Chapter 7 SEED STORAGE

Introduction

Storage may be defined as the preservation of viable seeds from the time of collection until they are required for sowing (Holmes and Buszewicz 1958). When seed for afforestation can be sown immediately after collection, no storage is needed. The best sowing date for a given species being raised in a nursery depends on (a) The anticipated date of planting, itself dependent on seasonal climate, (b) The time needed in the nursery for planting stock of that species to reach the right size for out-planting. Only rarely does best sowing date coincide with the best date for seed collection. More often it is necessary to store the seed for varying periods which may be

1. Up to one year when both seed production and afforestation are regular annual events, but it is necessary to await the best season for sowing.

2. 1 – 5 years or more when a species bears an abundant seed crop at intervals of several years and enough seed must be collected in a good year to cover annual afforestation needs in intermediate years of poor seed production.

3. Long-term storage for purposes of conserving genetic resources. The period of storage will vary according to the seed longevity of the species and the storage conditions, but will be measured in decades in species which are easy to store.

The storage facilities to be provided must be related to the amount of seeds and the period over which they are to be stored. It is a waste of money to create expensive facilities capable of maintaining viability for 10 years if the seeds will never stay longer than nine months between collection and sowing. It is equally wasteful to spend money on seed collection, extraction and cleaning if storage conditions are so inadequate that the seeds are 90 % dead before they reach the nursery.

For general accounts of seed storage of forest tree seeds, there are a number of useful references (Holmes and Buszewicz 1958, Magini 1962, Stein et al. 1974, Wang 1974, Barner 1975b). They deal mainly or exclusively with temperate species. More intensive studies have been made of the storage of agricultural seeds and there is good reason to accept that the general principles established for agricultural crops apply also to forest trees. An excellent recent account of the subject as applied to agricultural seeds is contained in “Principles and practices of seed storage” (Justice & Bass 1979) and there is also much useful information in the slightly older publications of Roberts (1972) and Harrington (1970, 1972, 1973). Long-term storage for gene resource conservation is well covered by Cromarty et al. (1982).

Natural Longevity of Tree Seeds

The period for which seed can remain viable without germinating is greatly affected by its quality at the time of collection, its treatment between collection and storage and the conditions in which it is stored. Nevertheless, seed longevity varies enormously from species to species even if they are given identical treatment and storage conditions. Ewart (1908) divided seeds into three biological classes according to the time for which they are capable of retaining viability under “good” storage conditions:

Microbiotic: seed life span not exceeding 3 years

Mesobiotic: seed life span from 3 to 15 years

Macrobiotic: seed life span from 15 to over 100 years.

Although Ewart's classes were useful in drawing attention to the differences in natural longevity of seeds of different species, his classification is too rigid to fit the variations between individuals, provenances and seed years in a single species, or the possible variations in storage conditions. It is not possible to define a standard set of “good” storage conditions equally suitable to all species, because species vary in their requirements for optimum conditions. Yet storage life of a given species will vary greatly according to the conditions in which it is stored.

Today two major classes of seed are recognised (Roberts 1973):

1. Orthodox. Seeds which can be dried down to a low MC of around 5% (wet basis) and successfully stored at low or sub-freezing temperatures for long periods.

2. Recalcitrant. Seeds which cannot survive drying below a relatively high moisture content (often in the range 20–50% wet basis) and which cannot be successfully stored for long periods.

Within these two classes some further subdivision may be made, for example between orthodox seeds with or without hard coats and between recalcitrant seeds which can or cannot withstand low temperatures of below around 10°C. Within each of the main classes there are still considerable differences between species in the period for which viability is maintained under a given set of conditions. There may also be a distinction between truly recalcitrant species and species that are just difficult; the latter may turn out to behave in an orthodox manner if, for example, particular attention is paid to the methods of drying them.

Hard-coated orthodox seeds

Most, if not all, of the species which have been recorded as maintaining seed viability over a period of decades are hard-seeded. They include a number of tropical leguminous species. Examples of species of which at least some seeds maintained viability after lengthy periods of storage in herbaria, cited by Harrington (1970) from the work of Ewart (1908) and Becquerel (1934), are:

Ambient conditions in herbaria storage can be considered good (fairly low relative humidity and temperature) but well short of the combination of low initial MC, sealed storage and sub-freezing temperature now considered ideal for longterm storage of orthodox species.

Recent research has provided more precise information on conditions of storage, initial germination and final germination of some species, but over shorter periods. Examples are:

Sources: 1) Ffolliot and Thames 1983.
2) Doran et al. 1983.

As explained later in this chapter, modern thinking has defined low MC, low temperature and low oxygen pressure as the three most important constituents of the storage conditions which man should provide to maximize seed longevity in orthodox species. In the impermeable seedcoat nature has provided two of these constituents, a low MC and exclusion of oxygen. Full-sized but green leguminous seeds, sown immediately without drying, may germinate at once, indicating that the seedcoat has not yet developed an impermeable layer; no doubt the development of impermeability is synchronized in nature with the reduction of seed moisture by natural drying to the optimum content for longevity. Hard seeds are thus a potent factor in extending seed life in all conditions of storage but confer their most important benefits when storage facilities are limited and during the potentially dangerous period between collection and entry into long-term storage.

Not all leguminous seeds are equally long-lived, for example Koompassia malaccensis seeds have thinner seedcoats and deteriorate more rapidly in storage than species such as Parkia javanica, and they need no pretreatment to overcome seedcoat dormancy (Sasaki 1980 a). In Sudan seeds of Dalbergia sissoo stored less well at room temperature than those of local Acacia, Albizzia and Tamarindus species (Wunder 1966), while in Australia seeds of Acacia harpophylla deteriorate rapidly unless stored in sealed containers at 2–4 °C (Turnbull 1983).

Orthodox seeds without hard seedcoats

Many species in important genera of forest trees fall into this group, e.g. Pinus, Picea, Eucalyptus. Experience in Australia is that mature seeds of all eucalypts can be kept viable for some years if stored with a low moisture content in sealed containers at 3° – 5° C. The majority of species can be stored for 10 years at room temperature with relatively little loss of viability (Turnbull 1975 f). Both E. deglupta and E. microtheca seeds deteriorate more rapidly if stored at room temperature, but storage life is improved if they are kept in air-tight containers at 3° – 5° C and recent evidence suggests that storage at -18°C is even better. In Thailand seeds of P. kesiya and P. merkusii retained good viability for four years if stored at below 8 % moisture content in sealed containers at 0° – 5° C (Bryndum 1975), while at least five years' good viability is possible with P. caribaea and P. oocarpa under similar conditions (Robbins 1983, a,b). Considerably longer periods have been recorded for some species of pine e.g. 30 years for Pinus resinosa in USA when stored in sealed containers at 1.1° – 2.2° C (Heit 1967b, Wang 1974). Tectona grandis is an orthodox tropical broadleaved species (Barner 1975b) but, since it produces good seed crops in most years, there has been little stimulus to investigate optimum conditions for long-term storage (Schubert 1974).

According to evidence summarized by Bowen and Whitmore (1980), most Agathis spp. are orthodox. For example, an appropriate treatment of A. australis in one study (dried to 6 % MC, then stored in sealed containers at 5° C) preserved viability for 6 years (79 % germination compared with the initial 88 %), while storage at below freezing temperature maintained a germination of about 60 % for up to 12 years (Preest 1979). The same seed stored at higher MC or temperatures (15 – 20 % MC. or 15° – 20° C) had lost all germination power within 14 months. A australis seed is inherently longer lived than A. robusta which in turn is longer lived than A. macrophylla. Initial trials with the tropical A. macrophylla indicated that good results could be obtained by drying fresh seeds from about 65 % to 20 % before despatch by air (period in transit 14 days) and by further drying in the recipient country for 5 days at 16° C and 14 % RH. final MC was 6 % and germination 75 %. However, later trials were inconsistent and less successful. With tropical species, it is likely that handling between collection and despatch and the largely uncontrollable conditions of air transit are more critical than in the easier temperate or subtropical species.

Orthodox species which rapidly lose viability unless they are given the optimum treatment include species in the mainly temperate genera Populus, Salix and Ulmus. Many of these lose viability within a few weeks under natural conditions or if stored in ambient conditions of temperature and humidity, but can be stored for months or years if maintained at low temperature and low moisture content. Examples are of Ulmus americana stored successfully for 15 years at 3 % moisture content and -4° C (Barton 1961), and Populus sieboldii stored for 6 years at -15° C over a desiccating agent in a sealed container (Sato 1949). In Populus balsamifera and Salix glauca, reduction in germination after two years of sealed storage at -10° C was less than 6.5 % of initial germination (Zasada and Densmore 1980); after three years there was very little change in Populus but up to 40 % reduction in Salix.

