Contents - Previous - Next

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

Food science

Egg flotation method: its assessment and development into a household technology for the accurate preparation of pickling brine in developing countries
Standardized food terminology: an essential element for preparing and using food consumption data on an international basis

Egg flotation method: its assessment and development into a household technology for the accurate preparation of pickling brine in developing countries

Aftim Acra and May Jurdi
Department of Environmental Health, Faculty of Health Sciences, American University of Beirut, Beirut, Lebanon


The domestic pickling of fresh vegetables in brine or a blend of brine and vinegar, an ancient method of food preservation, continues to be a popular practice in many parts of the world. However, preparation of the brine at the right concentration constitutes a practical problem among rural communities deprived of appropriate measuring devices. This shortcoming has been resolved in Lebanon and some other countries in the Middle East by applying the traditional egg flotation method (EFM) based on Archimedes' principle.

The procedure entails putting the desired amount of water into a vessel; immersing a hen's egg; gradually adding common salt, usually in the crude form; and, to dissolve the salt, carefully stirring after each addition until the egg just floats.

Experienced housewives insist on using freshly laid eggs to achieve satisfactory and uniform results. But, with the advent of modern poultry farming in the past few decades, fresh eggs from native chickens have gradually become scarcer and are now rarely available even in rural areas.

The impact of these changes on the use of EFM as an aid in the preparation of pickling brine warrants appraisal, particularly if this simple method is to remain in vogue under the present conditions. The need to evaluate the method is supported by studies on the use of EFM as an indirect means of measuring the shell thickness of poultry eggs. These studies have shown that physical characteristics of eggs such as weight and specific gravity tend to vary widely with each hen, breed, strain, diet, and season, as well as the age of an egg [4]. These observations suggest that the domestic use of EFM for pickling brine preparation may not be very reliable.

The purpose of the present study was to assess EFM using a variety of eggs in salt solutions of various concentrations in order to determine the optimum specific gravity and other characteristics of a standardized artificial egg for household use.


A total of 118 hens' eggs purchased from various sources were used in the experiments. Among these were 33 eggs obtained from two domestic sources, where chickens were reared and fed under non-commercial conditions, while the rest were purchased in small batches from several shops in Beirut. Most, if not all, of these eggs are believed to have been produced by various breeds of Leghorn hens that have dominated chicken production in this region in the past few decades. For experimental purposes, these eggs were distributed into categories according to source of production (commercial or domestic), shell colour (white or brown), and number of yolks. In addition, 72 quails' eggs were included in the study, as this kind of egg has recently been introduced into the local market.

To minimize possible changes, the unrefrigerated eggs were tested for all parameters, except for the measurement of circumference, on the day they were received at the laboratory. Circumferences were measured the next day. Neither the type of hen nor the age of each egg could be ascertained. This shortcoming, however, does reflect the actual situation that housewives face when using eggs obtained from commercial sources for pickling brine preparation.


The long (L) and short (S) circumferences of each egg were measured by tightly winding sewing thread around the egg, allowing for some overlap where the thread was cut at a convenient point with a sharp scalpel. The length of the thread cuttings were measured by stretching them tightly on a ruler. The mean of two readings was recorded.

Weight and Volume

Each egg was weighed with an Ohaus Dial-0-Gram scale.

TABLE 1. Means and standard deviations for egg characteristics and for the egg flotation results distributed by kind of bird and category of eggs


Egg characteristics

Egg flotation (EFM)

