Contents - Previous - Next

This is the old United Nations University website. Visit the new site at

Hunger, technology, and society

Preparation of weaning foods with high nutrient density using flour of germinated cereals

Alexander C. Mosha
Tanzania Food and Nutrition Centre, Dar-es-Salaam, Tanzania

Ulf Svanberg
Department of Medical Biochemistry, University of Göteborg, Göteborg, Sweden


Observations of traditional child-feeding practices in many developing countries reveal that the weaning period, defined as the whole period during which breast milk is being replaced by other foods, usually starts when the infant is 4 to 6 months old and is extended to the age of two to three years 11). During this time, special foods are rarely available for the children. Consequently, they have to depend on the same types of foods as those eaten by adults. In the poor countries, these foods are mainly starchy tubers like cassava and sweet potato, or cereals like maize, rice, wheat, sorghum, and millet. Small children are normally given these staples in the form of gruels, i.e., boiled with water. When prepared in this way, the starch structures bind large amounts of water, which results in gruels of high viscosity (2). Such gruels need to be further diluted with water in order to give a consistency that is appropriate for child feeding. This dilution, however, decreases the energy and nutrient density of the gruel, and the child has to eat large amounts in order to satisfy his requirements.

This high volume/viscosity characteristic of a diet is usually referred to as "dietary bulk", and the importance of this factor in relation to child feeding has earlier been investigated in our laboratories and reported in a series of papers (2-5).

Industrial manufacture of cereal-based weaning foods often includes operations intended to reduce the dietary bulk, e. 9. enzyme (amylase) treatment, pre-cooking, or extrusion. These processes modify the starch structures and hence result in lower water-binding in the gruels. However, such sophisticated technologies make rather expensive products even when low-cost alternatives are developed (6), and in poor countries these products are normally only available to urban children of higher income families.

An alternative to industrial processes for reducing the dietary bulk of cereal-based weaning foods seems to be increased use of improved traditional food preparation procedures that will also modify starch structures. One such procedure that is widely known and used is germination. Germination of cereals is mainly associated with the preparation of local alcoholic beverages, but there are also a few examples where this procedure is used in preparing local weaning foods with seemingly low dietary bulk (7).

In the many situations where dietary bulk is an important factor in the aetiology of protein-energy malnutrition in pre-school children, the promotion, possibly after modifications, of germination procedures in the preparation of weaning foods may constitute an effective alleviation. The purpose of this paper is to provide further information on some alternative methods that can be applied.


Sorghum was the main cereal included in this study. The reason for this is two-fold: First, Tanzania has now embarked on a major national sorghum-promoting campaign aimed at increasing its production and consumption as a means of alleviating local food shortages. Second, sorghum is already widely used for germination, primarily for preparing locally brewed beer.

Four varieties of sorghum were selected: two traditional varieties that are grown in large areas in the semiarid parts of Tanzania, and two improved varieties that have recently been developed at the llonga Agricultural Research Institute (IARI) and now ready to be released for promotion.

One each of the traditional and improved varieties were of the white (low-tannin) type of sorghum and the others were brown (high-tannin) types. Their respective generic names are:

  "White" "Brown"
Traditional (local) Lugugu Udo Msonga
Improved ET-35 5DX 135/1313/1

In addition to the sorghum varieties, one type of maize (llonga white composite) was also studied This is a white "flint" maize and is the variety most commonly grown in Tanzania at present.

FIG. 1. Viscosity Curves of Various Gruel Concentrations of Ungerminated and Germinated Grains. (1) Improved brown sorghum, ungerminated, (2) improved brown sorghum, germinated, (3) improved white sorghum, ungerminated, (4) white maize, ungerminated, (5) white maize, germinated, (6) improved white sorghum, germinated.

The maize and the improved sorghum varieties were all sampled at the IARI. The traditional sorghum varieties were collected from a few villages in the Dodoma region of Tanzania.

A small test was also performed on cassava flour. This was purchased at a market in Dar-es-Salaam and had been prepared by milling sun-dried pieces of cassava root.

In this study, all of the sorghum and maize flour used for preparing the gruels had been milled (whole grains) in a laboratory plate mill (80 per cent passing a 60-mesh sieve) either with or without prior germination.