In the tropics Aucoumea klaineana is a good example of an orthodox species which is short-lived under ambient conditions. Germination of fresh seed is often over 90 %, but after 30 days' storage in room conditions there is a significant drop in germination and this falls to zero after 100 days. Storage at 0 – 5° C and 7 – 8 % MC in sealed containers with a chemical desiccator, Actigel, maintains over 50 % germination for at least 30 months (Deval 1976). There is some indication that a further reduction of MC will conserve viability even better. Thus one seed lot of initial germination 76 % in the laboratory and 79 % in sand, when stored in sealed containers with Actigel, had an MC of 4.6 % and a germination of 70 % in the laboratory and 79 % in sand after 30 months; the same seed lot stored in sealed containers without Actigel had an MC of 9.9 – 10.4 % and a germination of 54 – 63 % in the laboratory and 62 – 67 % in sand. Other species of this type include Entandrophragma angolense which has a seed life of 6 weeks in room conditions but up to 6 years in cold storage (Olatoye 1968) and Cedrela odorata which loses all germination capacity in 10 months at room temperature but suffers no loss in 14 months if stored at 5° C in a sealed jar (Lamprecht 1956).

Some species may need special treatment to prolong viability for more than a few months. Fagus sylvatica can be conserved overwinter by maintaining MC at 20 – 30 % and storing part-filled in sealed polythene bags at 0 – 5°C for 100 days. It is then suitable for sowing because such storage conditions constitute a suitable pretreatment to break dormancy. If longer storage is intended, MC should be reduced to 8–10 % by drying in a current of air at room temperature (15–20°C). The nuts are then placed in sealed containers and stored at -5° to -10° C and will keep for several years (Nyholm 1960, Suszka 1974, Rudolf and Leak 1974). Later research in France and Poland confirmed the above MC of 8–10 % and the advantages of sealed containers for long term storage (Bonnet-Masimbert and Muller 1975, Suszka and Kluczynska 1980). This technique has been successfully applied on a large scale (17 tons of beechnuts from 51 different sources) in France. Germination has been maintained over periods of 4 to 6 years (Muller and Bonnet-Masimbert 1982).

Where storage conditions leave much to be desired, the longevity of orthodox seeds without hard coats can be expected to be much inferior to the hard-coated species. The nearer that conditions of storage approach the ideal for a given non-hardseeded species, the less the difference between its longevity and that of a hardseeded species. The best combination of MC and temperature will vary to some extent between species, for example the above-quoted 8–10 % MC for Fagus sylvatica is considerably higher than the 5–6 % considered ideal for many forest and agricultural seeds.

Recalcitrant seeds

Recalcitrant seeds include a number of large seeds that cannot withstand appreciable drying without injury; it is of interest that the overwhelming majority of recalcitrant species listed by King and Roberts (1979) are woody. Temperate species such as Quercus and Castanea are commonly stored moist only for short periods over winter. Reduction of storage temperature to near freezing will prolong longevity. Bonner (1973 a) found that it was possible to store acorns of Quercus falcata for 30 months and still obtain over 90 % germination at the end of the period, provided that temperature was maintained at 3 ° C and MC between 33 % (initial) and 37 % (final). A lower MC or a higher temperature (8° C) both reduced germination. For Quercus robur MC should be maintained above 40 % (Holmes and Buszewicz 1956, Suszka and Tylkowski 1980). Recent research in Poland has demonstrated good results from storing this species at >40 % MC in air-dry peat or air-dry sawdust in milk cans at -1° C. It is important to allow free entry of oxygen and this is ensured by inserting several strips of cardboard at intervals between the lid and the edge of the can. In these conditions germination after 3 winters was in the range of 38 – 75 % and after 5 winters was still about 12 % (Suszka and Tylkowski 1980). Temperatures below -5° C killed all the acorns, while a temperature of +1° C encouraged excessive pregermination (60 – 75 % after 3 winters, with radicles up to 25 cm long, compared with 12 % and radicles <0.5 cm long at -1°C). There may be possibilities of storing seeds after emergence of radicles (see p. 152). Recent research in Poland (Suszka and Tylkowski 1982) has indicated that best results are obtained with the recalcitrant Acer saccharinum by maintaining MC at the same percentage (50–52%) as when the seeds were freshly collected. For A. pseudoplatanus in the UK a minimum MC of 35 % is recommended (Gordon and Rowe 1982), while in Poland an MC of 24–32% and a temperature of -3°C have proved suitable to store samaras over three winters (Suszka 1978a).

Most short-lived recalcitrant tropical species are constituents of the moist tropical forests, where conditions conducive to immediate germination (high humidity and high temperature) are prevalent throughout the year. Typical genera are Hevea, Swietenia, Terminalia and Triplochiton, as well as a number of Dipterocarp genera such as Dryabalanops, Dipterocarpus and Shorea and some species of Araucaria. Dryabalanops is injured if dried below 35 % moisture content (MC) but still survives only about three weeks at over 35 % MC (King and Roberts 1979). Triplochiton seed is naturally short-lived but can be stored for up to 22 months at a temperature of around 6° C and a moisture content of between 12 and 25 % (Bowen and Jones 1975). Azadirachta indica seeds also have a short period of viability, although the species occurs in dry, not moist, tropical forests and it is not clear whether it is a genuine recalcitrant or simply a short-lived orthodox species.

Orthodox and recalcitrant species sometimes occur within the same genus. In Acer and Ulmus, genera in which both orthodox and recalcitrant seed behaviour occur, the distinction in North American species is clearly between spring- and fallseeders. A. rubrum and A. saccharinum flower and seed in the spring. Their seeds are not dormant, and their storage behaviour is clearly recalcitrant. Other Acer species have fall-maturing seeds, which are dormant and orthodox in nature at maturity. The same occurs in Ulmus. Seeds of U. crassifolia and U. serotina mature in the fall, and are orthodox in storage behaviour. Spring-seeding species of Ulmus are “weakly” recalcitrant (Bonner 1984b). In Araucaria, A. cunninghamii and other spp. in the Eutacta taxonomic group behave as orthodox. In Queensland seeds of A. cunninghamii of 5 provenances were air dried and stored at varying temperatures in sealed and unsealed containers. At the higher temperatures, +1.7°C and -3.9°C, germination started to drop after 17 months' storage and after 8 years was down to about half the initial germination rate in sealed containers and about one third in unsealed containers. At the lower temperatures of -9.4°C and -15°C, germination after 8 years' storage was little changed from the initial (41–44% compared with initial 49%) (Shea and Armstrong 1978), and there was virtually no difference between sealed and unsealed containers. The rate of viability loss at the higher storage temperatures varied from provenance to provenance, but all stored better at the lower temperatures. Moisture content was not recorded but under local conditions air-dry seed is normally in the range of 16–23% (Kleinschmidt 1980, cited in Tompsett 1982). Later trials with Papua New Guinea A. cunninghamii have shown that seeds can be dried from 21% to 7% MC without any effect on initial germination rate; effects on storage life are still under investigation (Tompsett 1982). A. hunsteinii in the Intermedia group and A. angustifolia, A. araucana and A. bidwillii in the Colymbea group are apparently recalcitrant. Arentz (1980) found that high viability of A. hunsteinii could be maintained for at least 6 months by storage at 3.5°C and high MC; 37% was significantly better than 32%. Research reported by Tompsett (1982) confirmed that MC should be maintained above 32%. Placing the seed in one polythene bag of 25 microns thickness inside a second bag is effective in maintaining viability. The double thickness of polythene maintains a high MC but allows for some exchange of oxygen which is necessary to preserve viability of A. hunsteinii. A. angustifolia also needs a high MC; seeds died if dried to less than 25–30% (Tompsett, in press).