Categorya and (no.) Circumferences (cm) Weight(g) Volume (cm) Specific gravity at 2020C Brine concentration (percentage w/v) Brine specific gravity at 2020C
Long(L) Short(S) Ratio(L/S)
A. Hens' eggs
CB 15.99 13.86 1.16 63.61 59.43 1.070 11.852 1.076
(27) 0.41 0.45 0.03 4.63 4.17 0.007 0.924 0.006
CW 15.92 13.53 1.18 61.47 56.92 1.077 13.365 1.086
(30 0.59 0.43 0.03 4.88 4.76 0.020 1.012 0.006
CB + CW 15.97 13.69 1.17 62.15 58.11 1.076 12.641 1.081
(57) 0.52 0.46 0.03 4.73 4.63 0.013 1.230 0.008
D 15.35 13.14 1.17 55.40 51.05 1.085 14.006 1.090
(33) 0.53 0.40 0.04 4.90 4.54 0.011 1.032 0.007
SY 15.73 13.48 1.17 59.89 55.52 1.079 13.146 1.084
(90) 0.59 0.52 0.04 5.93 5.71 0.013 1.328 0.008
DY 16.70 13.36 1.25 64.39 60.59 1.063 11.596 1.074
(28) 0.37 0.36 0.03 3.89 3.78 0.009 0.733 0.005
SY + DY 15.96 13.45 1.19 60.96 56.72 1.075 12.778 1.082
(118) 0.69 0.48 0.05 5.83 5.73 0.014 1.380 0.008
B. Quails' eggs
Q 8.84 7.88 1.12 10.40 9.65 1.090 10.751 1.069
(72) 0.36 0.32 0.03 1.56 1.17 0.052 0.888 0.006

a. Categories of eggs are designated by symbols D = domestic; CB = commericial, brown shells CW = commercial, white shells CB + CW = commercial egg total: SY = single-yolk hens' eggs DY = double-yolk hens' eggs; Q = quails' eggs.

The egg volume was determined by weighing a rigid plastic container (100 ml capacity) provided with a tightly fitting screw cap completely filled with distilled water at 20C and carefully capped to avoid entrapped air bubbles or an air gap.

Each egg was at room temperature (20 2C) and was immersed in the water inside the container, which was again refilled with distilled water and tightly capped under the specified conditions. The container and its contents were then weighed after wiping its external surfaces, and the volume of the egg was estimated from the weight of the distilled water displaced. The specific gravity was computed in each case from the weight and volume obtained.

Appraisal of EFM

A series of 23 solutions differing in salt concentration by 0.30 per cent weight per volume (w/v) was prepared with unrefined common salt and tap water at room temperature (20 2C) to simulate domestic practices. The salt concentrations ranged from 8.75 to 15.65 per cent w/v. Each egg was dipped into the salt solutions and carefully blotted with absorbent paper between transfers. The concentration Per cent w/v) of the solution at which an egg just floated — when the broad end barely emerged at the surface of the salt solution as viewed from above — was recorded, as well as the specific gravity at 20 2C of the same solution as determined with a precision hydrometer calibrated at 15.6C.

Improvised Hydrometers

Several hydrometers were improvised to simulate eggs intended for use as indicators of pickling brine concentration. The hydrometers were used to determine the precision of the EFM results.

The prototypic hydrometers were prepared from 20 ml glass bottles provided with rubber stoppers sealed with aluminium seals. Such bottles containing normal saline solutions are commonly found in hospitals.

Each of the bottles was calibrated against a salt solution with a concentration corresponding to one of those in the series prepared for egg flotation testing. For calibration, tap water was injected into the bottle through the rubber cap by means of a syringe and needle to bring the total weight to a value that would yield a specific gravity equivalent to that of a given salt solution.

TABLE 2. Levels of significance of differences between the means of two egg categories as assessed by the t-test

Categories compared and (no.)a Circumferences (cm) Weight (g) Volume (cm3) Specific gravity at 20 2c Brine concentration (percentage w/v) Brine specific gravity at 202C
Long Short Ratio
(L) (S) (L/S)
CB (27 v. CW (30) NS ** * NS * ** *** ***
CB+CW(57) v. D(33) *** *** NS *** *** *** *** ***
SY(90) v. DY(28) *** NS *** *** *** *** *** ***
SY+DY v. Q *** *** *** *** *** *** *** ***

* = P<0 05; ** = P<0 01; *** = P<0.001; NS = not significant.
a. Symbols defined in table 1.

After each injection of water the bottle was placed in a one-litre beaker containing brine with a specified salt content. Water in the bottle was adjusted by adding or removing water depending upon whether the bottle sank or rose when placed in the beaker. This adjustment process was repeated as necessary until the bottle just floated. Fine adjustments were made by adding or removing water drop by drop.

Once a bottle was standardized in the manner described, replicates could be more easily prepared by injecting an equivalent volume of water into each of a series of bottles to be calibrated and then making fine adjustments drop by drop. The use of a balance to adjust the total weight to a predetermined value would facilitate the routine procedure before the final adjustment stage using the standard salt solution. With some experience the whole procedure would take only a few minutes to complete.