The traditional germination procedures at the village level in Tanzania 18) were adapted to laboratory conditions as follows before washing, the seeds were soaked in 70 per cent ethanol for two minutes to prevent growth of microorganisms. The seeds were then soaked in the same amount of water and kept in the dark for 12 to 20 hours. After washing in distilled water, the seeds were spread in a 1 cm. thick layer between wet cotton cloths and left to germinate for 48 hours at room temperature (20°C). The sprouted seeds were dried on a plate in an air-stream at 50°C, and the dried seeds with their vegetative portion were milled in the laboratory mill.

Preparation of Gruels

Flour and distilled water were mixed in a glass beaker and heated in a boiling water-bath to reach a cooking temperature of 95"C within 7 to 10 minutes, and then kept at cooking temperature for 15 minutes. The gruel was left at room temperature to cool. Gruels were prepared from either ungerminated flours or germinated flours, or ungerminated flour mixed with 5 to 10 per cent germinated flour added before or after cooking.

Viscosity Measurements

The viscosity of the gruels was measured using a Haake Rotovisco model RV1 with an SCII profiled measuring system and a shear rate of 54 rpm. The measuring temperature was kept at 40°C in order to allow comparisons of different types of gruels. Lower temperatures are commonly used for feeding infants, and this normally means higher viscosities.


Gruels Prepared from Ungerminated and Germinated Flours

Figure 1 shows the viscosities of gruels prepared from different ungerminated and germinated flours in increasing amounts (5 to 25 percent counted as dry matter). All the ungerminated flours have very similar concentration-viscosity relationships, where a suitable eating consistency for small children 11.000 to 3,000 cps*) is obtained with gruels containing about 10 per cent flour.

With germinated flours, large differences are observed among the varieties studied: for the improved white sorghum variety and the maize, the amount of flour in the gruels can be increased considerably (two to three times) before the viscosity becomes unacceptably high. This is evidently a result of starch degradation caused by the action of the alpha and beta amylases that have developed during the germination procedure 19, 10).

For the improved brown sorghum variety, however, germination for 48 hours had no observed effect on the concentration-viscosity relationship. After extending the germination time to 96 hours, a small reduction in viscosity was detected. The amylolytic activity obviously develops more slowly in the hightanning sorghum variety. The explanation for this may be the inhibitory effect exerted by the tannins on the activity of amylolytic enzymes in the seed /11, 1 2).

FIG. 2. Viscosity of Gruels with Ungerminated Flour, Concentration 15% and 40°C . After addition of germinated flour at 40°C and viscosity measured after 10 minutes Acceptable eating consistency for children of approximately 1 - 3 years .

Gruels Prepared from Ungerminated Flours with Germinated Flours Added

As is evident from the results above, germinated cereals of certain varieties contain active amylolytic enzyme. if these enzymes could be used for degrading starch in flours that have not been germinated, this would considerably decrease the amount of work needed for preparing weaning foods with improved dietary bulk. Addition of germinated (enzyme-rich) flour to gruels prepared from ungerminated flours was therefore tested both before and after cooking.

Addition after cooking. Figure 2 shows how the viscosities of thick gruels are affected by the addition of germinated flour when cooled down to 40"C after cooking. Prior to addition, the four gruels prepared from ungerminated sorghums (15 per cent dry matter) had viscosities ranging from 9,000 to 16,000 cps at 40°C., i.e., very thick consistencies. With the addition of a small amount of germinated flour of the white /low-tanning) sorghum varieties, the viscosity decreased within 5 minutes to below 1,000 cps, i.e., a semi-liquid consistency suitable for child-feeding. Only about 5 per cent addition (of the total amount of flour) of germinated flour was required for this effect, and both the white (low tannin) sorghums were equally effective. If brown sorghums are added, the germination period must be extended to at least 96 hours in order to obtain enough amylolytic activity in the flour.

The same effect was observed when germinated maize flour was added to both ungerminated maize flour gruel and a gruel prepared from cassava flour.