For some temperate recalcitrant species, as indicated above, a relatively low temperature (just above or just below 0°C) has been found beneficial in extending the life of the seeds; low temperature to some extent compensates for the high MC which must be maintained to prevent the early loss of viability. In some tropical species, seeds are quickly killed if temperature is reduced too low, just as they are quickly killed if MC is reduced too low. Among woody species cited in King and Roberts (1979) are Theobroma cacao (killed below +10°C), Mangifera indica (damaged below +3° to +6°C) and, among the dipterocarps, Hopea helferi, Hopea odorata and Shorea ovalis (damaged below, respectively, +5°C, +10°C and +15°C). This susceptibility to chilling damage at temperatures above 0°C compounds the difficulty of storing these recalcitrant species, which seldom maintain viability for more than a few weeks or at most months. This compares with a normal seeding periodicity of several years in most dipterocarps, so there is no possibility as yet of conserving seeds in a viable condition from one good seed year to the next.

Unlike orthodox species, in which viability is best preserved by maintaining a minimal respiration rate, it appears that active respiration is necessary to survival of seeds of most recalcitrants. Thus damage to recalcitrant seeds has been reported not only from inadequate MC and too low a temperature but also from lack of oxygen e.g. in Araucaria hunsteinii (Tompsett 1983), Hevea brasiliensis and Quercus spp. (cited in King and Roberts 1979).

Whereas some temperate recalcitrant species have been stored successfully for several years, seed longevity in tropical recalcitrants can be measured in days or weeks. The amount of research on tropical species is still small, especially on forest species, and it is possible that seed longevity could be prolonged beyond a few weeks if the best combination of seed maturity, speed, conditions and degree of drying, and most suitable storage temperature could be determined for each species. King and Roberts (1979) suggest a research strategy.

Factors Affecting Longevity in Storage

Seed condition

Even in ideal storage conditions seed will soon lose viability if it is defective from the start. Factors to be considered are:

Seed maturity. Fully ripened seeds retain viability longer than seeds collected when immature (Stein et al. 1974, Harrington 1970). Certain biochemical compounds, essential for preserving viability, may not be formed until the final stages of seed ripening. These include dormancy-inducing compounds in certain species, and dormancy is sometimes associated with seed longevity. In a few species e.g. Gingko biloba, Fraxinus excelsior, seed embryos are underdeveloped when the seed is shed. Maturation of these embryos is necessary before sowing but need not be done before storage. In Fraxinus excelsior drying of freshly collected samaras to 9–10% MC, followed by storage in sealed containers at -3° C, gives satisfactory results provided that successive moist warm and moist cold treatments are applied after storage (Suszka 1978a). For details of treatments see p. 184.

Parental and annual effects. In seed harvest, quantity and quality often go together. The percentage of sound seeds in a high-yielding mother tree is usually higher than in one with a poor crop. Similarly, a given mother tree will have a higher percentage of sound seeds in a good crop year than in a poor one. Collection from high-cropping mother trees in a seed year is likely to yield seeds with the best longevity in storage. On the other hand, high-cropping “wolf” trees should be avoided because of their potentially undesirable wood properties, even though they may produce long-storing seeds.

Freedom from mechanical damage. Seeds damaged mechanically in extraction, cleaning, dewinging etc. rapidly lose viability. The danger is greatest for species which have thin or soft seedcoats. Excessive heat during extraction or drying also damages seed. Care should be taken to use the minimum times, lowest temperatures and minimum machine speeds necessary during the preparation of seed for storage (Stein et al. 1974). In some species, damage during dewinging may be reduced by partly restoring the moisture content between extraction from cones and dewinging, since moist seeds suffer less mechanical damage than dry ones (Nilsson 1963, Barner 1975b).

Freedom from physiological deterioration. Poor handling in the forest, during transit or during processing causes physiological deterioration of seeds even if mechanical and fungal damage are absent. Adequate ventilation of orthodox seeds is necessary to avoid rapid respiration and overheating, while recalcitrant seeds must be protected against excessive drying.

Freedom from fungi and insects. For species stored at low temperature and low moisture content, the storage conditions themselves should prevent the development of fungi and insects. It is necessary, however, to avoid collection of crops showing a high incidence of fungal or insect attack and to carry out all operations of collection, transport, processing etc. as quickly as possible to ensure seed is not already damaged before it goes into storage. Attack by fungi and insects is most rapid on the forest floor, so collection from the ground should be carried out as soon after fruit fall as possible. Fungicidal treatment cannot be generally recommended since it can be harmful to seeds (Magini 1962); many fungicides are only effective when dissolved in water and are inappropriate for dry storage. Insects are usually killed if seeds are dried at temperatures above 40° – 42° C. For seeds which cannot be dried, other measures may be needed. For example seeds of Quercus are fumigated with serafume or other chemicals or heated in warm water for control of weevils (Belcher 1966, Olson 1957), while methyl bromide or carbon bisulphide are also commonly used to kill insects (Boland et al. 1980).

Initial viability. Seed lots with high initial viability and germinative capacity have a higher longevity in storage than those with low initial viability. Germination tests, preceded if necessary by appropriate pretreatment to overcome dormancy, should be carried out on a sample of each seed lot before storage, in order to determine how long the seed is likely to retain viability in storage. Longevity of the viable seeds is correlated with the percentage which germinate in the initial test. As an example, samples of two seed lots of the same species, from which 80 % germination of fresh seeds is normally expected, might give results of 90 % and 50 % initial germination. Not only would storage of the second seed lot involve wasting space in storing dead seed, but even the 50 % of initially viable seeds are likely to lose their viability more quickly in storage than would the 90 % viable seeds in the first seed lot. Deterioration in initial viability may not be serious if the seeds are to be sown within a few weeks or months, but only good quality seed should be stored for long periods (Holmes and Buszewicz 1958, Magini 1962). In long-term storage of agricultural seeds for genetic conservation, it is recommended that no seeds should be accepted for storage which have an initial viability of less than 85 % of that considered typical for the species or variety in question (IBPGR 1976). It may be noted that initial viability and germinative capacity are frequently the resultant of the factors described in previous paragraphs (seed maturity, mechanical damage, fungal or insect attack).

Storage conditions and ageing of seeds

In common with all other living things, seeds are subject to ageing and, eventually, to death. In the case of orthodox seeds, the process of ageing and deterioration is so greatly affected by the conditions of storage that the “age” of seeds, expressed solely in terms of the period elapsed since ripening and harvesting, is an inadequate measure of the degree to which they have “aged” in the sense of losing viability and progressing towards the irreversible deterioration of death. The term “physiological age” is commonly used to describe the degree of deterioration of seeds measured by their reduced capacity for germination. Nomographs for the effects of temperature and MC on physiological ageing of seeds have been constructed for several agricultural crops (Ellis and Roberts 1981). As an example, the nomograph for barley indicates that the same degree of deterioration (from initial 95 % to final 50 % germination) would occur in about 16 days in seeds stored at 25° C and 21 % MC, but in about 100 years in seeds stored at 8° C and 8 % MC. Both seed lots would have the identical physiological age, though stored for very different periods of time. Similar effects can be expected in orthodox seeds of forest trees.

A number of physiological changes in cell tissues may be associated with physiological ageing in seeds. They include (1) Loss of food reserves caused by respiration, e.g. decrease in proteins and non-reducing sugars, accompanied by increase in reducing sugars and free fatty acids (2) Accumulation of toxic or growth-inhibiting by-products of respiration (3) Loss of activity of enzyme systems (4) Loss of ability of dried proteid molecules to recombine to form active protoplasmic molecules on subsequent rehydration (5) Deterioration of semi-permeable cell membranes (6) Lipid peroxidation, leading to production of free radicals which react with, and damage, other components in the cell (7) Alterations to DNA in cell nucleus, causing genetic mutations as well as physiological damage (Roberts 1972, Harrington 1973, Villiers 1973). It is still uncertain to what extent these various effects are the causes or only the symptoms of deterioration, but it has been suggested (Villiers 1973) that the production of free radicals is the first effect of ageing and that damage to the several systems in the cells is the subsequent result of the release of free radicals.

Whatever the exact mechanism of seed deterioration, there is a consensus that, in orthodox seeds, loss of seed viability is largely governed by the rate of respiration. Any measures which reduce the rate of respiration without otherwise damaging the seed are likely to be effective in extending longevity during storage. These are the control of oxygen, the control of moisture content and the control of temperature. In recalcitrant seeds the safe minimum levels of oxygen, moisture content and temperature, and hence of respiration, are all considerably higher than those for orthodox seeds but, provided levels are maintained above the safe minima for each species, it appears that longevity can be extended by keeping them as close to the minima as possible in order to avoid an excessively high respiration rate.