Statistical Methods

Mean and standard deviation values for each group of data were computed in the usual manner. The t-test was applied to evaluate the significance of differences between group means. Linear regression analysis was used to derive the linear regression equations relating two of the variables in each category of eggs.


Egg Characteristics

A fundamental part of this study involved the determination of the salient physical characteristics of the variety of eggs at present available to a housewife in the local market. Marked differences in some of these characteristics arising from intrinsic or extrinsic factors associated with the egg producing birds could potentially influence the performance of the eggs in household or other applications of EFM.

Apart from these considerations, we wanted to determine the optimum characteristics of an ideal egg that would assure uniform and sufficiently accurate results of domestic applications of EFM. Unfortunately, the literature [4] provided only incomplete information of direct relevance to this task, as investigators had focused on the use of EFM as an indirect measure of the shell thickness of eggs.

Table 1 summarizes the values for means, standard deviations, and coefficients of variation for the following egg characteristics: long circumference (L), short circumference (S), ratio of L/S, weight, volume, specific gravity at 2020C, brine concentration (BC) percentage of weight per volume (w/v), and brine specific gravity at 20 2C.

The mean values for most egg characteristics varied markedly with egg colour-brown or white-and number of yolks, as well as the kind of birds and their mode of rearing, including dietary and nutritional conditions (table 21. The differences were not significant between: the long circumferences (L) of commercial brown and white eggs; the short circumferences (S) of single-yolk and double-yolk eggs; the ratios of L/S of the commercial and the domestic eggs; and the weights of commercial brown and white eggs. The mode of bird rearing, breed or strain of birds, and the number of yolks per egg had the greatest effects on egg characteristics. The differences between the means for eggs from commercial sources and those from domestically reared hens were highly significant, as were the differences between the mean values for hens' eggs with a single yolk and those with double yolks. For all the characteristics tested, quails' eggs were significantly different from hens' eggs.

According to Wells [4], shell pigmentation of hens' eggs varies with the age of the bird. In an attempt to correlate shell mottling with the specific gravity of eggs, Baker and Curtiss [1] studied eggs from Single-comb White Leghorns and concluded that there is very little relation between the two things. In our study the mean values for the specific gravities of brown and white eggs from commercial sources (table 1) were significantly different (table 2, p<0.01).

It is possible that the highly significant differences between the mean values of the study parameters for eggs from commercial sources and those from domestic sources (tables 1 and 2) may be partially attributable to dietary factors.

Egg Flotation Results

Evidence from the literature indicates that the specific gravity of eggs as measured by EFM could vary with certain traits of birds or of their eggs [1-5]. Similar inferences can be drawn from the mean values in table 1 for brine concentration (percentage of w/v) and brine specific gravity (20 2C). The levels of statistical validity of the differences between these means as assessed by the test are presented in table 2. The graphs in figure 1 represent the linear relationship between the differences in brine concentration (percentage of w/v) and egg specific gravity (20 2C) for different kinds of hens' eggs.

With a set of 38 freshly laid eggs from a mixed collection of hens, Tyler and Geake [3] showed that the mean specific gravity as measured by EFM changed from 1.0852 to 1.0817 in 24 hours. This represented a decrease of about 3 per cent, which corresponded to a drop of about 4 per cent in the brine concentration, from 13.31 to 12.76 per cent w/v as computed from the conversion expression in figure 1. These findings raise the issue of the reliability of EFM. Since the age of the experimental eggs used in the present study could not be determined, we could not evaluate this pattern of change in our sample.

FIG. 1. Graphic representation of the linear relationship between the percentage brine concentration at 202C as determined by the EFM and the specific gravity values obtained for the eggs in each category designated by the same symbols as in table 1

Wells published data on the differences in the specific gravities of eggs from individual hens [5] . The mean values ranged from 1.0621 to 1.0857 by EFM for a total of 2,541 eggs from 23 White Leghorn type hens. The overall mean value was 1.0769. We used the conversion equation in figure 1 to compute the corresponding values for brine concentrations from Wells' data and obtained values for brine concentration of 9,70, 13.39, and 12.02 per cent w/v respectively. Because of the differences between our study and Wells' in experimental design and conditions, a detailed comparison of these results with those in table 1 would be unjustifiable. It is worth noting, however, that despite the differences in the studies, there seems to be some general agreement between the two sets of mean values for brine concentration.

