PROSEA, Introduction to Carbohydrates

Plant Resources of South-East Asia Introduction

List of species

Introduction

Definition and species diversity

The plants yielding non-seed carbohydrates have been brought together into one group based on their main product: carbohydrates. Pulses and cereals that store starch in their seed have been excluded. They are dealt with in Prosea volumes 1 and 10, respectively. The plants considered in this volume produce and often also store carbohydrates, i.e. starches and/or sugars, as a reserve plant food. The starches and sugars form the main source of food energy for both humans and animals and - especially starches - are useful as a basic material for industry.

In terms of starch production, the starch crops are interchangeable. They have an approximately equal value as food, feed and also as industrial raw material, but may differ importantly in quality. The same holds for the sugar-producing crops. The source of refined sucrose can only be deduced from impurities present in the sugar. Starch crops may be substituted for sugar crops to a certain extent because plants can convert starches into sugars. A few of the tapped palms, e.g. sugar palm (Arenga pinnata (Wurmb) Merrill) are able to convert starch into sugar on an appreciable scale. This conversion can also be done in industrial processes, e.g the high-fructose syrup (HFS) production process. Although plants may convert sugar into starch, it is not yet possible to do so on a large scale, either by exploiting plants or by means of industrial processes.

The major crops in the group include annuals and perennials. The annual crops store starch in their roots or tubers. Of the perennial genera, one palm genus (Metroxylon Rottboell) stores starch in the trunk, to be converted into sugar for development of the inflorescence. Although most sugar-producing palm genera usually store some starch in the trunk, most of the sugar is produced directly for growth and development of their inflorescences. One genus (Musa L.) stores starch first in the stem, for use later for both starch and sugars in the fruits. And one genus of the grass family (Saccharum L.) stores sugar in its stem.

This commodity group is very diverse. It contains monocotyledons as well as dicotyledons, from different families. The underground storage organs mostly vary from roots to tubers and stolons. The aboveground storage organs are mostly stems, more rarely trunks, and fruits. In this volume 54 important species are dealt with comprehensively in 33 papers. In addition, about 50 species that are minor producers of non-seed carbohydrates are described briefly in Chapter 3, and 107 with another primary use are listed in Chapter 4. The important species with another main use include breadfruit (Artocarpus altilis (Parkinson) Fosberg) and coconut (Cocos nucifera L.). Breadfruit (see Prosea 2) is a major staple crop on a number of Pacific islands, and coconut palm is an oil crop (see Prosea 14) which is also widely tapped for sugar.

Role of plants yielding non-seed carbohydrates

Geographic distribution

Most of the crops described in this volume have been cultivated since time immemorial. All the palms in the group have been domesticated in South-East Asia. The root and tuber crops have been domesticated there too, with a few notable exceptions such as the Irish potato (Solanum tuberosum L.), sweet potato (Ipomoea batatas (L.) Lamk) and cassava (Manihot esculenta Crantz), which were domesticated in Central and South America. Sweet potato is an old introduction (15-16th Centuries), cassava was introduced into South-East Asia some 200 years ago, but Irish potato has been introduced more recently. Cassava has been very successful and is now a common crop. Irish potato is quickly growing in importance in the tropics. At present, both crops are well dispersed throughout South-East Asia. Other crops originating outside the region include xanthosoma (Xanthosoma spp.) and some yams (Dioscorea spp.).

Outside South-East Asia the dispersal of sugar cane (Saccharum officinarum L.) is closely interwoven with the history of the slave trade, especially in Central and South America. Nipa palm (Nypa fruticans Wurmb) is now also found in West Africa. It was planted in the coastal belt of Nigeria in the 1940s. Sago palm (Metroxylon sagu Rottboell) was brought to West Africa and Central America in the 1980s for research purposes.

The present distribution of the main crops yielding non-seed carbohydrates over developing and industrialized countries is presented in Table 1. The distinction into developing and industrialized countries does not follow the climatic zones exactly, but most developing countries are either tropical or subtropical. Despite its limitations, Table 1 presents an acceptable estimate of the distribution of the crops over the tropics and subtropics as compared with the rest of the world. Table 1 also presents production in terms of weight and in energy, as the crops differ considerably in energy content. Only the edible part was considered when determining the energy content. In energy terms, sugar cane is the most important in the whole world, cassava is second, Irish potato a close third, and sweet potato a close fourth. All other crops are far less important. If only the developing countries are considered (in energy terms), the picture changes: sugar cane remains the first, cassava remains second, but sweet potato becomes third, Irish potato fourth and yam fifth, just before plantain. It is remarkable that Irish potato has achieved the fourth place in the tropics in recent years.

In total, the crops yielding non-seed carbohydrates provide food for an estimated 15% of the world population, compared with only 10% of the population in the tropics.

Area and production

The area under the crops and their production and yield over the countries in South-East Asia are presented in Table 2.

The FAO statistics on the crops yielding non-seed carbohydrates are less than satisfactory. Many of these are only smallholders' crops and are therefore overlooked by the enumerators. The statistics are sometimes dubious. For example, they show that Indonesia (and thus the Indonesian part of New Guinea) does not produce any taro and yam, but the half of New Guinea that is part of Papua New Guinea, produces quite a lot! It is also remarkable that plantains and cooking bananas feature little in South-East Asia except in Burma (Myanmar). Yet according to the statistics, that country does not produce dessert bananas! The production of plantains and other cooking bananas is estimated to be 40% of the total production of bananas (Valmayor & Wagih, this volume).

Nutritional aspects

There is a widespread misconception that the consumption of the crops yielding non-seed carbohydrates as a staple food leads to nutritional problems, especially with respect to the protein requirements of humans. In Table 3 the main constituents on a dry-matter basis of a number of these crops have been compared with rice. Except for cassava, sago palm, plantain and of course sugar cane, the crops contain a reasonable amount of protein. The protein content is highest in crops whose harvested produce may also be used for vegetative propagation.

If functional, structural and storage proteins are distinguished, the picture improves even more. Functional proteins, which are found in the contents of live cells, are the most nutritious. Their quality approaches that of animal protein. Structural proteins are found in cell wall material. They are also of good quality, but are usually difficult to digest. Storage proteins are usually of inferior quality. They are hardly found in the plants yielding non-seed carbohydrates. In young storage organs both the functional and structural proteins are usually already present, and therefore these organs have a higher percentage of good quality protein. In older storage organs the starch content increases, which leads to a relative decrease in protein.