Storage atmosphere

The most obvious method of reducing the rate of aerobic respiration is to exclude oxygen from the atmosphere surrounding the seeds. This can be done by replacing oxygen by other gases such as CO₂ or nitrogen, or by using a partial or complete vacuum. In an example with lettuce cited by Roberts (1972), seeds were stored in sealed containers at 6 % MC and 18° C. After 3 years, seed stored in an atmosphere of pure oxygen had 8 % viability, those in air 57 %, those in nitrogen or argon or CO₂ 78 % and those in a vacuum 77 %. The value of excluding oxygen during storage of dry orthodox seeds has also been demonstrated in Pinus radiata (Shrestha, Shepherd and Turnbull 1984). Best results were obtained with a storage atmosphere of nitrogen, followed by CO₂, while vacuum and air both gave poorer results. At the highest temperature used, 35°C, at which deterioration in viability was most rapid, the loss in final germination after 50 weeks' storage in sealed containers at 8% MC was 8% in nitrogen, 14% in CO₂, 21% in vacuum and 29% in air. The same ranking was obtained by comparing the speed of germination and the vigour of germinated seedlings (measured as dry weight 49 days after sowing). Although increases in seed longevity of this magnitude have been achieved experimentally, some of the methods are expensive to apply and the effects on seed life are less dramatic than the effects of differences in temperature and humidity (Goldbach 1979). Exclusion of oxygen will prevent aerobic, but not anaerobic, respiration, whereas reduced MC and temperature will decrease the level of both. While systematic predictions have been made of seed longevity under a range of temperature and MC for several agricultural crops (Ellis and Roberts 1981), similar quantitative predictions of the effect of oxygen levels on longevity are lacking.

One simple method which is recommended is to fill sealed containers as nearly full as possible. If there is only a small amount of air inside the container as compared with the volume occupied by the seeds, oxygen will be consumed and CO₂ produced. The resulting high CO₂/O₂ ratio is probably favourable for seed longevity in orthodox seeds (Goldbach 1979).

Whereas the complete exclusion of oxygen from the storage atmosphere appears beneficial to most dry orthodox seeds, there is evidence that some oxygen is necessary for recalcitrant seeds. Seeds of Araucaria hunsteinii, which had an initial germination of 56%, had all died within a month if stored in pure nitrogen, in two months if stored in 1% oxygen, and in three months if stored in 5% oxygen, while germination was still 18% after four months' storage in 10% oxygen (Tompsett 1983, 1984). Storage in a polythene bag of 25 micron thickness, which was ventilated (21% oxygen) periodically when opened for extraction of samples, gave results very similar to those in 10% oxygen. King and Roberts (1979) record a general consensus that adequate ventilation (i.e. adequate oxygen), is necessary for the successful storage of recalcitrant seeds at relatively high MC, as well as for storage of imbibed seeds of orthodox species.

Seed moisture content

The relationships of seed moisture content on a wet weight or fresh weight basis to seed MC on a dry weight basis, and of the equilibrium moisture content of seeds to the relative humidity of the surrounding atmosphere, are important in seed processing and are explained on pp. 121–125. They are equally important in seed storage. In the first case manipulation of RH can effectively change MC of seeds to the optimum for storage, in the second case MC can be maintained at or near that optimum by maintaining a suitable RH in the atmosphere around and between the seeds.

Effect of MC. In orthodox seeds, moisture content is probably the most important single factor in determining seed longevity (Holmes and Buszewicz 1958). Reduction in MC causes a reduction in respiration and thus slows down ageing of the seed and prolongs viability. Harrington (1959), cited by Barner (1975b), has related MC to various processes within and around the seed as follows:

Prevention of fungal activity is more easily achieved by controlling MC than by controlling temperature. If MC and RH are high enough, fungal activity is possible between -8° C and +80° C (Roberts 1972) and it is easier to keep MC below 12–14 % (or RH to the equilibrium of around 65 %) than to maintain sub-zero temperatures.

Within the range of 4 to 14 % MC, Harrington (1963, 1970) has suggested a rule of thumb applicable to many agricultural species - the life of the seed is doubled for every 1 % decrease in MC, Schönborn (1965) found a relationship of a similar order when measuring the respiration rate, expressed in terms of CO₂ production, of Picea abies. At 20° C and 20 % MC the seed gave off 80 ml. CO₂ per hour per kg of seed; at 20° C and 5 % MC, the rate of CO₂ production was reduced to 0.11 ml/hr/kg, a reduction of nearly a thousand times for a difference of 15 percentage points in MC.

An MC of 4 – 8 % is considered safe for most orthodox species; 5 % ± 1 % is recommended for long-term storage for genetic conservation (IBPGR 1976). Oily seeds will usually tolerate drying to a somewhat lower moisture content (calculated on the basis of total fresh weight) than will non-oily seeds (Harrington 1970). Drying below 4 % can lead to damage or more rapid loss of viability in some species, although certain species can be dried to a considerably lower MC. Betula papyrifera was successfully stored at 0.6 % MC without injury (Joseph 1929 cited in Holmes and Buszewicz 1958), while Schönborn (1965) succeeded in drying small samples of Picea abies, Pinus sylvestris, Pseudotsuga menziesii and Larix decidua down to 0 % MC without any observable drop in germination after 6 months, compared with germination at the normally applied 6 – 8 % MC. Drying in this case was done, not by exposing the seeds to high temperatures but by leading a current of dry air through the seeds at 20° C. The same treatment killed Pinus strobus and Abies alba, and earlier attempts by Barton (1961) to store seeds of several species of Pinus and Picea at 0 % MC also failed. Below about 2 % MC desiccation injury becomes a strong possibility in many species. Drying to very low MC is also more costly than drying to the usual 4 – 8 % and is likely to be used only in exceptional cases. Methods of drying are described on pp. 95–107.

Some orthodox forest trees store best at appreciably higher MC. As mentioned on p. 134, 8–10% is recommended for Fagus sylvatica. For seeds of Abies spp., an MC of 12–13 % is recommended for storage of one to three years, but for longer periods this should be reduced to 7–9 % (Barner 1975b). Species which benefit from storage at higher than average MC also need particular care in the timing and speed of drying.

Fluctuation in the moisture content of seed in storage due to open storage without humidity control or to frequent opening and resealing of sealed containers results in deterioration in the germinability of the seeds (Wang 1974, Stein et al. 1974). In fact a steady MC slightly above the optimum is usually less harmful than one which fluctuates between the optimum and a higher moisture content.

Some cases have been reported in which the usual trend of decreasing seed longevity associated with increasing MC is reversed at or near the moisture content of fully imbibed seeds. If the species in question needs exposure to light in order to germinate, it is possible to store fully imbibed but ungerminated seeds for some time in the dark. Fraxinus americana was stored at 22° C at varying MCs with the following results (Villiers 1973):

It has been postulated that imbibed seeds can repair damage to cell membranes, enzymes and DNA in the cell nucleus caused by free radicals in a way which is not possible for seeds at lower MC. Prolonged imbibed storage may, however, be difficult in practice, because of the need to maintain constant high moisture for imbibition and adequate oxygen without allowing the seeds to germinate or encouraging the multiplication of fungi and bacteria (Roberts 1981).

Moisture content is also important in recalcitrant seeds, but in this case the critical MC is the minimum to which it is allowable to dry the seeds rather than the maximum content for prolonged storage. For many of the large temperate hardwood seeds, moisture contents in the range of 25 – 79 % are appropriate (Wang 1974). Storage should be carried out at close to the minimum safe MC, since the higher the MC the higher the respiration rate and the more rapid the loss of viability. Higher respiration rates release higher amounts of energy and there is a risk of overheating and death of the seed, unless great care is taken to provide adequate aeration. High MC also increases fungal activity and the spread of rot. Wang (1974) quotes results from two seed lots of Acer saccharinum; germinability of one lot stored at 58 % MC and 1 – 2° C dropped from 94 % to 12 % after 6 months' storage, while that of the other, stored at 45 % MC at the same temperature, was still 78 % after 16 months. Loss of viability in this species may be sudden. Tylkowski found that seeds stored in sealed bottles at 50–52 % MC and -1° to -3°C had over 90 % germination after 18 months but this dropped to nearly zero after 24 months (Suszka and Tylkowski 1982).