From experiments involving eggs from 20 strains of White Leghorn type hens, Baker and Curtiss concluded that there was a highly significant difference (p=0.01) in egg specific gravity between strains of hens [1] . Unfortunately, no specific gravity values were reported to enable a comparison to be made.

For different breeds and strains of hens, Tyler and Geake observed seasonal fluctuations in eggshell thickness, which are generally considered to be directly related to egg specific gravity [3, 4]. Their data show that maximum and minimum values do occur in winter and summer respectively. in the present study, the experimental work was done in March and April 1984, when the trend in specific gravities was expected to be on the decline. Because of such seasonal fluctuations in egg specific gravity, eggs may not be accurate indicators of the salt content of domestically prepared pickling brine.

EFM has been used routinely by poultry farmers to screen eggs for shell thickness and cracking tendency. Tyler and Geake experimentally determined the results for egg specific gravity using both EFM and the more tedious and time-consuming Archimedes' method and found the results from both methods to be in close agreement [2]. Linear regression analysis of their specific gravity data revealed a statistically significant correlation between the two methods (n=6, r=0.9914; p<,0.005). Their observations were confirmed by the high linear correlation obtained in the present study, when the specific gravity of brine measured by EFM was regressed against the specific gravity of eggs obtained by the Archimedes' method (n=90, r=0.6335, p<,0.005). The sample consisted of single-yolk hens' eggs. Similar results were obtained for the various categories of hens' eggs, but the eggs from quails failed to show a statistically valid correlation (n=72, r=-0.0127, p>0.25). In fact, all of the measured characteristics, as well as the EFM performance of quails' eggs, were distinctly different from those for hens' eggs (tables 1 and 2).

TABLE 3. Linear regression expressions relating pairs of parameters for single-yolk hens, eggs

Expressionsa Correlation coefficients (r)b
Sp. gr. = 1.0018 + 0.0068 C 0.6405
V = 237.7326 - 168.8620 Sp. gr. - 0.3581
W = 2.5607 + 1.0326 v 0.9937
L = 10.7191 + 0.0905 V 0.8562
S = 0.1507 + 0.0780 V 0.8702

a. sp. gr. = egg specific gravity at 202C; C = per cent brine concentration: V = egg volume (cm3); W = egg weight ago; L = long circumference (cm); S = short circumference (cm).
b. Significant at P<0.005 (n = 90).

Improvised Hydrometers

Each of the improvised hydrometers calibrated at brine concentration intervals of 0.3 per cent w/v when used as an egg substitute in repeated EFM trials consistently gave exactly the same results. However, care should be taken to perform the tests at brine temperatures close to that at which the hydrometers were calibrated.

The Artificial Egg

The EFM applied routinely for fresh-egg screening in egg producing poultry farms is sufficiently accurate for the purpose. However, its household application for pickling brine preparation suffers from marked differences in results. The development of an artificial egg of standard dimensions and specific gravity is needed to ensure accuracy and precision in domestic applications.

The linear correlation expressions in table 3 derived from data obtained for 90 eggs with single yolks collected from various sources would be useful for computing values for the essential egg parameters for various brine concentrations. For example, if the desired concentration of the pickling brine is 12.00 per cent w/v at 20C, then the computed values for the parameters of the standard artificial egg would be as follows: specific gravity at 20 C equal to 1.0714; volume equal to 56.184 cm; weight equal to 60.576 9; long circumference equal to 15.80 cm; and short circumference equal to 13.53 cm. The prototype egg should be tested for accuracy by means of the EFM.

The development of an artificial egg that is egg-shaped and white in colour would probably be most appealing to people in the Middle East and would maintain an age-old household tradition.


The authors wish to thank the UNICEF Regional Office for the Middle East and North Africa for funding this project. Funding does not, however, imply that UNICEF necessarily endorses the comments, conclusions, and recommendations made by the authors. The authors are also grateful to Dr. Leila Barraj for her valuable help in the statistical analyses of the data.


1. R. C. Baker and R. Curtiss, "Strain Differences in Egg Shell Mottling, Internal Quality, Shell Thickness, Specific Gravity, and the Interrelationships between These Factors,"Poultry Science" 37: 1086-1090 (1958).

