Small-scale seed productionForeword Foreword Seed production and the maintenance of crop cultivars by small farmers is a subject that has attracted increasing attention over the past decade. The increasing dominance of large multinationals in the seed trade, the controversy over genetic engineering, and the recognition of farmers rights over cultivars developed by them over the course of many years have all highlighted the importance of the maintenance of farmer capacity and capability in seed production. This Agrodok hopes to contribute to the skills and references at the farmers disposal. It has been written with frontline extension staff and skilled small-scale farmers in mind. It deals with the general principles and practices of cultivar maintenance and seed production, and makes reference to specific issues regarding cereal and legume seeds. It is hoped that follow-up booklets will deal with specific requirements of other important crop groups. The author would like to thank Niels Louwaars for his constructive comments, and Conny Almekinders and Roy Keijzer for help with references and addresses and encouragement throughout. For the illustrations many thanks are due to Barbera Oranje. And last, but not the least, Agromisa would like to thank KERKINAKTIE for the financial support that made the realization of this publication possible. Harry van den Burg 4 Small-scale seed production Contents 1 Introduction 6 1.1 Reasons for producing ones own seed 6 1.2 Seed production and cultivar development 7 1.3 Open-pollinated versus hybrid cultivars 8 2 What you must know about inheritance 10 2.1 Self-pollination versus cross-pollination 10 2.2 Genetic variation in cultivars 13 2.3 Selection criteria 17 2.4 Selection methods 18 2.5 Increasing genetic variation 23 3 Components of seed quality 28 3.1 Moisture 28 3.2 Cleanliness 29 3.3 Germination 30 3.4 Vigour 31 3.5 Seed health 32 4 Seed production of cereals and pulses 35 4.1 Choice of field 35 4.2 Crop husbandry for seed production 39 4.3 Selecting self-pollinators in the field 46 4.4 Selecting cross-pollinators in the field 49 4.5 Selecting intermediate crops 51 4.6 Examples in seed production 51 5 Harvesting seed crops 58 5.1 Timing 58 5.2 Threshing, cleaning, selecting 61 5.3 Avoiding admixture 64 Contents 5 6 Post-harvest care and storage of seed 65 6.1 Safe moisture content 65 6.2 Chemical protection and seed dressing 67 6.3 Difficult cases 70 6.4 Good housekeeping 71 7 Seed sales as a small business 73 7.1 Business potential 73 7.2 Regulatory issues 82 List of cereal and pulse crops by pollination type 84 Further reading 86 Useful addresses 88 Glossary 89 6 Small-scale seed production 1 Introduction 1.1 Reasons for producing ones own seed Keeping seed from ones own crop was standard practice for farmers throughout nearly the whole history of agriculture. Swapping or passing on new types of seed between farmers must have been common, but only in the case of crop failure or other disasters would farmers have obtained all their seed from others. Occasionally, someone may have found a plant type that was better than the one normally cultivated, and some farmers may have been better at producing good seed than others. From these very early differences the modern seed industry slowly developed. Nowadays, most technologically advanced farmers buy their seed every year. They recognize that specialized seed companies offer better quality seed of new, continually improved, cultivars than what they can produce themselves. The cost is more than offset by the benefits that they obtain. But in many countries there is no modern seed industry. Or, if there is one, it concentrates only on certain areas in the country, or on certain, mostly richer, groups of farmers. It is also common for modern seed companies to concentrate on only certain crops, for which there is a steady, large market, and not on small crops with fluctuating market sizes. The cultivars (varieties) that these companies produce are often also only suitable for certain groups of farmers. The seed may be expensive, it may be hybrid (this often goes together), or the cultivars may not have the characteristics that the smaller farmers are looking for. These are all good reasons for why farmers may want to keep their own seed. This booklet is meant to assist farmers and extension workers in applying the right methods to obtain the best possible seed qualIntroduction 7 ity. It explains the principles of seed production and indicates methods that can be used by resource-poor farmers. Because of the history of on-farm seed production, there may well be individual farmers who have developed their own, different methods of seed production. These can be very valuable when developing locally adapted methods. By checking them against the general principles explained here, it will become clear whether and how they promote the same result: good quality seed of the right variety. Likewise, the products of the formal, modern seed industry are not necessarily always wrong for the small farmer. The physical seed quality is often excellent, in most cases assisted by official certification schemes. The cultivars are designed to meet the demands of buyers other than the small farmer, but sometimes, quite by chance, these cultivars have characteristics that are of interest to the small farmer as well. It is therefore always wise to keep an eye on what the formal sector offers, and to try out what might look promising. 1.2 Seed production and cultivar development Producing seed should always go hand in hand with selection, with choosing the best and discarding the worst. This can very easily have an impact on the characteristics of the cultivars, on the way they look and perform from year to year. The identity of cultivars may slowly change over time. This is in fact how our cultivars, and even our crops, have come to look the way they do now, starting thousands of years ago with plants taken from the wild. The farmer who wants to keep his or her own seed must bear this in mind. It is one thing to want to keep a cultivar the way it is, but improving it or developing new cultivars out of it is something else. In the formal seed industry it is very important to maintain the identity of a cultivar, and official certification schemes are strictly applied for this purpose. This is because the buyers of the seed and of the end product want to know exactly what they are getting. If the buyer is a 8 Small-scale seed production processor of potatoes for instance, it is essential that