Thus, provided care is taken to eat a balanced diet, these crops are fully acceptable as a human food. Refined sugar is almost exclusively comprised of carbohydrates. But, as shown in Table 4, unrefined sugar (jaggery) may contain other nutritious compounds. These and other compounds are also found in sap tapped from the inflorescences of palms (Table 5), and in palm wine and in the vinegar derived from it, but are almost completely absent from the distilled product.

Sometimes, other parts of the crops are consumed too. The leaves of root and tuber crops, especially of the Araceae, cassava and sweet potato are often used as a vegetable, thus supplying valuable protein. Banana flowers are also used as a vegetable. The growing point of a number of palms is eaten as palm cabbage.

Quality aspects

Very little needs to be said about the quality of sugar. Sugar is largely comprised of sucrose, a compound molecule which can be split into two molecules, fructose and glucose, the reducing sugars. Refined sugar consists solely of sucrose. Its source can only be recognized from cell remains in the sucrose, thus only by means of impurities.

The quality of starch is another matter. Starch is a polysaccharide, a glucose polymer with a high molecular weight. Various factors affect starch quality:

• Size of the starch grains. Large starch grains settle faster than smaller ones. Usually, the starch grains of younger plant parts are smaller than those of older ones: to a certain extent the starch grains increase in size with age. Small starch grains are easier to digest than larger ones.
• Amylose versus amylopectin. Starches contain some amylose, a non-branched polymer, and the rest is amylopectin, a branched polymer. Amylose is soluble in water and turns blue in the presence of iodine. Amylopectin is insoluble in water and turns brown in the presence of iodine. When gelatinized at high temperatures amylopectin yields a paste of a certain viscosity, which retrogrades with time. Gelatinization temperature, maximum viscosity and speed of retrogradation are the main factors determining the industrial uses of the starch.
• Purity of the product. Starch, especially starch for industrial purposes, should be white and clean and free from impurities. This purity depends on the washing process and thus also on the purity of the water used.
• Erosion of starch grains. Starch grains kept too long in water may erode irregularly because of microbial activity. This reduces the quality.

The properties of starch can be represented diagrammatically in an amylograph.

In general, it can be said that all starches are economically valuable, provided they are pure and are produced in sufficiently large batches of constant quality. Only then are they of interest to the starch industry. If the starches are impure, or if the batches fluctuate in quality, they will fetch a lower price on the international starch market. This does not necessarily apply to starches that are converted into food products or high fructose syrup.

Inulin, the main carbohydrate in the Jerusalem artichoke (Helianthus tuberosus L.) cannot be digested by man, except when broken down by micro-organisms in the rectum. Hence its importance for diabetics, which in turn has led to the confusion with insulin, a hormone from the pancreas necessary to metabolize sugar in the human body.

Other uses

Palm leaves are often used for roofing. The leaves of the true sago palm are the best. If properly prepared and handled they may last for some twelve years. The leaves of the nipa palm are second best. They may last up to eight years. The long rachides of these two palms may be used to construct walls. The leaves of the toddy palm (Borassus flabellifer L.) may be used as well, but the fan leaves cannot easily be made into proper thatch ("atap"). The leaves of sugar palm and of coconut are only occasionally used as roofing, e.g. for livestock sheds. They are much less durable.

Palm wood can be very hard and thus durable. Only the outer bark of sago palm and of sugar palm is used, often as flooring. The wood of the toddy palm is reasonably durable.

Fruits of the sugar palm are irritating to the skin, even though they are eaten as a candy. Fruits of toddy palm can be used in the same way as those of coconut. Fruits of the sago palm are only used for ornamental purposes. Those of the nipa palm are hardly used at all.

Average and actual yield levels

In Table 6 the average world yields of the most important tropical crops yielding non-seed carbohydrates are presented, and compared with the high yields found in the literature. The common names of the crops are given, together with the literature reference for the high yield figures (column 1) and the reported high yields per crop combined with the number of days in which these were obtained (column 2). To facilitate comparison, these yields have been expressed in kiloJoules edible portion per ha and per day (column 5), using the tables compiled by Platt (1971) (columns 3 and 4). The average yields of the most important crops in 1992 have been taken from the FAO Production Yearbook (FAO, 1993), and the cropping period has been estimated (column 6). The average world yield of each crop per ha and per day of cultivation has been determined (column 7) using the same procedure as was used for the high yields.

This calculation procedure has some severe limitations. The world yields presented by FAO are averaged from highly divergent figures. The estimates of the duration of growth (column 6) are of limited value, as the duration of growth of a crop may vary greatly. Furthermore, the figures given by Platt are of limited value. For instance, young roots and tubers usually contain less dry matter than older ones, thus the amount of kiloJoules of the edible part may vary with age.

The calculation procedure is only valid for regions where crops can be grown throughout the year, i.e. tropical and subtropical regions with sufficient water. Soil preparation has been ignored, yet this appears to increase in importance as the cropping period becomes shorter. For instance, three crops of Irish potato may be grown in one year, but plantain and palms may occupy the same land for a number of years. Moreover, the palms have a long unproductive period: neither sago palm nor sugar palm start producing until after at least 8 years. Sago palm continues to produce for perhaps up to 40 years without replanting, whereas sugar palm produces for 3-4 years and then needs to be replanted. Despite these limitations, Table 6 suggests some general trends. Root and tuber crops produce much more bulk than rice, but mainly because of their high water content. Their nutritional value in terms of energy amounts to approximately 30% of that of rice per unit of weight of the edible product. All starch crops except for Irish potato and sweet potato have approximately the same average world yield (in energy terms) per ha per day (column 7). The rather high average world yield of Irish potato is caused by this crop's high yields in the temperate zone, where both breeding and cultivation are of a high standard. Sweet potato shows an even higher average world yield, probably thanks to the high yields in the United States, mainland China and Japan.

The average world yields of the sugar crops approach the yields of sweet potato and Irish potato. Palms do well, especially considering that they receive little attention in terms of research for cultivation and breeding. But the long unproductive period reduces average production.

The highest experimental yields in the crops shown in Table 6 are of the same order of magnitude, with the exception of plantain and yam, which yield much less.

Potential yield levels

If the other conditions are optimal, the dry matter production of a crop is determined by photosynthesis and thus by sunlight. In the tropics, temperature is more important for respiration than for photosynthesis. The gross potential dry matter production, assuming a half overcast sky, may be estimated at roughly 275 kg carbohydrate equivalents per ha and per day. This equals 46.10⁵ kJ ha^-1 day^-1. However, net food production is much lower, because of the following factors:

• The crop canopy does not close immediately, and therefore the use of sunlight is not optimised. This can be remedied to some extent by spacing the plants more closely.
• Nocturnal respiration uses up part of the carbohydrates.
• The edible percentage of the dry matter produced varies between the crops. In cereals, a maximum of 40% of the dry matter produced is edible. Some 50% is needed by the aboveground structure that carries the edible product and the remaining 10% is used for the root system. It may be possible to change this pattern somewhat by breeding in favour of the edible dry matter, but it will never reach the 80% of edible dry matter attained by the root and tuber crops. The root and tuber crops produce the edible dry matter in the soil and thus do not need an elaborate aboveground structure to carry it. It may thus be assumed that the potential yield of edible dry matter of the root and tuber crops is twice that of the cereal crops.