2. C. Tyler and F. H. Geake, "Studies on Egg Shells: XV. Critical Appraisal of Various Methods of Assessing Shell Thickness," Journal of the Science of Food and Agriculture, 12: 281 -289 (1961).

3. C. Tyler and F. H. Geake, "Studies on Egg Shells: XIII. Influence of Individuality, Breed, Season and Age on Certain Characteristics of Egg Shells," Journal of the Science of Food and Agriculture, 11: 535-547 (1960).

4. R. G. Wells, "The Measurement of Certain Egg Quality Characteristics: A Review," in T. C. Carter, ea., Egg Quality: A Study of the Hen's Egg (Oliver & Boyd, Edinburgh, 1968).

5. R. G. Wells, "Egg-Shell Strength: 1. The Relationship between Egg Breakage in the Field and Certain Laboratory Assessments of Shell Strength," British Poultry Science, 8: 131-139 (1967).

UN-ACC Subcommittee on Nutrition Statement on Vitamin A and Mortality

The following statement is published because it represents the best available scientific judgement of a study of great potential significance not yet available for general review. It should be cautioned that these results apply to a population in which vitamin A deficiency is severe enough to be associated with many cases of xeropthalmia and in which there is a relatively high morbidity and mortality from infectious disease.

The extent to which significant effects will be observed in populations with lesser severities of vitamin A deficiency and different patterns of infectious disease remains to be determined. Fortunately there are a number of studies now being initiated that will provide further information on these issues.

* * *

In response to a request from SON members, the AGN has reviewed a draft manuscript of an important but still unpublished study in Indonesia -"Impact of Vitamin A Supplementation on Childhood Mortality: A Randomized Controlled Community Trial," by A. Sommer, I. Tarwotjo, E. Djunaedi, K. P. West, A. A. Loeden, R. Tilden, L. Mele, and the ACEH Study Group. This study presents an analysis of the effects of a high-dosage vitamin A supplementation programme, of the type normally implemented for the control of xerophthalmia in Indonesia, on mortality in young children. As a result of this review, the AGN concludes that:

1. The approximately 30 per cent difference in mortality in pre-school children (one to five years of age) between treated and control villages was probably attributable to the vitamin A supplementation.

2. There is reason to expect that effects of this magnitude would be seen in other settings with similar conditions-including at least severe vitamin A deficiency with associated xerophthalmia, high prevalences of childhood morbidity and mortality, and a comparable xerophthalmia control programme.

3. Countries mounting high-dosage vitamin A programmes for the control of xerophthalmia should be advised that a reduction in childhood mortality is a reasonable expectation and provides further justification for such programmes.

The AGN notes that confirmatory trials are now in process and urges that these be monitored to ensure that the experience in Indonesia is reproducible in other regions which may have different patterns of morbidity or other conditioning factors.

The AGN expresses the hope that the trials now in process, or if necessary future trials, will also examine the relationship between the severity of vitamin A deficiency in a population and the effect of vitamin A supplementation on morbidity and mortality in pre-school children and infants It is important to find out whether significant effects are seen in populations where vitamin A deficiency is present without a prevalence of xerophthalmia.

Standardized food terminology: an essential element for preparing and using food consumption data on an international basis

Wanda Polacchi
MEDIFOODS, Rome, Italy


The preparation of reliable food composition data requires a precise nomenclature and a detailed description of foods. Even data of good quality can be a source of error if they are derived from foods that are not clearly defined.

Ideally, a food item is defined unequivocally by a food name and a certain number of descriptive terms or "descriptors." The number of descriptors that are necessary depends on the level of aggregation of the food item. For example, different varieties of apples can be aggregated together in the food item "apple," and the varietal descriptors dispensed with. There are, however, some descriptors that are essential, such as the part of the food analysed and any processing the food has undergone. In general, food items that differ with regard to one of these descriptors cannot be aggregated-for example, the seed of a plant should not be aggregated with the leaves of the same plant, and dried and raw fish should not be aggregated.