the processing characteristics of the cultivar do not change, otherwise his chips or potato flour will not look or taste the same. It is also important because other seed companies may have a very similar cultivar. By allowing a cultivar to change, it may turn into somebody elses, and rules of ownership may be infringed. The small farmer who produces seed for his own use does not need to worry about all that. In fact, he most likely will be actively looking to improve his cultivar all the time. But the situation changes if he chooses to sell some of the seed. An improvement for one farmer may be a disadvantage for a buyer who farms in a different area, or for a different purpose. It is important to always be aware of what the user of the seed is looking for. In such cases, it is often better to have separate fields for the maintenance of the cultivar and for trying out new improvements. We will look separately at the methods for producing seed and at those for improving cultivars. 1.3 Open-pollinated versus hybrid cultivars Nowadays, for certain crops, modern seed companies market mostly hybrid seeds. Among the crops we deal with in this booklet, this practice applies mostly to maize and sorghum. Hybrid cultivars are made by planting two cultivars in the same field, allowing only one parent (the male parent) to produce pollen, and harvesting the seed only from the other (the female) parent. If the parents are chosen correctly, the offspring (the hybrid cultivar) will perform much better than the average of the parents, or even better than each of the parents. This is called heterosis, or hybrid vigour. It is very difficult and time-consuming to develop and choose just the right parents, who together will produce the maximum hybrid vigour. This is why the seed is expensive. It is also very difficult to copy a hybrid. If the farmer keeps and plants seed from the harvest of a hybrid, the (worse performing) parent types will appear again among the following years crop, and most of the hybrid vigour will be lost. HyIntroduction 9 brid seed production is a job for professionals, and we will not deal with it in this booklet. Farmers who keep their own seed normally work with non-hybrid or open pollinated (OP) cultivars. The plants are allowed to pollinate freely, and seed can be harvested from all plants. The only exception to this rule involves certain selection methods used in cultivar improvement, which will be described later. 10 Small-scale seed production 2 What you must know about inheritance 2.1 Self-pollination versus cross-pollination The offspring of a pair of parents, whether they are people, plants or animals, often look alike, and almost always have a number of things in common with their parents. We say that individuals inherit these things (traits or characteristics) from their parents. Whereas with animals and people there are always separate male and female parents, with plants that is not always the case. The pollen (the fine yellow powder produced by flowers, which has the same function as sperm in animals) that fertilizes a flower can be produced by a flower on a different plant. These two plants then obviously are the two parents. But it can also be produced by a flower on the same plant, or even the very same flower! Figure 1 shows the three types of flowers that exist. Figure 1: Schematic representation of a flower. a: Complete flower (impression), b: Complete flower (schematic), c: Male flower, d: Female flower If a plant is able to pollinate its own flowers and does so most of the time, we say it is self-pollinating. If there is any reason why that is not happening, we say the plant is cross-pollinating (because the pollen has to come from a different plant). Figure 2 shows some examples of how self-pollination and cross-pollination work. What you must know about inheritance 11 Figure 2: Self and cross-pollination between different types of flowers. a: Complete flower, self-pollinator, b: Dioecius, cross- pollinator, c: Monoecious, cross-pollinator, d: Complete flower, cross-pollinator Examples of self-pollinators are most plants in the bean family, as well as wheat, rice, barley and finger millet. 12 Small-scale seed production There are also plants that usually do not propagate through true seed, that is, seed produced by flowers. These are said to propagate vegetatively. Examples are sugarcane, cassava, potato and sweet potato. We will deal with these in a different booklet. There can be many reasons why a plant is a cross-pollinator. Some species have plants that produce only male or only female flowers, so self-pollination is physically impossible. These species are called dio- ecious (from a Greek word meaning: two houses). Examples are date palm and many papaya cultivars. In other cases, the plant has both male flowers and female flowers, but they are in different parts of the plant, and tend not to pollinate flowers on the same plant. Such plants are called monoecious (one house), and examples are maize, adlay and oil palm. In many such cases though, self-pollination is in fact possible, and occurs to a limited degree. In yet other plants, there are complicated genetic mechanisms that prevent pollen from one plant from fertilizing its own flowers even if it lands on them, such as in cabbages. This is called self-incompatibility, and it always results in cross-pollination. If under normal conditions less than 5 percent cross-pollination occurs, a crop is called a self-pollinator. The list in Appendix 1 shows which cereal and pulse crops are self-pollinators or cross-pollinators. It also deals with those crops that use both methods to a significant degree, i.e. between 5 and 20 percent cross-pollination. These are called intermediate pollinators. Any plant with over 20 percent crosspollination is treated as a full cross-pollinator. When producing seed, it is very important to know how your crop pollinates. Cross-pollinators as a rule are more difficult to handle if you want to keep cultivars separate, and if you want to improve cultivars through selection. You have to pay attention to pollen that may reach your fields from outside. It will also take longer to get rid of traits you dont want in your cultivar. With self-pollinators it is sometimes difficult to combine different characteristics into one cultivar. For most of this booklet, self- and cross-pollinators will be treated separately. What you must know about inheritance 13 2.2 Genetic variation in cultivars Farmers have different words for the types of seed varieties they