Less research has been done on tropical recalcitrant species, but there is some evidence e.g. in Triplochiton, that viability may be significantly prolonged if the minimum MC for the species can be determined and if particular care is taken over the period of drying down to that MC (Bowen and Jones 1975). Trials of Shorea platyclados in Malaysia indicated that a gradual reduction of MC to 20 – 27 %, followed by sealing in charcoal, sawdust or vermiculite at 15° – 22° C allowed storage for at least one month, compared with the week or so of natural viability (Tang 1971). On the basis of experiments carried out on Shorea parvifolia and Dipterocarpus humeratus, Maury-Lechon et al. (1981) recommended reduction of MC to between one quarter and one half of the initial MC in freshly collected fruits. Although tropical recalcitrant seeds cannot yet be stored for more than short periods, there is a growing body of useful research on the problem. King and Roberts (1979) contains a good summary of achievements and possible approaches.

Storage temperature

Temperature, like moisture content, is negatively correlated with seed longevity; the lower the temperature the lower the rate of respiration and thus the longer the life-span of the seed in storage. Harrington (1963, 1970) suggested another rule of thumb for agricultural seeds - between 50° C and 0° C, every 5 ° C lowering of storage temperature doubles the life of the seed. For orthodox seeds, which can be dried to a low moisture content, still greater longevity can be assured by storage at sub-freezing temperatures. For long-term storage for genetic conservation of agricultural seeds, a temperature of -18° C has been recommended as the “preferred” standard for most species and -10° C as an “acceptable” standard for those species known to have an intrinsic high viability (IBPGR 1976). Much lower temperatures have been used with success on an experimental basis, e.g. in liquid helium at -269° C, but the high cost of maintaining such low temperatures for a long period would outweigh any (as yet unproven) advantages in increased longevity.

Choice of storage temperature varies considerably according to species and the period for which the seed is to be stored. The lower the temperature that has to be maintained in a cold store, the higher the cost, and provision of subfreezing temperatures may be unnecessary if seed is to be stored for only a year or two for afforestation projects. Holmes and Buszewicz (1958) noted that, in a number of experiments on conifers, the superiority of sub-freezing temperatures became evident only after prolonged storage over periods of about 5 years or more. Sub-freezing temperatures (-18°C) appear to prolong viability in tropical orthodox species such as Eucalyptus deglupta and Flindersia brayleyana (Turnbull 1983).

Some seeds keep well at room temperature, e.g. many leguminous and rosaceous genera, Eucalyptus, Tilia and many other hard-seeded or stone fruits. Most species, however, only keep well for longer periods at lower temperatures. In 3 – 5 years' storage of most conifers and Alnus and Betula, the crucial temperature seems to lie at a maximum of +4° C. The temperature should therefore be kept at 1 to 4° C. For longer periods of storage, say 5 – 15 years, the temperature should be -4 to -10° C. For Abies, however, a temperature of -4° C is used for short storage periods and -10° to -20° C for longer periods (Barner 1975b).

Temperature and moisture factors are so interrelated that it is very difficult to separate them. Seeds at a relatively high moisture level can be stored for considerably longer periods at near freezing temperatures than at higher temperatures, while higher storage temperatures (30° C) are less harmful when the moisture content of the seed is low. In short, it can be said that the critical moisture content lies at a higher level when storage temperatures are low than when they are intermediate or high, i.e. to some extent, a low temperature can compensate for a high moisture content, and vice versa. (Holmes and Buszewicz 1958). It is, however, necessary to avoid any risk of freezing damage caused by ice formation in seeds of high MC. Roberts (1981) has suggested that 20% MC may be the critical upper limit for storage at 0°C, 15 % for -20°C and 13% for -196°C. If seeds are dried to 4–8% MC as commonly recommended for orthodox species, there should be no danger of freezing damage, even at temperatures well below zero.

As mentioned in Chapter 6, the equilibrium moisture content of many seeds at a given RH varies with temperature. Barton and Crocker (1948) have shown that, at a range of RH from 35 % to 76 %, the amount of water contained by seeds increased progressively as temperature decreased from 30° C to 10° C. Species included Pinus as well as several agricultural crops. At low (35%) RH moisture absorption by dry seeds was approximately the same at 5°C as at 10°C, but at higher (55% and 76%) RH the seeds absorbed less moisture at 5°C than at 10°C, thus reversing the trend above 10°C. Change of equilibrium moisture content with change of temperature can be of importance in open storage. In sealed containers the effect is minimal because the final EMC is dominated by the initial MC of the seeds and not the moisture of the enclosed air.

As with moisture content, repeated fluctuations in temperature lead to loss of viability. As far as possible, temperature should be maintained at a uniform level.

The effect of temperature on seed longevity of temperate recalcitrant species is similar to that of orthodox species - within certain limits the lower the temperature the longer the period of viability. Some tropical species are killed by temperatures above freezing e.g. some dipterocarps at <14° C (Gordon 1981), cacao at <10 °C and mango at < 3–6 ° C (King and Roberts 1979). Seeds of Hopea helferi stored at 15° C with high moisture content in unsealed polythene bags retained 98 % germination after 37 days and 80 % after 60 days (Tang and Tamari 1973). Germination was much reduced if temperature was dropped to 10° C or raised to 25°–28° C. Shorea ovalis is another species which does not withstand low temperatures; it stores best at 21° C. Shorea talura, in contrast, stores well at 4° C and 40 % MC (wet weight basis); after six months germination was reduced from an initial 95 % to 69 % (Sasaki 1980 b). Other species of dipterocarp have a much shorter seed life.

For most temperate recalcitrant species the lower the temperature down to 0° C the longer the safe storage period. Below-freezing temperatures, on the other hand, often kill recalcitrant seeds which need to be stored at a high moisture content (Harrington 1970, Wang 1974). Some success has been achieved in storing Quercus seeds in the USA by maintaining them at 35 – 45 % moisture content and -1° to +3° C temperature (Bonner 1978). Temperature can be critical because less than -1° C will usually kill the seeds, while temperatures above 2° or 3° C cause excessive germination. In Europe more northerly species of Quercus can be stored at slightly lower temperature -1° to -3° C and 38 – 45 % MC (Suszka & Tylkowski 1980).

Light

Light, particularly ultra-violet light, is reported to be harmful to seed (Harrington 1970), but very few studies have been made. Use of opaque metal containers would be preferable to glass jars or bottles for species which are affected by light. But light appears to be much less important than either moisture content or temperature.

Choice of Storage Method

A number of different storage methods are available, as described below. The main factors affecting choice are the seed characteristics of the species in question, the period for which it is to be stored and the cost. If more than one method is suitable to maintain viability for the required period, the simplest and cheapest will normally be chosen.

Storage at ambient temperature and humidity

Seeds can be stored in piles, single layers, sacks or open containers, under shelter against rain, well ventilated and protected against rodents (Holmes and Buszewicz 1958, Magini 1962, Stein et al. 1974). Best results are obtained in cool, dry climates. In these conditions several species of Pinus, Eucalyptus Pseudotsuga and Tectona will store satisfactorily for at least six months, while leguminous trees with impermeable seedcoats and naturally low MC, e.g. Acacia Prosopis, Robinia will retain viability for years (Magini 1962, Stein et al. 1974).

Dry storage with control of MC but not of temperature

Orthodox seeds will retain viability longer, when dried to a low moisture content (4 – 8 %), as described on pp. 126–128, and then stored in a sealed container or in a room in which humidity is controlled, than when stored in equilibrium with ambient air humidity. Storage life is further prolonged if cool, but not controlled, temperature conditions can be provided, e.g. at high latitude or altitude and in a cellar or other room screened from direct sunlight.

Seed is sometimes stored in open containers in a room maintained at an RH of 15 – 20 % by dehumidifying machinery. In forestry it is more common to rely on predrying of the seed to the correct MC, followed by storage in full, sealed containers. Provided that the containers are not opened too frequently and that the sealing is effective, the method will maintain a low MC for many years. It is cheaper than using a humidity-controlled room, especially during periods when only a little seed is in store, and it is not subject to hazards of mechanical breakdown.

This method is suitable for a range of species including many Pinus and Eucalyptus species, for which it should maintain viability for one or more years.