The average yields in the world and the high experimental yields (energy terms) in Table 6 have been combined in Figure 1. The average net potential production of dry matter has been estimated as about 23.10⁵ kJ ha^-1 day^-1, which is half the gross potential production. The distribution pattern of the dry matter has been estimated at its theoretical maximum. For plantain, the palms and sugar cane the distribution pattern has not been estimated in Figure 1.

These rough estimates are open to the following criticisms. Firstly, the crops differ in photosynthetic efficiency. Sugar cane, being a C₄ plant, has a higher photosynthetic efficiency than rice, especially at higher temperatures. This means that sugar cane can produce better than the other crops, especially at higher levels of irradiation. All the other crops are C₃ plants, like rice, and have approximately the same photosynthetic efficiency. Secondly it is not known whether the pattern of dry matter distribution is comparable within each group of crops. The root and tuber crops in particular may contain discrepancies, and even less is known in this respect about plantain, palms and sugar cane.

These estimates of the highest possible yields are not absolute maxima. First, yields may be considerably higher in exceptionally favourable regions, where a lesser cloud cover is supplemented by ample water.

The figure clearly shows that the root and tuber crops have a far higher yield potential than cereal crops. And some root and tuber crops, especially yam, lag far behind in breeding (as does plantain). Improved cultivation techniques for palms could produce surprising results.

Root and tuber crops compared with cereals

The main advantage of root and tuber crops over the cereals is thus the partition of the dry matter production. They are able to produce twice as much useful dry matter as cereals. But this yield potential is not realized because research results are not disseminated adequately.

Cereals, however, have a distinct advantage because of their low (about 12%) water content, which facilitates their storage and transportation. Root and tuber crops with a moisture content of 60-80% are difficult to transport and can only be stored reasonably well if they have a dormant period, found only in Irish potato and in certain yams.

Perennial crops

Plantains and palms probably have a less advantageous partition of dry matter than root and tuber crops. But these perennial crops have a distinct agronomic advantage over annual crops: they remain in place for long periods. Tillering palms and plantain may be permanent land cover, and as such have an ecological advantage comparable to forest. On the other hand, these crops all have an unproductive period, during which they become established. This period may last 8 years in sago and sugar palms. Sugar cane is actually a perennial, but is usually grown as an annual crop. Thanks to detailed knowledge of its agronomy it can be grown continuously as a monocrop on the same piece of land. It is usually grown as a ratoon for a number of years and replanted when yields become unsatisfactory.

Sweetening agents of plant origin

Sugar is the ideal sweetener, because it easily dissolves in water, its sweet taste has no unpleasant effects of bitterness or saltiness, and it is rather cheap. It shows, however, considerable disadvantages. For instance, it is a major cause of dental decay and it contributes to obesity. Therefore, there has been a continuing search for sweetening agents that are low in energy value and even more sweet than ordinary sugar. There are some artificial sweeteners like saccharin (300-500 times as sweet as sucrose), cyclamate (30 times as sweet as sucrose), and aspartame (100-200 times as sweet as sucrose) (Fox & Cameron, 1977).

The following sweetening agents of plant origin are found in the following plants (Fox & Cameron, 1977; Rehm & Espig, 1976):

• miraculin, in the fruits of Richardella dulcifica (Schum. & Thonn.) Baehni (syn. Synsepalum dulcificum (Schum. & Thonn.) Daniell), which is able to make sour-tasting food taste sweet;
• monellin, in the fruits of Dioscoreophyllum cumminsii (Stapf) Diels, which is 3000 times sweeter than sugar;
• thaumatin, in the arillus of fruits of Thaumatococcus daniellii (Benn.) Benth., which is about 3 times as sweet as saccharin;
• stevioside, in the dried leaves of Stevia rebaudiana (Bertoni) Hemsley, being 200-300 times as sweet as sucrose (Mohede & van Son, 1989).

Since the sweetening agents of plant origin mentioned above are not of carbohydrate origin, these natural sweeteners are not treated here but in the Prosea volume on spices.

Botany

Taxonomy

Taxonomically, this commodity group is extremely variable, comprising plants from many different families. Table 7 gives an overview of the major plants yielding non-seed carbohydrates, arranged according to plant parts used. The most important crops are taro, sweet potato, yams, cassava, Irish potato, sugar cane, sago palm, sugar palm and plantain.

Growth and development

The growth cycle

The crops yielding non-seed carbohydrates are usually propagated vegetatively, except for some of the palms that have to be propagated from seed, such as sugar palm, toddy palm and coconut. Most of the other crops are propagated from stem parts, suckers or tubers. Some have to be propagated from the produce harvested, which is clearly disadvantageous. They include yam, the aroids and Irish potato.

In general, the following consecutive phases can be distinguished in the growth cycle of the crops yielding non-seed carbohydrates:

• Establishment. The parts used for propagation establish by forming roots and shoots; their reserves, especially of carbohydrates, should be adequate for quick establishment. Water should be available. Strong sunshine is usually not favourable because high temperatures may result in water stress.
• Development of leaf area. The plants produce the leaf area necessary for optimal growth. Water, sunshine and nitrogen should be adequate. Weed competition may be especially important.
• Accumulation of reserve food. After forming the leaf area the plants form their sink and start accumulating reserves of food. High sunshine is usually favourable. But somewhat lower night temperatures and a sufficient supply of potassium are considered to be beneficial for translocation to the sink.
• Ripening. Some of the crops show marked symptoms of ripening: a diminishing leaf area, accompanied by slackening accumulation and even a cessation of the accumulation of reserve food.

Carbohydrate accumulation

Yam and Irish potato are the only crops in which sink formation and carbohydrate accumulation may start before the optimum leaf area has been attained. But in these crops too, carbohydrates accumulate quickest after the optimum leaf area has been reached. A comprehensive survey of literature data on root and tuber crops is shown in Table 8 (Wilson, 1977). Note that the data given may not be generalized for the crop, because they represent only one cultivar for each of the crops. Moreover, these data are only estimates. Consequently, the figures given show some marked limitations, which are less strict if the common name refers to one species only, but may be important where the common name denotes a number of species in one genus (yams, Dioscorea) or even a number of genera in one family (aroids, Araceae).