can distinguish. They may talk about breeds, types, strains, lines, or any number of words in languages other than English. What they mean to say is that they can recognize different seed types or varieties by their outward appearance, or by the way they perform under given conditions. These differences are very often large enough for farmers to identify varieties by name, just as is done in the commercial seed sector. The word used throughout this book for such varieties is culti- var, which comes from cultivated variety. A cultivar is a group of plants within a crop, which retains its particular characteristics when multiplied in the way that is usual for that crop. Although a cultivar as a whole will remain roughly the same from year to year when grown under the same or similar conditions, it does not always mean that all plants in it are the same. The degree of ge- netic variation within each cultivar, that is, the differences between plants that are caused by characteristics inherited from their parents, can be quite large. This is an important aspect of both seed production and cultivar improvement. Often (but not always) the success of a cultivar in performing well under widely differing conditions is due to the genetic variation present within it. To put it simply, no matter what the conditions are, there are always some plant types within the cultivar that can perform well, while others lag behind, only to take over when conditions change. Such stability of performance is often of much more value to small farmers than the ability to perform exceptionally well under very specific and constant conditions. But please note that stability of performance can in most crops also be fixed in individual plants, through genes that make the carrier ecologically stable! So you dont always need genetic variation to achieve stability of performance. On the other hand, when a farmer is able to purposely set about improving his farming operations, for instance by improving his tillage, using fertilizer or lime, perhaps even using chemicals or irrigation, 14 Small-scale seed production then such genetic variation may hold back his progress. There may be too many plants in the cultivar that do not respond sufficiently to the inputs to justify the cost of applying them. However, having a cultivar with large genetic variation as starting material offers many opportunities for cultivar development. By selecting certain types within such a cultivar, new cultivars can be created that may be suitable for specific purposes or conditions, and perform better in some respects. Often, such variable cultivars are found where farmers have been multiplying their own seed for many generations without applying much in the way of selection, or where they have actively encouraged diversity. The particular combination of soils, climate, other environmental factors, and farmer-imposed factors then determines which plants prosper and produce seed and which dont. Such a cultivar is called a landrace. Landraces are normally very well adapted to the soils, climate and farming system of a certain region, and have a lot of genetic variation in them. As a general rule, the genetic variation within a cultivar of a crosspollinating crop is usually larger than within a self-pollinator. Moreover, with self-pollinators you can identify and isolate the components of this variation more easily. This is because each individual plant inherits all its characteristics in a fixed combination from its one parent. Provided that all parents roughly produce the same amount of seed, this means that the same combinations of genetic traits are coming back unchanged from generation to generation. With cross-pollinators the combinations of traits shift constantly, as individual plants inherit different traits from separate parents. As a whole within the cultivar though, the traits occur in the same ratios from generation to generation. It is just more difficult to catch them within one plant. Box 1 on Mendels peas elaborates this difference between self- and cross-pollinators. What you must know about inheritance 15 Box 1: Mendels pea experiments Mendel did most of his work with peas (Pisum sativum L.), which are selfpollinators. He cross-pollinated by hand (= crossing) several varieties that differed in many traits. Traits are characteristics that occur in two or more forms. In the case of Mendels peas, seed shape was round or wrinkled. When Mendel crossed two varieties that differed in seed shape, he found that all offspring resembled the one parent with the round seed but none resembled the other with the wrinkled seed. Therefore, Mendel called the round form dominant and the wrinkled form, which wasnt present in the offspring, recessive. Which form is dominant is different for each trait and can also be different for different crops. Some traits dont have a dominant form, like yield or seed size. The parental plants are called the P (parental) generation or the F0 generation. Their hybrid offspring are called the F1 (first filial) generation. The offspring of the F1-generation are called the F2-generation, the offspring of the F2-generation are called the F3-generation, and so on. When Mendel produced such F2 generations, he saw both round and wrinkled forms present again. The wrinkled seed that seemed to have disappeared in the F1, reappeared in the F2 generation. Mendel repeated this experiment with many other traits which had a dominant and a recessive form and each time he found that one form disappeared in the F1, but was back in the F2 generation with a ratio dominant:recessive of close to 3:1. Later it was found that each pea plant has two copies of the gene responsible for seed shape, but that pollen grains and ovules only have one of these two copies. After fertilization the copy of a pollen grain and of an ovule come together again and the resulting seed has two copies, one coming from the pollen parent, the other one coming from the ovule parent. Seed shape is regulated by one gene. Also seed colour is often also regulated by one gene, but it is not clear yet how many genes regulate yield. The different forms of a gene are also called alleles. The dominant allele is indicated with a capital letter, while the recessive allele is indicated with a small letter. In the example of seed shape, round seed shape is indicated with an R while wrinkled seed shape is given a r. If both alleles of the gene for seed shape are the same (either RR or rr), a plant is called homozygous for seed shape. If a plant has both a dominant and a recessive allele (Rr), the plant is called heterozygous for seed shape, but still has only round seeds, the dominant form. If a homozygous pea plant self-fertilizes itself, it either produces round seed or wrinkled seed. If a heterozygous pea plant fertilizes itself, it produces round and wrinkled seed in a ratio of 3:1. By continued self-fertilization the number of heterozygous plants declines, and after 5-6 generations it will be very small. 