Dry storage with control of both MC and temperature

This storage treatment is the standard practice for many orthodox species which have periodicity of seeding but which are planted annually in large-scale afforestation projects. For many species a combination of 4 – 8 % MC and 0 to +5° C temperature will maintain viability for 5 years or more. Some cool-temperate genera benefit by storage at sub-freezing temperatures, e.g. -4°C or lower for Abies (Barner 1982), -10°C for Fagus (Suszka 1966, 1974), -5°C for Fagus (Muller and Bonnet-Masimbert 1982), -18°C for Pinus strobus, Populus deltoides and others (Wang 1980). Pinus merkusii is an example of a tropical pine which responds well to storage at low temperature and MC. 80 % germination was obtained after 3 years' storage at 2° C and 6 – 10 % MC of seeds of the Zambales (Philippines) provenance, while seeds stored in equilibrium with room temperature and humidity showed a significant loss of germination after 3 – 4 months (Gordon et al. 1972). P. caribaea and P. oocarpa behave similarly in this respect. In addition to true seeds, this method is also suitable for certain types of fruits. For example in the Jari project of Brazil depulped, cleaned and dried stones of Gmelina arborea are successfully stored in sealed containers at 5° C and 6 – 10 % MC (Woessner and McNabb 1979). The fresh stones have a germination of 90 % and after two years of storage germination is still 80 %.

Dry storage for long-term gene conservation

The preferred storage treatment for long-term conservation of gene resources of orthodox agricultural seeds is -18° C temperature and 5 % +-1 % MC (IBPGR 1976). This is likely to be equally appropriate for orthodox seeds of forest trees requiring storage for genetic conservation. The quantity of seeds requiring this standard of storage is small in comparison with the quantities used each year for operational afforestation, and the cost per kg of seed is higher. For many countries it would therefore be desirable that forest and agricultural crop genetic resources should share a common long-term storage facility. A good example is that of the Banco Latino Americano de Semillas Forestales at CATIE, Turrialba which has its own seed store (55 m³ capacity at 5°C) for short or medium term storage, but also has access to the long-term storage facilities (at -20°C) of the Regional Genetic Resources Unit (described in appendix 3), which are also at Turrialba (Chang 1980).

Loss of viability in storage, in addition to reducing the number of plants which can be produced by a given seed lot, may result in a shift in the genetic constitution of the seed being stored. This could be particularly important in forest trees which are predominantly outbreeding, variable populations. First, loss of viability may occur more rapidly in some genotypes than others; if losses are high, say 50 % of the total, the genotypes with short-lived seeds may be eliminated altogether. Yet they may have valuable traits for adaptation, growth or disease resistance as growing trees, and in any case they contribute to genetic variation in the species which it is the purpose of genetic conservation to preserve. Secondly, it is an accepted fact in agriculture that chromosome damage or change occurs and accumulates in the seed in storage, and that the risk of such heritable gene mutations depends not so much on the age of the seed as on changes in its viability (Roberts 1972, Barner 1975b). A seed lot which has suffered a serious loss of viability is likely to have experienced some gene mutations among the survivors, but there is very little direct evidence that heritable mutations are induced under good storage conditions which lead to only small losses in viability.

The high standard of storage conditions recommended by the IBPGR and referred to above, if combined with regular testing of seed and regeneration as soon as germination falls to 85 % of the initial germination rate (Ellis et al. 1980) should minimize the risk of genetic change in storage. It is possible that still lower temperatures would increase longevity even more. Research on storage in liquid nitrogen has been pursued for some years and considerable progress has been made, but testing for several more years will be needed before the method can be recommended for general adoption in gene banks (IBPGR 1981).

Moist storage without control of MC or temperature

Suitable for storage of recalcitrant seeds for a few months over winter. Seeds may be stored in heaps on the ground, in shallow pits in well-drained soils or in layers in well ventilated sheds, often covered or mixed with leaves, moist sand, peat or other porous materials (Holmes and Buszewicz 1958, Magini 1962). Seeds stored outdoors are kept moist by rain or snow, but those under shelter may need to be moistened periodically (Stein et al. 1974). The aim is to maintain moist and cool conditions, combined with good aeration to avoid overheating which may result from the relatively high rates of respiration associated with moist storage. This may be accomplished by regular turning of the heaps of seed (Aldhous 1972) or by inserting bundles of straw or twigs into them (Magini 1962).

This method is suitable for short-term storage of large-seeded hardwood species in the temperate zone e.g. Quercus, Castanea, Aesculus. It is unlikely to be suitable for tropical recalcitrant species because ambient temperature is too high.

Outdoor stratification, a method of overcoming internal dormancy, is described on pp. 178–180. It is properly to be considered as a seed pretreatment, but it serves the incidental function of storing the seed for a few weeks or months and the method used is closely akin to those described in this section.

Moist cold storage, with control of temperature

This method implies controlled low temperatures just above freezing or, less commonly, just below freezing (Magini 1962). Moisture can be controlled within approximate limits by adding moist media e.g. sand, peat or a mixture of both to the seed, in the proportions of one part media to 1 part seed by volume, and remoistening periodically, or more accurately (but more rarely) by controlling the relative humidity in the cold store. The latter type of control is often too expensive (Magini 1962, Holmes and Buszewicz 1958). Respiration rate is reduced and storage life prolonged by the low temperature, but seed should not be stored in sealed gas-proof containers which would limit oxygen supply. Closed polyethylene bags of 4 – 10 mil (100–250 microns) thickness will allow exchange of oxygen and CO₂ with air outside, while severely restricting exchange of moisture (Stein et al. 1974).

The method is suitable for the same temperate recalcitrant genera as listed in the previous section, and with temperatures of 0 – 5° C should extend viability up to 1 ½ to 2 years. Sub-freezing temperatures have given improved results in a few cases but frequently injure seeds with high MC and should be used only after research has demonstrated their applicability to the species in question.

Less is known about the application of this method to tropical species, but it merits much more investigation that it has so far received, for the dipterocarps and genera such as Araucaria, Agathis and Triplochiton. As mentioned earlier there is evidence that some species are killed at low but above-freezing temperatures and Gordon (1981) has proposed a division of recalcitrant seeds into those which can withstand temperatures below 10° C without loss of viability and those which cannot. Tamari (1976), summing up several years of research on dipterocarps in Malaysia, concluded that the best treatment for several species was (1) Dry at a temperature not above 35° C, to reduce MC to 35 % (2) Seal with a fungicide inside polythene bags (3) Store at 15° C, or for 3 weeks at 15° C followed by further storage at 10° C. This treatment has been successful in extending longevity from a week or two up to two months in Hopea helferi, but this is still a long way from providing safe storage between seed years, reported to vary from 3 – 6 years apart in many dipterocarps (Tang 1971). Storage at 3.5°C and MC over 32 % over periods of at least 6 months has been successful on the recalcitrant Araucaria hunsteinii (Arentz 1980).

In some recalcitrant species newly germinated seeds may retain viability better under moist cool storage conditions than ungerminated seeds. Gordon (1981) reported that some pregerminated seedlots of Quercus spp. showed no significant change in the number of living seedlings after one year's storage in 500 gauge (125 microns) polythene bags lightly sealed at 3° C, whereas a large proportion of the seeds which were viable but ungerminated when placed in the same bags died during the same period.

Other methods

Other storage methods have been used in the past, but are not yet of wide application. They include (Magini 1962, Stein et al. 1974):

1. Storage of recalcitrant seeds in running (not stagnant) water.

2. Storage under partial vacuum.

3. Storage in gases other than air e.g. nitrogen or CO₂.

4. Coating individual large seeds with paraffin or latex to prevent moisture exchange. This method may also be used to maintain moisture content during shipment.

Storage Containers

Some form of container is necessary for most seed storage, to facilitate access to, and handling of, individual seed lots while keeping them separate, to make the best possible use of storage space, to provide protection against animal and insect pests and, for some seeds, to prevent passage of moisture and gases between the enclosed and the outside atmosphere. Many types of container have been used for tree seeds; they may be conveniently divided into (1) Materials freely permeable to moisture and gases (2) Materials completely impermeable, when sealed, to moisture and gases (3) Materials resistant, but not completely impermeable, to moisture.

Materials freely permeable to moisture and gases

These include hessian or burlap sacks, cotton bags and containers of paper, cardboard and fibreboard. Hessian and cotton have the advantage that seed triers can be inserted through the cloth mesh to withdraw samples for testing without the need to open the mouth of the container. The resilience of the cloth will close the hole and avoid subsequent loss of seed, which is not possible with containers based on paper or paperboard (Harrington 1973). Hessian and cotton are also robust materials which can be used more than once.

None of these materials is entirely proof against insect and rodent pests, and all are freely permeable to water vapour and gases. For orthodox seeds in uncontrolled conditions they are therefore suitable only for rather short storage periods; these can be extended in the case of hardcoated seeds or where ambient conditions are cool and dry. If seeds are stored in large containers after drying to the correct MC, the outer seeds themselves provide some barrier to the passage of moisture. Viability of the inner seeds may thus be preserved for a period even though there is some deterioration from increased MC in the outer layers. If a seed store has facilities for controlling both temperature and relative humidity, then permeable containers can be safely used for orthodox seeds for several years, provided that pests can be excluded.

