green pod in the spring after the flowers drop and the larvae hatch some weeks later and penetrate the young seed. The larvae grow and pupate inside the seed which ripen in the meantime. The adult bruchids emerge from the seed by cutting a round hole through the seed coat. Shortly after emerging some adults may lay eggs which provide a basis for further generations of bruchids in the dry seeds during ensuing months. Seed infested by bruchids remains viable only if the embryo is undamaged and if only a relatively small amount of the cotyledons is eaten. In the small fraction of infested seed that remains viable, the bruchid exit holes greatly increase the permeability of the seedcoat and thus facilitate more rapid germination.
Because of the destruction of a large fraction of the seed on or beneath the tree by succeeding generations of bruchids, rapid dispersal of seed is possibly critical to the survival of some acacia species (Ross 1979). Certain indehiscent species (e.g. A. albida, A. nilotica and A. tortilis) have achieved this by their adaptation for seed dispersal by large mammals. The shape, size and nutritional content of their pods are attractive to large mammals. By this means the hard-coated, rounded seeds escape some of the bruchid predators and are eventually passed out relatively undamaged in the animal faeces.
Not all species of acacia are affected to the same degree by bruchids and some species appear to be unacceptable to bruchid larvae. Bruchids are not significant parasites of acacia in Australia. Seeds in Australia are attacked by other insects e.g. a chalcid wasp infests A. aneura (Preece 1971b).
Birds can also cause significant losses. Fruits of A. aneura are frequently eaten by parrots before they are fully ripe. This can result in the loss of the whole crop from individual trees (Davies 1976). The flowers and fruit of A. aneura are also an important source of food for emus. A variety of other birds also eat acacia seeds.