16 Small-scale seed production If peas were a cross-pollinating crop (which they arent), the ratios would be different. In that case the ratio of homozygous and heterozygous pea plants would remain stable over all generations. Also the ratio of round and wrinkled seed would remain 3:1 in all generations. Whereas a homozygous selffertilizing plant always produces the same seed type, a homozygous crossfertilizing plant can produce all seed types if the surrounding plants have different alleles. The following figures give schematic illustrations of the inheritance of dominant and recessive alleles of the trait seed shape in a selfpollinator (Figure 3, showing that after several generations the heterozygous form decreases and the ratio of round to wrinkled seeds approaches 50:50) and a cross-pollinator (Figure 4, showing that after several generations the heterozygous form doesnt decrease and the ratio of round to wrinkled seed remains 75:25). Figure 3: Inheritance in a self-pollinator What you must know about inheritance 17 Figure 4: Inheritance in a cross-pollinator 2.3 Selection criteria Whether you are producing seed and maintaining the cultivar as it is, actively improving your cultivar or developing new ones, you must be clear in your mind as to what you are looking for. What are your selection criteria? Ask yourself which traits are essential, which are important, and which are merely desirable, and write them all down according to importance. Try wherever possible to quantify them, for instance: at least four tillers per plant instead of good tillering. Consider all aspects of the crop: ? strong seedlings ? quick early growth ? tillering ? reaction to diseases ? reaction to temperature extremes 18 Small-scale seed production ? reaction to drought ? reaction to poor soil ? reaction to fertilizer ? time of flowering ? amount of seed set (number of pods/ears/cobs) ? time to maturity ? resistance to lodging ? drying off ? ease of harvest (shattering!) ? yield (remember by-products, e.g. straw!) ? storage characteristics (insect damage) ? processing and consumption quality ? consumers preferences (size, colour, taste, etc.) It is also important, especially with cross-pollinators, to distinguish between traits you can judge before or only after flowering. When maintaining an existing cultivar you should concentrate on listing and maintaining its strong points and most important characteristics. A bit of progress can often be made in the weak points by selecting against them (i.e. eliminate those plants that show those weak points), especially if there is still a lot of genetic variation in the cultivar, but you should guard against losing the typical characteristics that define the cultivar. Bear in mind that the perfect cultivar does not exist, and that being overly critical when selecting is not going to get you anywhere. Concentrate on the essential and the more important traits, and consider the rest a bonus. As you progress, and your cultivars improve in the essential aspects, you can then pay more attention to traits of lesser priority. 2.4 Selection methods So far we have pretended that everything about a plants performance is inherited. This is of course not true. When two plants in a field difWhat you must know about inheritance 19 fer in yield the cause can be genetic (the yield potential inherited from the plants parents), but it can also be a result of a difference in, for instance, soil fertility. We say then that the variation is environmental. Especially traits such as consumption quality and yield are often strongly influenced by the environment, and not so much by genetics. We say that they have a low heritability. Even very strong selection for a trait with low heritability will not give us much progress, unless it is done on a large scale and with scientific methods. Again, it is better in such cases not to be too critical. In chapter 4 we will look at selection methods that minimize interference from environmental variation. The traits plants inherit from their parents are laid down in genes, tens of thousands of them. Even the largest selection programme will only focus on a fraction of these. The others are inherited pretty much at random (by chance), even the ones we are happy with. This opens up the risk of random drift, i.e. the statistical chance that we lose a gene because the group of plants we select to continue with by chance does not contain that gene. To guard against this loss through random drift it is essential to stick with minimum numbers of selected plants, particularly in cross-pollinators. Wherever it is appropriate, we will mention with each selection method the minimum numbers of plants needed to avoid random drift. The simplest method to maintain an existing cultivar is to merely remove whatever is undesirable, and harvest the rest of the crop in bulk. This is called negative mass selection. Undesirable may refer to plants that belong to the cultivar but show unwanted characteristics, or they may be plants resulting from cross-pollination from outside, accidental admixture of seed, or mutation (the spontaneous changing of genetic information). Even untrained farmers who have been keeping their own seed for many years often instinctively practise this method. If only a part of the harvested crop is used for seed (as is often the case), it should be taken at random from the harvested crop. If this does not happen and further choices are made (for instance selecting nice-looking cobs of maize), this method becomes what we call posi20 Small-scale seed production tive mass selection (described below). This is not necessarily a bad thing, but the objective and method should be clearly defined ahead of time. Negative mass selection can be adequate to maintain cultivars of self-pollinators, but with cross-pollinators it is not very effective, especially with traits that are only visible after flowering. It is not a suitable method to improve a cultivar or develop a new one, only to maintain an existing one. With negative mass selection there is no risk of random