Sugar cane may follow the same pattern as yam and Irish potato and so does sago palm, albeit during a much longer life span. The situation is quite different for banana, because of the dual nature of its sink. First, the pseudostem acts as a sink, but later this function is taken over by the fruits. In Ethiopia, the pseudostem of another member of the Musaceae, Ensete ventricosum (Welwitsch) Cheesman, is actually used as a starchy staple (Westphal, 1975).

Ripening

It is unclear whether ripening is caused by external factors (e.g. climate), which is an exogenous rhythm, or by internal factors of the plant, an endogenous rhythm. Once-flowering plants such as sago palm, plantain and sugar cane have an endogenous rhythm; they end their growth by flowering and fruiting and suckers may take over. Sago palm should be harvested before it starts flowering, then suckers take over. The same may hold for the Araceae and the edible canna.

In the other members of the group, except for most tapped palms, it is thought that both rhythms exist and that, ideally, they are mutually attuned. Usually, ripening can be described as an imbalance between the functions of the roots, the leaves and the sink. Root function becomes too limited to provide nutrients and/or water for both sink and photosynthetic apparatus. A choice is then usually made in favour of the sink. Clearly, such a phase of ripening would be strongly influenced by a water shortage and should be called exogenous. On the other hand, a programmed destruction of both photosynthetic apparatus and roots may exist, as appears to be the case in species of yam and in Irish potato. This form of ripening should be called endogenous.

In general, the phases of plant establishment and leaf area development should be made as short as possible. Prolonging the phase of rapid carbohydrate accumulation is probably advantageous, if at all possible. If there is an endogenous factor initiating a possible phase of ripening, then such a lengthening may only be realized through breeding. Sugar cane, plantain and sago palm have this endogenous rhythm, and in these crops carbohydrate accumulation is ended by flowering followed by fruiting. Irish potato and yam probably also possess an endogenous rhythmicity, as shown by their dormancy. The duration of growth in these crops is probably a varietal characteristic.

In Irish potato and yam there may be a connection between the dormancy before germination of the tubers, the endogenous rhythmicity and the onset of sink formation before maximum leaf area is attained. In sweet potato, the aroids and cassava the situation is less clear and needs further research. If there is no such endogenous rhythmicity in these crops, then their production could be increased by prolonging the phase of starch accumulation through agronomic measures.

More agronomic research needs to be done on the feasibility of prolonging the phase of carbohydrate accumulation, especially in the root and tuber crops. This phase is probably closely connected with the phase of ripening. To study this complex phenomenon the crops need to be grown under carefully controlled conditions.

The palms that are tapped in their inflorescences differ from the other crops yielding non-seed carbohydrates. In most tapped palms the trunk acts to a certain extent as a sink. At flowering, carbohydrates stored in the trunk are converted into sugars and transported to the developing inflorescence. These carbohydrates are augmented by the carbohydrate production of the still functional leaves. The products are intended for the development of flowers and fruits. This is especially important in the case of the sago palm, which flowers only once and then dies. In sugar palm, which also flowers for an extended period at the end of its life cycle, the production of carbohydrates may be increased by prolonging its ripening phase. Then all carbohydrates in the previous sink, the trunk, will be exhausted, supplemented by the carbohydrates produced by the leaves during ripening.

Toddy palm, coconut and nipa palm are different again, since they are able to flower continuously. Their carbo¬hydrates are produced directly for the growth of flowers and fruits. These carbohydrates are tapped from the flowering stalks as a sugary solution.

Ecology

The crops yielding non-seed carbohydrates differ in their optimal environmental requirements. It is therefore important to describe and compare their ecological potential and to define the range of optimal conditions for a high production of carbohydrates per unit of time and per unit of land area. And, in addition, it is useful to know how they are likely to perform outside these optimum ranges.

Climatic factors

Light

Influence of light usually is separated into light intensity and daylength. Both terms are interrelated in the total irradiation received.

Short days stimulate tuberization in cassava, yam and taro as well as in Irish potato. Such influence is unlikely in sweet potato, as this is a summer crop in the subtropics. In Irish potato cultivars grown in the temperate zone, the influence of short days on tuberization has been diminished by breeding. Short days are unlikely to stimulate flowering in perennial crops, except for sugar cane, where daylength fully controls flowering.

Total irradiation is of course important for all the crops. As shown in Figure 1, some highest yields already approach the theoretical maximum. This implies that the crop is at peak productivity during the growth phase of carbohydrate accumulation. This is probably also true for plantain, where carbohydrates accumulate initially in the pseudostem and only later in the fruits during the relatively short period of ripening.

It is often stated that aroid crops tolerate shade. The highest experimental yields as given in Table 6 and Figure 1 suggest otherwise. It is possible that the growth pattern of these crops is controlled more by temperature than by light.

Apart from the phase of rapid carbohydrate accumulation the crops may differ in their light requirements in other phases too. Some shade during establishment may have a slight detrimental effect (or even a positive effect) on plants that tiller near the mother plant (e.g. plantain, sago palm and the aroids). This does not hold for sugar cane, however. And in sago palm too, the shade of the mother plant is detrimental to the next generation after the phase of establishment.

Temperature

Experimental data on temperature requirements of the crops yielding non-seed carbo¬hydrates are scarce, except for potato and banana. A generalized survey of literature data on temperature ranges and optimum temperature is presented in Table 9 (Westphal, 1985). The altitudinal ranges have also been taken into account, as temperatures decrease with increasing altitude. Most crops thrive in the lowland tropics, with the exception of Irish potato which favours cooler conditions.

Differences between day and night temperatures within the optimal range are usually advantageous for storage of dry matter in the sink. This is probably caused by two factors: circumstances favour sink development (Tsuno, 1970) and diminishing respiration.

The positive influence of differences between day and night temperatures has not been proven for palms and plantain.

Water

Water requirements of the crops may be given in comparison to the potential evapotranspiration (E₀). The actual water need of a crop (E_t) usually differs from E₀ because the evaporating surface differs from an open water surface. For instance, in a tropical rain forest E_t may exceed E₀ because of the well spread and dense leaf canopy. And in young crops E_t may be considerably lower than E₀, especially in the first period when the leaf canopy is not yet closed.