In order to determine the heat sensitivity of the amylolytic enzymes in the germinated flours, additions were made at different temperatures. Figure 3 shows that the germinated flour can be added at any temperature between 20"C and 70°C with the same viscosity-reducing effect as that described above for 40"C Above 70°C, however, the enzymes seem to be inactivated by heat, although some activity still remains even at 90°C (40 per cent).

Germinated flour of the white /low tannin) sorghum varieties that had been prepared in the villages was also tested in this experiment, and gave the same results as the laboratory-germinated flours described above. In this case, however, some mouldy seeds were observed, and in practical applications these should be removed in order to avoid mycotoxin hazards. If the microbiological quality of the added germinated flour is questionable, it is possible to re-heat the gruel to boiling temperature after allowing time for enzyme activity. This will not affect the viscosity.

FIG 3. Reduction of Viscosity by Addition of Germinated Flour to Gruels of Ungerminated Flour (15% Concentration) at Different Temperatures.

Addition before cooking. Figure 4 shows that, when germinated flour was added to ungerminated flour immediately before cooking the gruel, no viscosity-reducing effect could be observed. In this case, a 10 per cent addition of germinated white sorghum flours was used.

Figure 4 also shows the effect of leaving the mixtures of ungerminated and germinated flours (90:10) to soak in water over night before cooking. Here the amylolytic activity of the germinated sorghums has evidently been effective. The gruels prepared from these soaked flour mixtures had markedly reduced viscosities, measured at 40°C. It is interesting to note that the improved white sorghum variety gave a much more liquid gruel (1,130 cps) than the sorghum variety that is traditionally grown (7,700 cps). If this approach to producing low-density gruels is considered, it is thus important to select varieties that will give the desired effect.

No viscosity reduction was observed if only ungerminated flour was left to soak over night before cooking.

Practical Implications

Table 1 demonstrates the beneficial effects of using germinated flours for reducing the dietary bulk of gruels. The amounts of different white sorghum gruels needed to cover 60 per cent of the daily energy require meets of a one-year-old child (13) are calculated for gruels with a viscosity of 1,000 cps (a liquid gruel). When ungerminated flour is used, 2,500 ml of prepared gruel has to be consumed by the child. If 5 per cent germinated flour is added after cooking, the amount needed is 1,200 ml of gruel, and if only germinated flour is used, 800 ml will be enough. The dietary bulk of the gruels could be further improved if energy-rich components like cooking oil and groundnuts were added (14). The protein, vitamin, and mineral requirements should obviously also be considered.

FIG. 4. Viscosity of Sorghum Gruels with Germinated Flour added Immediately before Cooking and after Soaking Overnight before Cooking Acceptable eating consistency for children of approximately 1-3 years


The viscosity-reducing effect of using germinated cereals for preparing gruels intended for infants and small children has been reported earlier by us (4, 7) and others (15), and is evidently also recognized in some traditional communities (7, 8). One might ask why this beneficial practice is not more widely used, considering that the technology and raw materials are common in most village situations, at least in sub-Saharan Africa. One reason may be that the preparation of germinated flours is a rather time-consuming operation, and if the work burden of women is already heavy ( 16), new tasks may not easily be accepted. In such cases, our proposed method of using germinated flour as an additive to liquefy the prepared gruel may be a viable option, because germinated flour needs to be prepared less often, and small portions can actually be set aside when local beer is produced. There is also reason to believe that the other preparations that have amylolytic activity can be used for reducing the viscosity of starch-based gruels in the same manner as we have described for germinated cereals. Such preparations could include fermented products like cassava (17) or cultured milk, for example. These could also easily be prepared and used at village/household level, and open up further possibilities for promoting high-density foods in starchy staple areas.

TABLE 1 Estimated Consumed Volume of White Local Sorghum Gruels Needed to Cover 60 Per Cent of Daily Energy Needs for a One-Year-Old Child

  Ungerminated Flour Germinated Flour Germinated Flour Added to Ungerminated Flour*
9 flour**/100 g gruel 8 25 17
Viscosity of gruels (cps) 1,000 1,000 1,000
Volume to cover 60 % of daily      
energy needs (ml)*** 2,500 800 1,200

* 5% of total dry matter.
** Energy content of sorghum flour, 350 kcal/100 g,
*** Energy requirements for a one-year-old child,1,180 kcal/day (13).