drift. Positive mass selection is going one step further. We select and mark individual plants that are as close as possible to the typical description of the cultivar we are maintaining, or to the ideal we have in mind. At harvest time, the yield of all selected plants is kept separate from the bulk and used as seed. Further selection sometimes takes place after harvest, for instance maize cob size and shape. Now the danger of random drift appears. It is very important never to use less than 50 plants of a cross-pollinator, or 30 of a self-pollinator for the next generation. You will have to select at least twice that number in the field while the crop is growing, to allow for ones that are lost or are discarded later for various reasons. If you still end up with too much seed for next year, then be sure to take what you need at random, without selecting further. Positive mass selection is the standard method for maintaining cultivars of both self- and cross-pollinators, as long as we are happy with the cultivar and the way it maintains itself, and there are no major problems. There will come a time when we will want to correct flaws that have crept into our cultivar, whether through cross-pollination, admixture, or mutation, or for any other reason. Then it becomes important to know the parent (or parents) of the plants we have selected. Most of the time we can only do this by looking at their offspring. We then select plants as described above under positive mass selection, but keep the seeds of each plant separate at harvest. The next season we plant the seed of each plant in a separate row, as a family. In the offspring we can now see where the unwanted traits are. We then discard the entire row of plants (the whole family) where we found the unWhat you must know about inheritance 21 wanted trait, even if it only appeared in one of the plants in the row. This is because the others may very well have that trait too, even if they dont show it. In the end, you must still be left with your minimum number of families (50 or 30, as above) to mix for the new planting, so you must make sure you had plenty to start with! You then harvest all the remaining families, and mix the seed for next seasons planting. Do not select further, even if you have too much seed! With self-pollinators this method is called line selection, with crosspollinators half-sib (HS) family selection (Figure 5). Figure 5: Schematic example of line or HS-family selection For cross-pollinators, never use this method for more than two years in a row with the same material; after that go back to mass selection. It 22 Small-scale seed production is quite a lot of work anyway, so you wont want to do it too often, and one year will normally be enough. Line selection and half-sib family selection are also the methods to use if you spot a different plant in your field that looks as if it could be a good new cultivar. Harvest it separately and plant its seed in its own little row, so you can have a good look at it and see if the good traits show themselves again. You can again select individual plants from the family if you want to. If you are trying to select for or against a trait that can only be seen after flowering, there is an additional problem with cross-pollinators. By the time you have spotted where the culprits are, they already have pollinated the rest and spread their unwanted trait. Or the good plants you want to keep have been pollinated by the bad ones. You can then use the remnant seed method. (See Figure 6.) Figure 6: Schematic example of remnant seed selection What you must know about inheritance 23 This is the same as half-sib family selection, except that you only plant half the seed of each separately harvested plant, and keep the other half in storage packets. Once you have worked out which families were the good ones, you go back to the seed you kept of them in the packets, and mix that together for next years planting. The observation field with family rows is then not used for seed at all. With this method your record keeping must be very good (give packets and corresponding rows the same numbers), because you must be able to link each little family row in the field to the right packet of seed still sitting in your storage! And of course the storage must be good too the remnant seed is kept for a second season. If used properly, you should only need to use the remnant seed method once before going back to mass selection. The remnant seed method should be used for serious problems only. Again, after having spotted and discarded the unwanted families, you must still be able to continue with at least 50 packets of remnant seed from your storage. We will look further into these different selection methods and the ways to use them in sections 4.3 and 4.4. 2.5 Increasing genetic variation All the above selection methods result eventually in reduced genetic variation within the cultivar. The line/half-sib family selection and remnant seed methods work faster, and maintain less variation than the mass selection methods. There may come a time when you want more genetic variation, either for more choice of characteristics to develop new cultivars, or more variation within your cultivar to withstand widely differing conditions. Or a new disease may show up, and you want to make your cultivar, with which you are otherwise happy, more resistant to it. Or your customers demand additional traits in the product. You will then need to introduce new characteristics. 24 Small-scale seed production The first step is to look for sources of genetic variation: seeds of other plants of the same crop that have the traits you want. Good sources are usually farmers in areas with similar conditions to yours but located far away. You can also try government or international research stations, certain NGOs or aid organizations, or commercial seed companies. If you are looking for sources of resistance to a new disease you may have to search for wild forms of your crop, if they exist. A list of some public institutions you can try is in Appendix 3. Once you have seeds of promising new types, the new traits have to be incorporated into your cultivar(s). This can be done through either mixing or crossing. Mixing Mixing is by far the easiest way: you basically let the crop do the work! Before you mix new cultivar seeds with yours though, it is advisable to try at least a bit of the new seed out for a season, separate from but at the same time as your crop. In that way you can see if the growing periods