Potential evapotranspiration in the tropics is in the order of 3-5 mm water per day, with a mean of 4 mm per day (Monteith, 1977). Much higher rates of up to 10-15 mm per day are reported for irrigated crops in the semi-arid tropics and the subtropics, but this situation is excluded here. So, in general it may be concluded that only in well spread and dense canopies with adequate air circulation, can E_t exceed E₀. This situation may occur in the mature and well regulated canopy of palms, where the leaf area index (LAI) may reach a value of 6.5 (Flach, 1977). It may also be the case in a mature canopy of banana like "Gros Michel" (Moreau, 1965). In none of the other crops does LAI normally exceed a value of approximately 4 (Wilson, 1977). Moreover, the leaf canopy is either too low or too dense, or both, to allow the necessary air circulation. It may therefore be concluded that among the crops yielding non-seed carbohydrates only mature plantings of palms and plantain may exceed E₀ in their water use. In all other crops E_t is at most equal to E₀ and often lower. See Table 9 for the water requirements of some of the crops.

Excess of water

If excess water does not run off, the crop has to tolerate waterlogging. Taro is such a crop. It may be cultivated at a high production level under conditions closely resembling those for wet rice.

The natural habitat of sago palm includes swamps, where it possesses a competitive advantage over other plants. It may produce pneumatophores under extremely wet conditions as a means to supply oxygen to the submerged roots. But under such conditions of permanent flooding the palm reportedly produces far less starch than under drier conditions. Although sago palm may also grow well under occasional flooding, even with brackish water, it grows best under less extreme conditions. Nipa palm is another crop that can withstand extreme wet conditions. It occupies large tracts of mangrove under tidal influence with brackish water. But this crop will produce less or may even stop production when air humidity becomes too low. It seems that under such conditions transpiration prevails over production and transport of sugar. None of the other crops yielding non-seed carbohydrates will do well under conditions where the available water exceeds Et.

Shortage of water

Plants/crops have various survival strategies when water is short (May & Milthorpe, 1962):

• ceasing growth and production before a serious water shortage develops (Irish potato, yam);
• deep rooting in order to use otherwise unavailable water in the soil (cassava, sweet potato, yam);
• temporarily diminishing transpiration (cassava, yam);
• temporarily lowering the water content of tissues.

Moreover, the ability to survive dry conditions may be influenced by methods of cultivation. For instance, all crops need water during establishment, but the amount needed and the urgency of timely availability is influenced by:

• the size of the parts used for vegetative propagation;
• whether these parts are planted aboveground or belowground;
• the presence of leaves at planting.

Planting large sets underground without leaves is an appropriate strategy under dry conditions. Only yam is able to survive a drought of 2-3 months (Onwueme, 1978). The yam sets may remain dormant. When they start to grow they first develop roots and only later some xerophytic vines. However, even such adaptations will not prevent a loss, at the very least, in production per unit of time, compared with optimal conditions. During the development of leaf area a shortage of water will at the least result in slower growth, again leading to a decrease in production per unit of time. But the effects of drought may be somewhat retarded in deep-rooting crops. Cassava's strategy of shedding leaves at drought and resuming growth later will also result in lower yields.

It is generally believed that a slight water deficit during the phase of carbohydrate accumulation is advantageous, whereas plenty of water will promote vegetative development. The advantage of a slight water deficit is probably attributable to the limited cloudiness during drier periods. And, without clouds there is usually more irradiation, whereas night temperature tends to be lower. The phase of ripening is usually hastened by a water shortage and may be retarded by plenty of water.

Soil factors

Soil types

Root and tuber crops need an especially well structured and friable soil because this promotes the growth of their underground harvestable organs and facilitates harvesting.

Acidity

In general the optimum pH is between 5.5 and 6.5. The upper limit for cassava, sweet potato (Tsuno, 1970) and plantain (de Geus, 1973) is approximately 8, with varietal differences. The Irish potato shows a wider range, from 4.5 to 7.5. The coconut palm has a lower limit of 4.5. The lower limit for cassava and sago palm is as low as 4. Data on the extreme limits of aroid crops and yams are incomplete.

Salinity

Salinity is usually presented as the electrical conductivity (EC) of a saturated (soil) extract and expressed in milliSiemens per cm (mS/cm). The average salt concentration of sea water is 3.5%. A salt concentration of 0.35% corresponds with an EC slightly above 8 mS/cm.

Nipa palm is the most salinity-tolerant crop among the plants yielding non-seed carbohydrates. It grows well under tidal influence and probably needs brackish water, with an EC of 10 mS/cm or more. All other palms in the group still grow and produce reasonably well up to an EC of around 8 mS/cm, which is considered to be moderately salt tolerant.

Irish potato and banana are also moderately salt tolerant, but both cassava and yam are only poorly salt tolerant.

Agronomy

Cropping systems

Some plants yielding non-seed carbohydrates are collected from the wild, while others are cultivated in shifting cultivation and fallow systems, permanent upland cultivation, perennial crop cultivation, home gardens and corporate plantations (which may or may not be irrigated).

Most root and tuber crops are grown on a small scale for subsistence in shifting cultivation, fallow systems and in home gardens. In shifting cultivation cassava, with plantain and/or banana, is usually the last crop when the plot is in transition to fallow. Only cassava and sweet potato are also grown in permanent upland cultivation. Irish potato is only grown in permanent upland cultivation at higher altitudes as a cash crop. Taro may also be grown on permanently flooded land, as in wet rice.

Sugar cane is found in shifting cultivation and home gardens, where it is often grown on a small scale for chewing. In permanent upland cultivation the scale is larger and it may be grown either for on-farm production of raw sugar or the cane may be sold to a cane factory. In corporate plantations husbandry is of a high standard and the highest yields are reached.

Palms are still in the transition from collecting to perennial crop cultivation. Sago palm is a clear example. It has always been planted for subsistence on a small scale, with planting material obtained from especially selected very productive types. This has been possible because of the easy vegetative propagation by means of suckers. Sago palm is now being planted on quite a large scale in Sarawak on undrained deep peat. Some planting has also been started in Indonesia. But most sago starch is harvested from wild stands or from old neglected plantings. Coconut palm, of course, is planted regularly. Some of the palms in a stand may be tapped. Sugar palm and toddy palm are saved for production whenever they arise from discarded seed, although sometimes they are planted deliberately. Only the nipa palm growing in the extensive wild stands in brackish water under tidal influence in South-East Asia is exploited for collection of its sugary sap. Work is in progress to improve and increase production of all these palms.

Propagation

Most crops can be easily propagated vegetatively (Table 8).

Vegetative propagation gives rise to more or less genetically identical populations. In some of the root and tuber crops, e.g. aroids, canna, yam and Irish potato, pieces of the harvested product are used for planting. Up to 20% of the yam harvest may have to be used as planting material. Only cassava and sweet potato are propagated from pieces of stem, thus not diminishing the harvest. If these two crops are grown in areas with a pronounced dry season, special measures may be necessary to save and/or produce sufficient planting material.