The first step in preparing food composition tables is the collection of data from the published literature, unpublished documents, and, where possible, direct analysis. In order to use data for a specific food, one needs, at least, a food name and the essential descriptors. Additional descriptors that may be given in the source documents, such as colour, variety, length of storage, and fertilizer use, should be recorded to the extent that they are expected to be necessary. It should be kept in mind that many different kinds of uses may be made of the data, and the recording of descriptors should err on the side of over-recording.


Although in theory it may appear easy to establish or interpret food names and descriptors, in practice ambiguity and lack of specificity are often problems. Several aspects of this problem are illustrated in this section.

1. Non-correspondence between the local name, the English name, and the Latin name of foods. An article may report the names for a food in the local language, in Latin, and in English, but none of these names may correspond. For example, in the article "The Fatty Acid Composition of Edible Marine Fish Oils," a fish is named surmai in the local language (Pakistani), striped mackerel in English, and Pelamys chilensis in Latin [1]. But Pelamys chilensis, which is synonymous with Sarda chiliensis, is the Latin name of the Eastern Pacific bonito, which is distributed in the Eastern Pacific Ocean and therefore is not consumed in Pakistan. Surmai, on the other hand, is the local name of Rastrelliger kanagurta, which in English is called Indian mackerel [21. It is interesting to note that the length of Pelamys chilensis ranges from 47 to 102 cm, while the maximum length of Rastrelliger kanagurta is 35 cm.

2. Use of only the local names for foods. In addition to the difficulty of determining the appropriate English or Latin names, which may not even exist, it can happen that the local name refers to different foods in different areas. For example, battikha in Arabic means watermelon in the Middle East and melon in North Africa.

3. The same English name indicating different foods having different Latin names. The English name catfish is used for several fish, which have the following Latin names: Bagrus bayad, Clarias Iazera, Clupisoma garua, Eutropiichtys vacha, Heteropneustes fossilis, Mystus corsula, Mystus gulio, Mystus vittatus, Silonia silonia, Silurus triostegus, Synodontis spp., and Wallag attu [3]. Most of these fish not only belong to different genera, but the genera are in the following different families: Siluridae, Bagridae, Synodidae Ariadae, Schilbeidae, Claridae.

4. Incorrect translation from one language to another. In two unpublished documents from a nutrition institute, which had been translated from Arabic into English, a single set of nutritive values is ascribed to potatoes in one document and to sweet potatoes in the other. In one of these documents the food name was obviously translated incorrectly.

5. Indeterminacy of the description of the part analysed. For example, a food sample may be referred to as meat, without the part being specified. The indication of the part of the meat is important and should always be given, because the nutritive values of various tissues can be considerably different. For example, the USDA Handbook, no. 8/5, reports the fat content for parts of chicken (broilers or fryers, raw) as: 11.1 per cent in light meat with skin, 18.3 per cent in dark meat with skin, 1.7 per cent in light meat without skin, and 4.3 per cent in dark meat without skin [4]

6. Indeterminacy of description of the process the sample has undergone. A food sample is described as cooked, but the method of cooking is not specified. The indication of the cooking method is important because it can make a

TABLE 1. Description of the same food item by different terms and classification in different groups

Food item Food group Source
1. Bean, mung, green gram, rawa Vegetables [a]
2. Mung bean, Indian bean, red bean, green gram, golden gram (Phaseolus aureus, Vigna radiate), whole seeds, dried Grain resumes and legume products [6]
3. Beans, mung, mature seeds, dry, raw - b [7]
4. Mung bean (Phaseolus aureus, P. mango),
whole seeds, dry
Poises, nuts and oilseeds [8]

a. Dry seed: water content of 12 per cent h. Foods are not classified in food groups hut are listed in alphabetical order. considerable difference to the nutritive values and water content of a food.


More than one food composition data table is frequently needed to find data on unusual foods or to compare different countries or regions in international studies. The diversity in presentation, terminology, and classification often makes it difficult to draw data from different tables Such non-standardization results not only in difficulty in finding information, but also in potential misunderstanding. Some examples of these problems are given below.

1. The same food item is described by varying terms and classified into different groups. Table 1 shows how differently mung beans are described and classified by four different sources.