generally coincide. Having parts of your crop ripen much earlier or later can give problems at harvest, and with crosspollinators the time of flowering is also important if you want the two types to blend. After the mixing of seeds of different origins, the cross-pollination will ensure that the new genes are spread throughout the population. All you have to do is look out for the individuals that combine the best of both original types. One round of HS family selection is usually enough to get these to the fore and lose most of the less successful combinations. The speed at which the mixing happens depends on the percentage of cross-pollination in the crop. With complete or nearly complete self-pollinators this will not happen. You will continue to see plants of the different origins growing side by side in the field year after year. Whether or not this matters much depends on your objective. If your aim was for instance to make the cultivar more tolerant to weather fluctuations, this will work just fine. In What you must know about inheritance 25 a dry year the drought-tolerant ones yield better, in a wet year the flood-tolerant ones. Even disease pressure on susceptible plants will be reduced when there are resistant plants in the field, and the crop as a whole may be tolerant enough for your purpose. If, however, it is essential that the characteristics of the different sources are combined in the same plant, you have no alternative to artificial crossing. Crossing Crossing in cross-pollinators, as set out in the previous paragraph, is basically automatic. There are however exceptions, where certain genetic factors prevent some individuals from crossing with others. In the cereals and pulses this is rare, but it might surface when you attempt cross-pollination with wild forms of these crops. These crossing barriers are hard to overcome without resorting to fairly high-tech methods, so if you happen to run into one of these barriers, you have reached a dead-end. It may sound surprising, but most self-pollinators cross quite easily when the right manual techniques are used. This normally involves opening a bud well before its normal time, removing the anthers (male organs), and applying pollen from your chosen male parent to the stigma (female organ). It would be too detailed to explain all these techniques here, but in the literature list you will find some references. In general: the larger the flower, the simpler the technique. Soya beans are quite difficult, but most bean (Phaseolus) species are easy. You will need to label hand-pollinated flowers individually, showing which plant was used as male parent. Small paper tags with cotton yarn on them, such as often used for price tags in clothing or jewellery shops, are ideal. Seed from crosses should then be planted according to the line selection method. All seed involving the same two parent plants crossed in the same direction (i.e. with the same plant as mother) can be planted as one line, even if different flowers were involved. You can then assess the results, and decide with which ones you want to continue. You will also see that sometimes your cross was not successful, and your offspring is a copy of the mother plant! 26 Small-scale seed production Maize is a special case. (See Figure 7 and Figure 8.) Because this crop is monoecious, and the male and female parts are well separated, it is quite easy to make directed crosses. This means that you can choose the male and female parent individually and cross them by hand, rather than rely on the natural crossing in the open field. Again, mark your crosses carefully. Figure 7: Self-fertilization of maize (adapted from Almekinders and Louwaars, 1999) What you must know about inheritance 27 However, since individual plants of cross-pollinators contain a lot of genetic variation, your offspring will be even more variable than the parents. There is not much point in separating the single plant progenies. Rather plant them as a block, and do one round of HS family selection followed by positive mass selection in the following generation. You should then have the basis of the new, improved cultivar. Figure 8: Cross-fertilization of maize (adapted from Almekinders and Louwaars, 1999) 28 Small-scale seed production 3 Components of seed quality 3.1 Moisture Through their seeds, plants are able to survive from season to season. One of the most essential functions of seed is therefore storability, which is determined by its moisture content. As a general rule, most seeds can be stored longer when they are drier. Exceptions to this rule can mostly be found among seeds of certain trees, including fruit trees such as mango and avocado, but also coffee, cocoa, rubber and oil palm, which die very soon no matter what you do. Because such seeds seem to resist storage, they are often referred to as recalcitrant seeds. Also some non-recalcitrant or orthodox seeds do not keep very well. The most important of those treated in this book are soya beans and groundnuts, including Bambara and Kerstings groundnuts. For most cereals the maximum safe moisture content for storage is around 12 to 13 percent, for pulses it can be around 2 percent higher. It should never be more. The drier the seed is, the longer you can keep it. Very dry legume seed poses other problems though: it becomes very brittle and very easily damaged when handled. The practical moisture content of legumes is therefore between 11 and 14 percent. Assessing moisture content is often one of the largest problems faced by the small-scale seed producer. There is no accurate method to judge moisture other than with small electronic moisture meters, and they are usually quite expensive. However, if you can afford them, it is definitely worthwhile investing in one. Some experienced farmers and seed handlers can judge seed moisture content by biting the seed. If they crack, rather than cut, they are dry enough for storage. If the seed is not dry enough for storage, the first casualties are germination potential and vigour. The drier the seed, the longer it will maintain a good germination potential and strong seedling growth. Roughly, for every 1 percent reduction in moisture content the seed stores approximately twice as long. Components of seed quality 29 If seed is very wet when put into storage, it provides an ideal food source for moulds and insects. The activity of these storage pests raises the temperature in your stockpile or bags, and within quite a short time the seed is totally spoiled, both by direct damage (rot and feeding) and by destruction of the germination capacity through high temperature. Ensuring that your seed is dry is therefore the first and most important measure to achieve quality. Most crops will dry sufficiently if left in the field, provided the weather is suitable. However, long periods of field drying may expose the seeds to other dangers such as lodging, unseasonable rain resulting in sprouting on the plant, attack by insects or other animals, or even theft. It is often better to harvest a little earlier, and use some form of on-farm drying. More information about moisture content, drying and safe storage of seeds can be found in Chapter 6, and in Agrodok no. 31: The storage of tropical agricultural products. 