We are grateful to Brita Brandtzaeg, Björn Ljungquist, and Olof Mellander for valuable advice in preparing this Paper.


1. "Nutritional Problems in the Weaning Period: Report of a Seminar in Addis Ababa, Ethiopia, 3-15 March 1969." Monograph No. 11, J. Trop. Pediat 16: 211 (1970).

2. Å. Hellstrom, A.M. Hermansson, A, Karlsson, B. Liungquist, O. Mellander, and U. Svanberg, "Dietary Bulk as a Limiting Factor for Nutrient Intake in Pre-School Children. II. Consistency as Related to Dietary Bulk-A Model Study." J. Trop Pediat 27: 127 (1981).

3. B. Ljungquist, O. Mellander, and U. Svanberg, "Dietary Bulk as a Limiting Factor for Nutrient Intake in Pre-School Children. I. A Problem Description," J. Trop. Pediat 27: 68 (1981).

4. B. Brandtzaeg, N. G. Malleshi, U. Svanberg, H. S. B. Desikachar, and O. Mellander, "Dietary Bulk as a Limiting Factor for Nutrient Intake in Pre-School Children. III. Studies of Malted Flour from Ragi, Sorghum, and Green Gram." J. Trop. Pediat 27: 184 (1981).

5. A. Karlsson and U. Svanberg, "Dietary Bulk as a Limiting Factor for Nutrient Intake in Pre-School Children. IV. Effect of Digestive Enzymes on the Viscosity of Starch-Based Weaning Foods," J. Trop. Pediat 28: 230 (1982).

6. E. Orr, The Use of Protein-Rich Foods for the Relief of Malnutrition in Developing Countries: An Analysis of Experience, Tropical Products Institute, G73, London (1972).

7. B. Brandzaeg, "Nutritional and Technological Evaluation of Malted Flours from Ragi, Sorghum, and Green Gram for Local Processing of Supplementary and Weaning Foods," Report to the World Hunger Programme, Post-Harvest Food Conservation and Processing, The United Nations University, Tokyo (1979).

8. J. Chisawilo and F. Wandema, "Traditional Food Processing and Preservation," Tanzania Food and Nutrition Centre (TFNC) Report No. 640, Dar-es-Salaam, Tanzania (1980).

9. D. R. Lineback and S. Ponpipom, "Effects of Germination of Wheat, Oats, and Pearl Millet on Alphe-Amylase Activity and Starch Degradation," Die Stärke 29: 52 (1977).

10. V. S. Sheorain and D. S. Wagle, ''Beta-Amylase Activity in Germinated Bajra and Barley Varieties," J. Food Sci. Tech. 10: 184 (1973).

11. M. Tamir and E. Alumot, "Inhibition of Digestive Enzymes by Condensed Tannins from Green and Ripe Carobs," J. Sci. Food Agric. 20: 199 (1969).

12. B. L. Milic, S. Stojanovic, and N. Vucurevic, "Lucerne Tannins, III. Isolation of Tannins from Lucerne, Their Nature and Influence on the Digestive Enzymes in Vitro," J. Sci. Food Agric. 23: 1157 (1972).

13. Energy and Protein Requirements. WHO Tech. Rep. Ser. No, 522, WHO, Geneva (1973).

14. C. Dearden, P. Harman, and D. Morley, "Eating More Fats and Oils as a Step Towards Overcoming Malnutrition," Tropical Doctor 10: 137 (1980).

15. H. S. R. Desikachar, "Development of Weaning Foods with High Caloric Density and Low Hot-Paste Viscosity Using Traditional Technologies," UNU Food and Nutrition Bulletin 2 (4): 21 (1980).

16. B. Brandtzaeg, "The Role and Status of Women in Post Harvest Food Conservation," UNU food and Nutrition Bulletin 4 (1): 33 (1982).

17. C. V. Ramakrishnan, "Studies on Indian Fermented Foods, "Baroda J. Nutr. 6:1 (1979),

Contents - Previous - Next