2. The description and/or classification are confusing. Table 2 shows five different entries for pigeon peas (Cajanus cajan). Item 1 is dry seed, as the water content indicates, but it is classified in the group of vegetables. The refuse description indicates that item 2 is immature seed, but it is classified in the group of pulses, nuts and oilseeds. Item 3 is immature seed, as the water content and refuse description indicate; for items 4 and 5, the description is clear. This example shows that the food description given in different tables may be confusing and that other factors (in this case water content and refuse description) need to be considered in order to identify the food item.

Table 3 shows two items as described by the same source table in confusing and contradictory terms [8]. As indicated by the water content, the seeds of both are dry and uncooked; however, in item 1 the seed is described as "dry" while in item 2 it is described as "raw."

TABLE 2. Confusing classification and apparent inconsistencies in analysis of food items

Food item Food group Water (% of edible portion) Refuse (%) Source
1. Peas, red pigeon, raw Vegetables 10.0 0 [5]
2. Pigeon peas, Congo pea (Cajanus cajan), green immature Pulses, nuts and
69.4 61
3. Pigeon peas (Cajanus cajan), raw Vegetables 65.9 52
4. Pigeon pea (Cajanus cajan), immature pods and seeds, raw Vegetables 68.9   [6]
5. Pigeon pea (Cajanus cajan), immature pods and seeds, raw Vegetables 64.4 38
(Ends, strims, and stems)

TABLE 3. Confusing descriptions and classifications

Food item Food group Water (% of edible portion) Source
Peas, green pea
(Pisum sativum)
1. Whole seeds, dry   11.7 [8]
2. Split pea, withoutseed coat, raw   9.3  
3. Fennel Spices and condiments 90.0 [8]

In the same food table [B] the classification of fennel is confusing, since it is listed under the classificatory heading "Spices and condiments" while also being described as having 90 per cent water. From this and its other values, obviously the data refer to fennel as a vegetable.

Confusing Terminology

In some food composition tables the term "raw" is not used, and if there is no indication of processing the food has not undergone any. In other tables, where the term "raw" is used, sometimes it means "uncooked," sometimes "not dry," and at other times "not processed." In some tables "raw" is used for some foods but not for others.


The problem of food terminology is not the difficulty of finding the best terms or the best ways of classifying foods, but the fact that differing, inconsistent, and often incompatible terminologies are used. Fundamentally, what is needed is a global standardization of terminology and classification. Such an international system would solve many of the problems arising from the misidentification of foods. However, the development of such a system is a very complex and difficult task. Its characteristics must include:

- flexibility in accepting new terms and names;
- flexibility in retrieval of information;
- ease of use and understanding; and
- explicit recipe algorithms where necessary.

The importance of the whole field of food composition data to the modern world of health and nutrition and the growing scarcity of resources demands that procedures be developed to permit the available data that are being gathered to be optimally used. A key aspect of this endeavour is the development of an international standardized food terminology.


1. Q. Khalid et al., The Fatty Acid Composition of Edible Marine Fish Oils," Am, Oil Chem. Sac. J., 45: 247-249 (1968).

2. FAO Fisheries Synopsis, no.125, vol. 2 (FAO, Rome, 1983).

3. United States Department of Agriculture (USDA), Scientific and Common Names of Foods, for Use with the USDA Nutrient Data Bank (USDA, Washington, D.C., 1983).

4. United States Department of Agriculture (USDA), Composition of Foods, ``Poultry Products, Raw, Processed, and Prepared," Agriculture Handbook, no. 8/5 (USDA, Washington, D.C., 1979).

5. A. A. Paul and D. A. T. Southgate, McCance and Widdowson's: The Composition of Foods, 4th revised and extended edition of the MRC Special Report, no, 297 (MRC, London, 1978).

6. United States Department of Health, Education, and Welfare and FAO, Food Composition Tables for Use in East Asia (USDA, Washington, D.C., 1972).

7. United States Department of Agriculture (USDA), Composition of Foods, Raw, Processed, and Prepared, Agriculture Handbook, no. 8 (USDA, Washington, D.C., 1963).

8. The Caribbean Food and Nutrition Institute (CFNI), Food Composition Tables for Use in the English-speaking Caribbean (CFNI, Jamaica, 1974).

9. United States Department of Agriculture (USDA), Composition of Foods: Vegetables and Vegetable Products, Raw, Processed, and Prepared, Agriculture Handbook, no. 8/11 (USDA, Washington, D.C., 1984).

Contents - Previous - Next