3.2 Cleanliness An important benefit of good, clean seed is the reduced spread of weeds. Crop seed should not contain any weed seeds, soil, stones, chaff or other bits of plants or broken pieces of seed. If the farmer produces seed for his own use he may not mind a little dirt, but nobody who buys seed wants to pay for rubbish! With some tall, large seeded crops (e.g. maize) cleanliness is quite easy to achieve, but with shorter plants and smaller seeded grains and pulses the seeds of wild grasses and legumes can cause problems, as can soil clods and stones. Some diseases that normally only survive in the ground can be spread to other fields too, if disease-carrying soil or plant debris is mixed with the seed. In official certification schemes the minimum required seed purity (cleanliness) is almost always 99 percent. That means there must not be more than 1 percent of all the above impurities, by weight, mixed 30 Small-scale seed production in with the seed. In addition, seeds of certain troublesome weeds are often listed, and 100 percent freedom from these is required. It is good even for the small seed producer to strive for these kinds of standards. The benefits to the user of the seed will be very great. A weed-free field is the best guarantee for a weed-free seed crop. This is especially important with cereals such as sorghum, wheat, barley, rye and oats, because closely related weeds such as Johnson grass, shattercane and wild oats can cause big problems. Even unrelated weeds, such as morning glory and bindweed, can give problems in small-seeded cowpeas, soya beans and other crops with nearly round seeds. Careful handling during harvesting, threshing and winnowing will minimize the percentage of broken seed as well as the amount of soil and stones that get mixed in with the seed. In section 4.2 we will pay further attention to achieving seed cleanliness in the field, and in chapters 5 and 6 to methods to improve seed cleanliness during and after harvest. 3.3 Germination The most important job of the seed is, of course, to germinate! Good quality seed has a high percentage of seeds that are capable of germination. For cereals this should be a minimum of 85 or 90 percent, depending on the crop. For pulses the minimum is usually 80 percent. Again, these are official seed certification standards which the smallscale producer would do well to try and achieve. A small seed producer can test the germination of his seed quite easily. Any type of plastic, steel or enamel container with a tight-fitting lid can be used. (Figure 9 shows an example.) Avoid aluminium, cast iron or anything with signs of rust, or anything that cannot be thoroughly cleaned, such as unglazed pottery or calabashes. Clean one or more containers, depending on their size, with boiling water and soap. Line the container with a few layers of tissue paper (preferably without any printing). Even toilet tissue can be used. Moisten the paper, and lay Components of seed quality 31 out at least 100 seeds, well spaced. Use more than one container if necessary. Cover with another layer of tissue, moisten again, close the lid and put the container away in a cool place, out of direct sunlight. Do not overwater, the seed must be able to breathe air. For very small seeds, the top layer of tissue can be left out. Check every day, and add a bit of water if required. After one week, count how many seeds have formed a complete, normal-looking seedling that shows no sign of disease. Those are then said to have germinated. Anything that looks strange, diseased, has parts missing, or is just very far behind the rest does not count. You can then calculate your germination percentage. Instead of tissue, plain river sand can also be used, but it should be sifted first, and then sterilized by boiling it in a pot with water. Let it cool before using, and plant the seed to the depth of the thickness of one seed. Figure 9: Example of seed germination testing on tissue paper A less exact way is by planting at least one hundred seeds in a properly tilled piece of ground with adequate watering near the house. You will however not be able to control things like excessive rain, disease, insects, or animals. Your soil must also be suitable, not too heavy and without stones or too much manure. 3.4 Vigour When talking about minimum germination, we normally refer to the germination percentage under a set of favourable conditions (soft soil, the right amount of water, optimum temperature). In the field the conditions are not always ideal, and even good seed can show less germi32 Small-scale seed production nation. The ability of seed to germinate even under less than ideal conditions, and form a strong seedling, is called seed vigour. Fresh, healthy, well-grown seed almost always has a high vigour. Even if the seed has been infected with disease, if it was produced on weak parent plants, or if it has been kept for a long time, it may still show a good germination under ideal conditions. However, as soon as the going gets tougher, lack of vigour will clearly show in greatly reduced emergence. Sound production methods are the best way of ensuring that vigorous seed is produced. If you starve your plants, you cant expect them to produce a strong seed! Maintaining good germination percentage and vigour is the most important function of seed storage. This requires a dry, cool environment that is out of direct sunlight, and free from insects, rats, and any other storage pest. You can find more information about these in Chapter 6, and in Agrodok no. 31: The storage of tropical agricultural products. 