For moist storage of recalcitrant seeds, open or freely permeable containers such as hessian sacks should be used in order to allow free exchange of air and so avoid the overheating which can occur if moist, rapidly respiring seeds are enclosed without adequate ventilation. Periodic spraying of the sacks may be necessary to maintain the high MC which is appropriate for this type of seeds.

Materials completely impermeable, when sealed, to moisture and gases

After drying of orthodox seeds to the correct MC, the MC may be maintained in storage by dehumidifying the whole storage space. Another very efficient way, commonly used in storing forest seeds, is to place the seed in sealed moistureproof containers. This avoids the need for expensive dehumidification equipment. For long-term storage the most effective method is a combination of moistureproof containers with controlled low temperatures provided by refrigeration. An added advantage of most materials in this type is that they also exclude oxygen and so reduce still further the rate of respiration. Impermeable sealed containers are not suitable for storing recalcitrant seeds nor are they suitable for orthodox seeds at high MC, which deteriorate more rapidly in sealed than in open storage. Some seeds absorb moisture quickly, so it is important that they be sealed inside the container as soon as possible after drying is complete, preferably within the drying room itself.

Moistureproof containers include tin or aluminium cans and drums, glass jars of the Mason or Kilner types, plastic vials and laminated aluminium foil packages. Rigid and unbreakable metal cans provide maximum protection against mechanical damage to the seeds and are equally suitable for storage and subsequent shipment. Containers are only as moistureproof as their sealing. For rigid containers screw-top or clamp-down gasketed lids should be used, if periodic opening for seed extraction and subsequent resealing are anticipated; aluminium foil should be heat-sealed. The effectiveness of sealing is particularly important in long-term storage. Three types of container are considered suitable for hermetically sealed long-term storage of agricultural seeds: glass jars or vials; metal cans; and laminated foil packets. They should be equally appropriate for orthodox forest seeds. But the report by IBPGR (1976) recommended sealed metal cans as the most reliable and convenient. It noted that the seals on screw-cap jars are not always perfect and that further experience of the lasting qualities of laminated foil packets is needed before they can be recommended for general use in storage which will often last for several decades.

Materials resistant, but not completely impermeable, to moisture

They include polyethylene and other plastic films and aluminium foil. These materials are resistant to the passage of moisture but, over a long period of time, there will be a slow passage of water vapour tending to equilibrate the RH inside with that outside the container. Some of the figures quoted by Justice and Bass (1979) for transmittal of water vapour appear surprisingly high, e.g. 0.13 g per 100 square inches (645 cm²) per 24 hours for low density polyethylene film 10 mil (250 microns) thick and about ten times this figure for low density film 1 mil (25 microns) thick. However, the standard conditions for testing these materials are 0% RH on one side and 90–100% on the other. The RH gradient during storage is never so severe as this and hence the rate of passage of water vapour is much less rapid in practice. In one test using 6 mil (150 microns) high density polyethylene the rate of passage over two years from an outside RH of 95–100% at 20°/30°C was four times that from an outside RH of 50% at 10°C (Justice and Bass 1979). The thicker the film, the greater the resistance to passage of water vapour and, for a given thickness, high density polyethylene is more resistant than low density.

Although polyethylene is not suitable for long-term storage of orthodox seeds for genetic conservation, it is very suitable for short-or medium-term storage and has given excellent results for up to 5 years' storage of Pinus caribaea and P. oocarpa seeds in Honduras, with no significant change in MC. For Honduran conditions a thickness of at least 4–5 mil (100–125 microns) is recommended; thinner polythene can permit a significant passage of water vapour in time and is also subject to mechanical damage in handling (Robbins 1983 a, b). Harrington (1973) considered 3 mil (75 microns) high density or 5 mil (125 microns) regular suitable for temperate conditions and 7 mil (175 microns) high density or 10 mil (250 microns) regular as adequate for even severe tropical conditions. Proper sealing of bags is essential and can be done by a combination of heat and pressure. In the past hot irons were used, but sealing can now be done more efficiently and conveniently by commercial heat sealers, of which a number of different models is now on the market.

Different materials, each alone slowly permeable to water vapour, may become completely impermeable when laminated together. Various combinations of laminated polyethylene, aluminium foil and kraft paper proved completely impermeable to water vapour over a two year period, even when there was a high differential between the inside and outside RH (Justice and Bass 1979).

Use of desiccants in containers

If orthodox seeds are dried to the correct MC and stored in sealed impermeable containers, the MC should remain constant for years. If, however, the seeds are stored in moisture resistant but not completely impermeable material such as polythene bags, or if it is necessary to open and reclose the containers periodically to extract seeds, there will be a slow build-up of moisture in time. A convenient way to prevent this is to enclose some desiccant such as silica gel in the containers. The capacity of silica gel to adsorb moisture depends on the relative humidity of the ambient air, as shown in the following table (after Harrington 1972):

Moisture content of silica gel in equilibrium with various relative humidities

A convenient method is to use silica gel treated with cobalt chloride, which changes colour from blue to pink about 45% RH; the corresponding equilibrium MC for many orthodox species would be 7–9% (see graphs Figs. 6.23 and 6.24 ). Dried silica gel is enclosed with the seeds and, whenever the granules turn pink, the silica gel is removed and reactivated by drying in an oven at 175° C and cooling in a sealed container before reuse. A weight of silica gel equal to one tenth the weight of seeds is recommended (Harrington 1972). Care should be taken not to include too much silica gel which could lead to overdrying of the seeds. Even with silica gel of one tenth the weight of seeds, the MC of seeds enclosed at 6% would be lowered to below 5% during the initial phase of storage. More frequent reactivation of silica gel would preserve an equilibrium of RH and seed MC at lower levels than the 45% and 7–9% mentioned above but would forego the convenience of the colour indicator.

Another use for desiccants is where the MC of seeds is known to be higher than the optimum for sealed storage, for example because only air-drying is possible. As mentioned on p. 127, the enclosure with the seed of approximately an equal weight of silica gel in sealed containers should reduce the MC of the seed to a suitable level and maintain it. As an example

Therefore a weight of silica gel equal to the weight of seed will reduce the initial 19% MC to just over 6% MC for storage.

Choice and use of container

The following factors, which should be considered when choosing the best storage container for a given use, are based on those listed by Stein et al. (1974): When seed requires further drying in storage, do not use a tight-closing container because enclosing excess moisture is harmful to the seed (Barton 1961). Use a tight-closing container if gain in seed moisture content can be damaging and relative humidity in the storage facility is high.

Containers and seed can quickly gather unwanted condensation when brought out of cool or subfreezing storage. Warming to room temperature is recommended before opening a container brought out of such storage.

4 to 10 mil (100 – 250 microns) polyethylene bags will greatly restrict exchange of moisture, but still allow exchange of oxygen and carbon dioxide with air outside. Such exchange may be beneficial or harmful, depending on species.

A container that is easy to open and close is desirable when quantities of seed are likely to be added or removed repeatedly. In order to minimize temperature and relative humidity fluctuations, open only when necessary. Alternatively, store seed in small containers, so that the entire contents can be stored or emptied at one time.

For orthodox seeds, fill containers completely to ensure minimum exchange of moisture between the seed and the entrapped air and, more importantly, to limit the amount of oxygen enclosed.

When exchange of moisture through the container walls must be eliminated or restricted, the container must be made impermeable or of moisture resistant material. The longer the storage period and the higher the differential between external RH and RH within the container, the more impermeable the material must be.

When seeds are fragile and easily damaged, a rigid-walled container should be used. Moisture-proof plastic bags are often used as liners for rigid containers.

Choose a container shape and stacking arrangement which facilitates uniform temperature and aeration throughout the storage facility.

Some containers may be made of substances that are harmful to tree and shrub seeds (Barton 1954). Unproven containers should be tested for toxicity.

7.1 Airtight containers used for storing seed, Division of Forest Research, CSIRO Canberra. (FAO/Division of Forest Research, CSIRO Canberra)

	7.2 Interior view of cold storage room at Humlebaek, Denmark. (DANIDA Forest Seed Centre)
7.3 Examples of different types of container used for storage or shipment in Denmark. (DANIDA Forest Seed Centre)

Static electricity can build up slowly on some materials such as PVC. This makes them difficult to clean between exhaustion of one seed batch and insertion of the next.