Sugar cane is propagated from pieces of stem. The upper part of the stem which contains less sugar, may be used. But in corporate plantations special seed gardens are often maintained, because of the rather strict planting schedule.

Both sago palm and plantain are propagated from well-sized suckers, taken from productive mother plants. This reduces the juvenile stage in both crops, although this is far more important in sago palm than in plantain.

Sugar palm, toddy palm and coconut are only propagated from seed. With the exception of coconut, in these palms it is uncommon to select planting material. However, propagating sugar and toddy palm from seed from palms that are not harvested may lead to negative mass selection.

Husbandry

Root and tuber crops are grown similarly, albeit in different soils and also often at different altitudes. Vegetative parts are planted.

Nutrient removal and immobilization

The use of nutrients by plants can be put into two broad categories:

• nutrients taken up by the storage organ, the sink, are usually removed from the field at harvesting and thus lost for cultivation;
• nutrients in other plant parts are mostly immobilized during the growth cycle and are thus not available for recycling during growth. This does not hold for fallen leaves, which in perennial crops (e.g. plantain, palms) may play an important role in recycling.

Clearly, the nutrients removed when harvesting have to be replenished if they are not available in the soil. The same holds for the nutrients immobilized in other plant parts. All crops yielding non-seed carbohydrates need high potassium levels for optimal production. On average, starch crops need some 10-20 kg of K per t of harvested dry matter and in addition 5-10 kg of N, and 2-5 kg each of P, Ca and Mg. Thus, each t dry matter in the main product removes nutrients in the order of magnitude of 10 kg N, 5 kg P, 20 kg K, 5 kg Ca and 5 kg Mg.

Cassava is a crop that can still be grown profitably on poor soils. It is able to take up the remaining nutrients, probably largely because of its symbiosis with a mycorrhiza. This is an advantage, but it may lead to severe soil exhaustion.

Crop rotation

Sugar cane may be grown successfully on the same land for a number of years, especially if the crop is ratooned.

In the dry season most crops are grown intercropped after wet rice. Only cassava and sweet potato are also grown as sole crops in larger plots.

Cereals are the first crop to be planted after a fallow has been cleared. The cereals can then use the available nitrogen. Root and tuber crops follow immediately as intercrops. Often a combination of long duration crops like cassava and plantain or banana is planted as the last crop, closing the cultivation period. Both cassava and plantain are able to compete to some extent with the returning forest.

If the fallow contains much grass, root and tuber crops are usually planted first, because grasses suffer from the same diseases and pests as cereals. Moreover, root and tuber crops make it easier to control grassy weeds.

In Hawaii, taro is grown like paddy in flooded fields, in rotation with wet rice, with excellent results. This enables good weed control.

Crop protection

Crop protection of sugar cane is rather well developed, because only a few clones, selected for their productivity, are grown industrially throughout the world.

Most of the other crops are not yet cultivated intensively and are usually only grown in small plots. Diseases and pests are therefore usually limited in their occurrence and are not a great risk. As crops become cultivated more intensively, occupying larger contiguous areas, they become increasingly prone to diseases and pests. A good example of this is cassava; its diseases and pests are becoming more of a risk. Crop protection should therefore be developed concurrently with intensifying cultivation and breeding.

How frequently a crop needs to be weeded often depends on how quickly its canopy closes. Cassava, for instance, needs a long time to do so, but sweet potato quickly covers the soil, smothering weeds.

Harvesting and post-harvest handling

Sugar cane

The best time to harvest cane can be determined fairly accurately. Once harvested, sugar cane can be kept for only a few days without appreciable losses. If the cane has been burned before harvesting, deterioration sets in even more rapidly. A large cane sugar factory therefore needs to organize harvesting in a strict schedule. Small farms delivering their own cane at will to such a factory may adversely affect sugar content and recovery.

Root and tuber crops

The harvesting period of roots and tubers varies to a certain extent. The best way to preserve the roots and tubers is to leave them in the soil as long as possible.

At harvest, they need to be lifted from the soil. Care should be taken to harvest the product undamaged, as all damage leads to infection by micro-organisms and thus to deterioration. The roots or tubers should be stored in a dark, cool and well-ventilated spot. Even then most crops cannot be stored well, except for yam and Irish potato, which become dormant.

Damaged cassava deteriorates especially rapidly and can be kept for only a few days. Sweet potato is somewhat easier to store. Undamaged tubers can be kept for a number of weeks.

Sago palm

The sago palm has a clear advantage over the root and tuber crops, i.e. that the period of harvesting can be lengthened at will to over one year, without appreciable losses. The best time for harvesting is at flower initiation, but the amount of starch in the trunk increases slowly until the seed starts to form. The only disadvantage of harvesting at this time is that the next crop from the suckers will mature later.

Palm tapping

In principle the inflorescences of all palms may be tapped for their phloem sap. Whether they are exploited depends on:

• whether other products of the palm may have more important economic uses,
• the growth habit of the palm,
• whether the structure of the inflorescence makes tapping possible,
• the ease of reaching the inflorescence.

What actually happens in palm tapping is that the phloem sap intended for the production of flowers and fruits is used for the production of sugary sap. For tapping, the main axis of the inflorescence must be of a certain length. Usually the axis is pretreated so that it can be bent over and to enhance sap flow. The mechanism of pretreatment is not yet fully understood.

If a palm contains reserve food in the trunk, this may be removed from the trunk by tapping. The direct products of photo¬synthesis may also be removed by tapping. Both processes occur in the sugar palm. In lowlands this palm grows for 7-8 years and only then starts to flower, from top to bottom. The trunk may contain up to 125 kg of starch, which may be extracted in the same way as from the sago palm. During tapping, however, the starch collected in the trunk, and the products of ongoing photosynthesis in the attached leaves are converted into tappable sugar. Over the years of tapping, more sugar is therefore usually produced than is contained in the starch in the trunk. Cultivation should therefore aim at keeping the leaves productive for as long as possible. In the nipa palm only the direct products of photosynthesis are tapped, as little or no starch appears to be present in the underground stem.

Sometimes the growing points of trunks are also tapped. Then the phloem sap intended for the production of leaves is used for sugar production. This sap has a somewhat different composition. Such tapping is usually more detrimental to palm growth than tapping of the inflorescences. Even the trunks of felled palms lying on the ground may be tapped, though only for relatively short periods.