3.5 Seed health Compared to vegetative propagation methods (cuttings, shoots, tubers, etc.), cultivation from seed is a good way to raise healthy plants as most plant diseases are not carried with the seed. Nevertheless, a number of important diseases are transmitted through the seed, and a seed producer should be on his guard for them. Healthy seed is the first requirement if you want to produce a healthy crop, and the battle will be half lost already if the farmer starts off with disease in the seed. Diseases can be transmitted either inside the seed, or by riding along on the outside. As mentioned in 3.2, soil mixed with the seed can carry disease too, but the disease is then not considered to be seed-borne. If the disease is on the outside of the seed it can usually be eliminated by chemical seed dressings. Chapter 6 will explain more about these. You must realize, though, that treated seed can no longer be used for consumption when the germination percentage drops or when you Components of seed quality 33 have produced too much. In the case of pulses for dry grain production, seed treatment is often not done for this reason. In many pulses germination can drop quickly if storage conditions are not right. Treatment just before planting can sometimes be a solution. The most problematic diseases are those which are contained inside the seed, where common chemicals cant reach them. They are mostly transmitted by fungi but sometimes by bacteria or viruses. Some of the fungi and bacteria can be controlled by systemic chemicals. These are chemicals that are sprayed on the seed crop, and that are taken up by the plants and work inside the seed. They tend to be quite expensive. The only way of controlling viruses, and the best way to control fungi and bacteria without the use of chemicals, is through strict disease control and sanitation practices in the field. In section 4.2 we will go more deeply into husbandry methods to assist with disease control. Table 1 shows a list that includes seed-borne diseases that are either widespread or cause large amounts of damage in certain areas only. Local diseases with relatively low impact are not listed. This list looks positively frightening! Bear in mind, however, that it covers important seed-borne diseases from all over the world. Fortunately, there is no country (yet?) where they all occur at the same time, and although some may be very widespread their level of damage is not all that high. An example is ear-cockle of wheat (which also happens to be a nematode disease). And large parts of the important maize growing area of Southern Africa are free of maize downy mildew. The list does show, however, that it is very important that prospective seed producers learn as much as they can about the important seedborne diseases that can affect their seed crops in their area. It would go beyond the scope of this booklet to describe all of the possible diseases. The list of further reading includes a number of handy booklets that will help you with identification. Try to get more information from extension officers and any other people experienced in the crop you are dealing with. 34 Small-scale seed production Table 1: Important seed-borne diseases by crop Crop External Internal Barley smut loose smut *) downy mildew barley stripe mosaic virus Foxtail millet green ear Maize Diplodia cob rot several seedling wilts and blights downy mildew *) Stewarts wilt Pearl millet ergot grain smut green ear *) Rice brown spot foot rot bacterial blight blast Sorghum bacterial streak bacterial stripe target spot covered smut loose smut downy mildew anthracnose Wheat common bunt ear-cockle flag smut scab Fusarium rot loose smut *) spot blotch tan spot Broad bean Ascochyta Fusarium Chickpea Ascochyta blight (gram blight) Cowpea anthracnose Ascochyta blackeye cowpea mosaic bacterial blight bacterial pustule brown blotch cowpea aphid-borne mosaic cowpea mottle virus cowpea severe mosaic cucumber mosaic southern bean mosaic Groundnuts crown rot yellow mould Fusarium seed rot Lentils Ascochyta Fusarium wilt viruses Peas Ascochyta Phaseolus bean Ascochyta angular leaf spot anthracnose charcoal rot halo blight bean common mosaic virus common bean blight Pigeon pea anthracnose Soya bean anthracnose downy mildew sclerotinia stem rot soybean mosaic virus wildfire bacterial pustule bacterial blight *) = Internal disease that is treatable with systemic fungicides Seed production of cereals and pulses 35 4 Seed production of cereals and pulses 4.1 Choice of field Uniformity The most important thing to consider when choosing a field for seed production is uniformity. The reason is that you will be selecting individual plants for different stages of the multiplication process. If you select (or reject) a plant, you want to be sure that you do so based on its inherited characteristics, and not because it is standing in a much better (or worse) spot in the field. A poor plant in a fertile spot could then be selected without you noticing it! This is especially important with cross-pollinators, which tend to adapt to the environment faster than self-pollinators. A uniform field will be as level and even as possible (lower spots tend to be wetter, higher ones will be drier), does not lie on a slope, has the same type of soil throughout, and does not have tall trees close to it. It would also have the same degree of fertility throughout. A perfectly uniform field is almost impossible to find, but try to guard against extremes of all of the above. Note carefully that we did not say it has to be a very good field! Remember that selection of seeds will make your cultivar more suitable to the place where it is being selected. If that is the best field in the village, the cultivar will eventually be less suited to poorer conditions. An average field, not too fertile, not too poor, not too dry, not too wet, etc., is your best bet. This also applies when you are producing seed for use on your own farm only: dont use your most fertile field, unless it is also the most uniform. Of course a very poor field is also not suitable: poor fields and weak crops produce weak seeds. If you produce seed as part of your normal crop, and dont want to get involved with selection for genetic improvement, you can choose a 36 Small-scale seed production corner of the field where plants look strong and healthy to harvest for seed. This is better practice than choosing from the grain bulk after harvest. The seeds will most likely be stronger and healthier than if they were taken from other parts of the field. Crop rotation Growing one crop on the same field year after year is not good farming practice.
Posted on: Fri, 30 Aug 2013 23:06:43 +0000
Trending Topics
Recently Viewed Topics
© 2015