It should be stressed that no one container or packaging material can be the best for all sizes, conditions and objectives that occur in seed packaging. The relative merits of the various containers will have to be weighed against their disadvantages and costs before deciding the final choice.

Design and Engineering of Seed Storage Facilities

Storage capacity

The weight of seeds to be kept in store can be estimated in the manner indicated in Chapter 3 and will depend on the annual planting area, the maximum number of years' seed supply to be stored at any one time because of seeding periodicity, and the number of seeds per kg, for each species. Weight of seed in kg can be converted to net volume in litres (or in g to cm³) by a factor related to average specific gravity. An average factor of 2.0 is appropriate for many forest species and corresponds to an apparent specific gravity of 0.5 (true specific gravity would be slightly higher because of the air-spaces between the seeds).

For conversion from net volume to gross storage space, allowing for shelving, ventilation, air spaces within and between containers, access and fittings within the cold room, a factor of about X8 is commonly used (Magini 1962); this is with fixed shelving. Use of mobile shelving may double the quantity of seed which may be stored in a given space (IBPGR 1976); in this case a factor of about X4 is appropriate. Thus 500 kg of seed of S.G. 0.5 would need a gross storage space of 500 × 2 × 8 = 8000 litres or 8 m³ if fixed shelving were used and 4 m³ with mobile shelving. Where relatively few seed lots and large quantities of each lot are being stored, it is possible to use standard sizes of container, each filled to the brim, and for shelving space to be adapted to fit container size exactly. Under these conditions considerable savings in storage space can be effected. Thus in the Danish seed store at Humlebaek, which uses fixed shelving, a factor of only 3.12 has been calculated (Barner 1982 a).

Design and equipment

The design and machinery for refrigerated storage is a matter for refrigeration engineers. Some guidance as to the features which should be included in any quotation for installation may be obtained from the excerpts from the IBPGR (1976) report which appears in Appendix 2 and the example of the facilities installed by the Regional Genetic Resources Project at Turrialba (Goldbach 1979) which appears in Appendix 3. It should be noted that both these documents refer to long-term storage of agricultural seeds for purposes of genetic conservation.

It is essential that designs and equipment be adapted to local conditions and local resources. The best installation in the world is of little use if it cannot be maintained, so it is essential to investigate the local provision for servicing and spares before committing oneself to any particular item. The reliability of mains electricity services and the need for a voltage controller and standby generator are of primary importance. Ready availability of a spare compressor may also be necessary.

The correct siting of a seed store may reduce the need for much expensive equipment. For example a tropical country with variable climate and topography might solve many problems by moving its store from a hot humid coastal site to the dry rain-shadow side of a mountain at 2000 m. In such a case a well-ventilated room might provide perfectly suitable conditions for several years' storage for relatively “easy” species such as pines and eucalypts and could be supplemented by one or more deep-freeze chests for small quantities of more “difficult” species requiring sub-freezing temperature. The value of deep-freeze chests was stressed by the IBPGR (1976) and its comments are reproduced in Appendix 4.

Seed Shipment

The benefits of exemplary seed collection, processing and storage methods may be largely lost if care is not taken over shipment from seed store to nursery. It is seed viability at the time of sowing, rather than at the time of despatch from the seed store, which determines the number of healthy plants produced from a particular seed lot. It is therefore essential to provide shipment methods which will ensure the minimum loss of viability in the interval between storage and sowing. The selection of appropriate packing material will depend on the characteristics of the species, the quantity to be shipped, the length of time in transit, the mode of transport and the temperature and moisture conditions to which the shipment will be exposed (Baldwin 1955).

High and fluctuating temperatures and adverse humidity are the chief causes of viability losses during shipment (Stein et al. 1974). These factors are identical with those that cause deterioration in freshly collected fruits between the collecting site and the processing depot, as described on pp. 83–84. However, seeds between storage and sowing should start with advantage of having had optimum conditions of temperature and moisture content during the storage period. In fact, maintenance of storage conditions during transit would be ideal, but is often not possible (Stein et al. 1974).

Provided that the initial moisture content of the seeds is correct, it can be easily maintained during transit by the use of sealed containers. In some cases the seeds can be despatched in the same containers in which they were stored. In others it may be advisable to transfer them from a large container in storage to a smaller container for despatch. Individual nurseries may require only a small quantity of a given seed lot. In addition, small and light packages are often less subject to mechanical damage in transit than large, heavy ones. Magini (1962) recommends separate packages of 1 – 20 kg but not larger. A variety of moisture-proof or moisture-resistant material is available, as described earlier in this chapter under storage containers. Polyethylene of 4–8 mil (100–200 microns) has the advantage of restricting moisture passage while allowing exchange of oxygen and CO₂.

Sealed containers are highly suitable for orthodox species, of which the seeds must be kept dry during transit. The addition of a desiccant such as silica gel may be a useful additional insurance if there is any risk that the seeds may absorb moisture while being transferred from storage container to shipment container. Seeds of recalcitrant species, on the other hand, are best left unsealed, since the effect of some loss of moisture is less harmful than that of the overheating which can occur as a result of rapid respiration in sealed bags at ambient temperatures. They should be well-mixed with pulverized sphagnum moss, ground peat, coconut fibre or sawdust, that has been moistened and squeezed dry. A mixture of equal weights of dry packing and water will give adequate moisture content to these materials (Baldwin 1955). In the case of international transit, however, an inert non-organic substance such as moist vermiculite is likely to be more acceptable to quarantine authorities.

Sealed moisture-proof containers should always be used for long journeys, e.g. form one country to another, of orthodox species of short longevity, provided that the initial MC is correct. But if orthodox seeds are being forwarded soon after collection and without having been dried to the appropriate MC for storage, it is preferable to ship in bags permeable to air rather than to seal with excessively high MC. A number of species with resistant seedcoats or pericarps, such as Tectona and many leguminous species, are able to withstand prolonged periods in ambient conditions; cotton or paper bags or hessian sacks are perfectly suitable for these species.

Large, moist seeds can be sealed individually with paraffin wax or latex. In the method described by Baldwin (1955), paraffin wax is heated to 71°–77° C and seeds or nuts dipped for a few seconds in a screen-type container, which should be shaken vigorously during the immersion. The waxed seeds should be packed in soft material so that the wax is not scraped off during transit. At the time of sowing the wax must be partly scraped off to permit the entry of water.

Protection against high or rapidly fluctuating temperatures is more difficult, but care should be taken to avoid placing the seeds close to local hotspots such as radiators and hot pipes. For very sensitive seeds, temperature effects can be mitigated by the use of insulating material in the packaging. Sub-zero temperatures do not usually affect dry seeds but may cause damage to recalcitrant seeds which must be kept moist. Premature germination is another risk which affects moist seeds. During storage, germination can be restricted by the use of low temperatures just above freezing, but the higher temperatures encountered during transit may induce germination in a substantial number of seeds. Seeds which are prone to germinate when held in moist packing may be treated with an inhibitor such as maleic hydrazide (Baldwin 1955).

No matter what type of seeds is being despatched, it is necessary to take precautions against mechanical damage to seeds and against losses due to damage to the containers in transit. Double wrapping is often advisable, for example a sealed polythene bag should be placed inside a stout canvas bag. Stout drum cartons with sealed polythene or aluminium-foil containers inside provide an especially effective combination for seeds which need to be kept dry. If the inner bag is labelled, this is also an insurance against accidental defacement of the outer label. Clear labelling is essential and the consignee should be advised of despatch by means of an appropriate seed consignment note or seed issue form (see Appendix 1).

Stein et al. (1974) have provided a useful check list of helpful practices in seed shipment, reproduced hereunder:

1. Double wrap the seed. Enclose the seed container in a sturdy, preferably rigid, outer container.

2. Small or moderate size containers generally withstand shipment better than large containers.

3. Fill containers completely to minimize air content and jostling of seeds during shipment.

4. All packages should bear a good identifying label on the innermost covering and another one within the container.

5. For long distances, shipment of sensitive seeds by air is desirable.

6. Seed packages should permit ready opening and reclosing if destined for export to a country requiring fumigation. In addition a copy of the phytosanitary certificate should be readily available to quarantine authorities e.g. by sealing it in an envelope which is firmly attached to the outside of the package.

Seed storage facilities at nursery sites or district forest stations are inferior to those at the central seed store. Shipments should therefore be timed so that seeds can be sown with the minimum delay after receipt.