Processing and utilization

Sugar cane

The processing of sugar cane both industrially and in home production on small farms follows the same pathway. The juice is pressed out and the sap is boiled until the sugar concentrates, using the pressed-out cane (bagasse) as a fuel. In factories the thickened sap is then purified, ultimately producing crystallized sugar. Small farms usually produce jaggery, a brown sticky and hygroscopic sugar, still containing all dissolved parts of the sap.

In the factory processing of sugar cane a residue of reducing sugars is produced: the molasses. This is often used for the production of alcohol or as an animal feed.

Root and tuber crops

Cassava is often washed, cut and dried if it is to be stored. Thin slices dry easily and can be stored as such. When sliced more coarsely it is often sold as an animal feed. It may also be pressed mechanically into pellets for animal feed.

In Africa, a number of processed products are often made from roots and tubers, such as gari. Gari, prepared from cassava, consists of washed, peeled and grated roots fermented in jute sacks under heavy weights to press out the excess water. Fermentation increases the protein content, as a result of micro-organism action. After drying, the gari keeps for some time.

Cassava can also be used for the production of starch, because of its low sugar and protein content. Flakes may be prepared from all kinds of root and tuber crops. These also keep after drying. Unfortunately, gari and flakes are little known in South-East Asia.

In industrialized countries, sweet potato may be preserved by canning. Although the crop is also used for starch production, especially in Japan, it is less suitable for this because of its usually high sugar content.

Sago palm

The sago palm trunk needs to be ground for processing, in order to obtain the starch. The core or pith of the trunk cannot be eaten fresh by man. Grated debarked trunk, however, can be fed to animals, especially horses, pigs and chickens. Under the 2 cm layer of bark, the pith is softer than in Irish potato and thus easy to grate. The starch can then be washed out of the pith. The starch is dried for industrial purposes. In areas where it is the staple, the starch may be preserved under water, where it deteriorates slowly. If kept wet but out of water it acquires a characteristic smell of lactic acid. The most common way of preparing this product for eating is to add hot water to wet starch and then stir with a stick. The resulting glue-like mass is eaten with a number of side dishes of vegetables and meat.

The starch can also be baked into various shapes. The product obtained in this way can be preserved easily and be eaten dipped in coffee, tea or other liquids such as a soup.

Tapped palms

For sugar production, the sap of palm tapping is collected and dehydrated by cooking in an open pan, which also results in jaggery. The sap from tapping contains all the elements that were intended for the production of flowers and fruits. This provides an excellent substrate for the growth of micro-organisms, especially yeasts. Therefore, potassium bisulphite is often used to stop their growth. The sugary sap is then handled in the same way as sugar cane sap. It is even possible to produce crystallized sugar from palm sap.

The juice from palm tapping may also be sold fresh. Each kind of palm juice possesses its own, often characteristic taste, and contains up to 16% sugars, mainly sucrose. Some of the sucrose may have broken down into reducing sugars, depending on the cleanliness of the tapping vessels and the time between harvesting and consumption.

Immediately after tapping, the sugar starts to convert into alcohol (palm wine) because of the action of yeasts. After two to three days the alcohol percentage of the sap may reach 12-15%. This product is called wine or beer. It mostly also already contains some acetic acid and tastes sour. The palm wine may be distilled to give higher, but usually largely unknown, percentages of alcohol (between 20% and 60%). The palm wine can also be converted into vinegar. If the product is then kept for some time, the concentration of acetic acid increases. The final product may then be diluted with water to obtain vinegar.

Wine and vinegar still contain most of the nutrients that were present in the fresh sap and are thus quite nutritious.

Genetic resources and breeding

Sugar cane breeding has a long history and therefore there are good collections of cultivars in the national research stations throughout the world. Sugar cane breeding is also well developed. Collections of genetic material of the main root and tuber crops are maintained by the relevant international research institutes under the auspices of the Consultative Group for International Agricultural Research (CGIAR). The Centro Internacional de Agricultura Tropical (CIAT) in Colombia maintains a collection of cassava. The International Institute of Tropical Agriculture (IITA) in Nigeria does likewise for yams. The Centro Internacional de la Papa (CIP) in Peru possesses a collection of Irish potato and the Asian Vegetable Research and Development Center (AVRDC) in Taiwan maintains a collection of sweet potato. The same institutions, alone and in combination, also conduct research on the crops and exchange cultivars.

Very little work on maintenance of collections is being done on the other crops in the group. Often, small collections are available in the countries where the crops are grown, but breeding of these crops is almost non-existent. The International Plant Genetic Resources Institute (IPGRI) in Rome, Italy, is encouraging the establishment of gene banks, especially for the neglected crops.

Prospects

Non-seed carbohydrates as food

Despite having a much higher potential yield than the cereals, the root and tuber crops are still mainly subsistence crops and are used as a fresh product on only a limited scale. Their high water content (60-90%) compared with 12-15% for cereals makes them difficult to store and heavy to transport. Only the products favoured for their taste have a good chance of retaining and even improving their position. Yam and Irish potato have good prospects: both of these crops can be stored more easily than the others.

The increasing prosperity of most of the developing countries means that the use of non-seed carbohydrates may decrease further as they tend to be replaced by cereal crops, notably rice and also by mostly imported wheat. Rice is already gaining in importance, replacing the starch crops that are often considered to be "poor man's food". Bread, mostly produced from wheat, is gaining an important position, especially for breakfast and mainly in cities. It is expected to increase in importance throughout South-East Asia.

The "poor man's crops" are important at subsistence level. In some areas (e.g. New Guinea), people subsist almost entirely on the starch crops. It has been rightly stated that they are important for both household and national food security (Onwueme & Charles, 1994). Their importance probably outweighs the production and trade figures. The crops do not receive the infrastructural support they need for promotion and development. Their strategic food importance should be recognized and appropriate food policies should be designed and developed to improve their supply systems to make them competitive. This could be helped by making them into products that are easier to store. Sugar cane will probably retain its important position in the world mainly as an industrial crop. All countries will, if possible, try to produce their own sugar, if only for strategic reasons.

The tapped palms also produce sugar. Their sugar (and also the unrefined cane sugar produced by small farmers) possesses some special qualities not to be found in refined cane sugar. There will always be a market, albeit limited, for this product, however hygroscopic it may be. Nowadays it is even exported to Europe on a limited scale. Although the tapped palms produce more energy per unit area and per unit of time than sugar cane, production is so difficult to mechanize that replacement of sugar cane by either nipa, sugar or toddy palm is unlikely. This may only happen on a limited scale in countries with much cheap labour and with climates and soils less suited to sugar cane.

Non-seed carbohydrates as industrial raw material

The other trend is for the starch crops to become more important whenever their products (e.g. starch and high fructose syrup) can be used directly as an industrial raw material for food, feed and other purposes.

The market for prime quality starches is growing fast. The starches must be produced in sufficiently large quantities and be of a consistently good quality. For instance, up to 50% of wheat flour in bread can be replaced by pure starches, provided the starch is white, clean and has no peculiar smell. This would provide developing countries with an opportunity to reduce wheat imports. High fructose syrup, which can easily be produced from starch, is also gaining importance, especially in bakery products and in the production of soft drinks. Starch is also used as a filler in pills, to size paper and textile, and in the manufacture of glues. There is also potential in the market for animal feed. As this product should contain some protein, the starch need not be pure. Both these potential markets are only for crops that are produced competitively (primarily cassava and sago palm).

Non-seed carbohydrates in energy cropping

Back in the 1920s in what is now Sarawak, a research project was completed on the production of fuel alcohol from the sap tapped from nipa palm (Dennett, 1927). It proved that this was an economically viable proposition, although the fuel was more expensive than gasoline. Unfortunately this pilot project was never implemented on a practical scale, because fuel prices in the region fell after the discovery of the huge oil reserves in present-day Brunei.

As mentioned above, starch can be converted into sugar and sugar into alcohol. And alcohol can be used as gasohol, a mixture of 20% ethanol and 80% gasoline, as a replacement for pure gasoline. 96% ethanol, called power alcohol, can be used pure in specially modified engines. Theoretically, 1 kg of sucrose yields 0.65 l of ethanol, whereas 1 kg of starch can be hydrolysed into 1.1 kg of sucrose and consequently be converted into 0.71 l of alcohol. Up to 10% may be lost during conversion due to the formation of yeasts. Thus, the factor for converting sugar into ethanol can be set at 0.6 and that for converting starch into ethanol at 0.64. The energy content of mineral oil is around 36 MJ/l, compared with 21 MJ/l for 96% ethanol.

In the 1970s the feasibility of producing fluid energy to replace mineral oil again became the focus of attention, because of the oil crisis. Some theoretical and practical research on this was conducted in the decade after this crisis. Although oil prices subsequently fell again, which led to a diminishing of research, it now seems that the production of substitutes for oil will gain new impetus, because of increasing prosperity in South-East Asia and also because of increasing concern about environmental aspects (power alcohol is a renewable resource and its use produces fewer noxious by-products than oil does).

The production of power alcohol from crops is feasible, albeit under a number of restrictions. First, one needs to consider the energy balance. The production process should, in total, produce more energy than it consumes. The industrial process of converting starch into sugar, followed by conversion of sugar into alcohol and finally distilling the low grade alcohol, the beer, into 96% alcohol, usually costs about as much energy as it yields. Only if the crop also yields another - combustible - product, a residue, is the energy balance positive. This is because only the combustible residue is converted into additional energy. This picture will change only if important technological advances are made in techniques to separate alcohol and water, to replace the energy-consuming distillation. Thus crops could already be selected for the presence of a combustible residue.

Crops producing appreciable amounts of additional combustible residue include the perennial crops sugar cane and sago palm. Sugar cane bagasse can be used. However, sugar cane is usually produced in a highly mechanized operation, which also requires energy. Sago palm bark is combustible and possibly so is the residue ("hampas") left after washing out the starch, if dried. Nipa palm gives only a slightly positive energy balance, because the sugary sap is obtained solely by human labour. Most other starch and sugar crops need fuel from other resources for the production process.

The second requirement is for prices for starches and/or sugar to be low, otherwise these products will not be used for conversion into energy. This may change, however, if oil prices rise.

Third, the operation must be economically feasible. This implies that the production of fluid energy will be limited to areas where labour is plentiful and cheap.

Fourth, other crops (such as oil palm and other oil crops) may well show a far more interesting energy balance.

Fifth, the production of energy crops should not compete with the production of food and feed. Sugar cane is at a disadvantage in this, as it requires prime soils which could be used for various other crops. Sago palm thrives on badly drained heavy soils that cannot be used for other crops without expensive improvements. Nipa palm grows on soils under tidal influence of brackish water - the potentially acid sulphate soils that can only be used for other crops after prolonged and expensive treatment.

In the not too distant future the production of alternative fluid energy will probably be taken up again, because of the rapidly growing demand for energy in the world and the diminishing oil reserves.

Research

The research efforts on the crops yielding non-seed carbohydrates are governed by:

• the importance of the crops in industrialized countries. This is especially true for Irish potato and sweet potato. Most of the research on Irish potato has been done in the temperate industrialized countries, although this is now changing thanks to an international research institute, the Centro Internacional de la Papa, in Peru. Most of the research on sweet potato has been done in Japan and in the United States. Recently, research has been developed in mainland China.
• commercial interest in the crop. A good example is the research on sugar cane. Sugar cane is one of the best researched crops in the world, although it is a typical tropical crop.

All other crops are mainly of local interest. Only if research is developed locally will these crops receive the attention they so badly need. This has been clearly demonstrated for taro in Hawaii and, more recently, for sago palm.

Today, the International Society for Tropical Root and Tuber Crops (ISTRC) is important in maintaining contacts between researchers on all aspects of root and tuber crops. The emphasis used to be on organizing meetings once every four or five years for scientists working on root and tuber crops from all over the world. Nowadays, smaller regional meetings are preferred. There are scientific organizations for other crops in the commodity group, such as the International Association for Research on the Plantain and other Cooking Bananas (IARPCB). Every four to five years a scientific symposium is organized on sago palm. Such meetings could be important in advancing research.

Except for sugar cane, the crops yielding non-seed carbohydrates still largely form a neglected group. This is partly because of social factors. Most crops are grown in small plots and largely by subsistence farmers. In mainly rice-consuming South-East Asia the crops are considered as the food for the poorer part of the population. Moreover, as a food, these crops are still associated - wrongly so - with low quality, poor in protein.

The main economic disadvantages of the crops are their high water content, associated with bulk, and their poor keeping quality.

The potential of the crops, especially their potential high yield, is slowly being recognized. Starch is the main energy source for humans and animals and is becoming an even more important industrial raw material.

One could envisage a development in the future in which the production of fuel alcohol is the primary basis of the market for starch and sugar, mainly as a renewable and clean source of fluid energy. A second somewhat higher price level would then be attained with its use as a feedstock, for industrial products and animal feed. The highest price level would be for human consumption, fresh or prepared.

Much needs to be done in developing the entire chain from production to consumption if this scenario is to be achieved.

Table 1

Table 2

Table 3

Table 4

Table 5

Table 6

Table 7

Table 8

Table 9

Figure 1